diff --git a/.gitattributes b/.gitattributes index e69e05420f877a6883e476670fd40024577ff68e..eddc2e4853225d9802491e8690f00d6e2274c75e 100644 --- a/.gitattributes +++ b/.gitattributes @@ -242,3 +242,8 @@ babylm_shuffle_local3_10M_seed0/runs/checkpoint-400/tokenizer.json filter=lfs di babylm_shuffle_local3_10M_seed0/runs/checkpoint-1300/tokenizer.json filter=lfs diff=lfs merge=lfs -text babylm_shuffle_local3_10M_seed0/runs/checkpoint-100/tokenizer.json filter=lfs diff=lfs merge=lfs -text babylm_shuffle_local3_10M_seed0/runs/checkpoint-1000/tokenizer.json filter=lfs diff=lfs merge=lfs -text +babylm_shuffle_local3_10M_seed0/runs/checkpoint-300/tokenizer.json filter=lfs diff=lfs merge=lfs -text +babylm_shuffle_local3_10M_seed0/runs/checkpoint-1600/tokenizer.json filter=lfs diff=lfs merge=lfs -text +babylm_shuffle_local3_10M_seed0/runs/checkpoint-2100/tokenizer.json filter=lfs diff=lfs merge=lfs -text +babylm_shuffle_control_10M_seed0/artifacts/models--meta-llama--Llama-3.2-3B/blobs/4719a04514ec2f060240711b7c33ab21187cac730ecaba3040b7a0fd95a9cefb filter=lfs diff=lfs merge=lfs -text +babylm_shuffle_control_10M_seed0/artifacts/models--meta-llama--Llama-3.2-3B/blobs/584d8d3e3f82f7964955174dfe5e3b1cf117a9d859f022cfdf7fcb884856e002 filter=lfs diff=lfs merge=lfs -text diff --git a/babylm_shuffle_control_10M_seed0/artifacts/models--meta-llama--Llama-3.2-3B/blobs/4719a04514ec2f060240711b7c33ab21187cac730ecaba3040b7a0fd95a9cefb b/babylm_shuffle_control_10M_seed0/artifacts/models--meta-llama--Llama-3.2-3B/blobs/4719a04514ec2f060240711b7c33ab21187cac730ecaba3040b7a0fd95a9cefb new file mode 100644 index 0000000000000000000000000000000000000000..82a877359be75c9c7ad1226b362efeeabf2edef4 --- /dev/null +++ b/babylm_shuffle_control_10M_seed0/artifacts/models--meta-llama--Llama-3.2-3B/blobs/4719a04514ec2f060240711b7c33ab21187cac730ecaba3040b7a0fd95a9cefb @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:4719a04514ec2f060240711b7c33ab21187cac730ecaba3040b7a0fd95a9cefb +size 1459729952 diff --git a/babylm_shuffle_control_10M_seed0/artifacts/models--meta-llama--Llama-3.2-3B/blobs/584d8d3e3f82f7964955174dfe5e3b1cf117a9d859f022cfdf7fcb884856e002 b/babylm_shuffle_control_10M_seed0/artifacts/models--meta-llama--Llama-3.2-3B/blobs/584d8d3e3f82f7964955174dfe5e3b1cf117a9d859f022cfdf7fcb884856e002 new file mode 100644 index 0000000000000000000000000000000000000000..e362b81bf93cef8c6d96b08171bab63416808c70 --- /dev/null +++ b/babylm_shuffle_control_10M_seed0/artifacts/models--meta-llama--Llama-3.2-3B/blobs/584d8d3e3f82f7964955174dfe5e3b1cf117a9d859f022cfdf7fcb884856e002 @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:584d8d3e3f82f7964955174dfe5e3b1cf117a9d859f022cfdf7fcb884856e002 +size 4965799096 diff --git a/babylm_shuffle_control_10M_seed0/artifacts/models--meta-llama--Llama-3.2-3B/snapshots/13afe5124825b4f3751f836b40dafda64c1ed062/model-00001-of-00002.safetensors b/babylm_shuffle_control_10M_seed0/artifacts/models--meta-llama--Llama-3.2-3B/snapshots/13afe5124825b4f3751f836b40dafda64c1ed062/model-00001-of-00002.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..e362b81bf93cef8c6d96b08171bab63416808c70 --- /dev/null +++ b/babylm_shuffle_control_10M_seed0/artifacts/models--meta-llama--Llama-3.2-3B/snapshots/13afe5124825b4f3751f836b40dafda64c1ed062/model-00001-of-00002.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:584d8d3e3f82f7964955174dfe5e3b1cf117a9d859f022cfdf7fcb884856e002 +size 4965799096 diff --git a/babylm_shuffle_control_10M_seed0/artifacts/models--meta-llama--Llama-3.2-3B/snapshots/13afe5124825b4f3751f836b40dafda64c1ed062/model-00002-of-00002.safetensors b/babylm_shuffle_control_10M_seed0/artifacts/models--meta-llama--Llama-3.2-3B/snapshots/13afe5124825b4f3751f836b40dafda64c1ed062/model-00002-of-00002.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..82a877359be75c9c7ad1226b362efeeabf2edef4 --- /dev/null +++ b/babylm_shuffle_control_10M_seed0/artifacts/models--meta-llama--Llama-3.2-3B/snapshots/13afe5124825b4f3751f836b40dafda64c1ed062/model-00002-of-00002.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:4719a04514ec2f060240711b7c33ab21187cac730ecaba3040b7a0fd95a9cefb +size 1459729952 diff --git a/babylm_shuffle_control_10M_seed0/runs/checkpoint-1800/scheduler.pt b/babylm_shuffle_control_10M_seed0/runs/checkpoint-1800/scheduler.pt new file mode 100644 index 0000000000000000000000000000000000000000..384b3da234871141f0964eff67b960336491a262 --- /dev/null +++ b/babylm_shuffle_control_10M_seed0/runs/checkpoint-1800/scheduler.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:9c1df20c7d32914c4894472a32b4a9b1d35b678dd12cba3553d9d38bba385a50 +size 627 diff --git a/babylm_shuffle_local10_10M_seed0/runs/checkpoint-1800/model-00001-of-00002.safetensors b/babylm_shuffle_local10_10M_seed0/runs/checkpoint-1800/model-00001-of-00002.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..352a4ee05183da6f02584074dd22011c2ca7b6fd --- /dev/null +++ b/babylm_shuffle_local10_10M_seed0/runs/checkpoint-1800/model-00001-of-00002.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:5883867978ff8842f42c50b608477cfb69fa91cf08fd43f7290722a7f9b6fced +size 4965798912 diff --git a/babylm_shuffle_local10_10M_seed0/runs/checkpoint-1900/model-00001-of-00002.safetensors b/babylm_shuffle_local10_10M_seed0/runs/checkpoint-1900/model-00001-of-00002.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..e4033c2df09ca888268d5484be2e7f642c712c9a --- /dev/null +++ b/babylm_shuffle_local10_10M_seed0/runs/checkpoint-1900/model-00001-of-00002.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:d86ebae8b3547a3d33bd508d92e946fc9bbe995342209d7090a61c8b044bbcee +size 4965798912 diff --git a/babylm_shuffle_local10_10M_seed0/runs/checkpoint-2400/model-00001-of-00002.safetensors b/babylm_shuffle_local10_10M_seed0/runs/checkpoint-2400/model-00001-of-00002.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..e7d30027c0bcac08a84ebb5fb0fc108769f437cc --- /dev/null +++ b/babylm_shuffle_local10_10M_seed0/runs/checkpoint-2400/model-00001-of-00002.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:a4a187617b252e8e556c409ededadfa25ac636fac58977289f4c2b9346570a8c +size 4965798912 diff --git a/babylm_shuffle_local3_10M_seed0/runs/checkpoint-1100/model-00002-of-00002.safetensors b/babylm_shuffle_local3_10M_seed0/runs/checkpoint-1100/model-00002-of-00002.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..652de82728050f49d5e47fdffdee7f4605df1f36 --- /dev/null +++ b/babylm_shuffle_local3_10M_seed0/runs/checkpoint-1100/model-00002-of-00002.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:323ca22ec7f619420727a70c9a7936593bdf11dbfe549c4d581989c9df88faa9 +size 2247734920 diff --git a/babylm_shuffle_local3_10M_seed0/runs/checkpoint-1600/rng_state_0.pth b/babylm_shuffle_local3_10M_seed0/runs/checkpoint-1600/rng_state_0.pth new file mode 100644 index 0000000000000000000000000000000000000000..5509f0e1ece25901639e38b63337f0f28e33b98d --- /dev/null +++ b/babylm_shuffle_local3_10M_seed0/runs/checkpoint-1600/rng_state_0.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:0123f1b841cbb90ad2cc1a10c38027bc63872dc02a831b18d254a6186f0ffff0 +size 16567 diff --git a/babylm_shuffle_local3_10M_seed0/runs/checkpoint-1600/rng_state_1.pth b/babylm_shuffle_local3_10M_seed0/runs/checkpoint-1600/rng_state_1.pth new file mode 100644 index 0000000000000000000000000000000000000000..bdaddc4981c690246dfa665b10da40b38a679bc3 --- /dev/null +++ b/babylm_shuffle_local3_10M_seed0/runs/checkpoint-1600/rng_state_1.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:86e82dbca3ee98b4aff4d87464271202c99a718b90530a7719851b8b6a191547 +size 16567 diff --git a/babylm_shuffle_local3_10M_seed0/runs/checkpoint-1600/rng_state_2.pth b/babylm_shuffle_local3_10M_seed0/runs/checkpoint-1600/rng_state_2.pth new file mode 100644 index 0000000000000000000000000000000000000000..fd1ca9202bb0004e7be63108d043e05db918daf3 --- /dev/null +++ b/babylm_shuffle_local3_10M_seed0/runs/checkpoint-1600/rng_state_2.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:fdc0a29e8caac820aa227d61191352d70074e7bc59c8656d34721d88d568ccb2 +size 16567 diff --git a/babylm_shuffle_local3_10M_seed0/runs/checkpoint-1600/scheduler.pt b/babylm_shuffle_local3_10M_seed0/runs/checkpoint-1600/scheduler.pt new file mode 100644 index 0000000000000000000000000000000000000000..dc327edece7a7a7c40b3b9a69e4152e20661e7ba --- /dev/null +++ b/babylm_shuffle_local3_10M_seed0/runs/checkpoint-1600/scheduler.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:7f530e5786027673f67db19426a9a978b2bdb02bc1800d92b2b2f57eeaca551c +size 627 diff --git a/babylm_shuffle_local3_10M_seed0/runs/checkpoint-1600/tokenizer.json b/babylm_shuffle_local3_10M_seed0/runs/checkpoint-1600/tokenizer.json new file mode 100644 index 0000000000000000000000000000000000000000..309789d5ccd06f418fbcfb7472a6a9affda68101 --- /dev/null +++ b/babylm_shuffle_local3_10M_seed0/runs/checkpoint-1600/tokenizer.json @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:d422bc7d90212ffb8070a743cdbe8ff2dfc5634b1fce8f2e3fa671ee441989eb +size 17210374 diff --git a/babylm_shuffle_local3_10M_seed0/runs/checkpoint-1600/training_args.bin b/babylm_shuffle_local3_10M_seed0/runs/checkpoint-1600/training_args.bin new file mode 100644 index 0000000000000000000000000000000000000000..6db59a81c34edebb00ef6a59f5669f3babfbb622 --- /dev/null +++ b/babylm_shuffle_local3_10M_seed0/runs/checkpoint-1600/training_args.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:484dadb3af9420a36d698e37e41887c1bddbd89850c9b141ca33808576dc9968 +size 6011 diff --git a/babylm_shuffle_local3_10M_seed0/runs/checkpoint-2100/rng_state_0.pth b/babylm_shuffle_local3_10M_seed0/runs/checkpoint-2100/rng_state_0.pth new file mode 100644 index 0000000000000000000000000000000000000000..9829253a45a30486360d23ef44a31e49a0fbb37c --- /dev/null +++ b/babylm_shuffle_local3_10M_seed0/runs/checkpoint-2100/rng_state_0.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:e1200c6256bb3c5211e1fa1cfcea6721e1b8871fd3c8953a59bf49a59f67404c +size 16567 diff --git a/babylm_shuffle_local3_10M_seed0/runs/checkpoint-2100/rng_state_1.pth b/babylm_shuffle_local3_10M_seed0/runs/checkpoint-2100/rng_state_1.pth new file mode 100644 index 0000000000000000000000000000000000000000..ed05b8eb12ac744b29eb407adca7ab9af789a745 --- /dev/null +++ b/babylm_shuffle_local3_10M_seed0/runs/checkpoint-2100/rng_state_1.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:4b59dfa224b09069bb1265a6a5cb86c22c34f53669b17ebd9335621f959bf7a4 +size 16567 diff --git a/babylm_shuffle_local3_10M_seed0/runs/checkpoint-2100/rng_state_2.pth b/babylm_shuffle_local3_10M_seed0/runs/checkpoint-2100/rng_state_2.pth new file mode 100644 index 0000000000000000000000000000000000000000..a26252516294710c0550a7751d8c4bc072b73762 --- /dev/null +++ b/babylm_shuffle_local3_10M_seed0/runs/checkpoint-2100/rng_state_2.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:316a81119e1ab6e1b3e5b9d890ce1d0ca9ea9223094844fdcf8b7d9f0f4bb6c2 +size 16567 diff --git a/babylm_shuffle_local3_10M_seed0/runs/checkpoint-2100/scheduler.pt b/babylm_shuffle_local3_10M_seed0/runs/checkpoint-2100/scheduler.pt new file mode 100644 index 0000000000000000000000000000000000000000..f58e49518f66b11f5c4d07a04bd0a5f134c1d2e6 --- /dev/null +++ b/babylm_shuffle_local3_10M_seed0/runs/checkpoint-2100/scheduler.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:6bfa44765a8aa41d7ea84057f001ffa2e36d36de41206bec5ed89cb556546105 +size 627 diff --git a/babylm_shuffle_local3_10M_seed0/runs/checkpoint-2100/tokenizer.json b/babylm_shuffle_local3_10M_seed0/runs/checkpoint-2100/tokenizer.json new file mode 100644 index 0000000000000000000000000000000000000000..309789d5ccd06f418fbcfb7472a6a9affda68101 --- /dev/null +++ b/babylm_shuffle_local3_10M_seed0/runs/checkpoint-2100/tokenizer.json @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:d422bc7d90212ffb8070a743cdbe8ff2dfc5634b1fce8f2e3fa671ee441989eb +size 17210374 diff --git a/babylm_shuffle_local3_10M_seed0/runs/checkpoint-2100/training_args.bin b/babylm_shuffle_local3_10M_seed0/runs/checkpoint-2100/training_args.bin new file mode 100644 index 0000000000000000000000000000000000000000..6db59a81c34edebb00ef6a59f5669f3babfbb622 --- /dev/null +++ b/babylm_shuffle_local3_10M_seed0/runs/checkpoint-2100/training_args.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:484dadb3af9420a36d698e37e41887c1bddbd89850c9b141ca33808576dc9968 +size 6011 diff --git a/babylm_shuffle_local3_10M_seed0/runs/checkpoint-2736/model-00002-of-00002.safetensors b/babylm_shuffle_local3_10M_seed0/runs/checkpoint-2736/model-00002-of-00002.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..b9c342a118d2b85299f0105063f4778ba0d41117 --- /dev/null +++ b/babylm_shuffle_local3_10M_seed0/runs/checkpoint-2736/model-00002-of-00002.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:f641aeadffe6d9e9995b6884e270a33fcf8d72d361646c75c15b167c62a0f17b +size 2247734920 diff --git a/babylm_shuffle_local3_10M_seed0/runs/checkpoint-300/rng_state_0.pth b/babylm_shuffle_local3_10M_seed0/runs/checkpoint-300/rng_state_0.pth new file mode 100644 index 0000000000000000000000000000000000000000..9015a2b291768ac1cbe0fb417354be7e82849ff8 --- /dev/null +++ b/babylm_shuffle_local3_10M_seed0/runs/checkpoint-300/rng_state_0.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:c47a2aec04adfb2b774bed1d02a5453693feb66e4636bdf7cd908de748732842 +size 16567 diff --git a/babylm_shuffle_local3_10M_seed0/runs/checkpoint-300/rng_state_1.pth b/babylm_shuffle_local3_10M_seed0/runs/checkpoint-300/rng_state_1.pth new file mode 100644 index 0000000000000000000000000000000000000000..a17e6802aa777eb35577f53d24a38a0381fecbb7 --- /dev/null +++ b/babylm_shuffle_local3_10M_seed0/runs/checkpoint-300/rng_state_1.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:1e375d2646839f0fc8836323cd9f6c6fec1f1316a3f21b7366755c68ab3c469c +size 16567 diff --git a/babylm_shuffle_local3_10M_seed0/runs/checkpoint-300/rng_state_2.pth b/babylm_shuffle_local3_10M_seed0/runs/checkpoint-300/rng_state_2.pth new file mode 100644 index 0000000000000000000000000000000000000000..8aab09805997622aec2edae7b565ed82010cfd53 --- /dev/null +++ b/babylm_shuffle_local3_10M_seed0/runs/checkpoint-300/rng_state_2.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:6802d0e782c931221303c74054cdedbab70d850d6310dc553b393a4f3f848807 +size 16567 diff --git a/babylm_shuffle_local3_10M_seed0/runs/checkpoint-300/tokenizer.json b/babylm_shuffle_local3_10M_seed0/runs/checkpoint-300/tokenizer.json new file mode 100644 index 0000000000000000000000000000000000000000..309789d5ccd06f418fbcfb7472a6a9affda68101 --- /dev/null +++ b/babylm_shuffle_local3_10M_seed0/runs/checkpoint-300/tokenizer.json @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:d422bc7d90212ffb8070a743cdbe8ff2dfc5634b1fce8f2e3fa671ee441989eb +size 17210374 diff --git a/babylm_shuffle_local3_10M_seed0/runs/checkpoint-300/training_args.bin b/babylm_shuffle_local3_10M_seed0/runs/checkpoint-300/training_args.bin new file mode 100644 index 0000000000000000000000000000000000000000..6db59a81c34edebb00ef6a59f5669f3babfbb622 --- /dev/null +++ b/babylm_shuffle_local3_10M_seed0/runs/checkpoint-300/training_args.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:484dadb3af9420a36d698e37e41887c1bddbd89850c9b141ca33808576dc9968 +size 6011 diff --git a/babylm_shuffle_local3_10M_seed0/runs/checkpoint-600/model-00002-of-00002.safetensors b/babylm_shuffle_local3_10M_seed0/runs/checkpoint-600/model-00002-of-00002.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..9156e8bd020b25f8cf2c873a1fd2ba16d5832f86 --- /dev/null +++ b/babylm_shuffle_local3_10M_seed0/runs/checkpoint-600/model-00002-of-00002.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:0b5880959a2fa59d270f0821857945e44507e78908fd22debd420971a72ae4c2 +size 2247734920 diff --git a/babylm_shuffle_nondeterministic_10M_seed0/runs/checkpoint-1100/config.json b/babylm_shuffle_nondeterministic_10M_seed0/runs/checkpoint-1100/config.json new file mode 100644 index 0000000000000000000000000000000000000000..fb0f9a53a507ffcdb60410deb71d60ef801bf350 --- /dev/null +++ b/babylm_shuffle_nondeterministic_10M_seed0/runs/checkpoint-1100/config.json @@ -0,0 +1,36 @@ +{ + "_name_or_path": "meta-llama/Llama-3.2-3B", + "architectures": [ + "LlamaForCausalLM" + ], + "attention_bias": false, + "attention_dropout": 0.0, + "bos_token_id": 128000, + "eos_token_id": 128001, + "head_dim": 128, + "hidden_act": "silu", + "hidden_size": 3072, + "initializer_range": 0.02, + "intermediate_size": 8192, + "max_position_embeddings": 131072, + "mlp_bias": false, + "model_type": "llama", + "num_attention_heads": 24, + "num_hidden_layers": 28, + "num_key_value_heads": 8, + "pretraining_tp": 1, + "rms_norm_eps": 1e-05, + "rope_scaling": { + "factor": 32.0, + "high_freq_factor": 4.0, + "low_freq_factor": 1.0, + "original_max_position_embeddings": 8192, + "rope_type": "llama3" + }, + "rope_theta": 500000.0, + "tie_word_embeddings": true, + "torch_dtype": "float16", + "transformers_version": "4.45.1", + "use_cache": true, + "vocab_size": 128256 +} diff --git a/babylm_shuffle_nondeterministic_10M_seed0/runs/checkpoint-1100/generation_config.json b/babylm_shuffle_nondeterministic_10M_seed0/runs/checkpoint-1100/generation_config.json new file mode 100644 index 0000000000000000000000000000000000000000..9c389b841a9a29ddd904e02d1eb0e08dcce82ad9 --- /dev/null +++ b/babylm_shuffle_nondeterministic_10M_seed0/runs/checkpoint-1100/generation_config.json @@ -0,0 +1,9 @@ +{ + "_from_model_config": true, + "bos_token_id": 128000, + "do_sample": true, + "eos_token_id": 128001, + "temperature": 0.6, + "top_p": 0.9, + "transformers_version": "4.45.1" +} diff --git a/babylm_shuffle_nondeterministic_10M_seed0/runs/checkpoint-1100/latest b/babylm_shuffle_nondeterministic_10M_seed0/runs/checkpoint-1100/latest new file mode 100644 index 0000000000000000000000000000000000000000..22cd5c3402316b70299aed2025d7943595f5d495 --- /dev/null +++ b/babylm_shuffle_nondeterministic_10M_seed0/runs/checkpoint-1100/latest @@ -0,0 +1 @@ +global_step1100 \ No newline at end of file diff --git a/babylm_shuffle_nondeterministic_10M_seed0/runs/checkpoint-1100/model.safetensors.index.json b/babylm_shuffle_nondeterministic_10M_seed0/runs/checkpoint-1100/model.safetensors.index.json new file mode 100644 index 0000000000000000000000000000000000000000..ed64de846d720b9a7859dc20575fea8e8ca51940 --- /dev/null +++ b/babylm_shuffle_nondeterministic_10M_seed0/runs/checkpoint-1100/model.safetensors.index.json @@ -0,0 +1,262 @@ +{ + "metadata": { + "total_size": 7213504512 + }, + "weight_map": { + "lm_head.weight": "model-00002-of-00002.safetensors", + "model.embed_tokens.weight": "model-00001-of-00002.safetensors", + "model.layers.0.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.0.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.0.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.0.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.0.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.0.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.0.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.0.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.0.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.1.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.1.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.1.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.1.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.1.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.1.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.1.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.1.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.1.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.10.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.10.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.10.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.10.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.10.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.10.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.10.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.10.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.10.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.11.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.11.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.11.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.11.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.11.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.11.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.11.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.11.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.11.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.12.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.12.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.12.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.12.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.12.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.12.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.12.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.12.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.12.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.13.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.13.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.13.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.13.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.13.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.13.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.13.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.13.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.13.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.14.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.14.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.14.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.14.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.14.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.14.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.14.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.14.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.14.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.15.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.15.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.15.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.15.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.15.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.15.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.15.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.15.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.15.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.16.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.16.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.16.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.16.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.16.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.16.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.16.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.16.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.16.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.17.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.17.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.17.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.17.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.17.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.17.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.17.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.17.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.17.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.18.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.18.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.18.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.18.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.18.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.18.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.18.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.18.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.18.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.19.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.19.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.19.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.19.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.19.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.19.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.19.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.19.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.19.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.2.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.2.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.2.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.2.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.2.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.2.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.2.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.2.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.2.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.20.input_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.20.mlp.down_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.20.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.20.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.20.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.20.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.20.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.20.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.20.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.21.input_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.21.mlp.down_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.21.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.21.mlp.up_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.21.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.21.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.21.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.21.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.21.self_attn.v_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.22.input_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.22.mlp.down_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.22.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.22.mlp.up_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.22.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.22.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.22.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.22.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.22.self_attn.v_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.23.input_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.23.mlp.down_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.23.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.23.mlp.up_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.23.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.23.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.23.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.23.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.23.self_attn.v_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.24.input_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.24.mlp.down_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.24.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.24.mlp.up_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.24.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.24.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.24.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.24.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.24.self_attn.v_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.25.input_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.25.mlp.down_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.25.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.25.mlp.up_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.25.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.25.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.25.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.25.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.25.self_attn.v_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.26.input_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.26.mlp.down_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.26.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.26.mlp.up_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.26.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.26.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.26.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.26.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.26.self_attn.v_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.27.input_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.27.mlp.down_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.27.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.27.mlp.up_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.27.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.27.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.27.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.27.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.27.self_attn.v_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.3.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.3.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.3.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.3.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.3.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.3.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.3.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.3.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.3.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.4.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.4.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.4.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.4.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.4.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.4.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.4.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.4.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.4.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.5.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.5.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.5.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.5.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.5.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.5.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.5.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.5.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.5.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.6.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.6.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.6.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.6.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.6.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.6.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.6.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.6.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.6.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.7.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.7.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.7.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.7.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.7.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.7.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.7.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.7.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.7.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.8.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.8.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.8.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.8.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.8.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.8.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.8.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.8.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.8.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.9.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.9.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.9.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.9.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.9.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.9.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.9.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.9.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.9.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.norm.weight": "model-00002-of-00002.safetensors" + } +} diff --git a/babylm_shuffle_nondeterministic_10M_seed0/runs/checkpoint-1100/special_tokens_map.json b/babylm_shuffle_nondeterministic_10M_seed0/runs/checkpoint-1100/special_tokens_map.json new file mode 100644 index 0000000000000000000000000000000000000000..6949c5975ee0e961ef61cf31010dce04df0a03f8 --- /dev/null +++ b/babylm_shuffle_nondeterministic_10M_seed0/runs/checkpoint-1100/special_tokens_map.json @@ -0,0 +1,23 @@ +{ + "bos_token": { + "content": "<|begin_of_text|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false + }, + "eos_token": { + "content": "<|end_of_text|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false + }, + "pad_token": { + "content": "[PAD]", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false + } +} diff --git a/babylm_shuffle_nondeterministic_10M_seed0/runs/checkpoint-1100/tokenizer_config.json b/babylm_shuffle_nondeterministic_10M_seed0/runs/checkpoint-1100/tokenizer_config.json new file mode 100644 index 0000000000000000000000000000000000000000..5684b14e2756b56aa397c1324e239b442162b34a --- /dev/null +++ b/babylm_shuffle_nondeterministic_10M_seed0/runs/checkpoint-1100/tokenizer_config.json @@ -0,0 +1,2070 @@ +{ + "added_tokens_decoder": { + "128000": { + "content": "<|begin_of_text|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128001": { + "content": "<|end_of_text|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128002": { + "content": "<|reserved_special_token_0|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128003": { + "content": "<|reserved_special_token_1|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128004": { + "content": "<|finetune_right_pad_id|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128005": { + "content": "<|reserved_special_token_2|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128006": { + "content": "<|start_header_id|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128007": { + "content": "<|end_header_id|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128008": { + "content": "<|eom_id|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128009": { + "content": "<|eot_id|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128010": { + "content": "<|python_tag|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128011": { + "content": "<|reserved_special_token_3|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128012": { + "content": "<|reserved_special_token_4|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128013": { + "content": "<|reserved_special_token_5|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128014": { + "content": "<|reserved_special_token_6|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128015": { + "content": "<|reserved_special_token_7|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128016": { + "content": "<|reserved_special_token_8|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128017": { + "content": "<|reserved_special_token_9|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128018": { + "content": "<|reserved_special_token_10|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128019": { + "content": "<|reserved_special_token_11|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128020": { + "content": "<|reserved_special_token_12|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128021": { + "content": "<|reserved_special_token_13|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128022": { + "content": "<|reserved_special_token_14|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128023": { + "content": "<|reserved_special_token_15|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128024": { + "content": "<|reserved_special_token_16|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128025": { + "content": "<|reserved_special_token_17|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128026": { + "content": "<|reserved_special_token_18|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128027": { + "content": "<|reserved_special_token_19|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128028": { + "content": "<|reserved_special_token_20|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128029": { + "content": "<|reserved_special_token_21|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128030": { + "content": "<|reserved_special_token_22|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128031": { + "content": "<|reserved_special_token_23|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128032": { + "content": "<|reserved_special_token_24|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128033": { + "content": "<|reserved_special_token_25|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128034": { + "content": "<|reserved_special_token_26|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128035": { + "content": "<|reserved_special_token_27|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128036": { + "content": "<|reserved_special_token_28|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128037": { + "content": "<|reserved_special_token_29|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128038": { + "content": "<|reserved_special_token_30|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128039": { + "content": "<|reserved_special_token_31|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128040": { + "content": "<|reserved_special_token_32|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128041": { + "content": "<|reserved_special_token_33|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128042": { + "content": "<|reserved_special_token_34|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128043": { + "content": "<|reserved_special_token_35|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128044": { + "content": "<|reserved_special_token_36|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128045": { + "content": "<|reserved_special_token_37|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128046": { + "content": "<|reserved_special_token_38|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128047": { + "content": "<|reserved_special_token_39|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128048": { + "content": "<|reserved_special_token_40|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128049": { + "content": "<|reserved_special_token_41|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128050": { + "content": "<|reserved_special_token_42|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128051": { + "content": "<|reserved_special_token_43|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128052": { + "content": "<|reserved_special_token_44|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128053": { + "content": "<|reserved_special_token_45|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128054": { + "content": "<|reserved_special_token_46|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128055": { + "content": "<|reserved_special_token_47|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128056": { + "content": "<|reserved_special_token_48|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128057": { + "content": "<|reserved_special_token_49|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128058": { + "content": "<|reserved_special_token_50|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128059": { + "content": "<|reserved_special_token_51|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128060": { + "content": "<|reserved_special_token_52|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128061": { + "content": "<|reserved_special_token_53|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128062": { + "content": "<|reserved_special_token_54|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128063": { + "content": "<|reserved_special_token_55|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128064": { + "content": "<|reserved_special_token_56|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128065": { + "content": "<|reserved_special_token_57|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128066": { + "content": "<|reserved_special_token_58|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128067": { + "content": "<|reserved_special_token_59|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128068": { + "content": "<|reserved_special_token_60|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128069": { + "content": "<|reserved_special_token_61|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128070": { + "content": "<|reserved_special_token_62|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128071": { + "content": "<|reserved_special_token_63|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128072": { + "content": "<|reserved_special_token_64|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128073": { + "content": "<|reserved_special_token_65|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128074": { + "content": "<|reserved_special_token_66|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128075": { + "content": "<|reserved_special_token_67|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128076": { + "content": "<|reserved_special_token_68|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128077": { + "content": "<|reserved_special_token_69|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128078": { + "content": "<|reserved_special_token_70|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128079": { + "content": "<|reserved_special_token_71|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128080": { + "content": "<|reserved_special_token_72|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128081": { + "content": "<|reserved_special_token_73|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128082": { + "content": "<|reserved_special_token_74|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128083": { + "content": "<|reserved_special_token_75|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128084": { + "content": "<|reserved_special_token_76|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128085": { + "content": "<|reserved_special_token_77|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128086": { + "content": "<|reserved_special_token_78|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128087": { + "content": "<|reserved_special_token_79|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128088": { + "content": "<|reserved_special_token_80|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128089": { + "content": "<|reserved_special_token_81|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128090": { + "content": "<|reserved_special_token_82|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128091": { + "content": "<|reserved_special_token_83|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128092": { + "content": "<|reserved_special_token_84|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128093": { + "content": "<|reserved_special_token_85|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128094": { + "content": "<|reserved_special_token_86|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128095": { + "content": "<|reserved_special_token_87|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128096": { + "content": "<|reserved_special_token_88|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128097": { + "content": "<|reserved_special_token_89|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128098": { + "content": "<|reserved_special_token_90|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128099": { + "content": "<|reserved_special_token_91|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128100": { + "content": "<|reserved_special_token_92|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128101": { + "content": "<|reserved_special_token_93|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128102": { + "content": "<|reserved_special_token_94|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128103": { + "content": "<|reserved_special_token_95|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128104": { + "content": "<|reserved_special_token_96|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128105": { + "content": "<|reserved_special_token_97|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128106": { + "content": "<|reserved_special_token_98|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128107": { + "content": "<|reserved_special_token_99|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128108": { + "content": "<|reserved_special_token_100|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128109": { + "content": "<|reserved_special_token_101|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128110": { + "content": "<|reserved_special_token_102|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128111": { + "content": "<|reserved_special_token_103|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128112": { + "content": "<|reserved_special_token_104|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128113": { + "content": "<|reserved_special_token_105|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128114": { + "content": "<|reserved_special_token_106|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128115": { + "content": "<|reserved_special_token_107|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128116": { + "content": "<|reserved_special_token_108|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128117": { + "content": "<|reserved_special_token_109|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128118": { + "content": "<|reserved_special_token_110|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128119": { + "content": "<|reserved_special_token_111|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128120": { + "content": "<|reserved_special_token_112|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128121": { + "content": "<|reserved_special_token_113|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128122": { + "content": "<|reserved_special_token_114|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128123": { + "content": "<|reserved_special_token_115|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128124": { + "content": "<|reserved_special_token_116|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128125": { + "content": "<|reserved_special_token_117|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128126": { + "content": "<|reserved_special_token_118|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128127": { + "content": "<|reserved_special_token_119|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128128": { + "content": "<|reserved_special_token_120|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128129": { + "content": "<|reserved_special_token_121|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128130": { + "content": "<|reserved_special_token_122|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128131": { + "content": "<|reserved_special_token_123|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128132": { + "content": "<|reserved_special_token_124|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128133": { + "content": "<|reserved_special_token_125|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128134": { + "content": "<|reserved_special_token_126|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128135": { + "content": "<|reserved_special_token_127|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128136": { + "content": "<|reserved_special_token_128|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128137": { + "content": "<|reserved_special_token_129|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128138": { + "content": "<|reserved_special_token_130|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128139": { + "content": "<|reserved_special_token_131|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128140": { + "content": "<|reserved_special_token_132|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128141": { + "content": "<|reserved_special_token_133|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128142": { + "content": "<|reserved_special_token_134|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128143": { + "content": "<|reserved_special_token_135|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128144": { + "content": "<|reserved_special_token_136|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128145": { + "content": "<|reserved_special_token_137|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128146": { + "content": "<|reserved_special_token_138|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128147": { + "content": "<|reserved_special_token_139|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128148": { + "content": "<|reserved_special_token_140|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128149": { + "content": "<|reserved_special_token_141|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128150": { + "content": "<|reserved_special_token_142|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128151": { + "content": "<|reserved_special_token_143|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128152": { + "content": "<|reserved_special_token_144|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128153": { + "content": "<|reserved_special_token_145|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128154": { + "content": "<|reserved_special_token_146|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128155": { + "content": "<|reserved_special_token_147|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128156": { + "content": "<|reserved_special_token_148|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128157": { + "content": "<|reserved_special_token_149|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128158": { + "content": "<|reserved_special_token_150|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128159": { + "content": "<|reserved_special_token_151|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128160": { + "content": "<|reserved_special_token_152|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128161": { + "content": "<|reserved_special_token_153|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128162": { + "content": "<|reserved_special_token_154|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128163": { + "content": "<|reserved_special_token_155|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128164": { + "content": "<|reserved_special_token_156|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128165": { + "content": "<|reserved_special_token_157|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128166": { + "content": "<|reserved_special_token_158|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128167": { + "content": "<|reserved_special_token_159|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128168": { + "content": "<|reserved_special_token_160|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128169": { + "content": "<|reserved_special_token_161|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128170": { + "content": "<|reserved_special_token_162|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128171": { + "content": "<|reserved_special_token_163|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128172": { + "content": "<|reserved_special_token_164|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128173": { + "content": "<|reserved_special_token_165|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128174": { + "content": "<|reserved_special_token_166|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128175": { + "content": "<|reserved_special_token_167|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128176": { + "content": "<|reserved_special_token_168|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128177": { + "content": "<|reserved_special_token_169|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128178": { + "content": "<|reserved_special_token_170|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128179": { + "content": "<|reserved_special_token_171|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128180": { + "content": "<|reserved_special_token_172|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128181": { + "content": "<|reserved_special_token_173|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128182": { + "content": "<|reserved_special_token_174|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128183": { + "content": "<|reserved_special_token_175|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128184": { + "content": "<|reserved_special_token_176|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128185": { + "content": "<|reserved_special_token_177|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128186": { + "content": "<|reserved_special_token_178|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128187": { + "content": "<|reserved_special_token_179|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128188": { + "content": "<|reserved_special_token_180|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128189": { + "content": "<|reserved_special_token_181|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128190": { + "content": "<|reserved_special_token_182|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128191": { + "content": "<|reserved_special_token_183|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128192": { + "content": "<|reserved_special_token_184|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128193": { + "content": "<|reserved_special_token_185|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128194": { + "content": "<|reserved_special_token_186|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128195": { + "content": "<|reserved_special_token_187|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128196": { + "content": "<|reserved_special_token_188|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128197": { + "content": "<|reserved_special_token_189|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128198": { + "content": "<|reserved_special_token_190|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128199": { + "content": "<|reserved_special_token_191|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128200": { + "content": "<|reserved_special_token_192|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128201": { + "content": "<|reserved_special_token_193|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128202": { + "content": "<|reserved_special_token_194|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128203": { + "content": "<|reserved_special_token_195|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128204": { + "content": "<|reserved_special_token_196|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128205": { + "content": "<|reserved_special_token_197|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128206": { + "content": "<|reserved_special_token_198|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128207": { + "content": "<|reserved_special_token_199|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128208": { + "content": "<|reserved_special_token_200|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128209": { + "content": "<|reserved_special_token_201|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128210": { + "content": "<|reserved_special_token_202|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128211": { + "content": "<|reserved_special_token_203|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128212": { + "content": "<|reserved_special_token_204|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128213": { + "content": "<|reserved_special_token_205|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128214": { + "content": "<|reserved_special_token_206|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128215": { + "content": "<|reserved_special_token_207|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128216": { + "content": "<|reserved_special_token_208|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128217": { + "content": "<|reserved_special_token_209|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128218": { + "content": "<|reserved_special_token_210|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128219": { + "content": "<|reserved_special_token_211|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128220": { + "content": "<|reserved_special_token_212|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128221": { + "content": "<|reserved_special_token_213|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128222": { + "content": "<|reserved_special_token_214|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128223": { + "content": "<|reserved_special_token_215|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128224": { + "content": "<|reserved_special_token_216|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128225": { + "content": "<|reserved_special_token_217|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128226": { + "content": "<|reserved_special_token_218|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128227": { + "content": "<|reserved_special_token_219|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128228": { + "content": "<|reserved_special_token_220|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128229": { + "content": "<|reserved_special_token_221|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128230": { + "content": "<|reserved_special_token_222|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128231": { + "content": "<|reserved_special_token_223|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128232": { + "content": "<|reserved_special_token_224|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128233": { + "content": "<|reserved_special_token_225|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128234": { + "content": "<|reserved_special_token_226|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128235": { + "content": "<|reserved_special_token_227|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128236": { + "content": "<|reserved_special_token_228|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128237": { + "content": "<|reserved_special_token_229|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128238": { + "content": "<|reserved_special_token_230|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128239": { + "content": "<|reserved_special_token_231|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128240": { + "content": "<|reserved_special_token_232|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128241": { + "content": "<|reserved_special_token_233|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128242": { + "content": "<|reserved_special_token_234|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128243": { + "content": "<|reserved_special_token_235|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128244": { + "content": "<|reserved_special_token_236|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128245": { + "content": "<|reserved_special_token_237|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128246": { + "content": "<|reserved_special_token_238|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128247": { + "content": "<|reserved_special_token_239|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128248": { + "content": "<|reserved_special_token_240|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128249": { + "content": "<|reserved_special_token_241|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128250": { + "content": "<|reserved_special_token_242|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128251": { + "content": "<|reserved_special_token_243|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128252": { + "content": "<|reserved_special_token_244|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128253": { + "content": "<|reserved_special_token_245|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128254": { + "content": "<|reserved_special_token_246|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128255": { + "content": "<|reserved_special_token_247|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128256": { + "content": "[PAD]", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + } + }, + "bos_token": "<|begin_of_text|>", + "clean_up_tokenization_spaces": true, + "eos_token": "<|end_of_text|>", + "model_input_names": [ + "input_ids", + "attention_mask" + ], + "model_max_length": 131072, + "pad_token": "[PAD]", + "tokenizer_class": "PreTrainedTokenizerFast" +} diff --git a/babylm_shuffle_nondeterministic_10M_seed0/runs/checkpoint-1100/trainer_state.json b/babylm_shuffle_nondeterministic_10M_seed0/runs/checkpoint-1100/trainer_state.json new file mode 100644 index 0000000000000000000000000000000000000000..0ac3a962c47366c323db86e90e2910735c6f6071 --- /dev/null +++ b/babylm_shuffle_nondeterministic_10M_seed0/runs/checkpoint-1100/trainer_state.json @@ -0,0 +1,8613 @@ +{ + "best_metric": null, + "best_model_checkpoint": null, + "epoch": 1.2054794520547945, + "eval_steps": 10, + "global_step": 1100, + "is_hyper_param_search": false, + "is_local_process_zero": true, + "is_world_process_zero": true, + "log_history": [ + { + "epoch": 0.001095890410958904, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.0998, + "step": 1 + }, + { + "epoch": 0.002191780821917808, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.0858, + "step": 2 + }, + { + "epoch": 0.0032876712328767125, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.0672, + "step": 3 + }, + { + "epoch": 0.004383561643835616, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.0802, + "step": 4 + }, + { + "epoch": 0.005479452054794521, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.0636, + "step": 5 + }, + { + "epoch": 0.006575342465753425, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.1672, + "step": 6 + }, + { + "epoch": 0.007671232876712329, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.0961, + "step": 7 + }, + { + "epoch": 0.008767123287671232, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.1258, + "step": 8 + }, + { + "epoch": 0.009863013698630137, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.0953, + "step": 9 + }, + { + "epoch": 0.010958904109589041, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.1404, + "step": 10 + }, + { + "epoch": 0.010958904109589041, + "eval_loss": 3.12593412399292, + "eval_runtime": 42.732, + "eval_samples_per_second": 23.402, + "eval_steps_per_second": 0.983, + "step": 10 + }, + { + "epoch": 0.012054794520547946, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.1187, + "step": 11 + }, + { + "epoch": 0.01315068493150685, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.0593, + "step": 12 + }, + { + "epoch": 0.014246575342465753, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.111, + "step": 13 + }, + { + "epoch": 0.015342465753424657, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.0974, + "step": 14 + }, + { + "epoch": 0.01643835616438356, + "grad_norm": 5.991453170776367, + "learning_rate": 1.8248175182481753e-08, + "loss": 3.135, + "step": 15 + }, + { + "epoch": 0.017534246575342465, + "grad_norm": 6.274210453033447, + "learning_rate": 3.6496350364963505e-08, + "loss": 3.083, + "step": 16 + }, + { + "epoch": 0.01863013698630137, + "grad_norm": 6.644786834716797, + "learning_rate": 5.4744525547445264e-08, + "loss": 3.0856, + "step": 17 + }, + { + "epoch": 0.019726027397260273, + "grad_norm": 6.769156455993652, + "learning_rate": 7.299270072992701e-08, + "loss": 3.1082, + "step": 18 + }, + { + "epoch": 0.020821917808219178, + "grad_norm": 6.771246910095215, + "learning_rate": 9.124087591240876e-08, + "loss": 3.0753, + "step": 19 + }, + { + "epoch": 0.021917808219178082, + "grad_norm": 6.468441486358643, + "learning_rate": 1.0948905109489053e-07, + "loss": 3.1478, + "step": 20 + }, + { + "epoch": 0.021917808219178082, + "eval_loss": 3.1245107650756836, + "eval_runtime": 43.6251, + "eval_samples_per_second": 22.923, + "eval_steps_per_second": 0.963, + "step": 20 + }, + { + "epoch": 0.023013698630136987, + "grad_norm": 6.83078670501709, + "learning_rate": 1.2773722627737227e-07, + "loss": 3.1213, + "step": 21 + }, + { + "epoch": 0.02410958904109589, + "grad_norm": 6.221287250518799, + "learning_rate": 1.4598540145985402e-07, + "loss": 3.0549, + "step": 22 + }, + { + "epoch": 0.025205479452054796, + "grad_norm": 5.564266204833984, + "learning_rate": 1.642335766423358e-07, + "loss": 3.0769, + "step": 23 + }, + { + "epoch": 0.0263013698630137, + "grad_norm": 5.735209941864014, + "learning_rate": 1.8248175182481753e-07, + "loss": 3.1124, + "step": 24 + }, + { + "epoch": 0.0273972602739726, + "grad_norm": 5.577925682067871, + "learning_rate": 2.0072992700729928e-07, + "loss": 3.1191, + "step": 25 + }, + { + "epoch": 0.028493150684931506, + "grad_norm": 4.814139366149902, + "learning_rate": 2.1897810218978106e-07, + "loss": 3.0947, + "step": 26 + }, + { + "epoch": 0.02958904109589041, + "grad_norm": 4.69775390625, + "learning_rate": 2.3722627737226278e-07, + "loss": 3.0997, + "step": 27 + }, + { + "epoch": 0.030684931506849315, + "grad_norm": 4.811586380004883, + "learning_rate": 2.5547445255474454e-07, + "loss": 3.0749, + "step": 28 + }, + { + "epoch": 0.03178082191780822, + "grad_norm": 4.623377323150635, + "learning_rate": 2.737226277372263e-07, + "loss": 3.0619, + "step": 29 + }, + { + "epoch": 0.03287671232876712, + "grad_norm": 4.721253395080566, + "learning_rate": 2.9197080291970804e-07, + "loss": 3.0371, + "step": 30 + }, + { + "epoch": 0.03287671232876712, + "eval_loss": 3.082857847213745, + "eval_runtime": 44.127, + "eval_samples_per_second": 22.662, + "eval_steps_per_second": 0.952, + "step": 30 + }, + { + "epoch": 0.03397260273972603, + "grad_norm": 3.968630790710449, + "learning_rate": 3.102189781021898e-07, + "loss": 3.0535, + "step": 31 + }, + { + "epoch": 0.03506849315068493, + "grad_norm": 4.4373779296875, + "learning_rate": 3.284671532846716e-07, + "loss": 3.0476, + "step": 32 + }, + { + "epoch": 0.03616438356164384, + "grad_norm": 4.479320049285889, + "learning_rate": 3.467153284671533e-07, + "loss": 3.0215, + "step": 33 + }, + { + "epoch": 0.03726027397260274, + "grad_norm": 4.122231960296631, + "learning_rate": 3.6496350364963505e-07, + "loss": 3.0106, + "step": 34 + }, + { + "epoch": 0.038356164383561646, + "grad_norm": 4.153995513916016, + "learning_rate": 3.8321167883211686e-07, + "loss": 3.0318, + "step": 35 + }, + { + "epoch": 0.03945205479452055, + "grad_norm": 3.6949076652526855, + "learning_rate": 4.0145985401459856e-07, + "loss": 3.046, + "step": 36 + }, + { + "epoch": 0.040547945205479455, + "grad_norm": 3.4610488414764404, + "learning_rate": 4.197080291970803e-07, + "loss": 3.002, + "step": 37 + }, + { + "epoch": 0.041643835616438356, + "grad_norm": 3.9523138999938965, + "learning_rate": 4.379562043795621e-07, + "loss": 2.9665, + "step": 38 + }, + { + "epoch": 0.042739726027397264, + "grad_norm": 3.68516206741333, + "learning_rate": 4.562043795620438e-07, + "loss": 3.0411, + "step": 39 + }, + { + "epoch": 0.043835616438356165, + "grad_norm": 3.692396402359009, + "learning_rate": 4.7445255474452557e-07, + "loss": 2.9666, + "step": 40 + }, + { + "epoch": 0.043835616438356165, + "eval_loss": 3.015512704849243, + "eval_runtime": 48.2438, + "eval_samples_per_second": 20.728, + "eval_steps_per_second": 0.871, + "step": 40 + }, + { + "epoch": 0.044931506849315066, + "grad_norm": 3.3765921592712402, + "learning_rate": 4.927007299270073e-07, + "loss": 2.9831, + "step": 41 + }, + { + "epoch": 0.046027397260273974, + "grad_norm": 2.9631125926971436, + "learning_rate": 5.109489051094891e-07, + "loss": 3.0126, + "step": 42 + }, + { + "epoch": 0.047123287671232875, + "grad_norm": 2.969730854034424, + "learning_rate": 5.291970802919708e-07, + "loss": 2.9803, + "step": 43 + }, + { + "epoch": 0.04821917808219178, + "grad_norm": 3.185237407684326, + "learning_rate": 5.474452554744526e-07, + "loss": 2.9888, + "step": 44 + }, + { + "epoch": 0.049315068493150684, + "grad_norm": 3.055579423904419, + "learning_rate": 5.656934306569343e-07, + "loss": 2.9715, + "step": 45 + }, + { + "epoch": 0.05041095890410959, + "grad_norm": 3.083284616470337, + "learning_rate": 5.839416058394161e-07, + "loss": 2.9293, + "step": 46 + }, + { + "epoch": 0.05150684931506849, + "grad_norm": 2.984781503677368, + "learning_rate": 6.021897810218979e-07, + "loss": 2.9228, + "step": 47 + }, + { + "epoch": 0.0526027397260274, + "grad_norm": 2.756173849105835, + "learning_rate": 6.204379562043796e-07, + "loss": 2.9406, + "step": 48 + }, + { + "epoch": 0.0536986301369863, + "grad_norm": 2.911132574081421, + "learning_rate": 6.386861313868614e-07, + "loss": 2.8829, + "step": 49 + }, + { + "epoch": 0.0547945205479452, + "grad_norm": 2.592890977859497, + "learning_rate": 6.569343065693432e-07, + "loss": 2.9839, + "step": 50 + }, + { + "epoch": 0.0547945205479452, + "eval_loss": 2.9409468173980713, + "eval_runtime": 47.1045, + "eval_samples_per_second": 21.229, + "eval_steps_per_second": 0.892, + "step": 50 + }, + { + "epoch": 0.05589041095890411, + "grad_norm": 3.0112955570220947, + "learning_rate": 6.751824817518248e-07, + "loss": 2.8847, + "step": 51 + }, + { + "epoch": 0.05698630136986301, + "grad_norm": 2.3701798915863037, + "learning_rate": 6.934306569343066e-07, + "loss": 2.9109, + "step": 52 + }, + { + "epoch": 0.05808219178082192, + "grad_norm": 2.268850088119507, + "learning_rate": 7.116788321167883e-07, + "loss": 2.8483, + "step": 53 + }, + { + "epoch": 0.05917808219178082, + "grad_norm": 2.419015645980835, + "learning_rate": 7.299270072992701e-07, + "loss": 2.8491, + "step": 54 + }, + { + "epoch": 0.06027397260273973, + "grad_norm": 2.4582653045654297, + "learning_rate": 7.48175182481752e-07, + "loss": 2.8622, + "step": 55 + }, + { + "epoch": 0.06136986301369863, + "grad_norm": 2.3471553325653076, + "learning_rate": 7.664233576642337e-07, + "loss": 2.8738, + "step": 56 + }, + { + "epoch": 0.06246575342465754, + "grad_norm": 2.479177236557007, + "learning_rate": 7.846715328467154e-07, + "loss": 2.8843, + "step": 57 + }, + { + "epoch": 0.06356164383561644, + "grad_norm": 2.1676292419433594, + "learning_rate": 8.029197080291971e-07, + "loss": 2.8674, + "step": 58 + }, + { + "epoch": 0.06465753424657535, + "grad_norm": 1.9862910509109497, + "learning_rate": 8.211678832116789e-07, + "loss": 2.8197, + "step": 59 + }, + { + "epoch": 0.06575342465753424, + "grad_norm": 2.211313486099243, + "learning_rate": 8.394160583941606e-07, + "loss": 2.7961, + "step": 60 + }, + { + "epoch": 0.06575342465753424, + "eval_loss": 2.880824327468872, + "eval_runtime": 51.3235, + "eval_samples_per_second": 19.484, + "eval_steps_per_second": 0.818, + "step": 60 + }, + { + "epoch": 0.06684931506849315, + "grad_norm": 2.3497281074523926, + "learning_rate": 8.576642335766425e-07, + "loss": 2.8233, + "step": 61 + }, + { + "epoch": 0.06794520547945206, + "grad_norm": 2.158515214920044, + "learning_rate": 8.759124087591242e-07, + "loss": 2.8136, + "step": 62 + }, + { + "epoch": 0.06904109589041096, + "grad_norm": 2.0219955444335938, + "learning_rate": 8.94160583941606e-07, + "loss": 2.8231, + "step": 63 + }, + { + "epoch": 0.07013698630136986, + "grad_norm": 1.8989133834838867, + "learning_rate": 9.124087591240876e-07, + "loss": 2.8268, + "step": 64 + }, + { + "epoch": 0.07123287671232877, + "grad_norm": 2.243851661682129, + "learning_rate": 9.306569343065694e-07, + "loss": 2.7886, + "step": 65 + }, + { + "epoch": 0.07232876712328767, + "grad_norm": 2.122464418411255, + "learning_rate": 9.489051094890511e-07, + "loss": 2.8166, + "step": 66 + }, + { + "epoch": 0.07342465753424658, + "grad_norm": 1.899857997894287, + "learning_rate": 9.67153284671533e-07, + "loss": 2.7997, + "step": 67 + }, + { + "epoch": 0.07452054794520548, + "grad_norm": 2.4424173831939697, + "learning_rate": 9.854014598540146e-07, + "loss": 2.8342, + "step": 68 + }, + { + "epoch": 0.07561643835616438, + "grad_norm": 2.5677411556243896, + "learning_rate": 1.0036496350364965e-06, + "loss": 2.7786, + "step": 69 + }, + { + "epoch": 0.07671232876712329, + "grad_norm": 1.9989523887634277, + "learning_rate": 1.0218978102189781e-06, + "loss": 2.7411, + "step": 70 + }, + { + "epoch": 0.07671232876712329, + "eval_loss": 2.8185322284698486, + "eval_runtime": 50.4998, + "eval_samples_per_second": 19.802, + "eval_steps_per_second": 0.832, + "step": 70 + }, + { + "epoch": 0.07780821917808219, + "grad_norm": 1.9376521110534668, + "learning_rate": 1.04014598540146e-06, + "loss": 2.7988, + "step": 71 + }, + { + "epoch": 0.0789041095890411, + "grad_norm": 2.5629725456237793, + "learning_rate": 1.0583941605839416e-06, + "loss": 2.7079, + "step": 72 + }, + { + "epoch": 0.08, + "grad_norm": 2.1634678840637207, + "learning_rate": 1.0766423357664233e-06, + "loss": 2.7049, + "step": 73 + }, + { + "epoch": 0.08109589041095891, + "grad_norm": 2.2143008708953857, + "learning_rate": 1.0948905109489052e-06, + "loss": 2.72, + "step": 74 + }, + { + "epoch": 0.0821917808219178, + "grad_norm": 2.2701101303100586, + "learning_rate": 1.113138686131387e-06, + "loss": 2.7418, + "step": 75 + }, + { + "epoch": 0.08328767123287671, + "grad_norm": 1.716569423675537, + "learning_rate": 1.1313868613138687e-06, + "loss": 2.7558, + "step": 76 + }, + { + "epoch": 0.08438356164383562, + "grad_norm": 2.79757022857666, + "learning_rate": 1.1496350364963505e-06, + "loss": 2.6865, + "step": 77 + }, + { + "epoch": 0.08547945205479453, + "grad_norm": 2.2189838886260986, + "learning_rate": 1.1678832116788322e-06, + "loss": 2.6684, + "step": 78 + }, + { + "epoch": 0.08657534246575342, + "grad_norm": 2.0695579051971436, + "learning_rate": 1.1861313868613138e-06, + "loss": 2.6779, + "step": 79 + }, + { + "epoch": 0.08767123287671233, + "grad_norm": 2.372326374053955, + "learning_rate": 1.2043795620437959e-06, + "loss": 2.6711, + "step": 80 + }, + { + "epoch": 0.08767123287671233, + "eval_loss": 2.7537453174591064, + "eval_runtime": 47.7274, + "eval_samples_per_second": 20.952, + "eval_steps_per_second": 0.88, + "step": 80 + }, + { + "epoch": 0.08876712328767124, + "grad_norm": 2.3398947715759277, + "learning_rate": 1.2226277372262775e-06, + "loss": 2.6776, + "step": 81 + }, + { + "epoch": 0.08986301369863013, + "grad_norm": 2.6619036197662354, + "learning_rate": 1.2408759124087592e-06, + "loss": 2.7077, + "step": 82 + }, + { + "epoch": 0.09095890410958904, + "grad_norm": 1.8245115280151367, + "learning_rate": 1.259124087591241e-06, + "loss": 2.662, + "step": 83 + }, + { + "epoch": 0.09205479452054795, + "grad_norm": 2.2627148628234863, + "learning_rate": 1.2773722627737229e-06, + "loss": 2.677, + "step": 84 + }, + { + "epoch": 0.09315068493150686, + "grad_norm": 2.4983835220336914, + "learning_rate": 1.2956204379562045e-06, + "loss": 2.6601, + "step": 85 + }, + { + "epoch": 0.09424657534246575, + "grad_norm": 2.147010087966919, + "learning_rate": 1.3138686131386864e-06, + "loss": 2.6213, + "step": 86 + }, + { + "epoch": 0.09534246575342466, + "grad_norm": 2.791682481765747, + "learning_rate": 1.332116788321168e-06, + "loss": 2.6281, + "step": 87 + }, + { + "epoch": 0.09643835616438357, + "grad_norm": 2.023970365524292, + "learning_rate": 1.3503649635036497e-06, + "loss": 2.5766, + "step": 88 + }, + { + "epoch": 0.09753424657534246, + "grad_norm": 2.8143956661224365, + "learning_rate": 1.3686131386861315e-06, + "loss": 2.6086, + "step": 89 + }, + { + "epoch": 0.09863013698630137, + "grad_norm": 2.226106882095337, + "learning_rate": 1.3868613138686132e-06, + "loss": 2.5477, + "step": 90 + }, + { + "epoch": 0.09863013698630137, + "eval_loss": 2.6880786418914795, + "eval_runtime": 47.8417, + "eval_samples_per_second": 20.902, + "eval_steps_per_second": 0.878, + "step": 90 + }, + { + "epoch": 0.09972602739726028, + "grad_norm": 2.5248148441314697, + "learning_rate": 1.405109489051095e-06, + "loss": 2.6232, + "step": 91 + }, + { + "epoch": 0.10082191780821918, + "grad_norm": 2.421330690383911, + "learning_rate": 1.4233576642335767e-06, + "loss": 2.599, + "step": 92 + }, + { + "epoch": 0.10191780821917808, + "grad_norm": 2.5562429428100586, + "learning_rate": 1.4416058394160583e-06, + "loss": 2.6197, + "step": 93 + }, + { + "epoch": 0.10301369863013699, + "grad_norm": 2.856201171875, + "learning_rate": 1.4598540145985402e-06, + "loss": 2.5825, + "step": 94 + }, + { + "epoch": 0.10410958904109589, + "grad_norm": 2.6576807498931885, + "learning_rate": 1.478102189781022e-06, + "loss": 2.5496, + "step": 95 + }, + { + "epoch": 0.1052054794520548, + "grad_norm": 3.1408190727233887, + "learning_rate": 1.496350364963504e-06, + "loss": 2.6142, + "step": 96 + }, + { + "epoch": 0.1063013698630137, + "grad_norm": 2.3270952701568604, + "learning_rate": 1.5145985401459856e-06, + "loss": 2.584, + "step": 97 + }, + { + "epoch": 0.1073972602739726, + "grad_norm": 2.3976614475250244, + "learning_rate": 1.5328467153284674e-06, + "loss": 2.5941, + "step": 98 + }, + { + "epoch": 0.10849315068493151, + "grad_norm": 2.783485174179077, + "learning_rate": 1.551094890510949e-06, + "loss": 2.5626, + "step": 99 + }, + { + "epoch": 0.1095890410958904, + "grad_norm": 2.2522788047790527, + "learning_rate": 1.5693430656934307e-06, + "loss": 2.517, + "step": 100 + }, + { + "epoch": 0.1095890410958904, + "eval_loss": 2.637315034866333, + "eval_runtime": 48.2107, + "eval_samples_per_second": 20.742, + "eval_steps_per_second": 0.871, + "step": 100 + }, + { + "epoch": 0.11068493150684931, + "grad_norm": 2.917799949645996, + "learning_rate": 1.5875912408759126e-06, + "loss": 2.5159, + "step": 101 + }, + { + "epoch": 0.11178082191780822, + "grad_norm": 2.955815076828003, + "learning_rate": 1.6058394160583942e-06, + "loss": 2.563, + "step": 102 + }, + { + "epoch": 0.11287671232876713, + "grad_norm": 2.8096001148223877, + "learning_rate": 1.624087591240876e-06, + "loss": 2.5995, + "step": 103 + }, + { + "epoch": 0.11397260273972602, + "grad_norm": 2.4967427253723145, + "learning_rate": 1.6423357664233577e-06, + "loss": 2.5336, + "step": 104 + }, + { + "epoch": 0.11506849315068493, + "grad_norm": 2.740717887878418, + "learning_rate": 1.6605839416058394e-06, + "loss": 2.4778, + "step": 105 + }, + { + "epoch": 0.11616438356164384, + "grad_norm": 2.5064632892608643, + "learning_rate": 1.6788321167883212e-06, + "loss": 2.5214, + "step": 106 + }, + { + "epoch": 0.11726027397260275, + "grad_norm": 2.2923007011413574, + "learning_rate": 1.6970802919708029e-06, + "loss": 2.5315, + "step": 107 + }, + { + "epoch": 0.11835616438356164, + "grad_norm": 3.339569568634033, + "learning_rate": 1.715328467153285e-06, + "loss": 2.4661, + "step": 108 + }, + { + "epoch": 0.11945205479452055, + "grad_norm": 2.1650891304016113, + "learning_rate": 1.7335766423357666e-06, + "loss": 2.5072, + "step": 109 + }, + { + "epoch": 0.12054794520547946, + "grad_norm": 3.1216914653778076, + "learning_rate": 1.7518248175182485e-06, + "loss": 2.5164, + "step": 110 + }, + { + "epoch": 0.12054794520547946, + "eval_loss": 2.594250202178955, + "eval_runtime": 41.8939, + "eval_samples_per_second": 23.87, + "eval_steps_per_second": 1.003, + "step": 110 + }, + { + "epoch": 0.12164383561643835, + "grad_norm": 2.191063404083252, + "learning_rate": 1.77007299270073e-06, + "loss": 2.5422, + "step": 111 + }, + { + "epoch": 0.12273972602739726, + "grad_norm": 2.922083854675293, + "learning_rate": 1.788321167883212e-06, + "loss": 2.5258, + "step": 112 + }, + { + "epoch": 0.12383561643835617, + "grad_norm": 2.518002986907959, + "learning_rate": 1.8065693430656936e-06, + "loss": 2.5588, + "step": 113 + }, + { + "epoch": 0.12493150684931507, + "grad_norm": 2.515194892883301, + "learning_rate": 1.8248175182481753e-06, + "loss": 2.5079, + "step": 114 + }, + { + "epoch": 0.12602739726027398, + "grad_norm": 2.5530874729156494, + "learning_rate": 1.8430656934306571e-06, + "loss": 2.4615, + "step": 115 + }, + { + "epoch": 0.12712328767123288, + "grad_norm": 2.3550198078155518, + "learning_rate": 1.8613138686131388e-06, + "loss": 2.4825, + "step": 116 + }, + { + "epoch": 0.12821917808219177, + "grad_norm": 2.295145034790039, + "learning_rate": 1.8795620437956206e-06, + "loss": 2.4749, + "step": 117 + }, + { + "epoch": 0.1293150684931507, + "grad_norm": 2.0158982276916504, + "learning_rate": 1.8978102189781023e-06, + "loss": 2.4756, + "step": 118 + }, + { + "epoch": 0.1304109589041096, + "grad_norm": 2.213057518005371, + "learning_rate": 1.916058394160584e-06, + "loss": 2.4744, + "step": 119 + }, + { + "epoch": 0.13150684931506848, + "grad_norm": 2.186222553253174, + "learning_rate": 1.934306569343066e-06, + "loss": 2.4957, + "step": 120 + }, + { + "epoch": 0.13150684931506848, + "eval_loss": 2.563931465148926, + "eval_runtime": 44.8714, + "eval_samples_per_second": 22.286, + "eval_steps_per_second": 0.936, + "step": 120 + }, + { + "epoch": 0.1326027397260274, + "grad_norm": 2.3143310546875, + "learning_rate": 1.952554744525548e-06, + "loss": 2.4, + "step": 121 + }, + { + "epoch": 0.1336986301369863, + "grad_norm": 2.0223965644836426, + "learning_rate": 1.9708029197080293e-06, + "loss": 2.4453, + "step": 122 + }, + { + "epoch": 0.13479452054794522, + "grad_norm": 2.2025411128997803, + "learning_rate": 1.989051094890511e-06, + "loss": 2.4463, + "step": 123 + }, + { + "epoch": 0.1358904109589041, + "grad_norm": 2.707930564880371, + "learning_rate": 2.007299270072993e-06, + "loss": 2.4589, + "step": 124 + }, + { + "epoch": 0.136986301369863, + "grad_norm": 2.146662473678589, + "learning_rate": 2.025547445255475e-06, + "loss": 2.5023, + "step": 125 + }, + { + "epoch": 0.13808219178082193, + "grad_norm": 2.520660161972046, + "learning_rate": 2.0437956204379563e-06, + "loss": 2.4464, + "step": 126 + }, + { + "epoch": 0.13917808219178082, + "grad_norm": 2.50632381439209, + "learning_rate": 2.062043795620438e-06, + "loss": 2.4556, + "step": 127 + }, + { + "epoch": 0.14027397260273972, + "grad_norm": 2.2919397354125977, + "learning_rate": 2.08029197080292e-06, + "loss": 2.4479, + "step": 128 + }, + { + "epoch": 0.14136986301369864, + "grad_norm": 2.521475315093994, + "learning_rate": 2.0985401459854014e-06, + "loss": 2.4982, + "step": 129 + }, + { + "epoch": 0.14246575342465753, + "grad_norm": 2.3812668323516846, + "learning_rate": 2.1167883211678833e-06, + "loss": 2.4163, + "step": 130 + }, + { + "epoch": 0.14246575342465753, + "eval_loss": 2.5390918254852295, + "eval_runtime": 43.8746, + "eval_samples_per_second": 22.792, + "eval_steps_per_second": 0.957, + "step": 130 + }, + { + "epoch": 0.14356164383561643, + "grad_norm": 2.302286148071289, + "learning_rate": 2.135036496350365e-06, + "loss": 2.4588, + "step": 131 + }, + { + "epoch": 0.14465753424657535, + "grad_norm": 2.4222044944763184, + "learning_rate": 2.1532846715328466e-06, + "loss": 2.492, + "step": 132 + }, + { + "epoch": 0.14575342465753424, + "grad_norm": 2.605835199356079, + "learning_rate": 2.171532846715329e-06, + "loss": 2.4055, + "step": 133 + }, + { + "epoch": 0.14684931506849316, + "grad_norm": 3.6261861324310303, + "learning_rate": 2.1897810218978103e-06, + "loss": 2.4221, + "step": 134 + }, + { + "epoch": 0.14794520547945206, + "grad_norm": 2.7647221088409424, + "learning_rate": 2.208029197080292e-06, + "loss": 2.4549, + "step": 135 + }, + { + "epoch": 0.14904109589041095, + "grad_norm": 3.2977371215820312, + "learning_rate": 2.226277372262774e-06, + "loss": 2.4331, + "step": 136 + }, + { + "epoch": 0.15013698630136987, + "grad_norm": 3.147094488143921, + "learning_rate": 2.244525547445256e-06, + "loss": 2.4268, + "step": 137 + }, + { + "epoch": 0.15123287671232877, + "grad_norm": 2.6363413333892822, + "learning_rate": 2.2627737226277373e-06, + "loss": 2.4524, + "step": 138 + }, + { + "epoch": 0.15232876712328766, + "grad_norm": 2.9288904666900635, + "learning_rate": 2.281021897810219e-06, + "loss": 2.4534, + "step": 139 + }, + { + "epoch": 0.15342465753424658, + "grad_norm": 2.5604870319366455, + "learning_rate": 2.299270072992701e-06, + "loss": 2.4138, + "step": 140 + }, + { + "epoch": 0.15342465753424658, + "eval_loss": 2.524790048599243, + "eval_runtime": 43.8851, + "eval_samples_per_second": 22.787, + "eval_steps_per_second": 0.957, + "step": 140 + }, + { + "epoch": 0.15452054794520548, + "grad_norm": 3.2746663093566895, + "learning_rate": 2.3175182481751825e-06, + "loss": 2.4094, + "step": 141 + }, + { + "epoch": 0.15561643835616437, + "grad_norm": 2.691483497619629, + "learning_rate": 2.3357664233576643e-06, + "loss": 2.4265, + "step": 142 + }, + { + "epoch": 0.1567123287671233, + "grad_norm": 2.5850117206573486, + "learning_rate": 2.354014598540146e-06, + "loss": 2.3922, + "step": 143 + }, + { + "epoch": 0.1578082191780822, + "grad_norm": 2.68204927444458, + "learning_rate": 2.3722627737226276e-06, + "loss": 2.4262, + "step": 144 + }, + { + "epoch": 0.1589041095890411, + "grad_norm": 2.952584981918335, + "learning_rate": 2.39051094890511e-06, + "loss": 2.4267, + "step": 145 + }, + { + "epoch": 0.16, + "grad_norm": 3.2354788780212402, + "learning_rate": 2.4087591240875918e-06, + "loss": 2.3836, + "step": 146 + }, + { + "epoch": 0.1610958904109589, + "grad_norm": 2.634521007537842, + "learning_rate": 2.427007299270073e-06, + "loss": 2.3953, + "step": 147 + }, + { + "epoch": 0.16219178082191782, + "grad_norm": 3.1115682125091553, + "learning_rate": 2.445255474452555e-06, + "loss": 2.409, + "step": 148 + }, + { + "epoch": 0.1632876712328767, + "grad_norm": 2.6824135780334473, + "learning_rate": 2.463503649635037e-06, + "loss": 2.4459, + "step": 149 + }, + { + "epoch": 0.1643835616438356, + "grad_norm": 2.617614507675171, + "learning_rate": 2.4817518248175183e-06, + "loss": 2.366, + "step": 150 + }, + { + "epoch": 0.1643835616438356, + "eval_loss": 2.5088160037994385, + "eval_runtime": 44.0888, + "eval_samples_per_second": 22.682, + "eval_steps_per_second": 0.953, + "step": 150 + }, + { + "epoch": 0.16547945205479453, + "grad_norm": 3.976992130279541, + "learning_rate": 2.5e-06, + "loss": 2.3517, + "step": 151 + }, + { + "epoch": 0.16657534246575342, + "grad_norm": 3.1433017253875732, + "learning_rate": 2.518248175182482e-06, + "loss": 2.4489, + "step": 152 + }, + { + "epoch": 0.16767123287671232, + "grad_norm": 4.100688934326172, + "learning_rate": 2.5364963503649635e-06, + "loss": 2.3683, + "step": 153 + }, + { + "epoch": 0.16876712328767124, + "grad_norm": 4.316102504730225, + "learning_rate": 2.5547445255474458e-06, + "loss": 2.3663, + "step": 154 + }, + { + "epoch": 0.16986301369863013, + "grad_norm": 2.3103132247924805, + "learning_rate": 2.5729927007299272e-06, + "loss": 2.4009, + "step": 155 + }, + { + "epoch": 0.17095890410958905, + "grad_norm": 3.576810836791992, + "learning_rate": 2.591240875912409e-06, + "loss": 2.3983, + "step": 156 + }, + { + "epoch": 0.17205479452054795, + "grad_norm": 2.337080955505371, + "learning_rate": 2.6094890510948905e-06, + "loss": 2.3849, + "step": 157 + }, + { + "epoch": 0.17315068493150684, + "grad_norm": 3.2523117065429688, + "learning_rate": 2.627737226277373e-06, + "loss": 2.4171, + "step": 158 + }, + { + "epoch": 0.17424657534246576, + "grad_norm": 2.529175281524658, + "learning_rate": 2.6459854014598542e-06, + "loss": 2.4303, + "step": 159 + }, + { + "epoch": 0.17534246575342466, + "grad_norm": 3.052212953567505, + "learning_rate": 2.664233576642336e-06, + "loss": 2.3895, + "step": 160 + }, + { + "epoch": 0.17534246575342466, + "eval_loss": 2.492485761642456, + "eval_runtime": 45.6738, + "eval_samples_per_second": 21.894, + "eval_steps_per_second": 0.92, + "step": 160 + }, + { + "epoch": 0.17643835616438355, + "grad_norm": 2.1544973850250244, + "learning_rate": 2.6824817518248175e-06, + "loss": 2.3649, + "step": 161 + }, + { + "epoch": 0.17753424657534247, + "grad_norm": 2.7351951599121094, + "learning_rate": 2.7007299270072994e-06, + "loss": 2.3908, + "step": 162 + }, + { + "epoch": 0.17863013698630137, + "grad_norm": 2.8431951999664307, + "learning_rate": 2.718978102189781e-06, + "loss": 2.374, + "step": 163 + }, + { + "epoch": 0.17972602739726026, + "grad_norm": 3.083190441131592, + "learning_rate": 2.737226277372263e-06, + "loss": 2.4193, + "step": 164 + }, + { + "epoch": 0.18082191780821918, + "grad_norm": 2.6523995399475098, + "learning_rate": 2.755474452554745e-06, + "loss": 2.3969, + "step": 165 + }, + { + "epoch": 0.18191780821917808, + "grad_norm": 2.626047372817993, + "learning_rate": 2.7737226277372264e-06, + "loss": 2.3742, + "step": 166 + }, + { + "epoch": 0.18301369863013697, + "grad_norm": 3.300917863845825, + "learning_rate": 2.7919708029197082e-06, + "loss": 2.4546, + "step": 167 + }, + { + "epoch": 0.1841095890410959, + "grad_norm": 2.797304391860962, + "learning_rate": 2.81021897810219e-06, + "loss": 2.4161, + "step": 168 + }, + { + "epoch": 0.1852054794520548, + "grad_norm": 2.777117967605591, + "learning_rate": 2.828467153284672e-06, + "loss": 2.3572, + "step": 169 + }, + { + "epoch": 0.1863013698630137, + "grad_norm": 2.887470006942749, + "learning_rate": 2.8467153284671534e-06, + "loss": 2.3887, + "step": 170 + }, + { + "epoch": 0.1863013698630137, + "eval_loss": 2.4854257106781006, + "eval_runtime": 44.5644, + "eval_samples_per_second": 22.439, + "eval_steps_per_second": 0.942, + "step": 170 + }, + { + "epoch": 0.1873972602739726, + "grad_norm": 3.0706076622009277, + "learning_rate": 2.8649635036496353e-06, + "loss": 2.3162, + "step": 171 + }, + { + "epoch": 0.1884931506849315, + "grad_norm": 3.6129744052886963, + "learning_rate": 2.8832116788321167e-06, + "loss": 2.3685, + "step": 172 + }, + { + "epoch": 0.18958904109589042, + "grad_norm": 2.4936370849609375, + "learning_rate": 2.901459854014599e-06, + "loss": 2.3297, + "step": 173 + }, + { + "epoch": 0.19068493150684931, + "grad_norm": 2.1041860580444336, + "learning_rate": 2.9197080291970804e-06, + "loss": 2.4085, + "step": 174 + }, + { + "epoch": 0.1917808219178082, + "grad_norm": 2.2066831588745117, + "learning_rate": 2.9379562043795623e-06, + "loss": 2.3905, + "step": 175 + }, + { + "epoch": 0.19287671232876713, + "grad_norm": 2.444652795791626, + "learning_rate": 2.956204379562044e-06, + "loss": 2.378, + "step": 176 + }, + { + "epoch": 0.19397260273972602, + "grad_norm": 2.501939058303833, + "learning_rate": 2.9744525547445256e-06, + "loss": 2.4136, + "step": 177 + }, + { + "epoch": 0.19506849315068492, + "grad_norm": 2.2167556285858154, + "learning_rate": 2.992700729927008e-06, + "loss": 2.385, + "step": 178 + }, + { + "epoch": 0.19616438356164384, + "grad_norm": 2.5298545360565186, + "learning_rate": 3.0109489051094893e-06, + "loss": 2.3508, + "step": 179 + }, + { + "epoch": 0.19726027397260273, + "grad_norm": 2.3774642944335938, + "learning_rate": 3.029197080291971e-06, + "loss": 2.3421, + "step": 180 + }, + { + "epoch": 0.19726027397260273, + "eval_loss": 2.476067304611206, + "eval_runtime": 44.3761, + "eval_samples_per_second": 22.535, + "eval_steps_per_second": 0.946, + "step": 180 + }, + { + "epoch": 0.19835616438356166, + "grad_norm": 2.9108712673187256, + "learning_rate": 3.0474452554744526e-06, + "loss": 2.3983, + "step": 181 + }, + { + "epoch": 0.19945205479452055, + "grad_norm": 2.5232033729553223, + "learning_rate": 3.065693430656935e-06, + "loss": 2.3946, + "step": 182 + }, + { + "epoch": 0.20054794520547944, + "grad_norm": 2.158168077468872, + "learning_rate": 3.0839416058394163e-06, + "loss": 2.3805, + "step": 183 + }, + { + "epoch": 0.20164383561643837, + "grad_norm": 2.3405354022979736, + "learning_rate": 3.102189781021898e-06, + "loss": 2.3974, + "step": 184 + }, + { + "epoch": 0.20273972602739726, + "grad_norm": 2.5308327674865723, + "learning_rate": 3.1204379562043796e-06, + "loss": 2.384, + "step": 185 + }, + { + "epoch": 0.20383561643835615, + "grad_norm": 2.6328237056732178, + "learning_rate": 3.1386861313868614e-06, + "loss": 2.3345, + "step": 186 + }, + { + "epoch": 0.20493150684931508, + "grad_norm": 3.0725512504577637, + "learning_rate": 3.156934306569343e-06, + "loss": 2.3191, + "step": 187 + }, + { + "epoch": 0.20602739726027397, + "grad_norm": 2.196218729019165, + "learning_rate": 3.175182481751825e-06, + "loss": 2.4073, + "step": 188 + }, + { + "epoch": 0.20712328767123286, + "grad_norm": 2.7722432613372803, + "learning_rate": 3.193430656934307e-06, + "loss": 2.3775, + "step": 189 + }, + { + "epoch": 0.20821917808219179, + "grad_norm": 2.4848241806030273, + "learning_rate": 3.2116788321167884e-06, + "loss": 2.3538, + "step": 190 + }, + { + "epoch": 0.20821917808219179, + "eval_loss": 2.462618112564087, + "eval_runtime": 43.9461, + "eval_samples_per_second": 22.755, + "eval_steps_per_second": 0.956, + "step": 190 + }, + { + "epoch": 0.20931506849315068, + "grad_norm": 2.4682741165161133, + "learning_rate": 3.2299270072992707e-06, + "loss": 2.3688, + "step": 191 + }, + { + "epoch": 0.2104109589041096, + "grad_norm": 2.518923282623291, + "learning_rate": 3.248175182481752e-06, + "loss": 2.3556, + "step": 192 + }, + { + "epoch": 0.2115068493150685, + "grad_norm": 2.4023561477661133, + "learning_rate": 3.266423357664234e-06, + "loss": 2.292, + "step": 193 + }, + { + "epoch": 0.2126027397260274, + "grad_norm": 2.5196943283081055, + "learning_rate": 3.2846715328467155e-06, + "loss": 2.3703, + "step": 194 + }, + { + "epoch": 0.2136986301369863, + "grad_norm": 2.762709617614746, + "learning_rate": 3.3029197080291973e-06, + "loss": 2.3578, + "step": 195 + }, + { + "epoch": 0.2147945205479452, + "grad_norm": 2.1287691593170166, + "learning_rate": 3.3211678832116788e-06, + "loss": 2.3687, + "step": 196 + }, + { + "epoch": 0.2158904109589041, + "grad_norm": 2.477341651916504, + "learning_rate": 3.339416058394161e-06, + "loss": 2.3545, + "step": 197 + }, + { + "epoch": 0.21698630136986302, + "grad_norm": 2.23933744430542, + "learning_rate": 3.3576642335766425e-06, + "loss": 2.3511, + "step": 198 + }, + { + "epoch": 0.21808219178082192, + "grad_norm": 2.2440133094787598, + "learning_rate": 3.3759124087591243e-06, + "loss": 2.3539, + "step": 199 + }, + { + "epoch": 0.2191780821917808, + "grad_norm": 2.5603177547454834, + "learning_rate": 3.3941605839416058e-06, + "loss": 2.36, + "step": 200 + }, + { + "epoch": 0.2191780821917808, + "eval_loss": 2.4507246017456055, + "eval_runtime": 43.7786, + "eval_samples_per_second": 22.842, + "eval_steps_per_second": 0.959, + "step": 200 + }, + { + "epoch": 0.22027397260273973, + "grad_norm": 2.4570510387420654, + "learning_rate": 3.412408759124088e-06, + "loss": 2.3702, + "step": 201 + }, + { + "epoch": 0.22136986301369863, + "grad_norm": 2.1975133419036865, + "learning_rate": 3.43065693430657e-06, + "loss": 2.3667, + "step": 202 + }, + { + "epoch": 0.22246575342465755, + "grad_norm": 2.710186004638672, + "learning_rate": 3.4489051094890513e-06, + "loss": 2.3159, + "step": 203 + }, + { + "epoch": 0.22356164383561644, + "grad_norm": 2.7799506187438965, + "learning_rate": 3.467153284671533e-06, + "loss": 2.3519, + "step": 204 + }, + { + "epoch": 0.22465753424657534, + "grad_norm": 2.2479400634765625, + "learning_rate": 3.4854014598540146e-06, + "loss": 2.3103, + "step": 205 + }, + { + "epoch": 0.22575342465753426, + "grad_norm": 3.1538405418395996, + "learning_rate": 3.503649635036497e-06, + "loss": 2.3506, + "step": 206 + }, + { + "epoch": 0.22684931506849315, + "grad_norm": 2.7607762813568115, + "learning_rate": 3.5218978102189783e-06, + "loss": 2.3386, + "step": 207 + }, + { + "epoch": 0.22794520547945205, + "grad_norm": 2.371105909347534, + "learning_rate": 3.54014598540146e-06, + "loss": 2.3583, + "step": 208 + }, + { + "epoch": 0.22904109589041097, + "grad_norm": 2.793933391571045, + "learning_rate": 3.5583941605839416e-06, + "loss": 2.3782, + "step": 209 + }, + { + "epoch": 0.23013698630136986, + "grad_norm": 2.372732400894165, + "learning_rate": 3.576642335766424e-06, + "loss": 2.417, + "step": 210 + }, + { + "epoch": 0.23013698630136986, + "eval_loss": 2.451474189758301, + "eval_runtime": 41.7197, + "eval_samples_per_second": 23.97, + "eval_steps_per_second": 1.007, + "step": 210 + }, + { + "epoch": 0.23123287671232876, + "grad_norm": 2.6989448070526123, + "learning_rate": 3.5948905109489054e-06, + "loss": 2.393, + "step": 211 + }, + { + "epoch": 0.23232876712328768, + "grad_norm": 2.4324285984039307, + "learning_rate": 3.6131386861313872e-06, + "loss": 2.2738, + "step": 212 + }, + { + "epoch": 0.23342465753424657, + "grad_norm": 3.5063016414642334, + "learning_rate": 3.6313868613138687e-06, + "loss": 2.2823, + "step": 213 + }, + { + "epoch": 0.2345205479452055, + "grad_norm": 3.953951358795166, + "learning_rate": 3.6496350364963505e-06, + "loss": 2.3132, + "step": 214 + }, + { + "epoch": 0.2356164383561644, + "grad_norm": 2.3483071327209473, + "learning_rate": 3.667883211678833e-06, + "loss": 2.3244, + "step": 215 + }, + { + "epoch": 0.23671232876712328, + "grad_norm": 2.884199380874634, + "learning_rate": 3.6861313868613142e-06, + "loss": 2.3067, + "step": 216 + }, + { + "epoch": 0.2378082191780822, + "grad_norm": 2.9514520168304443, + "learning_rate": 3.704379562043796e-06, + "loss": 2.304, + "step": 217 + }, + { + "epoch": 0.2389041095890411, + "grad_norm": 2.7530996799468994, + "learning_rate": 3.7226277372262775e-06, + "loss": 2.3691, + "step": 218 + }, + { + "epoch": 0.24, + "grad_norm": 2.8753647804260254, + "learning_rate": 3.7408759124087594e-06, + "loss": 2.3406, + "step": 219 + }, + { + "epoch": 0.2410958904109589, + "grad_norm": 2.670574903488159, + "learning_rate": 3.7591240875912412e-06, + "loss": 2.304, + "step": 220 + }, + { + "epoch": 0.2410958904109589, + "eval_loss": 2.4395809173583984, + "eval_runtime": 46.6723, + "eval_samples_per_second": 21.426, + "eval_steps_per_second": 0.9, + "step": 220 + }, + { + "epoch": 0.2421917808219178, + "grad_norm": 2.4152438640594482, + "learning_rate": 3.777372262773723e-06, + "loss": 2.3294, + "step": 221 + }, + { + "epoch": 0.2432876712328767, + "grad_norm": 2.8398540019989014, + "learning_rate": 3.7956204379562045e-06, + "loss": 2.3216, + "step": 222 + }, + { + "epoch": 0.24438356164383562, + "grad_norm": 2.6036765575408936, + "learning_rate": 3.8138686131386864e-06, + "loss": 2.3109, + "step": 223 + }, + { + "epoch": 0.24547945205479452, + "grad_norm": 2.7073473930358887, + "learning_rate": 3.832116788321168e-06, + "loss": 2.3457, + "step": 224 + }, + { + "epoch": 0.2465753424657534, + "grad_norm": 2.584404230117798, + "learning_rate": 3.85036496350365e-06, + "loss": 2.3159, + "step": 225 + }, + { + "epoch": 0.24767123287671233, + "grad_norm": 2.4220666885375977, + "learning_rate": 3.868613138686132e-06, + "loss": 2.2944, + "step": 226 + }, + { + "epoch": 0.24876712328767123, + "grad_norm": 2.5636494159698486, + "learning_rate": 3.886861313868613e-06, + "loss": 2.3219, + "step": 227 + }, + { + "epoch": 0.24986301369863015, + "grad_norm": 2.2990903854370117, + "learning_rate": 3.905109489051096e-06, + "loss": 2.3342, + "step": 228 + }, + { + "epoch": 0.25095890410958904, + "grad_norm": 2.50913143157959, + "learning_rate": 3.923357664233577e-06, + "loss": 2.3619, + "step": 229 + }, + { + "epoch": 0.25205479452054796, + "grad_norm": 2.1687769889831543, + "learning_rate": 3.9416058394160585e-06, + "loss": 2.3153, + "step": 230 + }, + { + "epoch": 0.25205479452054796, + "eval_loss": 2.4315240383148193, + "eval_runtime": 48.3542, + "eval_samples_per_second": 20.681, + "eval_steps_per_second": 0.869, + "step": 230 + }, + { + "epoch": 0.25315068493150683, + "grad_norm": 2.842125177383423, + "learning_rate": 3.95985401459854e-06, + "loss": 2.3644, + "step": 231 + }, + { + "epoch": 0.25424657534246575, + "grad_norm": 2.258049964904785, + "learning_rate": 3.978102189781022e-06, + "loss": 2.3727, + "step": 232 + }, + { + "epoch": 0.2553424657534247, + "grad_norm": 2.4630489349365234, + "learning_rate": 3.996350364963504e-06, + "loss": 2.267, + "step": 233 + }, + { + "epoch": 0.25643835616438354, + "grad_norm": 3.1469743251800537, + "learning_rate": 4.014598540145986e-06, + "loss": 2.3432, + "step": 234 + }, + { + "epoch": 0.25753424657534246, + "grad_norm": 2.6277477741241455, + "learning_rate": 4.032846715328467e-06, + "loss": 2.2545, + "step": 235 + }, + { + "epoch": 0.2586301369863014, + "grad_norm": 2.404141902923584, + "learning_rate": 4.05109489051095e-06, + "loss": 2.3109, + "step": 236 + }, + { + "epoch": 0.25972602739726025, + "grad_norm": 2.4545605182647705, + "learning_rate": 4.069343065693431e-06, + "loss": 2.3112, + "step": 237 + }, + { + "epoch": 0.2608219178082192, + "grad_norm": 2.237638235092163, + "learning_rate": 4.0875912408759126e-06, + "loss": 2.3, + "step": 238 + }, + { + "epoch": 0.2619178082191781, + "grad_norm": 2.192811965942383, + "learning_rate": 4.105839416058395e-06, + "loss": 2.3108, + "step": 239 + }, + { + "epoch": 0.26301369863013696, + "grad_norm": 2.5058720111846924, + "learning_rate": 4.124087591240876e-06, + "loss": 2.3635, + "step": 240 + }, + { + "epoch": 0.26301369863013696, + "eval_loss": 2.4219377040863037, + "eval_runtime": 45.3561, + "eval_samples_per_second": 22.048, + "eval_steps_per_second": 0.926, + "step": 240 + }, + { + "epoch": 0.2641095890410959, + "grad_norm": 2.0708935260772705, + "learning_rate": 4.1423357664233586e-06, + "loss": 2.3129, + "step": 241 + }, + { + "epoch": 0.2652054794520548, + "grad_norm": 2.6353116035461426, + "learning_rate": 4.16058394160584e-06, + "loss": 2.3011, + "step": 242 + }, + { + "epoch": 0.2663013698630137, + "grad_norm": 2.8392436504364014, + "learning_rate": 4.1788321167883214e-06, + "loss": 2.3643, + "step": 243 + }, + { + "epoch": 0.2673972602739726, + "grad_norm": 3.1773295402526855, + "learning_rate": 4.197080291970803e-06, + "loss": 2.2642, + "step": 244 + }, + { + "epoch": 0.2684931506849315, + "grad_norm": 2.4757354259490967, + "learning_rate": 4.215328467153285e-06, + "loss": 2.3211, + "step": 245 + }, + { + "epoch": 0.26958904109589044, + "grad_norm": 2.5799412727355957, + "learning_rate": 4.233576642335767e-06, + "loss": 2.2909, + "step": 246 + }, + { + "epoch": 0.2706849315068493, + "grad_norm": 2.368502378463745, + "learning_rate": 4.251824817518249e-06, + "loss": 2.2951, + "step": 247 + }, + { + "epoch": 0.2717808219178082, + "grad_norm": 2.114969253540039, + "learning_rate": 4.27007299270073e-06, + "loss": 2.3124, + "step": 248 + }, + { + "epoch": 0.27287671232876715, + "grad_norm": 3.056820869445801, + "learning_rate": 4.288321167883212e-06, + "loss": 2.2923, + "step": 249 + }, + { + "epoch": 0.273972602739726, + "grad_norm": 2.7752203941345215, + "learning_rate": 4.306569343065693e-06, + "loss": 2.2874, + "step": 250 + }, + { + "epoch": 0.273972602739726, + "eval_loss": 2.4229912757873535, + "eval_runtime": 52.5941, + "eval_samples_per_second": 19.014, + "eval_steps_per_second": 0.799, + "step": 250 + }, + { + "epoch": 0.27506849315068493, + "grad_norm": 3.045740842819214, + "learning_rate": 4.3248175182481755e-06, + "loss": 2.2686, + "step": 251 + }, + { + "epoch": 0.27616438356164386, + "grad_norm": 2.646402359008789, + "learning_rate": 4.343065693430658e-06, + "loss": 2.3247, + "step": 252 + }, + { + "epoch": 0.2772602739726027, + "grad_norm": 3.274994134902954, + "learning_rate": 4.361313868613139e-06, + "loss": 2.3093, + "step": 253 + }, + { + "epoch": 0.27835616438356164, + "grad_norm": 2.352246046066284, + "learning_rate": 4.379562043795621e-06, + "loss": 2.3124, + "step": 254 + }, + { + "epoch": 0.27945205479452057, + "grad_norm": 3.248717784881592, + "learning_rate": 4.397810218978102e-06, + "loss": 2.3417, + "step": 255 + }, + { + "epoch": 0.28054794520547943, + "grad_norm": 3.0919013023376465, + "learning_rate": 4.416058394160584e-06, + "loss": 2.3093, + "step": 256 + }, + { + "epoch": 0.28164383561643835, + "grad_norm": 2.1174702644348145, + "learning_rate": 4.434306569343066e-06, + "loss": 2.305, + "step": 257 + }, + { + "epoch": 0.2827397260273973, + "grad_norm": 2.913797378540039, + "learning_rate": 4.452554744525548e-06, + "loss": 2.2831, + "step": 258 + }, + { + "epoch": 0.28383561643835614, + "grad_norm": 2.693089723587036, + "learning_rate": 4.4708029197080295e-06, + "loss": 2.3206, + "step": 259 + }, + { + "epoch": 0.28493150684931506, + "grad_norm": 2.521411895751953, + "learning_rate": 4.489051094890512e-06, + "loss": 2.3196, + "step": 260 + }, + { + "epoch": 0.28493150684931506, + "eval_loss": 2.4178109169006348, + "eval_runtime": 50.6753, + "eval_samples_per_second": 19.733, + "eval_steps_per_second": 0.829, + "step": 260 + }, + { + "epoch": 0.286027397260274, + "grad_norm": 2.346346855163574, + "learning_rate": 4.507299270072993e-06, + "loss": 2.3433, + "step": 261 + }, + { + "epoch": 0.28712328767123285, + "grad_norm": 2.520368814468384, + "learning_rate": 4.525547445255475e-06, + "loss": 2.3174, + "step": 262 + }, + { + "epoch": 0.2882191780821918, + "grad_norm": 2.550347328186035, + "learning_rate": 4.543795620437956e-06, + "loss": 2.3072, + "step": 263 + }, + { + "epoch": 0.2893150684931507, + "grad_norm": 2.268662691116333, + "learning_rate": 4.562043795620438e-06, + "loss": 2.2763, + "step": 264 + }, + { + "epoch": 0.29041095890410956, + "grad_norm": 2.445798873901367, + "learning_rate": 4.580291970802921e-06, + "loss": 2.3395, + "step": 265 + }, + { + "epoch": 0.2915068493150685, + "grad_norm": 2.5391135215759277, + "learning_rate": 4.598540145985402e-06, + "loss": 2.3496, + "step": 266 + }, + { + "epoch": 0.2926027397260274, + "grad_norm": 2.3553097248077393, + "learning_rate": 4.6167883211678835e-06, + "loss": 2.3243, + "step": 267 + }, + { + "epoch": 0.2936986301369863, + "grad_norm": 2.448089599609375, + "learning_rate": 4.635036496350365e-06, + "loss": 2.2887, + "step": 268 + }, + { + "epoch": 0.2947945205479452, + "grad_norm": 2.1600170135498047, + "learning_rate": 4.653284671532847e-06, + "loss": 2.2816, + "step": 269 + }, + { + "epoch": 0.2958904109589041, + "grad_norm": 2.3783180713653564, + "learning_rate": 4.671532846715329e-06, + "loss": 2.3275, + "step": 270 + }, + { + "epoch": 0.2958904109589041, + "eval_loss": 2.413832902908325, + "eval_runtime": 48.7524, + "eval_samples_per_second": 20.512, + "eval_steps_per_second": 0.861, + "step": 270 + }, + { + "epoch": 0.29698630136986304, + "grad_norm": 2.468320369720459, + "learning_rate": 4.689781021897811e-06, + "loss": 2.3358, + "step": 271 + }, + { + "epoch": 0.2980821917808219, + "grad_norm": 2.401634454727173, + "learning_rate": 4.708029197080292e-06, + "loss": 2.2991, + "step": 272 + }, + { + "epoch": 0.2991780821917808, + "grad_norm": 2.4604203701019287, + "learning_rate": 4.726277372262774e-06, + "loss": 2.3496, + "step": 273 + }, + { + "epoch": 0.30027397260273975, + "grad_norm": 2.0338871479034424, + "learning_rate": 4.744525547445255e-06, + "loss": 2.3115, + "step": 274 + }, + { + "epoch": 0.3013698630136986, + "grad_norm": 2.5404436588287354, + "learning_rate": 4.7627737226277375e-06, + "loss": 2.2665, + "step": 275 + }, + { + "epoch": 0.30246575342465754, + "grad_norm": 2.5369458198547363, + "learning_rate": 4.78102189781022e-06, + "loss": 2.3072, + "step": 276 + }, + { + "epoch": 0.30356164383561646, + "grad_norm": 2.471522569656372, + "learning_rate": 4.799270072992701e-06, + "loss": 2.3113, + "step": 277 + }, + { + "epoch": 0.3046575342465753, + "grad_norm": 2.3890135288238525, + "learning_rate": 4.8175182481751835e-06, + "loss": 2.2838, + "step": 278 + }, + { + "epoch": 0.30575342465753425, + "grad_norm": 2.2485053539276123, + "learning_rate": 4.835766423357665e-06, + "loss": 2.3052, + "step": 279 + }, + { + "epoch": 0.30684931506849317, + "grad_norm": 2.1090753078460693, + "learning_rate": 4.854014598540146e-06, + "loss": 2.3054, + "step": 280 + }, + { + "epoch": 0.30684931506849317, + "eval_loss": 2.412548780441284, + "eval_runtime": 48.5034, + "eval_samples_per_second": 20.617, + "eval_steps_per_second": 0.866, + "step": 280 + }, + { + "epoch": 0.30794520547945203, + "grad_norm": 2.299133777618408, + "learning_rate": 4.872262773722628e-06, + "loss": 2.2599, + "step": 281 + }, + { + "epoch": 0.30904109589041096, + "grad_norm": 2.177905797958374, + "learning_rate": 4.89051094890511e-06, + "loss": 2.2745, + "step": 282 + }, + { + "epoch": 0.3101369863013699, + "grad_norm": 2.3178462982177734, + "learning_rate": 4.9087591240875915e-06, + "loss": 2.2562, + "step": 283 + }, + { + "epoch": 0.31123287671232874, + "grad_norm": 2.5088376998901367, + "learning_rate": 4.927007299270074e-06, + "loss": 2.2945, + "step": 284 + }, + { + "epoch": 0.31232876712328766, + "grad_norm": 2.095379590988159, + "learning_rate": 4.945255474452555e-06, + "loss": 2.3038, + "step": 285 + }, + { + "epoch": 0.3134246575342466, + "grad_norm": 2.1792995929718018, + "learning_rate": 4.963503649635037e-06, + "loss": 2.2702, + "step": 286 + }, + { + "epoch": 0.31452054794520545, + "grad_norm": 2.3001186847686768, + "learning_rate": 4.981751824817518e-06, + "loss": 2.3482, + "step": 287 + }, + { + "epoch": 0.3156164383561644, + "grad_norm": 2.1576106548309326, + "learning_rate": 5e-06, + "loss": 2.3391, + "step": 288 + }, + { + "epoch": 0.3167123287671233, + "grad_norm": 2.3715577125549316, + "learning_rate": 4.997969130787978e-06, + "loss": 2.3049, + "step": 289 + }, + { + "epoch": 0.3178082191780822, + "grad_norm": 2.0316715240478516, + "learning_rate": 4.995938261575955e-06, + "loss": 2.3111, + "step": 290 + }, + { + "epoch": 0.3178082191780822, + "eval_loss": 2.40299654006958, + "eval_runtime": 50.0873, + "eval_samples_per_second": 19.965, + "eval_steps_per_second": 0.839, + "step": 290 + }, + { + "epoch": 0.3189041095890411, + "grad_norm": 2.1135756969451904, + "learning_rate": 4.993907392363932e-06, + "loss": 2.2465, + "step": 291 + }, + { + "epoch": 0.32, + "grad_norm": 2.335655927658081, + "learning_rate": 4.9918765231519095e-06, + "loss": 2.31, + "step": 292 + }, + { + "epoch": 0.32109589041095893, + "grad_norm": 2.3255794048309326, + "learning_rate": 4.989845653939887e-06, + "loss": 2.2531, + "step": 293 + }, + { + "epoch": 0.3221917808219178, + "grad_norm": 2.111055612564087, + "learning_rate": 4.987814784727864e-06, + "loss": 2.2799, + "step": 294 + }, + { + "epoch": 0.3232876712328767, + "grad_norm": 2.1773171424865723, + "learning_rate": 4.985783915515841e-06, + "loss": 2.2884, + "step": 295 + }, + { + "epoch": 0.32438356164383564, + "grad_norm": 2.3559885025024414, + "learning_rate": 4.983753046303819e-06, + "loss": 2.2711, + "step": 296 + }, + { + "epoch": 0.3254794520547945, + "grad_norm": 2.120802879333496, + "learning_rate": 4.981722177091796e-06, + "loss": 2.2946, + "step": 297 + }, + { + "epoch": 0.3265753424657534, + "grad_norm": 2.2319388389587402, + "learning_rate": 4.979691307879773e-06, + "loss": 2.2771, + "step": 298 + }, + { + "epoch": 0.32767123287671235, + "grad_norm": 2.208740711212158, + "learning_rate": 4.9776604386677504e-06, + "loss": 2.3003, + "step": 299 + }, + { + "epoch": 0.3287671232876712, + "grad_norm": 2.543060541152954, + "learning_rate": 4.975629569455728e-06, + "loss": 2.2891, + "step": 300 + }, + { + "epoch": 0.3287671232876712, + "eval_loss": 2.398495674133301, + "eval_runtime": 51.1973, + "eval_samples_per_second": 19.532, + "eval_steps_per_second": 0.82, + "step": 300 + }, + { + "epoch": 0.32986301369863014, + "grad_norm": 2.086219072341919, + "learning_rate": 4.973598700243705e-06, + "loss": 2.3286, + "step": 301 + }, + { + "epoch": 0.33095890410958906, + "grad_norm": 2.3673319816589355, + "learning_rate": 4.971567831031681e-06, + "loss": 2.2672, + "step": 302 + }, + { + "epoch": 0.3320547945205479, + "grad_norm": 2.167593240737915, + "learning_rate": 4.9695369618196595e-06, + "loss": 2.264, + "step": 303 + }, + { + "epoch": 0.33315068493150685, + "grad_norm": 2.7218785285949707, + "learning_rate": 4.967506092607636e-06, + "loss": 2.1737, + "step": 304 + }, + { + "epoch": 0.33424657534246577, + "grad_norm": 2.673245429992676, + "learning_rate": 4.965475223395614e-06, + "loss": 2.2439, + "step": 305 + }, + { + "epoch": 0.33534246575342463, + "grad_norm": 2.0341989994049072, + "learning_rate": 4.9634443541835905e-06, + "loss": 2.2614, + "step": 306 + }, + { + "epoch": 0.33643835616438356, + "grad_norm": 1.9314247369766235, + "learning_rate": 4.961413484971569e-06, + "loss": 2.2592, + "step": 307 + }, + { + "epoch": 0.3375342465753425, + "grad_norm": 2.537794589996338, + "learning_rate": 4.959382615759545e-06, + "loss": 2.2859, + "step": 308 + }, + { + "epoch": 0.33863013698630134, + "grad_norm": 2.0811166763305664, + "learning_rate": 4.957351746547523e-06, + "loss": 2.2868, + "step": 309 + }, + { + "epoch": 0.33972602739726027, + "grad_norm": 2.783701181411743, + "learning_rate": 4.9553208773355e-06, + "loss": 2.2649, + "step": 310 + }, + { + "epoch": 0.33972602739726027, + "eval_loss": 2.400538921356201, + "eval_runtime": 43.185, + "eval_samples_per_second": 23.156, + "eval_steps_per_second": 0.973, + "step": 310 + }, + { + "epoch": 0.3408219178082192, + "grad_norm": 2.8541319370269775, + "learning_rate": 4.953290008123478e-06, + "loss": 2.3729, + "step": 311 + }, + { + "epoch": 0.3419178082191781, + "grad_norm": 2.1852617263793945, + "learning_rate": 4.951259138911454e-06, + "loss": 2.2885, + "step": 312 + }, + { + "epoch": 0.343013698630137, + "grad_norm": 2.872201919555664, + "learning_rate": 4.9492282696994315e-06, + "loss": 2.2353, + "step": 313 + }, + { + "epoch": 0.3441095890410959, + "grad_norm": 2.5140910148620605, + "learning_rate": 4.947197400487409e-06, + "loss": 2.3364, + "step": 314 + }, + { + "epoch": 0.3452054794520548, + "grad_norm": 2.619974374771118, + "learning_rate": 4.945166531275386e-06, + "loss": 2.2721, + "step": 315 + }, + { + "epoch": 0.3463013698630137, + "grad_norm": 2.4567618370056152, + "learning_rate": 4.943135662063363e-06, + "loss": 2.26, + "step": 316 + }, + { + "epoch": 0.3473972602739726, + "grad_norm": 2.9603192806243896, + "learning_rate": 4.9411047928513406e-06, + "loss": 2.2861, + "step": 317 + }, + { + "epoch": 0.34849315068493153, + "grad_norm": 2.115262985229492, + "learning_rate": 4.939073923639318e-06, + "loss": 2.3125, + "step": 318 + }, + { + "epoch": 0.3495890410958904, + "grad_norm": 2.693024158477783, + "learning_rate": 4.937043054427295e-06, + "loss": 2.3054, + "step": 319 + }, + { + "epoch": 0.3506849315068493, + "grad_norm": 2.3439996242523193, + "learning_rate": 4.935012185215272e-06, + "loss": 2.2576, + "step": 320 + }, + { + "epoch": 0.3506849315068493, + "eval_loss": 2.3920650482177734, + "eval_runtime": 47.7451, + "eval_samples_per_second": 20.945, + "eval_steps_per_second": 0.88, + "step": 320 + }, + { + "epoch": 0.35178082191780824, + "grad_norm": 2.3009634017944336, + "learning_rate": 4.93298131600325e-06, + "loss": 2.2454, + "step": 321 + }, + { + "epoch": 0.3528767123287671, + "grad_norm": 2.593203544616699, + "learning_rate": 4.930950446791227e-06, + "loss": 2.2965, + "step": 322 + }, + { + "epoch": 0.353972602739726, + "grad_norm": 2.9069650173187256, + "learning_rate": 4.928919577579204e-06, + "loss": 2.2785, + "step": 323 + }, + { + "epoch": 0.35506849315068495, + "grad_norm": 2.1021759510040283, + "learning_rate": 4.9268887083671815e-06, + "loss": 2.2857, + "step": 324 + }, + { + "epoch": 0.3561643835616438, + "grad_norm": 2.0330185890197754, + "learning_rate": 4.924857839155159e-06, + "loss": 2.2697, + "step": 325 + }, + { + "epoch": 0.35726027397260274, + "grad_norm": 2.1318063735961914, + "learning_rate": 4.922826969943136e-06, + "loss": 2.2698, + "step": 326 + }, + { + "epoch": 0.35835616438356166, + "grad_norm": 2.0157387256622314, + "learning_rate": 4.920796100731113e-06, + "loss": 2.292, + "step": 327 + }, + { + "epoch": 0.3594520547945205, + "grad_norm": 2.3266026973724365, + "learning_rate": 4.918765231519091e-06, + "loss": 2.2215, + "step": 328 + }, + { + "epoch": 0.36054794520547945, + "grad_norm": 2.066305160522461, + "learning_rate": 4.916734362307068e-06, + "loss": 2.2758, + "step": 329 + }, + { + "epoch": 0.36164383561643837, + "grad_norm": 2.0219993591308594, + "learning_rate": 4.914703493095045e-06, + "loss": 2.2381, + "step": 330 + }, + { + "epoch": 0.36164383561643837, + "eval_loss": 2.394064426422119, + "eval_runtime": 51.5482, + "eval_samples_per_second": 19.399, + "eval_steps_per_second": 0.815, + "step": 330 + }, + { + "epoch": 0.36273972602739724, + "grad_norm": 2.3964319229125977, + "learning_rate": 4.912672623883022e-06, + "loss": 2.2384, + "step": 331 + }, + { + "epoch": 0.36383561643835616, + "grad_norm": 2.5466530323028564, + "learning_rate": 4.910641754671e-06, + "loss": 2.2428, + "step": 332 + }, + { + "epoch": 0.3649315068493151, + "grad_norm": 2.236152410507202, + "learning_rate": 4.908610885458977e-06, + "loss": 2.2639, + "step": 333 + }, + { + "epoch": 0.36602739726027395, + "grad_norm": 2.7267303466796875, + "learning_rate": 4.906580016246954e-06, + "loss": 2.2608, + "step": 334 + }, + { + "epoch": 0.36712328767123287, + "grad_norm": 1.9403765201568604, + "learning_rate": 4.9045491470349315e-06, + "loss": 2.2962, + "step": 335 + }, + { + "epoch": 0.3682191780821918, + "grad_norm": 2.89070463180542, + "learning_rate": 4.902518277822909e-06, + "loss": 2.2847, + "step": 336 + }, + { + "epoch": 0.3693150684931507, + "grad_norm": 2.6427228450775146, + "learning_rate": 4.900487408610886e-06, + "loss": 2.2734, + "step": 337 + }, + { + "epoch": 0.3704109589041096, + "grad_norm": 2.1680362224578857, + "learning_rate": 4.898456539398863e-06, + "loss": 2.2869, + "step": 338 + }, + { + "epoch": 0.3715068493150685, + "grad_norm": 3.0160160064697266, + "learning_rate": 4.89642567018684e-06, + "loss": 2.3043, + "step": 339 + }, + { + "epoch": 0.3726027397260274, + "grad_norm": 1.9858418703079224, + "learning_rate": 4.894394800974818e-06, + "loss": 2.3422, + "step": 340 + }, + { + "epoch": 0.3726027397260274, + "eval_loss": 2.387559652328491, + "eval_runtime": 52.2455, + "eval_samples_per_second": 19.14, + "eval_steps_per_second": 0.804, + "step": 340 + }, + { + "epoch": 0.3736986301369863, + "grad_norm": 2.243699073791504, + "learning_rate": 4.892363931762794e-06, + "loss": 2.2394, + "step": 341 + }, + { + "epoch": 0.3747945205479452, + "grad_norm": 2.371299982070923, + "learning_rate": 4.8903330625507725e-06, + "loss": 2.2856, + "step": 342 + }, + { + "epoch": 0.37589041095890413, + "grad_norm": 2.3769736289978027, + "learning_rate": 4.888302193338749e-06, + "loss": 2.246, + "step": 343 + }, + { + "epoch": 0.376986301369863, + "grad_norm": 2.3209030628204346, + "learning_rate": 4.886271324126727e-06, + "loss": 2.2331, + "step": 344 + }, + { + "epoch": 0.3780821917808219, + "grad_norm": 2.747319221496582, + "learning_rate": 4.8842404549147034e-06, + "loss": 2.3115, + "step": 345 + }, + { + "epoch": 0.37917808219178084, + "grad_norm": 1.8599231243133545, + "learning_rate": 4.8822095857026816e-06, + "loss": 2.2994, + "step": 346 + }, + { + "epoch": 0.3802739726027397, + "grad_norm": 2.0358943939208984, + "learning_rate": 4.880178716490658e-06, + "loss": 2.271, + "step": 347 + }, + { + "epoch": 0.38136986301369863, + "grad_norm": 2.409106731414795, + "learning_rate": 4.878147847278636e-06, + "loss": 2.2281, + "step": 348 + }, + { + "epoch": 0.38246575342465755, + "grad_norm": 2.1641807556152344, + "learning_rate": 4.8761169780666125e-06, + "loss": 2.2437, + "step": 349 + }, + { + "epoch": 0.3835616438356164, + "grad_norm": 2.6917359828948975, + "learning_rate": 4.874086108854591e-06, + "loss": 2.226, + "step": 350 + }, + { + "epoch": 0.3835616438356164, + "eval_loss": 2.382991313934326, + "eval_runtime": 48.9534, + "eval_samples_per_second": 20.428, + "eval_steps_per_second": 0.858, + "step": 350 + }, + { + "epoch": 0.38465753424657534, + "grad_norm": 2.364215612411499, + "learning_rate": 4.872055239642567e-06, + "loss": 2.2319, + "step": 351 + }, + { + "epoch": 0.38575342465753426, + "grad_norm": 2.317643165588379, + "learning_rate": 4.870024370430544e-06, + "loss": 2.2096, + "step": 352 + }, + { + "epoch": 0.3868493150684931, + "grad_norm": 2.2283246517181396, + "learning_rate": 4.867993501218522e-06, + "loss": 2.2634, + "step": 353 + }, + { + "epoch": 0.38794520547945205, + "grad_norm": 2.498391628265381, + "learning_rate": 4.865962632006499e-06, + "loss": 2.2813, + "step": 354 + }, + { + "epoch": 0.38904109589041097, + "grad_norm": 2.3889403343200684, + "learning_rate": 4.863931762794476e-06, + "loss": 2.2361, + "step": 355 + }, + { + "epoch": 0.39013698630136984, + "grad_norm": 2.2049975395202637, + "learning_rate": 4.8619008935824535e-06, + "loss": 2.2583, + "step": 356 + }, + { + "epoch": 0.39123287671232876, + "grad_norm": 2.132840871810913, + "learning_rate": 4.859870024370431e-06, + "loss": 2.2823, + "step": 357 + }, + { + "epoch": 0.3923287671232877, + "grad_norm": 2.2692184448242188, + "learning_rate": 4.857839155158408e-06, + "loss": 2.2939, + "step": 358 + }, + { + "epoch": 0.3934246575342466, + "grad_norm": 2.0501933097839355, + "learning_rate": 4.855808285946385e-06, + "loss": 2.2976, + "step": 359 + }, + { + "epoch": 0.39452054794520547, + "grad_norm": 2.170236587524414, + "learning_rate": 4.8537774167343626e-06, + "loss": 2.2748, + "step": 360 + }, + { + "epoch": 0.39452054794520547, + "eval_loss": 2.3783886432647705, + "eval_runtime": 46.7324, + "eval_samples_per_second": 21.398, + "eval_steps_per_second": 0.899, + "step": 360 + }, + { + "epoch": 0.3956164383561644, + "grad_norm": 2.283137083053589, + "learning_rate": 4.85174654752234e-06, + "loss": 2.2705, + "step": 361 + }, + { + "epoch": 0.3967123287671233, + "grad_norm": 2.195984363555908, + "learning_rate": 4.849715678310317e-06, + "loss": 2.26, + "step": 362 + }, + { + "epoch": 0.3978082191780822, + "grad_norm": 2.399935007095337, + "learning_rate": 4.847684809098294e-06, + "loss": 2.2639, + "step": 363 + }, + { + "epoch": 0.3989041095890411, + "grad_norm": 2.411280393600464, + "learning_rate": 4.845653939886272e-06, + "loss": 2.2523, + "step": 364 + }, + { + "epoch": 0.4, + "grad_norm": 2.2033896446228027, + "learning_rate": 4.843623070674249e-06, + "loss": 2.2722, + "step": 365 + }, + { + "epoch": 0.4010958904109589, + "grad_norm": 2.2990238666534424, + "learning_rate": 4.841592201462226e-06, + "loss": 2.2411, + "step": 366 + }, + { + "epoch": 0.4021917808219178, + "grad_norm": 1.911453127861023, + "learning_rate": 4.8395613322502035e-06, + "loss": 2.2921, + "step": 367 + }, + { + "epoch": 0.40328767123287673, + "grad_norm": 2.453195095062256, + "learning_rate": 4.837530463038181e-06, + "loss": 2.218, + "step": 368 + }, + { + "epoch": 0.4043835616438356, + "grad_norm": 2.126992702484131, + "learning_rate": 4.835499593826158e-06, + "loss": 2.2628, + "step": 369 + }, + { + "epoch": 0.4054794520547945, + "grad_norm": 1.9305835962295532, + "learning_rate": 4.833468724614135e-06, + "loss": 2.2726, + "step": 370 + }, + { + "epoch": 0.4054794520547945, + "eval_loss": 2.3741114139556885, + "eval_runtime": 49.2956, + "eval_samples_per_second": 20.286, + "eval_steps_per_second": 0.852, + "step": 370 + }, + { + "epoch": 0.40657534246575344, + "grad_norm": 2.222024917602539, + "learning_rate": 4.831437855402113e-06, + "loss": 2.2634, + "step": 371 + }, + { + "epoch": 0.4076712328767123, + "grad_norm": 2.473958969116211, + "learning_rate": 4.82940698619009e-06, + "loss": 2.1817, + "step": 372 + }, + { + "epoch": 0.40876712328767123, + "grad_norm": 1.9208838939666748, + "learning_rate": 4.827376116978067e-06, + "loss": 2.2353, + "step": 373 + }, + { + "epoch": 0.40986301369863015, + "grad_norm": 2.7226450443267822, + "learning_rate": 4.8253452477660444e-06, + "loss": 2.2831, + "step": 374 + }, + { + "epoch": 0.410958904109589, + "grad_norm": 1.9078222513198853, + "learning_rate": 4.823314378554022e-06, + "loss": 2.264, + "step": 375 + }, + { + "epoch": 0.41205479452054794, + "grad_norm": 2.3946311473846436, + "learning_rate": 4.821283509341998e-06, + "loss": 2.251, + "step": 376 + }, + { + "epoch": 0.41315068493150686, + "grad_norm": 1.847343921661377, + "learning_rate": 4.819252640129976e-06, + "loss": 2.2441, + "step": 377 + }, + { + "epoch": 0.41424657534246573, + "grad_norm": 2.475385904312134, + "learning_rate": 4.817221770917953e-06, + "loss": 2.2639, + "step": 378 + }, + { + "epoch": 0.41534246575342465, + "grad_norm": 2.1396560668945312, + "learning_rate": 4.815190901705931e-06, + "loss": 2.2353, + "step": 379 + }, + { + "epoch": 0.41643835616438357, + "grad_norm": 2.3868885040283203, + "learning_rate": 4.813160032493907e-06, + "loss": 2.2551, + "step": 380 + }, + { + "epoch": 0.41643835616438357, + "eval_loss": 2.369168519973755, + "eval_runtime": 45.6131, + "eval_samples_per_second": 21.924, + "eval_steps_per_second": 0.921, + "step": 380 + }, + { + "epoch": 0.41753424657534244, + "grad_norm": 2.0193872451782227, + "learning_rate": 4.811129163281885e-06, + "loss": 2.2399, + "step": 381 + }, + { + "epoch": 0.41863013698630136, + "grad_norm": 2.4655935764312744, + "learning_rate": 4.809098294069862e-06, + "loss": 2.2729, + "step": 382 + }, + { + "epoch": 0.4197260273972603, + "grad_norm": 1.880974292755127, + "learning_rate": 4.80706742485784e-06, + "loss": 2.2681, + "step": 383 + }, + { + "epoch": 0.4208219178082192, + "grad_norm": 2.1912078857421875, + "learning_rate": 4.805036555645816e-06, + "loss": 2.2961, + "step": 384 + }, + { + "epoch": 0.42191780821917807, + "grad_norm": 1.985931634902954, + "learning_rate": 4.8030056864337945e-06, + "loss": 2.2407, + "step": 385 + }, + { + "epoch": 0.423013698630137, + "grad_norm": 2.4353220462799072, + "learning_rate": 4.800974817221771e-06, + "loss": 2.2867, + "step": 386 + }, + { + "epoch": 0.4241095890410959, + "grad_norm": 1.8810508251190186, + "learning_rate": 4.798943948009749e-06, + "loss": 2.2498, + "step": 387 + }, + { + "epoch": 0.4252054794520548, + "grad_norm": 2.195033073425293, + "learning_rate": 4.7969130787977254e-06, + "loss": 2.2532, + "step": 388 + }, + { + "epoch": 0.4263013698630137, + "grad_norm": 1.855372428894043, + "learning_rate": 4.794882209585703e-06, + "loss": 2.2538, + "step": 389 + }, + { + "epoch": 0.4273972602739726, + "grad_norm": 2.4465434551239014, + "learning_rate": 4.79285134037368e-06, + "loss": 2.2724, + "step": 390 + }, + { + "epoch": 0.4273972602739726, + "eval_loss": 2.364928960800171, + "eval_runtime": 48.0245, + "eval_samples_per_second": 20.823, + "eval_steps_per_second": 0.875, + "step": 390 + }, + { + "epoch": 0.4284931506849315, + "grad_norm": 1.8911314010620117, + "learning_rate": 4.790820471161657e-06, + "loss": 2.2291, + "step": 391 + }, + { + "epoch": 0.4295890410958904, + "grad_norm": 2.1975977420806885, + "learning_rate": 4.7887896019496345e-06, + "loss": 2.2373, + "step": 392 + }, + { + "epoch": 0.43068493150684933, + "grad_norm": 2.015331506729126, + "learning_rate": 4.786758732737612e-06, + "loss": 2.2483, + "step": 393 + }, + { + "epoch": 0.4317808219178082, + "grad_norm": 2.039339065551758, + "learning_rate": 4.784727863525589e-06, + "loss": 2.2746, + "step": 394 + }, + { + "epoch": 0.4328767123287671, + "grad_norm": 1.977513074874878, + "learning_rate": 4.782696994313566e-06, + "loss": 2.2275, + "step": 395 + }, + { + "epoch": 0.43397260273972604, + "grad_norm": 2.147217035293579, + "learning_rate": 4.780666125101544e-06, + "loss": 2.252, + "step": 396 + }, + { + "epoch": 0.4350684931506849, + "grad_norm": 2.1799919605255127, + "learning_rate": 4.778635255889521e-06, + "loss": 2.1975, + "step": 397 + }, + { + "epoch": 0.43616438356164383, + "grad_norm": 2.582240343093872, + "learning_rate": 4.776604386677498e-06, + "loss": 2.2294, + "step": 398 + }, + { + "epoch": 0.43726027397260275, + "grad_norm": 1.9127846956253052, + "learning_rate": 4.7745735174654755e-06, + "loss": 2.2319, + "step": 399 + }, + { + "epoch": 0.4383561643835616, + "grad_norm": 2.2444543838500977, + "learning_rate": 4.772542648253453e-06, + "loss": 2.2857, + "step": 400 + }, + { + "epoch": 0.4383561643835616, + "eval_loss": 2.3653724193573, + "eval_runtime": 43.8304, + "eval_samples_per_second": 22.815, + "eval_steps_per_second": 0.958, + "step": 400 + }, + { + "epoch": 0.43945205479452054, + "grad_norm": 1.9913676977157593, + "learning_rate": 4.77051177904143e-06, + "loss": 2.274, + "step": 401 + }, + { + "epoch": 0.44054794520547946, + "grad_norm": 2.2850067615509033, + "learning_rate": 4.768480909829407e-06, + "loss": 2.2209, + "step": 402 + }, + { + "epoch": 0.44164383561643833, + "grad_norm": 2.4984688758850098, + "learning_rate": 4.766450040617385e-06, + "loss": 2.263, + "step": 403 + }, + { + "epoch": 0.44273972602739725, + "grad_norm": 2.3244640827178955, + "learning_rate": 4.764419171405362e-06, + "loss": 2.2368, + "step": 404 + }, + { + "epoch": 0.4438356164383562, + "grad_norm": 2.7579681873321533, + "learning_rate": 4.762388302193339e-06, + "loss": 2.2762, + "step": 405 + }, + { + "epoch": 0.4449315068493151, + "grad_norm": 2.1996467113494873, + "learning_rate": 4.760357432981316e-06, + "loss": 2.2536, + "step": 406 + }, + { + "epoch": 0.44602739726027396, + "grad_norm": 2.3767571449279785, + "learning_rate": 4.758326563769294e-06, + "loss": 2.2353, + "step": 407 + }, + { + "epoch": 0.4471232876712329, + "grad_norm": 1.9148359298706055, + "learning_rate": 4.756295694557271e-06, + "loss": 2.2416, + "step": 408 + }, + { + "epoch": 0.4482191780821918, + "grad_norm": 2.6499507427215576, + "learning_rate": 4.754264825345248e-06, + "loss": 2.2641, + "step": 409 + }, + { + "epoch": 0.44931506849315067, + "grad_norm": 1.9890286922454834, + "learning_rate": 4.7522339561332255e-06, + "loss": 2.2148, + "step": 410 + }, + { + "epoch": 0.44931506849315067, + "eval_loss": 2.3670520782470703, + "eval_runtime": 41.1722, + "eval_samples_per_second": 24.288, + "eval_steps_per_second": 1.02, + "step": 410 + }, + { + "epoch": 0.4504109589041096, + "grad_norm": 2.575968027114868, + "learning_rate": 4.750203086921203e-06, + "loss": 2.2254, + "step": 411 + }, + { + "epoch": 0.4515068493150685, + "grad_norm": 2.015174388885498, + "learning_rate": 4.74817221770918e-06, + "loss": 2.1881, + "step": 412 + }, + { + "epoch": 0.4526027397260274, + "grad_norm": 2.44292950630188, + "learning_rate": 4.7461413484971565e-06, + "loss": 2.256, + "step": 413 + }, + { + "epoch": 0.4536986301369863, + "grad_norm": 1.9808441400527954, + "learning_rate": 4.744110479285135e-06, + "loss": 2.285, + "step": 414 + }, + { + "epoch": 0.4547945205479452, + "grad_norm": 2.009481906890869, + "learning_rate": 4.742079610073111e-06, + "loss": 2.2018, + "step": 415 + }, + { + "epoch": 0.4558904109589041, + "grad_norm": 2.1466031074523926, + "learning_rate": 4.740048740861089e-06, + "loss": 2.224, + "step": 416 + }, + { + "epoch": 0.456986301369863, + "grad_norm": 2.1337907314300537, + "learning_rate": 4.738017871649066e-06, + "loss": 2.2404, + "step": 417 + }, + { + "epoch": 0.45808219178082193, + "grad_norm": 1.94125235080719, + "learning_rate": 4.735987002437044e-06, + "loss": 2.2474, + "step": 418 + }, + { + "epoch": 0.4591780821917808, + "grad_norm": 2.379099130630493, + "learning_rate": 4.73395613322502e-06, + "loss": 2.2858, + "step": 419 + }, + { + "epoch": 0.4602739726027397, + "grad_norm": 1.844056248664856, + "learning_rate": 4.731925264012998e-06, + "loss": 2.2967, + "step": 420 + }, + { + "epoch": 0.4602739726027397, + "eval_loss": 2.3622241020202637, + "eval_runtime": 45.4674, + "eval_samples_per_second": 21.994, + "eval_steps_per_second": 0.924, + "step": 420 + }, + { + "epoch": 0.46136986301369864, + "grad_norm": 2.411895513534546, + "learning_rate": 4.729894394800975e-06, + "loss": 2.274, + "step": 421 + }, + { + "epoch": 0.4624657534246575, + "grad_norm": 2.014885425567627, + "learning_rate": 4.727863525588953e-06, + "loss": 2.2144, + "step": 422 + }, + { + "epoch": 0.46356164383561643, + "grad_norm": 2.112009048461914, + "learning_rate": 4.725832656376929e-06, + "loss": 2.1824, + "step": 423 + }, + { + "epoch": 0.46465753424657535, + "grad_norm": 2.111392021179199, + "learning_rate": 4.723801787164907e-06, + "loss": 2.2261, + "step": 424 + }, + { + "epoch": 0.4657534246575342, + "grad_norm": 2.0349185466766357, + "learning_rate": 4.721770917952885e-06, + "loss": 2.2622, + "step": 425 + }, + { + "epoch": 0.46684931506849314, + "grad_norm": 2.30222487449646, + "learning_rate": 4.719740048740861e-06, + "loss": 2.2206, + "step": 426 + }, + { + "epoch": 0.46794520547945206, + "grad_norm": 2.018789529800415, + "learning_rate": 4.717709179528839e-06, + "loss": 2.2196, + "step": 427 + }, + { + "epoch": 0.469041095890411, + "grad_norm": 2.046649932861328, + "learning_rate": 4.715678310316816e-06, + "loss": 2.2584, + "step": 428 + }, + { + "epoch": 0.47013698630136985, + "grad_norm": 2.0393869876861572, + "learning_rate": 4.713647441104794e-06, + "loss": 2.2537, + "step": 429 + }, + { + "epoch": 0.4712328767123288, + "grad_norm": 2.2788116931915283, + "learning_rate": 4.71161657189277e-06, + "loss": 2.2145, + "step": 430 + }, + { + "epoch": 0.4712328767123288, + "eval_loss": 2.3575785160064697, + "eval_runtime": 44.6687, + "eval_samples_per_second": 22.387, + "eval_steps_per_second": 0.94, + "step": 430 + }, + { + "epoch": 0.4723287671232877, + "grad_norm": 1.9950047731399536, + "learning_rate": 4.709585702680748e-06, + "loss": 2.2072, + "step": 431 + }, + { + "epoch": 0.47342465753424656, + "grad_norm": 1.9222451448440552, + "learning_rate": 4.707554833468725e-06, + "loss": 2.2654, + "step": 432 + }, + { + "epoch": 0.4745205479452055, + "grad_norm": 1.9735718965530396, + "learning_rate": 4.705523964256703e-06, + "loss": 2.3037, + "step": 433 + }, + { + "epoch": 0.4756164383561644, + "grad_norm": 2.3209776878356934, + "learning_rate": 4.703493095044679e-06, + "loss": 2.2083, + "step": 434 + }, + { + "epoch": 0.4767123287671233, + "grad_norm": 2.0993995666503906, + "learning_rate": 4.701462225832657e-06, + "loss": 2.2275, + "step": 435 + }, + { + "epoch": 0.4778082191780822, + "grad_norm": 2.334463119506836, + "learning_rate": 4.699431356620634e-06, + "loss": 2.1852, + "step": 436 + }, + { + "epoch": 0.4789041095890411, + "grad_norm": 1.9098213911056519, + "learning_rate": 4.697400487408612e-06, + "loss": 2.2829, + "step": 437 + }, + { + "epoch": 0.48, + "grad_norm": 2.3005056381225586, + "learning_rate": 4.695369618196588e-06, + "loss": 2.2532, + "step": 438 + }, + { + "epoch": 0.4810958904109589, + "grad_norm": 1.9069781303405762, + "learning_rate": 4.693338748984566e-06, + "loss": 2.2374, + "step": 439 + }, + { + "epoch": 0.4821917808219178, + "grad_norm": 2.050032138824463, + "learning_rate": 4.691307879772543e-06, + "loss": 2.2256, + "step": 440 + }, + { + "epoch": 0.4821917808219178, + "eval_loss": 2.3587422370910645, + "eval_runtime": 45.0941, + "eval_samples_per_second": 22.176, + "eval_steps_per_second": 0.931, + "step": 440 + }, + { + "epoch": 0.4832876712328767, + "grad_norm": 2.0563836097717285, + "learning_rate": 4.68927701056052e-06, + "loss": 2.1918, + "step": 441 + }, + { + "epoch": 0.4843835616438356, + "grad_norm": 2.1598386764526367, + "learning_rate": 4.6872461413484975e-06, + "loss": 2.2618, + "step": 442 + }, + { + "epoch": 0.48547945205479454, + "grad_norm": 1.9236539602279663, + "learning_rate": 4.685215272136475e-06, + "loss": 2.2477, + "step": 443 + }, + { + "epoch": 0.4865753424657534, + "grad_norm": 1.9923924207687378, + "learning_rate": 4.683184402924452e-06, + "loss": 2.2313, + "step": 444 + }, + { + "epoch": 0.4876712328767123, + "grad_norm": 2.162627696990967, + "learning_rate": 4.681153533712429e-06, + "loss": 2.2367, + "step": 445 + }, + { + "epoch": 0.48876712328767125, + "grad_norm": 2.083523988723755, + "learning_rate": 4.679122664500407e-06, + "loss": 2.2946, + "step": 446 + }, + { + "epoch": 0.4898630136986301, + "grad_norm": 2.504408359527588, + "learning_rate": 4.677091795288384e-06, + "loss": 2.2104, + "step": 447 + }, + { + "epoch": 0.49095890410958903, + "grad_norm": 1.8316532373428345, + "learning_rate": 4.675060926076361e-06, + "loss": 2.2027, + "step": 448 + }, + { + "epoch": 0.49205479452054796, + "grad_norm": 2.3337953090667725, + "learning_rate": 4.673030056864338e-06, + "loss": 2.278, + "step": 449 + }, + { + "epoch": 0.4931506849315068, + "grad_norm": 2.24102783203125, + "learning_rate": 4.670999187652316e-06, + "loss": 2.2018, + "step": 450 + }, + { + "epoch": 0.4931506849315068, + "eval_loss": 2.3545279502868652, + "eval_runtime": 45.1041, + "eval_samples_per_second": 22.171, + "eval_steps_per_second": 0.931, + "step": 450 + }, + { + "epoch": 0.49424657534246574, + "grad_norm": 2.07834792137146, + "learning_rate": 4.668968318440293e-06, + "loss": 2.1975, + "step": 451 + }, + { + "epoch": 0.49534246575342467, + "grad_norm": 1.7681723833084106, + "learning_rate": 4.66693744922827e-06, + "loss": 2.2033, + "step": 452 + }, + { + "epoch": 0.4964383561643836, + "grad_norm": 2.062134027481079, + "learning_rate": 4.6649065800162475e-06, + "loss": 2.2646, + "step": 453 + }, + { + "epoch": 0.49753424657534245, + "grad_norm": 1.94647216796875, + "learning_rate": 4.662875710804225e-06, + "loss": 2.1748, + "step": 454 + }, + { + "epoch": 0.4986301369863014, + "grad_norm": 1.9173872470855713, + "learning_rate": 4.660844841592202e-06, + "loss": 2.2435, + "step": 455 + }, + { + "epoch": 0.4997260273972603, + "grad_norm": 1.88880455493927, + "learning_rate": 4.658813972380179e-06, + "loss": 2.1711, + "step": 456 + }, + { + "epoch": 0.5008219178082192, + "grad_norm": 1.9628506898880005, + "learning_rate": 4.656783103168157e-06, + "loss": 2.2701, + "step": 457 + }, + { + "epoch": 0.5019178082191781, + "grad_norm": 2.0515286922454834, + "learning_rate": 4.654752233956134e-06, + "loss": 2.1656, + "step": 458 + }, + { + "epoch": 0.503013698630137, + "grad_norm": 1.978522539138794, + "learning_rate": 4.652721364744111e-06, + "loss": 2.246, + "step": 459 + }, + { + "epoch": 0.5041095890410959, + "grad_norm": 1.7993966341018677, + "learning_rate": 4.6506904955320885e-06, + "loss": 2.214, + "step": 460 + }, + { + "epoch": 0.5041095890410959, + "eval_loss": 2.3568837642669678, + "eval_runtime": 45.2086, + "eval_samples_per_second": 22.12, + "eval_steps_per_second": 0.929, + "step": 460 + }, + { + "epoch": 0.5052054794520547, + "grad_norm": 1.7809834480285645, + "learning_rate": 4.648659626320066e-06, + "loss": 2.2525, + "step": 461 + }, + { + "epoch": 0.5063013698630137, + "grad_norm": 1.9256166219711304, + "learning_rate": 4.646628757108043e-06, + "loss": 2.2611, + "step": 462 + }, + { + "epoch": 0.5073972602739726, + "grad_norm": 1.94192636013031, + "learning_rate": 4.6445978878960194e-06, + "loss": 2.2282, + "step": 463 + }, + { + "epoch": 0.5084931506849315, + "grad_norm": 1.8891818523406982, + "learning_rate": 4.6425670186839976e-06, + "loss": 2.2036, + "step": 464 + }, + { + "epoch": 0.5095890410958904, + "grad_norm": 2.352078676223755, + "learning_rate": 4.640536149471974e-06, + "loss": 2.2293, + "step": 465 + }, + { + "epoch": 0.5106849315068493, + "grad_norm": 1.8141659498214722, + "learning_rate": 4.638505280259952e-06, + "loss": 2.2551, + "step": 466 + }, + { + "epoch": 0.5117808219178083, + "grad_norm": 1.9492480754852295, + "learning_rate": 4.6364744110479285e-06, + "loss": 2.2641, + "step": 467 + }, + { + "epoch": 0.5128767123287671, + "grad_norm": 2.0210182666778564, + "learning_rate": 4.634443541835907e-06, + "loss": 2.2367, + "step": 468 + }, + { + "epoch": 0.513972602739726, + "grad_norm": 1.8936505317687988, + "learning_rate": 4.632412672623883e-06, + "loss": 2.265, + "step": 469 + }, + { + "epoch": 0.5150684931506849, + "grad_norm": 1.845888614654541, + "learning_rate": 4.630381803411861e-06, + "loss": 2.2169, + "step": 470 + }, + { + "epoch": 0.5150684931506849, + "eval_loss": 2.353489875793457, + "eval_runtime": 46.8739, + "eval_samples_per_second": 21.334, + "eval_steps_per_second": 0.896, + "step": 470 + }, + { + "epoch": 0.5161643835616438, + "grad_norm": 1.889675259590149, + "learning_rate": 4.628350934199838e-06, + "loss": 2.1383, + "step": 471 + }, + { + "epoch": 0.5172602739726028, + "grad_norm": 1.9799216985702515, + "learning_rate": 4.626320064987816e-06, + "loss": 2.2312, + "step": 472 + }, + { + "epoch": 0.5183561643835617, + "grad_norm": 2.0128607749938965, + "learning_rate": 4.624289195775792e-06, + "loss": 2.2529, + "step": 473 + }, + { + "epoch": 0.5194520547945205, + "grad_norm": 1.8791241645812988, + "learning_rate": 4.62225832656377e-06, + "loss": 2.226, + "step": 474 + }, + { + "epoch": 0.5205479452054794, + "grad_norm": 2.0497279167175293, + "learning_rate": 4.620227457351747e-06, + "loss": 2.2283, + "step": 475 + }, + { + "epoch": 0.5216438356164383, + "grad_norm": 1.8097424507141113, + "learning_rate": 4.618196588139724e-06, + "loss": 2.222, + "step": 476 + }, + { + "epoch": 0.5227397260273973, + "grad_norm": 1.8960500955581665, + "learning_rate": 4.616165718927701e-06, + "loss": 2.2342, + "step": 477 + }, + { + "epoch": 0.5238356164383562, + "grad_norm": 2.020730495452881, + "learning_rate": 4.6141348497156786e-06, + "loss": 2.2106, + "step": 478 + }, + { + "epoch": 0.5249315068493151, + "grad_norm": 1.8955297470092773, + "learning_rate": 4.612103980503656e-06, + "loss": 2.2755, + "step": 479 + }, + { + "epoch": 0.5260273972602739, + "grad_norm": 1.9767351150512695, + "learning_rate": 4.610073111291633e-06, + "loss": 2.2554, + "step": 480 + }, + { + "epoch": 0.5260273972602739, + "eval_loss": 2.347705841064453, + "eval_runtime": 46.9125, + "eval_samples_per_second": 21.316, + "eval_steps_per_second": 0.895, + "step": 480 + }, + { + "epoch": 0.5271232876712328, + "grad_norm": 1.936457633972168, + "learning_rate": 4.60804224207961e-06, + "loss": 2.2323, + "step": 481 + }, + { + "epoch": 0.5282191780821918, + "grad_norm": 1.9271862506866455, + "learning_rate": 4.606011372867588e-06, + "loss": 2.2151, + "step": 482 + }, + { + "epoch": 0.5293150684931507, + "grad_norm": 1.878153681755066, + "learning_rate": 4.603980503655565e-06, + "loss": 2.2322, + "step": 483 + }, + { + "epoch": 0.5304109589041096, + "grad_norm": 2.1009249687194824, + "learning_rate": 4.601949634443542e-06, + "loss": 2.2181, + "step": 484 + }, + { + "epoch": 0.5315068493150685, + "grad_norm": 1.9089181423187256, + "learning_rate": 4.5999187652315195e-06, + "loss": 2.1712, + "step": 485 + }, + { + "epoch": 0.5326027397260275, + "grad_norm": 2.207282781600952, + "learning_rate": 4.597887896019497e-06, + "loss": 2.1847, + "step": 486 + }, + { + "epoch": 0.5336986301369863, + "grad_norm": 2.001584053039551, + "learning_rate": 4.595857026807474e-06, + "loss": 2.214, + "step": 487 + }, + { + "epoch": 0.5347945205479452, + "grad_norm": 1.966700553894043, + "learning_rate": 4.593826157595451e-06, + "loss": 2.2237, + "step": 488 + }, + { + "epoch": 0.5358904109589041, + "grad_norm": 1.842469334602356, + "learning_rate": 4.591795288383429e-06, + "loss": 2.2089, + "step": 489 + }, + { + "epoch": 0.536986301369863, + "grad_norm": 2.0412986278533936, + "learning_rate": 4.589764419171406e-06, + "loss": 2.2734, + "step": 490 + }, + { + "epoch": 0.536986301369863, + "eval_loss": 2.3460772037506104, + "eval_runtime": 50.4048, + "eval_samples_per_second": 19.839, + "eval_steps_per_second": 0.833, + "step": 490 + }, + { + "epoch": 0.538082191780822, + "grad_norm": 1.9337031841278076, + "learning_rate": 4.587733549959383e-06, + "loss": 2.2314, + "step": 491 + }, + { + "epoch": 0.5391780821917809, + "grad_norm": 2.4448320865631104, + "learning_rate": 4.5857026807473604e-06, + "loss": 2.2098, + "step": 492 + }, + { + "epoch": 0.5402739726027397, + "grad_norm": 1.83162522315979, + "learning_rate": 4.583671811535338e-06, + "loss": 2.1902, + "step": 493 + }, + { + "epoch": 0.5413698630136986, + "grad_norm": 2.3614656925201416, + "learning_rate": 4.581640942323315e-06, + "loss": 2.1737, + "step": 494 + }, + { + "epoch": 0.5424657534246575, + "grad_norm": 2.001187801361084, + "learning_rate": 4.579610073111292e-06, + "loss": 2.2485, + "step": 495 + }, + { + "epoch": 0.5435616438356164, + "grad_norm": 2.3479933738708496, + "learning_rate": 4.5775792038992695e-06, + "loss": 2.2595, + "step": 496 + }, + { + "epoch": 0.5446575342465754, + "grad_norm": 2.135856866836548, + "learning_rate": 4.575548334687247e-06, + "loss": 2.2528, + "step": 497 + }, + { + "epoch": 0.5457534246575343, + "grad_norm": 1.983126163482666, + "learning_rate": 4.573517465475224e-06, + "loss": 2.2788, + "step": 498 + }, + { + "epoch": 0.5468493150684931, + "grad_norm": 2.18208909034729, + "learning_rate": 4.571486596263201e-06, + "loss": 2.1995, + "step": 499 + }, + { + "epoch": 0.547945205479452, + "grad_norm": 1.848747968673706, + "learning_rate": 4.569455727051178e-06, + "loss": 2.2403, + "step": 500 + }, + { + "epoch": 0.547945205479452, + "eval_loss": 2.344066619873047, + "eval_runtime": 51.1213, + "eval_samples_per_second": 19.561, + "eval_steps_per_second": 0.822, + "step": 500 + }, + { + "epoch": 0.549041095890411, + "grad_norm": 2.427860736846924, + "learning_rate": 4.567424857839156e-06, + "loss": 2.246, + "step": 501 + }, + { + "epoch": 0.5501369863013699, + "grad_norm": 1.913405418395996, + "learning_rate": 4.565393988627132e-06, + "loss": 2.2296, + "step": 502 + }, + { + "epoch": 0.5512328767123288, + "grad_norm": 2.0106029510498047, + "learning_rate": 4.5633631194151105e-06, + "loss": 2.2657, + "step": 503 + }, + { + "epoch": 0.5523287671232877, + "grad_norm": 2.2719595432281494, + "learning_rate": 4.561332250203087e-06, + "loss": 2.2341, + "step": 504 + }, + { + "epoch": 0.5534246575342465, + "grad_norm": 2.109786033630371, + "learning_rate": 4.559301380991065e-06, + "loss": 2.2181, + "step": 505 + }, + { + "epoch": 0.5545205479452054, + "grad_norm": 2.189779281616211, + "learning_rate": 4.5572705117790414e-06, + "loss": 2.1821, + "step": 506 + }, + { + "epoch": 0.5556164383561644, + "grad_norm": 2.0263590812683105, + "learning_rate": 4.5552396425670196e-06, + "loss": 2.2072, + "step": 507 + }, + { + "epoch": 0.5567123287671233, + "grad_norm": 2.215289354324341, + "learning_rate": 4.553208773354996e-06, + "loss": 2.2513, + "step": 508 + }, + { + "epoch": 0.5578082191780822, + "grad_norm": 2.3031351566314697, + "learning_rate": 4.551177904142974e-06, + "loss": 2.1988, + "step": 509 + }, + { + "epoch": 0.5589041095890411, + "grad_norm": 2.1830058097839355, + "learning_rate": 4.5491470349309505e-06, + "loss": 2.269, + "step": 510 + }, + { + "epoch": 0.5589041095890411, + "eval_loss": 2.342463731765747, + "eval_runtime": 43.1956, + "eval_samples_per_second": 23.151, + "eval_steps_per_second": 0.972, + "step": 510 + }, + { + "epoch": 0.56, + "grad_norm": 2.3210105895996094, + "learning_rate": 4.547116165718929e-06, + "loss": 2.2051, + "step": 511 + }, + { + "epoch": 0.5610958904109589, + "grad_norm": 1.9175361394882202, + "learning_rate": 4.545085296506905e-06, + "loss": 2.2304, + "step": 512 + }, + { + "epoch": 0.5621917808219178, + "grad_norm": 2.302947759628296, + "learning_rate": 4.543054427294882e-06, + "loss": 2.215, + "step": 513 + }, + { + "epoch": 0.5632876712328767, + "grad_norm": 1.8947396278381348, + "learning_rate": 4.54102355808286e-06, + "loss": 2.1926, + "step": 514 + }, + { + "epoch": 0.5643835616438356, + "grad_norm": 2.195268154144287, + "learning_rate": 4.538992688870837e-06, + "loss": 2.1937, + "step": 515 + }, + { + "epoch": 0.5654794520547946, + "grad_norm": 2.178874969482422, + "learning_rate": 4.536961819658814e-06, + "loss": 2.2248, + "step": 516 + }, + { + "epoch": 0.5665753424657535, + "grad_norm": 1.9856929779052734, + "learning_rate": 4.5349309504467915e-06, + "loss": 2.2077, + "step": 517 + }, + { + "epoch": 0.5676712328767123, + "grad_norm": 2.0686943531036377, + "learning_rate": 4.532900081234769e-06, + "loss": 2.2328, + "step": 518 + }, + { + "epoch": 0.5687671232876712, + "grad_norm": 1.9554919004440308, + "learning_rate": 4.530869212022746e-06, + "loss": 2.198, + "step": 519 + }, + { + "epoch": 0.5698630136986301, + "grad_norm": 1.9489670991897583, + "learning_rate": 4.528838342810723e-06, + "loss": 2.2287, + "step": 520 + }, + { + "epoch": 0.5698630136986301, + "eval_loss": 2.3415298461914062, + "eval_runtime": 48.7918, + "eval_samples_per_second": 20.495, + "eval_steps_per_second": 0.861, + "step": 520 + }, + { + "epoch": 0.570958904109589, + "grad_norm": 1.9868658781051636, + "learning_rate": 4.526807473598701e-06, + "loss": 2.2205, + "step": 521 + }, + { + "epoch": 0.572054794520548, + "grad_norm": 2.0965914726257324, + "learning_rate": 4.524776604386678e-06, + "loss": 2.2239, + "step": 522 + }, + { + "epoch": 0.5731506849315069, + "grad_norm": 2.041471481323242, + "learning_rate": 4.522745735174655e-06, + "loss": 2.2035, + "step": 523 + }, + { + "epoch": 0.5742465753424657, + "grad_norm": 1.885293960571289, + "learning_rate": 4.520714865962632e-06, + "loss": 2.2469, + "step": 524 + }, + { + "epoch": 0.5753424657534246, + "grad_norm": 1.9107296466827393, + "learning_rate": 4.51868399675061e-06, + "loss": 2.2386, + "step": 525 + }, + { + "epoch": 0.5764383561643835, + "grad_norm": 1.9945271015167236, + "learning_rate": 4.516653127538587e-06, + "loss": 2.235, + "step": 526 + }, + { + "epoch": 0.5775342465753425, + "grad_norm": 1.9787253141403198, + "learning_rate": 4.514622258326564e-06, + "loss": 2.2034, + "step": 527 + }, + { + "epoch": 0.5786301369863014, + "grad_norm": 2.0300636291503906, + "learning_rate": 4.5125913891145415e-06, + "loss": 2.1849, + "step": 528 + }, + { + "epoch": 0.5797260273972603, + "grad_norm": 2.0348970890045166, + "learning_rate": 4.510560519902519e-06, + "loss": 2.2174, + "step": 529 + }, + { + "epoch": 0.5808219178082191, + "grad_norm": 1.8589537143707275, + "learning_rate": 4.508529650690496e-06, + "loss": 2.2351, + "step": 530 + }, + { + "epoch": 0.5808219178082191, + "eval_loss": 2.3450186252593994, + "eval_runtime": 52.4548, + "eval_samples_per_second": 19.064, + "eval_steps_per_second": 0.801, + "step": 530 + }, + { + "epoch": 0.581917808219178, + "grad_norm": 2.0496020317077637, + "learning_rate": 4.506498781478473e-06, + "loss": 2.182, + "step": 531 + }, + { + "epoch": 0.583013698630137, + "grad_norm": 2.0649805068969727, + "learning_rate": 4.504467912266451e-06, + "loss": 2.1654, + "step": 532 + }, + { + "epoch": 0.5841095890410959, + "grad_norm": 1.9809486865997314, + "learning_rate": 4.502437043054428e-06, + "loss": 2.2182, + "step": 533 + }, + { + "epoch": 0.5852054794520548, + "grad_norm": 2.21915340423584, + "learning_rate": 4.500406173842405e-06, + "loss": 2.2697, + "step": 534 + }, + { + "epoch": 0.5863013698630137, + "grad_norm": 1.9822649955749512, + "learning_rate": 4.4983753046303824e-06, + "loss": 2.2297, + "step": 535 + }, + { + "epoch": 0.5873972602739727, + "grad_norm": 1.972575306892395, + "learning_rate": 4.49634443541836e-06, + "loss": 2.2164, + "step": 536 + }, + { + "epoch": 0.5884931506849315, + "grad_norm": 1.948667049407959, + "learning_rate": 4.494313566206336e-06, + "loss": 2.2087, + "step": 537 + }, + { + "epoch": 0.5895890410958904, + "grad_norm": 2.7486417293548584, + "learning_rate": 4.492282696994314e-06, + "loss": 2.2602, + "step": 538 + }, + { + "epoch": 0.5906849315068493, + "grad_norm": 2.3374552726745605, + "learning_rate": 4.490251827782291e-06, + "loss": 2.2054, + "step": 539 + }, + { + "epoch": 0.5917808219178082, + "grad_norm": 2.222116231918335, + "learning_rate": 4.488220958570269e-06, + "loss": 2.1868, + "step": 540 + }, + { + "epoch": 0.5917808219178082, + "eval_loss": 2.345618963241577, + "eval_runtime": 53.0302, + "eval_samples_per_second": 18.857, + "eval_steps_per_second": 0.792, + "step": 540 + }, + { + "epoch": 0.5928767123287672, + "grad_norm": 2.00752854347229, + "learning_rate": 4.486190089358245e-06, + "loss": 2.187, + "step": 541 + }, + { + "epoch": 0.5939726027397261, + "grad_norm": 2.2466132640838623, + "learning_rate": 4.484159220146223e-06, + "loss": 2.1611, + "step": 542 + }, + { + "epoch": 0.5950684931506849, + "grad_norm": 2.1279385089874268, + "learning_rate": 4.4821283509342e-06, + "loss": 2.248, + "step": 543 + }, + { + "epoch": 0.5961643835616438, + "grad_norm": 1.9140571355819702, + "learning_rate": 4.480097481722178e-06, + "loss": 2.2314, + "step": 544 + }, + { + "epoch": 0.5972602739726027, + "grad_norm": 2.120741605758667, + "learning_rate": 4.478066612510154e-06, + "loss": 2.212, + "step": 545 + }, + { + "epoch": 0.5983561643835617, + "grad_norm": 2.003713369369507, + "learning_rate": 4.4760357432981325e-06, + "loss": 2.1824, + "step": 546 + }, + { + "epoch": 0.5994520547945206, + "grad_norm": 1.9161221981048584, + "learning_rate": 4.474004874086109e-06, + "loss": 2.2015, + "step": 547 + }, + { + "epoch": 0.6005479452054795, + "grad_norm": 2.367110252380371, + "learning_rate": 4.471974004874087e-06, + "loss": 2.3122, + "step": 548 + }, + { + "epoch": 0.6016438356164383, + "grad_norm": 2.113380193710327, + "learning_rate": 4.4699431356620635e-06, + "loss": 2.1692, + "step": 549 + }, + { + "epoch": 0.6027397260273972, + "grad_norm": 1.974181890487671, + "learning_rate": 4.467912266450041e-06, + "loss": 2.2509, + "step": 550 + }, + { + "epoch": 0.6027397260273972, + "eval_loss": 2.342583656311035, + "eval_runtime": 50.7362, + "eval_samples_per_second": 19.71, + "eval_steps_per_second": 0.828, + "step": 550 + }, + { + "epoch": 0.6038356164383561, + "grad_norm": 1.889894962310791, + "learning_rate": 4.465881397238018e-06, + "loss": 2.2214, + "step": 551 + }, + { + "epoch": 0.6049315068493151, + "grad_norm": 1.8982343673706055, + "learning_rate": 4.463850528025995e-06, + "loss": 2.2135, + "step": 552 + }, + { + "epoch": 0.606027397260274, + "grad_norm": 1.9462087154388428, + "learning_rate": 4.4618196588139726e-06, + "loss": 2.2255, + "step": 553 + }, + { + "epoch": 0.6071232876712329, + "grad_norm": 1.8729931116104126, + "learning_rate": 4.45978878960195e-06, + "loss": 2.2358, + "step": 554 + }, + { + "epoch": 0.6082191780821918, + "grad_norm": 1.9230246543884277, + "learning_rate": 4.457757920389927e-06, + "loss": 2.222, + "step": 555 + }, + { + "epoch": 0.6093150684931506, + "grad_norm": 1.9495900869369507, + "learning_rate": 4.455727051177904e-06, + "loss": 2.1779, + "step": 556 + }, + { + "epoch": 0.6104109589041096, + "grad_norm": 1.983487606048584, + "learning_rate": 4.453696181965882e-06, + "loss": 2.2007, + "step": 557 + }, + { + "epoch": 0.6115068493150685, + "grad_norm": 2.1541049480438232, + "learning_rate": 4.451665312753859e-06, + "loss": 2.2513, + "step": 558 + }, + { + "epoch": 0.6126027397260274, + "grad_norm": 1.9133352041244507, + "learning_rate": 4.449634443541836e-06, + "loss": 2.2288, + "step": 559 + }, + { + "epoch": 0.6136986301369863, + "grad_norm": 2.3325741291046143, + "learning_rate": 4.4476035743298135e-06, + "loss": 2.2539, + "step": 560 + }, + { + "epoch": 0.6136986301369863, + "eval_loss": 2.3387348651885986, + "eval_runtime": 53.4301, + "eval_samples_per_second": 18.716, + "eval_steps_per_second": 0.786, + "step": 560 + }, + { + "epoch": 0.6147945205479453, + "grad_norm": 1.8099673986434937, + "learning_rate": 4.445572705117791e-06, + "loss": 2.1845, + "step": 561 + }, + { + "epoch": 0.6158904109589041, + "grad_norm": 2.201901912689209, + "learning_rate": 4.443541835905768e-06, + "loss": 2.195, + "step": 562 + }, + { + "epoch": 0.616986301369863, + "grad_norm": 1.8412714004516602, + "learning_rate": 4.441510966693745e-06, + "loss": 2.2551, + "step": 563 + }, + { + "epoch": 0.6180821917808219, + "grad_norm": 1.8761520385742188, + "learning_rate": 4.439480097481723e-06, + "loss": 2.1965, + "step": 564 + }, + { + "epoch": 0.6191780821917808, + "grad_norm": 1.9123467206954956, + "learning_rate": 4.4374492282697e-06, + "loss": 2.2502, + "step": 565 + }, + { + "epoch": 0.6202739726027398, + "grad_norm": 2.0631234645843506, + "learning_rate": 4.435418359057677e-06, + "loss": 2.1582, + "step": 566 + }, + { + "epoch": 0.6213698630136987, + "grad_norm": 2.3191616535186768, + "learning_rate": 4.433387489845654e-06, + "loss": 2.2271, + "step": 567 + }, + { + "epoch": 0.6224657534246575, + "grad_norm": 2.043057918548584, + "learning_rate": 4.431356620633632e-06, + "loss": 2.2292, + "step": 568 + }, + { + "epoch": 0.6235616438356164, + "grad_norm": 2.0716068744659424, + "learning_rate": 4.429325751421609e-06, + "loss": 2.196, + "step": 569 + }, + { + "epoch": 0.6246575342465753, + "grad_norm": 1.7669776678085327, + "learning_rate": 4.427294882209586e-06, + "loss": 2.1953, + "step": 570 + }, + { + "epoch": 0.6246575342465753, + "eval_loss": 2.336578607559204, + "eval_runtime": 50.0083, + "eval_samples_per_second": 19.997, + "eval_steps_per_second": 0.84, + "step": 570 + }, + { + "epoch": 0.6257534246575343, + "grad_norm": 1.8165771961212158, + "learning_rate": 4.4252640129975635e-06, + "loss": 2.2155, + "step": 571 + }, + { + "epoch": 0.6268493150684932, + "grad_norm": 1.8640249967575073, + "learning_rate": 4.423233143785541e-06, + "loss": 2.1648, + "step": 572 + }, + { + "epoch": 0.6279452054794521, + "grad_norm": 2.013869285583496, + "learning_rate": 4.421202274573518e-06, + "loss": 2.2131, + "step": 573 + }, + { + "epoch": 0.6290410958904109, + "grad_norm": 2.2650580406188965, + "learning_rate": 4.4191714053614945e-06, + "loss": 2.2287, + "step": 574 + }, + { + "epoch": 0.6301369863013698, + "grad_norm": 1.9232103824615479, + "learning_rate": 4.417140536149473e-06, + "loss": 2.1339, + "step": 575 + }, + { + "epoch": 0.6312328767123287, + "grad_norm": 1.910134196281433, + "learning_rate": 4.415109666937449e-06, + "loss": 2.1755, + "step": 576 + }, + { + "epoch": 0.6323287671232877, + "grad_norm": 2.2492005825042725, + "learning_rate": 4.413078797725427e-06, + "loss": 2.2362, + "step": 577 + }, + { + "epoch": 0.6334246575342466, + "grad_norm": 1.8584033250808716, + "learning_rate": 4.411047928513404e-06, + "loss": 2.1911, + "step": 578 + }, + { + "epoch": 0.6345205479452055, + "grad_norm": 2.156273126602173, + "learning_rate": 4.409017059301382e-06, + "loss": 2.2109, + "step": 579 + }, + { + "epoch": 0.6356164383561644, + "grad_norm": 1.7977535724639893, + "learning_rate": 4.406986190089358e-06, + "loss": 2.2195, + "step": 580 + }, + { + "epoch": 0.6356164383561644, + "eval_loss": 2.33555269241333, + "eval_runtime": 48.3919, + "eval_samples_per_second": 20.665, + "eval_steps_per_second": 0.868, + "step": 580 + }, + { + "epoch": 0.6367123287671232, + "grad_norm": 1.8262702226638794, + "learning_rate": 4.404955320877336e-06, + "loss": 2.2345, + "step": 581 + }, + { + "epoch": 0.6378082191780822, + "grad_norm": 1.9389383792877197, + "learning_rate": 4.402924451665313e-06, + "loss": 2.2347, + "step": 582 + }, + { + "epoch": 0.6389041095890411, + "grad_norm": 1.9548988342285156, + "learning_rate": 4.400893582453291e-06, + "loss": 2.1967, + "step": 583 + }, + { + "epoch": 0.64, + "grad_norm": 1.898092269897461, + "learning_rate": 4.398862713241267e-06, + "loss": 2.1795, + "step": 584 + }, + { + "epoch": 0.6410958904109589, + "grad_norm": 1.8183610439300537, + "learning_rate": 4.396831844029245e-06, + "loss": 2.2031, + "step": 585 + }, + { + "epoch": 0.6421917808219179, + "grad_norm": 2.1520016193389893, + "learning_rate": 4.394800974817222e-06, + "loss": 2.2051, + "step": 586 + }, + { + "epoch": 0.6432876712328767, + "grad_norm": 1.9930012226104736, + "learning_rate": 4.392770105605199e-06, + "loss": 2.2053, + "step": 587 + }, + { + "epoch": 0.6443835616438356, + "grad_norm": 2.311450958251953, + "learning_rate": 4.390739236393176e-06, + "loss": 2.2252, + "step": 588 + }, + { + "epoch": 0.6454794520547945, + "grad_norm": 2.0168697834014893, + "learning_rate": 4.388708367181154e-06, + "loss": 2.2286, + "step": 589 + }, + { + "epoch": 0.6465753424657534, + "grad_norm": 2.055964469909668, + "learning_rate": 4.386677497969131e-06, + "loss": 2.2278, + "step": 590 + }, + { + "epoch": 0.6465753424657534, + "eval_loss": 2.334367275238037, + "eval_runtime": 48.9414, + "eval_samples_per_second": 20.433, + "eval_steps_per_second": 0.858, + "step": 590 + }, + { + "epoch": 0.6476712328767124, + "grad_norm": 2.0161590576171875, + "learning_rate": 4.384646628757108e-06, + "loss": 2.2102, + "step": 591 + }, + { + "epoch": 0.6487671232876713, + "grad_norm": 2.0827667713165283, + "learning_rate": 4.3826157595450855e-06, + "loss": 2.2531, + "step": 592 + }, + { + "epoch": 0.6498630136986301, + "grad_norm": 2.077637195587158, + "learning_rate": 4.380584890333063e-06, + "loss": 2.2091, + "step": 593 + }, + { + "epoch": 0.650958904109589, + "grad_norm": 2.023892879486084, + "learning_rate": 4.37855402112104e-06, + "loss": 2.1684, + "step": 594 + }, + { + "epoch": 0.6520547945205479, + "grad_norm": 1.8509857654571533, + "learning_rate": 4.376523151909017e-06, + "loss": 2.1439, + "step": 595 + }, + { + "epoch": 0.6531506849315069, + "grad_norm": 2.1649608612060547, + "learning_rate": 4.3744922826969946e-06, + "loss": 2.2557, + "step": 596 + }, + { + "epoch": 0.6542465753424658, + "grad_norm": 2.1351773738861084, + "learning_rate": 4.372461413484972e-06, + "loss": 2.177, + "step": 597 + }, + { + "epoch": 0.6553424657534247, + "grad_norm": 2.1713974475860596, + "learning_rate": 4.370430544272949e-06, + "loss": 2.2223, + "step": 598 + }, + { + "epoch": 0.6564383561643835, + "grad_norm": 2.2905397415161133, + "learning_rate": 4.368399675060926e-06, + "loss": 2.2373, + "step": 599 + }, + { + "epoch": 0.6575342465753424, + "grad_norm": 2.007160186767578, + "learning_rate": 4.366368805848904e-06, + "loss": 2.2355, + "step": 600 + }, + { + "epoch": 0.6575342465753424, + "eval_loss": 2.3322324752807617, + "eval_runtime": 46.2769, + "eval_samples_per_second": 21.609, + "eval_steps_per_second": 0.908, + "step": 600 + }, + { + "epoch": 0.6586301369863014, + "grad_norm": 2.309783697128296, + "learning_rate": 4.364337936636881e-06, + "loss": 2.2052, + "step": 601 + }, + { + "epoch": 0.6597260273972603, + "grad_norm": 2.498034954071045, + "learning_rate": 4.362307067424858e-06, + "loss": 2.1567, + "step": 602 + }, + { + "epoch": 0.6608219178082192, + "grad_norm": 1.8302545547485352, + "learning_rate": 4.3602761982128355e-06, + "loss": 2.2117, + "step": 603 + }, + { + "epoch": 0.6619178082191781, + "grad_norm": 2.226947546005249, + "learning_rate": 4.358245329000813e-06, + "loss": 2.1881, + "step": 604 + }, + { + "epoch": 0.663013698630137, + "grad_norm": 1.8484126329421997, + "learning_rate": 4.35621445978879e-06, + "loss": 2.1955, + "step": 605 + }, + { + "epoch": 0.6641095890410958, + "grad_norm": 1.9844266176223755, + "learning_rate": 4.354183590576767e-06, + "loss": 2.2063, + "step": 606 + }, + { + "epoch": 0.6652054794520548, + "grad_norm": 1.9292365312576294, + "learning_rate": 4.352152721364745e-06, + "loss": 2.205, + "step": 607 + }, + { + "epoch": 0.6663013698630137, + "grad_norm": 1.980895757675171, + "learning_rate": 4.350121852152722e-06, + "loss": 2.1727, + "step": 608 + }, + { + "epoch": 0.6673972602739726, + "grad_norm": 1.8331713676452637, + "learning_rate": 4.348090982940699e-06, + "loss": 2.226, + "step": 609 + }, + { + "epoch": 0.6684931506849315, + "grad_norm": 1.9666537046432495, + "learning_rate": 4.3460601137286764e-06, + "loss": 2.2198, + "step": 610 + }, + { + "epoch": 0.6684931506849315, + "eval_loss": 2.3297338485717773, + "eval_runtime": 42.6274, + "eval_samples_per_second": 23.459, + "eval_steps_per_second": 0.985, + "step": 610 + }, + { + "epoch": 0.6695890410958905, + "grad_norm": 1.9741110801696777, + "learning_rate": 4.344029244516654e-06, + "loss": 2.2132, + "step": 611 + }, + { + "epoch": 0.6706849315068493, + "grad_norm": 2.0747334957122803, + "learning_rate": 4.341998375304631e-06, + "loss": 2.2081, + "step": 612 + }, + { + "epoch": 0.6717808219178082, + "grad_norm": 2.031503915786743, + "learning_rate": 4.339967506092607e-06, + "loss": 2.2642, + "step": 613 + }, + { + "epoch": 0.6728767123287671, + "grad_norm": 1.8513256311416626, + "learning_rate": 4.3379366368805855e-06, + "loss": 2.2141, + "step": 614 + }, + { + "epoch": 0.673972602739726, + "grad_norm": 1.9098315238952637, + "learning_rate": 4.335905767668562e-06, + "loss": 2.2064, + "step": 615 + }, + { + "epoch": 0.675068493150685, + "grad_norm": 1.9704065322875977, + "learning_rate": 4.33387489845654e-06, + "loss": 2.1997, + "step": 616 + }, + { + "epoch": 0.6761643835616439, + "grad_norm": 2.1520018577575684, + "learning_rate": 4.3318440292445165e-06, + "loss": 2.2278, + "step": 617 + }, + { + "epoch": 0.6772602739726027, + "grad_norm": 1.898164987564087, + "learning_rate": 4.329813160032495e-06, + "loss": 2.2147, + "step": 618 + }, + { + "epoch": 0.6783561643835616, + "grad_norm": 1.8955962657928467, + "learning_rate": 4.327782290820471e-06, + "loss": 2.1825, + "step": 619 + }, + { + "epoch": 0.6794520547945205, + "grad_norm": 1.7579313516616821, + "learning_rate": 4.325751421608449e-06, + "loss": 2.1619, + "step": 620 + }, + { + "epoch": 0.6794520547945205, + "eval_loss": 2.329692840576172, + "eval_runtime": 46.3187, + "eval_samples_per_second": 21.59, + "eval_steps_per_second": 0.907, + "step": 620 + }, + { + "epoch": 0.6805479452054795, + "grad_norm": 1.8716063499450684, + "learning_rate": 4.323720552396426e-06, + "loss": 2.2289, + "step": 621 + }, + { + "epoch": 0.6816438356164384, + "grad_norm": 1.8798290491104126, + "learning_rate": 4.321689683184404e-06, + "loss": 2.1944, + "step": 622 + }, + { + "epoch": 0.6827397260273973, + "grad_norm": 1.8438303470611572, + "learning_rate": 4.31965881397238e-06, + "loss": 2.1879, + "step": 623 + }, + { + "epoch": 0.6838356164383562, + "grad_norm": 1.7844786643981934, + "learning_rate": 4.3176279447603574e-06, + "loss": 2.1936, + "step": 624 + }, + { + "epoch": 0.684931506849315, + "grad_norm": 1.955529808998108, + "learning_rate": 4.315597075548335e-06, + "loss": 2.2258, + "step": 625 + }, + { + "epoch": 0.686027397260274, + "grad_norm": 2.3826634883880615, + "learning_rate": 4.313566206336312e-06, + "loss": 2.2135, + "step": 626 + }, + { + "epoch": 0.6871232876712329, + "grad_norm": 1.8453123569488525, + "learning_rate": 4.311535337124289e-06, + "loss": 2.2362, + "step": 627 + }, + { + "epoch": 0.6882191780821918, + "grad_norm": 2.742682933807373, + "learning_rate": 4.3095044679122665e-06, + "loss": 2.194, + "step": 628 + }, + { + "epoch": 0.6893150684931507, + "grad_norm": 2.0299901962280273, + "learning_rate": 4.307473598700244e-06, + "loss": 2.2029, + "step": 629 + }, + { + "epoch": 0.6904109589041096, + "grad_norm": 2.29683780670166, + "learning_rate": 4.305442729488221e-06, + "loss": 2.2141, + "step": 630 + }, + { + "epoch": 0.6904109589041096, + "eval_loss": 2.329368829727173, + "eval_runtime": 45.1947, + "eval_samples_per_second": 22.126, + "eval_steps_per_second": 0.929, + "step": 630 + }, + { + "epoch": 0.6915068493150685, + "grad_norm": 2.208280086517334, + "learning_rate": 4.303411860276198e-06, + "loss": 2.1637, + "step": 631 + }, + { + "epoch": 0.6926027397260274, + "grad_norm": 1.943514108657837, + "learning_rate": 4.301380991064176e-06, + "loss": 2.234, + "step": 632 + }, + { + "epoch": 0.6936986301369863, + "grad_norm": 2.486440420150757, + "learning_rate": 4.299350121852153e-06, + "loss": 2.2239, + "step": 633 + }, + { + "epoch": 0.6947945205479452, + "grad_norm": 1.7243690490722656, + "learning_rate": 4.29731925264013e-06, + "loss": 2.2428, + "step": 634 + }, + { + "epoch": 0.6958904109589041, + "grad_norm": 2.31227445602417, + "learning_rate": 4.2952883834281075e-06, + "loss": 2.2403, + "step": 635 + }, + { + "epoch": 0.6969863013698631, + "grad_norm": 1.7361263036727905, + "learning_rate": 4.293257514216085e-06, + "loss": 2.1975, + "step": 636 + }, + { + "epoch": 0.6980821917808219, + "grad_norm": 2.010831356048584, + "learning_rate": 4.291226645004062e-06, + "loss": 2.1633, + "step": 637 + }, + { + "epoch": 0.6991780821917808, + "grad_norm": 2.2796833515167236, + "learning_rate": 4.289195775792039e-06, + "loss": 2.2307, + "step": 638 + }, + { + "epoch": 0.7002739726027397, + "grad_norm": 2.1069483757019043, + "learning_rate": 4.287164906580017e-06, + "loss": 2.2389, + "step": 639 + }, + { + "epoch": 0.7013698630136986, + "grad_norm": 2.19460129737854, + "learning_rate": 4.285134037367994e-06, + "loss": 2.1992, + "step": 640 + }, + { + "epoch": 0.7013698630136986, + "eval_loss": 2.3223369121551514, + "eval_runtime": 43.1093, + "eval_samples_per_second": 23.197, + "eval_steps_per_second": 0.974, + "step": 640 + }, + { + "epoch": 0.7024657534246576, + "grad_norm": 1.875261902809143, + "learning_rate": 4.283103168155971e-06, + "loss": 2.1767, + "step": 641 + }, + { + "epoch": 0.7035616438356165, + "grad_norm": 1.906762957572937, + "learning_rate": 4.281072298943948e-06, + "loss": 2.1734, + "step": 642 + }, + { + "epoch": 0.7046575342465753, + "grad_norm": 2.1114180088043213, + "learning_rate": 4.279041429731926e-06, + "loss": 2.1734, + "step": 643 + }, + { + "epoch": 0.7057534246575342, + "grad_norm": 1.844373106956482, + "learning_rate": 4.277010560519903e-06, + "loss": 2.1967, + "step": 644 + }, + { + "epoch": 0.7068493150684931, + "grad_norm": 2.3463754653930664, + "learning_rate": 4.27497969130788e-06, + "loss": 2.1735, + "step": 645 + }, + { + "epoch": 0.707945205479452, + "grad_norm": 2.3530960083007812, + "learning_rate": 4.2729488220958575e-06, + "loss": 2.189, + "step": 646 + }, + { + "epoch": 0.709041095890411, + "grad_norm": 2.110963821411133, + "learning_rate": 4.270917952883835e-06, + "loss": 2.218, + "step": 647 + }, + { + "epoch": 0.7101369863013699, + "grad_norm": 2.625005006790161, + "learning_rate": 4.268887083671812e-06, + "loss": 2.2235, + "step": 648 + }, + { + "epoch": 0.7112328767123288, + "grad_norm": 1.8411654233932495, + "learning_rate": 4.266856214459789e-06, + "loss": 2.1757, + "step": 649 + }, + { + "epoch": 0.7123287671232876, + "grad_norm": 2.0817489624023438, + "learning_rate": 4.264825345247766e-06, + "loss": 2.2698, + "step": 650 + }, + { + "epoch": 0.7123287671232876, + "eval_loss": 2.322544574737549, + "eval_runtime": 43.0771, + "eval_samples_per_second": 23.214, + "eval_steps_per_second": 0.975, + "step": 650 + }, + { + "epoch": 0.7134246575342466, + "grad_norm": 1.8953847885131836, + "learning_rate": 4.262794476035744e-06, + "loss": 2.1745, + "step": 651 + }, + { + "epoch": 0.7145205479452055, + "grad_norm": 1.8463294506072998, + "learning_rate": 4.26076360682372e-06, + "loss": 2.2042, + "step": 652 + }, + { + "epoch": 0.7156164383561644, + "grad_norm": 1.8776047229766846, + "learning_rate": 4.2587327376116984e-06, + "loss": 2.1512, + "step": 653 + }, + { + "epoch": 0.7167123287671233, + "grad_norm": 1.8949921131134033, + "learning_rate": 4.256701868399675e-06, + "loss": 2.1895, + "step": 654 + }, + { + "epoch": 0.7178082191780822, + "grad_norm": 1.8172632455825806, + "learning_rate": 4.254670999187653e-06, + "loss": 2.2, + "step": 655 + }, + { + "epoch": 0.718904109589041, + "grad_norm": 2.105499505996704, + "learning_rate": 4.252640129975629e-06, + "loss": 2.202, + "step": 656 + }, + { + "epoch": 0.72, + "grad_norm": 1.9515924453735352, + "learning_rate": 4.2506092607636075e-06, + "loss": 2.2116, + "step": 657 + }, + { + "epoch": 0.7210958904109589, + "grad_norm": 1.761770248413086, + "learning_rate": 4.248578391551584e-06, + "loss": 2.1921, + "step": 658 + }, + { + "epoch": 0.7221917808219178, + "grad_norm": 1.9848097562789917, + "learning_rate": 4.246547522339562e-06, + "loss": 2.1553, + "step": 659 + }, + { + "epoch": 0.7232876712328767, + "grad_norm": 2.041987180709839, + "learning_rate": 4.2445166531275385e-06, + "loss": 2.1975, + "step": 660 + }, + { + "epoch": 0.7232876712328767, + "eval_loss": 2.325580358505249, + "eval_runtime": 43.6207, + "eval_samples_per_second": 22.925, + "eval_steps_per_second": 0.963, + "step": 660 + }, + { + "epoch": 0.7243835616438357, + "grad_norm": 1.9755775928497314, + "learning_rate": 4.242485783915517e-06, + "loss": 2.1955, + "step": 661 + }, + { + "epoch": 0.7254794520547945, + "grad_norm": 1.9621502161026, + "learning_rate": 4.240454914703493e-06, + "loss": 2.1554, + "step": 662 + }, + { + "epoch": 0.7265753424657534, + "grad_norm": 1.848013162612915, + "learning_rate": 4.23842404549147e-06, + "loss": 2.16, + "step": 663 + }, + { + "epoch": 0.7276712328767123, + "grad_norm": 1.7370128631591797, + "learning_rate": 4.236393176279448e-06, + "loss": 2.1696, + "step": 664 + }, + { + "epoch": 0.7287671232876712, + "grad_norm": 2.0708158016204834, + "learning_rate": 4.234362307067425e-06, + "loss": 2.1592, + "step": 665 + }, + { + "epoch": 0.7298630136986302, + "grad_norm": 1.9864391088485718, + "learning_rate": 4.232331437855402e-06, + "loss": 2.2121, + "step": 666 + }, + { + "epoch": 0.7309589041095891, + "grad_norm": 1.9988250732421875, + "learning_rate": 4.2303005686433795e-06, + "loss": 2.2306, + "step": 667 + }, + { + "epoch": 0.7320547945205479, + "grad_norm": 1.89108407497406, + "learning_rate": 4.228269699431357e-06, + "loss": 2.222, + "step": 668 + }, + { + "epoch": 0.7331506849315068, + "grad_norm": 1.7941503524780273, + "learning_rate": 4.226238830219334e-06, + "loss": 2.1975, + "step": 669 + }, + { + "epoch": 0.7342465753424657, + "grad_norm": 1.8058613538742065, + "learning_rate": 4.224207961007311e-06, + "loss": 2.2037, + "step": 670 + }, + { + "epoch": 0.7342465753424657, + "eval_loss": 2.3266475200653076, + "eval_runtime": 47.6623, + "eval_samples_per_second": 20.981, + "eval_steps_per_second": 0.881, + "step": 670 + }, + { + "epoch": 0.7353424657534247, + "grad_norm": 1.764448642730713, + "learning_rate": 4.2221770917952886e-06, + "loss": 2.1904, + "step": 671 + }, + { + "epoch": 0.7364383561643836, + "grad_norm": 1.8252112865447998, + "learning_rate": 4.220146222583266e-06, + "loss": 2.1959, + "step": 672 + }, + { + "epoch": 0.7375342465753425, + "grad_norm": 2.02898907661438, + "learning_rate": 4.218115353371243e-06, + "loss": 2.1937, + "step": 673 + }, + { + "epoch": 0.7386301369863014, + "grad_norm": 1.8766248226165771, + "learning_rate": 4.21608448415922e-06, + "loss": 2.195, + "step": 674 + }, + { + "epoch": 0.7397260273972602, + "grad_norm": 2.2350332736968994, + "learning_rate": 4.214053614947198e-06, + "loss": 2.1798, + "step": 675 + }, + { + "epoch": 0.7408219178082192, + "grad_norm": 1.820399522781372, + "learning_rate": 4.212022745735175e-06, + "loss": 2.2371, + "step": 676 + }, + { + "epoch": 0.7419178082191781, + "grad_norm": 2.3411312103271484, + "learning_rate": 4.209991876523152e-06, + "loss": 2.1603, + "step": 677 + }, + { + "epoch": 0.743013698630137, + "grad_norm": 1.8562874794006348, + "learning_rate": 4.2079610073111295e-06, + "loss": 2.2082, + "step": 678 + }, + { + "epoch": 0.7441095890410959, + "grad_norm": 2.274214506149292, + "learning_rate": 4.205930138099107e-06, + "loss": 2.1716, + "step": 679 + }, + { + "epoch": 0.7452054794520548, + "grad_norm": 1.9173530340194702, + "learning_rate": 4.203899268887084e-06, + "loss": 2.2461, + "step": 680 + }, + { + "epoch": 0.7452054794520548, + "eval_loss": 2.3253841400146484, + "eval_runtime": 46.5732, + "eval_samples_per_second": 21.472, + "eval_steps_per_second": 0.902, + "step": 680 + }, + { + "epoch": 0.7463013698630137, + "grad_norm": 1.9732033014297485, + "learning_rate": 4.201868399675061e-06, + "loss": 2.2016, + "step": 681 + }, + { + "epoch": 0.7473972602739726, + "grad_norm": 1.8850247859954834, + "learning_rate": 4.199837530463039e-06, + "loss": 2.223, + "step": 682 + }, + { + "epoch": 0.7484931506849315, + "grad_norm": 1.8936313390731812, + "learning_rate": 4.197806661251016e-06, + "loss": 2.1381, + "step": 683 + }, + { + "epoch": 0.7495890410958904, + "grad_norm": 2.11995530128479, + "learning_rate": 4.195775792038993e-06, + "loss": 2.1962, + "step": 684 + }, + { + "epoch": 0.7506849315068493, + "grad_norm": 1.8207820653915405, + "learning_rate": 4.19374492282697e-06, + "loss": 2.1835, + "step": 685 + }, + { + "epoch": 0.7517808219178083, + "grad_norm": 2.06030535697937, + "learning_rate": 4.191714053614948e-06, + "loss": 2.1347, + "step": 686 + }, + { + "epoch": 0.7528767123287671, + "grad_norm": 2.048990488052368, + "learning_rate": 4.189683184402924e-06, + "loss": 2.2189, + "step": 687 + }, + { + "epoch": 0.753972602739726, + "grad_norm": 1.976369023323059, + "learning_rate": 4.187652315190902e-06, + "loss": 2.1805, + "step": 688 + }, + { + "epoch": 0.7550684931506849, + "grad_norm": 2.1160926818847656, + "learning_rate": 4.185621445978879e-06, + "loss": 2.2322, + "step": 689 + }, + { + "epoch": 0.7561643835616438, + "grad_norm": 1.9444516897201538, + "learning_rate": 4.183590576766857e-06, + "loss": 2.1696, + "step": 690 + }, + { + "epoch": 0.7561643835616438, + "eval_loss": 2.3242011070251465, + "eval_runtime": 48.7528, + "eval_samples_per_second": 20.512, + "eval_steps_per_second": 0.861, + "step": 690 + }, + { + "epoch": 0.7572602739726028, + "grad_norm": 1.9567183256149292, + "learning_rate": 4.181559707554833e-06, + "loss": 2.1891, + "step": 691 + }, + { + "epoch": 0.7583561643835617, + "grad_norm": 1.9185020923614502, + "learning_rate": 4.179528838342811e-06, + "loss": 2.2318, + "step": 692 + }, + { + "epoch": 0.7594520547945205, + "grad_norm": 1.833487629890442, + "learning_rate": 4.177497969130788e-06, + "loss": 2.1705, + "step": 693 + }, + { + "epoch": 0.7605479452054794, + "grad_norm": 2.2949867248535156, + "learning_rate": 4.175467099918766e-06, + "loss": 2.1712, + "step": 694 + }, + { + "epoch": 0.7616438356164383, + "grad_norm": 1.8663084506988525, + "learning_rate": 4.173436230706742e-06, + "loss": 2.2262, + "step": 695 + }, + { + "epoch": 0.7627397260273973, + "grad_norm": 2.4633052349090576, + "learning_rate": 4.1714053614947205e-06, + "loss": 2.185, + "step": 696 + }, + { + "epoch": 0.7638356164383562, + "grad_norm": 2.036524772644043, + "learning_rate": 4.169374492282697e-06, + "loss": 2.1697, + "step": 697 + }, + { + "epoch": 0.7649315068493151, + "grad_norm": 2.013810634613037, + "learning_rate": 4.167343623070675e-06, + "loss": 2.1826, + "step": 698 + }, + { + "epoch": 0.766027397260274, + "grad_norm": 2.214109182357788, + "learning_rate": 4.165312753858652e-06, + "loss": 2.2162, + "step": 699 + }, + { + "epoch": 0.7671232876712328, + "grad_norm": 1.8463964462280273, + "learning_rate": 4.163281884646629e-06, + "loss": 2.1865, + "step": 700 + }, + { + "epoch": 0.7671232876712328, + "eval_loss": 2.323822498321533, + "eval_runtime": 48.3307, + "eval_samples_per_second": 20.691, + "eval_steps_per_second": 0.869, + "step": 700 + }, + { + "epoch": 0.7682191780821918, + "grad_norm": 2.321021556854248, + "learning_rate": 4.161251015434607e-06, + "loss": 2.1758, + "step": 701 + }, + { + "epoch": 0.7693150684931507, + "grad_norm": 1.8213255405426025, + "learning_rate": 4.159220146222583e-06, + "loss": 2.2002, + "step": 702 + }, + { + "epoch": 0.7704109589041096, + "grad_norm": 1.8925111293792725, + "learning_rate": 4.157189277010561e-06, + "loss": 2.1575, + "step": 703 + }, + { + "epoch": 0.7715068493150685, + "grad_norm": 1.8050175905227661, + "learning_rate": 4.155158407798538e-06, + "loss": 2.179, + "step": 704 + }, + { + "epoch": 0.7726027397260274, + "grad_norm": 1.873228907585144, + "learning_rate": 4.153127538586516e-06, + "loss": 2.2232, + "step": 705 + }, + { + "epoch": 0.7736986301369863, + "grad_norm": 1.954910397529602, + "learning_rate": 4.151096669374492e-06, + "loss": 2.1992, + "step": 706 + }, + { + "epoch": 0.7747945205479452, + "grad_norm": 1.8913410902023315, + "learning_rate": 4.1490658001624705e-06, + "loss": 2.2086, + "step": 707 + }, + { + "epoch": 0.7758904109589041, + "grad_norm": 1.8883898258209229, + "learning_rate": 4.147034930950447e-06, + "loss": 2.1801, + "step": 708 + }, + { + "epoch": 0.776986301369863, + "grad_norm": 2.054351806640625, + "learning_rate": 4.145004061738425e-06, + "loss": 2.1908, + "step": 709 + }, + { + "epoch": 0.7780821917808219, + "grad_norm": 2.029036045074463, + "learning_rate": 4.1429731925264015e-06, + "loss": 2.1891, + "step": 710 + }, + { + "epoch": 0.7780821917808219, + "eval_loss": 2.3196680545806885, + "eval_runtime": 42.8326, + "eval_samples_per_second": 23.347, + "eval_steps_per_second": 0.981, + "step": 710 + }, + { + "epoch": 0.7791780821917809, + "grad_norm": 2.27848219871521, + "learning_rate": 4.140942323314379e-06, + "loss": 2.1165, + "step": 711 + }, + { + "epoch": 0.7802739726027397, + "grad_norm": 2.134625196456909, + "learning_rate": 4.138911454102356e-06, + "loss": 2.1508, + "step": 712 + }, + { + "epoch": 0.7813698630136986, + "grad_norm": 2.103895902633667, + "learning_rate": 4.136880584890333e-06, + "loss": 2.1746, + "step": 713 + }, + { + "epoch": 0.7824657534246575, + "grad_norm": 1.713189721107483, + "learning_rate": 4.1348497156783106e-06, + "loss": 2.2023, + "step": 714 + }, + { + "epoch": 0.7835616438356164, + "grad_norm": 2.0390899181365967, + "learning_rate": 4.132818846466288e-06, + "loss": 2.2039, + "step": 715 + }, + { + "epoch": 0.7846575342465754, + "grad_norm": 1.9643651247024536, + "learning_rate": 4.130787977254265e-06, + "loss": 2.199, + "step": 716 + }, + { + "epoch": 0.7857534246575343, + "grad_norm": 1.9173564910888672, + "learning_rate": 4.128757108042242e-06, + "loss": 2.2166, + "step": 717 + }, + { + "epoch": 0.7868493150684932, + "grad_norm": 1.7923007011413574, + "learning_rate": 4.12672623883022e-06, + "loss": 2.2144, + "step": 718 + }, + { + "epoch": 0.787945205479452, + "grad_norm": 2.2235090732574463, + "learning_rate": 4.124695369618197e-06, + "loss": 2.2586, + "step": 719 + }, + { + "epoch": 0.7890410958904109, + "grad_norm": 1.780316710472107, + "learning_rate": 4.122664500406174e-06, + "loss": 2.1862, + "step": 720 + }, + { + "epoch": 0.7890410958904109, + "eval_loss": 2.318016767501831, + "eval_runtime": 50.5082, + "eval_samples_per_second": 19.799, + "eval_steps_per_second": 0.832, + "step": 720 + }, + { + "epoch": 0.7901369863013699, + "grad_norm": 2.0788493156433105, + "learning_rate": 4.1206336311941515e-06, + "loss": 2.1845, + "step": 721 + }, + { + "epoch": 0.7912328767123288, + "grad_norm": 1.9053714275360107, + "learning_rate": 4.118602761982129e-06, + "loss": 2.2005, + "step": 722 + }, + { + "epoch": 0.7923287671232877, + "grad_norm": 1.9328513145446777, + "learning_rate": 4.116571892770106e-06, + "loss": 2.1475, + "step": 723 + }, + { + "epoch": 0.7934246575342466, + "grad_norm": 2.059760570526123, + "learning_rate": 4.114541023558083e-06, + "loss": 2.1386, + "step": 724 + }, + { + "epoch": 0.7945205479452054, + "grad_norm": 2.0093464851379395, + "learning_rate": 4.112510154346061e-06, + "loss": 2.2242, + "step": 725 + }, + { + "epoch": 0.7956164383561644, + "grad_norm": 1.8513909578323364, + "learning_rate": 4.110479285134038e-06, + "loss": 2.178, + "step": 726 + }, + { + "epoch": 0.7967123287671233, + "grad_norm": 1.9481958150863647, + "learning_rate": 4.108448415922015e-06, + "loss": 2.1961, + "step": 727 + }, + { + "epoch": 0.7978082191780822, + "grad_norm": 1.9229767322540283, + "learning_rate": 4.1064175467099924e-06, + "loss": 2.2116, + "step": 728 + }, + { + "epoch": 0.7989041095890411, + "grad_norm": 1.9819737672805786, + "learning_rate": 4.10438667749797e-06, + "loss": 2.2097, + "step": 729 + }, + { + "epoch": 0.8, + "grad_norm": 1.9382416009902954, + "learning_rate": 4.102355808285947e-06, + "loss": 2.1099, + "step": 730 + }, + { + "epoch": 0.8, + "eval_loss": 2.31795334815979, + "eval_runtime": 51.8961, + "eval_samples_per_second": 19.269, + "eval_steps_per_second": 0.809, + "step": 730 + }, + { + "epoch": 0.8010958904109589, + "grad_norm": 2.006890058517456, + "learning_rate": 4.100324939073924e-06, + "loss": 2.2606, + "step": 731 + }, + { + "epoch": 0.8021917808219178, + "grad_norm": 2.164473295211792, + "learning_rate": 4.0982940698619015e-06, + "loss": 2.1869, + "step": 732 + }, + { + "epoch": 0.8032876712328767, + "grad_norm": 2.04368257522583, + "learning_rate": 4.096263200649879e-06, + "loss": 2.1846, + "step": 733 + }, + { + "epoch": 0.8043835616438356, + "grad_norm": 1.8363912105560303, + "learning_rate": 4.094232331437856e-06, + "loss": 2.1938, + "step": 734 + }, + { + "epoch": 0.8054794520547945, + "grad_norm": 1.9260189533233643, + "learning_rate": 4.092201462225833e-06, + "loss": 2.1682, + "step": 735 + }, + { + "epoch": 0.8065753424657535, + "grad_norm": 2.0688912868499756, + "learning_rate": 4.090170593013811e-06, + "loss": 2.1772, + "step": 736 + }, + { + "epoch": 0.8076712328767123, + "grad_norm": 1.859474539756775, + "learning_rate": 4.088139723801787e-06, + "loss": 2.2013, + "step": 737 + }, + { + "epoch": 0.8087671232876712, + "grad_norm": 2.0915753841400146, + "learning_rate": 4.086108854589765e-06, + "loss": 2.1519, + "step": 738 + }, + { + "epoch": 0.8098630136986301, + "grad_norm": 2.2019519805908203, + "learning_rate": 4.084077985377742e-06, + "loss": 2.1608, + "step": 739 + }, + { + "epoch": 0.810958904109589, + "grad_norm": 1.8727316856384277, + "learning_rate": 4.08204711616572e-06, + "loss": 2.1823, + "step": 740 + }, + { + "epoch": 0.810958904109589, + "eval_loss": 2.318638324737549, + "eval_runtime": 52.3669, + "eval_samples_per_second": 19.096, + "eval_steps_per_second": 0.802, + "step": 740 + }, + { + "epoch": 0.812054794520548, + "grad_norm": 2.059250831604004, + "learning_rate": 4.080016246953696e-06, + "loss": 2.2285, + "step": 741 + }, + { + "epoch": 0.8131506849315069, + "grad_norm": 1.9466749429702759, + "learning_rate": 4.077985377741674e-06, + "loss": 2.1496, + "step": 742 + }, + { + "epoch": 0.8142465753424658, + "grad_norm": 1.8967524766921997, + "learning_rate": 4.075954508529651e-06, + "loss": 2.2066, + "step": 743 + }, + { + "epoch": 0.8153424657534246, + "grad_norm": 1.976780891418457, + "learning_rate": 4.073923639317629e-06, + "loss": 2.2194, + "step": 744 + }, + { + "epoch": 0.8164383561643835, + "grad_norm": 2.077683687210083, + "learning_rate": 4.071892770105605e-06, + "loss": 2.2209, + "step": 745 + }, + { + "epoch": 0.8175342465753425, + "grad_norm": 1.9075323343276978, + "learning_rate": 4.069861900893583e-06, + "loss": 2.1727, + "step": 746 + }, + { + "epoch": 0.8186301369863014, + "grad_norm": 1.6940832138061523, + "learning_rate": 4.06783103168156e-06, + "loss": 2.1437, + "step": 747 + }, + { + "epoch": 0.8197260273972603, + "grad_norm": 1.9035768508911133, + "learning_rate": 4.065800162469538e-06, + "loss": 2.165, + "step": 748 + }, + { + "epoch": 0.8208219178082192, + "grad_norm": 1.8775745630264282, + "learning_rate": 4.063769293257514e-06, + "loss": 2.1757, + "step": 749 + }, + { + "epoch": 0.821917808219178, + "grad_norm": 1.9622418880462646, + "learning_rate": 4.061738424045492e-06, + "loss": 2.197, + "step": 750 + }, + { + "epoch": 0.821917808219178, + "eval_loss": 2.317519426345825, + "eval_runtime": 50.2699, + "eval_samples_per_second": 19.893, + "eval_steps_per_second": 0.835, + "step": 750 + }, + { + "epoch": 0.823013698630137, + "grad_norm": 1.9646214246749878, + "learning_rate": 4.059707554833469e-06, + "loss": 2.1783, + "step": 751 + }, + { + "epoch": 0.8241095890410959, + "grad_norm": 1.890342354774475, + "learning_rate": 4.057676685621446e-06, + "loss": 2.1925, + "step": 752 + }, + { + "epoch": 0.8252054794520548, + "grad_norm": 1.8977251052856445, + "learning_rate": 4.0556458164094235e-06, + "loss": 2.219, + "step": 753 + }, + { + "epoch": 0.8263013698630137, + "grad_norm": 1.8251665830612183, + "learning_rate": 4.053614947197401e-06, + "loss": 2.189, + "step": 754 + }, + { + "epoch": 0.8273972602739726, + "grad_norm": 1.7953882217407227, + "learning_rate": 4.051584077985378e-06, + "loss": 2.1237, + "step": 755 + }, + { + "epoch": 0.8284931506849315, + "grad_norm": 1.8247836828231812, + "learning_rate": 4.049553208773355e-06, + "loss": 2.1962, + "step": 756 + }, + { + "epoch": 0.8295890410958904, + "grad_norm": 1.8537967205047607, + "learning_rate": 4.047522339561333e-06, + "loss": 2.2201, + "step": 757 + }, + { + "epoch": 0.8306849315068493, + "grad_norm": 1.865315318107605, + "learning_rate": 4.04549147034931e-06, + "loss": 2.1819, + "step": 758 + }, + { + "epoch": 0.8317808219178082, + "grad_norm": 1.8620522022247314, + "learning_rate": 4.043460601137287e-06, + "loss": 2.1717, + "step": 759 + }, + { + "epoch": 0.8328767123287671, + "grad_norm": 1.9475384950637817, + "learning_rate": 4.041429731925264e-06, + "loss": 2.2718, + "step": 760 + }, + { + "epoch": 0.8328767123287671, + "eval_loss": 2.314429759979248, + "eval_runtime": 54.2324, + "eval_samples_per_second": 18.439, + "eval_steps_per_second": 0.774, + "step": 760 + }, + { + "epoch": 0.8339726027397261, + "grad_norm": 1.7170753479003906, + "learning_rate": 4.039398862713242e-06, + "loss": 2.1785, + "step": 761 + }, + { + "epoch": 0.8350684931506849, + "grad_norm": 1.968220591545105, + "learning_rate": 4.037367993501219e-06, + "loss": 2.1798, + "step": 762 + }, + { + "epoch": 0.8361643835616438, + "grad_norm": 1.7927111387252808, + "learning_rate": 4.035337124289196e-06, + "loss": 2.2372, + "step": 763 + }, + { + "epoch": 0.8372602739726027, + "grad_norm": 1.7871185541152954, + "learning_rate": 4.0333062550771735e-06, + "loss": 2.182, + "step": 764 + }, + { + "epoch": 0.8383561643835616, + "grad_norm": 1.9415812492370605, + "learning_rate": 4.031275385865151e-06, + "loss": 2.1527, + "step": 765 + }, + { + "epoch": 0.8394520547945206, + "grad_norm": 1.7327690124511719, + "learning_rate": 4.029244516653128e-06, + "loss": 2.1744, + "step": 766 + }, + { + "epoch": 0.8405479452054795, + "grad_norm": 1.9123411178588867, + "learning_rate": 4.027213647441105e-06, + "loss": 2.2049, + "step": 767 + }, + { + "epoch": 0.8416438356164384, + "grad_norm": 2.011591911315918, + "learning_rate": 4.025182778229083e-06, + "loss": 2.1391, + "step": 768 + }, + { + "epoch": 0.8427397260273972, + "grad_norm": 1.9169542789459229, + "learning_rate": 4.02315190901706e-06, + "loss": 2.1562, + "step": 769 + }, + { + "epoch": 0.8438356164383561, + "grad_norm": 1.8020051717758179, + "learning_rate": 4.021121039805037e-06, + "loss": 2.1795, + "step": 770 + }, + { + "epoch": 0.8438356164383561, + "eval_loss": 2.3199150562286377, + "eval_runtime": 52.468, + "eval_samples_per_second": 19.059, + "eval_steps_per_second": 0.8, + "step": 770 + }, + { + "epoch": 0.8449315068493151, + "grad_norm": 1.927666425704956, + "learning_rate": 4.0190901705930144e-06, + "loss": 2.1834, + "step": 771 + }, + { + "epoch": 0.846027397260274, + "grad_norm": 2.0998382568359375, + "learning_rate": 4.017059301380992e-06, + "loss": 2.2541, + "step": 772 + }, + { + "epoch": 0.8471232876712329, + "grad_norm": 1.88541579246521, + "learning_rate": 4.015028432168969e-06, + "loss": 2.1631, + "step": 773 + }, + { + "epoch": 0.8482191780821918, + "grad_norm": 1.9546253681182861, + "learning_rate": 4.012997562956945e-06, + "loss": 2.2266, + "step": 774 + }, + { + "epoch": 0.8493150684931506, + "grad_norm": 1.7969225645065308, + "learning_rate": 4.0109666937449235e-06, + "loss": 2.1526, + "step": 775 + }, + { + "epoch": 0.8504109589041096, + "grad_norm": 1.7237955331802368, + "learning_rate": 4.0089358245329e-06, + "loss": 2.1618, + "step": 776 + }, + { + "epoch": 0.8515068493150685, + "grad_norm": 1.8355857133865356, + "learning_rate": 4.006904955320878e-06, + "loss": 2.1673, + "step": 777 + }, + { + "epoch": 0.8526027397260274, + "grad_norm": 1.8049591779708862, + "learning_rate": 4.0048740861088545e-06, + "loss": 2.1676, + "step": 778 + }, + { + "epoch": 0.8536986301369863, + "grad_norm": 1.8143601417541504, + "learning_rate": 4.002843216896833e-06, + "loss": 2.214, + "step": 779 + }, + { + "epoch": 0.8547945205479452, + "grad_norm": 1.8344221115112305, + "learning_rate": 4.000812347684809e-06, + "loss": 2.1958, + "step": 780 + }, + { + "epoch": 0.8547945205479452, + "eval_loss": 2.314359426498413, + "eval_runtime": 51.8717, + "eval_samples_per_second": 19.278, + "eval_steps_per_second": 0.81, + "step": 780 + }, + { + "epoch": 0.8558904109589041, + "grad_norm": 1.944402813911438, + "learning_rate": 3.998781478472787e-06, + "loss": 2.1497, + "step": 781 + }, + { + "epoch": 0.856986301369863, + "grad_norm": 1.93898606300354, + "learning_rate": 3.996750609260764e-06, + "loss": 2.1737, + "step": 782 + }, + { + "epoch": 0.8580821917808219, + "grad_norm": 1.9499506950378418, + "learning_rate": 3.994719740048742e-06, + "loss": 2.1545, + "step": 783 + }, + { + "epoch": 0.8591780821917808, + "grad_norm": 1.9311052560806274, + "learning_rate": 3.992688870836718e-06, + "loss": 2.2559, + "step": 784 + }, + { + "epoch": 0.8602739726027397, + "grad_norm": 1.864862084388733, + "learning_rate": 3.990658001624696e-06, + "loss": 2.1996, + "step": 785 + }, + { + "epoch": 0.8613698630136987, + "grad_norm": 1.8248569965362549, + "learning_rate": 3.988627132412673e-06, + "loss": 2.1591, + "step": 786 + }, + { + "epoch": 0.8624657534246576, + "grad_norm": 2.2370970249176025, + "learning_rate": 3.98659626320065e-06, + "loss": 2.2095, + "step": 787 + }, + { + "epoch": 0.8635616438356164, + "grad_norm": 1.8877084255218506, + "learning_rate": 3.984565393988627e-06, + "loss": 2.1813, + "step": 788 + }, + { + "epoch": 0.8646575342465753, + "grad_norm": 2.1722517013549805, + "learning_rate": 3.9825345247766046e-06, + "loss": 2.1866, + "step": 789 + }, + { + "epoch": 0.8657534246575342, + "grad_norm": 1.9640655517578125, + "learning_rate": 3.980503655564582e-06, + "loss": 2.1818, + "step": 790 + }, + { + "epoch": 0.8657534246575342, + "eval_loss": 2.314087390899658, + "eval_runtime": 51.9778, + "eval_samples_per_second": 19.239, + "eval_steps_per_second": 0.808, + "step": 790 + }, + { + "epoch": 0.8668493150684932, + "grad_norm": 2.0583789348602295, + "learning_rate": 3.978472786352559e-06, + "loss": 2.1956, + "step": 791 + }, + { + "epoch": 0.8679452054794521, + "grad_norm": 2.2128682136535645, + "learning_rate": 3.976441917140536e-06, + "loss": 2.204, + "step": 792 + }, + { + "epoch": 0.869041095890411, + "grad_norm": 1.917763113975525, + "learning_rate": 3.974411047928514e-06, + "loss": 2.1841, + "step": 793 + }, + { + "epoch": 0.8701369863013698, + "grad_norm": 2.0327019691467285, + "learning_rate": 3.972380178716491e-06, + "loss": 2.2351, + "step": 794 + }, + { + "epoch": 0.8712328767123287, + "grad_norm": 1.9498991966247559, + "learning_rate": 3.970349309504468e-06, + "loss": 2.2074, + "step": 795 + }, + { + "epoch": 0.8723287671232877, + "grad_norm": 1.8105775117874146, + "learning_rate": 3.9683184402924455e-06, + "loss": 2.2088, + "step": 796 + }, + { + "epoch": 0.8734246575342466, + "grad_norm": 1.9299930334091187, + "learning_rate": 3.966287571080423e-06, + "loss": 2.1469, + "step": 797 + }, + { + "epoch": 0.8745205479452055, + "grad_norm": 1.9346121549606323, + "learning_rate": 3.9642567018684e-06, + "loss": 2.1434, + "step": 798 + }, + { + "epoch": 0.8756164383561644, + "grad_norm": 1.8202977180480957, + "learning_rate": 3.962225832656377e-06, + "loss": 2.194, + "step": 799 + }, + { + "epoch": 0.8767123287671232, + "grad_norm": 1.6922450065612793, + "learning_rate": 3.960194963444355e-06, + "loss": 2.1821, + "step": 800 + }, + { + "epoch": 0.8767123287671232, + "eval_loss": 2.310209035873413, + "eval_runtime": 46.1102, + "eval_samples_per_second": 21.687, + "eval_steps_per_second": 0.911, + "step": 800 + }, + { + "epoch": 0.8778082191780822, + "grad_norm": 1.9258904457092285, + "learning_rate": 3.958164094232332e-06, + "loss": 2.1326, + "step": 801 + }, + { + "epoch": 0.8789041095890411, + "grad_norm": 1.8485561609268188, + "learning_rate": 3.956133225020309e-06, + "loss": 2.1888, + "step": 802 + }, + { + "epoch": 0.88, + "grad_norm": 1.9802418947219849, + "learning_rate": 3.954102355808286e-06, + "loss": 2.1876, + "step": 803 + }, + { + "epoch": 0.8810958904109589, + "grad_norm": 1.8365259170532227, + "learning_rate": 3.952071486596264e-06, + "loss": 2.1804, + "step": 804 + }, + { + "epoch": 0.8821917808219178, + "grad_norm": 2.023909568786621, + "learning_rate": 3.950040617384241e-06, + "loss": 2.1801, + "step": 805 + }, + { + "epoch": 0.8832876712328767, + "grad_norm": 1.8889018297195435, + "learning_rate": 3.948009748172218e-06, + "loss": 2.1657, + "step": 806 + }, + { + "epoch": 0.8843835616438356, + "grad_norm": 1.7983578443527222, + "learning_rate": 3.9459788789601955e-06, + "loss": 2.1589, + "step": 807 + }, + { + "epoch": 0.8854794520547945, + "grad_norm": 2.157043218612671, + "learning_rate": 3.943948009748173e-06, + "loss": 2.2258, + "step": 808 + }, + { + "epoch": 0.8865753424657534, + "grad_norm": 1.7387373447418213, + "learning_rate": 3.94191714053615e-06, + "loss": 2.1878, + "step": 809 + }, + { + "epoch": 0.8876712328767123, + "grad_norm": 1.963233232498169, + "learning_rate": 3.939886271324127e-06, + "loss": 2.186, + "step": 810 + }, + { + "epoch": 0.8876712328767123, + "eval_loss": 2.306187152862549, + "eval_runtime": 42.563, + "eval_samples_per_second": 23.495, + "eval_steps_per_second": 0.987, + "step": 810 + }, + { + "epoch": 0.8887671232876713, + "grad_norm": 1.8932961225509644, + "learning_rate": 3.937855402112104e-06, + "loss": 2.198, + "step": 811 + }, + { + "epoch": 0.8898630136986302, + "grad_norm": 2.0431783199310303, + "learning_rate": 3.935824532900082e-06, + "loss": 2.1505, + "step": 812 + }, + { + "epoch": 0.890958904109589, + "grad_norm": 1.8954466581344604, + "learning_rate": 3.933793663688058e-06, + "loss": 2.1494, + "step": 813 + }, + { + "epoch": 0.8920547945205479, + "grad_norm": 1.841184139251709, + "learning_rate": 3.9317627944760365e-06, + "loss": 2.1932, + "step": 814 + }, + { + "epoch": 0.8931506849315068, + "grad_norm": 1.9083219766616821, + "learning_rate": 3.929731925264013e-06, + "loss": 2.2081, + "step": 815 + }, + { + "epoch": 0.8942465753424658, + "grad_norm": 1.9737226963043213, + "learning_rate": 3.927701056051991e-06, + "loss": 2.1405, + "step": 816 + }, + { + "epoch": 0.8953424657534247, + "grad_norm": 1.869194746017456, + "learning_rate": 3.9256701868399674e-06, + "loss": 2.1642, + "step": 817 + }, + { + "epoch": 0.8964383561643836, + "grad_norm": 1.9553205966949463, + "learning_rate": 3.9236393176279456e-06, + "loss": 2.1761, + "step": 818 + }, + { + "epoch": 0.8975342465753424, + "grad_norm": 1.8355534076690674, + "learning_rate": 3.921608448415922e-06, + "loss": 2.1551, + "step": 819 + }, + { + "epoch": 0.8986301369863013, + "grad_norm": 1.9770171642303467, + "learning_rate": 3.9195775792039e-06, + "loss": 2.1424, + "step": 820 + }, + { + "epoch": 0.8986301369863013, + "eval_loss": 2.307065010070801, + "eval_runtime": 44.5816, + "eval_samples_per_second": 22.431, + "eval_steps_per_second": 0.942, + "step": 820 + }, + { + "epoch": 0.8997260273972603, + "grad_norm": 1.7375415563583374, + "learning_rate": 3.9175467099918765e-06, + "loss": 2.1895, + "step": 821 + }, + { + "epoch": 0.9008219178082192, + "grad_norm": 1.8910484313964844, + "learning_rate": 3.915515840779855e-06, + "loss": 2.1685, + "step": 822 + }, + { + "epoch": 0.9019178082191781, + "grad_norm": 1.9191482067108154, + "learning_rate": 3.913484971567831e-06, + "loss": 2.2074, + "step": 823 + }, + { + "epoch": 0.903013698630137, + "grad_norm": 1.7864760160446167, + "learning_rate": 3.911454102355808e-06, + "loss": 2.1972, + "step": 824 + }, + { + "epoch": 0.9041095890410958, + "grad_norm": 1.9044650793075562, + "learning_rate": 3.909423233143786e-06, + "loss": 2.1237, + "step": 825 + }, + { + "epoch": 0.9052054794520548, + "grad_norm": 1.8378455638885498, + "learning_rate": 3.907392363931763e-06, + "loss": 2.1907, + "step": 826 + }, + { + "epoch": 0.9063013698630137, + "grad_norm": 1.8708094358444214, + "learning_rate": 3.90536149471974e-06, + "loss": 2.1883, + "step": 827 + }, + { + "epoch": 0.9073972602739726, + "grad_norm": 1.7802201509475708, + "learning_rate": 3.9033306255077175e-06, + "loss": 2.2356, + "step": 828 + }, + { + "epoch": 0.9084931506849315, + "grad_norm": 1.9374496936798096, + "learning_rate": 3.901299756295695e-06, + "loss": 2.2546, + "step": 829 + }, + { + "epoch": 0.9095890410958904, + "grad_norm": 1.9246270656585693, + "learning_rate": 3.899268887083672e-06, + "loss": 2.2684, + "step": 830 + }, + { + "epoch": 0.9095890410958904, + "eval_loss": 2.3065874576568604, + "eval_runtime": 45.18, + "eval_samples_per_second": 22.134, + "eval_steps_per_second": 0.93, + "step": 830 + }, + { + "epoch": 0.9106849315068493, + "grad_norm": 1.8681883811950684, + "learning_rate": 3.897238017871649e-06, + "loss": 2.1737, + "step": 831 + }, + { + "epoch": 0.9117808219178082, + "grad_norm": 1.9293930530548096, + "learning_rate": 3.8952071486596266e-06, + "loss": 2.1813, + "step": 832 + }, + { + "epoch": 0.9128767123287671, + "grad_norm": 1.844586968421936, + "learning_rate": 3.893176279447604e-06, + "loss": 2.1788, + "step": 833 + }, + { + "epoch": 0.913972602739726, + "grad_norm": 1.7658660411834717, + "learning_rate": 3.891145410235581e-06, + "loss": 2.1494, + "step": 834 + }, + { + "epoch": 0.915068493150685, + "grad_norm": 2.05153751373291, + "learning_rate": 3.889114541023558e-06, + "loss": 2.241, + "step": 835 + }, + { + "epoch": 0.9161643835616439, + "grad_norm": 1.7869952917099, + "learning_rate": 3.887083671811536e-06, + "loss": 2.1866, + "step": 836 + }, + { + "epoch": 0.9172602739726028, + "grad_norm": 1.99534273147583, + "learning_rate": 3.885052802599513e-06, + "loss": 2.1682, + "step": 837 + }, + { + "epoch": 0.9183561643835616, + "grad_norm": 1.9593979120254517, + "learning_rate": 3.88302193338749e-06, + "loss": 2.1357, + "step": 838 + }, + { + "epoch": 0.9194520547945205, + "grad_norm": 1.8441778421401978, + "learning_rate": 3.8809910641754675e-06, + "loss": 2.2283, + "step": 839 + }, + { + "epoch": 0.9205479452054794, + "grad_norm": 1.9153512716293335, + "learning_rate": 3.878960194963445e-06, + "loss": 2.1899, + "step": 840 + }, + { + "epoch": 0.9205479452054794, + "eval_loss": 2.307375907897949, + "eval_runtime": 48.0216, + "eval_samples_per_second": 20.824, + "eval_steps_per_second": 0.875, + "step": 840 + }, + { + "epoch": 0.9216438356164384, + "grad_norm": 2.004232883453369, + "learning_rate": 3.876929325751422e-06, + "loss": 2.173, + "step": 841 + }, + { + "epoch": 0.9227397260273973, + "grad_norm": 2.12992262840271, + "learning_rate": 3.874898456539399e-06, + "loss": 2.2164, + "step": 842 + }, + { + "epoch": 0.9238356164383562, + "grad_norm": 2.0505411624908447, + "learning_rate": 3.872867587327377e-06, + "loss": 2.1432, + "step": 843 + }, + { + "epoch": 0.924931506849315, + "grad_norm": 1.767324686050415, + "learning_rate": 3.870836718115354e-06, + "loss": 2.1838, + "step": 844 + }, + { + "epoch": 0.9260273972602739, + "grad_norm": 2.189107894897461, + "learning_rate": 3.868805848903331e-06, + "loss": 2.1623, + "step": 845 + }, + { + "epoch": 0.9271232876712329, + "grad_norm": 1.9781970977783203, + "learning_rate": 3.8667749796913084e-06, + "loss": 2.1738, + "step": 846 + }, + { + "epoch": 0.9282191780821918, + "grad_norm": 1.8734674453735352, + "learning_rate": 3.864744110479286e-06, + "loss": 2.1829, + "step": 847 + }, + { + "epoch": 0.9293150684931507, + "grad_norm": 1.7699538469314575, + "learning_rate": 3.862713241267262e-06, + "loss": 2.186, + "step": 848 + }, + { + "epoch": 0.9304109589041096, + "grad_norm": 1.8354437351226807, + "learning_rate": 3.86068237205524e-06, + "loss": 2.2149, + "step": 849 + }, + { + "epoch": 0.9315068493150684, + "grad_norm": 1.913249135017395, + "learning_rate": 3.858651502843217e-06, + "loss": 2.2208, + "step": 850 + }, + { + "epoch": 0.9315068493150684, + "eval_loss": 2.307549238204956, + "eval_runtime": 43.1706, + "eval_samples_per_second": 23.164, + "eval_steps_per_second": 0.973, + "step": 850 + }, + { + "epoch": 0.9326027397260274, + "grad_norm": 1.92095148563385, + "learning_rate": 3.856620633631195e-06, + "loss": 2.1839, + "step": 851 + }, + { + "epoch": 0.9336986301369863, + "grad_norm": 1.8833959102630615, + "learning_rate": 3.854589764419171e-06, + "loss": 2.1372, + "step": 852 + }, + { + "epoch": 0.9347945205479452, + "grad_norm": 1.9410737752914429, + "learning_rate": 3.852558895207149e-06, + "loss": 2.1726, + "step": 853 + }, + { + "epoch": 0.9358904109589041, + "grad_norm": 1.8328828811645508, + "learning_rate": 3.850528025995126e-06, + "loss": 2.2425, + "step": 854 + }, + { + "epoch": 0.936986301369863, + "grad_norm": 1.89652419090271, + "learning_rate": 3.848497156783104e-06, + "loss": 2.1956, + "step": 855 + }, + { + "epoch": 0.938082191780822, + "grad_norm": 1.9026265144348145, + "learning_rate": 3.84646628757108e-06, + "loss": 2.167, + "step": 856 + }, + { + "epoch": 0.9391780821917808, + "grad_norm": 2.010864734649658, + "learning_rate": 3.8444354183590585e-06, + "loss": 2.1725, + "step": 857 + }, + { + "epoch": 0.9402739726027397, + "grad_norm": 2.293717384338379, + "learning_rate": 3.842404549147035e-06, + "loss": 2.1668, + "step": 858 + }, + { + "epoch": 0.9413698630136986, + "grad_norm": 1.8209116458892822, + "learning_rate": 3.840373679935013e-06, + "loss": 2.1649, + "step": 859 + }, + { + "epoch": 0.9424657534246575, + "grad_norm": 2.061277389526367, + "learning_rate": 3.8383428107229894e-06, + "loss": 2.1479, + "step": 860 + }, + { + "epoch": 0.9424657534246575, + "eval_loss": 2.3073670864105225, + "eval_runtime": 44.3781, + "eval_samples_per_second": 22.534, + "eval_steps_per_second": 0.946, + "step": 860 + }, + { + "epoch": 0.9435616438356165, + "grad_norm": 1.8032523393630981, + "learning_rate": 3.836311941510967e-06, + "loss": 2.1665, + "step": 861 + }, + { + "epoch": 0.9446575342465754, + "grad_norm": 1.989533543586731, + "learning_rate": 3.834281072298944e-06, + "loss": 2.1612, + "step": 862 + }, + { + "epoch": 0.9457534246575342, + "grad_norm": 2.017179012298584, + "learning_rate": 3.832250203086921e-06, + "loss": 2.201, + "step": 863 + }, + { + "epoch": 0.9468493150684931, + "grad_norm": 2.0291004180908203, + "learning_rate": 3.8302193338748985e-06, + "loss": 2.1508, + "step": 864 + }, + { + "epoch": 0.947945205479452, + "grad_norm": 1.8693255186080933, + "learning_rate": 3.828188464662876e-06, + "loss": 2.2106, + "step": 865 + }, + { + "epoch": 0.949041095890411, + "grad_norm": 1.8890169858932495, + "learning_rate": 3.826157595450853e-06, + "loss": 2.2467, + "step": 866 + }, + { + "epoch": 0.9501369863013699, + "grad_norm": 2.0550642013549805, + "learning_rate": 3.82412672623883e-06, + "loss": 2.1899, + "step": 867 + }, + { + "epoch": 0.9512328767123288, + "grad_norm": 1.8681174516677856, + "learning_rate": 3.822095857026808e-06, + "loss": 2.1428, + "step": 868 + }, + { + "epoch": 0.9523287671232876, + "grad_norm": 1.9133951663970947, + "learning_rate": 3.820064987814785e-06, + "loss": 2.1947, + "step": 869 + }, + { + "epoch": 0.9534246575342465, + "grad_norm": 1.8717621564865112, + "learning_rate": 3.818034118602762e-06, + "loss": 2.1542, + "step": 870 + }, + { + "epoch": 0.9534246575342465, + "eval_loss": 2.307584047317505, + "eval_runtime": 43.718, + "eval_samples_per_second": 22.874, + "eval_steps_per_second": 0.961, + "step": 870 + }, + { + "epoch": 0.9545205479452055, + "grad_norm": 1.9843461513519287, + "learning_rate": 3.8160032493907395e-06, + "loss": 2.1547, + "step": 871 + }, + { + "epoch": 0.9556164383561644, + "grad_norm": 1.725468397140503, + "learning_rate": 3.813972380178717e-06, + "loss": 2.1679, + "step": 872 + }, + { + "epoch": 0.9567123287671233, + "grad_norm": 1.8076086044311523, + "learning_rate": 3.811941510966694e-06, + "loss": 2.1654, + "step": 873 + }, + { + "epoch": 0.9578082191780822, + "grad_norm": 1.8879774808883667, + "learning_rate": 3.809910641754671e-06, + "loss": 2.191, + "step": 874 + }, + { + "epoch": 0.958904109589041, + "grad_norm": 1.943313717842102, + "learning_rate": 3.8078797725426486e-06, + "loss": 2.1448, + "step": 875 + }, + { + "epoch": 0.96, + "grad_norm": 1.8877806663513184, + "learning_rate": 3.8058489033306254e-06, + "loss": 2.1129, + "step": 876 + }, + { + "epoch": 0.9610958904109589, + "grad_norm": 1.787083625793457, + "learning_rate": 3.803818034118603e-06, + "loss": 2.1854, + "step": 877 + }, + { + "epoch": 0.9621917808219178, + "grad_norm": 2.0925140380859375, + "learning_rate": 3.80178716490658e-06, + "loss": 2.1614, + "step": 878 + }, + { + "epoch": 0.9632876712328767, + "grad_norm": 1.9043725728988647, + "learning_rate": 3.7997562956945577e-06, + "loss": 2.2161, + "step": 879 + }, + { + "epoch": 0.9643835616438357, + "grad_norm": 1.8173545598983765, + "learning_rate": 3.7977254264825345e-06, + "loss": 2.2067, + "step": 880 + }, + { + "epoch": 0.9643835616438357, + "eval_loss": 2.302727460861206, + "eval_runtime": 44.2015, + "eval_samples_per_second": 22.624, + "eval_steps_per_second": 0.95, + "step": 880 + }, + { + "epoch": 0.9654794520547946, + "grad_norm": 1.9871562719345093, + "learning_rate": 3.7956945572705122e-06, + "loss": 2.1658, + "step": 881 + }, + { + "epoch": 0.9665753424657534, + "grad_norm": 2.103553295135498, + "learning_rate": 3.793663688058489e-06, + "loss": 2.1468, + "step": 882 + }, + { + "epoch": 0.9676712328767123, + "grad_norm": 1.9089381694793701, + "learning_rate": 3.791632818846467e-06, + "loss": 2.2289, + "step": 883 + }, + { + "epoch": 0.9687671232876712, + "grad_norm": 2.3238179683685303, + "learning_rate": 3.7896019496344436e-06, + "loss": 2.2179, + "step": 884 + }, + { + "epoch": 0.9698630136986301, + "grad_norm": 1.902161955833435, + "learning_rate": 3.787571080422421e-06, + "loss": 2.2086, + "step": 885 + }, + { + "epoch": 0.9709589041095891, + "grad_norm": 1.8587497472763062, + "learning_rate": 3.785540211210398e-06, + "loss": 2.2175, + "step": 886 + }, + { + "epoch": 0.972054794520548, + "grad_norm": 2.029719829559326, + "learning_rate": 3.7835093419983755e-06, + "loss": 2.207, + "step": 887 + }, + { + "epoch": 0.9731506849315068, + "grad_norm": 2.0228402614593506, + "learning_rate": 3.7814784727863527e-06, + "loss": 2.1851, + "step": 888 + }, + { + "epoch": 0.9742465753424657, + "grad_norm": 1.9885436296463013, + "learning_rate": 3.77944760357433e-06, + "loss": 2.1667, + "step": 889 + }, + { + "epoch": 0.9753424657534246, + "grad_norm": 1.8621459007263184, + "learning_rate": 3.7774167343623073e-06, + "loss": 2.2529, + "step": 890 + }, + { + "epoch": 0.9753424657534246, + "eval_loss": 2.303697347640991, + "eval_runtime": 44.9459, + "eval_samples_per_second": 22.249, + "eval_steps_per_second": 0.934, + "step": 890 + }, + { + "epoch": 0.9764383561643836, + "grad_norm": 2.0311460494995117, + "learning_rate": 3.7753858651502846e-06, + "loss": 2.2185, + "step": 891 + }, + { + "epoch": 0.9775342465753425, + "grad_norm": 2.2017149925231934, + "learning_rate": 3.773354995938262e-06, + "loss": 2.164, + "step": 892 + }, + { + "epoch": 0.9786301369863014, + "grad_norm": 1.949684977531433, + "learning_rate": 3.771324126726239e-06, + "loss": 2.2128, + "step": 893 + }, + { + "epoch": 0.9797260273972602, + "grad_norm": 1.9395498037338257, + "learning_rate": 3.7692932575142164e-06, + "loss": 2.1957, + "step": 894 + }, + { + "epoch": 0.9808219178082191, + "grad_norm": 2.096432685852051, + "learning_rate": 3.7672623883021937e-06, + "loss": 2.219, + "step": 895 + }, + { + "epoch": 0.9819178082191781, + "grad_norm": 2.0942952632904053, + "learning_rate": 3.765231519090171e-06, + "loss": 2.1546, + "step": 896 + }, + { + "epoch": 0.983013698630137, + "grad_norm": 2.0097291469573975, + "learning_rate": 3.7632006498781482e-06, + "loss": 2.1021, + "step": 897 + }, + { + "epoch": 0.9841095890410959, + "grad_norm": 2.0909335613250732, + "learning_rate": 3.761169780666125e-06, + "loss": 2.1862, + "step": 898 + }, + { + "epoch": 0.9852054794520548, + "grad_norm": 2.187309980392456, + "learning_rate": 3.7591389114541028e-06, + "loss": 2.1511, + "step": 899 + }, + { + "epoch": 0.9863013698630136, + "grad_norm": 2.194276809692383, + "learning_rate": 3.7571080422420796e-06, + "loss": 2.196, + "step": 900 + }, + { + "epoch": 0.9863013698630136, + "eval_loss": 2.3047611713409424, + "eval_runtime": 45.609, + "eval_samples_per_second": 21.925, + "eval_steps_per_second": 0.921, + "step": 900 + }, + { + "epoch": 0.9873972602739726, + "grad_norm": 2.1132829189300537, + "learning_rate": 3.7550771730300573e-06, + "loss": 2.2046, + "step": 901 + }, + { + "epoch": 0.9884931506849315, + "grad_norm": 2.1946468353271484, + "learning_rate": 3.753046303818034e-06, + "loss": 2.1656, + "step": 902 + }, + { + "epoch": 0.9895890410958904, + "grad_norm": 2.2210536003112793, + "learning_rate": 3.751015434606012e-06, + "loss": 2.2186, + "step": 903 + }, + { + "epoch": 0.9906849315068493, + "grad_norm": 1.913695216178894, + "learning_rate": 3.7489845653939887e-06, + "loss": 2.1933, + "step": 904 + }, + { + "epoch": 0.9917808219178083, + "grad_norm": 2.1088368892669678, + "learning_rate": 3.7469536961819664e-06, + "loss": 2.2406, + "step": 905 + }, + { + "epoch": 0.9928767123287672, + "grad_norm": 2.04239821434021, + "learning_rate": 3.7449228269699433e-06, + "loss": 2.142, + "step": 906 + }, + { + "epoch": 0.993972602739726, + "grad_norm": 1.7852507829666138, + "learning_rate": 3.742891957757921e-06, + "loss": 2.1701, + "step": 907 + }, + { + "epoch": 0.9950684931506849, + "grad_norm": 2.0275776386260986, + "learning_rate": 3.740861088545898e-06, + "loss": 2.2004, + "step": 908 + }, + { + "epoch": 0.9961643835616438, + "grad_norm": 1.9028904438018799, + "learning_rate": 3.7388302193338755e-06, + "loss": 2.177, + "step": 909 + }, + { + "epoch": 0.9972602739726028, + "grad_norm": 1.7815115451812744, + "learning_rate": 3.7367993501218524e-06, + "loss": 2.207, + "step": 910 + }, + { + "epoch": 0.9972602739726028, + "eval_loss": 2.302318811416626, + "eval_runtime": 42.9296, + "eval_samples_per_second": 23.294, + "eval_steps_per_second": 0.978, + "step": 910 + }, + { + "epoch": 0.9983561643835617, + "grad_norm": 1.913689374923706, + "learning_rate": 3.7347684809098297e-06, + "loss": 2.1594, + "step": 911 + }, + { + "epoch": 0.9994520547945206, + "grad_norm": 2.171222686767578, + "learning_rate": 3.732737611697807e-06, + "loss": 2.1689, + "step": 912 + }, + { + "epoch": 1.0005479452054795, + "grad_norm": 1.831642985343933, + "learning_rate": 3.7307067424857842e-06, + "loss": 2.1367, + "step": 913 + }, + { + "epoch": 1.0016438356164383, + "grad_norm": 2.1831955909729004, + "learning_rate": 3.7286758732737615e-06, + "loss": 2.1035, + "step": 914 + }, + { + "epoch": 1.0027397260273974, + "grad_norm": 2.038025379180908, + "learning_rate": 3.7266450040617388e-06, + "loss": 2.1461, + "step": 915 + }, + { + "epoch": 1.0038356164383562, + "grad_norm": 1.866502046585083, + "learning_rate": 3.724614134849716e-06, + "loss": 2.1322, + "step": 916 + }, + { + "epoch": 1.004931506849315, + "grad_norm": 2.1359357833862305, + "learning_rate": 3.7225832656376933e-06, + "loss": 2.2333, + "step": 917 + }, + { + "epoch": 1.006027397260274, + "grad_norm": 1.8689136505126953, + "learning_rate": 3.7205523964256706e-06, + "loss": 2.1176, + "step": 918 + }, + { + "epoch": 1.0071232876712328, + "grad_norm": 1.8869425058364868, + "learning_rate": 3.718521527213648e-06, + "loss": 2.2127, + "step": 919 + }, + { + "epoch": 1.0082191780821919, + "grad_norm": 1.9148075580596924, + "learning_rate": 3.716490658001625e-06, + "loss": 2.1846, + "step": 920 + }, + { + "epoch": 1.0082191780821919, + "eval_loss": 2.30368709564209, + "eval_runtime": 52.1845, + "eval_samples_per_second": 19.163, + "eval_steps_per_second": 0.805, + "step": 920 + }, + { + "epoch": 1.0093150684931507, + "grad_norm": 1.9502761363983154, + "learning_rate": 3.7144597887896024e-06, + "loss": 2.1731, + "step": 921 + }, + { + "epoch": 1.0104109589041095, + "grad_norm": 1.9093869924545288, + "learning_rate": 3.7124289195775797e-06, + "loss": 2.0864, + "step": 922 + }, + { + "epoch": 1.0115068493150685, + "grad_norm": 2.0871293544769287, + "learning_rate": 3.710398050365557e-06, + "loss": 2.1838, + "step": 923 + }, + { + "epoch": 1.0126027397260273, + "grad_norm": 2.0559985637664795, + "learning_rate": 3.708367181153534e-06, + "loss": 2.1764, + "step": 924 + }, + { + "epoch": 1.0136986301369864, + "grad_norm": 1.8968807458877563, + "learning_rate": 3.7063363119415115e-06, + "loss": 2.1515, + "step": 925 + }, + { + "epoch": 1.0147945205479452, + "grad_norm": 1.8613216876983643, + "learning_rate": 3.7043054427294884e-06, + "loss": 2.1694, + "step": 926 + }, + { + "epoch": 1.0158904109589042, + "grad_norm": 2.002044439315796, + "learning_rate": 3.702274573517466e-06, + "loss": 2.148, + "step": 927 + }, + { + "epoch": 1.016986301369863, + "grad_norm": 2.1138675212860107, + "learning_rate": 3.700243704305443e-06, + "loss": 2.1401, + "step": 928 + }, + { + "epoch": 1.0180821917808218, + "grad_norm": 1.7620841264724731, + "learning_rate": 3.6982128350934206e-06, + "loss": 2.1399, + "step": 929 + }, + { + "epoch": 1.0191780821917809, + "grad_norm": 1.8970746994018555, + "learning_rate": 3.6961819658813975e-06, + "loss": 2.1404, + "step": 930 + }, + { + "epoch": 1.0191780821917809, + "eval_loss": 2.3068811893463135, + "eval_runtime": 53.0188, + "eval_samples_per_second": 18.861, + "eval_steps_per_second": 0.792, + "step": 930 + }, + { + "epoch": 1.0202739726027397, + "grad_norm": 1.9405478239059448, + "learning_rate": 3.694151096669375e-06, + "loss": 2.1286, + "step": 931 + }, + { + "epoch": 1.0213698630136987, + "grad_norm": 2.112222909927368, + "learning_rate": 3.692120227457352e-06, + "loss": 2.2246, + "step": 932 + }, + { + "epoch": 1.0224657534246575, + "grad_norm": 1.8792580366134644, + "learning_rate": 3.6900893582453297e-06, + "loss": 2.148, + "step": 933 + }, + { + "epoch": 1.0235616438356165, + "grad_norm": 1.9530014991760254, + "learning_rate": 3.6880584890333066e-06, + "loss": 2.1707, + "step": 934 + }, + { + "epoch": 1.0246575342465754, + "grad_norm": 2.1778736114501953, + "learning_rate": 3.6860276198212834e-06, + "loss": 2.0879, + "step": 935 + }, + { + "epoch": 1.0257534246575342, + "grad_norm": 1.8960413932800293, + "learning_rate": 3.683996750609261e-06, + "loss": 2.1659, + "step": 936 + }, + { + "epoch": 1.0268493150684932, + "grad_norm": 1.9805221557617188, + "learning_rate": 3.681965881397238e-06, + "loss": 2.1674, + "step": 937 + }, + { + "epoch": 1.027945205479452, + "grad_norm": 1.9999068975448608, + "learning_rate": 3.6799350121852157e-06, + "loss": 2.1677, + "step": 938 + }, + { + "epoch": 1.029041095890411, + "grad_norm": 2.016911029815674, + "learning_rate": 3.6779041429731925e-06, + "loss": 2.1841, + "step": 939 + }, + { + "epoch": 1.0301369863013699, + "grad_norm": 1.7200374603271484, + "learning_rate": 3.6758732737611702e-06, + "loss": 2.1391, + "step": 940 + }, + { + "epoch": 1.0301369863013699, + "eval_loss": 2.3022048473358154, + "eval_runtime": 51.2908, + "eval_samples_per_second": 19.497, + "eval_steps_per_second": 0.819, + "step": 940 + }, + { + "epoch": 1.0312328767123287, + "grad_norm": 2.249377489089966, + "learning_rate": 3.673842404549147e-06, + "loss": 2.1618, + "step": 941 + }, + { + "epoch": 1.0323287671232877, + "grad_norm": 1.838922381401062, + "learning_rate": 3.671811535337125e-06, + "loss": 2.1527, + "step": 942 + }, + { + "epoch": 1.0334246575342465, + "grad_norm": 1.9571045637130737, + "learning_rate": 3.6697806661251016e-06, + "loss": 2.149, + "step": 943 + }, + { + "epoch": 1.0345205479452055, + "grad_norm": 1.9929102659225464, + "learning_rate": 3.6677497969130793e-06, + "loss": 2.159, + "step": 944 + }, + { + "epoch": 1.0356164383561643, + "grad_norm": 2.004830837249756, + "learning_rate": 3.665718927701056e-06, + "loss": 2.2369, + "step": 945 + }, + { + "epoch": 1.0367123287671234, + "grad_norm": 1.9605739116668701, + "learning_rate": 3.663688058489034e-06, + "loss": 2.1795, + "step": 946 + }, + { + "epoch": 1.0378082191780822, + "grad_norm": 1.9499164819717407, + "learning_rate": 3.6616571892770107e-06, + "loss": 2.1979, + "step": 947 + }, + { + "epoch": 1.038904109589041, + "grad_norm": 1.867347240447998, + "learning_rate": 3.659626320064988e-06, + "loss": 2.1616, + "step": 948 + }, + { + "epoch": 1.04, + "grad_norm": 1.9342801570892334, + "learning_rate": 3.6575954508529653e-06, + "loss": 2.1044, + "step": 949 + }, + { + "epoch": 1.0410958904109588, + "grad_norm": 2.0009584426879883, + "learning_rate": 3.6555645816409426e-06, + "loss": 2.1543, + "step": 950 + }, + { + "epoch": 1.0410958904109588, + "eval_loss": 2.3028438091278076, + "eval_runtime": 46.7426, + "eval_samples_per_second": 21.394, + "eval_steps_per_second": 0.899, + "step": 950 + }, + { + "epoch": 1.0421917808219179, + "grad_norm": 2.0411503314971924, + "learning_rate": 3.65353371242892e-06, + "loss": 2.1465, + "step": 951 + }, + { + "epoch": 1.0432876712328767, + "grad_norm": 2.0512070655822754, + "learning_rate": 3.651502843216897e-06, + "loss": 2.1828, + "step": 952 + }, + { + "epoch": 1.0443835616438357, + "grad_norm": 2.0948643684387207, + "learning_rate": 3.6494719740048744e-06, + "loss": 2.0671, + "step": 953 + }, + { + "epoch": 1.0454794520547945, + "grad_norm": 1.9103025197982788, + "learning_rate": 3.6474411047928517e-06, + "loss": 2.1651, + "step": 954 + }, + { + "epoch": 1.0465753424657533, + "grad_norm": 2.1500349044799805, + "learning_rate": 3.645410235580829e-06, + "loss": 2.137, + "step": 955 + }, + { + "epoch": 1.0476712328767124, + "grad_norm": 1.7760887145996094, + "learning_rate": 3.6433793663688062e-06, + "loss": 2.1738, + "step": 956 + }, + { + "epoch": 1.0487671232876712, + "grad_norm": 2.0111730098724365, + "learning_rate": 3.6413484971567835e-06, + "loss": 2.1569, + "step": 957 + }, + { + "epoch": 1.0498630136986302, + "grad_norm": 2.0040462017059326, + "learning_rate": 3.6393176279447608e-06, + "loss": 2.1483, + "step": 958 + }, + { + "epoch": 1.050958904109589, + "grad_norm": 1.9583499431610107, + "learning_rate": 3.637286758732738e-06, + "loss": 2.1575, + "step": 959 + }, + { + "epoch": 1.0520547945205478, + "grad_norm": 1.8842004537582397, + "learning_rate": 3.6352558895207153e-06, + "loss": 2.1797, + "step": 960 + }, + { + "epoch": 1.0520547945205478, + "eval_loss": 2.301715850830078, + "eval_runtime": 52.7345, + "eval_samples_per_second": 18.963, + "eval_steps_per_second": 0.796, + "step": 960 + }, + { + "epoch": 1.0531506849315069, + "grad_norm": 2.020909309387207, + "learning_rate": 3.633225020308692e-06, + "loss": 2.1939, + "step": 961 + }, + { + "epoch": 1.0542465753424657, + "grad_norm": 2.09222149848938, + "learning_rate": 3.63119415109667e-06, + "loss": 2.1777, + "step": 962 + }, + { + "epoch": 1.0553424657534247, + "grad_norm": 1.8582347631454468, + "learning_rate": 3.6291632818846467e-06, + "loss": 2.1505, + "step": 963 + }, + { + "epoch": 1.0564383561643835, + "grad_norm": 1.9468824863433838, + "learning_rate": 3.6271324126726244e-06, + "loss": 2.1762, + "step": 964 + }, + { + "epoch": 1.0575342465753426, + "grad_norm": 1.9484895467758179, + "learning_rate": 3.6251015434606013e-06, + "loss": 2.1102, + "step": 965 + }, + { + "epoch": 1.0586301369863014, + "grad_norm": 2.11462140083313, + "learning_rate": 3.623070674248579e-06, + "loss": 2.1114, + "step": 966 + }, + { + "epoch": 1.0597260273972602, + "grad_norm": 2.2038373947143555, + "learning_rate": 3.621039805036556e-06, + "loss": 2.1513, + "step": 967 + }, + { + "epoch": 1.0608219178082192, + "grad_norm": 2.1271684169769287, + "learning_rate": 3.6190089358245335e-06, + "loss": 2.137, + "step": 968 + }, + { + "epoch": 1.061917808219178, + "grad_norm": 1.8624194860458374, + "learning_rate": 3.6169780666125104e-06, + "loss": 2.1701, + "step": 969 + }, + { + "epoch": 1.063013698630137, + "grad_norm": 1.935445785522461, + "learning_rate": 3.614947197400488e-06, + "loss": 2.1726, + "step": 970 + }, + { + "epoch": 1.063013698630137, + "eval_loss": 2.302633047103882, + "eval_runtime": 51.8284, + "eval_samples_per_second": 19.294, + "eval_steps_per_second": 0.81, + "step": 970 + }, + { + "epoch": 1.0641095890410959, + "grad_norm": 2.048907518386841, + "learning_rate": 3.612916328188465e-06, + "loss": 2.1712, + "step": 971 + }, + { + "epoch": 1.0652054794520547, + "grad_norm": 1.850943684577942, + "learning_rate": 3.610885458976442e-06, + "loss": 2.102, + "step": 972 + }, + { + "epoch": 1.0663013698630137, + "grad_norm": 1.8364315032958984, + "learning_rate": 3.6088545897644195e-06, + "loss": 2.1903, + "step": 973 + }, + { + "epoch": 1.0673972602739725, + "grad_norm": 1.8227031230926514, + "learning_rate": 3.6068237205523963e-06, + "loss": 2.1787, + "step": 974 + }, + { + "epoch": 1.0684931506849316, + "grad_norm": 1.9136372804641724, + "learning_rate": 3.604792851340374e-06, + "loss": 2.1244, + "step": 975 + }, + { + "epoch": 1.0695890410958904, + "grad_norm": 1.8585968017578125, + "learning_rate": 3.602761982128351e-06, + "loss": 2.1287, + "step": 976 + }, + { + "epoch": 1.0706849315068494, + "grad_norm": 2.139573812484741, + "learning_rate": 3.6007311129163286e-06, + "loss": 2.1938, + "step": 977 + }, + { + "epoch": 1.0717808219178082, + "grad_norm": 2.1584925651550293, + "learning_rate": 3.5987002437043054e-06, + "loss": 2.1626, + "step": 978 + }, + { + "epoch": 1.072876712328767, + "grad_norm": 1.8905949592590332, + "learning_rate": 3.596669374492283e-06, + "loss": 2.1219, + "step": 979 + }, + { + "epoch": 1.073972602739726, + "grad_norm": 2.099651575088501, + "learning_rate": 3.59463850528026e-06, + "loss": 2.1385, + "step": 980 + }, + { + "epoch": 1.073972602739726, + "eval_loss": 2.3004090785980225, + "eval_runtime": 52.7406, + "eval_samples_per_second": 18.961, + "eval_steps_per_second": 0.796, + "step": 980 + }, + { + "epoch": 1.0750684931506849, + "grad_norm": 2.1456780433654785, + "learning_rate": 3.5926076360682377e-06, + "loss": 2.1973, + "step": 981 + }, + { + "epoch": 1.076164383561644, + "grad_norm": 1.929372787475586, + "learning_rate": 3.5905767668562145e-06, + "loss": 2.1115, + "step": 982 + }, + { + "epoch": 1.0772602739726027, + "grad_norm": 2.08174991607666, + "learning_rate": 3.5885458976441922e-06, + "loss": 2.1785, + "step": 983 + }, + { + "epoch": 1.0783561643835617, + "grad_norm": 1.9355952739715576, + "learning_rate": 3.586515028432169e-06, + "loss": 2.1567, + "step": 984 + }, + { + "epoch": 1.0794520547945206, + "grad_norm": 1.9983466863632202, + "learning_rate": 3.5844841592201464e-06, + "loss": 2.1913, + "step": 985 + }, + { + "epoch": 1.0805479452054794, + "grad_norm": 1.8714051246643066, + "learning_rate": 3.5824532900081237e-06, + "loss": 2.1938, + "step": 986 + }, + { + "epoch": 1.0816438356164384, + "grad_norm": 2.007049798965454, + "learning_rate": 3.580422420796101e-06, + "loss": 2.1093, + "step": 987 + }, + { + "epoch": 1.0827397260273972, + "grad_norm": 1.965572476387024, + "learning_rate": 3.578391551584078e-06, + "loss": 2.1426, + "step": 988 + }, + { + "epoch": 1.0838356164383562, + "grad_norm": 1.9040261507034302, + "learning_rate": 3.5763606823720555e-06, + "loss": 2.1627, + "step": 989 + }, + { + "epoch": 1.084931506849315, + "grad_norm": 2.001929759979248, + "learning_rate": 3.5743298131600328e-06, + "loss": 2.0983, + "step": 990 + }, + { + "epoch": 1.084931506849315, + "eval_loss": 2.2979862689971924, + "eval_runtime": 50.6325, + "eval_samples_per_second": 19.75, + "eval_steps_per_second": 0.83, + "step": 990 + }, + { + "epoch": 1.086027397260274, + "grad_norm": 1.9204787015914917, + "learning_rate": 3.57229894394801e-06, + "loss": 2.1571, + "step": 991 + }, + { + "epoch": 1.087123287671233, + "grad_norm": 1.9628651142120361, + "learning_rate": 3.5702680747359873e-06, + "loss": 2.1736, + "step": 992 + }, + { + "epoch": 1.0882191780821917, + "grad_norm": 1.9888029098510742, + "learning_rate": 3.5682372055239646e-06, + "loss": 2.17, + "step": 993 + }, + { + "epoch": 1.0893150684931507, + "grad_norm": 2.049769401550293, + "learning_rate": 3.566206336311942e-06, + "loss": 2.1352, + "step": 994 + }, + { + "epoch": 1.0904109589041096, + "grad_norm": 2.0208375453948975, + "learning_rate": 3.564175467099919e-06, + "loss": 2.1564, + "step": 995 + }, + { + "epoch": 1.0915068493150686, + "grad_norm": 1.976775050163269, + "learning_rate": 3.5621445978878964e-06, + "loss": 2.1292, + "step": 996 + }, + { + "epoch": 1.0926027397260274, + "grad_norm": 2.0502822399139404, + "learning_rate": 3.5601137286758737e-06, + "loss": 2.1115, + "step": 997 + }, + { + "epoch": 1.0936986301369862, + "grad_norm": 2.1184189319610596, + "learning_rate": 3.5580828594638505e-06, + "loss": 2.1641, + "step": 998 + }, + { + "epoch": 1.0947945205479452, + "grad_norm": 1.9662559032440186, + "learning_rate": 3.5560519902518282e-06, + "loss": 2.1338, + "step": 999 + }, + { + "epoch": 1.095890410958904, + "grad_norm": 2.13386869430542, + "learning_rate": 3.554021121039805e-06, + "loss": 2.158, + "step": 1000 + }, + { + "epoch": 1.095890410958904, + "eval_loss": 2.3001608848571777, + "eval_runtime": 45.9219, + "eval_samples_per_second": 21.776, + "eval_steps_per_second": 0.915, + "step": 1000 + }, + { + "epoch": 1.096986301369863, + "grad_norm": 1.8791636228561401, + "learning_rate": 3.551990251827783e-06, + "loss": 2.1752, + "step": 1001 + }, + { + "epoch": 1.098082191780822, + "grad_norm": 1.9105546474456787, + "learning_rate": 3.5499593826157596e-06, + "loss": 2.1543, + "step": 1002 + }, + { + "epoch": 1.099178082191781, + "grad_norm": 1.887668251991272, + "learning_rate": 3.5479285134037373e-06, + "loss": 2.1226, + "step": 1003 + }, + { + "epoch": 1.1002739726027397, + "grad_norm": 2.2854599952697754, + "learning_rate": 3.545897644191714e-06, + "loss": 2.1425, + "step": 1004 + }, + { + "epoch": 1.1013698630136985, + "grad_norm": 2.03892183303833, + "learning_rate": 3.543866774979692e-06, + "loss": 2.135, + "step": 1005 + }, + { + "epoch": 1.1024657534246576, + "grad_norm": 1.9144858121871948, + "learning_rate": 3.5418359057676687e-06, + "loss": 2.1323, + "step": 1006 + }, + { + "epoch": 1.1035616438356164, + "grad_norm": 2.1402337551116943, + "learning_rate": 3.5398050365556464e-06, + "loss": 2.1668, + "step": 1007 + }, + { + "epoch": 1.1046575342465754, + "grad_norm": 2.084592819213867, + "learning_rate": 3.5377741673436233e-06, + "loss": 2.1331, + "step": 1008 + }, + { + "epoch": 1.1057534246575342, + "grad_norm": 2.0012383460998535, + "learning_rate": 3.535743298131601e-06, + "loss": 2.1677, + "step": 1009 + }, + { + "epoch": 1.106849315068493, + "grad_norm": 2.272204637527466, + "learning_rate": 3.533712428919578e-06, + "loss": 2.13, + "step": 1010 + }, + { + "epoch": 1.106849315068493, + "eval_loss": 2.3001229763031006, + "eval_runtime": 42.0444, + "eval_samples_per_second": 23.784, + "eval_steps_per_second": 0.999, + "step": 1010 + }, + { + "epoch": 1.107945205479452, + "grad_norm": 2.1061534881591797, + "learning_rate": 3.5316815597075547e-06, + "loss": 2.1581, + "step": 1011 + }, + { + "epoch": 1.1090410958904109, + "grad_norm": 2.159306764602661, + "learning_rate": 3.5296506904955324e-06, + "loss": 2.1868, + "step": 1012 + }, + { + "epoch": 1.11013698630137, + "grad_norm": 2.1429545879364014, + "learning_rate": 3.5276198212835093e-06, + "loss": 2.1274, + "step": 1013 + }, + { + "epoch": 1.1112328767123287, + "grad_norm": 1.8494454622268677, + "learning_rate": 3.525588952071487e-06, + "loss": 2.1513, + "step": 1014 + }, + { + "epoch": 1.1123287671232878, + "grad_norm": 1.8494454622268677, + "learning_rate": 3.525588952071487e-06, + "loss": 2.1883, + "step": 1015 + }, + { + "epoch": 1.1134246575342466, + "grad_norm": 2.2244861125946045, + "learning_rate": 3.523558082859464e-06, + "loss": 2.1896, + "step": 1016 + }, + { + "epoch": 1.1145205479452054, + "grad_norm": 2.181950569152832, + "learning_rate": 3.5215272136474415e-06, + "loss": 2.1561, + "step": 1017 + }, + { + "epoch": 1.1156164383561644, + "grad_norm": 1.7740849256515503, + "learning_rate": 3.5194963444354184e-06, + "loss": 2.1358, + "step": 1018 + }, + { + "epoch": 1.1167123287671232, + "grad_norm": 2.3540139198303223, + "learning_rate": 3.517465475223396e-06, + "loss": 2.1241, + "step": 1019 + }, + { + "epoch": 1.1178082191780823, + "grad_norm": 2.2737820148468018, + "learning_rate": 3.515434606011373e-06, + "loss": 2.1117, + "step": 1020 + }, + { + "epoch": 1.1178082191780823, + "eval_loss": 2.2998275756835938, + "eval_runtime": 46.2142, + "eval_samples_per_second": 21.638, + "eval_steps_per_second": 0.909, + "step": 1020 + }, + { + "epoch": 1.118904109589041, + "grad_norm": 2.0705313682556152, + "learning_rate": 3.5134037367993506e-06, + "loss": 2.1743, + "step": 1021 + }, + { + "epoch": 1.12, + "grad_norm": 2.2605583667755127, + "learning_rate": 3.5113728675873275e-06, + "loss": 2.2082, + "step": 1022 + }, + { + "epoch": 1.121095890410959, + "grad_norm": 2.028203248977661, + "learning_rate": 3.5093419983753047e-06, + "loss": 2.1722, + "step": 1023 + }, + { + "epoch": 1.1221917808219177, + "grad_norm": 1.9794270992279053, + "learning_rate": 3.507311129163282e-06, + "loss": 2.1581, + "step": 1024 + }, + { + "epoch": 1.1232876712328768, + "grad_norm": 2.1830344200134277, + "learning_rate": 3.5052802599512593e-06, + "loss": 2.1142, + "step": 1025 + }, + { + "epoch": 1.1243835616438356, + "grad_norm": 1.926081895828247, + "learning_rate": 3.5032493907392366e-06, + "loss": 2.1532, + "step": 1026 + }, + { + "epoch": 1.1254794520547946, + "grad_norm": 1.9909394979476929, + "learning_rate": 3.501218521527214e-06, + "loss": 2.153, + "step": 1027 + }, + { + "epoch": 1.1265753424657534, + "grad_norm": 1.9685133695602417, + "learning_rate": 3.499187652315191e-06, + "loss": 2.0983, + "step": 1028 + }, + { + "epoch": 1.1276712328767124, + "grad_norm": 1.9185279607772827, + "learning_rate": 3.4971567831031684e-06, + "loss": 2.1495, + "step": 1029 + }, + { + "epoch": 1.1287671232876713, + "grad_norm": 2.0214717388153076, + "learning_rate": 3.4951259138911457e-06, + "loss": 2.176, + "step": 1030 + }, + { + "epoch": 1.1287671232876713, + "eval_loss": 2.2963690757751465, + "eval_runtime": 44.2655, + "eval_samples_per_second": 22.591, + "eval_steps_per_second": 0.949, + "step": 1030 + }, + { + "epoch": 1.12986301369863, + "grad_norm": 1.9134252071380615, + "learning_rate": 3.493095044679123e-06, + "loss": 2.1034, + "step": 1031 + }, + { + "epoch": 1.130958904109589, + "grad_norm": 1.8448772430419922, + "learning_rate": 3.4910641754671002e-06, + "loss": 2.1947, + "step": 1032 + }, + { + "epoch": 1.132054794520548, + "grad_norm": 1.856341004371643, + "learning_rate": 3.4890333062550775e-06, + "loss": 2.0676, + "step": 1033 + }, + { + "epoch": 1.133150684931507, + "grad_norm": 2.039062261581421, + "learning_rate": 3.4870024370430548e-06, + "loss": 2.1618, + "step": 1034 + }, + { + "epoch": 1.1342465753424658, + "grad_norm": 1.8440991640090942, + "learning_rate": 3.484971567831032e-06, + "loss": 2.0956, + "step": 1035 + }, + { + "epoch": 1.1353424657534246, + "grad_norm": 1.8483991622924805, + "learning_rate": 3.482940698619009e-06, + "loss": 2.1364, + "step": 1036 + }, + { + "epoch": 1.1364383561643836, + "grad_norm": 1.9045069217681885, + "learning_rate": 3.4809098294069866e-06, + "loss": 2.1786, + "step": 1037 + }, + { + "epoch": 1.1375342465753424, + "grad_norm": 1.8203203678131104, + "learning_rate": 3.4788789601949634e-06, + "loss": 2.1438, + "step": 1038 + }, + { + "epoch": 1.1386301369863014, + "grad_norm": 2.1084606647491455, + "learning_rate": 3.476848090982941e-06, + "loss": 2.1553, + "step": 1039 + }, + { + "epoch": 1.1397260273972603, + "grad_norm": 2.1041061878204346, + "learning_rate": 3.474817221770918e-06, + "loss": 2.1435, + "step": 1040 + }, + { + "epoch": 1.1397260273972603, + "eval_loss": 2.2970690727233887, + "eval_runtime": 43.2304, + "eval_samples_per_second": 23.132, + "eval_steps_per_second": 0.972, + "step": 1040 + }, + { + "epoch": 1.140821917808219, + "grad_norm": 1.9701104164123535, + "learning_rate": 3.4727863525588957e-06, + "loss": 2.2201, + "step": 1041 + }, + { + "epoch": 1.141917808219178, + "grad_norm": 1.8816475868225098, + "learning_rate": 3.4707554833468725e-06, + "loss": 2.1643, + "step": 1042 + }, + { + "epoch": 1.143013698630137, + "grad_norm": 2.096355438232422, + "learning_rate": 3.4687246141348502e-06, + "loss": 2.1514, + "step": 1043 + }, + { + "epoch": 1.144109589041096, + "grad_norm": 1.9975471496582031, + "learning_rate": 3.466693744922827e-06, + "loss": 2.1246, + "step": 1044 + }, + { + "epoch": 1.1452054794520548, + "grad_norm": 2.0587127208709717, + "learning_rate": 3.464662875710805e-06, + "loss": 2.1371, + "step": 1045 + }, + { + "epoch": 1.1463013698630138, + "grad_norm": 1.9275381565093994, + "learning_rate": 3.4626320064987817e-06, + "loss": 2.1518, + "step": 1046 + }, + { + "epoch": 1.1473972602739726, + "grad_norm": 2.1690948009490967, + "learning_rate": 3.4606011372867594e-06, + "loss": 2.1528, + "step": 1047 + }, + { + "epoch": 1.1484931506849314, + "grad_norm": 1.970973014831543, + "learning_rate": 3.458570268074736e-06, + "loss": 2.1569, + "step": 1048 + }, + { + "epoch": 1.1495890410958904, + "grad_norm": 1.8897690773010254, + "learning_rate": 3.4565393988627135e-06, + "loss": 2.1401, + "step": 1049 + }, + { + "epoch": 1.1506849315068493, + "grad_norm": 1.9245911836624146, + "learning_rate": 3.4545085296506908e-06, + "loss": 2.1311, + "step": 1050 + }, + { + "epoch": 1.1506849315068493, + "eval_loss": 2.2956528663635254, + "eval_runtime": 44.1778, + "eval_samples_per_second": 22.636, + "eval_steps_per_second": 0.951, + "step": 1050 + }, + { + "epoch": 1.1517808219178083, + "grad_norm": 2.142209529876709, + "learning_rate": 3.452477660438668e-06, + "loss": 2.1144, + "step": 1051 + }, + { + "epoch": 1.152876712328767, + "grad_norm": 1.932369351387024, + "learning_rate": 3.4504467912266453e-06, + "loss": 2.1739, + "step": 1052 + }, + { + "epoch": 1.1539726027397261, + "grad_norm": 2.259359359741211, + "learning_rate": 3.4484159220146226e-06, + "loss": 2.1036, + "step": 1053 + }, + { + "epoch": 1.155068493150685, + "grad_norm": 2.112485647201538, + "learning_rate": 3.4463850528026e-06, + "loss": 2.1429, + "step": 1054 + }, + { + "epoch": 1.1561643835616437, + "grad_norm": 2.6211049556732178, + "learning_rate": 3.444354183590577e-06, + "loss": 2.1421, + "step": 1055 + }, + { + "epoch": 1.1572602739726028, + "grad_norm": 1.8724851608276367, + "learning_rate": 3.4423233143785544e-06, + "loss": 2.1622, + "step": 1056 + }, + { + "epoch": 1.1583561643835616, + "grad_norm": 1.9292882680892944, + "learning_rate": 3.4402924451665317e-06, + "loss": 2.1071, + "step": 1057 + }, + { + "epoch": 1.1594520547945206, + "grad_norm": 1.9832521677017212, + "learning_rate": 3.438261575954509e-06, + "loss": 2.0879, + "step": 1058 + }, + { + "epoch": 1.1605479452054794, + "grad_norm": 1.9720780849456787, + "learning_rate": 3.4362307067424862e-06, + "loss": 2.1603, + "step": 1059 + }, + { + "epoch": 1.1616438356164385, + "grad_norm": 1.8179587125778198, + "learning_rate": 3.434199837530463e-06, + "loss": 2.1851, + "step": 1060 + }, + { + "epoch": 1.1616438356164385, + "eval_loss": 2.2950079441070557, + "eval_runtime": 42.9625, + "eval_samples_per_second": 23.276, + "eval_steps_per_second": 0.978, + "step": 1060 + }, + { + "epoch": 1.1627397260273973, + "grad_norm": 1.8762129545211792, + "learning_rate": 3.432168968318441e-06, + "loss": 2.1527, + "step": 1061 + }, + { + "epoch": 1.163835616438356, + "grad_norm": 2.173297643661499, + "learning_rate": 3.4301380991064176e-06, + "loss": 2.1023, + "step": 1062 + }, + { + "epoch": 1.1649315068493151, + "grad_norm": 1.9358222484588623, + "learning_rate": 3.4281072298943953e-06, + "loss": 2.1638, + "step": 1063 + }, + { + "epoch": 1.166027397260274, + "grad_norm": 1.9421939849853516, + "learning_rate": 3.426076360682372e-06, + "loss": 2.1794, + "step": 1064 + }, + { + "epoch": 1.167123287671233, + "grad_norm": 2.212996006011963, + "learning_rate": 3.42404549147035e-06, + "loss": 2.1877, + "step": 1065 + }, + { + "epoch": 1.1682191780821918, + "grad_norm": 2.036180257797241, + "learning_rate": 3.4220146222583267e-06, + "loss": 2.1534, + "step": 1066 + }, + { + "epoch": 1.1693150684931506, + "grad_norm": 1.8819142580032349, + "learning_rate": 3.4199837530463044e-06, + "loss": 2.099, + "step": 1067 + }, + { + "epoch": 1.1704109589041096, + "grad_norm": 2.0781915187835693, + "learning_rate": 3.4179528838342813e-06, + "loss": 2.1326, + "step": 1068 + }, + { + "epoch": 1.1715068493150684, + "grad_norm": 2.034895181655884, + "learning_rate": 3.415922014622259e-06, + "loss": 2.2107, + "step": 1069 + }, + { + "epoch": 1.1726027397260275, + "grad_norm": 1.92604660987854, + "learning_rate": 3.413891145410236e-06, + "loss": 2.1773, + "step": 1070 + }, + { + "epoch": 1.1726027397260275, + "eval_loss": 2.2959954738616943, + "eval_runtime": 47.2123, + "eval_samples_per_second": 21.181, + "eval_steps_per_second": 0.89, + "step": 1070 + }, + { + "epoch": 1.1736986301369863, + "grad_norm": 1.762730598449707, + "learning_rate": 3.4118602761982135e-06, + "loss": 2.1803, + "step": 1071 + }, + { + "epoch": 1.174794520547945, + "grad_norm": 2.0661215782165527, + "learning_rate": 3.4098294069861904e-06, + "loss": 2.1919, + "step": 1072 + }, + { + "epoch": 1.1758904109589041, + "grad_norm": 2.079982280731201, + "learning_rate": 3.4077985377741673e-06, + "loss": 2.1536, + "step": 1073 + }, + { + "epoch": 1.176986301369863, + "grad_norm": 1.9044359922409058, + "learning_rate": 3.405767668562145e-06, + "loss": 2.1403, + "step": 1074 + }, + { + "epoch": 1.178082191780822, + "grad_norm": 1.8870306015014648, + "learning_rate": 3.403736799350122e-06, + "loss": 2.1845, + "step": 1075 + }, + { + "epoch": 1.1791780821917808, + "grad_norm": 1.9305400848388672, + "learning_rate": 3.4017059301380995e-06, + "loss": 2.1786, + "step": 1076 + }, + { + "epoch": 1.1802739726027398, + "grad_norm": 1.987713098526001, + "learning_rate": 3.3996750609260764e-06, + "loss": 2.1214, + "step": 1077 + }, + { + "epoch": 1.1813698630136986, + "grad_norm": 1.8673466444015503, + "learning_rate": 3.397644191714054e-06, + "loss": 2.2027, + "step": 1078 + }, + { + "epoch": 1.1824657534246574, + "grad_norm": 2.009316921234131, + "learning_rate": 3.395613322502031e-06, + "loss": 2.139, + "step": 1079 + }, + { + "epoch": 1.1835616438356165, + "grad_norm": 2.1024956703186035, + "learning_rate": 3.3935824532900086e-06, + "loss": 2.1438, + "step": 1080 + }, + { + "epoch": 1.1835616438356165, + "eval_loss": 2.2966299057006836, + "eval_runtime": 44.8946, + "eval_samples_per_second": 22.274, + "eval_steps_per_second": 0.936, + "step": 1080 + }, + { + "epoch": 1.1846575342465753, + "grad_norm": 1.8911967277526855, + "learning_rate": 3.3915515840779855e-06, + "loss": 2.1238, + "step": 1081 + }, + { + "epoch": 1.1857534246575343, + "grad_norm": 2.0649192333221436, + "learning_rate": 3.389520714865963e-06, + "loss": 2.1534, + "step": 1082 + }, + { + "epoch": 1.1868493150684931, + "grad_norm": 1.9236847162246704, + "learning_rate": 3.38748984565394e-06, + "loss": 2.1427, + "step": 1083 + }, + { + "epoch": 1.1879452054794521, + "grad_norm": 1.8662182092666626, + "learning_rate": 3.3854589764419177e-06, + "loss": 2.1699, + "step": 1084 + }, + { + "epoch": 1.189041095890411, + "grad_norm": 1.9226020574569702, + "learning_rate": 3.3834281072298946e-06, + "loss": 2.1192, + "step": 1085 + }, + { + "epoch": 1.1901369863013698, + "grad_norm": 1.9659301042556763, + "learning_rate": 3.381397238017872e-06, + "loss": 2.1529, + "step": 1086 + }, + { + "epoch": 1.1912328767123288, + "grad_norm": 2.2714693546295166, + "learning_rate": 3.379366368805849e-06, + "loss": 2.1273, + "step": 1087 + }, + { + "epoch": 1.1923287671232876, + "grad_norm": 2.1853270530700684, + "learning_rate": 3.3773354995938264e-06, + "loss": 2.2029, + "step": 1088 + }, + { + "epoch": 1.1934246575342466, + "grad_norm": 1.8302830457687378, + "learning_rate": 3.3753046303818037e-06, + "loss": 2.1599, + "step": 1089 + }, + { + "epoch": 1.1945205479452055, + "grad_norm": 2.005591869354248, + "learning_rate": 3.373273761169781e-06, + "loss": 2.139, + "step": 1090 + }, + { + "epoch": 1.1945205479452055, + "eval_loss": 2.295994520187378, + "eval_runtime": 45.4802, + "eval_samples_per_second": 21.988, + "eval_steps_per_second": 0.923, + "step": 1090 + }, + { + "epoch": 1.1956164383561645, + "grad_norm": 2.139267683029175, + "learning_rate": 3.3712428919577582e-06, + "loss": 2.151, + "step": 1091 + }, + { + "epoch": 1.1967123287671233, + "grad_norm": 1.8547207117080688, + "learning_rate": 3.3692120227457355e-06, + "loss": 2.1249, + "step": 1092 + }, + { + "epoch": 1.1978082191780821, + "grad_norm": 2.214583396911621, + "learning_rate": 3.3671811535337128e-06, + "loss": 2.1656, + "step": 1093 + }, + { + "epoch": 1.1989041095890411, + "grad_norm": 2.1608800888061523, + "learning_rate": 3.36515028432169e-06, + "loss": 2.1056, + "step": 1094 + }, + { + "epoch": 1.2, + "grad_norm": 2.0850605964660645, + "learning_rate": 3.3631194151096673e-06, + "loss": 2.1215, + "step": 1095 + }, + { + "epoch": 1.201095890410959, + "grad_norm": 1.9199103116989136, + "learning_rate": 3.3610885458976446e-06, + "loss": 2.1584, + "step": 1096 + }, + { + "epoch": 1.2021917808219178, + "grad_norm": 1.9766334295272827, + "learning_rate": 3.359057676685622e-06, + "loss": 2.1821, + "step": 1097 + }, + { + "epoch": 1.2032876712328768, + "grad_norm": 1.948988676071167, + "learning_rate": 3.357026807473599e-06, + "loss": 2.182, + "step": 1098 + }, + { + "epoch": 1.2043835616438356, + "grad_norm": 2.0419397354125977, + "learning_rate": 3.354995938261576e-06, + "loss": 2.1051, + "step": 1099 + }, + { + "epoch": 1.2054794520547945, + "grad_norm": 1.9751213788986206, + "learning_rate": 3.3529650690495537e-06, + "loss": 2.1255, + "step": 1100 + }, + { + "epoch": 1.2054794520547945, + "eval_loss": 2.293811798095703, + "eval_runtime": 46.1182, + "eval_samples_per_second": 21.683, + "eval_steps_per_second": 0.911, + "step": 1100 + } + ], + "logging_steps": 1, + "max_steps": 2736, + "num_input_tokens_seen": 0, + "num_train_epochs": 3, + "save_steps": 100, + "stateful_callbacks": { + "TrainerControl": { + "args": { + "should_epoch_stop": false, + "should_evaluate": false, + "should_log": false, + "should_save": true, + "should_training_stop": false + }, + "attributes": {} + } + }, + "total_flos": 3.429040350298112e+17, + "train_batch_size": 3, + "trial_name": null, + "trial_params": null +} diff --git a/babylm_shuffle_nondeterministic_10M_seed0/runs/checkpoint-1100/zero_to_fp32.py b/babylm_shuffle_nondeterministic_10M_seed0/runs/checkpoint-1100/zero_to_fp32.py new file mode 100644 index 0000000000000000000000000000000000000000..24cc342e78d1a006c782b3a4cd68d9ce786d8fd8 --- /dev/null +++ b/babylm_shuffle_nondeterministic_10M_seed0/runs/checkpoint-1100/zero_to_fp32.py @@ -0,0 +1,604 @@ +#!/usr/bin/env python + +# Copyright (c) Microsoft Corporation. +# SPDX-License-Identifier: Apache-2.0 + +# DeepSpeed Team + +# This script extracts fp32 consolidated weights from a zero 1, 2 and 3 DeepSpeed checkpoints. It gets +# copied into the top level checkpoint dir, so the user can easily do the conversion at any point in +# the future. Once extracted, the weights don't require DeepSpeed and can be used in any +# application. +# +# example: python zero_to_fp32.py . pytorch_model.bin + +import argparse +import torch +import glob +import math +import os +import re +from collections import OrderedDict +from dataclasses import dataclass + +# while this script doesn't use deepspeed to recover data, since the checkpoints are pickled with +# DeepSpeed data structures it has to be available in the current python environment. +from deepspeed.utils import logger +from deepspeed.checkpoint.constants import (DS_VERSION, OPTIMIZER_STATE_DICT, SINGLE_PARTITION_OF_FP32_GROUPS, + FP32_FLAT_GROUPS, ZERO_STAGE, PARTITION_COUNT, PARAM_SHAPES, BUFFER_NAMES, + FROZEN_PARAM_SHAPES, FROZEN_PARAM_FRAGMENTS) + + +@dataclass +class zero_model_state: + buffers: dict() + param_shapes: dict() + shared_params: list + ds_version: int + frozen_param_shapes: dict() + frozen_param_fragments: dict() + + +debug = 0 + +# load to cpu +device = torch.device('cpu') + + +def atoi(text): + return int(text) if text.isdigit() else text + + +def natural_keys(text): + ''' + alist.sort(key=natural_keys) sorts in human order + http://nedbatchelder.com/blog/200712/human_sorting.html + (See Toothy's implementation in the comments) + ''' + return [atoi(c) for c in re.split(r'(\d+)', text)] + + +def get_model_state_file(checkpoint_dir, zero_stage): + if not os.path.isdir(checkpoint_dir): + raise FileNotFoundError(f"Directory '{checkpoint_dir}' doesn't exist") + + # there should be only one file + if zero_stage <= 2: + file = os.path.join(checkpoint_dir, "mp_rank_00_model_states.pt") + elif zero_stage == 3: + file = os.path.join(checkpoint_dir, "zero_pp_rank_0_mp_rank_00_model_states.pt") + + if not os.path.exists(file): + raise FileNotFoundError(f"can't find model states file at '{file}'") + + return file + + +def get_checkpoint_files(checkpoint_dir, glob_pattern): + # XXX: need to test that this simple glob rule works for multi-node setup too + ckpt_files = sorted(glob.glob(os.path.join(checkpoint_dir, glob_pattern)), key=natural_keys) + + if len(ckpt_files) == 0: + raise FileNotFoundError(f"can't find {glob_pattern} files in directory '{checkpoint_dir}'") + + return ckpt_files + + +def get_optim_files(checkpoint_dir): + return get_checkpoint_files(checkpoint_dir, "*_optim_states.pt") + + +def get_model_state_files(checkpoint_dir): + return get_checkpoint_files(checkpoint_dir, "*_model_states.pt") + + +def parse_model_states(files): + zero_model_states = [] + for file in files: + state_dict = torch.load(file, map_location=device) + + if BUFFER_NAMES not in state_dict: + raise ValueError(f"{file} is not a model state checkpoint") + buffer_names = state_dict[BUFFER_NAMES] + if debug: + print("Found buffers:", buffer_names) + + # recover just the buffers while restoring them to fp32 if they were saved in fp16 + buffers = {k: v.float() for k, v in state_dict["module"].items() if k in buffer_names} + param_shapes = state_dict[PARAM_SHAPES] + + # collect parameters that are included in param_shapes + param_names = [] + for s in param_shapes: + for name in s.keys(): + param_names.append(name) + + # update with frozen parameters + frozen_param_shapes = state_dict.get(FROZEN_PARAM_SHAPES, None) + if frozen_param_shapes is not None: + if debug: + print(f"Found frozen_param_shapes: {frozen_param_shapes}") + param_names += list(frozen_param_shapes.keys()) + + # handle shared params + shared_params = [[k, v] for k, v in state_dict["shared_params"].items()] + + ds_version = state_dict.get(DS_VERSION, None) + + frozen_param_fragments = state_dict.get(FROZEN_PARAM_FRAGMENTS, None) + + z_model_state = zero_model_state(buffers=buffers, + param_shapes=param_shapes, + shared_params=shared_params, + ds_version=ds_version, + frozen_param_shapes=frozen_param_shapes, + frozen_param_fragments=frozen_param_fragments) + zero_model_states.append(z_model_state) + + return zero_model_states + + +def parse_optim_states(files, ds_checkpoint_dir): + + total_files = len(files) + state_dicts = [] + for f in files: + state_dict = torch.load(f, map_location=device) + # immediately discard the potentially huge 2 optimizer states as we only care for fp32 master weights + # and also handle the case where it was already removed by another helper script + state_dict["optimizer_state_dict"].pop("optimizer_state_dict", None) + state_dicts.append(state_dict) + + if not ZERO_STAGE in state_dicts[0][OPTIMIZER_STATE_DICT]: + raise ValueError(f"{files[0]} is not a zero checkpoint") + zero_stage = state_dicts[0][OPTIMIZER_STATE_DICT][ZERO_STAGE] + world_size = state_dicts[0][OPTIMIZER_STATE_DICT][PARTITION_COUNT] + + # For ZeRO-2 each param group can have different partition_count as data parallelism for expert + # parameters can be different from data parallelism for non-expert parameters. So we can just + # use the max of the partition_count to get the dp world_size. + + if type(world_size) is list: + world_size = max(world_size) + + if world_size != total_files: + raise ValueError( + f"Expected {world_size} of '*_optim_states.pt' under '{ds_checkpoint_dir}' but found {total_files} files. " + "Possibly due to an overwrite of an old checkpoint, or a checkpoint didn't get saved by one or more processes." + ) + + # the groups are named differently in each stage + if zero_stage <= 2: + fp32_groups_key = SINGLE_PARTITION_OF_FP32_GROUPS + elif zero_stage == 3: + fp32_groups_key = FP32_FLAT_GROUPS + else: + raise ValueError(f"unknown zero stage {zero_stage}") + + if zero_stage <= 2: + fp32_flat_groups = [state_dicts[i][OPTIMIZER_STATE_DICT][fp32_groups_key] for i in range(len(state_dicts))] + elif zero_stage == 3: + # if there is more than one param group, there will be multiple flattened tensors - one + # flattened tensor per group - for simplicity merge them into a single tensor + # + # XXX: could make the script more memory efficient for when there are multiple groups - it + # will require matching the sub-lists of param_shapes for each param group flattened tensor + + fp32_flat_groups = [ + torch.cat(state_dicts[i][OPTIMIZER_STATE_DICT][fp32_groups_key], 0) for i in range(len(state_dicts)) + ] + + return zero_stage, world_size, fp32_flat_groups + + +def _get_fp32_state_dict_from_zero_checkpoint(ds_checkpoint_dir, exclude_frozen_parameters): + """ + Returns fp32 state_dict reconstructed from ds checkpoint + + Args: + - ``ds_checkpoint_dir``: path to the deepspeed checkpoint folder (where the optimizer files are) + + """ + print(f"Processing zero checkpoint '{ds_checkpoint_dir}'") + + optim_files = get_optim_files(ds_checkpoint_dir) + zero_stage, world_size, fp32_flat_groups = parse_optim_states(optim_files, ds_checkpoint_dir) + print(f"Detected checkpoint of type zero stage {zero_stage}, world_size: {world_size}") + + model_files = get_model_state_files(ds_checkpoint_dir) + + zero_model_states = parse_model_states(model_files) + print(f'Parsing checkpoint created by deepspeed=={zero_model_states[0].ds_version}') + + if zero_stage <= 2: + return _get_fp32_state_dict_from_zero2_checkpoint(world_size, fp32_flat_groups, zero_model_states, + exclude_frozen_parameters) + elif zero_stage == 3: + return _get_fp32_state_dict_from_zero3_checkpoint(world_size, fp32_flat_groups, zero_model_states, + exclude_frozen_parameters) + + +def _zero2_merge_frozen_params(state_dict, zero_model_states): + if zero_model_states[0].frozen_param_shapes is None or len(zero_model_states[0].frozen_param_shapes) == 0: + return + + frozen_param_shapes = zero_model_states[0].frozen_param_shapes + frozen_param_fragments = zero_model_states[0].frozen_param_fragments + + if debug: + num_elem = sum(s.numel() for s in frozen_param_shapes.values()) + print(f'rank 0: {FROZEN_PARAM_SHAPES}.numel = {num_elem}') + + wanted_params = len(frozen_param_shapes) + wanted_numel = sum(s.numel() for s in frozen_param_shapes.values()) + avail_numel = sum([p.numel() for p in frozen_param_fragments.values()]) + print(f'Frozen params: Have {avail_numel} numels to process.') + print(f'Frozen params: Need {wanted_numel} numels in {wanted_params} params') + + total_params = 0 + total_numel = 0 + for name, shape in frozen_param_shapes.items(): + total_params += 1 + unpartitioned_numel = shape.numel() + total_numel += unpartitioned_numel + + state_dict[name] = frozen_param_fragments[name] + + if debug: + print(f"{name} full shape: {shape} unpartitioned numel {unpartitioned_numel} ") + + print(f"Reconstructed Frozen fp32 state dict with {total_params} params {total_numel} elements") + + +def _has_callable(obj, fn): + attr = getattr(obj, fn, None) + return callable(attr) + + +def _zero2_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states): + param_shapes = zero_model_states[0].param_shapes + + # Reconstruction protocol: + # + # XXX: document this + + if debug: + for i in range(world_size): + for j in range(len(fp32_flat_groups[0])): + print(f"{FP32_FLAT_GROUPS}[{i}][{j}].shape={fp32_flat_groups[i][j].shape}") + + # XXX: memory usage doubles here (zero2) + num_param_groups = len(fp32_flat_groups[0]) + merged_single_partition_of_fp32_groups = [] + for i in range(num_param_groups): + merged_partitions = [sd[i] for sd in fp32_flat_groups] + full_single_fp32_vector = torch.cat(merged_partitions, 0) + merged_single_partition_of_fp32_groups.append(full_single_fp32_vector) + avail_numel = sum( + [full_single_fp32_vector.numel() for full_single_fp32_vector in merged_single_partition_of_fp32_groups]) + + if debug: + wanted_params = sum([len(shapes) for shapes in param_shapes]) + wanted_numel = sum([sum(shape.numel() for shape in shapes.values()) for shapes in param_shapes]) + # not asserting if there is a mismatch due to possible padding + print(f"Have {avail_numel} numels to process.") + print(f"Need {wanted_numel} numels in {wanted_params} params.") + + # params + # XXX: for huge models that can't fit into the host's RAM we will have to recode this to support + # out-of-core computing solution + total_numel = 0 + total_params = 0 + for shapes, full_single_fp32_vector in zip(param_shapes, merged_single_partition_of_fp32_groups): + offset = 0 + avail_numel = full_single_fp32_vector.numel() + for name, shape in shapes.items(): + + unpartitioned_numel = shape.numel() if _has_callable(shape, 'numel') else math.prod(shape) + total_numel += unpartitioned_numel + total_params += 1 + + if debug: + print(f"{name} full shape: {shape} unpartitioned numel {unpartitioned_numel} ") + state_dict[name] = full_single_fp32_vector.narrow(0, offset, unpartitioned_numel).view(shape) + offset += unpartitioned_numel + + # Z2 started to align to 2*world_size to improve nccl performance. Therefore both offset and + # avail_numel can differ by anywhere between 0..2*world_size. Due to two unrelated complex + # paddings performed in the code it's almost impossible to predict the exact numbers w/o the + # live optimizer object, so we are checking that the numbers are within the right range + align_to = 2 * world_size + + def zero2_align(x): + return align_to * math.ceil(x / align_to) + + if debug: + print(f"original offset={offset}, avail_numel={avail_numel}") + + offset = zero2_align(offset) + avail_numel = zero2_align(avail_numel) + + if debug: + print(f"aligned offset={offset}, avail_numel={avail_numel}") + + # Sanity check + if offset != avail_numel: + raise ValueError(f"consumed {offset} numels out of {avail_numel} - something is wrong") + + print(f"Reconstructed fp32 state dict with {total_params} params {total_numel} elements") + + +def _get_fp32_state_dict_from_zero2_checkpoint(world_size, fp32_flat_groups, zero_model_states, + exclude_frozen_parameters): + state_dict = OrderedDict() + + # buffers + buffers = zero_model_states[0].buffers + state_dict.update(buffers) + if debug: + print(f"added {len(buffers)} buffers") + + if not exclude_frozen_parameters: + _zero2_merge_frozen_params(state_dict, zero_model_states) + + _zero2_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states) + + # recover shared parameters + for pair in zero_model_states[0].shared_params: + if pair[1] in state_dict: + state_dict[pair[0]] = state_dict[pair[1]] + + return state_dict + + +def zero3_partitioned_param_info(unpartitioned_numel, world_size): + remainder = unpartitioned_numel % world_size + padding_numel = (world_size - remainder) if remainder else 0 + partitioned_numel = math.ceil(unpartitioned_numel / world_size) + return partitioned_numel, padding_numel + + +def _zero3_merge_frozen_params(state_dict, world_size, zero_model_states): + if zero_model_states[0].frozen_param_shapes is None or len(zero_model_states[0].frozen_param_shapes) == 0: + return + + if debug: + for i in range(world_size): + num_elem = sum(s.numel() for s in zero_model_states[i].frozen_param_fragments.values()) + print(f'rank {i}: {FROZEN_PARAM_SHAPES}.numel = {num_elem}') + + frozen_param_shapes = zero_model_states[0].frozen_param_shapes + wanted_params = len(frozen_param_shapes) + wanted_numel = sum(s.numel() for s in frozen_param_shapes.values()) + avail_numel = sum([p.numel() for p in zero_model_states[0].frozen_param_fragments.values()]) * world_size + print(f'Frozen params: Have {avail_numel} numels to process.') + print(f'Frozen params: Need {wanted_numel} numels in {wanted_params} params') + + total_params = 0 + total_numel = 0 + for name, shape in zero_model_states[0].frozen_param_shapes.items(): + total_params += 1 + unpartitioned_numel = shape.numel() + total_numel += unpartitioned_numel + + param_frags = tuple(model_state.frozen_param_fragments[name] for model_state in zero_model_states) + state_dict[name] = torch.cat(param_frags, 0).narrow(0, 0, unpartitioned_numel).view(shape) + + partitioned_numel, partitioned_padding_numel = zero3_partitioned_param_info(unpartitioned_numel, world_size) + + if debug: + print( + f"Frozen params: {total_params} {name} full shape: {shape} partition0 numel={partitioned_numel} partitioned_padding_numel={partitioned_padding_numel}" + ) + + print(f"Reconstructed Frozen fp32 state dict with {total_params} params {total_numel} elements") + + +def _zero3_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states): + param_shapes = zero_model_states[0].param_shapes + avail_numel = fp32_flat_groups[0].numel() * world_size + # Reconstruction protocol: For zero3 we need to zip the partitions together at boundary of each + # param, re-consolidating each param, while dealing with padding if any + + # merge list of dicts, preserving order + param_shapes = {k: v for d in param_shapes for k, v in d.items()} + + if debug: + for i in range(world_size): + print(f"{FP32_FLAT_GROUPS}[{i}].shape={fp32_flat_groups[i].shape}") + + wanted_params = len(param_shapes) + wanted_numel = sum(shape.numel() for shape in param_shapes.values()) + # not asserting if there is a mismatch due to possible padding + avail_numel = fp32_flat_groups[0].numel() * world_size + print(f"Trainable params: Have {avail_numel} numels to process.") + print(f"Trainable params: Need {wanted_numel} numels in {wanted_params} params.") + + # params + # XXX: for huge models that can't fit into the host's RAM we will have to recode this to support + # out-of-core computing solution + offset = 0 + total_numel = 0 + total_params = 0 + for name, shape in param_shapes.items(): + + unpartitioned_numel = shape.numel() + total_numel += unpartitioned_numel + total_params += 1 + + partitioned_numel, partitioned_padding_numel = zero3_partitioned_param_info(unpartitioned_numel, world_size) + + if debug: + print( + f"Trainable params: {total_params} {name} full shape: {shape} partition0 numel={partitioned_numel} partitioned_padding_numel={partitioned_padding_numel}" + ) + + # XXX: memory usage doubles here + state_dict[name] = torch.cat( + tuple(fp32_flat_groups[i].narrow(0, offset, partitioned_numel) for i in range(world_size)), + 0).narrow(0, 0, unpartitioned_numel).view(shape) + offset += partitioned_numel + + offset *= world_size + + # Sanity check + if offset != avail_numel: + raise ValueError(f"consumed {offset} numels out of {avail_numel} - something is wrong") + + print(f"Reconstructed Trainable fp32 state dict with {total_params} params {total_numel} elements") + + +def _get_fp32_state_dict_from_zero3_checkpoint(world_size, fp32_flat_groups, zero_model_states, + exclude_frozen_parameters): + state_dict = OrderedDict() + + # buffers + buffers = zero_model_states[0].buffers + state_dict.update(buffers) + if debug: + print(f"added {len(buffers)} buffers") + + if not exclude_frozen_parameters: + _zero3_merge_frozen_params(state_dict, world_size, zero_model_states) + + _zero3_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states) + + # recover shared parameters + for pair in zero_model_states[0].shared_params: + if pair[1] in state_dict: + state_dict[pair[0]] = state_dict[pair[1]] + + return state_dict + + +def get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir, tag=None, exclude_frozen_parameters=False): + """ + Convert ZeRO 2 or 3 checkpoint into a single fp32 consolidated state_dict that can be loaded with + ``load_state_dict()`` and used for training without DeepSpeed or shared with others, for example + via a model hub. + + Args: + - ``checkpoint_dir``: path to the desired checkpoint folder + - ``tag``: checkpoint tag used as a unique identifier for checkpoint. If not provided will attempt to load tag in 'latest' file. e.g., ``global_step14`` + - ``exclude_frozen_parameters``: exclude frozen parameters + + Returns: + - pytorch ``state_dict`` + + Note: this approach may not work if your application doesn't have sufficient free CPU memory and + you may need to use the offline approach using the ``zero_to_fp32.py`` script that is saved with + the checkpoint. + + A typical usage might be :: + + from deepspeed.utils.zero_to_fp32 import get_fp32_state_dict_from_zero_checkpoint + # do the training and checkpoint saving + state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir) # already on cpu + model = model.cpu() # move to cpu + model.load_state_dict(state_dict) + # submit to model hub or save the model to share with others + + In this example the ``model`` will no longer be usable in the deepspeed context of the same + application. i.e. you will need to re-initialize the deepspeed engine, since + ``model.load_state_dict(state_dict)`` will remove all the deepspeed magic from it. + + If you want it all done for you, use ``load_state_dict_from_zero_checkpoint`` instead. + + """ + if tag is None: + latest_path = os.path.join(checkpoint_dir, 'latest') + if os.path.isfile(latest_path): + with open(latest_path, 'r') as fd: + tag = fd.read().strip() + else: + raise ValueError(f"Unable to find 'latest' file at {latest_path}") + + ds_checkpoint_dir = os.path.join(checkpoint_dir, tag) + + if not os.path.isdir(ds_checkpoint_dir): + raise FileNotFoundError(f"Directory '{ds_checkpoint_dir}' doesn't exist") + + return _get_fp32_state_dict_from_zero_checkpoint(ds_checkpoint_dir, exclude_frozen_parameters) + + +def convert_zero_checkpoint_to_fp32_state_dict(checkpoint_dir, output_file, tag=None, exclude_frozen_parameters=False): + """ + Convert ZeRO 2 or 3 checkpoint into a single fp32 consolidated ``state_dict`` file that can be + loaded with ``torch.load(file)`` + ``load_state_dict()`` and used for training without DeepSpeed. + + Args: + - ``checkpoint_dir``: path to the desired checkpoint folder. (one that contains the tag-folder, like ``global_step14``) + - ``output_file``: path to the pytorch fp32 state_dict output file (e.g. path/pytorch_model.bin) + - ``tag``: checkpoint tag used as a unique identifier for checkpoint. If not provided will attempt to load tag in the file named ``latest`` in the checkpoint folder, e.g., ``global_step14`` + - ``exclude_frozen_parameters``: exclude frozen parameters + """ + + state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir, tag, exclude_frozen_parameters) + print(f"Saving fp32 state dict to {output_file}") + torch.save(state_dict, output_file) + + +def load_state_dict_from_zero_checkpoint(model, checkpoint_dir, tag=None): + """ + 1. Put the provided model to cpu + 2. Convert ZeRO 2 or 3 checkpoint into a single fp32 consolidated ``state_dict`` + 3. Load it into the provided model + + Args: + - ``model``: the model object to update + - ``checkpoint_dir``: path to the desired checkpoint folder. (one that contains the tag-folder, like ``global_step14``) + - ``tag``: checkpoint tag used as a unique identifier for checkpoint. If not provided will attempt to load tag in the file named ``latest`` in the checkpoint folder, e.g., ``global_step14`` + + Returns: + - ``model`: modified model + + Make sure you have plenty of CPU memory available before you call this function. If you don't + have enough use the ``zero_to_fp32.py`` utility to do the conversion. You will find it + conveniently placed for you in the checkpoint folder. + + A typical usage might be :: + + from deepspeed.utils.zero_to_fp32 import load_state_dict_from_zero_checkpoint + model = load_state_dict_from_zero_checkpoint(trainer.model, checkpoint_dir) + # submit to model hub or save the model to share with others + + Note, that once this was run, the ``model`` will no longer be usable in the deepspeed context + of the same application. i.e. you will need to re-initialize the deepspeed engine, since + ``model.load_state_dict(state_dict)`` will remove all the deepspeed magic from it. + + """ + logger.info(f"Extracting fp32 weights") + state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir, tag) + + logger.info(f"Overwriting model with fp32 weights") + model = model.cpu() + model.load_state_dict(state_dict, strict=False) + + return model + + +if __name__ == "__main__": + + parser = argparse.ArgumentParser() + parser.add_argument("checkpoint_dir", + type=str, + help="path to the desired checkpoint folder, e.g., path/checkpoint-12") + parser.add_argument( + "output_file", + type=str, + help="path to the pytorch fp32 state_dict output file (e.g. path/checkpoint-12/pytorch_model.bin)") + parser.add_argument("-t", + "--tag", + type=str, + default=None, + help="checkpoint tag used as a unique identifier for checkpoint. e.g., global_step1") + parser.add_argument("--exclude_frozen_parameters", action='store_true', help="exclude frozen parameters") + parser.add_argument("-d", "--debug", action='store_true', help="enable debug") + args = parser.parse_args() + + debug = args.debug + + convert_zero_checkpoint_to_fp32_state_dict(args.checkpoint_dir, + args.output_file, + tag=args.tag, + exclude_frozen_parameters=args.exclude_frozen_parameters) diff --git a/babylm_shuffle_nondeterministic_10M_seed0/runs/checkpoint-1200/config.json b/babylm_shuffle_nondeterministic_10M_seed0/runs/checkpoint-1200/config.json new file mode 100644 index 0000000000000000000000000000000000000000..fb0f9a53a507ffcdb60410deb71d60ef801bf350 --- /dev/null +++ b/babylm_shuffle_nondeterministic_10M_seed0/runs/checkpoint-1200/config.json @@ -0,0 +1,36 @@ +{ + "_name_or_path": "meta-llama/Llama-3.2-3B", + "architectures": [ + "LlamaForCausalLM" + ], + "attention_bias": false, + "attention_dropout": 0.0, + "bos_token_id": 128000, + "eos_token_id": 128001, + "head_dim": 128, + "hidden_act": "silu", + "hidden_size": 3072, + "initializer_range": 0.02, + "intermediate_size": 8192, + "max_position_embeddings": 131072, + "mlp_bias": false, + "model_type": "llama", + "num_attention_heads": 24, + "num_hidden_layers": 28, + "num_key_value_heads": 8, + "pretraining_tp": 1, + "rms_norm_eps": 1e-05, + "rope_scaling": { + "factor": 32.0, + "high_freq_factor": 4.0, + "low_freq_factor": 1.0, + "original_max_position_embeddings": 8192, + "rope_type": "llama3" + }, + "rope_theta": 500000.0, + "tie_word_embeddings": true, + "torch_dtype": "float16", + "transformers_version": "4.45.1", + "use_cache": true, + "vocab_size": 128256 +} diff --git a/babylm_shuffle_nondeterministic_10M_seed0/runs/checkpoint-1200/generation_config.json b/babylm_shuffle_nondeterministic_10M_seed0/runs/checkpoint-1200/generation_config.json new file mode 100644 index 0000000000000000000000000000000000000000..9c389b841a9a29ddd904e02d1eb0e08dcce82ad9 --- /dev/null +++ b/babylm_shuffle_nondeterministic_10M_seed0/runs/checkpoint-1200/generation_config.json @@ -0,0 +1,9 @@ +{ + "_from_model_config": true, + "bos_token_id": 128000, + "do_sample": true, + "eos_token_id": 128001, + "temperature": 0.6, + "top_p": 0.9, + "transformers_version": "4.45.1" +} diff --git a/babylm_shuffle_nondeterministic_10M_seed0/runs/checkpoint-1200/latest b/babylm_shuffle_nondeterministic_10M_seed0/runs/checkpoint-1200/latest new file mode 100644 index 0000000000000000000000000000000000000000..873b08355ae55e303606c474ab5fba25f9e8262a --- /dev/null +++ b/babylm_shuffle_nondeterministic_10M_seed0/runs/checkpoint-1200/latest @@ -0,0 +1 @@ +global_step1200 \ No newline at end of file diff --git a/babylm_shuffle_nondeterministic_10M_seed0/runs/checkpoint-1200/model.safetensors.index.json b/babylm_shuffle_nondeterministic_10M_seed0/runs/checkpoint-1200/model.safetensors.index.json new file mode 100644 index 0000000000000000000000000000000000000000..ed64de846d720b9a7859dc20575fea8e8ca51940 --- /dev/null +++ b/babylm_shuffle_nondeterministic_10M_seed0/runs/checkpoint-1200/model.safetensors.index.json @@ -0,0 +1,262 @@ +{ + "metadata": { + "total_size": 7213504512 + }, + "weight_map": { + "lm_head.weight": "model-00002-of-00002.safetensors", + "model.embed_tokens.weight": "model-00001-of-00002.safetensors", + "model.layers.0.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.0.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.0.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.0.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.0.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.0.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.0.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.0.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.0.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.1.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.1.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.1.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.1.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.1.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.1.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.1.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.1.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.1.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.10.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.10.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.10.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.10.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.10.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.10.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.10.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.10.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.10.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.11.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.11.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.11.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.11.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.11.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.11.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.11.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.11.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.11.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.12.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.12.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.12.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.12.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.12.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.12.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.12.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.12.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.12.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.13.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.13.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.13.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.13.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.13.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.13.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.13.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.13.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.13.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.14.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.14.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.14.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.14.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.14.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.14.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.14.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.14.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.14.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.15.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.15.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.15.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.15.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.15.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.15.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.15.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.15.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.15.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.16.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.16.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.16.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.16.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.16.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.16.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.16.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.16.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.16.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.17.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.17.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.17.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.17.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.17.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.17.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.17.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.17.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.17.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.18.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.18.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.18.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.18.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.18.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.18.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.18.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.18.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.18.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.19.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.19.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.19.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.19.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.19.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.19.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.19.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.19.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.19.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.2.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.2.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.2.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.2.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.2.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.2.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.2.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.2.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.2.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.20.input_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.20.mlp.down_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.20.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.20.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.20.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.20.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.20.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.20.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.20.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.21.input_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.21.mlp.down_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.21.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.21.mlp.up_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.21.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.21.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.21.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.21.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.21.self_attn.v_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.22.input_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.22.mlp.down_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.22.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.22.mlp.up_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.22.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.22.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.22.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.22.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.22.self_attn.v_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.23.input_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.23.mlp.down_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.23.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.23.mlp.up_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.23.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.23.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.23.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.23.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.23.self_attn.v_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.24.input_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.24.mlp.down_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.24.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.24.mlp.up_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.24.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.24.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.24.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.24.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.24.self_attn.v_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.25.input_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.25.mlp.down_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.25.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.25.mlp.up_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.25.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.25.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.25.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.25.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.25.self_attn.v_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.26.input_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.26.mlp.down_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.26.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.26.mlp.up_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.26.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.26.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.26.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.26.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.26.self_attn.v_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.27.input_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.27.mlp.down_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.27.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.27.mlp.up_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.27.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.27.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.27.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.27.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.27.self_attn.v_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.3.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.3.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.3.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.3.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.3.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.3.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.3.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.3.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.3.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.4.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.4.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.4.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.4.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.4.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.4.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.4.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.4.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.4.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.5.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.5.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.5.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.5.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.5.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.5.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.5.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.5.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.5.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.6.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.6.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.6.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.6.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.6.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.6.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.6.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.6.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.6.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.7.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.7.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.7.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.7.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.7.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.7.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.7.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.7.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.7.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.8.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.8.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.8.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.8.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.8.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.8.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.8.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.8.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.8.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.9.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.9.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.9.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.9.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.9.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.9.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.9.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.9.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.9.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.norm.weight": "model-00002-of-00002.safetensors" + } +} diff --git a/babylm_shuffle_nondeterministic_10M_seed0/runs/checkpoint-1200/special_tokens_map.json b/babylm_shuffle_nondeterministic_10M_seed0/runs/checkpoint-1200/special_tokens_map.json new file mode 100644 index 0000000000000000000000000000000000000000..6949c5975ee0e961ef61cf31010dce04df0a03f8 --- /dev/null +++ b/babylm_shuffle_nondeterministic_10M_seed0/runs/checkpoint-1200/special_tokens_map.json @@ -0,0 +1,23 @@ +{ + "bos_token": { + "content": "<|begin_of_text|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false + }, + "eos_token": { + "content": "<|end_of_text|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false + }, + "pad_token": { + "content": "[PAD]", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false + } +} diff --git a/babylm_shuffle_nondeterministic_10M_seed0/runs/checkpoint-1200/tokenizer_config.json b/babylm_shuffle_nondeterministic_10M_seed0/runs/checkpoint-1200/tokenizer_config.json new file mode 100644 index 0000000000000000000000000000000000000000..5684b14e2756b56aa397c1324e239b442162b34a --- /dev/null +++ b/babylm_shuffle_nondeterministic_10M_seed0/runs/checkpoint-1200/tokenizer_config.json @@ -0,0 +1,2070 @@ +{ + "added_tokens_decoder": { + "128000": { + "content": "<|begin_of_text|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128001": { + "content": "<|end_of_text|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128002": { + "content": "<|reserved_special_token_0|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128003": { + "content": "<|reserved_special_token_1|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128004": { + "content": "<|finetune_right_pad_id|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128005": { + "content": "<|reserved_special_token_2|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128006": { + "content": "<|start_header_id|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128007": { + "content": "<|end_header_id|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128008": { + "content": "<|eom_id|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128009": { + "content": "<|eot_id|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128010": { + "content": "<|python_tag|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128011": { + "content": "<|reserved_special_token_3|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128012": { + "content": "<|reserved_special_token_4|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128013": { + "content": "<|reserved_special_token_5|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128014": { + "content": "<|reserved_special_token_6|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128015": { + "content": "<|reserved_special_token_7|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128016": { + "content": "<|reserved_special_token_8|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128017": { + "content": "<|reserved_special_token_9|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128018": { + "content": "<|reserved_special_token_10|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128019": { + "content": "<|reserved_special_token_11|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128020": { + "content": "<|reserved_special_token_12|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128021": { + "content": "<|reserved_special_token_13|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128022": { + "content": "<|reserved_special_token_14|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128023": { + "content": "<|reserved_special_token_15|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128024": { + "content": "<|reserved_special_token_16|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128025": { + "content": "<|reserved_special_token_17|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128026": { + "content": "<|reserved_special_token_18|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128027": { + "content": "<|reserved_special_token_19|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128028": { + "content": "<|reserved_special_token_20|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128029": { + "content": "<|reserved_special_token_21|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128030": { + "content": "<|reserved_special_token_22|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128031": { + "content": "<|reserved_special_token_23|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128032": { + "content": "<|reserved_special_token_24|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128033": { + "content": "<|reserved_special_token_25|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128034": { + "content": "<|reserved_special_token_26|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128035": { + "content": "<|reserved_special_token_27|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128036": { + "content": "<|reserved_special_token_28|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128037": { + "content": "<|reserved_special_token_29|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128038": { + "content": "<|reserved_special_token_30|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128039": { + "content": "<|reserved_special_token_31|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128040": { + "content": "<|reserved_special_token_32|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128041": { + "content": "<|reserved_special_token_33|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128042": { + "content": "<|reserved_special_token_34|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128043": { + "content": "<|reserved_special_token_35|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128044": { + "content": "<|reserved_special_token_36|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128045": { + "content": "<|reserved_special_token_37|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128046": { + "content": "<|reserved_special_token_38|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128047": { + "content": "<|reserved_special_token_39|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128048": { + "content": "<|reserved_special_token_40|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128049": { + "content": "<|reserved_special_token_41|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128050": { + "content": "<|reserved_special_token_42|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128051": { + "content": "<|reserved_special_token_43|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128052": { + "content": "<|reserved_special_token_44|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128053": { + "content": "<|reserved_special_token_45|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128054": { + "content": "<|reserved_special_token_46|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128055": { + "content": "<|reserved_special_token_47|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128056": { + "content": "<|reserved_special_token_48|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128057": { + "content": "<|reserved_special_token_49|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128058": { + "content": "<|reserved_special_token_50|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128059": { + "content": "<|reserved_special_token_51|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128060": { + "content": "<|reserved_special_token_52|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128061": { + "content": "<|reserved_special_token_53|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128062": { + "content": "<|reserved_special_token_54|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128063": { + "content": "<|reserved_special_token_55|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128064": { + "content": "<|reserved_special_token_56|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128065": { + "content": "<|reserved_special_token_57|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128066": { + "content": "<|reserved_special_token_58|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128067": { + "content": "<|reserved_special_token_59|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128068": { + "content": "<|reserved_special_token_60|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128069": { + "content": "<|reserved_special_token_61|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128070": { + "content": "<|reserved_special_token_62|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128071": { + "content": "<|reserved_special_token_63|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128072": { + "content": "<|reserved_special_token_64|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128073": { + "content": "<|reserved_special_token_65|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128074": { + "content": "<|reserved_special_token_66|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128075": { + "content": "<|reserved_special_token_67|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128076": { + "content": "<|reserved_special_token_68|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128077": { + "content": "<|reserved_special_token_69|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128078": { + "content": "<|reserved_special_token_70|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128079": { + "content": "<|reserved_special_token_71|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128080": { + "content": "<|reserved_special_token_72|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128081": { + "content": "<|reserved_special_token_73|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128082": { + "content": "<|reserved_special_token_74|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128083": { + "content": "<|reserved_special_token_75|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128084": { + "content": "<|reserved_special_token_76|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128085": { + "content": "<|reserved_special_token_77|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128086": { + "content": "<|reserved_special_token_78|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128087": { + "content": "<|reserved_special_token_79|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128088": { + "content": "<|reserved_special_token_80|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128089": { + "content": "<|reserved_special_token_81|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128090": { + "content": "<|reserved_special_token_82|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128091": { + "content": "<|reserved_special_token_83|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128092": { + "content": "<|reserved_special_token_84|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128093": { + "content": "<|reserved_special_token_85|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128094": { + "content": "<|reserved_special_token_86|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128095": { + "content": "<|reserved_special_token_87|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128096": { + "content": "<|reserved_special_token_88|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128097": { + "content": "<|reserved_special_token_89|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128098": { + "content": "<|reserved_special_token_90|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128099": { + "content": "<|reserved_special_token_91|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128100": { + "content": "<|reserved_special_token_92|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128101": { + "content": "<|reserved_special_token_93|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128102": { + "content": "<|reserved_special_token_94|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128103": { + "content": "<|reserved_special_token_95|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128104": { + "content": "<|reserved_special_token_96|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128105": { + "content": "<|reserved_special_token_97|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128106": { + "content": "<|reserved_special_token_98|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128107": { + "content": "<|reserved_special_token_99|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128108": { + "content": "<|reserved_special_token_100|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128109": { + "content": "<|reserved_special_token_101|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128110": { + "content": "<|reserved_special_token_102|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128111": { + "content": "<|reserved_special_token_103|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128112": { + "content": "<|reserved_special_token_104|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128113": { + "content": "<|reserved_special_token_105|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128114": { + "content": "<|reserved_special_token_106|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128115": { + "content": "<|reserved_special_token_107|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128116": { + "content": "<|reserved_special_token_108|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128117": { + "content": "<|reserved_special_token_109|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128118": { + "content": "<|reserved_special_token_110|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128119": { + "content": "<|reserved_special_token_111|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128120": { + "content": "<|reserved_special_token_112|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128121": { + "content": "<|reserved_special_token_113|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128122": { + "content": "<|reserved_special_token_114|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128123": { + "content": "<|reserved_special_token_115|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128124": { + "content": "<|reserved_special_token_116|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128125": { + "content": "<|reserved_special_token_117|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128126": { + "content": "<|reserved_special_token_118|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128127": { + "content": "<|reserved_special_token_119|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128128": { + "content": "<|reserved_special_token_120|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128129": { + "content": "<|reserved_special_token_121|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128130": { + "content": "<|reserved_special_token_122|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128131": { + "content": "<|reserved_special_token_123|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128132": { + "content": "<|reserved_special_token_124|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128133": { + "content": "<|reserved_special_token_125|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128134": { + "content": "<|reserved_special_token_126|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128135": { + "content": "<|reserved_special_token_127|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128136": { + "content": "<|reserved_special_token_128|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128137": { + "content": "<|reserved_special_token_129|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128138": { + "content": "<|reserved_special_token_130|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128139": { + "content": "<|reserved_special_token_131|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128140": { + "content": "<|reserved_special_token_132|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128141": { + "content": "<|reserved_special_token_133|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128142": { + "content": "<|reserved_special_token_134|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128143": { + "content": "<|reserved_special_token_135|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128144": { + "content": "<|reserved_special_token_136|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128145": { + "content": "<|reserved_special_token_137|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128146": { + "content": "<|reserved_special_token_138|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128147": { + "content": "<|reserved_special_token_139|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128148": { + "content": "<|reserved_special_token_140|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128149": { + "content": "<|reserved_special_token_141|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128150": { + "content": "<|reserved_special_token_142|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128151": { + "content": "<|reserved_special_token_143|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128152": { + "content": "<|reserved_special_token_144|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128153": { + "content": "<|reserved_special_token_145|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128154": { + "content": "<|reserved_special_token_146|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128155": { + "content": "<|reserved_special_token_147|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128156": { + "content": "<|reserved_special_token_148|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128157": { + "content": "<|reserved_special_token_149|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128158": { + "content": "<|reserved_special_token_150|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128159": { + "content": "<|reserved_special_token_151|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128160": { + "content": "<|reserved_special_token_152|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128161": { + "content": "<|reserved_special_token_153|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128162": { + "content": "<|reserved_special_token_154|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128163": { + "content": "<|reserved_special_token_155|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128164": { + "content": "<|reserved_special_token_156|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128165": { + "content": "<|reserved_special_token_157|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128166": { + "content": "<|reserved_special_token_158|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128167": { + "content": "<|reserved_special_token_159|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128168": { + "content": "<|reserved_special_token_160|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128169": { + "content": "<|reserved_special_token_161|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128170": { + "content": "<|reserved_special_token_162|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128171": { + "content": "<|reserved_special_token_163|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128172": { + "content": "<|reserved_special_token_164|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128173": { + "content": "<|reserved_special_token_165|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128174": { + "content": "<|reserved_special_token_166|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128175": { + "content": "<|reserved_special_token_167|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128176": { + "content": "<|reserved_special_token_168|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128177": { + "content": "<|reserved_special_token_169|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128178": { + "content": "<|reserved_special_token_170|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128179": { + "content": "<|reserved_special_token_171|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128180": { + "content": "<|reserved_special_token_172|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128181": { + "content": "<|reserved_special_token_173|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128182": { + "content": "<|reserved_special_token_174|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128183": { + "content": "<|reserved_special_token_175|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128184": { + "content": "<|reserved_special_token_176|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128185": { + "content": "<|reserved_special_token_177|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128186": { + "content": "<|reserved_special_token_178|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128187": { + "content": "<|reserved_special_token_179|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128188": { + "content": "<|reserved_special_token_180|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128189": { + "content": "<|reserved_special_token_181|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128190": { + "content": "<|reserved_special_token_182|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128191": { + "content": "<|reserved_special_token_183|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128192": { + "content": "<|reserved_special_token_184|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128193": { + "content": "<|reserved_special_token_185|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128194": { + "content": "<|reserved_special_token_186|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128195": { + "content": "<|reserved_special_token_187|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128196": { + "content": "<|reserved_special_token_188|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128197": { + "content": "<|reserved_special_token_189|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128198": { + "content": "<|reserved_special_token_190|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128199": { + "content": "<|reserved_special_token_191|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128200": { + "content": "<|reserved_special_token_192|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128201": { + "content": "<|reserved_special_token_193|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128202": { + "content": "<|reserved_special_token_194|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128203": { + "content": "<|reserved_special_token_195|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128204": { + "content": "<|reserved_special_token_196|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128205": { + "content": "<|reserved_special_token_197|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128206": { + "content": "<|reserved_special_token_198|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128207": { + "content": "<|reserved_special_token_199|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128208": { + "content": "<|reserved_special_token_200|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128209": { + "content": "<|reserved_special_token_201|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128210": { + "content": "<|reserved_special_token_202|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128211": { + "content": "<|reserved_special_token_203|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128212": { + "content": "<|reserved_special_token_204|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128213": { + "content": "<|reserved_special_token_205|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128214": { + "content": "<|reserved_special_token_206|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128215": { + "content": "<|reserved_special_token_207|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128216": { + "content": "<|reserved_special_token_208|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128217": { + "content": "<|reserved_special_token_209|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128218": { + "content": "<|reserved_special_token_210|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128219": { + "content": "<|reserved_special_token_211|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128220": { + "content": "<|reserved_special_token_212|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128221": { + "content": "<|reserved_special_token_213|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128222": { + "content": "<|reserved_special_token_214|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128223": { + "content": "<|reserved_special_token_215|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128224": { + "content": "<|reserved_special_token_216|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128225": { + "content": "<|reserved_special_token_217|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128226": { + "content": "<|reserved_special_token_218|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128227": { + "content": "<|reserved_special_token_219|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128228": { + "content": "<|reserved_special_token_220|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128229": { + "content": "<|reserved_special_token_221|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128230": { + "content": "<|reserved_special_token_222|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128231": { + "content": "<|reserved_special_token_223|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128232": { + "content": "<|reserved_special_token_224|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128233": { + "content": "<|reserved_special_token_225|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128234": { + "content": "<|reserved_special_token_226|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128235": { + "content": "<|reserved_special_token_227|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128236": { + "content": "<|reserved_special_token_228|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128237": { + "content": "<|reserved_special_token_229|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128238": { + "content": "<|reserved_special_token_230|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128239": { + "content": "<|reserved_special_token_231|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128240": { + "content": "<|reserved_special_token_232|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128241": { + "content": "<|reserved_special_token_233|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128242": { + "content": "<|reserved_special_token_234|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128243": { + "content": "<|reserved_special_token_235|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128244": { + "content": "<|reserved_special_token_236|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128245": { + "content": "<|reserved_special_token_237|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128246": { + "content": "<|reserved_special_token_238|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128247": { + "content": "<|reserved_special_token_239|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128248": { + "content": "<|reserved_special_token_240|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128249": { + "content": "<|reserved_special_token_241|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128250": { + "content": "<|reserved_special_token_242|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128251": { + "content": "<|reserved_special_token_243|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128252": { + "content": "<|reserved_special_token_244|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128253": { + "content": "<|reserved_special_token_245|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128254": { + "content": "<|reserved_special_token_246|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128255": { + "content": "<|reserved_special_token_247|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128256": { + "content": "[PAD]", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + } + }, + "bos_token": "<|begin_of_text|>", + "clean_up_tokenization_spaces": true, + "eos_token": "<|end_of_text|>", + "model_input_names": [ + "input_ids", + "attention_mask" + ], + "model_max_length": 131072, + "pad_token": "[PAD]", + "tokenizer_class": "PreTrainedTokenizerFast" +} diff --git a/babylm_shuffle_nondeterministic_10M_seed0/runs/checkpoint-1200/trainer_state.json b/babylm_shuffle_nondeterministic_10M_seed0/runs/checkpoint-1200/trainer_state.json new file mode 100644 index 0000000000000000000000000000000000000000..26c80cd0638c1ff7ae6707e59769b4c7e921c746 --- /dev/null +++ b/babylm_shuffle_nondeterministic_10M_seed0/runs/checkpoint-1200/trainer_state.json @@ -0,0 +1,9393 @@ +{ + "best_metric": null, + "best_model_checkpoint": null, + "epoch": 1.3150684931506849, + "eval_steps": 10, + "global_step": 1200, + "is_hyper_param_search": false, + "is_local_process_zero": true, + "is_world_process_zero": true, + "log_history": [ + { + "epoch": 0.001095890410958904, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.0998, + "step": 1 + }, + { + "epoch": 0.002191780821917808, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.0858, + "step": 2 + }, + { + "epoch": 0.0032876712328767125, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.0672, + "step": 3 + }, + { + "epoch": 0.004383561643835616, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.0802, + "step": 4 + }, + { + "epoch": 0.005479452054794521, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.0636, + "step": 5 + }, + { + "epoch": 0.006575342465753425, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.1672, + "step": 6 + }, + { + "epoch": 0.007671232876712329, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.0961, + "step": 7 + }, + { + "epoch": 0.008767123287671232, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.1258, + "step": 8 + }, + { + "epoch": 0.009863013698630137, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.0953, + "step": 9 + }, + { + "epoch": 0.010958904109589041, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.1404, + "step": 10 + }, + { + "epoch": 0.010958904109589041, + "eval_loss": 3.12593412399292, + "eval_runtime": 42.732, + "eval_samples_per_second": 23.402, + "eval_steps_per_second": 0.983, + "step": 10 + }, + { + "epoch": 0.012054794520547946, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.1187, + "step": 11 + }, + { + "epoch": 0.01315068493150685, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.0593, + "step": 12 + }, + { + "epoch": 0.014246575342465753, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.111, + "step": 13 + }, + { + "epoch": 0.015342465753424657, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.0974, + "step": 14 + }, + { + "epoch": 0.01643835616438356, + "grad_norm": 5.991453170776367, + "learning_rate": 1.8248175182481753e-08, + "loss": 3.135, + "step": 15 + }, + { + "epoch": 0.017534246575342465, + "grad_norm": 6.274210453033447, + "learning_rate": 3.6496350364963505e-08, + "loss": 3.083, + "step": 16 + }, + { + "epoch": 0.01863013698630137, + "grad_norm": 6.644786834716797, + "learning_rate": 5.4744525547445264e-08, + "loss": 3.0856, + "step": 17 + }, + { + "epoch": 0.019726027397260273, + "grad_norm": 6.769156455993652, + "learning_rate": 7.299270072992701e-08, + "loss": 3.1082, + "step": 18 + }, + { + "epoch": 0.020821917808219178, + "grad_norm": 6.771246910095215, + "learning_rate": 9.124087591240876e-08, + "loss": 3.0753, + "step": 19 + }, + { + "epoch": 0.021917808219178082, + "grad_norm": 6.468441486358643, + "learning_rate": 1.0948905109489053e-07, + "loss": 3.1478, + "step": 20 + }, + { + "epoch": 0.021917808219178082, + "eval_loss": 3.1245107650756836, + "eval_runtime": 43.6251, + "eval_samples_per_second": 22.923, + "eval_steps_per_second": 0.963, + "step": 20 + }, + { + "epoch": 0.023013698630136987, + "grad_norm": 6.83078670501709, + "learning_rate": 1.2773722627737227e-07, + "loss": 3.1213, + "step": 21 + }, + { + "epoch": 0.02410958904109589, + "grad_norm": 6.221287250518799, + "learning_rate": 1.4598540145985402e-07, + "loss": 3.0549, + "step": 22 + }, + { + "epoch": 0.025205479452054796, + "grad_norm": 5.564266204833984, + "learning_rate": 1.642335766423358e-07, + "loss": 3.0769, + "step": 23 + }, + { + "epoch": 0.0263013698630137, + "grad_norm": 5.735209941864014, + "learning_rate": 1.8248175182481753e-07, + "loss": 3.1124, + "step": 24 + }, + { + "epoch": 0.0273972602739726, + "grad_norm": 5.577925682067871, + "learning_rate": 2.0072992700729928e-07, + "loss": 3.1191, + "step": 25 + }, + { + "epoch": 0.028493150684931506, + "grad_norm": 4.814139366149902, + "learning_rate": 2.1897810218978106e-07, + "loss": 3.0947, + "step": 26 + }, + { + "epoch": 0.02958904109589041, + "grad_norm": 4.69775390625, + "learning_rate": 2.3722627737226278e-07, + "loss": 3.0997, + "step": 27 + }, + { + "epoch": 0.030684931506849315, + "grad_norm": 4.811586380004883, + "learning_rate": 2.5547445255474454e-07, + "loss": 3.0749, + "step": 28 + }, + { + "epoch": 0.03178082191780822, + "grad_norm": 4.623377323150635, + "learning_rate": 2.737226277372263e-07, + "loss": 3.0619, + "step": 29 + }, + { + "epoch": 0.03287671232876712, + "grad_norm": 4.721253395080566, + "learning_rate": 2.9197080291970804e-07, + "loss": 3.0371, + "step": 30 + }, + { + "epoch": 0.03287671232876712, + "eval_loss": 3.082857847213745, + "eval_runtime": 44.127, + "eval_samples_per_second": 22.662, + "eval_steps_per_second": 0.952, + "step": 30 + }, + { + "epoch": 0.03397260273972603, + "grad_norm": 3.968630790710449, + "learning_rate": 3.102189781021898e-07, + "loss": 3.0535, + "step": 31 + }, + { + "epoch": 0.03506849315068493, + "grad_norm": 4.4373779296875, + "learning_rate": 3.284671532846716e-07, + "loss": 3.0476, + "step": 32 + }, + { + "epoch": 0.03616438356164384, + "grad_norm": 4.479320049285889, + "learning_rate": 3.467153284671533e-07, + "loss": 3.0215, + "step": 33 + }, + { + "epoch": 0.03726027397260274, + "grad_norm": 4.122231960296631, + "learning_rate": 3.6496350364963505e-07, + "loss": 3.0106, + "step": 34 + }, + { + "epoch": 0.038356164383561646, + "grad_norm": 4.153995513916016, + "learning_rate": 3.8321167883211686e-07, + "loss": 3.0318, + "step": 35 + }, + { + "epoch": 0.03945205479452055, + "grad_norm": 3.6949076652526855, + "learning_rate": 4.0145985401459856e-07, + "loss": 3.046, + "step": 36 + }, + { + "epoch": 0.040547945205479455, + "grad_norm": 3.4610488414764404, + "learning_rate": 4.197080291970803e-07, + "loss": 3.002, + "step": 37 + }, + { + "epoch": 0.041643835616438356, + "grad_norm": 3.9523138999938965, + "learning_rate": 4.379562043795621e-07, + "loss": 2.9665, + "step": 38 + }, + { + "epoch": 0.042739726027397264, + "grad_norm": 3.68516206741333, + "learning_rate": 4.562043795620438e-07, + "loss": 3.0411, + "step": 39 + }, + { + "epoch": 0.043835616438356165, + "grad_norm": 3.692396402359009, + "learning_rate": 4.7445255474452557e-07, + "loss": 2.9666, + "step": 40 + }, + { + "epoch": 0.043835616438356165, + "eval_loss": 3.015512704849243, + "eval_runtime": 48.2438, + "eval_samples_per_second": 20.728, + "eval_steps_per_second": 0.871, + "step": 40 + }, + { + "epoch": 0.044931506849315066, + "grad_norm": 3.3765921592712402, + "learning_rate": 4.927007299270073e-07, + "loss": 2.9831, + "step": 41 + }, + { + "epoch": 0.046027397260273974, + "grad_norm": 2.9631125926971436, + "learning_rate": 5.109489051094891e-07, + "loss": 3.0126, + "step": 42 + }, + { + "epoch": 0.047123287671232875, + "grad_norm": 2.969730854034424, + "learning_rate": 5.291970802919708e-07, + "loss": 2.9803, + "step": 43 + }, + { + "epoch": 0.04821917808219178, + "grad_norm": 3.185237407684326, + "learning_rate": 5.474452554744526e-07, + "loss": 2.9888, + "step": 44 + }, + { + "epoch": 0.049315068493150684, + "grad_norm": 3.055579423904419, + "learning_rate": 5.656934306569343e-07, + "loss": 2.9715, + "step": 45 + }, + { + "epoch": 0.05041095890410959, + "grad_norm": 3.083284616470337, + "learning_rate": 5.839416058394161e-07, + "loss": 2.9293, + "step": 46 + }, + { + "epoch": 0.05150684931506849, + "grad_norm": 2.984781503677368, + "learning_rate": 6.021897810218979e-07, + "loss": 2.9228, + "step": 47 + }, + { + "epoch": 0.0526027397260274, + "grad_norm": 2.756173849105835, + "learning_rate": 6.204379562043796e-07, + "loss": 2.9406, + "step": 48 + }, + { + "epoch": 0.0536986301369863, + "grad_norm": 2.911132574081421, + "learning_rate": 6.386861313868614e-07, + "loss": 2.8829, + "step": 49 + }, + { + "epoch": 0.0547945205479452, + "grad_norm": 2.592890977859497, + "learning_rate": 6.569343065693432e-07, + "loss": 2.9839, + "step": 50 + }, + { + "epoch": 0.0547945205479452, + "eval_loss": 2.9409468173980713, + "eval_runtime": 47.1045, + "eval_samples_per_second": 21.229, + "eval_steps_per_second": 0.892, + "step": 50 + }, + { + "epoch": 0.05589041095890411, + "grad_norm": 3.0112955570220947, + "learning_rate": 6.751824817518248e-07, + "loss": 2.8847, + "step": 51 + }, + { + "epoch": 0.05698630136986301, + "grad_norm": 2.3701798915863037, + "learning_rate": 6.934306569343066e-07, + "loss": 2.9109, + "step": 52 + }, + { + "epoch": 0.05808219178082192, + "grad_norm": 2.268850088119507, + "learning_rate": 7.116788321167883e-07, + "loss": 2.8483, + "step": 53 + }, + { + "epoch": 0.05917808219178082, + "grad_norm": 2.419015645980835, + "learning_rate": 7.299270072992701e-07, + "loss": 2.8491, + "step": 54 + }, + { + "epoch": 0.06027397260273973, + "grad_norm": 2.4582653045654297, + "learning_rate": 7.48175182481752e-07, + "loss": 2.8622, + "step": 55 + }, + { + "epoch": 0.06136986301369863, + "grad_norm": 2.3471553325653076, + "learning_rate": 7.664233576642337e-07, + "loss": 2.8738, + "step": 56 + }, + { + "epoch": 0.06246575342465754, + "grad_norm": 2.479177236557007, + "learning_rate": 7.846715328467154e-07, + "loss": 2.8843, + "step": 57 + }, + { + "epoch": 0.06356164383561644, + "grad_norm": 2.1676292419433594, + "learning_rate": 8.029197080291971e-07, + "loss": 2.8674, + "step": 58 + }, + { + "epoch": 0.06465753424657535, + "grad_norm": 1.9862910509109497, + "learning_rate": 8.211678832116789e-07, + "loss": 2.8197, + "step": 59 + }, + { + "epoch": 0.06575342465753424, + "grad_norm": 2.211313486099243, + "learning_rate": 8.394160583941606e-07, + "loss": 2.7961, + "step": 60 + }, + { + "epoch": 0.06575342465753424, + "eval_loss": 2.880824327468872, + "eval_runtime": 51.3235, + "eval_samples_per_second": 19.484, + "eval_steps_per_second": 0.818, + "step": 60 + }, + { + "epoch": 0.06684931506849315, + "grad_norm": 2.3497281074523926, + "learning_rate": 8.576642335766425e-07, + "loss": 2.8233, + "step": 61 + }, + { + "epoch": 0.06794520547945206, + "grad_norm": 2.158515214920044, + "learning_rate": 8.759124087591242e-07, + "loss": 2.8136, + "step": 62 + }, + { + "epoch": 0.06904109589041096, + "grad_norm": 2.0219955444335938, + "learning_rate": 8.94160583941606e-07, + "loss": 2.8231, + "step": 63 + }, + { + "epoch": 0.07013698630136986, + "grad_norm": 1.8989133834838867, + "learning_rate": 9.124087591240876e-07, + "loss": 2.8268, + "step": 64 + }, + { + "epoch": 0.07123287671232877, + "grad_norm": 2.243851661682129, + "learning_rate": 9.306569343065694e-07, + "loss": 2.7886, + "step": 65 + }, + { + "epoch": 0.07232876712328767, + "grad_norm": 2.122464418411255, + "learning_rate": 9.489051094890511e-07, + "loss": 2.8166, + "step": 66 + }, + { + "epoch": 0.07342465753424658, + "grad_norm": 1.899857997894287, + "learning_rate": 9.67153284671533e-07, + "loss": 2.7997, + "step": 67 + }, + { + "epoch": 0.07452054794520548, + "grad_norm": 2.4424173831939697, + "learning_rate": 9.854014598540146e-07, + "loss": 2.8342, + "step": 68 + }, + { + "epoch": 0.07561643835616438, + "grad_norm": 2.5677411556243896, + "learning_rate": 1.0036496350364965e-06, + "loss": 2.7786, + "step": 69 + }, + { + "epoch": 0.07671232876712329, + "grad_norm": 1.9989523887634277, + "learning_rate": 1.0218978102189781e-06, + "loss": 2.7411, + "step": 70 + }, + { + "epoch": 0.07671232876712329, + "eval_loss": 2.8185322284698486, + "eval_runtime": 50.4998, + "eval_samples_per_second": 19.802, + "eval_steps_per_second": 0.832, + "step": 70 + }, + { + "epoch": 0.07780821917808219, + "grad_norm": 1.9376521110534668, + "learning_rate": 1.04014598540146e-06, + "loss": 2.7988, + "step": 71 + }, + { + "epoch": 0.0789041095890411, + "grad_norm": 2.5629725456237793, + "learning_rate": 1.0583941605839416e-06, + "loss": 2.7079, + "step": 72 + }, + { + "epoch": 0.08, + "grad_norm": 2.1634678840637207, + "learning_rate": 1.0766423357664233e-06, + "loss": 2.7049, + "step": 73 + }, + { + "epoch": 0.08109589041095891, + "grad_norm": 2.2143008708953857, + "learning_rate": 1.0948905109489052e-06, + "loss": 2.72, + "step": 74 + }, + { + "epoch": 0.0821917808219178, + "grad_norm": 2.2701101303100586, + "learning_rate": 1.113138686131387e-06, + "loss": 2.7418, + "step": 75 + }, + { + "epoch": 0.08328767123287671, + "grad_norm": 1.716569423675537, + "learning_rate": 1.1313868613138687e-06, + "loss": 2.7558, + "step": 76 + }, + { + "epoch": 0.08438356164383562, + "grad_norm": 2.79757022857666, + "learning_rate": 1.1496350364963505e-06, + "loss": 2.6865, + "step": 77 + }, + { + "epoch": 0.08547945205479453, + "grad_norm": 2.2189838886260986, + "learning_rate": 1.1678832116788322e-06, + "loss": 2.6684, + "step": 78 + }, + { + "epoch": 0.08657534246575342, + "grad_norm": 2.0695579051971436, + "learning_rate": 1.1861313868613138e-06, + "loss": 2.6779, + "step": 79 + }, + { + "epoch": 0.08767123287671233, + "grad_norm": 2.372326374053955, + "learning_rate": 1.2043795620437959e-06, + "loss": 2.6711, + "step": 80 + }, + { + "epoch": 0.08767123287671233, + "eval_loss": 2.7537453174591064, + "eval_runtime": 47.7274, + "eval_samples_per_second": 20.952, + "eval_steps_per_second": 0.88, + "step": 80 + }, + { + "epoch": 0.08876712328767124, + "grad_norm": 2.3398947715759277, + "learning_rate": 1.2226277372262775e-06, + "loss": 2.6776, + "step": 81 + }, + { + "epoch": 0.08986301369863013, + "grad_norm": 2.6619036197662354, + "learning_rate": 1.2408759124087592e-06, + "loss": 2.7077, + "step": 82 + }, + { + "epoch": 0.09095890410958904, + "grad_norm": 1.8245115280151367, + "learning_rate": 1.259124087591241e-06, + "loss": 2.662, + "step": 83 + }, + { + "epoch": 0.09205479452054795, + "grad_norm": 2.2627148628234863, + "learning_rate": 1.2773722627737229e-06, + "loss": 2.677, + "step": 84 + }, + { + "epoch": 0.09315068493150686, + "grad_norm": 2.4983835220336914, + "learning_rate": 1.2956204379562045e-06, + "loss": 2.6601, + "step": 85 + }, + { + "epoch": 0.09424657534246575, + "grad_norm": 2.147010087966919, + "learning_rate": 1.3138686131386864e-06, + "loss": 2.6213, + "step": 86 + }, + { + "epoch": 0.09534246575342466, + "grad_norm": 2.791682481765747, + "learning_rate": 1.332116788321168e-06, + "loss": 2.6281, + "step": 87 + }, + { + "epoch": 0.09643835616438357, + "grad_norm": 2.023970365524292, + "learning_rate": 1.3503649635036497e-06, + "loss": 2.5766, + "step": 88 + }, + { + "epoch": 0.09753424657534246, + "grad_norm": 2.8143956661224365, + "learning_rate": 1.3686131386861315e-06, + "loss": 2.6086, + "step": 89 + }, + { + "epoch": 0.09863013698630137, + "grad_norm": 2.226106882095337, + "learning_rate": 1.3868613138686132e-06, + "loss": 2.5477, + "step": 90 + }, + { + "epoch": 0.09863013698630137, + "eval_loss": 2.6880786418914795, + "eval_runtime": 47.8417, + "eval_samples_per_second": 20.902, + "eval_steps_per_second": 0.878, + "step": 90 + }, + { + "epoch": 0.09972602739726028, + "grad_norm": 2.5248148441314697, + "learning_rate": 1.405109489051095e-06, + "loss": 2.6232, + "step": 91 + }, + { + "epoch": 0.10082191780821918, + "grad_norm": 2.421330690383911, + "learning_rate": 1.4233576642335767e-06, + "loss": 2.599, + "step": 92 + }, + { + "epoch": 0.10191780821917808, + "grad_norm": 2.5562429428100586, + "learning_rate": 1.4416058394160583e-06, + "loss": 2.6197, + "step": 93 + }, + { + "epoch": 0.10301369863013699, + "grad_norm": 2.856201171875, + "learning_rate": 1.4598540145985402e-06, + "loss": 2.5825, + "step": 94 + }, + { + "epoch": 0.10410958904109589, + "grad_norm": 2.6576807498931885, + "learning_rate": 1.478102189781022e-06, + "loss": 2.5496, + "step": 95 + }, + { + "epoch": 0.1052054794520548, + "grad_norm": 3.1408190727233887, + "learning_rate": 1.496350364963504e-06, + "loss": 2.6142, + "step": 96 + }, + { + "epoch": 0.1063013698630137, + "grad_norm": 2.3270952701568604, + "learning_rate": 1.5145985401459856e-06, + "loss": 2.584, + "step": 97 + }, + { + "epoch": 0.1073972602739726, + "grad_norm": 2.3976614475250244, + "learning_rate": 1.5328467153284674e-06, + "loss": 2.5941, + "step": 98 + }, + { + "epoch": 0.10849315068493151, + "grad_norm": 2.783485174179077, + "learning_rate": 1.551094890510949e-06, + "loss": 2.5626, + "step": 99 + }, + { + "epoch": 0.1095890410958904, + "grad_norm": 2.2522788047790527, + "learning_rate": 1.5693430656934307e-06, + "loss": 2.517, + "step": 100 + }, + { + "epoch": 0.1095890410958904, + "eval_loss": 2.637315034866333, + "eval_runtime": 48.2107, + "eval_samples_per_second": 20.742, + "eval_steps_per_second": 0.871, + "step": 100 + }, + { + "epoch": 0.11068493150684931, + "grad_norm": 2.917799949645996, + "learning_rate": 1.5875912408759126e-06, + "loss": 2.5159, + "step": 101 + }, + { + "epoch": 0.11178082191780822, + "grad_norm": 2.955815076828003, + "learning_rate": 1.6058394160583942e-06, + "loss": 2.563, + "step": 102 + }, + { + "epoch": 0.11287671232876713, + "grad_norm": 2.8096001148223877, + "learning_rate": 1.624087591240876e-06, + "loss": 2.5995, + "step": 103 + }, + { + "epoch": 0.11397260273972602, + "grad_norm": 2.4967427253723145, + "learning_rate": 1.6423357664233577e-06, + "loss": 2.5336, + "step": 104 + }, + { + "epoch": 0.11506849315068493, + "grad_norm": 2.740717887878418, + "learning_rate": 1.6605839416058394e-06, + "loss": 2.4778, + "step": 105 + }, + { + "epoch": 0.11616438356164384, + "grad_norm": 2.5064632892608643, + "learning_rate": 1.6788321167883212e-06, + "loss": 2.5214, + "step": 106 + }, + { + "epoch": 0.11726027397260275, + "grad_norm": 2.2923007011413574, + "learning_rate": 1.6970802919708029e-06, + "loss": 2.5315, + "step": 107 + }, + { + "epoch": 0.11835616438356164, + "grad_norm": 3.339569568634033, + "learning_rate": 1.715328467153285e-06, + "loss": 2.4661, + "step": 108 + }, + { + "epoch": 0.11945205479452055, + "grad_norm": 2.1650891304016113, + "learning_rate": 1.7335766423357666e-06, + "loss": 2.5072, + "step": 109 + }, + { + "epoch": 0.12054794520547946, + "grad_norm": 3.1216914653778076, + "learning_rate": 1.7518248175182485e-06, + "loss": 2.5164, + "step": 110 + }, + { + "epoch": 0.12054794520547946, + "eval_loss": 2.594250202178955, + "eval_runtime": 41.8939, + "eval_samples_per_second": 23.87, + "eval_steps_per_second": 1.003, + "step": 110 + }, + { + "epoch": 0.12164383561643835, + "grad_norm": 2.191063404083252, + "learning_rate": 1.77007299270073e-06, + "loss": 2.5422, + "step": 111 + }, + { + "epoch": 0.12273972602739726, + "grad_norm": 2.922083854675293, + "learning_rate": 1.788321167883212e-06, + "loss": 2.5258, + "step": 112 + }, + { + "epoch": 0.12383561643835617, + "grad_norm": 2.518002986907959, + "learning_rate": 1.8065693430656936e-06, + "loss": 2.5588, + "step": 113 + }, + { + "epoch": 0.12493150684931507, + "grad_norm": 2.515194892883301, + "learning_rate": 1.8248175182481753e-06, + "loss": 2.5079, + "step": 114 + }, + { + "epoch": 0.12602739726027398, + "grad_norm": 2.5530874729156494, + "learning_rate": 1.8430656934306571e-06, + "loss": 2.4615, + "step": 115 + }, + { + "epoch": 0.12712328767123288, + "grad_norm": 2.3550198078155518, + "learning_rate": 1.8613138686131388e-06, + "loss": 2.4825, + "step": 116 + }, + { + "epoch": 0.12821917808219177, + "grad_norm": 2.295145034790039, + "learning_rate": 1.8795620437956206e-06, + "loss": 2.4749, + "step": 117 + }, + { + "epoch": 0.1293150684931507, + "grad_norm": 2.0158982276916504, + "learning_rate": 1.8978102189781023e-06, + "loss": 2.4756, + "step": 118 + }, + { + "epoch": 0.1304109589041096, + "grad_norm": 2.213057518005371, + "learning_rate": 1.916058394160584e-06, + "loss": 2.4744, + "step": 119 + }, + { + "epoch": 0.13150684931506848, + "grad_norm": 2.186222553253174, + "learning_rate": 1.934306569343066e-06, + "loss": 2.4957, + "step": 120 + }, + { + "epoch": 0.13150684931506848, + "eval_loss": 2.563931465148926, + "eval_runtime": 44.8714, + "eval_samples_per_second": 22.286, + "eval_steps_per_second": 0.936, + "step": 120 + }, + { + "epoch": 0.1326027397260274, + "grad_norm": 2.3143310546875, + "learning_rate": 1.952554744525548e-06, + "loss": 2.4, + "step": 121 + }, + { + "epoch": 0.1336986301369863, + "grad_norm": 2.0223965644836426, + "learning_rate": 1.9708029197080293e-06, + "loss": 2.4453, + "step": 122 + }, + { + "epoch": 0.13479452054794522, + "grad_norm": 2.2025411128997803, + "learning_rate": 1.989051094890511e-06, + "loss": 2.4463, + "step": 123 + }, + { + "epoch": 0.1358904109589041, + "grad_norm": 2.707930564880371, + "learning_rate": 2.007299270072993e-06, + "loss": 2.4589, + "step": 124 + }, + { + "epoch": 0.136986301369863, + "grad_norm": 2.146662473678589, + "learning_rate": 2.025547445255475e-06, + "loss": 2.5023, + "step": 125 + }, + { + "epoch": 0.13808219178082193, + "grad_norm": 2.520660161972046, + "learning_rate": 2.0437956204379563e-06, + "loss": 2.4464, + "step": 126 + }, + { + "epoch": 0.13917808219178082, + "grad_norm": 2.50632381439209, + "learning_rate": 2.062043795620438e-06, + "loss": 2.4556, + "step": 127 + }, + { + "epoch": 0.14027397260273972, + "grad_norm": 2.2919397354125977, + "learning_rate": 2.08029197080292e-06, + "loss": 2.4479, + "step": 128 + }, + { + "epoch": 0.14136986301369864, + "grad_norm": 2.521475315093994, + "learning_rate": 2.0985401459854014e-06, + "loss": 2.4982, + "step": 129 + }, + { + "epoch": 0.14246575342465753, + "grad_norm": 2.3812668323516846, + "learning_rate": 2.1167883211678833e-06, + "loss": 2.4163, + "step": 130 + }, + { + "epoch": 0.14246575342465753, + "eval_loss": 2.5390918254852295, + "eval_runtime": 43.8746, + "eval_samples_per_second": 22.792, + "eval_steps_per_second": 0.957, + "step": 130 + }, + { + "epoch": 0.14356164383561643, + "grad_norm": 2.302286148071289, + "learning_rate": 2.135036496350365e-06, + "loss": 2.4588, + "step": 131 + }, + { + "epoch": 0.14465753424657535, + "grad_norm": 2.4222044944763184, + "learning_rate": 2.1532846715328466e-06, + "loss": 2.492, + "step": 132 + }, + { + "epoch": 0.14575342465753424, + "grad_norm": 2.605835199356079, + "learning_rate": 2.171532846715329e-06, + "loss": 2.4055, + "step": 133 + }, + { + "epoch": 0.14684931506849316, + "grad_norm": 3.6261861324310303, + "learning_rate": 2.1897810218978103e-06, + "loss": 2.4221, + "step": 134 + }, + { + "epoch": 0.14794520547945206, + "grad_norm": 2.7647221088409424, + "learning_rate": 2.208029197080292e-06, + "loss": 2.4549, + "step": 135 + }, + { + "epoch": 0.14904109589041095, + "grad_norm": 3.2977371215820312, + "learning_rate": 2.226277372262774e-06, + "loss": 2.4331, + "step": 136 + }, + { + "epoch": 0.15013698630136987, + "grad_norm": 3.147094488143921, + "learning_rate": 2.244525547445256e-06, + "loss": 2.4268, + "step": 137 + }, + { + "epoch": 0.15123287671232877, + "grad_norm": 2.6363413333892822, + "learning_rate": 2.2627737226277373e-06, + "loss": 2.4524, + "step": 138 + }, + { + "epoch": 0.15232876712328766, + "grad_norm": 2.9288904666900635, + "learning_rate": 2.281021897810219e-06, + "loss": 2.4534, + "step": 139 + }, + { + "epoch": 0.15342465753424658, + "grad_norm": 2.5604870319366455, + "learning_rate": 2.299270072992701e-06, + "loss": 2.4138, + "step": 140 + }, + { + "epoch": 0.15342465753424658, + "eval_loss": 2.524790048599243, + "eval_runtime": 43.8851, + "eval_samples_per_second": 22.787, + "eval_steps_per_second": 0.957, + "step": 140 + }, + { + "epoch": 0.15452054794520548, + "grad_norm": 3.2746663093566895, + "learning_rate": 2.3175182481751825e-06, + "loss": 2.4094, + "step": 141 + }, + { + "epoch": 0.15561643835616437, + "grad_norm": 2.691483497619629, + "learning_rate": 2.3357664233576643e-06, + "loss": 2.4265, + "step": 142 + }, + { + "epoch": 0.1567123287671233, + "grad_norm": 2.5850117206573486, + "learning_rate": 2.354014598540146e-06, + "loss": 2.3922, + "step": 143 + }, + { + "epoch": 0.1578082191780822, + "grad_norm": 2.68204927444458, + "learning_rate": 2.3722627737226276e-06, + "loss": 2.4262, + "step": 144 + }, + { + "epoch": 0.1589041095890411, + "grad_norm": 2.952584981918335, + "learning_rate": 2.39051094890511e-06, + "loss": 2.4267, + "step": 145 + }, + { + "epoch": 0.16, + "grad_norm": 3.2354788780212402, + "learning_rate": 2.4087591240875918e-06, + "loss": 2.3836, + "step": 146 + }, + { + "epoch": 0.1610958904109589, + "grad_norm": 2.634521007537842, + "learning_rate": 2.427007299270073e-06, + "loss": 2.3953, + "step": 147 + }, + { + "epoch": 0.16219178082191782, + "grad_norm": 3.1115682125091553, + "learning_rate": 2.445255474452555e-06, + "loss": 2.409, + "step": 148 + }, + { + "epoch": 0.1632876712328767, + "grad_norm": 2.6824135780334473, + "learning_rate": 2.463503649635037e-06, + "loss": 2.4459, + "step": 149 + }, + { + "epoch": 0.1643835616438356, + "grad_norm": 2.617614507675171, + "learning_rate": 2.4817518248175183e-06, + "loss": 2.366, + "step": 150 + }, + { + "epoch": 0.1643835616438356, + "eval_loss": 2.5088160037994385, + "eval_runtime": 44.0888, + "eval_samples_per_second": 22.682, + "eval_steps_per_second": 0.953, + "step": 150 + }, + { + "epoch": 0.16547945205479453, + "grad_norm": 3.976992130279541, + "learning_rate": 2.5e-06, + "loss": 2.3517, + "step": 151 + }, + { + "epoch": 0.16657534246575342, + "grad_norm": 3.1433017253875732, + "learning_rate": 2.518248175182482e-06, + "loss": 2.4489, + "step": 152 + }, + { + "epoch": 0.16767123287671232, + "grad_norm": 4.100688934326172, + "learning_rate": 2.5364963503649635e-06, + "loss": 2.3683, + "step": 153 + }, + { + "epoch": 0.16876712328767124, + "grad_norm": 4.316102504730225, + "learning_rate": 2.5547445255474458e-06, + "loss": 2.3663, + "step": 154 + }, + { + "epoch": 0.16986301369863013, + "grad_norm": 2.3103132247924805, + "learning_rate": 2.5729927007299272e-06, + "loss": 2.4009, + "step": 155 + }, + { + "epoch": 0.17095890410958905, + "grad_norm": 3.576810836791992, + "learning_rate": 2.591240875912409e-06, + "loss": 2.3983, + "step": 156 + }, + { + "epoch": 0.17205479452054795, + "grad_norm": 2.337080955505371, + "learning_rate": 2.6094890510948905e-06, + "loss": 2.3849, + "step": 157 + }, + { + "epoch": 0.17315068493150684, + "grad_norm": 3.2523117065429688, + "learning_rate": 2.627737226277373e-06, + "loss": 2.4171, + "step": 158 + }, + { + "epoch": 0.17424657534246576, + "grad_norm": 2.529175281524658, + "learning_rate": 2.6459854014598542e-06, + "loss": 2.4303, + "step": 159 + }, + { + "epoch": 0.17534246575342466, + "grad_norm": 3.052212953567505, + "learning_rate": 2.664233576642336e-06, + "loss": 2.3895, + "step": 160 + }, + { + "epoch": 0.17534246575342466, + "eval_loss": 2.492485761642456, + "eval_runtime": 45.6738, + "eval_samples_per_second": 21.894, + "eval_steps_per_second": 0.92, + "step": 160 + }, + { + "epoch": 0.17643835616438355, + "grad_norm": 2.1544973850250244, + "learning_rate": 2.6824817518248175e-06, + "loss": 2.3649, + "step": 161 + }, + { + "epoch": 0.17753424657534247, + "grad_norm": 2.7351951599121094, + "learning_rate": 2.7007299270072994e-06, + "loss": 2.3908, + "step": 162 + }, + { + "epoch": 0.17863013698630137, + "grad_norm": 2.8431951999664307, + "learning_rate": 2.718978102189781e-06, + "loss": 2.374, + "step": 163 + }, + { + "epoch": 0.17972602739726026, + "grad_norm": 3.083190441131592, + "learning_rate": 2.737226277372263e-06, + "loss": 2.4193, + "step": 164 + }, + { + "epoch": 0.18082191780821918, + "grad_norm": 2.6523995399475098, + "learning_rate": 2.755474452554745e-06, + "loss": 2.3969, + "step": 165 + }, + { + "epoch": 0.18191780821917808, + "grad_norm": 2.626047372817993, + "learning_rate": 2.7737226277372264e-06, + "loss": 2.3742, + "step": 166 + }, + { + "epoch": 0.18301369863013697, + "grad_norm": 3.300917863845825, + "learning_rate": 2.7919708029197082e-06, + "loss": 2.4546, + "step": 167 + }, + { + "epoch": 0.1841095890410959, + "grad_norm": 2.797304391860962, + "learning_rate": 2.81021897810219e-06, + "loss": 2.4161, + "step": 168 + }, + { + "epoch": 0.1852054794520548, + "grad_norm": 2.777117967605591, + "learning_rate": 2.828467153284672e-06, + "loss": 2.3572, + "step": 169 + }, + { + "epoch": 0.1863013698630137, + "grad_norm": 2.887470006942749, + "learning_rate": 2.8467153284671534e-06, + "loss": 2.3887, + "step": 170 + }, + { + "epoch": 0.1863013698630137, + "eval_loss": 2.4854257106781006, + "eval_runtime": 44.5644, + "eval_samples_per_second": 22.439, + "eval_steps_per_second": 0.942, + "step": 170 + }, + { + "epoch": 0.1873972602739726, + "grad_norm": 3.0706076622009277, + "learning_rate": 2.8649635036496353e-06, + "loss": 2.3162, + "step": 171 + }, + { + "epoch": 0.1884931506849315, + "grad_norm": 3.6129744052886963, + "learning_rate": 2.8832116788321167e-06, + "loss": 2.3685, + "step": 172 + }, + { + "epoch": 0.18958904109589042, + "grad_norm": 2.4936370849609375, + "learning_rate": 2.901459854014599e-06, + "loss": 2.3297, + "step": 173 + }, + { + "epoch": 0.19068493150684931, + "grad_norm": 2.1041860580444336, + "learning_rate": 2.9197080291970804e-06, + "loss": 2.4085, + "step": 174 + }, + { + "epoch": 0.1917808219178082, + "grad_norm": 2.2066831588745117, + "learning_rate": 2.9379562043795623e-06, + "loss": 2.3905, + "step": 175 + }, + { + "epoch": 0.19287671232876713, + "grad_norm": 2.444652795791626, + "learning_rate": 2.956204379562044e-06, + "loss": 2.378, + "step": 176 + }, + { + "epoch": 0.19397260273972602, + "grad_norm": 2.501939058303833, + "learning_rate": 2.9744525547445256e-06, + "loss": 2.4136, + "step": 177 + }, + { + "epoch": 0.19506849315068492, + "grad_norm": 2.2167556285858154, + "learning_rate": 2.992700729927008e-06, + "loss": 2.385, + "step": 178 + }, + { + "epoch": 0.19616438356164384, + "grad_norm": 2.5298545360565186, + "learning_rate": 3.0109489051094893e-06, + "loss": 2.3508, + "step": 179 + }, + { + "epoch": 0.19726027397260273, + "grad_norm": 2.3774642944335938, + "learning_rate": 3.029197080291971e-06, + "loss": 2.3421, + "step": 180 + }, + { + "epoch": 0.19726027397260273, + "eval_loss": 2.476067304611206, + "eval_runtime": 44.3761, + "eval_samples_per_second": 22.535, + "eval_steps_per_second": 0.946, + "step": 180 + }, + { + "epoch": 0.19835616438356166, + "grad_norm": 2.9108712673187256, + "learning_rate": 3.0474452554744526e-06, + "loss": 2.3983, + "step": 181 + }, + { + "epoch": 0.19945205479452055, + "grad_norm": 2.5232033729553223, + "learning_rate": 3.065693430656935e-06, + "loss": 2.3946, + "step": 182 + }, + { + "epoch": 0.20054794520547944, + "grad_norm": 2.158168077468872, + "learning_rate": 3.0839416058394163e-06, + "loss": 2.3805, + "step": 183 + }, + { + "epoch": 0.20164383561643837, + "grad_norm": 2.3405354022979736, + "learning_rate": 3.102189781021898e-06, + "loss": 2.3974, + "step": 184 + }, + { + "epoch": 0.20273972602739726, + "grad_norm": 2.5308327674865723, + "learning_rate": 3.1204379562043796e-06, + "loss": 2.384, + "step": 185 + }, + { + "epoch": 0.20383561643835615, + "grad_norm": 2.6328237056732178, + "learning_rate": 3.1386861313868614e-06, + "loss": 2.3345, + "step": 186 + }, + { + "epoch": 0.20493150684931508, + "grad_norm": 3.0725512504577637, + "learning_rate": 3.156934306569343e-06, + "loss": 2.3191, + "step": 187 + }, + { + "epoch": 0.20602739726027397, + "grad_norm": 2.196218729019165, + "learning_rate": 3.175182481751825e-06, + "loss": 2.4073, + "step": 188 + }, + { + "epoch": 0.20712328767123286, + "grad_norm": 2.7722432613372803, + "learning_rate": 3.193430656934307e-06, + "loss": 2.3775, + "step": 189 + }, + { + "epoch": 0.20821917808219179, + "grad_norm": 2.4848241806030273, + "learning_rate": 3.2116788321167884e-06, + "loss": 2.3538, + "step": 190 + }, + { + "epoch": 0.20821917808219179, + "eval_loss": 2.462618112564087, + "eval_runtime": 43.9461, + "eval_samples_per_second": 22.755, + "eval_steps_per_second": 0.956, + "step": 190 + }, + { + "epoch": 0.20931506849315068, + "grad_norm": 2.4682741165161133, + "learning_rate": 3.2299270072992707e-06, + "loss": 2.3688, + "step": 191 + }, + { + "epoch": 0.2104109589041096, + "grad_norm": 2.518923282623291, + "learning_rate": 3.248175182481752e-06, + "loss": 2.3556, + "step": 192 + }, + { + "epoch": 0.2115068493150685, + "grad_norm": 2.4023561477661133, + "learning_rate": 3.266423357664234e-06, + "loss": 2.292, + "step": 193 + }, + { + "epoch": 0.2126027397260274, + "grad_norm": 2.5196943283081055, + "learning_rate": 3.2846715328467155e-06, + "loss": 2.3703, + "step": 194 + }, + { + "epoch": 0.2136986301369863, + "grad_norm": 2.762709617614746, + "learning_rate": 3.3029197080291973e-06, + "loss": 2.3578, + "step": 195 + }, + { + "epoch": 0.2147945205479452, + "grad_norm": 2.1287691593170166, + "learning_rate": 3.3211678832116788e-06, + "loss": 2.3687, + "step": 196 + }, + { + "epoch": 0.2158904109589041, + "grad_norm": 2.477341651916504, + "learning_rate": 3.339416058394161e-06, + "loss": 2.3545, + "step": 197 + }, + { + "epoch": 0.21698630136986302, + "grad_norm": 2.23933744430542, + "learning_rate": 3.3576642335766425e-06, + "loss": 2.3511, + "step": 198 + }, + { + "epoch": 0.21808219178082192, + "grad_norm": 2.2440133094787598, + "learning_rate": 3.3759124087591243e-06, + "loss": 2.3539, + "step": 199 + }, + { + "epoch": 0.2191780821917808, + "grad_norm": 2.5603177547454834, + "learning_rate": 3.3941605839416058e-06, + "loss": 2.36, + "step": 200 + }, + { + "epoch": 0.2191780821917808, + "eval_loss": 2.4507246017456055, + "eval_runtime": 43.7786, + "eval_samples_per_second": 22.842, + "eval_steps_per_second": 0.959, + "step": 200 + }, + { + "epoch": 0.22027397260273973, + "grad_norm": 2.4570510387420654, + "learning_rate": 3.412408759124088e-06, + "loss": 2.3702, + "step": 201 + }, + { + "epoch": 0.22136986301369863, + "grad_norm": 2.1975133419036865, + "learning_rate": 3.43065693430657e-06, + "loss": 2.3667, + "step": 202 + }, + { + "epoch": 0.22246575342465755, + "grad_norm": 2.710186004638672, + "learning_rate": 3.4489051094890513e-06, + "loss": 2.3159, + "step": 203 + }, + { + "epoch": 0.22356164383561644, + "grad_norm": 2.7799506187438965, + "learning_rate": 3.467153284671533e-06, + "loss": 2.3519, + "step": 204 + }, + { + "epoch": 0.22465753424657534, + "grad_norm": 2.2479400634765625, + "learning_rate": 3.4854014598540146e-06, + "loss": 2.3103, + "step": 205 + }, + { + "epoch": 0.22575342465753426, + "grad_norm": 3.1538405418395996, + "learning_rate": 3.503649635036497e-06, + "loss": 2.3506, + "step": 206 + }, + { + "epoch": 0.22684931506849315, + "grad_norm": 2.7607762813568115, + "learning_rate": 3.5218978102189783e-06, + "loss": 2.3386, + "step": 207 + }, + { + "epoch": 0.22794520547945205, + "grad_norm": 2.371105909347534, + "learning_rate": 3.54014598540146e-06, + "loss": 2.3583, + "step": 208 + }, + { + "epoch": 0.22904109589041097, + "grad_norm": 2.793933391571045, + "learning_rate": 3.5583941605839416e-06, + "loss": 2.3782, + "step": 209 + }, + { + "epoch": 0.23013698630136986, + "grad_norm": 2.372732400894165, + "learning_rate": 3.576642335766424e-06, + "loss": 2.417, + "step": 210 + }, + { + "epoch": 0.23013698630136986, + "eval_loss": 2.451474189758301, + "eval_runtime": 41.7197, + "eval_samples_per_second": 23.97, + "eval_steps_per_second": 1.007, + "step": 210 + }, + { + "epoch": 0.23123287671232876, + "grad_norm": 2.6989448070526123, + "learning_rate": 3.5948905109489054e-06, + "loss": 2.393, + "step": 211 + }, + { + "epoch": 0.23232876712328768, + "grad_norm": 2.4324285984039307, + "learning_rate": 3.6131386861313872e-06, + "loss": 2.2738, + "step": 212 + }, + { + "epoch": 0.23342465753424657, + "grad_norm": 3.5063016414642334, + "learning_rate": 3.6313868613138687e-06, + "loss": 2.2823, + "step": 213 + }, + { + "epoch": 0.2345205479452055, + "grad_norm": 3.953951358795166, + "learning_rate": 3.6496350364963505e-06, + "loss": 2.3132, + "step": 214 + }, + { + "epoch": 0.2356164383561644, + "grad_norm": 2.3483071327209473, + "learning_rate": 3.667883211678833e-06, + "loss": 2.3244, + "step": 215 + }, + { + "epoch": 0.23671232876712328, + "grad_norm": 2.884199380874634, + "learning_rate": 3.6861313868613142e-06, + "loss": 2.3067, + "step": 216 + }, + { + "epoch": 0.2378082191780822, + "grad_norm": 2.9514520168304443, + "learning_rate": 3.704379562043796e-06, + "loss": 2.304, + "step": 217 + }, + { + "epoch": 0.2389041095890411, + "grad_norm": 2.7530996799468994, + "learning_rate": 3.7226277372262775e-06, + "loss": 2.3691, + "step": 218 + }, + { + "epoch": 0.24, + "grad_norm": 2.8753647804260254, + "learning_rate": 3.7408759124087594e-06, + "loss": 2.3406, + "step": 219 + }, + { + "epoch": 0.2410958904109589, + "grad_norm": 2.670574903488159, + "learning_rate": 3.7591240875912412e-06, + "loss": 2.304, + "step": 220 + }, + { + "epoch": 0.2410958904109589, + "eval_loss": 2.4395809173583984, + "eval_runtime": 46.6723, + "eval_samples_per_second": 21.426, + "eval_steps_per_second": 0.9, + "step": 220 + }, + { + "epoch": 0.2421917808219178, + "grad_norm": 2.4152438640594482, + "learning_rate": 3.777372262773723e-06, + "loss": 2.3294, + "step": 221 + }, + { + "epoch": 0.2432876712328767, + "grad_norm": 2.8398540019989014, + "learning_rate": 3.7956204379562045e-06, + "loss": 2.3216, + "step": 222 + }, + { + "epoch": 0.24438356164383562, + "grad_norm": 2.6036765575408936, + "learning_rate": 3.8138686131386864e-06, + "loss": 2.3109, + "step": 223 + }, + { + "epoch": 0.24547945205479452, + "grad_norm": 2.7073473930358887, + "learning_rate": 3.832116788321168e-06, + "loss": 2.3457, + "step": 224 + }, + { + "epoch": 0.2465753424657534, + "grad_norm": 2.584404230117798, + "learning_rate": 3.85036496350365e-06, + "loss": 2.3159, + "step": 225 + }, + { + "epoch": 0.24767123287671233, + "grad_norm": 2.4220666885375977, + "learning_rate": 3.868613138686132e-06, + "loss": 2.2944, + "step": 226 + }, + { + "epoch": 0.24876712328767123, + "grad_norm": 2.5636494159698486, + "learning_rate": 3.886861313868613e-06, + "loss": 2.3219, + "step": 227 + }, + { + "epoch": 0.24986301369863015, + "grad_norm": 2.2990903854370117, + "learning_rate": 3.905109489051096e-06, + "loss": 2.3342, + "step": 228 + }, + { + "epoch": 0.25095890410958904, + "grad_norm": 2.50913143157959, + "learning_rate": 3.923357664233577e-06, + "loss": 2.3619, + "step": 229 + }, + { + "epoch": 0.25205479452054796, + "grad_norm": 2.1687769889831543, + "learning_rate": 3.9416058394160585e-06, + "loss": 2.3153, + "step": 230 + }, + { + "epoch": 0.25205479452054796, + "eval_loss": 2.4315240383148193, + "eval_runtime": 48.3542, + "eval_samples_per_second": 20.681, + "eval_steps_per_second": 0.869, + "step": 230 + }, + { + "epoch": 0.25315068493150683, + "grad_norm": 2.842125177383423, + "learning_rate": 3.95985401459854e-06, + "loss": 2.3644, + "step": 231 + }, + { + "epoch": 0.25424657534246575, + "grad_norm": 2.258049964904785, + "learning_rate": 3.978102189781022e-06, + "loss": 2.3727, + "step": 232 + }, + { + "epoch": 0.2553424657534247, + "grad_norm": 2.4630489349365234, + "learning_rate": 3.996350364963504e-06, + "loss": 2.267, + "step": 233 + }, + { + "epoch": 0.25643835616438354, + "grad_norm": 3.1469743251800537, + "learning_rate": 4.014598540145986e-06, + "loss": 2.3432, + "step": 234 + }, + { + "epoch": 0.25753424657534246, + "grad_norm": 2.6277477741241455, + "learning_rate": 4.032846715328467e-06, + "loss": 2.2545, + "step": 235 + }, + { + "epoch": 0.2586301369863014, + "grad_norm": 2.404141902923584, + "learning_rate": 4.05109489051095e-06, + "loss": 2.3109, + "step": 236 + }, + { + "epoch": 0.25972602739726025, + "grad_norm": 2.4545605182647705, + "learning_rate": 4.069343065693431e-06, + "loss": 2.3112, + "step": 237 + }, + { + "epoch": 0.2608219178082192, + "grad_norm": 2.237638235092163, + "learning_rate": 4.0875912408759126e-06, + "loss": 2.3, + "step": 238 + }, + { + "epoch": 0.2619178082191781, + "grad_norm": 2.192811965942383, + "learning_rate": 4.105839416058395e-06, + "loss": 2.3108, + "step": 239 + }, + { + "epoch": 0.26301369863013696, + "grad_norm": 2.5058720111846924, + "learning_rate": 4.124087591240876e-06, + "loss": 2.3635, + "step": 240 + }, + { + "epoch": 0.26301369863013696, + "eval_loss": 2.4219377040863037, + "eval_runtime": 45.3561, + "eval_samples_per_second": 22.048, + "eval_steps_per_second": 0.926, + "step": 240 + }, + { + "epoch": 0.2641095890410959, + "grad_norm": 2.0708935260772705, + "learning_rate": 4.1423357664233586e-06, + "loss": 2.3129, + "step": 241 + }, + { + "epoch": 0.2652054794520548, + "grad_norm": 2.6353116035461426, + "learning_rate": 4.16058394160584e-06, + "loss": 2.3011, + "step": 242 + }, + { + "epoch": 0.2663013698630137, + "grad_norm": 2.8392436504364014, + "learning_rate": 4.1788321167883214e-06, + "loss": 2.3643, + "step": 243 + }, + { + "epoch": 0.2673972602739726, + "grad_norm": 3.1773295402526855, + "learning_rate": 4.197080291970803e-06, + "loss": 2.2642, + "step": 244 + }, + { + "epoch": 0.2684931506849315, + "grad_norm": 2.4757354259490967, + "learning_rate": 4.215328467153285e-06, + "loss": 2.3211, + "step": 245 + }, + { + "epoch": 0.26958904109589044, + "grad_norm": 2.5799412727355957, + "learning_rate": 4.233576642335767e-06, + "loss": 2.2909, + "step": 246 + }, + { + "epoch": 0.2706849315068493, + "grad_norm": 2.368502378463745, + "learning_rate": 4.251824817518249e-06, + "loss": 2.2951, + "step": 247 + }, + { + "epoch": 0.2717808219178082, + "grad_norm": 2.114969253540039, + "learning_rate": 4.27007299270073e-06, + "loss": 2.3124, + "step": 248 + }, + { + "epoch": 0.27287671232876715, + "grad_norm": 3.056820869445801, + "learning_rate": 4.288321167883212e-06, + "loss": 2.2923, + "step": 249 + }, + { + "epoch": 0.273972602739726, + "grad_norm": 2.7752203941345215, + "learning_rate": 4.306569343065693e-06, + "loss": 2.2874, + "step": 250 + }, + { + "epoch": 0.273972602739726, + "eval_loss": 2.4229912757873535, + "eval_runtime": 52.5941, + "eval_samples_per_second": 19.014, + "eval_steps_per_second": 0.799, + "step": 250 + }, + { + "epoch": 0.27506849315068493, + "grad_norm": 3.045740842819214, + "learning_rate": 4.3248175182481755e-06, + "loss": 2.2686, + "step": 251 + }, + { + "epoch": 0.27616438356164386, + "grad_norm": 2.646402359008789, + "learning_rate": 4.343065693430658e-06, + "loss": 2.3247, + "step": 252 + }, + { + "epoch": 0.2772602739726027, + "grad_norm": 3.274994134902954, + "learning_rate": 4.361313868613139e-06, + "loss": 2.3093, + "step": 253 + }, + { + "epoch": 0.27835616438356164, + "grad_norm": 2.352246046066284, + "learning_rate": 4.379562043795621e-06, + "loss": 2.3124, + "step": 254 + }, + { + "epoch": 0.27945205479452057, + "grad_norm": 3.248717784881592, + "learning_rate": 4.397810218978102e-06, + "loss": 2.3417, + "step": 255 + }, + { + "epoch": 0.28054794520547943, + "grad_norm": 3.0919013023376465, + "learning_rate": 4.416058394160584e-06, + "loss": 2.3093, + "step": 256 + }, + { + "epoch": 0.28164383561643835, + "grad_norm": 2.1174702644348145, + "learning_rate": 4.434306569343066e-06, + "loss": 2.305, + "step": 257 + }, + { + "epoch": 0.2827397260273973, + "grad_norm": 2.913797378540039, + "learning_rate": 4.452554744525548e-06, + "loss": 2.2831, + "step": 258 + }, + { + "epoch": 0.28383561643835614, + "grad_norm": 2.693089723587036, + "learning_rate": 4.4708029197080295e-06, + "loss": 2.3206, + "step": 259 + }, + { + "epoch": 0.28493150684931506, + "grad_norm": 2.521411895751953, + "learning_rate": 4.489051094890512e-06, + "loss": 2.3196, + "step": 260 + }, + { + "epoch": 0.28493150684931506, + "eval_loss": 2.4178109169006348, + "eval_runtime": 50.6753, + "eval_samples_per_second": 19.733, + "eval_steps_per_second": 0.829, + "step": 260 + }, + { + "epoch": 0.286027397260274, + "grad_norm": 2.346346855163574, + "learning_rate": 4.507299270072993e-06, + "loss": 2.3433, + "step": 261 + }, + { + "epoch": 0.28712328767123285, + "grad_norm": 2.520368814468384, + "learning_rate": 4.525547445255475e-06, + "loss": 2.3174, + "step": 262 + }, + { + "epoch": 0.2882191780821918, + "grad_norm": 2.550347328186035, + "learning_rate": 4.543795620437956e-06, + "loss": 2.3072, + "step": 263 + }, + { + "epoch": 0.2893150684931507, + "grad_norm": 2.268662691116333, + "learning_rate": 4.562043795620438e-06, + "loss": 2.2763, + "step": 264 + }, + { + "epoch": 0.29041095890410956, + "grad_norm": 2.445798873901367, + "learning_rate": 4.580291970802921e-06, + "loss": 2.3395, + "step": 265 + }, + { + "epoch": 0.2915068493150685, + "grad_norm": 2.5391135215759277, + "learning_rate": 4.598540145985402e-06, + "loss": 2.3496, + "step": 266 + }, + { + "epoch": 0.2926027397260274, + "grad_norm": 2.3553097248077393, + "learning_rate": 4.6167883211678835e-06, + "loss": 2.3243, + "step": 267 + }, + { + "epoch": 0.2936986301369863, + "grad_norm": 2.448089599609375, + "learning_rate": 4.635036496350365e-06, + "loss": 2.2887, + "step": 268 + }, + { + "epoch": 0.2947945205479452, + "grad_norm": 2.1600170135498047, + "learning_rate": 4.653284671532847e-06, + "loss": 2.2816, + "step": 269 + }, + { + "epoch": 0.2958904109589041, + "grad_norm": 2.3783180713653564, + "learning_rate": 4.671532846715329e-06, + "loss": 2.3275, + "step": 270 + }, + { + "epoch": 0.2958904109589041, + "eval_loss": 2.413832902908325, + "eval_runtime": 48.7524, + "eval_samples_per_second": 20.512, + "eval_steps_per_second": 0.861, + "step": 270 + }, + { + "epoch": 0.29698630136986304, + "grad_norm": 2.468320369720459, + "learning_rate": 4.689781021897811e-06, + "loss": 2.3358, + "step": 271 + }, + { + "epoch": 0.2980821917808219, + "grad_norm": 2.401634454727173, + "learning_rate": 4.708029197080292e-06, + "loss": 2.2991, + "step": 272 + }, + { + "epoch": 0.2991780821917808, + "grad_norm": 2.4604203701019287, + "learning_rate": 4.726277372262774e-06, + "loss": 2.3496, + "step": 273 + }, + { + "epoch": 0.30027397260273975, + "grad_norm": 2.0338871479034424, + "learning_rate": 4.744525547445255e-06, + "loss": 2.3115, + "step": 274 + }, + { + "epoch": 0.3013698630136986, + "grad_norm": 2.5404436588287354, + "learning_rate": 4.7627737226277375e-06, + "loss": 2.2665, + "step": 275 + }, + { + "epoch": 0.30246575342465754, + "grad_norm": 2.5369458198547363, + "learning_rate": 4.78102189781022e-06, + "loss": 2.3072, + "step": 276 + }, + { + "epoch": 0.30356164383561646, + "grad_norm": 2.471522569656372, + "learning_rate": 4.799270072992701e-06, + "loss": 2.3113, + "step": 277 + }, + { + "epoch": 0.3046575342465753, + "grad_norm": 2.3890135288238525, + "learning_rate": 4.8175182481751835e-06, + "loss": 2.2838, + "step": 278 + }, + { + "epoch": 0.30575342465753425, + "grad_norm": 2.2485053539276123, + "learning_rate": 4.835766423357665e-06, + "loss": 2.3052, + "step": 279 + }, + { + "epoch": 0.30684931506849317, + "grad_norm": 2.1090753078460693, + "learning_rate": 4.854014598540146e-06, + "loss": 2.3054, + "step": 280 + }, + { + "epoch": 0.30684931506849317, + "eval_loss": 2.412548780441284, + "eval_runtime": 48.5034, + "eval_samples_per_second": 20.617, + "eval_steps_per_second": 0.866, + "step": 280 + }, + { + "epoch": 0.30794520547945203, + "grad_norm": 2.299133777618408, + "learning_rate": 4.872262773722628e-06, + "loss": 2.2599, + "step": 281 + }, + { + "epoch": 0.30904109589041096, + "grad_norm": 2.177905797958374, + "learning_rate": 4.89051094890511e-06, + "loss": 2.2745, + "step": 282 + }, + { + "epoch": 0.3101369863013699, + "grad_norm": 2.3178462982177734, + "learning_rate": 4.9087591240875915e-06, + "loss": 2.2562, + "step": 283 + }, + { + "epoch": 0.31123287671232874, + "grad_norm": 2.5088376998901367, + "learning_rate": 4.927007299270074e-06, + "loss": 2.2945, + "step": 284 + }, + { + "epoch": 0.31232876712328766, + "grad_norm": 2.095379590988159, + "learning_rate": 4.945255474452555e-06, + "loss": 2.3038, + "step": 285 + }, + { + "epoch": 0.3134246575342466, + "grad_norm": 2.1792995929718018, + "learning_rate": 4.963503649635037e-06, + "loss": 2.2702, + "step": 286 + }, + { + "epoch": 0.31452054794520545, + "grad_norm": 2.3001186847686768, + "learning_rate": 4.981751824817518e-06, + "loss": 2.3482, + "step": 287 + }, + { + "epoch": 0.3156164383561644, + "grad_norm": 2.1576106548309326, + "learning_rate": 5e-06, + "loss": 2.3391, + "step": 288 + }, + { + "epoch": 0.3167123287671233, + "grad_norm": 2.3715577125549316, + "learning_rate": 4.997969130787978e-06, + "loss": 2.3049, + "step": 289 + }, + { + "epoch": 0.3178082191780822, + "grad_norm": 2.0316715240478516, + "learning_rate": 4.995938261575955e-06, + "loss": 2.3111, + "step": 290 + }, + { + "epoch": 0.3178082191780822, + "eval_loss": 2.40299654006958, + "eval_runtime": 50.0873, + "eval_samples_per_second": 19.965, + "eval_steps_per_second": 0.839, + "step": 290 + }, + { + "epoch": 0.3189041095890411, + "grad_norm": 2.1135756969451904, + "learning_rate": 4.993907392363932e-06, + "loss": 2.2465, + "step": 291 + }, + { + "epoch": 0.32, + "grad_norm": 2.335655927658081, + "learning_rate": 4.9918765231519095e-06, + "loss": 2.31, + "step": 292 + }, + { + "epoch": 0.32109589041095893, + "grad_norm": 2.3255794048309326, + "learning_rate": 4.989845653939887e-06, + "loss": 2.2531, + "step": 293 + }, + { + "epoch": 0.3221917808219178, + "grad_norm": 2.111055612564087, + "learning_rate": 4.987814784727864e-06, + "loss": 2.2799, + "step": 294 + }, + { + "epoch": 0.3232876712328767, + "grad_norm": 2.1773171424865723, + "learning_rate": 4.985783915515841e-06, + "loss": 2.2884, + "step": 295 + }, + { + "epoch": 0.32438356164383564, + "grad_norm": 2.3559885025024414, + "learning_rate": 4.983753046303819e-06, + "loss": 2.2711, + "step": 296 + }, + { + "epoch": 0.3254794520547945, + "grad_norm": 2.120802879333496, + "learning_rate": 4.981722177091796e-06, + "loss": 2.2946, + "step": 297 + }, + { + "epoch": 0.3265753424657534, + "grad_norm": 2.2319388389587402, + "learning_rate": 4.979691307879773e-06, + "loss": 2.2771, + "step": 298 + }, + { + "epoch": 0.32767123287671235, + "grad_norm": 2.208740711212158, + "learning_rate": 4.9776604386677504e-06, + "loss": 2.3003, + "step": 299 + }, + { + "epoch": 0.3287671232876712, + "grad_norm": 2.543060541152954, + "learning_rate": 4.975629569455728e-06, + "loss": 2.2891, + "step": 300 + }, + { + "epoch": 0.3287671232876712, + "eval_loss": 2.398495674133301, + "eval_runtime": 51.1973, + "eval_samples_per_second": 19.532, + "eval_steps_per_second": 0.82, + "step": 300 + }, + { + "epoch": 0.32986301369863014, + "grad_norm": 2.086219072341919, + "learning_rate": 4.973598700243705e-06, + "loss": 2.3286, + "step": 301 + }, + { + "epoch": 0.33095890410958906, + "grad_norm": 2.3673319816589355, + "learning_rate": 4.971567831031681e-06, + "loss": 2.2672, + "step": 302 + }, + { + "epoch": 0.3320547945205479, + "grad_norm": 2.167593240737915, + "learning_rate": 4.9695369618196595e-06, + "loss": 2.264, + "step": 303 + }, + { + "epoch": 0.33315068493150685, + "grad_norm": 2.7218785285949707, + "learning_rate": 4.967506092607636e-06, + "loss": 2.1737, + "step": 304 + }, + { + "epoch": 0.33424657534246577, + "grad_norm": 2.673245429992676, + "learning_rate": 4.965475223395614e-06, + "loss": 2.2439, + "step": 305 + }, + { + "epoch": 0.33534246575342463, + "grad_norm": 2.0341989994049072, + "learning_rate": 4.9634443541835905e-06, + "loss": 2.2614, + "step": 306 + }, + { + "epoch": 0.33643835616438356, + "grad_norm": 1.9314247369766235, + "learning_rate": 4.961413484971569e-06, + "loss": 2.2592, + "step": 307 + }, + { + "epoch": 0.3375342465753425, + "grad_norm": 2.537794589996338, + "learning_rate": 4.959382615759545e-06, + "loss": 2.2859, + "step": 308 + }, + { + "epoch": 0.33863013698630134, + "grad_norm": 2.0811166763305664, + "learning_rate": 4.957351746547523e-06, + "loss": 2.2868, + "step": 309 + }, + { + "epoch": 0.33972602739726027, + "grad_norm": 2.783701181411743, + "learning_rate": 4.9553208773355e-06, + "loss": 2.2649, + "step": 310 + }, + { + "epoch": 0.33972602739726027, + "eval_loss": 2.400538921356201, + "eval_runtime": 43.185, + "eval_samples_per_second": 23.156, + "eval_steps_per_second": 0.973, + "step": 310 + }, + { + "epoch": 0.3408219178082192, + "grad_norm": 2.8541319370269775, + "learning_rate": 4.953290008123478e-06, + "loss": 2.3729, + "step": 311 + }, + { + "epoch": 0.3419178082191781, + "grad_norm": 2.1852617263793945, + "learning_rate": 4.951259138911454e-06, + "loss": 2.2885, + "step": 312 + }, + { + "epoch": 0.343013698630137, + "grad_norm": 2.872201919555664, + "learning_rate": 4.9492282696994315e-06, + "loss": 2.2353, + "step": 313 + }, + { + "epoch": 0.3441095890410959, + "grad_norm": 2.5140910148620605, + "learning_rate": 4.947197400487409e-06, + "loss": 2.3364, + "step": 314 + }, + { + "epoch": 0.3452054794520548, + "grad_norm": 2.619974374771118, + "learning_rate": 4.945166531275386e-06, + "loss": 2.2721, + "step": 315 + }, + { + "epoch": 0.3463013698630137, + "grad_norm": 2.4567618370056152, + "learning_rate": 4.943135662063363e-06, + "loss": 2.26, + "step": 316 + }, + { + "epoch": 0.3473972602739726, + "grad_norm": 2.9603192806243896, + "learning_rate": 4.9411047928513406e-06, + "loss": 2.2861, + "step": 317 + }, + { + "epoch": 0.34849315068493153, + "grad_norm": 2.115262985229492, + "learning_rate": 4.939073923639318e-06, + "loss": 2.3125, + "step": 318 + }, + { + "epoch": 0.3495890410958904, + "grad_norm": 2.693024158477783, + "learning_rate": 4.937043054427295e-06, + "loss": 2.3054, + "step": 319 + }, + { + "epoch": 0.3506849315068493, + "grad_norm": 2.3439996242523193, + "learning_rate": 4.935012185215272e-06, + "loss": 2.2576, + "step": 320 + }, + { + "epoch": 0.3506849315068493, + "eval_loss": 2.3920650482177734, + "eval_runtime": 47.7451, + "eval_samples_per_second": 20.945, + "eval_steps_per_second": 0.88, + "step": 320 + }, + { + "epoch": 0.35178082191780824, + "grad_norm": 2.3009634017944336, + "learning_rate": 4.93298131600325e-06, + "loss": 2.2454, + "step": 321 + }, + { + "epoch": 0.3528767123287671, + "grad_norm": 2.593203544616699, + "learning_rate": 4.930950446791227e-06, + "loss": 2.2965, + "step": 322 + }, + { + "epoch": 0.353972602739726, + "grad_norm": 2.9069650173187256, + "learning_rate": 4.928919577579204e-06, + "loss": 2.2785, + "step": 323 + }, + { + "epoch": 0.35506849315068495, + "grad_norm": 2.1021759510040283, + "learning_rate": 4.9268887083671815e-06, + "loss": 2.2857, + "step": 324 + }, + { + "epoch": 0.3561643835616438, + "grad_norm": 2.0330185890197754, + "learning_rate": 4.924857839155159e-06, + "loss": 2.2697, + "step": 325 + }, + { + "epoch": 0.35726027397260274, + "grad_norm": 2.1318063735961914, + "learning_rate": 4.922826969943136e-06, + "loss": 2.2698, + "step": 326 + }, + { + "epoch": 0.35835616438356166, + "grad_norm": 2.0157387256622314, + "learning_rate": 4.920796100731113e-06, + "loss": 2.292, + "step": 327 + }, + { + "epoch": 0.3594520547945205, + "grad_norm": 2.3266026973724365, + "learning_rate": 4.918765231519091e-06, + "loss": 2.2215, + "step": 328 + }, + { + "epoch": 0.36054794520547945, + "grad_norm": 2.066305160522461, + "learning_rate": 4.916734362307068e-06, + "loss": 2.2758, + "step": 329 + }, + { + "epoch": 0.36164383561643837, + "grad_norm": 2.0219993591308594, + "learning_rate": 4.914703493095045e-06, + "loss": 2.2381, + "step": 330 + }, + { + "epoch": 0.36164383561643837, + "eval_loss": 2.394064426422119, + "eval_runtime": 51.5482, + "eval_samples_per_second": 19.399, + "eval_steps_per_second": 0.815, + "step": 330 + }, + { + "epoch": 0.36273972602739724, + "grad_norm": 2.3964319229125977, + "learning_rate": 4.912672623883022e-06, + "loss": 2.2384, + "step": 331 + }, + { + "epoch": 0.36383561643835616, + "grad_norm": 2.5466530323028564, + "learning_rate": 4.910641754671e-06, + "loss": 2.2428, + "step": 332 + }, + { + "epoch": 0.3649315068493151, + "grad_norm": 2.236152410507202, + "learning_rate": 4.908610885458977e-06, + "loss": 2.2639, + "step": 333 + }, + { + "epoch": 0.36602739726027395, + "grad_norm": 2.7267303466796875, + "learning_rate": 4.906580016246954e-06, + "loss": 2.2608, + "step": 334 + }, + { + "epoch": 0.36712328767123287, + "grad_norm": 1.9403765201568604, + "learning_rate": 4.9045491470349315e-06, + "loss": 2.2962, + "step": 335 + }, + { + "epoch": 0.3682191780821918, + "grad_norm": 2.89070463180542, + "learning_rate": 4.902518277822909e-06, + "loss": 2.2847, + "step": 336 + }, + { + "epoch": 0.3693150684931507, + "grad_norm": 2.6427228450775146, + "learning_rate": 4.900487408610886e-06, + "loss": 2.2734, + "step": 337 + }, + { + "epoch": 0.3704109589041096, + "grad_norm": 2.1680362224578857, + "learning_rate": 4.898456539398863e-06, + "loss": 2.2869, + "step": 338 + }, + { + "epoch": 0.3715068493150685, + "grad_norm": 3.0160160064697266, + "learning_rate": 4.89642567018684e-06, + "loss": 2.3043, + "step": 339 + }, + { + "epoch": 0.3726027397260274, + "grad_norm": 1.9858418703079224, + "learning_rate": 4.894394800974818e-06, + "loss": 2.3422, + "step": 340 + }, + { + "epoch": 0.3726027397260274, + "eval_loss": 2.387559652328491, + "eval_runtime": 52.2455, + "eval_samples_per_second": 19.14, + "eval_steps_per_second": 0.804, + "step": 340 + }, + { + "epoch": 0.3736986301369863, + "grad_norm": 2.243699073791504, + "learning_rate": 4.892363931762794e-06, + "loss": 2.2394, + "step": 341 + }, + { + "epoch": 0.3747945205479452, + "grad_norm": 2.371299982070923, + "learning_rate": 4.8903330625507725e-06, + "loss": 2.2856, + "step": 342 + }, + { + "epoch": 0.37589041095890413, + "grad_norm": 2.3769736289978027, + "learning_rate": 4.888302193338749e-06, + "loss": 2.246, + "step": 343 + }, + { + "epoch": 0.376986301369863, + "grad_norm": 2.3209030628204346, + "learning_rate": 4.886271324126727e-06, + "loss": 2.2331, + "step": 344 + }, + { + "epoch": 0.3780821917808219, + "grad_norm": 2.747319221496582, + "learning_rate": 4.8842404549147034e-06, + "loss": 2.3115, + "step": 345 + }, + { + "epoch": 0.37917808219178084, + "grad_norm": 1.8599231243133545, + "learning_rate": 4.8822095857026816e-06, + "loss": 2.2994, + "step": 346 + }, + { + "epoch": 0.3802739726027397, + "grad_norm": 2.0358943939208984, + "learning_rate": 4.880178716490658e-06, + "loss": 2.271, + "step": 347 + }, + { + "epoch": 0.38136986301369863, + "grad_norm": 2.409106731414795, + "learning_rate": 4.878147847278636e-06, + "loss": 2.2281, + "step": 348 + }, + { + "epoch": 0.38246575342465755, + "grad_norm": 2.1641807556152344, + "learning_rate": 4.8761169780666125e-06, + "loss": 2.2437, + "step": 349 + }, + { + "epoch": 0.3835616438356164, + "grad_norm": 2.6917359828948975, + "learning_rate": 4.874086108854591e-06, + "loss": 2.226, + "step": 350 + }, + { + "epoch": 0.3835616438356164, + "eval_loss": 2.382991313934326, + "eval_runtime": 48.9534, + "eval_samples_per_second": 20.428, + "eval_steps_per_second": 0.858, + "step": 350 + }, + { + "epoch": 0.38465753424657534, + "grad_norm": 2.364215612411499, + "learning_rate": 4.872055239642567e-06, + "loss": 2.2319, + "step": 351 + }, + { + "epoch": 0.38575342465753426, + "grad_norm": 2.317643165588379, + "learning_rate": 4.870024370430544e-06, + "loss": 2.2096, + "step": 352 + }, + { + "epoch": 0.3868493150684931, + "grad_norm": 2.2283246517181396, + "learning_rate": 4.867993501218522e-06, + "loss": 2.2634, + "step": 353 + }, + { + "epoch": 0.38794520547945205, + "grad_norm": 2.498391628265381, + "learning_rate": 4.865962632006499e-06, + "loss": 2.2813, + "step": 354 + }, + { + "epoch": 0.38904109589041097, + "grad_norm": 2.3889403343200684, + "learning_rate": 4.863931762794476e-06, + "loss": 2.2361, + "step": 355 + }, + { + "epoch": 0.39013698630136984, + "grad_norm": 2.2049975395202637, + "learning_rate": 4.8619008935824535e-06, + "loss": 2.2583, + "step": 356 + }, + { + "epoch": 0.39123287671232876, + "grad_norm": 2.132840871810913, + "learning_rate": 4.859870024370431e-06, + "loss": 2.2823, + "step": 357 + }, + { + "epoch": 0.3923287671232877, + "grad_norm": 2.2692184448242188, + "learning_rate": 4.857839155158408e-06, + "loss": 2.2939, + "step": 358 + }, + { + "epoch": 0.3934246575342466, + "grad_norm": 2.0501933097839355, + "learning_rate": 4.855808285946385e-06, + "loss": 2.2976, + "step": 359 + }, + { + "epoch": 0.39452054794520547, + "grad_norm": 2.170236587524414, + "learning_rate": 4.8537774167343626e-06, + "loss": 2.2748, + "step": 360 + }, + { + "epoch": 0.39452054794520547, + "eval_loss": 2.3783886432647705, + "eval_runtime": 46.7324, + "eval_samples_per_second": 21.398, + "eval_steps_per_second": 0.899, + "step": 360 + }, + { + "epoch": 0.3956164383561644, + "grad_norm": 2.283137083053589, + "learning_rate": 4.85174654752234e-06, + "loss": 2.2705, + "step": 361 + }, + { + "epoch": 0.3967123287671233, + "grad_norm": 2.195984363555908, + "learning_rate": 4.849715678310317e-06, + "loss": 2.26, + "step": 362 + }, + { + "epoch": 0.3978082191780822, + "grad_norm": 2.399935007095337, + "learning_rate": 4.847684809098294e-06, + "loss": 2.2639, + "step": 363 + }, + { + "epoch": 0.3989041095890411, + "grad_norm": 2.411280393600464, + "learning_rate": 4.845653939886272e-06, + "loss": 2.2523, + "step": 364 + }, + { + "epoch": 0.4, + "grad_norm": 2.2033896446228027, + "learning_rate": 4.843623070674249e-06, + "loss": 2.2722, + "step": 365 + }, + { + "epoch": 0.4010958904109589, + "grad_norm": 2.2990238666534424, + "learning_rate": 4.841592201462226e-06, + "loss": 2.2411, + "step": 366 + }, + { + "epoch": 0.4021917808219178, + "grad_norm": 1.911453127861023, + "learning_rate": 4.8395613322502035e-06, + "loss": 2.2921, + "step": 367 + }, + { + "epoch": 0.40328767123287673, + "grad_norm": 2.453195095062256, + "learning_rate": 4.837530463038181e-06, + "loss": 2.218, + "step": 368 + }, + { + "epoch": 0.4043835616438356, + "grad_norm": 2.126992702484131, + "learning_rate": 4.835499593826158e-06, + "loss": 2.2628, + "step": 369 + }, + { + "epoch": 0.4054794520547945, + "grad_norm": 1.9305835962295532, + "learning_rate": 4.833468724614135e-06, + "loss": 2.2726, + "step": 370 + }, + { + "epoch": 0.4054794520547945, + "eval_loss": 2.3741114139556885, + "eval_runtime": 49.2956, + "eval_samples_per_second": 20.286, + "eval_steps_per_second": 0.852, + "step": 370 + }, + { + "epoch": 0.40657534246575344, + "grad_norm": 2.222024917602539, + "learning_rate": 4.831437855402113e-06, + "loss": 2.2634, + "step": 371 + }, + { + "epoch": 0.4076712328767123, + "grad_norm": 2.473958969116211, + "learning_rate": 4.82940698619009e-06, + "loss": 2.1817, + "step": 372 + }, + { + "epoch": 0.40876712328767123, + "grad_norm": 1.9208838939666748, + "learning_rate": 4.827376116978067e-06, + "loss": 2.2353, + "step": 373 + }, + { + "epoch": 0.40986301369863015, + "grad_norm": 2.7226450443267822, + "learning_rate": 4.8253452477660444e-06, + "loss": 2.2831, + "step": 374 + }, + { + "epoch": 0.410958904109589, + "grad_norm": 1.9078222513198853, + "learning_rate": 4.823314378554022e-06, + "loss": 2.264, + "step": 375 + }, + { + "epoch": 0.41205479452054794, + "grad_norm": 2.3946311473846436, + "learning_rate": 4.821283509341998e-06, + "loss": 2.251, + "step": 376 + }, + { + "epoch": 0.41315068493150686, + "grad_norm": 1.847343921661377, + "learning_rate": 4.819252640129976e-06, + "loss": 2.2441, + "step": 377 + }, + { + "epoch": 0.41424657534246573, + "grad_norm": 2.475385904312134, + "learning_rate": 4.817221770917953e-06, + "loss": 2.2639, + "step": 378 + }, + { + "epoch": 0.41534246575342465, + "grad_norm": 2.1396560668945312, + "learning_rate": 4.815190901705931e-06, + "loss": 2.2353, + "step": 379 + }, + { + "epoch": 0.41643835616438357, + "grad_norm": 2.3868885040283203, + "learning_rate": 4.813160032493907e-06, + "loss": 2.2551, + "step": 380 + }, + { + "epoch": 0.41643835616438357, + "eval_loss": 2.369168519973755, + "eval_runtime": 45.6131, + "eval_samples_per_second": 21.924, + "eval_steps_per_second": 0.921, + "step": 380 + }, + { + "epoch": 0.41753424657534244, + "grad_norm": 2.0193872451782227, + "learning_rate": 4.811129163281885e-06, + "loss": 2.2399, + "step": 381 + }, + { + "epoch": 0.41863013698630136, + "grad_norm": 2.4655935764312744, + "learning_rate": 4.809098294069862e-06, + "loss": 2.2729, + "step": 382 + }, + { + "epoch": 0.4197260273972603, + "grad_norm": 1.880974292755127, + "learning_rate": 4.80706742485784e-06, + "loss": 2.2681, + "step": 383 + }, + { + "epoch": 0.4208219178082192, + "grad_norm": 2.1912078857421875, + "learning_rate": 4.805036555645816e-06, + "loss": 2.2961, + "step": 384 + }, + { + "epoch": 0.42191780821917807, + "grad_norm": 1.985931634902954, + "learning_rate": 4.8030056864337945e-06, + "loss": 2.2407, + "step": 385 + }, + { + "epoch": 0.423013698630137, + "grad_norm": 2.4353220462799072, + "learning_rate": 4.800974817221771e-06, + "loss": 2.2867, + "step": 386 + }, + { + "epoch": 0.4241095890410959, + "grad_norm": 1.8810508251190186, + "learning_rate": 4.798943948009749e-06, + "loss": 2.2498, + "step": 387 + }, + { + "epoch": 0.4252054794520548, + "grad_norm": 2.195033073425293, + "learning_rate": 4.7969130787977254e-06, + "loss": 2.2532, + "step": 388 + }, + { + "epoch": 0.4263013698630137, + "grad_norm": 1.855372428894043, + "learning_rate": 4.794882209585703e-06, + "loss": 2.2538, + "step": 389 + }, + { + "epoch": 0.4273972602739726, + "grad_norm": 2.4465434551239014, + "learning_rate": 4.79285134037368e-06, + "loss": 2.2724, + "step": 390 + }, + { + "epoch": 0.4273972602739726, + "eval_loss": 2.364928960800171, + "eval_runtime": 48.0245, + "eval_samples_per_second": 20.823, + "eval_steps_per_second": 0.875, + "step": 390 + }, + { + "epoch": 0.4284931506849315, + "grad_norm": 1.8911314010620117, + "learning_rate": 4.790820471161657e-06, + "loss": 2.2291, + "step": 391 + }, + { + "epoch": 0.4295890410958904, + "grad_norm": 2.1975977420806885, + "learning_rate": 4.7887896019496345e-06, + "loss": 2.2373, + "step": 392 + }, + { + "epoch": 0.43068493150684933, + "grad_norm": 2.015331506729126, + "learning_rate": 4.786758732737612e-06, + "loss": 2.2483, + "step": 393 + }, + { + "epoch": 0.4317808219178082, + "grad_norm": 2.039339065551758, + "learning_rate": 4.784727863525589e-06, + "loss": 2.2746, + "step": 394 + }, + { + "epoch": 0.4328767123287671, + "grad_norm": 1.977513074874878, + "learning_rate": 4.782696994313566e-06, + "loss": 2.2275, + "step": 395 + }, + { + "epoch": 0.43397260273972604, + "grad_norm": 2.147217035293579, + "learning_rate": 4.780666125101544e-06, + "loss": 2.252, + "step": 396 + }, + { + "epoch": 0.4350684931506849, + "grad_norm": 2.1799919605255127, + "learning_rate": 4.778635255889521e-06, + "loss": 2.1975, + "step": 397 + }, + { + "epoch": 0.43616438356164383, + "grad_norm": 2.582240343093872, + "learning_rate": 4.776604386677498e-06, + "loss": 2.2294, + "step": 398 + }, + { + "epoch": 0.43726027397260275, + "grad_norm": 1.9127846956253052, + "learning_rate": 4.7745735174654755e-06, + "loss": 2.2319, + "step": 399 + }, + { + "epoch": 0.4383561643835616, + "grad_norm": 2.2444543838500977, + "learning_rate": 4.772542648253453e-06, + "loss": 2.2857, + "step": 400 + }, + { + "epoch": 0.4383561643835616, + "eval_loss": 2.3653724193573, + "eval_runtime": 43.8304, + "eval_samples_per_second": 22.815, + "eval_steps_per_second": 0.958, + "step": 400 + }, + { + "epoch": 0.43945205479452054, + "grad_norm": 1.9913676977157593, + "learning_rate": 4.77051177904143e-06, + "loss": 2.274, + "step": 401 + }, + { + "epoch": 0.44054794520547946, + "grad_norm": 2.2850067615509033, + "learning_rate": 4.768480909829407e-06, + "loss": 2.2209, + "step": 402 + }, + { + "epoch": 0.44164383561643833, + "grad_norm": 2.4984688758850098, + "learning_rate": 4.766450040617385e-06, + "loss": 2.263, + "step": 403 + }, + { + "epoch": 0.44273972602739725, + "grad_norm": 2.3244640827178955, + "learning_rate": 4.764419171405362e-06, + "loss": 2.2368, + "step": 404 + }, + { + "epoch": 0.4438356164383562, + "grad_norm": 2.7579681873321533, + "learning_rate": 4.762388302193339e-06, + "loss": 2.2762, + "step": 405 + }, + { + "epoch": 0.4449315068493151, + "grad_norm": 2.1996467113494873, + "learning_rate": 4.760357432981316e-06, + "loss": 2.2536, + "step": 406 + }, + { + "epoch": 0.44602739726027396, + "grad_norm": 2.3767571449279785, + "learning_rate": 4.758326563769294e-06, + "loss": 2.2353, + "step": 407 + }, + { + "epoch": 0.4471232876712329, + "grad_norm": 1.9148359298706055, + "learning_rate": 4.756295694557271e-06, + "loss": 2.2416, + "step": 408 + }, + { + "epoch": 0.4482191780821918, + "grad_norm": 2.6499507427215576, + "learning_rate": 4.754264825345248e-06, + "loss": 2.2641, + "step": 409 + }, + { + "epoch": 0.44931506849315067, + "grad_norm": 1.9890286922454834, + "learning_rate": 4.7522339561332255e-06, + "loss": 2.2148, + "step": 410 + }, + { + "epoch": 0.44931506849315067, + "eval_loss": 2.3670520782470703, + "eval_runtime": 41.1722, + "eval_samples_per_second": 24.288, + "eval_steps_per_second": 1.02, + "step": 410 + }, + { + "epoch": 0.4504109589041096, + "grad_norm": 2.575968027114868, + "learning_rate": 4.750203086921203e-06, + "loss": 2.2254, + "step": 411 + }, + { + "epoch": 0.4515068493150685, + "grad_norm": 2.015174388885498, + "learning_rate": 4.74817221770918e-06, + "loss": 2.1881, + "step": 412 + }, + { + "epoch": 0.4526027397260274, + "grad_norm": 2.44292950630188, + "learning_rate": 4.7461413484971565e-06, + "loss": 2.256, + "step": 413 + }, + { + "epoch": 0.4536986301369863, + "grad_norm": 1.9808441400527954, + "learning_rate": 4.744110479285135e-06, + "loss": 2.285, + "step": 414 + }, + { + "epoch": 0.4547945205479452, + "grad_norm": 2.009481906890869, + "learning_rate": 4.742079610073111e-06, + "loss": 2.2018, + "step": 415 + }, + { + "epoch": 0.4558904109589041, + "grad_norm": 2.1466031074523926, + "learning_rate": 4.740048740861089e-06, + "loss": 2.224, + "step": 416 + }, + { + "epoch": 0.456986301369863, + "grad_norm": 2.1337907314300537, + "learning_rate": 4.738017871649066e-06, + "loss": 2.2404, + "step": 417 + }, + { + "epoch": 0.45808219178082193, + "grad_norm": 1.94125235080719, + "learning_rate": 4.735987002437044e-06, + "loss": 2.2474, + "step": 418 + }, + { + "epoch": 0.4591780821917808, + "grad_norm": 2.379099130630493, + "learning_rate": 4.73395613322502e-06, + "loss": 2.2858, + "step": 419 + }, + { + "epoch": 0.4602739726027397, + "grad_norm": 1.844056248664856, + "learning_rate": 4.731925264012998e-06, + "loss": 2.2967, + "step": 420 + }, + { + "epoch": 0.4602739726027397, + "eval_loss": 2.3622241020202637, + "eval_runtime": 45.4674, + "eval_samples_per_second": 21.994, + "eval_steps_per_second": 0.924, + "step": 420 + }, + { + "epoch": 0.46136986301369864, + "grad_norm": 2.411895513534546, + "learning_rate": 4.729894394800975e-06, + "loss": 2.274, + "step": 421 + }, + { + "epoch": 0.4624657534246575, + "grad_norm": 2.014885425567627, + "learning_rate": 4.727863525588953e-06, + "loss": 2.2144, + "step": 422 + }, + { + "epoch": 0.46356164383561643, + "grad_norm": 2.112009048461914, + "learning_rate": 4.725832656376929e-06, + "loss": 2.1824, + "step": 423 + }, + { + "epoch": 0.46465753424657535, + "grad_norm": 2.111392021179199, + "learning_rate": 4.723801787164907e-06, + "loss": 2.2261, + "step": 424 + }, + { + "epoch": 0.4657534246575342, + "grad_norm": 2.0349185466766357, + "learning_rate": 4.721770917952885e-06, + "loss": 2.2622, + "step": 425 + }, + { + "epoch": 0.46684931506849314, + "grad_norm": 2.30222487449646, + "learning_rate": 4.719740048740861e-06, + "loss": 2.2206, + "step": 426 + }, + { + "epoch": 0.46794520547945206, + "grad_norm": 2.018789529800415, + "learning_rate": 4.717709179528839e-06, + "loss": 2.2196, + "step": 427 + }, + { + "epoch": 0.469041095890411, + "grad_norm": 2.046649932861328, + "learning_rate": 4.715678310316816e-06, + "loss": 2.2584, + "step": 428 + }, + { + "epoch": 0.47013698630136985, + "grad_norm": 2.0393869876861572, + "learning_rate": 4.713647441104794e-06, + "loss": 2.2537, + "step": 429 + }, + { + "epoch": 0.4712328767123288, + "grad_norm": 2.2788116931915283, + "learning_rate": 4.71161657189277e-06, + "loss": 2.2145, + "step": 430 + }, + { + "epoch": 0.4712328767123288, + "eval_loss": 2.3575785160064697, + "eval_runtime": 44.6687, + "eval_samples_per_second": 22.387, + "eval_steps_per_second": 0.94, + "step": 430 + }, + { + "epoch": 0.4723287671232877, + "grad_norm": 1.9950047731399536, + "learning_rate": 4.709585702680748e-06, + "loss": 2.2072, + "step": 431 + }, + { + "epoch": 0.47342465753424656, + "grad_norm": 1.9222451448440552, + "learning_rate": 4.707554833468725e-06, + "loss": 2.2654, + "step": 432 + }, + { + "epoch": 0.4745205479452055, + "grad_norm": 1.9735718965530396, + "learning_rate": 4.705523964256703e-06, + "loss": 2.3037, + "step": 433 + }, + { + "epoch": 0.4756164383561644, + "grad_norm": 2.3209776878356934, + "learning_rate": 4.703493095044679e-06, + "loss": 2.2083, + "step": 434 + }, + { + "epoch": 0.4767123287671233, + "grad_norm": 2.0993995666503906, + "learning_rate": 4.701462225832657e-06, + "loss": 2.2275, + "step": 435 + }, + { + "epoch": 0.4778082191780822, + "grad_norm": 2.334463119506836, + "learning_rate": 4.699431356620634e-06, + "loss": 2.1852, + "step": 436 + }, + { + "epoch": 0.4789041095890411, + "grad_norm": 1.9098213911056519, + "learning_rate": 4.697400487408612e-06, + "loss": 2.2829, + "step": 437 + }, + { + "epoch": 0.48, + "grad_norm": 2.3005056381225586, + "learning_rate": 4.695369618196588e-06, + "loss": 2.2532, + "step": 438 + }, + { + "epoch": 0.4810958904109589, + "grad_norm": 1.9069781303405762, + "learning_rate": 4.693338748984566e-06, + "loss": 2.2374, + "step": 439 + }, + { + "epoch": 0.4821917808219178, + "grad_norm": 2.050032138824463, + "learning_rate": 4.691307879772543e-06, + "loss": 2.2256, + "step": 440 + }, + { + "epoch": 0.4821917808219178, + "eval_loss": 2.3587422370910645, + "eval_runtime": 45.0941, + "eval_samples_per_second": 22.176, + "eval_steps_per_second": 0.931, + "step": 440 + }, + { + "epoch": 0.4832876712328767, + "grad_norm": 2.0563836097717285, + "learning_rate": 4.68927701056052e-06, + "loss": 2.1918, + "step": 441 + }, + { + "epoch": 0.4843835616438356, + "grad_norm": 2.1598386764526367, + "learning_rate": 4.6872461413484975e-06, + "loss": 2.2618, + "step": 442 + }, + { + "epoch": 0.48547945205479454, + "grad_norm": 1.9236539602279663, + "learning_rate": 4.685215272136475e-06, + "loss": 2.2477, + "step": 443 + }, + { + "epoch": 0.4865753424657534, + "grad_norm": 1.9923924207687378, + "learning_rate": 4.683184402924452e-06, + "loss": 2.2313, + "step": 444 + }, + { + "epoch": 0.4876712328767123, + "grad_norm": 2.162627696990967, + "learning_rate": 4.681153533712429e-06, + "loss": 2.2367, + "step": 445 + }, + { + "epoch": 0.48876712328767125, + "grad_norm": 2.083523988723755, + "learning_rate": 4.679122664500407e-06, + "loss": 2.2946, + "step": 446 + }, + { + "epoch": 0.4898630136986301, + "grad_norm": 2.504408359527588, + "learning_rate": 4.677091795288384e-06, + "loss": 2.2104, + "step": 447 + }, + { + "epoch": 0.49095890410958903, + "grad_norm": 1.8316532373428345, + "learning_rate": 4.675060926076361e-06, + "loss": 2.2027, + "step": 448 + }, + { + "epoch": 0.49205479452054796, + "grad_norm": 2.3337953090667725, + "learning_rate": 4.673030056864338e-06, + "loss": 2.278, + "step": 449 + }, + { + "epoch": 0.4931506849315068, + "grad_norm": 2.24102783203125, + "learning_rate": 4.670999187652316e-06, + "loss": 2.2018, + "step": 450 + }, + { + "epoch": 0.4931506849315068, + "eval_loss": 2.3545279502868652, + "eval_runtime": 45.1041, + "eval_samples_per_second": 22.171, + "eval_steps_per_second": 0.931, + "step": 450 + }, + { + "epoch": 0.49424657534246574, + "grad_norm": 2.07834792137146, + "learning_rate": 4.668968318440293e-06, + "loss": 2.1975, + "step": 451 + }, + { + "epoch": 0.49534246575342467, + "grad_norm": 1.7681723833084106, + "learning_rate": 4.66693744922827e-06, + "loss": 2.2033, + "step": 452 + }, + { + "epoch": 0.4964383561643836, + "grad_norm": 2.062134027481079, + "learning_rate": 4.6649065800162475e-06, + "loss": 2.2646, + "step": 453 + }, + { + "epoch": 0.49753424657534245, + "grad_norm": 1.94647216796875, + "learning_rate": 4.662875710804225e-06, + "loss": 2.1748, + "step": 454 + }, + { + "epoch": 0.4986301369863014, + "grad_norm": 1.9173872470855713, + "learning_rate": 4.660844841592202e-06, + "loss": 2.2435, + "step": 455 + }, + { + "epoch": 0.4997260273972603, + "grad_norm": 1.88880455493927, + "learning_rate": 4.658813972380179e-06, + "loss": 2.1711, + "step": 456 + }, + { + "epoch": 0.5008219178082192, + "grad_norm": 1.9628506898880005, + "learning_rate": 4.656783103168157e-06, + "loss": 2.2701, + "step": 457 + }, + { + "epoch": 0.5019178082191781, + "grad_norm": 2.0515286922454834, + "learning_rate": 4.654752233956134e-06, + "loss": 2.1656, + "step": 458 + }, + { + "epoch": 0.503013698630137, + "grad_norm": 1.978522539138794, + "learning_rate": 4.652721364744111e-06, + "loss": 2.246, + "step": 459 + }, + { + "epoch": 0.5041095890410959, + "grad_norm": 1.7993966341018677, + "learning_rate": 4.6506904955320885e-06, + "loss": 2.214, + "step": 460 + }, + { + "epoch": 0.5041095890410959, + "eval_loss": 2.3568837642669678, + "eval_runtime": 45.2086, + "eval_samples_per_second": 22.12, + "eval_steps_per_second": 0.929, + "step": 460 + }, + { + "epoch": 0.5052054794520547, + "grad_norm": 1.7809834480285645, + "learning_rate": 4.648659626320066e-06, + "loss": 2.2525, + "step": 461 + }, + { + "epoch": 0.5063013698630137, + "grad_norm": 1.9256166219711304, + "learning_rate": 4.646628757108043e-06, + "loss": 2.2611, + "step": 462 + }, + { + "epoch": 0.5073972602739726, + "grad_norm": 1.94192636013031, + "learning_rate": 4.6445978878960194e-06, + "loss": 2.2282, + "step": 463 + }, + { + "epoch": 0.5084931506849315, + "grad_norm": 1.8891818523406982, + "learning_rate": 4.6425670186839976e-06, + "loss": 2.2036, + "step": 464 + }, + { + "epoch": 0.5095890410958904, + "grad_norm": 2.352078676223755, + "learning_rate": 4.640536149471974e-06, + "loss": 2.2293, + "step": 465 + }, + { + "epoch": 0.5106849315068493, + "grad_norm": 1.8141659498214722, + "learning_rate": 4.638505280259952e-06, + "loss": 2.2551, + "step": 466 + }, + { + "epoch": 0.5117808219178083, + "grad_norm": 1.9492480754852295, + "learning_rate": 4.6364744110479285e-06, + "loss": 2.2641, + "step": 467 + }, + { + "epoch": 0.5128767123287671, + "grad_norm": 2.0210182666778564, + "learning_rate": 4.634443541835907e-06, + "loss": 2.2367, + "step": 468 + }, + { + "epoch": 0.513972602739726, + "grad_norm": 1.8936505317687988, + "learning_rate": 4.632412672623883e-06, + "loss": 2.265, + "step": 469 + }, + { + "epoch": 0.5150684931506849, + "grad_norm": 1.845888614654541, + "learning_rate": 4.630381803411861e-06, + "loss": 2.2169, + "step": 470 + }, + { + "epoch": 0.5150684931506849, + "eval_loss": 2.353489875793457, + "eval_runtime": 46.8739, + "eval_samples_per_second": 21.334, + "eval_steps_per_second": 0.896, + "step": 470 + }, + { + "epoch": 0.5161643835616438, + "grad_norm": 1.889675259590149, + "learning_rate": 4.628350934199838e-06, + "loss": 2.1383, + "step": 471 + }, + { + "epoch": 0.5172602739726028, + "grad_norm": 1.9799216985702515, + "learning_rate": 4.626320064987816e-06, + "loss": 2.2312, + "step": 472 + }, + { + "epoch": 0.5183561643835617, + "grad_norm": 2.0128607749938965, + "learning_rate": 4.624289195775792e-06, + "loss": 2.2529, + "step": 473 + }, + { + "epoch": 0.5194520547945205, + "grad_norm": 1.8791241645812988, + "learning_rate": 4.62225832656377e-06, + "loss": 2.226, + "step": 474 + }, + { + "epoch": 0.5205479452054794, + "grad_norm": 2.0497279167175293, + "learning_rate": 4.620227457351747e-06, + "loss": 2.2283, + "step": 475 + }, + { + "epoch": 0.5216438356164383, + "grad_norm": 1.8097424507141113, + "learning_rate": 4.618196588139724e-06, + "loss": 2.222, + "step": 476 + }, + { + "epoch": 0.5227397260273973, + "grad_norm": 1.8960500955581665, + "learning_rate": 4.616165718927701e-06, + "loss": 2.2342, + "step": 477 + }, + { + "epoch": 0.5238356164383562, + "grad_norm": 2.020730495452881, + "learning_rate": 4.6141348497156786e-06, + "loss": 2.2106, + "step": 478 + }, + { + "epoch": 0.5249315068493151, + "grad_norm": 1.8955297470092773, + "learning_rate": 4.612103980503656e-06, + "loss": 2.2755, + "step": 479 + }, + { + "epoch": 0.5260273972602739, + "grad_norm": 1.9767351150512695, + "learning_rate": 4.610073111291633e-06, + "loss": 2.2554, + "step": 480 + }, + { + "epoch": 0.5260273972602739, + "eval_loss": 2.347705841064453, + "eval_runtime": 46.9125, + "eval_samples_per_second": 21.316, + "eval_steps_per_second": 0.895, + "step": 480 + }, + { + "epoch": 0.5271232876712328, + "grad_norm": 1.936457633972168, + "learning_rate": 4.60804224207961e-06, + "loss": 2.2323, + "step": 481 + }, + { + "epoch": 0.5282191780821918, + "grad_norm": 1.9271862506866455, + "learning_rate": 4.606011372867588e-06, + "loss": 2.2151, + "step": 482 + }, + { + "epoch": 0.5293150684931507, + "grad_norm": 1.878153681755066, + "learning_rate": 4.603980503655565e-06, + "loss": 2.2322, + "step": 483 + }, + { + "epoch": 0.5304109589041096, + "grad_norm": 2.1009249687194824, + "learning_rate": 4.601949634443542e-06, + "loss": 2.2181, + "step": 484 + }, + { + "epoch": 0.5315068493150685, + "grad_norm": 1.9089181423187256, + "learning_rate": 4.5999187652315195e-06, + "loss": 2.1712, + "step": 485 + }, + { + "epoch": 0.5326027397260275, + "grad_norm": 2.207282781600952, + "learning_rate": 4.597887896019497e-06, + "loss": 2.1847, + "step": 486 + }, + { + "epoch": 0.5336986301369863, + "grad_norm": 2.001584053039551, + "learning_rate": 4.595857026807474e-06, + "loss": 2.214, + "step": 487 + }, + { + "epoch": 0.5347945205479452, + "grad_norm": 1.966700553894043, + "learning_rate": 4.593826157595451e-06, + "loss": 2.2237, + "step": 488 + }, + { + "epoch": 0.5358904109589041, + "grad_norm": 1.842469334602356, + "learning_rate": 4.591795288383429e-06, + "loss": 2.2089, + "step": 489 + }, + { + "epoch": 0.536986301369863, + "grad_norm": 2.0412986278533936, + "learning_rate": 4.589764419171406e-06, + "loss": 2.2734, + "step": 490 + }, + { + "epoch": 0.536986301369863, + "eval_loss": 2.3460772037506104, + "eval_runtime": 50.4048, + "eval_samples_per_second": 19.839, + "eval_steps_per_second": 0.833, + "step": 490 + }, + { + "epoch": 0.538082191780822, + "grad_norm": 1.9337031841278076, + "learning_rate": 4.587733549959383e-06, + "loss": 2.2314, + "step": 491 + }, + { + "epoch": 0.5391780821917809, + "grad_norm": 2.4448320865631104, + "learning_rate": 4.5857026807473604e-06, + "loss": 2.2098, + "step": 492 + }, + { + "epoch": 0.5402739726027397, + "grad_norm": 1.83162522315979, + "learning_rate": 4.583671811535338e-06, + "loss": 2.1902, + "step": 493 + }, + { + "epoch": 0.5413698630136986, + "grad_norm": 2.3614656925201416, + "learning_rate": 4.581640942323315e-06, + "loss": 2.1737, + "step": 494 + }, + { + "epoch": 0.5424657534246575, + "grad_norm": 2.001187801361084, + "learning_rate": 4.579610073111292e-06, + "loss": 2.2485, + "step": 495 + }, + { + "epoch": 0.5435616438356164, + "grad_norm": 2.3479933738708496, + "learning_rate": 4.5775792038992695e-06, + "loss": 2.2595, + "step": 496 + }, + { + "epoch": 0.5446575342465754, + "grad_norm": 2.135856866836548, + "learning_rate": 4.575548334687247e-06, + "loss": 2.2528, + "step": 497 + }, + { + "epoch": 0.5457534246575343, + "grad_norm": 1.983126163482666, + "learning_rate": 4.573517465475224e-06, + "loss": 2.2788, + "step": 498 + }, + { + "epoch": 0.5468493150684931, + "grad_norm": 2.18208909034729, + "learning_rate": 4.571486596263201e-06, + "loss": 2.1995, + "step": 499 + }, + { + "epoch": 0.547945205479452, + "grad_norm": 1.848747968673706, + "learning_rate": 4.569455727051178e-06, + "loss": 2.2403, + "step": 500 + }, + { + "epoch": 0.547945205479452, + "eval_loss": 2.344066619873047, + "eval_runtime": 51.1213, + "eval_samples_per_second": 19.561, + "eval_steps_per_second": 0.822, + "step": 500 + }, + { + "epoch": 0.549041095890411, + "grad_norm": 2.427860736846924, + "learning_rate": 4.567424857839156e-06, + "loss": 2.246, + "step": 501 + }, + { + "epoch": 0.5501369863013699, + "grad_norm": 1.913405418395996, + "learning_rate": 4.565393988627132e-06, + "loss": 2.2296, + "step": 502 + }, + { + "epoch": 0.5512328767123288, + "grad_norm": 2.0106029510498047, + "learning_rate": 4.5633631194151105e-06, + "loss": 2.2657, + "step": 503 + }, + { + "epoch": 0.5523287671232877, + "grad_norm": 2.2719595432281494, + "learning_rate": 4.561332250203087e-06, + "loss": 2.2341, + "step": 504 + }, + { + "epoch": 0.5534246575342465, + "grad_norm": 2.109786033630371, + "learning_rate": 4.559301380991065e-06, + "loss": 2.2181, + "step": 505 + }, + { + "epoch": 0.5545205479452054, + "grad_norm": 2.189779281616211, + "learning_rate": 4.5572705117790414e-06, + "loss": 2.1821, + "step": 506 + }, + { + "epoch": 0.5556164383561644, + "grad_norm": 2.0263590812683105, + "learning_rate": 4.5552396425670196e-06, + "loss": 2.2072, + "step": 507 + }, + { + "epoch": 0.5567123287671233, + "grad_norm": 2.215289354324341, + "learning_rate": 4.553208773354996e-06, + "loss": 2.2513, + "step": 508 + }, + { + "epoch": 0.5578082191780822, + "grad_norm": 2.3031351566314697, + "learning_rate": 4.551177904142974e-06, + "loss": 2.1988, + "step": 509 + }, + { + "epoch": 0.5589041095890411, + "grad_norm": 2.1830058097839355, + "learning_rate": 4.5491470349309505e-06, + "loss": 2.269, + "step": 510 + }, + { + "epoch": 0.5589041095890411, + "eval_loss": 2.342463731765747, + "eval_runtime": 43.1956, + "eval_samples_per_second": 23.151, + "eval_steps_per_second": 0.972, + "step": 510 + }, + { + "epoch": 0.56, + "grad_norm": 2.3210105895996094, + "learning_rate": 4.547116165718929e-06, + "loss": 2.2051, + "step": 511 + }, + { + "epoch": 0.5610958904109589, + "grad_norm": 1.9175361394882202, + "learning_rate": 4.545085296506905e-06, + "loss": 2.2304, + "step": 512 + }, + { + "epoch": 0.5621917808219178, + "grad_norm": 2.302947759628296, + "learning_rate": 4.543054427294882e-06, + "loss": 2.215, + "step": 513 + }, + { + "epoch": 0.5632876712328767, + "grad_norm": 1.8947396278381348, + "learning_rate": 4.54102355808286e-06, + "loss": 2.1926, + "step": 514 + }, + { + "epoch": 0.5643835616438356, + "grad_norm": 2.195268154144287, + "learning_rate": 4.538992688870837e-06, + "loss": 2.1937, + "step": 515 + }, + { + "epoch": 0.5654794520547946, + "grad_norm": 2.178874969482422, + "learning_rate": 4.536961819658814e-06, + "loss": 2.2248, + "step": 516 + }, + { + "epoch": 0.5665753424657535, + "grad_norm": 1.9856929779052734, + "learning_rate": 4.5349309504467915e-06, + "loss": 2.2077, + "step": 517 + }, + { + "epoch": 0.5676712328767123, + "grad_norm": 2.0686943531036377, + "learning_rate": 4.532900081234769e-06, + "loss": 2.2328, + "step": 518 + }, + { + "epoch": 0.5687671232876712, + "grad_norm": 1.9554919004440308, + "learning_rate": 4.530869212022746e-06, + "loss": 2.198, + "step": 519 + }, + { + "epoch": 0.5698630136986301, + "grad_norm": 1.9489670991897583, + "learning_rate": 4.528838342810723e-06, + "loss": 2.2287, + "step": 520 + }, + { + "epoch": 0.5698630136986301, + "eval_loss": 2.3415298461914062, + "eval_runtime": 48.7918, + "eval_samples_per_second": 20.495, + "eval_steps_per_second": 0.861, + "step": 520 + }, + { + "epoch": 0.570958904109589, + "grad_norm": 1.9868658781051636, + "learning_rate": 4.526807473598701e-06, + "loss": 2.2205, + "step": 521 + }, + { + "epoch": 0.572054794520548, + "grad_norm": 2.0965914726257324, + "learning_rate": 4.524776604386678e-06, + "loss": 2.2239, + "step": 522 + }, + { + "epoch": 0.5731506849315069, + "grad_norm": 2.041471481323242, + "learning_rate": 4.522745735174655e-06, + "loss": 2.2035, + "step": 523 + }, + { + "epoch": 0.5742465753424657, + "grad_norm": 1.885293960571289, + "learning_rate": 4.520714865962632e-06, + "loss": 2.2469, + "step": 524 + }, + { + "epoch": 0.5753424657534246, + "grad_norm": 1.9107296466827393, + "learning_rate": 4.51868399675061e-06, + "loss": 2.2386, + "step": 525 + }, + { + "epoch": 0.5764383561643835, + "grad_norm": 1.9945271015167236, + "learning_rate": 4.516653127538587e-06, + "loss": 2.235, + "step": 526 + }, + { + "epoch": 0.5775342465753425, + "grad_norm": 1.9787253141403198, + "learning_rate": 4.514622258326564e-06, + "loss": 2.2034, + "step": 527 + }, + { + "epoch": 0.5786301369863014, + "grad_norm": 2.0300636291503906, + "learning_rate": 4.5125913891145415e-06, + "loss": 2.1849, + "step": 528 + }, + { + "epoch": 0.5797260273972603, + "grad_norm": 2.0348970890045166, + "learning_rate": 4.510560519902519e-06, + "loss": 2.2174, + "step": 529 + }, + { + "epoch": 0.5808219178082191, + "grad_norm": 1.8589537143707275, + "learning_rate": 4.508529650690496e-06, + "loss": 2.2351, + "step": 530 + }, + { + "epoch": 0.5808219178082191, + "eval_loss": 2.3450186252593994, + "eval_runtime": 52.4548, + "eval_samples_per_second": 19.064, + "eval_steps_per_second": 0.801, + "step": 530 + }, + { + "epoch": 0.581917808219178, + "grad_norm": 2.0496020317077637, + "learning_rate": 4.506498781478473e-06, + "loss": 2.182, + "step": 531 + }, + { + "epoch": 0.583013698630137, + "grad_norm": 2.0649805068969727, + "learning_rate": 4.504467912266451e-06, + "loss": 2.1654, + "step": 532 + }, + { + "epoch": 0.5841095890410959, + "grad_norm": 1.9809486865997314, + "learning_rate": 4.502437043054428e-06, + "loss": 2.2182, + "step": 533 + }, + { + "epoch": 0.5852054794520548, + "grad_norm": 2.21915340423584, + "learning_rate": 4.500406173842405e-06, + "loss": 2.2697, + "step": 534 + }, + { + "epoch": 0.5863013698630137, + "grad_norm": 1.9822649955749512, + "learning_rate": 4.4983753046303824e-06, + "loss": 2.2297, + "step": 535 + }, + { + "epoch": 0.5873972602739727, + "grad_norm": 1.972575306892395, + "learning_rate": 4.49634443541836e-06, + "loss": 2.2164, + "step": 536 + }, + { + "epoch": 0.5884931506849315, + "grad_norm": 1.948667049407959, + "learning_rate": 4.494313566206336e-06, + "loss": 2.2087, + "step": 537 + }, + { + "epoch": 0.5895890410958904, + "grad_norm": 2.7486417293548584, + "learning_rate": 4.492282696994314e-06, + "loss": 2.2602, + "step": 538 + }, + { + "epoch": 0.5906849315068493, + "grad_norm": 2.3374552726745605, + "learning_rate": 4.490251827782291e-06, + "loss": 2.2054, + "step": 539 + }, + { + "epoch": 0.5917808219178082, + "grad_norm": 2.222116231918335, + "learning_rate": 4.488220958570269e-06, + "loss": 2.1868, + "step": 540 + }, + { + "epoch": 0.5917808219178082, + "eval_loss": 2.345618963241577, + "eval_runtime": 53.0302, + "eval_samples_per_second": 18.857, + "eval_steps_per_second": 0.792, + "step": 540 + }, + { + "epoch": 0.5928767123287672, + "grad_norm": 2.00752854347229, + "learning_rate": 4.486190089358245e-06, + "loss": 2.187, + "step": 541 + }, + { + "epoch": 0.5939726027397261, + "grad_norm": 2.2466132640838623, + "learning_rate": 4.484159220146223e-06, + "loss": 2.1611, + "step": 542 + }, + { + "epoch": 0.5950684931506849, + "grad_norm": 2.1279385089874268, + "learning_rate": 4.4821283509342e-06, + "loss": 2.248, + "step": 543 + }, + { + "epoch": 0.5961643835616438, + "grad_norm": 1.9140571355819702, + "learning_rate": 4.480097481722178e-06, + "loss": 2.2314, + "step": 544 + }, + { + "epoch": 0.5972602739726027, + "grad_norm": 2.120741605758667, + "learning_rate": 4.478066612510154e-06, + "loss": 2.212, + "step": 545 + }, + { + "epoch": 0.5983561643835617, + "grad_norm": 2.003713369369507, + "learning_rate": 4.4760357432981325e-06, + "loss": 2.1824, + "step": 546 + }, + { + "epoch": 0.5994520547945206, + "grad_norm": 1.9161221981048584, + "learning_rate": 4.474004874086109e-06, + "loss": 2.2015, + "step": 547 + }, + { + "epoch": 0.6005479452054795, + "grad_norm": 2.367110252380371, + "learning_rate": 4.471974004874087e-06, + "loss": 2.3122, + "step": 548 + }, + { + "epoch": 0.6016438356164383, + "grad_norm": 2.113380193710327, + "learning_rate": 4.4699431356620635e-06, + "loss": 2.1692, + "step": 549 + }, + { + "epoch": 0.6027397260273972, + "grad_norm": 1.974181890487671, + "learning_rate": 4.467912266450041e-06, + "loss": 2.2509, + "step": 550 + }, + { + "epoch": 0.6027397260273972, + "eval_loss": 2.342583656311035, + "eval_runtime": 50.7362, + "eval_samples_per_second": 19.71, + "eval_steps_per_second": 0.828, + "step": 550 + }, + { + "epoch": 0.6038356164383561, + "grad_norm": 1.889894962310791, + "learning_rate": 4.465881397238018e-06, + "loss": 2.2214, + "step": 551 + }, + { + "epoch": 0.6049315068493151, + "grad_norm": 1.8982343673706055, + "learning_rate": 4.463850528025995e-06, + "loss": 2.2135, + "step": 552 + }, + { + "epoch": 0.606027397260274, + "grad_norm": 1.9462087154388428, + "learning_rate": 4.4618196588139726e-06, + "loss": 2.2255, + "step": 553 + }, + { + "epoch": 0.6071232876712329, + "grad_norm": 1.8729931116104126, + "learning_rate": 4.45978878960195e-06, + "loss": 2.2358, + "step": 554 + }, + { + "epoch": 0.6082191780821918, + "grad_norm": 1.9230246543884277, + "learning_rate": 4.457757920389927e-06, + "loss": 2.222, + "step": 555 + }, + { + "epoch": 0.6093150684931506, + "grad_norm": 1.9495900869369507, + "learning_rate": 4.455727051177904e-06, + "loss": 2.1779, + "step": 556 + }, + { + "epoch": 0.6104109589041096, + "grad_norm": 1.983487606048584, + "learning_rate": 4.453696181965882e-06, + "loss": 2.2007, + "step": 557 + }, + { + "epoch": 0.6115068493150685, + "grad_norm": 2.1541049480438232, + "learning_rate": 4.451665312753859e-06, + "loss": 2.2513, + "step": 558 + }, + { + "epoch": 0.6126027397260274, + "grad_norm": 1.9133352041244507, + "learning_rate": 4.449634443541836e-06, + "loss": 2.2288, + "step": 559 + }, + { + "epoch": 0.6136986301369863, + "grad_norm": 2.3325741291046143, + "learning_rate": 4.4476035743298135e-06, + "loss": 2.2539, + "step": 560 + }, + { + "epoch": 0.6136986301369863, + "eval_loss": 2.3387348651885986, + "eval_runtime": 53.4301, + "eval_samples_per_second": 18.716, + "eval_steps_per_second": 0.786, + "step": 560 + }, + { + "epoch": 0.6147945205479453, + "grad_norm": 1.8099673986434937, + "learning_rate": 4.445572705117791e-06, + "loss": 2.1845, + "step": 561 + }, + { + "epoch": 0.6158904109589041, + "grad_norm": 2.201901912689209, + "learning_rate": 4.443541835905768e-06, + "loss": 2.195, + "step": 562 + }, + { + "epoch": 0.616986301369863, + "grad_norm": 1.8412714004516602, + "learning_rate": 4.441510966693745e-06, + "loss": 2.2551, + "step": 563 + }, + { + "epoch": 0.6180821917808219, + "grad_norm": 1.8761520385742188, + "learning_rate": 4.439480097481723e-06, + "loss": 2.1965, + "step": 564 + }, + { + "epoch": 0.6191780821917808, + "grad_norm": 1.9123467206954956, + "learning_rate": 4.4374492282697e-06, + "loss": 2.2502, + "step": 565 + }, + { + "epoch": 0.6202739726027398, + "grad_norm": 2.0631234645843506, + "learning_rate": 4.435418359057677e-06, + "loss": 2.1582, + "step": 566 + }, + { + "epoch": 0.6213698630136987, + "grad_norm": 2.3191616535186768, + "learning_rate": 4.433387489845654e-06, + "loss": 2.2271, + "step": 567 + }, + { + "epoch": 0.6224657534246575, + "grad_norm": 2.043057918548584, + "learning_rate": 4.431356620633632e-06, + "loss": 2.2292, + "step": 568 + }, + { + "epoch": 0.6235616438356164, + "grad_norm": 2.0716068744659424, + "learning_rate": 4.429325751421609e-06, + "loss": 2.196, + "step": 569 + }, + { + "epoch": 0.6246575342465753, + "grad_norm": 1.7669776678085327, + "learning_rate": 4.427294882209586e-06, + "loss": 2.1953, + "step": 570 + }, + { + "epoch": 0.6246575342465753, + "eval_loss": 2.336578607559204, + "eval_runtime": 50.0083, + "eval_samples_per_second": 19.997, + "eval_steps_per_second": 0.84, + "step": 570 + }, + { + "epoch": 0.6257534246575343, + "grad_norm": 1.8165771961212158, + "learning_rate": 4.4252640129975635e-06, + "loss": 2.2155, + "step": 571 + }, + { + "epoch": 0.6268493150684932, + "grad_norm": 1.8640249967575073, + "learning_rate": 4.423233143785541e-06, + "loss": 2.1648, + "step": 572 + }, + { + "epoch": 0.6279452054794521, + "grad_norm": 2.013869285583496, + "learning_rate": 4.421202274573518e-06, + "loss": 2.2131, + "step": 573 + }, + { + "epoch": 0.6290410958904109, + "grad_norm": 2.2650580406188965, + "learning_rate": 4.4191714053614945e-06, + "loss": 2.2287, + "step": 574 + }, + { + "epoch": 0.6301369863013698, + "grad_norm": 1.9232103824615479, + "learning_rate": 4.417140536149473e-06, + "loss": 2.1339, + "step": 575 + }, + { + "epoch": 0.6312328767123287, + "grad_norm": 1.910134196281433, + "learning_rate": 4.415109666937449e-06, + "loss": 2.1755, + "step": 576 + }, + { + "epoch": 0.6323287671232877, + "grad_norm": 2.2492005825042725, + "learning_rate": 4.413078797725427e-06, + "loss": 2.2362, + "step": 577 + }, + { + "epoch": 0.6334246575342466, + "grad_norm": 1.8584033250808716, + "learning_rate": 4.411047928513404e-06, + "loss": 2.1911, + "step": 578 + }, + { + "epoch": 0.6345205479452055, + "grad_norm": 2.156273126602173, + "learning_rate": 4.409017059301382e-06, + "loss": 2.2109, + "step": 579 + }, + { + "epoch": 0.6356164383561644, + "grad_norm": 1.7977535724639893, + "learning_rate": 4.406986190089358e-06, + "loss": 2.2195, + "step": 580 + }, + { + "epoch": 0.6356164383561644, + "eval_loss": 2.33555269241333, + "eval_runtime": 48.3919, + "eval_samples_per_second": 20.665, + "eval_steps_per_second": 0.868, + "step": 580 + }, + { + "epoch": 0.6367123287671232, + "grad_norm": 1.8262702226638794, + "learning_rate": 4.404955320877336e-06, + "loss": 2.2345, + "step": 581 + }, + { + "epoch": 0.6378082191780822, + "grad_norm": 1.9389383792877197, + "learning_rate": 4.402924451665313e-06, + "loss": 2.2347, + "step": 582 + }, + { + "epoch": 0.6389041095890411, + "grad_norm": 1.9548988342285156, + "learning_rate": 4.400893582453291e-06, + "loss": 2.1967, + "step": 583 + }, + { + "epoch": 0.64, + "grad_norm": 1.898092269897461, + "learning_rate": 4.398862713241267e-06, + "loss": 2.1795, + "step": 584 + }, + { + "epoch": 0.6410958904109589, + "grad_norm": 1.8183610439300537, + "learning_rate": 4.396831844029245e-06, + "loss": 2.2031, + "step": 585 + }, + { + "epoch": 0.6421917808219179, + "grad_norm": 2.1520016193389893, + "learning_rate": 4.394800974817222e-06, + "loss": 2.2051, + "step": 586 + }, + { + "epoch": 0.6432876712328767, + "grad_norm": 1.9930012226104736, + "learning_rate": 4.392770105605199e-06, + "loss": 2.2053, + "step": 587 + }, + { + "epoch": 0.6443835616438356, + "grad_norm": 2.311450958251953, + "learning_rate": 4.390739236393176e-06, + "loss": 2.2252, + "step": 588 + }, + { + "epoch": 0.6454794520547945, + "grad_norm": 2.0168697834014893, + "learning_rate": 4.388708367181154e-06, + "loss": 2.2286, + "step": 589 + }, + { + "epoch": 0.6465753424657534, + "grad_norm": 2.055964469909668, + "learning_rate": 4.386677497969131e-06, + "loss": 2.2278, + "step": 590 + }, + { + "epoch": 0.6465753424657534, + "eval_loss": 2.334367275238037, + "eval_runtime": 48.9414, + "eval_samples_per_second": 20.433, + "eval_steps_per_second": 0.858, + "step": 590 + }, + { + "epoch": 0.6476712328767124, + "grad_norm": 2.0161590576171875, + "learning_rate": 4.384646628757108e-06, + "loss": 2.2102, + "step": 591 + }, + { + "epoch": 0.6487671232876713, + "grad_norm": 2.0827667713165283, + "learning_rate": 4.3826157595450855e-06, + "loss": 2.2531, + "step": 592 + }, + { + "epoch": 0.6498630136986301, + "grad_norm": 2.077637195587158, + "learning_rate": 4.380584890333063e-06, + "loss": 2.2091, + "step": 593 + }, + { + "epoch": 0.650958904109589, + "grad_norm": 2.023892879486084, + "learning_rate": 4.37855402112104e-06, + "loss": 2.1684, + "step": 594 + }, + { + "epoch": 0.6520547945205479, + "grad_norm": 1.8509857654571533, + "learning_rate": 4.376523151909017e-06, + "loss": 2.1439, + "step": 595 + }, + { + "epoch": 0.6531506849315069, + "grad_norm": 2.1649608612060547, + "learning_rate": 4.3744922826969946e-06, + "loss": 2.2557, + "step": 596 + }, + { + "epoch": 0.6542465753424658, + "grad_norm": 2.1351773738861084, + "learning_rate": 4.372461413484972e-06, + "loss": 2.177, + "step": 597 + }, + { + "epoch": 0.6553424657534247, + "grad_norm": 2.1713974475860596, + "learning_rate": 4.370430544272949e-06, + "loss": 2.2223, + "step": 598 + }, + { + "epoch": 0.6564383561643835, + "grad_norm": 2.2905397415161133, + "learning_rate": 4.368399675060926e-06, + "loss": 2.2373, + "step": 599 + }, + { + "epoch": 0.6575342465753424, + "grad_norm": 2.007160186767578, + "learning_rate": 4.366368805848904e-06, + "loss": 2.2355, + "step": 600 + }, + { + "epoch": 0.6575342465753424, + "eval_loss": 2.3322324752807617, + "eval_runtime": 46.2769, + "eval_samples_per_second": 21.609, + "eval_steps_per_second": 0.908, + "step": 600 + }, + { + "epoch": 0.6586301369863014, + "grad_norm": 2.309783697128296, + "learning_rate": 4.364337936636881e-06, + "loss": 2.2052, + "step": 601 + }, + { + "epoch": 0.6597260273972603, + "grad_norm": 2.498034954071045, + "learning_rate": 4.362307067424858e-06, + "loss": 2.1567, + "step": 602 + }, + { + "epoch": 0.6608219178082192, + "grad_norm": 1.8302545547485352, + "learning_rate": 4.3602761982128355e-06, + "loss": 2.2117, + "step": 603 + }, + { + "epoch": 0.6619178082191781, + "grad_norm": 2.226947546005249, + "learning_rate": 4.358245329000813e-06, + "loss": 2.1881, + "step": 604 + }, + { + "epoch": 0.663013698630137, + "grad_norm": 1.8484126329421997, + "learning_rate": 4.35621445978879e-06, + "loss": 2.1955, + "step": 605 + }, + { + "epoch": 0.6641095890410958, + "grad_norm": 1.9844266176223755, + "learning_rate": 4.354183590576767e-06, + "loss": 2.2063, + "step": 606 + }, + { + "epoch": 0.6652054794520548, + "grad_norm": 1.9292365312576294, + "learning_rate": 4.352152721364745e-06, + "loss": 2.205, + "step": 607 + }, + { + "epoch": 0.6663013698630137, + "grad_norm": 1.980895757675171, + "learning_rate": 4.350121852152722e-06, + "loss": 2.1727, + "step": 608 + }, + { + "epoch": 0.6673972602739726, + "grad_norm": 1.8331713676452637, + "learning_rate": 4.348090982940699e-06, + "loss": 2.226, + "step": 609 + }, + { + "epoch": 0.6684931506849315, + "grad_norm": 1.9666537046432495, + "learning_rate": 4.3460601137286764e-06, + "loss": 2.2198, + "step": 610 + }, + { + "epoch": 0.6684931506849315, + "eval_loss": 2.3297338485717773, + "eval_runtime": 42.6274, + "eval_samples_per_second": 23.459, + "eval_steps_per_second": 0.985, + "step": 610 + }, + { + "epoch": 0.6695890410958905, + "grad_norm": 1.9741110801696777, + "learning_rate": 4.344029244516654e-06, + "loss": 2.2132, + "step": 611 + }, + { + "epoch": 0.6706849315068493, + "grad_norm": 2.0747334957122803, + "learning_rate": 4.341998375304631e-06, + "loss": 2.2081, + "step": 612 + }, + { + "epoch": 0.6717808219178082, + "grad_norm": 2.031503915786743, + "learning_rate": 4.339967506092607e-06, + "loss": 2.2642, + "step": 613 + }, + { + "epoch": 0.6728767123287671, + "grad_norm": 1.8513256311416626, + "learning_rate": 4.3379366368805855e-06, + "loss": 2.2141, + "step": 614 + }, + { + "epoch": 0.673972602739726, + "grad_norm": 1.9098315238952637, + "learning_rate": 4.335905767668562e-06, + "loss": 2.2064, + "step": 615 + }, + { + "epoch": 0.675068493150685, + "grad_norm": 1.9704065322875977, + "learning_rate": 4.33387489845654e-06, + "loss": 2.1997, + "step": 616 + }, + { + "epoch": 0.6761643835616439, + "grad_norm": 2.1520018577575684, + "learning_rate": 4.3318440292445165e-06, + "loss": 2.2278, + "step": 617 + }, + { + "epoch": 0.6772602739726027, + "grad_norm": 1.898164987564087, + "learning_rate": 4.329813160032495e-06, + "loss": 2.2147, + "step": 618 + }, + { + "epoch": 0.6783561643835616, + "grad_norm": 1.8955962657928467, + "learning_rate": 4.327782290820471e-06, + "loss": 2.1825, + "step": 619 + }, + { + "epoch": 0.6794520547945205, + "grad_norm": 1.7579313516616821, + "learning_rate": 4.325751421608449e-06, + "loss": 2.1619, + "step": 620 + }, + { + "epoch": 0.6794520547945205, + "eval_loss": 2.329692840576172, + "eval_runtime": 46.3187, + "eval_samples_per_second": 21.59, + "eval_steps_per_second": 0.907, + "step": 620 + }, + { + "epoch": 0.6805479452054795, + "grad_norm": 1.8716063499450684, + "learning_rate": 4.323720552396426e-06, + "loss": 2.2289, + "step": 621 + }, + { + "epoch": 0.6816438356164384, + "grad_norm": 1.8798290491104126, + "learning_rate": 4.321689683184404e-06, + "loss": 2.1944, + "step": 622 + }, + { + "epoch": 0.6827397260273973, + "grad_norm": 1.8438303470611572, + "learning_rate": 4.31965881397238e-06, + "loss": 2.1879, + "step": 623 + }, + { + "epoch": 0.6838356164383562, + "grad_norm": 1.7844786643981934, + "learning_rate": 4.3176279447603574e-06, + "loss": 2.1936, + "step": 624 + }, + { + "epoch": 0.684931506849315, + "grad_norm": 1.955529808998108, + "learning_rate": 4.315597075548335e-06, + "loss": 2.2258, + "step": 625 + }, + { + "epoch": 0.686027397260274, + "grad_norm": 2.3826634883880615, + "learning_rate": 4.313566206336312e-06, + "loss": 2.2135, + "step": 626 + }, + { + "epoch": 0.6871232876712329, + "grad_norm": 1.8453123569488525, + "learning_rate": 4.311535337124289e-06, + "loss": 2.2362, + "step": 627 + }, + { + "epoch": 0.6882191780821918, + "grad_norm": 2.742682933807373, + "learning_rate": 4.3095044679122665e-06, + "loss": 2.194, + "step": 628 + }, + { + "epoch": 0.6893150684931507, + "grad_norm": 2.0299901962280273, + "learning_rate": 4.307473598700244e-06, + "loss": 2.2029, + "step": 629 + }, + { + "epoch": 0.6904109589041096, + "grad_norm": 2.29683780670166, + "learning_rate": 4.305442729488221e-06, + "loss": 2.2141, + "step": 630 + }, + { + "epoch": 0.6904109589041096, + "eval_loss": 2.329368829727173, + "eval_runtime": 45.1947, + "eval_samples_per_second": 22.126, + "eval_steps_per_second": 0.929, + "step": 630 + }, + { + "epoch": 0.6915068493150685, + "grad_norm": 2.208280086517334, + "learning_rate": 4.303411860276198e-06, + "loss": 2.1637, + "step": 631 + }, + { + "epoch": 0.6926027397260274, + "grad_norm": 1.943514108657837, + "learning_rate": 4.301380991064176e-06, + "loss": 2.234, + "step": 632 + }, + { + "epoch": 0.6936986301369863, + "grad_norm": 2.486440420150757, + "learning_rate": 4.299350121852153e-06, + "loss": 2.2239, + "step": 633 + }, + { + "epoch": 0.6947945205479452, + "grad_norm": 1.7243690490722656, + "learning_rate": 4.29731925264013e-06, + "loss": 2.2428, + "step": 634 + }, + { + "epoch": 0.6958904109589041, + "grad_norm": 2.31227445602417, + "learning_rate": 4.2952883834281075e-06, + "loss": 2.2403, + "step": 635 + }, + { + "epoch": 0.6969863013698631, + "grad_norm": 1.7361263036727905, + "learning_rate": 4.293257514216085e-06, + "loss": 2.1975, + "step": 636 + }, + { + "epoch": 0.6980821917808219, + "grad_norm": 2.010831356048584, + "learning_rate": 4.291226645004062e-06, + "loss": 2.1633, + "step": 637 + }, + { + "epoch": 0.6991780821917808, + "grad_norm": 2.2796833515167236, + "learning_rate": 4.289195775792039e-06, + "loss": 2.2307, + "step": 638 + }, + { + "epoch": 0.7002739726027397, + "grad_norm": 2.1069483757019043, + "learning_rate": 4.287164906580017e-06, + "loss": 2.2389, + "step": 639 + }, + { + "epoch": 0.7013698630136986, + "grad_norm": 2.19460129737854, + "learning_rate": 4.285134037367994e-06, + "loss": 2.1992, + "step": 640 + }, + { + "epoch": 0.7013698630136986, + "eval_loss": 2.3223369121551514, + "eval_runtime": 43.1093, + "eval_samples_per_second": 23.197, + "eval_steps_per_second": 0.974, + "step": 640 + }, + { + "epoch": 0.7024657534246576, + "grad_norm": 1.875261902809143, + "learning_rate": 4.283103168155971e-06, + "loss": 2.1767, + "step": 641 + }, + { + "epoch": 0.7035616438356165, + "grad_norm": 1.906762957572937, + "learning_rate": 4.281072298943948e-06, + "loss": 2.1734, + "step": 642 + }, + { + "epoch": 0.7046575342465753, + "grad_norm": 2.1114180088043213, + "learning_rate": 4.279041429731926e-06, + "loss": 2.1734, + "step": 643 + }, + { + "epoch": 0.7057534246575342, + "grad_norm": 1.844373106956482, + "learning_rate": 4.277010560519903e-06, + "loss": 2.1967, + "step": 644 + }, + { + "epoch": 0.7068493150684931, + "grad_norm": 2.3463754653930664, + "learning_rate": 4.27497969130788e-06, + "loss": 2.1735, + "step": 645 + }, + { + "epoch": 0.707945205479452, + "grad_norm": 2.3530960083007812, + "learning_rate": 4.2729488220958575e-06, + "loss": 2.189, + "step": 646 + }, + { + "epoch": 0.709041095890411, + "grad_norm": 2.110963821411133, + "learning_rate": 4.270917952883835e-06, + "loss": 2.218, + "step": 647 + }, + { + "epoch": 0.7101369863013699, + "grad_norm": 2.625005006790161, + "learning_rate": 4.268887083671812e-06, + "loss": 2.2235, + "step": 648 + }, + { + "epoch": 0.7112328767123288, + "grad_norm": 1.8411654233932495, + "learning_rate": 4.266856214459789e-06, + "loss": 2.1757, + "step": 649 + }, + { + "epoch": 0.7123287671232876, + "grad_norm": 2.0817489624023438, + "learning_rate": 4.264825345247766e-06, + "loss": 2.2698, + "step": 650 + }, + { + "epoch": 0.7123287671232876, + "eval_loss": 2.322544574737549, + "eval_runtime": 43.0771, + "eval_samples_per_second": 23.214, + "eval_steps_per_second": 0.975, + "step": 650 + }, + { + "epoch": 0.7134246575342466, + "grad_norm": 1.8953847885131836, + "learning_rate": 4.262794476035744e-06, + "loss": 2.1745, + "step": 651 + }, + { + "epoch": 0.7145205479452055, + "grad_norm": 1.8463294506072998, + "learning_rate": 4.26076360682372e-06, + "loss": 2.2042, + "step": 652 + }, + { + "epoch": 0.7156164383561644, + "grad_norm": 1.8776047229766846, + "learning_rate": 4.2587327376116984e-06, + "loss": 2.1512, + "step": 653 + }, + { + "epoch": 0.7167123287671233, + "grad_norm": 1.8949921131134033, + "learning_rate": 4.256701868399675e-06, + "loss": 2.1895, + "step": 654 + }, + { + "epoch": 0.7178082191780822, + "grad_norm": 1.8172632455825806, + "learning_rate": 4.254670999187653e-06, + "loss": 2.2, + "step": 655 + }, + { + "epoch": 0.718904109589041, + "grad_norm": 2.105499505996704, + "learning_rate": 4.252640129975629e-06, + "loss": 2.202, + "step": 656 + }, + { + "epoch": 0.72, + "grad_norm": 1.9515924453735352, + "learning_rate": 4.2506092607636075e-06, + "loss": 2.2116, + "step": 657 + }, + { + "epoch": 0.7210958904109589, + "grad_norm": 1.761770248413086, + "learning_rate": 4.248578391551584e-06, + "loss": 2.1921, + "step": 658 + }, + { + "epoch": 0.7221917808219178, + "grad_norm": 1.9848097562789917, + "learning_rate": 4.246547522339562e-06, + "loss": 2.1553, + "step": 659 + }, + { + "epoch": 0.7232876712328767, + "grad_norm": 2.041987180709839, + "learning_rate": 4.2445166531275385e-06, + "loss": 2.1975, + "step": 660 + }, + { + "epoch": 0.7232876712328767, + "eval_loss": 2.325580358505249, + "eval_runtime": 43.6207, + "eval_samples_per_second": 22.925, + "eval_steps_per_second": 0.963, + "step": 660 + }, + { + "epoch": 0.7243835616438357, + "grad_norm": 1.9755775928497314, + "learning_rate": 4.242485783915517e-06, + "loss": 2.1955, + "step": 661 + }, + { + "epoch": 0.7254794520547945, + "grad_norm": 1.9621502161026, + "learning_rate": 4.240454914703493e-06, + "loss": 2.1554, + "step": 662 + }, + { + "epoch": 0.7265753424657534, + "grad_norm": 1.848013162612915, + "learning_rate": 4.23842404549147e-06, + "loss": 2.16, + "step": 663 + }, + { + "epoch": 0.7276712328767123, + "grad_norm": 1.7370128631591797, + "learning_rate": 4.236393176279448e-06, + "loss": 2.1696, + "step": 664 + }, + { + "epoch": 0.7287671232876712, + "grad_norm": 2.0708158016204834, + "learning_rate": 4.234362307067425e-06, + "loss": 2.1592, + "step": 665 + }, + { + "epoch": 0.7298630136986302, + "grad_norm": 1.9864391088485718, + "learning_rate": 4.232331437855402e-06, + "loss": 2.2121, + "step": 666 + }, + { + "epoch": 0.7309589041095891, + "grad_norm": 1.9988250732421875, + "learning_rate": 4.2303005686433795e-06, + "loss": 2.2306, + "step": 667 + }, + { + "epoch": 0.7320547945205479, + "grad_norm": 1.89108407497406, + "learning_rate": 4.228269699431357e-06, + "loss": 2.222, + "step": 668 + }, + { + "epoch": 0.7331506849315068, + "grad_norm": 1.7941503524780273, + "learning_rate": 4.226238830219334e-06, + "loss": 2.1975, + "step": 669 + }, + { + "epoch": 0.7342465753424657, + "grad_norm": 1.8058613538742065, + "learning_rate": 4.224207961007311e-06, + "loss": 2.2037, + "step": 670 + }, + { + "epoch": 0.7342465753424657, + "eval_loss": 2.3266475200653076, + "eval_runtime": 47.6623, + "eval_samples_per_second": 20.981, + "eval_steps_per_second": 0.881, + "step": 670 + }, + { + "epoch": 0.7353424657534247, + "grad_norm": 1.764448642730713, + "learning_rate": 4.2221770917952886e-06, + "loss": 2.1904, + "step": 671 + }, + { + "epoch": 0.7364383561643836, + "grad_norm": 1.8252112865447998, + "learning_rate": 4.220146222583266e-06, + "loss": 2.1959, + "step": 672 + }, + { + "epoch": 0.7375342465753425, + "grad_norm": 2.02898907661438, + "learning_rate": 4.218115353371243e-06, + "loss": 2.1937, + "step": 673 + }, + { + "epoch": 0.7386301369863014, + "grad_norm": 1.8766248226165771, + "learning_rate": 4.21608448415922e-06, + "loss": 2.195, + "step": 674 + }, + { + "epoch": 0.7397260273972602, + "grad_norm": 2.2350332736968994, + "learning_rate": 4.214053614947198e-06, + "loss": 2.1798, + "step": 675 + }, + { + "epoch": 0.7408219178082192, + "grad_norm": 1.820399522781372, + "learning_rate": 4.212022745735175e-06, + "loss": 2.2371, + "step": 676 + }, + { + "epoch": 0.7419178082191781, + "grad_norm": 2.3411312103271484, + "learning_rate": 4.209991876523152e-06, + "loss": 2.1603, + "step": 677 + }, + { + "epoch": 0.743013698630137, + "grad_norm": 1.8562874794006348, + "learning_rate": 4.2079610073111295e-06, + "loss": 2.2082, + "step": 678 + }, + { + "epoch": 0.7441095890410959, + "grad_norm": 2.274214506149292, + "learning_rate": 4.205930138099107e-06, + "loss": 2.1716, + "step": 679 + }, + { + "epoch": 0.7452054794520548, + "grad_norm": 1.9173530340194702, + "learning_rate": 4.203899268887084e-06, + "loss": 2.2461, + "step": 680 + }, + { + "epoch": 0.7452054794520548, + "eval_loss": 2.3253841400146484, + "eval_runtime": 46.5732, + "eval_samples_per_second": 21.472, + "eval_steps_per_second": 0.902, + "step": 680 + }, + { + "epoch": 0.7463013698630137, + "grad_norm": 1.9732033014297485, + "learning_rate": 4.201868399675061e-06, + "loss": 2.2016, + "step": 681 + }, + { + "epoch": 0.7473972602739726, + "grad_norm": 1.8850247859954834, + "learning_rate": 4.199837530463039e-06, + "loss": 2.223, + "step": 682 + }, + { + "epoch": 0.7484931506849315, + "grad_norm": 1.8936313390731812, + "learning_rate": 4.197806661251016e-06, + "loss": 2.1381, + "step": 683 + }, + { + "epoch": 0.7495890410958904, + "grad_norm": 2.11995530128479, + "learning_rate": 4.195775792038993e-06, + "loss": 2.1962, + "step": 684 + }, + { + "epoch": 0.7506849315068493, + "grad_norm": 1.8207820653915405, + "learning_rate": 4.19374492282697e-06, + "loss": 2.1835, + "step": 685 + }, + { + "epoch": 0.7517808219178083, + "grad_norm": 2.06030535697937, + "learning_rate": 4.191714053614948e-06, + "loss": 2.1347, + "step": 686 + }, + { + "epoch": 0.7528767123287671, + "grad_norm": 2.048990488052368, + "learning_rate": 4.189683184402924e-06, + "loss": 2.2189, + "step": 687 + }, + { + "epoch": 0.753972602739726, + "grad_norm": 1.976369023323059, + "learning_rate": 4.187652315190902e-06, + "loss": 2.1805, + "step": 688 + }, + { + "epoch": 0.7550684931506849, + "grad_norm": 2.1160926818847656, + "learning_rate": 4.185621445978879e-06, + "loss": 2.2322, + "step": 689 + }, + { + "epoch": 0.7561643835616438, + "grad_norm": 1.9444516897201538, + "learning_rate": 4.183590576766857e-06, + "loss": 2.1696, + "step": 690 + }, + { + "epoch": 0.7561643835616438, + "eval_loss": 2.3242011070251465, + "eval_runtime": 48.7528, + "eval_samples_per_second": 20.512, + "eval_steps_per_second": 0.861, + "step": 690 + }, + { + "epoch": 0.7572602739726028, + "grad_norm": 1.9567183256149292, + "learning_rate": 4.181559707554833e-06, + "loss": 2.1891, + "step": 691 + }, + { + "epoch": 0.7583561643835617, + "grad_norm": 1.9185020923614502, + "learning_rate": 4.179528838342811e-06, + "loss": 2.2318, + "step": 692 + }, + { + "epoch": 0.7594520547945205, + "grad_norm": 1.833487629890442, + "learning_rate": 4.177497969130788e-06, + "loss": 2.1705, + "step": 693 + }, + { + "epoch": 0.7605479452054794, + "grad_norm": 2.2949867248535156, + "learning_rate": 4.175467099918766e-06, + "loss": 2.1712, + "step": 694 + }, + { + "epoch": 0.7616438356164383, + "grad_norm": 1.8663084506988525, + "learning_rate": 4.173436230706742e-06, + "loss": 2.2262, + "step": 695 + }, + { + "epoch": 0.7627397260273973, + "grad_norm": 2.4633052349090576, + "learning_rate": 4.1714053614947205e-06, + "loss": 2.185, + "step": 696 + }, + { + "epoch": 0.7638356164383562, + "grad_norm": 2.036524772644043, + "learning_rate": 4.169374492282697e-06, + "loss": 2.1697, + "step": 697 + }, + { + "epoch": 0.7649315068493151, + "grad_norm": 2.013810634613037, + "learning_rate": 4.167343623070675e-06, + "loss": 2.1826, + "step": 698 + }, + { + "epoch": 0.766027397260274, + "grad_norm": 2.214109182357788, + "learning_rate": 4.165312753858652e-06, + "loss": 2.2162, + "step": 699 + }, + { + "epoch": 0.7671232876712328, + "grad_norm": 1.8463964462280273, + "learning_rate": 4.163281884646629e-06, + "loss": 2.1865, + "step": 700 + }, + { + "epoch": 0.7671232876712328, + "eval_loss": 2.323822498321533, + "eval_runtime": 48.3307, + "eval_samples_per_second": 20.691, + "eval_steps_per_second": 0.869, + "step": 700 + }, + { + "epoch": 0.7682191780821918, + "grad_norm": 2.321021556854248, + "learning_rate": 4.161251015434607e-06, + "loss": 2.1758, + "step": 701 + }, + { + "epoch": 0.7693150684931507, + "grad_norm": 1.8213255405426025, + "learning_rate": 4.159220146222583e-06, + "loss": 2.2002, + "step": 702 + }, + { + "epoch": 0.7704109589041096, + "grad_norm": 1.8925111293792725, + "learning_rate": 4.157189277010561e-06, + "loss": 2.1575, + "step": 703 + }, + { + "epoch": 0.7715068493150685, + "grad_norm": 1.8050175905227661, + "learning_rate": 4.155158407798538e-06, + "loss": 2.179, + "step": 704 + }, + { + "epoch": 0.7726027397260274, + "grad_norm": 1.873228907585144, + "learning_rate": 4.153127538586516e-06, + "loss": 2.2232, + "step": 705 + }, + { + "epoch": 0.7736986301369863, + "grad_norm": 1.954910397529602, + "learning_rate": 4.151096669374492e-06, + "loss": 2.1992, + "step": 706 + }, + { + "epoch": 0.7747945205479452, + "grad_norm": 1.8913410902023315, + "learning_rate": 4.1490658001624705e-06, + "loss": 2.2086, + "step": 707 + }, + { + "epoch": 0.7758904109589041, + "grad_norm": 1.8883898258209229, + "learning_rate": 4.147034930950447e-06, + "loss": 2.1801, + "step": 708 + }, + { + "epoch": 0.776986301369863, + "grad_norm": 2.054351806640625, + "learning_rate": 4.145004061738425e-06, + "loss": 2.1908, + "step": 709 + }, + { + "epoch": 0.7780821917808219, + "grad_norm": 2.029036045074463, + "learning_rate": 4.1429731925264015e-06, + "loss": 2.1891, + "step": 710 + }, + { + "epoch": 0.7780821917808219, + "eval_loss": 2.3196680545806885, + "eval_runtime": 42.8326, + "eval_samples_per_second": 23.347, + "eval_steps_per_second": 0.981, + "step": 710 + }, + { + "epoch": 0.7791780821917809, + "grad_norm": 2.27848219871521, + "learning_rate": 4.140942323314379e-06, + "loss": 2.1165, + "step": 711 + }, + { + "epoch": 0.7802739726027397, + "grad_norm": 2.134625196456909, + "learning_rate": 4.138911454102356e-06, + "loss": 2.1508, + "step": 712 + }, + { + "epoch": 0.7813698630136986, + "grad_norm": 2.103895902633667, + "learning_rate": 4.136880584890333e-06, + "loss": 2.1746, + "step": 713 + }, + { + "epoch": 0.7824657534246575, + "grad_norm": 1.713189721107483, + "learning_rate": 4.1348497156783106e-06, + "loss": 2.2023, + "step": 714 + }, + { + "epoch": 0.7835616438356164, + "grad_norm": 2.0390899181365967, + "learning_rate": 4.132818846466288e-06, + "loss": 2.2039, + "step": 715 + }, + { + "epoch": 0.7846575342465754, + "grad_norm": 1.9643651247024536, + "learning_rate": 4.130787977254265e-06, + "loss": 2.199, + "step": 716 + }, + { + "epoch": 0.7857534246575343, + "grad_norm": 1.9173564910888672, + "learning_rate": 4.128757108042242e-06, + "loss": 2.2166, + "step": 717 + }, + { + "epoch": 0.7868493150684932, + "grad_norm": 1.7923007011413574, + "learning_rate": 4.12672623883022e-06, + "loss": 2.2144, + "step": 718 + }, + { + "epoch": 0.787945205479452, + "grad_norm": 2.2235090732574463, + "learning_rate": 4.124695369618197e-06, + "loss": 2.2586, + "step": 719 + }, + { + "epoch": 0.7890410958904109, + "grad_norm": 1.780316710472107, + "learning_rate": 4.122664500406174e-06, + "loss": 2.1862, + "step": 720 + }, + { + "epoch": 0.7890410958904109, + "eval_loss": 2.318016767501831, + "eval_runtime": 50.5082, + "eval_samples_per_second": 19.799, + "eval_steps_per_second": 0.832, + "step": 720 + }, + { + "epoch": 0.7901369863013699, + "grad_norm": 2.0788493156433105, + "learning_rate": 4.1206336311941515e-06, + "loss": 2.1845, + "step": 721 + }, + { + "epoch": 0.7912328767123288, + "grad_norm": 1.9053714275360107, + "learning_rate": 4.118602761982129e-06, + "loss": 2.2005, + "step": 722 + }, + { + "epoch": 0.7923287671232877, + "grad_norm": 1.9328513145446777, + "learning_rate": 4.116571892770106e-06, + "loss": 2.1475, + "step": 723 + }, + { + "epoch": 0.7934246575342466, + "grad_norm": 2.059760570526123, + "learning_rate": 4.114541023558083e-06, + "loss": 2.1386, + "step": 724 + }, + { + "epoch": 0.7945205479452054, + "grad_norm": 2.0093464851379395, + "learning_rate": 4.112510154346061e-06, + "loss": 2.2242, + "step": 725 + }, + { + "epoch": 0.7956164383561644, + "grad_norm": 1.8513909578323364, + "learning_rate": 4.110479285134038e-06, + "loss": 2.178, + "step": 726 + }, + { + "epoch": 0.7967123287671233, + "grad_norm": 1.9481958150863647, + "learning_rate": 4.108448415922015e-06, + "loss": 2.1961, + "step": 727 + }, + { + "epoch": 0.7978082191780822, + "grad_norm": 1.9229767322540283, + "learning_rate": 4.1064175467099924e-06, + "loss": 2.2116, + "step": 728 + }, + { + "epoch": 0.7989041095890411, + "grad_norm": 1.9819737672805786, + "learning_rate": 4.10438667749797e-06, + "loss": 2.2097, + "step": 729 + }, + { + "epoch": 0.8, + "grad_norm": 1.9382416009902954, + "learning_rate": 4.102355808285947e-06, + "loss": 2.1099, + "step": 730 + }, + { + "epoch": 0.8, + "eval_loss": 2.31795334815979, + "eval_runtime": 51.8961, + "eval_samples_per_second": 19.269, + "eval_steps_per_second": 0.809, + "step": 730 + }, + { + "epoch": 0.8010958904109589, + "grad_norm": 2.006890058517456, + "learning_rate": 4.100324939073924e-06, + "loss": 2.2606, + "step": 731 + }, + { + "epoch": 0.8021917808219178, + "grad_norm": 2.164473295211792, + "learning_rate": 4.0982940698619015e-06, + "loss": 2.1869, + "step": 732 + }, + { + "epoch": 0.8032876712328767, + "grad_norm": 2.04368257522583, + "learning_rate": 4.096263200649879e-06, + "loss": 2.1846, + "step": 733 + }, + { + "epoch": 0.8043835616438356, + "grad_norm": 1.8363912105560303, + "learning_rate": 4.094232331437856e-06, + "loss": 2.1938, + "step": 734 + }, + { + "epoch": 0.8054794520547945, + "grad_norm": 1.9260189533233643, + "learning_rate": 4.092201462225833e-06, + "loss": 2.1682, + "step": 735 + }, + { + "epoch": 0.8065753424657535, + "grad_norm": 2.0688912868499756, + "learning_rate": 4.090170593013811e-06, + "loss": 2.1772, + "step": 736 + }, + { + "epoch": 0.8076712328767123, + "grad_norm": 1.859474539756775, + "learning_rate": 4.088139723801787e-06, + "loss": 2.2013, + "step": 737 + }, + { + "epoch": 0.8087671232876712, + "grad_norm": 2.0915753841400146, + "learning_rate": 4.086108854589765e-06, + "loss": 2.1519, + "step": 738 + }, + { + "epoch": 0.8098630136986301, + "grad_norm": 2.2019519805908203, + "learning_rate": 4.084077985377742e-06, + "loss": 2.1608, + "step": 739 + }, + { + "epoch": 0.810958904109589, + "grad_norm": 1.8727316856384277, + "learning_rate": 4.08204711616572e-06, + "loss": 2.1823, + "step": 740 + }, + { + "epoch": 0.810958904109589, + "eval_loss": 2.318638324737549, + "eval_runtime": 52.3669, + "eval_samples_per_second": 19.096, + "eval_steps_per_second": 0.802, + "step": 740 + }, + { + "epoch": 0.812054794520548, + "grad_norm": 2.059250831604004, + "learning_rate": 4.080016246953696e-06, + "loss": 2.2285, + "step": 741 + }, + { + "epoch": 0.8131506849315069, + "grad_norm": 1.9466749429702759, + "learning_rate": 4.077985377741674e-06, + "loss": 2.1496, + "step": 742 + }, + { + "epoch": 0.8142465753424658, + "grad_norm": 1.8967524766921997, + "learning_rate": 4.075954508529651e-06, + "loss": 2.2066, + "step": 743 + }, + { + "epoch": 0.8153424657534246, + "grad_norm": 1.976780891418457, + "learning_rate": 4.073923639317629e-06, + "loss": 2.2194, + "step": 744 + }, + { + "epoch": 0.8164383561643835, + "grad_norm": 2.077683687210083, + "learning_rate": 4.071892770105605e-06, + "loss": 2.2209, + "step": 745 + }, + { + "epoch": 0.8175342465753425, + "grad_norm": 1.9075323343276978, + "learning_rate": 4.069861900893583e-06, + "loss": 2.1727, + "step": 746 + }, + { + "epoch": 0.8186301369863014, + "grad_norm": 1.6940832138061523, + "learning_rate": 4.06783103168156e-06, + "loss": 2.1437, + "step": 747 + }, + { + "epoch": 0.8197260273972603, + "grad_norm": 1.9035768508911133, + "learning_rate": 4.065800162469538e-06, + "loss": 2.165, + "step": 748 + }, + { + "epoch": 0.8208219178082192, + "grad_norm": 1.8775745630264282, + "learning_rate": 4.063769293257514e-06, + "loss": 2.1757, + "step": 749 + }, + { + "epoch": 0.821917808219178, + "grad_norm": 1.9622418880462646, + "learning_rate": 4.061738424045492e-06, + "loss": 2.197, + "step": 750 + }, + { + "epoch": 0.821917808219178, + "eval_loss": 2.317519426345825, + "eval_runtime": 50.2699, + "eval_samples_per_second": 19.893, + "eval_steps_per_second": 0.835, + "step": 750 + }, + { + "epoch": 0.823013698630137, + "grad_norm": 1.9646214246749878, + "learning_rate": 4.059707554833469e-06, + "loss": 2.1783, + "step": 751 + }, + { + "epoch": 0.8241095890410959, + "grad_norm": 1.890342354774475, + "learning_rate": 4.057676685621446e-06, + "loss": 2.1925, + "step": 752 + }, + { + "epoch": 0.8252054794520548, + "grad_norm": 1.8977251052856445, + "learning_rate": 4.0556458164094235e-06, + "loss": 2.219, + "step": 753 + }, + { + "epoch": 0.8263013698630137, + "grad_norm": 1.8251665830612183, + "learning_rate": 4.053614947197401e-06, + "loss": 2.189, + "step": 754 + }, + { + "epoch": 0.8273972602739726, + "grad_norm": 1.7953882217407227, + "learning_rate": 4.051584077985378e-06, + "loss": 2.1237, + "step": 755 + }, + { + "epoch": 0.8284931506849315, + "grad_norm": 1.8247836828231812, + "learning_rate": 4.049553208773355e-06, + "loss": 2.1962, + "step": 756 + }, + { + "epoch": 0.8295890410958904, + "grad_norm": 1.8537967205047607, + "learning_rate": 4.047522339561333e-06, + "loss": 2.2201, + "step": 757 + }, + { + "epoch": 0.8306849315068493, + "grad_norm": 1.865315318107605, + "learning_rate": 4.04549147034931e-06, + "loss": 2.1819, + "step": 758 + }, + { + "epoch": 0.8317808219178082, + "grad_norm": 1.8620522022247314, + "learning_rate": 4.043460601137287e-06, + "loss": 2.1717, + "step": 759 + }, + { + "epoch": 0.8328767123287671, + "grad_norm": 1.9475384950637817, + "learning_rate": 4.041429731925264e-06, + "loss": 2.2718, + "step": 760 + }, + { + "epoch": 0.8328767123287671, + "eval_loss": 2.314429759979248, + "eval_runtime": 54.2324, + "eval_samples_per_second": 18.439, + "eval_steps_per_second": 0.774, + "step": 760 + }, + { + "epoch": 0.8339726027397261, + "grad_norm": 1.7170753479003906, + "learning_rate": 4.039398862713242e-06, + "loss": 2.1785, + "step": 761 + }, + { + "epoch": 0.8350684931506849, + "grad_norm": 1.968220591545105, + "learning_rate": 4.037367993501219e-06, + "loss": 2.1798, + "step": 762 + }, + { + "epoch": 0.8361643835616438, + "grad_norm": 1.7927111387252808, + "learning_rate": 4.035337124289196e-06, + "loss": 2.2372, + "step": 763 + }, + { + "epoch": 0.8372602739726027, + "grad_norm": 1.7871185541152954, + "learning_rate": 4.0333062550771735e-06, + "loss": 2.182, + "step": 764 + }, + { + "epoch": 0.8383561643835616, + "grad_norm": 1.9415812492370605, + "learning_rate": 4.031275385865151e-06, + "loss": 2.1527, + "step": 765 + }, + { + "epoch": 0.8394520547945206, + "grad_norm": 1.7327690124511719, + "learning_rate": 4.029244516653128e-06, + "loss": 2.1744, + "step": 766 + }, + { + "epoch": 0.8405479452054795, + "grad_norm": 1.9123411178588867, + "learning_rate": 4.027213647441105e-06, + "loss": 2.2049, + "step": 767 + }, + { + "epoch": 0.8416438356164384, + "grad_norm": 2.011591911315918, + "learning_rate": 4.025182778229083e-06, + "loss": 2.1391, + "step": 768 + }, + { + "epoch": 0.8427397260273972, + "grad_norm": 1.9169542789459229, + "learning_rate": 4.02315190901706e-06, + "loss": 2.1562, + "step": 769 + }, + { + "epoch": 0.8438356164383561, + "grad_norm": 1.8020051717758179, + "learning_rate": 4.021121039805037e-06, + "loss": 2.1795, + "step": 770 + }, + { + "epoch": 0.8438356164383561, + "eval_loss": 2.3199150562286377, + "eval_runtime": 52.468, + "eval_samples_per_second": 19.059, + "eval_steps_per_second": 0.8, + "step": 770 + }, + { + "epoch": 0.8449315068493151, + "grad_norm": 1.927666425704956, + "learning_rate": 4.0190901705930144e-06, + "loss": 2.1834, + "step": 771 + }, + { + "epoch": 0.846027397260274, + "grad_norm": 2.0998382568359375, + "learning_rate": 4.017059301380992e-06, + "loss": 2.2541, + "step": 772 + }, + { + "epoch": 0.8471232876712329, + "grad_norm": 1.88541579246521, + "learning_rate": 4.015028432168969e-06, + "loss": 2.1631, + "step": 773 + }, + { + "epoch": 0.8482191780821918, + "grad_norm": 1.9546253681182861, + "learning_rate": 4.012997562956945e-06, + "loss": 2.2266, + "step": 774 + }, + { + "epoch": 0.8493150684931506, + "grad_norm": 1.7969225645065308, + "learning_rate": 4.0109666937449235e-06, + "loss": 2.1526, + "step": 775 + }, + { + "epoch": 0.8504109589041096, + "grad_norm": 1.7237955331802368, + "learning_rate": 4.0089358245329e-06, + "loss": 2.1618, + "step": 776 + }, + { + "epoch": 0.8515068493150685, + "grad_norm": 1.8355857133865356, + "learning_rate": 4.006904955320878e-06, + "loss": 2.1673, + "step": 777 + }, + { + "epoch": 0.8526027397260274, + "grad_norm": 1.8049591779708862, + "learning_rate": 4.0048740861088545e-06, + "loss": 2.1676, + "step": 778 + }, + { + "epoch": 0.8536986301369863, + "grad_norm": 1.8143601417541504, + "learning_rate": 4.002843216896833e-06, + "loss": 2.214, + "step": 779 + }, + { + "epoch": 0.8547945205479452, + "grad_norm": 1.8344221115112305, + "learning_rate": 4.000812347684809e-06, + "loss": 2.1958, + "step": 780 + }, + { + "epoch": 0.8547945205479452, + "eval_loss": 2.314359426498413, + "eval_runtime": 51.8717, + "eval_samples_per_second": 19.278, + "eval_steps_per_second": 0.81, + "step": 780 + }, + { + "epoch": 0.8558904109589041, + "grad_norm": 1.944402813911438, + "learning_rate": 3.998781478472787e-06, + "loss": 2.1497, + "step": 781 + }, + { + "epoch": 0.856986301369863, + "grad_norm": 1.93898606300354, + "learning_rate": 3.996750609260764e-06, + "loss": 2.1737, + "step": 782 + }, + { + "epoch": 0.8580821917808219, + "grad_norm": 1.9499506950378418, + "learning_rate": 3.994719740048742e-06, + "loss": 2.1545, + "step": 783 + }, + { + "epoch": 0.8591780821917808, + "grad_norm": 1.9311052560806274, + "learning_rate": 3.992688870836718e-06, + "loss": 2.2559, + "step": 784 + }, + { + "epoch": 0.8602739726027397, + "grad_norm": 1.864862084388733, + "learning_rate": 3.990658001624696e-06, + "loss": 2.1996, + "step": 785 + }, + { + "epoch": 0.8613698630136987, + "grad_norm": 1.8248569965362549, + "learning_rate": 3.988627132412673e-06, + "loss": 2.1591, + "step": 786 + }, + { + "epoch": 0.8624657534246576, + "grad_norm": 2.2370970249176025, + "learning_rate": 3.98659626320065e-06, + "loss": 2.2095, + "step": 787 + }, + { + "epoch": 0.8635616438356164, + "grad_norm": 1.8877084255218506, + "learning_rate": 3.984565393988627e-06, + "loss": 2.1813, + "step": 788 + }, + { + "epoch": 0.8646575342465753, + "grad_norm": 2.1722517013549805, + "learning_rate": 3.9825345247766046e-06, + "loss": 2.1866, + "step": 789 + }, + { + "epoch": 0.8657534246575342, + "grad_norm": 1.9640655517578125, + "learning_rate": 3.980503655564582e-06, + "loss": 2.1818, + "step": 790 + }, + { + "epoch": 0.8657534246575342, + "eval_loss": 2.314087390899658, + "eval_runtime": 51.9778, + "eval_samples_per_second": 19.239, + "eval_steps_per_second": 0.808, + "step": 790 + }, + { + "epoch": 0.8668493150684932, + "grad_norm": 2.0583789348602295, + "learning_rate": 3.978472786352559e-06, + "loss": 2.1956, + "step": 791 + }, + { + "epoch": 0.8679452054794521, + "grad_norm": 2.2128682136535645, + "learning_rate": 3.976441917140536e-06, + "loss": 2.204, + "step": 792 + }, + { + "epoch": 0.869041095890411, + "grad_norm": 1.917763113975525, + "learning_rate": 3.974411047928514e-06, + "loss": 2.1841, + "step": 793 + }, + { + "epoch": 0.8701369863013698, + "grad_norm": 2.0327019691467285, + "learning_rate": 3.972380178716491e-06, + "loss": 2.2351, + "step": 794 + }, + { + "epoch": 0.8712328767123287, + "grad_norm": 1.9498991966247559, + "learning_rate": 3.970349309504468e-06, + "loss": 2.2074, + "step": 795 + }, + { + "epoch": 0.8723287671232877, + "grad_norm": 1.8105775117874146, + "learning_rate": 3.9683184402924455e-06, + "loss": 2.2088, + "step": 796 + }, + { + "epoch": 0.8734246575342466, + "grad_norm": 1.9299930334091187, + "learning_rate": 3.966287571080423e-06, + "loss": 2.1469, + "step": 797 + }, + { + "epoch": 0.8745205479452055, + "grad_norm": 1.9346121549606323, + "learning_rate": 3.9642567018684e-06, + "loss": 2.1434, + "step": 798 + }, + { + "epoch": 0.8756164383561644, + "grad_norm": 1.8202977180480957, + "learning_rate": 3.962225832656377e-06, + "loss": 2.194, + "step": 799 + }, + { + "epoch": 0.8767123287671232, + "grad_norm": 1.6922450065612793, + "learning_rate": 3.960194963444355e-06, + "loss": 2.1821, + "step": 800 + }, + { + "epoch": 0.8767123287671232, + "eval_loss": 2.310209035873413, + "eval_runtime": 46.1102, + "eval_samples_per_second": 21.687, + "eval_steps_per_second": 0.911, + "step": 800 + }, + { + "epoch": 0.8778082191780822, + "grad_norm": 1.9258904457092285, + "learning_rate": 3.958164094232332e-06, + "loss": 2.1326, + "step": 801 + }, + { + "epoch": 0.8789041095890411, + "grad_norm": 1.8485561609268188, + "learning_rate": 3.956133225020309e-06, + "loss": 2.1888, + "step": 802 + }, + { + "epoch": 0.88, + "grad_norm": 1.9802418947219849, + "learning_rate": 3.954102355808286e-06, + "loss": 2.1876, + "step": 803 + }, + { + "epoch": 0.8810958904109589, + "grad_norm": 1.8365259170532227, + "learning_rate": 3.952071486596264e-06, + "loss": 2.1804, + "step": 804 + }, + { + "epoch": 0.8821917808219178, + "grad_norm": 2.023909568786621, + "learning_rate": 3.950040617384241e-06, + "loss": 2.1801, + "step": 805 + }, + { + "epoch": 0.8832876712328767, + "grad_norm": 1.8889018297195435, + "learning_rate": 3.948009748172218e-06, + "loss": 2.1657, + "step": 806 + }, + { + "epoch": 0.8843835616438356, + "grad_norm": 1.7983578443527222, + "learning_rate": 3.9459788789601955e-06, + "loss": 2.1589, + "step": 807 + }, + { + "epoch": 0.8854794520547945, + "grad_norm": 2.157043218612671, + "learning_rate": 3.943948009748173e-06, + "loss": 2.2258, + "step": 808 + }, + { + "epoch": 0.8865753424657534, + "grad_norm": 1.7387373447418213, + "learning_rate": 3.94191714053615e-06, + "loss": 2.1878, + "step": 809 + }, + { + "epoch": 0.8876712328767123, + "grad_norm": 1.963233232498169, + "learning_rate": 3.939886271324127e-06, + "loss": 2.186, + "step": 810 + }, + { + "epoch": 0.8876712328767123, + "eval_loss": 2.306187152862549, + "eval_runtime": 42.563, + "eval_samples_per_second": 23.495, + "eval_steps_per_second": 0.987, + "step": 810 + }, + { + "epoch": 0.8887671232876713, + "grad_norm": 1.8932961225509644, + "learning_rate": 3.937855402112104e-06, + "loss": 2.198, + "step": 811 + }, + { + "epoch": 0.8898630136986302, + "grad_norm": 2.0431783199310303, + "learning_rate": 3.935824532900082e-06, + "loss": 2.1505, + "step": 812 + }, + { + "epoch": 0.890958904109589, + "grad_norm": 1.8954466581344604, + "learning_rate": 3.933793663688058e-06, + "loss": 2.1494, + "step": 813 + }, + { + "epoch": 0.8920547945205479, + "grad_norm": 1.841184139251709, + "learning_rate": 3.9317627944760365e-06, + "loss": 2.1932, + "step": 814 + }, + { + "epoch": 0.8931506849315068, + "grad_norm": 1.9083219766616821, + "learning_rate": 3.929731925264013e-06, + "loss": 2.2081, + "step": 815 + }, + { + "epoch": 0.8942465753424658, + "grad_norm": 1.9737226963043213, + "learning_rate": 3.927701056051991e-06, + "loss": 2.1405, + "step": 816 + }, + { + "epoch": 0.8953424657534247, + "grad_norm": 1.869194746017456, + "learning_rate": 3.9256701868399674e-06, + "loss": 2.1642, + "step": 817 + }, + { + "epoch": 0.8964383561643836, + "grad_norm": 1.9553205966949463, + "learning_rate": 3.9236393176279456e-06, + "loss": 2.1761, + "step": 818 + }, + { + "epoch": 0.8975342465753424, + "grad_norm": 1.8355534076690674, + "learning_rate": 3.921608448415922e-06, + "loss": 2.1551, + "step": 819 + }, + { + "epoch": 0.8986301369863013, + "grad_norm": 1.9770171642303467, + "learning_rate": 3.9195775792039e-06, + "loss": 2.1424, + "step": 820 + }, + { + "epoch": 0.8986301369863013, + "eval_loss": 2.307065010070801, + "eval_runtime": 44.5816, + "eval_samples_per_second": 22.431, + "eval_steps_per_second": 0.942, + "step": 820 + }, + { + "epoch": 0.8997260273972603, + "grad_norm": 1.7375415563583374, + "learning_rate": 3.9175467099918765e-06, + "loss": 2.1895, + "step": 821 + }, + { + "epoch": 0.9008219178082192, + "grad_norm": 1.8910484313964844, + "learning_rate": 3.915515840779855e-06, + "loss": 2.1685, + "step": 822 + }, + { + "epoch": 0.9019178082191781, + "grad_norm": 1.9191482067108154, + "learning_rate": 3.913484971567831e-06, + "loss": 2.2074, + "step": 823 + }, + { + "epoch": 0.903013698630137, + "grad_norm": 1.7864760160446167, + "learning_rate": 3.911454102355808e-06, + "loss": 2.1972, + "step": 824 + }, + { + "epoch": 0.9041095890410958, + "grad_norm": 1.9044650793075562, + "learning_rate": 3.909423233143786e-06, + "loss": 2.1237, + "step": 825 + }, + { + "epoch": 0.9052054794520548, + "grad_norm": 1.8378455638885498, + "learning_rate": 3.907392363931763e-06, + "loss": 2.1907, + "step": 826 + }, + { + "epoch": 0.9063013698630137, + "grad_norm": 1.8708094358444214, + "learning_rate": 3.90536149471974e-06, + "loss": 2.1883, + "step": 827 + }, + { + "epoch": 0.9073972602739726, + "grad_norm": 1.7802201509475708, + "learning_rate": 3.9033306255077175e-06, + "loss": 2.2356, + "step": 828 + }, + { + "epoch": 0.9084931506849315, + "grad_norm": 1.9374496936798096, + "learning_rate": 3.901299756295695e-06, + "loss": 2.2546, + "step": 829 + }, + { + "epoch": 0.9095890410958904, + "grad_norm": 1.9246270656585693, + "learning_rate": 3.899268887083672e-06, + "loss": 2.2684, + "step": 830 + }, + { + "epoch": 0.9095890410958904, + "eval_loss": 2.3065874576568604, + "eval_runtime": 45.18, + "eval_samples_per_second": 22.134, + "eval_steps_per_second": 0.93, + "step": 830 + }, + { + "epoch": 0.9106849315068493, + "grad_norm": 1.8681883811950684, + "learning_rate": 3.897238017871649e-06, + "loss": 2.1737, + "step": 831 + }, + { + "epoch": 0.9117808219178082, + "grad_norm": 1.9293930530548096, + "learning_rate": 3.8952071486596266e-06, + "loss": 2.1813, + "step": 832 + }, + { + "epoch": 0.9128767123287671, + "grad_norm": 1.844586968421936, + "learning_rate": 3.893176279447604e-06, + "loss": 2.1788, + "step": 833 + }, + { + "epoch": 0.913972602739726, + "grad_norm": 1.7658660411834717, + "learning_rate": 3.891145410235581e-06, + "loss": 2.1494, + "step": 834 + }, + { + "epoch": 0.915068493150685, + "grad_norm": 2.05153751373291, + "learning_rate": 3.889114541023558e-06, + "loss": 2.241, + "step": 835 + }, + { + "epoch": 0.9161643835616439, + "grad_norm": 1.7869952917099, + "learning_rate": 3.887083671811536e-06, + "loss": 2.1866, + "step": 836 + }, + { + "epoch": 0.9172602739726028, + "grad_norm": 1.99534273147583, + "learning_rate": 3.885052802599513e-06, + "loss": 2.1682, + "step": 837 + }, + { + "epoch": 0.9183561643835616, + "grad_norm": 1.9593979120254517, + "learning_rate": 3.88302193338749e-06, + "loss": 2.1357, + "step": 838 + }, + { + "epoch": 0.9194520547945205, + "grad_norm": 1.8441778421401978, + "learning_rate": 3.8809910641754675e-06, + "loss": 2.2283, + "step": 839 + }, + { + "epoch": 0.9205479452054794, + "grad_norm": 1.9153512716293335, + "learning_rate": 3.878960194963445e-06, + "loss": 2.1899, + "step": 840 + }, + { + "epoch": 0.9205479452054794, + "eval_loss": 2.307375907897949, + "eval_runtime": 48.0216, + "eval_samples_per_second": 20.824, + "eval_steps_per_second": 0.875, + "step": 840 + }, + { + "epoch": 0.9216438356164384, + "grad_norm": 2.004232883453369, + "learning_rate": 3.876929325751422e-06, + "loss": 2.173, + "step": 841 + }, + { + "epoch": 0.9227397260273973, + "grad_norm": 2.12992262840271, + "learning_rate": 3.874898456539399e-06, + "loss": 2.2164, + "step": 842 + }, + { + "epoch": 0.9238356164383562, + "grad_norm": 2.0505411624908447, + "learning_rate": 3.872867587327377e-06, + "loss": 2.1432, + "step": 843 + }, + { + "epoch": 0.924931506849315, + "grad_norm": 1.767324686050415, + "learning_rate": 3.870836718115354e-06, + "loss": 2.1838, + "step": 844 + }, + { + "epoch": 0.9260273972602739, + "grad_norm": 2.189107894897461, + "learning_rate": 3.868805848903331e-06, + "loss": 2.1623, + "step": 845 + }, + { + "epoch": 0.9271232876712329, + "grad_norm": 1.9781970977783203, + "learning_rate": 3.8667749796913084e-06, + "loss": 2.1738, + "step": 846 + }, + { + "epoch": 0.9282191780821918, + "grad_norm": 1.8734674453735352, + "learning_rate": 3.864744110479286e-06, + "loss": 2.1829, + "step": 847 + }, + { + "epoch": 0.9293150684931507, + "grad_norm": 1.7699538469314575, + "learning_rate": 3.862713241267262e-06, + "loss": 2.186, + "step": 848 + }, + { + "epoch": 0.9304109589041096, + "grad_norm": 1.8354437351226807, + "learning_rate": 3.86068237205524e-06, + "loss": 2.2149, + "step": 849 + }, + { + "epoch": 0.9315068493150684, + "grad_norm": 1.913249135017395, + "learning_rate": 3.858651502843217e-06, + "loss": 2.2208, + "step": 850 + }, + { + "epoch": 0.9315068493150684, + "eval_loss": 2.307549238204956, + "eval_runtime": 43.1706, + "eval_samples_per_second": 23.164, + "eval_steps_per_second": 0.973, + "step": 850 + }, + { + "epoch": 0.9326027397260274, + "grad_norm": 1.92095148563385, + "learning_rate": 3.856620633631195e-06, + "loss": 2.1839, + "step": 851 + }, + { + "epoch": 0.9336986301369863, + "grad_norm": 1.8833959102630615, + "learning_rate": 3.854589764419171e-06, + "loss": 2.1372, + "step": 852 + }, + { + "epoch": 0.9347945205479452, + "grad_norm": 1.9410737752914429, + "learning_rate": 3.852558895207149e-06, + "loss": 2.1726, + "step": 853 + }, + { + "epoch": 0.9358904109589041, + "grad_norm": 1.8328828811645508, + "learning_rate": 3.850528025995126e-06, + "loss": 2.2425, + "step": 854 + }, + { + "epoch": 0.936986301369863, + "grad_norm": 1.89652419090271, + "learning_rate": 3.848497156783104e-06, + "loss": 2.1956, + "step": 855 + }, + { + "epoch": 0.938082191780822, + "grad_norm": 1.9026265144348145, + "learning_rate": 3.84646628757108e-06, + "loss": 2.167, + "step": 856 + }, + { + "epoch": 0.9391780821917808, + "grad_norm": 2.010864734649658, + "learning_rate": 3.8444354183590585e-06, + "loss": 2.1725, + "step": 857 + }, + { + "epoch": 0.9402739726027397, + "grad_norm": 2.293717384338379, + "learning_rate": 3.842404549147035e-06, + "loss": 2.1668, + "step": 858 + }, + { + "epoch": 0.9413698630136986, + "grad_norm": 1.8209116458892822, + "learning_rate": 3.840373679935013e-06, + "loss": 2.1649, + "step": 859 + }, + { + "epoch": 0.9424657534246575, + "grad_norm": 2.061277389526367, + "learning_rate": 3.8383428107229894e-06, + "loss": 2.1479, + "step": 860 + }, + { + "epoch": 0.9424657534246575, + "eval_loss": 2.3073670864105225, + "eval_runtime": 44.3781, + "eval_samples_per_second": 22.534, + "eval_steps_per_second": 0.946, + "step": 860 + }, + { + "epoch": 0.9435616438356165, + "grad_norm": 1.8032523393630981, + "learning_rate": 3.836311941510967e-06, + "loss": 2.1665, + "step": 861 + }, + { + "epoch": 0.9446575342465754, + "grad_norm": 1.989533543586731, + "learning_rate": 3.834281072298944e-06, + "loss": 2.1612, + "step": 862 + }, + { + "epoch": 0.9457534246575342, + "grad_norm": 2.017179012298584, + "learning_rate": 3.832250203086921e-06, + "loss": 2.201, + "step": 863 + }, + { + "epoch": 0.9468493150684931, + "grad_norm": 2.0291004180908203, + "learning_rate": 3.8302193338748985e-06, + "loss": 2.1508, + "step": 864 + }, + { + "epoch": 0.947945205479452, + "grad_norm": 1.8693255186080933, + "learning_rate": 3.828188464662876e-06, + "loss": 2.2106, + "step": 865 + }, + { + "epoch": 0.949041095890411, + "grad_norm": 1.8890169858932495, + "learning_rate": 3.826157595450853e-06, + "loss": 2.2467, + "step": 866 + }, + { + "epoch": 0.9501369863013699, + "grad_norm": 2.0550642013549805, + "learning_rate": 3.82412672623883e-06, + "loss": 2.1899, + "step": 867 + }, + { + "epoch": 0.9512328767123288, + "grad_norm": 1.8681174516677856, + "learning_rate": 3.822095857026808e-06, + "loss": 2.1428, + "step": 868 + }, + { + "epoch": 0.9523287671232876, + "grad_norm": 1.9133951663970947, + "learning_rate": 3.820064987814785e-06, + "loss": 2.1947, + "step": 869 + }, + { + "epoch": 0.9534246575342465, + "grad_norm": 1.8717621564865112, + "learning_rate": 3.818034118602762e-06, + "loss": 2.1542, + "step": 870 + }, + { + "epoch": 0.9534246575342465, + "eval_loss": 2.307584047317505, + "eval_runtime": 43.718, + "eval_samples_per_second": 22.874, + "eval_steps_per_second": 0.961, + "step": 870 + }, + { + "epoch": 0.9545205479452055, + "grad_norm": 1.9843461513519287, + "learning_rate": 3.8160032493907395e-06, + "loss": 2.1547, + "step": 871 + }, + { + "epoch": 0.9556164383561644, + "grad_norm": 1.725468397140503, + "learning_rate": 3.813972380178717e-06, + "loss": 2.1679, + "step": 872 + }, + { + "epoch": 0.9567123287671233, + "grad_norm": 1.8076086044311523, + "learning_rate": 3.811941510966694e-06, + "loss": 2.1654, + "step": 873 + }, + { + "epoch": 0.9578082191780822, + "grad_norm": 1.8879774808883667, + "learning_rate": 3.809910641754671e-06, + "loss": 2.191, + "step": 874 + }, + { + "epoch": 0.958904109589041, + "grad_norm": 1.943313717842102, + "learning_rate": 3.8078797725426486e-06, + "loss": 2.1448, + "step": 875 + }, + { + "epoch": 0.96, + "grad_norm": 1.8877806663513184, + "learning_rate": 3.8058489033306254e-06, + "loss": 2.1129, + "step": 876 + }, + { + "epoch": 0.9610958904109589, + "grad_norm": 1.787083625793457, + "learning_rate": 3.803818034118603e-06, + "loss": 2.1854, + "step": 877 + }, + { + "epoch": 0.9621917808219178, + "grad_norm": 2.0925140380859375, + "learning_rate": 3.80178716490658e-06, + "loss": 2.1614, + "step": 878 + }, + { + "epoch": 0.9632876712328767, + "grad_norm": 1.9043725728988647, + "learning_rate": 3.7997562956945577e-06, + "loss": 2.2161, + "step": 879 + }, + { + "epoch": 0.9643835616438357, + "grad_norm": 1.8173545598983765, + "learning_rate": 3.7977254264825345e-06, + "loss": 2.2067, + "step": 880 + }, + { + "epoch": 0.9643835616438357, + "eval_loss": 2.302727460861206, + "eval_runtime": 44.2015, + "eval_samples_per_second": 22.624, + "eval_steps_per_second": 0.95, + "step": 880 + }, + { + "epoch": 0.9654794520547946, + "grad_norm": 1.9871562719345093, + "learning_rate": 3.7956945572705122e-06, + "loss": 2.1658, + "step": 881 + }, + { + "epoch": 0.9665753424657534, + "grad_norm": 2.103553295135498, + "learning_rate": 3.793663688058489e-06, + "loss": 2.1468, + "step": 882 + }, + { + "epoch": 0.9676712328767123, + "grad_norm": 1.9089381694793701, + "learning_rate": 3.791632818846467e-06, + "loss": 2.2289, + "step": 883 + }, + { + "epoch": 0.9687671232876712, + "grad_norm": 2.3238179683685303, + "learning_rate": 3.7896019496344436e-06, + "loss": 2.2179, + "step": 884 + }, + { + "epoch": 0.9698630136986301, + "grad_norm": 1.902161955833435, + "learning_rate": 3.787571080422421e-06, + "loss": 2.2086, + "step": 885 + }, + { + "epoch": 0.9709589041095891, + "grad_norm": 1.8587497472763062, + "learning_rate": 3.785540211210398e-06, + "loss": 2.2175, + "step": 886 + }, + { + "epoch": 0.972054794520548, + "grad_norm": 2.029719829559326, + "learning_rate": 3.7835093419983755e-06, + "loss": 2.207, + "step": 887 + }, + { + "epoch": 0.9731506849315068, + "grad_norm": 2.0228402614593506, + "learning_rate": 3.7814784727863527e-06, + "loss": 2.1851, + "step": 888 + }, + { + "epoch": 0.9742465753424657, + "grad_norm": 1.9885436296463013, + "learning_rate": 3.77944760357433e-06, + "loss": 2.1667, + "step": 889 + }, + { + "epoch": 0.9753424657534246, + "grad_norm": 1.8621459007263184, + "learning_rate": 3.7774167343623073e-06, + "loss": 2.2529, + "step": 890 + }, + { + "epoch": 0.9753424657534246, + "eval_loss": 2.303697347640991, + "eval_runtime": 44.9459, + "eval_samples_per_second": 22.249, + "eval_steps_per_second": 0.934, + "step": 890 + }, + { + "epoch": 0.9764383561643836, + "grad_norm": 2.0311460494995117, + "learning_rate": 3.7753858651502846e-06, + "loss": 2.2185, + "step": 891 + }, + { + "epoch": 0.9775342465753425, + "grad_norm": 2.2017149925231934, + "learning_rate": 3.773354995938262e-06, + "loss": 2.164, + "step": 892 + }, + { + "epoch": 0.9786301369863014, + "grad_norm": 1.949684977531433, + "learning_rate": 3.771324126726239e-06, + "loss": 2.2128, + "step": 893 + }, + { + "epoch": 0.9797260273972602, + "grad_norm": 1.9395498037338257, + "learning_rate": 3.7692932575142164e-06, + "loss": 2.1957, + "step": 894 + }, + { + "epoch": 0.9808219178082191, + "grad_norm": 2.096432685852051, + "learning_rate": 3.7672623883021937e-06, + "loss": 2.219, + "step": 895 + }, + { + "epoch": 0.9819178082191781, + "grad_norm": 2.0942952632904053, + "learning_rate": 3.765231519090171e-06, + "loss": 2.1546, + "step": 896 + }, + { + "epoch": 0.983013698630137, + "grad_norm": 2.0097291469573975, + "learning_rate": 3.7632006498781482e-06, + "loss": 2.1021, + "step": 897 + }, + { + "epoch": 0.9841095890410959, + "grad_norm": 2.0909335613250732, + "learning_rate": 3.761169780666125e-06, + "loss": 2.1862, + "step": 898 + }, + { + "epoch": 0.9852054794520548, + "grad_norm": 2.187309980392456, + "learning_rate": 3.7591389114541028e-06, + "loss": 2.1511, + "step": 899 + }, + { + "epoch": 0.9863013698630136, + "grad_norm": 2.194276809692383, + "learning_rate": 3.7571080422420796e-06, + "loss": 2.196, + "step": 900 + }, + { + "epoch": 0.9863013698630136, + "eval_loss": 2.3047611713409424, + "eval_runtime": 45.609, + "eval_samples_per_second": 21.925, + "eval_steps_per_second": 0.921, + "step": 900 + }, + { + "epoch": 0.9873972602739726, + "grad_norm": 2.1132829189300537, + "learning_rate": 3.7550771730300573e-06, + "loss": 2.2046, + "step": 901 + }, + { + "epoch": 0.9884931506849315, + "grad_norm": 2.1946468353271484, + "learning_rate": 3.753046303818034e-06, + "loss": 2.1656, + "step": 902 + }, + { + "epoch": 0.9895890410958904, + "grad_norm": 2.2210536003112793, + "learning_rate": 3.751015434606012e-06, + "loss": 2.2186, + "step": 903 + }, + { + "epoch": 0.9906849315068493, + "grad_norm": 1.913695216178894, + "learning_rate": 3.7489845653939887e-06, + "loss": 2.1933, + "step": 904 + }, + { + "epoch": 0.9917808219178083, + "grad_norm": 2.1088368892669678, + "learning_rate": 3.7469536961819664e-06, + "loss": 2.2406, + "step": 905 + }, + { + "epoch": 0.9928767123287672, + "grad_norm": 2.04239821434021, + "learning_rate": 3.7449228269699433e-06, + "loss": 2.142, + "step": 906 + }, + { + "epoch": 0.993972602739726, + "grad_norm": 1.7852507829666138, + "learning_rate": 3.742891957757921e-06, + "loss": 2.1701, + "step": 907 + }, + { + "epoch": 0.9950684931506849, + "grad_norm": 2.0275776386260986, + "learning_rate": 3.740861088545898e-06, + "loss": 2.2004, + "step": 908 + }, + { + "epoch": 0.9961643835616438, + "grad_norm": 1.9028904438018799, + "learning_rate": 3.7388302193338755e-06, + "loss": 2.177, + "step": 909 + }, + { + "epoch": 0.9972602739726028, + "grad_norm": 1.7815115451812744, + "learning_rate": 3.7367993501218524e-06, + "loss": 2.207, + "step": 910 + }, + { + "epoch": 0.9972602739726028, + "eval_loss": 2.302318811416626, + "eval_runtime": 42.9296, + "eval_samples_per_second": 23.294, + "eval_steps_per_second": 0.978, + "step": 910 + }, + { + "epoch": 0.9983561643835617, + "grad_norm": 1.913689374923706, + "learning_rate": 3.7347684809098297e-06, + "loss": 2.1594, + "step": 911 + }, + { + "epoch": 0.9994520547945206, + "grad_norm": 2.171222686767578, + "learning_rate": 3.732737611697807e-06, + "loss": 2.1689, + "step": 912 + }, + { + "epoch": 1.0005479452054795, + "grad_norm": 1.831642985343933, + "learning_rate": 3.7307067424857842e-06, + "loss": 2.1367, + "step": 913 + }, + { + "epoch": 1.0016438356164383, + "grad_norm": 2.1831955909729004, + "learning_rate": 3.7286758732737615e-06, + "loss": 2.1035, + "step": 914 + }, + { + "epoch": 1.0027397260273974, + "grad_norm": 2.038025379180908, + "learning_rate": 3.7266450040617388e-06, + "loss": 2.1461, + "step": 915 + }, + { + "epoch": 1.0038356164383562, + "grad_norm": 1.866502046585083, + "learning_rate": 3.724614134849716e-06, + "loss": 2.1322, + "step": 916 + }, + { + "epoch": 1.004931506849315, + "grad_norm": 2.1359357833862305, + "learning_rate": 3.7225832656376933e-06, + "loss": 2.2333, + "step": 917 + }, + { + "epoch": 1.006027397260274, + "grad_norm": 1.8689136505126953, + "learning_rate": 3.7205523964256706e-06, + "loss": 2.1176, + "step": 918 + }, + { + "epoch": 1.0071232876712328, + "grad_norm": 1.8869425058364868, + "learning_rate": 3.718521527213648e-06, + "loss": 2.2127, + "step": 919 + }, + { + "epoch": 1.0082191780821919, + "grad_norm": 1.9148075580596924, + "learning_rate": 3.716490658001625e-06, + "loss": 2.1846, + "step": 920 + }, + { + "epoch": 1.0082191780821919, + "eval_loss": 2.30368709564209, + "eval_runtime": 52.1845, + "eval_samples_per_second": 19.163, + "eval_steps_per_second": 0.805, + "step": 920 + }, + { + "epoch": 1.0093150684931507, + "grad_norm": 1.9502761363983154, + "learning_rate": 3.7144597887896024e-06, + "loss": 2.1731, + "step": 921 + }, + { + "epoch": 1.0104109589041095, + "grad_norm": 1.9093869924545288, + "learning_rate": 3.7124289195775797e-06, + "loss": 2.0864, + "step": 922 + }, + { + "epoch": 1.0115068493150685, + "grad_norm": 2.0871293544769287, + "learning_rate": 3.710398050365557e-06, + "loss": 2.1838, + "step": 923 + }, + { + "epoch": 1.0126027397260273, + "grad_norm": 2.0559985637664795, + "learning_rate": 3.708367181153534e-06, + "loss": 2.1764, + "step": 924 + }, + { + "epoch": 1.0136986301369864, + "grad_norm": 1.8968807458877563, + "learning_rate": 3.7063363119415115e-06, + "loss": 2.1515, + "step": 925 + }, + { + "epoch": 1.0147945205479452, + "grad_norm": 1.8613216876983643, + "learning_rate": 3.7043054427294884e-06, + "loss": 2.1694, + "step": 926 + }, + { + "epoch": 1.0158904109589042, + "grad_norm": 2.002044439315796, + "learning_rate": 3.702274573517466e-06, + "loss": 2.148, + "step": 927 + }, + { + "epoch": 1.016986301369863, + "grad_norm": 2.1138675212860107, + "learning_rate": 3.700243704305443e-06, + "loss": 2.1401, + "step": 928 + }, + { + "epoch": 1.0180821917808218, + "grad_norm": 1.7620841264724731, + "learning_rate": 3.6982128350934206e-06, + "loss": 2.1399, + "step": 929 + }, + { + "epoch": 1.0191780821917809, + "grad_norm": 1.8970746994018555, + "learning_rate": 3.6961819658813975e-06, + "loss": 2.1404, + "step": 930 + }, + { + "epoch": 1.0191780821917809, + "eval_loss": 2.3068811893463135, + "eval_runtime": 53.0188, + "eval_samples_per_second": 18.861, + "eval_steps_per_second": 0.792, + "step": 930 + }, + { + "epoch": 1.0202739726027397, + "grad_norm": 1.9405478239059448, + "learning_rate": 3.694151096669375e-06, + "loss": 2.1286, + "step": 931 + }, + { + "epoch": 1.0213698630136987, + "grad_norm": 2.112222909927368, + "learning_rate": 3.692120227457352e-06, + "loss": 2.2246, + "step": 932 + }, + { + "epoch": 1.0224657534246575, + "grad_norm": 1.8792580366134644, + "learning_rate": 3.6900893582453297e-06, + "loss": 2.148, + "step": 933 + }, + { + "epoch": 1.0235616438356165, + "grad_norm": 1.9530014991760254, + "learning_rate": 3.6880584890333066e-06, + "loss": 2.1707, + "step": 934 + }, + { + "epoch": 1.0246575342465754, + "grad_norm": 2.1778736114501953, + "learning_rate": 3.6860276198212834e-06, + "loss": 2.0879, + "step": 935 + }, + { + "epoch": 1.0257534246575342, + "grad_norm": 1.8960413932800293, + "learning_rate": 3.683996750609261e-06, + "loss": 2.1659, + "step": 936 + }, + { + "epoch": 1.0268493150684932, + "grad_norm": 1.9805221557617188, + "learning_rate": 3.681965881397238e-06, + "loss": 2.1674, + "step": 937 + }, + { + "epoch": 1.027945205479452, + "grad_norm": 1.9999068975448608, + "learning_rate": 3.6799350121852157e-06, + "loss": 2.1677, + "step": 938 + }, + { + "epoch": 1.029041095890411, + "grad_norm": 2.016911029815674, + "learning_rate": 3.6779041429731925e-06, + "loss": 2.1841, + "step": 939 + }, + { + "epoch": 1.0301369863013699, + "grad_norm": 1.7200374603271484, + "learning_rate": 3.6758732737611702e-06, + "loss": 2.1391, + "step": 940 + }, + { + "epoch": 1.0301369863013699, + "eval_loss": 2.3022048473358154, + "eval_runtime": 51.2908, + "eval_samples_per_second": 19.497, + "eval_steps_per_second": 0.819, + "step": 940 + }, + { + "epoch": 1.0312328767123287, + "grad_norm": 2.249377489089966, + "learning_rate": 3.673842404549147e-06, + "loss": 2.1618, + "step": 941 + }, + { + "epoch": 1.0323287671232877, + "grad_norm": 1.838922381401062, + "learning_rate": 3.671811535337125e-06, + "loss": 2.1527, + "step": 942 + }, + { + "epoch": 1.0334246575342465, + "grad_norm": 1.9571045637130737, + "learning_rate": 3.6697806661251016e-06, + "loss": 2.149, + "step": 943 + }, + { + "epoch": 1.0345205479452055, + "grad_norm": 1.9929102659225464, + "learning_rate": 3.6677497969130793e-06, + "loss": 2.159, + "step": 944 + }, + { + "epoch": 1.0356164383561643, + "grad_norm": 2.004830837249756, + "learning_rate": 3.665718927701056e-06, + "loss": 2.2369, + "step": 945 + }, + { + "epoch": 1.0367123287671234, + "grad_norm": 1.9605739116668701, + "learning_rate": 3.663688058489034e-06, + "loss": 2.1795, + "step": 946 + }, + { + "epoch": 1.0378082191780822, + "grad_norm": 1.9499164819717407, + "learning_rate": 3.6616571892770107e-06, + "loss": 2.1979, + "step": 947 + }, + { + "epoch": 1.038904109589041, + "grad_norm": 1.867347240447998, + "learning_rate": 3.659626320064988e-06, + "loss": 2.1616, + "step": 948 + }, + { + "epoch": 1.04, + "grad_norm": 1.9342801570892334, + "learning_rate": 3.6575954508529653e-06, + "loss": 2.1044, + "step": 949 + }, + { + "epoch": 1.0410958904109588, + "grad_norm": 2.0009584426879883, + "learning_rate": 3.6555645816409426e-06, + "loss": 2.1543, + "step": 950 + }, + { + "epoch": 1.0410958904109588, + "eval_loss": 2.3028438091278076, + "eval_runtime": 46.7426, + "eval_samples_per_second": 21.394, + "eval_steps_per_second": 0.899, + "step": 950 + }, + { + "epoch": 1.0421917808219179, + "grad_norm": 2.0411503314971924, + "learning_rate": 3.65353371242892e-06, + "loss": 2.1465, + "step": 951 + }, + { + "epoch": 1.0432876712328767, + "grad_norm": 2.0512070655822754, + "learning_rate": 3.651502843216897e-06, + "loss": 2.1828, + "step": 952 + }, + { + "epoch": 1.0443835616438357, + "grad_norm": 2.0948643684387207, + "learning_rate": 3.6494719740048744e-06, + "loss": 2.0671, + "step": 953 + }, + { + "epoch": 1.0454794520547945, + "grad_norm": 1.9103025197982788, + "learning_rate": 3.6474411047928517e-06, + "loss": 2.1651, + "step": 954 + }, + { + "epoch": 1.0465753424657533, + "grad_norm": 2.1500349044799805, + "learning_rate": 3.645410235580829e-06, + "loss": 2.137, + "step": 955 + }, + { + "epoch": 1.0476712328767124, + "grad_norm": 1.7760887145996094, + "learning_rate": 3.6433793663688062e-06, + "loss": 2.1738, + "step": 956 + }, + { + "epoch": 1.0487671232876712, + "grad_norm": 2.0111730098724365, + "learning_rate": 3.6413484971567835e-06, + "loss": 2.1569, + "step": 957 + }, + { + "epoch": 1.0498630136986302, + "grad_norm": 2.0040462017059326, + "learning_rate": 3.6393176279447608e-06, + "loss": 2.1483, + "step": 958 + }, + { + "epoch": 1.050958904109589, + "grad_norm": 1.9583499431610107, + "learning_rate": 3.637286758732738e-06, + "loss": 2.1575, + "step": 959 + }, + { + "epoch": 1.0520547945205478, + "grad_norm": 1.8842004537582397, + "learning_rate": 3.6352558895207153e-06, + "loss": 2.1797, + "step": 960 + }, + { + "epoch": 1.0520547945205478, + "eval_loss": 2.301715850830078, + "eval_runtime": 52.7345, + "eval_samples_per_second": 18.963, + "eval_steps_per_second": 0.796, + "step": 960 + }, + { + "epoch": 1.0531506849315069, + "grad_norm": 2.020909309387207, + "learning_rate": 3.633225020308692e-06, + "loss": 2.1939, + "step": 961 + }, + { + "epoch": 1.0542465753424657, + "grad_norm": 2.09222149848938, + "learning_rate": 3.63119415109667e-06, + "loss": 2.1777, + "step": 962 + }, + { + "epoch": 1.0553424657534247, + "grad_norm": 1.8582347631454468, + "learning_rate": 3.6291632818846467e-06, + "loss": 2.1505, + "step": 963 + }, + { + "epoch": 1.0564383561643835, + "grad_norm": 1.9468824863433838, + "learning_rate": 3.6271324126726244e-06, + "loss": 2.1762, + "step": 964 + }, + { + "epoch": 1.0575342465753426, + "grad_norm": 1.9484895467758179, + "learning_rate": 3.6251015434606013e-06, + "loss": 2.1102, + "step": 965 + }, + { + "epoch": 1.0586301369863014, + "grad_norm": 2.11462140083313, + "learning_rate": 3.623070674248579e-06, + "loss": 2.1114, + "step": 966 + }, + { + "epoch": 1.0597260273972602, + "grad_norm": 2.2038373947143555, + "learning_rate": 3.621039805036556e-06, + "loss": 2.1513, + "step": 967 + }, + { + "epoch": 1.0608219178082192, + "grad_norm": 2.1271684169769287, + "learning_rate": 3.6190089358245335e-06, + "loss": 2.137, + "step": 968 + }, + { + "epoch": 1.061917808219178, + "grad_norm": 1.8624194860458374, + "learning_rate": 3.6169780666125104e-06, + "loss": 2.1701, + "step": 969 + }, + { + "epoch": 1.063013698630137, + "grad_norm": 1.935445785522461, + "learning_rate": 3.614947197400488e-06, + "loss": 2.1726, + "step": 970 + }, + { + "epoch": 1.063013698630137, + "eval_loss": 2.302633047103882, + "eval_runtime": 51.8284, + "eval_samples_per_second": 19.294, + "eval_steps_per_second": 0.81, + "step": 970 + }, + { + "epoch": 1.0641095890410959, + "grad_norm": 2.048907518386841, + "learning_rate": 3.612916328188465e-06, + "loss": 2.1712, + "step": 971 + }, + { + "epoch": 1.0652054794520547, + "grad_norm": 1.850943684577942, + "learning_rate": 3.610885458976442e-06, + "loss": 2.102, + "step": 972 + }, + { + "epoch": 1.0663013698630137, + "grad_norm": 1.8364315032958984, + "learning_rate": 3.6088545897644195e-06, + "loss": 2.1903, + "step": 973 + }, + { + "epoch": 1.0673972602739725, + "grad_norm": 1.8227031230926514, + "learning_rate": 3.6068237205523963e-06, + "loss": 2.1787, + "step": 974 + }, + { + "epoch": 1.0684931506849316, + "grad_norm": 1.9136372804641724, + "learning_rate": 3.604792851340374e-06, + "loss": 2.1244, + "step": 975 + }, + { + "epoch": 1.0695890410958904, + "grad_norm": 1.8585968017578125, + "learning_rate": 3.602761982128351e-06, + "loss": 2.1287, + "step": 976 + }, + { + "epoch": 1.0706849315068494, + "grad_norm": 2.139573812484741, + "learning_rate": 3.6007311129163286e-06, + "loss": 2.1938, + "step": 977 + }, + { + "epoch": 1.0717808219178082, + "grad_norm": 2.1584925651550293, + "learning_rate": 3.5987002437043054e-06, + "loss": 2.1626, + "step": 978 + }, + { + "epoch": 1.072876712328767, + "grad_norm": 1.8905949592590332, + "learning_rate": 3.596669374492283e-06, + "loss": 2.1219, + "step": 979 + }, + { + "epoch": 1.073972602739726, + "grad_norm": 2.099651575088501, + "learning_rate": 3.59463850528026e-06, + "loss": 2.1385, + "step": 980 + }, + { + "epoch": 1.073972602739726, + "eval_loss": 2.3004090785980225, + "eval_runtime": 52.7406, + "eval_samples_per_second": 18.961, + "eval_steps_per_second": 0.796, + "step": 980 + }, + { + "epoch": 1.0750684931506849, + "grad_norm": 2.1456780433654785, + "learning_rate": 3.5926076360682377e-06, + "loss": 2.1973, + "step": 981 + }, + { + "epoch": 1.076164383561644, + "grad_norm": 1.929372787475586, + "learning_rate": 3.5905767668562145e-06, + "loss": 2.1115, + "step": 982 + }, + { + "epoch": 1.0772602739726027, + "grad_norm": 2.08174991607666, + "learning_rate": 3.5885458976441922e-06, + "loss": 2.1785, + "step": 983 + }, + { + "epoch": 1.0783561643835617, + "grad_norm": 1.9355952739715576, + "learning_rate": 3.586515028432169e-06, + "loss": 2.1567, + "step": 984 + }, + { + "epoch": 1.0794520547945206, + "grad_norm": 1.9983466863632202, + "learning_rate": 3.5844841592201464e-06, + "loss": 2.1913, + "step": 985 + }, + { + "epoch": 1.0805479452054794, + "grad_norm": 1.8714051246643066, + "learning_rate": 3.5824532900081237e-06, + "loss": 2.1938, + "step": 986 + }, + { + "epoch": 1.0816438356164384, + "grad_norm": 2.007049798965454, + "learning_rate": 3.580422420796101e-06, + "loss": 2.1093, + "step": 987 + }, + { + "epoch": 1.0827397260273972, + "grad_norm": 1.965572476387024, + "learning_rate": 3.578391551584078e-06, + "loss": 2.1426, + "step": 988 + }, + { + "epoch": 1.0838356164383562, + "grad_norm": 1.9040261507034302, + "learning_rate": 3.5763606823720555e-06, + "loss": 2.1627, + "step": 989 + }, + { + "epoch": 1.084931506849315, + "grad_norm": 2.001929759979248, + "learning_rate": 3.5743298131600328e-06, + "loss": 2.0983, + "step": 990 + }, + { + "epoch": 1.084931506849315, + "eval_loss": 2.2979862689971924, + "eval_runtime": 50.6325, + "eval_samples_per_second": 19.75, + "eval_steps_per_second": 0.83, + "step": 990 + }, + { + "epoch": 1.086027397260274, + "grad_norm": 1.9204787015914917, + "learning_rate": 3.57229894394801e-06, + "loss": 2.1571, + "step": 991 + }, + { + "epoch": 1.087123287671233, + "grad_norm": 1.9628651142120361, + "learning_rate": 3.5702680747359873e-06, + "loss": 2.1736, + "step": 992 + }, + { + "epoch": 1.0882191780821917, + "grad_norm": 1.9888029098510742, + "learning_rate": 3.5682372055239646e-06, + "loss": 2.17, + "step": 993 + }, + { + "epoch": 1.0893150684931507, + "grad_norm": 2.049769401550293, + "learning_rate": 3.566206336311942e-06, + "loss": 2.1352, + "step": 994 + }, + { + "epoch": 1.0904109589041096, + "grad_norm": 2.0208375453948975, + "learning_rate": 3.564175467099919e-06, + "loss": 2.1564, + "step": 995 + }, + { + "epoch": 1.0915068493150686, + "grad_norm": 1.976775050163269, + "learning_rate": 3.5621445978878964e-06, + "loss": 2.1292, + "step": 996 + }, + { + "epoch": 1.0926027397260274, + "grad_norm": 2.0502822399139404, + "learning_rate": 3.5601137286758737e-06, + "loss": 2.1115, + "step": 997 + }, + { + "epoch": 1.0936986301369862, + "grad_norm": 2.1184189319610596, + "learning_rate": 3.5580828594638505e-06, + "loss": 2.1641, + "step": 998 + }, + { + "epoch": 1.0947945205479452, + "grad_norm": 1.9662559032440186, + "learning_rate": 3.5560519902518282e-06, + "loss": 2.1338, + "step": 999 + }, + { + "epoch": 1.095890410958904, + "grad_norm": 2.13386869430542, + "learning_rate": 3.554021121039805e-06, + "loss": 2.158, + "step": 1000 + }, + { + "epoch": 1.095890410958904, + "eval_loss": 2.3001608848571777, + "eval_runtime": 45.9219, + "eval_samples_per_second": 21.776, + "eval_steps_per_second": 0.915, + "step": 1000 + }, + { + "epoch": 1.096986301369863, + "grad_norm": 1.8791636228561401, + "learning_rate": 3.551990251827783e-06, + "loss": 2.1752, + "step": 1001 + }, + { + "epoch": 1.098082191780822, + "grad_norm": 1.9105546474456787, + "learning_rate": 3.5499593826157596e-06, + "loss": 2.1543, + "step": 1002 + }, + { + "epoch": 1.099178082191781, + "grad_norm": 1.887668251991272, + "learning_rate": 3.5479285134037373e-06, + "loss": 2.1226, + "step": 1003 + }, + { + "epoch": 1.1002739726027397, + "grad_norm": 2.2854599952697754, + "learning_rate": 3.545897644191714e-06, + "loss": 2.1425, + "step": 1004 + }, + { + "epoch": 1.1013698630136985, + "grad_norm": 2.03892183303833, + "learning_rate": 3.543866774979692e-06, + "loss": 2.135, + "step": 1005 + }, + { + "epoch": 1.1024657534246576, + "grad_norm": 1.9144858121871948, + "learning_rate": 3.5418359057676687e-06, + "loss": 2.1323, + "step": 1006 + }, + { + "epoch": 1.1035616438356164, + "grad_norm": 2.1402337551116943, + "learning_rate": 3.5398050365556464e-06, + "loss": 2.1668, + "step": 1007 + }, + { + "epoch": 1.1046575342465754, + "grad_norm": 2.084592819213867, + "learning_rate": 3.5377741673436233e-06, + "loss": 2.1331, + "step": 1008 + }, + { + "epoch": 1.1057534246575342, + "grad_norm": 2.0012383460998535, + "learning_rate": 3.535743298131601e-06, + "loss": 2.1677, + "step": 1009 + }, + { + "epoch": 1.106849315068493, + "grad_norm": 2.272204637527466, + "learning_rate": 3.533712428919578e-06, + "loss": 2.13, + "step": 1010 + }, + { + "epoch": 1.106849315068493, + "eval_loss": 2.3001229763031006, + "eval_runtime": 42.0444, + "eval_samples_per_second": 23.784, + "eval_steps_per_second": 0.999, + "step": 1010 + }, + { + "epoch": 1.107945205479452, + "grad_norm": 2.1061534881591797, + "learning_rate": 3.5316815597075547e-06, + "loss": 2.1581, + "step": 1011 + }, + { + "epoch": 1.1090410958904109, + "grad_norm": 2.159306764602661, + "learning_rate": 3.5296506904955324e-06, + "loss": 2.1868, + "step": 1012 + }, + { + "epoch": 1.11013698630137, + "grad_norm": 2.1429545879364014, + "learning_rate": 3.5276198212835093e-06, + "loss": 2.1274, + "step": 1013 + }, + { + "epoch": 1.1112328767123287, + "grad_norm": 1.8494454622268677, + "learning_rate": 3.525588952071487e-06, + "loss": 2.1513, + "step": 1014 + }, + { + "epoch": 1.1123287671232878, + "grad_norm": 1.8494454622268677, + "learning_rate": 3.525588952071487e-06, + "loss": 2.1883, + "step": 1015 + }, + { + "epoch": 1.1134246575342466, + "grad_norm": 2.2244861125946045, + "learning_rate": 3.523558082859464e-06, + "loss": 2.1896, + "step": 1016 + }, + { + "epoch": 1.1145205479452054, + "grad_norm": 2.181950569152832, + "learning_rate": 3.5215272136474415e-06, + "loss": 2.1561, + "step": 1017 + }, + { + "epoch": 1.1156164383561644, + "grad_norm": 1.7740849256515503, + "learning_rate": 3.5194963444354184e-06, + "loss": 2.1358, + "step": 1018 + }, + { + "epoch": 1.1167123287671232, + "grad_norm": 2.3540139198303223, + "learning_rate": 3.517465475223396e-06, + "loss": 2.1241, + "step": 1019 + }, + { + "epoch": 1.1178082191780823, + "grad_norm": 2.2737820148468018, + "learning_rate": 3.515434606011373e-06, + "loss": 2.1117, + "step": 1020 + }, + { + "epoch": 1.1178082191780823, + "eval_loss": 2.2998275756835938, + "eval_runtime": 46.2142, + "eval_samples_per_second": 21.638, + "eval_steps_per_second": 0.909, + "step": 1020 + }, + { + "epoch": 1.118904109589041, + "grad_norm": 2.0705313682556152, + "learning_rate": 3.5134037367993506e-06, + "loss": 2.1743, + "step": 1021 + }, + { + "epoch": 1.12, + "grad_norm": 2.2605583667755127, + "learning_rate": 3.5113728675873275e-06, + "loss": 2.2082, + "step": 1022 + }, + { + "epoch": 1.121095890410959, + "grad_norm": 2.028203248977661, + "learning_rate": 3.5093419983753047e-06, + "loss": 2.1722, + "step": 1023 + }, + { + "epoch": 1.1221917808219177, + "grad_norm": 1.9794270992279053, + "learning_rate": 3.507311129163282e-06, + "loss": 2.1581, + "step": 1024 + }, + { + "epoch": 1.1232876712328768, + "grad_norm": 2.1830344200134277, + "learning_rate": 3.5052802599512593e-06, + "loss": 2.1142, + "step": 1025 + }, + { + "epoch": 1.1243835616438356, + "grad_norm": 1.926081895828247, + "learning_rate": 3.5032493907392366e-06, + "loss": 2.1532, + "step": 1026 + }, + { + "epoch": 1.1254794520547946, + "grad_norm": 1.9909394979476929, + "learning_rate": 3.501218521527214e-06, + "loss": 2.153, + "step": 1027 + }, + { + "epoch": 1.1265753424657534, + "grad_norm": 1.9685133695602417, + "learning_rate": 3.499187652315191e-06, + "loss": 2.0983, + "step": 1028 + }, + { + "epoch": 1.1276712328767124, + "grad_norm": 1.9185279607772827, + "learning_rate": 3.4971567831031684e-06, + "loss": 2.1495, + "step": 1029 + }, + { + "epoch": 1.1287671232876713, + "grad_norm": 2.0214717388153076, + "learning_rate": 3.4951259138911457e-06, + "loss": 2.176, + "step": 1030 + }, + { + "epoch": 1.1287671232876713, + "eval_loss": 2.2963690757751465, + "eval_runtime": 44.2655, + "eval_samples_per_second": 22.591, + "eval_steps_per_second": 0.949, + "step": 1030 + }, + { + "epoch": 1.12986301369863, + "grad_norm": 1.9134252071380615, + "learning_rate": 3.493095044679123e-06, + "loss": 2.1034, + "step": 1031 + }, + { + "epoch": 1.130958904109589, + "grad_norm": 1.8448772430419922, + "learning_rate": 3.4910641754671002e-06, + "loss": 2.1947, + "step": 1032 + }, + { + "epoch": 1.132054794520548, + "grad_norm": 1.856341004371643, + "learning_rate": 3.4890333062550775e-06, + "loss": 2.0676, + "step": 1033 + }, + { + "epoch": 1.133150684931507, + "grad_norm": 2.039062261581421, + "learning_rate": 3.4870024370430548e-06, + "loss": 2.1618, + "step": 1034 + }, + { + "epoch": 1.1342465753424658, + "grad_norm": 1.8440991640090942, + "learning_rate": 3.484971567831032e-06, + "loss": 2.0956, + "step": 1035 + }, + { + "epoch": 1.1353424657534246, + "grad_norm": 1.8483991622924805, + "learning_rate": 3.482940698619009e-06, + "loss": 2.1364, + "step": 1036 + }, + { + "epoch": 1.1364383561643836, + "grad_norm": 1.9045069217681885, + "learning_rate": 3.4809098294069866e-06, + "loss": 2.1786, + "step": 1037 + }, + { + "epoch": 1.1375342465753424, + "grad_norm": 1.8203203678131104, + "learning_rate": 3.4788789601949634e-06, + "loss": 2.1438, + "step": 1038 + }, + { + "epoch": 1.1386301369863014, + "grad_norm": 2.1084606647491455, + "learning_rate": 3.476848090982941e-06, + "loss": 2.1553, + "step": 1039 + }, + { + "epoch": 1.1397260273972603, + "grad_norm": 2.1041061878204346, + "learning_rate": 3.474817221770918e-06, + "loss": 2.1435, + "step": 1040 + }, + { + "epoch": 1.1397260273972603, + "eval_loss": 2.2970690727233887, + "eval_runtime": 43.2304, + "eval_samples_per_second": 23.132, + "eval_steps_per_second": 0.972, + "step": 1040 + }, + { + "epoch": 1.140821917808219, + "grad_norm": 1.9701104164123535, + "learning_rate": 3.4727863525588957e-06, + "loss": 2.2201, + "step": 1041 + }, + { + "epoch": 1.141917808219178, + "grad_norm": 1.8816475868225098, + "learning_rate": 3.4707554833468725e-06, + "loss": 2.1643, + "step": 1042 + }, + { + "epoch": 1.143013698630137, + "grad_norm": 2.096355438232422, + "learning_rate": 3.4687246141348502e-06, + "loss": 2.1514, + "step": 1043 + }, + { + "epoch": 1.144109589041096, + "grad_norm": 1.9975471496582031, + "learning_rate": 3.466693744922827e-06, + "loss": 2.1246, + "step": 1044 + }, + { + "epoch": 1.1452054794520548, + "grad_norm": 2.0587127208709717, + "learning_rate": 3.464662875710805e-06, + "loss": 2.1371, + "step": 1045 + }, + { + "epoch": 1.1463013698630138, + "grad_norm": 1.9275381565093994, + "learning_rate": 3.4626320064987817e-06, + "loss": 2.1518, + "step": 1046 + }, + { + "epoch": 1.1473972602739726, + "grad_norm": 2.1690948009490967, + "learning_rate": 3.4606011372867594e-06, + "loss": 2.1528, + "step": 1047 + }, + { + "epoch": 1.1484931506849314, + "grad_norm": 1.970973014831543, + "learning_rate": 3.458570268074736e-06, + "loss": 2.1569, + "step": 1048 + }, + { + "epoch": 1.1495890410958904, + "grad_norm": 1.8897690773010254, + "learning_rate": 3.4565393988627135e-06, + "loss": 2.1401, + "step": 1049 + }, + { + "epoch": 1.1506849315068493, + "grad_norm": 1.9245911836624146, + "learning_rate": 3.4545085296506908e-06, + "loss": 2.1311, + "step": 1050 + }, + { + "epoch": 1.1506849315068493, + "eval_loss": 2.2956528663635254, + "eval_runtime": 44.1778, + "eval_samples_per_second": 22.636, + "eval_steps_per_second": 0.951, + "step": 1050 + }, + { + "epoch": 1.1517808219178083, + "grad_norm": 2.142209529876709, + "learning_rate": 3.452477660438668e-06, + "loss": 2.1144, + "step": 1051 + }, + { + "epoch": 1.152876712328767, + "grad_norm": 1.932369351387024, + "learning_rate": 3.4504467912266453e-06, + "loss": 2.1739, + "step": 1052 + }, + { + "epoch": 1.1539726027397261, + "grad_norm": 2.259359359741211, + "learning_rate": 3.4484159220146226e-06, + "loss": 2.1036, + "step": 1053 + }, + { + "epoch": 1.155068493150685, + "grad_norm": 2.112485647201538, + "learning_rate": 3.4463850528026e-06, + "loss": 2.1429, + "step": 1054 + }, + { + "epoch": 1.1561643835616437, + "grad_norm": 2.6211049556732178, + "learning_rate": 3.444354183590577e-06, + "loss": 2.1421, + "step": 1055 + }, + { + "epoch": 1.1572602739726028, + "grad_norm": 1.8724851608276367, + "learning_rate": 3.4423233143785544e-06, + "loss": 2.1622, + "step": 1056 + }, + { + "epoch": 1.1583561643835616, + "grad_norm": 1.9292882680892944, + "learning_rate": 3.4402924451665317e-06, + "loss": 2.1071, + "step": 1057 + }, + { + "epoch": 1.1594520547945206, + "grad_norm": 1.9832521677017212, + "learning_rate": 3.438261575954509e-06, + "loss": 2.0879, + "step": 1058 + }, + { + "epoch": 1.1605479452054794, + "grad_norm": 1.9720780849456787, + "learning_rate": 3.4362307067424862e-06, + "loss": 2.1603, + "step": 1059 + }, + { + "epoch": 1.1616438356164385, + "grad_norm": 1.8179587125778198, + "learning_rate": 3.434199837530463e-06, + "loss": 2.1851, + "step": 1060 + }, + { + "epoch": 1.1616438356164385, + "eval_loss": 2.2950079441070557, + "eval_runtime": 42.9625, + "eval_samples_per_second": 23.276, + "eval_steps_per_second": 0.978, + "step": 1060 + }, + { + "epoch": 1.1627397260273973, + "grad_norm": 1.8762129545211792, + "learning_rate": 3.432168968318441e-06, + "loss": 2.1527, + "step": 1061 + }, + { + "epoch": 1.163835616438356, + "grad_norm": 2.173297643661499, + "learning_rate": 3.4301380991064176e-06, + "loss": 2.1023, + "step": 1062 + }, + { + "epoch": 1.1649315068493151, + "grad_norm": 1.9358222484588623, + "learning_rate": 3.4281072298943953e-06, + "loss": 2.1638, + "step": 1063 + }, + { + "epoch": 1.166027397260274, + "grad_norm": 1.9421939849853516, + "learning_rate": 3.426076360682372e-06, + "loss": 2.1794, + "step": 1064 + }, + { + "epoch": 1.167123287671233, + "grad_norm": 2.212996006011963, + "learning_rate": 3.42404549147035e-06, + "loss": 2.1877, + "step": 1065 + }, + { + "epoch": 1.1682191780821918, + "grad_norm": 2.036180257797241, + "learning_rate": 3.4220146222583267e-06, + "loss": 2.1534, + "step": 1066 + }, + { + "epoch": 1.1693150684931506, + "grad_norm": 1.8819142580032349, + "learning_rate": 3.4199837530463044e-06, + "loss": 2.099, + "step": 1067 + }, + { + "epoch": 1.1704109589041096, + "grad_norm": 2.0781915187835693, + "learning_rate": 3.4179528838342813e-06, + "loss": 2.1326, + "step": 1068 + }, + { + "epoch": 1.1715068493150684, + "grad_norm": 2.034895181655884, + "learning_rate": 3.415922014622259e-06, + "loss": 2.2107, + "step": 1069 + }, + { + "epoch": 1.1726027397260275, + "grad_norm": 1.92604660987854, + "learning_rate": 3.413891145410236e-06, + "loss": 2.1773, + "step": 1070 + }, + { + "epoch": 1.1726027397260275, + "eval_loss": 2.2959954738616943, + "eval_runtime": 47.2123, + "eval_samples_per_second": 21.181, + "eval_steps_per_second": 0.89, + "step": 1070 + }, + { + "epoch": 1.1736986301369863, + "grad_norm": 1.762730598449707, + "learning_rate": 3.4118602761982135e-06, + "loss": 2.1803, + "step": 1071 + }, + { + "epoch": 1.174794520547945, + "grad_norm": 2.0661215782165527, + "learning_rate": 3.4098294069861904e-06, + "loss": 2.1919, + "step": 1072 + }, + { + "epoch": 1.1758904109589041, + "grad_norm": 2.079982280731201, + "learning_rate": 3.4077985377741673e-06, + "loss": 2.1536, + "step": 1073 + }, + { + "epoch": 1.176986301369863, + "grad_norm": 1.9044359922409058, + "learning_rate": 3.405767668562145e-06, + "loss": 2.1403, + "step": 1074 + }, + { + "epoch": 1.178082191780822, + "grad_norm": 1.8870306015014648, + "learning_rate": 3.403736799350122e-06, + "loss": 2.1845, + "step": 1075 + }, + { + "epoch": 1.1791780821917808, + "grad_norm": 1.9305400848388672, + "learning_rate": 3.4017059301380995e-06, + "loss": 2.1786, + "step": 1076 + }, + { + "epoch": 1.1802739726027398, + "grad_norm": 1.987713098526001, + "learning_rate": 3.3996750609260764e-06, + "loss": 2.1214, + "step": 1077 + }, + { + "epoch": 1.1813698630136986, + "grad_norm": 1.8673466444015503, + "learning_rate": 3.397644191714054e-06, + "loss": 2.2027, + "step": 1078 + }, + { + "epoch": 1.1824657534246574, + "grad_norm": 2.009316921234131, + "learning_rate": 3.395613322502031e-06, + "loss": 2.139, + "step": 1079 + }, + { + "epoch": 1.1835616438356165, + "grad_norm": 2.1024956703186035, + "learning_rate": 3.3935824532900086e-06, + "loss": 2.1438, + "step": 1080 + }, + { + "epoch": 1.1835616438356165, + "eval_loss": 2.2966299057006836, + "eval_runtime": 44.8946, + "eval_samples_per_second": 22.274, + "eval_steps_per_second": 0.936, + "step": 1080 + }, + { + "epoch": 1.1846575342465753, + "grad_norm": 1.8911967277526855, + "learning_rate": 3.3915515840779855e-06, + "loss": 2.1238, + "step": 1081 + }, + { + "epoch": 1.1857534246575343, + "grad_norm": 2.0649192333221436, + "learning_rate": 3.389520714865963e-06, + "loss": 2.1534, + "step": 1082 + }, + { + "epoch": 1.1868493150684931, + "grad_norm": 1.9236847162246704, + "learning_rate": 3.38748984565394e-06, + "loss": 2.1427, + "step": 1083 + }, + { + "epoch": 1.1879452054794521, + "grad_norm": 1.8662182092666626, + "learning_rate": 3.3854589764419177e-06, + "loss": 2.1699, + "step": 1084 + }, + { + "epoch": 1.189041095890411, + "grad_norm": 1.9226020574569702, + "learning_rate": 3.3834281072298946e-06, + "loss": 2.1192, + "step": 1085 + }, + { + "epoch": 1.1901369863013698, + "grad_norm": 1.9659301042556763, + "learning_rate": 3.381397238017872e-06, + "loss": 2.1529, + "step": 1086 + }, + { + "epoch": 1.1912328767123288, + "grad_norm": 2.2714693546295166, + "learning_rate": 3.379366368805849e-06, + "loss": 2.1273, + "step": 1087 + }, + { + "epoch": 1.1923287671232876, + "grad_norm": 2.1853270530700684, + "learning_rate": 3.3773354995938264e-06, + "loss": 2.2029, + "step": 1088 + }, + { + "epoch": 1.1934246575342466, + "grad_norm": 1.8302830457687378, + "learning_rate": 3.3753046303818037e-06, + "loss": 2.1599, + "step": 1089 + }, + { + "epoch": 1.1945205479452055, + "grad_norm": 2.005591869354248, + "learning_rate": 3.373273761169781e-06, + "loss": 2.139, + "step": 1090 + }, + { + "epoch": 1.1945205479452055, + "eval_loss": 2.295994520187378, + "eval_runtime": 45.4802, + "eval_samples_per_second": 21.988, + "eval_steps_per_second": 0.923, + "step": 1090 + }, + { + "epoch": 1.1956164383561645, + "grad_norm": 2.139267683029175, + "learning_rate": 3.3712428919577582e-06, + "loss": 2.151, + "step": 1091 + }, + { + "epoch": 1.1967123287671233, + "grad_norm": 1.8547207117080688, + "learning_rate": 3.3692120227457355e-06, + "loss": 2.1249, + "step": 1092 + }, + { + "epoch": 1.1978082191780821, + "grad_norm": 2.214583396911621, + "learning_rate": 3.3671811535337128e-06, + "loss": 2.1656, + "step": 1093 + }, + { + "epoch": 1.1989041095890411, + "grad_norm": 2.1608800888061523, + "learning_rate": 3.36515028432169e-06, + "loss": 2.1056, + "step": 1094 + }, + { + "epoch": 1.2, + "grad_norm": 2.0850605964660645, + "learning_rate": 3.3631194151096673e-06, + "loss": 2.1215, + "step": 1095 + }, + { + "epoch": 1.201095890410959, + "grad_norm": 1.9199103116989136, + "learning_rate": 3.3610885458976446e-06, + "loss": 2.1584, + "step": 1096 + }, + { + "epoch": 1.2021917808219178, + "grad_norm": 1.9766334295272827, + "learning_rate": 3.359057676685622e-06, + "loss": 2.1821, + "step": 1097 + }, + { + "epoch": 1.2032876712328768, + "grad_norm": 1.948988676071167, + "learning_rate": 3.357026807473599e-06, + "loss": 2.182, + "step": 1098 + }, + { + "epoch": 1.2043835616438356, + "grad_norm": 2.0419397354125977, + "learning_rate": 3.354995938261576e-06, + "loss": 2.1051, + "step": 1099 + }, + { + "epoch": 1.2054794520547945, + "grad_norm": 1.9751213788986206, + "learning_rate": 3.3529650690495537e-06, + "loss": 2.1255, + "step": 1100 + }, + { + "epoch": 1.2054794520547945, + "eval_loss": 2.293811798095703, + "eval_runtime": 46.1182, + "eval_samples_per_second": 21.683, + "eval_steps_per_second": 0.911, + "step": 1100 + }, + { + "epoch": 1.2065753424657535, + "grad_norm": 1.9806510210037231, + "learning_rate": 3.3509341998375305e-06, + "loss": 2.1021, + "step": 1101 + }, + { + "epoch": 1.2076712328767123, + "grad_norm": 2.0570414066314697, + "learning_rate": 3.3489033306255082e-06, + "loss": 2.1417, + "step": 1102 + }, + { + "epoch": 1.208767123287671, + "grad_norm": 1.868134617805481, + "learning_rate": 3.346872461413485e-06, + "loss": 2.1494, + "step": 1103 + }, + { + "epoch": 1.2098630136986301, + "grad_norm": 2.0156848430633545, + "learning_rate": 3.344841592201463e-06, + "loss": 2.1238, + "step": 1104 + }, + { + "epoch": 1.210958904109589, + "grad_norm": 2.082301616668701, + "learning_rate": 3.3428107229894397e-06, + "loss": 2.1105, + "step": 1105 + }, + { + "epoch": 1.212054794520548, + "grad_norm": 1.8853904008865356, + "learning_rate": 3.3407798537774174e-06, + "loss": 2.1406, + "step": 1106 + }, + { + "epoch": 1.2131506849315068, + "grad_norm": 2.0410518646240234, + "learning_rate": 3.338748984565394e-06, + "loss": 2.1567, + "step": 1107 + }, + { + "epoch": 1.2142465753424658, + "grad_norm": 1.9901232719421387, + "learning_rate": 3.336718115353372e-06, + "loss": 2.1778, + "step": 1108 + }, + { + "epoch": 1.2153424657534246, + "grad_norm": 1.9840792417526245, + "learning_rate": 3.3346872461413488e-06, + "loss": 2.1935, + "step": 1109 + }, + { + "epoch": 1.2164383561643834, + "grad_norm": 1.8854445219039917, + "learning_rate": 3.3326563769293256e-06, + "loss": 2.137, + "step": 1110 + }, + { + "epoch": 1.2164383561643834, + "eval_loss": 2.294447660446167, + "eval_runtime": 42.6423, + "eval_samples_per_second": 23.451, + "eval_steps_per_second": 0.985, + "step": 1110 + }, + { + "epoch": 1.2175342465753425, + "grad_norm": 2.0948665142059326, + "learning_rate": 3.3306255077173033e-06, + "loss": 2.1288, + "step": 1111 + }, + { + "epoch": 1.2186301369863013, + "grad_norm": 1.859448790550232, + "learning_rate": 3.32859463850528e-06, + "loss": 2.1403, + "step": 1112 + }, + { + "epoch": 1.2197260273972603, + "grad_norm": 1.92074453830719, + "learning_rate": 3.326563769293258e-06, + "loss": 2.141, + "step": 1113 + }, + { + "epoch": 1.2208219178082191, + "grad_norm": 1.8377320766448975, + "learning_rate": 3.3245329000812347e-06, + "loss": 2.0889, + "step": 1114 + }, + { + "epoch": 1.2219178082191782, + "grad_norm": 1.9320917129516602, + "learning_rate": 3.3225020308692124e-06, + "loss": 2.1254, + "step": 1115 + }, + { + "epoch": 1.223013698630137, + "grad_norm": 1.9691251516342163, + "learning_rate": 3.3204711616571893e-06, + "loss": 2.1409, + "step": 1116 + }, + { + "epoch": 1.2241095890410958, + "grad_norm": 1.8214130401611328, + "learning_rate": 3.318440292445167e-06, + "loss": 2.1046, + "step": 1117 + }, + { + "epoch": 1.2252054794520548, + "grad_norm": 1.959455966949463, + "learning_rate": 3.316409423233144e-06, + "loss": 2.1418, + "step": 1118 + }, + { + "epoch": 1.2263013698630136, + "grad_norm": 1.9920449256896973, + "learning_rate": 3.3143785540211215e-06, + "loss": 2.1192, + "step": 1119 + }, + { + "epoch": 1.2273972602739727, + "grad_norm": 2.0276854038238525, + "learning_rate": 3.3123476848090984e-06, + "loss": 2.1939, + "step": 1120 + }, + { + "epoch": 1.2273972602739727, + "eval_loss": 2.295330047607422, + "eval_runtime": 49.8691, + "eval_samples_per_second": 20.052, + "eval_steps_per_second": 0.842, + "step": 1120 + }, + { + "epoch": 1.2284931506849315, + "grad_norm": 1.9765129089355469, + "learning_rate": 3.310316815597076e-06, + "loss": 2.1545, + "step": 1121 + }, + { + "epoch": 1.2295890410958905, + "grad_norm": 1.8449276685714722, + "learning_rate": 3.308285946385053e-06, + "loss": 2.1459, + "step": 1122 + }, + { + "epoch": 1.2306849315068493, + "grad_norm": 1.953072190284729, + "learning_rate": 3.30625507717303e-06, + "loss": 2.2046, + "step": 1123 + }, + { + "epoch": 1.2317808219178081, + "grad_norm": 1.9570913314819336, + "learning_rate": 3.3042242079610075e-06, + "loss": 2.1015, + "step": 1124 + }, + { + "epoch": 1.2328767123287672, + "grad_norm": 2.00569224357605, + "learning_rate": 3.3021933387489847e-06, + "loss": 2.1681, + "step": 1125 + }, + { + "epoch": 1.233972602739726, + "grad_norm": 1.9131590127944946, + "learning_rate": 3.300162469536962e-06, + "loss": 2.1765, + "step": 1126 + }, + { + "epoch": 1.235068493150685, + "grad_norm": 1.8375864028930664, + "learning_rate": 3.2981316003249393e-06, + "loss": 2.1175, + "step": 1127 + }, + { + "epoch": 1.2361643835616438, + "grad_norm": 1.852847933769226, + "learning_rate": 3.2961007311129166e-06, + "loss": 2.1838, + "step": 1128 + }, + { + "epoch": 1.2372602739726029, + "grad_norm": 1.8772693872451782, + "learning_rate": 3.294069861900894e-06, + "loss": 2.1524, + "step": 1129 + }, + { + "epoch": 1.2383561643835617, + "grad_norm": 1.8593205213546753, + "learning_rate": 3.292038992688871e-06, + "loss": 2.2059, + "step": 1130 + }, + { + "epoch": 1.2383561643835617, + "eval_loss": 2.294188976287842, + "eval_runtime": 50.4549, + "eval_samples_per_second": 19.82, + "eval_steps_per_second": 0.832, + "step": 1130 + }, + { + "epoch": 1.2394520547945205, + "grad_norm": 1.8967972993850708, + "learning_rate": 3.2900081234768484e-06, + "loss": 2.1551, + "step": 1131 + }, + { + "epoch": 1.2405479452054795, + "grad_norm": 1.9799107313156128, + "learning_rate": 3.2879772542648257e-06, + "loss": 2.1644, + "step": 1132 + }, + { + "epoch": 1.2416438356164383, + "grad_norm": 1.978883147239685, + "learning_rate": 3.285946385052803e-06, + "loss": 2.182, + "step": 1133 + }, + { + "epoch": 1.2427397260273974, + "grad_norm": 1.8705304861068726, + "learning_rate": 3.2839155158407802e-06, + "loss": 2.0981, + "step": 1134 + }, + { + "epoch": 1.2438356164383562, + "grad_norm": 2.0146889686584473, + "learning_rate": 3.2818846466287575e-06, + "loss": 2.1686, + "step": 1135 + }, + { + "epoch": 1.244931506849315, + "grad_norm": 1.94461989402771, + "learning_rate": 3.2798537774167344e-06, + "loss": 2.093, + "step": 1136 + }, + { + "epoch": 1.246027397260274, + "grad_norm": 1.7823240756988525, + "learning_rate": 3.277822908204712e-06, + "loss": 2.198, + "step": 1137 + }, + { + "epoch": 1.2471232876712328, + "grad_norm": 1.89297354221344, + "learning_rate": 3.275792038992689e-06, + "loss": 2.1359, + "step": 1138 + }, + { + "epoch": 1.2482191780821918, + "grad_norm": 1.8670459985733032, + "learning_rate": 3.2737611697806666e-06, + "loss": 2.0956, + "step": 1139 + }, + { + "epoch": 1.2493150684931507, + "grad_norm": 1.813568353652954, + "learning_rate": 3.2717303005686435e-06, + "loss": 2.1448, + "step": 1140 + }, + { + "epoch": 1.2493150684931507, + "eval_loss": 2.295804023742676, + "eval_runtime": 53.286, + "eval_samples_per_second": 18.767, + "eval_steps_per_second": 0.788, + "step": 1140 + }, + { + "epoch": 1.2504109589041095, + "grad_norm": 1.912916898727417, + "learning_rate": 3.269699431356621e-06, + "loss": 2.1159, + "step": 1141 + }, + { + "epoch": 1.2515068493150685, + "grad_norm": 1.7939883470535278, + "learning_rate": 3.267668562144598e-06, + "loss": 2.1566, + "step": 1142 + }, + { + "epoch": 1.2526027397260273, + "grad_norm": 1.8494806289672852, + "learning_rate": 3.2656376929325757e-06, + "loss": 2.1496, + "step": 1143 + }, + { + "epoch": 1.2536986301369863, + "grad_norm": 1.8468523025512695, + "learning_rate": 3.2636068237205526e-06, + "loss": 2.1159, + "step": 1144 + }, + { + "epoch": 1.2547945205479452, + "grad_norm": 1.8962055444717407, + "learning_rate": 3.2615759545085303e-06, + "loss": 2.1712, + "step": 1145 + }, + { + "epoch": 1.2558904109589042, + "grad_norm": 1.800353765487671, + "learning_rate": 3.259545085296507e-06, + "loss": 2.1486, + "step": 1146 + }, + { + "epoch": 1.256986301369863, + "grad_norm": 2.124359607696533, + "learning_rate": 3.257514216084484e-06, + "loss": 2.1913, + "step": 1147 + }, + { + "epoch": 1.2580821917808218, + "grad_norm": 1.9248145818710327, + "learning_rate": 3.2554833468724617e-06, + "loss": 2.1128, + "step": 1148 + }, + { + "epoch": 1.2591780821917808, + "grad_norm": 1.9990876913070679, + "learning_rate": 3.2534524776604385e-06, + "loss": 2.1854, + "step": 1149 + }, + { + "epoch": 1.2602739726027397, + "grad_norm": 1.8600568771362305, + "learning_rate": 3.2514216084484162e-06, + "loss": 2.1565, + "step": 1150 + }, + { + "epoch": 1.2602739726027397, + "eval_loss": 2.2942006587982178, + "eval_runtime": 51.4021, + "eval_samples_per_second": 19.454, + "eval_steps_per_second": 0.817, + "step": 1150 + }, + { + "epoch": 1.2613698630136987, + "grad_norm": 1.8621101379394531, + "learning_rate": 3.249390739236393e-06, + "loss": 2.1438, + "step": 1151 + }, + { + "epoch": 1.2624657534246575, + "grad_norm": 1.9290356636047363, + "learning_rate": 3.2473598700243708e-06, + "loss": 2.1577, + "step": 1152 + }, + { + "epoch": 1.2635616438356165, + "grad_norm": 1.9007307291030884, + "learning_rate": 3.2453290008123476e-06, + "loss": 2.1295, + "step": 1153 + }, + { + "epoch": 1.2646575342465753, + "grad_norm": 1.7647960186004639, + "learning_rate": 3.2432981316003253e-06, + "loss": 2.1472, + "step": 1154 + }, + { + "epoch": 1.2657534246575342, + "grad_norm": 1.9796825647354126, + "learning_rate": 3.241267262388302e-06, + "loss": 2.1542, + "step": 1155 + }, + { + "epoch": 1.2668493150684932, + "grad_norm": 1.9505796432495117, + "learning_rate": 3.23923639317628e-06, + "loss": 2.1093, + "step": 1156 + }, + { + "epoch": 1.267945205479452, + "grad_norm": 1.9152203798294067, + "learning_rate": 3.2372055239642567e-06, + "loss": 2.1456, + "step": 1157 + }, + { + "epoch": 1.269041095890411, + "grad_norm": 1.9094963073730469, + "learning_rate": 3.2351746547522344e-06, + "loss": 2.1149, + "step": 1158 + }, + { + "epoch": 1.2701369863013698, + "grad_norm": 1.8872679471969604, + "learning_rate": 3.2331437855402113e-06, + "loss": 2.1545, + "step": 1159 + }, + { + "epoch": 1.2712328767123289, + "grad_norm": 2.171703338623047, + "learning_rate": 3.2311129163281885e-06, + "loss": 2.2108, + "step": 1160 + }, + { + "epoch": 1.2712328767123289, + "eval_loss": 2.2910356521606445, + "eval_runtime": 52.5176, + "eval_samples_per_second": 19.041, + "eval_steps_per_second": 0.8, + "step": 1160 + }, + { + "epoch": 1.2723287671232877, + "grad_norm": 1.8578304052352905, + "learning_rate": 3.229082047116166e-06, + "loss": 2.1158, + "step": 1161 + }, + { + "epoch": 1.2734246575342465, + "grad_norm": 1.8873049020767212, + "learning_rate": 3.227051177904143e-06, + "loss": 2.1618, + "step": 1162 + }, + { + "epoch": 1.2745205479452055, + "grad_norm": 1.8442409038543701, + "learning_rate": 3.2250203086921204e-06, + "loss": 2.1454, + "step": 1163 + }, + { + "epoch": 1.2756164383561643, + "grad_norm": 1.9609211683273315, + "learning_rate": 3.2229894394800977e-06, + "loss": 2.1519, + "step": 1164 + }, + { + "epoch": 1.2767123287671232, + "grad_norm": 1.9093680381774902, + "learning_rate": 3.220958570268075e-06, + "loss": 2.1775, + "step": 1165 + }, + { + "epoch": 1.2778082191780822, + "grad_norm": 1.8355298042297363, + "learning_rate": 3.218927701056052e-06, + "loss": 2.0871, + "step": 1166 + }, + { + "epoch": 1.2789041095890412, + "grad_norm": 1.9497056007385254, + "learning_rate": 3.2168968318440295e-06, + "loss": 2.1441, + "step": 1167 + }, + { + "epoch": 1.28, + "grad_norm": 1.8635729551315308, + "learning_rate": 3.2148659626320068e-06, + "loss": 2.0748, + "step": 1168 + }, + { + "epoch": 1.2810958904109588, + "grad_norm": 2.0972132682800293, + "learning_rate": 3.212835093419984e-06, + "loss": 2.1843, + "step": 1169 + }, + { + "epoch": 1.2821917808219179, + "grad_norm": 1.9235926866531372, + "learning_rate": 3.2108042242079613e-06, + "loss": 2.1194, + "step": 1170 + }, + { + "epoch": 1.2821917808219179, + "eval_loss": 2.291309118270874, + "eval_runtime": 55.6408, + "eval_samples_per_second": 17.972, + "eval_steps_per_second": 0.755, + "step": 1170 + }, + { + "epoch": 1.2832876712328767, + "grad_norm": 1.8674591779708862, + "learning_rate": 3.2087733549959386e-06, + "loss": 2.1264, + "step": 1171 + }, + { + "epoch": 1.2843835616438355, + "grad_norm": 1.810589075088501, + "learning_rate": 3.206742485783916e-06, + "loss": 2.1781, + "step": 1172 + }, + { + "epoch": 1.2854794520547945, + "grad_norm": 1.9794505834579468, + "learning_rate": 3.2047116165718927e-06, + "loss": 2.1089, + "step": 1173 + }, + { + "epoch": 1.2865753424657536, + "grad_norm": 1.8487070798873901, + "learning_rate": 3.2026807473598704e-06, + "loss": 2.097, + "step": 1174 + }, + { + "epoch": 1.2876712328767124, + "grad_norm": 1.9668854475021362, + "learning_rate": 3.2006498781478473e-06, + "loss": 2.1516, + "step": 1175 + }, + { + "epoch": 1.2887671232876712, + "grad_norm": 1.855817198753357, + "learning_rate": 3.198619008935825e-06, + "loss": 2.1678, + "step": 1176 + }, + { + "epoch": 1.2898630136986302, + "grad_norm": 2.381211757659912, + "learning_rate": 3.196588139723802e-06, + "loss": 2.099, + "step": 1177 + }, + { + "epoch": 1.290958904109589, + "grad_norm": 2.1524250507354736, + "learning_rate": 3.1945572705117795e-06, + "loss": 2.1345, + "step": 1178 + }, + { + "epoch": 1.2920547945205478, + "grad_norm": 1.9120781421661377, + "learning_rate": 3.1925264012997564e-06, + "loss": 2.1586, + "step": 1179 + }, + { + "epoch": 1.2931506849315069, + "grad_norm": 2.153817653656006, + "learning_rate": 3.190495532087734e-06, + "loss": 2.1597, + "step": 1180 + }, + { + "epoch": 1.2931506849315069, + "eval_loss": 2.2936320304870605, + "eval_runtime": 53.6426, + "eval_samples_per_second": 18.642, + "eval_steps_per_second": 0.783, + "step": 1180 + }, + { + "epoch": 1.2942465753424657, + "grad_norm": 2.074873447418213, + "learning_rate": 3.188464662875711e-06, + "loss": 2.1572, + "step": 1181 + }, + { + "epoch": 1.2953424657534247, + "grad_norm": 1.9313570261001587, + "learning_rate": 3.1864337936636886e-06, + "loss": 2.1986, + "step": 1182 + }, + { + "epoch": 1.2964383561643835, + "grad_norm": 1.9836641550064087, + "learning_rate": 3.1844029244516655e-06, + "loss": 2.13, + "step": 1183 + }, + { + "epoch": 1.2975342465753426, + "grad_norm": 2.0893137454986572, + "learning_rate": 3.182372055239643e-06, + "loss": 2.1138, + "step": 1184 + }, + { + "epoch": 1.2986301369863014, + "grad_norm": 2.1667087078094482, + "learning_rate": 3.18034118602762e-06, + "loss": 2.1119, + "step": 1185 + }, + { + "epoch": 1.2997260273972602, + "grad_norm": 2.093961000442505, + "learning_rate": 3.1783103168155973e-06, + "loss": 2.1728, + "step": 1186 + }, + { + "epoch": 1.3008219178082192, + "grad_norm": 1.919267177581787, + "learning_rate": 3.1762794476035746e-06, + "loss": 2.1469, + "step": 1187 + }, + { + "epoch": 1.301917808219178, + "grad_norm": 2.017967939376831, + "learning_rate": 3.174248578391552e-06, + "loss": 2.1307, + "step": 1188 + }, + { + "epoch": 1.303013698630137, + "grad_norm": 1.921632170677185, + "learning_rate": 3.172217709179529e-06, + "loss": 2.13, + "step": 1189 + }, + { + "epoch": 1.3041095890410959, + "grad_norm": 2.0171291828155518, + "learning_rate": 3.1701868399675064e-06, + "loss": 2.135, + "step": 1190 + }, + { + "epoch": 1.3041095890410959, + "eval_loss": 2.2934045791625977, + "eval_runtime": 48.0821, + "eval_samples_per_second": 20.798, + "eval_steps_per_second": 0.874, + "step": 1190 + }, + { + "epoch": 1.305205479452055, + "grad_norm": 1.8685601949691772, + "learning_rate": 3.1681559707554837e-06, + "loss": 2.1329, + "step": 1191 + }, + { + "epoch": 1.3063013698630137, + "grad_norm": 1.8909096717834473, + "learning_rate": 3.166125101543461e-06, + "loss": 2.1004, + "step": 1192 + }, + { + "epoch": 1.3073972602739725, + "grad_norm": 1.798614740371704, + "learning_rate": 3.1640942323314382e-06, + "loss": 2.1555, + "step": 1193 + }, + { + "epoch": 1.3084931506849315, + "grad_norm": 1.8296692371368408, + "learning_rate": 3.1620633631194155e-06, + "loss": 2.1549, + "step": 1194 + }, + { + "epoch": 1.3095890410958904, + "grad_norm": 1.8676908016204834, + "learning_rate": 3.1600324939073928e-06, + "loss": 2.1081, + "step": 1195 + }, + { + "epoch": 1.3106849315068494, + "grad_norm": 1.8682451248168945, + "learning_rate": 3.15800162469537e-06, + "loss": 2.138, + "step": 1196 + }, + { + "epoch": 1.3117808219178082, + "grad_norm": 2.0049314498901367, + "learning_rate": 3.155970755483347e-06, + "loss": 2.144, + "step": 1197 + }, + { + "epoch": 1.3128767123287672, + "grad_norm": 2.00691556930542, + "learning_rate": 3.1539398862713246e-06, + "loss": 2.1539, + "step": 1198 + }, + { + "epoch": 1.313972602739726, + "grad_norm": 1.8521753549575806, + "learning_rate": 3.1519090170593015e-06, + "loss": 2.1597, + "step": 1199 + }, + { + "epoch": 1.3150684931506849, + "grad_norm": 2.0974042415618896, + "learning_rate": 3.149878147847279e-06, + "loss": 2.147, + "step": 1200 + }, + { + "epoch": 1.3150684931506849, + "eval_loss": 2.2891385555267334, + "eval_runtime": 48.1618, + "eval_samples_per_second": 20.763, + "eval_steps_per_second": 0.872, + "step": 1200 + } + ], + "logging_steps": 1, + "max_steps": 2736, + "num_input_tokens_seen": 0, + "num_train_epochs": 3, + "save_steps": 100, + "stateful_callbacks": { + "TrainerControl": { + "args": { + "should_epoch_stop": false, + "should_evaluate": false, + "should_log": false, + "should_save": true, + "should_training_stop": false + }, + "attributes": {} + } + }, + "total_flos": 3.740771291234304e+17, + "train_batch_size": 3, + "trial_name": null, + "trial_params": null +} diff --git a/babylm_shuffle_nondeterministic_10M_seed0/runs/checkpoint-1200/zero_to_fp32.py b/babylm_shuffle_nondeterministic_10M_seed0/runs/checkpoint-1200/zero_to_fp32.py new file mode 100644 index 0000000000000000000000000000000000000000..24cc342e78d1a006c782b3a4cd68d9ce786d8fd8 --- /dev/null +++ b/babylm_shuffle_nondeterministic_10M_seed0/runs/checkpoint-1200/zero_to_fp32.py @@ -0,0 +1,604 @@ +#!/usr/bin/env python + +# Copyright (c) Microsoft Corporation. +# SPDX-License-Identifier: Apache-2.0 + +# DeepSpeed Team + +# This script extracts fp32 consolidated weights from a zero 1, 2 and 3 DeepSpeed checkpoints. It gets +# copied into the top level checkpoint dir, so the user can easily do the conversion at any point in +# the future. Once extracted, the weights don't require DeepSpeed and can be used in any +# application. +# +# example: python zero_to_fp32.py . pytorch_model.bin + +import argparse +import torch +import glob +import math +import os +import re +from collections import OrderedDict +from dataclasses import dataclass + +# while this script doesn't use deepspeed to recover data, since the checkpoints are pickled with +# DeepSpeed data structures it has to be available in the current python environment. +from deepspeed.utils import logger +from deepspeed.checkpoint.constants import (DS_VERSION, OPTIMIZER_STATE_DICT, SINGLE_PARTITION_OF_FP32_GROUPS, + FP32_FLAT_GROUPS, ZERO_STAGE, PARTITION_COUNT, PARAM_SHAPES, BUFFER_NAMES, + FROZEN_PARAM_SHAPES, FROZEN_PARAM_FRAGMENTS) + + +@dataclass +class zero_model_state: + buffers: dict() + param_shapes: dict() + shared_params: list + ds_version: int + frozen_param_shapes: dict() + frozen_param_fragments: dict() + + +debug = 0 + +# load to cpu +device = torch.device('cpu') + + +def atoi(text): + return int(text) if text.isdigit() else text + + +def natural_keys(text): + ''' + alist.sort(key=natural_keys) sorts in human order + http://nedbatchelder.com/blog/200712/human_sorting.html + (See Toothy's implementation in the comments) + ''' + return [atoi(c) for c in re.split(r'(\d+)', text)] + + +def get_model_state_file(checkpoint_dir, zero_stage): + if not os.path.isdir(checkpoint_dir): + raise FileNotFoundError(f"Directory '{checkpoint_dir}' doesn't exist") + + # there should be only one file + if zero_stage <= 2: + file = os.path.join(checkpoint_dir, "mp_rank_00_model_states.pt") + elif zero_stage == 3: + file = os.path.join(checkpoint_dir, "zero_pp_rank_0_mp_rank_00_model_states.pt") + + if not os.path.exists(file): + raise FileNotFoundError(f"can't find model states file at '{file}'") + + return file + + +def get_checkpoint_files(checkpoint_dir, glob_pattern): + # XXX: need to test that this simple glob rule works for multi-node setup too + ckpt_files = sorted(glob.glob(os.path.join(checkpoint_dir, glob_pattern)), key=natural_keys) + + if len(ckpt_files) == 0: + raise FileNotFoundError(f"can't find {glob_pattern} files in directory '{checkpoint_dir}'") + + return ckpt_files + + +def get_optim_files(checkpoint_dir): + return get_checkpoint_files(checkpoint_dir, "*_optim_states.pt") + + +def get_model_state_files(checkpoint_dir): + return get_checkpoint_files(checkpoint_dir, "*_model_states.pt") + + +def parse_model_states(files): + zero_model_states = [] + for file in files: + state_dict = torch.load(file, map_location=device) + + if BUFFER_NAMES not in state_dict: + raise ValueError(f"{file} is not a model state checkpoint") + buffer_names = state_dict[BUFFER_NAMES] + if debug: + print("Found buffers:", buffer_names) + + # recover just the buffers while restoring them to fp32 if they were saved in fp16 + buffers = {k: v.float() for k, v in state_dict["module"].items() if k in buffer_names} + param_shapes = state_dict[PARAM_SHAPES] + + # collect parameters that are included in param_shapes + param_names = [] + for s in param_shapes: + for name in s.keys(): + param_names.append(name) + + # update with frozen parameters + frozen_param_shapes = state_dict.get(FROZEN_PARAM_SHAPES, None) + if frozen_param_shapes is not None: + if debug: + print(f"Found frozen_param_shapes: {frozen_param_shapes}") + param_names += list(frozen_param_shapes.keys()) + + # handle shared params + shared_params = [[k, v] for k, v in state_dict["shared_params"].items()] + + ds_version = state_dict.get(DS_VERSION, None) + + frozen_param_fragments = state_dict.get(FROZEN_PARAM_FRAGMENTS, None) + + z_model_state = zero_model_state(buffers=buffers, + param_shapes=param_shapes, + shared_params=shared_params, + ds_version=ds_version, + frozen_param_shapes=frozen_param_shapes, + frozen_param_fragments=frozen_param_fragments) + zero_model_states.append(z_model_state) + + return zero_model_states + + +def parse_optim_states(files, ds_checkpoint_dir): + + total_files = len(files) + state_dicts = [] + for f in files: + state_dict = torch.load(f, map_location=device) + # immediately discard the potentially huge 2 optimizer states as we only care for fp32 master weights + # and also handle the case where it was already removed by another helper script + state_dict["optimizer_state_dict"].pop("optimizer_state_dict", None) + state_dicts.append(state_dict) + + if not ZERO_STAGE in state_dicts[0][OPTIMIZER_STATE_DICT]: + raise ValueError(f"{files[0]} is not a zero checkpoint") + zero_stage = state_dicts[0][OPTIMIZER_STATE_DICT][ZERO_STAGE] + world_size = state_dicts[0][OPTIMIZER_STATE_DICT][PARTITION_COUNT] + + # For ZeRO-2 each param group can have different partition_count as data parallelism for expert + # parameters can be different from data parallelism for non-expert parameters. So we can just + # use the max of the partition_count to get the dp world_size. + + if type(world_size) is list: + world_size = max(world_size) + + if world_size != total_files: + raise ValueError( + f"Expected {world_size} of '*_optim_states.pt' under '{ds_checkpoint_dir}' but found {total_files} files. " + "Possibly due to an overwrite of an old checkpoint, or a checkpoint didn't get saved by one or more processes." + ) + + # the groups are named differently in each stage + if zero_stage <= 2: + fp32_groups_key = SINGLE_PARTITION_OF_FP32_GROUPS + elif zero_stage == 3: + fp32_groups_key = FP32_FLAT_GROUPS + else: + raise ValueError(f"unknown zero stage {zero_stage}") + + if zero_stage <= 2: + fp32_flat_groups = [state_dicts[i][OPTIMIZER_STATE_DICT][fp32_groups_key] for i in range(len(state_dicts))] + elif zero_stage == 3: + # if there is more than one param group, there will be multiple flattened tensors - one + # flattened tensor per group - for simplicity merge them into a single tensor + # + # XXX: could make the script more memory efficient for when there are multiple groups - it + # will require matching the sub-lists of param_shapes for each param group flattened tensor + + fp32_flat_groups = [ + torch.cat(state_dicts[i][OPTIMIZER_STATE_DICT][fp32_groups_key], 0) for i in range(len(state_dicts)) + ] + + return zero_stage, world_size, fp32_flat_groups + + +def _get_fp32_state_dict_from_zero_checkpoint(ds_checkpoint_dir, exclude_frozen_parameters): + """ + Returns fp32 state_dict reconstructed from ds checkpoint + + Args: + - ``ds_checkpoint_dir``: path to the deepspeed checkpoint folder (where the optimizer files are) + + """ + print(f"Processing zero checkpoint '{ds_checkpoint_dir}'") + + optim_files = get_optim_files(ds_checkpoint_dir) + zero_stage, world_size, fp32_flat_groups = parse_optim_states(optim_files, ds_checkpoint_dir) + print(f"Detected checkpoint of type zero stage {zero_stage}, world_size: {world_size}") + + model_files = get_model_state_files(ds_checkpoint_dir) + + zero_model_states = parse_model_states(model_files) + print(f'Parsing checkpoint created by deepspeed=={zero_model_states[0].ds_version}') + + if zero_stage <= 2: + return _get_fp32_state_dict_from_zero2_checkpoint(world_size, fp32_flat_groups, zero_model_states, + exclude_frozen_parameters) + elif zero_stage == 3: + return _get_fp32_state_dict_from_zero3_checkpoint(world_size, fp32_flat_groups, zero_model_states, + exclude_frozen_parameters) + + +def _zero2_merge_frozen_params(state_dict, zero_model_states): + if zero_model_states[0].frozen_param_shapes is None or len(zero_model_states[0].frozen_param_shapes) == 0: + return + + frozen_param_shapes = zero_model_states[0].frozen_param_shapes + frozen_param_fragments = zero_model_states[0].frozen_param_fragments + + if debug: + num_elem = sum(s.numel() for s in frozen_param_shapes.values()) + print(f'rank 0: {FROZEN_PARAM_SHAPES}.numel = {num_elem}') + + wanted_params = len(frozen_param_shapes) + wanted_numel = sum(s.numel() for s in frozen_param_shapes.values()) + avail_numel = sum([p.numel() for p in frozen_param_fragments.values()]) + print(f'Frozen params: Have {avail_numel} numels to process.') + print(f'Frozen params: Need {wanted_numel} numels in {wanted_params} params') + + total_params = 0 + total_numel = 0 + for name, shape in frozen_param_shapes.items(): + total_params += 1 + unpartitioned_numel = shape.numel() + total_numel += unpartitioned_numel + + state_dict[name] = frozen_param_fragments[name] + + if debug: + print(f"{name} full shape: {shape} unpartitioned numel {unpartitioned_numel} ") + + print(f"Reconstructed Frozen fp32 state dict with {total_params} params {total_numel} elements") + + +def _has_callable(obj, fn): + attr = getattr(obj, fn, None) + return callable(attr) + + +def _zero2_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states): + param_shapes = zero_model_states[0].param_shapes + + # Reconstruction protocol: + # + # XXX: document this + + if debug: + for i in range(world_size): + for j in range(len(fp32_flat_groups[0])): + print(f"{FP32_FLAT_GROUPS}[{i}][{j}].shape={fp32_flat_groups[i][j].shape}") + + # XXX: memory usage doubles here (zero2) + num_param_groups = len(fp32_flat_groups[0]) + merged_single_partition_of_fp32_groups = [] + for i in range(num_param_groups): + merged_partitions = [sd[i] for sd in fp32_flat_groups] + full_single_fp32_vector = torch.cat(merged_partitions, 0) + merged_single_partition_of_fp32_groups.append(full_single_fp32_vector) + avail_numel = sum( + [full_single_fp32_vector.numel() for full_single_fp32_vector in merged_single_partition_of_fp32_groups]) + + if debug: + wanted_params = sum([len(shapes) for shapes in param_shapes]) + wanted_numel = sum([sum(shape.numel() for shape in shapes.values()) for shapes in param_shapes]) + # not asserting if there is a mismatch due to possible padding + print(f"Have {avail_numel} numels to process.") + print(f"Need {wanted_numel} numels in {wanted_params} params.") + + # params + # XXX: for huge models that can't fit into the host's RAM we will have to recode this to support + # out-of-core computing solution + total_numel = 0 + total_params = 0 + for shapes, full_single_fp32_vector in zip(param_shapes, merged_single_partition_of_fp32_groups): + offset = 0 + avail_numel = full_single_fp32_vector.numel() + for name, shape in shapes.items(): + + unpartitioned_numel = shape.numel() if _has_callable(shape, 'numel') else math.prod(shape) + total_numel += unpartitioned_numel + total_params += 1 + + if debug: + print(f"{name} full shape: {shape} unpartitioned numel {unpartitioned_numel} ") + state_dict[name] = full_single_fp32_vector.narrow(0, offset, unpartitioned_numel).view(shape) + offset += unpartitioned_numel + + # Z2 started to align to 2*world_size to improve nccl performance. Therefore both offset and + # avail_numel can differ by anywhere between 0..2*world_size. Due to two unrelated complex + # paddings performed in the code it's almost impossible to predict the exact numbers w/o the + # live optimizer object, so we are checking that the numbers are within the right range + align_to = 2 * world_size + + def zero2_align(x): + return align_to * math.ceil(x / align_to) + + if debug: + print(f"original offset={offset}, avail_numel={avail_numel}") + + offset = zero2_align(offset) + avail_numel = zero2_align(avail_numel) + + if debug: + print(f"aligned offset={offset}, avail_numel={avail_numel}") + + # Sanity check + if offset != avail_numel: + raise ValueError(f"consumed {offset} numels out of {avail_numel} - something is wrong") + + print(f"Reconstructed fp32 state dict with {total_params} params {total_numel} elements") + + +def _get_fp32_state_dict_from_zero2_checkpoint(world_size, fp32_flat_groups, zero_model_states, + exclude_frozen_parameters): + state_dict = OrderedDict() + + # buffers + buffers = zero_model_states[0].buffers + state_dict.update(buffers) + if debug: + print(f"added {len(buffers)} buffers") + + if not exclude_frozen_parameters: + _zero2_merge_frozen_params(state_dict, zero_model_states) + + _zero2_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states) + + # recover shared parameters + for pair in zero_model_states[0].shared_params: + if pair[1] in state_dict: + state_dict[pair[0]] = state_dict[pair[1]] + + return state_dict + + +def zero3_partitioned_param_info(unpartitioned_numel, world_size): + remainder = unpartitioned_numel % world_size + padding_numel = (world_size - remainder) if remainder else 0 + partitioned_numel = math.ceil(unpartitioned_numel / world_size) + return partitioned_numel, padding_numel + + +def _zero3_merge_frozen_params(state_dict, world_size, zero_model_states): + if zero_model_states[0].frozen_param_shapes is None or len(zero_model_states[0].frozen_param_shapes) == 0: + return + + if debug: + for i in range(world_size): + num_elem = sum(s.numel() for s in zero_model_states[i].frozen_param_fragments.values()) + print(f'rank {i}: {FROZEN_PARAM_SHAPES}.numel = {num_elem}') + + frozen_param_shapes = zero_model_states[0].frozen_param_shapes + wanted_params = len(frozen_param_shapes) + wanted_numel = sum(s.numel() for s in frozen_param_shapes.values()) + avail_numel = sum([p.numel() for p in zero_model_states[0].frozen_param_fragments.values()]) * world_size + print(f'Frozen params: Have {avail_numel} numels to process.') + print(f'Frozen params: Need {wanted_numel} numels in {wanted_params} params') + + total_params = 0 + total_numel = 0 + for name, shape in zero_model_states[0].frozen_param_shapes.items(): + total_params += 1 + unpartitioned_numel = shape.numel() + total_numel += unpartitioned_numel + + param_frags = tuple(model_state.frozen_param_fragments[name] for model_state in zero_model_states) + state_dict[name] = torch.cat(param_frags, 0).narrow(0, 0, unpartitioned_numel).view(shape) + + partitioned_numel, partitioned_padding_numel = zero3_partitioned_param_info(unpartitioned_numel, world_size) + + if debug: + print( + f"Frozen params: {total_params} {name} full shape: {shape} partition0 numel={partitioned_numel} partitioned_padding_numel={partitioned_padding_numel}" + ) + + print(f"Reconstructed Frozen fp32 state dict with {total_params} params {total_numel} elements") + + +def _zero3_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states): + param_shapes = zero_model_states[0].param_shapes + avail_numel = fp32_flat_groups[0].numel() * world_size + # Reconstruction protocol: For zero3 we need to zip the partitions together at boundary of each + # param, re-consolidating each param, while dealing with padding if any + + # merge list of dicts, preserving order + param_shapes = {k: v for d in param_shapes for k, v in d.items()} + + if debug: + for i in range(world_size): + print(f"{FP32_FLAT_GROUPS}[{i}].shape={fp32_flat_groups[i].shape}") + + wanted_params = len(param_shapes) + wanted_numel = sum(shape.numel() for shape in param_shapes.values()) + # not asserting if there is a mismatch due to possible padding + avail_numel = fp32_flat_groups[0].numel() * world_size + print(f"Trainable params: Have {avail_numel} numels to process.") + print(f"Trainable params: Need {wanted_numel} numels in {wanted_params} params.") + + # params + # XXX: for huge models that can't fit into the host's RAM we will have to recode this to support + # out-of-core computing solution + offset = 0 + total_numel = 0 + total_params = 0 + for name, shape in param_shapes.items(): + + unpartitioned_numel = shape.numel() + total_numel += unpartitioned_numel + total_params += 1 + + partitioned_numel, partitioned_padding_numel = zero3_partitioned_param_info(unpartitioned_numel, world_size) + + if debug: + print( + f"Trainable params: {total_params} {name} full shape: {shape} partition0 numel={partitioned_numel} partitioned_padding_numel={partitioned_padding_numel}" + ) + + # XXX: memory usage doubles here + state_dict[name] = torch.cat( + tuple(fp32_flat_groups[i].narrow(0, offset, partitioned_numel) for i in range(world_size)), + 0).narrow(0, 0, unpartitioned_numel).view(shape) + offset += partitioned_numel + + offset *= world_size + + # Sanity check + if offset != avail_numel: + raise ValueError(f"consumed {offset} numels out of {avail_numel} - something is wrong") + + print(f"Reconstructed Trainable fp32 state dict with {total_params} params {total_numel} elements") + + +def _get_fp32_state_dict_from_zero3_checkpoint(world_size, fp32_flat_groups, zero_model_states, + exclude_frozen_parameters): + state_dict = OrderedDict() + + # buffers + buffers = zero_model_states[0].buffers + state_dict.update(buffers) + if debug: + print(f"added {len(buffers)} buffers") + + if not exclude_frozen_parameters: + _zero3_merge_frozen_params(state_dict, world_size, zero_model_states) + + _zero3_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states) + + # recover shared parameters + for pair in zero_model_states[0].shared_params: + if pair[1] in state_dict: + state_dict[pair[0]] = state_dict[pair[1]] + + return state_dict + + +def get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir, tag=None, exclude_frozen_parameters=False): + """ + Convert ZeRO 2 or 3 checkpoint into a single fp32 consolidated state_dict that can be loaded with + ``load_state_dict()`` and used for training without DeepSpeed or shared with others, for example + via a model hub. + + Args: + - ``checkpoint_dir``: path to the desired checkpoint folder + - ``tag``: checkpoint tag used as a unique identifier for checkpoint. If not provided will attempt to load tag in 'latest' file. e.g., ``global_step14`` + - ``exclude_frozen_parameters``: exclude frozen parameters + + Returns: + - pytorch ``state_dict`` + + Note: this approach may not work if your application doesn't have sufficient free CPU memory and + you may need to use the offline approach using the ``zero_to_fp32.py`` script that is saved with + the checkpoint. + + A typical usage might be :: + + from deepspeed.utils.zero_to_fp32 import get_fp32_state_dict_from_zero_checkpoint + # do the training and checkpoint saving + state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir) # already on cpu + model = model.cpu() # move to cpu + model.load_state_dict(state_dict) + # submit to model hub or save the model to share with others + + In this example the ``model`` will no longer be usable in the deepspeed context of the same + application. i.e. you will need to re-initialize the deepspeed engine, since + ``model.load_state_dict(state_dict)`` will remove all the deepspeed magic from it. + + If you want it all done for you, use ``load_state_dict_from_zero_checkpoint`` instead. + + """ + if tag is None: + latest_path = os.path.join(checkpoint_dir, 'latest') + if os.path.isfile(latest_path): + with open(latest_path, 'r') as fd: + tag = fd.read().strip() + else: + raise ValueError(f"Unable to find 'latest' file at {latest_path}") + + ds_checkpoint_dir = os.path.join(checkpoint_dir, tag) + + if not os.path.isdir(ds_checkpoint_dir): + raise FileNotFoundError(f"Directory '{ds_checkpoint_dir}' doesn't exist") + + return _get_fp32_state_dict_from_zero_checkpoint(ds_checkpoint_dir, exclude_frozen_parameters) + + +def convert_zero_checkpoint_to_fp32_state_dict(checkpoint_dir, output_file, tag=None, exclude_frozen_parameters=False): + """ + Convert ZeRO 2 or 3 checkpoint into a single fp32 consolidated ``state_dict`` file that can be + loaded with ``torch.load(file)`` + ``load_state_dict()`` and used for training without DeepSpeed. + + Args: + - ``checkpoint_dir``: path to the desired checkpoint folder. (one that contains the tag-folder, like ``global_step14``) + - ``output_file``: path to the pytorch fp32 state_dict output file (e.g. path/pytorch_model.bin) + - ``tag``: checkpoint tag used as a unique identifier for checkpoint. If not provided will attempt to load tag in the file named ``latest`` in the checkpoint folder, e.g., ``global_step14`` + - ``exclude_frozen_parameters``: exclude frozen parameters + """ + + state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir, tag, exclude_frozen_parameters) + print(f"Saving fp32 state dict to {output_file}") + torch.save(state_dict, output_file) + + +def load_state_dict_from_zero_checkpoint(model, checkpoint_dir, tag=None): + """ + 1. Put the provided model to cpu + 2. Convert ZeRO 2 or 3 checkpoint into a single fp32 consolidated ``state_dict`` + 3. Load it into the provided model + + Args: + - ``model``: the model object to update + - ``checkpoint_dir``: path to the desired checkpoint folder. (one that contains the tag-folder, like ``global_step14``) + - ``tag``: checkpoint tag used as a unique identifier for checkpoint. If not provided will attempt to load tag in the file named ``latest`` in the checkpoint folder, e.g., ``global_step14`` + + Returns: + - ``model`: modified model + + Make sure you have plenty of CPU memory available before you call this function. If you don't + have enough use the ``zero_to_fp32.py`` utility to do the conversion. You will find it + conveniently placed for you in the checkpoint folder. + + A typical usage might be :: + + from deepspeed.utils.zero_to_fp32 import load_state_dict_from_zero_checkpoint + model = load_state_dict_from_zero_checkpoint(trainer.model, checkpoint_dir) + # submit to model hub or save the model to share with others + + Note, that once this was run, the ``model`` will no longer be usable in the deepspeed context + of the same application. i.e. you will need to re-initialize the deepspeed engine, since + ``model.load_state_dict(state_dict)`` will remove all the deepspeed magic from it. + + """ + logger.info(f"Extracting fp32 weights") + state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir, tag) + + logger.info(f"Overwriting model with fp32 weights") + model = model.cpu() + model.load_state_dict(state_dict, strict=False) + + return model + + +if __name__ == "__main__": + + parser = argparse.ArgumentParser() + parser.add_argument("checkpoint_dir", + type=str, + help="path to the desired checkpoint folder, e.g., path/checkpoint-12") + parser.add_argument( + "output_file", + type=str, + help="path to the pytorch fp32 state_dict output file (e.g. path/checkpoint-12/pytorch_model.bin)") + parser.add_argument("-t", + "--tag", + type=str, + default=None, + help="checkpoint tag used as a unique identifier for checkpoint. e.g., global_step1") + parser.add_argument("--exclude_frozen_parameters", action='store_true', help="exclude frozen parameters") + parser.add_argument("-d", "--debug", action='store_true', help="enable debug") + args = parser.parse_args() + + debug = args.debug + + convert_zero_checkpoint_to_fp32_state_dict(args.checkpoint_dir, + args.output_file, + tag=args.tag, + exclude_frozen_parameters=args.exclude_frozen_parameters) diff --git a/babylm_shuffle_nondeterministic_10M_seed0/runs/checkpoint-1400/config.json b/babylm_shuffle_nondeterministic_10M_seed0/runs/checkpoint-1400/config.json new file mode 100644 index 0000000000000000000000000000000000000000..fb0f9a53a507ffcdb60410deb71d60ef801bf350 --- /dev/null +++ b/babylm_shuffle_nondeterministic_10M_seed0/runs/checkpoint-1400/config.json @@ -0,0 +1,36 @@ +{ + "_name_or_path": "meta-llama/Llama-3.2-3B", + "architectures": [ + "LlamaForCausalLM" + ], + "attention_bias": false, + "attention_dropout": 0.0, + "bos_token_id": 128000, + "eos_token_id": 128001, + "head_dim": 128, + "hidden_act": "silu", + "hidden_size": 3072, + "initializer_range": 0.02, + "intermediate_size": 8192, + "max_position_embeddings": 131072, + "mlp_bias": false, + "model_type": "llama", + "num_attention_heads": 24, + "num_hidden_layers": 28, + "num_key_value_heads": 8, + "pretraining_tp": 1, + "rms_norm_eps": 1e-05, + "rope_scaling": { + "factor": 32.0, + "high_freq_factor": 4.0, + "low_freq_factor": 1.0, + "original_max_position_embeddings": 8192, + "rope_type": "llama3" + }, + "rope_theta": 500000.0, + "tie_word_embeddings": true, + "torch_dtype": "float16", + "transformers_version": "4.45.1", + "use_cache": true, + "vocab_size": 128256 +} diff --git a/babylm_shuffle_nondeterministic_10M_seed0/runs/checkpoint-1400/generation_config.json b/babylm_shuffle_nondeterministic_10M_seed0/runs/checkpoint-1400/generation_config.json new file mode 100644 index 0000000000000000000000000000000000000000..9c389b841a9a29ddd904e02d1eb0e08dcce82ad9 --- /dev/null +++ b/babylm_shuffle_nondeterministic_10M_seed0/runs/checkpoint-1400/generation_config.json @@ -0,0 +1,9 @@ +{ + "_from_model_config": true, + "bos_token_id": 128000, + "do_sample": true, + "eos_token_id": 128001, + "temperature": 0.6, + "top_p": 0.9, + "transformers_version": "4.45.1" +} diff --git a/babylm_shuffle_nondeterministic_10M_seed0/runs/checkpoint-1400/latest b/babylm_shuffle_nondeterministic_10M_seed0/runs/checkpoint-1400/latest new file mode 100644 index 0000000000000000000000000000000000000000..c2a0b5b17b996514a9bea0debe9403a641844bce --- /dev/null +++ b/babylm_shuffle_nondeterministic_10M_seed0/runs/checkpoint-1400/latest @@ -0,0 +1 @@ +global_step1400 \ No newline at end of file diff --git a/babylm_shuffle_nondeterministic_10M_seed0/runs/checkpoint-1400/model.safetensors.index.json b/babylm_shuffle_nondeterministic_10M_seed0/runs/checkpoint-1400/model.safetensors.index.json new file mode 100644 index 0000000000000000000000000000000000000000..ed64de846d720b9a7859dc20575fea8e8ca51940 --- /dev/null +++ b/babylm_shuffle_nondeterministic_10M_seed0/runs/checkpoint-1400/model.safetensors.index.json @@ -0,0 +1,262 @@ +{ + "metadata": { + "total_size": 7213504512 + }, + "weight_map": { + "lm_head.weight": "model-00002-of-00002.safetensors", + "model.embed_tokens.weight": "model-00001-of-00002.safetensors", + "model.layers.0.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.0.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.0.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.0.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.0.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.0.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.0.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.0.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.0.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.1.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.1.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.1.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.1.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.1.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.1.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.1.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.1.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.1.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.10.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.10.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.10.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.10.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.10.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.10.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.10.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.10.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.10.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.11.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.11.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.11.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.11.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.11.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.11.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.11.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.11.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.11.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.12.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.12.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.12.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.12.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.12.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.12.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.12.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.12.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.12.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.13.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.13.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.13.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.13.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.13.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.13.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.13.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.13.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.13.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.14.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.14.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.14.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.14.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.14.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.14.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.14.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.14.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.14.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.15.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.15.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.15.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.15.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.15.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.15.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.15.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.15.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.15.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.16.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.16.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.16.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.16.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.16.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.16.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.16.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.16.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.16.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.17.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.17.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.17.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.17.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.17.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.17.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.17.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.17.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.17.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.18.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.18.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.18.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.18.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.18.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.18.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.18.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.18.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.18.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.19.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.19.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.19.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.19.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.19.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.19.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.19.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.19.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.19.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.2.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.2.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.2.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.2.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.2.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.2.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.2.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.2.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.2.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.20.input_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.20.mlp.down_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.20.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.20.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.20.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.20.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.20.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.20.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.20.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.21.input_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.21.mlp.down_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.21.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.21.mlp.up_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.21.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.21.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.21.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.21.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.21.self_attn.v_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.22.input_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.22.mlp.down_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.22.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.22.mlp.up_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.22.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.22.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.22.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.22.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.22.self_attn.v_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.23.input_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.23.mlp.down_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.23.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.23.mlp.up_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.23.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.23.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.23.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.23.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.23.self_attn.v_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.24.input_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.24.mlp.down_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.24.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.24.mlp.up_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.24.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.24.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.24.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.24.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.24.self_attn.v_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.25.input_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.25.mlp.down_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.25.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.25.mlp.up_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.25.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.25.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.25.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.25.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.25.self_attn.v_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.26.input_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.26.mlp.down_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.26.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.26.mlp.up_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.26.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.26.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.26.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.26.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.26.self_attn.v_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.27.input_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.27.mlp.down_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.27.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.27.mlp.up_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.27.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.27.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.27.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.27.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.27.self_attn.v_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.3.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.3.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.3.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.3.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.3.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.3.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.3.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.3.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.3.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.4.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.4.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.4.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.4.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.4.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.4.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.4.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.4.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.4.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.5.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.5.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.5.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.5.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.5.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.5.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.5.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.5.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.5.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.6.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.6.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.6.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.6.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.6.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.6.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.6.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.6.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.6.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.7.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.7.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.7.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.7.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.7.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.7.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.7.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.7.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.7.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.8.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.8.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.8.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.8.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.8.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.8.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.8.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.8.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.8.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.9.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.9.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.9.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.9.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.9.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.9.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.9.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.9.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.9.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.norm.weight": "model-00002-of-00002.safetensors" + } +} diff --git a/babylm_shuffle_nondeterministic_10M_seed0/runs/checkpoint-1400/special_tokens_map.json b/babylm_shuffle_nondeterministic_10M_seed0/runs/checkpoint-1400/special_tokens_map.json new file mode 100644 index 0000000000000000000000000000000000000000..6949c5975ee0e961ef61cf31010dce04df0a03f8 --- /dev/null +++ b/babylm_shuffle_nondeterministic_10M_seed0/runs/checkpoint-1400/special_tokens_map.json @@ -0,0 +1,23 @@ +{ + "bos_token": { + "content": "<|begin_of_text|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false + }, + "eos_token": { + "content": "<|end_of_text|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false + }, + "pad_token": { + "content": "[PAD]", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false + } +} diff --git a/babylm_shuffle_nondeterministic_10M_seed0/runs/checkpoint-1400/tokenizer_config.json b/babylm_shuffle_nondeterministic_10M_seed0/runs/checkpoint-1400/tokenizer_config.json new file mode 100644 index 0000000000000000000000000000000000000000..5684b14e2756b56aa397c1324e239b442162b34a --- /dev/null +++ b/babylm_shuffle_nondeterministic_10M_seed0/runs/checkpoint-1400/tokenizer_config.json @@ -0,0 +1,2070 @@ +{ + "added_tokens_decoder": { + "128000": { + "content": "<|begin_of_text|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128001": { + "content": "<|end_of_text|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128002": { + "content": "<|reserved_special_token_0|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128003": { + "content": "<|reserved_special_token_1|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128004": { + "content": "<|finetune_right_pad_id|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128005": { + "content": "<|reserved_special_token_2|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128006": { + "content": "<|start_header_id|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128007": { + "content": "<|end_header_id|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128008": { + "content": "<|eom_id|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128009": { + "content": "<|eot_id|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128010": { + "content": "<|python_tag|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128011": { + "content": "<|reserved_special_token_3|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128012": { + "content": "<|reserved_special_token_4|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128013": { + "content": "<|reserved_special_token_5|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128014": { + "content": "<|reserved_special_token_6|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128015": { + "content": "<|reserved_special_token_7|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128016": { + "content": "<|reserved_special_token_8|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128017": { + "content": "<|reserved_special_token_9|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128018": { + "content": "<|reserved_special_token_10|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128019": { + "content": "<|reserved_special_token_11|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128020": { + "content": "<|reserved_special_token_12|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128021": { + "content": "<|reserved_special_token_13|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128022": { + "content": "<|reserved_special_token_14|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128023": { + "content": "<|reserved_special_token_15|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128024": { + "content": "<|reserved_special_token_16|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128025": { + "content": "<|reserved_special_token_17|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128026": { + "content": "<|reserved_special_token_18|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128027": { + "content": "<|reserved_special_token_19|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128028": { + "content": "<|reserved_special_token_20|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128029": { + "content": "<|reserved_special_token_21|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128030": { + "content": "<|reserved_special_token_22|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128031": { + "content": "<|reserved_special_token_23|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128032": { + "content": "<|reserved_special_token_24|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128033": { + "content": "<|reserved_special_token_25|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128034": { + "content": "<|reserved_special_token_26|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128035": { + "content": "<|reserved_special_token_27|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128036": { + "content": "<|reserved_special_token_28|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128037": { + "content": "<|reserved_special_token_29|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128038": { + "content": "<|reserved_special_token_30|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128039": { + "content": "<|reserved_special_token_31|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128040": { + "content": "<|reserved_special_token_32|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128041": { + "content": "<|reserved_special_token_33|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128042": { + "content": "<|reserved_special_token_34|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128043": { + "content": "<|reserved_special_token_35|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128044": { + "content": "<|reserved_special_token_36|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128045": { + "content": "<|reserved_special_token_37|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128046": { + "content": "<|reserved_special_token_38|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128047": { + "content": "<|reserved_special_token_39|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128048": { + "content": "<|reserved_special_token_40|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128049": { + "content": "<|reserved_special_token_41|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128050": { + "content": "<|reserved_special_token_42|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128051": { + "content": "<|reserved_special_token_43|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128052": { + "content": "<|reserved_special_token_44|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128053": { + "content": "<|reserved_special_token_45|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128054": { + "content": "<|reserved_special_token_46|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128055": { + "content": "<|reserved_special_token_47|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128056": { + "content": "<|reserved_special_token_48|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128057": { + "content": "<|reserved_special_token_49|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128058": { + "content": "<|reserved_special_token_50|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128059": { + "content": "<|reserved_special_token_51|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128060": { + "content": "<|reserved_special_token_52|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128061": { + "content": "<|reserved_special_token_53|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128062": { + "content": "<|reserved_special_token_54|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128063": { + "content": "<|reserved_special_token_55|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128064": { + "content": "<|reserved_special_token_56|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128065": { + "content": "<|reserved_special_token_57|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128066": { + "content": "<|reserved_special_token_58|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128067": { + "content": "<|reserved_special_token_59|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128068": { + "content": "<|reserved_special_token_60|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128069": { + "content": "<|reserved_special_token_61|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128070": { + "content": "<|reserved_special_token_62|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128071": { + "content": "<|reserved_special_token_63|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128072": { + "content": "<|reserved_special_token_64|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128073": { + "content": "<|reserved_special_token_65|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128074": { + "content": "<|reserved_special_token_66|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128075": { + "content": "<|reserved_special_token_67|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128076": { + "content": "<|reserved_special_token_68|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128077": { + "content": "<|reserved_special_token_69|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128078": { + "content": "<|reserved_special_token_70|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128079": { + "content": "<|reserved_special_token_71|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128080": { + "content": "<|reserved_special_token_72|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128081": { + "content": "<|reserved_special_token_73|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128082": { + "content": "<|reserved_special_token_74|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128083": { + "content": "<|reserved_special_token_75|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128084": { + "content": "<|reserved_special_token_76|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128085": { + "content": "<|reserved_special_token_77|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128086": { + "content": "<|reserved_special_token_78|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128087": { + "content": "<|reserved_special_token_79|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128088": { + "content": "<|reserved_special_token_80|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128089": { + "content": "<|reserved_special_token_81|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128090": { + "content": "<|reserved_special_token_82|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128091": { + "content": "<|reserved_special_token_83|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128092": { + "content": "<|reserved_special_token_84|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128093": { + "content": "<|reserved_special_token_85|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128094": { + "content": "<|reserved_special_token_86|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128095": { + "content": "<|reserved_special_token_87|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128096": { + "content": "<|reserved_special_token_88|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128097": { + "content": "<|reserved_special_token_89|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128098": { + "content": "<|reserved_special_token_90|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128099": { + "content": "<|reserved_special_token_91|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128100": { + "content": "<|reserved_special_token_92|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128101": { + "content": "<|reserved_special_token_93|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128102": { + "content": "<|reserved_special_token_94|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128103": { + "content": "<|reserved_special_token_95|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128104": { + "content": "<|reserved_special_token_96|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128105": { + "content": "<|reserved_special_token_97|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128106": { + "content": "<|reserved_special_token_98|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128107": { + "content": "<|reserved_special_token_99|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128108": { + "content": "<|reserved_special_token_100|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128109": { + "content": "<|reserved_special_token_101|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128110": { + "content": "<|reserved_special_token_102|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128111": { + "content": "<|reserved_special_token_103|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128112": { + "content": "<|reserved_special_token_104|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128113": { + "content": "<|reserved_special_token_105|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128114": { + "content": "<|reserved_special_token_106|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128115": { + "content": "<|reserved_special_token_107|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128116": { + "content": "<|reserved_special_token_108|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128117": { + "content": "<|reserved_special_token_109|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128118": { + "content": "<|reserved_special_token_110|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128119": { + "content": "<|reserved_special_token_111|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128120": { + "content": "<|reserved_special_token_112|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128121": { + "content": "<|reserved_special_token_113|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128122": { + "content": "<|reserved_special_token_114|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128123": { + "content": "<|reserved_special_token_115|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128124": { + "content": "<|reserved_special_token_116|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128125": { + "content": "<|reserved_special_token_117|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128126": { + "content": "<|reserved_special_token_118|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128127": { + "content": "<|reserved_special_token_119|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128128": { + "content": "<|reserved_special_token_120|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128129": { + "content": "<|reserved_special_token_121|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128130": { + "content": "<|reserved_special_token_122|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128131": { + "content": "<|reserved_special_token_123|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128132": { + "content": "<|reserved_special_token_124|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128133": { + "content": "<|reserved_special_token_125|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128134": { + "content": "<|reserved_special_token_126|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128135": { + "content": "<|reserved_special_token_127|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128136": { + "content": "<|reserved_special_token_128|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128137": { + "content": "<|reserved_special_token_129|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128138": { + "content": "<|reserved_special_token_130|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128139": { + "content": "<|reserved_special_token_131|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128140": { + "content": "<|reserved_special_token_132|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128141": { + "content": "<|reserved_special_token_133|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128142": { + "content": "<|reserved_special_token_134|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128143": { + "content": "<|reserved_special_token_135|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128144": { + "content": "<|reserved_special_token_136|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128145": { + "content": "<|reserved_special_token_137|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128146": { + "content": "<|reserved_special_token_138|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128147": { + "content": "<|reserved_special_token_139|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128148": { + "content": "<|reserved_special_token_140|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128149": { + "content": "<|reserved_special_token_141|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128150": { + "content": "<|reserved_special_token_142|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128151": { + "content": "<|reserved_special_token_143|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128152": { + "content": "<|reserved_special_token_144|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128153": { + "content": "<|reserved_special_token_145|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128154": { + "content": "<|reserved_special_token_146|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128155": { + "content": "<|reserved_special_token_147|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128156": { + "content": "<|reserved_special_token_148|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128157": { + "content": "<|reserved_special_token_149|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128158": { + "content": "<|reserved_special_token_150|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128159": { + "content": "<|reserved_special_token_151|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128160": { + "content": "<|reserved_special_token_152|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128161": { + "content": "<|reserved_special_token_153|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128162": { + "content": "<|reserved_special_token_154|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128163": { + "content": "<|reserved_special_token_155|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128164": { + "content": "<|reserved_special_token_156|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128165": { + "content": "<|reserved_special_token_157|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128166": { + "content": "<|reserved_special_token_158|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128167": { + "content": "<|reserved_special_token_159|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128168": { + "content": "<|reserved_special_token_160|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128169": { + "content": "<|reserved_special_token_161|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128170": { + "content": "<|reserved_special_token_162|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128171": { + "content": "<|reserved_special_token_163|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128172": { + "content": "<|reserved_special_token_164|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128173": { + "content": "<|reserved_special_token_165|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128174": { + "content": "<|reserved_special_token_166|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128175": { + "content": "<|reserved_special_token_167|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128176": { + "content": "<|reserved_special_token_168|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128177": { + "content": "<|reserved_special_token_169|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128178": { + "content": "<|reserved_special_token_170|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128179": { + "content": "<|reserved_special_token_171|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128180": { + "content": "<|reserved_special_token_172|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128181": { + "content": "<|reserved_special_token_173|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128182": { + "content": "<|reserved_special_token_174|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128183": { + "content": "<|reserved_special_token_175|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128184": { + "content": "<|reserved_special_token_176|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128185": { + "content": "<|reserved_special_token_177|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128186": { + "content": "<|reserved_special_token_178|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128187": { + "content": "<|reserved_special_token_179|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128188": { + "content": "<|reserved_special_token_180|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128189": { + "content": "<|reserved_special_token_181|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128190": { + "content": "<|reserved_special_token_182|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128191": { + "content": "<|reserved_special_token_183|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128192": { + "content": "<|reserved_special_token_184|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128193": { + "content": "<|reserved_special_token_185|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128194": { + "content": "<|reserved_special_token_186|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128195": { + "content": "<|reserved_special_token_187|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128196": { + "content": "<|reserved_special_token_188|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128197": { + "content": "<|reserved_special_token_189|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128198": { + "content": "<|reserved_special_token_190|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128199": { + "content": "<|reserved_special_token_191|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128200": { + "content": "<|reserved_special_token_192|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128201": { + "content": "<|reserved_special_token_193|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128202": { + "content": "<|reserved_special_token_194|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128203": { + "content": "<|reserved_special_token_195|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128204": { + "content": "<|reserved_special_token_196|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128205": { + "content": "<|reserved_special_token_197|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128206": { + "content": "<|reserved_special_token_198|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128207": { + "content": "<|reserved_special_token_199|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128208": { + "content": "<|reserved_special_token_200|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128209": { + "content": "<|reserved_special_token_201|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128210": { + "content": "<|reserved_special_token_202|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128211": { + "content": "<|reserved_special_token_203|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128212": { + "content": "<|reserved_special_token_204|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128213": { + "content": "<|reserved_special_token_205|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128214": { + "content": "<|reserved_special_token_206|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128215": { + "content": "<|reserved_special_token_207|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128216": { + "content": "<|reserved_special_token_208|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128217": { + "content": "<|reserved_special_token_209|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128218": { + "content": "<|reserved_special_token_210|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128219": { + "content": "<|reserved_special_token_211|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128220": { + "content": "<|reserved_special_token_212|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128221": { + "content": "<|reserved_special_token_213|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128222": { + "content": "<|reserved_special_token_214|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128223": { + "content": "<|reserved_special_token_215|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128224": { + "content": "<|reserved_special_token_216|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128225": { + "content": "<|reserved_special_token_217|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128226": { + "content": "<|reserved_special_token_218|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128227": { + "content": "<|reserved_special_token_219|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128228": { + "content": "<|reserved_special_token_220|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128229": { + "content": "<|reserved_special_token_221|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128230": { + "content": "<|reserved_special_token_222|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128231": { + "content": "<|reserved_special_token_223|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128232": { + "content": "<|reserved_special_token_224|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128233": { + "content": "<|reserved_special_token_225|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128234": { + "content": "<|reserved_special_token_226|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128235": { + "content": "<|reserved_special_token_227|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128236": { + "content": "<|reserved_special_token_228|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128237": { + "content": "<|reserved_special_token_229|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128238": { + "content": "<|reserved_special_token_230|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128239": { + "content": "<|reserved_special_token_231|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128240": { + "content": "<|reserved_special_token_232|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128241": { + "content": "<|reserved_special_token_233|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128242": { + "content": "<|reserved_special_token_234|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128243": { + "content": "<|reserved_special_token_235|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128244": { + "content": "<|reserved_special_token_236|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128245": { + "content": "<|reserved_special_token_237|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128246": { + "content": "<|reserved_special_token_238|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128247": { + "content": "<|reserved_special_token_239|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128248": { + "content": "<|reserved_special_token_240|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128249": { + "content": "<|reserved_special_token_241|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128250": { + "content": "<|reserved_special_token_242|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128251": { + "content": "<|reserved_special_token_243|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128252": { + "content": "<|reserved_special_token_244|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128253": { + "content": "<|reserved_special_token_245|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128254": { + "content": "<|reserved_special_token_246|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128255": { + "content": "<|reserved_special_token_247|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128256": { + "content": "[PAD]", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + } + }, + "bos_token": "<|begin_of_text|>", + "clean_up_tokenization_spaces": true, + "eos_token": "<|end_of_text|>", + "model_input_names": [ + "input_ids", + "attention_mask" + ], + "model_max_length": 131072, + "pad_token": "[PAD]", + "tokenizer_class": "PreTrainedTokenizerFast" +} diff --git a/babylm_shuffle_nondeterministic_10M_seed0/runs/checkpoint-1400/trainer_state.json b/babylm_shuffle_nondeterministic_10M_seed0/runs/checkpoint-1400/trainer_state.json new file mode 100644 index 0000000000000000000000000000000000000000..ac3b5b9fbb56a560adf73268f3006173f5ac91ab --- /dev/null +++ b/babylm_shuffle_nondeterministic_10M_seed0/runs/checkpoint-1400/trainer_state.json @@ -0,0 +1,10953 @@ +{ + "best_metric": null, + "best_model_checkpoint": null, + "epoch": 1.5342465753424657, + "eval_steps": 10, + "global_step": 1400, + "is_hyper_param_search": false, + "is_local_process_zero": true, + "is_world_process_zero": true, + "log_history": [ + { + "epoch": 0.001095890410958904, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.0998, + "step": 1 + }, + { + "epoch": 0.002191780821917808, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.0858, + "step": 2 + }, + { + "epoch": 0.0032876712328767125, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.0672, + "step": 3 + }, + { + "epoch": 0.004383561643835616, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.0802, + "step": 4 + }, + { + "epoch": 0.005479452054794521, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.0636, + "step": 5 + }, + { + "epoch": 0.006575342465753425, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.1672, + "step": 6 + }, + { + "epoch": 0.007671232876712329, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.0961, + "step": 7 + }, + { + "epoch": 0.008767123287671232, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.1258, + "step": 8 + }, + { + "epoch": 0.009863013698630137, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.0953, + "step": 9 + }, + { + "epoch": 0.010958904109589041, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.1404, + "step": 10 + }, + { + "epoch": 0.010958904109589041, + "eval_loss": 3.12593412399292, + "eval_runtime": 42.732, + "eval_samples_per_second": 23.402, + "eval_steps_per_second": 0.983, + "step": 10 + }, + { + "epoch": 0.012054794520547946, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.1187, + "step": 11 + }, + { + "epoch": 0.01315068493150685, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.0593, + "step": 12 + }, + { + "epoch": 0.014246575342465753, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.111, + "step": 13 + }, + { + "epoch": 0.015342465753424657, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.0974, + "step": 14 + }, + { + "epoch": 0.01643835616438356, + "grad_norm": 5.991453170776367, + "learning_rate": 1.8248175182481753e-08, + "loss": 3.135, + "step": 15 + }, + { + "epoch": 0.017534246575342465, + "grad_norm": 6.274210453033447, + "learning_rate": 3.6496350364963505e-08, + "loss": 3.083, + "step": 16 + }, + { + "epoch": 0.01863013698630137, + "grad_norm": 6.644786834716797, + "learning_rate": 5.4744525547445264e-08, + "loss": 3.0856, + "step": 17 + }, + { + "epoch": 0.019726027397260273, + "grad_norm": 6.769156455993652, + "learning_rate": 7.299270072992701e-08, + "loss": 3.1082, + "step": 18 + }, + { + "epoch": 0.020821917808219178, + "grad_norm": 6.771246910095215, + "learning_rate": 9.124087591240876e-08, + "loss": 3.0753, + "step": 19 + }, + { + "epoch": 0.021917808219178082, + "grad_norm": 6.468441486358643, + "learning_rate": 1.0948905109489053e-07, + "loss": 3.1478, + "step": 20 + }, + { + "epoch": 0.021917808219178082, + "eval_loss": 3.1245107650756836, + "eval_runtime": 43.6251, + "eval_samples_per_second": 22.923, + "eval_steps_per_second": 0.963, + "step": 20 + }, + { + "epoch": 0.023013698630136987, + "grad_norm": 6.83078670501709, + "learning_rate": 1.2773722627737227e-07, + "loss": 3.1213, + "step": 21 + }, + { + "epoch": 0.02410958904109589, + "grad_norm": 6.221287250518799, + "learning_rate": 1.4598540145985402e-07, + "loss": 3.0549, + "step": 22 + }, + { + "epoch": 0.025205479452054796, + "grad_norm": 5.564266204833984, + "learning_rate": 1.642335766423358e-07, + "loss": 3.0769, + "step": 23 + }, + { + "epoch": 0.0263013698630137, + "grad_norm": 5.735209941864014, + "learning_rate": 1.8248175182481753e-07, + "loss": 3.1124, + "step": 24 + }, + { + "epoch": 0.0273972602739726, + "grad_norm": 5.577925682067871, + "learning_rate": 2.0072992700729928e-07, + "loss": 3.1191, + "step": 25 + }, + { + "epoch": 0.028493150684931506, + "grad_norm": 4.814139366149902, + "learning_rate": 2.1897810218978106e-07, + "loss": 3.0947, + "step": 26 + }, + { + "epoch": 0.02958904109589041, + "grad_norm": 4.69775390625, + "learning_rate": 2.3722627737226278e-07, + "loss": 3.0997, + "step": 27 + }, + { + "epoch": 0.030684931506849315, + "grad_norm": 4.811586380004883, + "learning_rate": 2.5547445255474454e-07, + "loss": 3.0749, + "step": 28 + }, + { + "epoch": 0.03178082191780822, + "grad_norm": 4.623377323150635, + "learning_rate": 2.737226277372263e-07, + "loss": 3.0619, + "step": 29 + }, + { + "epoch": 0.03287671232876712, + "grad_norm": 4.721253395080566, + "learning_rate": 2.9197080291970804e-07, + "loss": 3.0371, + "step": 30 + }, + { + "epoch": 0.03287671232876712, + "eval_loss": 3.082857847213745, + "eval_runtime": 44.127, + "eval_samples_per_second": 22.662, + "eval_steps_per_second": 0.952, + "step": 30 + }, + { + "epoch": 0.03397260273972603, + "grad_norm": 3.968630790710449, + "learning_rate": 3.102189781021898e-07, + "loss": 3.0535, + "step": 31 + }, + { + "epoch": 0.03506849315068493, + "grad_norm": 4.4373779296875, + "learning_rate": 3.284671532846716e-07, + "loss": 3.0476, + "step": 32 + }, + { + "epoch": 0.03616438356164384, + "grad_norm": 4.479320049285889, + "learning_rate": 3.467153284671533e-07, + "loss": 3.0215, + "step": 33 + }, + { + "epoch": 0.03726027397260274, + "grad_norm": 4.122231960296631, + "learning_rate": 3.6496350364963505e-07, + "loss": 3.0106, + "step": 34 + }, + { + "epoch": 0.038356164383561646, + "grad_norm": 4.153995513916016, + "learning_rate": 3.8321167883211686e-07, + "loss": 3.0318, + "step": 35 + }, + { + "epoch": 0.03945205479452055, + "grad_norm": 3.6949076652526855, + "learning_rate": 4.0145985401459856e-07, + "loss": 3.046, + "step": 36 + }, + { + "epoch": 0.040547945205479455, + "grad_norm": 3.4610488414764404, + "learning_rate": 4.197080291970803e-07, + "loss": 3.002, + "step": 37 + }, + { + "epoch": 0.041643835616438356, + "grad_norm": 3.9523138999938965, + "learning_rate": 4.379562043795621e-07, + "loss": 2.9665, + "step": 38 + }, + { + "epoch": 0.042739726027397264, + "grad_norm": 3.68516206741333, + "learning_rate": 4.562043795620438e-07, + "loss": 3.0411, + "step": 39 + }, + { + "epoch": 0.043835616438356165, + "grad_norm": 3.692396402359009, + "learning_rate": 4.7445255474452557e-07, + "loss": 2.9666, + "step": 40 + }, + { + "epoch": 0.043835616438356165, + "eval_loss": 3.015512704849243, + "eval_runtime": 48.2438, + "eval_samples_per_second": 20.728, + "eval_steps_per_second": 0.871, + "step": 40 + }, + { + "epoch": 0.044931506849315066, + "grad_norm": 3.3765921592712402, + "learning_rate": 4.927007299270073e-07, + "loss": 2.9831, + "step": 41 + }, + { + "epoch": 0.046027397260273974, + "grad_norm": 2.9631125926971436, + "learning_rate": 5.109489051094891e-07, + "loss": 3.0126, + "step": 42 + }, + { + "epoch": 0.047123287671232875, + "grad_norm": 2.969730854034424, + "learning_rate": 5.291970802919708e-07, + "loss": 2.9803, + "step": 43 + }, + { + "epoch": 0.04821917808219178, + "grad_norm": 3.185237407684326, + "learning_rate": 5.474452554744526e-07, + "loss": 2.9888, + "step": 44 + }, + { + "epoch": 0.049315068493150684, + "grad_norm": 3.055579423904419, + "learning_rate": 5.656934306569343e-07, + "loss": 2.9715, + "step": 45 + }, + { + "epoch": 0.05041095890410959, + "grad_norm": 3.083284616470337, + "learning_rate": 5.839416058394161e-07, + "loss": 2.9293, + "step": 46 + }, + { + "epoch": 0.05150684931506849, + "grad_norm": 2.984781503677368, + "learning_rate": 6.021897810218979e-07, + "loss": 2.9228, + "step": 47 + }, + { + "epoch": 0.0526027397260274, + "grad_norm": 2.756173849105835, + "learning_rate": 6.204379562043796e-07, + "loss": 2.9406, + "step": 48 + }, + { + "epoch": 0.0536986301369863, + "grad_norm": 2.911132574081421, + "learning_rate": 6.386861313868614e-07, + "loss": 2.8829, + "step": 49 + }, + { + "epoch": 0.0547945205479452, + "grad_norm": 2.592890977859497, + "learning_rate": 6.569343065693432e-07, + "loss": 2.9839, + "step": 50 + }, + { + "epoch": 0.0547945205479452, + "eval_loss": 2.9409468173980713, + "eval_runtime": 47.1045, + "eval_samples_per_second": 21.229, + "eval_steps_per_second": 0.892, + "step": 50 + }, + { + "epoch": 0.05589041095890411, + "grad_norm": 3.0112955570220947, + "learning_rate": 6.751824817518248e-07, + "loss": 2.8847, + "step": 51 + }, + { + "epoch": 0.05698630136986301, + "grad_norm": 2.3701798915863037, + "learning_rate": 6.934306569343066e-07, + "loss": 2.9109, + "step": 52 + }, + { + "epoch": 0.05808219178082192, + "grad_norm": 2.268850088119507, + "learning_rate": 7.116788321167883e-07, + "loss": 2.8483, + "step": 53 + }, + { + "epoch": 0.05917808219178082, + "grad_norm": 2.419015645980835, + "learning_rate": 7.299270072992701e-07, + "loss": 2.8491, + "step": 54 + }, + { + "epoch": 0.06027397260273973, + "grad_norm": 2.4582653045654297, + "learning_rate": 7.48175182481752e-07, + "loss": 2.8622, + "step": 55 + }, + { + "epoch": 0.06136986301369863, + "grad_norm": 2.3471553325653076, + "learning_rate": 7.664233576642337e-07, + "loss": 2.8738, + "step": 56 + }, + { + "epoch": 0.06246575342465754, + "grad_norm": 2.479177236557007, + "learning_rate": 7.846715328467154e-07, + "loss": 2.8843, + "step": 57 + }, + { + "epoch": 0.06356164383561644, + "grad_norm": 2.1676292419433594, + "learning_rate": 8.029197080291971e-07, + "loss": 2.8674, + "step": 58 + }, + { + "epoch": 0.06465753424657535, + "grad_norm": 1.9862910509109497, + "learning_rate": 8.211678832116789e-07, + "loss": 2.8197, + "step": 59 + }, + { + "epoch": 0.06575342465753424, + "grad_norm": 2.211313486099243, + "learning_rate": 8.394160583941606e-07, + "loss": 2.7961, + "step": 60 + }, + { + "epoch": 0.06575342465753424, + "eval_loss": 2.880824327468872, + "eval_runtime": 51.3235, + "eval_samples_per_second": 19.484, + "eval_steps_per_second": 0.818, + "step": 60 + }, + { + "epoch": 0.06684931506849315, + "grad_norm": 2.3497281074523926, + "learning_rate": 8.576642335766425e-07, + "loss": 2.8233, + "step": 61 + }, + { + "epoch": 0.06794520547945206, + "grad_norm": 2.158515214920044, + "learning_rate": 8.759124087591242e-07, + "loss": 2.8136, + "step": 62 + }, + { + "epoch": 0.06904109589041096, + "grad_norm": 2.0219955444335938, + "learning_rate": 8.94160583941606e-07, + "loss": 2.8231, + "step": 63 + }, + { + "epoch": 0.07013698630136986, + "grad_norm": 1.8989133834838867, + "learning_rate": 9.124087591240876e-07, + "loss": 2.8268, + "step": 64 + }, + { + "epoch": 0.07123287671232877, + "grad_norm": 2.243851661682129, + "learning_rate": 9.306569343065694e-07, + "loss": 2.7886, + "step": 65 + }, + { + "epoch": 0.07232876712328767, + "grad_norm": 2.122464418411255, + "learning_rate": 9.489051094890511e-07, + "loss": 2.8166, + "step": 66 + }, + { + "epoch": 0.07342465753424658, + "grad_norm": 1.899857997894287, + "learning_rate": 9.67153284671533e-07, + "loss": 2.7997, + "step": 67 + }, + { + "epoch": 0.07452054794520548, + "grad_norm": 2.4424173831939697, + "learning_rate": 9.854014598540146e-07, + "loss": 2.8342, + "step": 68 + }, + { + "epoch": 0.07561643835616438, + "grad_norm": 2.5677411556243896, + "learning_rate": 1.0036496350364965e-06, + "loss": 2.7786, + "step": 69 + }, + { + "epoch": 0.07671232876712329, + "grad_norm": 1.9989523887634277, + "learning_rate": 1.0218978102189781e-06, + "loss": 2.7411, + "step": 70 + }, + { + "epoch": 0.07671232876712329, + "eval_loss": 2.8185322284698486, + "eval_runtime": 50.4998, + "eval_samples_per_second": 19.802, + "eval_steps_per_second": 0.832, + "step": 70 + }, + { + "epoch": 0.07780821917808219, + "grad_norm": 1.9376521110534668, + "learning_rate": 1.04014598540146e-06, + "loss": 2.7988, + "step": 71 + }, + { + "epoch": 0.0789041095890411, + "grad_norm": 2.5629725456237793, + "learning_rate": 1.0583941605839416e-06, + "loss": 2.7079, + "step": 72 + }, + { + "epoch": 0.08, + "grad_norm": 2.1634678840637207, + "learning_rate": 1.0766423357664233e-06, + "loss": 2.7049, + "step": 73 + }, + { + "epoch": 0.08109589041095891, + "grad_norm": 2.2143008708953857, + "learning_rate": 1.0948905109489052e-06, + "loss": 2.72, + "step": 74 + }, + { + "epoch": 0.0821917808219178, + "grad_norm": 2.2701101303100586, + "learning_rate": 1.113138686131387e-06, + "loss": 2.7418, + "step": 75 + }, + { + "epoch": 0.08328767123287671, + "grad_norm": 1.716569423675537, + "learning_rate": 1.1313868613138687e-06, + "loss": 2.7558, + "step": 76 + }, + { + "epoch": 0.08438356164383562, + "grad_norm": 2.79757022857666, + "learning_rate": 1.1496350364963505e-06, + "loss": 2.6865, + "step": 77 + }, + { + "epoch": 0.08547945205479453, + "grad_norm": 2.2189838886260986, + "learning_rate": 1.1678832116788322e-06, + "loss": 2.6684, + "step": 78 + }, + { + "epoch": 0.08657534246575342, + "grad_norm": 2.0695579051971436, + "learning_rate": 1.1861313868613138e-06, + "loss": 2.6779, + "step": 79 + }, + { + "epoch": 0.08767123287671233, + "grad_norm": 2.372326374053955, + "learning_rate": 1.2043795620437959e-06, + "loss": 2.6711, + "step": 80 + }, + { + "epoch": 0.08767123287671233, + "eval_loss": 2.7537453174591064, + "eval_runtime": 47.7274, + "eval_samples_per_second": 20.952, + "eval_steps_per_second": 0.88, + "step": 80 + }, + { + "epoch": 0.08876712328767124, + "grad_norm": 2.3398947715759277, + "learning_rate": 1.2226277372262775e-06, + "loss": 2.6776, + "step": 81 + }, + { + "epoch": 0.08986301369863013, + "grad_norm": 2.6619036197662354, + "learning_rate": 1.2408759124087592e-06, + "loss": 2.7077, + "step": 82 + }, + { + "epoch": 0.09095890410958904, + "grad_norm": 1.8245115280151367, + "learning_rate": 1.259124087591241e-06, + "loss": 2.662, + "step": 83 + }, + { + "epoch": 0.09205479452054795, + "grad_norm": 2.2627148628234863, + "learning_rate": 1.2773722627737229e-06, + "loss": 2.677, + "step": 84 + }, + { + "epoch": 0.09315068493150686, + "grad_norm": 2.4983835220336914, + "learning_rate": 1.2956204379562045e-06, + "loss": 2.6601, + "step": 85 + }, + { + "epoch": 0.09424657534246575, + "grad_norm": 2.147010087966919, + "learning_rate": 1.3138686131386864e-06, + "loss": 2.6213, + "step": 86 + }, + { + "epoch": 0.09534246575342466, + "grad_norm": 2.791682481765747, + "learning_rate": 1.332116788321168e-06, + "loss": 2.6281, + "step": 87 + }, + { + "epoch": 0.09643835616438357, + "grad_norm": 2.023970365524292, + "learning_rate": 1.3503649635036497e-06, + "loss": 2.5766, + "step": 88 + }, + { + "epoch": 0.09753424657534246, + "grad_norm": 2.8143956661224365, + "learning_rate": 1.3686131386861315e-06, + "loss": 2.6086, + "step": 89 + }, + { + "epoch": 0.09863013698630137, + "grad_norm": 2.226106882095337, + "learning_rate": 1.3868613138686132e-06, + "loss": 2.5477, + "step": 90 + }, + { + "epoch": 0.09863013698630137, + "eval_loss": 2.6880786418914795, + "eval_runtime": 47.8417, + "eval_samples_per_second": 20.902, + "eval_steps_per_second": 0.878, + "step": 90 + }, + { + "epoch": 0.09972602739726028, + "grad_norm": 2.5248148441314697, + "learning_rate": 1.405109489051095e-06, + "loss": 2.6232, + "step": 91 + }, + { + "epoch": 0.10082191780821918, + "grad_norm": 2.421330690383911, + "learning_rate": 1.4233576642335767e-06, + "loss": 2.599, + "step": 92 + }, + { + "epoch": 0.10191780821917808, + "grad_norm": 2.5562429428100586, + "learning_rate": 1.4416058394160583e-06, + "loss": 2.6197, + "step": 93 + }, + { + "epoch": 0.10301369863013699, + "grad_norm": 2.856201171875, + "learning_rate": 1.4598540145985402e-06, + "loss": 2.5825, + "step": 94 + }, + { + "epoch": 0.10410958904109589, + "grad_norm": 2.6576807498931885, + "learning_rate": 1.478102189781022e-06, + "loss": 2.5496, + "step": 95 + }, + { + "epoch": 0.1052054794520548, + "grad_norm": 3.1408190727233887, + "learning_rate": 1.496350364963504e-06, + "loss": 2.6142, + "step": 96 + }, + { + "epoch": 0.1063013698630137, + "grad_norm": 2.3270952701568604, + "learning_rate": 1.5145985401459856e-06, + "loss": 2.584, + "step": 97 + }, + { + "epoch": 0.1073972602739726, + "grad_norm": 2.3976614475250244, + "learning_rate": 1.5328467153284674e-06, + "loss": 2.5941, + "step": 98 + }, + { + "epoch": 0.10849315068493151, + "grad_norm": 2.783485174179077, + "learning_rate": 1.551094890510949e-06, + "loss": 2.5626, + "step": 99 + }, + { + "epoch": 0.1095890410958904, + "grad_norm": 2.2522788047790527, + "learning_rate": 1.5693430656934307e-06, + "loss": 2.517, + "step": 100 + }, + { + "epoch": 0.1095890410958904, + "eval_loss": 2.637315034866333, + "eval_runtime": 48.2107, + "eval_samples_per_second": 20.742, + "eval_steps_per_second": 0.871, + "step": 100 + }, + { + "epoch": 0.11068493150684931, + "grad_norm": 2.917799949645996, + "learning_rate": 1.5875912408759126e-06, + "loss": 2.5159, + "step": 101 + }, + { + "epoch": 0.11178082191780822, + "grad_norm": 2.955815076828003, + "learning_rate": 1.6058394160583942e-06, + "loss": 2.563, + "step": 102 + }, + { + "epoch": 0.11287671232876713, + "grad_norm": 2.8096001148223877, + "learning_rate": 1.624087591240876e-06, + "loss": 2.5995, + "step": 103 + }, + { + "epoch": 0.11397260273972602, + "grad_norm": 2.4967427253723145, + "learning_rate": 1.6423357664233577e-06, + "loss": 2.5336, + "step": 104 + }, + { + "epoch": 0.11506849315068493, + "grad_norm": 2.740717887878418, + "learning_rate": 1.6605839416058394e-06, + "loss": 2.4778, + "step": 105 + }, + { + "epoch": 0.11616438356164384, + "grad_norm": 2.5064632892608643, + "learning_rate": 1.6788321167883212e-06, + "loss": 2.5214, + "step": 106 + }, + { + "epoch": 0.11726027397260275, + "grad_norm": 2.2923007011413574, + "learning_rate": 1.6970802919708029e-06, + "loss": 2.5315, + "step": 107 + }, + { + "epoch": 0.11835616438356164, + "grad_norm": 3.339569568634033, + "learning_rate": 1.715328467153285e-06, + "loss": 2.4661, + "step": 108 + }, + { + "epoch": 0.11945205479452055, + "grad_norm": 2.1650891304016113, + "learning_rate": 1.7335766423357666e-06, + "loss": 2.5072, + "step": 109 + }, + { + "epoch": 0.12054794520547946, + "grad_norm": 3.1216914653778076, + "learning_rate": 1.7518248175182485e-06, + "loss": 2.5164, + "step": 110 + }, + { + "epoch": 0.12054794520547946, + "eval_loss": 2.594250202178955, + "eval_runtime": 41.8939, + "eval_samples_per_second": 23.87, + "eval_steps_per_second": 1.003, + "step": 110 + }, + { + "epoch": 0.12164383561643835, + "grad_norm": 2.191063404083252, + "learning_rate": 1.77007299270073e-06, + "loss": 2.5422, + "step": 111 + }, + { + "epoch": 0.12273972602739726, + "grad_norm": 2.922083854675293, + "learning_rate": 1.788321167883212e-06, + "loss": 2.5258, + "step": 112 + }, + { + "epoch": 0.12383561643835617, + "grad_norm": 2.518002986907959, + "learning_rate": 1.8065693430656936e-06, + "loss": 2.5588, + "step": 113 + }, + { + "epoch": 0.12493150684931507, + "grad_norm": 2.515194892883301, + "learning_rate": 1.8248175182481753e-06, + "loss": 2.5079, + "step": 114 + }, + { + "epoch": 0.12602739726027398, + "grad_norm": 2.5530874729156494, + "learning_rate": 1.8430656934306571e-06, + "loss": 2.4615, + "step": 115 + }, + { + "epoch": 0.12712328767123288, + "grad_norm": 2.3550198078155518, + "learning_rate": 1.8613138686131388e-06, + "loss": 2.4825, + "step": 116 + }, + { + "epoch": 0.12821917808219177, + "grad_norm": 2.295145034790039, + "learning_rate": 1.8795620437956206e-06, + "loss": 2.4749, + "step": 117 + }, + { + "epoch": 0.1293150684931507, + "grad_norm": 2.0158982276916504, + "learning_rate": 1.8978102189781023e-06, + "loss": 2.4756, + "step": 118 + }, + { + "epoch": 0.1304109589041096, + "grad_norm": 2.213057518005371, + "learning_rate": 1.916058394160584e-06, + "loss": 2.4744, + "step": 119 + }, + { + "epoch": 0.13150684931506848, + "grad_norm": 2.186222553253174, + "learning_rate": 1.934306569343066e-06, + "loss": 2.4957, + "step": 120 + }, + { + "epoch": 0.13150684931506848, + "eval_loss": 2.563931465148926, + "eval_runtime": 44.8714, + "eval_samples_per_second": 22.286, + "eval_steps_per_second": 0.936, + "step": 120 + }, + { + "epoch": 0.1326027397260274, + "grad_norm": 2.3143310546875, + "learning_rate": 1.952554744525548e-06, + "loss": 2.4, + "step": 121 + }, + { + "epoch": 0.1336986301369863, + "grad_norm": 2.0223965644836426, + "learning_rate": 1.9708029197080293e-06, + "loss": 2.4453, + "step": 122 + }, + { + "epoch": 0.13479452054794522, + "grad_norm": 2.2025411128997803, + "learning_rate": 1.989051094890511e-06, + "loss": 2.4463, + "step": 123 + }, + { + "epoch": 0.1358904109589041, + "grad_norm": 2.707930564880371, + "learning_rate": 2.007299270072993e-06, + "loss": 2.4589, + "step": 124 + }, + { + "epoch": 0.136986301369863, + "grad_norm": 2.146662473678589, + "learning_rate": 2.025547445255475e-06, + "loss": 2.5023, + "step": 125 + }, + { + "epoch": 0.13808219178082193, + "grad_norm": 2.520660161972046, + "learning_rate": 2.0437956204379563e-06, + "loss": 2.4464, + "step": 126 + }, + { + "epoch": 0.13917808219178082, + "grad_norm": 2.50632381439209, + "learning_rate": 2.062043795620438e-06, + "loss": 2.4556, + "step": 127 + }, + { + "epoch": 0.14027397260273972, + "grad_norm": 2.2919397354125977, + "learning_rate": 2.08029197080292e-06, + "loss": 2.4479, + "step": 128 + }, + { + "epoch": 0.14136986301369864, + "grad_norm": 2.521475315093994, + "learning_rate": 2.0985401459854014e-06, + "loss": 2.4982, + "step": 129 + }, + { + "epoch": 0.14246575342465753, + "grad_norm": 2.3812668323516846, + "learning_rate": 2.1167883211678833e-06, + "loss": 2.4163, + "step": 130 + }, + { + "epoch": 0.14246575342465753, + "eval_loss": 2.5390918254852295, + "eval_runtime": 43.8746, + "eval_samples_per_second": 22.792, + "eval_steps_per_second": 0.957, + "step": 130 + }, + { + "epoch": 0.14356164383561643, + "grad_norm": 2.302286148071289, + "learning_rate": 2.135036496350365e-06, + "loss": 2.4588, + "step": 131 + }, + { + "epoch": 0.14465753424657535, + "grad_norm": 2.4222044944763184, + "learning_rate": 2.1532846715328466e-06, + "loss": 2.492, + "step": 132 + }, + { + "epoch": 0.14575342465753424, + "grad_norm": 2.605835199356079, + "learning_rate": 2.171532846715329e-06, + "loss": 2.4055, + "step": 133 + }, + { + "epoch": 0.14684931506849316, + "grad_norm": 3.6261861324310303, + "learning_rate": 2.1897810218978103e-06, + "loss": 2.4221, + "step": 134 + }, + { + "epoch": 0.14794520547945206, + "grad_norm": 2.7647221088409424, + "learning_rate": 2.208029197080292e-06, + "loss": 2.4549, + "step": 135 + }, + { + "epoch": 0.14904109589041095, + "grad_norm": 3.2977371215820312, + "learning_rate": 2.226277372262774e-06, + "loss": 2.4331, + "step": 136 + }, + { + "epoch": 0.15013698630136987, + "grad_norm": 3.147094488143921, + "learning_rate": 2.244525547445256e-06, + "loss": 2.4268, + "step": 137 + }, + { + "epoch": 0.15123287671232877, + "grad_norm": 2.6363413333892822, + "learning_rate": 2.2627737226277373e-06, + "loss": 2.4524, + "step": 138 + }, + { + "epoch": 0.15232876712328766, + "grad_norm": 2.9288904666900635, + "learning_rate": 2.281021897810219e-06, + "loss": 2.4534, + "step": 139 + }, + { + "epoch": 0.15342465753424658, + "grad_norm": 2.5604870319366455, + "learning_rate": 2.299270072992701e-06, + "loss": 2.4138, + "step": 140 + }, + { + "epoch": 0.15342465753424658, + "eval_loss": 2.524790048599243, + "eval_runtime": 43.8851, + "eval_samples_per_second": 22.787, + "eval_steps_per_second": 0.957, + "step": 140 + }, + { + "epoch": 0.15452054794520548, + "grad_norm": 3.2746663093566895, + "learning_rate": 2.3175182481751825e-06, + "loss": 2.4094, + "step": 141 + }, + { + "epoch": 0.15561643835616437, + "grad_norm": 2.691483497619629, + "learning_rate": 2.3357664233576643e-06, + "loss": 2.4265, + "step": 142 + }, + { + "epoch": 0.1567123287671233, + "grad_norm": 2.5850117206573486, + "learning_rate": 2.354014598540146e-06, + "loss": 2.3922, + "step": 143 + }, + { + "epoch": 0.1578082191780822, + "grad_norm": 2.68204927444458, + "learning_rate": 2.3722627737226276e-06, + "loss": 2.4262, + "step": 144 + }, + { + "epoch": 0.1589041095890411, + "grad_norm": 2.952584981918335, + "learning_rate": 2.39051094890511e-06, + "loss": 2.4267, + "step": 145 + }, + { + "epoch": 0.16, + "grad_norm": 3.2354788780212402, + "learning_rate": 2.4087591240875918e-06, + "loss": 2.3836, + "step": 146 + }, + { + "epoch": 0.1610958904109589, + "grad_norm": 2.634521007537842, + "learning_rate": 2.427007299270073e-06, + "loss": 2.3953, + "step": 147 + }, + { + "epoch": 0.16219178082191782, + "grad_norm": 3.1115682125091553, + "learning_rate": 2.445255474452555e-06, + "loss": 2.409, + "step": 148 + }, + { + "epoch": 0.1632876712328767, + "grad_norm": 2.6824135780334473, + "learning_rate": 2.463503649635037e-06, + "loss": 2.4459, + "step": 149 + }, + { + "epoch": 0.1643835616438356, + "grad_norm": 2.617614507675171, + "learning_rate": 2.4817518248175183e-06, + "loss": 2.366, + "step": 150 + }, + { + "epoch": 0.1643835616438356, + "eval_loss": 2.5088160037994385, + "eval_runtime": 44.0888, + "eval_samples_per_second": 22.682, + "eval_steps_per_second": 0.953, + "step": 150 + }, + { + "epoch": 0.16547945205479453, + "grad_norm": 3.976992130279541, + "learning_rate": 2.5e-06, + "loss": 2.3517, + "step": 151 + }, + { + "epoch": 0.16657534246575342, + "grad_norm": 3.1433017253875732, + "learning_rate": 2.518248175182482e-06, + "loss": 2.4489, + "step": 152 + }, + { + "epoch": 0.16767123287671232, + "grad_norm": 4.100688934326172, + "learning_rate": 2.5364963503649635e-06, + "loss": 2.3683, + "step": 153 + }, + { + "epoch": 0.16876712328767124, + "grad_norm": 4.316102504730225, + "learning_rate": 2.5547445255474458e-06, + "loss": 2.3663, + "step": 154 + }, + { + "epoch": 0.16986301369863013, + "grad_norm": 2.3103132247924805, + "learning_rate": 2.5729927007299272e-06, + "loss": 2.4009, + "step": 155 + }, + { + "epoch": 0.17095890410958905, + "grad_norm": 3.576810836791992, + "learning_rate": 2.591240875912409e-06, + "loss": 2.3983, + "step": 156 + }, + { + "epoch": 0.17205479452054795, + "grad_norm": 2.337080955505371, + "learning_rate": 2.6094890510948905e-06, + "loss": 2.3849, + "step": 157 + }, + { + "epoch": 0.17315068493150684, + "grad_norm": 3.2523117065429688, + "learning_rate": 2.627737226277373e-06, + "loss": 2.4171, + "step": 158 + }, + { + "epoch": 0.17424657534246576, + "grad_norm": 2.529175281524658, + "learning_rate": 2.6459854014598542e-06, + "loss": 2.4303, + "step": 159 + }, + { + "epoch": 0.17534246575342466, + "grad_norm": 3.052212953567505, + "learning_rate": 2.664233576642336e-06, + "loss": 2.3895, + "step": 160 + }, + { + "epoch": 0.17534246575342466, + "eval_loss": 2.492485761642456, + "eval_runtime": 45.6738, + "eval_samples_per_second": 21.894, + "eval_steps_per_second": 0.92, + "step": 160 + }, + { + "epoch": 0.17643835616438355, + "grad_norm": 2.1544973850250244, + "learning_rate": 2.6824817518248175e-06, + "loss": 2.3649, + "step": 161 + }, + { + "epoch": 0.17753424657534247, + "grad_norm": 2.7351951599121094, + "learning_rate": 2.7007299270072994e-06, + "loss": 2.3908, + "step": 162 + }, + { + "epoch": 0.17863013698630137, + "grad_norm": 2.8431951999664307, + "learning_rate": 2.718978102189781e-06, + "loss": 2.374, + "step": 163 + }, + { + "epoch": 0.17972602739726026, + "grad_norm": 3.083190441131592, + "learning_rate": 2.737226277372263e-06, + "loss": 2.4193, + "step": 164 + }, + { + "epoch": 0.18082191780821918, + "grad_norm": 2.6523995399475098, + "learning_rate": 2.755474452554745e-06, + "loss": 2.3969, + "step": 165 + }, + { + "epoch": 0.18191780821917808, + "grad_norm": 2.626047372817993, + "learning_rate": 2.7737226277372264e-06, + "loss": 2.3742, + "step": 166 + }, + { + "epoch": 0.18301369863013697, + "grad_norm": 3.300917863845825, + "learning_rate": 2.7919708029197082e-06, + "loss": 2.4546, + "step": 167 + }, + { + "epoch": 0.1841095890410959, + "grad_norm": 2.797304391860962, + "learning_rate": 2.81021897810219e-06, + "loss": 2.4161, + "step": 168 + }, + { + "epoch": 0.1852054794520548, + "grad_norm": 2.777117967605591, + "learning_rate": 2.828467153284672e-06, + "loss": 2.3572, + "step": 169 + }, + { + "epoch": 0.1863013698630137, + "grad_norm": 2.887470006942749, + "learning_rate": 2.8467153284671534e-06, + "loss": 2.3887, + "step": 170 + }, + { + "epoch": 0.1863013698630137, + "eval_loss": 2.4854257106781006, + "eval_runtime": 44.5644, + "eval_samples_per_second": 22.439, + "eval_steps_per_second": 0.942, + "step": 170 + }, + { + "epoch": 0.1873972602739726, + "grad_norm": 3.0706076622009277, + "learning_rate": 2.8649635036496353e-06, + "loss": 2.3162, + "step": 171 + }, + { + "epoch": 0.1884931506849315, + "grad_norm": 3.6129744052886963, + "learning_rate": 2.8832116788321167e-06, + "loss": 2.3685, + "step": 172 + }, + { + "epoch": 0.18958904109589042, + "grad_norm": 2.4936370849609375, + "learning_rate": 2.901459854014599e-06, + "loss": 2.3297, + "step": 173 + }, + { + "epoch": 0.19068493150684931, + "grad_norm": 2.1041860580444336, + "learning_rate": 2.9197080291970804e-06, + "loss": 2.4085, + "step": 174 + }, + { + "epoch": 0.1917808219178082, + "grad_norm": 2.2066831588745117, + "learning_rate": 2.9379562043795623e-06, + "loss": 2.3905, + "step": 175 + }, + { + "epoch": 0.19287671232876713, + "grad_norm": 2.444652795791626, + "learning_rate": 2.956204379562044e-06, + "loss": 2.378, + "step": 176 + }, + { + "epoch": 0.19397260273972602, + "grad_norm": 2.501939058303833, + "learning_rate": 2.9744525547445256e-06, + "loss": 2.4136, + "step": 177 + }, + { + "epoch": 0.19506849315068492, + "grad_norm": 2.2167556285858154, + "learning_rate": 2.992700729927008e-06, + "loss": 2.385, + "step": 178 + }, + { + "epoch": 0.19616438356164384, + "grad_norm": 2.5298545360565186, + "learning_rate": 3.0109489051094893e-06, + "loss": 2.3508, + "step": 179 + }, + { + "epoch": 0.19726027397260273, + "grad_norm": 2.3774642944335938, + "learning_rate": 3.029197080291971e-06, + "loss": 2.3421, + "step": 180 + }, + { + "epoch": 0.19726027397260273, + "eval_loss": 2.476067304611206, + "eval_runtime": 44.3761, + "eval_samples_per_second": 22.535, + "eval_steps_per_second": 0.946, + "step": 180 + }, + { + "epoch": 0.19835616438356166, + "grad_norm": 2.9108712673187256, + "learning_rate": 3.0474452554744526e-06, + "loss": 2.3983, + "step": 181 + }, + { + "epoch": 0.19945205479452055, + "grad_norm": 2.5232033729553223, + "learning_rate": 3.065693430656935e-06, + "loss": 2.3946, + "step": 182 + }, + { + "epoch": 0.20054794520547944, + "grad_norm": 2.158168077468872, + "learning_rate": 3.0839416058394163e-06, + "loss": 2.3805, + "step": 183 + }, + { + "epoch": 0.20164383561643837, + "grad_norm": 2.3405354022979736, + "learning_rate": 3.102189781021898e-06, + "loss": 2.3974, + "step": 184 + }, + { + "epoch": 0.20273972602739726, + "grad_norm": 2.5308327674865723, + "learning_rate": 3.1204379562043796e-06, + "loss": 2.384, + "step": 185 + }, + { + "epoch": 0.20383561643835615, + "grad_norm": 2.6328237056732178, + "learning_rate": 3.1386861313868614e-06, + "loss": 2.3345, + "step": 186 + }, + { + "epoch": 0.20493150684931508, + "grad_norm": 3.0725512504577637, + "learning_rate": 3.156934306569343e-06, + "loss": 2.3191, + "step": 187 + }, + { + "epoch": 0.20602739726027397, + "grad_norm": 2.196218729019165, + "learning_rate": 3.175182481751825e-06, + "loss": 2.4073, + "step": 188 + }, + { + "epoch": 0.20712328767123286, + "grad_norm": 2.7722432613372803, + "learning_rate": 3.193430656934307e-06, + "loss": 2.3775, + "step": 189 + }, + { + "epoch": 0.20821917808219179, + "grad_norm": 2.4848241806030273, + "learning_rate": 3.2116788321167884e-06, + "loss": 2.3538, + "step": 190 + }, + { + "epoch": 0.20821917808219179, + "eval_loss": 2.462618112564087, + "eval_runtime": 43.9461, + "eval_samples_per_second": 22.755, + "eval_steps_per_second": 0.956, + "step": 190 + }, + { + "epoch": 0.20931506849315068, + "grad_norm": 2.4682741165161133, + "learning_rate": 3.2299270072992707e-06, + "loss": 2.3688, + "step": 191 + }, + { + "epoch": 0.2104109589041096, + "grad_norm": 2.518923282623291, + "learning_rate": 3.248175182481752e-06, + "loss": 2.3556, + "step": 192 + }, + { + "epoch": 0.2115068493150685, + "grad_norm": 2.4023561477661133, + "learning_rate": 3.266423357664234e-06, + "loss": 2.292, + "step": 193 + }, + { + "epoch": 0.2126027397260274, + "grad_norm": 2.5196943283081055, + "learning_rate": 3.2846715328467155e-06, + "loss": 2.3703, + "step": 194 + }, + { + "epoch": 0.2136986301369863, + "grad_norm": 2.762709617614746, + "learning_rate": 3.3029197080291973e-06, + "loss": 2.3578, + "step": 195 + }, + { + "epoch": 0.2147945205479452, + "grad_norm": 2.1287691593170166, + "learning_rate": 3.3211678832116788e-06, + "loss": 2.3687, + "step": 196 + }, + { + "epoch": 0.2158904109589041, + "grad_norm": 2.477341651916504, + "learning_rate": 3.339416058394161e-06, + "loss": 2.3545, + "step": 197 + }, + { + "epoch": 0.21698630136986302, + "grad_norm": 2.23933744430542, + "learning_rate": 3.3576642335766425e-06, + "loss": 2.3511, + "step": 198 + }, + { + "epoch": 0.21808219178082192, + "grad_norm": 2.2440133094787598, + "learning_rate": 3.3759124087591243e-06, + "loss": 2.3539, + "step": 199 + }, + { + "epoch": 0.2191780821917808, + "grad_norm": 2.5603177547454834, + "learning_rate": 3.3941605839416058e-06, + "loss": 2.36, + "step": 200 + }, + { + "epoch": 0.2191780821917808, + "eval_loss": 2.4507246017456055, + "eval_runtime": 43.7786, + "eval_samples_per_second": 22.842, + "eval_steps_per_second": 0.959, + "step": 200 + }, + { + "epoch": 0.22027397260273973, + "grad_norm": 2.4570510387420654, + "learning_rate": 3.412408759124088e-06, + "loss": 2.3702, + "step": 201 + }, + { + "epoch": 0.22136986301369863, + "grad_norm": 2.1975133419036865, + "learning_rate": 3.43065693430657e-06, + "loss": 2.3667, + "step": 202 + }, + { + "epoch": 0.22246575342465755, + "grad_norm": 2.710186004638672, + "learning_rate": 3.4489051094890513e-06, + "loss": 2.3159, + "step": 203 + }, + { + "epoch": 0.22356164383561644, + "grad_norm": 2.7799506187438965, + "learning_rate": 3.467153284671533e-06, + "loss": 2.3519, + "step": 204 + }, + { + "epoch": 0.22465753424657534, + "grad_norm": 2.2479400634765625, + "learning_rate": 3.4854014598540146e-06, + "loss": 2.3103, + "step": 205 + }, + { + "epoch": 0.22575342465753426, + "grad_norm": 3.1538405418395996, + "learning_rate": 3.503649635036497e-06, + "loss": 2.3506, + "step": 206 + }, + { + "epoch": 0.22684931506849315, + "grad_norm": 2.7607762813568115, + "learning_rate": 3.5218978102189783e-06, + "loss": 2.3386, + "step": 207 + }, + { + "epoch": 0.22794520547945205, + "grad_norm": 2.371105909347534, + "learning_rate": 3.54014598540146e-06, + "loss": 2.3583, + "step": 208 + }, + { + "epoch": 0.22904109589041097, + "grad_norm": 2.793933391571045, + "learning_rate": 3.5583941605839416e-06, + "loss": 2.3782, + "step": 209 + }, + { + "epoch": 0.23013698630136986, + "grad_norm": 2.372732400894165, + "learning_rate": 3.576642335766424e-06, + "loss": 2.417, + "step": 210 + }, + { + "epoch": 0.23013698630136986, + "eval_loss": 2.451474189758301, + "eval_runtime": 41.7197, + "eval_samples_per_second": 23.97, + "eval_steps_per_second": 1.007, + "step": 210 + }, + { + "epoch": 0.23123287671232876, + "grad_norm": 2.6989448070526123, + "learning_rate": 3.5948905109489054e-06, + "loss": 2.393, + "step": 211 + }, + { + "epoch": 0.23232876712328768, + "grad_norm": 2.4324285984039307, + "learning_rate": 3.6131386861313872e-06, + "loss": 2.2738, + "step": 212 + }, + { + "epoch": 0.23342465753424657, + "grad_norm": 3.5063016414642334, + "learning_rate": 3.6313868613138687e-06, + "loss": 2.2823, + "step": 213 + }, + { + "epoch": 0.2345205479452055, + "grad_norm": 3.953951358795166, + "learning_rate": 3.6496350364963505e-06, + "loss": 2.3132, + "step": 214 + }, + { + "epoch": 0.2356164383561644, + "grad_norm": 2.3483071327209473, + "learning_rate": 3.667883211678833e-06, + "loss": 2.3244, + "step": 215 + }, + { + "epoch": 0.23671232876712328, + "grad_norm": 2.884199380874634, + "learning_rate": 3.6861313868613142e-06, + "loss": 2.3067, + "step": 216 + }, + { + "epoch": 0.2378082191780822, + "grad_norm": 2.9514520168304443, + "learning_rate": 3.704379562043796e-06, + "loss": 2.304, + "step": 217 + }, + { + "epoch": 0.2389041095890411, + "grad_norm": 2.7530996799468994, + "learning_rate": 3.7226277372262775e-06, + "loss": 2.3691, + "step": 218 + }, + { + "epoch": 0.24, + "grad_norm": 2.8753647804260254, + "learning_rate": 3.7408759124087594e-06, + "loss": 2.3406, + "step": 219 + }, + { + "epoch": 0.2410958904109589, + "grad_norm": 2.670574903488159, + "learning_rate": 3.7591240875912412e-06, + "loss": 2.304, + "step": 220 + }, + { + "epoch": 0.2410958904109589, + "eval_loss": 2.4395809173583984, + "eval_runtime": 46.6723, + "eval_samples_per_second": 21.426, + "eval_steps_per_second": 0.9, + "step": 220 + }, + { + "epoch": 0.2421917808219178, + "grad_norm": 2.4152438640594482, + "learning_rate": 3.777372262773723e-06, + "loss": 2.3294, + "step": 221 + }, + { + "epoch": 0.2432876712328767, + "grad_norm": 2.8398540019989014, + "learning_rate": 3.7956204379562045e-06, + "loss": 2.3216, + "step": 222 + }, + { + "epoch": 0.24438356164383562, + "grad_norm": 2.6036765575408936, + "learning_rate": 3.8138686131386864e-06, + "loss": 2.3109, + "step": 223 + }, + { + "epoch": 0.24547945205479452, + "grad_norm": 2.7073473930358887, + "learning_rate": 3.832116788321168e-06, + "loss": 2.3457, + "step": 224 + }, + { + "epoch": 0.2465753424657534, + "grad_norm": 2.584404230117798, + "learning_rate": 3.85036496350365e-06, + "loss": 2.3159, + "step": 225 + }, + { + "epoch": 0.24767123287671233, + "grad_norm": 2.4220666885375977, + "learning_rate": 3.868613138686132e-06, + "loss": 2.2944, + "step": 226 + }, + { + "epoch": 0.24876712328767123, + "grad_norm": 2.5636494159698486, + "learning_rate": 3.886861313868613e-06, + "loss": 2.3219, + "step": 227 + }, + { + "epoch": 0.24986301369863015, + "grad_norm": 2.2990903854370117, + "learning_rate": 3.905109489051096e-06, + "loss": 2.3342, + "step": 228 + }, + { + "epoch": 0.25095890410958904, + "grad_norm": 2.50913143157959, + "learning_rate": 3.923357664233577e-06, + "loss": 2.3619, + "step": 229 + }, + { + "epoch": 0.25205479452054796, + "grad_norm": 2.1687769889831543, + "learning_rate": 3.9416058394160585e-06, + "loss": 2.3153, + "step": 230 + }, + { + "epoch": 0.25205479452054796, + "eval_loss": 2.4315240383148193, + "eval_runtime": 48.3542, + "eval_samples_per_second": 20.681, + "eval_steps_per_second": 0.869, + "step": 230 + }, + { + "epoch": 0.25315068493150683, + "grad_norm": 2.842125177383423, + "learning_rate": 3.95985401459854e-06, + "loss": 2.3644, + "step": 231 + }, + { + "epoch": 0.25424657534246575, + "grad_norm": 2.258049964904785, + "learning_rate": 3.978102189781022e-06, + "loss": 2.3727, + "step": 232 + }, + { + "epoch": 0.2553424657534247, + "grad_norm": 2.4630489349365234, + "learning_rate": 3.996350364963504e-06, + "loss": 2.267, + "step": 233 + }, + { + "epoch": 0.25643835616438354, + "grad_norm": 3.1469743251800537, + "learning_rate": 4.014598540145986e-06, + "loss": 2.3432, + "step": 234 + }, + { + "epoch": 0.25753424657534246, + "grad_norm": 2.6277477741241455, + "learning_rate": 4.032846715328467e-06, + "loss": 2.2545, + "step": 235 + }, + { + "epoch": 0.2586301369863014, + "grad_norm": 2.404141902923584, + "learning_rate": 4.05109489051095e-06, + "loss": 2.3109, + "step": 236 + }, + { + "epoch": 0.25972602739726025, + "grad_norm": 2.4545605182647705, + "learning_rate": 4.069343065693431e-06, + "loss": 2.3112, + "step": 237 + }, + { + "epoch": 0.2608219178082192, + "grad_norm": 2.237638235092163, + "learning_rate": 4.0875912408759126e-06, + "loss": 2.3, + "step": 238 + }, + { + "epoch": 0.2619178082191781, + "grad_norm": 2.192811965942383, + "learning_rate": 4.105839416058395e-06, + "loss": 2.3108, + "step": 239 + }, + { + "epoch": 0.26301369863013696, + "grad_norm": 2.5058720111846924, + "learning_rate": 4.124087591240876e-06, + "loss": 2.3635, + "step": 240 + }, + { + "epoch": 0.26301369863013696, + "eval_loss": 2.4219377040863037, + "eval_runtime": 45.3561, + "eval_samples_per_second": 22.048, + "eval_steps_per_second": 0.926, + "step": 240 + }, + { + "epoch": 0.2641095890410959, + "grad_norm": 2.0708935260772705, + "learning_rate": 4.1423357664233586e-06, + "loss": 2.3129, + "step": 241 + }, + { + "epoch": 0.2652054794520548, + "grad_norm": 2.6353116035461426, + "learning_rate": 4.16058394160584e-06, + "loss": 2.3011, + "step": 242 + }, + { + "epoch": 0.2663013698630137, + "grad_norm": 2.8392436504364014, + "learning_rate": 4.1788321167883214e-06, + "loss": 2.3643, + "step": 243 + }, + { + "epoch": 0.2673972602739726, + "grad_norm": 3.1773295402526855, + "learning_rate": 4.197080291970803e-06, + "loss": 2.2642, + "step": 244 + }, + { + "epoch": 0.2684931506849315, + "grad_norm": 2.4757354259490967, + "learning_rate": 4.215328467153285e-06, + "loss": 2.3211, + "step": 245 + }, + { + "epoch": 0.26958904109589044, + "grad_norm": 2.5799412727355957, + "learning_rate": 4.233576642335767e-06, + "loss": 2.2909, + "step": 246 + }, + { + "epoch": 0.2706849315068493, + "grad_norm": 2.368502378463745, + "learning_rate": 4.251824817518249e-06, + "loss": 2.2951, + "step": 247 + }, + { + "epoch": 0.2717808219178082, + "grad_norm": 2.114969253540039, + "learning_rate": 4.27007299270073e-06, + "loss": 2.3124, + "step": 248 + }, + { + "epoch": 0.27287671232876715, + "grad_norm": 3.056820869445801, + "learning_rate": 4.288321167883212e-06, + "loss": 2.2923, + "step": 249 + }, + { + "epoch": 0.273972602739726, + "grad_norm": 2.7752203941345215, + "learning_rate": 4.306569343065693e-06, + "loss": 2.2874, + "step": 250 + }, + { + "epoch": 0.273972602739726, + "eval_loss": 2.4229912757873535, + "eval_runtime": 52.5941, + "eval_samples_per_second": 19.014, + "eval_steps_per_second": 0.799, + "step": 250 + }, + { + "epoch": 0.27506849315068493, + "grad_norm": 3.045740842819214, + "learning_rate": 4.3248175182481755e-06, + "loss": 2.2686, + "step": 251 + }, + { + "epoch": 0.27616438356164386, + "grad_norm": 2.646402359008789, + "learning_rate": 4.343065693430658e-06, + "loss": 2.3247, + "step": 252 + }, + { + "epoch": 0.2772602739726027, + "grad_norm": 3.274994134902954, + "learning_rate": 4.361313868613139e-06, + "loss": 2.3093, + "step": 253 + }, + { + "epoch": 0.27835616438356164, + "grad_norm": 2.352246046066284, + "learning_rate": 4.379562043795621e-06, + "loss": 2.3124, + "step": 254 + }, + { + "epoch": 0.27945205479452057, + "grad_norm": 3.248717784881592, + "learning_rate": 4.397810218978102e-06, + "loss": 2.3417, + "step": 255 + }, + { + "epoch": 0.28054794520547943, + "grad_norm": 3.0919013023376465, + "learning_rate": 4.416058394160584e-06, + "loss": 2.3093, + "step": 256 + }, + { + "epoch": 0.28164383561643835, + "grad_norm": 2.1174702644348145, + "learning_rate": 4.434306569343066e-06, + "loss": 2.305, + "step": 257 + }, + { + "epoch": 0.2827397260273973, + "grad_norm": 2.913797378540039, + "learning_rate": 4.452554744525548e-06, + "loss": 2.2831, + "step": 258 + }, + { + "epoch": 0.28383561643835614, + "grad_norm": 2.693089723587036, + "learning_rate": 4.4708029197080295e-06, + "loss": 2.3206, + "step": 259 + }, + { + "epoch": 0.28493150684931506, + "grad_norm": 2.521411895751953, + "learning_rate": 4.489051094890512e-06, + "loss": 2.3196, + "step": 260 + }, + { + "epoch": 0.28493150684931506, + "eval_loss": 2.4178109169006348, + "eval_runtime": 50.6753, + "eval_samples_per_second": 19.733, + "eval_steps_per_second": 0.829, + "step": 260 + }, + { + "epoch": 0.286027397260274, + "grad_norm": 2.346346855163574, + "learning_rate": 4.507299270072993e-06, + "loss": 2.3433, + "step": 261 + }, + { + "epoch": 0.28712328767123285, + "grad_norm": 2.520368814468384, + "learning_rate": 4.525547445255475e-06, + "loss": 2.3174, + "step": 262 + }, + { + "epoch": 0.2882191780821918, + "grad_norm": 2.550347328186035, + "learning_rate": 4.543795620437956e-06, + "loss": 2.3072, + "step": 263 + }, + { + "epoch": 0.2893150684931507, + "grad_norm": 2.268662691116333, + "learning_rate": 4.562043795620438e-06, + "loss": 2.2763, + "step": 264 + }, + { + "epoch": 0.29041095890410956, + "grad_norm": 2.445798873901367, + "learning_rate": 4.580291970802921e-06, + "loss": 2.3395, + "step": 265 + }, + { + "epoch": 0.2915068493150685, + "grad_norm": 2.5391135215759277, + "learning_rate": 4.598540145985402e-06, + "loss": 2.3496, + "step": 266 + }, + { + "epoch": 0.2926027397260274, + "grad_norm": 2.3553097248077393, + "learning_rate": 4.6167883211678835e-06, + "loss": 2.3243, + "step": 267 + }, + { + "epoch": 0.2936986301369863, + "grad_norm": 2.448089599609375, + "learning_rate": 4.635036496350365e-06, + "loss": 2.2887, + "step": 268 + }, + { + "epoch": 0.2947945205479452, + "grad_norm": 2.1600170135498047, + "learning_rate": 4.653284671532847e-06, + "loss": 2.2816, + "step": 269 + }, + { + "epoch": 0.2958904109589041, + "grad_norm": 2.3783180713653564, + "learning_rate": 4.671532846715329e-06, + "loss": 2.3275, + "step": 270 + }, + { + "epoch": 0.2958904109589041, + "eval_loss": 2.413832902908325, + "eval_runtime": 48.7524, + "eval_samples_per_second": 20.512, + "eval_steps_per_second": 0.861, + "step": 270 + }, + { + "epoch": 0.29698630136986304, + "grad_norm": 2.468320369720459, + "learning_rate": 4.689781021897811e-06, + "loss": 2.3358, + "step": 271 + }, + { + "epoch": 0.2980821917808219, + "grad_norm": 2.401634454727173, + "learning_rate": 4.708029197080292e-06, + "loss": 2.2991, + "step": 272 + }, + { + "epoch": 0.2991780821917808, + "grad_norm": 2.4604203701019287, + "learning_rate": 4.726277372262774e-06, + "loss": 2.3496, + "step": 273 + }, + { + "epoch": 0.30027397260273975, + "grad_norm": 2.0338871479034424, + "learning_rate": 4.744525547445255e-06, + "loss": 2.3115, + "step": 274 + }, + { + "epoch": 0.3013698630136986, + "grad_norm": 2.5404436588287354, + "learning_rate": 4.7627737226277375e-06, + "loss": 2.2665, + "step": 275 + }, + { + "epoch": 0.30246575342465754, + "grad_norm": 2.5369458198547363, + "learning_rate": 4.78102189781022e-06, + "loss": 2.3072, + "step": 276 + }, + { + "epoch": 0.30356164383561646, + "grad_norm": 2.471522569656372, + "learning_rate": 4.799270072992701e-06, + "loss": 2.3113, + "step": 277 + }, + { + "epoch": 0.3046575342465753, + "grad_norm": 2.3890135288238525, + "learning_rate": 4.8175182481751835e-06, + "loss": 2.2838, + "step": 278 + }, + { + "epoch": 0.30575342465753425, + "grad_norm": 2.2485053539276123, + "learning_rate": 4.835766423357665e-06, + "loss": 2.3052, + "step": 279 + }, + { + "epoch": 0.30684931506849317, + "grad_norm": 2.1090753078460693, + "learning_rate": 4.854014598540146e-06, + "loss": 2.3054, + "step": 280 + }, + { + "epoch": 0.30684931506849317, + "eval_loss": 2.412548780441284, + "eval_runtime": 48.5034, + "eval_samples_per_second": 20.617, + "eval_steps_per_second": 0.866, + "step": 280 + }, + { + "epoch": 0.30794520547945203, + "grad_norm": 2.299133777618408, + "learning_rate": 4.872262773722628e-06, + "loss": 2.2599, + "step": 281 + }, + { + "epoch": 0.30904109589041096, + "grad_norm": 2.177905797958374, + "learning_rate": 4.89051094890511e-06, + "loss": 2.2745, + "step": 282 + }, + { + "epoch": 0.3101369863013699, + "grad_norm": 2.3178462982177734, + "learning_rate": 4.9087591240875915e-06, + "loss": 2.2562, + "step": 283 + }, + { + "epoch": 0.31123287671232874, + "grad_norm": 2.5088376998901367, + "learning_rate": 4.927007299270074e-06, + "loss": 2.2945, + "step": 284 + }, + { + "epoch": 0.31232876712328766, + "grad_norm": 2.095379590988159, + "learning_rate": 4.945255474452555e-06, + "loss": 2.3038, + "step": 285 + }, + { + "epoch": 0.3134246575342466, + "grad_norm": 2.1792995929718018, + "learning_rate": 4.963503649635037e-06, + "loss": 2.2702, + "step": 286 + }, + { + "epoch": 0.31452054794520545, + "grad_norm": 2.3001186847686768, + "learning_rate": 4.981751824817518e-06, + "loss": 2.3482, + "step": 287 + }, + { + "epoch": 0.3156164383561644, + "grad_norm": 2.1576106548309326, + "learning_rate": 5e-06, + "loss": 2.3391, + "step": 288 + }, + { + "epoch": 0.3167123287671233, + "grad_norm": 2.3715577125549316, + "learning_rate": 4.997969130787978e-06, + "loss": 2.3049, + "step": 289 + }, + { + "epoch": 0.3178082191780822, + "grad_norm": 2.0316715240478516, + "learning_rate": 4.995938261575955e-06, + "loss": 2.3111, + "step": 290 + }, + { + "epoch": 0.3178082191780822, + "eval_loss": 2.40299654006958, + "eval_runtime": 50.0873, + "eval_samples_per_second": 19.965, + "eval_steps_per_second": 0.839, + "step": 290 + }, + { + "epoch": 0.3189041095890411, + "grad_norm": 2.1135756969451904, + "learning_rate": 4.993907392363932e-06, + "loss": 2.2465, + "step": 291 + }, + { + "epoch": 0.32, + "grad_norm": 2.335655927658081, + "learning_rate": 4.9918765231519095e-06, + "loss": 2.31, + "step": 292 + }, + { + "epoch": 0.32109589041095893, + "grad_norm": 2.3255794048309326, + "learning_rate": 4.989845653939887e-06, + "loss": 2.2531, + "step": 293 + }, + { + "epoch": 0.3221917808219178, + "grad_norm": 2.111055612564087, + "learning_rate": 4.987814784727864e-06, + "loss": 2.2799, + "step": 294 + }, + { + "epoch": 0.3232876712328767, + "grad_norm": 2.1773171424865723, + "learning_rate": 4.985783915515841e-06, + "loss": 2.2884, + "step": 295 + }, + { + "epoch": 0.32438356164383564, + "grad_norm": 2.3559885025024414, + "learning_rate": 4.983753046303819e-06, + "loss": 2.2711, + "step": 296 + }, + { + "epoch": 0.3254794520547945, + "grad_norm": 2.120802879333496, + "learning_rate": 4.981722177091796e-06, + "loss": 2.2946, + "step": 297 + }, + { + "epoch": 0.3265753424657534, + "grad_norm": 2.2319388389587402, + "learning_rate": 4.979691307879773e-06, + "loss": 2.2771, + "step": 298 + }, + { + "epoch": 0.32767123287671235, + "grad_norm": 2.208740711212158, + "learning_rate": 4.9776604386677504e-06, + "loss": 2.3003, + "step": 299 + }, + { + "epoch": 0.3287671232876712, + "grad_norm": 2.543060541152954, + "learning_rate": 4.975629569455728e-06, + "loss": 2.2891, + "step": 300 + }, + { + "epoch": 0.3287671232876712, + "eval_loss": 2.398495674133301, + "eval_runtime": 51.1973, + "eval_samples_per_second": 19.532, + "eval_steps_per_second": 0.82, + "step": 300 + }, + { + "epoch": 0.32986301369863014, + "grad_norm": 2.086219072341919, + "learning_rate": 4.973598700243705e-06, + "loss": 2.3286, + "step": 301 + }, + { + "epoch": 0.33095890410958906, + "grad_norm": 2.3673319816589355, + "learning_rate": 4.971567831031681e-06, + "loss": 2.2672, + "step": 302 + }, + { + "epoch": 0.3320547945205479, + "grad_norm": 2.167593240737915, + "learning_rate": 4.9695369618196595e-06, + "loss": 2.264, + "step": 303 + }, + { + "epoch": 0.33315068493150685, + "grad_norm": 2.7218785285949707, + "learning_rate": 4.967506092607636e-06, + "loss": 2.1737, + "step": 304 + }, + { + "epoch": 0.33424657534246577, + "grad_norm": 2.673245429992676, + "learning_rate": 4.965475223395614e-06, + "loss": 2.2439, + "step": 305 + }, + { + "epoch": 0.33534246575342463, + "grad_norm": 2.0341989994049072, + "learning_rate": 4.9634443541835905e-06, + "loss": 2.2614, + "step": 306 + }, + { + "epoch": 0.33643835616438356, + "grad_norm": 1.9314247369766235, + "learning_rate": 4.961413484971569e-06, + "loss": 2.2592, + "step": 307 + }, + { + "epoch": 0.3375342465753425, + "grad_norm": 2.537794589996338, + "learning_rate": 4.959382615759545e-06, + "loss": 2.2859, + "step": 308 + }, + { + "epoch": 0.33863013698630134, + "grad_norm": 2.0811166763305664, + "learning_rate": 4.957351746547523e-06, + "loss": 2.2868, + "step": 309 + }, + { + "epoch": 0.33972602739726027, + "grad_norm": 2.783701181411743, + "learning_rate": 4.9553208773355e-06, + "loss": 2.2649, + "step": 310 + }, + { + "epoch": 0.33972602739726027, + "eval_loss": 2.400538921356201, + "eval_runtime": 43.185, + "eval_samples_per_second": 23.156, + "eval_steps_per_second": 0.973, + "step": 310 + }, + { + "epoch": 0.3408219178082192, + "grad_norm": 2.8541319370269775, + "learning_rate": 4.953290008123478e-06, + "loss": 2.3729, + "step": 311 + }, + { + "epoch": 0.3419178082191781, + "grad_norm": 2.1852617263793945, + "learning_rate": 4.951259138911454e-06, + "loss": 2.2885, + "step": 312 + }, + { + "epoch": 0.343013698630137, + "grad_norm": 2.872201919555664, + "learning_rate": 4.9492282696994315e-06, + "loss": 2.2353, + "step": 313 + }, + { + "epoch": 0.3441095890410959, + "grad_norm": 2.5140910148620605, + "learning_rate": 4.947197400487409e-06, + "loss": 2.3364, + "step": 314 + }, + { + "epoch": 0.3452054794520548, + "grad_norm": 2.619974374771118, + "learning_rate": 4.945166531275386e-06, + "loss": 2.2721, + "step": 315 + }, + { + "epoch": 0.3463013698630137, + "grad_norm": 2.4567618370056152, + "learning_rate": 4.943135662063363e-06, + "loss": 2.26, + "step": 316 + }, + { + "epoch": 0.3473972602739726, + "grad_norm": 2.9603192806243896, + "learning_rate": 4.9411047928513406e-06, + "loss": 2.2861, + "step": 317 + }, + { + "epoch": 0.34849315068493153, + "grad_norm": 2.115262985229492, + "learning_rate": 4.939073923639318e-06, + "loss": 2.3125, + "step": 318 + }, + { + "epoch": 0.3495890410958904, + "grad_norm": 2.693024158477783, + "learning_rate": 4.937043054427295e-06, + "loss": 2.3054, + "step": 319 + }, + { + "epoch": 0.3506849315068493, + "grad_norm": 2.3439996242523193, + "learning_rate": 4.935012185215272e-06, + "loss": 2.2576, + "step": 320 + }, + { + "epoch": 0.3506849315068493, + "eval_loss": 2.3920650482177734, + "eval_runtime": 47.7451, + "eval_samples_per_second": 20.945, + "eval_steps_per_second": 0.88, + "step": 320 + }, + { + "epoch": 0.35178082191780824, + "grad_norm": 2.3009634017944336, + "learning_rate": 4.93298131600325e-06, + "loss": 2.2454, + "step": 321 + }, + { + "epoch": 0.3528767123287671, + "grad_norm": 2.593203544616699, + "learning_rate": 4.930950446791227e-06, + "loss": 2.2965, + "step": 322 + }, + { + "epoch": 0.353972602739726, + "grad_norm": 2.9069650173187256, + "learning_rate": 4.928919577579204e-06, + "loss": 2.2785, + "step": 323 + }, + { + "epoch": 0.35506849315068495, + "grad_norm": 2.1021759510040283, + "learning_rate": 4.9268887083671815e-06, + "loss": 2.2857, + "step": 324 + }, + { + "epoch": 0.3561643835616438, + "grad_norm": 2.0330185890197754, + "learning_rate": 4.924857839155159e-06, + "loss": 2.2697, + "step": 325 + }, + { + "epoch": 0.35726027397260274, + "grad_norm": 2.1318063735961914, + "learning_rate": 4.922826969943136e-06, + "loss": 2.2698, + "step": 326 + }, + { + "epoch": 0.35835616438356166, + "grad_norm": 2.0157387256622314, + "learning_rate": 4.920796100731113e-06, + "loss": 2.292, + "step": 327 + }, + { + "epoch": 0.3594520547945205, + "grad_norm": 2.3266026973724365, + "learning_rate": 4.918765231519091e-06, + "loss": 2.2215, + "step": 328 + }, + { + "epoch": 0.36054794520547945, + "grad_norm": 2.066305160522461, + "learning_rate": 4.916734362307068e-06, + "loss": 2.2758, + "step": 329 + }, + { + "epoch": 0.36164383561643837, + "grad_norm": 2.0219993591308594, + "learning_rate": 4.914703493095045e-06, + "loss": 2.2381, + "step": 330 + }, + { + "epoch": 0.36164383561643837, + "eval_loss": 2.394064426422119, + "eval_runtime": 51.5482, + "eval_samples_per_second": 19.399, + "eval_steps_per_second": 0.815, + "step": 330 + }, + { + "epoch": 0.36273972602739724, + "grad_norm": 2.3964319229125977, + "learning_rate": 4.912672623883022e-06, + "loss": 2.2384, + "step": 331 + }, + { + "epoch": 0.36383561643835616, + "grad_norm": 2.5466530323028564, + "learning_rate": 4.910641754671e-06, + "loss": 2.2428, + "step": 332 + }, + { + "epoch": 0.3649315068493151, + "grad_norm": 2.236152410507202, + "learning_rate": 4.908610885458977e-06, + "loss": 2.2639, + "step": 333 + }, + { + "epoch": 0.36602739726027395, + "grad_norm": 2.7267303466796875, + "learning_rate": 4.906580016246954e-06, + "loss": 2.2608, + "step": 334 + }, + { + "epoch": 0.36712328767123287, + "grad_norm": 1.9403765201568604, + "learning_rate": 4.9045491470349315e-06, + "loss": 2.2962, + "step": 335 + }, + { + "epoch": 0.3682191780821918, + "grad_norm": 2.89070463180542, + "learning_rate": 4.902518277822909e-06, + "loss": 2.2847, + "step": 336 + }, + { + "epoch": 0.3693150684931507, + "grad_norm": 2.6427228450775146, + "learning_rate": 4.900487408610886e-06, + "loss": 2.2734, + "step": 337 + }, + { + "epoch": 0.3704109589041096, + "grad_norm": 2.1680362224578857, + "learning_rate": 4.898456539398863e-06, + "loss": 2.2869, + "step": 338 + }, + { + "epoch": 0.3715068493150685, + "grad_norm": 3.0160160064697266, + "learning_rate": 4.89642567018684e-06, + "loss": 2.3043, + "step": 339 + }, + { + "epoch": 0.3726027397260274, + "grad_norm": 1.9858418703079224, + "learning_rate": 4.894394800974818e-06, + "loss": 2.3422, + "step": 340 + }, + { + "epoch": 0.3726027397260274, + "eval_loss": 2.387559652328491, + "eval_runtime": 52.2455, + "eval_samples_per_second": 19.14, + "eval_steps_per_second": 0.804, + "step": 340 + }, + { + "epoch": 0.3736986301369863, + "grad_norm": 2.243699073791504, + "learning_rate": 4.892363931762794e-06, + "loss": 2.2394, + "step": 341 + }, + { + "epoch": 0.3747945205479452, + "grad_norm": 2.371299982070923, + "learning_rate": 4.8903330625507725e-06, + "loss": 2.2856, + "step": 342 + }, + { + "epoch": 0.37589041095890413, + "grad_norm": 2.3769736289978027, + "learning_rate": 4.888302193338749e-06, + "loss": 2.246, + "step": 343 + }, + { + "epoch": 0.376986301369863, + "grad_norm": 2.3209030628204346, + "learning_rate": 4.886271324126727e-06, + "loss": 2.2331, + "step": 344 + }, + { + "epoch": 0.3780821917808219, + "grad_norm": 2.747319221496582, + "learning_rate": 4.8842404549147034e-06, + "loss": 2.3115, + "step": 345 + }, + { + "epoch": 0.37917808219178084, + "grad_norm": 1.8599231243133545, + "learning_rate": 4.8822095857026816e-06, + "loss": 2.2994, + "step": 346 + }, + { + "epoch": 0.3802739726027397, + "grad_norm": 2.0358943939208984, + "learning_rate": 4.880178716490658e-06, + "loss": 2.271, + "step": 347 + }, + { + "epoch": 0.38136986301369863, + "grad_norm": 2.409106731414795, + "learning_rate": 4.878147847278636e-06, + "loss": 2.2281, + "step": 348 + }, + { + "epoch": 0.38246575342465755, + "grad_norm": 2.1641807556152344, + "learning_rate": 4.8761169780666125e-06, + "loss": 2.2437, + "step": 349 + }, + { + "epoch": 0.3835616438356164, + "grad_norm": 2.6917359828948975, + "learning_rate": 4.874086108854591e-06, + "loss": 2.226, + "step": 350 + }, + { + "epoch": 0.3835616438356164, + "eval_loss": 2.382991313934326, + "eval_runtime": 48.9534, + "eval_samples_per_second": 20.428, + "eval_steps_per_second": 0.858, + "step": 350 + }, + { + "epoch": 0.38465753424657534, + "grad_norm": 2.364215612411499, + "learning_rate": 4.872055239642567e-06, + "loss": 2.2319, + "step": 351 + }, + { + "epoch": 0.38575342465753426, + "grad_norm": 2.317643165588379, + "learning_rate": 4.870024370430544e-06, + "loss": 2.2096, + "step": 352 + }, + { + "epoch": 0.3868493150684931, + "grad_norm": 2.2283246517181396, + "learning_rate": 4.867993501218522e-06, + "loss": 2.2634, + "step": 353 + }, + { + "epoch": 0.38794520547945205, + "grad_norm": 2.498391628265381, + "learning_rate": 4.865962632006499e-06, + "loss": 2.2813, + "step": 354 + }, + { + "epoch": 0.38904109589041097, + "grad_norm": 2.3889403343200684, + "learning_rate": 4.863931762794476e-06, + "loss": 2.2361, + "step": 355 + }, + { + "epoch": 0.39013698630136984, + "grad_norm": 2.2049975395202637, + "learning_rate": 4.8619008935824535e-06, + "loss": 2.2583, + "step": 356 + }, + { + "epoch": 0.39123287671232876, + "grad_norm": 2.132840871810913, + "learning_rate": 4.859870024370431e-06, + "loss": 2.2823, + "step": 357 + }, + { + "epoch": 0.3923287671232877, + "grad_norm": 2.2692184448242188, + "learning_rate": 4.857839155158408e-06, + "loss": 2.2939, + "step": 358 + }, + { + "epoch": 0.3934246575342466, + "grad_norm": 2.0501933097839355, + "learning_rate": 4.855808285946385e-06, + "loss": 2.2976, + "step": 359 + }, + { + "epoch": 0.39452054794520547, + "grad_norm": 2.170236587524414, + "learning_rate": 4.8537774167343626e-06, + "loss": 2.2748, + "step": 360 + }, + { + "epoch": 0.39452054794520547, + "eval_loss": 2.3783886432647705, + "eval_runtime": 46.7324, + "eval_samples_per_second": 21.398, + "eval_steps_per_second": 0.899, + "step": 360 + }, + { + "epoch": 0.3956164383561644, + "grad_norm": 2.283137083053589, + "learning_rate": 4.85174654752234e-06, + "loss": 2.2705, + "step": 361 + }, + { + "epoch": 0.3967123287671233, + "grad_norm": 2.195984363555908, + "learning_rate": 4.849715678310317e-06, + "loss": 2.26, + "step": 362 + }, + { + "epoch": 0.3978082191780822, + "grad_norm": 2.399935007095337, + "learning_rate": 4.847684809098294e-06, + "loss": 2.2639, + "step": 363 + }, + { + "epoch": 0.3989041095890411, + "grad_norm": 2.411280393600464, + "learning_rate": 4.845653939886272e-06, + "loss": 2.2523, + "step": 364 + }, + { + "epoch": 0.4, + "grad_norm": 2.2033896446228027, + "learning_rate": 4.843623070674249e-06, + "loss": 2.2722, + "step": 365 + }, + { + "epoch": 0.4010958904109589, + "grad_norm": 2.2990238666534424, + "learning_rate": 4.841592201462226e-06, + "loss": 2.2411, + "step": 366 + }, + { + "epoch": 0.4021917808219178, + "grad_norm": 1.911453127861023, + "learning_rate": 4.8395613322502035e-06, + "loss": 2.2921, + "step": 367 + }, + { + "epoch": 0.40328767123287673, + "grad_norm": 2.453195095062256, + "learning_rate": 4.837530463038181e-06, + "loss": 2.218, + "step": 368 + }, + { + "epoch": 0.4043835616438356, + "grad_norm": 2.126992702484131, + "learning_rate": 4.835499593826158e-06, + "loss": 2.2628, + "step": 369 + }, + { + "epoch": 0.4054794520547945, + "grad_norm": 1.9305835962295532, + "learning_rate": 4.833468724614135e-06, + "loss": 2.2726, + "step": 370 + }, + { + "epoch": 0.4054794520547945, + "eval_loss": 2.3741114139556885, + "eval_runtime": 49.2956, + "eval_samples_per_second": 20.286, + "eval_steps_per_second": 0.852, + "step": 370 + }, + { + "epoch": 0.40657534246575344, + "grad_norm": 2.222024917602539, + "learning_rate": 4.831437855402113e-06, + "loss": 2.2634, + "step": 371 + }, + { + "epoch": 0.4076712328767123, + "grad_norm": 2.473958969116211, + "learning_rate": 4.82940698619009e-06, + "loss": 2.1817, + "step": 372 + }, + { + "epoch": 0.40876712328767123, + "grad_norm": 1.9208838939666748, + "learning_rate": 4.827376116978067e-06, + "loss": 2.2353, + "step": 373 + }, + { + "epoch": 0.40986301369863015, + "grad_norm": 2.7226450443267822, + "learning_rate": 4.8253452477660444e-06, + "loss": 2.2831, + "step": 374 + }, + { + "epoch": 0.410958904109589, + "grad_norm": 1.9078222513198853, + "learning_rate": 4.823314378554022e-06, + "loss": 2.264, + "step": 375 + }, + { + "epoch": 0.41205479452054794, + "grad_norm": 2.3946311473846436, + "learning_rate": 4.821283509341998e-06, + "loss": 2.251, + "step": 376 + }, + { + "epoch": 0.41315068493150686, + "grad_norm": 1.847343921661377, + "learning_rate": 4.819252640129976e-06, + "loss": 2.2441, + "step": 377 + }, + { + "epoch": 0.41424657534246573, + "grad_norm": 2.475385904312134, + "learning_rate": 4.817221770917953e-06, + "loss": 2.2639, + "step": 378 + }, + { + "epoch": 0.41534246575342465, + "grad_norm": 2.1396560668945312, + "learning_rate": 4.815190901705931e-06, + "loss": 2.2353, + "step": 379 + }, + { + "epoch": 0.41643835616438357, + "grad_norm": 2.3868885040283203, + "learning_rate": 4.813160032493907e-06, + "loss": 2.2551, + "step": 380 + }, + { + "epoch": 0.41643835616438357, + "eval_loss": 2.369168519973755, + "eval_runtime": 45.6131, + "eval_samples_per_second": 21.924, + "eval_steps_per_second": 0.921, + "step": 380 + }, + { + "epoch": 0.41753424657534244, + "grad_norm": 2.0193872451782227, + "learning_rate": 4.811129163281885e-06, + "loss": 2.2399, + "step": 381 + }, + { + "epoch": 0.41863013698630136, + "grad_norm": 2.4655935764312744, + "learning_rate": 4.809098294069862e-06, + "loss": 2.2729, + "step": 382 + }, + { + "epoch": 0.4197260273972603, + "grad_norm": 1.880974292755127, + "learning_rate": 4.80706742485784e-06, + "loss": 2.2681, + "step": 383 + }, + { + "epoch": 0.4208219178082192, + "grad_norm": 2.1912078857421875, + "learning_rate": 4.805036555645816e-06, + "loss": 2.2961, + "step": 384 + }, + { + "epoch": 0.42191780821917807, + "grad_norm": 1.985931634902954, + "learning_rate": 4.8030056864337945e-06, + "loss": 2.2407, + "step": 385 + }, + { + "epoch": 0.423013698630137, + "grad_norm": 2.4353220462799072, + "learning_rate": 4.800974817221771e-06, + "loss": 2.2867, + "step": 386 + }, + { + "epoch": 0.4241095890410959, + "grad_norm": 1.8810508251190186, + "learning_rate": 4.798943948009749e-06, + "loss": 2.2498, + "step": 387 + }, + { + "epoch": 0.4252054794520548, + "grad_norm": 2.195033073425293, + "learning_rate": 4.7969130787977254e-06, + "loss": 2.2532, + "step": 388 + }, + { + "epoch": 0.4263013698630137, + "grad_norm": 1.855372428894043, + "learning_rate": 4.794882209585703e-06, + "loss": 2.2538, + "step": 389 + }, + { + "epoch": 0.4273972602739726, + "grad_norm": 2.4465434551239014, + "learning_rate": 4.79285134037368e-06, + "loss": 2.2724, + "step": 390 + }, + { + "epoch": 0.4273972602739726, + "eval_loss": 2.364928960800171, + "eval_runtime": 48.0245, + "eval_samples_per_second": 20.823, + "eval_steps_per_second": 0.875, + "step": 390 + }, + { + "epoch": 0.4284931506849315, + "grad_norm": 1.8911314010620117, + "learning_rate": 4.790820471161657e-06, + "loss": 2.2291, + "step": 391 + }, + { + "epoch": 0.4295890410958904, + "grad_norm": 2.1975977420806885, + "learning_rate": 4.7887896019496345e-06, + "loss": 2.2373, + "step": 392 + }, + { + "epoch": 0.43068493150684933, + "grad_norm": 2.015331506729126, + "learning_rate": 4.786758732737612e-06, + "loss": 2.2483, + "step": 393 + }, + { + "epoch": 0.4317808219178082, + "grad_norm": 2.039339065551758, + "learning_rate": 4.784727863525589e-06, + "loss": 2.2746, + "step": 394 + }, + { + "epoch": 0.4328767123287671, + "grad_norm": 1.977513074874878, + "learning_rate": 4.782696994313566e-06, + "loss": 2.2275, + "step": 395 + }, + { + "epoch": 0.43397260273972604, + "grad_norm": 2.147217035293579, + "learning_rate": 4.780666125101544e-06, + "loss": 2.252, + "step": 396 + }, + { + "epoch": 0.4350684931506849, + "grad_norm": 2.1799919605255127, + "learning_rate": 4.778635255889521e-06, + "loss": 2.1975, + "step": 397 + }, + { + "epoch": 0.43616438356164383, + "grad_norm": 2.582240343093872, + "learning_rate": 4.776604386677498e-06, + "loss": 2.2294, + "step": 398 + }, + { + "epoch": 0.43726027397260275, + "grad_norm": 1.9127846956253052, + "learning_rate": 4.7745735174654755e-06, + "loss": 2.2319, + "step": 399 + }, + { + "epoch": 0.4383561643835616, + "grad_norm": 2.2444543838500977, + "learning_rate": 4.772542648253453e-06, + "loss": 2.2857, + "step": 400 + }, + { + "epoch": 0.4383561643835616, + "eval_loss": 2.3653724193573, + "eval_runtime": 43.8304, + "eval_samples_per_second": 22.815, + "eval_steps_per_second": 0.958, + "step": 400 + }, + { + "epoch": 0.43945205479452054, + "grad_norm": 1.9913676977157593, + "learning_rate": 4.77051177904143e-06, + "loss": 2.274, + "step": 401 + }, + { + "epoch": 0.44054794520547946, + "grad_norm": 2.2850067615509033, + "learning_rate": 4.768480909829407e-06, + "loss": 2.2209, + "step": 402 + }, + { + "epoch": 0.44164383561643833, + "grad_norm": 2.4984688758850098, + "learning_rate": 4.766450040617385e-06, + "loss": 2.263, + "step": 403 + }, + { + "epoch": 0.44273972602739725, + "grad_norm": 2.3244640827178955, + "learning_rate": 4.764419171405362e-06, + "loss": 2.2368, + "step": 404 + }, + { + "epoch": 0.4438356164383562, + "grad_norm": 2.7579681873321533, + "learning_rate": 4.762388302193339e-06, + "loss": 2.2762, + "step": 405 + }, + { + "epoch": 0.4449315068493151, + "grad_norm": 2.1996467113494873, + "learning_rate": 4.760357432981316e-06, + "loss": 2.2536, + "step": 406 + }, + { + "epoch": 0.44602739726027396, + "grad_norm": 2.3767571449279785, + "learning_rate": 4.758326563769294e-06, + "loss": 2.2353, + "step": 407 + }, + { + "epoch": 0.4471232876712329, + "grad_norm": 1.9148359298706055, + "learning_rate": 4.756295694557271e-06, + "loss": 2.2416, + "step": 408 + }, + { + "epoch": 0.4482191780821918, + "grad_norm": 2.6499507427215576, + "learning_rate": 4.754264825345248e-06, + "loss": 2.2641, + "step": 409 + }, + { + "epoch": 0.44931506849315067, + "grad_norm": 1.9890286922454834, + "learning_rate": 4.7522339561332255e-06, + "loss": 2.2148, + "step": 410 + }, + { + "epoch": 0.44931506849315067, + "eval_loss": 2.3670520782470703, + "eval_runtime": 41.1722, + "eval_samples_per_second": 24.288, + "eval_steps_per_second": 1.02, + "step": 410 + }, + { + "epoch": 0.4504109589041096, + "grad_norm": 2.575968027114868, + "learning_rate": 4.750203086921203e-06, + "loss": 2.2254, + "step": 411 + }, + { + "epoch": 0.4515068493150685, + "grad_norm": 2.015174388885498, + "learning_rate": 4.74817221770918e-06, + "loss": 2.1881, + "step": 412 + }, + { + "epoch": 0.4526027397260274, + "grad_norm": 2.44292950630188, + "learning_rate": 4.7461413484971565e-06, + "loss": 2.256, + "step": 413 + }, + { + "epoch": 0.4536986301369863, + "grad_norm": 1.9808441400527954, + "learning_rate": 4.744110479285135e-06, + "loss": 2.285, + "step": 414 + }, + { + "epoch": 0.4547945205479452, + "grad_norm": 2.009481906890869, + "learning_rate": 4.742079610073111e-06, + "loss": 2.2018, + "step": 415 + }, + { + "epoch": 0.4558904109589041, + "grad_norm": 2.1466031074523926, + "learning_rate": 4.740048740861089e-06, + "loss": 2.224, + "step": 416 + }, + { + "epoch": 0.456986301369863, + "grad_norm": 2.1337907314300537, + "learning_rate": 4.738017871649066e-06, + "loss": 2.2404, + "step": 417 + }, + { + "epoch": 0.45808219178082193, + "grad_norm": 1.94125235080719, + "learning_rate": 4.735987002437044e-06, + "loss": 2.2474, + "step": 418 + }, + { + "epoch": 0.4591780821917808, + "grad_norm": 2.379099130630493, + "learning_rate": 4.73395613322502e-06, + "loss": 2.2858, + "step": 419 + }, + { + "epoch": 0.4602739726027397, + "grad_norm": 1.844056248664856, + "learning_rate": 4.731925264012998e-06, + "loss": 2.2967, + "step": 420 + }, + { + "epoch": 0.4602739726027397, + "eval_loss": 2.3622241020202637, + "eval_runtime": 45.4674, + "eval_samples_per_second": 21.994, + "eval_steps_per_second": 0.924, + "step": 420 + }, + { + "epoch": 0.46136986301369864, + "grad_norm": 2.411895513534546, + "learning_rate": 4.729894394800975e-06, + "loss": 2.274, + "step": 421 + }, + { + "epoch": 0.4624657534246575, + "grad_norm": 2.014885425567627, + "learning_rate": 4.727863525588953e-06, + "loss": 2.2144, + "step": 422 + }, + { + "epoch": 0.46356164383561643, + "grad_norm": 2.112009048461914, + "learning_rate": 4.725832656376929e-06, + "loss": 2.1824, + "step": 423 + }, + { + "epoch": 0.46465753424657535, + "grad_norm": 2.111392021179199, + "learning_rate": 4.723801787164907e-06, + "loss": 2.2261, + "step": 424 + }, + { + "epoch": 0.4657534246575342, + "grad_norm": 2.0349185466766357, + "learning_rate": 4.721770917952885e-06, + "loss": 2.2622, + "step": 425 + }, + { + "epoch": 0.46684931506849314, + "grad_norm": 2.30222487449646, + "learning_rate": 4.719740048740861e-06, + "loss": 2.2206, + "step": 426 + }, + { + "epoch": 0.46794520547945206, + "grad_norm": 2.018789529800415, + "learning_rate": 4.717709179528839e-06, + "loss": 2.2196, + "step": 427 + }, + { + "epoch": 0.469041095890411, + "grad_norm": 2.046649932861328, + "learning_rate": 4.715678310316816e-06, + "loss": 2.2584, + "step": 428 + }, + { + "epoch": 0.47013698630136985, + "grad_norm": 2.0393869876861572, + "learning_rate": 4.713647441104794e-06, + "loss": 2.2537, + "step": 429 + }, + { + "epoch": 0.4712328767123288, + "grad_norm": 2.2788116931915283, + "learning_rate": 4.71161657189277e-06, + "loss": 2.2145, + "step": 430 + }, + { + "epoch": 0.4712328767123288, + "eval_loss": 2.3575785160064697, + "eval_runtime": 44.6687, + "eval_samples_per_second": 22.387, + "eval_steps_per_second": 0.94, + "step": 430 + }, + { + "epoch": 0.4723287671232877, + "grad_norm": 1.9950047731399536, + "learning_rate": 4.709585702680748e-06, + "loss": 2.2072, + "step": 431 + }, + { + "epoch": 0.47342465753424656, + "grad_norm": 1.9222451448440552, + "learning_rate": 4.707554833468725e-06, + "loss": 2.2654, + "step": 432 + }, + { + "epoch": 0.4745205479452055, + "grad_norm": 1.9735718965530396, + "learning_rate": 4.705523964256703e-06, + "loss": 2.3037, + "step": 433 + }, + { + "epoch": 0.4756164383561644, + "grad_norm": 2.3209776878356934, + "learning_rate": 4.703493095044679e-06, + "loss": 2.2083, + "step": 434 + }, + { + "epoch": 0.4767123287671233, + "grad_norm": 2.0993995666503906, + "learning_rate": 4.701462225832657e-06, + "loss": 2.2275, + "step": 435 + }, + { + "epoch": 0.4778082191780822, + "grad_norm": 2.334463119506836, + "learning_rate": 4.699431356620634e-06, + "loss": 2.1852, + "step": 436 + }, + { + "epoch": 0.4789041095890411, + "grad_norm": 1.9098213911056519, + "learning_rate": 4.697400487408612e-06, + "loss": 2.2829, + "step": 437 + }, + { + "epoch": 0.48, + "grad_norm": 2.3005056381225586, + "learning_rate": 4.695369618196588e-06, + "loss": 2.2532, + "step": 438 + }, + { + "epoch": 0.4810958904109589, + "grad_norm": 1.9069781303405762, + "learning_rate": 4.693338748984566e-06, + "loss": 2.2374, + "step": 439 + }, + { + "epoch": 0.4821917808219178, + "grad_norm": 2.050032138824463, + "learning_rate": 4.691307879772543e-06, + "loss": 2.2256, + "step": 440 + }, + { + "epoch": 0.4821917808219178, + "eval_loss": 2.3587422370910645, + "eval_runtime": 45.0941, + "eval_samples_per_second": 22.176, + "eval_steps_per_second": 0.931, + "step": 440 + }, + { + "epoch": 0.4832876712328767, + "grad_norm": 2.0563836097717285, + "learning_rate": 4.68927701056052e-06, + "loss": 2.1918, + "step": 441 + }, + { + "epoch": 0.4843835616438356, + "grad_norm": 2.1598386764526367, + "learning_rate": 4.6872461413484975e-06, + "loss": 2.2618, + "step": 442 + }, + { + "epoch": 0.48547945205479454, + "grad_norm": 1.9236539602279663, + "learning_rate": 4.685215272136475e-06, + "loss": 2.2477, + "step": 443 + }, + { + "epoch": 0.4865753424657534, + "grad_norm": 1.9923924207687378, + "learning_rate": 4.683184402924452e-06, + "loss": 2.2313, + "step": 444 + }, + { + "epoch": 0.4876712328767123, + "grad_norm": 2.162627696990967, + "learning_rate": 4.681153533712429e-06, + "loss": 2.2367, + "step": 445 + }, + { + "epoch": 0.48876712328767125, + "grad_norm": 2.083523988723755, + "learning_rate": 4.679122664500407e-06, + "loss": 2.2946, + "step": 446 + }, + { + "epoch": 0.4898630136986301, + "grad_norm": 2.504408359527588, + "learning_rate": 4.677091795288384e-06, + "loss": 2.2104, + "step": 447 + }, + { + "epoch": 0.49095890410958903, + "grad_norm": 1.8316532373428345, + "learning_rate": 4.675060926076361e-06, + "loss": 2.2027, + "step": 448 + }, + { + "epoch": 0.49205479452054796, + "grad_norm": 2.3337953090667725, + "learning_rate": 4.673030056864338e-06, + "loss": 2.278, + "step": 449 + }, + { + "epoch": 0.4931506849315068, + "grad_norm": 2.24102783203125, + "learning_rate": 4.670999187652316e-06, + "loss": 2.2018, + "step": 450 + }, + { + "epoch": 0.4931506849315068, + "eval_loss": 2.3545279502868652, + "eval_runtime": 45.1041, + "eval_samples_per_second": 22.171, + "eval_steps_per_second": 0.931, + "step": 450 + }, + { + "epoch": 0.49424657534246574, + "grad_norm": 2.07834792137146, + "learning_rate": 4.668968318440293e-06, + "loss": 2.1975, + "step": 451 + }, + { + "epoch": 0.49534246575342467, + "grad_norm": 1.7681723833084106, + "learning_rate": 4.66693744922827e-06, + "loss": 2.2033, + "step": 452 + }, + { + "epoch": 0.4964383561643836, + "grad_norm": 2.062134027481079, + "learning_rate": 4.6649065800162475e-06, + "loss": 2.2646, + "step": 453 + }, + { + "epoch": 0.49753424657534245, + "grad_norm": 1.94647216796875, + "learning_rate": 4.662875710804225e-06, + "loss": 2.1748, + "step": 454 + }, + { + "epoch": 0.4986301369863014, + "grad_norm": 1.9173872470855713, + "learning_rate": 4.660844841592202e-06, + "loss": 2.2435, + "step": 455 + }, + { + "epoch": 0.4997260273972603, + "grad_norm": 1.88880455493927, + "learning_rate": 4.658813972380179e-06, + "loss": 2.1711, + "step": 456 + }, + { + "epoch": 0.5008219178082192, + "grad_norm": 1.9628506898880005, + "learning_rate": 4.656783103168157e-06, + "loss": 2.2701, + "step": 457 + }, + { + "epoch": 0.5019178082191781, + "grad_norm": 2.0515286922454834, + "learning_rate": 4.654752233956134e-06, + "loss": 2.1656, + "step": 458 + }, + { + "epoch": 0.503013698630137, + "grad_norm": 1.978522539138794, + "learning_rate": 4.652721364744111e-06, + "loss": 2.246, + "step": 459 + }, + { + "epoch": 0.5041095890410959, + "grad_norm": 1.7993966341018677, + "learning_rate": 4.6506904955320885e-06, + "loss": 2.214, + "step": 460 + }, + { + "epoch": 0.5041095890410959, + "eval_loss": 2.3568837642669678, + "eval_runtime": 45.2086, + "eval_samples_per_second": 22.12, + "eval_steps_per_second": 0.929, + "step": 460 + }, + { + "epoch": 0.5052054794520547, + "grad_norm": 1.7809834480285645, + "learning_rate": 4.648659626320066e-06, + "loss": 2.2525, + "step": 461 + }, + { + "epoch": 0.5063013698630137, + "grad_norm": 1.9256166219711304, + "learning_rate": 4.646628757108043e-06, + "loss": 2.2611, + "step": 462 + }, + { + "epoch": 0.5073972602739726, + "grad_norm": 1.94192636013031, + "learning_rate": 4.6445978878960194e-06, + "loss": 2.2282, + "step": 463 + }, + { + "epoch": 0.5084931506849315, + "grad_norm": 1.8891818523406982, + "learning_rate": 4.6425670186839976e-06, + "loss": 2.2036, + "step": 464 + }, + { + "epoch": 0.5095890410958904, + "grad_norm": 2.352078676223755, + "learning_rate": 4.640536149471974e-06, + "loss": 2.2293, + "step": 465 + }, + { + "epoch": 0.5106849315068493, + "grad_norm": 1.8141659498214722, + "learning_rate": 4.638505280259952e-06, + "loss": 2.2551, + "step": 466 + }, + { + "epoch": 0.5117808219178083, + "grad_norm": 1.9492480754852295, + "learning_rate": 4.6364744110479285e-06, + "loss": 2.2641, + "step": 467 + }, + { + "epoch": 0.5128767123287671, + "grad_norm": 2.0210182666778564, + "learning_rate": 4.634443541835907e-06, + "loss": 2.2367, + "step": 468 + }, + { + "epoch": 0.513972602739726, + "grad_norm": 1.8936505317687988, + "learning_rate": 4.632412672623883e-06, + "loss": 2.265, + "step": 469 + }, + { + "epoch": 0.5150684931506849, + "grad_norm": 1.845888614654541, + "learning_rate": 4.630381803411861e-06, + "loss": 2.2169, + "step": 470 + }, + { + "epoch": 0.5150684931506849, + "eval_loss": 2.353489875793457, + "eval_runtime": 46.8739, + "eval_samples_per_second": 21.334, + "eval_steps_per_second": 0.896, + "step": 470 + }, + { + "epoch": 0.5161643835616438, + "grad_norm": 1.889675259590149, + "learning_rate": 4.628350934199838e-06, + "loss": 2.1383, + "step": 471 + }, + { + "epoch": 0.5172602739726028, + "grad_norm": 1.9799216985702515, + "learning_rate": 4.626320064987816e-06, + "loss": 2.2312, + "step": 472 + }, + { + "epoch": 0.5183561643835617, + "grad_norm": 2.0128607749938965, + "learning_rate": 4.624289195775792e-06, + "loss": 2.2529, + "step": 473 + }, + { + "epoch": 0.5194520547945205, + "grad_norm": 1.8791241645812988, + "learning_rate": 4.62225832656377e-06, + "loss": 2.226, + "step": 474 + }, + { + "epoch": 0.5205479452054794, + "grad_norm": 2.0497279167175293, + "learning_rate": 4.620227457351747e-06, + "loss": 2.2283, + "step": 475 + }, + { + "epoch": 0.5216438356164383, + "grad_norm": 1.8097424507141113, + "learning_rate": 4.618196588139724e-06, + "loss": 2.222, + "step": 476 + }, + { + "epoch": 0.5227397260273973, + "grad_norm": 1.8960500955581665, + "learning_rate": 4.616165718927701e-06, + "loss": 2.2342, + "step": 477 + }, + { + "epoch": 0.5238356164383562, + "grad_norm": 2.020730495452881, + "learning_rate": 4.6141348497156786e-06, + "loss": 2.2106, + "step": 478 + }, + { + "epoch": 0.5249315068493151, + "grad_norm": 1.8955297470092773, + "learning_rate": 4.612103980503656e-06, + "loss": 2.2755, + "step": 479 + }, + { + "epoch": 0.5260273972602739, + "grad_norm": 1.9767351150512695, + "learning_rate": 4.610073111291633e-06, + "loss": 2.2554, + "step": 480 + }, + { + "epoch": 0.5260273972602739, + "eval_loss": 2.347705841064453, + "eval_runtime": 46.9125, + "eval_samples_per_second": 21.316, + "eval_steps_per_second": 0.895, + "step": 480 + }, + { + "epoch": 0.5271232876712328, + "grad_norm": 1.936457633972168, + "learning_rate": 4.60804224207961e-06, + "loss": 2.2323, + "step": 481 + }, + { + "epoch": 0.5282191780821918, + "grad_norm": 1.9271862506866455, + "learning_rate": 4.606011372867588e-06, + "loss": 2.2151, + "step": 482 + }, + { + "epoch": 0.5293150684931507, + "grad_norm": 1.878153681755066, + "learning_rate": 4.603980503655565e-06, + "loss": 2.2322, + "step": 483 + }, + { + "epoch": 0.5304109589041096, + "grad_norm": 2.1009249687194824, + "learning_rate": 4.601949634443542e-06, + "loss": 2.2181, + "step": 484 + }, + { + "epoch": 0.5315068493150685, + "grad_norm": 1.9089181423187256, + "learning_rate": 4.5999187652315195e-06, + "loss": 2.1712, + "step": 485 + }, + { + "epoch": 0.5326027397260275, + "grad_norm": 2.207282781600952, + "learning_rate": 4.597887896019497e-06, + "loss": 2.1847, + "step": 486 + }, + { + "epoch": 0.5336986301369863, + "grad_norm": 2.001584053039551, + "learning_rate": 4.595857026807474e-06, + "loss": 2.214, + "step": 487 + }, + { + "epoch": 0.5347945205479452, + "grad_norm": 1.966700553894043, + "learning_rate": 4.593826157595451e-06, + "loss": 2.2237, + "step": 488 + }, + { + "epoch": 0.5358904109589041, + "grad_norm": 1.842469334602356, + "learning_rate": 4.591795288383429e-06, + "loss": 2.2089, + "step": 489 + }, + { + "epoch": 0.536986301369863, + "grad_norm": 2.0412986278533936, + "learning_rate": 4.589764419171406e-06, + "loss": 2.2734, + "step": 490 + }, + { + "epoch": 0.536986301369863, + "eval_loss": 2.3460772037506104, + "eval_runtime": 50.4048, + "eval_samples_per_second": 19.839, + "eval_steps_per_second": 0.833, + "step": 490 + }, + { + "epoch": 0.538082191780822, + "grad_norm": 1.9337031841278076, + "learning_rate": 4.587733549959383e-06, + "loss": 2.2314, + "step": 491 + }, + { + "epoch": 0.5391780821917809, + "grad_norm": 2.4448320865631104, + "learning_rate": 4.5857026807473604e-06, + "loss": 2.2098, + "step": 492 + }, + { + "epoch": 0.5402739726027397, + "grad_norm": 1.83162522315979, + "learning_rate": 4.583671811535338e-06, + "loss": 2.1902, + "step": 493 + }, + { + "epoch": 0.5413698630136986, + "grad_norm": 2.3614656925201416, + "learning_rate": 4.581640942323315e-06, + "loss": 2.1737, + "step": 494 + }, + { + "epoch": 0.5424657534246575, + "grad_norm": 2.001187801361084, + "learning_rate": 4.579610073111292e-06, + "loss": 2.2485, + "step": 495 + }, + { + "epoch": 0.5435616438356164, + "grad_norm": 2.3479933738708496, + "learning_rate": 4.5775792038992695e-06, + "loss": 2.2595, + "step": 496 + }, + { + "epoch": 0.5446575342465754, + "grad_norm": 2.135856866836548, + "learning_rate": 4.575548334687247e-06, + "loss": 2.2528, + "step": 497 + }, + { + "epoch": 0.5457534246575343, + "grad_norm": 1.983126163482666, + "learning_rate": 4.573517465475224e-06, + "loss": 2.2788, + "step": 498 + }, + { + "epoch": 0.5468493150684931, + "grad_norm": 2.18208909034729, + "learning_rate": 4.571486596263201e-06, + "loss": 2.1995, + "step": 499 + }, + { + "epoch": 0.547945205479452, + "grad_norm": 1.848747968673706, + "learning_rate": 4.569455727051178e-06, + "loss": 2.2403, + "step": 500 + }, + { + "epoch": 0.547945205479452, + "eval_loss": 2.344066619873047, + "eval_runtime": 51.1213, + "eval_samples_per_second": 19.561, + "eval_steps_per_second": 0.822, + "step": 500 + }, + { + "epoch": 0.549041095890411, + "grad_norm": 2.427860736846924, + "learning_rate": 4.567424857839156e-06, + "loss": 2.246, + "step": 501 + }, + { + "epoch": 0.5501369863013699, + "grad_norm": 1.913405418395996, + "learning_rate": 4.565393988627132e-06, + "loss": 2.2296, + "step": 502 + }, + { + "epoch": 0.5512328767123288, + "grad_norm": 2.0106029510498047, + "learning_rate": 4.5633631194151105e-06, + "loss": 2.2657, + "step": 503 + }, + { + "epoch": 0.5523287671232877, + "grad_norm": 2.2719595432281494, + "learning_rate": 4.561332250203087e-06, + "loss": 2.2341, + "step": 504 + }, + { + "epoch": 0.5534246575342465, + "grad_norm": 2.109786033630371, + "learning_rate": 4.559301380991065e-06, + "loss": 2.2181, + "step": 505 + }, + { + "epoch": 0.5545205479452054, + "grad_norm": 2.189779281616211, + "learning_rate": 4.5572705117790414e-06, + "loss": 2.1821, + "step": 506 + }, + { + "epoch": 0.5556164383561644, + "grad_norm": 2.0263590812683105, + "learning_rate": 4.5552396425670196e-06, + "loss": 2.2072, + "step": 507 + }, + { + "epoch": 0.5567123287671233, + "grad_norm": 2.215289354324341, + "learning_rate": 4.553208773354996e-06, + "loss": 2.2513, + "step": 508 + }, + { + "epoch": 0.5578082191780822, + "grad_norm": 2.3031351566314697, + "learning_rate": 4.551177904142974e-06, + "loss": 2.1988, + "step": 509 + }, + { + "epoch": 0.5589041095890411, + "grad_norm": 2.1830058097839355, + "learning_rate": 4.5491470349309505e-06, + "loss": 2.269, + "step": 510 + }, + { + "epoch": 0.5589041095890411, + "eval_loss": 2.342463731765747, + "eval_runtime": 43.1956, + "eval_samples_per_second": 23.151, + "eval_steps_per_second": 0.972, + "step": 510 + }, + { + "epoch": 0.56, + "grad_norm": 2.3210105895996094, + "learning_rate": 4.547116165718929e-06, + "loss": 2.2051, + "step": 511 + }, + { + "epoch": 0.5610958904109589, + "grad_norm": 1.9175361394882202, + "learning_rate": 4.545085296506905e-06, + "loss": 2.2304, + "step": 512 + }, + { + "epoch": 0.5621917808219178, + "grad_norm": 2.302947759628296, + "learning_rate": 4.543054427294882e-06, + "loss": 2.215, + "step": 513 + }, + { + "epoch": 0.5632876712328767, + "grad_norm": 1.8947396278381348, + "learning_rate": 4.54102355808286e-06, + "loss": 2.1926, + "step": 514 + }, + { + "epoch": 0.5643835616438356, + "grad_norm": 2.195268154144287, + "learning_rate": 4.538992688870837e-06, + "loss": 2.1937, + "step": 515 + }, + { + "epoch": 0.5654794520547946, + "grad_norm": 2.178874969482422, + "learning_rate": 4.536961819658814e-06, + "loss": 2.2248, + "step": 516 + }, + { + "epoch": 0.5665753424657535, + "grad_norm": 1.9856929779052734, + "learning_rate": 4.5349309504467915e-06, + "loss": 2.2077, + "step": 517 + }, + { + "epoch": 0.5676712328767123, + "grad_norm": 2.0686943531036377, + "learning_rate": 4.532900081234769e-06, + "loss": 2.2328, + "step": 518 + }, + { + "epoch": 0.5687671232876712, + "grad_norm": 1.9554919004440308, + "learning_rate": 4.530869212022746e-06, + "loss": 2.198, + "step": 519 + }, + { + "epoch": 0.5698630136986301, + "grad_norm": 1.9489670991897583, + "learning_rate": 4.528838342810723e-06, + "loss": 2.2287, + "step": 520 + }, + { + "epoch": 0.5698630136986301, + "eval_loss": 2.3415298461914062, + "eval_runtime": 48.7918, + "eval_samples_per_second": 20.495, + "eval_steps_per_second": 0.861, + "step": 520 + }, + { + "epoch": 0.570958904109589, + "grad_norm": 1.9868658781051636, + "learning_rate": 4.526807473598701e-06, + "loss": 2.2205, + "step": 521 + }, + { + "epoch": 0.572054794520548, + "grad_norm": 2.0965914726257324, + "learning_rate": 4.524776604386678e-06, + "loss": 2.2239, + "step": 522 + }, + { + "epoch": 0.5731506849315069, + "grad_norm": 2.041471481323242, + "learning_rate": 4.522745735174655e-06, + "loss": 2.2035, + "step": 523 + }, + { + "epoch": 0.5742465753424657, + "grad_norm": 1.885293960571289, + "learning_rate": 4.520714865962632e-06, + "loss": 2.2469, + "step": 524 + }, + { + "epoch": 0.5753424657534246, + "grad_norm": 1.9107296466827393, + "learning_rate": 4.51868399675061e-06, + "loss": 2.2386, + "step": 525 + }, + { + "epoch": 0.5764383561643835, + "grad_norm": 1.9945271015167236, + "learning_rate": 4.516653127538587e-06, + "loss": 2.235, + "step": 526 + }, + { + "epoch": 0.5775342465753425, + "grad_norm": 1.9787253141403198, + "learning_rate": 4.514622258326564e-06, + "loss": 2.2034, + "step": 527 + }, + { + "epoch": 0.5786301369863014, + "grad_norm": 2.0300636291503906, + "learning_rate": 4.5125913891145415e-06, + "loss": 2.1849, + "step": 528 + }, + { + "epoch": 0.5797260273972603, + "grad_norm": 2.0348970890045166, + "learning_rate": 4.510560519902519e-06, + "loss": 2.2174, + "step": 529 + }, + { + "epoch": 0.5808219178082191, + "grad_norm": 1.8589537143707275, + "learning_rate": 4.508529650690496e-06, + "loss": 2.2351, + "step": 530 + }, + { + "epoch": 0.5808219178082191, + "eval_loss": 2.3450186252593994, + "eval_runtime": 52.4548, + "eval_samples_per_second": 19.064, + "eval_steps_per_second": 0.801, + "step": 530 + }, + { + "epoch": 0.581917808219178, + "grad_norm": 2.0496020317077637, + "learning_rate": 4.506498781478473e-06, + "loss": 2.182, + "step": 531 + }, + { + "epoch": 0.583013698630137, + "grad_norm": 2.0649805068969727, + "learning_rate": 4.504467912266451e-06, + "loss": 2.1654, + "step": 532 + }, + { + "epoch": 0.5841095890410959, + "grad_norm": 1.9809486865997314, + "learning_rate": 4.502437043054428e-06, + "loss": 2.2182, + "step": 533 + }, + { + "epoch": 0.5852054794520548, + "grad_norm": 2.21915340423584, + "learning_rate": 4.500406173842405e-06, + "loss": 2.2697, + "step": 534 + }, + { + "epoch": 0.5863013698630137, + "grad_norm": 1.9822649955749512, + "learning_rate": 4.4983753046303824e-06, + "loss": 2.2297, + "step": 535 + }, + { + "epoch": 0.5873972602739727, + "grad_norm": 1.972575306892395, + "learning_rate": 4.49634443541836e-06, + "loss": 2.2164, + "step": 536 + }, + { + "epoch": 0.5884931506849315, + "grad_norm": 1.948667049407959, + "learning_rate": 4.494313566206336e-06, + "loss": 2.2087, + "step": 537 + }, + { + "epoch": 0.5895890410958904, + "grad_norm": 2.7486417293548584, + "learning_rate": 4.492282696994314e-06, + "loss": 2.2602, + "step": 538 + }, + { + "epoch": 0.5906849315068493, + "grad_norm": 2.3374552726745605, + "learning_rate": 4.490251827782291e-06, + "loss": 2.2054, + "step": 539 + }, + { + "epoch": 0.5917808219178082, + "grad_norm": 2.222116231918335, + "learning_rate": 4.488220958570269e-06, + "loss": 2.1868, + "step": 540 + }, + { + "epoch": 0.5917808219178082, + "eval_loss": 2.345618963241577, + "eval_runtime": 53.0302, + "eval_samples_per_second": 18.857, + "eval_steps_per_second": 0.792, + "step": 540 + }, + { + "epoch": 0.5928767123287672, + "grad_norm": 2.00752854347229, + "learning_rate": 4.486190089358245e-06, + "loss": 2.187, + "step": 541 + }, + { + "epoch": 0.5939726027397261, + "grad_norm": 2.2466132640838623, + "learning_rate": 4.484159220146223e-06, + "loss": 2.1611, + "step": 542 + }, + { + "epoch": 0.5950684931506849, + "grad_norm": 2.1279385089874268, + "learning_rate": 4.4821283509342e-06, + "loss": 2.248, + "step": 543 + }, + { + "epoch": 0.5961643835616438, + "grad_norm": 1.9140571355819702, + "learning_rate": 4.480097481722178e-06, + "loss": 2.2314, + "step": 544 + }, + { + "epoch": 0.5972602739726027, + "grad_norm": 2.120741605758667, + "learning_rate": 4.478066612510154e-06, + "loss": 2.212, + "step": 545 + }, + { + "epoch": 0.5983561643835617, + "grad_norm": 2.003713369369507, + "learning_rate": 4.4760357432981325e-06, + "loss": 2.1824, + "step": 546 + }, + { + "epoch": 0.5994520547945206, + "grad_norm": 1.9161221981048584, + "learning_rate": 4.474004874086109e-06, + "loss": 2.2015, + "step": 547 + }, + { + "epoch": 0.6005479452054795, + "grad_norm": 2.367110252380371, + "learning_rate": 4.471974004874087e-06, + "loss": 2.3122, + "step": 548 + }, + { + "epoch": 0.6016438356164383, + "grad_norm": 2.113380193710327, + "learning_rate": 4.4699431356620635e-06, + "loss": 2.1692, + "step": 549 + }, + { + "epoch": 0.6027397260273972, + "grad_norm": 1.974181890487671, + "learning_rate": 4.467912266450041e-06, + "loss": 2.2509, + "step": 550 + }, + { + "epoch": 0.6027397260273972, + "eval_loss": 2.342583656311035, + "eval_runtime": 50.7362, + "eval_samples_per_second": 19.71, + "eval_steps_per_second": 0.828, + "step": 550 + }, + { + "epoch": 0.6038356164383561, + "grad_norm": 1.889894962310791, + "learning_rate": 4.465881397238018e-06, + "loss": 2.2214, + "step": 551 + }, + { + "epoch": 0.6049315068493151, + "grad_norm": 1.8982343673706055, + "learning_rate": 4.463850528025995e-06, + "loss": 2.2135, + "step": 552 + }, + { + "epoch": 0.606027397260274, + "grad_norm": 1.9462087154388428, + "learning_rate": 4.4618196588139726e-06, + "loss": 2.2255, + "step": 553 + }, + { + "epoch": 0.6071232876712329, + "grad_norm": 1.8729931116104126, + "learning_rate": 4.45978878960195e-06, + "loss": 2.2358, + "step": 554 + }, + { + "epoch": 0.6082191780821918, + "grad_norm": 1.9230246543884277, + "learning_rate": 4.457757920389927e-06, + "loss": 2.222, + "step": 555 + }, + { + "epoch": 0.6093150684931506, + "grad_norm": 1.9495900869369507, + "learning_rate": 4.455727051177904e-06, + "loss": 2.1779, + "step": 556 + }, + { + "epoch": 0.6104109589041096, + "grad_norm": 1.983487606048584, + "learning_rate": 4.453696181965882e-06, + "loss": 2.2007, + "step": 557 + }, + { + "epoch": 0.6115068493150685, + "grad_norm": 2.1541049480438232, + "learning_rate": 4.451665312753859e-06, + "loss": 2.2513, + "step": 558 + }, + { + "epoch": 0.6126027397260274, + "grad_norm": 1.9133352041244507, + "learning_rate": 4.449634443541836e-06, + "loss": 2.2288, + "step": 559 + }, + { + "epoch": 0.6136986301369863, + "grad_norm": 2.3325741291046143, + "learning_rate": 4.4476035743298135e-06, + "loss": 2.2539, + "step": 560 + }, + { + "epoch": 0.6136986301369863, + "eval_loss": 2.3387348651885986, + "eval_runtime": 53.4301, + "eval_samples_per_second": 18.716, + "eval_steps_per_second": 0.786, + "step": 560 + }, + { + "epoch": 0.6147945205479453, + "grad_norm": 1.8099673986434937, + "learning_rate": 4.445572705117791e-06, + "loss": 2.1845, + "step": 561 + }, + { + "epoch": 0.6158904109589041, + "grad_norm": 2.201901912689209, + "learning_rate": 4.443541835905768e-06, + "loss": 2.195, + "step": 562 + }, + { + "epoch": 0.616986301369863, + "grad_norm": 1.8412714004516602, + "learning_rate": 4.441510966693745e-06, + "loss": 2.2551, + "step": 563 + }, + { + "epoch": 0.6180821917808219, + "grad_norm": 1.8761520385742188, + "learning_rate": 4.439480097481723e-06, + "loss": 2.1965, + "step": 564 + }, + { + "epoch": 0.6191780821917808, + "grad_norm": 1.9123467206954956, + "learning_rate": 4.4374492282697e-06, + "loss": 2.2502, + "step": 565 + }, + { + "epoch": 0.6202739726027398, + "grad_norm": 2.0631234645843506, + "learning_rate": 4.435418359057677e-06, + "loss": 2.1582, + "step": 566 + }, + { + "epoch": 0.6213698630136987, + "grad_norm": 2.3191616535186768, + "learning_rate": 4.433387489845654e-06, + "loss": 2.2271, + "step": 567 + }, + { + "epoch": 0.6224657534246575, + "grad_norm": 2.043057918548584, + "learning_rate": 4.431356620633632e-06, + "loss": 2.2292, + "step": 568 + }, + { + "epoch": 0.6235616438356164, + "grad_norm": 2.0716068744659424, + "learning_rate": 4.429325751421609e-06, + "loss": 2.196, + "step": 569 + }, + { + "epoch": 0.6246575342465753, + "grad_norm": 1.7669776678085327, + "learning_rate": 4.427294882209586e-06, + "loss": 2.1953, + "step": 570 + }, + { + "epoch": 0.6246575342465753, + "eval_loss": 2.336578607559204, + "eval_runtime": 50.0083, + "eval_samples_per_second": 19.997, + "eval_steps_per_second": 0.84, + "step": 570 + }, + { + "epoch": 0.6257534246575343, + "grad_norm": 1.8165771961212158, + "learning_rate": 4.4252640129975635e-06, + "loss": 2.2155, + "step": 571 + }, + { + "epoch": 0.6268493150684932, + "grad_norm": 1.8640249967575073, + "learning_rate": 4.423233143785541e-06, + "loss": 2.1648, + "step": 572 + }, + { + "epoch": 0.6279452054794521, + "grad_norm": 2.013869285583496, + "learning_rate": 4.421202274573518e-06, + "loss": 2.2131, + "step": 573 + }, + { + "epoch": 0.6290410958904109, + "grad_norm": 2.2650580406188965, + "learning_rate": 4.4191714053614945e-06, + "loss": 2.2287, + "step": 574 + }, + { + "epoch": 0.6301369863013698, + "grad_norm": 1.9232103824615479, + "learning_rate": 4.417140536149473e-06, + "loss": 2.1339, + "step": 575 + }, + { + "epoch": 0.6312328767123287, + "grad_norm": 1.910134196281433, + "learning_rate": 4.415109666937449e-06, + "loss": 2.1755, + "step": 576 + }, + { + "epoch": 0.6323287671232877, + "grad_norm": 2.2492005825042725, + "learning_rate": 4.413078797725427e-06, + "loss": 2.2362, + "step": 577 + }, + { + "epoch": 0.6334246575342466, + "grad_norm": 1.8584033250808716, + "learning_rate": 4.411047928513404e-06, + "loss": 2.1911, + "step": 578 + }, + { + "epoch": 0.6345205479452055, + "grad_norm": 2.156273126602173, + "learning_rate": 4.409017059301382e-06, + "loss": 2.2109, + "step": 579 + }, + { + "epoch": 0.6356164383561644, + "grad_norm": 1.7977535724639893, + "learning_rate": 4.406986190089358e-06, + "loss": 2.2195, + "step": 580 + }, + { + "epoch": 0.6356164383561644, + "eval_loss": 2.33555269241333, + "eval_runtime": 48.3919, + "eval_samples_per_second": 20.665, + "eval_steps_per_second": 0.868, + "step": 580 + }, + { + "epoch": 0.6367123287671232, + "grad_norm": 1.8262702226638794, + "learning_rate": 4.404955320877336e-06, + "loss": 2.2345, + "step": 581 + }, + { + "epoch": 0.6378082191780822, + "grad_norm": 1.9389383792877197, + "learning_rate": 4.402924451665313e-06, + "loss": 2.2347, + "step": 582 + }, + { + "epoch": 0.6389041095890411, + "grad_norm": 1.9548988342285156, + "learning_rate": 4.400893582453291e-06, + "loss": 2.1967, + "step": 583 + }, + { + "epoch": 0.64, + "grad_norm": 1.898092269897461, + "learning_rate": 4.398862713241267e-06, + "loss": 2.1795, + "step": 584 + }, + { + "epoch": 0.6410958904109589, + "grad_norm": 1.8183610439300537, + "learning_rate": 4.396831844029245e-06, + "loss": 2.2031, + "step": 585 + }, + { + "epoch": 0.6421917808219179, + "grad_norm": 2.1520016193389893, + "learning_rate": 4.394800974817222e-06, + "loss": 2.2051, + "step": 586 + }, + { + "epoch": 0.6432876712328767, + "grad_norm": 1.9930012226104736, + "learning_rate": 4.392770105605199e-06, + "loss": 2.2053, + "step": 587 + }, + { + "epoch": 0.6443835616438356, + "grad_norm": 2.311450958251953, + "learning_rate": 4.390739236393176e-06, + "loss": 2.2252, + "step": 588 + }, + { + "epoch": 0.6454794520547945, + "grad_norm": 2.0168697834014893, + "learning_rate": 4.388708367181154e-06, + "loss": 2.2286, + "step": 589 + }, + { + "epoch": 0.6465753424657534, + "grad_norm": 2.055964469909668, + "learning_rate": 4.386677497969131e-06, + "loss": 2.2278, + "step": 590 + }, + { + "epoch": 0.6465753424657534, + "eval_loss": 2.334367275238037, + "eval_runtime": 48.9414, + "eval_samples_per_second": 20.433, + "eval_steps_per_second": 0.858, + "step": 590 + }, + { + "epoch": 0.6476712328767124, + "grad_norm": 2.0161590576171875, + "learning_rate": 4.384646628757108e-06, + "loss": 2.2102, + "step": 591 + }, + { + "epoch": 0.6487671232876713, + "grad_norm": 2.0827667713165283, + "learning_rate": 4.3826157595450855e-06, + "loss": 2.2531, + "step": 592 + }, + { + "epoch": 0.6498630136986301, + "grad_norm": 2.077637195587158, + "learning_rate": 4.380584890333063e-06, + "loss": 2.2091, + "step": 593 + }, + { + "epoch": 0.650958904109589, + "grad_norm": 2.023892879486084, + "learning_rate": 4.37855402112104e-06, + "loss": 2.1684, + "step": 594 + }, + { + "epoch": 0.6520547945205479, + "grad_norm": 1.8509857654571533, + "learning_rate": 4.376523151909017e-06, + "loss": 2.1439, + "step": 595 + }, + { + "epoch": 0.6531506849315069, + "grad_norm": 2.1649608612060547, + "learning_rate": 4.3744922826969946e-06, + "loss": 2.2557, + "step": 596 + }, + { + "epoch": 0.6542465753424658, + "grad_norm": 2.1351773738861084, + "learning_rate": 4.372461413484972e-06, + "loss": 2.177, + "step": 597 + }, + { + "epoch": 0.6553424657534247, + "grad_norm": 2.1713974475860596, + "learning_rate": 4.370430544272949e-06, + "loss": 2.2223, + "step": 598 + }, + { + "epoch": 0.6564383561643835, + "grad_norm": 2.2905397415161133, + "learning_rate": 4.368399675060926e-06, + "loss": 2.2373, + "step": 599 + }, + { + "epoch": 0.6575342465753424, + "grad_norm": 2.007160186767578, + "learning_rate": 4.366368805848904e-06, + "loss": 2.2355, + "step": 600 + }, + { + "epoch": 0.6575342465753424, + "eval_loss": 2.3322324752807617, + "eval_runtime": 46.2769, + "eval_samples_per_second": 21.609, + "eval_steps_per_second": 0.908, + "step": 600 + }, + { + "epoch": 0.6586301369863014, + "grad_norm": 2.309783697128296, + "learning_rate": 4.364337936636881e-06, + "loss": 2.2052, + "step": 601 + }, + { + "epoch": 0.6597260273972603, + "grad_norm": 2.498034954071045, + "learning_rate": 4.362307067424858e-06, + "loss": 2.1567, + "step": 602 + }, + { + "epoch": 0.6608219178082192, + "grad_norm": 1.8302545547485352, + "learning_rate": 4.3602761982128355e-06, + "loss": 2.2117, + "step": 603 + }, + { + "epoch": 0.6619178082191781, + "grad_norm": 2.226947546005249, + "learning_rate": 4.358245329000813e-06, + "loss": 2.1881, + "step": 604 + }, + { + "epoch": 0.663013698630137, + "grad_norm": 1.8484126329421997, + "learning_rate": 4.35621445978879e-06, + "loss": 2.1955, + "step": 605 + }, + { + "epoch": 0.6641095890410958, + "grad_norm": 1.9844266176223755, + "learning_rate": 4.354183590576767e-06, + "loss": 2.2063, + "step": 606 + }, + { + "epoch": 0.6652054794520548, + "grad_norm": 1.9292365312576294, + "learning_rate": 4.352152721364745e-06, + "loss": 2.205, + "step": 607 + }, + { + "epoch": 0.6663013698630137, + "grad_norm": 1.980895757675171, + "learning_rate": 4.350121852152722e-06, + "loss": 2.1727, + "step": 608 + }, + { + "epoch": 0.6673972602739726, + "grad_norm": 1.8331713676452637, + "learning_rate": 4.348090982940699e-06, + "loss": 2.226, + "step": 609 + }, + { + "epoch": 0.6684931506849315, + "grad_norm": 1.9666537046432495, + "learning_rate": 4.3460601137286764e-06, + "loss": 2.2198, + "step": 610 + }, + { + "epoch": 0.6684931506849315, + "eval_loss": 2.3297338485717773, + "eval_runtime": 42.6274, + "eval_samples_per_second": 23.459, + "eval_steps_per_second": 0.985, + "step": 610 + }, + { + "epoch": 0.6695890410958905, + "grad_norm": 1.9741110801696777, + "learning_rate": 4.344029244516654e-06, + "loss": 2.2132, + "step": 611 + }, + { + "epoch": 0.6706849315068493, + "grad_norm": 2.0747334957122803, + "learning_rate": 4.341998375304631e-06, + "loss": 2.2081, + "step": 612 + }, + { + "epoch": 0.6717808219178082, + "grad_norm": 2.031503915786743, + "learning_rate": 4.339967506092607e-06, + "loss": 2.2642, + "step": 613 + }, + { + "epoch": 0.6728767123287671, + "grad_norm": 1.8513256311416626, + "learning_rate": 4.3379366368805855e-06, + "loss": 2.2141, + "step": 614 + }, + { + "epoch": 0.673972602739726, + "grad_norm": 1.9098315238952637, + "learning_rate": 4.335905767668562e-06, + "loss": 2.2064, + "step": 615 + }, + { + "epoch": 0.675068493150685, + "grad_norm": 1.9704065322875977, + "learning_rate": 4.33387489845654e-06, + "loss": 2.1997, + "step": 616 + }, + { + "epoch": 0.6761643835616439, + "grad_norm": 2.1520018577575684, + "learning_rate": 4.3318440292445165e-06, + "loss": 2.2278, + "step": 617 + }, + { + "epoch": 0.6772602739726027, + "grad_norm": 1.898164987564087, + "learning_rate": 4.329813160032495e-06, + "loss": 2.2147, + "step": 618 + }, + { + "epoch": 0.6783561643835616, + "grad_norm": 1.8955962657928467, + "learning_rate": 4.327782290820471e-06, + "loss": 2.1825, + "step": 619 + }, + { + "epoch": 0.6794520547945205, + "grad_norm": 1.7579313516616821, + "learning_rate": 4.325751421608449e-06, + "loss": 2.1619, + "step": 620 + }, + { + "epoch": 0.6794520547945205, + "eval_loss": 2.329692840576172, + "eval_runtime": 46.3187, + "eval_samples_per_second": 21.59, + "eval_steps_per_second": 0.907, + "step": 620 + }, + { + "epoch": 0.6805479452054795, + "grad_norm": 1.8716063499450684, + "learning_rate": 4.323720552396426e-06, + "loss": 2.2289, + "step": 621 + }, + { + "epoch": 0.6816438356164384, + "grad_norm": 1.8798290491104126, + "learning_rate": 4.321689683184404e-06, + "loss": 2.1944, + "step": 622 + }, + { + "epoch": 0.6827397260273973, + "grad_norm": 1.8438303470611572, + "learning_rate": 4.31965881397238e-06, + "loss": 2.1879, + "step": 623 + }, + { + "epoch": 0.6838356164383562, + "grad_norm": 1.7844786643981934, + "learning_rate": 4.3176279447603574e-06, + "loss": 2.1936, + "step": 624 + }, + { + "epoch": 0.684931506849315, + "grad_norm": 1.955529808998108, + "learning_rate": 4.315597075548335e-06, + "loss": 2.2258, + "step": 625 + }, + { + "epoch": 0.686027397260274, + "grad_norm": 2.3826634883880615, + "learning_rate": 4.313566206336312e-06, + "loss": 2.2135, + "step": 626 + }, + { + "epoch": 0.6871232876712329, + "grad_norm": 1.8453123569488525, + "learning_rate": 4.311535337124289e-06, + "loss": 2.2362, + "step": 627 + }, + { + "epoch": 0.6882191780821918, + "grad_norm": 2.742682933807373, + "learning_rate": 4.3095044679122665e-06, + "loss": 2.194, + "step": 628 + }, + { + "epoch": 0.6893150684931507, + "grad_norm": 2.0299901962280273, + "learning_rate": 4.307473598700244e-06, + "loss": 2.2029, + "step": 629 + }, + { + "epoch": 0.6904109589041096, + "grad_norm": 2.29683780670166, + "learning_rate": 4.305442729488221e-06, + "loss": 2.2141, + "step": 630 + }, + { + "epoch": 0.6904109589041096, + "eval_loss": 2.329368829727173, + "eval_runtime": 45.1947, + "eval_samples_per_second": 22.126, + "eval_steps_per_second": 0.929, + "step": 630 + }, + { + "epoch": 0.6915068493150685, + "grad_norm": 2.208280086517334, + "learning_rate": 4.303411860276198e-06, + "loss": 2.1637, + "step": 631 + }, + { + "epoch": 0.6926027397260274, + "grad_norm": 1.943514108657837, + "learning_rate": 4.301380991064176e-06, + "loss": 2.234, + "step": 632 + }, + { + "epoch": 0.6936986301369863, + "grad_norm": 2.486440420150757, + "learning_rate": 4.299350121852153e-06, + "loss": 2.2239, + "step": 633 + }, + { + "epoch": 0.6947945205479452, + "grad_norm": 1.7243690490722656, + "learning_rate": 4.29731925264013e-06, + "loss": 2.2428, + "step": 634 + }, + { + "epoch": 0.6958904109589041, + "grad_norm": 2.31227445602417, + "learning_rate": 4.2952883834281075e-06, + "loss": 2.2403, + "step": 635 + }, + { + "epoch": 0.6969863013698631, + "grad_norm": 1.7361263036727905, + "learning_rate": 4.293257514216085e-06, + "loss": 2.1975, + "step": 636 + }, + { + "epoch": 0.6980821917808219, + "grad_norm": 2.010831356048584, + "learning_rate": 4.291226645004062e-06, + "loss": 2.1633, + "step": 637 + }, + { + "epoch": 0.6991780821917808, + "grad_norm": 2.2796833515167236, + "learning_rate": 4.289195775792039e-06, + "loss": 2.2307, + "step": 638 + }, + { + "epoch": 0.7002739726027397, + "grad_norm": 2.1069483757019043, + "learning_rate": 4.287164906580017e-06, + "loss": 2.2389, + "step": 639 + }, + { + "epoch": 0.7013698630136986, + "grad_norm": 2.19460129737854, + "learning_rate": 4.285134037367994e-06, + "loss": 2.1992, + "step": 640 + }, + { + "epoch": 0.7013698630136986, + "eval_loss": 2.3223369121551514, + "eval_runtime": 43.1093, + "eval_samples_per_second": 23.197, + "eval_steps_per_second": 0.974, + "step": 640 + }, + { + "epoch": 0.7024657534246576, + "grad_norm": 1.875261902809143, + "learning_rate": 4.283103168155971e-06, + "loss": 2.1767, + "step": 641 + }, + { + "epoch": 0.7035616438356165, + "grad_norm": 1.906762957572937, + "learning_rate": 4.281072298943948e-06, + "loss": 2.1734, + "step": 642 + }, + { + "epoch": 0.7046575342465753, + "grad_norm": 2.1114180088043213, + "learning_rate": 4.279041429731926e-06, + "loss": 2.1734, + "step": 643 + }, + { + "epoch": 0.7057534246575342, + "grad_norm": 1.844373106956482, + "learning_rate": 4.277010560519903e-06, + "loss": 2.1967, + "step": 644 + }, + { + "epoch": 0.7068493150684931, + "grad_norm": 2.3463754653930664, + "learning_rate": 4.27497969130788e-06, + "loss": 2.1735, + "step": 645 + }, + { + "epoch": 0.707945205479452, + "grad_norm": 2.3530960083007812, + "learning_rate": 4.2729488220958575e-06, + "loss": 2.189, + "step": 646 + }, + { + "epoch": 0.709041095890411, + "grad_norm": 2.110963821411133, + "learning_rate": 4.270917952883835e-06, + "loss": 2.218, + "step": 647 + }, + { + "epoch": 0.7101369863013699, + "grad_norm": 2.625005006790161, + "learning_rate": 4.268887083671812e-06, + "loss": 2.2235, + "step": 648 + }, + { + "epoch": 0.7112328767123288, + "grad_norm": 1.8411654233932495, + "learning_rate": 4.266856214459789e-06, + "loss": 2.1757, + "step": 649 + }, + { + "epoch": 0.7123287671232876, + "grad_norm": 2.0817489624023438, + "learning_rate": 4.264825345247766e-06, + "loss": 2.2698, + "step": 650 + }, + { + "epoch": 0.7123287671232876, + "eval_loss": 2.322544574737549, + "eval_runtime": 43.0771, + "eval_samples_per_second": 23.214, + "eval_steps_per_second": 0.975, + "step": 650 + }, + { + "epoch": 0.7134246575342466, + "grad_norm": 1.8953847885131836, + "learning_rate": 4.262794476035744e-06, + "loss": 2.1745, + "step": 651 + }, + { + "epoch": 0.7145205479452055, + "grad_norm": 1.8463294506072998, + "learning_rate": 4.26076360682372e-06, + "loss": 2.2042, + "step": 652 + }, + { + "epoch": 0.7156164383561644, + "grad_norm": 1.8776047229766846, + "learning_rate": 4.2587327376116984e-06, + "loss": 2.1512, + "step": 653 + }, + { + "epoch": 0.7167123287671233, + "grad_norm": 1.8949921131134033, + "learning_rate": 4.256701868399675e-06, + "loss": 2.1895, + "step": 654 + }, + { + "epoch": 0.7178082191780822, + "grad_norm": 1.8172632455825806, + "learning_rate": 4.254670999187653e-06, + "loss": 2.2, + "step": 655 + }, + { + "epoch": 0.718904109589041, + "grad_norm": 2.105499505996704, + "learning_rate": 4.252640129975629e-06, + "loss": 2.202, + "step": 656 + }, + { + "epoch": 0.72, + "grad_norm": 1.9515924453735352, + "learning_rate": 4.2506092607636075e-06, + "loss": 2.2116, + "step": 657 + }, + { + "epoch": 0.7210958904109589, + "grad_norm": 1.761770248413086, + "learning_rate": 4.248578391551584e-06, + "loss": 2.1921, + "step": 658 + }, + { + "epoch": 0.7221917808219178, + "grad_norm": 1.9848097562789917, + "learning_rate": 4.246547522339562e-06, + "loss": 2.1553, + "step": 659 + }, + { + "epoch": 0.7232876712328767, + "grad_norm": 2.041987180709839, + "learning_rate": 4.2445166531275385e-06, + "loss": 2.1975, + "step": 660 + }, + { + "epoch": 0.7232876712328767, + "eval_loss": 2.325580358505249, + "eval_runtime": 43.6207, + "eval_samples_per_second": 22.925, + "eval_steps_per_second": 0.963, + "step": 660 + }, + { + "epoch": 0.7243835616438357, + "grad_norm": 1.9755775928497314, + "learning_rate": 4.242485783915517e-06, + "loss": 2.1955, + "step": 661 + }, + { + "epoch": 0.7254794520547945, + "grad_norm": 1.9621502161026, + "learning_rate": 4.240454914703493e-06, + "loss": 2.1554, + "step": 662 + }, + { + "epoch": 0.7265753424657534, + "grad_norm": 1.848013162612915, + "learning_rate": 4.23842404549147e-06, + "loss": 2.16, + "step": 663 + }, + { + "epoch": 0.7276712328767123, + "grad_norm": 1.7370128631591797, + "learning_rate": 4.236393176279448e-06, + "loss": 2.1696, + "step": 664 + }, + { + "epoch": 0.7287671232876712, + "grad_norm": 2.0708158016204834, + "learning_rate": 4.234362307067425e-06, + "loss": 2.1592, + "step": 665 + }, + { + "epoch": 0.7298630136986302, + "grad_norm": 1.9864391088485718, + "learning_rate": 4.232331437855402e-06, + "loss": 2.2121, + "step": 666 + }, + { + "epoch": 0.7309589041095891, + "grad_norm": 1.9988250732421875, + "learning_rate": 4.2303005686433795e-06, + "loss": 2.2306, + "step": 667 + }, + { + "epoch": 0.7320547945205479, + "grad_norm": 1.89108407497406, + "learning_rate": 4.228269699431357e-06, + "loss": 2.222, + "step": 668 + }, + { + "epoch": 0.7331506849315068, + "grad_norm": 1.7941503524780273, + "learning_rate": 4.226238830219334e-06, + "loss": 2.1975, + "step": 669 + }, + { + "epoch": 0.7342465753424657, + "grad_norm": 1.8058613538742065, + "learning_rate": 4.224207961007311e-06, + "loss": 2.2037, + "step": 670 + }, + { + "epoch": 0.7342465753424657, + "eval_loss": 2.3266475200653076, + "eval_runtime": 47.6623, + "eval_samples_per_second": 20.981, + "eval_steps_per_second": 0.881, + "step": 670 + }, + { + "epoch": 0.7353424657534247, + "grad_norm": 1.764448642730713, + "learning_rate": 4.2221770917952886e-06, + "loss": 2.1904, + "step": 671 + }, + { + "epoch": 0.7364383561643836, + "grad_norm": 1.8252112865447998, + "learning_rate": 4.220146222583266e-06, + "loss": 2.1959, + "step": 672 + }, + { + "epoch": 0.7375342465753425, + "grad_norm": 2.02898907661438, + "learning_rate": 4.218115353371243e-06, + "loss": 2.1937, + "step": 673 + }, + { + "epoch": 0.7386301369863014, + "grad_norm": 1.8766248226165771, + "learning_rate": 4.21608448415922e-06, + "loss": 2.195, + "step": 674 + }, + { + "epoch": 0.7397260273972602, + "grad_norm": 2.2350332736968994, + "learning_rate": 4.214053614947198e-06, + "loss": 2.1798, + "step": 675 + }, + { + "epoch": 0.7408219178082192, + "grad_norm": 1.820399522781372, + "learning_rate": 4.212022745735175e-06, + "loss": 2.2371, + "step": 676 + }, + { + "epoch": 0.7419178082191781, + "grad_norm": 2.3411312103271484, + "learning_rate": 4.209991876523152e-06, + "loss": 2.1603, + "step": 677 + }, + { + "epoch": 0.743013698630137, + "grad_norm": 1.8562874794006348, + "learning_rate": 4.2079610073111295e-06, + "loss": 2.2082, + "step": 678 + }, + { + "epoch": 0.7441095890410959, + "grad_norm": 2.274214506149292, + "learning_rate": 4.205930138099107e-06, + "loss": 2.1716, + "step": 679 + }, + { + "epoch": 0.7452054794520548, + "grad_norm": 1.9173530340194702, + "learning_rate": 4.203899268887084e-06, + "loss": 2.2461, + "step": 680 + }, + { + "epoch": 0.7452054794520548, + "eval_loss": 2.3253841400146484, + "eval_runtime": 46.5732, + "eval_samples_per_second": 21.472, + "eval_steps_per_second": 0.902, + "step": 680 + }, + { + "epoch": 0.7463013698630137, + "grad_norm": 1.9732033014297485, + "learning_rate": 4.201868399675061e-06, + "loss": 2.2016, + "step": 681 + }, + { + "epoch": 0.7473972602739726, + "grad_norm": 1.8850247859954834, + "learning_rate": 4.199837530463039e-06, + "loss": 2.223, + "step": 682 + }, + { + "epoch": 0.7484931506849315, + "grad_norm": 1.8936313390731812, + "learning_rate": 4.197806661251016e-06, + "loss": 2.1381, + "step": 683 + }, + { + "epoch": 0.7495890410958904, + "grad_norm": 2.11995530128479, + "learning_rate": 4.195775792038993e-06, + "loss": 2.1962, + "step": 684 + }, + { + "epoch": 0.7506849315068493, + "grad_norm": 1.8207820653915405, + "learning_rate": 4.19374492282697e-06, + "loss": 2.1835, + "step": 685 + }, + { + "epoch": 0.7517808219178083, + "grad_norm": 2.06030535697937, + "learning_rate": 4.191714053614948e-06, + "loss": 2.1347, + "step": 686 + }, + { + "epoch": 0.7528767123287671, + "grad_norm": 2.048990488052368, + "learning_rate": 4.189683184402924e-06, + "loss": 2.2189, + "step": 687 + }, + { + "epoch": 0.753972602739726, + "grad_norm": 1.976369023323059, + "learning_rate": 4.187652315190902e-06, + "loss": 2.1805, + "step": 688 + }, + { + "epoch": 0.7550684931506849, + "grad_norm": 2.1160926818847656, + "learning_rate": 4.185621445978879e-06, + "loss": 2.2322, + "step": 689 + }, + { + "epoch": 0.7561643835616438, + "grad_norm": 1.9444516897201538, + "learning_rate": 4.183590576766857e-06, + "loss": 2.1696, + "step": 690 + }, + { + "epoch": 0.7561643835616438, + "eval_loss": 2.3242011070251465, + "eval_runtime": 48.7528, + "eval_samples_per_second": 20.512, + "eval_steps_per_second": 0.861, + "step": 690 + }, + { + "epoch": 0.7572602739726028, + "grad_norm": 1.9567183256149292, + "learning_rate": 4.181559707554833e-06, + "loss": 2.1891, + "step": 691 + }, + { + "epoch": 0.7583561643835617, + "grad_norm": 1.9185020923614502, + "learning_rate": 4.179528838342811e-06, + "loss": 2.2318, + "step": 692 + }, + { + "epoch": 0.7594520547945205, + "grad_norm": 1.833487629890442, + "learning_rate": 4.177497969130788e-06, + "loss": 2.1705, + "step": 693 + }, + { + "epoch": 0.7605479452054794, + "grad_norm": 2.2949867248535156, + "learning_rate": 4.175467099918766e-06, + "loss": 2.1712, + "step": 694 + }, + { + "epoch": 0.7616438356164383, + "grad_norm": 1.8663084506988525, + "learning_rate": 4.173436230706742e-06, + "loss": 2.2262, + "step": 695 + }, + { + "epoch": 0.7627397260273973, + "grad_norm": 2.4633052349090576, + "learning_rate": 4.1714053614947205e-06, + "loss": 2.185, + "step": 696 + }, + { + "epoch": 0.7638356164383562, + "grad_norm": 2.036524772644043, + "learning_rate": 4.169374492282697e-06, + "loss": 2.1697, + "step": 697 + }, + { + "epoch": 0.7649315068493151, + "grad_norm": 2.013810634613037, + "learning_rate": 4.167343623070675e-06, + "loss": 2.1826, + "step": 698 + }, + { + "epoch": 0.766027397260274, + "grad_norm": 2.214109182357788, + "learning_rate": 4.165312753858652e-06, + "loss": 2.2162, + "step": 699 + }, + { + "epoch": 0.7671232876712328, + "grad_norm": 1.8463964462280273, + "learning_rate": 4.163281884646629e-06, + "loss": 2.1865, + "step": 700 + }, + { + "epoch": 0.7671232876712328, + "eval_loss": 2.323822498321533, + "eval_runtime": 48.3307, + "eval_samples_per_second": 20.691, + "eval_steps_per_second": 0.869, + "step": 700 + }, + { + "epoch": 0.7682191780821918, + "grad_norm": 2.321021556854248, + "learning_rate": 4.161251015434607e-06, + "loss": 2.1758, + "step": 701 + }, + { + "epoch": 0.7693150684931507, + "grad_norm": 1.8213255405426025, + "learning_rate": 4.159220146222583e-06, + "loss": 2.2002, + "step": 702 + }, + { + "epoch": 0.7704109589041096, + "grad_norm": 1.8925111293792725, + "learning_rate": 4.157189277010561e-06, + "loss": 2.1575, + "step": 703 + }, + { + "epoch": 0.7715068493150685, + "grad_norm": 1.8050175905227661, + "learning_rate": 4.155158407798538e-06, + "loss": 2.179, + "step": 704 + }, + { + "epoch": 0.7726027397260274, + "grad_norm": 1.873228907585144, + "learning_rate": 4.153127538586516e-06, + "loss": 2.2232, + "step": 705 + }, + { + "epoch": 0.7736986301369863, + "grad_norm": 1.954910397529602, + "learning_rate": 4.151096669374492e-06, + "loss": 2.1992, + "step": 706 + }, + { + "epoch": 0.7747945205479452, + "grad_norm": 1.8913410902023315, + "learning_rate": 4.1490658001624705e-06, + "loss": 2.2086, + "step": 707 + }, + { + "epoch": 0.7758904109589041, + "grad_norm": 1.8883898258209229, + "learning_rate": 4.147034930950447e-06, + "loss": 2.1801, + "step": 708 + }, + { + "epoch": 0.776986301369863, + "grad_norm": 2.054351806640625, + "learning_rate": 4.145004061738425e-06, + "loss": 2.1908, + "step": 709 + }, + { + "epoch": 0.7780821917808219, + "grad_norm": 2.029036045074463, + "learning_rate": 4.1429731925264015e-06, + "loss": 2.1891, + "step": 710 + }, + { + "epoch": 0.7780821917808219, + "eval_loss": 2.3196680545806885, + "eval_runtime": 42.8326, + "eval_samples_per_second": 23.347, + "eval_steps_per_second": 0.981, + "step": 710 + }, + { + "epoch": 0.7791780821917809, + "grad_norm": 2.27848219871521, + "learning_rate": 4.140942323314379e-06, + "loss": 2.1165, + "step": 711 + }, + { + "epoch": 0.7802739726027397, + "grad_norm": 2.134625196456909, + "learning_rate": 4.138911454102356e-06, + "loss": 2.1508, + "step": 712 + }, + { + "epoch": 0.7813698630136986, + "grad_norm": 2.103895902633667, + "learning_rate": 4.136880584890333e-06, + "loss": 2.1746, + "step": 713 + }, + { + "epoch": 0.7824657534246575, + "grad_norm": 1.713189721107483, + "learning_rate": 4.1348497156783106e-06, + "loss": 2.2023, + "step": 714 + }, + { + "epoch": 0.7835616438356164, + "grad_norm": 2.0390899181365967, + "learning_rate": 4.132818846466288e-06, + "loss": 2.2039, + "step": 715 + }, + { + "epoch": 0.7846575342465754, + "grad_norm": 1.9643651247024536, + "learning_rate": 4.130787977254265e-06, + "loss": 2.199, + "step": 716 + }, + { + "epoch": 0.7857534246575343, + "grad_norm": 1.9173564910888672, + "learning_rate": 4.128757108042242e-06, + "loss": 2.2166, + "step": 717 + }, + { + "epoch": 0.7868493150684932, + "grad_norm": 1.7923007011413574, + "learning_rate": 4.12672623883022e-06, + "loss": 2.2144, + "step": 718 + }, + { + "epoch": 0.787945205479452, + "grad_norm": 2.2235090732574463, + "learning_rate": 4.124695369618197e-06, + "loss": 2.2586, + "step": 719 + }, + { + "epoch": 0.7890410958904109, + "grad_norm": 1.780316710472107, + "learning_rate": 4.122664500406174e-06, + "loss": 2.1862, + "step": 720 + }, + { + "epoch": 0.7890410958904109, + "eval_loss": 2.318016767501831, + "eval_runtime": 50.5082, + "eval_samples_per_second": 19.799, + "eval_steps_per_second": 0.832, + "step": 720 + }, + { + "epoch": 0.7901369863013699, + "grad_norm": 2.0788493156433105, + "learning_rate": 4.1206336311941515e-06, + "loss": 2.1845, + "step": 721 + }, + { + "epoch": 0.7912328767123288, + "grad_norm": 1.9053714275360107, + "learning_rate": 4.118602761982129e-06, + "loss": 2.2005, + "step": 722 + }, + { + "epoch": 0.7923287671232877, + "grad_norm": 1.9328513145446777, + "learning_rate": 4.116571892770106e-06, + "loss": 2.1475, + "step": 723 + }, + { + "epoch": 0.7934246575342466, + "grad_norm": 2.059760570526123, + "learning_rate": 4.114541023558083e-06, + "loss": 2.1386, + "step": 724 + }, + { + "epoch": 0.7945205479452054, + "grad_norm": 2.0093464851379395, + "learning_rate": 4.112510154346061e-06, + "loss": 2.2242, + "step": 725 + }, + { + "epoch": 0.7956164383561644, + "grad_norm": 1.8513909578323364, + "learning_rate": 4.110479285134038e-06, + "loss": 2.178, + "step": 726 + }, + { + "epoch": 0.7967123287671233, + "grad_norm": 1.9481958150863647, + "learning_rate": 4.108448415922015e-06, + "loss": 2.1961, + "step": 727 + }, + { + "epoch": 0.7978082191780822, + "grad_norm": 1.9229767322540283, + "learning_rate": 4.1064175467099924e-06, + "loss": 2.2116, + "step": 728 + }, + { + "epoch": 0.7989041095890411, + "grad_norm": 1.9819737672805786, + "learning_rate": 4.10438667749797e-06, + "loss": 2.2097, + "step": 729 + }, + { + "epoch": 0.8, + "grad_norm": 1.9382416009902954, + "learning_rate": 4.102355808285947e-06, + "loss": 2.1099, + "step": 730 + }, + { + "epoch": 0.8, + "eval_loss": 2.31795334815979, + "eval_runtime": 51.8961, + "eval_samples_per_second": 19.269, + "eval_steps_per_second": 0.809, + "step": 730 + }, + { + "epoch": 0.8010958904109589, + "grad_norm": 2.006890058517456, + "learning_rate": 4.100324939073924e-06, + "loss": 2.2606, + "step": 731 + }, + { + "epoch": 0.8021917808219178, + "grad_norm": 2.164473295211792, + "learning_rate": 4.0982940698619015e-06, + "loss": 2.1869, + "step": 732 + }, + { + "epoch": 0.8032876712328767, + "grad_norm": 2.04368257522583, + "learning_rate": 4.096263200649879e-06, + "loss": 2.1846, + "step": 733 + }, + { + "epoch": 0.8043835616438356, + "grad_norm": 1.8363912105560303, + "learning_rate": 4.094232331437856e-06, + "loss": 2.1938, + "step": 734 + }, + { + "epoch": 0.8054794520547945, + "grad_norm": 1.9260189533233643, + "learning_rate": 4.092201462225833e-06, + "loss": 2.1682, + "step": 735 + }, + { + "epoch": 0.8065753424657535, + "grad_norm": 2.0688912868499756, + "learning_rate": 4.090170593013811e-06, + "loss": 2.1772, + "step": 736 + }, + { + "epoch": 0.8076712328767123, + "grad_norm": 1.859474539756775, + "learning_rate": 4.088139723801787e-06, + "loss": 2.2013, + "step": 737 + }, + { + "epoch": 0.8087671232876712, + "grad_norm": 2.0915753841400146, + "learning_rate": 4.086108854589765e-06, + "loss": 2.1519, + "step": 738 + }, + { + "epoch": 0.8098630136986301, + "grad_norm": 2.2019519805908203, + "learning_rate": 4.084077985377742e-06, + "loss": 2.1608, + "step": 739 + }, + { + "epoch": 0.810958904109589, + "grad_norm": 1.8727316856384277, + "learning_rate": 4.08204711616572e-06, + "loss": 2.1823, + "step": 740 + }, + { + "epoch": 0.810958904109589, + "eval_loss": 2.318638324737549, + "eval_runtime": 52.3669, + "eval_samples_per_second": 19.096, + "eval_steps_per_second": 0.802, + "step": 740 + }, + { + "epoch": 0.812054794520548, + "grad_norm": 2.059250831604004, + "learning_rate": 4.080016246953696e-06, + "loss": 2.2285, + "step": 741 + }, + { + "epoch": 0.8131506849315069, + "grad_norm": 1.9466749429702759, + "learning_rate": 4.077985377741674e-06, + "loss": 2.1496, + "step": 742 + }, + { + "epoch": 0.8142465753424658, + "grad_norm": 1.8967524766921997, + "learning_rate": 4.075954508529651e-06, + "loss": 2.2066, + "step": 743 + }, + { + "epoch": 0.8153424657534246, + "grad_norm": 1.976780891418457, + "learning_rate": 4.073923639317629e-06, + "loss": 2.2194, + "step": 744 + }, + { + "epoch": 0.8164383561643835, + "grad_norm": 2.077683687210083, + "learning_rate": 4.071892770105605e-06, + "loss": 2.2209, + "step": 745 + }, + { + "epoch": 0.8175342465753425, + "grad_norm": 1.9075323343276978, + "learning_rate": 4.069861900893583e-06, + "loss": 2.1727, + "step": 746 + }, + { + "epoch": 0.8186301369863014, + "grad_norm": 1.6940832138061523, + "learning_rate": 4.06783103168156e-06, + "loss": 2.1437, + "step": 747 + }, + { + "epoch": 0.8197260273972603, + "grad_norm": 1.9035768508911133, + "learning_rate": 4.065800162469538e-06, + "loss": 2.165, + "step": 748 + }, + { + "epoch": 0.8208219178082192, + "grad_norm": 1.8775745630264282, + "learning_rate": 4.063769293257514e-06, + "loss": 2.1757, + "step": 749 + }, + { + "epoch": 0.821917808219178, + "grad_norm": 1.9622418880462646, + "learning_rate": 4.061738424045492e-06, + "loss": 2.197, + "step": 750 + }, + { + "epoch": 0.821917808219178, + "eval_loss": 2.317519426345825, + "eval_runtime": 50.2699, + "eval_samples_per_second": 19.893, + "eval_steps_per_second": 0.835, + "step": 750 + }, + { + "epoch": 0.823013698630137, + "grad_norm": 1.9646214246749878, + "learning_rate": 4.059707554833469e-06, + "loss": 2.1783, + "step": 751 + }, + { + "epoch": 0.8241095890410959, + "grad_norm": 1.890342354774475, + "learning_rate": 4.057676685621446e-06, + "loss": 2.1925, + "step": 752 + }, + { + "epoch": 0.8252054794520548, + "grad_norm": 1.8977251052856445, + "learning_rate": 4.0556458164094235e-06, + "loss": 2.219, + "step": 753 + }, + { + "epoch": 0.8263013698630137, + "grad_norm": 1.8251665830612183, + "learning_rate": 4.053614947197401e-06, + "loss": 2.189, + "step": 754 + }, + { + "epoch": 0.8273972602739726, + "grad_norm": 1.7953882217407227, + "learning_rate": 4.051584077985378e-06, + "loss": 2.1237, + "step": 755 + }, + { + "epoch": 0.8284931506849315, + "grad_norm": 1.8247836828231812, + "learning_rate": 4.049553208773355e-06, + "loss": 2.1962, + "step": 756 + }, + { + "epoch": 0.8295890410958904, + "grad_norm": 1.8537967205047607, + "learning_rate": 4.047522339561333e-06, + "loss": 2.2201, + "step": 757 + }, + { + "epoch": 0.8306849315068493, + "grad_norm": 1.865315318107605, + "learning_rate": 4.04549147034931e-06, + "loss": 2.1819, + "step": 758 + }, + { + "epoch": 0.8317808219178082, + "grad_norm": 1.8620522022247314, + "learning_rate": 4.043460601137287e-06, + "loss": 2.1717, + "step": 759 + }, + { + "epoch": 0.8328767123287671, + "grad_norm": 1.9475384950637817, + "learning_rate": 4.041429731925264e-06, + "loss": 2.2718, + "step": 760 + }, + { + "epoch": 0.8328767123287671, + "eval_loss": 2.314429759979248, + "eval_runtime": 54.2324, + "eval_samples_per_second": 18.439, + "eval_steps_per_second": 0.774, + "step": 760 + }, + { + "epoch": 0.8339726027397261, + "grad_norm": 1.7170753479003906, + "learning_rate": 4.039398862713242e-06, + "loss": 2.1785, + "step": 761 + }, + { + "epoch": 0.8350684931506849, + "grad_norm": 1.968220591545105, + "learning_rate": 4.037367993501219e-06, + "loss": 2.1798, + "step": 762 + }, + { + "epoch": 0.8361643835616438, + "grad_norm": 1.7927111387252808, + "learning_rate": 4.035337124289196e-06, + "loss": 2.2372, + "step": 763 + }, + { + "epoch": 0.8372602739726027, + "grad_norm": 1.7871185541152954, + "learning_rate": 4.0333062550771735e-06, + "loss": 2.182, + "step": 764 + }, + { + "epoch": 0.8383561643835616, + "grad_norm": 1.9415812492370605, + "learning_rate": 4.031275385865151e-06, + "loss": 2.1527, + "step": 765 + }, + { + "epoch": 0.8394520547945206, + "grad_norm": 1.7327690124511719, + "learning_rate": 4.029244516653128e-06, + "loss": 2.1744, + "step": 766 + }, + { + "epoch": 0.8405479452054795, + "grad_norm": 1.9123411178588867, + "learning_rate": 4.027213647441105e-06, + "loss": 2.2049, + "step": 767 + }, + { + "epoch": 0.8416438356164384, + "grad_norm": 2.011591911315918, + "learning_rate": 4.025182778229083e-06, + "loss": 2.1391, + "step": 768 + }, + { + "epoch": 0.8427397260273972, + "grad_norm": 1.9169542789459229, + "learning_rate": 4.02315190901706e-06, + "loss": 2.1562, + "step": 769 + }, + { + "epoch": 0.8438356164383561, + "grad_norm": 1.8020051717758179, + "learning_rate": 4.021121039805037e-06, + "loss": 2.1795, + "step": 770 + }, + { + "epoch": 0.8438356164383561, + "eval_loss": 2.3199150562286377, + "eval_runtime": 52.468, + "eval_samples_per_second": 19.059, + "eval_steps_per_second": 0.8, + "step": 770 + }, + { + "epoch": 0.8449315068493151, + "grad_norm": 1.927666425704956, + "learning_rate": 4.0190901705930144e-06, + "loss": 2.1834, + "step": 771 + }, + { + "epoch": 0.846027397260274, + "grad_norm": 2.0998382568359375, + "learning_rate": 4.017059301380992e-06, + "loss": 2.2541, + "step": 772 + }, + { + "epoch": 0.8471232876712329, + "grad_norm": 1.88541579246521, + "learning_rate": 4.015028432168969e-06, + "loss": 2.1631, + "step": 773 + }, + { + "epoch": 0.8482191780821918, + "grad_norm": 1.9546253681182861, + "learning_rate": 4.012997562956945e-06, + "loss": 2.2266, + "step": 774 + }, + { + "epoch": 0.8493150684931506, + "grad_norm": 1.7969225645065308, + "learning_rate": 4.0109666937449235e-06, + "loss": 2.1526, + "step": 775 + }, + { + "epoch": 0.8504109589041096, + "grad_norm": 1.7237955331802368, + "learning_rate": 4.0089358245329e-06, + "loss": 2.1618, + "step": 776 + }, + { + "epoch": 0.8515068493150685, + "grad_norm": 1.8355857133865356, + "learning_rate": 4.006904955320878e-06, + "loss": 2.1673, + "step": 777 + }, + { + "epoch": 0.8526027397260274, + "grad_norm": 1.8049591779708862, + "learning_rate": 4.0048740861088545e-06, + "loss": 2.1676, + "step": 778 + }, + { + "epoch": 0.8536986301369863, + "grad_norm": 1.8143601417541504, + "learning_rate": 4.002843216896833e-06, + "loss": 2.214, + "step": 779 + }, + { + "epoch": 0.8547945205479452, + "grad_norm": 1.8344221115112305, + "learning_rate": 4.000812347684809e-06, + "loss": 2.1958, + "step": 780 + }, + { + "epoch": 0.8547945205479452, + "eval_loss": 2.314359426498413, + "eval_runtime": 51.8717, + "eval_samples_per_second": 19.278, + "eval_steps_per_second": 0.81, + "step": 780 + }, + { + "epoch": 0.8558904109589041, + "grad_norm": 1.944402813911438, + "learning_rate": 3.998781478472787e-06, + "loss": 2.1497, + "step": 781 + }, + { + "epoch": 0.856986301369863, + "grad_norm": 1.93898606300354, + "learning_rate": 3.996750609260764e-06, + "loss": 2.1737, + "step": 782 + }, + { + "epoch": 0.8580821917808219, + "grad_norm": 1.9499506950378418, + "learning_rate": 3.994719740048742e-06, + "loss": 2.1545, + "step": 783 + }, + { + "epoch": 0.8591780821917808, + "grad_norm": 1.9311052560806274, + "learning_rate": 3.992688870836718e-06, + "loss": 2.2559, + "step": 784 + }, + { + "epoch": 0.8602739726027397, + "grad_norm": 1.864862084388733, + "learning_rate": 3.990658001624696e-06, + "loss": 2.1996, + "step": 785 + }, + { + "epoch": 0.8613698630136987, + "grad_norm": 1.8248569965362549, + "learning_rate": 3.988627132412673e-06, + "loss": 2.1591, + "step": 786 + }, + { + "epoch": 0.8624657534246576, + "grad_norm": 2.2370970249176025, + "learning_rate": 3.98659626320065e-06, + "loss": 2.2095, + "step": 787 + }, + { + "epoch": 0.8635616438356164, + "grad_norm": 1.8877084255218506, + "learning_rate": 3.984565393988627e-06, + "loss": 2.1813, + "step": 788 + }, + { + "epoch": 0.8646575342465753, + "grad_norm": 2.1722517013549805, + "learning_rate": 3.9825345247766046e-06, + "loss": 2.1866, + "step": 789 + }, + { + "epoch": 0.8657534246575342, + "grad_norm": 1.9640655517578125, + "learning_rate": 3.980503655564582e-06, + "loss": 2.1818, + "step": 790 + }, + { + "epoch": 0.8657534246575342, + "eval_loss": 2.314087390899658, + "eval_runtime": 51.9778, + "eval_samples_per_second": 19.239, + "eval_steps_per_second": 0.808, + "step": 790 + }, + { + "epoch": 0.8668493150684932, + "grad_norm": 2.0583789348602295, + "learning_rate": 3.978472786352559e-06, + "loss": 2.1956, + "step": 791 + }, + { + "epoch": 0.8679452054794521, + "grad_norm": 2.2128682136535645, + "learning_rate": 3.976441917140536e-06, + "loss": 2.204, + "step": 792 + }, + { + "epoch": 0.869041095890411, + "grad_norm": 1.917763113975525, + "learning_rate": 3.974411047928514e-06, + "loss": 2.1841, + "step": 793 + }, + { + "epoch": 0.8701369863013698, + "grad_norm": 2.0327019691467285, + "learning_rate": 3.972380178716491e-06, + "loss": 2.2351, + "step": 794 + }, + { + "epoch": 0.8712328767123287, + "grad_norm": 1.9498991966247559, + "learning_rate": 3.970349309504468e-06, + "loss": 2.2074, + "step": 795 + }, + { + "epoch": 0.8723287671232877, + "grad_norm": 1.8105775117874146, + "learning_rate": 3.9683184402924455e-06, + "loss": 2.2088, + "step": 796 + }, + { + "epoch": 0.8734246575342466, + "grad_norm": 1.9299930334091187, + "learning_rate": 3.966287571080423e-06, + "loss": 2.1469, + "step": 797 + }, + { + "epoch": 0.8745205479452055, + "grad_norm": 1.9346121549606323, + "learning_rate": 3.9642567018684e-06, + "loss": 2.1434, + "step": 798 + }, + { + "epoch": 0.8756164383561644, + "grad_norm": 1.8202977180480957, + "learning_rate": 3.962225832656377e-06, + "loss": 2.194, + "step": 799 + }, + { + "epoch": 0.8767123287671232, + "grad_norm": 1.6922450065612793, + "learning_rate": 3.960194963444355e-06, + "loss": 2.1821, + "step": 800 + }, + { + "epoch": 0.8767123287671232, + "eval_loss": 2.310209035873413, + "eval_runtime": 46.1102, + "eval_samples_per_second": 21.687, + "eval_steps_per_second": 0.911, + "step": 800 + }, + { + "epoch": 0.8778082191780822, + "grad_norm": 1.9258904457092285, + "learning_rate": 3.958164094232332e-06, + "loss": 2.1326, + "step": 801 + }, + { + "epoch": 0.8789041095890411, + "grad_norm": 1.8485561609268188, + "learning_rate": 3.956133225020309e-06, + "loss": 2.1888, + "step": 802 + }, + { + "epoch": 0.88, + "grad_norm": 1.9802418947219849, + "learning_rate": 3.954102355808286e-06, + "loss": 2.1876, + "step": 803 + }, + { + "epoch": 0.8810958904109589, + "grad_norm": 1.8365259170532227, + "learning_rate": 3.952071486596264e-06, + "loss": 2.1804, + "step": 804 + }, + { + "epoch": 0.8821917808219178, + "grad_norm": 2.023909568786621, + "learning_rate": 3.950040617384241e-06, + "loss": 2.1801, + "step": 805 + }, + { + "epoch": 0.8832876712328767, + "grad_norm": 1.8889018297195435, + "learning_rate": 3.948009748172218e-06, + "loss": 2.1657, + "step": 806 + }, + { + "epoch": 0.8843835616438356, + "grad_norm": 1.7983578443527222, + "learning_rate": 3.9459788789601955e-06, + "loss": 2.1589, + "step": 807 + }, + { + "epoch": 0.8854794520547945, + "grad_norm": 2.157043218612671, + "learning_rate": 3.943948009748173e-06, + "loss": 2.2258, + "step": 808 + }, + { + "epoch": 0.8865753424657534, + "grad_norm": 1.7387373447418213, + "learning_rate": 3.94191714053615e-06, + "loss": 2.1878, + "step": 809 + }, + { + "epoch": 0.8876712328767123, + "grad_norm": 1.963233232498169, + "learning_rate": 3.939886271324127e-06, + "loss": 2.186, + "step": 810 + }, + { + "epoch": 0.8876712328767123, + "eval_loss": 2.306187152862549, + "eval_runtime": 42.563, + "eval_samples_per_second": 23.495, + "eval_steps_per_second": 0.987, + "step": 810 + }, + { + "epoch": 0.8887671232876713, + "grad_norm": 1.8932961225509644, + "learning_rate": 3.937855402112104e-06, + "loss": 2.198, + "step": 811 + }, + { + "epoch": 0.8898630136986302, + "grad_norm": 2.0431783199310303, + "learning_rate": 3.935824532900082e-06, + "loss": 2.1505, + "step": 812 + }, + { + "epoch": 0.890958904109589, + "grad_norm": 1.8954466581344604, + "learning_rate": 3.933793663688058e-06, + "loss": 2.1494, + "step": 813 + }, + { + "epoch": 0.8920547945205479, + "grad_norm": 1.841184139251709, + "learning_rate": 3.9317627944760365e-06, + "loss": 2.1932, + "step": 814 + }, + { + "epoch": 0.8931506849315068, + "grad_norm": 1.9083219766616821, + "learning_rate": 3.929731925264013e-06, + "loss": 2.2081, + "step": 815 + }, + { + "epoch": 0.8942465753424658, + "grad_norm": 1.9737226963043213, + "learning_rate": 3.927701056051991e-06, + "loss": 2.1405, + "step": 816 + }, + { + "epoch": 0.8953424657534247, + "grad_norm": 1.869194746017456, + "learning_rate": 3.9256701868399674e-06, + "loss": 2.1642, + "step": 817 + }, + { + "epoch": 0.8964383561643836, + "grad_norm": 1.9553205966949463, + "learning_rate": 3.9236393176279456e-06, + "loss": 2.1761, + "step": 818 + }, + { + "epoch": 0.8975342465753424, + "grad_norm": 1.8355534076690674, + "learning_rate": 3.921608448415922e-06, + "loss": 2.1551, + "step": 819 + }, + { + "epoch": 0.8986301369863013, + "grad_norm": 1.9770171642303467, + "learning_rate": 3.9195775792039e-06, + "loss": 2.1424, + "step": 820 + }, + { + "epoch": 0.8986301369863013, + "eval_loss": 2.307065010070801, + "eval_runtime": 44.5816, + "eval_samples_per_second": 22.431, + "eval_steps_per_second": 0.942, + "step": 820 + }, + { + "epoch": 0.8997260273972603, + "grad_norm": 1.7375415563583374, + "learning_rate": 3.9175467099918765e-06, + "loss": 2.1895, + "step": 821 + }, + { + "epoch": 0.9008219178082192, + "grad_norm": 1.8910484313964844, + "learning_rate": 3.915515840779855e-06, + "loss": 2.1685, + "step": 822 + }, + { + "epoch": 0.9019178082191781, + "grad_norm": 1.9191482067108154, + "learning_rate": 3.913484971567831e-06, + "loss": 2.2074, + "step": 823 + }, + { + "epoch": 0.903013698630137, + "grad_norm": 1.7864760160446167, + "learning_rate": 3.911454102355808e-06, + "loss": 2.1972, + "step": 824 + }, + { + "epoch": 0.9041095890410958, + "grad_norm": 1.9044650793075562, + "learning_rate": 3.909423233143786e-06, + "loss": 2.1237, + "step": 825 + }, + { + "epoch": 0.9052054794520548, + "grad_norm": 1.8378455638885498, + "learning_rate": 3.907392363931763e-06, + "loss": 2.1907, + "step": 826 + }, + { + "epoch": 0.9063013698630137, + "grad_norm": 1.8708094358444214, + "learning_rate": 3.90536149471974e-06, + "loss": 2.1883, + "step": 827 + }, + { + "epoch": 0.9073972602739726, + "grad_norm": 1.7802201509475708, + "learning_rate": 3.9033306255077175e-06, + "loss": 2.2356, + "step": 828 + }, + { + "epoch": 0.9084931506849315, + "grad_norm": 1.9374496936798096, + "learning_rate": 3.901299756295695e-06, + "loss": 2.2546, + "step": 829 + }, + { + "epoch": 0.9095890410958904, + "grad_norm": 1.9246270656585693, + "learning_rate": 3.899268887083672e-06, + "loss": 2.2684, + "step": 830 + }, + { + "epoch": 0.9095890410958904, + "eval_loss": 2.3065874576568604, + "eval_runtime": 45.18, + "eval_samples_per_second": 22.134, + "eval_steps_per_second": 0.93, + "step": 830 + }, + { + "epoch": 0.9106849315068493, + "grad_norm": 1.8681883811950684, + "learning_rate": 3.897238017871649e-06, + "loss": 2.1737, + "step": 831 + }, + { + "epoch": 0.9117808219178082, + "grad_norm": 1.9293930530548096, + "learning_rate": 3.8952071486596266e-06, + "loss": 2.1813, + "step": 832 + }, + { + "epoch": 0.9128767123287671, + "grad_norm": 1.844586968421936, + "learning_rate": 3.893176279447604e-06, + "loss": 2.1788, + "step": 833 + }, + { + "epoch": 0.913972602739726, + "grad_norm": 1.7658660411834717, + "learning_rate": 3.891145410235581e-06, + "loss": 2.1494, + "step": 834 + }, + { + "epoch": 0.915068493150685, + "grad_norm": 2.05153751373291, + "learning_rate": 3.889114541023558e-06, + "loss": 2.241, + "step": 835 + }, + { + "epoch": 0.9161643835616439, + "grad_norm": 1.7869952917099, + "learning_rate": 3.887083671811536e-06, + "loss": 2.1866, + "step": 836 + }, + { + "epoch": 0.9172602739726028, + "grad_norm": 1.99534273147583, + "learning_rate": 3.885052802599513e-06, + "loss": 2.1682, + "step": 837 + }, + { + "epoch": 0.9183561643835616, + "grad_norm": 1.9593979120254517, + "learning_rate": 3.88302193338749e-06, + "loss": 2.1357, + "step": 838 + }, + { + "epoch": 0.9194520547945205, + "grad_norm": 1.8441778421401978, + "learning_rate": 3.8809910641754675e-06, + "loss": 2.2283, + "step": 839 + }, + { + "epoch": 0.9205479452054794, + "grad_norm": 1.9153512716293335, + "learning_rate": 3.878960194963445e-06, + "loss": 2.1899, + "step": 840 + }, + { + "epoch": 0.9205479452054794, + "eval_loss": 2.307375907897949, + "eval_runtime": 48.0216, + "eval_samples_per_second": 20.824, + "eval_steps_per_second": 0.875, + "step": 840 + }, + { + "epoch": 0.9216438356164384, + "grad_norm": 2.004232883453369, + "learning_rate": 3.876929325751422e-06, + "loss": 2.173, + "step": 841 + }, + { + "epoch": 0.9227397260273973, + "grad_norm": 2.12992262840271, + "learning_rate": 3.874898456539399e-06, + "loss": 2.2164, + "step": 842 + }, + { + "epoch": 0.9238356164383562, + "grad_norm": 2.0505411624908447, + "learning_rate": 3.872867587327377e-06, + "loss": 2.1432, + "step": 843 + }, + { + "epoch": 0.924931506849315, + "grad_norm": 1.767324686050415, + "learning_rate": 3.870836718115354e-06, + "loss": 2.1838, + "step": 844 + }, + { + "epoch": 0.9260273972602739, + "grad_norm": 2.189107894897461, + "learning_rate": 3.868805848903331e-06, + "loss": 2.1623, + "step": 845 + }, + { + "epoch": 0.9271232876712329, + "grad_norm": 1.9781970977783203, + "learning_rate": 3.8667749796913084e-06, + "loss": 2.1738, + "step": 846 + }, + { + "epoch": 0.9282191780821918, + "grad_norm": 1.8734674453735352, + "learning_rate": 3.864744110479286e-06, + "loss": 2.1829, + "step": 847 + }, + { + "epoch": 0.9293150684931507, + "grad_norm": 1.7699538469314575, + "learning_rate": 3.862713241267262e-06, + "loss": 2.186, + "step": 848 + }, + { + "epoch": 0.9304109589041096, + "grad_norm": 1.8354437351226807, + "learning_rate": 3.86068237205524e-06, + "loss": 2.2149, + "step": 849 + }, + { + "epoch": 0.9315068493150684, + "grad_norm": 1.913249135017395, + "learning_rate": 3.858651502843217e-06, + "loss": 2.2208, + "step": 850 + }, + { + "epoch": 0.9315068493150684, + "eval_loss": 2.307549238204956, + "eval_runtime": 43.1706, + "eval_samples_per_second": 23.164, + "eval_steps_per_second": 0.973, + "step": 850 + }, + { + "epoch": 0.9326027397260274, + "grad_norm": 1.92095148563385, + "learning_rate": 3.856620633631195e-06, + "loss": 2.1839, + "step": 851 + }, + { + "epoch": 0.9336986301369863, + "grad_norm": 1.8833959102630615, + "learning_rate": 3.854589764419171e-06, + "loss": 2.1372, + "step": 852 + }, + { + "epoch": 0.9347945205479452, + "grad_norm": 1.9410737752914429, + "learning_rate": 3.852558895207149e-06, + "loss": 2.1726, + "step": 853 + }, + { + "epoch": 0.9358904109589041, + "grad_norm": 1.8328828811645508, + "learning_rate": 3.850528025995126e-06, + "loss": 2.2425, + "step": 854 + }, + { + "epoch": 0.936986301369863, + "grad_norm": 1.89652419090271, + "learning_rate": 3.848497156783104e-06, + "loss": 2.1956, + "step": 855 + }, + { + "epoch": 0.938082191780822, + "grad_norm": 1.9026265144348145, + "learning_rate": 3.84646628757108e-06, + "loss": 2.167, + "step": 856 + }, + { + "epoch": 0.9391780821917808, + "grad_norm": 2.010864734649658, + "learning_rate": 3.8444354183590585e-06, + "loss": 2.1725, + "step": 857 + }, + { + "epoch": 0.9402739726027397, + "grad_norm": 2.293717384338379, + "learning_rate": 3.842404549147035e-06, + "loss": 2.1668, + "step": 858 + }, + { + "epoch": 0.9413698630136986, + "grad_norm": 1.8209116458892822, + "learning_rate": 3.840373679935013e-06, + "loss": 2.1649, + "step": 859 + }, + { + "epoch": 0.9424657534246575, + "grad_norm": 2.061277389526367, + "learning_rate": 3.8383428107229894e-06, + "loss": 2.1479, + "step": 860 + }, + { + "epoch": 0.9424657534246575, + "eval_loss": 2.3073670864105225, + "eval_runtime": 44.3781, + "eval_samples_per_second": 22.534, + "eval_steps_per_second": 0.946, + "step": 860 + }, + { + "epoch": 0.9435616438356165, + "grad_norm": 1.8032523393630981, + "learning_rate": 3.836311941510967e-06, + "loss": 2.1665, + "step": 861 + }, + { + "epoch": 0.9446575342465754, + "grad_norm": 1.989533543586731, + "learning_rate": 3.834281072298944e-06, + "loss": 2.1612, + "step": 862 + }, + { + "epoch": 0.9457534246575342, + "grad_norm": 2.017179012298584, + "learning_rate": 3.832250203086921e-06, + "loss": 2.201, + "step": 863 + }, + { + "epoch": 0.9468493150684931, + "grad_norm": 2.0291004180908203, + "learning_rate": 3.8302193338748985e-06, + "loss": 2.1508, + "step": 864 + }, + { + "epoch": 0.947945205479452, + "grad_norm": 1.8693255186080933, + "learning_rate": 3.828188464662876e-06, + "loss": 2.2106, + "step": 865 + }, + { + "epoch": 0.949041095890411, + "grad_norm": 1.8890169858932495, + "learning_rate": 3.826157595450853e-06, + "loss": 2.2467, + "step": 866 + }, + { + "epoch": 0.9501369863013699, + "grad_norm": 2.0550642013549805, + "learning_rate": 3.82412672623883e-06, + "loss": 2.1899, + "step": 867 + }, + { + "epoch": 0.9512328767123288, + "grad_norm": 1.8681174516677856, + "learning_rate": 3.822095857026808e-06, + "loss": 2.1428, + "step": 868 + }, + { + "epoch": 0.9523287671232876, + "grad_norm": 1.9133951663970947, + "learning_rate": 3.820064987814785e-06, + "loss": 2.1947, + "step": 869 + }, + { + "epoch": 0.9534246575342465, + "grad_norm": 1.8717621564865112, + "learning_rate": 3.818034118602762e-06, + "loss": 2.1542, + "step": 870 + }, + { + "epoch": 0.9534246575342465, + "eval_loss": 2.307584047317505, + "eval_runtime": 43.718, + "eval_samples_per_second": 22.874, + "eval_steps_per_second": 0.961, + "step": 870 + }, + { + "epoch": 0.9545205479452055, + "grad_norm": 1.9843461513519287, + "learning_rate": 3.8160032493907395e-06, + "loss": 2.1547, + "step": 871 + }, + { + "epoch": 0.9556164383561644, + "grad_norm": 1.725468397140503, + "learning_rate": 3.813972380178717e-06, + "loss": 2.1679, + "step": 872 + }, + { + "epoch": 0.9567123287671233, + "grad_norm": 1.8076086044311523, + "learning_rate": 3.811941510966694e-06, + "loss": 2.1654, + "step": 873 + }, + { + "epoch": 0.9578082191780822, + "grad_norm": 1.8879774808883667, + "learning_rate": 3.809910641754671e-06, + "loss": 2.191, + "step": 874 + }, + { + "epoch": 0.958904109589041, + "grad_norm": 1.943313717842102, + "learning_rate": 3.8078797725426486e-06, + "loss": 2.1448, + "step": 875 + }, + { + "epoch": 0.96, + "grad_norm": 1.8877806663513184, + "learning_rate": 3.8058489033306254e-06, + "loss": 2.1129, + "step": 876 + }, + { + "epoch": 0.9610958904109589, + "grad_norm": 1.787083625793457, + "learning_rate": 3.803818034118603e-06, + "loss": 2.1854, + "step": 877 + }, + { + "epoch": 0.9621917808219178, + "grad_norm": 2.0925140380859375, + "learning_rate": 3.80178716490658e-06, + "loss": 2.1614, + "step": 878 + }, + { + "epoch": 0.9632876712328767, + "grad_norm": 1.9043725728988647, + "learning_rate": 3.7997562956945577e-06, + "loss": 2.2161, + "step": 879 + }, + { + "epoch": 0.9643835616438357, + "grad_norm": 1.8173545598983765, + "learning_rate": 3.7977254264825345e-06, + "loss": 2.2067, + "step": 880 + }, + { + "epoch": 0.9643835616438357, + "eval_loss": 2.302727460861206, + "eval_runtime": 44.2015, + "eval_samples_per_second": 22.624, + "eval_steps_per_second": 0.95, + "step": 880 + }, + { + "epoch": 0.9654794520547946, + "grad_norm": 1.9871562719345093, + "learning_rate": 3.7956945572705122e-06, + "loss": 2.1658, + "step": 881 + }, + { + "epoch": 0.9665753424657534, + "grad_norm": 2.103553295135498, + "learning_rate": 3.793663688058489e-06, + "loss": 2.1468, + "step": 882 + }, + { + "epoch": 0.9676712328767123, + "grad_norm": 1.9089381694793701, + "learning_rate": 3.791632818846467e-06, + "loss": 2.2289, + "step": 883 + }, + { + "epoch": 0.9687671232876712, + "grad_norm": 2.3238179683685303, + "learning_rate": 3.7896019496344436e-06, + "loss": 2.2179, + "step": 884 + }, + { + "epoch": 0.9698630136986301, + "grad_norm": 1.902161955833435, + "learning_rate": 3.787571080422421e-06, + "loss": 2.2086, + "step": 885 + }, + { + "epoch": 0.9709589041095891, + "grad_norm": 1.8587497472763062, + "learning_rate": 3.785540211210398e-06, + "loss": 2.2175, + "step": 886 + }, + { + "epoch": 0.972054794520548, + "grad_norm": 2.029719829559326, + "learning_rate": 3.7835093419983755e-06, + "loss": 2.207, + "step": 887 + }, + { + "epoch": 0.9731506849315068, + "grad_norm": 2.0228402614593506, + "learning_rate": 3.7814784727863527e-06, + "loss": 2.1851, + "step": 888 + }, + { + "epoch": 0.9742465753424657, + "grad_norm": 1.9885436296463013, + "learning_rate": 3.77944760357433e-06, + "loss": 2.1667, + "step": 889 + }, + { + "epoch": 0.9753424657534246, + "grad_norm": 1.8621459007263184, + "learning_rate": 3.7774167343623073e-06, + "loss": 2.2529, + "step": 890 + }, + { + "epoch": 0.9753424657534246, + "eval_loss": 2.303697347640991, + "eval_runtime": 44.9459, + "eval_samples_per_second": 22.249, + "eval_steps_per_second": 0.934, + "step": 890 + }, + { + "epoch": 0.9764383561643836, + "grad_norm": 2.0311460494995117, + "learning_rate": 3.7753858651502846e-06, + "loss": 2.2185, + "step": 891 + }, + { + "epoch": 0.9775342465753425, + "grad_norm": 2.2017149925231934, + "learning_rate": 3.773354995938262e-06, + "loss": 2.164, + "step": 892 + }, + { + "epoch": 0.9786301369863014, + "grad_norm": 1.949684977531433, + "learning_rate": 3.771324126726239e-06, + "loss": 2.2128, + "step": 893 + }, + { + "epoch": 0.9797260273972602, + "grad_norm": 1.9395498037338257, + "learning_rate": 3.7692932575142164e-06, + "loss": 2.1957, + "step": 894 + }, + { + "epoch": 0.9808219178082191, + "grad_norm": 2.096432685852051, + "learning_rate": 3.7672623883021937e-06, + "loss": 2.219, + "step": 895 + }, + { + "epoch": 0.9819178082191781, + "grad_norm": 2.0942952632904053, + "learning_rate": 3.765231519090171e-06, + "loss": 2.1546, + "step": 896 + }, + { + "epoch": 0.983013698630137, + "grad_norm": 2.0097291469573975, + "learning_rate": 3.7632006498781482e-06, + "loss": 2.1021, + "step": 897 + }, + { + "epoch": 0.9841095890410959, + "grad_norm": 2.0909335613250732, + "learning_rate": 3.761169780666125e-06, + "loss": 2.1862, + "step": 898 + }, + { + "epoch": 0.9852054794520548, + "grad_norm": 2.187309980392456, + "learning_rate": 3.7591389114541028e-06, + "loss": 2.1511, + "step": 899 + }, + { + "epoch": 0.9863013698630136, + "grad_norm": 2.194276809692383, + "learning_rate": 3.7571080422420796e-06, + "loss": 2.196, + "step": 900 + }, + { + "epoch": 0.9863013698630136, + "eval_loss": 2.3047611713409424, + "eval_runtime": 45.609, + "eval_samples_per_second": 21.925, + "eval_steps_per_second": 0.921, + "step": 900 + }, + { + "epoch": 0.9873972602739726, + "grad_norm": 2.1132829189300537, + "learning_rate": 3.7550771730300573e-06, + "loss": 2.2046, + "step": 901 + }, + { + "epoch": 0.9884931506849315, + "grad_norm": 2.1946468353271484, + "learning_rate": 3.753046303818034e-06, + "loss": 2.1656, + "step": 902 + }, + { + "epoch": 0.9895890410958904, + "grad_norm": 2.2210536003112793, + "learning_rate": 3.751015434606012e-06, + "loss": 2.2186, + "step": 903 + }, + { + "epoch": 0.9906849315068493, + "grad_norm": 1.913695216178894, + "learning_rate": 3.7489845653939887e-06, + "loss": 2.1933, + "step": 904 + }, + { + "epoch": 0.9917808219178083, + "grad_norm": 2.1088368892669678, + "learning_rate": 3.7469536961819664e-06, + "loss": 2.2406, + "step": 905 + }, + { + "epoch": 0.9928767123287672, + "grad_norm": 2.04239821434021, + "learning_rate": 3.7449228269699433e-06, + "loss": 2.142, + "step": 906 + }, + { + "epoch": 0.993972602739726, + "grad_norm": 1.7852507829666138, + "learning_rate": 3.742891957757921e-06, + "loss": 2.1701, + "step": 907 + }, + { + "epoch": 0.9950684931506849, + "grad_norm": 2.0275776386260986, + "learning_rate": 3.740861088545898e-06, + "loss": 2.2004, + "step": 908 + }, + { + "epoch": 0.9961643835616438, + "grad_norm": 1.9028904438018799, + "learning_rate": 3.7388302193338755e-06, + "loss": 2.177, + "step": 909 + }, + { + "epoch": 0.9972602739726028, + "grad_norm": 1.7815115451812744, + "learning_rate": 3.7367993501218524e-06, + "loss": 2.207, + "step": 910 + }, + { + "epoch": 0.9972602739726028, + "eval_loss": 2.302318811416626, + "eval_runtime": 42.9296, + "eval_samples_per_second": 23.294, + "eval_steps_per_second": 0.978, + "step": 910 + }, + { + "epoch": 0.9983561643835617, + "grad_norm": 1.913689374923706, + "learning_rate": 3.7347684809098297e-06, + "loss": 2.1594, + "step": 911 + }, + { + "epoch": 0.9994520547945206, + "grad_norm": 2.171222686767578, + "learning_rate": 3.732737611697807e-06, + "loss": 2.1689, + "step": 912 + }, + { + "epoch": 1.0005479452054795, + "grad_norm": 1.831642985343933, + "learning_rate": 3.7307067424857842e-06, + "loss": 2.1367, + "step": 913 + }, + { + "epoch": 1.0016438356164383, + "grad_norm": 2.1831955909729004, + "learning_rate": 3.7286758732737615e-06, + "loss": 2.1035, + "step": 914 + }, + { + "epoch": 1.0027397260273974, + "grad_norm": 2.038025379180908, + "learning_rate": 3.7266450040617388e-06, + "loss": 2.1461, + "step": 915 + }, + { + "epoch": 1.0038356164383562, + "grad_norm": 1.866502046585083, + "learning_rate": 3.724614134849716e-06, + "loss": 2.1322, + "step": 916 + }, + { + "epoch": 1.004931506849315, + "grad_norm": 2.1359357833862305, + "learning_rate": 3.7225832656376933e-06, + "loss": 2.2333, + "step": 917 + }, + { + "epoch": 1.006027397260274, + "grad_norm": 1.8689136505126953, + "learning_rate": 3.7205523964256706e-06, + "loss": 2.1176, + "step": 918 + }, + { + "epoch": 1.0071232876712328, + "grad_norm": 1.8869425058364868, + "learning_rate": 3.718521527213648e-06, + "loss": 2.2127, + "step": 919 + }, + { + "epoch": 1.0082191780821919, + "grad_norm": 1.9148075580596924, + "learning_rate": 3.716490658001625e-06, + "loss": 2.1846, + "step": 920 + }, + { + "epoch": 1.0082191780821919, + "eval_loss": 2.30368709564209, + "eval_runtime": 52.1845, + "eval_samples_per_second": 19.163, + "eval_steps_per_second": 0.805, + "step": 920 + }, + { + "epoch": 1.0093150684931507, + "grad_norm": 1.9502761363983154, + "learning_rate": 3.7144597887896024e-06, + "loss": 2.1731, + "step": 921 + }, + { + "epoch": 1.0104109589041095, + "grad_norm": 1.9093869924545288, + "learning_rate": 3.7124289195775797e-06, + "loss": 2.0864, + "step": 922 + }, + { + "epoch": 1.0115068493150685, + "grad_norm": 2.0871293544769287, + "learning_rate": 3.710398050365557e-06, + "loss": 2.1838, + "step": 923 + }, + { + "epoch": 1.0126027397260273, + "grad_norm": 2.0559985637664795, + "learning_rate": 3.708367181153534e-06, + "loss": 2.1764, + "step": 924 + }, + { + "epoch": 1.0136986301369864, + "grad_norm": 1.8968807458877563, + "learning_rate": 3.7063363119415115e-06, + "loss": 2.1515, + "step": 925 + }, + { + "epoch": 1.0147945205479452, + "grad_norm": 1.8613216876983643, + "learning_rate": 3.7043054427294884e-06, + "loss": 2.1694, + "step": 926 + }, + { + "epoch": 1.0158904109589042, + "grad_norm": 2.002044439315796, + "learning_rate": 3.702274573517466e-06, + "loss": 2.148, + "step": 927 + }, + { + "epoch": 1.016986301369863, + "grad_norm": 2.1138675212860107, + "learning_rate": 3.700243704305443e-06, + "loss": 2.1401, + "step": 928 + }, + { + "epoch": 1.0180821917808218, + "grad_norm": 1.7620841264724731, + "learning_rate": 3.6982128350934206e-06, + "loss": 2.1399, + "step": 929 + }, + { + "epoch": 1.0191780821917809, + "grad_norm": 1.8970746994018555, + "learning_rate": 3.6961819658813975e-06, + "loss": 2.1404, + "step": 930 + }, + { + "epoch": 1.0191780821917809, + "eval_loss": 2.3068811893463135, + "eval_runtime": 53.0188, + "eval_samples_per_second": 18.861, + "eval_steps_per_second": 0.792, + "step": 930 + }, + { + "epoch": 1.0202739726027397, + "grad_norm": 1.9405478239059448, + "learning_rate": 3.694151096669375e-06, + "loss": 2.1286, + "step": 931 + }, + { + "epoch": 1.0213698630136987, + "grad_norm": 2.112222909927368, + "learning_rate": 3.692120227457352e-06, + "loss": 2.2246, + "step": 932 + }, + { + "epoch": 1.0224657534246575, + "grad_norm": 1.8792580366134644, + "learning_rate": 3.6900893582453297e-06, + "loss": 2.148, + "step": 933 + }, + { + "epoch": 1.0235616438356165, + "grad_norm": 1.9530014991760254, + "learning_rate": 3.6880584890333066e-06, + "loss": 2.1707, + "step": 934 + }, + { + "epoch": 1.0246575342465754, + "grad_norm": 2.1778736114501953, + "learning_rate": 3.6860276198212834e-06, + "loss": 2.0879, + "step": 935 + }, + { + "epoch": 1.0257534246575342, + "grad_norm": 1.8960413932800293, + "learning_rate": 3.683996750609261e-06, + "loss": 2.1659, + "step": 936 + }, + { + "epoch": 1.0268493150684932, + "grad_norm": 1.9805221557617188, + "learning_rate": 3.681965881397238e-06, + "loss": 2.1674, + "step": 937 + }, + { + "epoch": 1.027945205479452, + "grad_norm": 1.9999068975448608, + "learning_rate": 3.6799350121852157e-06, + "loss": 2.1677, + "step": 938 + }, + { + "epoch": 1.029041095890411, + "grad_norm": 2.016911029815674, + "learning_rate": 3.6779041429731925e-06, + "loss": 2.1841, + "step": 939 + }, + { + "epoch": 1.0301369863013699, + "grad_norm": 1.7200374603271484, + "learning_rate": 3.6758732737611702e-06, + "loss": 2.1391, + "step": 940 + }, + { + "epoch": 1.0301369863013699, + "eval_loss": 2.3022048473358154, + "eval_runtime": 51.2908, + "eval_samples_per_second": 19.497, + "eval_steps_per_second": 0.819, + "step": 940 + }, + { + "epoch": 1.0312328767123287, + "grad_norm": 2.249377489089966, + "learning_rate": 3.673842404549147e-06, + "loss": 2.1618, + "step": 941 + }, + { + "epoch": 1.0323287671232877, + "grad_norm": 1.838922381401062, + "learning_rate": 3.671811535337125e-06, + "loss": 2.1527, + "step": 942 + }, + { + "epoch": 1.0334246575342465, + "grad_norm": 1.9571045637130737, + "learning_rate": 3.6697806661251016e-06, + "loss": 2.149, + "step": 943 + }, + { + "epoch": 1.0345205479452055, + "grad_norm": 1.9929102659225464, + "learning_rate": 3.6677497969130793e-06, + "loss": 2.159, + "step": 944 + }, + { + "epoch": 1.0356164383561643, + "grad_norm": 2.004830837249756, + "learning_rate": 3.665718927701056e-06, + "loss": 2.2369, + "step": 945 + }, + { + "epoch": 1.0367123287671234, + "grad_norm": 1.9605739116668701, + "learning_rate": 3.663688058489034e-06, + "loss": 2.1795, + "step": 946 + }, + { + "epoch": 1.0378082191780822, + "grad_norm": 1.9499164819717407, + "learning_rate": 3.6616571892770107e-06, + "loss": 2.1979, + "step": 947 + }, + { + "epoch": 1.038904109589041, + "grad_norm": 1.867347240447998, + "learning_rate": 3.659626320064988e-06, + "loss": 2.1616, + "step": 948 + }, + { + "epoch": 1.04, + "grad_norm": 1.9342801570892334, + "learning_rate": 3.6575954508529653e-06, + "loss": 2.1044, + "step": 949 + }, + { + "epoch": 1.0410958904109588, + "grad_norm": 2.0009584426879883, + "learning_rate": 3.6555645816409426e-06, + "loss": 2.1543, + "step": 950 + }, + { + "epoch": 1.0410958904109588, + "eval_loss": 2.3028438091278076, + "eval_runtime": 46.7426, + "eval_samples_per_second": 21.394, + "eval_steps_per_second": 0.899, + "step": 950 + }, + { + "epoch": 1.0421917808219179, + "grad_norm": 2.0411503314971924, + "learning_rate": 3.65353371242892e-06, + "loss": 2.1465, + "step": 951 + }, + { + "epoch": 1.0432876712328767, + "grad_norm": 2.0512070655822754, + "learning_rate": 3.651502843216897e-06, + "loss": 2.1828, + "step": 952 + }, + { + "epoch": 1.0443835616438357, + "grad_norm": 2.0948643684387207, + "learning_rate": 3.6494719740048744e-06, + "loss": 2.0671, + "step": 953 + }, + { + "epoch": 1.0454794520547945, + "grad_norm": 1.9103025197982788, + "learning_rate": 3.6474411047928517e-06, + "loss": 2.1651, + "step": 954 + }, + { + "epoch": 1.0465753424657533, + "grad_norm": 2.1500349044799805, + "learning_rate": 3.645410235580829e-06, + "loss": 2.137, + "step": 955 + }, + { + "epoch": 1.0476712328767124, + "grad_norm": 1.7760887145996094, + "learning_rate": 3.6433793663688062e-06, + "loss": 2.1738, + "step": 956 + }, + { + "epoch": 1.0487671232876712, + "grad_norm": 2.0111730098724365, + "learning_rate": 3.6413484971567835e-06, + "loss": 2.1569, + "step": 957 + }, + { + "epoch": 1.0498630136986302, + "grad_norm": 2.0040462017059326, + "learning_rate": 3.6393176279447608e-06, + "loss": 2.1483, + "step": 958 + }, + { + "epoch": 1.050958904109589, + "grad_norm": 1.9583499431610107, + "learning_rate": 3.637286758732738e-06, + "loss": 2.1575, + "step": 959 + }, + { + "epoch": 1.0520547945205478, + "grad_norm": 1.8842004537582397, + "learning_rate": 3.6352558895207153e-06, + "loss": 2.1797, + "step": 960 + }, + { + "epoch": 1.0520547945205478, + "eval_loss": 2.301715850830078, + "eval_runtime": 52.7345, + "eval_samples_per_second": 18.963, + "eval_steps_per_second": 0.796, + "step": 960 + }, + { + "epoch": 1.0531506849315069, + "grad_norm": 2.020909309387207, + "learning_rate": 3.633225020308692e-06, + "loss": 2.1939, + "step": 961 + }, + { + "epoch": 1.0542465753424657, + "grad_norm": 2.09222149848938, + "learning_rate": 3.63119415109667e-06, + "loss": 2.1777, + "step": 962 + }, + { + "epoch": 1.0553424657534247, + "grad_norm": 1.8582347631454468, + "learning_rate": 3.6291632818846467e-06, + "loss": 2.1505, + "step": 963 + }, + { + "epoch": 1.0564383561643835, + "grad_norm": 1.9468824863433838, + "learning_rate": 3.6271324126726244e-06, + "loss": 2.1762, + "step": 964 + }, + { + "epoch": 1.0575342465753426, + "grad_norm": 1.9484895467758179, + "learning_rate": 3.6251015434606013e-06, + "loss": 2.1102, + "step": 965 + }, + { + "epoch": 1.0586301369863014, + "grad_norm": 2.11462140083313, + "learning_rate": 3.623070674248579e-06, + "loss": 2.1114, + "step": 966 + }, + { + "epoch": 1.0597260273972602, + "grad_norm": 2.2038373947143555, + "learning_rate": 3.621039805036556e-06, + "loss": 2.1513, + "step": 967 + }, + { + "epoch": 1.0608219178082192, + "grad_norm": 2.1271684169769287, + "learning_rate": 3.6190089358245335e-06, + "loss": 2.137, + "step": 968 + }, + { + "epoch": 1.061917808219178, + "grad_norm": 1.8624194860458374, + "learning_rate": 3.6169780666125104e-06, + "loss": 2.1701, + "step": 969 + }, + { + "epoch": 1.063013698630137, + "grad_norm": 1.935445785522461, + "learning_rate": 3.614947197400488e-06, + "loss": 2.1726, + "step": 970 + }, + { + "epoch": 1.063013698630137, + "eval_loss": 2.302633047103882, + "eval_runtime": 51.8284, + "eval_samples_per_second": 19.294, + "eval_steps_per_second": 0.81, + "step": 970 + }, + { + "epoch": 1.0641095890410959, + "grad_norm": 2.048907518386841, + "learning_rate": 3.612916328188465e-06, + "loss": 2.1712, + "step": 971 + }, + { + "epoch": 1.0652054794520547, + "grad_norm": 1.850943684577942, + "learning_rate": 3.610885458976442e-06, + "loss": 2.102, + "step": 972 + }, + { + "epoch": 1.0663013698630137, + "grad_norm": 1.8364315032958984, + "learning_rate": 3.6088545897644195e-06, + "loss": 2.1903, + "step": 973 + }, + { + "epoch": 1.0673972602739725, + "grad_norm": 1.8227031230926514, + "learning_rate": 3.6068237205523963e-06, + "loss": 2.1787, + "step": 974 + }, + { + "epoch": 1.0684931506849316, + "grad_norm": 1.9136372804641724, + "learning_rate": 3.604792851340374e-06, + "loss": 2.1244, + "step": 975 + }, + { + "epoch": 1.0695890410958904, + "grad_norm": 1.8585968017578125, + "learning_rate": 3.602761982128351e-06, + "loss": 2.1287, + "step": 976 + }, + { + "epoch": 1.0706849315068494, + "grad_norm": 2.139573812484741, + "learning_rate": 3.6007311129163286e-06, + "loss": 2.1938, + "step": 977 + }, + { + "epoch": 1.0717808219178082, + "grad_norm": 2.1584925651550293, + "learning_rate": 3.5987002437043054e-06, + "loss": 2.1626, + "step": 978 + }, + { + "epoch": 1.072876712328767, + "grad_norm": 1.8905949592590332, + "learning_rate": 3.596669374492283e-06, + "loss": 2.1219, + "step": 979 + }, + { + "epoch": 1.073972602739726, + "grad_norm": 2.099651575088501, + "learning_rate": 3.59463850528026e-06, + "loss": 2.1385, + "step": 980 + }, + { + "epoch": 1.073972602739726, + "eval_loss": 2.3004090785980225, + "eval_runtime": 52.7406, + "eval_samples_per_second": 18.961, + "eval_steps_per_second": 0.796, + "step": 980 + }, + { + "epoch": 1.0750684931506849, + "grad_norm": 2.1456780433654785, + "learning_rate": 3.5926076360682377e-06, + "loss": 2.1973, + "step": 981 + }, + { + "epoch": 1.076164383561644, + "grad_norm": 1.929372787475586, + "learning_rate": 3.5905767668562145e-06, + "loss": 2.1115, + "step": 982 + }, + { + "epoch": 1.0772602739726027, + "grad_norm": 2.08174991607666, + "learning_rate": 3.5885458976441922e-06, + "loss": 2.1785, + "step": 983 + }, + { + "epoch": 1.0783561643835617, + "grad_norm": 1.9355952739715576, + "learning_rate": 3.586515028432169e-06, + "loss": 2.1567, + "step": 984 + }, + { + "epoch": 1.0794520547945206, + "grad_norm": 1.9983466863632202, + "learning_rate": 3.5844841592201464e-06, + "loss": 2.1913, + "step": 985 + }, + { + "epoch": 1.0805479452054794, + "grad_norm": 1.8714051246643066, + "learning_rate": 3.5824532900081237e-06, + "loss": 2.1938, + "step": 986 + }, + { + "epoch": 1.0816438356164384, + "grad_norm": 2.007049798965454, + "learning_rate": 3.580422420796101e-06, + "loss": 2.1093, + "step": 987 + }, + { + "epoch": 1.0827397260273972, + "grad_norm": 1.965572476387024, + "learning_rate": 3.578391551584078e-06, + "loss": 2.1426, + "step": 988 + }, + { + "epoch": 1.0838356164383562, + "grad_norm": 1.9040261507034302, + "learning_rate": 3.5763606823720555e-06, + "loss": 2.1627, + "step": 989 + }, + { + "epoch": 1.084931506849315, + "grad_norm": 2.001929759979248, + "learning_rate": 3.5743298131600328e-06, + "loss": 2.0983, + "step": 990 + }, + { + "epoch": 1.084931506849315, + "eval_loss": 2.2979862689971924, + "eval_runtime": 50.6325, + "eval_samples_per_second": 19.75, + "eval_steps_per_second": 0.83, + "step": 990 + }, + { + "epoch": 1.086027397260274, + "grad_norm": 1.9204787015914917, + "learning_rate": 3.57229894394801e-06, + "loss": 2.1571, + "step": 991 + }, + { + "epoch": 1.087123287671233, + "grad_norm": 1.9628651142120361, + "learning_rate": 3.5702680747359873e-06, + "loss": 2.1736, + "step": 992 + }, + { + "epoch": 1.0882191780821917, + "grad_norm": 1.9888029098510742, + "learning_rate": 3.5682372055239646e-06, + "loss": 2.17, + "step": 993 + }, + { + "epoch": 1.0893150684931507, + "grad_norm": 2.049769401550293, + "learning_rate": 3.566206336311942e-06, + "loss": 2.1352, + "step": 994 + }, + { + "epoch": 1.0904109589041096, + "grad_norm": 2.0208375453948975, + "learning_rate": 3.564175467099919e-06, + "loss": 2.1564, + "step": 995 + }, + { + "epoch": 1.0915068493150686, + "grad_norm": 1.976775050163269, + "learning_rate": 3.5621445978878964e-06, + "loss": 2.1292, + "step": 996 + }, + { + "epoch": 1.0926027397260274, + "grad_norm": 2.0502822399139404, + "learning_rate": 3.5601137286758737e-06, + "loss": 2.1115, + "step": 997 + }, + { + "epoch": 1.0936986301369862, + "grad_norm": 2.1184189319610596, + "learning_rate": 3.5580828594638505e-06, + "loss": 2.1641, + "step": 998 + }, + { + "epoch": 1.0947945205479452, + "grad_norm": 1.9662559032440186, + "learning_rate": 3.5560519902518282e-06, + "loss": 2.1338, + "step": 999 + }, + { + "epoch": 1.095890410958904, + "grad_norm": 2.13386869430542, + "learning_rate": 3.554021121039805e-06, + "loss": 2.158, + "step": 1000 + }, + { + "epoch": 1.095890410958904, + "eval_loss": 2.3001608848571777, + "eval_runtime": 45.9219, + "eval_samples_per_second": 21.776, + "eval_steps_per_second": 0.915, + "step": 1000 + }, + { + "epoch": 1.096986301369863, + "grad_norm": 1.8791636228561401, + "learning_rate": 3.551990251827783e-06, + "loss": 2.1752, + "step": 1001 + }, + { + "epoch": 1.098082191780822, + "grad_norm": 1.9105546474456787, + "learning_rate": 3.5499593826157596e-06, + "loss": 2.1543, + "step": 1002 + }, + { + "epoch": 1.099178082191781, + "grad_norm": 1.887668251991272, + "learning_rate": 3.5479285134037373e-06, + "loss": 2.1226, + "step": 1003 + }, + { + "epoch": 1.1002739726027397, + "grad_norm": 2.2854599952697754, + "learning_rate": 3.545897644191714e-06, + "loss": 2.1425, + "step": 1004 + }, + { + "epoch": 1.1013698630136985, + "grad_norm": 2.03892183303833, + "learning_rate": 3.543866774979692e-06, + "loss": 2.135, + "step": 1005 + }, + { + "epoch": 1.1024657534246576, + "grad_norm": 1.9144858121871948, + "learning_rate": 3.5418359057676687e-06, + "loss": 2.1323, + "step": 1006 + }, + { + "epoch": 1.1035616438356164, + "grad_norm": 2.1402337551116943, + "learning_rate": 3.5398050365556464e-06, + "loss": 2.1668, + "step": 1007 + }, + { + "epoch": 1.1046575342465754, + "grad_norm": 2.084592819213867, + "learning_rate": 3.5377741673436233e-06, + "loss": 2.1331, + "step": 1008 + }, + { + "epoch": 1.1057534246575342, + "grad_norm": 2.0012383460998535, + "learning_rate": 3.535743298131601e-06, + "loss": 2.1677, + "step": 1009 + }, + { + "epoch": 1.106849315068493, + "grad_norm": 2.272204637527466, + "learning_rate": 3.533712428919578e-06, + "loss": 2.13, + "step": 1010 + }, + { + "epoch": 1.106849315068493, + "eval_loss": 2.3001229763031006, + "eval_runtime": 42.0444, + "eval_samples_per_second": 23.784, + "eval_steps_per_second": 0.999, + "step": 1010 + }, + { + "epoch": 1.107945205479452, + "grad_norm": 2.1061534881591797, + "learning_rate": 3.5316815597075547e-06, + "loss": 2.1581, + "step": 1011 + }, + { + "epoch": 1.1090410958904109, + "grad_norm": 2.159306764602661, + "learning_rate": 3.5296506904955324e-06, + "loss": 2.1868, + "step": 1012 + }, + { + "epoch": 1.11013698630137, + "grad_norm": 2.1429545879364014, + "learning_rate": 3.5276198212835093e-06, + "loss": 2.1274, + "step": 1013 + }, + { + "epoch": 1.1112328767123287, + "grad_norm": 1.8494454622268677, + "learning_rate": 3.525588952071487e-06, + "loss": 2.1513, + "step": 1014 + }, + { + "epoch": 1.1123287671232878, + "grad_norm": 1.8494454622268677, + "learning_rate": 3.525588952071487e-06, + "loss": 2.1883, + "step": 1015 + }, + { + "epoch": 1.1134246575342466, + "grad_norm": 2.2244861125946045, + "learning_rate": 3.523558082859464e-06, + "loss": 2.1896, + "step": 1016 + }, + { + "epoch": 1.1145205479452054, + "grad_norm": 2.181950569152832, + "learning_rate": 3.5215272136474415e-06, + "loss": 2.1561, + "step": 1017 + }, + { + "epoch": 1.1156164383561644, + "grad_norm": 1.7740849256515503, + "learning_rate": 3.5194963444354184e-06, + "loss": 2.1358, + "step": 1018 + }, + { + "epoch": 1.1167123287671232, + "grad_norm": 2.3540139198303223, + "learning_rate": 3.517465475223396e-06, + "loss": 2.1241, + "step": 1019 + }, + { + "epoch": 1.1178082191780823, + "grad_norm": 2.2737820148468018, + "learning_rate": 3.515434606011373e-06, + "loss": 2.1117, + "step": 1020 + }, + { + "epoch": 1.1178082191780823, + "eval_loss": 2.2998275756835938, + "eval_runtime": 46.2142, + "eval_samples_per_second": 21.638, + "eval_steps_per_second": 0.909, + "step": 1020 + }, + { + "epoch": 1.118904109589041, + "grad_norm": 2.0705313682556152, + "learning_rate": 3.5134037367993506e-06, + "loss": 2.1743, + "step": 1021 + }, + { + "epoch": 1.12, + "grad_norm": 2.2605583667755127, + "learning_rate": 3.5113728675873275e-06, + "loss": 2.2082, + "step": 1022 + }, + { + "epoch": 1.121095890410959, + "grad_norm": 2.028203248977661, + "learning_rate": 3.5093419983753047e-06, + "loss": 2.1722, + "step": 1023 + }, + { + "epoch": 1.1221917808219177, + "grad_norm": 1.9794270992279053, + "learning_rate": 3.507311129163282e-06, + "loss": 2.1581, + "step": 1024 + }, + { + "epoch": 1.1232876712328768, + "grad_norm": 2.1830344200134277, + "learning_rate": 3.5052802599512593e-06, + "loss": 2.1142, + "step": 1025 + }, + { + "epoch": 1.1243835616438356, + "grad_norm": 1.926081895828247, + "learning_rate": 3.5032493907392366e-06, + "loss": 2.1532, + "step": 1026 + }, + { + "epoch": 1.1254794520547946, + "grad_norm": 1.9909394979476929, + "learning_rate": 3.501218521527214e-06, + "loss": 2.153, + "step": 1027 + }, + { + "epoch": 1.1265753424657534, + "grad_norm": 1.9685133695602417, + "learning_rate": 3.499187652315191e-06, + "loss": 2.0983, + "step": 1028 + }, + { + "epoch": 1.1276712328767124, + "grad_norm": 1.9185279607772827, + "learning_rate": 3.4971567831031684e-06, + "loss": 2.1495, + "step": 1029 + }, + { + "epoch": 1.1287671232876713, + "grad_norm": 2.0214717388153076, + "learning_rate": 3.4951259138911457e-06, + "loss": 2.176, + "step": 1030 + }, + { + "epoch": 1.1287671232876713, + "eval_loss": 2.2963690757751465, + "eval_runtime": 44.2655, + "eval_samples_per_second": 22.591, + "eval_steps_per_second": 0.949, + "step": 1030 + }, + { + "epoch": 1.12986301369863, + "grad_norm": 1.9134252071380615, + "learning_rate": 3.493095044679123e-06, + "loss": 2.1034, + "step": 1031 + }, + { + "epoch": 1.130958904109589, + "grad_norm": 1.8448772430419922, + "learning_rate": 3.4910641754671002e-06, + "loss": 2.1947, + "step": 1032 + }, + { + "epoch": 1.132054794520548, + "grad_norm": 1.856341004371643, + "learning_rate": 3.4890333062550775e-06, + "loss": 2.0676, + "step": 1033 + }, + { + "epoch": 1.133150684931507, + "grad_norm": 2.039062261581421, + "learning_rate": 3.4870024370430548e-06, + "loss": 2.1618, + "step": 1034 + }, + { + "epoch": 1.1342465753424658, + "grad_norm": 1.8440991640090942, + "learning_rate": 3.484971567831032e-06, + "loss": 2.0956, + "step": 1035 + }, + { + "epoch": 1.1353424657534246, + "grad_norm": 1.8483991622924805, + "learning_rate": 3.482940698619009e-06, + "loss": 2.1364, + "step": 1036 + }, + { + "epoch": 1.1364383561643836, + "grad_norm": 1.9045069217681885, + "learning_rate": 3.4809098294069866e-06, + "loss": 2.1786, + "step": 1037 + }, + { + "epoch": 1.1375342465753424, + "grad_norm": 1.8203203678131104, + "learning_rate": 3.4788789601949634e-06, + "loss": 2.1438, + "step": 1038 + }, + { + "epoch": 1.1386301369863014, + "grad_norm": 2.1084606647491455, + "learning_rate": 3.476848090982941e-06, + "loss": 2.1553, + "step": 1039 + }, + { + "epoch": 1.1397260273972603, + "grad_norm": 2.1041061878204346, + "learning_rate": 3.474817221770918e-06, + "loss": 2.1435, + "step": 1040 + }, + { + "epoch": 1.1397260273972603, + "eval_loss": 2.2970690727233887, + "eval_runtime": 43.2304, + "eval_samples_per_second": 23.132, + "eval_steps_per_second": 0.972, + "step": 1040 + }, + { + "epoch": 1.140821917808219, + "grad_norm": 1.9701104164123535, + "learning_rate": 3.4727863525588957e-06, + "loss": 2.2201, + "step": 1041 + }, + { + "epoch": 1.141917808219178, + "grad_norm": 1.8816475868225098, + "learning_rate": 3.4707554833468725e-06, + "loss": 2.1643, + "step": 1042 + }, + { + "epoch": 1.143013698630137, + "grad_norm": 2.096355438232422, + "learning_rate": 3.4687246141348502e-06, + "loss": 2.1514, + "step": 1043 + }, + { + "epoch": 1.144109589041096, + "grad_norm": 1.9975471496582031, + "learning_rate": 3.466693744922827e-06, + "loss": 2.1246, + "step": 1044 + }, + { + "epoch": 1.1452054794520548, + "grad_norm": 2.0587127208709717, + "learning_rate": 3.464662875710805e-06, + "loss": 2.1371, + "step": 1045 + }, + { + "epoch": 1.1463013698630138, + "grad_norm": 1.9275381565093994, + "learning_rate": 3.4626320064987817e-06, + "loss": 2.1518, + "step": 1046 + }, + { + "epoch": 1.1473972602739726, + "grad_norm": 2.1690948009490967, + "learning_rate": 3.4606011372867594e-06, + "loss": 2.1528, + "step": 1047 + }, + { + "epoch": 1.1484931506849314, + "grad_norm": 1.970973014831543, + "learning_rate": 3.458570268074736e-06, + "loss": 2.1569, + "step": 1048 + }, + { + "epoch": 1.1495890410958904, + "grad_norm": 1.8897690773010254, + "learning_rate": 3.4565393988627135e-06, + "loss": 2.1401, + "step": 1049 + }, + { + "epoch": 1.1506849315068493, + "grad_norm": 1.9245911836624146, + "learning_rate": 3.4545085296506908e-06, + "loss": 2.1311, + "step": 1050 + }, + { + "epoch": 1.1506849315068493, + "eval_loss": 2.2956528663635254, + "eval_runtime": 44.1778, + "eval_samples_per_second": 22.636, + "eval_steps_per_second": 0.951, + "step": 1050 + }, + { + "epoch": 1.1517808219178083, + "grad_norm": 2.142209529876709, + "learning_rate": 3.452477660438668e-06, + "loss": 2.1144, + "step": 1051 + }, + { + "epoch": 1.152876712328767, + "grad_norm": 1.932369351387024, + "learning_rate": 3.4504467912266453e-06, + "loss": 2.1739, + "step": 1052 + }, + { + "epoch": 1.1539726027397261, + "grad_norm": 2.259359359741211, + "learning_rate": 3.4484159220146226e-06, + "loss": 2.1036, + "step": 1053 + }, + { + "epoch": 1.155068493150685, + "grad_norm": 2.112485647201538, + "learning_rate": 3.4463850528026e-06, + "loss": 2.1429, + "step": 1054 + }, + { + "epoch": 1.1561643835616437, + "grad_norm": 2.6211049556732178, + "learning_rate": 3.444354183590577e-06, + "loss": 2.1421, + "step": 1055 + }, + { + "epoch": 1.1572602739726028, + "grad_norm": 1.8724851608276367, + "learning_rate": 3.4423233143785544e-06, + "loss": 2.1622, + "step": 1056 + }, + { + "epoch": 1.1583561643835616, + "grad_norm": 1.9292882680892944, + "learning_rate": 3.4402924451665317e-06, + "loss": 2.1071, + "step": 1057 + }, + { + "epoch": 1.1594520547945206, + "grad_norm": 1.9832521677017212, + "learning_rate": 3.438261575954509e-06, + "loss": 2.0879, + "step": 1058 + }, + { + "epoch": 1.1605479452054794, + "grad_norm": 1.9720780849456787, + "learning_rate": 3.4362307067424862e-06, + "loss": 2.1603, + "step": 1059 + }, + { + "epoch": 1.1616438356164385, + "grad_norm": 1.8179587125778198, + "learning_rate": 3.434199837530463e-06, + "loss": 2.1851, + "step": 1060 + }, + { + "epoch": 1.1616438356164385, + "eval_loss": 2.2950079441070557, + "eval_runtime": 42.9625, + "eval_samples_per_second": 23.276, + "eval_steps_per_second": 0.978, + "step": 1060 + }, + { + "epoch": 1.1627397260273973, + "grad_norm": 1.8762129545211792, + "learning_rate": 3.432168968318441e-06, + "loss": 2.1527, + "step": 1061 + }, + { + "epoch": 1.163835616438356, + "grad_norm": 2.173297643661499, + "learning_rate": 3.4301380991064176e-06, + "loss": 2.1023, + "step": 1062 + }, + { + "epoch": 1.1649315068493151, + "grad_norm": 1.9358222484588623, + "learning_rate": 3.4281072298943953e-06, + "loss": 2.1638, + "step": 1063 + }, + { + "epoch": 1.166027397260274, + "grad_norm": 1.9421939849853516, + "learning_rate": 3.426076360682372e-06, + "loss": 2.1794, + "step": 1064 + }, + { + "epoch": 1.167123287671233, + "grad_norm": 2.212996006011963, + "learning_rate": 3.42404549147035e-06, + "loss": 2.1877, + "step": 1065 + }, + { + "epoch": 1.1682191780821918, + "grad_norm": 2.036180257797241, + "learning_rate": 3.4220146222583267e-06, + "loss": 2.1534, + "step": 1066 + }, + { + "epoch": 1.1693150684931506, + "grad_norm": 1.8819142580032349, + "learning_rate": 3.4199837530463044e-06, + "loss": 2.099, + "step": 1067 + }, + { + "epoch": 1.1704109589041096, + "grad_norm": 2.0781915187835693, + "learning_rate": 3.4179528838342813e-06, + "loss": 2.1326, + "step": 1068 + }, + { + "epoch": 1.1715068493150684, + "grad_norm": 2.034895181655884, + "learning_rate": 3.415922014622259e-06, + "loss": 2.2107, + "step": 1069 + }, + { + "epoch": 1.1726027397260275, + "grad_norm": 1.92604660987854, + "learning_rate": 3.413891145410236e-06, + "loss": 2.1773, + "step": 1070 + }, + { + "epoch": 1.1726027397260275, + "eval_loss": 2.2959954738616943, + "eval_runtime": 47.2123, + "eval_samples_per_second": 21.181, + "eval_steps_per_second": 0.89, + "step": 1070 + }, + { + "epoch": 1.1736986301369863, + "grad_norm": 1.762730598449707, + "learning_rate": 3.4118602761982135e-06, + "loss": 2.1803, + "step": 1071 + }, + { + "epoch": 1.174794520547945, + "grad_norm": 2.0661215782165527, + "learning_rate": 3.4098294069861904e-06, + "loss": 2.1919, + "step": 1072 + }, + { + "epoch": 1.1758904109589041, + "grad_norm": 2.079982280731201, + "learning_rate": 3.4077985377741673e-06, + "loss": 2.1536, + "step": 1073 + }, + { + "epoch": 1.176986301369863, + "grad_norm": 1.9044359922409058, + "learning_rate": 3.405767668562145e-06, + "loss": 2.1403, + "step": 1074 + }, + { + "epoch": 1.178082191780822, + "grad_norm": 1.8870306015014648, + "learning_rate": 3.403736799350122e-06, + "loss": 2.1845, + "step": 1075 + }, + { + "epoch": 1.1791780821917808, + "grad_norm": 1.9305400848388672, + "learning_rate": 3.4017059301380995e-06, + "loss": 2.1786, + "step": 1076 + }, + { + "epoch": 1.1802739726027398, + "grad_norm": 1.987713098526001, + "learning_rate": 3.3996750609260764e-06, + "loss": 2.1214, + "step": 1077 + }, + { + "epoch": 1.1813698630136986, + "grad_norm": 1.8673466444015503, + "learning_rate": 3.397644191714054e-06, + "loss": 2.2027, + "step": 1078 + }, + { + "epoch": 1.1824657534246574, + "grad_norm": 2.009316921234131, + "learning_rate": 3.395613322502031e-06, + "loss": 2.139, + "step": 1079 + }, + { + "epoch": 1.1835616438356165, + "grad_norm": 2.1024956703186035, + "learning_rate": 3.3935824532900086e-06, + "loss": 2.1438, + "step": 1080 + }, + { + "epoch": 1.1835616438356165, + "eval_loss": 2.2966299057006836, + "eval_runtime": 44.8946, + "eval_samples_per_second": 22.274, + "eval_steps_per_second": 0.936, + "step": 1080 + }, + { + "epoch": 1.1846575342465753, + "grad_norm": 1.8911967277526855, + "learning_rate": 3.3915515840779855e-06, + "loss": 2.1238, + "step": 1081 + }, + { + "epoch": 1.1857534246575343, + "grad_norm": 2.0649192333221436, + "learning_rate": 3.389520714865963e-06, + "loss": 2.1534, + "step": 1082 + }, + { + "epoch": 1.1868493150684931, + "grad_norm": 1.9236847162246704, + "learning_rate": 3.38748984565394e-06, + "loss": 2.1427, + "step": 1083 + }, + { + "epoch": 1.1879452054794521, + "grad_norm": 1.8662182092666626, + "learning_rate": 3.3854589764419177e-06, + "loss": 2.1699, + "step": 1084 + }, + { + "epoch": 1.189041095890411, + "grad_norm": 1.9226020574569702, + "learning_rate": 3.3834281072298946e-06, + "loss": 2.1192, + "step": 1085 + }, + { + "epoch": 1.1901369863013698, + "grad_norm": 1.9659301042556763, + "learning_rate": 3.381397238017872e-06, + "loss": 2.1529, + "step": 1086 + }, + { + "epoch": 1.1912328767123288, + "grad_norm": 2.2714693546295166, + "learning_rate": 3.379366368805849e-06, + "loss": 2.1273, + "step": 1087 + }, + { + "epoch": 1.1923287671232876, + "grad_norm": 2.1853270530700684, + "learning_rate": 3.3773354995938264e-06, + "loss": 2.2029, + "step": 1088 + }, + { + "epoch": 1.1934246575342466, + "grad_norm": 1.8302830457687378, + "learning_rate": 3.3753046303818037e-06, + "loss": 2.1599, + "step": 1089 + }, + { + "epoch": 1.1945205479452055, + "grad_norm": 2.005591869354248, + "learning_rate": 3.373273761169781e-06, + "loss": 2.139, + "step": 1090 + }, + { + "epoch": 1.1945205479452055, + "eval_loss": 2.295994520187378, + "eval_runtime": 45.4802, + "eval_samples_per_second": 21.988, + "eval_steps_per_second": 0.923, + "step": 1090 + }, + { + "epoch": 1.1956164383561645, + "grad_norm": 2.139267683029175, + "learning_rate": 3.3712428919577582e-06, + "loss": 2.151, + "step": 1091 + }, + { + "epoch": 1.1967123287671233, + "grad_norm": 1.8547207117080688, + "learning_rate": 3.3692120227457355e-06, + "loss": 2.1249, + "step": 1092 + }, + { + "epoch": 1.1978082191780821, + "grad_norm": 2.214583396911621, + "learning_rate": 3.3671811535337128e-06, + "loss": 2.1656, + "step": 1093 + }, + { + "epoch": 1.1989041095890411, + "grad_norm": 2.1608800888061523, + "learning_rate": 3.36515028432169e-06, + "loss": 2.1056, + "step": 1094 + }, + { + "epoch": 1.2, + "grad_norm": 2.0850605964660645, + "learning_rate": 3.3631194151096673e-06, + "loss": 2.1215, + "step": 1095 + }, + { + "epoch": 1.201095890410959, + "grad_norm": 1.9199103116989136, + "learning_rate": 3.3610885458976446e-06, + "loss": 2.1584, + "step": 1096 + }, + { + "epoch": 1.2021917808219178, + "grad_norm": 1.9766334295272827, + "learning_rate": 3.359057676685622e-06, + "loss": 2.1821, + "step": 1097 + }, + { + "epoch": 1.2032876712328768, + "grad_norm": 1.948988676071167, + "learning_rate": 3.357026807473599e-06, + "loss": 2.182, + "step": 1098 + }, + { + "epoch": 1.2043835616438356, + "grad_norm": 2.0419397354125977, + "learning_rate": 3.354995938261576e-06, + "loss": 2.1051, + "step": 1099 + }, + { + "epoch": 1.2054794520547945, + "grad_norm": 1.9751213788986206, + "learning_rate": 3.3529650690495537e-06, + "loss": 2.1255, + "step": 1100 + }, + { + "epoch": 1.2054794520547945, + "eval_loss": 2.293811798095703, + "eval_runtime": 46.1182, + "eval_samples_per_second": 21.683, + "eval_steps_per_second": 0.911, + "step": 1100 + }, + { + "epoch": 1.2065753424657535, + "grad_norm": 1.9806510210037231, + "learning_rate": 3.3509341998375305e-06, + "loss": 2.1021, + "step": 1101 + }, + { + "epoch": 1.2076712328767123, + "grad_norm": 2.0570414066314697, + "learning_rate": 3.3489033306255082e-06, + "loss": 2.1417, + "step": 1102 + }, + { + "epoch": 1.208767123287671, + "grad_norm": 1.868134617805481, + "learning_rate": 3.346872461413485e-06, + "loss": 2.1494, + "step": 1103 + }, + { + "epoch": 1.2098630136986301, + "grad_norm": 2.0156848430633545, + "learning_rate": 3.344841592201463e-06, + "loss": 2.1238, + "step": 1104 + }, + { + "epoch": 1.210958904109589, + "grad_norm": 2.082301616668701, + "learning_rate": 3.3428107229894397e-06, + "loss": 2.1105, + "step": 1105 + }, + { + "epoch": 1.212054794520548, + "grad_norm": 1.8853904008865356, + "learning_rate": 3.3407798537774174e-06, + "loss": 2.1406, + "step": 1106 + }, + { + "epoch": 1.2131506849315068, + "grad_norm": 2.0410518646240234, + "learning_rate": 3.338748984565394e-06, + "loss": 2.1567, + "step": 1107 + }, + { + "epoch": 1.2142465753424658, + "grad_norm": 1.9901232719421387, + "learning_rate": 3.336718115353372e-06, + "loss": 2.1778, + "step": 1108 + }, + { + "epoch": 1.2153424657534246, + "grad_norm": 1.9840792417526245, + "learning_rate": 3.3346872461413488e-06, + "loss": 2.1935, + "step": 1109 + }, + { + "epoch": 1.2164383561643834, + "grad_norm": 1.8854445219039917, + "learning_rate": 3.3326563769293256e-06, + "loss": 2.137, + "step": 1110 + }, + { + "epoch": 1.2164383561643834, + "eval_loss": 2.294447660446167, + "eval_runtime": 42.6423, + "eval_samples_per_second": 23.451, + "eval_steps_per_second": 0.985, + "step": 1110 + }, + { + "epoch": 1.2175342465753425, + "grad_norm": 2.0948665142059326, + "learning_rate": 3.3306255077173033e-06, + "loss": 2.1288, + "step": 1111 + }, + { + "epoch": 1.2186301369863013, + "grad_norm": 1.859448790550232, + "learning_rate": 3.32859463850528e-06, + "loss": 2.1403, + "step": 1112 + }, + { + "epoch": 1.2197260273972603, + "grad_norm": 1.92074453830719, + "learning_rate": 3.326563769293258e-06, + "loss": 2.141, + "step": 1113 + }, + { + "epoch": 1.2208219178082191, + "grad_norm": 1.8377320766448975, + "learning_rate": 3.3245329000812347e-06, + "loss": 2.0889, + "step": 1114 + }, + { + "epoch": 1.2219178082191782, + "grad_norm": 1.9320917129516602, + "learning_rate": 3.3225020308692124e-06, + "loss": 2.1254, + "step": 1115 + }, + { + "epoch": 1.223013698630137, + "grad_norm": 1.9691251516342163, + "learning_rate": 3.3204711616571893e-06, + "loss": 2.1409, + "step": 1116 + }, + { + "epoch": 1.2241095890410958, + "grad_norm": 1.8214130401611328, + "learning_rate": 3.318440292445167e-06, + "loss": 2.1046, + "step": 1117 + }, + { + "epoch": 1.2252054794520548, + "grad_norm": 1.959455966949463, + "learning_rate": 3.316409423233144e-06, + "loss": 2.1418, + "step": 1118 + }, + { + "epoch": 1.2263013698630136, + "grad_norm": 1.9920449256896973, + "learning_rate": 3.3143785540211215e-06, + "loss": 2.1192, + "step": 1119 + }, + { + "epoch": 1.2273972602739727, + "grad_norm": 2.0276854038238525, + "learning_rate": 3.3123476848090984e-06, + "loss": 2.1939, + "step": 1120 + }, + { + "epoch": 1.2273972602739727, + "eval_loss": 2.295330047607422, + "eval_runtime": 49.8691, + "eval_samples_per_second": 20.052, + "eval_steps_per_second": 0.842, + "step": 1120 + }, + { + "epoch": 1.2284931506849315, + "grad_norm": 1.9765129089355469, + "learning_rate": 3.310316815597076e-06, + "loss": 2.1545, + "step": 1121 + }, + { + "epoch": 1.2295890410958905, + "grad_norm": 1.8449276685714722, + "learning_rate": 3.308285946385053e-06, + "loss": 2.1459, + "step": 1122 + }, + { + "epoch": 1.2306849315068493, + "grad_norm": 1.953072190284729, + "learning_rate": 3.30625507717303e-06, + "loss": 2.2046, + "step": 1123 + }, + { + "epoch": 1.2317808219178081, + "grad_norm": 1.9570913314819336, + "learning_rate": 3.3042242079610075e-06, + "loss": 2.1015, + "step": 1124 + }, + { + "epoch": 1.2328767123287672, + "grad_norm": 2.00569224357605, + "learning_rate": 3.3021933387489847e-06, + "loss": 2.1681, + "step": 1125 + }, + { + "epoch": 1.233972602739726, + "grad_norm": 1.9131590127944946, + "learning_rate": 3.300162469536962e-06, + "loss": 2.1765, + "step": 1126 + }, + { + "epoch": 1.235068493150685, + "grad_norm": 1.8375864028930664, + "learning_rate": 3.2981316003249393e-06, + "loss": 2.1175, + "step": 1127 + }, + { + "epoch": 1.2361643835616438, + "grad_norm": 1.852847933769226, + "learning_rate": 3.2961007311129166e-06, + "loss": 2.1838, + "step": 1128 + }, + { + "epoch": 1.2372602739726029, + "grad_norm": 1.8772693872451782, + "learning_rate": 3.294069861900894e-06, + "loss": 2.1524, + "step": 1129 + }, + { + "epoch": 1.2383561643835617, + "grad_norm": 1.8593205213546753, + "learning_rate": 3.292038992688871e-06, + "loss": 2.2059, + "step": 1130 + }, + { + "epoch": 1.2383561643835617, + "eval_loss": 2.294188976287842, + "eval_runtime": 50.4549, + "eval_samples_per_second": 19.82, + "eval_steps_per_second": 0.832, + "step": 1130 + }, + { + "epoch": 1.2394520547945205, + "grad_norm": 1.8967972993850708, + "learning_rate": 3.2900081234768484e-06, + "loss": 2.1551, + "step": 1131 + }, + { + "epoch": 1.2405479452054795, + "grad_norm": 1.9799107313156128, + "learning_rate": 3.2879772542648257e-06, + "loss": 2.1644, + "step": 1132 + }, + { + "epoch": 1.2416438356164383, + "grad_norm": 1.978883147239685, + "learning_rate": 3.285946385052803e-06, + "loss": 2.182, + "step": 1133 + }, + { + "epoch": 1.2427397260273974, + "grad_norm": 1.8705304861068726, + "learning_rate": 3.2839155158407802e-06, + "loss": 2.0981, + "step": 1134 + }, + { + "epoch": 1.2438356164383562, + "grad_norm": 2.0146889686584473, + "learning_rate": 3.2818846466287575e-06, + "loss": 2.1686, + "step": 1135 + }, + { + "epoch": 1.244931506849315, + "grad_norm": 1.94461989402771, + "learning_rate": 3.2798537774167344e-06, + "loss": 2.093, + "step": 1136 + }, + { + "epoch": 1.246027397260274, + "grad_norm": 1.7823240756988525, + "learning_rate": 3.277822908204712e-06, + "loss": 2.198, + "step": 1137 + }, + { + "epoch": 1.2471232876712328, + "grad_norm": 1.89297354221344, + "learning_rate": 3.275792038992689e-06, + "loss": 2.1359, + "step": 1138 + }, + { + "epoch": 1.2482191780821918, + "grad_norm": 1.8670459985733032, + "learning_rate": 3.2737611697806666e-06, + "loss": 2.0956, + "step": 1139 + }, + { + "epoch": 1.2493150684931507, + "grad_norm": 1.813568353652954, + "learning_rate": 3.2717303005686435e-06, + "loss": 2.1448, + "step": 1140 + }, + { + "epoch": 1.2493150684931507, + "eval_loss": 2.295804023742676, + "eval_runtime": 53.286, + "eval_samples_per_second": 18.767, + "eval_steps_per_second": 0.788, + "step": 1140 + }, + { + "epoch": 1.2504109589041095, + "grad_norm": 1.912916898727417, + "learning_rate": 3.269699431356621e-06, + "loss": 2.1159, + "step": 1141 + }, + { + "epoch": 1.2515068493150685, + "grad_norm": 1.7939883470535278, + "learning_rate": 3.267668562144598e-06, + "loss": 2.1566, + "step": 1142 + }, + { + "epoch": 1.2526027397260273, + "grad_norm": 1.8494806289672852, + "learning_rate": 3.2656376929325757e-06, + "loss": 2.1496, + "step": 1143 + }, + { + "epoch": 1.2536986301369863, + "grad_norm": 1.8468523025512695, + "learning_rate": 3.2636068237205526e-06, + "loss": 2.1159, + "step": 1144 + }, + { + "epoch": 1.2547945205479452, + "grad_norm": 1.8962055444717407, + "learning_rate": 3.2615759545085303e-06, + "loss": 2.1712, + "step": 1145 + }, + { + "epoch": 1.2558904109589042, + "grad_norm": 1.800353765487671, + "learning_rate": 3.259545085296507e-06, + "loss": 2.1486, + "step": 1146 + }, + { + "epoch": 1.256986301369863, + "grad_norm": 2.124359607696533, + "learning_rate": 3.257514216084484e-06, + "loss": 2.1913, + "step": 1147 + }, + { + "epoch": 1.2580821917808218, + "grad_norm": 1.9248145818710327, + "learning_rate": 3.2554833468724617e-06, + "loss": 2.1128, + "step": 1148 + }, + { + "epoch": 1.2591780821917808, + "grad_norm": 1.9990876913070679, + "learning_rate": 3.2534524776604385e-06, + "loss": 2.1854, + "step": 1149 + }, + { + "epoch": 1.2602739726027397, + "grad_norm": 1.8600568771362305, + "learning_rate": 3.2514216084484162e-06, + "loss": 2.1565, + "step": 1150 + }, + { + "epoch": 1.2602739726027397, + "eval_loss": 2.2942006587982178, + "eval_runtime": 51.4021, + "eval_samples_per_second": 19.454, + "eval_steps_per_second": 0.817, + "step": 1150 + }, + { + "epoch": 1.2613698630136987, + "grad_norm": 1.8621101379394531, + "learning_rate": 3.249390739236393e-06, + "loss": 2.1438, + "step": 1151 + }, + { + "epoch": 1.2624657534246575, + "grad_norm": 1.9290356636047363, + "learning_rate": 3.2473598700243708e-06, + "loss": 2.1577, + "step": 1152 + }, + { + "epoch": 1.2635616438356165, + "grad_norm": 1.9007307291030884, + "learning_rate": 3.2453290008123476e-06, + "loss": 2.1295, + "step": 1153 + }, + { + "epoch": 1.2646575342465753, + "grad_norm": 1.7647960186004639, + "learning_rate": 3.2432981316003253e-06, + "loss": 2.1472, + "step": 1154 + }, + { + "epoch": 1.2657534246575342, + "grad_norm": 1.9796825647354126, + "learning_rate": 3.241267262388302e-06, + "loss": 2.1542, + "step": 1155 + }, + { + "epoch": 1.2668493150684932, + "grad_norm": 1.9505796432495117, + "learning_rate": 3.23923639317628e-06, + "loss": 2.1093, + "step": 1156 + }, + { + "epoch": 1.267945205479452, + "grad_norm": 1.9152203798294067, + "learning_rate": 3.2372055239642567e-06, + "loss": 2.1456, + "step": 1157 + }, + { + "epoch": 1.269041095890411, + "grad_norm": 1.9094963073730469, + "learning_rate": 3.2351746547522344e-06, + "loss": 2.1149, + "step": 1158 + }, + { + "epoch": 1.2701369863013698, + "grad_norm": 1.8872679471969604, + "learning_rate": 3.2331437855402113e-06, + "loss": 2.1545, + "step": 1159 + }, + { + "epoch": 1.2712328767123289, + "grad_norm": 2.171703338623047, + "learning_rate": 3.2311129163281885e-06, + "loss": 2.2108, + "step": 1160 + }, + { + "epoch": 1.2712328767123289, + "eval_loss": 2.2910356521606445, + "eval_runtime": 52.5176, + "eval_samples_per_second": 19.041, + "eval_steps_per_second": 0.8, + "step": 1160 + }, + { + "epoch": 1.2723287671232877, + "grad_norm": 1.8578304052352905, + "learning_rate": 3.229082047116166e-06, + "loss": 2.1158, + "step": 1161 + }, + { + "epoch": 1.2734246575342465, + "grad_norm": 1.8873049020767212, + "learning_rate": 3.227051177904143e-06, + "loss": 2.1618, + "step": 1162 + }, + { + "epoch": 1.2745205479452055, + "grad_norm": 1.8442409038543701, + "learning_rate": 3.2250203086921204e-06, + "loss": 2.1454, + "step": 1163 + }, + { + "epoch": 1.2756164383561643, + "grad_norm": 1.9609211683273315, + "learning_rate": 3.2229894394800977e-06, + "loss": 2.1519, + "step": 1164 + }, + { + "epoch": 1.2767123287671232, + "grad_norm": 1.9093680381774902, + "learning_rate": 3.220958570268075e-06, + "loss": 2.1775, + "step": 1165 + }, + { + "epoch": 1.2778082191780822, + "grad_norm": 1.8355298042297363, + "learning_rate": 3.218927701056052e-06, + "loss": 2.0871, + "step": 1166 + }, + { + "epoch": 1.2789041095890412, + "grad_norm": 1.9497056007385254, + "learning_rate": 3.2168968318440295e-06, + "loss": 2.1441, + "step": 1167 + }, + { + "epoch": 1.28, + "grad_norm": 1.8635729551315308, + "learning_rate": 3.2148659626320068e-06, + "loss": 2.0748, + "step": 1168 + }, + { + "epoch": 1.2810958904109588, + "grad_norm": 2.0972132682800293, + "learning_rate": 3.212835093419984e-06, + "loss": 2.1843, + "step": 1169 + }, + { + "epoch": 1.2821917808219179, + "grad_norm": 1.9235926866531372, + "learning_rate": 3.2108042242079613e-06, + "loss": 2.1194, + "step": 1170 + }, + { + "epoch": 1.2821917808219179, + "eval_loss": 2.291309118270874, + "eval_runtime": 55.6408, + "eval_samples_per_second": 17.972, + "eval_steps_per_second": 0.755, + "step": 1170 + }, + { + "epoch": 1.2832876712328767, + "grad_norm": 1.8674591779708862, + "learning_rate": 3.2087733549959386e-06, + "loss": 2.1264, + "step": 1171 + }, + { + "epoch": 1.2843835616438355, + "grad_norm": 1.810589075088501, + "learning_rate": 3.206742485783916e-06, + "loss": 2.1781, + "step": 1172 + }, + { + "epoch": 1.2854794520547945, + "grad_norm": 1.9794505834579468, + "learning_rate": 3.2047116165718927e-06, + "loss": 2.1089, + "step": 1173 + }, + { + "epoch": 1.2865753424657536, + "grad_norm": 1.8487070798873901, + "learning_rate": 3.2026807473598704e-06, + "loss": 2.097, + "step": 1174 + }, + { + "epoch": 1.2876712328767124, + "grad_norm": 1.9668854475021362, + "learning_rate": 3.2006498781478473e-06, + "loss": 2.1516, + "step": 1175 + }, + { + "epoch": 1.2887671232876712, + "grad_norm": 1.855817198753357, + "learning_rate": 3.198619008935825e-06, + "loss": 2.1678, + "step": 1176 + }, + { + "epoch": 1.2898630136986302, + "grad_norm": 2.381211757659912, + "learning_rate": 3.196588139723802e-06, + "loss": 2.099, + "step": 1177 + }, + { + "epoch": 1.290958904109589, + "grad_norm": 2.1524250507354736, + "learning_rate": 3.1945572705117795e-06, + "loss": 2.1345, + "step": 1178 + }, + { + "epoch": 1.2920547945205478, + "grad_norm": 1.9120781421661377, + "learning_rate": 3.1925264012997564e-06, + "loss": 2.1586, + "step": 1179 + }, + { + "epoch": 1.2931506849315069, + "grad_norm": 2.153817653656006, + "learning_rate": 3.190495532087734e-06, + "loss": 2.1597, + "step": 1180 + }, + { + "epoch": 1.2931506849315069, + "eval_loss": 2.2936320304870605, + "eval_runtime": 53.6426, + "eval_samples_per_second": 18.642, + "eval_steps_per_second": 0.783, + "step": 1180 + }, + { + "epoch": 1.2942465753424657, + "grad_norm": 2.074873447418213, + "learning_rate": 3.188464662875711e-06, + "loss": 2.1572, + "step": 1181 + }, + { + "epoch": 1.2953424657534247, + "grad_norm": 1.9313570261001587, + "learning_rate": 3.1864337936636886e-06, + "loss": 2.1986, + "step": 1182 + }, + { + "epoch": 1.2964383561643835, + "grad_norm": 1.9836641550064087, + "learning_rate": 3.1844029244516655e-06, + "loss": 2.13, + "step": 1183 + }, + { + "epoch": 1.2975342465753426, + "grad_norm": 2.0893137454986572, + "learning_rate": 3.182372055239643e-06, + "loss": 2.1138, + "step": 1184 + }, + { + "epoch": 1.2986301369863014, + "grad_norm": 2.1667087078094482, + "learning_rate": 3.18034118602762e-06, + "loss": 2.1119, + "step": 1185 + }, + { + "epoch": 1.2997260273972602, + "grad_norm": 2.093961000442505, + "learning_rate": 3.1783103168155973e-06, + "loss": 2.1728, + "step": 1186 + }, + { + "epoch": 1.3008219178082192, + "grad_norm": 1.919267177581787, + "learning_rate": 3.1762794476035746e-06, + "loss": 2.1469, + "step": 1187 + }, + { + "epoch": 1.301917808219178, + "grad_norm": 2.017967939376831, + "learning_rate": 3.174248578391552e-06, + "loss": 2.1307, + "step": 1188 + }, + { + "epoch": 1.303013698630137, + "grad_norm": 1.921632170677185, + "learning_rate": 3.172217709179529e-06, + "loss": 2.13, + "step": 1189 + }, + { + "epoch": 1.3041095890410959, + "grad_norm": 2.0171291828155518, + "learning_rate": 3.1701868399675064e-06, + "loss": 2.135, + "step": 1190 + }, + { + "epoch": 1.3041095890410959, + "eval_loss": 2.2934045791625977, + "eval_runtime": 48.0821, + "eval_samples_per_second": 20.798, + "eval_steps_per_second": 0.874, + "step": 1190 + }, + { + "epoch": 1.305205479452055, + "grad_norm": 1.8685601949691772, + "learning_rate": 3.1681559707554837e-06, + "loss": 2.1329, + "step": 1191 + }, + { + "epoch": 1.3063013698630137, + "grad_norm": 1.8909096717834473, + "learning_rate": 3.166125101543461e-06, + "loss": 2.1004, + "step": 1192 + }, + { + "epoch": 1.3073972602739725, + "grad_norm": 1.798614740371704, + "learning_rate": 3.1640942323314382e-06, + "loss": 2.1555, + "step": 1193 + }, + { + "epoch": 1.3084931506849315, + "grad_norm": 1.8296692371368408, + "learning_rate": 3.1620633631194155e-06, + "loss": 2.1549, + "step": 1194 + }, + { + "epoch": 1.3095890410958904, + "grad_norm": 1.8676908016204834, + "learning_rate": 3.1600324939073928e-06, + "loss": 2.1081, + "step": 1195 + }, + { + "epoch": 1.3106849315068494, + "grad_norm": 1.8682451248168945, + "learning_rate": 3.15800162469537e-06, + "loss": 2.138, + "step": 1196 + }, + { + "epoch": 1.3117808219178082, + "grad_norm": 2.0049314498901367, + "learning_rate": 3.155970755483347e-06, + "loss": 2.144, + "step": 1197 + }, + { + "epoch": 1.3128767123287672, + "grad_norm": 2.00691556930542, + "learning_rate": 3.1539398862713246e-06, + "loss": 2.1539, + "step": 1198 + }, + { + "epoch": 1.313972602739726, + "grad_norm": 1.8521753549575806, + "learning_rate": 3.1519090170593015e-06, + "loss": 2.1597, + "step": 1199 + }, + { + "epoch": 1.3150684931506849, + "grad_norm": 2.0974042415618896, + "learning_rate": 3.149878147847279e-06, + "loss": 2.147, + "step": 1200 + }, + { + "epoch": 1.3150684931506849, + "eval_loss": 2.2891385555267334, + "eval_runtime": 48.1618, + "eval_samples_per_second": 20.763, + "eval_steps_per_second": 0.872, + "step": 1200 + }, + { + "epoch": 1.316164383561644, + "grad_norm": 1.9671052694320679, + "learning_rate": 3.147847278635256e-06, + "loss": 2.1838, + "step": 1201 + }, + { + "epoch": 1.3172602739726027, + "grad_norm": 1.9421573877334595, + "learning_rate": 3.1458164094232337e-06, + "loss": 2.2093, + "step": 1202 + }, + { + "epoch": 1.3183561643835615, + "grad_norm": 1.917198657989502, + "learning_rate": 3.1437855402112106e-06, + "loss": 2.136, + "step": 1203 + }, + { + "epoch": 1.3194520547945205, + "grad_norm": 1.8664982318878174, + "learning_rate": 3.1417546709991883e-06, + "loss": 2.089, + "step": 1204 + }, + { + "epoch": 1.3205479452054796, + "grad_norm": 2.043558359146118, + "learning_rate": 3.139723801787165e-06, + "loss": 2.159, + "step": 1205 + }, + { + "epoch": 1.3216438356164384, + "grad_norm": 1.8314157724380493, + "learning_rate": 3.137692932575143e-06, + "loss": 2.13, + "step": 1206 + }, + { + "epoch": 1.3227397260273972, + "grad_norm": 1.907884120941162, + "learning_rate": 3.1356620633631197e-06, + "loss": 2.1746, + "step": 1207 + }, + { + "epoch": 1.3238356164383562, + "grad_norm": 2.0570952892303467, + "learning_rate": 3.1336311941510974e-06, + "loss": 2.1187, + "step": 1208 + }, + { + "epoch": 1.324931506849315, + "grad_norm": 1.832292914390564, + "learning_rate": 3.1316003249390742e-06, + "loss": 2.1425, + "step": 1209 + }, + { + "epoch": 1.3260273972602739, + "grad_norm": 1.8307697772979736, + "learning_rate": 3.129569455727051e-06, + "loss": 2.1749, + "step": 1210 + }, + { + "epoch": 1.3260273972602739, + "eval_loss": 2.2865941524505615, + "eval_runtime": 42.2877, + "eval_samples_per_second": 23.648, + "eval_steps_per_second": 0.993, + "step": 1210 + }, + { + "epoch": 1.3271232876712329, + "grad_norm": 1.8292433023452759, + "learning_rate": 3.1275385865150288e-06, + "loss": 2.1702, + "step": 1211 + }, + { + "epoch": 1.3282191780821917, + "grad_norm": 1.9146367311477661, + "learning_rate": 3.1255077173030056e-06, + "loss": 2.16, + "step": 1212 + }, + { + "epoch": 1.3293150684931507, + "grad_norm": 1.809275507926941, + "learning_rate": 3.1234768480909833e-06, + "loss": 2.157, + "step": 1213 + }, + { + "epoch": 1.3304109589041095, + "grad_norm": 1.939081072807312, + "learning_rate": 3.12144597887896e-06, + "loss": 2.1496, + "step": 1214 + }, + { + "epoch": 1.3315068493150686, + "grad_norm": 1.8665997982025146, + "learning_rate": 3.119415109666938e-06, + "loss": 2.1535, + "step": 1215 + }, + { + "epoch": 1.3326027397260274, + "grad_norm": 1.8455616235733032, + "learning_rate": 3.1173842404549147e-06, + "loss": 2.1845, + "step": 1216 + }, + { + "epoch": 1.3336986301369862, + "grad_norm": 2.050560235977173, + "learning_rate": 3.1153533712428924e-06, + "loss": 2.1461, + "step": 1217 + }, + { + "epoch": 1.3347945205479452, + "grad_norm": 1.8440574407577515, + "learning_rate": 3.1133225020308693e-06, + "loss": 2.1476, + "step": 1218 + }, + { + "epoch": 1.335890410958904, + "grad_norm": 1.8801573514938354, + "learning_rate": 3.111291632818847e-06, + "loss": 2.1721, + "step": 1219 + }, + { + "epoch": 1.336986301369863, + "grad_norm": 1.9227471351623535, + "learning_rate": 3.109260763606824e-06, + "loss": 2.178, + "step": 1220 + }, + { + "epoch": 1.336986301369863, + "eval_loss": 2.2896478176116943, + "eval_runtime": 43.157, + "eval_samples_per_second": 23.171, + "eval_steps_per_second": 0.973, + "step": 1220 + }, + { + "epoch": 1.3380821917808219, + "grad_norm": 1.9979732036590576, + "learning_rate": 3.1072298943948015e-06, + "loss": 2.148, + "step": 1221 + }, + { + "epoch": 1.339178082191781, + "grad_norm": 1.9481284618377686, + "learning_rate": 3.1051990251827784e-06, + "loss": 2.165, + "step": 1222 + }, + { + "epoch": 1.3402739726027397, + "grad_norm": 1.8830820322036743, + "learning_rate": 3.1031681559707557e-06, + "loss": 2.1239, + "step": 1223 + }, + { + "epoch": 1.3413698630136985, + "grad_norm": 1.890512466430664, + "learning_rate": 3.101137286758733e-06, + "loss": 2.1761, + "step": 1224 + }, + { + "epoch": 1.3424657534246576, + "grad_norm": 2.0024659633636475, + "learning_rate": 3.09910641754671e-06, + "loss": 2.179, + "step": 1225 + }, + { + "epoch": 1.3435616438356164, + "grad_norm": 1.8659477233886719, + "learning_rate": 3.0970755483346875e-06, + "loss": 2.1722, + "step": 1226 + }, + { + "epoch": 1.3446575342465754, + "grad_norm": 1.93405020236969, + "learning_rate": 3.0950446791226648e-06, + "loss": 2.1545, + "step": 1227 + }, + { + "epoch": 1.3457534246575342, + "grad_norm": 1.9514124393463135, + "learning_rate": 3.093013809910642e-06, + "loss": 2.187, + "step": 1228 + }, + { + "epoch": 1.3468493150684933, + "grad_norm": 1.9571877717971802, + "learning_rate": 3.0909829406986193e-06, + "loss": 2.1507, + "step": 1229 + }, + { + "epoch": 1.347945205479452, + "grad_norm": 1.9434577226638794, + "learning_rate": 3.0889520714865966e-06, + "loss": 2.1953, + "step": 1230 + }, + { + "epoch": 1.347945205479452, + "eval_loss": 2.289128065109253, + "eval_runtime": 46.7151, + "eval_samples_per_second": 21.406, + "eval_steps_per_second": 0.899, + "step": 1230 + }, + { + "epoch": 1.3490410958904109, + "grad_norm": 1.8294577598571777, + "learning_rate": 3.086921202274574e-06, + "loss": 2.1684, + "step": 1231 + }, + { + "epoch": 1.35013698630137, + "grad_norm": 1.9432605504989624, + "learning_rate": 3.084890333062551e-06, + "loss": 2.1393, + "step": 1232 + }, + { + "epoch": 1.3512328767123287, + "grad_norm": 1.9030417203903198, + "learning_rate": 3.0828594638505284e-06, + "loss": 2.1181, + "step": 1233 + }, + { + "epoch": 1.3523287671232875, + "grad_norm": 1.9233827590942383, + "learning_rate": 3.0808285946385057e-06, + "loss": 2.1479, + "step": 1234 + }, + { + "epoch": 1.3534246575342466, + "grad_norm": 1.9549237489700317, + "learning_rate": 3.078797725426483e-06, + "loss": 2.1253, + "step": 1235 + }, + { + "epoch": 1.3545205479452056, + "grad_norm": 2.0584867000579834, + "learning_rate": 3.07676685621446e-06, + "loss": 2.0846, + "step": 1236 + }, + { + "epoch": 1.3556164383561644, + "grad_norm": 1.7543913125991821, + "learning_rate": 3.0747359870024375e-06, + "loss": 2.1495, + "step": 1237 + }, + { + "epoch": 1.3567123287671232, + "grad_norm": 1.750698447227478, + "learning_rate": 3.0727051177904144e-06, + "loss": 2.1179, + "step": 1238 + }, + { + "epoch": 1.3578082191780823, + "grad_norm": 1.9867119789123535, + "learning_rate": 3.070674248578392e-06, + "loss": 2.1313, + "step": 1239 + }, + { + "epoch": 1.358904109589041, + "grad_norm": 1.8228135108947754, + "learning_rate": 3.068643379366369e-06, + "loss": 2.158, + "step": 1240 + }, + { + "epoch": 1.358904109589041, + "eval_loss": 2.2861993312835693, + "eval_runtime": 44.081, + "eval_samples_per_second": 22.686, + "eval_steps_per_second": 0.953, + "step": 1240 + }, + { + "epoch": 1.3599999999999999, + "grad_norm": 1.8551791906356812, + "learning_rate": 3.0666125101543466e-06, + "loss": 2.129, + "step": 1241 + }, + { + "epoch": 1.361095890410959, + "grad_norm": 1.8729921579360962, + "learning_rate": 3.0645816409423235e-06, + "loss": 2.1069, + "step": 1242 + }, + { + "epoch": 1.362191780821918, + "grad_norm": 1.8203959465026855, + "learning_rate": 3.062550771730301e-06, + "loss": 2.1509, + "step": 1243 + }, + { + "epoch": 1.3632876712328768, + "grad_norm": 1.9498761892318726, + "learning_rate": 3.060519902518278e-06, + "loss": 2.1336, + "step": 1244 + }, + { + "epoch": 1.3643835616438356, + "grad_norm": 2.134521245956421, + "learning_rate": 3.0584890333062557e-06, + "loss": 2.1294, + "step": 1245 + }, + { + "epoch": 1.3654794520547946, + "grad_norm": 1.9321913719177246, + "learning_rate": 3.0564581640942326e-06, + "loss": 2.0823, + "step": 1246 + }, + { + "epoch": 1.3665753424657534, + "grad_norm": 1.8948324918746948, + "learning_rate": 3.0544272948822094e-06, + "loss": 2.1425, + "step": 1247 + }, + { + "epoch": 1.3676712328767122, + "grad_norm": 2.2063205242156982, + "learning_rate": 3.052396425670187e-06, + "loss": 2.1242, + "step": 1248 + }, + { + "epoch": 1.3687671232876713, + "grad_norm": 2.0060935020446777, + "learning_rate": 3.050365556458164e-06, + "loss": 2.1586, + "step": 1249 + }, + { + "epoch": 1.36986301369863, + "grad_norm": 1.8441938161849976, + "learning_rate": 3.0483346872461417e-06, + "loss": 2.1305, + "step": 1250 + }, + { + "epoch": 1.36986301369863, + "eval_loss": 2.2872142791748047, + "eval_runtime": 43.4998, + "eval_samples_per_second": 22.989, + "eval_steps_per_second": 0.966, + "step": 1250 + }, + { + "epoch": 1.370958904109589, + "grad_norm": 2.019545793533325, + "learning_rate": 3.0463038180341185e-06, + "loss": 2.1257, + "step": 1251 + }, + { + "epoch": 1.372054794520548, + "grad_norm": 1.981020212173462, + "learning_rate": 3.0442729488220962e-06, + "loss": 2.155, + "step": 1252 + }, + { + "epoch": 1.373150684931507, + "grad_norm": 1.9707001447677612, + "learning_rate": 3.042242079610073e-06, + "loss": 2.1671, + "step": 1253 + }, + { + "epoch": 1.3742465753424657, + "grad_norm": 1.8645458221435547, + "learning_rate": 3.0402112103980508e-06, + "loss": 2.1655, + "step": 1254 + }, + { + "epoch": 1.3753424657534246, + "grad_norm": 1.863911747932434, + "learning_rate": 3.0381803411860276e-06, + "loss": 2.0917, + "step": 1255 + }, + { + "epoch": 1.3764383561643836, + "grad_norm": 2.1079251766204834, + "learning_rate": 3.0361494719740053e-06, + "loss": 2.119, + "step": 1256 + }, + { + "epoch": 1.3775342465753424, + "grad_norm": 1.8242024183273315, + "learning_rate": 3.034118602761982e-06, + "loss": 2.1398, + "step": 1257 + }, + { + "epoch": 1.3786301369863014, + "grad_norm": 1.8663142919540405, + "learning_rate": 3.03208773354996e-06, + "loss": 2.1405, + "step": 1258 + }, + { + "epoch": 1.3797260273972602, + "grad_norm": 1.9437193870544434, + "learning_rate": 3.0300568643379367e-06, + "loss": 2.1156, + "step": 1259 + }, + { + "epoch": 1.3808219178082193, + "grad_norm": 2.0837693214416504, + "learning_rate": 3.028025995125914e-06, + "loss": 2.1731, + "step": 1260 + }, + { + "epoch": 1.3808219178082193, + "eval_loss": 2.283296823501587, + "eval_runtime": 45.341, + "eval_samples_per_second": 22.055, + "eval_steps_per_second": 0.926, + "step": 1260 + }, + { + "epoch": 1.381917808219178, + "grad_norm": 1.9430493116378784, + "learning_rate": 3.0259951259138913e-06, + "loss": 2.1623, + "step": 1261 + }, + { + "epoch": 1.383013698630137, + "grad_norm": 1.825675368309021, + "learning_rate": 3.0239642567018686e-06, + "loss": 2.1022, + "step": 1262 + }, + { + "epoch": 1.384109589041096, + "grad_norm": 2.0050806999206543, + "learning_rate": 3.021933387489846e-06, + "loss": 2.1443, + "step": 1263 + }, + { + "epoch": 1.3852054794520547, + "grad_norm": 1.9548894166946411, + "learning_rate": 3.019902518277823e-06, + "loss": 2.0983, + "step": 1264 + }, + { + "epoch": 1.3863013698630138, + "grad_norm": 1.9097814559936523, + "learning_rate": 3.0178716490658004e-06, + "loss": 2.1488, + "step": 1265 + }, + { + "epoch": 1.3873972602739726, + "grad_norm": 1.8977514505386353, + "learning_rate": 3.0158407798537777e-06, + "loss": 2.1641, + "step": 1266 + }, + { + "epoch": 1.3884931506849316, + "grad_norm": 1.9222278594970703, + "learning_rate": 3.013809910641755e-06, + "loss": 2.1043, + "step": 1267 + }, + { + "epoch": 1.3895890410958904, + "grad_norm": 1.9333430528640747, + "learning_rate": 3.0117790414297322e-06, + "loss": 2.1152, + "step": 1268 + }, + { + "epoch": 1.3906849315068492, + "grad_norm": 1.8405439853668213, + "learning_rate": 3.0097481722177095e-06, + "loss": 2.1046, + "step": 1269 + }, + { + "epoch": 1.3917808219178083, + "grad_norm": 2.0583181381225586, + "learning_rate": 3.0077173030056868e-06, + "loss": 2.1378, + "step": 1270 + }, + { + "epoch": 1.3917808219178083, + "eval_loss": 2.286198616027832, + "eval_runtime": 46.6914, + "eval_samples_per_second": 21.417, + "eval_steps_per_second": 0.9, + "step": 1270 + }, + { + "epoch": 1.392876712328767, + "grad_norm": 1.8485472202301025, + "learning_rate": 3.005686433793664e-06, + "loss": 2.1184, + "step": 1271 + }, + { + "epoch": 1.393972602739726, + "grad_norm": 2.0430188179016113, + "learning_rate": 3.0036555645816413e-06, + "loss": 2.1518, + "step": 1272 + }, + { + "epoch": 1.395068493150685, + "grad_norm": 2.307474374771118, + "learning_rate": 3.001624695369618e-06, + "loss": 2.1288, + "step": 1273 + }, + { + "epoch": 1.396164383561644, + "grad_norm": 1.85003662109375, + "learning_rate": 2.999593826157596e-06, + "loss": 2.1573, + "step": 1274 + }, + { + "epoch": 1.3972602739726028, + "grad_norm": 2.047619581222534, + "learning_rate": 2.9975629569455727e-06, + "loss": 2.1285, + "step": 1275 + }, + { + "epoch": 1.3983561643835616, + "grad_norm": 2.013644218444824, + "learning_rate": 2.9955320877335504e-06, + "loss": 2.1603, + "step": 1276 + }, + { + "epoch": 1.3994520547945206, + "grad_norm": 1.8814456462860107, + "learning_rate": 2.9935012185215273e-06, + "loss": 2.1333, + "step": 1277 + }, + { + "epoch": 1.4005479452054794, + "grad_norm": 2.0338361263275146, + "learning_rate": 2.991470349309505e-06, + "loss": 2.1564, + "step": 1278 + }, + { + "epoch": 1.4016438356164382, + "grad_norm": 1.972992181777954, + "learning_rate": 2.989439480097482e-06, + "loss": 2.1305, + "step": 1279 + }, + { + "epoch": 1.4027397260273973, + "grad_norm": 1.845928430557251, + "learning_rate": 2.9874086108854595e-06, + "loss": 2.1384, + "step": 1280 + }, + { + "epoch": 1.4027397260273973, + "eval_loss": 2.28485107421875, + "eval_runtime": 46.9757, + "eval_samples_per_second": 21.288, + "eval_steps_per_second": 0.894, + "step": 1280 + }, + { + "epoch": 1.403835616438356, + "grad_norm": 1.7601349353790283, + "learning_rate": 2.9853777416734364e-06, + "loss": 2.1219, + "step": 1281 + }, + { + "epoch": 1.4049315068493151, + "grad_norm": 2.2385566234588623, + "learning_rate": 2.983346872461414e-06, + "loss": 2.1581, + "step": 1282 + }, + { + "epoch": 1.406027397260274, + "grad_norm": 2.019174098968506, + "learning_rate": 2.981316003249391e-06, + "loss": 2.1753, + "step": 1283 + }, + { + "epoch": 1.407123287671233, + "grad_norm": 1.9927541017532349, + "learning_rate": 2.9792851340373678e-06, + "loss": 2.147, + "step": 1284 + }, + { + "epoch": 1.4082191780821918, + "grad_norm": 2.5143957138061523, + "learning_rate": 2.9772542648253455e-06, + "loss": 2.138, + "step": 1285 + }, + { + "epoch": 1.4093150684931506, + "grad_norm": 1.9925073385238647, + "learning_rate": 2.9752233956133223e-06, + "loss": 2.1642, + "step": 1286 + }, + { + "epoch": 1.4104109589041096, + "grad_norm": 2.1739184856414795, + "learning_rate": 2.9731925264013e-06, + "loss": 2.1258, + "step": 1287 + }, + { + "epoch": 1.4115068493150684, + "grad_norm": 2.1719982624053955, + "learning_rate": 2.971161657189277e-06, + "loss": 2.1323, + "step": 1288 + }, + { + "epoch": 1.4126027397260275, + "grad_norm": 1.9897956848144531, + "learning_rate": 2.9691307879772546e-06, + "loss": 2.1286, + "step": 1289 + }, + { + "epoch": 1.4136986301369863, + "grad_norm": 1.997477412223816, + "learning_rate": 2.9670999187652314e-06, + "loss": 2.125, + "step": 1290 + }, + { + "epoch": 1.4136986301369863, + "eval_loss": 2.2839529514312744, + "eval_runtime": 44.5005, + "eval_samples_per_second": 22.472, + "eval_steps_per_second": 0.944, + "step": 1290 + }, + { + "epoch": 1.4147945205479453, + "grad_norm": 1.9529412984848022, + "learning_rate": 2.965069049553209e-06, + "loss": 2.1074, + "step": 1291 + }, + { + "epoch": 1.415890410958904, + "grad_norm": 1.900692343711853, + "learning_rate": 2.963038180341186e-06, + "loss": 2.1796, + "step": 1292 + }, + { + "epoch": 1.416986301369863, + "grad_norm": 2.0269863605499268, + "learning_rate": 2.9610073111291637e-06, + "loss": 2.0998, + "step": 1293 + }, + { + "epoch": 1.418082191780822, + "grad_norm": 2.0290050506591797, + "learning_rate": 2.9589764419171405e-06, + "loss": 2.1638, + "step": 1294 + }, + { + "epoch": 1.4191780821917808, + "grad_norm": 1.8717535734176636, + "learning_rate": 2.9569455727051182e-06, + "loss": 2.1756, + "step": 1295 + }, + { + "epoch": 1.4202739726027398, + "grad_norm": 1.9470241069793701, + "learning_rate": 2.954914703493095e-06, + "loss": 2.116, + "step": 1296 + }, + { + "epoch": 1.4213698630136986, + "grad_norm": 2.1749250888824463, + "learning_rate": 2.9528838342810724e-06, + "loss": 2.1381, + "step": 1297 + }, + { + "epoch": 1.4224657534246576, + "grad_norm": 1.905742883682251, + "learning_rate": 2.9508529650690496e-06, + "loss": 2.118, + "step": 1298 + }, + { + "epoch": 1.4235616438356165, + "grad_norm": 2.1214921474456787, + "learning_rate": 2.948822095857027e-06, + "loss": 2.1223, + "step": 1299 + }, + { + "epoch": 1.4246575342465753, + "grad_norm": 1.894509196281433, + "learning_rate": 2.946791226645004e-06, + "loss": 2.1457, + "step": 1300 + }, + { + "epoch": 1.4246575342465753, + "eval_loss": 2.2824268341064453, + "eval_runtime": 47.8031, + "eval_samples_per_second": 20.919, + "eval_steps_per_second": 0.879, + "step": 1300 + }, + { + "epoch": 1.4257534246575343, + "grad_norm": 1.9588708877563477, + "learning_rate": 2.9447603574329815e-06, + "loss": 2.1113, + "step": 1301 + }, + { + "epoch": 1.426849315068493, + "grad_norm": 1.9709562063217163, + "learning_rate": 2.9427294882209587e-06, + "loss": 2.1086, + "step": 1302 + }, + { + "epoch": 1.427945205479452, + "grad_norm": 1.82259202003479, + "learning_rate": 2.940698619008936e-06, + "loss": 2.118, + "step": 1303 + }, + { + "epoch": 1.429041095890411, + "grad_norm": 1.9776451587677002, + "learning_rate": 2.9386677497969133e-06, + "loss": 2.1386, + "step": 1304 + }, + { + "epoch": 1.43013698630137, + "grad_norm": 2.147463321685791, + "learning_rate": 2.9366368805848906e-06, + "loss": 2.1309, + "step": 1305 + }, + { + "epoch": 1.4312328767123288, + "grad_norm": 1.9845961332321167, + "learning_rate": 2.934606011372868e-06, + "loss": 2.1586, + "step": 1306 + }, + { + "epoch": 1.4323287671232876, + "grad_norm": 1.9153180122375488, + "learning_rate": 2.932575142160845e-06, + "loss": 2.1588, + "step": 1307 + }, + { + "epoch": 1.4334246575342466, + "grad_norm": 1.9372929334640503, + "learning_rate": 2.9305442729488224e-06, + "loss": 2.1351, + "step": 1308 + }, + { + "epoch": 1.4345205479452054, + "grad_norm": 1.852669358253479, + "learning_rate": 2.9285134037367997e-06, + "loss": 2.1394, + "step": 1309 + }, + { + "epoch": 1.4356164383561643, + "grad_norm": 1.934251308441162, + "learning_rate": 2.9264825345247765e-06, + "loss": 2.1364, + "step": 1310 + }, + { + "epoch": 1.4356164383561643, + "eval_loss": 2.280508279800415, + "eval_runtime": 43.1734, + "eval_samples_per_second": 23.162, + "eval_steps_per_second": 0.973, + "step": 1310 + }, + { + "epoch": 1.4367123287671233, + "grad_norm": 1.8886998891830444, + "learning_rate": 2.9244516653127542e-06, + "loss": 2.1452, + "step": 1311 + }, + { + "epoch": 1.4378082191780823, + "grad_norm": 2.275813102722168, + "learning_rate": 2.922420796100731e-06, + "loss": 2.1465, + "step": 1312 + }, + { + "epoch": 1.4389041095890411, + "grad_norm": 1.9017852544784546, + "learning_rate": 2.9203899268887088e-06, + "loss": 2.0785, + "step": 1313 + }, + { + "epoch": 1.44, + "grad_norm": 1.8482435941696167, + "learning_rate": 2.9183590576766856e-06, + "loss": 2.1892, + "step": 1314 + }, + { + "epoch": 1.441095890410959, + "grad_norm": 2.2122418880462646, + "learning_rate": 2.9163281884646633e-06, + "loss": 2.1058, + "step": 1315 + }, + { + "epoch": 1.4421917808219178, + "grad_norm": 1.9721672534942627, + "learning_rate": 2.91429731925264e-06, + "loss": 2.1391, + "step": 1316 + }, + { + "epoch": 1.4432876712328766, + "grad_norm": 1.9381605386734009, + "learning_rate": 2.912266450040618e-06, + "loss": 2.1651, + "step": 1317 + }, + { + "epoch": 1.4443835616438356, + "grad_norm": 2.274630069732666, + "learning_rate": 2.9102355808285947e-06, + "loss": 2.1434, + "step": 1318 + }, + { + "epoch": 1.4454794520547944, + "grad_norm": 1.91957688331604, + "learning_rate": 2.9082047116165724e-06, + "loss": 2.1218, + "step": 1319 + }, + { + "epoch": 1.4465753424657535, + "grad_norm": 2.042171001434326, + "learning_rate": 2.9061738424045493e-06, + "loss": 2.1233, + "step": 1320 + }, + { + "epoch": 1.4465753424657535, + "eval_loss": 2.280571699142456, + "eval_runtime": 48.425, + "eval_samples_per_second": 20.65, + "eval_steps_per_second": 0.867, + "step": 1320 + }, + { + "epoch": 1.4476712328767123, + "grad_norm": 2.0493276119232178, + "learning_rate": 2.904142973192527e-06, + "loss": 2.1241, + "step": 1321 + }, + { + "epoch": 1.4487671232876713, + "grad_norm": 1.8865023851394653, + "learning_rate": 2.902112103980504e-06, + "loss": 2.1826, + "step": 1322 + }, + { + "epoch": 1.4498630136986301, + "grad_norm": 1.9723960161209106, + "learning_rate": 2.900081234768481e-06, + "loss": 2.1508, + "step": 1323 + }, + { + "epoch": 1.450958904109589, + "grad_norm": 1.8321057558059692, + "learning_rate": 2.8980503655564584e-06, + "loss": 2.1319, + "step": 1324 + }, + { + "epoch": 1.452054794520548, + "grad_norm": 1.97796630859375, + "learning_rate": 2.8960194963444357e-06, + "loss": 2.1289, + "step": 1325 + }, + { + "epoch": 1.4531506849315068, + "grad_norm": 1.7793136835098267, + "learning_rate": 2.893988627132413e-06, + "loss": 2.1059, + "step": 1326 + }, + { + "epoch": 1.4542465753424658, + "grad_norm": 1.878381609916687, + "learning_rate": 2.8919577579203902e-06, + "loss": 2.125, + "step": 1327 + }, + { + "epoch": 1.4553424657534246, + "grad_norm": 1.8854790925979614, + "learning_rate": 2.8899268887083675e-06, + "loss": 2.0908, + "step": 1328 + }, + { + "epoch": 1.4564383561643837, + "grad_norm": 1.940931797027588, + "learning_rate": 2.8878960194963448e-06, + "loss": 2.1338, + "step": 1329 + }, + { + "epoch": 1.4575342465753425, + "grad_norm": 1.8014280796051025, + "learning_rate": 2.885865150284322e-06, + "loss": 2.1384, + "step": 1330 + }, + { + "epoch": 1.4575342465753425, + "eval_loss": 2.2815349102020264, + "eval_runtime": 48.7646, + "eval_samples_per_second": 20.507, + "eval_steps_per_second": 0.861, + "step": 1330 + }, + { + "epoch": 1.4586301369863013, + "grad_norm": 1.9306361675262451, + "learning_rate": 2.8838342810722993e-06, + "loss": 2.1562, + "step": 1331 + }, + { + "epoch": 1.4597260273972603, + "grad_norm": 1.902762770652771, + "learning_rate": 2.8818034118602766e-06, + "loss": 2.1183, + "step": 1332 + }, + { + "epoch": 1.4608219178082191, + "grad_norm": 1.980449914932251, + "learning_rate": 2.879772542648254e-06, + "loss": 2.1066, + "step": 1333 + }, + { + "epoch": 1.4619178082191782, + "grad_norm": 2.0774567127227783, + "learning_rate": 2.8777416734362307e-06, + "loss": 2.157, + "step": 1334 + }, + { + "epoch": 1.463013698630137, + "grad_norm": 1.8203723430633545, + "learning_rate": 2.8757108042242084e-06, + "loss": 2.0792, + "step": 1335 + }, + { + "epoch": 1.464109589041096, + "grad_norm": 1.8849077224731445, + "learning_rate": 2.8736799350121853e-06, + "loss": 2.1917, + "step": 1336 + }, + { + "epoch": 1.4652054794520548, + "grad_norm": 1.8782836198806763, + "learning_rate": 2.871649065800163e-06, + "loss": 2.1744, + "step": 1337 + }, + { + "epoch": 1.4663013698630136, + "grad_norm": 1.8628056049346924, + "learning_rate": 2.86961819658814e-06, + "loss": 2.1788, + "step": 1338 + }, + { + "epoch": 1.4673972602739727, + "grad_norm": 1.865370750427246, + "learning_rate": 2.8675873273761175e-06, + "loss": 2.1159, + "step": 1339 + }, + { + "epoch": 1.4684931506849315, + "grad_norm": 1.8673044443130493, + "learning_rate": 2.8655564581640944e-06, + "loss": 2.1447, + "step": 1340 + }, + { + "epoch": 1.4684931506849315, + "eval_loss": 2.282686471939087, + "eval_runtime": 49.85, + "eval_samples_per_second": 20.06, + "eval_steps_per_second": 0.843, + "step": 1340 + }, + { + "epoch": 1.4695890410958903, + "grad_norm": 1.8928618431091309, + "learning_rate": 2.863525588952072e-06, + "loss": 2.1362, + "step": 1341 + }, + { + "epoch": 1.4706849315068493, + "grad_norm": 1.8964238166809082, + "learning_rate": 2.861494719740049e-06, + "loss": 2.1628, + "step": 1342 + }, + { + "epoch": 1.4717808219178083, + "grad_norm": 1.9889185428619385, + "learning_rate": 2.8594638505280266e-06, + "loss": 2.0922, + "step": 1343 + }, + { + "epoch": 1.4728767123287672, + "grad_norm": 1.9503910541534424, + "learning_rate": 2.8574329813160035e-06, + "loss": 2.0979, + "step": 1344 + }, + { + "epoch": 1.473972602739726, + "grad_norm": 1.8889869451522827, + "learning_rate": 2.855402112103981e-06, + "loss": 2.1329, + "step": 1345 + }, + { + "epoch": 1.475068493150685, + "grad_norm": 1.7875893115997314, + "learning_rate": 2.853371242891958e-06, + "loss": 2.123, + "step": 1346 + }, + { + "epoch": 1.4761643835616438, + "grad_norm": 1.9358372688293457, + "learning_rate": 2.851340373679935e-06, + "loss": 2.1284, + "step": 1347 + }, + { + "epoch": 1.4772602739726026, + "grad_norm": 2.125243663787842, + "learning_rate": 2.8493095044679126e-06, + "loss": 2.1386, + "step": 1348 + }, + { + "epoch": 1.4783561643835617, + "grad_norm": 1.927788257598877, + "learning_rate": 2.8472786352558894e-06, + "loss": 2.1835, + "step": 1349 + }, + { + "epoch": 1.4794520547945205, + "grad_norm": 1.9361577033996582, + "learning_rate": 2.845247766043867e-06, + "loss": 2.1411, + "step": 1350 + }, + { + "epoch": 1.4794520547945205, + "eval_loss": 2.281172275543213, + "eval_runtime": 47.4205, + "eval_samples_per_second": 21.088, + "eval_steps_per_second": 0.886, + "step": 1350 + }, + { + "epoch": 1.4805479452054795, + "grad_norm": 1.847364068031311, + "learning_rate": 2.843216896831844e-06, + "loss": 2.111, + "step": 1351 + }, + { + "epoch": 1.4816438356164383, + "grad_norm": 1.9927674531936646, + "learning_rate": 2.8411860276198217e-06, + "loss": 2.107, + "step": 1352 + }, + { + "epoch": 1.4827397260273973, + "grad_norm": 1.8565406799316406, + "learning_rate": 2.8391551584077985e-06, + "loss": 2.12, + "step": 1353 + }, + { + "epoch": 1.4838356164383562, + "grad_norm": 1.9477596282958984, + "learning_rate": 2.8371242891957762e-06, + "loss": 2.1661, + "step": 1354 + }, + { + "epoch": 1.484931506849315, + "grad_norm": 1.9354814291000366, + "learning_rate": 2.835093419983753e-06, + "loss": 2.168, + "step": 1355 + }, + { + "epoch": 1.486027397260274, + "grad_norm": 2.0603487491607666, + "learning_rate": 2.833062550771731e-06, + "loss": 2.1338, + "step": 1356 + }, + { + "epoch": 1.4871232876712328, + "grad_norm": 1.987990379333496, + "learning_rate": 2.8310316815597076e-06, + "loss": 2.1282, + "step": 1357 + }, + { + "epoch": 1.4882191780821918, + "grad_norm": 1.895529866218567, + "learning_rate": 2.8290008123476853e-06, + "loss": 2.1032, + "step": 1358 + }, + { + "epoch": 1.4893150684931507, + "grad_norm": 1.917994737625122, + "learning_rate": 2.826969943135662e-06, + "loss": 2.149, + "step": 1359 + }, + { + "epoch": 1.4904109589041097, + "grad_norm": 1.9238080978393555, + "learning_rate": 2.8249390739236395e-06, + "loss": 2.1275, + "step": 1360 + }, + { + "epoch": 1.4904109589041097, + "eval_loss": 2.280546188354492, + "eval_runtime": 46.9985, + "eval_samples_per_second": 21.277, + "eval_steps_per_second": 0.894, + "step": 1360 + }, + { + "epoch": 1.4915068493150685, + "grad_norm": 1.947768211364746, + "learning_rate": 2.8229082047116167e-06, + "loss": 2.1282, + "step": 1361 + }, + { + "epoch": 1.4926027397260273, + "grad_norm": 1.7888189554214478, + "learning_rate": 2.820877335499594e-06, + "loss": 2.1134, + "step": 1362 + }, + { + "epoch": 1.4936986301369863, + "grad_norm": 1.8480347394943237, + "learning_rate": 2.8188464662875713e-06, + "loss": 2.1194, + "step": 1363 + }, + { + "epoch": 1.4947945205479451, + "grad_norm": 1.8146322965621948, + "learning_rate": 2.8168155970755486e-06, + "loss": 2.0781, + "step": 1364 + }, + { + "epoch": 1.4958904109589042, + "grad_norm": 1.7994698286056519, + "learning_rate": 2.814784727863526e-06, + "loss": 2.1027, + "step": 1365 + }, + { + "epoch": 1.496986301369863, + "grad_norm": 1.9375026226043701, + "learning_rate": 2.812753858651503e-06, + "loss": 2.1273, + "step": 1366 + }, + { + "epoch": 1.498082191780822, + "grad_norm": 1.9606287479400635, + "learning_rate": 2.8107229894394804e-06, + "loss": 2.1438, + "step": 1367 + }, + { + "epoch": 1.4991780821917808, + "grad_norm": 2.1815521717071533, + "learning_rate": 2.8086921202274577e-06, + "loss": 2.1176, + "step": 1368 + }, + { + "epoch": 1.5002739726027396, + "grad_norm": 1.8796950578689575, + "learning_rate": 2.806661251015435e-06, + "loss": 2.096, + "step": 1369 + }, + { + "epoch": 1.5013698630136987, + "grad_norm": 2.016091823577881, + "learning_rate": 2.8046303818034122e-06, + "loss": 2.1451, + "step": 1370 + }, + { + "epoch": 1.5013698630136987, + "eval_loss": 2.280881643295288, + "eval_runtime": 53.6471, + "eval_samples_per_second": 18.64, + "eval_steps_per_second": 0.783, + "step": 1370 + }, + { + "epoch": 1.5024657534246575, + "grad_norm": 1.9303081035614014, + "learning_rate": 2.802599512591389e-06, + "loss": 2.1736, + "step": 1371 + }, + { + "epoch": 1.5035616438356163, + "grad_norm": 1.9831101894378662, + "learning_rate": 2.8005686433793668e-06, + "loss": 2.2147, + "step": 1372 + }, + { + "epoch": 1.5046575342465753, + "grad_norm": 1.7864863872528076, + "learning_rate": 2.7985377741673436e-06, + "loss": 2.1231, + "step": 1373 + }, + { + "epoch": 1.5057534246575344, + "grad_norm": 1.8945375680923462, + "learning_rate": 2.7965069049553213e-06, + "loss": 2.0866, + "step": 1374 + }, + { + "epoch": 1.5068493150684932, + "grad_norm": 1.9881761074066162, + "learning_rate": 2.794476035743298e-06, + "loss": 2.1398, + "step": 1375 + }, + { + "epoch": 1.507945205479452, + "grad_norm": 1.9888391494750977, + "learning_rate": 2.792445166531276e-06, + "loss": 2.172, + "step": 1376 + }, + { + "epoch": 1.509041095890411, + "grad_norm": 1.9845142364501953, + "learning_rate": 2.7904142973192527e-06, + "loss": 2.1378, + "step": 1377 + }, + { + "epoch": 1.5101369863013698, + "grad_norm": 2.0640218257904053, + "learning_rate": 2.7883834281072304e-06, + "loss": 2.16, + "step": 1378 + }, + { + "epoch": 1.5112328767123286, + "grad_norm": 1.9453262090682983, + "learning_rate": 2.7863525588952073e-06, + "loss": 2.1112, + "step": 1379 + }, + { + "epoch": 1.5123287671232877, + "grad_norm": 2.0385119915008545, + "learning_rate": 2.784321689683185e-06, + "loss": 2.1068, + "step": 1380 + }, + { + "epoch": 1.5123287671232877, + "eval_loss": 2.2804832458496094, + "eval_runtime": 50.1317, + "eval_samples_per_second": 19.947, + "eval_steps_per_second": 0.838, + "step": 1380 + }, + { + "epoch": 1.5134246575342467, + "grad_norm": 1.9497289657592773, + "learning_rate": 2.782290820471162e-06, + "loss": 2.1017, + "step": 1381 + }, + { + "epoch": 1.5145205479452055, + "grad_norm": 1.9402867555618286, + "learning_rate": 2.7802599512591395e-06, + "loss": 2.1141, + "step": 1382 + }, + { + "epoch": 1.5156164383561643, + "grad_norm": 2.008141279220581, + "learning_rate": 2.7782290820471164e-06, + "loss": 2.1601, + "step": 1383 + }, + { + "epoch": 1.5167123287671234, + "grad_norm": 1.9685825109481812, + "learning_rate": 2.7761982128350932e-06, + "loss": 2.1494, + "step": 1384 + }, + { + "epoch": 1.5178082191780822, + "grad_norm": 1.9075162410736084, + "learning_rate": 2.774167343623071e-06, + "loss": 2.147, + "step": 1385 + }, + { + "epoch": 1.518904109589041, + "grad_norm": 1.8381123542785645, + "learning_rate": 2.772136474411048e-06, + "loss": 2.1772, + "step": 1386 + }, + { + "epoch": 1.52, + "grad_norm": 1.806057095527649, + "learning_rate": 2.7701056051990255e-06, + "loss": 2.1077, + "step": 1387 + }, + { + "epoch": 1.521095890410959, + "grad_norm": 1.9030957221984863, + "learning_rate": 2.7680747359870023e-06, + "loss": 2.0735, + "step": 1388 + }, + { + "epoch": 1.5221917808219179, + "grad_norm": 1.9967273473739624, + "learning_rate": 2.76604386677498e-06, + "loss": 2.155, + "step": 1389 + }, + { + "epoch": 1.5232876712328767, + "grad_norm": 1.864998698234558, + "learning_rate": 2.764012997562957e-06, + "loss": 2.1319, + "step": 1390 + }, + { + "epoch": 1.5232876712328767, + "eval_loss": 2.279405355453491, + "eval_runtime": 49.6092, + "eval_samples_per_second": 20.158, + "eval_steps_per_second": 0.847, + "step": 1390 + }, + { + "epoch": 1.5243835616438357, + "grad_norm": 1.831916093826294, + "learning_rate": 2.7619821283509346e-06, + "loss": 2.1518, + "step": 1391 + }, + { + "epoch": 1.5254794520547945, + "grad_norm": 2.1456778049468994, + "learning_rate": 2.7599512591389114e-06, + "loss": 2.0955, + "step": 1392 + }, + { + "epoch": 1.5265753424657533, + "grad_norm": 1.9264456033706665, + "learning_rate": 2.757920389926889e-06, + "loss": 2.1594, + "step": 1393 + }, + { + "epoch": 1.5276712328767124, + "grad_norm": 1.9881576299667358, + "learning_rate": 2.755889520714866e-06, + "loss": 2.115, + "step": 1394 + }, + { + "epoch": 1.5287671232876714, + "grad_norm": 1.9104706048965454, + "learning_rate": 2.7538586515028437e-06, + "loss": 2.1074, + "step": 1395 + }, + { + "epoch": 1.52986301369863, + "grad_norm": 1.8815401792526245, + "learning_rate": 2.7518277822908205e-06, + "loss": 2.0779, + "step": 1396 + }, + { + "epoch": 1.530958904109589, + "grad_norm": 1.999870777130127, + "learning_rate": 2.749796913078798e-06, + "loss": 2.1409, + "step": 1397 + }, + { + "epoch": 1.532054794520548, + "grad_norm": 2.037266969680786, + "learning_rate": 2.747766043866775e-06, + "loss": 2.1565, + "step": 1398 + }, + { + "epoch": 1.5331506849315069, + "grad_norm": 1.865891695022583, + "learning_rate": 2.7457351746547524e-06, + "loss": 2.1269, + "step": 1399 + }, + { + "epoch": 1.5342465753424657, + "grad_norm": 1.959797739982605, + "learning_rate": 2.7437043054427297e-06, + "loss": 2.1674, + "step": 1400 + }, + { + "epoch": 1.5342465753424657, + "eval_loss": 2.279064893722534, + "eval_runtime": 50.0471, + "eval_samples_per_second": 19.981, + "eval_steps_per_second": 0.839, + "step": 1400 + } + ], + "logging_steps": 1, + "max_steps": 2736, + "num_input_tokens_seen": 0, + "num_train_epochs": 3, + "save_steps": 100, + "stateful_callbacks": { + "TrainerControl": { + "args": { + "should_epoch_stop": false, + "should_evaluate": false, + "should_log": false, + "should_save": true, + "should_training_stop": false + }, + "attributes": {} + } + }, + "total_flos": 4.364233173106688e+17, + "train_batch_size": 3, + "trial_name": null, + "trial_params": null +} diff --git a/babylm_shuffle_nondeterministic_10M_seed0/runs/checkpoint-1400/zero_to_fp32.py b/babylm_shuffle_nondeterministic_10M_seed0/runs/checkpoint-1400/zero_to_fp32.py new file mode 100644 index 0000000000000000000000000000000000000000..24cc342e78d1a006c782b3a4cd68d9ce786d8fd8 --- /dev/null +++ b/babylm_shuffle_nondeterministic_10M_seed0/runs/checkpoint-1400/zero_to_fp32.py @@ -0,0 +1,604 @@ +#!/usr/bin/env python + +# Copyright (c) Microsoft Corporation. +# SPDX-License-Identifier: Apache-2.0 + +# DeepSpeed Team + +# This script extracts fp32 consolidated weights from a zero 1, 2 and 3 DeepSpeed checkpoints. It gets +# copied into the top level checkpoint dir, so the user can easily do the conversion at any point in +# the future. Once extracted, the weights don't require DeepSpeed and can be used in any +# application. +# +# example: python zero_to_fp32.py . pytorch_model.bin + +import argparse +import torch +import glob +import math +import os +import re +from collections import OrderedDict +from dataclasses import dataclass + +# while this script doesn't use deepspeed to recover data, since the checkpoints are pickled with +# DeepSpeed data structures it has to be available in the current python environment. +from deepspeed.utils import logger +from deepspeed.checkpoint.constants import (DS_VERSION, OPTIMIZER_STATE_DICT, SINGLE_PARTITION_OF_FP32_GROUPS, + FP32_FLAT_GROUPS, ZERO_STAGE, PARTITION_COUNT, PARAM_SHAPES, BUFFER_NAMES, + FROZEN_PARAM_SHAPES, FROZEN_PARAM_FRAGMENTS) + + +@dataclass +class zero_model_state: + buffers: dict() + param_shapes: dict() + shared_params: list + ds_version: int + frozen_param_shapes: dict() + frozen_param_fragments: dict() + + +debug = 0 + +# load to cpu +device = torch.device('cpu') + + +def atoi(text): + return int(text) if text.isdigit() else text + + +def natural_keys(text): + ''' + alist.sort(key=natural_keys) sorts in human order + http://nedbatchelder.com/blog/200712/human_sorting.html + (See Toothy's implementation in the comments) + ''' + return [atoi(c) for c in re.split(r'(\d+)', text)] + + +def get_model_state_file(checkpoint_dir, zero_stage): + if not os.path.isdir(checkpoint_dir): + raise FileNotFoundError(f"Directory '{checkpoint_dir}' doesn't exist") + + # there should be only one file + if zero_stage <= 2: + file = os.path.join(checkpoint_dir, "mp_rank_00_model_states.pt") + elif zero_stage == 3: + file = os.path.join(checkpoint_dir, "zero_pp_rank_0_mp_rank_00_model_states.pt") + + if not os.path.exists(file): + raise FileNotFoundError(f"can't find model states file at '{file}'") + + return file + + +def get_checkpoint_files(checkpoint_dir, glob_pattern): + # XXX: need to test that this simple glob rule works for multi-node setup too + ckpt_files = sorted(glob.glob(os.path.join(checkpoint_dir, glob_pattern)), key=natural_keys) + + if len(ckpt_files) == 0: + raise FileNotFoundError(f"can't find {glob_pattern} files in directory '{checkpoint_dir}'") + + return ckpt_files + + +def get_optim_files(checkpoint_dir): + return get_checkpoint_files(checkpoint_dir, "*_optim_states.pt") + + +def get_model_state_files(checkpoint_dir): + return get_checkpoint_files(checkpoint_dir, "*_model_states.pt") + + +def parse_model_states(files): + zero_model_states = [] + for file in files: + state_dict = torch.load(file, map_location=device) + + if BUFFER_NAMES not in state_dict: + raise ValueError(f"{file} is not a model state checkpoint") + buffer_names = state_dict[BUFFER_NAMES] + if debug: + print("Found buffers:", buffer_names) + + # recover just the buffers while restoring them to fp32 if they were saved in fp16 + buffers = {k: v.float() for k, v in state_dict["module"].items() if k in buffer_names} + param_shapes = state_dict[PARAM_SHAPES] + + # collect parameters that are included in param_shapes + param_names = [] + for s in param_shapes: + for name in s.keys(): + param_names.append(name) + + # update with frozen parameters + frozen_param_shapes = state_dict.get(FROZEN_PARAM_SHAPES, None) + if frozen_param_shapes is not None: + if debug: + print(f"Found frozen_param_shapes: {frozen_param_shapes}") + param_names += list(frozen_param_shapes.keys()) + + # handle shared params + shared_params = [[k, v] for k, v in state_dict["shared_params"].items()] + + ds_version = state_dict.get(DS_VERSION, None) + + frozen_param_fragments = state_dict.get(FROZEN_PARAM_FRAGMENTS, None) + + z_model_state = zero_model_state(buffers=buffers, + param_shapes=param_shapes, + shared_params=shared_params, + ds_version=ds_version, + frozen_param_shapes=frozen_param_shapes, + frozen_param_fragments=frozen_param_fragments) + zero_model_states.append(z_model_state) + + return zero_model_states + + +def parse_optim_states(files, ds_checkpoint_dir): + + total_files = len(files) + state_dicts = [] + for f in files: + state_dict = torch.load(f, map_location=device) + # immediately discard the potentially huge 2 optimizer states as we only care for fp32 master weights + # and also handle the case where it was already removed by another helper script + state_dict["optimizer_state_dict"].pop("optimizer_state_dict", None) + state_dicts.append(state_dict) + + if not ZERO_STAGE in state_dicts[0][OPTIMIZER_STATE_DICT]: + raise ValueError(f"{files[0]} is not a zero checkpoint") + zero_stage = state_dicts[0][OPTIMIZER_STATE_DICT][ZERO_STAGE] + world_size = state_dicts[0][OPTIMIZER_STATE_DICT][PARTITION_COUNT] + + # For ZeRO-2 each param group can have different partition_count as data parallelism for expert + # parameters can be different from data parallelism for non-expert parameters. So we can just + # use the max of the partition_count to get the dp world_size. + + if type(world_size) is list: + world_size = max(world_size) + + if world_size != total_files: + raise ValueError( + f"Expected {world_size} of '*_optim_states.pt' under '{ds_checkpoint_dir}' but found {total_files} files. " + "Possibly due to an overwrite of an old checkpoint, or a checkpoint didn't get saved by one or more processes." + ) + + # the groups are named differently in each stage + if zero_stage <= 2: + fp32_groups_key = SINGLE_PARTITION_OF_FP32_GROUPS + elif zero_stage == 3: + fp32_groups_key = FP32_FLAT_GROUPS + else: + raise ValueError(f"unknown zero stage {zero_stage}") + + if zero_stage <= 2: + fp32_flat_groups = [state_dicts[i][OPTIMIZER_STATE_DICT][fp32_groups_key] for i in range(len(state_dicts))] + elif zero_stage == 3: + # if there is more than one param group, there will be multiple flattened tensors - one + # flattened tensor per group - for simplicity merge them into a single tensor + # + # XXX: could make the script more memory efficient for when there are multiple groups - it + # will require matching the sub-lists of param_shapes for each param group flattened tensor + + fp32_flat_groups = [ + torch.cat(state_dicts[i][OPTIMIZER_STATE_DICT][fp32_groups_key], 0) for i in range(len(state_dicts)) + ] + + return zero_stage, world_size, fp32_flat_groups + + +def _get_fp32_state_dict_from_zero_checkpoint(ds_checkpoint_dir, exclude_frozen_parameters): + """ + Returns fp32 state_dict reconstructed from ds checkpoint + + Args: + - ``ds_checkpoint_dir``: path to the deepspeed checkpoint folder (where the optimizer files are) + + """ + print(f"Processing zero checkpoint '{ds_checkpoint_dir}'") + + optim_files = get_optim_files(ds_checkpoint_dir) + zero_stage, world_size, fp32_flat_groups = parse_optim_states(optim_files, ds_checkpoint_dir) + print(f"Detected checkpoint of type zero stage {zero_stage}, world_size: {world_size}") + + model_files = get_model_state_files(ds_checkpoint_dir) + + zero_model_states = parse_model_states(model_files) + print(f'Parsing checkpoint created by deepspeed=={zero_model_states[0].ds_version}') + + if zero_stage <= 2: + return _get_fp32_state_dict_from_zero2_checkpoint(world_size, fp32_flat_groups, zero_model_states, + exclude_frozen_parameters) + elif zero_stage == 3: + return _get_fp32_state_dict_from_zero3_checkpoint(world_size, fp32_flat_groups, zero_model_states, + exclude_frozen_parameters) + + +def _zero2_merge_frozen_params(state_dict, zero_model_states): + if zero_model_states[0].frozen_param_shapes is None or len(zero_model_states[0].frozen_param_shapes) == 0: + return + + frozen_param_shapes = zero_model_states[0].frozen_param_shapes + frozen_param_fragments = zero_model_states[0].frozen_param_fragments + + if debug: + num_elem = sum(s.numel() for s in frozen_param_shapes.values()) + print(f'rank 0: {FROZEN_PARAM_SHAPES}.numel = {num_elem}') + + wanted_params = len(frozen_param_shapes) + wanted_numel = sum(s.numel() for s in frozen_param_shapes.values()) + avail_numel = sum([p.numel() for p in frozen_param_fragments.values()]) + print(f'Frozen params: Have {avail_numel} numels to process.') + print(f'Frozen params: Need {wanted_numel} numels in {wanted_params} params') + + total_params = 0 + total_numel = 0 + for name, shape in frozen_param_shapes.items(): + total_params += 1 + unpartitioned_numel = shape.numel() + total_numel += unpartitioned_numel + + state_dict[name] = frozen_param_fragments[name] + + if debug: + print(f"{name} full shape: {shape} unpartitioned numel {unpartitioned_numel} ") + + print(f"Reconstructed Frozen fp32 state dict with {total_params} params {total_numel} elements") + + +def _has_callable(obj, fn): + attr = getattr(obj, fn, None) + return callable(attr) + + +def _zero2_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states): + param_shapes = zero_model_states[0].param_shapes + + # Reconstruction protocol: + # + # XXX: document this + + if debug: + for i in range(world_size): + for j in range(len(fp32_flat_groups[0])): + print(f"{FP32_FLAT_GROUPS}[{i}][{j}].shape={fp32_flat_groups[i][j].shape}") + + # XXX: memory usage doubles here (zero2) + num_param_groups = len(fp32_flat_groups[0]) + merged_single_partition_of_fp32_groups = [] + for i in range(num_param_groups): + merged_partitions = [sd[i] for sd in fp32_flat_groups] + full_single_fp32_vector = torch.cat(merged_partitions, 0) + merged_single_partition_of_fp32_groups.append(full_single_fp32_vector) + avail_numel = sum( + [full_single_fp32_vector.numel() for full_single_fp32_vector in merged_single_partition_of_fp32_groups]) + + if debug: + wanted_params = sum([len(shapes) for shapes in param_shapes]) + wanted_numel = sum([sum(shape.numel() for shape in shapes.values()) for shapes in param_shapes]) + # not asserting if there is a mismatch due to possible padding + print(f"Have {avail_numel} numels to process.") + print(f"Need {wanted_numel} numels in {wanted_params} params.") + + # params + # XXX: for huge models that can't fit into the host's RAM we will have to recode this to support + # out-of-core computing solution + total_numel = 0 + total_params = 0 + for shapes, full_single_fp32_vector in zip(param_shapes, merged_single_partition_of_fp32_groups): + offset = 0 + avail_numel = full_single_fp32_vector.numel() + for name, shape in shapes.items(): + + unpartitioned_numel = shape.numel() if _has_callable(shape, 'numel') else math.prod(shape) + total_numel += unpartitioned_numel + total_params += 1 + + if debug: + print(f"{name} full shape: {shape} unpartitioned numel {unpartitioned_numel} ") + state_dict[name] = full_single_fp32_vector.narrow(0, offset, unpartitioned_numel).view(shape) + offset += unpartitioned_numel + + # Z2 started to align to 2*world_size to improve nccl performance. Therefore both offset and + # avail_numel can differ by anywhere between 0..2*world_size. Due to two unrelated complex + # paddings performed in the code it's almost impossible to predict the exact numbers w/o the + # live optimizer object, so we are checking that the numbers are within the right range + align_to = 2 * world_size + + def zero2_align(x): + return align_to * math.ceil(x / align_to) + + if debug: + print(f"original offset={offset}, avail_numel={avail_numel}") + + offset = zero2_align(offset) + avail_numel = zero2_align(avail_numel) + + if debug: + print(f"aligned offset={offset}, avail_numel={avail_numel}") + + # Sanity check + if offset != avail_numel: + raise ValueError(f"consumed {offset} numels out of {avail_numel} - something is wrong") + + print(f"Reconstructed fp32 state dict with {total_params} params {total_numel} elements") + + +def _get_fp32_state_dict_from_zero2_checkpoint(world_size, fp32_flat_groups, zero_model_states, + exclude_frozen_parameters): + state_dict = OrderedDict() + + # buffers + buffers = zero_model_states[0].buffers + state_dict.update(buffers) + if debug: + print(f"added {len(buffers)} buffers") + + if not exclude_frozen_parameters: + _zero2_merge_frozen_params(state_dict, zero_model_states) + + _zero2_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states) + + # recover shared parameters + for pair in zero_model_states[0].shared_params: + if pair[1] in state_dict: + state_dict[pair[0]] = state_dict[pair[1]] + + return state_dict + + +def zero3_partitioned_param_info(unpartitioned_numel, world_size): + remainder = unpartitioned_numel % world_size + padding_numel = (world_size - remainder) if remainder else 0 + partitioned_numel = math.ceil(unpartitioned_numel / world_size) + return partitioned_numel, padding_numel + + +def _zero3_merge_frozen_params(state_dict, world_size, zero_model_states): + if zero_model_states[0].frozen_param_shapes is None or len(zero_model_states[0].frozen_param_shapes) == 0: + return + + if debug: + for i in range(world_size): + num_elem = sum(s.numel() for s in zero_model_states[i].frozen_param_fragments.values()) + print(f'rank {i}: {FROZEN_PARAM_SHAPES}.numel = {num_elem}') + + frozen_param_shapes = zero_model_states[0].frozen_param_shapes + wanted_params = len(frozen_param_shapes) + wanted_numel = sum(s.numel() for s in frozen_param_shapes.values()) + avail_numel = sum([p.numel() for p in zero_model_states[0].frozen_param_fragments.values()]) * world_size + print(f'Frozen params: Have {avail_numel} numels to process.') + print(f'Frozen params: Need {wanted_numel} numels in {wanted_params} params') + + total_params = 0 + total_numel = 0 + for name, shape in zero_model_states[0].frozen_param_shapes.items(): + total_params += 1 + unpartitioned_numel = shape.numel() + total_numel += unpartitioned_numel + + param_frags = tuple(model_state.frozen_param_fragments[name] for model_state in zero_model_states) + state_dict[name] = torch.cat(param_frags, 0).narrow(0, 0, unpartitioned_numel).view(shape) + + partitioned_numel, partitioned_padding_numel = zero3_partitioned_param_info(unpartitioned_numel, world_size) + + if debug: + print( + f"Frozen params: {total_params} {name} full shape: {shape} partition0 numel={partitioned_numel} partitioned_padding_numel={partitioned_padding_numel}" + ) + + print(f"Reconstructed Frozen fp32 state dict with {total_params} params {total_numel} elements") + + +def _zero3_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states): + param_shapes = zero_model_states[0].param_shapes + avail_numel = fp32_flat_groups[0].numel() * world_size + # Reconstruction protocol: For zero3 we need to zip the partitions together at boundary of each + # param, re-consolidating each param, while dealing with padding if any + + # merge list of dicts, preserving order + param_shapes = {k: v for d in param_shapes for k, v in d.items()} + + if debug: + for i in range(world_size): + print(f"{FP32_FLAT_GROUPS}[{i}].shape={fp32_flat_groups[i].shape}") + + wanted_params = len(param_shapes) + wanted_numel = sum(shape.numel() for shape in param_shapes.values()) + # not asserting if there is a mismatch due to possible padding + avail_numel = fp32_flat_groups[0].numel() * world_size + print(f"Trainable params: Have {avail_numel} numels to process.") + print(f"Trainable params: Need {wanted_numel} numels in {wanted_params} params.") + + # params + # XXX: for huge models that can't fit into the host's RAM we will have to recode this to support + # out-of-core computing solution + offset = 0 + total_numel = 0 + total_params = 0 + for name, shape in param_shapes.items(): + + unpartitioned_numel = shape.numel() + total_numel += unpartitioned_numel + total_params += 1 + + partitioned_numel, partitioned_padding_numel = zero3_partitioned_param_info(unpartitioned_numel, world_size) + + if debug: + print( + f"Trainable params: {total_params} {name} full shape: {shape} partition0 numel={partitioned_numel} partitioned_padding_numel={partitioned_padding_numel}" + ) + + # XXX: memory usage doubles here + state_dict[name] = torch.cat( + tuple(fp32_flat_groups[i].narrow(0, offset, partitioned_numel) for i in range(world_size)), + 0).narrow(0, 0, unpartitioned_numel).view(shape) + offset += partitioned_numel + + offset *= world_size + + # Sanity check + if offset != avail_numel: + raise ValueError(f"consumed {offset} numels out of {avail_numel} - something is wrong") + + print(f"Reconstructed Trainable fp32 state dict with {total_params} params {total_numel} elements") + + +def _get_fp32_state_dict_from_zero3_checkpoint(world_size, fp32_flat_groups, zero_model_states, + exclude_frozen_parameters): + state_dict = OrderedDict() + + # buffers + buffers = zero_model_states[0].buffers + state_dict.update(buffers) + if debug: + print(f"added {len(buffers)} buffers") + + if not exclude_frozen_parameters: + _zero3_merge_frozen_params(state_dict, world_size, zero_model_states) + + _zero3_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states) + + # recover shared parameters + for pair in zero_model_states[0].shared_params: + if pair[1] in state_dict: + state_dict[pair[0]] = state_dict[pair[1]] + + return state_dict + + +def get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir, tag=None, exclude_frozen_parameters=False): + """ + Convert ZeRO 2 or 3 checkpoint into a single fp32 consolidated state_dict that can be loaded with + ``load_state_dict()`` and used for training without DeepSpeed or shared with others, for example + via a model hub. + + Args: + - ``checkpoint_dir``: path to the desired checkpoint folder + - ``tag``: checkpoint tag used as a unique identifier for checkpoint. If not provided will attempt to load tag in 'latest' file. e.g., ``global_step14`` + - ``exclude_frozen_parameters``: exclude frozen parameters + + Returns: + - pytorch ``state_dict`` + + Note: this approach may not work if your application doesn't have sufficient free CPU memory and + you may need to use the offline approach using the ``zero_to_fp32.py`` script that is saved with + the checkpoint. + + A typical usage might be :: + + from deepspeed.utils.zero_to_fp32 import get_fp32_state_dict_from_zero_checkpoint + # do the training and checkpoint saving + state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir) # already on cpu + model = model.cpu() # move to cpu + model.load_state_dict(state_dict) + # submit to model hub or save the model to share with others + + In this example the ``model`` will no longer be usable in the deepspeed context of the same + application. i.e. you will need to re-initialize the deepspeed engine, since + ``model.load_state_dict(state_dict)`` will remove all the deepspeed magic from it. + + If you want it all done for you, use ``load_state_dict_from_zero_checkpoint`` instead. + + """ + if tag is None: + latest_path = os.path.join(checkpoint_dir, 'latest') + if os.path.isfile(latest_path): + with open(latest_path, 'r') as fd: + tag = fd.read().strip() + else: + raise ValueError(f"Unable to find 'latest' file at {latest_path}") + + ds_checkpoint_dir = os.path.join(checkpoint_dir, tag) + + if not os.path.isdir(ds_checkpoint_dir): + raise FileNotFoundError(f"Directory '{ds_checkpoint_dir}' doesn't exist") + + return _get_fp32_state_dict_from_zero_checkpoint(ds_checkpoint_dir, exclude_frozen_parameters) + + +def convert_zero_checkpoint_to_fp32_state_dict(checkpoint_dir, output_file, tag=None, exclude_frozen_parameters=False): + """ + Convert ZeRO 2 or 3 checkpoint into a single fp32 consolidated ``state_dict`` file that can be + loaded with ``torch.load(file)`` + ``load_state_dict()`` and used for training without DeepSpeed. + + Args: + - ``checkpoint_dir``: path to the desired checkpoint folder. (one that contains the tag-folder, like ``global_step14``) + - ``output_file``: path to the pytorch fp32 state_dict output file (e.g. path/pytorch_model.bin) + - ``tag``: checkpoint tag used as a unique identifier for checkpoint. If not provided will attempt to load tag in the file named ``latest`` in the checkpoint folder, e.g., ``global_step14`` + - ``exclude_frozen_parameters``: exclude frozen parameters + """ + + state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir, tag, exclude_frozen_parameters) + print(f"Saving fp32 state dict to {output_file}") + torch.save(state_dict, output_file) + + +def load_state_dict_from_zero_checkpoint(model, checkpoint_dir, tag=None): + """ + 1. Put the provided model to cpu + 2. Convert ZeRO 2 or 3 checkpoint into a single fp32 consolidated ``state_dict`` + 3. Load it into the provided model + + Args: + - ``model``: the model object to update + - ``checkpoint_dir``: path to the desired checkpoint folder. (one that contains the tag-folder, like ``global_step14``) + - ``tag``: checkpoint tag used as a unique identifier for checkpoint. If not provided will attempt to load tag in the file named ``latest`` in the checkpoint folder, e.g., ``global_step14`` + + Returns: + - ``model`: modified model + + Make sure you have plenty of CPU memory available before you call this function. If you don't + have enough use the ``zero_to_fp32.py`` utility to do the conversion. You will find it + conveniently placed for you in the checkpoint folder. + + A typical usage might be :: + + from deepspeed.utils.zero_to_fp32 import load_state_dict_from_zero_checkpoint + model = load_state_dict_from_zero_checkpoint(trainer.model, checkpoint_dir) + # submit to model hub or save the model to share with others + + Note, that once this was run, the ``model`` will no longer be usable in the deepspeed context + of the same application. i.e. you will need to re-initialize the deepspeed engine, since + ``model.load_state_dict(state_dict)`` will remove all the deepspeed magic from it. + + """ + logger.info(f"Extracting fp32 weights") + state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir, tag) + + logger.info(f"Overwriting model with fp32 weights") + model = model.cpu() + model.load_state_dict(state_dict, strict=False) + + return model + + +if __name__ == "__main__": + + parser = argparse.ArgumentParser() + parser.add_argument("checkpoint_dir", + type=str, + help="path to the desired checkpoint folder, e.g., path/checkpoint-12") + parser.add_argument( + "output_file", + type=str, + help="path to the pytorch fp32 state_dict output file (e.g. path/checkpoint-12/pytorch_model.bin)") + parser.add_argument("-t", + "--tag", + type=str, + default=None, + help="checkpoint tag used as a unique identifier for checkpoint. e.g., global_step1") + parser.add_argument("--exclude_frozen_parameters", action='store_true', help="exclude frozen parameters") + parser.add_argument("-d", "--debug", action='store_true', help="enable debug") + args = parser.parse_args() + + debug = args.debug + + convert_zero_checkpoint_to_fp32_state_dict(args.checkpoint_dir, + args.output_file, + tag=args.tag, + exclude_frozen_parameters=args.exclude_frozen_parameters) diff --git a/babylm_shuffle_nondeterministic_10M_seed0/runs/checkpoint-1700/config.json b/babylm_shuffle_nondeterministic_10M_seed0/runs/checkpoint-1700/config.json new file mode 100644 index 0000000000000000000000000000000000000000..fb0f9a53a507ffcdb60410deb71d60ef801bf350 --- /dev/null +++ b/babylm_shuffle_nondeterministic_10M_seed0/runs/checkpoint-1700/config.json @@ -0,0 +1,36 @@ +{ + "_name_or_path": "meta-llama/Llama-3.2-3B", + "architectures": [ + "LlamaForCausalLM" + ], + "attention_bias": false, + "attention_dropout": 0.0, + "bos_token_id": 128000, + "eos_token_id": 128001, + "head_dim": 128, + "hidden_act": "silu", + "hidden_size": 3072, + "initializer_range": 0.02, + "intermediate_size": 8192, + "max_position_embeddings": 131072, + "mlp_bias": false, + "model_type": "llama", + "num_attention_heads": 24, + "num_hidden_layers": 28, + "num_key_value_heads": 8, + "pretraining_tp": 1, + "rms_norm_eps": 1e-05, + "rope_scaling": { + "factor": 32.0, + "high_freq_factor": 4.0, + "low_freq_factor": 1.0, + "original_max_position_embeddings": 8192, + "rope_type": "llama3" + }, + "rope_theta": 500000.0, + "tie_word_embeddings": true, + "torch_dtype": "float16", + "transformers_version": "4.45.1", + "use_cache": true, + "vocab_size": 128256 +} diff --git a/babylm_shuffle_nondeterministic_10M_seed0/runs/checkpoint-1700/generation_config.json b/babylm_shuffle_nondeterministic_10M_seed0/runs/checkpoint-1700/generation_config.json new file mode 100644 index 0000000000000000000000000000000000000000..9c389b841a9a29ddd904e02d1eb0e08dcce82ad9 --- /dev/null +++ b/babylm_shuffle_nondeterministic_10M_seed0/runs/checkpoint-1700/generation_config.json @@ -0,0 +1,9 @@ +{ + "_from_model_config": true, + "bos_token_id": 128000, + "do_sample": true, + "eos_token_id": 128001, + "temperature": 0.6, + "top_p": 0.9, + "transformers_version": "4.45.1" +} diff --git a/babylm_shuffle_nondeterministic_10M_seed0/runs/checkpoint-1700/latest b/babylm_shuffle_nondeterministic_10M_seed0/runs/checkpoint-1700/latest new file mode 100644 index 0000000000000000000000000000000000000000..cb24338af451be6d3ccddc18a950b58a70589761 --- /dev/null +++ b/babylm_shuffle_nondeterministic_10M_seed0/runs/checkpoint-1700/latest @@ -0,0 +1 @@ +global_step1700 \ No newline at end of file diff --git a/babylm_shuffle_nondeterministic_10M_seed0/runs/checkpoint-1700/model.safetensors.index.json b/babylm_shuffle_nondeterministic_10M_seed0/runs/checkpoint-1700/model.safetensors.index.json new file mode 100644 index 0000000000000000000000000000000000000000..ed64de846d720b9a7859dc20575fea8e8ca51940 --- /dev/null +++ b/babylm_shuffle_nondeterministic_10M_seed0/runs/checkpoint-1700/model.safetensors.index.json @@ -0,0 +1,262 @@ +{ + "metadata": { + "total_size": 7213504512 + }, + "weight_map": { + "lm_head.weight": "model-00002-of-00002.safetensors", + "model.embed_tokens.weight": "model-00001-of-00002.safetensors", + "model.layers.0.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.0.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.0.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.0.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.0.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.0.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.0.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.0.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.0.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.1.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.1.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.1.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.1.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.1.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.1.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.1.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.1.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.1.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.10.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.10.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.10.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.10.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.10.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.10.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.10.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.10.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.10.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.11.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.11.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.11.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.11.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.11.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.11.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.11.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.11.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.11.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.12.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.12.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.12.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.12.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.12.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.12.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.12.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.12.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.12.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.13.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.13.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.13.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.13.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.13.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.13.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.13.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.13.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.13.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.14.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.14.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.14.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.14.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.14.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.14.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.14.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.14.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.14.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.15.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.15.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.15.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.15.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.15.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.15.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.15.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.15.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.15.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.16.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.16.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.16.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.16.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.16.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.16.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.16.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.16.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.16.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.17.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.17.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.17.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.17.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.17.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.17.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.17.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.17.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.17.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.18.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.18.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.18.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.18.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.18.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.18.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.18.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.18.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.18.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.19.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.19.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.19.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.19.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.19.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.19.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.19.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.19.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.19.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.2.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.2.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.2.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.2.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.2.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.2.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.2.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.2.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.2.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.20.input_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.20.mlp.down_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.20.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.20.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.20.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.20.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.20.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.20.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.20.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.21.input_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.21.mlp.down_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.21.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.21.mlp.up_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.21.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.21.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.21.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.21.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.21.self_attn.v_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.22.input_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.22.mlp.down_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.22.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.22.mlp.up_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.22.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.22.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.22.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.22.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.22.self_attn.v_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.23.input_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.23.mlp.down_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.23.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.23.mlp.up_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.23.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.23.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.23.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.23.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.23.self_attn.v_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.24.input_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.24.mlp.down_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.24.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.24.mlp.up_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.24.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.24.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.24.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.24.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.24.self_attn.v_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.25.input_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.25.mlp.down_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.25.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.25.mlp.up_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.25.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.25.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.25.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.25.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.25.self_attn.v_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.26.input_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.26.mlp.down_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.26.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.26.mlp.up_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.26.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.26.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.26.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.26.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.26.self_attn.v_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.27.input_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.27.mlp.down_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.27.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.27.mlp.up_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.27.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.27.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.27.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.27.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.27.self_attn.v_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.3.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.3.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.3.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.3.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.3.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.3.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.3.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.3.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.3.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.4.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.4.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.4.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.4.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.4.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.4.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.4.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.4.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.4.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.5.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.5.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.5.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.5.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.5.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.5.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.5.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.5.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.5.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.6.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.6.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.6.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.6.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.6.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.6.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.6.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.6.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.6.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.7.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.7.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.7.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.7.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.7.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.7.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.7.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.7.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.7.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.8.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.8.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.8.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.8.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.8.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.8.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.8.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.8.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.8.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.9.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.9.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.9.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.9.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.9.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.9.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.9.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.9.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.9.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.norm.weight": "model-00002-of-00002.safetensors" + } +} diff --git a/babylm_shuffle_nondeterministic_10M_seed0/runs/checkpoint-1700/special_tokens_map.json b/babylm_shuffle_nondeterministic_10M_seed0/runs/checkpoint-1700/special_tokens_map.json new file mode 100644 index 0000000000000000000000000000000000000000..6949c5975ee0e961ef61cf31010dce04df0a03f8 --- /dev/null +++ b/babylm_shuffle_nondeterministic_10M_seed0/runs/checkpoint-1700/special_tokens_map.json @@ -0,0 +1,23 @@ +{ + "bos_token": { + "content": "<|begin_of_text|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false + }, + "eos_token": { + "content": "<|end_of_text|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false + }, + "pad_token": { + "content": "[PAD]", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false + } +} diff --git a/babylm_shuffle_nondeterministic_10M_seed0/runs/checkpoint-1700/tokenizer_config.json b/babylm_shuffle_nondeterministic_10M_seed0/runs/checkpoint-1700/tokenizer_config.json new file mode 100644 index 0000000000000000000000000000000000000000..5684b14e2756b56aa397c1324e239b442162b34a --- /dev/null +++ b/babylm_shuffle_nondeterministic_10M_seed0/runs/checkpoint-1700/tokenizer_config.json @@ -0,0 +1,2070 @@ +{ + "added_tokens_decoder": { + "128000": { + "content": "<|begin_of_text|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128001": { + "content": "<|end_of_text|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128002": { + "content": "<|reserved_special_token_0|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128003": { + "content": "<|reserved_special_token_1|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128004": { + "content": "<|finetune_right_pad_id|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128005": { + "content": "<|reserved_special_token_2|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128006": { + "content": "<|start_header_id|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128007": { + "content": "<|end_header_id|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128008": { + "content": "<|eom_id|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128009": { + "content": "<|eot_id|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128010": { + "content": "<|python_tag|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128011": { + "content": "<|reserved_special_token_3|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128012": { + "content": "<|reserved_special_token_4|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128013": { + "content": "<|reserved_special_token_5|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128014": { + "content": "<|reserved_special_token_6|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128015": { + "content": "<|reserved_special_token_7|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128016": { + "content": "<|reserved_special_token_8|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128017": { + "content": "<|reserved_special_token_9|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128018": { + "content": "<|reserved_special_token_10|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128019": { + "content": "<|reserved_special_token_11|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128020": { + "content": "<|reserved_special_token_12|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128021": { + "content": "<|reserved_special_token_13|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128022": { + "content": "<|reserved_special_token_14|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128023": { + "content": "<|reserved_special_token_15|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128024": { + "content": "<|reserved_special_token_16|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128025": { + "content": "<|reserved_special_token_17|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128026": { + "content": "<|reserved_special_token_18|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128027": { + "content": "<|reserved_special_token_19|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128028": { + "content": "<|reserved_special_token_20|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128029": { + "content": "<|reserved_special_token_21|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128030": { + "content": "<|reserved_special_token_22|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128031": { + "content": "<|reserved_special_token_23|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128032": { + "content": "<|reserved_special_token_24|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128033": { + "content": "<|reserved_special_token_25|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128034": { + "content": "<|reserved_special_token_26|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128035": { + "content": "<|reserved_special_token_27|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128036": { + "content": "<|reserved_special_token_28|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128037": { + "content": "<|reserved_special_token_29|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128038": { + "content": "<|reserved_special_token_30|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128039": { + "content": "<|reserved_special_token_31|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128040": { + "content": "<|reserved_special_token_32|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128041": { + "content": "<|reserved_special_token_33|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128042": { + "content": "<|reserved_special_token_34|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128043": { + "content": "<|reserved_special_token_35|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128044": { + "content": "<|reserved_special_token_36|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128045": { + "content": "<|reserved_special_token_37|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128046": { + "content": "<|reserved_special_token_38|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128047": { + "content": "<|reserved_special_token_39|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128048": { + "content": "<|reserved_special_token_40|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128049": { + "content": "<|reserved_special_token_41|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128050": { + "content": "<|reserved_special_token_42|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128051": { + "content": "<|reserved_special_token_43|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128052": { + "content": "<|reserved_special_token_44|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128053": { + "content": "<|reserved_special_token_45|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128054": { + "content": "<|reserved_special_token_46|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128055": { + "content": "<|reserved_special_token_47|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128056": { + "content": "<|reserved_special_token_48|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128057": { + "content": "<|reserved_special_token_49|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128058": { + "content": "<|reserved_special_token_50|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128059": { + "content": "<|reserved_special_token_51|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128060": { + "content": "<|reserved_special_token_52|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128061": { + "content": "<|reserved_special_token_53|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128062": { + "content": "<|reserved_special_token_54|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128063": { + "content": "<|reserved_special_token_55|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128064": { + "content": "<|reserved_special_token_56|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128065": { + "content": "<|reserved_special_token_57|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128066": { + "content": "<|reserved_special_token_58|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128067": { + "content": "<|reserved_special_token_59|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128068": { + "content": "<|reserved_special_token_60|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128069": { + "content": "<|reserved_special_token_61|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128070": { + "content": "<|reserved_special_token_62|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128071": { + "content": "<|reserved_special_token_63|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128072": { + "content": "<|reserved_special_token_64|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128073": { + "content": "<|reserved_special_token_65|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128074": { + "content": "<|reserved_special_token_66|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128075": { + "content": "<|reserved_special_token_67|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128076": { + "content": "<|reserved_special_token_68|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128077": { + "content": "<|reserved_special_token_69|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128078": { + "content": "<|reserved_special_token_70|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128079": { + "content": "<|reserved_special_token_71|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128080": { + "content": "<|reserved_special_token_72|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128081": { + "content": "<|reserved_special_token_73|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128082": { + "content": "<|reserved_special_token_74|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128083": { + "content": "<|reserved_special_token_75|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128084": { + "content": "<|reserved_special_token_76|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128085": { + "content": "<|reserved_special_token_77|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128086": { + "content": "<|reserved_special_token_78|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128087": { + "content": "<|reserved_special_token_79|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128088": { + "content": "<|reserved_special_token_80|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128089": { + "content": "<|reserved_special_token_81|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128090": { + "content": "<|reserved_special_token_82|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128091": { + "content": "<|reserved_special_token_83|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128092": { + "content": "<|reserved_special_token_84|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128093": { + "content": "<|reserved_special_token_85|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128094": { + "content": "<|reserved_special_token_86|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128095": { + "content": "<|reserved_special_token_87|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128096": { + "content": "<|reserved_special_token_88|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128097": { + "content": "<|reserved_special_token_89|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128098": { + "content": "<|reserved_special_token_90|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128099": { + "content": "<|reserved_special_token_91|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128100": { + "content": "<|reserved_special_token_92|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128101": { + "content": "<|reserved_special_token_93|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128102": { + "content": "<|reserved_special_token_94|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128103": { + "content": "<|reserved_special_token_95|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128104": { + "content": "<|reserved_special_token_96|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128105": { + "content": "<|reserved_special_token_97|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128106": { + "content": "<|reserved_special_token_98|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128107": { + "content": "<|reserved_special_token_99|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128108": { + "content": "<|reserved_special_token_100|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128109": { + "content": "<|reserved_special_token_101|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128110": { + "content": "<|reserved_special_token_102|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128111": { + "content": "<|reserved_special_token_103|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128112": { + "content": "<|reserved_special_token_104|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128113": { + "content": "<|reserved_special_token_105|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128114": { + "content": "<|reserved_special_token_106|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128115": { + "content": "<|reserved_special_token_107|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128116": { + "content": "<|reserved_special_token_108|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128117": { + "content": "<|reserved_special_token_109|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128118": { + "content": "<|reserved_special_token_110|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128119": { + "content": "<|reserved_special_token_111|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128120": { + "content": "<|reserved_special_token_112|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128121": { + "content": "<|reserved_special_token_113|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128122": { + "content": "<|reserved_special_token_114|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128123": { + "content": "<|reserved_special_token_115|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128124": { + "content": "<|reserved_special_token_116|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128125": { + "content": "<|reserved_special_token_117|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128126": { + "content": "<|reserved_special_token_118|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128127": { + "content": "<|reserved_special_token_119|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128128": { + "content": "<|reserved_special_token_120|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128129": { + "content": "<|reserved_special_token_121|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128130": { + "content": "<|reserved_special_token_122|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128131": { + "content": "<|reserved_special_token_123|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128132": { + "content": "<|reserved_special_token_124|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128133": { + "content": "<|reserved_special_token_125|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128134": { + "content": "<|reserved_special_token_126|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128135": { + "content": "<|reserved_special_token_127|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128136": { + "content": "<|reserved_special_token_128|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128137": { + "content": "<|reserved_special_token_129|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128138": { + "content": "<|reserved_special_token_130|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128139": { + "content": "<|reserved_special_token_131|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128140": { + "content": "<|reserved_special_token_132|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128141": { + "content": "<|reserved_special_token_133|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128142": { + "content": "<|reserved_special_token_134|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128143": { + "content": "<|reserved_special_token_135|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128144": { + "content": "<|reserved_special_token_136|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128145": { + "content": "<|reserved_special_token_137|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128146": { + "content": "<|reserved_special_token_138|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128147": { + "content": "<|reserved_special_token_139|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128148": { + "content": "<|reserved_special_token_140|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128149": { + "content": "<|reserved_special_token_141|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128150": { + "content": "<|reserved_special_token_142|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128151": { + "content": "<|reserved_special_token_143|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128152": { + "content": "<|reserved_special_token_144|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128153": { + "content": "<|reserved_special_token_145|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128154": { + "content": "<|reserved_special_token_146|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128155": { + "content": "<|reserved_special_token_147|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128156": { + "content": "<|reserved_special_token_148|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128157": { + "content": "<|reserved_special_token_149|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128158": { + "content": "<|reserved_special_token_150|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128159": { + "content": "<|reserved_special_token_151|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128160": { + "content": "<|reserved_special_token_152|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128161": { + "content": "<|reserved_special_token_153|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128162": { + "content": "<|reserved_special_token_154|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128163": { + "content": "<|reserved_special_token_155|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128164": { + "content": "<|reserved_special_token_156|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128165": { + "content": "<|reserved_special_token_157|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128166": { + "content": "<|reserved_special_token_158|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128167": { + "content": "<|reserved_special_token_159|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128168": { + "content": "<|reserved_special_token_160|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128169": { + "content": "<|reserved_special_token_161|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128170": { + "content": "<|reserved_special_token_162|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128171": { + "content": "<|reserved_special_token_163|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128172": { + "content": "<|reserved_special_token_164|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128173": { + "content": "<|reserved_special_token_165|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128174": { + "content": "<|reserved_special_token_166|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128175": { + "content": "<|reserved_special_token_167|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128176": { + "content": "<|reserved_special_token_168|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128177": { + "content": "<|reserved_special_token_169|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128178": { + "content": "<|reserved_special_token_170|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128179": { + "content": "<|reserved_special_token_171|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128180": { + "content": "<|reserved_special_token_172|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128181": { + "content": "<|reserved_special_token_173|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128182": { + "content": "<|reserved_special_token_174|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128183": { + "content": "<|reserved_special_token_175|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128184": { + "content": "<|reserved_special_token_176|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128185": { + "content": "<|reserved_special_token_177|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128186": { + "content": "<|reserved_special_token_178|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128187": { + "content": "<|reserved_special_token_179|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128188": { + "content": "<|reserved_special_token_180|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128189": { + "content": "<|reserved_special_token_181|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128190": { + "content": "<|reserved_special_token_182|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128191": { + "content": "<|reserved_special_token_183|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128192": { + "content": "<|reserved_special_token_184|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128193": { + "content": "<|reserved_special_token_185|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128194": { + "content": "<|reserved_special_token_186|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128195": { + "content": "<|reserved_special_token_187|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128196": { + "content": "<|reserved_special_token_188|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128197": { + "content": "<|reserved_special_token_189|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128198": { + "content": "<|reserved_special_token_190|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128199": { + "content": "<|reserved_special_token_191|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128200": { + "content": "<|reserved_special_token_192|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128201": { + "content": "<|reserved_special_token_193|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128202": { + "content": "<|reserved_special_token_194|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128203": { + "content": "<|reserved_special_token_195|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128204": { + "content": "<|reserved_special_token_196|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128205": { + "content": "<|reserved_special_token_197|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128206": { + "content": "<|reserved_special_token_198|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128207": { + "content": "<|reserved_special_token_199|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128208": { + "content": "<|reserved_special_token_200|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128209": { + "content": "<|reserved_special_token_201|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128210": { + "content": "<|reserved_special_token_202|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128211": { + "content": "<|reserved_special_token_203|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128212": { + "content": "<|reserved_special_token_204|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128213": { + "content": "<|reserved_special_token_205|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128214": { + "content": "<|reserved_special_token_206|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128215": { + "content": "<|reserved_special_token_207|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128216": { + "content": "<|reserved_special_token_208|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128217": { + "content": "<|reserved_special_token_209|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128218": { + "content": "<|reserved_special_token_210|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128219": { + "content": "<|reserved_special_token_211|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128220": { + "content": "<|reserved_special_token_212|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128221": { + "content": "<|reserved_special_token_213|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128222": { + "content": "<|reserved_special_token_214|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128223": { + "content": "<|reserved_special_token_215|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128224": { + "content": "<|reserved_special_token_216|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128225": { + "content": "<|reserved_special_token_217|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128226": { + "content": "<|reserved_special_token_218|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128227": { + "content": "<|reserved_special_token_219|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128228": { + "content": "<|reserved_special_token_220|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128229": { + "content": "<|reserved_special_token_221|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128230": { + "content": "<|reserved_special_token_222|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128231": { + "content": "<|reserved_special_token_223|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128232": { + "content": "<|reserved_special_token_224|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128233": { + "content": "<|reserved_special_token_225|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128234": { + "content": "<|reserved_special_token_226|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128235": { + "content": "<|reserved_special_token_227|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128236": { + "content": "<|reserved_special_token_228|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128237": { + "content": "<|reserved_special_token_229|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128238": { + "content": "<|reserved_special_token_230|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128239": { + "content": "<|reserved_special_token_231|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128240": { + "content": "<|reserved_special_token_232|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128241": { + "content": "<|reserved_special_token_233|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128242": { + "content": "<|reserved_special_token_234|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128243": { + "content": "<|reserved_special_token_235|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128244": { + "content": "<|reserved_special_token_236|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128245": { + "content": "<|reserved_special_token_237|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128246": { + "content": "<|reserved_special_token_238|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128247": { + "content": "<|reserved_special_token_239|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128248": { + "content": "<|reserved_special_token_240|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128249": { + "content": "<|reserved_special_token_241|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128250": { + "content": "<|reserved_special_token_242|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128251": { + "content": "<|reserved_special_token_243|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128252": { + "content": "<|reserved_special_token_244|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128253": { + "content": "<|reserved_special_token_245|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128254": { + "content": "<|reserved_special_token_246|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128255": { + "content": "<|reserved_special_token_247|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128256": { + "content": "[PAD]", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + } + }, + "bos_token": "<|begin_of_text|>", + "clean_up_tokenization_spaces": true, + "eos_token": "<|end_of_text|>", + "model_input_names": [ + "input_ids", + "attention_mask" + ], + "model_max_length": 131072, + "pad_token": "[PAD]", + "tokenizer_class": "PreTrainedTokenizerFast" +} diff --git a/babylm_shuffle_nondeterministic_10M_seed0/runs/checkpoint-1700/trainer_state.json b/babylm_shuffle_nondeterministic_10M_seed0/runs/checkpoint-1700/trainer_state.json new file mode 100644 index 0000000000000000000000000000000000000000..4e5c5cfe16fe567c60b44b6f55447823ecae55b1 --- /dev/null +++ b/babylm_shuffle_nondeterministic_10M_seed0/runs/checkpoint-1700/trainer_state.json @@ -0,0 +1,13293 @@ +{ + "best_metric": null, + "best_model_checkpoint": null, + "epoch": 1.8630136986301369, + "eval_steps": 10, + "global_step": 1700, + "is_hyper_param_search": false, + "is_local_process_zero": true, + "is_world_process_zero": true, + "log_history": [ + { + "epoch": 0.001095890410958904, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.0998, + "step": 1 + }, + { + "epoch": 0.002191780821917808, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.0858, + "step": 2 + }, + { + "epoch": 0.0032876712328767125, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.0672, + "step": 3 + }, + { + "epoch": 0.004383561643835616, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.0802, + "step": 4 + }, + { + "epoch": 0.005479452054794521, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.0636, + "step": 5 + }, + { + "epoch": 0.006575342465753425, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.1672, + "step": 6 + }, + { + "epoch": 0.007671232876712329, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.0961, + "step": 7 + }, + { + "epoch": 0.008767123287671232, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.1258, + "step": 8 + }, + { + "epoch": 0.009863013698630137, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.0953, + "step": 9 + }, + { + "epoch": 0.010958904109589041, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.1404, + "step": 10 + }, + { + "epoch": 0.010958904109589041, + "eval_loss": 3.12593412399292, + "eval_runtime": 42.732, + "eval_samples_per_second": 23.402, + "eval_steps_per_second": 0.983, + "step": 10 + }, + { + "epoch": 0.012054794520547946, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.1187, + "step": 11 + }, + { + "epoch": 0.01315068493150685, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.0593, + "step": 12 + }, + { + "epoch": 0.014246575342465753, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.111, + "step": 13 + }, + { + "epoch": 0.015342465753424657, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.0974, + "step": 14 + }, + { + "epoch": 0.01643835616438356, + "grad_norm": 5.991453170776367, + "learning_rate": 1.8248175182481753e-08, + "loss": 3.135, + "step": 15 + }, + { + "epoch": 0.017534246575342465, + "grad_norm": 6.274210453033447, + "learning_rate": 3.6496350364963505e-08, + "loss": 3.083, + "step": 16 + }, + { + "epoch": 0.01863013698630137, + "grad_norm": 6.644786834716797, + "learning_rate": 5.4744525547445264e-08, + "loss": 3.0856, + "step": 17 + }, + { + "epoch": 0.019726027397260273, + "grad_norm": 6.769156455993652, + "learning_rate": 7.299270072992701e-08, + "loss": 3.1082, + "step": 18 + }, + { + "epoch": 0.020821917808219178, + "grad_norm": 6.771246910095215, + "learning_rate": 9.124087591240876e-08, + "loss": 3.0753, + "step": 19 + }, + { + "epoch": 0.021917808219178082, + "grad_norm": 6.468441486358643, + "learning_rate": 1.0948905109489053e-07, + "loss": 3.1478, + "step": 20 + }, + { + "epoch": 0.021917808219178082, + "eval_loss": 3.1245107650756836, + "eval_runtime": 43.6251, + "eval_samples_per_second": 22.923, + "eval_steps_per_second": 0.963, + "step": 20 + }, + { + "epoch": 0.023013698630136987, + "grad_norm": 6.83078670501709, + "learning_rate": 1.2773722627737227e-07, + "loss": 3.1213, + "step": 21 + }, + { + "epoch": 0.02410958904109589, + "grad_norm": 6.221287250518799, + "learning_rate": 1.4598540145985402e-07, + "loss": 3.0549, + "step": 22 + }, + { + "epoch": 0.025205479452054796, + "grad_norm": 5.564266204833984, + "learning_rate": 1.642335766423358e-07, + "loss": 3.0769, + "step": 23 + }, + { + "epoch": 0.0263013698630137, + "grad_norm": 5.735209941864014, + "learning_rate": 1.8248175182481753e-07, + "loss": 3.1124, + "step": 24 + }, + { + "epoch": 0.0273972602739726, + "grad_norm": 5.577925682067871, + "learning_rate": 2.0072992700729928e-07, + "loss": 3.1191, + "step": 25 + }, + { + "epoch": 0.028493150684931506, + "grad_norm": 4.814139366149902, + "learning_rate": 2.1897810218978106e-07, + "loss": 3.0947, + "step": 26 + }, + { + "epoch": 0.02958904109589041, + "grad_norm": 4.69775390625, + "learning_rate": 2.3722627737226278e-07, + "loss": 3.0997, + "step": 27 + }, + { + "epoch": 0.030684931506849315, + "grad_norm": 4.811586380004883, + "learning_rate": 2.5547445255474454e-07, + "loss": 3.0749, + "step": 28 + }, + { + "epoch": 0.03178082191780822, + "grad_norm": 4.623377323150635, + "learning_rate": 2.737226277372263e-07, + "loss": 3.0619, + "step": 29 + }, + { + "epoch": 0.03287671232876712, + "grad_norm": 4.721253395080566, + "learning_rate": 2.9197080291970804e-07, + "loss": 3.0371, + "step": 30 + }, + { + "epoch": 0.03287671232876712, + "eval_loss": 3.082857847213745, + "eval_runtime": 44.127, + "eval_samples_per_second": 22.662, + "eval_steps_per_second": 0.952, + "step": 30 + }, + { + "epoch": 0.03397260273972603, + "grad_norm": 3.968630790710449, + "learning_rate": 3.102189781021898e-07, + "loss": 3.0535, + "step": 31 + }, + { + "epoch": 0.03506849315068493, + "grad_norm": 4.4373779296875, + "learning_rate": 3.284671532846716e-07, + "loss": 3.0476, + "step": 32 + }, + { + "epoch": 0.03616438356164384, + "grad_norm": 4.479320049285889, + "learning_rate": 3.467153284671533e-07, + "loss": 3.0215, + "step": 33 + }, + { + "epoch": 0.03726027397260274, + "grad_norm": 4.122231960296631, + "learning_rate": 3.6496350364963505e-07, + "loss": 3.0106, + "step": 34 + }, + { + "epoch": 0.038356164383561646, + "grad_norm": 4.153995513916016, + "learning_rate": 3.8321167883211686e-07, + "loss": 3.0318, + "step": 35 + }, + { + "epoch": 0.03945205479452055, + "grad_norm": 3.6949076652526855, + "learning_rate": 4.0145985401459856e-07, + "loss": 3.046, + "step": 36 + }, + { + "epoch": 0.040547945205479455, + "grad_norm": 3.4610488414764404, + "learning_rate": 4.197080291970803e-07, + "loss": 3.002, + "step": 37 + }, + { + "epoch": 0.041643835616438356, + "grad_norm": 3.9523138999938965, + "learning_rate": 4.379562043795621e-07, + "loss": 2.9665, + "step": 38 + }, + { + "epoch": 0.042739726027397264, + "grad_norm": 3.68516206741333, + "learning_rate": 4.562043795620438e-07, + "loss": 3.0411, + "step": 39 + }, + { + "epoch": 0.043835616438356165, + "grad_norm": 3.692396402359009, + "learning_rate": 4.7445255474452557e-07, + "loss": 2.9666, + "step": 40 + }, + { + "epoch": 0.043835616438356165, + "eval_loss": 3.015512704849243, + "eval_runtime": 48.2438, + "eval_samples_per_second": 20.728, + "eval_steps_per_second": 0.871, + "step": 40 + }, + { + "epoch": 0.044931506849315066, + "grad_norm": 3.3765921592712402, + "learning_rate": 4.927007299270073e-07, + "loss": 2.9831, + "step": 41 + }, + { + "epoch": 0.046027397260273974, + "grad_norm": 2.9631125926971436, + "learning_rate": 5.109489051094891e-07, + "loss": 3.0126, + "step": 42 + }, + { + "epoch": 0.047123287671232875, + "grad_norm": 2.969730854034424, + "learning_rate": 5.291970802919708e-07, + "loss": 2.9803, + "step": 43 + }, + { + "epoch": 0.04821917808219178, + "grad_norm": 3.185237407684326, + "learning_rate": 5.474452554744526e-07, + "loss": 2.9888, + "step": 44 + }, + { + "epoch": 0.049315068493150684, + "grad_norm": 3.055579423904419, + "learning_rate": 5.656934306569343e-07, + "loss": 2.9715, + "step": 45 + }, + { + "epoch": 0.05041095890410959, + "grad_norm": 3.083284616470337, + "learning_rate": 5.839416058394161e-07, + "loss": 2.9293, + "step": 46 + }, + { + "epoch": 0.05150684931506849, + "grad_norm": 2.984781503677368, + "learning_rate": 6.021897810218979e-07, + "loss": 2.9228, + "step": 47 + }, + { + "epoch": 0.0526027397260274, + "grad_norm": 2.756173849105835, + "learning_rate": 6.204379562043796e-07, + "loss": 2.9406, + "step": 48 + }, + { + "epoch": 0.0536986301369863, + "grad_norm": 2.911132574081421, + "learning_rate": 6.386861313868614e-07, + "loss": 2.8829, + "step": 49 + }, + { + "epoch": 0.0547945205479452, + "grad_norm": 2.592890977859497, + "learning_rate": 6.569343065693432e-07, + "loss": 2.9839, + "step": 50 + }, + { + "epoch": 0.0547945205479452, + "eval_loss": 2.9409468173980713, + "eval_runtime": 47.1045, + "eval_samples_per_second": 21.229, + "eval_steps_per_second": 0.892, + "step": 50 + }, + { + "epoch": 0.05589041095890411, + "grad_norm": 3.0112955570220947, + "learning_rate": 6.751824817518248e-07, + "loss": 2.8847, + "step": 51 + }, + { + "epoch": 0.05698630136986301, + "grad_norm": 2.3701798915863037, + "learning_rate": 6.934306569343066e-07, + "loss": 2.9109, + "step": 52 + }, + { + "epoch": 0.05808219178082192, + "grad_norm": 2.268850088119507, + "learning_rate": 7.116788321167883e-07, + "loss": 2.8483, + "step": 53 + }, + { + "epoch": 0.05917808219178082, + "grad_norm": 2.419015645980835, + "learning_rate": 7.299270072992701e-07, + "loss": 2.8491, + "step": 54 + }, + { + "epoch": 0.06027397260273973, + "grad_norm": 2.4582653045654297, + "learning_rate": 7.48175182481752e-07, + "loss": 2.8622, + "step": 55 + }, + { + "epoch": 0.06136986301369863, + "grad_norm": 2.3471553325653076, + "learning_rate": 7.664233576642337e-07, + "loss": 2.8738, + "step": 56 + }, + { + "epoch": 0.06246575342465754, + "grad_norm": 2.479177236557007, + "learning_rate": 7.846715328467154e-07, + "loss": 2.8843, + "step": 57 + }, + { + "epoch": 0.06356164383561644, + "grad_norm": 2.1676292419433594, + "learning_rate": 8.029197080291971e-07, + "loss": 2.8674, + "step": 58 + }, + { + "epoch": 0.06465753424657535, + "grad_norm": 1.9862910509109497, + "learning_rate": 8.211678832116789e-07, + "loss": 2.8197, + "step": 59 + }, + { + "epoch": 0.06575342465753424, + "grad_norm": 2.211313486099243, + "learning_rate": 8.394160583941606e-07, + "loss": 2.7961, + "step": 60 + }, + { + "epoch": 0.06575342465753424, + "eval_loss": 2.880824327468872, + "eval_runtime": 51.3235, + "eval_samples_per_second": 19.484, + "eval_steps_per_second": 0.818, + "step": 60 + }, + { + "epoch": 0.06684931506849315, + "grad_norm": 2.3497281074523926, + "learning_rate": 8.576642335766425e-07, + "loss": 2.8233, + "step": 61 + }, + { + "epoch": 0.06794520547945206, + "grad_norm": 2.158515214920044, + "learning_rate": 8.759124087591242e-07, + "loss": 2.8136, + "step": 62 + }, + { + "epoch": 0.06904109589041096, + "grad_norm": 2.0219955444335938, + "learning_rate": 8.94160583941606e-07, + "loss": 2.8231, + "step": 63 + }, + { + "epoch": 0.07013698630136986, + "grad_norm": 1.8989133834838867, + "learning_rate": 9.124087591240876e-07, + "loss": 2.8268, + "step": 64 + }, + { + "epoch": 0.07123287671232877, + "grad_norm": 2.243851661682129, + "learning_rate": 9.306569343065694e-07, + "loss": 2.7886, + "step": 65 + }, + { + "epoch": 0.07232876712328767, + "grad_norm": 2.122464418411255, + "learning_rate": 9.489051094890511e-07, + "loss": 2.8166, + "step": 66 + }, + { + "epoch": 0.07342465753424658, + "grad_norm": 1.899857997894287, + "learning_rate": 9.67153284671533e-07, + "loss": 2.7997, + "step": 67 + }, + { + "epoch": 0.07452054794520548, + "grad_norm": 2.4424173831939697, + "learning_rate": 9.854014598540146e-07, + "loss": 2.8342, + "step": 68 + }, + { + "epoch": 0.07561643835616438, + "grad_norm": 2.5677411556243896, + "learning_rate": 1.0036496350364965e-06, + "loss": 2.7786, + "step": 69 + }, + { + "epoch": 0.07671232876712329, + "grad_norm": 1.9989523887634277, + "learning_rate": 1.0218978102189781e-06, + "loss": 2.7411, + "step": 70 + }, + { + "epoch": 0.07671232876712329, + "eval_loss": 2.8185322284698486, + "eval_runtime": 50.4998, + "eval_samples_per_second": 19.802, + "eval_steps_per_second": 0.832, + "step": 70 + }, + { + "epoch": 0.07780821917808219, + "grad_norm": 1.9376521110534668, + "learning_rate": 1.04014598540146e-06, + "loss": 2.7988, + "step": 71 + }, + { + "epoch": 0.0789041095890411, + "grad_norm": 2.5629725456237793, + "learning_rate": 1.0583941605839416e-06, + "loss": 2.7079, + "step": 72 + }, + { + "epoch": 0.08, + "grad_norm": 2.1634678840637207, + "learning_rate": 1.0766423357664233e-06, + "loss": 2.7049, + "step": 73 + }, + { + "epoch": 0.08109589041095891, + "grad_norm": 2.2143008708953857, + "learning_rate": 1.0948905109489052e-06, + "loss": 2.72, + "step": 74 + }, + { + "epoch": 0.0821917808219178, + "grad_norm": 2.2701101303100586, + "learning_rate": 1.113138686131387e-06, + "loss": 2.7418, + "step": 75 + }, + { + "epoch": 0.08328767123287671, + "grad_norm": 1.716569423675537, + "learning_rate": 1.1313868613138687e-06, + "loss": 2.7558, + "step": 76 + }, + { + "epoch": 0.08438356164383562, + "grad_norm": 2.79757022857666, + "learning_rate": 1.1496350364963505e-06, + "loss": 2.6865, + "step": 77 + }, + { + "epoch": 0.08547945205479453, + "grad_norm": 2.2189838886260986, + "learning_rate": 1.1678832116788322e-06, + "loss": 2.6684, + "step": 78 + }, + { + "epoch": 0.08657534246575342, + "grad_norm": 2.0695579051971436, + "learning_rate": 1.1861313868613138e-06, + "loss": 2.6779, + "step": 79 + }, + { + "epoch": 0.08767123287671233, + "grad_norm": 2.372326374053955, + "learning_rate": 1.2043795620437959e-06, + "loss": 2.6711, + "step": 80 + }, + { + "epoch": 0.08767123287671233, + "eval_loss": 2.7537453174591064, + "eval_runtime": 47.7274, + "eval_samples_per_second": 20.952, + "eval_steps_per_second": 0.88, + "step": 80 + }, + { + "epoch": 0.08876712328767124, + "grad_norm": 2.3398947715759277, + "learning_rate": 1.2226277372262775e-06, + "loss": 2.6776, + "step": 81 + }, + { + "epoch": 0.08986301369863013, + "grad_norm": 2.6619036197662354, + "learning_rate": 1.2408759124087592e-06, + "loss": 2.7077, + "step": 82 + }, + { + "epoch": 0.09095890410958904, + "grad_norm": 1.8245115280151367, + "learning_rate": 1.259124087591241e-06, + "loss": 2.662, + "step": 83 + }, + { + "epoch": 0.09205479452054795, + "grad_norm": 2.2627148628234863, + "learning_rate": 1.2773722627737229e-06, + "loss": 2.677, + "step": 84 + }, + { + "epoch": 0.09315068493150686, + "grad_norm": 2.4983835220336914, + "learning_rate": 1.2956204379562045e-06, + "loss": 2.6601, + "step": 85 + }, + { + "epoch": 0.09424657534246575, + "grad_norm": 2.147010087966919, + "learning_rate": 1.3138686131386864e-06, + "loss": 2.6213, + "step": 86 + }, + { + "epoch": 0.09534246575342466, + "grad_norm": 2.791682481765747, + "learning_rate": 1.332116788321168e-06, + "loss": 2.6281, + "step": 87 + }, + { + "epoch": 0.09643835616438357, + "grad_norm": 2.023970365524292, + "learning_rate": 1.3503649635036497e-06, + "loss": 2.5766, + "step": 88 + }, + { + "epoch": 0.09753424657534246, + "grad_norm": 2.8143956661224365, + "learning_rate": 1.3686131386861315e-06, + "loss": 2.6086, + "step": 89 + }, + { + "epoch": 0.09863013698630137, + "grad_norm": 2.226106882095337, + "learning_rate": 1.3868613138686132e-06, + "loss": 2.5477, + "step": 90 + }, + { + "epoch": 0.09863013698630137, + "eval_loss": 2.6880786418914795, + "eval_runtime": 47.8417, + "eval_samples_per_second": 20.902, + "eval_steps_per_second": 0.878, + "step": 90 + }, + { + "epoch": 0.09972602739726028, + "grad_norm": 2.5248148441314697, + "learning_rate": 1.405109489051095e-06, + "loss": 2.6232, + "step": 91 + }, + { + "epoch": 0.10082191780821918, + "grad_norm": 2.421330690383911, + "learning_rate": 1.4233576642335767e-06, + "loss": 2.599, + "step": 92 + }, + { + "epoch": 0.10191780821917808, + "grad_norm": 2.5562429428100586, + "learning_rate": 1.4416058394160583e-06, + "loss": 2.6197, + "step": 93 + }, + { + "epoch": 0.10301369863013699, + "grad_norm": 2.856201171875, + "learning_rate": 1.4598540145985402e-06, + "loss": 2.5825, + "step": 94 + }, + { + "epoch": 0.10410958904109589, + "grad_norm": 2.6576807498931885, + "learning_rate": 1.478102189781022e-06, + "loss": 2.5496, + "step": 95 + }, + { + "epoch": 0.1052054794520548, + "grad_norm": 3.1408190727233887, + "learning_rate": 1.496350364963504e-06, + "loss": 2.6142, + "step": 96 + }, + { + "epoch": 0.1063013698630137, + "grad_norm": 2.3270952701568604, + "learning_rate": 1.5145985401459856e-06, + "loss": 2.584, + "step": 97 + }, + { + "epoch": 0.1073972602739726, + "grad_norm": 2.3976614475250244, + "learning_rate": 1.5328467153284674e-06, + "loss": 2.5941, + "step": 98 + }, + { + "epoch": 0.10849315068493151, + "grad_norm": 2.783485174179077, + "learning_rate": 1.551094890510949e-06, + "loss": 2.5626, + "step": 99 + }, + { + "epoch": 0.1095890410958904, + "grad_norm": 2.2522788047790527, + "learning_rate": 1.5693430656934307e-06, + "loss": 2.517, + "step": 100 + }, + { + "epoch": 0.1095890410958904, + "eval_loss": 2.637315034866333, + "eval_runtime": 48.2107, + "eval_samples_per_second": 20.742, + "eval_steps_per_second": 0.871, + "step": 100 + }, + { + "epoch": 0.11068493150684931, + "grad_norm": 2.917799949645996, + "learning_rate": 1.5875912408759126e-06, + "loss": 2.5159, + "step": 101 + }, + { + "epoch": 0.11178082191780822, + "grad_norm": 2.955815076828003, + "learning_rate": 1.6058394160583942e-06, + "loss": 2.563, + "step": 102 + }, + { + "epoch": 0.11287671232876713, + "grad_norm": 2.8096001148223877, + "learning_rate": 1.624087591240876e-06, + "loss": 2.5995, + "step": 103 + }, + { + "epoch": 0.11397260273972602, + "grad_norm": 2.4967427253723145, + "learning_rate": 1.6423357664233577e-06, + "loss": 2.5336, + "step": 104 + }, + { + "epoch": 0.11506849315068493, + "grad_norm": 2.740717887878418, + "learning_rate": 1.6605839416058394e-06, + "loss": 2.4778, + "step": 105 + }, + { + "epoch": 0.11616438356164384, + "grad_norm": 2.5064632892608643, + "learning_rate": 1.6788321167883212e-06, + "loss": 2.5214, + "step": 106 + }, + { + "epoch": 0.11726027397260275, + "grad_norm": 2.2923007011413574, + "learning_rate": 1.6970802919708029e-06, + "loss": 2.5315, + "step": 107 + }, + { + "epoch": 0.11835616438356164, + "grad_norm": 3.339569568634033, + "learning_rate": 1.715328467153285e-06, + "loss": 2.4661, + "step": 108 + }, + { + "epoch": 0.11945205479452055, + "grad_norm": 2.1650891304016113, + "learning_rate": 1.7335766423357666e-06, + "loss": 2.5072, + "step": 109 + }, + { + "epoch": 0.12054794520547946, + "grad_norm": 3.1216914653778076, + "learning_rate": 1.7518248175182485e-06, + "loss": 2.5164, + "step": 110 + }, + { + "epoch": 0.12054794520547946, + "eval_loss": 2.594250202178955, + "eval_runtime": 41.8939, + "eval_samples_per_second": 23.87, + "eval_steps_per_second": 1.003, + "step": 110 + }, + { + "epoch": 0.12164383561643835, + "grad_norm": 2.191063404083252, + "learning_rate": 1.77007299270073e-06, + "loss": 2.5422, + "step": 111 + }, + { + "epoch": 0.12273972602739726, + "grad_norm": 2.922083854675293, + "learning_rate": 1.788321167883212e-06, + "loss": 2.5258, + "step": 112 + }, + { + "epoch": 0.12383561643835617, + "grad_norm": 2.518002986907959, + "learning_rate": 1.8065693430656936e-06, + "loss": 2.5588, + "step": 113 + }, + { + "epoch": 0.12493150684931507, + "grad_norm": 2.515194892883301, + "learning_rate": 1.8248175182481753e-06, + "loss": 2.5079, + "step": 114 + }, + { + "epoch": 0.12602739726027398, + "grad_norm": 2.5530874729156494, + "learning_rate": 1.8430656934306571e-06, + "loss": 2.4615, + "step": 115 + }, + { + "epoch": 0.12712328767123288, + "grad_norm": 2.3550198078155518, + "learning_rate": 1.8613138686131388e-06, + "loss": 2.4825, + "step": 116 + }, + { + "epoch": 0.12821917808219177, + "grad_norm": 2.295145034790039, + "learning_rate": 1.8795620437956206e-06, + "loss": 2.4749, + "step": 117 + }, + { + "epoch": 0.1293150684931507, + "grad_norm": 2.0158982276916504, + "learning_rate": 1.8978102189781023e-06, + "loss": 2.4756, + "step": 118 + }, + { + "epoch": 0.1304109589041096, + "grad_norm": 2.213057518005371, + "learning_rate": 1.916058394160584e-06, + "loss": 2.4744, + "step": 119 + }, + { + "epoch": 0.13150684931506848, + "grad_norm": 2.186222553253174, + "learning_rate": 1.934306569343066e-06, + "loss": 2.4957, + "step": 120 + }, + { + "epoch": 0.13150684931506848, + "eval_loss": 2.563931465148926, + "eval_runtime": 44.8714, + "eval_samples_per_second": 22.286, + "eval_steps_per_second": 0.936, + "step": 120 + }, + { + "epoch": 0.1326027397260274, + "grad_norm": 2.3143310546875, + "learning_rate": 1.952554744525548e-06, + "loss": 2.4, + "step": 121 + }, + { + "epoch": 0.1336986301369863, + "grad_norm": 2.0223965644836426, + "learning_rate": 1.9708029197080293e-06, + "loss": 2.4453, + "step": 122 + }, + { + "epoch": 0.13479452054794522, + "grad_norm": 2.2025411128997803, + "learning_rate": 1.989051094890511e-06, + "loss": 2.4463, + "step": 123 + }, + { + "epoch": 0.1358904109589041, + "grad_norm": 2.707930564880371, + "learning_rate": 2.007299270072993e-06, + "loss": 2.4589, + "step": 124 + }, + { + "epoch": 0.136986301369863, + "grad_norm": 2.146662473678589, + "learning_rate": 2.025547445255475e-06, + "loss": 2.5023, + "step": 125 + }, + { + "epoch": 0.13808219178082193, + "grad_norm": 2.520660161972046, + "learning_rate": 2.0437956204379563e-06, + "loss": 2.4464, + "step": 126 + }, + { + "epoch": 0.13917808219178082, + "grad_norm": 2.50632381439209, + "learning_rate": 2.062043795620438e-06, + "loss": 2.4556, + "step": 127 + }, + { + "epoch": 0.14027397260273972, + "grad_norm": 2.2919397354125977, + "learning_rate": 2.08029197080292e-06, + "loss": 2.4479, + "step": 128 + }, + { + "epoch": 0.14136986301369864, + "grad_norm": 2.521475315093994, + "learning_rate": 2.0985401459854014e-06, + "loss": 2.4982, + "step": 129 + }, + { + "epoch": 0.14246575342465753, + "grad_norm": 2.3812668323516846, + "learning_rate": 2.1167883211678833e-06, + "loss": 2.4163, + "step": 130 + }, + { + "epoch": 0.14246575342465753, + "eval_loss": 2.5390918254852295, + "eval_runtime": 43.8746, + "eval_samples_per_second": 22.792, + "eval_steps_per_second": 0.957, + "step": 130 + }, + { + "epoch": 0.14356164383561643, + "grad_norm": 2.302286148071289, + "learning_rate": 2.135036496350365e-06, + "loss": 2.4588, + "step": 131 + }, + { + "epoch": 0.14465753424657535, + "grad_norm": 2.4222044944763184, + "learning_rate": 2.1532846715328466e-06, + "loss": 2.492, + "step": 132 + }, + { + "epoch": 0.14575342465753424, + "grad_norm": 2.605835199356079, + "learning_rate": 2.171532846715329e-06, + "loss": 2.4055, + "step": 133 + }, + { + "epoch": 0.14684931506849316, + "grad_norm": 3.6261861324310303, + "learning_rate": 2.1897810218978103e-06, + "loss": 2.4221, + "step": 134 + }, + { + "epoch": 0.14794520547945206, + "grad_norm": 2.7647221088409424, + "learning_rate": 2.208029197080292e-06, + "loss": 2.4549, + "step": 135 + }, + { + "epoch": 0.14904109589041095, + "grad_norm": 3.2977371215820312, + "learning_rate": 2.226277372262774e-06, + "loss": 2.4331, + "step": 136 + }, + { + "epoch": 0.15013698630136987, + "grad_norm": 3.147094488143921, + "learning_rate": 2.244525547445256e-06, + "loss": 2.4268, + "step": 137 + }, + { + "epoch": 0.15123287671232877, + "grad_norm": 2.6363413333892822, + "learning_rate": 2.2627737226277373e-06, + "loss": 2.4524, + "step": 138 + }, + { + "epoch": 0.15232876712328766, + "grad_norm": 2.9288904666900635, + "learning_rate": 2.281021897810219e-06, + "loss": 2.4534, + "step": 139 + }, + { + "epoch": 0.15342465753424658, + "grad_norm": 2.5604870319366455, + "learning_rate": 2.299270072992701e-06, + "loss": 2.4138, + "step": 140 + }, + { + "epoch": 0.15342465753424658, + "eval_loss": 2.524790048599243, + "eval_runtime": 43.8851, + "eval_samples_per_second": 22.787, + "eval_steps_per_second": 0.957, + "step": 140 + }, + { + "epoch": 0.15452054794520548, + "grad_norm": 3.2746663093566895, + "learning_rate": 2.3175182481751825e-06, + "loss": 2.4094, + "step": 141 + }, + { + "epoch": 0.15561643835616437, + "grad_norm": 2.691483497619629, + "learning_rate": 2.3357664233576643e-06, + "loss": 2.4265, + "step": 142 + }, + { + "epoch": 0.1567123287671233, + "grad_norm": 2.5850117206573486, + "learning_rate": 2.354014598540146e-06, + "loss": 2.3922, + "step": 143 + }, + { + "epoch": 0.1578082191780822, + "grad_norm": 2.68204927444458, + "learning_rate": 2.3722627737226276e-06, + "loss": 2.4262, + "step": 144 + }, + { + "epoch": 0.1589041095890411, + "grad_norm": 2.952584981918335, + "learning_rate": 2.39051094890511e-06, + "loss": 2.4267, + "step": 145 + }, + { + "epoch": 0.16, + "grad_norm": 3.2354788780212402, + "learning_rate": 2.4087591240875918e-06, + "loss": 2.3836, + "step": 146 + }, + { + "epoch": 0.1610958904109589, + "grad_norm": 2.634521007537842, + "learning_rate": 2.427007299270073e-06, + "loss": 2.3953, + "step": 147 + }, + { + "epoch": 0.16219178082191782, + "grad_norm": 3.1115682125091553, + "learning_rate": 2.445255474452555e-06, + "loss": 2.409, + "step": 148 + }, + { + "epoch": 0.1632876712328767, + "grad_norm": 2.6824135780334473, + "learning_rate": 2.463503649635037e-06, + "loss": 2.4459, + "step": 149 + }, + { + "epoch": 0.1643835616438356, + "grad_norm": 2.617614507675171, + "learning_rate": 2.4817518248175183e-06, + "loss": 2.366, + "step": 150 + }, + { + "epoch": 0.1643835616438356, + "eval_loss": 2.5088160037994385, + "eval_runtime": 44.0888, + "eval_samples_per_second": 22.682, + "eval_steps_per_second": 0.953, + "step": 150 + }, + { + "epoch": 0.16547945205479453, + "grad_norm": 3.976992130279541, + "learning_rate": 2.5e-06, + "loss": 2.3517, + "step": 151 + }, + { + "epoch": 0.16657534246575342, + "grad_norm": 3.1433017253875732, + "learning_rate": 2.518248175182482e-06, + "loss": 2.4489, + "step": 152 + }, + { + "epoch": 0.16767123287671232, + "grad_norm": 4.100688934326172, + "learning_rate": 2.5364963503649635e-06, + "loss": 2.3683, + "step": 153 + }, + { + "epoch": 0.16876712328767124, + "grad_norm": 4.316102504730225, + "learning_rate": 2.5547445255474458e-06, + "loss": 2.3663, + "step": 154 + }, + { + "epoch": 0.16986301369863013, + "grad_norm": 2.3103132247924805, + "learning_rate": 2.5729927007299272e-06, + "loss": 2.4009, + "step": 155 + }, + { + "epoch": 0.17095890410958905, + "grad_norm": 3.576810836791992, + "learning_rate": 2.591240875912409e-06, + "loss": 2.3983, + "step": 156 + }, + { + "epoch": 0.17205479452054795, + "grad_norm": 2.337080955505371, + "learning_rate": 2.6094890510948905e-06, + "loss": 2.3849, + "step": 157 + }, + { + "epoch": 0.17315068493150684, + "grad_norm": 3.2523117065429688, + "learning_rate": 2.627737226277373e-06, + "loss": 2.4171, + "step": 158 + }, + { + "epoch": 0.17424657534246576, + "grad_norm": 2.529175281524658, + "learning_rate": 2.6459854014598542e-06, + "loss": 2.4303, + "step": 159 + }, + { + "epoch": 0.17534246575342466, + "grad_norm": 3.052212953567505, + "learning_rate": 2.664233576642336e-06, + "loss": 2.3895, + "step": 160 + }, + { + "epoch": 0.17534246575342466, + "eval_loss": 2.492485761642456, + "eval_runtime": 45.6738, + "eval_samples_per_second": 21.894, + "eval_steps_per_second": 0.92, + "step": 160 + }, + { + "epoch": 0.17643835616438355, + "grad_norm": 2.1544973850250244, + "learning_rate": 2.6824817518248175e-06, + "loss": 2.3649, + "step": 161 + }, + { + "epoch": 0.17753424657534247, + "grad_norm": 2.7351951599121094, + "learning_rate": 2.7007299270072994e-06, + "loss": 2.3908, + "step": 162 + }, + { + "epoch": 0.17863013698630137, + "grad_norm": 2.8431951999664307, + "learning_rate": 2.718978102189781e-06, + "loss": 2.374, + "step": 163 + }, + { + "epoch": 0.17972602739726026, + "grad_norm": 3.083190441131592, + "learning_rate": 2.737226277372263e-06, + "loss": 2.4193, + "step": 164 + }, + { + "epoch": 0.18082191780821918, + "grad_norm": 2.6523995399475098, + "learning_rate": 2.755474452554745e-06, + "loss": 2.3969, + "step": 165 + }, + { + "epoch": 0.18191780821917808, + "grad_norm": 2.626047372817993, + "learning_rate": 2.7737226277372264e-06, + "loss": 2.3742, + "step": 166 + }, + { + "epoch": 0.18301369863013697, + "grad_norm": 3.300917863845825, + "learning_rate": 2.7919708029197082e-06, + "loss": 2.4546, + "step": 167 + }, + { + "epoch": 0.1841095890410959, + "grad_norm": 2.797304391860962, + "learning_rate": 2.81021897810219e-06, + "loss": 2.4161, + "step": 168 + }, + { + "epoch": 0.1852054794520548, + "grad_norm": 2.777117967605591, + "learning_rate": 2.828467153284672e-06, + "loss": 2.3572, + "step": 169 + }, + { + "epoch": 0.1863013698630137, + "grad_norm": 2.887470006942749, + "learning_rate": 2.8467153284671534e-06, + "loss": 2.3887, + "step": 170 + }, + { + "epoch": 0.1863013698630137, + "eval_loss": 2.4854257106781006, + "eval_runtime": 44.5644, + "eval_samples_per_second": 22.439, + "eval_steps_per_second": 0.942, + "step": 170 + }, + { + "epoch": 0.1873972602739726, + "grad_norm": 3.0706076622009277, + "learning_rate": 2.8649635036496353e-06, + "loss": 2.3162, + "step": 171 + }, + { + "epoch": 0.1884931506849315, + "grad_norm": 3.6129744052886963, + "learning_rate": 2.8832116788321167e-06, + "loss": 2.3685, + "step": 172 + }, + { + "epoch": 0.18958904109589042, + "grad_norm": 2.4936370849609375, + "learning_rate": 2.901459854014599e-06, + "loss": 2.3297, + "step": 173 + }, + { + "epoch": 0.19068493150684931, + "grad_norm": 2.1041860580444336, + "learning_rate": 2.9197080291970804e-06, + "loss": 2.4085, + "step": 174 + }, + { + "epoch": 0.1917808219178082, + "grad_norm": 2.2066831588745117, + "learning_rate": 2.9379562043795623e-06, + "loss": 2.3905, + "step": 175 + }, + { + "epoch": 0.19287671232876713, + "grad_norm": 2.444652795791626, + "learning_rate": 2.956204379562044e-06, + "loss": 2.378, + "step": 176 + }, + { + "epoch": 0.19397260273972602, + "grad_norm": 2.501939058303833, + "learning_rate": 2.9744525547445256e-06, + "loss": 2.4136, + "step": 177 + }, + { + "epoch": 0.19506849315068492, + "grad_norm": 2.2167556285858154, + "learning_rate": 2.992700729927008e-06, + "loss": 2.385, + "step": 178 + }, + { + "epoch": 0.19616438356164384, + "grad_norm": 2.5298545360565186, + "learning_rate": 3.0109489051094893e-06, + "loss": 2.3508, + "step": 179 + }, + { + "epoch": 0.19726027397260273, + "grad_norm": 2.3774642944335938, + "learning_rate": 3.029197080291971e-06, + "loss": 2.3421, + "step": 180 + }, + { + "epoch": 0.19726027397260273, + "eval_loss": 2.476067304611206, + "eval_runtime": 44.3761, + "eval_samples_per_second": 22.535, + "eval_steps_per_second": 0.946, + "step": 180 + }, + { + "epoch": 0.19835616438356166, + "grad_norm": 2.9108712673187256, + "learning_rate": 3.0474452554744526e-06, + "loss": 2.3983, + "step": 181 + }, + { + "epoch": 0.19945205479452055, + "grad_norm": 2.5232033729553223, + "learning_rate": 3.065693430656935e-06, + "loss": 2.3946, + "step": 182 + }, + { + "epoch": 0.20054794520547944, + "grad_norm": 2.158168077468872, + "learning_rate": 3.0839416058394163e-06, + "loss": 2.3805, + "step": 183 + }, + { + "epoch": 0.20164383561643837, + "grad_norm": 2.3405354022979736, + "learning_rate": 3.102189781021898e-06, + "loss": 2.3974, + "step": 184 + }, + { + "epoch": 0.20273972602739726, + "grad_norm": 2.5308327674865723, + "learning_rate": 3.1204379562043796e-06, + "loss": 2.384, + "step": 185 + }, + { + "epoch": 0.20383561643835615, + "grad_norm": 2.6328237056732178, + "learning_rate": 3.1386861313868614e-06, + "loss": 2.3345, + "step": 186 + }, + { + "epoch": 0.20493150684931508, + "grad_norm": 3.0725512504577637, + "learning_rate": 3.156934306569343e-06, + "loss": 2.3191, + "step": 187 + }, + { + "epoch": 0.20602739726027397, + "grad_norm": 2.196218729019165, + "learning_rate": 3.175182481751825e-06, + "loss": 2.4073, + "step": 188 + }, + { + "epoch": 0.20712328767123286, + "grad_norm": 2.7722432613372803, + "learning_rate": 3.193430656934307e-06, + "loss": 2.3775, + "step": 189 + }, + { + "epoch": 0.20821917808219179, + "grad_norm": 2.4848241806030273, + "learning_rate": 3.2116788321167884e-06, + "loss": 2.3538, + "step": 190 + }, + { + "epoch": 0.20821917808219179, + "eval_loss": 2.462618112564087, + "eval_runtime": 43.9461, + "eval_samples_per_second": 22.755, + "eval_steps_per_second": 0.956, + "step": 190 + }, + { + "epoch": 0.20931506849315068, + "grad_norm": 2.4682741165161133, + "learning_rate": 3.2299270072992707e-06, + "loss": 2.3688, + "step": 191 + }, + { + "epoch": 0.2104109589041096, + "grad_norm": 2.518923282623291, + "learning_rate": 3.248175182481752e-06, + "loss": 2.3556, + "step": 192 + }, + { + "epoch": 0.2115068493150685, + "grad_norm": 2.4023561477661133, + "learning_rate": 3.266423357664234e-06, + "loss": 2.292, + "step": 193 + }, + { + "epoch": 0.2126027397260274, + "grad_norm": 2.5196943283081055, + "learning_rate": 3.2846715328467155e-06, + "loss": 2.3703, + "step": 194 + }, + { + "epoch": 0.2136986301369863, + "grad_norm": 2.762709617614746, + "learning_rate": 3.3029197080291973e-06, + "loss": 2.3578, + "step": 195 + }, + { + "epoch": 0.2147945205479452, + "grad_norm": 2.1287691593170166, + "learning_rate": 3.3211678832116788e-06, + "loss": 2.3687, + "step": 196 + }, + { + "epoch": 0.2158904109589041, + "grad_norm": 2.477341651916504, + "learning_rate": 3.339416058394161e-06, + "loss": 2.3545, + "step": 197 + }, + { + "epoch": 0.21698630136986302, + "grad_norm": 2.23933744430542, + "learning_rate": 3.3576642335766425e-06, + "loss": 2.3511, + "step": 198 + }, + { + "epoch": 0.21808219178082192, + "grad_norm": 2.2440133094787598, + "learning_rate": 3.3759124087591243e-06, + "loss": 2.3539, + "step": 199 + }, + { + "epoch": 0.2191780821917808, + "grad_norm": 2.5603177547454834, + "learning_rate": 3.3941605839416058e-06, + "loss": 2.36, + "step": 200 + }, + { + "epoch": 0.2191780821917808, + "eval_loss": 2.4507246017456055, + "eval_runtime": 43.7786, + "eval_samples_per_second": 22.842, + "eval_steps_per_second": 0.959, + "step": 200 + }, + { + "epoch": 0.22027397260273973, + "grad_norm": 2.4570510387420654, + "learning_rate": 3.412408759124088e-06, + "loss": 2.3702, + "step": 201 + }, + { + "epoch": 0.22136986301369863, + "grad_norm": 2.1975133419036865, + "learning_rate": 3.43065693430657e-06, + "loss": 2.3667, + "step": 202 + }, + { + "epoch": 0.22246575342465755, + "grad_norm": 2.710186004638672, + "learning_rate": 3.4489051094890513e-06, + "loss": 2.3159, + "step": 203 + }, + { + "epoch": 0.22356164383561644, + "grad_norm": 2.7799506187438965, + "learning_rate": 3.467153284671533e-06, + "loss": 2.3519, + "step": 204 + }, + { + "epoch": 0.22465753424657534, + "grad_norm": 2.2479400634765625, + "learning_rate": 3.4854014598540146e-06, + "loss": 2.3103, + "step": 205 + }, + { + "epoch": 0.22575342465753426, + "grad_norm": 3.1538405418395996, + "learning_rate": 3.503649635036497e-06, + "loss": 2.3506, + "step": 206 + }, + { + "epoch": 0.22684931506849315, + "grad_norm": 2.7607762813568115, + "learning_rate": 3.5218978102189783e-06, + "loss": 2.3386, + "step": 207 + }, + { + "epoch": 0.22794520547945205, + "grad_norm": 2.371105909347534, + "learning_rate": 3.54014598540146e-06, + "loss": 2.3583, + "step": 208 + }, + { + "epoch": 0.22904109589041097, + "grad_norm": 2.793933391571045, + "learning_rate": 3.5583941605839416e-06, + "loss": 2.3782, + "step": 209 + }, + { + "epoch": 0.23013698630136986, + "grad_norm": 2.372732400894165, + "learning_rate": 3.576642335766424e-06, + "loss": 2.417, + "step": 210 + }, + { + "epoch": 0.23013698630136986, + "eval_loss": 2.451474189758301, + "eval_runtime": 41.7197, + "eval_samples_per_second": 23.97, + "eval_steps_per_second": 1.007, + "step": 210 + }, + { + "epoch": 0.23123287671232876, + "grad_norm": 2.6989448070526123, + "learning_rate": 3.5948905109489054e-06, + "loss": 2.393, + "step": 211 + }, + { + "epoch": 0.23232876712328768, + "grad_norm": 2.4324285984039307, + "learning_rate": 3.6131386861313872e-06, + "loss": 2.2738, + "step": 212 + }, + { + "epoch": 0.23342465753424657, + "grad_norm": 3.5063016414642334, + "learning_rate": 3.6313868613138687e-06, + "loss": 2.2823, + "step": 213 + }, + { + "epoch": 0.2345205479452055, + "grad_norm": 3.953951358795166, + "learning_rate": 3.6496350364963505e-06, + "loss": 2.3132, + "step": 214 + }, + { + "epoch": 0.2356164383561644, + "grad_norm": 2.3483071327209473, + "learning_rate": 3.667883211678833e-06, + "loss": 2.3244, + "step": 215 + }, + { + "epoch": 0.23671232876712328, + "grad_norm": 2.884199380874634, + "learning_rate": 3.6861313868613142e-06, + "loss": 2.3067, + "step": 216 + }, + { + "epoch": 0.2378082191780822, + "grad_norm": 2.9514520168304443, + "learning_rate": 3.704379562043796e-06, + "loss": 2.304, + "step": 217 + }, + { + "epoch": 0.2389041095890411, + "grad_norm": 2.7530996799468994, + "learning_rate": 3.7226277372262775e-06, + "loss": 2.3691, + "step": 218 + }, + { + "epoch": 0.24, + "grad_norm": 2.8753647804260254, + "learning_rate": 3.7408759124087594e-06, + "loss": 2.3406, + "step": 219 + }, + { + "epoch": 0.2410958904109589, + "grad_norm": 2.670574903488159, + "learning_rate": 3.7591240875912412e-06, + "loss": 2.304, + "step": 220 + }, + { + "epoch": 0.2410958904109589, + "eval_loss": 2.4395809173583984, + "eval_runtime": 46.6723, + "eval_samples_per_second": 21.426, + "eval_steps_per_second": 0.9, + "step": 220 + }, + { + "epoch": 0.2421917808219178, + "grad_norm": 2.4152438640594482, + "learning_rate": 3.777372262773723e-06, + "loss": 2.3294, + "step": 221 + }, + { + "epoch": 0.2432876712328767, + "grad_norm": 2.8398540019989014, + "learning_rate": 3.7956204379562045e-06, + "loss": 2.3216, + "step": 222 + }, + { + "epoch": 0.24438356164383562, + "grad_norm": 2.6036765575408936, + "learning_rate": 3.8138686131386864e-06, + "loss": 2.3109, + "step": 223 + }, + { + "epoch": 0.24547945205479452, + "grad_norm": 2.7073473930358887, + "learning_rate": 3.832116788321168e-06, + "loss": 2.3457, + "step": 224 + }, + { + "epoch": 0.2465753424657534, + "grad_norm": 2.584404230117798, + "learning_rate": 3.85036496350365e-06, + "loss": 2.3159, + "step": 225 + }, + { + "epoch": 0.24767123287671233, + "grad_norm": 2.4220666885375977, + "learning_rate": 3.868613138686132e-06, + "loss": 2.2944, + "step": 226 + }, + { + "epoch": 0.24876712328767123, + "grad_norm": 2.5636494159698486, + "learning_rate": 3.886861313868613e-06, + "loss": 2.3219, + "step": 227 + }, + { + "epoch": 0.24986301369863015, + "grad_norm": 2.2990903854370117, + "learning_rate": 3.905109489051096e-06, + "loss": 2.3342, + "step": 228 + }, + { + "epoch": 0.25095890410958904, + "grad_norm": 2.50913143157959, + "learning_rate": 3.923357664233577e-06, + "loss": 2.3619, + "step": 229 + }, + { + "epoch": 0.25205479452054796, + "grad_norm": 2.1687769889831543, + "learning_rate": 3.9416058394160585e-06, + "loss": 2.3153, + "step": 230 + }, + { + "epoch": 0.25205479452054796, + "eval_loss": 2.4315240383148193, + "eval_runtime": 48.3542, + "eval_samples_per_second": 20.681, + "eval_steps_per_second": 0.869, + "step": 230 + }, + { + "epoch": 0.25315068493150683, + "grad_norm": 2.842125177383423, + "learning_rate": 3.95985401459854e-06, + "loss": 2.3644, + "step": 231 + }, + { + "epoch": 0.25424657534246575, + "grad_norm": 2.258049964904785, + "learning_rate": 3.978102189781022e-06, + "loss": 2.3727, + "step": 232 + }, + { + "epoch": 0.2553424657534247, + "grad_norm": 2.4630489349365234, + "learning_rate": 3.996350364963504e-06, + "loss": 2.267, + "step": 233 + }, + { + "epoch": 0.25643835616438354, + "grad_norm": 3.1469743251800537, + "learning_rate": 4.014598540145986e-06, + "loss": 2.3432, + "step": 234 + }, + { + "epoch": 0.25753424657534246, + "grad_norm": 2.6277477741241455, + "learning_rate": 4.032846715328467e-06, + "loss": 2.2545, + "step": 235 + }, + { + "epoch": 0.2586301369863014, + "grad_norm": 2.404141902923584, + "learning_rate": 4.05109489051095e-06, + "loss": 2.3109, + "step": 236 + }, + { + "epoch": 0.25972602739726025, + "grad_norm": 2.4545605182647705, + "learning_rate": 4.069343065693431e-06, + "loss": 2.3112, + "step": 237 + }, + { + "epoch": 0.2608219178082192, + "grad_norm": 2.237638235092163, + "learning_rate": 4.0875912408759126e-06, + "loss": 2.3, + "step": 238 + }, + { + "epoch": 0.2619178082191781, + "grad_norm": 2.192811965942383, + "learning_rate": 4.105839416058395e-06, + "loss": 2.3108, + "step": 239 + }, + { + "epoch": 0.26301369863013696, + "grad_norm": 2.5058720111846924, + "learning_rate": 4.124087591240876e-06, + "loss": 2.3635, + "step": 240 + }, + { + "epoch": 0.26301369863013696, + "eval_loss": 2.4219377040863037, + "eval_runtime": 45.3561, + "eval_samples_per_second": 22.048, + "eval_steps_per_second": 0.926, + "step": 240 + }, + { + "epoch": 0.2641095890410959, + "grad_norm": 2.0708935260772705, + "learning_rate": 4.1423357664233586e-06, + "loss": 2.3129, + "step": 241 + }, + { + "epoch": 0.2652054794520548, + "grad_norm": 2.6353116035461426, + "learning_rate": 4.16058394160584e-06, + "loss": 2.3011, + "step": 242 + }, + { + "epoch": 0.2663013698630137, + "grad_norm": 2.8392436504364014, + "learning_rate": 4.1788321167883214e-06, + "loss": 2.3643, + "step": 243 + }, + { + "epoch": 0.2673972602739726, + "grad_norm": 3.1773295402526855, + "learning_rate": 4.197080291970803e-06, + "loss": 2.2642, + "step": 244 + }, + { + "epoch": 0.2684931506849315, + "grad_norm": 2.4757354259490967, + "learning_rate": 4.215328467153285e-06, + "loss": 2.3211, + "step": 245 + }, + { + "epoch": 0.26958904109589044, + "grad_norm": 2.5799412727355957, + "learning_rate": 4.233576642335767e-06, + "loss": 2.2909, + "step": 246 + }, + { + "epoch": 0.2706849315068493, + "grad_norm": 2.368502378463745, + "learning_rate": 4.251824817518249e-06, + "loss": 2.2951, + "step": 247 + }, + { + "epoch": 0.2717808219178082, + "grad_norm": 2.114969253540039, + "learning_rate": 4.27007299270073e-06, + "loss": 2.3124, + "step": 248 + }, + { + "epoch": 0.27287671232876715, + "grad_norm": 3.056820869445801, + "learning_rate": 4.288321167883212e-06, + "loss": 2.2923, + "step": 249 + }, + { + "epoch": 0.273972602739726, + "grad_norm": 2.7752203941345215, + "learning_rate": 4.306569343065693e-06, + "loss": 2.2874, + "step": 250 + }, + { + "epoch": 0.273972602739726, + "eval_loss": 2.4229912757873535, + "eval_runtime": 52.5941, + "eval_samples_per_second": 19.014, + "eval_steps_per_second": 0.799, + "step": 250 + }, + { + "epoch": 0.27506849315068493, + "grad_norm": 3.045740842819214, + "learning_rate": 4.3248175182481755e-06, + "loss": 2.2686, + "step": 251 + }, + { + "epoch": 0.27616438356164386, + "grad_norm": 2.646402359008789, + "learning_rate": 4.343065693430658e-06, + "loss": 2.3247, + "step": 252 + }, + { + "epoch": 0.2772602739726027, + "grad_norm": 3.274994134902954, + "learning_rate": 4.361313868613139e-06, + "loss": 2.3093, + "step": 253 + }, + { + "epoch": 0.27835616438356164, + "grad_norm": 2.352246046066284, + "learning_rate": 4.379562043795621e-06, + "loss": 2.3124, + "step": 254 + }, + { + "epoch": 0.27945205479452057, + "grad_norm": 3.248717784881592, + "learning_rate": 4.397810218978102e-06, + "loss": 2.3417, + "step": 255 + }, + { + "epoch": 0.28054794520547943, + "grad_norm": 3.0919013023376465, + "learning_rate": 4.416058394160584e-06, + "loss": 2.3093, + "step": 256 + }, + { + "epoch": 0.28164383561643835, + "grad_norm": 2.1174702644348145, + "learning_rate": 4.434306569343066e-06, + "loss": 2.305, + "step": 257 + }, + { + "epoch": 0.2827397260273973, + "grad_norm": 2.913797378540039, + "learning_rate": 4.452554744525548e-06, + "loss": 2.2831, + "step": 258 + }, + { + "epoch": 0.28383561643835614, + "grad_norm": 2.693089723587036, + "learning_rate": 4.4708029197080295e-06, + "loss": 2.3206, + "step": 259 + }, + { + "epoch": 0.28493150684931506, + "grad_norm": 2.521411895751953, + "learning_rate": 4.489051094890512e-06, + "loss": 2.3196, + "step": 260 + }, + { + "epoch": 0.28493150684931506, + "eval_loss": 2.4178109169006348, + "eval_runtime": 50.6753, + "eval_samples_per_second": 19.733, + "eval_steps_per_second": 0.829, + "step": 260 + }, + { + "epoch": 0.286027397260274, + "grad_norm": 2.346346855163574, + "learning_rate": 4.507299270072993e-06, + "loss": 2.3433, + "step": 261 + }, + { + "epoch": 0.28712328767123285, + "grad_norm": 2.520368814468384, + "learning_rate": 4.525547445255475e-06, + "loss": 2.3174, + "step": 262 + }, + { + "epoch": 0.2882191780821918, + "grad_norm": 2.550347328186035, + "learning_rate": 4.543795620437956e-06, + "loss": 2.3072, + "step": 263 + }, + { + "epoch": 0.2893150684931507, + "grad_norm": 2.268662691116333, + "learning_rate": 4.562043795620438e-06, + "loss": 2.2763, + "step": 264 + }, + { + "epoch": 0.29041095890410956, + "grad_norm": 2.445798873901367, + "learning_rate": 4.580291970802921e-06, + "loss": 2.3395, + "step": 265 + }, + { + "epoch": 0.2915068493150685, + "grad_norm": 2.5391135215759277, + "learning_rate": 4.598540145985402e-06, + "loss": 2.3496, + "step": 266 + }, + { + "epoch": 0.2926027397260274, + "grad_norm": 2.3553097248077393, + "learning_rate": 4.6167883211678835e-06, + "loss": 2.3243, + "step": 267 + }, + { + "epoch": 0.2936986301369863, + "grad_norm": 2.448089599609375, + "learning_rate": 4.635036496350365e-06, + "loss": 2.2887, + "step": 268 + }, + { + "epoch": 0.2947945205479452, + "grad_norm": 2.1600170135498047, + "learning_rate": 4.653284671532847e-06, + "loss": 2.2816, + "step": 269 + }, + { + "epoch": 0.2958904109589041, + "grad_norm": 2.3783180713653564, + "learning_rate": 4.671532846715329e-06, + "loss": 2.3275, + "step": 270 + }, + { + "epoch": 0.2958904109589041, + "eval_loss": 2.413832902908325, + "eval_runtime": 48.7524, + "eval_samples_per_second": 20.512, + "eval_steps_per_second": 0.861, + "step": 270 + }, + { + "epoch": 0.29698630136986304, + "grad_norm": 2.468320369720459, + "learning_rate": 4.689781021897811e-06, + "loss": 2.3358, + "step": 271 + }, + { + "epoch": 0.2980821917808219, + "grad_norm": 2.401634454727173, + "learning_rate": 4.708029197080292e-06, + "loss": 2.2991, + "step": 272 + }, + { + "epoch": 0.2991780821917808, + "grad_norm": 2.4604203701019287, + "learning_rate": 4.726277372262774e-06, + "loss": 2.3496, + "step": 273 + }, + { + "epoch": 0.30027397260273975, + "grad_norm": 2.0338871479034424, + "learning_rate": 4.744525547445255e-06, + "loss": 2.3115, + "step": 274 + }, + { + "epoch": 0.3013698630136986, + "grad_norm": 2.5404436588287354, + "learning_rate": 4.7627737226277375e-06, + "loss": 2.2665, + "step": 275 + }, + { + "epoch": 0.30246575342465754, + "grad_norm": 2.5369458198547363, + "learning_rate": 4.78102189781022e-06, + "loss": 2.3072, + "step": 276 + }, + { + "epoch": 0.30356164383561646, + "grad_norm": 2.471522569656372, + "learning_rate": 4.799270072992701e-06, + "loss": 2.3113, + "step": 277 + }, + { + "epoch": 0.3046575342465753, + "grad_norm": 2.3890135288238525, + "learning_rate": 4.8175182481751835e-06, + "loss": 2.2838, + "step": 278 + }, + { + "epoch": 0.30575342465753425, + "grad_norm": 2.2485053539276123, + "learning_rate": 4.835766423357665e-06, + "loss": 2.3052, + "step": 279 + }, + { + "epoch": 0.30684931506849317, + "grad_norm": 2.1090753078460693, + "learning_rate": 4.854014598540146e-06, + "loss": 2.3054, + "step": 280 + }, + { + "epoch": 0.30684931506849317, + "eval_loss": 2.412548780441284, + "eval_runtime": 48.5034, + "eval_samples_per_second": 20.617, + "eval_steps_per_second": 0.866, + "step": 280 + }, + { + "epoch": 0.30794520547945203, + "grad_norm": 2.299133777618408, + "learning_rate": 4.872262773722628e-06, + "loss": 2.2599, + "step": 281 + }, + { + "epoch": 0.30904109589041096, + "grad_norm": 2.177905797958374, + "learning_rate": 4.89051094890511e-06, + "loss": 2.2745, + "step": 282 + }, + { + "epoch": 0.3101369863013699, + "grad_norm": 2.3178462982177734, + "learning_rate": 4.9087591240875915e-06, + "loss": 2.2562, + "step": 283 + }, + { + "epoch": 0.31123287671232874, + "grad_norm": 2.5088376998901367, + "learning_rate": 4.927007299270074e-06, + "loss": 2.2945, + "step": 284 + }, + { + "epoch": 0.31232876712328766, + "grad_norm": 2.095379590988159, + "learning_rate": 4.945255474452555e-06, + "loss": 2.3038, + "step": 285 + }, + { + "epoch": 0.3134246575342466, + "grad_norm": 2.1792995929718018, + "learning_rate": 4.963503649635037e-06, + "loss": 2.2702, + "step": 286 + }, + { + "epoch": 0.31452054794520545, + "grad_norm": 2.3001186847686768, + "learning_rate": 4.981751824817518e-06, + "loss": 2.3482, + "step": 287 + }, + { + "epoch": 0.3156164383561644, + "grad_norm": 2.1576106548309326, + "learning_rate": 5e-06, + "loss": 2.3391, + "step": 288 + }, + { + "epoch": 0.3167123287671233, + "grad_norm": 2.3715577125549316, + "learning_rate": 4.997969130787978e-06, + "loss": 2.3049, + "step": 289 + }, + { + "epoch": 0.3178082191780822, + "grad_norm": 2.0316715240478516, + "learning_rate": 4.995938261575955e-06, + "loss": 2.3111, + "step": 290 + }, + { + "epoch": 0.3178082191780822, + "eval_loss": 2.40299654006958, + "eval_runtime": 50.0873, + "eval_samples_per_second": 19.965, + "eval_steps_per_second": 0.839, + "step": 290 + }, + { + "epoch": 0.3189041095890411, + "grad_norm": 2.1135756969451904, + "learning_rate": 4.993907392363932e-06, + "loss": 2.2465, + "step": 291 + }, + { + "epoch": 0.32, + "grad_norm": 2.335655927658081, + "learning_rate": 4.9918765231519095e-06, + "loss": 2.31, + "step": 292 + }, + { + "epoch": 0.32109589041095893, + "grad_norm": 2.3255794048309326, + "learning_rate": 4.989845653939887e-06, + "loss": 2.2531, + "step": 293 + }, + { + "epoch": 0.3221917808219178, + "grad_norm": 2.111055612564087, + "learning_rate": 4.987814784727864e-06, + "loss": 2.2799, + "step": 294 + }, + { + "epoch": 0.3232876712328767, + "grad_norm": 2.1773171424865723, + "learning_rate": 4.985783915515841e-06, + "loss": 2.2884, + "step": 295 + }, + { + "epoch": 0.32438356164383564, + "grad_norm": 2.3559885025024414, + "learning_rate": 4.983753046303819e-06, + "loss": 2.2711, + "step": 296 + }, + { + "epoch": 0.3254794520547945, + "grad_norm": 2.120802879333496, + "learning_rate": 4.981722177091796e-06, + "loss": 2.2946, + "step": 297 + }, + { + "epoch": 0.3265753424657534, + "grad_norm": 2.2319388389587402, + "learning_rate": 4.979691307879773e-06, + "loss": 2.2771, + "step": 298 + }, + { + "epoch": 0.32767123287671235, + "grad_norm": 2.208740711212158, + "learning_rate": 4.9776604386677504e-06, + "loss": 2.3003, + "step": 299 + }, + { + "epoch": 0.3287671232876712, + "grad_norm": 2.543060541152954, + "learning_rate": 4.975629569455728e-06, + "loss": 2.2891, + "step": 300 + }, + { + "epoch": 0.3287671232876712, + "eval_loss": 2.398495674133301, + "eval_runtime": 51.1973, + "eval_samples_per_second": 19.532, + "eval_steps_per_second": 0.82, + "step": 300 + }, + { + "epoch": 0.32986301369863014, + "grad_norm": 2.086219072341919, + "learning_rate": 4.973598700243705e-06, + "loss": 2.3286, + "step": 301 + }, + { + "epoch": 0.33095890410958906, + "grad_norm": 2.3673319816589355, + "learning_rate": 4.971567831031681e-06, + "loss": 2.2672, + "step": 302 + }, + { + "epoch": 0.3320547945205479, + "grad_norm": 2.167593240737915, + "learning_rate": 4.9695369618196595e-06, + "loss": 2.264, + "step": 303 + }, + { + "epoch": 0.33315068493150685, + "grad_norm": 2.7218785285949707, + "learning_rate": 4.967506092607636e-06, + "loss": 2.1737, + "step": 304 + }, + { + "epoch": 0.33424657534246577, + "grad_norm": 2.673245429992676, + "learning_rate": 4.965475223395614e-06, + "loss": 2.2439, + "step": 305 + }, + { + "epoch": 0.33534246575342463, + "grad_norm": 2.0341989994049072, + "learning_rate": 4.9634443541835905e-06, + "loss": 2.2614, + "step": 306 + }, + { + "epoch": 0.33643835616438356, + "grad_norm": 1.9314247369766235, + "learning_rate": 4.961413484971569e-06, + "loss": 2.2592, + "step": 307 + }, + { + "epoch": 0.3375342465753425, + "grad_norm": 2.537794589996338, + "learning_rate": 4.959382615759545e-06, + "loss": 2.2859, + "step": 308 + }, + { + "epoch": 0.33863013698630134, + "grad_norm": 2.0811166763305664, + "learning_rate": 4.957351746547523e-06, + "loss": 2.2868, + "step": 309 + }, + { + "epoch": 0.33972602739726027, + "grad_norm": 2.783701181411743, + "learning_rate": 4.9553208773355e-06, + "loss": 2.2649, + "step": 310 + }, + { + "epoch": 0.33972602739726027, + "eval_loss": 2.400538921356201, + "eval_runtime": 43.185, + "eval_samples_per_second": 23.156, + "eval_steps_per_second": 0.973, + "step": 310 + }, + { + "epoch": 0.3408219178082192, + "grad_norm": 2.8541319370269775, + "learning_rate": 4.953290008123478e-06, + "loss": 2.3729, + "step": 311 + }, + { + "epoch": 0.3419178082191781, + "grad_norm": 2.1852617263793945, + "learning_rate": 4.951259138911454e-06, + "loss": 2.2885, + "step": 312 + }, + { + "epoch": 0.343013698630137, + "grad_norm": 2.872201919555664, + "learning_rate": 4.9492282696994315e-06, + "loss": 2.2353, + "step": 313 + }, + { + "epoch": 0.3441095890410959, + "grad_norm": 2.5140910148620605, + "learning_rate": 4.947197400487409e-06, + "loss": 2.3364, + "step": 314 + }, + { + "epoch": 0.3452054794520548, + "grad_norm": 2.619974374771118, + "learning_rate": 4.945166531275386e-06, + "loss": 2.2721, + "step": 315 + }, + { + "epoch": 0.3463013698630137, + "grad_norm": 2.4567618370056152, + "learning_rate": 4.943135662063363e-06, + "loss": 2.26, + "step": 316 + }, + { + "epoch": 0.3473972602739726, + "grad_norm": 2.9603192806243896, + "learning_rate": 4.9411047928513406e-06, + "loss": 2.2861, + "step": 317 + }, + { + "epoch": 0.34849315068493153, + "grad_norm": 2.115262985229492, + "learning_rate": 4.939073923639318e-06, + "loss": 2.3125, + "step": 318 + }, + { + "epoch": 0.3495890410958904, + "grad_norm": 2.693024158477783, + "learning_rate": 4.937043054427295e-06, + "loss": 2.3054, + "step": 319 + }, + { + "epoch": 0.3506849315068493, + "grad_norm": 2.3439996242523193, + "learning_rate": 4.935012185215272e-06, + "loss": 2.2576, + "step": 320 + }, + { + "epoch": 0.3506849315068493, + "eval_loss": 2.3920650482177734, + "eval_runtime": 47.7451, + "eval_samples_per_second": 20.945, + "eval_steps_per_second": 0.88, + "step": 320 + }, + { + "epoch": 0.35178082191780824, + "grad_norm": 2.3009634017944336, + "learning_rate": 4.93298131600325e-06, + "loss": 2.2454, + "step": 321 + }, + { + "epoch": 0.3528767123287671, + "grad_norm": 2.593203544616699, + "learning_rate": 4.930950446791227e-06, + "loss": 2.2965, + "step": 322 + }, + { + "epoch": 0.353972602739726, + "grad_norm": 2.9069650173187256, + "learning_rate": 4.928919577579204e-06, + "loss": 2.2785, + "step": 323 + }, + { + "epoch": 0.35506849315068495, + "grad_norm": 2.1021759510040283, + "learning_rate": 4.9268887083671815e-06, + "loss": 2.2857, + "step": 324 + }, + { + "epoch": 0.3561643835616438, + "grad_norm": 2.0330185890197754, + "learning_rate": 4.924857839155159e-06, + "loss": 2.2697, + "step": 325 + }, + { + "epoch": 0.35726027397260274, + "grad_norm": 2.1318063735961914, + "learning_rate": 4.922826969943136e-06, + "loss": 2.2698, + "step": 326 + }, + { + "epoch": 0.35835616438356166, + "grad_norm": 2.0157387256622314, + "learning_rate": 4.920796100731113e-06, + "loss": 2.292, + "step": 327 + }, + { + "epoch": 0.3594520547945205, + "grad_norm": 2.3266026973724365, + "learning_rate": 4.918765231519091e-06, + "loss": 2.2215, + "step": 328 + }, + { + "epoch": 0.36054794520547945, + "grad_norm": 2.066305160522461, + "learning_rate": 4.916734362307068e-06, + "loss": 2.2758, + "step": 329 + }, + { + "epoch": 0.36164383561643837, + "grad_norm": 2.0219993591308594, + "learning_rate": 4.914703493095045e-06, + "loss": 2.2381, + "step": 330 + }, + { + "epoch": 0.36164383561643837, + "eval_loss": 2.394064426422119, + "eval_runtime": 51.5482, + "eval_samples_per_second": 19.399, + "eval_steps_per_second": 0.815, + "step": 330 + }, + { + "epoch": 0.36273972602739724, + "grad_norm": 2.3964319229125977, + "learning_rate": 4.912672623883022e-06, + "loss": 2.2384, + "step": 331 + }, + { + "epoch": 0.36383561643835616, + "grad_norm": 2.5466530323028564, + "learning_rate": 4.910641754671e-06, + "loss": 2.2428, + "step": 332 + }, + { + "epoch": 0.3649315068493151, + "grad_norm": 2.236152410507202, + "learning_rate": 4.908610885458977e-06, + "loss": 2.2639, + "step": 333 + }, + { + "epoch": 0.36602739726027395, + "grad_norm": 2.7267303466796875, + "learning_rate": 4.906580016246954e-06, + "loss": 2.2608, + "step": 334 + }, + { + "epoch": 0.36712328767123287, + "grad_norm": 1.9403765201568604, + "learning_rate": 4.9045491470349315e-06, + "loss": 2.2962, + "step": 335 + }, + { + "epoch": 0.3682191780821918, + "grad_norm": 2.89070463180542, + "learning_rate": 4.902518277822909e-06, + "loss": 2.2847, + "step": 336 + }, + { + "epoch": 0.3693150684931507, + "grad_norm": 2.6427228450775146, + "learning_rate": 4.900487408610886e-06, + "loss": 2.2734, + "step": 337 + }, + { + "epoch": 0.3704109589041096, + "grad_norm": 2.1680362224578857, + "learning_rate": 4.898456539398863e-06, + "loss": 2.2869, + "step": 338 + }, + { + "epoch": 0.3715068493150685, + "grad_norm": 3.0160160064697266, + "learning_rate": 4.89642567018684e-06, + "loss": 2.3043, + "step": 339 + }, + { + "epoch": 0.3726027397260274, + "grad_norm": 1.9858418703079224, + "learning_rate": 4.894394800974818e-06, + "loss": 2.3422, + "step": 340 + }, + { + "epoch": 0.3726027397260274, + "eval_loss": 2.387559652328491, + "eval_runtime": 52.2455, + "eval_samples_per_second": 19.14, + "eval_steps_per_second": 0.804, + "step": 340 + }, + { + "epoch": 0.3736986301369863, + "grad_norm": 2.243699073791504, + "learning_rate": 4.892363931762794e-06, + "loss": 2.2394, + "step": 341 + }, + { + "epoch": 0.3747945205479452, + "grad_norm": 2.371299982070923, + "learning_rate": 4.8903330625507725e-06, + "loss": 2.2856, + "step": 342 + }, + { + "epoch": 0.37589041095890413, + "grad_norm": 2.3769736289978027, + "learning_rate": 4.888302193338749e-06, + "loss": 2.246, + "step": 343 + }, + { + "epoch": 0.376986301369863, + "grad_norm": 2.3209030628204346, + "learning_rate": 4.886271324126727e-06, + "loss": 2.2331, + "step": 344 + }, + { + "epoch": 0.3780821917808219, + "grad_norm": 2.747319221496582, + "learning_rate": 4.8842404549147034e-06, + "loss": 2.3115, + "step": 345 + }, + { + "epoch": 0.37917808219178084, + "grad_norm": 1.8599231243133545, + "learning_rate": 4.8822095857026816e-06, + "loss": 2.2994, + "step": 346 + }, + { + "epoch": 0.3802739726027397, + "grad_norm": 2.0358943939208984, + "learning_rate": 4.880178716490658e-06, + "loss": 2.271, + "step": 347 + }, + { + "epoch": 0.38136986301369863, + "grad_norm": 2.409106731414795, + "learning_rate": 4.878147847278636e-06, + "loss": 2.2281, + "step": 348 + }, + { + "epoch": 0.38246575342465755, + "grad_norm": 2.1641807556152344, + "learning_rate": 4.8761169780666125e-06, + "loss": 2.2437, + "step": 349 + }, + { + "epoch": 0.3835616438356164, + "grad_norm": 2.6917359828948975, + "learning_rate": 4.874086108854591e-06, + "loss": 2.226, + "step": 350 + }, + { + "epoch": 0.3835616438356164, + "eval_loss": 2.382991313934326, + "eval_runtime": 48.9534, + "eval_samples_per_second": 20.428, + "eval_steps_per_second": 0.858, + "step": 350 + }, + { + "epoch": 0.38465753424657534, + "grad_norm": 2.364215612411499, + "learning_rate": 4.872055239642567e-06, + "loss": 2.2319, + "step": 351 + }, + { + "epoch": 0.38575342465753426, + "grad_norm": 2.317643165588379, + "learning_rate": 4.870024370430544e-06, + "loss": 2.2096, + "step": 352 + }, + { + "epoch": 0.3868493150684931, + "grad_norm": 2.2283246517181396, + "learning_rate": 4.867993501218522e-06, + "loss": 2.2634, + "step": 353 + }, + { + "epoch": 0.38794520547945205, + "grad_norm": 2.498391628265381, + "learning_rate": 4.865962632006499e-06, + "loss": 2.2813, + "step": 354 + }, + { + "epoch": 0.38904109589041097, + "grad_norm": 2.3889403343200684, + "learning_rate": 4.863931762794476e-06, + "loss": 2.2361, + "step": 355 + }, + { + "epoch": 0.39013698630136984, + "grad_norm": 2.2049975395202637, + "learning_rate": 4.8619008935824535e-06, + "loss": 2.2583, + "step": 356 + }, + { + "epoch": 0.39123287671232876, + "grad_norm": 2.132840871810913, + "learning_rate": 4.859870024370431e-06, + "loss": 2.2823, + "step": 357 + }, + { + "epoch": 0.3923287671232877, + "grad_norm": 2.2692184448242188, + "learning_rate": 4.857839155158408e-06, + "loss": 2.2939, + "step": 358 + }, + { + "epoch": 0.3934246575342466, + "grad_norm": 2.0501933097839355, + "learning_rate": 4.855808285946385e-06, + "loss": 2.2976, + "step": 359 + }, + { + "epoch": 0.39452054794520547, + "grad_norm": 2.170236587524414, + "learning_rate": 4.8537774167343626e-06, + "loss": 2.2748, + "step": 360 + }, + { + "epoch": 0.39452054794520547, + "eval_loss": 2.3783886432647705, + "eval_runtime": 46.7324, + "eval_samples_per_second": 21.398, + "eval_steps_per_second": 0.899, + "step": 360 + }, + { + "epoch": 0.3956164383561644, + "grad_norm": 2.283137083053589, + "learning_rate": 4.85174654752234e-06, + "loss": 2.2705, + "step": 361 + }, + { + "epoch": 0.3967123287671233, + "grad_norm": 2.195984363555908, + "learning_rate": 4.849715678310317e-06, + "loss": 2.26, + "step": 362 + }, + { + "epoch": 0.3978082191780822, + "grad_norm": 2.399935007095337, + "learning_rate": 4.847684809098294e-06, + "loss": 2.2639, + "step": 363 + }, + { + "epoch": 0.3989041095890411, + "grad_norm": 2.411280393600464, + "learning_rate": 4.845653939886272e-06, + "loss": 2.2523, + "step": 364 + }, + { + "epoch": 0.4, + "grad_norm": 2.2033896446228027, + "learning_rate": 4.843623070674249e-06, + "loss": 2.2722, + "step": 365 + }, + { + "epoch": 0.4010958904109589, + "grad_norm": 2.2990238666534424, + "learning_rate": 4.841592201462226e-06, + "loss": 2.2411, + "step": 366 + }, + { + "epoch": 0.4021917808219178, + "grad_norm": 1.911453127861023, + "learning_rate": 4.8395613322502035e-06, + "loss": 2.2921, + "step": 367 + }, + { + "epoch": 0.40328767123287673, + "grad_norm": 2.453195095062256, + "learning_rate": 4.837530463038181e-06, + "loss": 2.218, + "step": 368 + }, + { + "epoch": 0.4043835616438356, + "grad_norm": 2.126992702484131, + "learning_rate": 4.835499593826158e-06, + "loss": 2.2628, + "step": 369 + }, + { + "epoch": 0.4054794520547945, + "grad_norm": 1.9305835962295532, + "learning_rate": 4.833468724614135e-06, + "loss": 2.2726, + "step": 370 + }, + { + "epoch": 0.4054794520547945, + "eval_loss": 2.3741114139556885, + "eval_runtime": 49.2956, + "eval_samples_per_second": 20.286, + "eval_steps_per_second": 0.852, + "step": 370 + }, + { + "epoch": 0.40657534246575344, + "grad_norm": 2.222024917602539, + "learning_rate": 4.831437855402113e-06, + "loss": 2.2634, + "step": 371 + }, + { + "epoch": 0.4076712328767123, + "grad_norm": 2.473958969116211, + "learning_rate": 4.82940698619009e-06, + "loss": 2.1817, + "step": 372 + }, + { + "epoch": 0.40876712328767123, + "grad_norm": 1.9208838939666748, + "learning_rate": 4.827376116978067e-06, + "loss": 2.2353, + "step": 373 + }, + { + "epoch": 0.40986301369863015, + "grad_norm": 2.7226450443267822, + "learning_rate": 4.8253452477660444e-06, + "loss": 2.2831, + "step": 374 + }, + { + "epoch": 0.410958904109589, + "grad_norm": 1.9078222513198853, + "learning_rate": 4.823314378554022e-06, + "loss": 2.264, + "step": 375 + }, + { + "epoch": 0.41205479452054794, + "grad_norm": 2.3946311473846436, + "learning_rate": 4.821283509341998e-06, + "loss": 2.251, + "step": 376 + }, + { + "epoch": 0.41315068493150686, + "grad_norm": 1.847343921661377, + "learning_rate": 4.819252640129976e-06, + "loss": 2.2441, + "step": 377 + }, + { + "epoch": 0.41424657534246573, + "grad_norm": 2.475385904312134, + "learning_rate": 4.817221770917953e-06, + "loss": 2.2639, + "step": 378 + }, + { + "epoch": 0.41534246575342465, + "grad_norm": 2.1396560668945312, + "learning_rate": 4.815190901705931e-06, + "loss": 2.2353, + "step": 379 + }, + { + "epoch": 0.41643835616438357, + "grad_norm": 2.3868885040283203, + "learning_rate": 4.813160032493907e-06, + "loss": 2.2551, + "step": 380 + }, + { + "epoch": 0.41643835616438357, + "eval_loss": 2.369168519973755, + "eval_runtime": 45.6131, + "eval_samples_per_second": 21.924, + "eval_steps_per_second": 0.921, + "step": 380 + }, + { + "epoch": 0.41753424657534244, + "grad_norm": 2.0193872451782227, + "learning_rate": 4.811129163281885e-06, + "loss": 2.2399, + "step": 381 + }, + { + "epoch": 0.41863013698630136, + "grad_norm": 2.4655935764312744, + "learning_rate": 4.809098294069862e-06, + "loss": 2.2729, + "step": 382 + }, + { + "epoch": 0.4197260273972603, + "grad_norm": 1.880974292755127, + "learning_rate": 4.80706742485784e-06, + "loss": 2.2681, + "step": 383 + }, + { + "epoch": 0.4208219178082192, + "grad_norm": 2.1912078857421875, + "learning_rate": 4.805036555645816e-06, + "loss": 2.2961, + "step": 384 + }, + { + "epoch": 0.42191780821917807, + "grad_norm": 1.985931634902954, + "learning_rate": 4.8030056864337945e-06, + "loss": 2.2407, + "step": 385 + }, + { + "epoch": 0.423013698630137, + "grad_norm": 2.4353220462799072, + "learning_rate": 4.800974817221771e-06, + "loss": 2.2867, + "step": 386 + }, + { + "epoch": 0.4241095890410959, + "grad_norm": 1.8810508251190186, + "learning_rate": 4.798943948009749e-06, + "loss": 2.2498, + "step": 387 + }, + { + "epoch": 0.4252054794520548, + "grad_norm": 2.195033073425293, + "learning_rate": 4.7969130787977254e-06, + "loss": 2.2532, + "step": 388 + }, + { + "epoch": 0.4263013698630137, + "grad_norm": 1.855372428894043, + "learning_rate": 4.794882209585703e-06, + "loss": 2.2538, + "step": 389 + }, + { + "epoch": 0.4273972602739726, + "grad_norm": 2.4465434551239014, + "learning_rate": 4.79285134037368e-06, + "loss": 2.2724, + "step": 390 + }, + { + "epoch": 0.4273972602739726, + "eval_loss": 2.364928960800171, + "eval_runtime": 48.0245, + "eval_samples_per_second": 20.823, + "eval_steps_per_second": 0.875, + "step": 390 + }, + { + "epoch": 0.4284931506849315, + "grad_norm": 1.8911314010620117, + "learning_rate": 4.790820471161657e-06, + "loss": 2.2291, + "step": 391 + }, + { + "epoch": 0.4295890410958904, + "grad_norm": 2.1975977420806885, + "learning_rate": 4.7887896019496345e-06, + "loss": 2.2373, + "step": 392 + }, + { + "epoch": 0.43068493150684933, + "grad_norm": 2.015331506729126, + "learning_rate": 4.786758732737612e-06, + "loss": 2.2483, + "step": 393 + }, + { + "epoch": 0.4317808219178082, + "grad_norm": 2.039339065551758, + "learning_rate": 4.784727863525589e-06, + "loss": 2.2746, + "step": 394 + }, + { + "epoch": 0.4328767123287671, + "grad_norm": 1.977513074874878, + "learning_rate": 4.782696994313566e-06, + "loss": 2.2275, + "step": 395 + }, + { + "epoch": 0.43397260273972604, + "grad_norm": 2.147217035293579, + "learning_rate": 4.780666125101544e-06, + "loss": 2.252, + "step": 396 + }, + { + "epoch": 0.4350684931506849, + "grad_norm": 2.1799919605255127, + "learning_rate": 4.778635255889521e-06, + "loss": 2.1975, + "step": 397 + }, + { + "epoch": 0.43616438356164383, + "grad_norm": 2.582240343093872, + "learning_rate": 4.776604386677498e-06, + "loss": 2.2294, + "step": 398 + }, + { + "epoch": 0.43726027397260275, + "grad_norm": 1.9127846956253052, + "learning_rate": 4.7745735174654755e-06, + "loss": 2.2319, + "step": 399 + }, + { + "epoch": 0.4383561643835616, + "grad_norm": 2.2444543838500977, + "learning_rate": 4.772542648253453e-06, + "loss": 2.2857, + "step": 400 + }, + { + "epoch": 0.4383561643835616, + "eval_loss": 2.3653724193573, + "eval_runtime": 43.8304, + "eval_samples_per_second": 22.815, + "eval_steps_per_second": 0.958, + "step": 400 + }, + { + "epoch": 0.43945205479452054, + "grad_norm": 1.9913676977157593, + "learning_rate": 4.77051177904143e-06, + "loss": 2.274, + "step": 401 + }, + { + "epoch": 0.44054794520547946, + "grad_norm": 2.2850067615509033, + "learning_rate": 4.768480909829407e-06, + "loss": 2.2209, + "step": 402 + }, + { + "epoch": 0.44164383561643833, + "grad_norm": 2.4984688758850098, + "learning_rate": 4.766450040617385e-06, + "loss": 2.263, + "step": 403 + }, + { + "epoch": 0.44273972602739725, + "grad_norm": 2.3244640827178955, + "learning_rate": 4.764419171405362e-06, + "loss": 2.2368, + "step": 404 + }, + { + "epoch": 0.4438356164383562, + "grad_norm": 2.7579681873321533, + "learning_rate": 4.762388302193339e-06, + "loss": 2.2762, + "step": 405 + }, + { + "epoch": 0.4449315068493151, + "grad_norm": 2.1996467113494873, + "learning_rate": 4.760357432981316e-06, + "loss": 2.2536, + "step": 406 + }, + { + "epoch": 0.44602739726027396, + "grad_norm": 2.3767571449279785, + "learning_rate": 4.758326563769294e-06, + "loss": 2.2353, + "step": 407 + }, + { + "epoch": 0.4471232876712329, + "grad_norm": 1.9148359298706055, + "learning_rate": 4.756295694557271e-06, + "loss": 2.2416, + "step": 408 + }, + { + "epoch": 0.4482191780821918, + "grad_norm": 2.6499507427215576, + "learning_rate": 4.754264825345248e-06, + "loss": 2.2641, + "step": 409 + }, + { + "epoch": 0.44931506849315067, + "grad_norm": 1.9890286922454834, + "learning_rate": 4.7522339561332255e-06, + "loss": 2.2148, + "step": 410 + }, + { + "epoch": 0.44931506849315067, + "eval_loss": 2.3670520782470703, + "eval_runtime": 41.1722, + "eval_samples_per_second": 24.288, + "eval_steps_per_second": 1.02, + "step": 410 + }, + { + "epoch": 0.4504109589041096, + "grad_norm": 2.575968027114868, + "learning_rate": 4.750203086921203e-06, + "loss": 2.2254, + "step": 411 + }, + { + "epoch": 0.4515068493150685, + "grad_norm": 2.015174388885498, + "learning_rate": 4.74817221770918e-06, + "loss": 2.1881, + "step": 412 + }, + { + "epoch": 0.4526027397260274, + "grad_norm": 2.44292950630188, + "learning_rate": 4.7461413484971565e-06, + "loss": 2.256, + "step": 413 + }, + { + "epoch": 0.4536986301369863, + "grad_norm": 1.9808441400527954, + "learning_rate": 4.744110479285135e-06, + "loss": 2.285, + "step": 414 + }, + { + "epoch": 0.4547945205479452, + "grad_norm": 2.009481906890869, + "learning_rate": 4.742079610073111e-06, + "loss": 2.2018, + "step": 415 + }, + { + "epoch": 0.4558904109589041, + "grad_norm": 2.1466031074523926, + "learning_rate": 4.740048740861089e-06, + "loss": 2.224, + "step": 416 + }, + { + "epoch": 0.456986301369863, + "grad_norm": 2.1337907314300537, + "learning_rate": 4.738017871649066e-06, + "loss": 2.2404, + "step": 417 + }, + { + "epoch": 0.45808219178082193, + "grad_norm": 1.94125235080719, + "learning_rate": 4.735987002437044e-06, + "loss": 2.2474, + "step": 418 + }, + { + "epoch": 0.4591780821917808, + "grad_norm": 2.379099130630493, + "learning_rate": 4.73395613322502e-06, + "loss": 2.2858, + "step": 419 + }, + { + "epoch": 0.4602739726027397, + "grad_norm": 1.844056248664856, + "learning_rate": 4.731925264012998e-06, + "loss": 2.2967, + "step": 420 + }, + { + "epoch": 0.4602739726027397, + "eval_loss": 2.3622241020202637, + "eval_runtime": 45.4674, + "eval_samples_per_second": 21.994, + "eval_steps_per_second": 0.924, + "step": 420 + }, + { + "epoch": 0.46136986301369864, + "grad_norm": 2.411895513534546, + "learning_rate": 4.729894394800975e-06, + "loss": 2.274, + "step": 421 + }, + { + "epoch": 0.4624657534246575, + "grad_norm": 2.014885425567627, + "learning_rate": 4.727863525588953e-06, + "loss": 2.2144, + "step": 422 + }, + { + "epoch": 0.46356164383561643, + "grad_norm": 2.112009048461914, + "learning_rate": 4.725832656376929e-06, + "loss": 2.1824, + "step": 423 + }, + { + "epoch": 0.46465753424657535, + "grad_norm": 2.111392021179199, + "learning_rate": 4.723801787164907e-06, + "loss": 2.2261, + "step": 424 + }, + { + "epoch": 0.4657534246575342, + "grad_norm": 2.0349185466766357, + "learning_rate": 4.721770917952885e-06, + "loss": 2.2622, + "step": 425 + }, + { + "epoch": 0.46684931506849314, + "grad_norm": 2.30222487449646, + "learning_rate": 4.719740048740861e-06, + "loss": 2.2206, + "step": 426 + }, + { + "epoch": 0.46794520547945206, + "grad_norm": 2.018789529800415, + "learning_rate": 4.717709179528839e-06, + "loss": 2.2196, + "step": 427 + }, + { + "epoch": 0.469041095890411, + "grad_norm": 2.046649932861328, + "learning_rate": 4.715678310316816e-06, + "loss": 2.2584, + "step": 428 + }, + { + "epoch": 0.47013698630136985, + "grad_norm": 2.0393869876861572, + "learning_rate": 4.713647441104794e-06, + "loss": 2.2537, + "step": 429 + }, + { + "epoch": 0.4712328767123288, + "grad_norm": 2.2788116931915283, + "learning_rate": 4.71161657189277e-06, + "loss": 2.2145, + "step": 430 + }, + { + "epoch": 0.4712328767123288, + "eval_loss": 2.3575785160064697, + "eval_runtime": 44.6687, + "eval_samples_per_second": 22.387, + "eval_steps_per_second": 0.94, + "step": 430 + }, + { + "epoch": 0.4723287671232877, + "grad_norm": 1.9950047731399536, + "learning_rate": 4.709585702680748e-06, + "loss": 2.2072, + "step": 431 + }, + { + "epoch": 0.47342465753424656, + "grad_norm": 1.9222451448440552, + "learning_rate": 4.707554833468725e-06, + "loss": 2.2654, + "step": 432 + }, + { + "epoch": 0.4745205479452055, + "grad_norm": 1.9735718965530396, + "learning_rate": 4.705523964256703e-06, + "loss": 2.3037, + "step": 433 + }, + { + "epoch": 0.4756164383561644, + "grad_norm": 2.3209776878356934, + "learning_rate": 4.703493095044679e-06, + "loss": 2.2083, + "step": 434 + }, + { + "epoch": 0.4767123287671233, + "grad_norm": 2.0993995666503906, + "learning_rate": 4.701462225832657e-06, + "loss": 2.2275, + "step": 435 + }, + { + "epoch": 0.4778082191780822, + "grad_norm": 2.334463119506836, + "learning_rate": 4.699431356620634e-06, + "loss": 2.1852, + "step": 436 + }, + { + "epoch": 0.4789041095890411, + "grad_norm": 1.9098213911056519, + "learning_rate": 4.697400487408612e-06, + "loss": 2.2829, + "step": 437 + }, + { + "epoch": 0.48, + "grad_norm": 2.3005056381225586, + "learning_rate": 4.695369618196588e-06, + "loss": 2.2532, + "step": 438 + }, + { + "epoch": 0.4810958904109589, + "grad_norm": 1.9069781303405762, + "learning_rate": 4.693338748984566e-06, + "loss": 2.2374, + "step": 439 + }, + { + "epoch": 0.4821917808219178, + "grad_norm": 2.050032138824463, + "learning_rate": 4.691307879772543e-06, + "loss": 2.2256, + "step": 440 + }, + { + "epoch": 0.4821917808219178, + "eval_loss": 2.3587422370910645, + "eval_runtime": 45.0941, + "eval_samples_per_second": 22.176, + "eval_steps_per_second": 0.931, + "step": 440 + }, + { + "epoch": 0.4832876712328767, + "grad_norm": 2.0563836097717285, + "learning_rate": 4.68927701056052e-06, + "loss": 2.1918, + "step": 441 + }, + { + "epoch": 0.4843835616438356, + "grad_norm": 2.1598386764526367, + "learning_rate": 4.6872461413484975e-06, + "loss": 2.2618, + "step": 442 + }, + { + "epoch": 0.48547945205479454, + "grad_norm": 1.9236539602279663, + "learning_rate": 4.685215272136475e-06, + "loss": 2.2477, + "step": 443 + }, + { + "epoch": 0.4865753424657534, + "grad_norm": 1.9923924207687378, + "learning_rate": 4.683184402924452e-06, + "loss": 2.2313, + "step": 444 + }, + { + "epoch": 0.4876712328767123, + "grad_norm": 2.162627696990967, + "learning_rate": 4.681153533712429e-06, + "loss": 2.2367, + "step": 445 + }, + { + "epoch": 0.48876712328767125, + "grad_norm": 2.083523988723755, + "learning_rate": 4.679122664500407e-06, + "loss": 2.2946, + "step": 446 + }, + { + "epoch": 0.4898630136986301, + "grad_norm": 2.504408359527588, + "learning_rate": 4.677091795288384e-06, + "loss": 2.2104, + "step": 447 + }, + { + "epoch": 0.49095890410958903, + "grad_norm": 1.8316532373428345, + "learning_rate": 4.675060926076361e-06, + "loss": 2.2027, + "step": 448 + }, + { + "epoch": 0.49205479452054796, + "grad_norm": 2.3337953090667725, + "learning_rate": 4.673030056864338e-06, + "loss": 2.278, + "step": 449 + }, + { + "epoch": 0.4931506849315068, + "grad_norm": 2.24102783203125, + "learning_rate": 4.670999187652316e-06, + "loss": 2.2018, + "step": 450 + }, + { + "epoch": 0.4931506849315068, + "eval_loss": 2.3545279502868652, + "eval_runtime": 45.1041, + "eval_samples_per_second": 22.171, + "eval_steps_per_second": 0.931, + "step": 450 + }, + { + "epoch": 0.49424657534246574, + "grad_norm": 2.07834792137146, + "learning_rate": 4.668968318440293e-06, + "loss": 2.1975, + "step": 451 + }, + { + "epoch": 0.49534246575342467, + "grad_norm": 1.7681723833084106, + "learning_rate": 4.66693744922827e-06, + "loss": 2.2033, + "step": 452 + }, + { + "epoch": 0.4964383561643836, + "grad_norm": 2.062134027481079, + "learning_rate": 4.6649065800162475e-06, + "loss": 2.2646, + "step": 453 + }, + { + "epoch": 0.49753424657534245, + "grad_norm": 1.94647216796875, + "learning_rate": 4.662875710804225e-06, + "loss": 2.1748, + "step": 454 + }, + { + "epoch": 0.4986301369863014, + "grad_norm": 1.9173872470855713, + "learning_rate": 4.660844841592202e-06, + "loss": 2.2435, + "step": 455 + }, + { + "epoch": 0.4997260273972603, + "grad_norm": 1.88880455493927, + "learning_rate": 4.658813972380179e-06, + "loss": 2.1711, + "step": 456 + }, + { + "epoch": 0.5008219178082192, + "grad_norm": 1.9628506898880005, + "learning_rate": 4.656783103168157e-06, + "loss": 2.2701, + "step": 457 + }, + { + "epoch": 0.5019178082191781, + "grad_norm": 2.0515286922454834, + "learning_rate": 4.654752233956134e-06, + "loss": 2.1656, + "step": 458 + }, + { + "epoch": 0.503013698630137, + "grad_norm": 1.978522539138794, + "learning_rate": 4.652721364744111e-06, + "loss": 2.246, + "step": 459 + }, + { + "epoch": 0.5041095890410959, + "grad_norm": 1.7993966341018677, + "learning_rate": 4.6506904955320885e-06, + "loss": 2.214, + "step": 460 + }, + { + "epoch": 0.5041095890410959, + "eval_loss": 2.3568837642669678, + "eval_runtime": 45.2086, + "eval_samples_per_second": 22.12, + "eval_steps_per_second": 0.929, + "step": 460 + }, + { + "epoch": 0.5052054794520547, + "grad_norm": 1.7809834480285645, + "learning_rate": 4.648659626320066e-06, + "loss": 2.2525, + "step": 461 + }, + { + "epoch": 0.5063013698630137, + "grad_norm": 1.9256166219711304, + "learning_rate": 4.646628757108043e-06, + "loss": 2.2611, + "step": 462 + }, + { + "epoch": 0.5073972602739726, + "grad_norm": 1.94192636013031, + "learning_rate": 4.6445978878960194e-06, + "loss": 2.2282, + "step": 463 + }, + { + "epoch": 0.5084931506849315, + "grad_norm": 1.8891818523406982, + "learning_rate": 4.6425670186839976e-06, + "loss": 2.2036, + "step": 464 + }, + { + "epoch": 0.5095890410958904, + "grad_norm": 2.352078676223755, + "learning_rate": 4.640536149471974e-06, + "loss": 2.2293, + "step": 465 + }, + { + "epoch": 0.5106849315068493, + "grad_norm": 1.8141659498214722, + "learning_rate": 4.638505280259952e-06, + "loss": 2.2551, + "step": 466 + }, + { + "epoch": 0.5117808219178083, + "grad_norm": 1.9492480754852295, + "learning_rate": 4.6364744110479285e-06, + "loss": 2.2641, + "step": 467 + }, + { + "epoch": 0.5128767123287671, + "grad_norm": 2.0210182666778564, + "learning_rate": 4.634443541835907e-06, + "loss": 2.2367, + "step": 468 + }, + { + "epoch": 0.513972602739726, + "grad_norm": 1.8936505317687988, + "learning_rate": 4.632412672623883e-06, + "loss": 2.265, + "step": 469 + }, + { + "epoch": 0.5150684931506849, + "grad_norm": 1.845888614654541, + "learning_rate": 4.630381803411861e-06, + "loss": 2.2169, + "step": 470 + }, + { + "epoch": 0.5150684931506849, + "eval_loss": 2.353489875793457, + "eval_runtime": 46.8739, + "eval_samples_per_second": 21.334, + "eval_steps_per_second": 0.896, + "step": 470 + }, + { + "epoch": 0.5161643835616438, + "grad_norm": 1.889675259590149, + "learning_rate": 4.628350934199838e-06, + "loss": 2.1383, + "step": 471 + }, + { + "epoch": 0.5172602739726028, + "grad_norm": 1.9799216985702515, + "learning_rate": 4.626320064987816e-06, + "loss": 2.2312, + "step": 472 + }, + { + "epoch": 0.5183561643835617, + "grad_norm": 2.0128607749938965, + "learning_rate": 4.624289195775792e-06, + "loss": 2.2529, + "step": 473 + }, + { + "epoch": 0.5194520547945205, + "grad_norm": 1.8791241645812988, + "learning_rate": 4.62225832656377e-06, + "loss": 2.226, + "step": 474 + }, + { + "epoch": 0.5205479452054794, + "grad_norm": 2.0497279167175293, + "learning_rate": 4.620227457351747e-06, + "loss": 2.2283, + "step": 475 + }, + { + "epoch": 0.5216438356164383, + "grad_norm": 1.8097424507141113, + "learning_rate": 4.618196588139724e-06, + "loss": 2.222, + "step": 476 + }, + { + "epoch": 0.5227397260273973, + "grad_norm": 1.8960500955581665, + "learning_rate": 4.616165718927701e-06, + "loss": 2.2342, + "step": 477 + }, + { + "epoch": 0.5238356164383562, + "grad_norm": 2.020730495452881, + "learning_rate": 4.6141348497156786e-06, + "loss": 2.2106, + "step": 478 + }, + { + "epoch": 0.5249315068493151, + "grad_norm": 1.8955297470092773, + "learning_rate": 4.612103980503656e-06, + "loss": 2.2755, + "step": 479 + }, + { + "epoch": 0.5260273972602739, + "grad_norm": 1.9767351150512695, + "learning_rate": 4.610073111291633e-06, + "loss": 2.2554, + "step": 480 + }, + { + "epoch": 0.5260273972602739, + "eval_loss": 2.347705841064453, + "eval_runtime": 46.9125, + "eval_samples_per_second": 21.316, + "eval_steps_per_second": 0.895, + "step": 480 + }, + { + "epoch": 0.5271232876712328, + "grad_norm": 1.936457633972168, + "learning_rate": 4.60804224207961e-06, + "loss": 2.2323, + "step": 481 + }, + { + "epoch": 0.5282191780821918, + "grad_norm": 1.9271862506866455, + "learning_rate": 4.606011372867588e-06, + "loss": 2.2151, + "step": 482 + }, + { + "epoch": 0.5293150684931507, + "grad_norm": 1.878153681755066, + "learning_rate": 4.603980503655565e-06, + "loss": 2.2322, + "step": 483 + }, + { + "epoch": 0.5304109589041096, + "grad_norm": 2.1009249687194824, + "learning_rate": 4.601949634443542e-06, + "loss": 2.2181, + "step": 484 + }, + { + "epoch": 0.5315068493150685, + "grad_norm": 1.9089181423187256, + "learning_rate": 4.5999187652315195e-06, + "loss": 2.1712, + "step": 485 + }, + { + "epoch": 0.5326027397260275, + "grad_norm": 2.207282781600952, + "learning_rate": 4.597887896019497e-06, + "loss": 2.1847, + "step": 486 + }, + { + "epoch": 0.5336986301369863, + "grad_norm": 2.001584053039551, + "learning_rate": 4.595857026807474e-06, + "loss": 2.214, + "step": 487 + }, + { + "epoch": 0.5347945205479452, + "grad_norm": 1.966700553894043, + "learning_rate": 4.593826157595451e-06, + "loss": 2.2237, + "step": 488 + }, + { + "epoch": 0.5358904109589041, + "grad_norm": 1.842469334602356, + "learning_rate": 4.591795288383429e-06, + "loss": 2.2089, + "step": 489 + }, + { + "epoch": 0.536986301369863, + "grad_norm": 2.0412986278533936, + "learning_rate": 4.589764419171406e-06, + "loss": 2.2734, + "step": 490 + }, + { + "epoch": 0.536986301369863, + "eval_loss": 2.3460772037506104, + "eval_runtime": 50.4048, + "eval_samples_per_second": 19.839, + "eval_steps_per_second": 0.833, + "step": 490 + }, + { + "epoch": 0.538082191780822, + "grad_norm": 1.9337031841278076, + "learning_rate": 4.587733549959383e-06, + "loss": 2.2314, + "step": 491 + }, + { + "epoch": 0.5391780821917809, + "grad_norm": 2.4448320865631104, + "learning_rate": 4.5857026807473604e-06, + "loss": 2.2098, + "step": 492 + }, + { + "epoch": 0.5402739726027397, + "grad_norm": 1.83162522315979, + "learning_rate": 4.583671811535338e-06, + "loss": 2.1902, + "step": 493 + }, + { + "epoch": 0.5413698630136986, + "grad_norm": 2.3614656925201416, + "learning_rate": 4.581640942323315e-06, + "loss": 2.1737, + "step": 494 + }, + { + "epoch": 0.5424657534246575, + "grad_norm": 2.001187801361084, + "learning_rate": 4.579610073111292e-06, + "loss": 2.2485, + "step": 495 + }, + { + "epoch": 0.5435616438356164, + "grad_norm": 2.3479933738708496, + "learning_rate": 4.5775792038992695e-06, + "loss": 2.2595, + "step": 496 + }, + { + "epoch": 0.5446575342465754, + "grad_norm": 2.135856866836548, + "learning_rate": 4.575548334687247e-06, + "loss": 2.2528, + "step": 497 + }, + { + "epoch": 0.5457534246575343, + "grad_norm": 1.983126163482666, + "learning_rate": 4.573517465475224e-06, + "loss": 2.2788, + "step": 498 + }, + { + "epoch": 0.5468493150684931, + "grad_norm": 2.18208909034729, + "learning_rate": 4.571486596263201e-06, + "loss": 2.1995, + "step": 499 + }, + { + "epoch": 0.547945205479452, + "grad_norm": 1.848747968673706, + "learning_rate": 4.569455727051178e-06, + "loss": 2.2403, + "step": 500 + }, + { + "epoch": 0.547945205479452, + "eval_loss": 2.344066619873047, + "eval_runtime": 51.1213, + "eval_samples_per_second": 19.561, + "eval_steps_per_second": 0.822, + "step": 500 + }, + { + "epoch": 0.549041095890411, + "grad_norm": 2.427860736846924, + "learning_rate": 4.567424857839156e-06, + "loss": 2.246, + "step": 501 + }, + { + "epoch": 0.5501369863013699, + "grad_norm": 1.913405418395996, + "learning_rate": 4.565393988627132e-06, + "loss": 2.2296, + "step": 502 + }, + { + "epoch": 0.5512328767123288, + "grad_norm": 2.0106029510498047, + "learning_rate": 4.5633631194151105e-06, + "loss": 2.2657, + "step": 503 + }, + { + "epoch": 0.5523287671232877, + "grad_norm": 2.2719595432281494, + "learning_rate": 4.561332250203087e-06, + "loss": 2.2341, + "step": 504 + }, + { + "epoch": 0.5534246575342465, + "grad_norm": 2.109786033630371, + "learning_rate": 4.559301380991065e-06, + "loss": 2.2181, + "step": 505 + }, + { + "epoch": 0.5545205479452054, + "grad_norm": 2.189779281616211, + "learning_rate": 4.5572705117790414e-06, + "loss": 2.1821, + "step": 506 + }, + { + "epoch": 0.5556164383561644, + "grad_norm": 2.0263590812683105, + "learning_rate": 4.5552396425670196e-06, + "loss": 2.2072, + "step": 507 + }, + { + "epoch": 0.5567123287671233, + "grad_norm": 2.215289354324341, + "learning_rate": 4.553208773354996e-06, + "loss": 2.2513, + "step": 508 + }, + { + "epoch": 0.5578082191780822, + "grad_norm": 2.3031351566314697, + "learning_rate": 4.551177904142974e-06, + "loss": 2.1988, + "step": 509 + }, + { + "epoch": 0.5589041095890411, + "grad_norm": 2.1830058097839355, + "learning_rate": 4.5491470349309505e-06, + "loss": 2.269, + "step": 510 + }, + { + "epoch": 0.5589041095890411, + "eval_loss": 2.342463731765747, + "eval_runtime": 43.1956, + "eval_samples_per_second": 23.151, + "eval_steps_per_second": 0.972, + "step": 510 + }, + { + "epoch": 0.56, + "grad_norm": 2.3210105895996094, + "learning_rate": 4.547116165718929e-06, + "loss": 2.2051, + "step": 511 + }, + { + "epoch": 0.5610958904109589, + "grad_norm": 1.9175361394882202, + "learning_rate": 4.545085296506905e-06, + "loss": 2.2304, + "step": 512 + }, + { + "epoch": 0.5621917808219178, + "grad_norm": 2.302947759628296, + "learning_rate": 4.543054427294882e-06, + "loss": 2.215, + "step": 513 + }, + { + "epoch": 0.5632876712328767, + "grad_norm": 1.8947396278381348, + "learning_rate": 4.54102355808286e-06, + "loss": 2.1926, + "step": 514 + }, + { + "epoch": 0.5643835616438356, + "grad_norm": 2.195268154144287, + "learning_rate": 4.538992688870837e-06, + "loss": 2.1937, + "step": 515 + }, + { + "epoch": 0.5654794520547946, + "grad_norm": 2.178874969482422, + "learning_rate": 4.536961819658814e-06, + "loss": 2.2248, + "step": 516 + }, + { + "epoch": 0.5665753424657535, + "grad_norm": 1.9856929779052734, + "learning_rate": 4.5349309504467915e-06, + "loss": 2.2077, + "step": 517 + }, + { + "epoch": 0.5676712328767123, + "grad_norm": 2.0686943531036377, + "learning_rate": 4.532900081234769e-06, + "loss": 2.2328, + "step": 518 + }, + { + "epoch": 0.5687671232876712, + "grad_norm": 1.9554919004440308, + "learning_rate": 4.530869212022746e-06, + "loss": 2.198, + "step": 519 + }, + { + "epoch": 0.5698630136986301, + "grad_norm": 1.9489670991897583, + "learning_rate": 4.528838342810723e-06, + "loss": 2.2287, + "step": 520 + }, + { + "epoch": 0.5698630136986301, + "eval_loss": 2.3415298461914062, + "eval_runtime": 48.7918, + "eval_samples_per_second": 20.495, + "eval_steps_per_second": 0.861, + "step": 520 + }, + { + "epoch": 0.570958904109589, + "grad_norm": 1.9868658781051636, + "learning_rate": 4.526807473598701e-06, + "loss": 2.2205, + "step": 521 + }, + { + "epoch": 0.572054794520548, + "grad_norm": 2.0965914726257324, + "learning_rate": 4.524776604386678e-06, + "loss": 2.2239, + "step": 522 + }, + { + "epoch": 0.5731506849315069, + "grad_norm": 2.041471481323242, + "learning_rate": 4.522745735174655e-06, + "loss": 2.2035, + "step": 523 + }, + { + "epoch": 0.5742465753424657, + "grad_norm": 1.885293960571289, + "learning_rate": 4.520714865962632e-06, + "loss": 2.2469, + "step": 524 + }, + { + "epoch": 0.5753424657534246, + "grad_norm": 1.9107296466827393, + "learning_rate": 4.51868399675061e-06, + "loss": 2.2386, + "step": 525 + }, + { + "epoch": 0.5764383561643835, + "grad_norm": 1.9945271015167236, + "learning_rate": 4.516653127538587e-06, + "loss": 2.235, + "step": 526 + }, + { + "epoch": 0.5775342465753425, + "grad_norm": 1.9787253141403198, + "learning_rate": 4.514622258326564e-06, + "loss": 2.2034, + "step": 527 + }, + { + "epoch": 0.5786301369863014, + "grad_norm": 2.0300636291503906, + "learning_rate": 4.5125913891145415e-06, + "loss": 2.1849, + "step": 528 + }, + { + "epoch": 0.5797260273972603, + "grad_norm": 2.0348970890045166, + "learning_rate": 4.510560519902519e-06, + "loss": 2.2174, + "step": 529 + }, + { + "epoch": 0.5808219178082191, + "grad_norm": 1.8589537143707275, + "learning_rate": 4.508529650690496e-06, + "loss": 2.2351, + "step": 530 + }, + { + "epoch": 0.5808219178082191, + "eval_loss": 2.3450186252593994, + "eval_runtime": 52.4548, + "eval_samples_per_second": 19.064, + "eval_steps_per_second": 0.801, + "step": 530 + }, + { + "epoch": 0.581917808219178, + "grad_norm": 2.0496020317077637, + "learning_rate": 4.506498781478473e-06, + "loss": 2.182, + "step": 531 + }, + { + "epoch": 0.583013698630137, + "grad_norm": 2.0649805068969727, + "learning_rate": 4.504467912266451e-06, + "loss": 2.1654, + "step": 532 + }, + { + "epoch": 0.5841095890410959, + "grad_norm": 1.9809486865997314, + "learning_rate": 4.502437043054428e-06, + "loss": 2.2182, + "step": 533 + }, + { + "epoch": 0.5852054794520548, + "grad_norm": 2.21915340423584, + "learning_rate": 4.500406173842405e-06, + "loss": 2.2697, + "step": 534 + }, + { + "epoch": 0.5863013698630137, + "grad_norm": 1.9822649955749512, + "learning_rate": 4.4983753046303824e-06, + "loss": 2.2297, + "step": 535 + }, + { + "epoch": 0.5873972602739727, + "grad_norm": 1.972575306892395, + "learning_rate": 4.49634443541836e-06, + "loss": 2.2164, + "step": 536 + }, + { + "epoch": 0.5884931506849315, + "grad_norm": 1.948667049407959, + "learning_rate": 4.494313566206336e-06, + "loss": 2.2087, + "step": 537 + }, + { + "epoch": 0.5895890410958904, + "grad_norm": 2.7486417293548584, + "learning_rate": 4.492282696994314e-06, + "loss": 2.2602, + "step": 538 + }, + { + "epoch": 0.5906849315068493, + "grad_norm": 2.3374552726745605, + "learning_rate": 4.490251827782291e-06, + "loss": 2.2054, + "step": 539 + }, + { + "epoch": 0.5917808219178082, + "grad_norm": 2.222116231918335, + "learning_rate": 4.488220958570269e-06, + "loss": 2.1868, + "step": 540 + }, + { + "epoch": 0.5917808219178082, + "eval_loss": 2.345618963241577, + "eval_runtime": 53.0302, + "eval_samples_per_second": 18.857, + "eval_steps_per_second": 0.792, + "step": 540 + }, + { + "epoch": 0.5928767123287672, + "grad_norm": 2.00752854347229, + "learning_rate": 4.486190089358245e-06, + "loss": 2.187, + "step": 541 + }, + { + "epoch": 0.5939726027397261, + "grad_norm": 2.2466132640838623, + "learning_rate": 4.484159220146223e-06, + "loss": 2.1611, + "step": 542 + }, + { + "epoch": 0.5950684931506849, + "grad_norm": 2.1279385089874268, + "learning_rate": 4.4821283509342e-06, + "loss": 2.248, + "step": 543 + }, + { + "epoch": 0.5961643835616438, + "grad_norm": 1.9140571355819702, + "learning_rate": 4.480097481722178e-06, + "loss": 2.2314, + "step": 544 + }, + { + "epoch": 0.5972602739726027, + "grad_norm": 2.120741605758667, + "learning_rate": 4.478066612510154e-06, + "loss": 2.212, + "step": 545 + }, + { + "epoch": 0.5983561643835617, + "grad_norm": 2.003713369369507, + "learning_rate": 4.4760357432981325e-06, + "loss": 2.1824, + "step": 546 + }, + { + "epoch": 0.5994520547945206, + "grad_norm": 1.9161221981048584, + "learning_rate": 4.474004874086109e-06, + "loss": 2.2015, + "step": 547 + }, + { + "epoch": 0.6005479452054795, + "grad_norm": 2.367110252380371, + "learning_rate": 4.471974004874087e-06, + "loss": 2.3122, + "step": 548 + }, + { + "epoch": 0.6016438356164383, + "grad_norm": 2.113380193710327, + "learning_rate": 4.4699431356620635e-06, + "loss": 2.1692, + "step": 549 + }, + { + "epoch": 0.6027397260273972, + "grad_norm": 1.974181890487671, + "learning_rate": 4.467912266450041e-06, + "loss": 2.2509, + "step": 550 + }, + { + "epoch": 0.6027397260273972, + "eval_loss": 2.342583656311035, + "eval_runtime": 50.7362, + "eval_samples_per_second": 19.71, + "eval_steps_per_second": 0.828, + "step": 550 + }, + { + "epoch": 0.6038356164383561, + "grad_norm": 1.889894962310791, + "learning_rate": 4.465881397238018e-06, + "loss": 2.2214, + "step": 551 + }, + { + "epoch": 0.6049315068493151, + "grad_norm": 1.8982343673706055, + "learning_rate": 4.463850528025995e-06, + "loss": 2.2135, + "step": 552 + }, + { + "epoch": 0.606027397260274, + "grad_norm": 1.9462087154388428, + "learning_rate": 4.4618196588139726e-06, + "loss": 2.2255, + "step": 553 + }, + { + "epoch": 0.6071232876712329, + "grad_norm": 1.8729931116104126, + "learning_rate": 4.45978878960195e-06, + "loss": 2.2358, + "step": 554 + }, + { + "epoch": 0.6082191780821918, + "grad_norm": 1.9230246543884277, + "learning_rate": 4.457757920389927e-06, + "loss": 2.222, + "step": 555 + }, + { + "epoch": 0.6093150684931506, + "grad_norm": 1.9495900869369507, + "learning_rate": 4.455727051177904e-06, + "loss": 2.1779, + "step": 556 + }, + { + "epoch": 0.6104109589041096, + "grad_norm": 1.983487606048584, + "learning_rate": 4.453696181965882e-06, + "loss": 2.2007, + "step": 557 + }, + { + "epoch": 0.6115068493150685, + "grad_norm": 2.1541049480438232, + "learning_rate": 4.451665312753859e-06, + "loss": 2.2513, + "step": 558 + }, + { + "epoch": 0.6126027397260274, + "grad_norm": 1.9133352041244507, + "learning_rate": 4.449634443541836e-06, + "loss": 2.2288, + "step": 559 + }, + { + "epoch": 0.6136986301369863, + "grad_norm": 2.3325741291046143, + "learning_rate": 4.4476035743298135e-06, + "loss": 2.2539, + "step": 560 + }, + { + "epoch": 0.6136986301369863, + "eval_loss": 2.3387348651885986, + "eval_runtime": 53.4301, + "eval_samples_per_second": 18.716, + "eval_steps_per_second": 0.786, + "step": 560 + }, + { + "epoch": 0.6147945205479453, + "grad_norm": 1.8099673986434937, + "learning_rate": 4.445572705117791e-06, + "loss": 2.1845, + "step": 561 + }, + { + "epoch": 0.6158904109589041, + "grad_norm": 2.201901912689209, + "learning_rate": 4.443541835905768e-06, + "loss": 2.195, + "step": 562 + }, + { + "epoch": 0.616986301369863, + "grad_norm": 1.8412714004516602, + "learning_rate": 4.441510966693745e-06, + "loss": 2.2551, + "step": 563 + }, + { + "epoch": 0.6180821917808219, + "grad_norm": 1.8761520385742188, + "learning_rate": 4.439480097481723e-06, + "loss": 2.1965, + "step": 564 + }, + { + "epoch": 0.6191780821917808, + "grad_norm": 1.9123467206954956, + "learning_rate": 4.4374492282697e-06, + "loss": 2.2502, + "step": 565 + }, + { + "epoch": 0.6202739726027398, + "grad_norm": 2.0631234645843506, + "learning_rate": 4.435418359057677e-06, + "loss": 2.1582, + "step": 566 + }, + { + "epoch": 0.6213698630136987, + "grad_norm": 2.3191616535186768, + "learning_rate": 4.433387489845654e-06, + "loss": 2.2271, + "step": 567 + }, + { + "epoch": 0.6224657534246575, + "grad_norm": 2.043057918548584, + "learning_rate": 4.431356620633632e-06, + "loss": 2.2292, + "step": 568 + }, + { + "epoch": 0.6235616438356164, + "grad_norm": 2.0716068744659424, + "learning_rate": 4.429325751421609e-06, + "loss": 2.196, + "step": 569 + }, + { + "epoch": 0.6246575342465753, + "grad_norm": 1.7669776678085327, + "learning_rate": 4.427294882209586e-06, + "loss": 2.1953, + "step": 570 + }, + { + "epoch": 0.6246575342465753, + "eval_loss": 2.336578607559204, + "eval_runtime": 50.0083, + "eval_samples_per_second": 19.997, + "eval_steps_per_second": 0.84, + "step": 570 + }, + { + "epoch": 0.6257534246575343, + "grad_norm": 1.8165771961212158, + "learning_rate": 4.4252640129975635e-06, + "loss": 2.2155, + "step": 571 + }, + { + "epoch": 0.6268493150684932, + "grad_norm": 1.8640249967575073, + "learning_rate": 4.423233143785541e-06, + "loss": 2.1648, + "step": 572 + }, + { + "epoch": 0.6279452054794521, + "grad_norm": 2.013869285583496, + "learning_rate": 4.421202274573518e-06, + "loss": 2.2131, + "step": 573 + }, + { + "epoch": 0.6290410958904109, + "grad_norm": 2.2650580406188965, + "learning_rate": 4.4191714053614945e-06, + "loss": 2.2287, + "step": 574 + }, + { + "epoch": 0.6301369863013698, + "grad_norm": 1.9232103824615479, + "learning_rate": 4.417140536149473e-06, + "loss": 2.1339, + "step": 575 + }, + { + "epoch": 0.6312328767123287, + "grad_norm": 1.910134196281433, + "learning_rate": 4.415109666937449e-06, + "loss": 2.1755, + "step": 576 + }, + { + "epoch": 0.6323287671232877, + "grad_norm": 2.2492005825042725, + "learning_rate": 4.413078797725427e-06, + "loss": 2.2362, + "step": 577 + }, + { + "epoch": 0.6334246575342466, + "grad_norm": 1.8584033250808716, + "learning_rate": 4.411047928513404e-06, + "loss": 2.1911, + "step": 578 + }, + { + "epoch": 0.6345205479452055, + "grad_norm": 2.156273126602173, + "learning_rate": 4.409017059301382e-06, + "loss": 2.2109, + "step": 579 + }, + { + "epoch": 0.6356164383561644, + "grad_norm": 1.7977535724639893, + "learning_rate": 4.406986190089358e-06, + "loss": 2.2195, + "step": 580 + }, + { + "epoch": 0.6356164383561644, + "eval_loss": 2.33555269241333, + "eval_runtime": 48.3919, + "eval_samples_per_second": 20.665, + "eval_steps_per_second": 0.868, + "step": 580 + }, + { + "epoch": 0.6367123287671232, + "grad_norm": 1.8262702226638794, + "learning_rate": 4.404955320877336e-06, + "loss": 2.2345, + "step": 581 + }, + { + "epoch": 0.6378082191780822, + "grad_norm": 1.9389383792877197, + "learning_rate": 4.402924451665313e-06, + "loss": 2.2347, + "step": 582 + }, + { + "epoch": 0.6389041095890411, + "grad_norm": 1.9548988342285156, + "learning_rate": 4.400893582453291e-06, + "loss": 2.1967, + "step": 583 + }, + { + "epoch": 0.64, + "grad_norm": 1.898092269897461, + "learning_rate": 4.398862713241267e-06, + "loss": 2.1795, + "step": 584 + }, + { + "epoch": 0.6410958904109589, + "grad_norm": 1.8183610439300537, + "learning_rate": 4.396831844029245e-06, + "loss": 2.2031, + "step": 585 + }, + { + "epoch": 0.6421917808219179, + "grad_norm": 2.1520016193389893, + "learning_rate": 4.394800974817222e-06, + "loss": 2.2051, + "step": 586 + }, + { + "epoch": 0.6432876712328767, + "grad_norm": 1.9930012226104736, + "learning_rate": 4.392770105605199e-06, + "loss": 2.2053, + "step": 587 + }, + { + "epoch": 0.6443835616438356, + "grad_norm": 2.311450958251953, + "learning_rate": 4.390739236393176e-06, + "loss": 2.2252, + "step": 588 + }, + { + "epoch": 0.6454794520547945, + "grad_norm": 2.0168697834014893, + "learning_rate": 4.388708367181154e-06, + "loss": 2.2286, + "step": 589 + }, + { + "epoch": 0.6465753424657534, + "grad_norm": 2.055964469909668, + "learning_rate": 4.386677497969131e-06, + "loss": 2.2278, + "step": 590 + }, + { + "epoch": 0.6465753424657534, + "eval_loss": 2.334367275238037, + "eval_runtime": 48.9414, + "eval_samples_per_second": 20.433, + "eval_steps_per_second": 0.858, + "step": 590 + }, + { + "epoch": 0.6476712328767124, + "grad_norm": 2.0161590576171875, + "learning_rate": 4.384646628757108e-06, + "loss": 2.2102, + "step": 591 + }, + { + "epoch": 0.6487671232876713, + "grad_norm": 2.0827667713165283, + "learning_rate": 4.3826157595450855e-06, + "loss": 2.2531, + "step": 592 + }, + { + "epoch": 0.6498630136986301, + "grad_norm": 2.077637195587158, + "learning_rate": 4.380584890333063e-06, + "loss": 2.2091, + "step": 593 + }, + { + "epoch": 0.650958904109589, + "grad_norm": 2.023892879486084, + "learning_rate": 4.37855402112104e-06, + "loss": 2.1684, + "step": 594 + }, + { + "epoch": 0.6520547945205479, + "grad_norm": 1.8509857654571533, + "learning_rate": 4.376523151909017e-06, + "loss": 2.1439, + "step": 595 + }, + { + "epoch": 0.6531506849315069, + "grad_norm": 2.1649608612060547, + "learning_rate": 4.3744922826969946e-06, + "loss": 2.2557, + "step": 596 + }, + { + "epoch": 0.6542465753424658, + "grad_norm": 2.1351773738861084, + "learning_rate": 4.372461413484972e-06, + "loss": 2.177, + "step": 597 + }, + { + "epoch": 0.6553424657534247, + "grad_norm": 2.1713974475860596, + "learning_rate": 4.370430544272949e-06, + "loss": 2.2223, + "step": 598 + }, + { + "epoch": 0.6564383561643835, + "grad_norm": 2.2905397415161133, + "learning_rate": 4.368399675060926e-06, + "loss": 2.2373, + "step": 599 + }, + { + "epoch": 0.6575342465753424, + "grad_norm": 2.007160186767578, + "learning_rate": 4.366368805848904e-06, + "loss": 2.2355, + "step": 600 + }, + { + "epoch": 0.6575342465753424, + "eval_loss": 2.3322324752807617, + "eval_runtime": 46.2769, + "eval_samples_per_second": 21.609, + "eval_steps_per_second": 0.908, + "step": 600 + }, + { + "epoch": 0.6586301369863014, + "grad_norm": 2.309783697128296, + "learning_rate": 4.364337936636881e-06, + "loss": 2.2052, + "step": 601 + }, + { + "epoch": 0.6597260273972603, + "grad_norm": 2.498034954071045, + "learning_rate": 4.362307067424858e-06, + "loss": 2.1567, + "step": 602 + }, + { + "epoch": 0.6608219178082192, + "grad_norm": 1.8302545547485352, + "learning_rate": 4.3602761982128355e-06, + "loss": 2.2117, + "step": 603 + }, + { + "epoch": 0.6619178082191781, + "grad_norm": 2.226947546005249, + "learning_rate": 4.358245329000813e-06, + "loss": 2.1881, + "step": 604 + }, + { + "epoch": 0.663013698630137, + "grad_norm": 1.8484126329421997, + "learning_rate": 4.35621445978879e-06, + "loss": 2.1955, + "step": 605 + }, + { + "epoch": 0.6641095890410958, + "grad_norm": 1.9844266176223755, + "learning_rate": 4.354183590576767e-06, + "loss": 2.2063, + "step": 606 + }, + { + "epoch": 0.6652054794520548, + "grad_norm": 1.9292365312576294, + "learning_rate": 4.352152721364745e-06, + "loss": 2.205, + "step": 607 + }, + { + "epoch": 0.6663013698630137, + "grad_norm": 1.980895757675171, + "learning_rate": 4.350121852152722e-06, + "loss": 2.1727, + "step": 608 + }, + { + "epoch": 0.6673972602739726, + "grad_norm": 1.8331713676452637, + "learning_rate": 4.348090982940699e-06, + "loss": 2.226, + "step": 609 + }, + { + "epoch": 0.6684931506849315, + "grad_norm": 1.9666537046432495, + "learning_rate": 4.3460601137286764e-06, + "loss": 2.2198, + "step": 610 + }, + { + "epoch": 0.6684931506849315, + "eval_loss": 2.3297338485717773, + "eval_runtime": 42.6274, + "eval_samples_per_second": 23.459, + "eval_steps_per_second": 0.985, + "step": 610 + }, + { + "epoch": 0.6695890410958905, + "grad_norm": 1.9741110801696777, + "learning_rate": 4.344029244516654e-06, + "loss": 2.2132, + "step": 611 + }, + { + "epoch": 0.6706849315068493, + "grad_norm": 2.0747334957122803, + "learning_rate": 4.341998375304631e-06, + "loss": 2.2081, + "step": 612 + }, + { + "epoch": 0.6717808219178082, + "grad_norm": 2.031503915786743, + "learning_rate": 4.339967506092607e-06, + "loss": 2.2642, + "step": 613 + }, + { + "epoch": 0.6728767123287671, + "grad_norm": 1.8513256311416626, + "learning_rate": 4.3379366368805855e-06, + "loss": 2.2141, + "step": 614 + }, + { + "epoch": 0.673972602739726, + "grad_norm": 1.9098315238952637, + "learning_rate": 4.335905767668562e-06, + "loss": 2.2064, + "step": 615 + }, + { + "epoch": 0.675068493150685, + "grad_norm": 1.9704065322875977, + "learning_rate": 4.33387489845654e-06, + "loss": 2.1997, + "step": 616 + }, + { + "epoch": 0.6761643835616439, + "grad_norm": 2.1520018577575684, + "learning_rate": 4.3318440292445165e-06, + "loss": 2.2278, + "step": 617 + }, + { + "epoch": 0.6772602739726027, + "grad_norm": 1.898164987564087, + "learning_rate": 4.329813160032495e-06, + "loss": 2.2147, + "step": 618 + }, + { + "epoch": 0.6783561643835616, + "grad_norm": 1.8955962657928467, + "learning_rate": 4.327782290820471e-06, + "loss": 2.1825, + "step": 619 + }, + { + "epoch": 0.6794520547945205, + "grad_norm": 1.7579313516616821, + "learning_rate": 4.325751421608449e-06, + "loss": 2.1619, + "step": 620 + }, + { + "epoch": 0.6794520547945205, + "eval_loss": 2.329692840576172, + "eval_runtime": 46.3187, + "eval_samples_per_second": 21.59, + "eval_steps_per_second": 0.907, + "step": 620 + }, + { + "epoch": 0.6805479452054795, + "grad_norm": 1.8716063499450684, + "learning_rate": 4.323720552396426e-06, + "loss": 2.2289, + "step": 621 + }, + { + "epoch": 0.6816438356164384, + "grad_norm": 1.8798290491104126, + "learning_rate": 4.321689683184404e-06, + "loss": 2.1944, + "step": 622 + }, + { + "epoch": 0.6827397260273973, + "grad_norm": 1.8438303470611572, + "learning_rate": 4.31965881397238e-06, + "loss": 2.1879, + "step": 623 + }, + { + "epoch": 0.6838356164383562, + "grad_norm": 1.7844786643981934, + "learning_rate": 4.3176279447603574e-06, + "loss": 2.1936, + "step": 624 + }, + { + "epoch": 0.684931506849315, + "grad_norm": 1.955529808998108, + "learning_rate": 4.315597075548335e-06, + "loss": 2.2258, + "step": 625 + }, + { + "epoch": 0.686027397260274, + "grad_norm": 2.3826634883880615, + "learning_rate": 4.313566206336312e-06, + "loss": 2.2135, + "step": 626 + }, + { + "epoch": 0.6871232876712329, + "grad_norm": 1.8453123569488525, + "learning_rate": 4.311535337124289e-06, + "loss": 2.2362, + "step": 627 + }, + { + "epoch": 0.6882191780821918, + "grad_norm": 2.742682933807373, + "learning_rate": 4.3095044679122665e-06, + "loss": 2.194, + "step": 628 + }, + { + "epoch": 0.6893150684931507, + "grad_norm": 2.0299901962280273, + "learning_rate": 4.307473598700244e-06, + "loss": 2.2029, + "step": 629 + }, + { + "epoch": 0.6904109589041096, + "grad_norm": 2.29683780670166, + "learning_rate": 4.305442729488221e-06, + "loss": 2.2141, + "step": 630 + }, + { + "epoch": 0.6904109589041096, + "eval_loss": 2.329368829727173, + "eval_runtime": 45.1947, + "eval_samples_per_second": 22.126, + "eval_steps_per_second": 0.929, + "step": 630 + }, + { + "epoch": 0.6915068493150685, + "grad_norm": 2.208280086517334, + "learning_rate": 4.303411860276198e-06, + "loss": 2.1637, + "step": 631 + }, + { + "epoch": 0.6926027397260274, + "grad_norm": 1.943514108657837, + "learning_rate": 4.301380991064176e-06, + "loss": 2.234, + "step": 632 + }, + { + "epoch": 0.6936986301369863, + "grad_norm": 2.486440420150757, + "learning_rate": 4.299350121852153e-06, + "loss": 2.2239, + "step": 633 + }, + { + "epoch": 0.6947945205479452, + "grad_norm": 1.7243690490722656, + "learning_rate": 4.29731925264013e-06, + "loss": 2.2428, + "step": 634 + }, + { + "epoch": 0.6958904109589041, + "grad_norm": 2.31227445602417, + "learning_rate": 4.2952883834281075e-06, + "loss": 2.2403, + "step": 635 + }, + { + "epoch": 0.6969863013698631, + "grad_norm": 1.7361263036727905, + "learning_rate": 4.293257514216085e-06, + "loss": 2.1975, + "step": 636 + }, + { + "epoch": 0.6980821917808219, + "grad_norm": 2.010831356048584, + "learning_rate": 4.291226645004062e-06, + "loss": 2.1633, + "step": 637 + }, + { + "epoch": 0.6991780821917808, + "grad_norm": 2.2796833515167236, + "learning_rate": 4.289195775792039e-06, + "loss": 2.2307, + "step": 638 + }, + { + "epoch": 0.7002739726027397, + "grad_norm": 2.1069483757019043, + "learning_rate": 4.287164906580017e-06, + "loss": 2.2389, + "step": 639 + }, + { + "epoch": 0.7013698630136986, + "grad_norm": 2.19460129737854, + "learning_rate": 4.285134037367994e-06, + "loss": 2.1992, + "step": 640 + }, + { + "epoch": 0.7013698630136986, + "eval_loss": 2.3223369121551514, + "eval_runtime": 43.1093, + "eval_samples_per_second": 23.197, + "eval_steps_per_second": 0.974, + "step": 640 + }, + { + "epoch": 0.7024657534246576, + "grad_norm": 1.875261902809143, + "learning_rate": 4.283103168155971e-06, + "loss": 2.1767, + "step": 641 + }, + { + "epoch": 0.7035616438356165, + "grad_norm": 1.906762957572937, + "learning_rate": 4.281072298943948e-06, + "loss": 2.1734, + "step": 642 + }, + { + "epoch": 0.7046575342465753, + "grad_norm": 2.1114180088043213, + "learning_rate": 4.279041429731926e-06, + "loss": 2.1734, + "step": 643 + }, + { + "epoch": 0.7057534246575342, + "grad_norm": 1.844373106956482, + "learning_rate": 4.277010560519903e-06, + "loss": 2.1967, + "step": 644 + }, + { + "epoch": 0.7068493150684931, + "grad_norm": 2.3463754653930664, + "learning_rate": 4.27497969130788e-06, + "loss": 2.1735, + "step": 645 + }, + { + "epoch": 0.707945205479452, + "grad_norm": 2.3530960083007812, + "learning_rate": 4.2729488220958575e-06, + "loss": 2.189, + "step": 646 + }, + { + "epoch": 0.709041095890411, + "grad_norm": 2.110963821411133, + "learning_rate": 4.270917952883835e-06, + "loss": 2.218, + "step": 647 + }, + { + "epoch": 0.7101369863013699, + "grad_norm": 2.625005006790161, + "learning_rate": 4.268887083671812e-06, + "loss": 2.2235, + "step": 648 + }, + { + "epoch": 0.7112328767123288, + "grad_norm": 1.8411654233932495, + "learning_rate": 4.266856214459789e-06, + "loss": 2.1757, + "step": 649 + }, + { + "epoch": 0.7123287671232876, + "grad_norm": 2.0817489624023438, + "learning_rate": 4.264825345247766e-06, + "loss": 2.2698, + "step": 650 + }, + { + "epoch": 0.7123287671232876, + "eval_loss": 2.322544574737549, + "eval_runtime": 43.0771, + "eval_samples_per_second": 23.214, + "eval_steps_per_second": 0.975, + "step": 650 + }, + { + "epoch": 0.7134246575342466, + "grad_norm": 1.8953847885131836, + "learning_rate": 4.262794476035744e-06, + "loss": 2.1745, + "step": 651 + }, + { + "epoch": 0.7145205479452055, + "grad_norm": 1.8463294506072998, + "learning_rate": 4.26076360682372e-06, + "loss": 2.2042, + "step": 652 + }, + { + "epoch": 0.7156164383561644, + "grad_norm": 1.8776047229766846, + "learning_rate": 4.2587327376116984e-06, + "loss": 2.1512, + "step": 653 + }, + { + "epoch": 0.7167123287671233, + "grad_norm": 1.8949921131134033, + "learning_rate": 4.256701868399675e-06, + "loss": 2.1895, + "step": 654 + }, + { + "epoch": 0.7178082191780822, + "grad_norm": 1.8172632455825806, + "learning_rate": 4.254670999187653e-06, + "loss": 2.2, + "step": 655 + }, + { + "epoch": 0.718904109589041, + "grad_norm": 2.105499505996704, + "learning_rate": 4.252640129975629e-06, + "loss": 2.202, + "step": 656 + }, + { + "epoch": 0.72, + "grad_norm": 1.9515924453735352, + "learning_rate": 4.2506092607636075e-06, + "loss": 2.2116, + "step": 657 + }, + { + "epoch": 0.7210958904109589, + "grad_norm": 1.761770248413086, + "learning_rate": 4.248578391551584e-06, + "loss": 2.1921, + "step": 658 + }, + { + "epoch": 0.7221917808219178, + "grad_norm": 1.9848097562789917, + "learning_rate": 4.246547522339562e-06, + "loss": 2.1553, + "step": 659 + }, + { + "epoch": 0.7232876712328767, + "grad_norm": 2.041987180709839, + "learning_rate": 4.2445166531275385e-06, + "loss": 2.1975, + "step": 660 + }, + { + "epoch": 0.7232876712328767, + "eval_loss": 2.325580358505249, + "eval_runtime": 43.6207, + "eval_samples_per_second": 22.925, + "eval_steps_per_second": 0.963, + "step": 660 + }, + { + "epoch": 0.7243835616438357, + "grad_norm": 1.9755775928497314, + "learning_rate": 4.242485783915517e-06, + "loss": 2.1955, + "step": 661 + }, + { + "epoch": 0.7254794520547945, + "grad_norm": 1.9621502161026, + "learning_rate": 4.240454914703493e-06, + "loss": 2.1554, + "step": 662 + }, + { + "epoch": 0.7265753424657534, + "grad_norm": 1.848013162612915, + "learning_rate": 4.23842404549147e-06, + "loss": 2.16, + "step": 663 + }, + { + "epoch": 0.7276712328767123, + "grad_norm": 1.7370128631591797, + "learning_rate": 4.236393176279448e-06, + "loss": 2.1696, + "step": 664 + }, + { + "epoch": 0.7287671232876712, + "grad_norm": 2.0708158016204834, + "learning_rate": 4.234362307067425e-06, + "loss": 2.1592, + "step": 665 + }, + { + "epoch": 0.7298630136986302, + "grad_norm": 1.9864391088485718, + "learning_rate": 4.232331437855402e-06, + "loss": 2.2121, + "step": 666 + }, + { + "epoch": 0.7309589041095891, + "grad_norm": 1.9988250732421875, + "learning_rate": 4.2303005686433795e-06, + "loss": 2.2306, + "step": 667 + }, + { + "epoch": 0.7320547945205479, + "grad_norm": 1.89108407497406, + "learning_rate": 4.228269699431357e-06, + "loss": 2.222, + "step": 668 + }, + { + "epoch": 0.7331506849315068, + "grad_norm": 1.7941503524780273, + "learning_rate": 4.226238830219334e-06, + "loss": 2.1975, + "step": 669 + }, + { + "epoch": 0.7342465753424657, + "grad_norm": 1.8058613538742065, + "learning_rate": 4.224207961007311e-06, + "loss": 2.2037, + "step": 670 + }, + { + "epoch": 0.7342465753424657, + "eval_loss": 2.3266475200653076, + "eval_runtime": 47.6623, + "eval_samples_per_second": 20.981, + "eval_steps_per_second": 0.881, + "step": 670 + }, + { + "epoch": 0.7353424657534247, + "grad_norm": 1.764448642730713, + "learning_rate": 4.2221770917952886e-06, + "loss": 2.1904, + "step": 671 + }, + { + "epoch": 0.7364383561643836, + "grad_norm": 1.8252112865447998, + "learning_rate": 4.220146222583266e-06, + "loss": 2.1959, + "step": 672 + }, + { + "epoch": 0.7375342465753425, + "grad_norm": 2.02898907661438, + "learning_rate": 4.218115353371243e-06, + "loss": 2.1937, + "step": 673 + }, + { + "epoch": 0.7386301369863014, + "grad_norm": 1.8766248226165771, + "learning_rate": 4.21608448415922e-06, + "loss": 2.195, + "step": 674 + }, + { + "epoch": 0.7397260273972602, + "grad_norm": 2.2350332736968994, + "learning_rate": 4.214053614947198e-06, + "loss": 2.1798, + "step": 675 + }, + { + "epoch": 0.7408219178082192, + "grad_norm": 1.820399522781372, + "learning_rate": 4.212022745735175e-06, + "loss": 2.2371, + "step": 676 + }, + { + "epoch": 0.7419178082191781, + "grad_norm": 2.3411312103271484, + "learning_rate": 4.209991876523152e-06, + "loss": 2.1603, + "step": 677 + }, + { + "epoch": 0.743013698630137, + "grad_norm": 1.8562874794006348, + "learning_rate": 4.2079610073111295e-06, + "loss": 2.2082, + "step": 678 + }, + { + "epoch": 0.7441095890410959, + "grad_norm": 2.274214506149292, + "learning_rate": 4.205930138099107e-06, + "loss": 2.1716, + "step": 679 + }, + { + "epoch": 0.7452054794520548, + "grad_norm": 1.9173530340194702, + "learning_rate": 4.203899268887084e-06, + "loss": 2.2461, + "step": 680 + }, + { + "epoch": 0.7452054794520548, + "eval_loss": 2.3253841400146484, + "eval_runtime": 46.5732, + "eval_samples_per_second": 21.472, + "eval_steps_per_second": 0.902, + "step": 680 + }, + { + "epoch": 0.7463013698630137, + "grad_norm": 1.9732033014297485, + "learning_rate": 4.201868399675061e-06, + "loss": 2.2016, + "step": 681 + }, + { + "epoch": 0.7473972602739726, + "grad_norm": 1.8850247859954834, + "learning_rate": 4.199837530463039e-06, + "loss": 2.223, + "step": 682 + }, + { + "epoch": 0.7484931506849315, + "grad_norm": 1.8936313390731812, + "learning_rate": 4.197806661251016e-06, + "loss": 2.1381, + "step": 683 + }, + { + "epoch": 0.7495890410958904, + "grad_norm": 2.11995530128479, + "learning_rate": 4.195775792038993e-06, + "loss": 2.1962, + "step": 684 + }, + { + "epoch": 0.7506849315068493, + "grad_norm": 1.8207820653915405, + "learning_rate": 4.19374492282697e-06, + "loss": 2.1835, + "step": 685 + }, + { + "epoch": 0.7517808219178083, + "grad_norm": 2.06030535697937, + "learning_rate": 4.191714053614948e-06, + "loss": 2.1347, + "step": 686 + }, + { + "epoch": 0.7528767123287671, + "grad_norm": 2.048990488052368, + "learning_rate": 4.189683184402924e-06, + "loss": 2.2189, + "step": 687 + }, + { + "epoch": 0.753972602739726, + "grad_norm": 1.976369023323059, + "learning_rate": 4.187652315190902e-06, + "loss": 2.1805, + "step": 688 + }, + { + "epoch": 0.7550684931506849, + "grad_norm": 2.1160926818847656, + "learning_rate": 4.185621445978879e-06, + "loss": 2.2322, + "step": 689 + }, + { + "epoch": 0.7561643835616438, + "grad_norm": 1.9444516897201538, + "learning_rate": 4.183590576766857e-06, + "loss": 2.1696, + "step": 690 + }, + { + "epoch": 0.7561643835616438, + "eval_loss": 2.3242011070251465, + "eval_runtime": 48.7528, + "eval_samples_per_second": 20.512, + "eval_steps_per_second": 0.861, + "step": 690 + }, + { + "epoch": 0.7572602739726028, + "grad_norm": 1.9567183256149292, + "learning_rate": 4.181559707554833e-06, + "loss": 2.1891, + "step": 691 + }, + { + "epoch": 0.7583561643835617, + "grad_norm": 1.9185020923614502, + "learning_rate": 4.179528838342811e-06, + "loss": 2.2318, + "step": 692 + }, + { + "epoch": 0.7594520547945205, + "grad_norm": 1.833487629890442, + "learning_rate": 4.177497969130788e-06, + "loss": 2.1705, + "step": 693 + }, + { + "epoch": 0.7605479452054794, + "grad_norm": 2.2949867248535156, + "learning_rate": 4.175467099918766e-06, + "loss": 2.1712, + "step": 694 + }, + { + "epoch": 0.7616438356164383, + "grad_norm": 1.8663084506988525, + "learning_rate": 4.173436230706742e-06, + "loss": 2.2262, + "step": 695 + }, + { + "epoch": 0.7627397260273973, + "grad_norm": 2.4633052349090576, + "learning_rate": 4.1714053614947205e-06, + "loss": 2.185, + "step": 696 + }, + { + "epoch": 0.7638356164383562, + "grad_norm": 2.036524772644043, + "learning_rate": 4.169374492282697e-06, + "loss": 2.1697, + "step": 697 + }, + { + "epoch": 0.7649315068493151, + "grad_norm": 2.013810634613037, + "learning_rate": 4.167343623070675e-06, + "loss": 2.1826, + "step": 698 + }, + { + "epoch": 0.766027397260274, + "grad_norm": 2.214109182357788, + "learning_rate": 4.165312753858652e-06, + "loss": 2.2162, + "step": 699 + }, + { + "epoch": 0.7671232876712328, + "grad_norm": 1.8463964462280273, + "learning_rate": 4.163281884646629e-06, + "loss": 2.1865, + "step": 700 + }, + { + "epoch": 0.7671232876712328, + "eval_loss": 2.323822498321533, + "eval_runtime": 48.3307, + "eval_samples_per_second": 20.691, + "eval_steps_per_second": 0.869, + "step": 700 + }, + { + "epoch": 0.7682191780821918, + "grad_norm": 2.321021556854248, + "learning_rate": 4.161251015434607e-06, + "loss": 2.1758, + "step": 701 + }, + { + "epoch": 0.7693150684931507, + "grad_norm": 1.8213255405426025, + "learning_rate": 4.159220146222583e-06, + "loss": 2.2002, + "step": 702 + }, + { + "epoch": 0.7704109589041096, + "grad_norm": 1.8925111293792725, + "learning_rate": 4.157189277010561e-06, + "loss": 2.1575, + "step": 703 + }, + { + "epoch": 0.7715068493150685, + "grad_norm": 1.8050175905227661, + "learning_rate": 4.155158407798538e-06, + "loss": 2.179, + "step": 704 + }, + { + "epoch": 0.7726027397260274, + "grad_norm": 1.873228907585144, + "learning_rate": 4.153127538586516e-06, + "loss": 2.2232, + "step": 705 + }, + { + "epoch": 0.7736986301369863, + "grad_norm": 1.954910397529602, + "learning_rate": 4.151096669374492e-06, + "loss": 2.1992, + "step": 706 + }, + { + "epoch": 0.7747945205479452, + "grad_norm": 1.8913410902023315, + "learning_rate": 4.1490658001624705e-06, + "loss": 2.2086, + "step": 707 + }, + { + "epoch": 0.7758904109589041, + "grad_norm": 1.8883898258209229, + "learning_rate": 4.147034930950447e-06, + "loss": 2.1801, + "step": 708 + }, + { + "epoch": 0.776986301369863, + "grad_norm": 2.054351806640625, + "learning_rate": 4.145004061738425e-06, + "loss": 2.1908, + "step": 709 + }, + { + "epoch": 0.7780821917808219, + "grad_norm": 2.029036045074463, + "learning_rate": 4.1429731925264015e-06, + "loss": 2.1891, + "step": 710 + }, + { + "epoch": 0.7780821917808219, + "eval_loss": 2.3196680545806885, + "eval_runtime": 42.8326, + "eval_samples_per_second": 23.347, + "eval_steps_per_second": 0.981, + "step": 710 + }, + { + "epoch": 0.7791780821917809, + "grad_norm": 2.27848219871521, + "learning_rate": 4.140942323314379e-06, + "loss": 2.1165, + "step": 711 + }, + { + "epoch": 0.7802739726027397, + "grad_norm": 2.134625196456909, + "learning_rate": 4.138911454102356e-06, + "loss": 2.1508, + "step": 712 + }, + { + "epoch": 0.7813698630136986, + "grad_norm": 2.103895902633667, + "learning_rate": 4.136880584890333e-06, + "loss": 2.1746, + "step": 713 + }, + { + "epoch": 0.7824657534246575, + "grad_norm": 1.713189721107483, + "learning_rate": 4.1348497156783106e-06, + "loss": 2.2023, + "step": 714 + }, + { + "epoch": 0.7835616438356164, + "grad_norm": 2.0390899181365967, + "learning_rate": 4.132818846466288e-06, + "loss": 2.2039, + "step": 715 + }, + { + "epoch": 0.7846575342465754, + "grad_norm": 1.9643651247024536, + "learning_rate": 4.130787977254265e-06, + "loss": 2.199, + "step": 716 + }, + { + "epoch": 0.7857534246575343, + "grad_norm": 1.9173564910888672, + "learning_rate": 4.128757108042242e-06, + "loss": 2.2166, + "step": 717 + }, + { + "epoch": 0.7868493150684932, + "grad_norm": 1.7923007011413574, + "learning_rate": 4.12672623883022e-06, + "loss": 2.2144, + "step": 718 + }, + { + "epoch": 0.787945205479452, + "grad_norm": 2.2235090732574463, + "learning_rate": 4.124695369618197e-06, + "loss": 2.2586, + "step": 719 + }, + { + "epoch": 0.7890410958904109, + "grad_norm": 1.780316710472107, + "learning_rate": 4.122664500406174e-06, + "loss": 2.1862, + "step": 720 + }, + { + "epoch": 0.7890410958904109, + "eval_loss": 2.318016767501831, + "eval_runtime": 50.5082, + "eval_samples_per_second": 19.799, + "eval_steps_per_second": 0.832, + "step": 720 + }, + { + "epoch": 0.7901369863013699, + "grad_norm": 2.0788493156433105, + "learning_rate": 4.1206336311941515e-06, + "loss": 2.1845, + "step": 721 + }, + { + "epoch": 0.7912328767123288, + "grad_norm": 1.9053714275360107, + "learning_rate": 4.118602761982129e-06, + "loss": 2.2005, + "step": 722 + }, + { + "epoch": 0.7923287671232877, + "grad_norm": 1.9328513145446777, + "learning_rate": 4.116571892770106e-06, + "loss": 2.1475, + "step": 723 + }, + { + "epoch": 0.7934246575342466, + "grad_norm": 2.059760570526123, + "learning_rate": 4.114541023558083e-06, + "loss": 2.1386, + "step": 724 + }, + { + "epoch": 0.7945205479452054, + "grad_norm": 2.0093464851379395, + "learning_rate": 4.112510154346061e-06, + "loss": 2.2242, + "step": 725 + }, + { + "epoch": 0.7956164383561644, + "grad_norm": 1.8513909578323364, + "learning_rate": 4.110479285134038e-06, + "loss": 2.178, + "step": 726 + }, + { + "epoch": 0.7967123287671233, + "grad_norm": 1.9481958150863647, + "learning_rate": 4.108448415922015e-06, + "loss": 2.1961, + "step": 727 + }, + { + "epoch": 0.7978082191780822, + "grad_norm": 1.9229767322540283, + "learning_rate": 4.1064175467099924e-06, + "loss": 2.2116, + "step": 728 + }, + { + "epoch": 0.7989041095890411, + "grad_norm": 1.9819737672805786, + "learning_rate": 4.10438667749797e-06, + "loss": 2.2097, + "step": 729 + }, + { + "epoch": 0.8, + "grad_norm": 1.9382416009902954, + "learning_rate": 4.102355808285947e-06, + "loss": 2.1099, + "step": 730 + }, + { + "epoch": 0.8, + "eval_loss": 2.31795334815979, + "eval_runtime": 51.8961, + "eval_samples_per_second": 19.269, + "eval_steps_per_second": 0.809, + "step": 730 + }, + { + "epoch": 0.8010958904109589, + "grad_norm": 2.006890058517456, + "learning_rate": 4.100324939073924e-06, + "loss": 2.2606, + "step": 731 + }, + { + "epoch": 0.8021917808219178, + "grad_norm": 2.164473295211792, + "learning_rate": 4.0982940698619015e-06, + "loss": 2.1869, + "step": 732 + }, + { + "epoch": 0.8032876712328767, + "grad_norm": 2.04368257522583, + "learning_rate": 4.096263200649879e-06, + "loss": 2.1846, + "step": 733 + }, + { + "epoch": 0.8043835616438356, + "grad_norm": 1.8363912105560303, + "learning_rate": 4.094232331437856e-06, + "loss": 2.1938, + "step": 734 + }, + { + "epoch": 0.8054794520547945, + "grad_norm": 1.9260189533233643, + "learning_rate": 4.092201462225833e-06, + "loss": 2.1682, + "step": 735 + }, + { + "epoch": 0.8065753424657535, + "grad_norm": 2.0688912868499756, + "learning_rate": 4.090170593013811e-06, + "loss": 2.1772, + "step": 736 + }, + { + "epoch": 0.8076712328767123, + "grad_norm": 1.859474539756775, + "learning_rate": 4.088139723801787e-06, + "loss": 2.2013, + "step": 737 + }, + { + "epoch": 0.8087671232876712, + "grad_norm": 2.0915753841400146, + "learning_rate": 4.086108854589765e-06, + "loss": 2.1519, + "step": 738 + }, + { + "epoch": 0.8098630136986301, + "grad_norm": 2.2019519805908203, + "learning_rate": 4.084077985377742e-06, + "loss": 2.1608, + "step": 739 + }, + { + "epoch": 0.810958904109589, + "grad_norm": 1.8727316856384277, + "learning_rate": 4.08204711616572e-06, + "loss": 2.1823, + "step": 740 + }, + { + "epoch": 0.810958904109589, + "eval_loss": 2.318638324737549, + "eval_runtime": 52.3669, + "eval_samples_per_second": 19.096, + "eval_steps_per_second": 0.802, + "step": 740 + }, + { + "epoch": 0.812054794520548, + "grad_norm": 2.059250831604004, + "learning_rate": 4.080016246953696e-06, + "loss": 2.2285, + "step": 741 + }, + { + "epoch": 0.8131506849315069, + "grad_norm": 1.9466749429702759, + "learning_rate": 4.077985377741674e-06, + "loss": 2.1496, + "step": 742 + }, + { + "epoch": 0.8142465753424658, + "grad_norm": 1.8967524766921997, + "learning_rate": 4.075954508529651e-06, + "loss": 2.2066, + "step": 743 + }, + { + "epoch": 0.8153424657534246, + "grad_norm": 1.976780891418457, + "learning_rate": 4.073923639317629e-06, + "loss": 2.2194, + "step": 744 + }, + { + "epoch": 0.8164383561643835, + "grad_norm": 2.077683687210083, + "learning_rate": 4.071892770105605e-06, + "loss": 2.2209, + "step": 745 + }, + { + "epoch": 0.8175342465753425, + "grad_norm": 1.9075323343276978, + "learning_rate": 4.069861900893583e-06, + "loss": 2.1727, + "step": 746 + }, + { + "epoch": 0.8186301369863014, + "grad_norm": 1.6940832138061523, + "learning_rate": 4.06783103168156e-06, + "loss": 2.1437, + "step": 747 + }, + { + "epoch": 0.8197260273972603, + "grad_norm": 1.9035768508911133, + "learning_rate": 4.065800162469538e-06, + "loss": 2.165, + "step": 748 + }, + { + "epoch": 0.8208219178082192, + "grad_norm": 1.8775745630264282, + "learning_rate": 4.063769293257514e-06, + "loss": 2.1757, + "step": 749 + }, + { + "epoch": 0.821917808219178, + "grad_norm": 1.9622418880462646, + "learning_rate": 4.061738424045492e-06, + "loss": 2.197, + "step": 750 + }, + { + "epoch": 0.821917808219178, + "eval_loss": 2.317519426345825, + "eval_runtime": 50.2699, + "eval_samples_per_second": 19.893, + "eval_steps_per_second": 0.835, + "step": 750 + }, + { + "epoch": 0.823013698630137, + "grad_norm": 1.9646214246749878, + "learning_rate": 4.059707554833469e-06, + "loss": 2.1783, + "step": 751 + }, + { + "epoch": 0.8241095890410959, + "grad_norm": 1.890342354774475, + "learning_rate": 4.057676685621446e-06, + "loss": 2.1925, + "step": 752 + }, + { + "epoch": 0.8252054794520548, + "grad_norm": 1.8977251052856445, + "learning_rate": 4.0556458164094235e-06, + "loss": 2.219, + "step": 753 + }, + { + "epoch": 0.8263013698630137, + "grad_norm": 1.8251665830612183, + "learning_rate": 4.053614947197401e-06, + "loss": 2.189, + "step": 754 + }, + { + "epoch": 0.8273972602739726, + "grad_norm": 1.7953882217407227, + "learning_rate": 4.051584077985378e-06, + "loss": 2.1237, + "step": 755 + }, + { + "epoch": 0.8284931506849315, + "grad_norm": 1.8247836828231812, + "learning_rate": 4.049553208773355e-06, + "loss": 2.1962, + "step": 756 + }, + { + "epoch": 0.8295890410958904, + "grad_norm": 1.8537967205047607, + "learning_rate": 4.047522339561333e-06, + "loss": 2.2201, + "step": 757 + }, + { + "epoch": 0.8306849315068493, + "grad_norm": 1.865315318107605, + "learning_rate": 4.04549147034931e-06, + "loss": 2.1819, + "step": 758 + }, + { + "epoch": 0.8317808219178082, + "grad_norm": 1.8620522022247314, + "learning_rate": 4.043460601137287e-06, + "loss": 2.1717, + "step": 759 + }, + { + "epoch": 0.8328767123287671, + "grad_norm": 1.9475384950637817, + "learning_rate": 4.041429731925264e-06, + "loss": 2.2718, + "step": 760 + }, + { + "epoch": 0.8328767123287671, + "eval_loss": 2.314429759979248, + "eval_runtime": 54.2324, + "eval_samples_per_second": 18.439, + "eval_steps_per_second": 0.774, + "step": 760 + }, + { + "epoch": 0.8339726027397261, + "grad_norm": 1.7170753479003906, + "learning_rate": 4.039398862713242e-06, + "loss": 2.1785, + "step": 761 + }, + { + "epoch": 0.8350684931506849, + "grad_norm": 1.968220591545105, + "learning_rate": 4.037367993501219e-06, + "loss": 2.1798, + "step": 762 + }, + { + "epoch": 0.8361643835616438, + "grad_norm": 1.7927111387252808, + "learning_rate": 4.035337124289196e-06, + "loss": 2.2372, + "step": 763 + }, + { + "epoch": 0.8372602739726027, + "grad_norm": 1.7871185541152954, + "learning_rate": 4.0333062550771735e-06, + "loss": 2.182, + "step": 764 + }, + { + "epoch": 0.8383561643835616, + "grad_norm": 1.9415812492370605, + "learning_rate": 4.031275385865151e-06, + "loss": 2.1527, + "step": 765 + }, + { + "epoch": 0.8394520547945206, + "grad_norm": 1.7327690124511719, + "learning_rate": 4.029244516653128e-06, + "loss": 2.1744, + "step": 766 + }, + { + "epoch": 0.8405479452054795, + "grad_norm": 1.9123411178588867, + "learning_rate": 4.027213647441105e-06, + "loss": 2.2049, + "step": 767 + }, + { + "epoch": 0.8416438356164384, + "grad_norm": 2.011591911315918, + "learning_rate": 4.025182778229083e-06, + "loss": 2.1391, + "step": 768 + }, + { + "epoch": 0.8427397260273972, + "grad_norm": 1.9169542789459229, + "learning_rate": 4.02315190901706e-06, + "loss": 2.1562, + "step": 769 + }, + { + "epoch": 0.8438356164383561, + "grad_norm": 1.8020051717758179, + "learning_rate": 4.021121039805037e-06, + "loss": 2.1795, + "step": 770 + }, + { + "epoch": 0.8438356164383561, + "eval_loss": 2.3199150562286377, + "eval_runtime": 52.468, + "eval_samples_per_second": 19.059, + "eval_steps_per_second": 0.8, + "step": 770 + }, + { + "epoch": 0.8449315068493151, + "grad_norm": 1.927666425704956, + "learning_rate": 4.0190901705930144e-06, + "loss": 2.1834, + "step": 771 + }, + { + "epoch": 0.846027397260274, + "grad_norm": 2.0998382568359375, + "learning_rate": 4.017059301380992e-06, + "loss": 2.2541, + "step": 772 + }, + { + "epoch": 0.8471232876712329, + "grad_norm": 1.88541579246521, + "learning_rate": 4.015028432168969e-06, + "loss": 2.1631, + "step": 773 + }, + { + "epoch": 0.8482191780821918, + "grad_norm": 1.9546253681182861, + "learning_rate": 4.012997562956945e-06, + "loss": 2.2266, + "step": 774 + }, + { + "epoch": 0.8493150684931506, + "grad_norm": 1.7969225645065308, + "learning_rate": 4.0109666937449235e-06, + "loss": 2.1526, + "step": 775 + }, + { + "epoch": 0.8504109589041096, + "grad_norm": 1.7237955331802368, + "learning_rate": 4.0089358245329e-06, + "loss": 2.1618, + "step": 776 + }, + { + "epoch": 0.8515068493150685, + "grad_norm": 1.8355857133865356, + "learning_rate": 4.006904955320878e-06, + "loss": 2.1673, + "step": 777 + }, + { + "epoch": 0.8526027397260274, + "grad_norm": 1.8049591779708862, + "learning_rate": 4.0048740861088545e-06, + "loss": 2.1676, + "step": 778 + }, + { + "epoch": 0.8536986301369863, + "grad_norm": 1.8143601417541504, + "learning_rate": 4.002843216896833e-06, + "loss": 2.214, + "step": 779 + }, + { + "epoch": 0.8547945205479452, + "grad_norm": 1.8344221115112305, + "learning_rate": 4.000812347684809e-06, + "loss": 2.1958, + "step": 780 + }, + { + "epoch": 0.8547945205479452, + "eval_loss": 2.314359426498413, + "eval_runtime": 51.8717, + "eval_samples_per_second": 19.278, + "eval_steps_per_second": 0.81, + "step": 780 + }, + { + "epoch": 0.8558904109589041, + "grad_norm": 1.944402813911438, + "learning_rate": 3.998781478472787e-06, + "loss": 2.1497, + "step": 781 + }, + { + "epoch": 0.856986301369863, + "grad_norm": 1.93898606300354, + "learning_rate": 3.996750609260764e-06, + "loss": 2.1737, + "step": 782 + }, + { + "epoch": 0.8580821917808219, + "grad_norm": 1.9499506950378418, + "learning_rate": 3.994719740048742e-06, + "loss": 2.1545, + "step": 783 + }, + { + "epoch": 0.8591780821917808, + "grad_norm": 1.9311052560806274, + "learning_rate": 3.992688870836718e-06, + "loss": 2.2559, + "step": 784 + }, + { + "epoch": 0.8602739726027397, + "grad_norm": 1.864862084388733, + "learning_rate": 3.990658001624696e-06, + "loss": 2.1996, + "step": 785 + }, + { + "epoch": 0.8613698630136987, + "grad_norm": 1.8248569965362549, + "learning_rate": 3.988627132412673e-06, + "loss": 2.1591, + "step": 786 + }, + { + "epoch": 0.8624657534246576, + "grad_norm": 2.2370970249176025, + "learning_rate": 3.98659626320065e-06, + "loss": 2.2095, + "step": 787 + }, + { + "epoch": 0.8635616438356164, + "grad_norm": 1.8877084255218506, + "learning_rate": 3.984565393988627e-06, + "loss": 2.1813, + "step": 788 + }, + { + "epoch": 0.8646575342465753, + "grad_norm": 2.1722517013549805, + "learning_rate": 3.9825345247766046e-06, + "loss": 2.1866, + "step": 789 + }, + { + "epoch": 0.8657534246575342, + "grad_norm": 1.9640655517578125, + "learning_rate": 3.980503655564582e-06, + "loss": 2.1818, + "step": 790 + }, + { + "epoch": 0.8657534246575342, + "eval_loss": 2.314087390899658, + "eval_runtime": 51.9778, + "eval_samples_per_second": 19.239, + "eval_steps_per_second": 0.808, + "step": 790 + }, + { + "epoch": 0.8668493150684932, + "grad_norm": 2.0583789348602295, + "learning_rate": 3.978472786352559e-06, + "loss": 2.1956, + "step": 791 + }, + { + "epoch": 0.8679452054794521, + "grad_norm": 2.2128682136535645, + "learning_rate": 3.976441917140536e-06, + "loss": 2.204, + "step": 792 + }, + { + "epoch": 0.869041095890411, + "grad_norm": 1.917763113975525, + "learning_rate": 3.974411047928514e-06, + "loss": 2.1841, + "step": 793 + }, + { + "epoch": 0.8701369863013698, + "grad_norm": 2.0327019691467285, + "learning_rate": 3.972380178716491e-06, + "loss": 2.2351, + "step": 794 + }, + { + "epoch": 0.8712328767123287, + "grad_norm": 1.9498991966247559, + "learning_rate": 3.970349309504468e-06, + "loss": 2.2074, + "step": 795 + }, + { + "epoch": 0.8723287671232877, + "grad_norm": 1.8105775117874146, + "learning_rate": 3.9683184402924455e-06, + "loss": 2.2088, + "step": 796 + }, + { + "epoch": 0.8734246575342466, + "grad_norm": 1.9299930334091187, + "learning_rate": 3.966287571080423e-06, + "loss": 2.1469, + "step": 797 + }, + { + "epoch": 0.8745205479452055, + "grad_norm": 1.9346121549606323, + "learning_rate": 3.9642567018684e-06, + "loss": 2.1434, + "step": 798 + }, + { + "epoch": 0.8756164383561644, + "grad_norm": 1.8202977180480957, + "learning_rate": 3.962225832656377e-06, + "loss": 2.194, + "step": 799 + }, + { + "epoch": 0.8767123287671232, + "grad_norm": 1.6922450065612793, + "learning_rate": 3.960194963444355e-06, + "loss": 2.1821, + "step": 800 + }, + { + "epoch": 0.8767123287671232, + "eval_loss": 2.310209035873413, + "eval_runtime": 46.1102, + "eval_samples_per_second": 21.687, + "eval_steps_per_second": 0.911, + "step": 800 + }, + { + "epoch": 0.8778082191780822, + "grad_norm": 1.9258904457092285, + "learning_rate": 3.958164094232332e-06, + "loss": 2.1326, + "step": 801 + }, + { + "epoch": 0.8789041095890411, + "grad_norm": 1.8485561609268188, + "learning_rate": 3.956133225020309e-06, + "loss": 2.1888, + "step": 802 + }, + { + "epoch": 0.88, + "grad_norm": 1.9802418947219849, + "learning_rate": 3.954102355808286e-06, + "loss": 2.1876, + "step": 803 + }, + { + "epoch": 0.8810958904109589, + "grad_norm": 1.8365259170532227, + "learning_rate": 3.952071486596264e-06, + "loss": 2.1804, + "step": 804 + }, + { + "epoch": 0.8821917808219178, + "grad_norm": 2.023909568786621, + "learning_rate": 3.950040617384241e-06, + "loss": 2.1801, + "step": 805 + }, + { + "epoch": 0.8832876712328767, + "grad_norm": 1.8889018297195435, + "learning_rate": 3.948009748172218e-06, + "loss": 2.1657, + "step": 806 + }, + { + "epoch": 0.8843835616438356, + "grad_norm": 1.7983578443527222, + "learning_rate": 3.9459788789601955e-06, + "loss": 2.1589, + "step": 807 + }, + { + "epoch": 0.8854794520547945, + "grad_norm": 2.157043218612671, + "learning_rate": 3.943948009748173e-06, + "loss": 2.2258, + "step": 808 + }, + { + "epoch": 0.8865753424657534, + "grad_norm": 1.7387373447418213, + "learning_rate": 3.94191714053615e-06, + "loss": 2.1878, + "step": 809 + }, + { + "epoch": 0.8876712328767123, + "grad_norm": 1.963233232498169, + "learning_rate": 3.939886271324127e-06, + "loss": 2.186, + "step": 810 + }, + { + "epoch": 0.8876712328767123, + "eval_loss": 2.306187152862549, + "eval_runtime": 42.563, + "eval_samples_per_second": 23.495, + "eval_steps_per_second": 0.987, + "step": 810 + }, + { + "epoch": 0.8887671232876713, + "grad_norm": 1.8932961225509644, + "learning_rate": 3.937855402112104e-06, + "loss": 2.198, + "step": 811 + }, + { + "epoch": 0.8898630136986302, + "grad_norm": 2.0431783199310303, + "learning_rate": 3.935824532900082e-06, + "loss": 2.1505, + "step": 812 + }, + { + "epoch": 0.890958904109589, + "grad_norm": 1.8954466581344604, + "learning_rate": 3.933793663688058e-06, + "loss": 2.1494, + "step": 813 + }, + { + "epoch": 0.8920547945205479, + "grad_norm": 1.841184139251709, + "learning_rate": 3.9317627944760365e-06, + "loss": 2.1932, + "step": 814 + }, + { + "epoch": 0.8931506849315068, + "grad_norm": 1.9083219766616821, + "learning_rate": 3.929731925264013e-06, + "loss": 2.2081, + "step": 815 + }, + { + "epoch": 0.8942465753424658, + "grad_norm": 1.9737226963043213, + "learning_rate": 3.927701056051991e-06, + "loss": 2.1405, + "step": 816 + }, + { + "epoch": 0.8953424657534247, + "grad_norm": 1.869194746017456, + "learning_rate": 3.9256701868399674e-06, + "loss": 2.1642, + "step": 817 + }, + { + "epoch": 0.8964383561643836, + "grad_norm": 1.9553205966949463, + "learning_rate": 3.9236393176279456e-06, + "loss": 2.1761, + "step": 818 + }, + { + "epoch": 0.8975342465753424, + "grad_norm": 1.8355534076690674, + "learning_rate": 3.921608448415922e-06, + "loss": 2.1551, + "step": 819 + }, + { + "epoch": 0.8986301369863013, + "grad_norm": 1.9770171642303467, + "learning_rate": 3.9195775792039e-06, + "loss": 2.1424, + "step": 820 + }, + { + "epoch": 0.8986301369863013, + "eval_loss": 2.307065010070801, + "eval_runtime": 44.5816, + "eval_samples_per_second": 22.431, + "eval_steps_per_second": 0.942, + "step": 820 + }, + { + "epoch": 0.8997260273972603, + "grad_norm": 1.7375415563583374, + "learning_rate": 3.9175467099918765e-06, + "loss": 2.1895, + "step": 821 + }, + { + "epoch": 0.9008219178082192, + "grad_norm": 1.8910484313964844, + "learning_rate": 3.915515840779855e-06, + "loss": 2.1685, + "step": 822 + }, + { + "epoch": 0.9019178082191781, + "grad_norm": 1.9191482067108154, + "learning_rate": 3.913484971567831e-06, + "loss": 2.2074, + "step": 823 + }, + { + "epoch": 0.903013698630137, + "grad_norm": 1.7864760160446167, + "learning_rate": 3.911454102355808e-06, + "loss": 2.1972, + "step": 824 + }, + { + "epoch": 0.9041095890410958, + "grad_norm": 1.9044650793075562, + "learning_rate": 3.909423233143786e-06, + "loss": 2.1237, + "step": 825 + }, + { + "epoch": 0.9052054794520548, + "grad_norm": 1.8378455638885498, + "learning_rate": 3.907392363931763e-06, + "loss": 2.1907, + "step": 826 + }, + { + "epoch": 0.9063013698630137, + "grad_norm": 1.8708094358444214, + "learning_rate": 3.90536149471974e-06, + "loss": 2.1883, + "step": 827 + }, + { + "epoch": 0.9073972602739726, + "grad_norm": 1.7802201509475708, + "learning_rate": 3.9033306255077175e-06, + "loss": 2.2356, + "step": 828 + }, + { + "epoch": 0.9084931506849315, + "grad_norm": 1.9374496936798096, + "learning_rate": 3.901299756295695e-06, + "loss": 2.2546, + "step": 829 + }, + { + "epoch": 0.9095890410958904, + "grad_norm": 1.9246270656585693, + "learning_rate": 3.899268887083672e-06, + "loss": 2.2684, + "step": 830 + }, + { + "epoch": 0.9095890410958904, + "eval_loss": 2.3065874576568604, + "eval_runtime": 45.18, + "eval_samples_per_second": 22.134, + "eval_steps_per_second": 0.93, + "step": 830 + }, + { + "epoch": 0.9106849315068493, + "grad_norm": 1.8681883811950684, + "learning_rate": 3.897238017871649e-06, + "loss": 2.1737, + "step": 831 + }, + { + "epoch": 0.9117808219178082, + "grad_norm": 1.9293930530548096, + "learning_rate": 3.8952071486596266e-06, + "loss": 2.1813, + "step": 832 + }, + { + "epoch": 0.9128767123287671, + "grad_norm": 1.844586968421936, + "learning_rate": 3.893176279447604e-06, + "loss": 2.1788, + "step": 833 + }, + { + "epoch": 0.913972602739726, + "grad_norm": 1.7658660411834717, + "learning_rate": 3.891145410235581e-06, + "loss": 2.1494, + "step": 834 + }, + { + "epoch": 0.915068493150685, + "grad_norm": 2.05153751373291, + "learning_rate": 3.889114541023558e-06, + "loss": 2.241, + "step": 835 + }, + { + "epoch": 0.9161643835616439, + "grad_norm": 1.7869952917099, + "learning_rate": 3.887083671811536e-06, + "loss": 2.1866, + "step": 836 + }, + { + "epoch": 0.9172602739726028, + "grad_norm": 1.99534273147583, + "learning_rate": 3.885052802599513e-06, + "loss": 2.1682, + "step": 837 + }, + { + "epoch": 0.9183561643835616, + "grad_norm": 1.9593979120254517, + "learning_rate": 3.88302193338749e-06, + "loss": 2.1357, + "step": 838 + }, + { + "epoch": 0.9194520547945205, + "grad_norm": 1.8441778421401978, + "learning_rate": 3.8809910641754675e-06, + "loss": 2.2283, + "step": 839 + }, + { + "epoch": 0.9205479452054794, + "grad_norm": 1.9153512716293335, + "learning_rate": 3.878960194963445e-06, + "loss": 2.1899, + "step": 840 + }, + { + "epoch": 0.9205479452054794, + "eval_loss": 2.307375907897949, + "eval_runtime": 48.0216, + "eval_samples_per_second": 20.824, + "eval_steps_per_second": 0.875, + "step": 840 + }, + { + "epoch": 0.9216438356164384, + "grad_norm": 2.004232883453369, + "learning_rate": 3.876929325751422e-06, + "loss": 2.173, + "step": 841 + }, + { + "epoch": 0.9227397260273973, + "grad_norm": 2.12992262840271, + "learning_rate": 3.874898456539399e-06, + "loss": 2.2164, + "step": 842 + }, + { + "epoch": 0.9238356164383562, + "grad_norm": 2.0505411624908447, + "learning_rate": 3.872867587327377e-06, + "loss": 2.1432, + "step": 843 + }, + { + "epoch": 0.924931506849315, + "grad_norm": 1.767324686050415, + "learning_rate": 3.870836718115354e-06, + "loss": 2.1838, + "step": 844 + }, + { + "epoch": 0.9260273972602739, + "grad_norm": 2.189107894897461, + "learning_rate": 3.868805848903331e-06, + "loss": 2.1623, + "step": 845 + }, + { + "epoch": 0.9271232876712329, + "grad_norm": 1.9781970977783203, + "learning_rate": 3.8667749796913084e-06, + "loss": 2.1738, + "step": 846 + }, + { + "epoch": 0.9282191780821918, + "grad_norm": 1.8734674453735352, + "learning_rate": 3.864744110479286e-06, + "loss": 2.1829, + "step": 847 + }, + { + "epoch": 0.9293150684931507, + "grad_norm": 1.7699538469314575, + "learning_rate": 3.862713241267262e-06, + "loss": 2.186, + "step": 848 + }, + { + "epoch": 0.9304109589041096, + "grad_norm": 1.8354437351226807, + "learning_rate": 3.86068237205524e-06, + "loss": 2.2149, + "step": 849 + }, + { + "epoch": 0.9315068493150684, + "grad_norm": 1.913249135017395, + "learning_rate": 3.858651502843217e-06, + "loss": 2.2208, + "step": 850 + }, + { + "epoch": 0.9315068493150684, + "eval_loss": 2.307549238204956, + "eval_runtime": 43.1706, + "eval_samples_per_second": 23.164, + "eval_steps_per_second": 0.973, + "step": 850 + }, + { + "epoch": 0.9326027397260274, + "grad_norm": 1.92095148563385, + "learning_rate": 3.856620633631195e-06, + "loss": 2.1839, + "step": 851 + }, + { + "epoch": 0.9336986301369863, + "grad_norm": 1.8833959102630615, + "learning_rate": 3.854589764419171e-06, + "loss": 2.1372, + "step": 852 + }, + { + "epoch": 0.9347945205479452, + "grad_norm": 1.9410737752914429, + "learning_rate": 3.852558895207149e-06, + "loss": 2.1726, + "step": 853 + }, + { + "epoch": 0.9358904109589041, + "grad_norm": 1.8328828811645508, + "learning_rate": 3.850528025995126e-06, + "loss": 2.2425, + "step": 854 + }, + { + "epoch": 0.936986301369863, + "grad_norm": 1.89652419090271, + "learning_rate": 3.848497156783104e-06, + "loss": 2.1956, + "step": 855 + }, + { + "epoch": 0.938082191780822, + "grad_norm": 1.9026265144348145, + "learning_rate": 3.84646628757108e-06, + "loss": 2.167, + "step": 856 + }, + { + "epoch": 0.9391780821917808, + "grad_norm": 2.010864734649658, + "learning_rate": 3.8444354183590585e-06, + "loss": 2.1725, + "step": 857 + }, + { + "epoch": 0.9402739726027397, + "grad_norm": 2.293717384338379, + "learning_rate": 3.842404549147035e-06, + "loss": 2.1668, + "step": 858 + }, + { + "epoch": 0.9413698630136986, + "grad_norm": 1.8209116458892822, + "learning_rate": 3.840373679935013e-06, + "loss": 2.1649, + "step": 859 + }, + { + "epoch": 0.9424657534246575, + "grad_norm": 2.061277389526367, + "learning_rate": 3.8383428107229894e-06, + "loss": 2.1479, + "step": 860 + }, + { + "epoch": 0.9424657534246575, + "eval_loss": 2.3073670864105225, + "eval_runtime": 44.3781, + "eval_samples_per_second": 22.534, + "eval_steps_per_second": 0.946, + "step": 860 + }, + { + "epoch": 0.9435616438356165, + "grad_norm": 1.8032523393630981, + "learning_rate": 3.836311941510967e-06, + "loss": 2.1665, + "step": 861 + }, + { + "epoch": 0.9446575342465754, + "grad_norm": 1.989533543586731, + "learning_rate": 3.834281072298944e-06, + "loss": 2.1612, + "step": 862 + }, + { + "epoch": 0.9457534246575342, + "grad_norm": 2.017179012298584, + "learning_rate": 3.832250203086921e-06, + "loss": 2.201, + "step": 863 + }, + { + "epoch": 0.9468493150684931, + "grad_norm": 2.0291004180908203, + "learning_rate": 3.8302193338748985e-06, + "loss": 2.1508, + "step": 864 + }, + { + "epoch": 0.947945205479452, + "grad_norm": 1.8693255186080933, + "learning_rate": 3.828188464662876e-06, + "loss": 2.2106, + "step": 865 + }, + { + "epoch": 0.949041095890411, + "grad_norm": 1.8890169858932495, + "learning_rate": 3.826157595450853e-06, + "loss": 2.2467, + "step": 866 + }, + { + "epoch": 0.9501369863013699, + "grad_norm": 2.0550642013549805, + "learning_rate": 3.82412672623883e-06, + "loss": 2.1899, + "step": 867 + }, + { + "epoch": 0.9512328767123288, + "grad_norm": 1.8681174516677856, + "learning_rate": 3.822095857026808e-06, + "loss": 2.1428, + "step": 868 + }, + { + "epoch": 0.9523287671232876, + "grad_norm": 1.9133951663970947, + "learning_rate": 3.820064987814785e-06, + "loss": 2.1947, + "step": 869 + }, + { + "epoch": 0.9534246575342465, + "grad_norm": 1.8717621564865112, + "learning_rate": 3.818034118602762e-06, + "loss": 2.1542, + "step": 870 + }, + { + "epoch": 0.9534246575342465, + "eval_loss": 2.307584047317505, + "eval_runtime": 43.718, + "eval_samples_per_second": 22.874, + "eval_steps_per_second": 0.961, + "step": 870 + }, + { + "epoch": 0.9545205479452055, + "grad_norm": 1.9843461513519287, + "learning_rate": 3.8160032493907395e-06, + "loss": 2.1547, + "step": 871 + }, + { + "epoch": 0.9556164383561644, + "grad_norm": 1.725468397140503, + "learning_rate": 3.813972380178717e-06, + "loss": 2.1679, + "step": 872 + }, + { + "epoch": 0.9567123287671233, + "grad_norm": 1.8076086044311523, + "learning_rate": 3.811941510966694e-06, + "loss": 2.1654, + "step": 873 + }, + { + "epoch": 0.9578082191780822, + "grad_norm": 1.8879774808883667, + "learning_rate": 3.809910641754671e-06, + "loss": 2.191, + "step": 874 + }, + { + "epoch": 0.958904109589041, + "grad_norm": 1.943313717842102, + "learning_rate": 3.8078797725426486e-06, + "loss": 2.1448, + "step": 875 + }, + { + "epoch": 0.96, + "grad_norm": 1.8877806663513184, + "learning_rate": 3.8058489033306254e-06, + "loss": 2.1129, + "step": 876 + }, + { + "epoch": 0.9610958904109589, + "grad_norm": 1.787083625793457, + "learning_rate": 3.803818034118603e-06, + "loss": 2.1854, + "step": 877 + }, + { + "epoch": 0.9621917808219178, + "grad_norm": 2.0925140380859375, + "learning_rate": 3.80178716490658e-06, + "loss": 2.1614, + "step": 878 + }, + { + "epoch": 0.9632876712328767, + "grad_norm": 1.9043725728988647, + "learning_rate": 3.7997562956945577e-06, + "loss": 2.2161, + "step": 879 + }, + { + "epoch": 0.9643835616438357, + "grad_norm": 1.8173545598983765, + "learning_rate": 3.7977254264825345e-06, + "loss": 2.2067, + "step": 880 + }, + { + "epoch": 0.9643835616438357, + "eval_loss": 2.302727460861206, + "eval_runtime": 44.2015, + "eval_samples_per_second": 22.624, + "eval_steps_per_second": 0.95, + "step": 880 + }, + { + "epoch": 0.9654794520547946, + "grad_norm": 1.9871562719345093, + "learning_rate": 3.7956945572705122e-06, + "loss": 2.1658, + "step": 881 + }, + { + "epoch": 0.9665753424657534, + "grad_norm": 2.103553295135498, + "learning_rate": 3.793663688058489e-06, + "loss": 2.1468, + "step": 882 + }, + { + "epoch": 0.9676712328767123, + "grad_norm": 1.9089381694793701, + "learning_rate": 3.791632818846467e-06, + "loss": 2.2289, + "step": 883 + }, + { + "epoch": 0.9687671232876712, + "grad_norm": 2.3238179683685303, + "learning_rate": 3.7896019496344436e-06, + "loss": 2.2179, + "step": 884 + }, + { + "epoch": 0.9698630136986301, + "grad_norm": 1.902161955833435, + "learning_rate": 3.787571080422421e-06, + "loss": 2.2086, + "step": 885 + }, + { + "epoch": 0.9709589041095891, + "grad_norm": 1.8587497472763062, + "learning_rate": 3.785540211210398e-06, + "loss": 2.2175, + "step": 886 + }, + { + "epoch": 0.972054794520548, + "grad_norm": 2.029719829559326, + "learning_rate": 3.7835093419983755e-06, + "loss": 2.207, + "step": 887 + }, + { + "epoch": 0.9731506849315068, + "grad_norm": 2.0228402614593506, + "learning_rate": 3.7814784727863527e-06, + "loss": 2.1851, + "step": 888 + }, + { + "epoch": 0.9742465753424657, + "grad_norm": 1.9885436296463013, + "learning_rate": 3.77944760357433e-06, + "loss": 2.1667, + "step": 889 + }, + { + "epoch": 0.9753424657534246, + "grad_norm": 1.8621459007263184, + "learning_rate": 3.7774167343623073e-06, + "loss": 2.2529, + "step": 890 + }, + { + "epoch": 0.9753424657534246, + "eval_loss": 2.303697347640991, + "eval_runtime": 44.9459, + "eval_samples_per_second": 22.249, + "eval_steps_per_second": 0.934, + "step": 890 + }, + { + "epoch": 0.9764383561643836, + "grad_norm": 2.0311460494995117, + "learning_rate": 3.7753858651502846e-06, + "loss": 2.2185, + "step": 891 + }, + { + "epoch": 0.9775342465753425, + "grad_norm": 2.2017149925231934, + "learning_rate": 3.773354995938262e-06, + "loss": 2.164, + "step": 892 + }, + { + "epoch": 0.9786301369863014, + "grad_norm": 1.949684977531433, + "learning_rate": 3.771324126726239e-06, + "loss": 2.2128, + "step": 893 + }, + { + "epoch": 0.9797260273972602, + "grad_norm": 1.9395498037338257, + "learning_rate": 3.7692932575142164e-06, + "loss": 2.1957, + "step": 894 + }, + { + "epoch": 0.9808219178082191, + "grad_norm": 2.096432685852051, + "learning_rate": 3.7672623883021937e-06, + "loss": 2.219, + "step": 895 + }, + { + "epoch": 0.9819178082191781, + "grad_norm": 2.0942952632904053, + "learning_rate": 3.765231519090171e-06, + "loss": 2.1546, + "step": 896 + }, + { + "epoch": 0.983013698630137, + "grad_norm": 2.0097291469573975, + "learning_rate": 3.7632006498781482e-06, + "loss": 2.1021, + "step": 897 + }, + { + "epoch": 0.9841095890410959, + "grad_norm": 2.0909335613250732, + "learning_rate": 3.761169780666125e-06, + "loss": 2.1862, + "step": 898 + }, + { + "epoch": 0.9852054794520548, + "grad_norm": 2.187309980392456, + "learning_rate": 3.7591389114541028e-06, + "loss": 2.1511, + "step": 899 + }, + { + "epoch": 0.9863013698630136, + "grad_norm": 2.194276809692383, + "learning_rate": 3.7571080422420796e-06, + "loss": 2.196, + "step": 900 + }, + { + "epoch": 0.9863013698630136, + "eval_loss": 2.3047611713409424, + "eval_runtime": 45.609, + "eval_samples_per_second": 21.925, + "eval_steps_per_second": 0.921, + "step": 900 + }, + { + "epoch": 0.9873972602739726, + "grad_norm": 2.1132829189300537, + "learning_rate": 3.7550771730300573e-06, + "loss": 2.2046, + "step": 901 + }, + { + "epoch": 0.9884931506849315, + "grad_norm": 2.1946468353271484, + "learning_rate": 3.753046303818034e-06, + "loss": 2.1656, + "step": 902 + }, + { + "epoch": 0.9895890410958904, + "grad_norm": 2.2210536003112793, + "learning_rate": 3.751015434606012e-06, + "loss": 2.2186, + "step": 903 + }, + { + "epoch": 0.9906849315068493, + "grad_norm": 1.913695216178894, + "learning_rate": 3.7489845653939887e-06, + "loss": 2.1933, + "step": 904 + }, + { + "epoch": 0.9917808219178083, + "grad_norm": 2.1088368892669678, + "learning_rate": 3.7469536961819664e-06, + "loss": 2.2406, + "step": 905 + }, + { + "epoch": 0.9928767123287672, + "grad_norm": 2.04239821434021, + "learning_rate": 3.7449228269699433e-06, + "loss": 2.142, + "step": 906 + }, + { + "epoch": 0.993972602739726, + "grad_norm": 1.7852507829666138, + "learning_rate": 3.742891957757921e-06, + "loss": 2.1701, + "step": 907 + }, + { + "epoch": 0.9950684931506849, + "grad_norm": 2.0275776386260986, + "learning_rate": 3.740861088545898e-06, + "loss": 2.2004, + "step": 908 + }, + { + "epoch": 0.9961643835616438, + "grad_norm": 1.9028904438018799, + "learning_rate": 3.7388302193338755e-06, + "loss": 2.177, + "step": 909 + }, + { + "epoch": 0.9972602739726028, + "grad_norm": 1.7815115451812744, + "learning_rate": 3.7367993501218524e-06, + "loss": 2.207, + "step": 910 + }, + { + "epoch": 0.9972602739726028, + "eval_loss": 2.302318811416626, + "eval_runtime": 42.9296, + "eval_samples_per_second": 23.294, + "eval_steps_per_second": 0.978, + "step": 910 + }, + { + "epoch": 0.9983561643835617, + "grad_norm": 1.913689374923706, + "learning_rate": 3.7347684809098297e-06, + "loss": 2.1594, + "step": 911 + }, + { + "epoch": 0.9994520547945206, + "grad_norm": 2.171222686767578, + "learning_rate": 3.732737611697807e-06, + "loss": 2.1689, + "step": 912 + }, + { + "epoch": 1.0005479452054795, + "grad_norm": 1.831642985343933, + "learning_rate": 3.7307067424857842e-06, + "loss": 2.1367, + "step": 913 + }, + { + "epoch": 1.0016438356164383, + "grad_norm": 2.1831955909729004, + "learning_rate": 3.7286758732737615e-06, + "loss": 2.1035, + "step": 914 + }, + { + "epoch": 1.0027397260273974, + "grad_norm": 2.038025379180908, + "learning_rate": 3.7266450040617388e-06, + "loss": 2.1461, + "step": 915 + }, + { + "epoch": 1.0038356164383562, + "grad_norm": 1.866502046585083, + "learning_rate": 3.724614134849716e-06, + "loss": 2.1322, + "step": 916 + }, + { + "epoch": 1.004931506849315, + "grad_norm": 2.1359357833862305, + "learning_rate": 3.7225832656376933e-06, + "loss": 2.2333, + "step": 917 + }, + { + "epoch": 1.006027397260274, + "grad_norm": 1.8689136505126953, + "learning_rate": 3.7205523964256706e-06, + "loss": 2.1176, + "step": 918 + }, + { + "epoch": 1.0071232876712328, + "grad_norm": 1.8869425058364868, + "learning_rate": 3.718521527213648e-06, + "loss": 2.2127, + "step": 919 + }, + { + "epoch": 1.0082191780821919, + "grad_norm": 1.9148075580596924, + "learning_rate": 3.716490658001625e-06, + "loss": 2.1846, + "step": 920 + }, + { + "epoch": 1.0082191780821919, + "eval_loss": 2.30368709564209, + "eval_runtime": 52.1845, + "eval_samples_per_second": 19.163, + "eval_steps_per_second": 0.805, + "step": 920 + }, + { + "epoch": 1.0093150684931507, + "grad_norm": 1.9502761363983154, + "learning_rate": 3.7144597887896024e-06, + "loss": 2.1731, + "step": 921 + }, + { + "epoch": 1.0104109589041095, + "grad_norm": 1.9093869924545288, + "learning_rate": 3.7124289195775797e-06, + "loss": 2.0864, + "step": 922 + }, + { + "epoch": 1.0115068493150685, + "grad_norm": 2.0871293544769287, + "learning_rate": 3.710398050365557e-06, + "loss": 2.1838, + "step": 923 + }, + { + "epoch": 1.0126027397260273, + "grad_norm": 2.0559985637664795, + "learning_rate": 3.708367181153534e-06, + "loss": 2.1764, + "step": 924 + }, + { + "epoch": 1.0136986301369864, + "grad_norm": 1.8968807458877563, + "learning_rate": 3.7063363119415115e-06, + "loss": 2.1515, + "step": 925 + }, + { + "epoch": 1.0147945205479452, + "grad_norm": 1.8613216876983643, + "learning_rate": 3.7043054427294884e-06, + "loss": 2.1694, + "step": 926 + }, + { + "epoch": 1.0158904109589042, + "grad_norm": 2.002044439315796, + "learning_rate": 3.702274573517466e-06, + "loss": 2.148, + "step": 927 + }, + { + "epoch": 1.016986301369863, + "grad_norm": 2.1138675212860107, + "learning_rate": 3.700243704305443e-06, + "loss": 2.1401, + "step": 928 + }, + { + "epoch": 1.0180821917808218, + "grad_norm": 1.7620841264724731, + "learning_rate": 3.6982128350934206e-06, + "loss": 2.1399, + "step": 929 + }, + { + "epoch": 1.0191780821917809, + "grad_norm": 1.8970746994018555, + "learning_rate": 3.6961819658813975e-06, + "loss": 2.1404, + "step": 930 + }, + { + "epoch": 1.0191780821917809, + "eval_loss": 2.3068811893463135, + "eval_runtime": 53.0188, + "eval_samples_per_second": 18.861, + "eval_steps_per_second": 0.792, + "step": 930 + }, + { + "epoch": 1.0202739726027397, + "grad_norm": 1.9405478239059448, + "learning_rate": 3.694151096669375e-06, + "loss": 2.1286, + "step": 931 + }, + { + "epoch": 1.0213698630136987, + "grad_norm": 2.112222909927368, + "learning_rate": 3.692120227457352e-06, + "loss": 2.2246, + "step": 932 + }, + { + "epoch": 1.0224657534246575, + "grad_norm": 1.8792580366134644, + "learning_rate": 3.6900893582453297e-06, + "loss": 2.148, + "step": 933 + }, + { + "epoch": 1.0235616438356165, + "grad_norm": 1.9530014991760254, + "learning_rate": 3.6880584890333066e-06, + "loss": 2.1707, + "step": 934 + }, + { + "epoch": 1.0246575342465754, + "grad_norm": 2.1778736114501953, + "learning_rate": 3.6860276198212834e-06, + "loss": 2.0879, + "step": 935 + }, + { + "epoch": 1.0257534246575342, + "grad_norm": 1.8960413932800293, + "learning_rate": 3.683996750609261e-06, + "loss": 2.1659, + "step": 936 + }, + { + "epoch": 1.0268493150684932, + "grad_norm": 1.9805221557617188, + "learning_rate": 3.681965881397238e-06, + "loss": 2.1674, + "step": 937 + }, + { + "epoch": 1.027945205479452, + "grad_norm": 1.9999068975448608, + "learning_rate": 3.6799350121852157e-06, + "loss": 2.1677, + "step": 938 + }, + { + "epoch": 1.029041095890411, + "grad_norm": 2.016911029815674, + "learning_rate": 3.6779041429731925e-06, + "loss": 2.1841, + "step": 939 + }, + { + "epoch": 1.0301369863013699, + "grad_norm": 1.7200374603271484, + "learning_rate": 3.6758732737611702e-06, + "loss": 2.1391, + "step": 940 + }, + { + "epoch": 1.0301369863013699, + "eval_loss": 2.3022048473358154, + "eval_runtime": 51.2908, + "eval_samples_per_second": 19.497, + "eval_steps_per_second": 0.819, + "step": 940 + }, + { + "epoch": 1.0312328767123287, + "grad_norm": 2.249377489089966, + "learning_rate": 3.673842404549147e-06, + "loss": 2.1618, + "step": 941 + }, + { + "epoch": 1.0323287671232877, + "grad_norm": 1.838922381401062, + "learning_rate": 3.671811535337125e-06, + "loss": 2.1527, + "step": 942 + }, + { + "epoch": 1.0334246575342465, + "grad_norm": 1.9571045637130737, + "learning_rate": 3.6697806661251016e-06, + "loss": 2.149, + "step": 943 + }, + { + "epoch": 1.0345205479452055, + "grad_norm": 1.9929102659225464, + "learning_rate": 3.6677497969130793e-06, + "loss": 2.159, + "step": 944 + }, + { + "epoch": 1.0356164383561643, + "grad_norm": 2.004830837249756, + "learning_rate": 3.665718927701056e-06, + "loss": 2.2369, + "step": 945 + }, + { + "epoch": 1.0367123287671234, + "grad_norm": 1.9605739116668701, + "learning_rate": 3.663688058489034e-06, + "loss": 2.1795, + "step": 946 + }, + { + "epoch": 1.0378082191780822, + "grad_norm": 1.9499164819717407, + "learning_rate": 3.6616571892770107e-06, + "loss": 2.1979, + "step": 947 + }, + { + "epoch": 1.038904109589041, + "grad_norm": 1.867347240447998, + "learning_rate": 3.659626320064988e-06, + "loss": 2.1616, + "step": 948 + }, + { + "epoch": 1.04, + "grad_norm": 1.9342801570892334, + "learning_rate": 3.6575954508529653e-06, + "loss": 2.1044, + "step": 949 + }, + { + "epoch": 1.0410958904109588, + "grad_norm": 2.0009584426879883, + "learning_rate": 3.6555645816409426e-06, + "loss": 2.1543, + "step": 950 + }, + { + "epoch": 1.0410958904109588, + "eval_loss": 2.3028438091278076, + "eval_runtime": 46.7426, + "eval_samples_per_second": 21.394, + "eval_steps_per_second": 0.899, + "step": 950 + }, + { + "epoch": 1.0421917808219179, + "grad_norm": 2.0411503314971924, + "learning_rate": 3.65353371242892e-06, + "loss": 2.1465, + "step": 951 + }, + { + "epoch": 1.0432876712328767, + "grad_norm": 2.0512070655822754, + "learning_rate": 3.651502843216897e-06, + "loss": 2.1828, + "step": 952 + }, + { + "epoch": 1.0443835616438357, + "grad_norm": 2.0948643684387207, + "learning_rate": 3.6494719740048744e-06, + "loss": 2.0671, + "step": 953 + }, + { + "epoch": 1.0454794520547945, + "grad_norm": 1.9103025197982788, + "learning_rate": 3.6474411047928517e-06, + "loss": 2.1651, + "step": 954 + }, + { + "epoch": 1.0465753424657533, + "grad_norm": 2.1500349044799805, + "learning_rate": 3.645410235580829e-06, + "loss": 2.137, + "step": 955 + }, + { + "epoch": 1.0476712328767124, + "grad_norm": 1.7760887145996094, + "learning_rate": 3.6433793663688062e-06, + "loss": 2.1738, + "step": 956 + }, + { + "epoch": 1.0487671232876712, + "grad_norm": 2.0111730098724365, + "learning_rate": 3.6413484971567835e-06, + "loss": 2.1569, + "step": 957 + }, + { + "epoch": 1.0498630136986302, + "grad_norm": 2.0040462017059326, + "learning_rate": 3.6393176279447608e-06, + "loss": 2.1483, + "step": 958 + }, + { + "epoch": 1.050958904109589, + "grad_norm": 1.9583499431610107, + "learning_rate": 3.637286758732738e-06, + "loss": 2.1575, + "step": 959 + }, + { + "epoch": 1.0520547945205478, + "grad_norm": 1.8842004537582397, + "learning_rate": 3.6352558895207153e-06, + "loss": 2.1797, + "step": 960 + }, + { + "epoch": 1.0520547945205478, + "eval_loss": 2.301715850830078, + "eval_runtime": 52.7345, + "eval_samples_per_second": 18.963, + "eval_steps_per_second": 0.796, + "step": 960 + }, + { + "epoch": 1.0531506849315069, + "grad_norm": 2.020909309387207, + "learning_rate": 3.633225020308692e-06, + "loss": 2.1939, + "step": 961 + }, + { + "epoch": 1.0542465753424657, + "grad_norm": 2.09222149848938, + "learning_rate": 3.63119415109667e-06, + "loss": 2.1777, + "step": 962 + }, + { + "epoch": 1.0553424657534247, + "grad_norm": 1.8582347631454468, + "learning_rate": 3.6291632818846467e-06, + "loss": 2.1505, + "step": 963 + }, + { + "epoch": 1.0564383561643835, + "grad_norm": 1.9468824863433838, + "learning_rate": 3.6271324126726244e-06, + "loss": 2.1762, + "step": 964 + }, + { + "epoch": 1.0575342465753426, + "grad_norm": 1.9484895467758179, + "learning_rate": 3.6251015434606013e-06, + "loss": 2.1102, + "step": 965 + }, + { + "epoch": 1.0586301369863014, + "grad_norm": 2.11462140083313, + "learning_rate": 3.623070674248579e-06, + "loss": 2.1114, + "step": 966 + }, + { + "epoch": 1.0597260273972602, + "grad_norm": 2.2038373947143555, + "learning_rate": 3.621039805036556e-06, + "loss": 2.1513, + "step": 967 + }, + { + "epoch": 1.0608219178082192, + "grad_norm": 2.1271684169769287, + "learning_rate": 3.6190089358245335e-06, + "loss": 2.137, + "step": 968 + }, + { + "epoch": 1.061917808219178, + "grad_norm": 1.8624194860458374, + "learning_rate": 3.6169780666125104e-06, + "loss": 2.1701, + "step": 969 + }, + { + "epoch": 1.063013698630137, + "grad_norm": 1.935445785522461, + "learning_rate": 3.614947197400488e-06, + "loss": 2.1726, + "step": 970 + }, + { + "epoch": 1.063013698630137, + "eval_loss": 2.302633047103882, + "eval_runtime": 51.8284, + "eval_samples_per_second": 19.294, + "eval_steps_per_second": 0.81, + "step": 970 + }, + { + "epoch": 1.0641095890410959, + "grad_norm": 2.048907518386841, + "learning_rate": 3.612916328188465e-06, + "loss": 2.1712, + "step": 971 + }, + { + "epoch": 1.0652054794520547, + "grad_norm": 1.850943684577942, + "learning_rate": 3.610885458976442e-06, + "loss": 2.102, + "step": 972 + }, + { + "epoch": 1.0663013698630137, + "grad_norm": 1.8364315032958984, + "learning_rate": 3.6088545897644195e-06, + "loss": 2.1903, + "step": 973 + }, + { + "epoch": 1.0673972602739725, + "grad_norm": 1.8227031230926514, + "learning_rate": 3.6068237205523963e-06, + "loss": 2.1787, + "step": 974 + }, + { + "epoch": 1.0684931506849316, + "grad_norm": 1.9136372804641724, + "learning_rate": 3.604792851340374e-06, + "loss": 2.1244, + "step": 975 + }, + { + "epoch": 1.0695890410958904, + "grad_norm": 1.8585968017578125, + "learning_rate": 3.602761982128351e-06, + "loss": 2.1287, + "step": 976 + }, + { + "epoch": 1.0706849315068494, + "grad_norm": 2.139573812484741, + "learning_rate": 3.6007311129163286e-06, + "loss": 2.1938, + "step": 977 + }, + { + "epoch": 1.0717808219178082, + "grad_norm": 2.1584925651550293, + "learning_rate": 3.5987002437043054e-06, + "loss": 2.1626, + "step": 978 + }, + { + "epoch": 1.072876712328767, + "grad_norm": 1.8905949592590332, + "learning_rate": 3.596669374492283e-06, + "loss": 2.1219, + "step": 979 + }, + { + "epoch": 1.073972602739726, + "grad_norm": 2.099651575088501, + "learning_rate": 3.59463850528026e-06, + "loss": 2.1385, + "step": 980 + }, + { + "epoch": 1.073972602739726, + "eval_loss": 2.3004090785980225, + "eval_runtime": 52.7406, + "eval_samples_per_second": 18.961, + "eval_steps_per_second": 0.796, + "step": 980 + }, + { + "epoch": 1.0750684931506849, + "grad_norm": 2.1456780433654785, + "learning_rate": 3.5926076360682377e-06, + "loss": 2.1973, + "step": 981 + }, + { + "epoch": 1.076164383561644, + "grad_norm": 1.929372787475586, + "learning_rate": 3.5905767668562145e-06, + "loss": 2.1115, + "step": 982 + }, + { + "epoch": 1.0772602739726027, + "grad_norm": 2.08174991607666, + "learning_rate": 3.5885458976441922e-06, + "loss": 2.1785, + "step": 983 + }, + { + "epoch": 1.0783561643835617, + "grad_norm": 1.9355952739715576, + "learning_rate": 3.586515028432169e-06, + "loss": 2.1567, + "step": 984 + }, + { + "epoch": 1.0794520547945206, + "grad_norm": 1.9983466863632202, + "learning_rate": 3.5844841592201464e-06, + "loss": 2.1913, + "step": 985 + }, + { + "epoch": 1.0805479452054794, + "grad_norm": 1.8714051246643066, + "learning_rate": 3.5824532900081237e-06, + "loss": 2.1938, + "step": 986 + }, + { + "epoch": 1.0816438356164384, + "grad_norm": 2.007049798965454, + "learning_rate": 3.580422420796101e-06, + "loss": 2.1093, + "step": 987 + }, + { + "epoch": 1.0827397260273972, + "grad_norm": 1.965572476387024, + "learning_rate": 3.578391551584078e-06, + "loss": 2.1426, + "step": 988 + }, + { + "epoch": 1.0838356164383562, + "grad_norm": 1.9040261507034302, + "learning_rate": 3.5763606823720555e-06, + "loss": 2.1627, + "step": 989 + }, + { + "epoch": 1.084931506849315, + "grad_norm": 2.001929759979248, + "learning_rate": 3.5743298131600328e-06, + "loss": 2.0983, + "step": 990 + }, + { + "epoch": 1.084931506849315, + "eval_loss": 2.2979862689971924, + "eval_runtime": 50.6325, + "eval_samples_per_second": 19.75, + "eval_steps_per_second": 0.83, + "step": 990 + }, + { + "epoch": 1.086027397260274, + "grad_norm": 1.9204787015914917, + "learning_rate": 3.57229894394801e-06, + "loss": 2.1571, + "step": 991 + }, + { + "epoch": 1.087123287671233, + "grad_norm": 1.9628651142120361, + "learning_rate": 3.5702680747359873e-06, + "loss": 2.1736, + "step": 992 + }, + { + "epoch": 1.0882191780821917, + "grad_norm": 1.9888029098510742, + "learning_rate": 3.5682372055239646e-06, + "loss": 2.17, + "step": 993 + }, + { + "epoch": 1.0893150684931507, + "grad_norm": 2.049769401550293, + "learning_rate": 3.566206336311942e-06, + "loss": 2.1352, + "step": 994 + }, + { + "epoch": 1.0904109589041096, + "grad_norm": 2.0208375453948975, + "learning_rate": 3.564175467099919e-06, + "loss": 2.1564, + "step": 995 + }, + { + "epoch": 1.0915068493150686, + "grad_norm": 1.976775050163269, + "learning_rate": 3.5621445978878964e-06, + "loss": 2.1292, + "step": 996 + }, + { + "epoch": 1.0926027397260274, + "grad_norm": 2.0502822399139404, + "learning_rate": 3.5601137286758737e-06, + "loss": 2.1115, + "step": 997 + }, + { + "epoch": 1.0936986301369862, + "grad_norm": 2.1184189319610596, + "learning_rate": 3.5580828594638505e-06, + "loss": 2.1641, + "step": 998 + }, + { + "epoch": 1.0947945205479452, + "grad_norm": 1.9662559032440186, + "learning_rate": 3.5560519902518282e-06, + "loss": 2.1338, + "step": 999 + }, + { + "epoch": 1.095890410958904, + "grad_norm": 2.13386869430542, + "learning_rate": 3.554021121039805e-06, + "loss": 2.158, + "step": 1000 + }, + { + "epoch": 1.095890410958904, + "eval_loss": 2.3001608848571777, + "eval_runtime": 45.9219, + "eval_samples_per_second": 21.776, + "eval_steps_per_second": 0.915, + "step": 1000 + }, + { + "epoch": 1.096986301369863, + "grad_norm": 1.8791636228561401, + "learning_rate": 3.551990251827783e-06, + "loss": 2.1752, + "step": 1001 + }, + { + "epoch": 1.098082191780822, + "grad_norm": 1.9105546474456787, + "learning_rate": 3.5499593826157596e-06, + "loss": 2.1543, + "step": 1002 + }, + { + "epoch": 1.099178082191781, + "grad_norm": 1.887668251991272, + "learning_rate": 3.5479285134037373e-06, + "loss": 2.1226, + "step": 1003 + }, + { + "epoch": 1.1002739726027397, + "grad_norm": 2.2854599952697754, + "learning_rate": 3.545897644191714e-06, + "loss": 2.1425, + "step": 1004 + }, + { + "epoch": 1.1013698630136985, + "grad_norm": 2.03892183303833, + "learning_rate": 3.543866774979692e-06, + "loss": 2.135, + "step": 1005 + }, + { + "epoch": 1.1024657534246576, + "grad_norm": 1.9144858121871948, + "learning_rate": 3.5418359057676687e-06, + "loss": 2.1323, + "step": 1006 + }, + { + "epoch": 1.1035616438356164, + "grad_norm": 2.1402337551116943, + "learning_rate": 3.5398050365556464e-06, + "loss": 2.1668, + "step": 1007 + }, + { + "epoch": 1.1046575342465754, + "grad_norm": 2.084592819213867, + "learning_rate": 3.5377741673436233e-06, + "loss": 2.1331, + "step": 1008 + }, + { + "epoch": 1.1057534246575342, + "grad_norm": 2.0012383460998535, + "learning_rate": 3.535743298131601e-06, + "loss": 2.1677, + "step": 1009 + }, + { + "epoch": 1.106849315068493, + "grad_norm": 2.272204637527466, + "learning_rate": 3.533712428919578e-06, + "loss": 2.13, + "step": 1010 + }, + { + "epoch": 1.106849315068493, + "eval_loss": 2.3001229763031006, + "eval_runtime": 42.0444, + "eval_samples_per_second": 23.784, + "eval_steps_per_second": 0.999, + "step": 1010 + }, + { + "epoch": 1.107945205479452, + "grad_norm": 2.1061534881591797, + "learning_rate": 3.5316815597075547e-06, + "loss": 2.1581, + "step": 1011 + }, + { + "epoch": 1.1090410958904109, + "grad_norm": 2.159306764602661, + "learning_rate": 3.5296506904955324e-06, + "loss": 2.1868, + "step": 1012 + }, + { + "epoch": 1.11013698630137, + "grad_norm": 2.1429545879364014, + "learning_rate": 3.5276198212835093e-06, + "loss": 2.1274, + "step": 1013 + }, + { + "epoch": 1.1112328767123287, + "grad_norm": 1.8494454622268677, + "learning_rate": 3.525588952071487e-06, + "loss": 2.1513, + "step": 1014 + }, + { + "epoch": 1.1123287671232878, + "grad_norm": 1.8494454622268677, + "learning_rate": 3.525588952071487e-06, + "loss": 2.1883, + "step": 1015 + }, + { + "epoch": 1.1134246575342466, + "grad_norm": 2.2244861125946045, + "learning_rate": 3.523558082859464e-06, + "loss": 2.1896, + "step": 1016 + }, + { + "epoch": 1.1145205479452054, + "grad_norm": 2.181950569152832, + "learning_rate": 3.5215272136474415e-06, + "loss": 2.1561, + "step": 1017 + }, + { + "epoch": 1.1156164383561644, + "grad_norm": 1.7740849256515503, + "learning_rate": 3.5194963444354184e-06, + "loss": 2.1358, + "step": 1018 + }, + { + "epoch": 1.1167123287671232, + "grad_norm": 2.3540139198303223, + "learning_rate": 3.517465475223396e-06, + "loss": 2.1241, + "step": 1019 + }, + { + "epoch": 1.1178082191780823, + "grad_norm": 2.2737820148468018, + "learning_rate": 3.515434606011373e-06, + "loss": 2.1117, + "step": 1020 + }, + { + "epoch": 1.1178082191780823, + "eval_loss": 2.2998275756835938, + "eval_runtime": 46.2142, + "eval_samples_per_second": 21.638, + "eval_steps_per_second": 0.909, + "step": 1020 + }, + { + "epoch": 1.118904109589041, + "grad_norm": 2.0705313682556152, + "learning_rate": 3.5134037367993506e-06, + "loss": 2.1743, + "step": 1021 + }, + { + "epoch": 1.12, + "grad_norm": 2.2605583667755127, + "learning_rate": 3.5113728675873275e-06, + "loss": 2.2082, + "step": 1022 + }, + { + "epoch": 1.121095890410959, + "grad_norm": 2.028203248977661, + "learning_rate": 3.5093419983753047e-06, + "loss": 2.1722, + "step": 1023 + }, + { + "epoch": 1.1221917808219177, + "grad_norm": 1.9794270992279053, + "learning_rate": 3.507311129163282e-06, + "loss": 2.1581, + "step": 1024 + }, + { + "epoch": 1.1232876712328768, + "grad_norm": 2.1830344200134277, + "learning_rate": 3.5052802599512593e-06, + "loss": 2.1142, + "step": 1025 + }, + { + "epoch": 1.1243835616438356, + "grad_norm": 1.926081895828247, + "learning_rate": 3.5032493907392366e-06, + "loss": 2.1532, + "step": 1026 + }, + { + "epoch": 1.1254794520547946, + "grad_norm": 1.9909394979476929, + "learning_rate": 3.501218521527214e-06, + "loss": 2.153, + "step": 1027 + }, + { + "epoch": 1.1265753424657534, + "grad_norm": 1.9685133695602417, + "learning_rate": 3.499187652315191e-06, + "loss": 2.0983, + "step": 1028 + }, + { + "epoch": 1.1276712328767124, + "grad_norm": 1.9185279607772827, + "learning_rate": 3.4971567831031684e-06, + "loss": 2.1495, + "step": 1029 + }, + { + "epoch": 1.1287671232876713, + "grad_norm": 2.0214717388153076, + "learning_rate": 3.4951259138911457e-06, + "loss": 2.176, + "step": 1030 + }, + { + "epoch": 1.1287671232876713, + "eval_loss": 2.2963690757751465, + "eval_runtime": 44.2655, + "eval_samples_per_second": 22.591, + "eval_steps_per_second": 0.949, + "step": 1030 + }, + { + "epoch": 1.12986301369863, + "grad_norm": 1.9134252071380615, + "learning_rate": 3.493095044679123e-06, + "loss": 2.1034, + "step": 1031 + }, + { + "epoch": 1.130958904109589, + "grad_norm": 1.8448772430419922, + "learning_rate": 3.4910641754671002e-06, + "loss": 2.1947, + "step": 1032 + }, + { + "epoch": 1.132054794520548, + "grad_norm": 1.856341004371643, + "learning_rate": 3.4890333062550775e-06, + "loss": 2.0676, + "step": 1033 + }, + { + "epoch": 1.133150684931507, + "grad_norm": 2.039062261581421, + "learning_rate": 3.4870024370430548e-06, + "loss": 2.1618, + "step": 1034 + }, + { + "epoch": 1.1342465753424658, + "grad_norm": 1.8440991640090942, + "learning_rate": 3.484971567831032e-06, + "loss": 2.0956, + "step": 1035 + }, + { + "epoch": 1.1353424657534246, + "grad_norm": 1.8483991622924805, + "learning_rate": 3.482940698619009e-06, + "loss": 2.1364, + "step": 1036 + }, + { + "epoch": 1.1364383561643836, + "grad_norm": 1.9045069217681885, + "learning_rate": 3.4809098294069866e-06, + "loss": 2.1786, + "step": 1037 + }, + { + "epoch": 1.1375342465753424, + "grad_norm": 1.8203203678131104, + "learning_rate": 3.4788789601949634e-06, + "loss": 2.1438, + "step": 1038 + }, + { + "epoch": 1.1386301369863014, + "grad_norm": 2.1084606647491455, + "learning_rate": 3.476848090982941e-06, + "loss": 2.1553, + "step": 1039 + }, + { + "epoch": 1.1397260273972603, + "grad_norm": 2.1041061878204346, + "learning_rate": 3.474817221770918e-06, + "loss": 2.1435, + "step": 1040 + }, + { + "epoch": 1.1397260273972603, + "eval_loss": 2.2970690727233887, + "eval_runtime": 43.2304, + "eval_samples_per_second": 23.132, + "eval_steps_per_second": 0.972, + "step": 1040 + }, + { + "epoch": 1.140821917808219, + "grad_norm": 1.9701104164123535, + "learning_rate": 3.4727863525588957e-06, + "loss": 2.2201, + "step": 1041 + }, + { + "epoch": 1.141917808219178, + "grad_norm": 1.8816475868225098, + "learning_rate": 3.4707554833468725e-06, + "loss": 2.1643, + "step": 1042 + }, + { + "epoch": 1.143013698630137, + "grad_norm": 2.096355438232422, + "learning_rate": 3.4687246141348502e-06, + "loss": 2.1514, + "step": 1043 + }, + { + "epoch": 1.144109589041096, + "grad_norm": 1.9975471496582031, + "learning_rate": 3.466693744922827e-06, + "loss": 2.1246, + "step": 1044 + }, + { + "epoch": 1.1452054794520548, + "grad_norm": 2.0587127208709717, + "learning_rate": 3.464662875710805e-06, + "loss": 2.1371, + "step": 1045 + }, + { + "epoch": 1.1463013698630138, + "grad_norm": 1.9275381565093994, + "learning_rate": 3.4626320064987817e-06, + "loss": 2.1518, + "step": 1046 + }, + { + "epoch": 1.1473972602739726, + "grad_norm": 2.1690948009490967, + "learning_rate": 3.4606011372867594e-06, + "loss": 2.1528, + "step": 1047 + }, + { + "epoch": 1.1484931506849314, + "grad_norm": 1.970973014831543, + "learning_rate": 3.458570268074736e-06, + "loss": 2.1569, + "step": 1048 + }, + { + "epoch": 1.1495890410958904, + "grad_norm": 1.8897690773010254, + "learning_rate": 3.4565393988627135e-06, + "loss": 2.1401, + "step": 1049 + }, + { + "epoch": 1.1506849315068493, + "grad_norm": 1.9245911836624146, + "learning_rate": 3.4545085296506908e-06, + "loss": 2.1311, + "step": 1050 + }, + { + "epoch": 1.1506849315068493, + "eval_loss": 2.2956528663635254, + "eval_runtime": 44.1778, + "eval_samples_per_second": 22.636, + "eval_steps_per_second": 0.951, + "step": 1050 + }, + { + "epoch": 1.1517808219178083, + "grad_norm": 2.142209529876709, + "learning_rate": 3.452477660438668e-06, + "loss": 2.1144, + "step": 1051 + }, + { + "epoch": 1.152876712328767, + "grad_norm": 1.932369351387024, + "learning_rate": 3.4504467912266453e-06, + "loss": 2.1739, + "step": 1052 + }, + { + "epoch": 1.1539726027397261, + "grad_norm": 2.259359359741211, + "learning_rate": 3.4484159220146226e-06, + "loss": 2.1036, + "step": 1053 + }, + { + "epoch": 1.155068493150685, + "grad_norm": 2.112485647201538, + "learning_rate": 3.4463850528026e-06, + "loss": 2.1429, + "step": 1054 + }, + { + "epoch": 1.1561643835616437, + "grad_norm": 2.6211049556732178, + "learning_rate": 3.444354183590577e-06, + "loss": 2.1421, + "step": 1055 + }, + { + "epoch": 1.1572602739726028, + "grad_norm": 1.8724851608276367, + "learning_rate": 3.4423233143785544e-06, + "loss": 2.1622, + "step": 1056 + }, + { + "epoch": 1.1583561643835616, + "grad_norm": 1.9292882680892944, + "learning_rate": 3.4402924451665317e-06, + "loss": 2.1071, + "step": 1057 + }, + { + "epoch": 1.1594520547945206, + "grad_norm": 1.9832521677017212, + "learning_rate": 3.438261575954509e-06, + "loss": 2.0879, + "step": 1058 + }, + { + "epoch": 1.1605479452054794, + "grad_norm": 1.9720780849456787, + "learning_rate": 3.4362307067424862e-06, + "loss": 2.1603, + "step": 1059 + }, + { + "epoch": 1.1616438356164385, + "grad_norm": 1.8179587125778198, + "learning_rate": 3.434199837530463e-06, + "loss": 2.1851, + "step": 1060 + }, + { + "epoch": 1.1616438356164385, + "eval_loss": 2.2950079441070557, + "eval_runtime": 42.9625, + "eval_samples_per_second": 23.276, + "eval_steps_per_second": 0.978, + "step": 1060 + }, + { + "epoch": 1.1627397260273973, + "grad_norm": 1.8762129545211792, + "learning_rate": 3.432168968318441e-06, + "loss": 2.1527, + "step": 1061 + }, + { + "epoch": 1.163835616438356, + "grad_norm": 2.173297643661499, + "learning_rate": 3.4301380991064176e-06, + "loss": 2.1023, + "step": 1062 + }, + { + "epoch": 1.1649315068493151, + "grad_norm": 1.9358222484588623, + "learning_rate": 3.4281072298943953e-06, + "loss": 2.1638, + "step": 1063 + }, + { + "epoch": 1.166027397260274, + "grad_norm": 1.9421939849853516, + "learning_rate": 3.426076360682372e-06, + "loss": 2.1794, + "step": 1064 + }, + { + "epoch": 1.167123287671233, + "grad_norm": 2.212996006011963, + "learning_rate": 3.42404549147035e-06, + "loss": 2.1877, + "step": 1065 + }, + { + "epoch": 1.1682191780821918, + "grad_norm": 2.036180257797241, + "learning_rate": 3.4220146222583267e-06, + "loss": 2.1534, + "step": 1066 + }, + { + "epoch": 1.1693150684931506, + "grad_norm": 1.8819142580032349, + "learning_rate": 3.4199837530463044e-06, + "loss": 2.099, + "step": 1067 + }, + { + "epoch": 1.1704109589041096, + "grad_norm": 2.0781915187835693, + "learning_rate": 3.4179528838342813e-06, + "loss": 2.1326, + "step": 1068 + }, + { + "epoch": 1.1715068493150684, + "grad_norm": 2.034895181655884, + "learning_rate": 3.415922014622259e-06, + "loss": 2.2107, + "step": 1069 + }, + { + "epoch": 1.1726027397260275, + "grad_norm": 1.92604660987854, + "learning_rate": 3.413891145410236e-06, + "loss": 2.1773, + "step": 1070 + }, + { + "epoch": 1.1726027397260275, + "eval_loss": 2.2959954738616943, + "eval_runtime": 47.2123, + "eval_samples_per_second": 21.181, + "eval_steps_per_second": 0.89, + "step": 1070 + }, + { + "epoch": 1.1736986301369863, + "grad_norm": 1.762730598449707, + "learning_rate": 3.4118602761982135e-06, + "loss": 2.1803, + "step": 1071 + }, + { + "epoch": 1.174794520547945, + "grad_norm": 2.0661215782165527, + "learning_rate": 3.4098294069861904e-06, + "loss": 2.1919, + "step": 1072 + }, + { + "epoch": 1.1758904109589041, + "grad_norm": 2.079982280731201, + "learning_rate": 3.4077985377741673e-06, + "loss": 2.1536, + "step": 1073 + }, + { + "epoch": 1.176986301369863, + "grad_norm": 1.9044359922409058, + "learning_rate": 3.405767668562145e-06, + "loss": 2.1403, + "step": 1074 + }, + { + "epoch": 1.178082191780822, + "grad_norm": 1.8870306015014648, + "learning_rate": 3.403736799350122e-06, + "loss": 2.1845, + "step": 1075 + }, + { + "epoch": 1.1791780821917808, + "grad_norm": 1.9305400848388672, + "learning_rate": 3.4017059301380995e-06, + "loss": 2.1786, + "step": 1076 + }, + { + "epoch": 1.1802739726027398, + "grad_norm": 1.987713098526001, + "learning_rate": 3.3996750609260764e-06, + "loss": 2.1214, + "step": 1077 + }, + { + "epoch": 1.1813698630136986, + "grad_norm": 1.8673466444015503, + "learning_rate": 3.397644191714054e-06, + "loss": 2.2027, + "step": 1078 + }, + { + "epoch": 1.1824657534246574, + "grad_norm": 2.009316921234131, + "learning_rate": 3.395613322502031e-06, + "loss": 2.139, + "step": 1079 + }, + { + "epoch": 1.1835616438356165, + "grad_norm": 2.1024956703186035, + "learning_rate": 3.3935824532900086e-06, + "loss": 2.1438, + "step": 1080 + }, + { + "epoch": 1.1835616438356165, + "eval_loss": 2.2966299057006836, + "eval_runtime": 44.8946, + "eval_samples_per_second": 22.274, + "eval_steps_per_second": 0.936, + "step": 1080 + }, + { + "epoch": 1.1846575342465753, + "grad_norm": 1.8911967277526855, + "learning_rate": 3.3915515840779855e-06, + "loss": 2.1238, + "step": 1081 + }, + { + "epoch": 1.1857534246575343, + "grad_norm": 2.0649192333221436, + "learning_rate": 3.389520714865963e-06, + "loss": 2.1534, + "step": 1082 + }, + { + "epoch": 1.1868493150684931, + "grad_norm": 1.9236847162246704, + "learning_rate": 3.38748984565394e-06, + "loss": 2.1427, + "step": 1083 + }, + { + "epoch": 1.1879452054794521, + "grad_norm": 1.8662182092666626, + "learning_rate": 3.3854589764419177e-06, + "loss": 2.1699, + "step": 1084 + }, + { + "epoch": 1.189041095890411, + "grad_norm": 1.9226020574569702, + "learning_rate": 3.3834281072298946e-06, + "loss": 2.1192, + "step": 1085 + }, + { + "epoch": 1.1901369863013698, + "grad_norm": 1.9659301042556763, + "learning_rate": 3.381397238017872e-06, + "loss": 2.1529, + "step": 1086 + }, + { + "epoch": 1.1912328767123288, + "grad_norm": 2.2714693546295166, + "learning_rate": 3.379366368805849e-06, + "loss": 2.1273, + "step": 1087 + }, + { + "epoch": 1.1923287671232876, + "grad_norm": 2.1853270530700684, + "learning_rate": 3.3773354995938264e-06, + "loss": 2.2029, + "step": 1088 + }, + { + "epoch": 1.1934246575342466, + "grad_norm": 1.8302830457687378, + "learning_rate": 3.3753046303818037e-06, + "loss": 2.1599, + "step": 1089 + }, + { + "epoch": 1.1945205479452055, + "grad_norm": 2.005591869354248, + "learning_rate": 3.373273761169781e-06, + "loss": 2.139, + "step": 1090 + }, + { + "epoch": 1.1945205479452055, + "eval_loss": 2.295994520187378, + "eval_runtime": 45.4802, + "eval_samples_per_second": 21.988, + "eval_steps_per_second": 0.923, + "step": 1090 + }, + { + "epoch": 1.1956164383561645, + "grad_norm": 2.139267683029175, + "learning_rate": 3.3712428919577582e-06, + "loss": 2.151, + "step": 1091 + }, + { + "epoch": 1.1967123287671233, + "grad_norm": 1.8547207117080688, + "learning_rate": 3.3692120227457355e-06, + "loss": 2.1249, + "step": 1092 + }, + { + "epoch": 1.1978082191780821, + "grad_norm": 2.214583396911621, + "learning_rate": 3.3671811535337128e-06, + "loss": 2.1656, + "step": 1093 + }, + { + "epoch": 1.1989041095890411, + "grad_norm": 2.1608800888061523, + "learning_rate": 3.36515028432169e-06, + "loss": 2.1056, + "step": 1094 + }, + { + "epoch": 1.2, + "grad_norm": 2.0850605964660645, + "learning_rate": 3.3631194151096673e-06, + "loss": 2.1215, + "step": 1095 + }, + { + "epoch": 1.201095890410959, + "grad_norm": 1.9199103116989136, + "learning_rate": 3.3610885458976446e-06, + "loss": 2.1584, + "step": 1096 + }, + { + "epoch": 1.2021917808219178, + "grad_norm": 1.9766334295272827, + "learning_rate": 3.359057676685622e-06, + "loss": 2.1821, + "step": 1097 + }, + { + "epoch": 1.2032876712328768, + "grad_norm": 1.948988676071167, + "learning_rate": 3.357026807473599e-06, + "loss": 2.182, + "step": 1098 + }, + { + "epoch": 1.2043835616438356, + "grad_norm": 2.0419397354125977, + "learning_rate": 3.354995938261576e-06, + "loss": 2.1051, + "step": 1099 + }, + { + "epoch": 1.2054794520547945, + "grad_norm": 1.9751213788986206, + "learning_rate": 3.3529650690495537e-06, + "loss": 2.1255, + "step": 1100 + }, + { + "epoch": 1.2054794520547945, + "eval_loss": 2.293811798095703, + "eval_runtime": 46.1182, + "eval_samples_per_second": 21.683, + "eval_steps_per_second": 0.911, + "step": 1100 + }, + { + "epoch": 1.2065753424657535, + "grad_norm": 1.9806510210037231, + "learning_rate": 3.3509341998375305e-06, + "loss": 2.1021, + "step": 1101 + }, + { + "epoch": 1.2076712328767123, + "grad_norm": 2.0570414066314697, + "learning_rate": 3.3489033306255082e-06, + "loss": 2.1417, + "step": 1102 + }, + { + "epoch": 1.208767123287671, + "grad_norm": 1.868134617805481, + "learning_rate": 3.346872461413485e-06, + "loss": 2.1494, + "step": 1103 + }, + { + "epoch": 1.2098630136986301, + "grad_norm": 2.0156848430633545, + "learning_rate": 3.344841592201463e-06, + "loss": 2.1238, + "step": 1104 + }, + { + "epoch": 1.210958904109589, + "grad_norm": 2.082301616668701, + "learning_rate": 3.3428107229894397e-06, + "loss": 2.1105, + "step": 1105 + }, + { + "epoch": 1.212054794520548, + "grad_norm": 1.8853904008865356, + "learning_rate": 3.3407798537774174e-06, + "loss": 2.1406, + "step": 1106 + }, + { + "epoch": 1.2131506849315068, + "grad_norm": 2.0410518646240234, + "learning_rate": 3.338748984565394e-06, + "loss": 2.1567, + "step": 1107 + }, + { + "epoch": 1.2142465753424658, + "grad_norm": 1.9901232719421387, + "learning_rate": 3.336718115353372e-06, + "loss": 2.1778, + "step": 1108 + }, + { + "epoch": 1.2153424657534246, + "grad_norm": 1.9840792417526245, + "learning_rate": 3.3346872461413488e-06, + "loss": 2.1935, + "step": 1109 + }, + { + "epoch": 1.2164383561643834, + "grad_norm": 1.8854445219039917, + "learning_rate": 3.3326563769293256e-06, + "loss": 2.137, + "step": 1110 + }, + { + "epoch": 1.2164383561643834, + "eval_loss": 2.294447660446167, + "eval_runtime": 42.6423, + "eval_samples_per_second": 23.451, + "eval_steps_per_second": 0.985, + "step": 1110 + }, + { + "epoch": 1.2175342465753425, + "grad_norm": 2.0948665142059326, + "learning_rate": 3.3306255077173033e-06, + "loss": 2.1288, + "step": 1111 + }, + { + "epoch": 1.2186301369863013, + "grad_norm": 1.859448790550232, + "learning_rate": 3.32859463850528e-06, + "loss": 2.1403, + "step": 1112 + }, + { + "epoch": 1.2197260273972603, + "grad_norm": 1.92074453830719, + "learning_rate": 3.326563769293258e-06, + "loss": 2.141, + "step": 1113 + }, + { + "epoch": 1.2208219178082191, + "grad_norm": 1.8377320766448975, + "learning_rate": 3.3245329000812347e-06, + "loss": 2.0889, + "step": 1114 + }, + { + "epoch": 1.2219178082191782, + "grad_norm": 1.9320917129516602, + "learning_rate": 3.3225020308692124e-06, + "loss": 2.1254, + "step": 1115 + }, + { + "epoch": 1.223013698630137, + "grad_norm": 1.9691251516342163, + "learning_rate": 3.3204711616571893e-06, + "loss": 2.1409, + "step": 1116 + }, + { + "epoch": 1.2241095890410958, + "grad_norm": 1.8214130401611328, + "learning_rate": 3.318440292445167e-06, + "loss": 2.1046, + "step": 1117 + }, + { + "epoch": 1.2252054794520548, + "grad_norm": 1.959455966949463, + "learning_rate": 3.316409423233144e-06, + "loss": 2.1418, + "step": 1118 + }, + { + "epoch": 1.2263013698630136, + "grad_norm": 1.9920449256896973, + "learning_rate": 3.3143785540211215e-06, + "loss": 2.1192, + "step": 1119 + }, + { + "epoch": 1.2273972602739727, + "grad_norm": 2.0276854038238525, + "learning_rate": 3.3123476848090984e-06, + "loss": 2.1939, + "step": 1120 + }, + { + "epoch": 1.2273972602739727, + "eval_loss": 2.295330047607422, + "eval_runtime": 49.8691, + "eval_samples_per_second": 20.052, + "eval_steps_per_second": 0.842, + "step": 1120 + }, + { + "epoch": 1.2284931506849315, + "grad_norm": 1.9765129089355469, + "learning_rate": 3.310316815597076e-06, + "loss": 2.1545, + "step": 1121 + }, + { + "epoch": 1.2295890410958905, + "grad_norm": 1.8449276685714722, + "learning_rate": 3.308285946385053e-06, + "loss": 2.1459, + "step": 1122 + }, + { + "epoch": 1.2306849315068493, + "grad_norm": 1.953072190284729, + "learning_rate": 3.30625507717303e-06, + "loss": 2.2046, + "step": 1123 + }, + { + "epoch": 1.2317808219178081, + "grad_norm": 1.9570913314819336, + "learning_rate": 3.3042242079610075e-06, + "loss": 2.1015, + "step": 1124 + }, + { + "epoch": 1.2328767123287672, + "grad_norm": 2.00569224357605, + "learning_rate": 3.3021933387489847e-06, + "loss": 2.1681, + "step": 1125 + }, + { + "epoch": 1.233972602739726, + "grad_norm": 1.9131590127944946, + "learning_rate": 3.300162469536962e-06, + "loss": 2.1765, + "step": 1126 + }, + { + "epoch": 1.235068493150685, + "grad_norm": 1.8375864028930664, + "learning_rate": 3.2981316003249393e-06, + "loss": 2.1175, + "step": 1127 + }, + { + "epoch": 1.2361643835616438, + "grad_norm": 1.852847933769226, + "learning_rate": 3.2961007311129166e-06, + "loss": 2.1838, + "step": 1128 + }, + { + "epoch": 1.2372602739726029, + "grad_norm": 1.8772693872451782, + "learning_rate": 3.294069861900894e-06, + "loss": 2.1524, + "step": 1129 + }, + { + "epoch": 1.2383561643835617, + "grad_norm": 1.8593205213546753, + "learning_rate": 3.292038992688871e-06, + "loss": 2.2059, + "step": 1130 + }, + { + "epoch": 1.2383561643835617, + "eval_loss": 2.294188976287842, + "eval_runtime": 50.4549, + "eval_samples_per_second": 19.82, + "eval_steps_per_second": 0.832, + "step": 1130 + }, + { + "epoch": 1.2394520547945205, + "grad_norm": 1.8967972993850708, + "learning_rate": 3.2900081234768484e-06, + "loss": 2.1551, + "step": 1131 + }, + { + "epoch": 1.2405479452054795, + "grad_norm": 1.9799107313156128, + "learning_rate": 3.2879772542648257e-06, + "loss": 2.1644, + "step": 1132 + }, + { + "epoch": 1.2416438356164383, + "grad_norm": 1.978883147239685, + "learning_rate": 3.285946385052803e-06, + "loss": 2.182, + "step": 1133 + }, + { + "epoch": 1.2427397260273974, + "grad_norm": 1.8705304861068726, + "learning_rate": 3.2839155158407802e-06, + "loss": 2.0981, + "step": 1134 + }, + { + "epoch": 1.2438356164383562, + "grad_norm": 2.0146889686584473, + "learning_rate": 3.2818846466287575e-06, + "loss": 2.1686, + "step": 1135 + }, + { + "epoch": 1.244931506849315, + "grad_norm": 1.94461989402771, + "learning_rate": 3.2798537774167344e-06, + "loss": 2.093, + "step": 1136 + }, + { + "epoch": 1.246027397260274, + "grad_norm": 1.7823240756988525, + "learning_rate": 3.277822908204712e-06, + "loss": 2.198, + "step": 1137 + }, + { + "epoch": 1.2471232876712328, + "grad_norm": 1.89297354221344, + "learning_rate": 3.275792038992689e-06, + "loss": 2.1359, + "step": 1138 + }, + { + "epoch": 1.2482191780821918, + "grad_norm": 1.8670459985733032, + "learning_rate": 3.2737611697806666e-06, + "loss": 2.0956, + "step": 1139 + }, + { + "epoch": 1.2493150684931507, + "grad_norm": 1.813568353652954, + "learning_rate": 3.2717303005686435e-06, + "loss": 2.1448, + "step": 1140 + }, + { + "epoch": 1.2493150684931507, + "eval_loss": 2.295804023742676, + "eval_runtime": 53.286, + "eval_samples_per_second": 18.767, + "eval_steps_per_second": 0.788, + "step": 1140 + }, + { + "epoch": 1.2504109589041095, + "grad_norm": 1.912916898727417, + "learning_rate": 3.269699431356621e-06, + "loss": 2.1159, + "step": 1141 + }, + { + "epoch": 1.2515068493150685, + "grad_norm": 1.7939883470535278, + "learning_rate": 3.267668562144598e-06, + "loss": 2.1566, + "step": 1142 + }, + { + "epoch": 1.2526027397260273, + "grad_norm": 1.8494806289672852, + "learning_rate": 3.2656376929325757e-06, + "loss": 2.1496, + "step": 1143 + }, + { + "epoch": 1.2536986301369863, + "grad_norm": 1.8468523025512695, + "learning_rate": 3.2636068237205526e-06, + "loss": 2.1159, + "step": 1144 + }, + { + "epoch": 1.2547945205479452, + "grad_norm": 1.8962055444717407, + "learning_rate": 3.2615759545085303e-06, + "loss": 2.1712, + "step": 1145 + }, + { + "epoch": 1.2558904109589042, + "grad_norm": 1.800353765487671, + "learning_rate": 3.259545085296507e-06, + "loss": 2.1486, + "step": 1146 + }, + { + "epoch": 1.256986301369863, + "grad_norm": 2.124359607696533, + "learning_rate": 3.257514216084484e-06, + "loss": 2.1913, + "step": 1147 + }, + { + "epoch": 1.2580821917808218, + "grad_norm": 1.9248145818710327, + "learning_rate": 3.2554833468724617e-06, + "loss": 2.1128, + "step": 1148 + }, + { + "epoch": 1.2591780821917808, + "grad_norm": 1.9990876913070679, + "learning_rate": 3.2534524776604385e-06, + "loss": 2.1854, + "step": 1149 + }, + { + "epoch": 1.2602739726027397, + "grad_norm": 1.8600568771362305, + "learning_rate": 3.2514216084484162e-06, + "loss": 2.1565, + "step": 1150 + }, + { + "epoch": 1.2602739726027397, + "eval_loss": 2.2942006587982178, + "eval_runtime": 51.4021, + "eval_samples_per_second": 19.454, + "eval_steps_per_second": 0.817, + "step": 1150 + }, + { + "epoch": 1.2613698630136987, + "grad_norm": 1.8621101379394531, + "learning_rate": 3.249390739236393e-06, + "loss": 2.1438, + "step": 1151 + }, + { + "epoch": 1.2624657534246575, + "grad_norm": 1.9290356636047363, + "learning_rate": 3.2473598700243708e-06, + "loss": 2.1577, + "step": 1152 + }, + { + "epoch": 1.2635616438356165, + "grad_norm": 1.9007307291030884, + "learning_rate": 3.2453290008123476e-06, + "loss": 2.1295, + "step": 1153 + }, + { + "epoch": 1.2646575342465753, + "grad_norm": 1.7647960186004639, + "learning_rate": 3.2432981316003253e-06, + "loss": 2.1472, + "step": 1154 + }, + { + "epoch": 1.2657534246575342, + "grad_norm": 1.9796825647354126, + "learning_rate": 3.241267262388302e-06, + "loss": 2.1542, + "step": 1155 + }, + { + "epoch": 1.2668493150684932, + "grad_norm": 1.9505796432495117, + "learning_rate": 3.23923639317628e-06, + "loss": 2.1093, + "step": 1156 + }, + { + "epoch": 1.267945205479452, + "grad_norm": 1.9152203798294067, + "learning_rate": 3.2372055239642567e-06, + "loss": 2.1456, + "step": 1157 + }, + { + "epoch": 1.269041095890411, + "grad_norm": 1.9094963073730469, + "learning_rate": 3.2351746547522344e-06, + "loss": 2.1149, + "step": 1158 + }, + { + "epoch": 1.2701369863013698, + "grad_norm": 1.8872679471969604, + "learning_rate": 3.2331437855402113e-06, + "loss": 2.1545, + "step": 1159 + }, + { + "epoch": 1.2712328767123289, + "grad_norm": 2.171703338623047, + "learning_rate": 3.2311129163281885e-06, + "loss": 2.2108, + "step": 1160 + }, + { + "epoch": 1.2712328767123289, + "eval_loss": 2.2910356521606445, + "eval_runtime": 52.5176, + "eval_samples_per_second": 19.041, + "eval_steps_per_second": 0.8, + "step": 1160 + }, + { + "epoch": 1.2723287671232877, + "grad_norm": 1.8578304052352905, + "learning_rate": 3.229082047116166e-06, + "loss": 2.1158, + "step": 1161 + }, + { + "epoch": 1.2734246575342465, + "grad_norm": 1.8873049020767212, + "learning_rate": 3.227051177904143e-06, + "loss": 2.1618, + "step": 1162 + }, + { + "epoch": 1.2745205479452055, + "grad_norm": 1.8442409038543701, + "learning_rate": 3.2250203086921204e-06, + "loss": 2.1454, + "step": 1163 + }, + { + "epoch": 1.2756164383561643, + "grad_norm": 1.9609211683273315, + "learning_rate": 3.2229894394800977e-06, + "loss": 2.1519, + "step": 1164 + }, + { + "epoch": 1.2767123287671232, + "grad_norm": 1.9093680381774902, + "learning_rate": 3.220958570268075e-06, + "loss": 2.1775, + "step": 1165 + }, + { + "epoch": 1.2778082191780822, + "grad_norm": 1.8355298042297363, + "learning_rate": 3.218927701056052e-06, + "loss": 2.0871, + "step": 1166 + }, + { + "epoch": 1.2789041095890412, + "grad_norm": 1.9497056007385254, + "learning_rate": 3.2168968318440295e-06, + "loss": 2.1441, + "step": 1167 + }, + { + "epoch": 1.28, + "grad_norm": 1.8635729551315308, + "learning_rate": 3.2148659626320068e-06, + "loss": 2.0748, + "step": 1168 + }, + { + "epoch": 1.2810958904109588, + "grad_norm": 2.0972132682800293, + "learning_rate": 3.212835093419984e-06, + "loss": 2.1843, + "step": 1169 + }, + { + "epoch": 1.2821917808219179, + "grad_norm": 1.9235926866531372, + "learning_rate": 3.2108042242079613e-06, + "loss": 2.1194, + "step": 1170 + }, + { + "epoch": 1.2821917808219179, + "eval_loss": 2.291309118270874, + "eval_runtime": 55.6408, + "eval_samples_per_second": 17.972, + "eval_steps_per_second": 0.755, + "step": 1170 + }, + { + "epoch": 1.2832876712328767, + "grad_norm": 1.8674591779708862, + "learning_rate": 3.2087733549959386e-06, + "loss": 2.1264, + "step": 1171 + }, + { + "epoch": 1.2843835616438355, + "grad_norm": 1.810589075088501, + "learning_rate": 3.206742485783916e-06, + "loss": 2.1781, + "step": 1172 + }, + { + "epoch": 1.2854794520547945, + "grad_norm": 1.9794505834579468, + "learning_rate": 3.2047116165718927e-06, + "loss": 2.1089, + "step": 1173 + }, + { + "epoch": 1.2865753424657536, + "grad_norm": 1.8487070798873901, + "learning_rate": 3.2026807473598704e-06, + "loss": 2.097, + "step": 1174 + }, + { + "epoch": 1.2876712328767124, + "grad_norm": 1.9668854475021362, + "learning_rate": 3.2006498781478473e-06, + "loss": 2.1516, + "step": 1175 + }, + { + "epoch": 1.2887671232876712, + "grad_norm": 1.855817198753357, + "learning_rate": 3.198619008935825e-06, + "loss": 2.1678, + "step": 1176 + }, + { + "epoch": 1.2898630136986302, + "grad_norm": 2.381211757659912, + "learning_rate": 3.196588139723802e-06, + "loss": 2.099, + "step": 1177 + }, + { + "epoch": 1.290958904109589, + "grad_norm": 2.1524250507354736, + "learning_rate": 3.1945572705117795e-06, + "loss": 2.1345, + "step": 1178 + }, + { + "epoch": 1.2920547945205478, + "grad_norm": 1.9120781421661377, + "learning_rate": 3.1925264012997564e-06, + "loss": 2.1586, + "step": 1179 + }, + { + "epoch": 1.2931506849315069, + "grad_norm": 2.153817653656006, + "learning_rate": 3.190495532087734e-06, + "loss": 2.1597, + "step": 1180 + }, + { + "epoch": 1.2931506849315069, + "eval_loss": 2.2936320304870605, + "eval_runtime": 53.6426, + "eval_samples_per_second": 18.642, + "eval_steps_per_second": 0.783, + "step": 1180 + }, + { + "epoch": 1.2942465753424657, + "grad_norm": 2.074873447418213, + "learning_rate": 3.188464662875711e-06, + "loss": 2.1572, + "step": 1181 + }, + { + "epoch": 1.2953424657534247, + "grad_norm": 1.9313570261001587, + "learning_rate": 3.1864337936636886e-06, + "loss": 2.1986, + "step": 1182 + }, + { + "epoch": 1.2964383561643835, + "grad_norm": 1.9836641550064087, + "learning_rate": 3.1844029244516655e-06, + "loss": 2.13, + "step": 1183 + }, + { + "epoch": 1.2975342465753426, + "grad_norm": 2.0893137454986572, + "learning_rate": 3.182372055239643e-06, + "loss": 2.1138, + "step": 1184 + }, + { + "epoch": 1.2986301369863014, + "grad_norm": 2.1667087078094482, + "learning_rate": 3.18034118602762e-06, + "loss": 2.1119, + "step": 1185 + }, + { + "epoch": 1.2997260273972602, + "grad_norm": 2.093961000442505, + "learning_rate": 3.1783103168155973e-06, + "loss": 2.1728, + "step": 1186 + }, + { + "epoch": 1.3008219178082192, + "grad_norm": 1.919267177581787, + "learning_rate": 3.1762794476035746e-06, + "loss": 2.1469, + "step": 1187 + }, + { + "epoch": 1.301917808219178, + "grad_norm": 2.017967939376831, + "learning_rate": 3.174248578391552e-06, + "loss": 2.1307, + "step": 1188 + }, + { + "epoch": 1.303013698630137, + "grad_norm": 1.921632170677185, + "learning_rate": 3.172217709179529e-06, + "loss": 2.13, + "step": 1189 + }, + { + "epoch": 1.3041095890410959, + "grad_norm": 2.0171291828155518, + "learning_rate": 3.1701868399675064e-06, + "loss": 2.135, + "step": 1190 + }, + { + "epoch": 1.3041095890410959, + "eval_loss": 2.2934045791625977, + "eval_runtime": 48.0821, + "eval_samples_per_second": 20.798, + "eval_steps_per_second": 0.874, + "step": 1190 + }, + { + "epoch": 1.305205479452055, + "grad_norm": 1.8685601949691772, + "learning_rate": 3.1681559707554837e-06, + "loss": 2.1329, + "step": 1191 + }, + { + "epoch": 1.3063013698630137, + "grad_norm": 1.8909096717834473, + "learning_rate": 3.166125101543461e-06, + "loss": 2.1004, + "step": 1192 + }, + { + "epoch": 1.3073972602739725, + "grad_norm": 1.798614740371704, + "learning_rate": 3.1640942323314382e-06, + "loss": 2.1555, + "step": 1193 + }, + { + "epoch": 1.3084931506849315, + "grad_norm": 1.8296692371368408, + "learning_rate": 3.1620633631194155e-06, + "loss": 2.1549, + "step": 1194 + }, + { + "epoch": 1.3095890410958904, + "grad_norm": 1.8676908016204834, + "learning_rate": 3.1600324939073928e-06, + "loss": 2.1081, + "step": 1195 + }, + { + "epoch": 1.3106849315068494, + "grad_norm": 1.8682451248168945, + "learning_rate": 3.15800162469537e-06, + "loss": 2.138, + "step": 1196 + }, + { + "epoch": 1.3117808219178082, + "grad_norm": 2.0049314498901367, + "learning_rate": 3.155970755483347e-06, + "loss": 2.144, + "step": 1197 + }, + { + "epoch": 1.3128767123287672, + "grad_norm": 2.00691556930542, + "learning_rate": 3.1539398862713246e-06, + "loss": 2.1539, + "step": 1198 + }, + { + "epoch": 1.313972602739726, + "grad_norm": 1.8521753549575806, + "learning_rate": 3.1519090170593015e-06, + "loss": 2.1597, + "step": 1199 + }, + { + "epoch": 1.3150684931506849, + "grad_norm": 2.0974042415618896, + "learning_rate": 3.149878147847279e-06, + "loss": 2.147, + "step": 1200 + }, + { + "epoch": 1.3150684931506849, + "eval_loss": 2.2891385555267334, + "eval_runtime": 48.1618, + "eval_samples_per_second": 20.763, + "eval_steps_per_second": 0.872, + "step": 1200 + }, + { + "epoch": 1.316164383561644, + "grad_norm": 1.9671052694320679, + "learning_rate": 3.147847278635256e-06, + "loss": 2.1838, + "step": 1201 + }, + { + "epoch": 1.3172602739726027, + "grad_norm": 1.9421573877334595, + "learning_rate": 3.1458164094232337e-06, + "loss": 2.2093, + "step": 1202 + }, + { + "epoch": 1.3183561643835615, + "grad_norm": 1.917198657989502, + "learning_rate": 3.1437855402112106e-06, + "loss": 2.136, + "step": 1203 + }, + { + "epoch": 1.3194520547945205, + "grad_norm": 1.8664982318878174, + "learning_rate": 3.1417546709991883e-06, + "loss": 2.089, + "step": 1204 + }, + { + "epoch": 1.3205479452054796, + "grad_norm": 2.043558359146118, + "learning_rate": 3.139723801787165e-06, + "loss": 2.159, + "step": 1205 + }, + { + "epoch": 1.3216438356164384, + "grad_norm": 1.8314157724380493, + "learning_rate": 3.137692932575143e-06, + "loss": 2.13, + "step": 1206 + }, + { + "epoch": 1.3227397260273972, + "grad_norm": 1.907884120941162, + "learning_rate": 3.1356620633631197e-06, + "loss": 2.1746, + "step": 1207 + }, + { + "epoch": 1.3238356164383562, + "grad_norm": 2.0570952892303467, + "learning_rate": 3.1336311941510974e-06, + "loss": 2.1187, + "step": 1208 + }, + { + "epoch": 1.324931506849315, + "grad_norm": 1.832292914390564, + "learning_rate": 3.1316003249390742e-06, + "loss": 2.1425, + "step": 1209 + }, + { + "epoch": 1.3260273972602739, + "grad_norm": 1.8307697772979736, + "learning_rate": 3.129569455727051e-06, + "loss": 2.1749, + "step": 1210 + }, + { + "epoch": 1.3260273972602739, + "eval_loss": 2.2865941524505615, + "eval_runtime": 42.2877, + "eval_samples_per_second": 23.648, + "eval_steps_per_second": 0.993, + "step": 1210 + }, + { + "epoch": 1.3271232876712329, + "grad_norm": 1.8292433023452759, + "learning_rate": 3.1275385865150288e-06, + "loss": 2.1702, + "step": 1211 + }, + { + "epoch": 1.3282191780821917, + "grad_norm": 1.9146367311477661, + "learning_rate": 3.1255077173030056e-06, + "loss": 2.16, + "step": 1212 + }, + { + "epoch": 1.3293150684931507, + "grad_norm": 1.809275507926941, + "learning_rate": 3.1234768480909833e-06, + "loss": 2.157, + "step": 1213 + }, + { + "epoch": 1.3304109589041095, + "grad_norm": 1.939081072807312, + "learning_rate": 3.12144597887896e-06, + "loss": 2.1496, + "step": 1214 + }, + { + "epoch": 1.3315068493150686, + "grad_norm": 1.8665997982025146, + "learning_rate": 3.119415109666938e-06, + "loss": 2.1535, + "step": 1215 + }, + { + "epoch": 1.3326027397260274, + "grad_norm": 1.8455616235733032, + "learning_rate": 3.1173842404549147e-06, + "loss": 2.1845, + "step": 1216 + }, + { + "epoch": 1.3336986301369862, + "grad_norm": 2.050560235977173, + "learning_rate": 3.1153533712428924e-06, + "loss": 2.1461, + "step": 1217 + }, + { + "epoch": 1.3347945205479452, + "grad_norm": 1.8440574407577515, + "learning_rate": 3.1133225020308693e-06, + "loss": 2.1476, + "step": 1218 + }, + { + "epoch": 1.335890410958904, + "grad_norm": 1.8801573514938354, + "learning_rate": 3.111291632818847e-06, + "loss": 2.1721, + "step": 1219 + }, + { + "epoch": 1.336986301369863, + "grad_norm": 1.9227471351623535, + "learning_rate": 3.109260763606824e-06, + "loss": 2.178, + "step": 1220 + }, + { + "epoch": 1.336986301369863, + "eval_loss": 2.2896478176116943, + "eval_runtime": 43.157, + "eval_samples_per_second": 23.171, + "eval_steps_per_second": 0.973, + "step": 1220 + }, + { + "epoch": 1.3380821917808219, + "grad_norm": 1.9979732036590576, + "learning_rate": 3.1072298943948015e-06, + "loss": 2.148, + "step": 1221 + }, + { + "epoch": 1.339178082191781, + "grad_norm": 1.9481284618377686, + "learning_rate": 3.1051990251827784e-06, + "loss": 2.165, + "step": 1222 + }, + { + "epoch": 1.3402739726027397, + "grad_norm": 1.8830820322036743, + "learning_rate": 3.1031681559707557e-06, + "loss": 2.1239, + "step": 1223 + }, + { + "epoch": 1.3413698630136985, + "grad_norm": 1.890512466430664, + "learning_rate": 3.101137286758733e-06, + "loss": 2.1761, + "step": 1224 + }, + { + "epoch": 1.3424657534246576, + "grad_norm": 2.0024659633636475, + "learning_rate": 3.09910641754671e-06, + "loss": 2.179, + "step": 1225 + }, + { + "epoch": 1.3435616438356164, + "grad_norm": 1.8659477233886719, + "learning_rate": 3.0970755483346875e-06, + "loss": 2.1722, + "step": 1226 + }, + { + "epoch": 1.3446575342465754, + "grad_norm": 1.93405020236969, + "learning_rate": 3.0950446791226648e-06, + "loss": 2.1545, + "step": 1227 + }, + { + "epoch": 1.3457534246575342, + "grad_norm": 1.9514124393463135, + "learning_rate": 3.093013809910642e-06, + "loss": 2.187, + "step": 1228 + }, + { + "epoch": 1.3468493150684933, + "grad_norm": 1.9571877717971802, + "learning_rate": 3.0909829406986193e-06, + "loss": 2.1507, + "step": 1229 + }, + { + "epoch": 1.347945205479452, + "grad_norm": 1.9434577226638794, + "learning_rate": 3.0889520714865966e-06, + "loss": 2.1953, + "step": 1230 + }, + { + "epoch": 1.347945205479452, + "eval_loss": 2.289128065109253, + "eval_runtime": 46.7151, + "eval_samples_per_second": 21.406, + "eval_steps_per_second": 0.899, + "step": 1230 + }, + { + "epoch": 1.3490410958904109, + "grad_norm": 1.8294577598571777, + "learning_rate": 3.086921202274574e-06, + "loss": 2.1684, + "step": 1231 + }, + { + "epoch": 1.35013698630137, + "grad_norm": 1.9432605504989624, + "learning_rate": 3.084890333062551e-06, + "loss": 2.1393, + "step": 1232 + }, + { + "epoch": 1.3512328767123287, + "grad_norm": 1.9030417203903198, + "learning_rate": 3.0828594638505284e-06, + "loss": 2.1181, + "step": 1233 + }, + { + "epoch": 1.3523287671232875, + "grad_norm": 1.9233827590942383, + "learning_rate": 3.0808285946385057e-06, + "loss": 2.1479, + "step": 1234 + }, + { + "epoch": 1.3534246575342466, + "grad_norm": 1.9549237489700317, + "learning_rate": 3.078797725426483e-06, + "loss": 2.1253, + "step": 1235 + }, + { + "epoch": 1.3545205479452056, + "grad_norm": 2.0584867000579834, + "learning_rate": 3.07676685621446e-06, + "loss": 2.0846, + "step": 1236 + }, + { + "epoch": 1.3556164383561644, + "grad_norm": 1.7543913125991821, + "learning_rate": 3.0747359870024375e-06, + "loss": 2.1495, + "step": 1237 + }, + { + "epoch": 1.3567123287671232, + "grad_norm": 1.750698447227478, + "learning_rate": 3.0727051177904144e-06, + "loss": 2.1179, + "step": 1238 + }, + { + "epoch": 1.3578082191780823, + "grad_norm": 1.9867119789123535, + "learning_rate": 3.070674248578392e-06, + "loss": 2.1313, + "step": 1239 + }, + { + "epoch": 1.358904109589041, + "grad_norm": 1.8228135108947754, + "learning_rate": 3.068643379366369e-06, + "loss": 2.158, + "step": 1240 + }, + { + "epoch": 1.358904109589041, + "eval_loss": 2.2861993312835693, + "eval_runtime": 44.081, + "eval_samples_per_second": 22.686, + "eval_steps_per_second": 0.953, + "step": 1240 + }, + { + "epoch": 1.3599999999999999, + "grad_norm": 1.8551791906356812, + "learning_rate": 3.0666125101543466e-06, + "loss": 2.129, + "step": 1241 + }, + { + "epoch": 1.361095890410959, + "grad_norm": 1.8729921579360962, + "learning_rate": 3.0645816409423235e-06, + "loss": 2.1069, + "step": 1242 + }, + { + "epoch": 1.362191780821918, + "grad_norm": 1.8203959465026855, + "learning_rate": 3.062550771730301e-06, + "loss": 2.1509, + "step": 1243 + }, + { + "epoch": 1.3632876712328768, + "grad_norm": 1.9498761892318726, + "learning_rate": 3.060519902518278e-06, + "loss": 2.1336, + "step": 1244 + }, + { + "epoch": 1.3643835616438356, + "grad_norm": 2.134521245956421, + "learning_rate": 3.0584890333062557e-06, + "loss": 2.1294, + "step": 1245 + }, + { + "epoch": 1.3654794520547946, + "grad_norm": 1.9321913719177246, + "learning_rate": 3.0564581640942326e-06, + "loss": 2.0823, + "step": 1246 + }, + { + "epoch": 1.3665753424657534, + "grad_norm": 1.8948324918746948, + "learning_rate": 3.0544272948822094e-06, + "loss": 2.1425, + "step": 1247 + }, + { + "epoch": 1.3676712328767122, + "grad_norm": 2.2063205242156982, + "learning_rate": 3.052396425670187e-06, + "loss": 2.1242, + "step": 1248 + }, + { + "epoch": 1.3687671232876713, + "grad_norm": 2.0060935020446777, + "learning_rate": 3.050365556458164e-06, + "loss": 2.1586, + "step": 1249 + }, + { + "epoch": 1.36986301369863, + "grad_norm": 1.8441938161849976, + "learning_rate": 3.0483346872461417e-06, + "loss": 2.1305, + "step": 1250 + }, + { + "epoch": 1.36986301369863, + "eval_loss": 2.2872142791748047, + "eval_runtime": 43.4998, + "eval_samples_per_second": 22.989, + "eval_steps_per_second": 0.966, + "step": 1250 + }, + { + "epoch": 1.370958904109589, + "grad_norm": 2.019545793533325, + "learning_rate": 3.0463038180341185e-06, + "loss": 2.1257, + "step": 1251 + }, + { + "epoch": 1.372054794520548, + "grad_norm": 1.981020212173462, + "learning_rate": 3.0442729488220962e-06, + "loss": 2.155, + "step": 1252 + }, + { + "epoch": 1.373150684931507, + "grad_norm": 1.9707001447677612, + "learning_rate": 3.042242079610073e-06, + "loss": 2.1671, + "step": 1253 + }, + { + "epoch": 1.3742465753424657, + "grad_norm": 1.8645458221435547, + "learning_rate": 3.0402112103980508e-06, + "loss": 2.1655, + "step": 1254 + }, + { + "epoch": 1.3753424657534246, + "grad_norm": 1.863911747932434, + "learning_rate": 3.0381803411860276e-06, + "loss": 2.0917, + "step": 1255 + }, + { + "epoch": 1.3764383561643836, + "grad_norm": 2.1079251766204834, + "learning_rate": 3.0361494719740053e-06, + "loss": 2.119, + "step": 1256 + }, + { + "epoch": 1.3775342465753424, + "grad_norm": 1.8242024183273315, + "learning_rate": 3.034118602761982e-06, + "loss": 2.1398, + "step": 1257 + }, + { + "epoch": 1.3786301369863014, + "grad_norm": 1.8663142919540405, + "learning_rate": 3.03208773354996e-06, + "loss": 2.1405, + "step": 1258 + }, + { + "epoch": 1.3797260273972602, + "grad_norm": 1.9437193870544434, + "learning_rate": 3.0300568643379367e-06, + "loss": 2.1156, + "step": 1259 + }, + { + "epoch": 1.3808219178082193, + "grad_norm": 2.0837693214416504, + "learning_rate": 3.028025995125914e-06, + "loss": 2.1731, + "step": 1260 + }, + { + "epoch": 1.3808219178082193, + "eval_loss": 2.283296823501587, + "eval_runtime": 45.341, + "eval_samples_per_second": 22.055, + "eval_steps_per_second": 0.926, + "step": 1260 + }, + { + "epoch": 1.381917808219178, + "grad_norm": 1.9430493116378784, + "learning_rate": 3.0259951259138913e-06, + "loss": 2.1623, + "step": 1261 + }, + { + "epoch": 1.383013698630137, + "grad_norm": 1.825675368309021, + "learning_rate": 3.0239642567018686e-06, + "loss": 2.1022, + "step": 1262 + }, + { + "epoch": 1.384109589041096, + "grad_norm": 2.0050806999206543, + "learning_rate": 3.021933387489846e-06, + "loss": 2.1443, + "step": 1263 + }, + { + "epoch": 1.3852054794520547, + "grad_norm": 1.9548894166946411, + "learning_rate": 3.019902518277823e-06, + "loss": 2.0983, + "step": 1264 + }, + { + "epoch": 1.3863013698630138, + "grad_norm": 1.9097814559936523, + "learning_rate": 3.0178716490658004e-06, + "loss": 2.1488, + "step": 1265 + }, + { + "epoch": 1.3873972602739726, + "grad_norm": 1.8977514505386353, + "learning_rate": 3.0158407798537777e-06, + "loss": 2.1641, + "step": 1266 + }, + { + "epoch": 1.3884931506849316, + "grad_norm": 1.9222278594970703, + "learning_rate": 3.013809910641755e-06, + "loss": 2.1043, + "step": 1267 + }, + { + "epoch": 1.3895890410958904, + "grad_norm": 1.9333430528640747, + "learning_rate": 3.0117790414297322e-06, + "loss": 2.1152, + "step": 1268 + }, + { + "epoch": 1.3906849315068492, + "grad_norm": 1.8405439853668213, + "learning_rate": 3.0097481722177095e-06, + "loss": 2.1046, + "step": 1269 + }, + { + "epoch": 1.3917808219178083, + "grad_norm": 2.0583181381225586, + "learning_rate": 3.0077173030056868e-06, + "loss": 2.1378, + "step": 1270 + }, + { + "epoch": 1.3917808219178083, + "eval_loss": 2.286198616027832, + "eval_runtime": 46.6914, + "eval_samples_per_second": 21.417, + "eval_steps_per_second": 0.9, + "step": 1270 + }, + { + "epoch": 1.392876712328767, + "grad_norm": 1.8485472202301025, + "learning_rate": 3.005686433793664e-06, + "loss": 2.1184, + "step": 1271 + }, + { + "epoch": 1.393972602739726, + "grad_norm": 2.0430188179016113, + "learning_rate": 3.0036555645816413e-06, + "loss": 2.1518, + "step": 1272 + }, + { + "epoch": 1.395068493150685, + "grad_norm": 2.307474374771118, + "learning_rate": 3.001624695369618e-06, + "loss": 2.1288, + "step": 1273 + }, + { + "epoch": 1.396164383561644, + "grad_norm": 1.85003662109375, + "learning_rate": 2.999593826157596e-06, + "loss": 2.1573, + "step": 1274 + }, + { + "epoch": 1.3972602739726028, + "grad_norm": 2.047619581222534, + "learning_rate": 2.9975629569455727e-06, + "loss": 2.1285, + "step": 1275 + }, + { + "epoch": 1.3983561643835616, + "grad_norm": 2.013644218444824, + "learning_rate": 2.9955320877335504e-06, + "loss": 2.1603, + "step": 1276 + }, + { + "epoch": 1.3994520547945206, + "grad_norm": 1.8814456462860107, + "learning_rate": 2.9935012185215273e-06, + "loss": 2.1333, + "step": 1277 + }, + { + "epoch": 1.4005479452054794, + "grad_norm": 2.0338361263275146, + "learning_rate": 2.991470349309505e-06, + "loss": 2.1564, + "step": 1278 + }, + { + "epoch": 1.4016438356164382, + "grad_norm": 1.972992181777954, + "learning_rate": 2.989439480097482e-06, + "loss": 2.1305, + "step": 1279 + }, + { + "epoch": 1.4027397260273973, + "grad_norm": 1.845928430557251, + "learning_rate": 2.9874086108854595e-06, + "loss": 2.1384, + "step": 1280 + }, + { + "epoch": 1.4027397260273973, + "eval_loss": 2.28485107421875, + "eval_runtime": 46.9757, + "eval_samples_per_second": 21.288, + "eval_steps_per_second": 0.894, + "step": 1280 + }, + { + "epoch": 1.403835616438356, + "grad_norm": 1.7601349353790283, + "learning_rate": 2.9853777416734364e-06, + "loss": 2.1219, + "step": 1281 + }, + { + "epoch": 1.4049315068493151, + "grad_norm": 2.2385566234588623, + "learning_rate": 2.983346872461414e-06, + "loss": 2.1581, + "step": 1282 + }, + { + "epoch": 1.406027397260274, + "grad_norm": 2.019174098968506, + "learning_rate": 2.981316003249391e-06, + "loss": 2.1753, + "step": 1283 + }, + { + "epoch": 1.407123287671233, + "grad_norm": 1.9927541017532349, + "learning_rate": 2.9792851340373678e-06, + "loss": 2.147, + "step": 1284 + }, + { + "epoch": 1.4082191780821918, + "grad_norm": 2.5143957138061523, + "learning_rate": 2.9772542648253455e-06, + "loss": 2.138, + "step": 1285 + }, + { + "epoch": 1.4093150684931506, + "grad_norm": 1.9925073385238647, + "learning_rate": 2.9752233956133223e-06, + "loss": 2.1642, + "step": 1286 + }, + { + "epoch": 1.4104109589041096, + "grad_norm": 2.1739184856414795, + "learning_rate": 2.9731925264013e-06, + "loss": 2.1258, + "step": 1287 + }, + { + "epoch": 1.4115068493150684, + "grad_norm": 2.1719982624053955, + "learning_rate": 2.971161657189277e-06, + "loss": 2.1323, + "step": 1288 + }, + { + "epoch": 1.4126027397260275, + "grad_norm": 1.9897956848144531, + "learning_rate": 2.9691307879772546e-06, + "loss": 2.1286, + "step": 1289 + }, + { + "epoch": 1.4136986301369863, + "grad_norm": 1.997477412223816, + "learning_rate": 2.9670999187652314e-06, + "loss": 2.125, + "step": 1290 + }, + { + "epoch": 1.4136986301369863, + "eval_loss": 2.2839529514312744, + "eval_runtime": 44.5005, + "eval_samples_per_second": 22.472, + "eval_steps_per_second": 0.944, + "step": 1290 + }, + { + "epoch": 1.4147945205479453, + "grad_norm": 1.9529412984848022, + "learning_rate": 2.965069049553209e-06, + "loss": 2.1074, + "step": 1291 + }, + { + "epoch": 1.415890410958904, + "grad_norm": 1.900692343711853, + "learning_rate": 2.963038180341186e-06, + "loss": 2.1796, + "step": 1292 + }, + { + "epoch": 1.416986301369863, + "grad_norm": 2.0269863605499268, + "learning_rate": 2.9610073111291637e-06, + "loss": 2.0998, + "step": 1293 + }, + { + "epoch": 1.418082191780822, + "grad_norm": 2.0290050506591797, + "learning_rate": 2.9589764419171405e-06, + "loss": 2.1638, + "step": 1294 + }, + { + "epoch": 1.4191780821917808, + "grad_norm": 1.8717535734176636, + "learning_rate": 2.9569455727051182e-06, + "loss": 2.1756, + "step": 1295 + }, + { + "epoch": 1.4202739726027398, + "grad_norm": 1.9470241069793701, + "learning_rate": 2.954914703493095e-06, + "loss": 2.116, + "step": 1296 + }, + { + "epoch": 1.4213698630136986, + "grad_norm": 2.1749250888824463, + "learning_rate": 2.9528838342810724e-06, + "loss": 2.1381, + "step": 1297 + }, + { + "epoch": 1.4224657534246576, + "grad_norm": 1.905742883682251, + "learning_rate": 2.9508529650690496e-06, + "loss": 2.118, + "step": 1298 + }, + { + "epoch": 1.4235616438356165, + "grad_norm": 2.1214921474456787, + "learning_rate": 2.948822095857027e-06, + "loss": 2.1223, + "step": 1299 + }, + { + "epoch": 1.4246575342465753, + "grad_norm": 1.894509196281433, + "learning_rate": 2.946791226645004e-06, + "loss": 2.1457, + "step": 1300 + }, + { + "epoch": 1.4246575342465753, + "eval_loss": 2.2824268341064453, + "eval_runtime": 47.8031, + "eval_samples_per_second": 20.919, + "eval_steps_per_second": 0.879, + "step": 1300 + }, + { + "epoch": 1.4257534246575343, + "grad_norm": 1.9588708877563477, + "learning_rate": 2.9447603574329815e-06, + "loss": 2.1113, + "step": 1301 + }, + { + "epoch": 1.426849315068493, + "grad_norm": 1.9709562063217163, + "learning_rate": 2.9427294882209587e-06, + "loss": 2.1086, + "step": 1302 + }, + { + "epoch": 1.427945205479452, + "grad_norm": 1.82259202003479, + "learning_rate": 2.940698619008936e-06, + "loss": 2.118, + "step": 1303 + }, + { + "epoch": 1.429041095890411, + "grad_norm": 1.9776451587677002, + "learning_rate": 2.9386677497969133e-06, + "loss": 2.1386, + "step": 1304 + }, + { + "epoch": 1.43013698630137, + "grad_norm": 2.147463321685791, + "learning_rate": 2.9366368805848906e-06, + "loss": 2.1309, + "step": 1305 + }, + { + "epoch": 1.4312328767123288, + "grad_norm": 1.9845961332321167, + "learning_rate": 2.934606011372868e-06, + "loss": 2.1586, + "step": 1306 + }, + { + "epoch": 1.4323287671232876, + "grad_norm": 1.9153180122375488, + "learning_rate": 2.932575142160845e-06, + "loss": 2.1588, + "step": 1307 + }, + { + "epoch": 1.4334246575342466, + "grad_norm": 1.9372929334640503, + "learning_rate": 2.9305442729488224e-06, + "loss": 2.1351, + "step": 1308 + }, + { + "epoch": 1.4345205479452054, + "grad_norm": 1.852669358253479, + "learning_rate": 2.9285134037367997e-06, + "loss": 2.1394, + "step": 1309 + }, + { + "epoch": 1.4356164383561643, + "grad_norm": 1.934251308441162, + "learning_rate": 2.9264825345247765e-06, + "loss": 2.1364, + "step": 1310 + }, + { + "epoch": 1.4356164383561643, + "eval_loss": 2.280508279800415, + "eval_runtime": 43.1734, + "eval_samples_per_second": 23.162, + "eval_steps_per_second": 0.973, + "step": 1310 + }, + { + "epoch": 1.4367123287671233, + "grad_norm": 1.8886998891830444, + "learning_rate": 2.9244516653127542e-06, + "loss": 2.1452, + "step": 1311 + }, + { + "epoch": 1.4378082191780823, + "grad_norm": 2.275813102722168, + "learning_rate": 2.922420796100731e-06, + "loss": 2.1465, + "step": 1312 + }, + { + "epoch": 1.4389041095890411, + "grad_norm": 1.9017852544784546, + "learning_rate": 2.9203899268887088e-06, + "loss": 2.0785, + "step": 1313 + }, + { + "epoch": 1.44, + "grad_norm": 1.8482435941696167, + "learning_rate": 2.9183590576766856e-06, + "loss": 2.1892, + "step": 1314 + }, + { + "epoch": 1.441095890410959, + "grad_norm": 2.2122418880462646, + "learning_rate": 2.9163281884646633e-06, + "loss": 2.1058, + "step": 1315 + }, + { + "epoch": 1.4421917808219178, + "grad_norm": 1.9721672534942627, + "learning_rate": 2.91429731925264e-06, + "loss": 2.1391, + "step": 1316 + }, + { + "epoch": 1.4432876712328766, + "grad_norm": 1.9381605386734009, + "learning_rate": 2.912266450040618e-06, + "loss": 2.1651, + "step": 1317 + }, + { + "epoch": 1.4443835616438356, + "grad_norm": 2.274630069732666, + "learning_rate": 2.9102355808285947e-06, + "loss": 2.1434, + "step": 1318 + }, + { + "epoch": 1.4454794520547944, + "grad_norm": 1.91957688331604, + "learning_rate": 2.9082047116165724e-06, + "loss": 2.1218, + "step": 1319 + }, + { + "epoch": 1.4465753424657535, + "grad_norm": 2.042171001434326, + "learning_rate": 2.9061738424045493e-06, + "loss": 2.1233, + "step": 1320 + }, + { + "epoch": 1.4465753424657535, + "eval_loss": 2.280571699142456, + "eval_runtime": 48.425, + "eval_samples_per_second": 20.65, + "eval_steps_per_second": 0.867, + "step": 1320 + }, + { + "epoch": 1.4476712328767123, + "grad_norm": 2.0493276119232178, + "learning_rate": 2.904142973192527e-06, + "loss": 2.1241, + "step": 1321 + }, + { + "epoch": 1.4487671232876713, + "grad_norm": 1.8865023851394653, + "learning_rate": 2.902112103980504e-06, + "loss": 2.1826, + "step": 1322 + }, + { + "epoch": 1.4498630136986301, + "grad_norm": 1.9723960161209106, + "learning_rate": 2.900081234768481e-06, + "loss": 2.1508, + "step": 1323 + }, + { + "epoch": 1.450958904109589, + "grad_norm": 1.8321057558059692, + "learning_rate": 2.8980503655564584e-06, + "loss": 2.1319, + "step": 1324 + }, + { + "epoch": 1.452054794520548, + "grad_norm": 1.97796630859375, + "learning_rate": 2.8960194963444357e-06, + "loss": 2.1289, + "step": 1325 + }, + { + "epoch": 1.4531506849315068, + "grad_norm": 1.7793136835098267, + "learning_rate": 2.893988627132413e-06, + "loss": 2.1059, + "step": 1326 + }, + { + "epoch": 1.4542465753424658, + "grad_norm": 1.878381609916687, + "learning_rate": 2.8919577579203902e-06, + "loss": 2.125, + "step": 1327 + }, + { + "epoch": 1.4553424657534246, + "grad_norm": 1.8854790925979614, + "learning_rate": 2.8899268887083675e-06, + "loss": 2.0908, + "step": 1328 + }, + { + "epoch": 1.4564383561643837, + "grad_norm": 1.940931797027588, + "learning_rate": 2.8878960194963448e-06, + "loss": 2.1338, + "step": 1329 + }, + { + "epoch": 1.4575342465753425, + "grad_norm": 1.8014280796051025, + "learning_rate": 2.885865150284322e-06, + "loss": 2.1384, + "step": 1330 + }, + { + "epoch": 1.4575342465753425, + "eval_loss": 2.2815349102020264, + "eval_runtime": 48.7646, + "eval_samples_per_second": 20.507, + "eval_steps_per_second": 0.861, + "step": 1330 + }, + { + "epoch": 1.4586301369863013, + "grad_norm": 1.9306361675262451, + "learning_rate": 2.8838342810722993e-06, + "loss": 2.1562, + "step": 1331 + }, + { + "epoch": 1.4597260273972603, + "grad_norm": 1.902762770652771, + "learning_rate": 2.8818034118602766e-06, + "loss": 2.1183, + "step": 1332 + }, + { + "epoch": 1.4608219178082191, + "grad_norm": 1.980449914932251, + "learning_rate": 2.879772542648254e-06, + "loss": 2.1066, + "step": 1333 + }, + { + "epoch": 1.4619178082191782, + "grad_norm": 2.0774567127227783, + "learning_rate": 2.8777416734362307e-06, + "loss": 2.157, + "step": 1334 + }, + { + "epoch": 1.463013698630137, + "grad_norm": 1.8203723430633545, + "learning_rate": 2.8757108042242084e-06, + "loss": 2.0792, + "step": 1335 + }, + { + "epoch": 1.464109589041096, + "grad_norm": 1.8849077224731445, + "learning_rate": 2.8736799350121853e-06, + "loss": 2.1917, + "step": 1336 + }, + { + "epoch": 1.4652054794520548, + "grad_norm": 1.8782836198806763, + "learning_rate": 2.871649065800163e-06, + "loss": 2.1744, + "step": 1337 + }, + { + "epoch": 1.4663013698630136, + "grad_norm": 1.8628056049346924, + "learning_rate": 2.86961819658814e-06, + "loss": 2.1788, + "step": 1338 + }, + { + "epoch": 1.4673972602739727, + "grad_norm": 1.865370750427246, + "learning_rate": 2.8675873273761175e-06, + "loss": 2.1159, + "step": 1339 + }, + { + "epoch": 1.4684931506849315, + "grad_norm": 1.8673044443130493, + "learning_rate": 2.8655564581640944e-06, + "loss": 2.1447, + "step": 1340 + }, + { + "epoch": 1.4684931506849315, + "eval_loss": 2.282686471939087, + "eval_runtime": 49.85, + "eval_samples_per_second": 20.06, + "eval_steps_per_second": 0.843, + "step": 1340 + }, + { + "epoch": 1.4695890410958903, + "grad_norm": 1.8928618431091309, + "learning_rate": 2.863525588952072e-06, + "loss": 2.1362, + "step": 1341 + }, + { + "epoch": 1.4706849315068493, + "grad_norm": 1.8964238166809082, + "learning_rate": 2.861494719740049e-06, + "loss": 2.1628, + "step": 1342 + }, + { + "epoch": 1.4717808219178083, + "grad_norm": 1.9889185428619385, + "learning_rate": 2.8594638505280266e-06, + "loss": 2.0922, + "step": 1343 + }, + { + "epoch": 1.4728767123287672, + "grad_norm": 1.9503910541534424, + "learning_rate": 2.8574329813160035e-06, + "loss": 2.0979, + "step": 1344 + }, + { + "epoch": 1.473972602739726, + "grad_norm": 1.8889869451522827, + "learning_rate": 2.855402112103981e-06, + "loss": 2.1329, + "step": 1345 + }, + { + "epoch": 1.475068493150685, + "grad_norm": 1.7875893115997314, + "learning_rate": 2.853371242891958e-06, + "loss": 2.123, + "step": 1346 + }, + { + "epoch": 1.4761643835616438, + "grad_norm": 1.9358372688293457, + "learning_rate": 2.851340373679935e-06, + "loss": 2.1284, + "step": 1347 + }, + { + "epoch": 1.4772602739726026, + "grad_norm": 2.125243663787842, + "learning_rate": 2.8493095044679126e-06, + "loss": 2.1386, + "step": 1348 + }, + { + "epoch": 1.4783561643835617, + "grad_norm": 1.927788257598877, + "learning_rate": 2.8472786352558894e-06, + "loss": 2.1835, + "step": 1349 + }, + { + "epoch": 1.4794520547945205, + "grad_norm": 1.9361577033996582, + "learning_rate": 2.845247766043867e-06, + "loss": 2.1411, + "step": 1350 + }, + { + "epoch": 1.4794520547945205, + "eval_loss": 2.281172275543213, + "eval_runtime": 47.4205, + "eval_samples_per_second": 21.088, + "eval_steps_per_second": 0.886, + "step": 1350 + }, + { + "epoch": 1.4805479452054795, + "grad_norm": 1.847364068031311, + "learning_rate": 2.843216896831844e-06, + "loss": 2.111, + "step": 1351 + }, + { + "epoch": 1.4816438356164383, + "grad_norm": 1.9927674531936646, + "learning_rate": 2.8411860276198217e-06, + "loss": 2.107, + "step": 1352 + }, + { + "epoch": 1.4827397260273973, + "grad_norm": 1.8565406799316406, + "learning_rate": 2.8391551584077985e-06, + "loss": 2.12, + "step": 1353 + }, + { + "epoch": 1.4838356164383562, + "grad_norm": 1.9477596282958984, + "learning_rate": 2.8371242891957762e-06, + "loss": 2.1661, + "step": 1354 + }, + { + "epoch": 1.484931506849315, + "grad_norm": 1.9354814291000366, + "learning_rate": 2.835093419983753e-06, + "loss": 2.168, + "step": 1355 + }, + { + "epoch": 1.486027397260274, + "grad_norm": 2.0603487491607666, + "learning_rate": 2.833062550771731e-06, + "loss": 2.1338, + "step": 1356 + }, + { + "epoch": 1.4871232876712328, + "grad_norm": 1.987990379333496, + "learning_rate": 2.8310316815597076e-06, + "loss": 2.1282, + "step": 1357 + }, + { + "epoch": 1.4882191780821918, + "grad_norm": 1.895529866218567, + "learning_rate": 2.8290008123476853e-06, + "loss": 2.1032, + "step": 1358 + }, + { + "epoch": 1.4893150684931507, + "grad_norm": 1.917994737625122, + "learning_rate": 2.826969943135662e-06, + "loss": 2.149, + "step": 1359 + }, + { + "epoch": 1.4904109589041097, + "grad_norm": 1.9238080978393555, + "learning_rate": 2.8249390739236395e-06, + "loss": 2.1275, + "step": 1360 + }, + { + "epoch": 1.4904109589041097, + "eval_loss": 2.280546188354492, + "eval_runtime": 46.9985, + "eval_samples_per_second": 21.277, + "eval_steps_per_second": 0.894, + "step": 1360 + }, + { + "epoch": 1.4915068493150685, + "grad_norm": 1.947768211364746, + "learning_rate": 2.8229082047116167e-06, + "loss": 2.1282, + "step": 1361 + }, + { + "epoch": 1.4926027397260273, + "grad_norm": 1.7888189554214478, + "learning_rate": 2.820877335499594e-06, + "loss": 2.1134, + "step": 1362 + }, + { + "epoch": 1.4936986301369863, + "grad_norm": 1.8480347394943237, + "learning_rate": 2.8188464662875713e-06, + "loss": 2.1194, + "step": 1363 + }, + { + "epoch": 1.4947945205479451, + "grad_norm": 1.8146322965621948, + "learning_rate": 2.8168155970755486e-06, + "loss": 2.0781, + "step": 1364 + }, + { + "epoch": 1.4958904109589042, + "grad_norm": 1.7994698286056519, + "learning_rate": 2.814784727863526e-06, + "loss": 2.1027, + "step": 1365 + }, + { + "epoch": 1.496986301369863, + "grad_norm": 1.9375026226043701, + "learning_rate": 2.812753858651503e-06, + "loss": 2.1273, + "step": 1366 + }, + { + "epoch": 1.498082191780822, + "grad_norm": 1.9606287479400635, + "learning_rate": 2.8107229894394804e-06, + "loss": 2.1438, + "step": 1367 + }, + { + "epoch": 1.4991780821917808, + "grad_norm": 2.1815521717071533, + "learning_rate": 2.8086921202274577e-06, + "loss": 2.1176, + "step": 1368 + }, + { + "epoch": 1.5002739726027396, + "grad_norm": 1.8796950578689575, + "learning_rate": 2.806661251015435e-06, + "loss": 2.096, + "step": 1369 + }, + { + "epoch": 1.5013698630136987, + "grad_norm": 2.016091823577881, + "learning_rate": 2.8046303818034122e-06, + "loss": 2.1451, + "step": 1370 + }, + { + "epoch": 1.5013698630136987, + "eval_loss": 2.280881643295288, + "eval_runtime": 53.6471, + "eval_samples_per_second": 18.64, + "eval_steps_per_second": 0.783, + "step": 1370 + }, + { + "epoch": 1.5024657534246575, + "grad_norm": 1.9303081035614014, + "learning_rate": 2.802599512591389e-06, + "loss": 2.1736, + "step": 1371 + }, + { + "epoch": 1.5035616438356163, + "grad_norm": 1.9831101894378662, + "learning_rate": 2.8005686433793668e-06, + "loss": 2.2147, + "step": 1372 + }, + { + "epoch": 1.5046575342465753, + "grad_norm": 1.7864863872528076, + "learning_rate": 2.7985377741673436e-06, + "loss": 2.1231, + "step": 1373 + }, + { + "epoch": 1.5057534246575344, + "grad_norm": 1.8945375680923462, + "learning_rate": 2.7965069049553213e-06, + "loss": 2.0866, + "step": 1374 + }, + { + "epoch": 1.5068493150684932, + "grad_norm": 1.9881761074066162, + "learning_rate": 2.794476035743298e-06, + "loss": 2.1398, + "step": 1375 + }, + { + "epoch": 1.507945205479452, + "grad_norm": 1.9888391494750977, + "learning_rate": 2.792445166531276e-06, + "loss": 2.172, + "step": 1376 + }, + { + "epoch": 1.509041095890411, + "grad_norm": 1.9845142364501953, + "learning_rate": 2.7904142973192527e-06, + "loss": 2.1378, + "step": 1377 + }, + { + "epoch": 1.5101369863013698, + "grad_norm": 2.0640218257904053, + "learning_rate": 2.7883834281072304e-06, + "loss": 2.16, + "step": 1378 + }, + { + "epoch": 1.5112328767123286, + "grad_norm": 1.9453262090682983, + "learning_rate": 2.7863525588952073e-06, + "loss": 2.1112, + "step": 1379 + }, + { + "epoch": 1.5123287671232877, + "grad_norm": 2.0385119915008545, + "learning_rate": 2.784321689683185e-06, + "loss": 2.1068, + "step": 1380 + }, + { + "epoch": 1.5123287671232877, + "eval_loss": 2.2804832458496094, + "eval_runtime": 50.1317, + "eval_samples_per_second": 19.947, + "eval_steps_per_second": 0.838, + "step": 1380 + }, + { + "epoch": 1.5134246575342467, + "grad_norm": 1.9497289657592773, + "learning_rate": 2.782290820471162e-06, + "loss": 2.1017, + "step": 1381 + }, + { + "epoch": 1.5145205479452055, + "grad_norm": 1.9402867555618286, + "learning_rate": 2.7802599512591395e-06, + "loss": 2.1141, + "step": 1382 + }, + { + "epoch": 1.5156164383561643, + "grad_norm": 2.008141279220581, + "learning_rate": 2.7782290820471164e-06, + "loss": 2.1601, + "step": 1383 + }, + { + "epoch": 1.5167123287671234, + "grad_norm": 1.9685825109481812, + "learning_rate": 2.7761982128350932e-06, + "loss": 2.1494, + "step": 1384 + }, + { + "epoch": 1.5178082191780822, + "grad_norm": 1.9075162410736084, + "learning_rate": 2.774167343623071e-06, + "loss": 2.147, + "step": 1385 + }, + { + "epoch": 1.518904109589041, + "grad_norm": 1.8381123542785645, + "learning_rate": 2.772136474411048e-06, + "loss": 2.1772, + "step": 1386 + }, + { + "epoch": 1.52, + "grad_norm": 1.806057095527649, + "learning_rate": 2.7701056051990255e-06, + "loss": 2.1077, + "step": 1387 + }, + { + "epoch": 1.521095890410959, + "grad_norm": 1.9030957221984863, + "learning_rate": 2.7680747359870023e-06, + "loss": 2.0735, + "step": 1388 + }, + { + "epoch": 1.5221917808219179, + "grad_norm": 1.9967273473739624, + "learning_rate": 2.76604386677498e-06, + "loss": 2.155, + "step": 1389 + }, + { + "epoch": 1.5232876712328767, + "grad_norm": 1.864998698234558, + "learning_rate": 2.764012997562957e-06, + "loss": 2.1319, + "step": 1390 + }, + { + "epoch": 1.5232876712328767, + "eval_loss": 2.279405355453491, + "eval_runtime": 49.6092, + "eval_samples_per_second": 20.158, + "eval_steps_per_second": 0.847, + "step": 1390 + }, + { + "epoch": 1.5243835616438357, + "grad_norm": 1.831916093826294, + "learning_rate": 2.7619821283509346e-06, + "loss": 2.1518, + "step": 1391 + }, + { + "epoch": 1.5254794520547945, + "grad_norm": 2.1456778049468994, + "learning_rate": 2.7599512591389114e-06, + "loss": 2.0955, + "step": 1392 + }, + { + "epoch": 1.5265753424657533, + "grad_norm": 1.9264456033706665, + "learning_rate": 2.757920389926889e-06, + "loss": 2.1594, + "step": 1393 + }, + { + "epoch": 1.5276712328767124, + "grad_norm": 1.9881576299667358, + "learning_rate": 2.755889520714866e-06, + "loss": 2.115, + "step": 1394 + }, + { + "epoch": 1.5287671232876714, + "grad_norm": 1.9104706048965454, + "learning_rate": 2.7538586515028437e-06, + "loss": 2.1074, + "step": 1395 + }, + { + "epoch": 1.52986301369863, + "grad_norm": 1.8815401792526245, + "learning_rate": 2.7518277822908205e-06, + "loss": 2.0779, + "step": 1396 + }, + { + "epoch": 1.530958904109589, + "grad_norm": 1.999870777130127, + "learning_rate": 2.749796913078798e-06, + "loss": 2.1409, + "step": 1397 + }, + { + "epoch": 1.532054794520548, + "grad_norm": 2.037266969680786, + "learning_rate": 2.747766043866775e-06, + "loss": 2.1565, + "step": 1398 + }, + { + "epoch": 1.5331506849315069, + "grad_norm": 1.865891695022583, + "learning_rate": 2.7457351746547524e-06, + "loss": 2.1269, + "step": 1399 + }, + { + "epoch": 1.5342465753424657, + "grad_norm": 1.959797739982605, + "learning_rate": 2.7437043054427297e-06, + "loss": 2.1674, + "step": 1400 + }, + { + "epoch": 1.5342465753424657, + "eval_loss": 2.279064893722534, + "eval_runtime": 50.0471, + "eval_samples_per_second": 19.981, + "eval_steps_per_second": 0.839, + "step": 1400 + }, + { + "epoch": 1.5353424657534247, + "grad_norm": 2.021819829940796, + "learning_rate": 2.741673436230707e-06, + "loss": 2.2088, + "step": 1401 + }, + { + "epoch": 1.5364383561643835, + "grad_norm": 1.8549301624298096, + "learning_rate": 2.739642567018684e-06, + "loss": 2.1384, + "step": 1402 + }, + { + "epoch": 1.5375342465753423, + "grad_norm": 2.0776336193084717, + "learning_rate": 2.7376116978066615e-06, + "loss": 2.1068, + "step": 1403 + }, + { + "epoch": 1.5386301369863014, + "grad_norm": 1.9860707521438599, + "learning_rate": 2.7355808285946388e-06, + "loss": 2.1377, + "step": 1404 + }, + { + "epoch": 1.5397260273972604, + "grad_norm": 2.009176015853882, + "learning_rate": 2.733549959382616e-06, + "loss": 2.136, + "step": 1405 + }, + { + "epoch": 1.5408219178082192, + "grad_norm": 1.8771592378616333, + "learning_rate": 2.7315190901705933e-06, + "loss": 2.1104, + "step": 1406 + }, + { + "epoch": 1.541917808219178, + "grad_norm": 1.9801877737045288, + "learning_rate": 2.7294882209585706e-06, + "loss": 2.1803, + "step": 1407 + }, + { + "epoch": 1.543013698630137, + "grad_norm": 1.8967032432556152, + "learning_rate": 2.727457351746548e-06, + "loss": 2.141, + "step": 1408 + }, + { + "epoch": 1.5441095890410959, + "grad_norm": 1.8221750259399414, + "learning_rate": 2.725426482534525e-06, + "loss": 2.1181, + "step": 1409 + }, + { + "epoch": 1.5452054794520547, + "grad_norm": 1.972421407699585, + "learning_rate": 2.723395613322502e-06, + "loss": 2.1056, + "step": 1410 + }, + { + "epoch": 1.5452054794520547, + "eval_loss": 2.2803397178649902, + "eval_runtime": 42.0131, + "eval_samples_per_second": 23.802, + "eval_steps_per_second": 1.0, + "step": 1410 + }, + { + "epoch": 1.5463013698630137, + "grad_norm": 1.9438509941101074, + "learning_rate": 2.7213647441104797e-06, + "loss": 2.1378, + "step": 1411 + }, + { + "epoch": 1.5473972602739727, + "grad_norm": 1.8502674102783203, + "learning_rate": 2.7193338748984565e-06, + "loss": 2.1244, + "step": 1412 + }, + { + "epoch": 1.5484931506849315, + "grad_norm": 1.8806159496307373, + "learning_rate": 2.7173030056864342e-06, + "loss": 2.1371, + "step": 1413 + }, + { + "epoch": 1.5495890410958904, + "grad_norm": 1.823891282081604, + "learning_rate": 2.715272136474411e-06, + "loss": 2.2526, + "step": 1414 + }, + { + "epoch": 1.5506849315068494, + "grad_norm": 1.9969522953033447, + "learning_rate": 2.713241267262389e-06, + "loss": 2.0843, + "step": 1415 + }, + { + "epoch": 1.5517808219178082, + "grad_norm": 1.9167888164520264, + "learning_rate": 2.7112103980503656e-06, + "loss": 2.139, + "step": 1416 + }, + { + "epoch": 1.552876712328767, + "grad_norm": 1.8950753211975098, + "learning_rate": 2.7091795288383433e-06, + "loss": 2.1184, + "step": 1417 + }, + { + "epoch": 1.553972602739726, + "grad_norm": 1.8647392988204956, + "learning_rate": 2.70714865962632e-06, + "loss": 2.1466, + "step": 1418 + }, + { + "epoch": 1.555068493150685, + "grad_norm": 2.074019432067871, + "learning_rate": 2.705117790414298e-06, + "loss": 2.1154, + "step": 1419 + }, + { + "epoch": 1.5561643835616439, + "grad_norm": 1.883457064628601, + "learning_rate": 2.7030869212022747e-06, + "loss": 2.1251, + "step": 1420 + }, + { + "epoch": 1.5561643835616439, + "eval_loss": 2.279707908630371, + "eval_runtime": 43.6933, + "eval_samples_per_second": 22.887, + "eval_steps_per_second": 0.961, + "step": 1420 + }, + { + "epoch": 1.5572602739726027, + "grad_norm": 1.88644540309906, + "learning_rate": 2.7010560519902516e-06, + "loss": 2.1554, + "step": 1421 + }, + { + "epoch": 1.5583561643835617, + "grad_norm": 1.960942029953003, + "learning_rate": 2.6990251827782293e-06, + "loss": 2.1134, + "step": 1422 + }, + { + "epoch": 1.5594520547945205, + "grad_norm": 1.9701318740844727, + "learning_rate": 2.696994313566206e-06, + "loss": 2.1212, + "step": 1423 + }, + { + "epoch": 1.5605479452054793, + "grad_norm": 1.9110147953033447, + "learning_rate": 2.694963444354184e-06, + "loss": 2.1245, + "step": 1424 + }, + { + "epoch": 1.5616438356164384, + "grad_norm": 2.1578176021575928, + "learning_rate": 2.6929325751421607e-06, + "loss": 2.1809, + "step": 1425 + }, + { + "epoch": 1.5627397260273974, + "grad_norm": 1.9082913398742676, + "learning_rate": 2.6909017059301384e-06, + "loss": 2.1186, + "step": 1426 + }, + { + "epoch": 1.563835616438356, + "grad_norm": 1.9049410820007324, + "learning_rate": 2.6888708367181153e-06, + "loss": 2.1531, + "step": 1427 + }, + { + "epoch": 1.564931506849315, + "grad_norm": 2.041297197341919, + "learning_rate": 2.686839967506093e-06, + "loss": 2.0996, + "step": 1428 + }, + { + "epoch": 1.566027397260274, + "grad_norm": 1.9282728433609009, + "learning_rate": 2.68480909829407e-06, + "loss": 2.1492, + "step": 1429 + }, + { + "epoch": 1.5671232876712329, + "grad_norm": 1.908298134803772, + "learning_rate": 2.6827782290820475e-06, + "loss": 2.0726, + "step": 1430 + }, + { + "epoch": 1.5671232876712329, + "eval_loss": 2.278843402862549, + "eval_runtime": 43.3233, + "eval_samples_per_second": 23.082, + "eval_steps_per_second": 0.969, + "step": 1430 + }, + { + "epoch": 1.5682191780821917, + "grad_norm": 2.0996971130371094, + "learning_rate": 2.6807473598700244e-06, + "loss": 2.0829, + "step": 1431 + }, + { + "epoch": 1.5693150684931507, + "grad_norm": 1.8816159963607788, + "learning_rate": 2.678716490658002e-06, + "loss": 2.1659, + "step": 1432 + }, + { + "epoch": 1.5704109589041098, + "grad_norm": 1.91497004032135, + "learning_rate": 2.676685621445979e-06, + "loss": 2.1437, + "step": 1433 + }, + { + "epoch": 1.5715068493150683, + "grad_norm": 1.8993713855743408, + "learning_rate": 2.674654752233956e-06, + "loss": 2.1324, + "step": 1434 + }, + { + "epoch": 1.5726027397260274, + "grad_norm": 1.9361183643341064, + "learning_rate": 2.6726238830219335e-06, + "loss": 2.138, + "step": 1435 + }, + { + "epoch": 1.5736986301369864, + "grad_norm": 1.8992613554000854, + "learning_rate": 2.6705930138099107e-06, + "loss": 2.1309, + "step": 1436 + }, + { + "epoch": 1.5747945205479452, + "grad_norm": 1.9301854372024536, + "learning_rate": 2.668562144597888e-06, + "loss": 2.1645, + "step": 1437 + }, + { + "epoch": 1.575890410958904, + "grad_norm": 1.8373525142669678, + "learning_rate": 2.6665312753858653e-06, + "loss": 2.0699, + "step": 1438 + }, + { + "epoch": 1.576986301369863, + "grad_norm": 2.0120513439178467, + "learning_rate": 2.6645004061738426e-06, + "loss": 2.1472, + "step": 1439 + }, + { + "epoch": 1.5780821917808219, + "grad_norm": 2.007274627685547, + "learning_rate": 2.66246953696182e-06, + "loss": 2.1406, + "step": 1440 + }, + { + "epoch": 1.5780821917808219, + "eval_loss": 2.278693914413452, + "eval_runtime": 43.8767, + "eval_samples_per_second": 22.791, + "eval_steps_per_second": 0.957, + "step": 1440 + }, + { + "epoch": 1.5791780821917807, + "grad_norm": 1.9184056520462036, + "learning_rate": 2.660438667749797e-06, + "loss": 2.1076, + "step": 1441 + }, + { + "epoch": 1.5802739726027397, + "grad_norm": 1.9140634536743164, + "learning_rate": 2.6584077985377744e-06, + "loss": 2.0868, + "step": 1442 + }, + { + "epoch": 1.5813698630136988, + "grad_norm": 1.9146252870559692, + "learning_rate": 2.6563769293257517e-06, + "loss": 2.1205, + "step": 1443 + }, + { + "epoch": 1.5824657534246576, + "grad_norm": 1.8893204927444458, + "learning_rate": 2.654346060113729e-06, + "loss": 2.1249, + "step": 1444 + }, + { + "epoch": 1.5835616438356164, + "grad_norm": 2.0452592372894287, + "learning_rate": 2.6523151909017062e-06, + "loss": 2.1069, + "step": 1445 + }, + { + "epoch": 1.5846575342465754, + "grad_norm": 1.8728071451187134, + "learning_rate": 2.6502843216896835e-06, + "loss": 2.1161, + "step": 1446 + }, + { + "epoch": 1.5857534246575342, + "grad_norm": 2.1464476585388184, + "learning_rate": 2.6482534524776603e-06, + "loss": 2.1032, + "step": 1447 + }, + { + "epoch": 1.586849315068493, + "grad_norm": 1.9760676622390747, + "learning_rate": 2.646222583265638e-06, + "loss": 2.1754, + "step": 1448 + }, + { + "epoch": 1.587945205479452, + "grad_norm": 1.9876340627670288, + "learning_rate": 2.644191714053615e-06, + "loss": 2.1512, + "step": 1449 + }, + { + "epoch": 1.589041095890411, + "grad_norm": 2.039198160171509, + "learning_rate": 2.6421608448415926e-06, + "loss": 2.1416, + "step": 1450 + }, + { + "epoch": 1.589041095890411, + "eval_loss": 2.2811355590820312, + "eval_runtime": 44.209, + "eval_samples_per_second": 22.62, + "eval_steps_per_second": 0.95, + "step": 1450 + }, + { + "epoch": 1.59013698630137, + "grad_norm": 1.882146954536438, + "learning_rate": 2.6401299756295694e-06, + "loss": 2.1106, + "step": 1451 + }, + { + "epoch": 1.5912328767123287, + "grad_norm": 2.0288634300231934, + "learning_rate": 2.638099106417547e-06, + "loss": 2.1011, + "step": 1452 + }, + { + "epoch": 1.5923287671232877, + "grad_norm": 2.0438144207000732, + "learning_rate": 2.636068237205524e-06, + "loss": 2.1352, + "step": 1453 + }, + { + "epoch": 1.5934246575342466, + "grad_norm": 1.9043279886245728, + "learning_rate": 2.6340373679935017e-06, + "loss": 2.1044, + "step": 1454 + }, + { + "epoch": 1.5945205479452054, + "grad_norm": 2.112670660018921, + "learning_rate": 2.6320064987814785e-06, + "loss": 2.1724, + "step": 1455 + }, + { + "epoch": 1.5956164383561644, + "grad_norm": 1.8258006572723389, + "learning_rate": 2.6299756295694562e-06, + "loss": 2.0825, + "step": 1456 + }, + { + "epoch": 1.5967123287671234, + "grad_norm": 1.970802664756775, + "learning_rate": 2.627944760357433e-06, + "loss": 2.1431, + "step": 1457 + }, + { + "epoch": 1.5978082191780822, + "grad_norm": 1.9067145586013794, + "learning_rate": 2.6259138911454104e-06, + "loss": 2.1729, + "step": 1458 + }, + { + "epoch": 1.598904109589041, + "grad_norm": 1.9728924036026, + "learning_rate": 2.6238830219333877e-06, + "loss": 2.1773, + "step": 1459 + }, + { + "epoch": 1.6, + "grad_norm": 2.0142018795013428, + "learning_rate": 2.621852152721365e-06, + "loss": 2.1626, + "step": 1460 + }, + { + "epoch": 1.6, + "eval_loss": 2.2811660766601562, + "eval_runtime": 44.1735, + "eval_samples_per_second": 22.638, + "eval_steps_per_second": 0.951, + "step": 1460 + }, + { + "epoch": 1.601095890410959, + "grad_norm": 2.064955472946167, + "learning_rate": 2.619821283509342e-06, + "loss": 2.0833, + "step": 1461 + }, + { + "epoch": 1.6021917808219177, + "grad_norm": 1.9600436687469482, + "learning_rate": 2.6177904142973195e-06, + "loss": 2.1596, + "step": 1462 + }, + { + "epoch": 1.6032876712328767, + "grad_norm": 1.8413004875183105, + "learning_rate": 2.6157595450852968e-06, + "loss": 2.0957, + "step": 1463 + }, + { + "epoch": 1.6043835616438358, + "grad_norm": 2.1802690029144287, + "learning_rate": 2.613728675873274e-06, + "loss": 2.1432, + "step": 1464 + }, + { + "epoch": 1.6054794520547944, + "grad_norm": 1.992746353149414, + "learning_rate": 2.6116978066612513e-06, + "loss": 2.1631, + "step": 1465 + }, + { + "epoch": 1.6065753424657534, + "grad_norm": 1.9906671047210693, + "learning_rate": 2.6096669374492286e-06, + "loss": 2.1587, + "step": 1466 + }, + { + "epoch": 1.6076712328767124, + "grad_norm": 2.1134135723114014, + "learning_rate": 2.607636068237206e-06, + "loss": 2.1385, + "step": 1467 + }, + { + "epoch": 1.6087671232876712, + "grad_norm": 2.259915590286255, + "learning_rate": 2.605605199025183e-06, + "loss": 2.1042, + "step": 1468 + }, + { + "epoch": 1.60986301369863, + "grad_norm": 2.0705254077911377, + "learning_rate": 2.6035743298131604e-06, + "loss": 2.145, + "step": 1469 + }, + { + "epoch": 1.610958904109589, + "grad_norm": 2.0326738357543945, + "learning_rate": 2.6015434606011377e-06, + "loss": 2.133, + "step": 1470 + }, + { + "epoch": 1.610958904109589, + "eval_loss": 2.2812676429748535, + "eval_runtime": 43.4091, + "eval_samples_per_second": 23.037, + "eval_steps_per_second": 0.968, + "step": 1470 + }, + { + "epoch": 1.612054794520548, + "grad_norm": 2.034182071685791, + "learning_rate": 2.5995125913891145e-06, + "loss": 2.1507, + "step": 1471 + }, + { + "epoch": 1.6131506849315067, + "grad_norm": 2.069373846054077, + "learning_rate": 2.5974817221770922e-06, + "loss": 2.152, + "step": 1472 + }, + { + "epoch": 1.6142465753424657, + "grad_norm": 2.2003285884857178, + "learning_rate": 2.595450852965069e-06, + "loss": 2.1347, + "step": 1473 + }, + { + "epoch": 1.6153424657534248, + "grad_norm": 1.919945478439331, + "learning_rate": 2.593419983753047e-06, + "loss": 2.1476, + "step": 1474 + }, + { + "epoch": 1.6164383561643836, + "grad_norm": 1.825148344039917, + "learning_rate": 2.5913891145410236e-06, + "loss": 2.1757, + "step": 1475 + }, + { + "epoch": 1.6175342465753424, + "grad_norm": 2.0014286041259766, + "learning_rate": 2.5893582453290013e-06, + "loss": 2.1208, + "step": 1476 + }, + { + "epoch": 1.6186301369863014, + "grad_norm": 2.0642764568328857, + "learning_rate": 2.587327376116978e-06, + "loss": 2.1619, + "step": 1477 + }, + { + "epoch": 1.6197260273972602, + "grad_norm": 2.002497911453247, + "learning_rate": 2.585296506904956e-06, + "loss": 2.1198, + "step": 1478 + }, + { + "epoch": 1.620821917808219, + "grad_norm": 1.7852193117141724, + "learning_rate": 2.5832656376929327e-06, + "loss": 2.1247, + "step": 1479 + }, + { + "epoch": 1.621917808219178, + "grad_norm": 1.9539082050323486, + "learning_rate": 2.5812347684809104e-06, + "loss": 2.1046, + "step": 1480 + }, + { + "epoch": 1.621917808219178, + "eval_loss": 2.2807998657226562, + "eval_runtime": 44.2896, + "eval_samples_per_second": 22.579, + "eval_steps_per_second": 0.948, + "step": 1480 + }, + { + "epoch": 1.6230136986301371, + "grad_norm": 2.095280647277832, + "learning_rate": 2.5792038992688873e-06, + "loss": 2.0685, + "step": 1481 + }, + { + "epoch": 1.624109589041096, + "grad_norm": 2.230388879776001, + "learning_rate": 2.577173030056865e-06, + "loss": 2.1976, + "step": 1482 + }, + { + "epoch": 1.6252054794520547, + "grad_norm": 1.9304745197296143, + "learning_rate": 2.575142160844842e-06, + "loss": 2.0647, + "step": 1483 + }, + { + "epoch": 1.6263013698630138, + "grad_norm": 1.856342077255249, + "learning_rate": 2.5731112916328187e-06, + "loss": 2.1444, + "step": 1484 + }, + { + "epoch": 1.6273972602739726, + "grad_norm": 2.117694854736328, + "learning_rate": 2.5710804224207964e-06, + "loss": 2.1536, + "step": 1485 + }, + { + "epoch": 1.6284931506849314, + "grad_norm": 2.043900489807129, + "learning_rate": 2.5690495532087733e-06, + "loss": 2.106, + "step": 1486 + }, + { + "epoch": 1.6295890410958904, + "grad_norm": 1.9283581972122192, + "learning_rate": 2.567018683996751e-06, + "loss": 2.0667, + "step": 1487 + }, + { + "epoch": 1.6306849315068495, + "grad_norm": 1.8138651847839355, + "learning_rate": 2.564987814784728e-06, + "loss": 2.1386, + "step": 1488 + }, + { + "epoch": 1.6317808219178083, + "grad_norm": 2.0534632205963135, + "learning_rate": 2.5629569455727055e-06, + "loss": 2.105, + "step": 1489 + }, + { + "epoch": 1.632876712328767, + "grad_norm": 1.9466804265975952, + "learning_rate": 2.5609260763606824e-06, + "loss": 2.1211, + "step": 1490 + }, + { + "epoch": 1.632876712328767, + "eval_loss": 2.275447368621826, + "eval_runtime": 44.2319, + "eval_samples_per_second": 22.608, + "eval_steps_per_second": 0.95, + "step": 1490 + }, + { + "epoch": 1.633972602739726, + "grad_norm": 1.8767902851104736, + "learning_rate": 2.55889520714866e-06, + "loss": 2.116, + "step": 1491 + }, + { + "epoch": 1.635068493150685, + "grad_norm": 1.9194284677505493, + "learning_rate": 2.556864337936637e-06, + "loss": 2.1928, + "step": 1492 + }, + { + "epoch": 1.6361643835616437, + "grad_norm": 2.0496408939361572, + "learning_rate": 2.5548334687246146e-06, + "loss": 2.1182, + "step": 1493 + }, + { + "epoch": 1.6372602739726028, + "grad_norm": 1.8577388525009155, + "learning_rate": 2.5528025995125915e-06, + "loss": 2.1313, + "step": 1494 + }, + { + "epoch": 1.6383561643835618, + "grad_norm": 2.020623207092285, + "learning_rate": 2.550771730300569e-06, + "loss": 2.1627, + "step": 1495 + }, + { + "epoch": 1.6394520547945204, + "grad_norm": 1.8472096920013428, + "learning_rate": 2.548740861088546e-06, + "loss": 2.1514, + "step": 1496 + }, + { + "epoch": 1.6405479452054794, + "grad_norm": 1.9105209112167358, + "learning_rate": 2.5467099918765233e-06, + "loss": 2.1558, + "step": 1497 + }, + { + "epoch": 1.6416438356164385, + "grad_norm": 1.9635803699493408, + "learning_rate": 2.5446791226645006e-06, + "loss": 2.08, + "step": 1498 + }, + { + "epoch": 1.6427397260273973, + "grad_norm": 1.8911043405532837, + "learning_rate": 2.542648253452478e-06, + "loss": 2.0648, + "step": 1499 + }, + { + "epoch": 1.643835616438356, + "grad_norm": 1.9337847232818604, + "learning_rate": 2.540617384240455e-06, + "loss": 2.1526, + "step": 1500 + }, + { + "epoch": 1.643835616438356, + "eval_loss": 2.2771341800689697, + "eval_runtime": 44.2753, + "eval_samples_per_second": 22.586, + "eval_steps_per_second": 0.949, + "step": 1500 + }, + { + "epoch": 1.644931506849315, + "grad_norm": 2.152970552444458, + "learning_rate": 2.5385865150284324e-06, + "loss": 2.1328, + "step": 1501 + }, + { + "epoch": 1.6460273972602741, + "grad_norm": 1.8749687671661377, + "learning_rate": 2.5365556458164097e-06, + "loss": 2.1093, + "step": 1502 + }, + { + "epoch": 1.6471232876712327, + "grad_norm": 1.8745299577713013, + "learning_rate": 2.534524776604387e-06, + "loss": 2.1172, + "step": 1503 + }, + { + "epoch": 1.6482191780821918, + "grad_norm": 2.006110906600952, + "learning_rate": 2.5324939073923642e-06, + "loss": 2.1318, + "step": 1504 + }, + { + "epoch": 1.6493150684931508, + "grad_norm": 2.055590867996216, + "learning_rate": 2.5304630381803415e-06, + "loss": 2.0463, + "step": 1505 + }, + { + "epoch": 1.6504109589041096, + "grad_norm": 1.9270331859588623, + "learning_rate": 2.5284321689683188e-06, + "loss": 2.1195, + "step": 1506 + }, + { + "epoch": 1.6515068493150684, + "grad_norm": 1.9981428384780884, + "learning_rate": 2.526401299756296e-06, + "loss": 2.1282, + "step": 1507 + }, + { + "epoch": 1.6526027397260274, + "grad_norm": 2.051969528198242, + "learning_rate": 2.524370430544273e-06, + "loss": 2.1013, + "step": 1508 + }, + { + "epoch": 1.6536986301369863, + "grad_norm": 2.067718744277954, + "learning_rate": 2.5223395613322506e-06, + "loss": 2.1265, + "step": 1509 + }, + { + "epoch": 1.654794520547945, + "grad_norm": 2.0843262672424316, + "learning_rate": 2.5203086921202274e-06, + "loss": 2.0994, + "step": 1510 + }, + { + "epoch": 1.654794520547945, + "eval_loss": 2.27781081199646, + "eval_runtime": 42.8213, + "eval_samples_per_second": 23.353, + "eval_steps_per_second": 0.981, + "step": 1510 + }, + { + "epoch": 1.655890410958904, + "grad_norm": 2.146921396255493, + "learning_rate": 2.518277822908205e-06, + "loss": 2.1456, + "step": 1511 + }, + { + "epoch": 1.6569863013698631, + "grad_norm": 1.9604636430740356, + "learning_rate": 2.516246953696182e-06, + "loss": 2.1118, + "step": 1512 + }, + { + "epoch": 1.658082191780822, + "grad_norm": 1.9613317251205444, + "learning_rate": 2.5142160844841597e-06, + "loss": 2.1401, + "step": 1513 + }, + { + "epoch": 1.6591780821917808, + "grad_norm": 2.0410237312316895, + "learning_rate": 2.5121852152721365e-06, + "loss": 2.1532, + "step": 1514 + }, + { + "epoch": 1.6602739726027398, + "grad_norm": 1.8603837490081787, + "learning_rate": 2.5101543460601142e-06, + "loss": 2.1437, + "step": 1515 + }, + { + "epoch": 1.6613698630136986, + "grad_norm": 1.9498400688171387, + "learning_rate": 2.508123476848091e-06, + "loss": 2.1384, + "step": 1516 + }, + { + "epoch": 1.6624657534246574, + "grad_norm": 1.8603938817977905, + "learning_rate": 2.506092607636069e-06, + "loss": 2.0903, + "step": 1517 + }, + { + "epoch": 1.6635616438356164, + "grad_norm": 1.8238614797592163, + "learning_rate": 2.5040617384240457e-06, + "loss": 2.1644, + "step": 1518 + }, + { + "epoch": 1.6646575342465755, + "grad_norm": 1.9574072360992432, + "learning_rate": 2.5020308692120234e-06, + "loss": 2.1072, + "step": 1519 + }, + { + "epoch": 1.6657534246575343, + "grad_norm": 2.100829601287842, + "learning_rate": 2.5e-06, + "loss": 2.111, + "step": 1520 + }, + { + "epoch": 1.6657534246575343, + "eval_loss": 2.278552293777466, + "eval_runtime": 46.587, + "eval_samples_per_second": 21.465, + "eval_steps_per_second": 0.902, + "step": 1520 + }, + { + "epoch": 1.666849315068493, + "grad_norm": 1.921262264251709, + "learning_rate": 2.4979691307879775e-06, + "loss": 2.1667, + "step": 1521 + }, + { + "epoch": 1.6679452054794521, + "grad_norm": 1.9328664541244507, + "learning_rate": 2.4959382615759548e-06, + "loss": 2.1175, + "step": 1522 + }, + { + "epoch": 1.669041095890411, + "grad_norm": 1.909544587135315, + "learning_rate": 2.493907392363932e-06, + "loss": 2.0905, + "step": 1523 + }, + { + "epoch": 1.6701369863013698, + "grad_norm": 1.928046464920044, + "learning_rate": 2.4918765231519093e-06, + "loss": 2.1757, + "step": 1524 + }, + { + "epoch": 1.6712328767123288, + "grad_norm": 1.8839048147201538, + "learning_rate": 2.4898456539398866e-06, + "loss": 2.1424, + "step": 1525 + }, + { + "epoch": 1.6723287671232878, + "grad_norm": 1.913812518119812, + "learning_rate": 2.487814784727864e-06, + "loss": 2.1178, + "step": 1526 + }, + { + "epoch": 1.6734246575342466, + "grad_norm": 1.7485435009002686, + "learning_rate": 2.4857839155158407e-06, + "loss": 2.1794, + "step": 1527 + }, + { + "epoch": 1.6745205479452054, + "grad_norm": 1.8693492412567139, + "learning_rate": 2.483753046303818e-06, + "loss": 2.1265, + "step": 1528 + }, + { + "epoch": 1.6756164383561645, + "grad_norm": 1.8982772827148438, + "learning_rate": 2.4817221770917953e-06, + "loss": 2.1254, + "step": 1529 + }, + { + "epoch": 1.6767123287671233, + "grad_norm": 1.9917978048324585, + "learning_rate": 2.4796913078797725e-06, + "loss": 2.13, + "step": 1530 + }, + { + "epoch": 1.6767123287671233, + "eval_loss": 2.277986764907837, + "eval_runtime": 52.1464, + "eval_samples_per_second": 19.177, + "eval_steps_per_second": 0.805, + "step": 1530 + }, + { + "epoch": 1.677808219178082, + "grad_norm": 1.8703829050064087, + "learning_rate": 2.47766043866775e-06, + "loss": 2.1527, + "step": 1531 + }, + { + "epoch": 1.6789041095890411, + "grad_norm": 1.950713872909546, + "learning_rate": 2.475629569455727e-06, + "loss": 2.1481, + "step": 1532 + }, + { + "epoch": 1.6800000000000002, + "grad_norm": 1.8339334726333618, + "learning_rate": 2.4735987002437044e-06, + "loss": 2.1422, + "step": 1533 + }, + { + "epoch": 1.6810958904109587, + "grad_norm": 1.8423672914505005, + "learning_rate": 2.4715678310316816e-06, + "loss": 2.1026, + "step": 1534 + }, + { + "epoch": 1.6821917808219178, + "grad_norm": 1.9338176250457764, + "learning_rate": 2.469536961819659e-06, + "loss": 2.1095, + "step": 1535 + }, + { + "epoch": 1.6832876712328768, + "grad_norm": 1.8666706085205078, + "learning_rate": 2.467506092607636e-06, + "loss": 2.1176, + "step": 1536 + }, + { + "epoch": 1.6843835616438356, + "grad_norm": 2.0762805938720703, + "learning_rate": 2.4654752233956135e-06, + "loss": 2.127, + "step": 1537 + }, + { + "epoch": 1.6854794520547944, + "grad_norm": 1.9729993343353271, + "learning_rate": 2.4634443541835907e-06, + "loss": 2.1194, + "step": 1538 + }, + { + "epoch": 1.6865753424657535, + "grad_norm": 1.910555362701416, + "learning_rate": 2.461413484971568e-06, + "loss": 2.1745, + "step": 1539 + }, + { + "epoch": 1.6876712328767123, + "grad_norm": 1.953911304473877, + "learning_rate": 2.4593826157595453e-06, + "loss": 2.1236, + "step": 1540 + }, + { + "epoch": 1.6876712328767123, + "eval_loss": 2.277575731277466, + "eval_runtime": 48.0126, + "eval_samples_per_second": 20.828, + "eval_steps_per_second": 0.875, + "step": 1540 + }, + { + "epoch": 1.688767123287671, + "grad_norm": 2.0469322204589844, + "learning_rate": 2.4573517465475226e-06, + "loss": 2.1999, + "step": 1541 + }, + { + "epoch": 1.6898630136986301, + "grad_norm": 1.9962514638900757, + "learning_rate": 2.4553208773355e-06, + "loss": 2.0865, + "step": 1542 + }, + { + "epoch": 1.6909589041095892, + "grad_norm": 1.911471962928772, + "learning_rate": 2.453290008123477e-06, + "loss": 2.1107, + "step": 1543 + }, + { + "epoch": 1.692054794520548, + "grad_norm": 1.9354206323623657, + "learning_rate": 2.4512591389114544e-06, + "loss": 2.1215, + "step": 1544 + }, + { + "epoch": 1.6931506849315068, + "grad_norm": 1.938178539276123, + "learning_rate": 2.4492282696994317e-06, + "loss": 2.1378, + "step": 1545 + }, + { + "epoch": 1.6942465753424658, + "grad_norm": 2.003552198410034, + "learning_rate": 2.447197400487409e-06, + "loss": 2.1361, + "step": 1546 + }, + { + "epoch": 1.6953424657534246, + "grad_norm": 1.9754462242126465, + "learning_rate": 2.4451665312753862e-06, + "loss": 2.134, + "step": 1547 + }, + { + "epoch": 1.6964383561643834, + "grad_norm": 1.924299955368042, + "learning_rate": 2.4431356620633635e-06, + "loss": 2.1497, + "step": 1548 + }, + { + "epoch": 1.6975342465753425, + "grad_norm": 2.1239371299743652, + "learning_rate": 2.4411047928513408e-06, + "loss": 2.1427, + "step": 1549 + }, + { + "epoch": 1.6986301369863015, + "grad_norm": 1.984377384185791, + "learning_rate": 2.439073923639318e-06, + "loss": 2.1268, + "step": 1550 + }, + { + "epoch": 1.6986301369863015, + "eval_loss": 2.2747738361358643, + "eval_runtime": 53.2249, + "eval_samples_per_second": 18.788, + "eval_steps_per_second": 0.789, + "step": 1550 + }, + { + "epoch": 1.6997260273972603, + "grad_norm": 1.9375758171081543, + "learning_rate": 2.4370430544272953e-06, + "loss": 2.1369, + "step": 1551 + }, + { + "epoch": 1.7008219178082191, + "grad_norm": 1.9191703796386719, + "learning_rate": 2.435012185215272e-06, + "loss": 2.1061, + "step": 1552 + }, + { + "epoch": 1.7019178082191782, + "grad_norm": 2.1356418132781982, + "learning_rate": 2.4329813160032495e-06, + "loss": 2.0923, + "step": 1553 + }, + { + "epoch": 1.703013698630137, + "grad_norm": 2.02870774269104, + "learning_rate": 2.4309504467912267e-06, + "loss": 2.105, + "step": 1554 + }, + { + "epoch": 1.7041095890410958, + "grad_norm": 1.9436805248260498, + "learning_rate": 2.428919577579204e-06, + "loss": 2.1194, + "step": 1555 + }, + { + "epoch": 1.7052054794520548, + "grad_norm": 1.997019648551941, + "learning_rate": 2.4268887083671813e-06, + "loss": 2.1114, + "step": 1556 + }, + { + "epoch": 1.7063013698630138, + "grad_norm": 2.0759427547454834, + "learning_rate": 2.4248578391551586e-06, + "loss": 2.1955, + "step": 1557 + }, + { + "epoch": 1.7073972602739727, + "grad_norm": 2.0526654720306396, + "learning_rate": 2.422826969943136e-06, + "loss": 2.1103, + "step": 1558 + }, + { + "epoch": 1.7084931506849315, + "grad_norm": 1.8765089511871338, + "learning_rate": 2.420796100731113e-06, + "loss": 2.1068, + "step": 1559 + }, + { + "epoch": 1.7095890410958905, + "grad_norm": 1.9164950847625732, + "learning_rate": 2.4187652315190904e-06, + "loss": 2.082, + "step": 1560 + }, + { + "epoch": 1.7095890410958905, + "eval_loss": 2.272998332977295, + "eval_runtime": 49.0487, + "eval_samples_per_second": 20.388, + "eval_steps_per_second": 0.856, + "step": 1560 + }, + { + "epoch": 1.7106849315068493, + "grad_norm": 1.9921822547912598, + "learning_rate": 2.4167343623070677e-06, + "loss": 2.1285, + "step": 1561 + }, + { + "epoch": 1.7117808219178081, + "grad_norm": 1.8890291452407837, + "learning_rate": 2.414703493095045e-06, + "loss": 2.1385, + "step": 1562 + }, + { + "epoch": 1.7128767123287671, + "grad_norm": 1.9233667850494385, + "learning_rate": 2.4126726238830222e-06, + "loss": 2.1282, + "step": 1563 + }, + { + "epoch": 1.7139726027397262, + "grad_norm": 2.0368800163269043, + "learning_rate": 2.410641754670999e-06, + "loss": 2.1366, + "step": 1564 + }, + { + "epoch": 1.7150684931506848, + "grad_norm": 2.0321929454803467, + "learning_rate": 2.4086108854589763e-06, + "loss": 2.1163, + "step": 1565 + }, + { + "epoch": 1.7161643835616438, + "grad_norm": 2.042773485183716, + "learning_rate": 2.4065800162469536e-06, + "loss": 2.1529, + "step": 1566 + }, + { + "epoch": 1.7172602739726028, + "grad_norm": 1.937940239906311, + "learning_rate": 2.404549147034931e-06, + "loss": 2.106, + "step": 1567 + }, + { + "epoch": 1.7183561643835616, + "grad_norm": 1.878649353981018, + "learning_rate": 2.402518277822908e-06, + "loss": 2.1179, + "step": 1568 + }, + { + "epoch": 1.7194520547945205, + "grad_norm": 1.997246265411377, + "learning_rate": 2.4004874086108854e-06, + "loss": 2.1672, + "step": 1569 + }, + { + "epoch": 1.7205479452054795, + "grad_norm": 1.9837015867233276, + "learning_rate": 2.3984565393988627e-06, + "loss": 2.0816, + "step": 1570 + }, + { + "epoch": 1.7205479452054795, + "eval_loss": 2.2730956077575684, + "eval_runtime": 52.1793, + "eval_samples_per_second": 19.165, + "eval_steps_per_second": 0.805, + "step": 1570 + }, + { + "epoch": 1.7216438356164385, + "grad_norm": 1.8919317722320557, + "learning_rate": 2.39642567018684e-06, + "loss": 2.1541, + "step": 1571 + }, + { + "epoch": 1.7227397260273971, + "grad_norm": 1.8129284381866455, + "learning_rate": 2.3943948009748173e-06, + "loss": 2.1033, + "step": 1572 + }, + { + "epoch": 1.7238356164383561, + "grad_norm": 1.79408860206604, + "learning_rate": 2.3923639317627945e-06, + "loss": 2.1163, + "step": 1573 + }, + { + "epoch": 1.7249315068493152, + "grad_norm": 1.9178274869918823, + "learning_rate": 2.390333062550772e-06, + "loss": 2.1383, + "step": 1574 + }, + { + "epoch": 1.726027397260274, + "grad_norm": 1.9113222360610962, + "learning_rate": 2.388302193338749e-06, + "loss": 2.0987, + "step": 1575 + }, + { + "epoch": 1.7271232876712328, + "grad_norm": 1.915170431137085, + "learning_rate": 2.3862713241267264e-06, + "loss": 2.0235, + "step": 1576 + }, + { + "epoch": 1.7282191780821918, + "grad_norm": 1.9789308309555054, + "learning_rate": 2.3842404549147037e-06, + "loss": 2.1942, + "step": 1577 + }, + { + "epoch": 1.7293150684931506, + "grad_norm": 1.8410224914550781, + "learning_rate": 2.382209585702681e-06, + "loss": 2.1105, + "step": 1578 + }, + { + "epoch": 1.7304109589041095, + "grad_norm": 1.9388853311538696, + "learning_rate": 2.380178716490658e-06, + "loss": 2.1021, + "step": 1579 + }, + { + "epoch": 1.7315068493150685, + "grad_norm": 1.9328628778457642, + "learning_rate": 2.3781478472786355e-06, + "loss": 2.1584, + "step": 1580 + }, + { + "epoch": 1.7315068493150685, + "eval_loss": 2.271240472793579, + "eval_runtime": 47.9206, + "eval_samples_per_second": 20.868, + "eval_steps_per_second": 0.876, + "step": 1580 + }, + { + "epoch": 1.7326027397260275, + "grad_norm": 1.8846843242645264, + "learning_rate": 2.3761169780666128e-06, + "loss": 2.1269, + "step": 1581 + }, + { + "epoch": 1.7336986301369863, + "grad_norm": 1.8232983350753784, + "learning_rate": 2.37408610885459e-06, + "loss": 2.1522, + "step": 1582 + }, + { + "epoch": 1.7347945205479451, + "grad_norm": 2.023982286453247, + "learning_rate": 2.3720552396425673e-06, + "loss": 2.1571, + "step": 1583 + }, + { + "epoch": 1.7358904109589042, + "grad_norm": 1.861229658126831, + "learning_rate": 2.3700243704305446e-06, + "loss": 2.103, + "step": 1584 + }, + { + "epoch": 1.736986301369863, + "grad_norm": 1.8577842712402344, + "learning_rate": 2.367993501218522e-06, + "loss": 2.1766, + "step": 1585 + }, + { + "epoch": 1.7380821917808218, + "grad_norm": 1.915722131729126, + "learning_rate": 2.365962632006499e-06, + "loss": 2.1306, + "step": 1586 + }, + { + "epoch": 1.7391780821917808, + "grad_norm": 1.8876861333847046, + "learning_rate": 2.3639317627944764e-06, + "loss": 2.1303, + "step": 1587 + }, + { + "epoch": 1.7402739726027399, + "grad_norm": 1.9599688053131104, + "learning_rate": 2.3619008935824537e-06, + "loss": 2.1513, + "step": 1588 + }, + { + "epoch": 1.7413698630136987, + "grad_norm": 1.90068519115448, + "learning_rate": 2.3598700243704305e-06, + "loss": 2.1459, + "step": 1589 + }, + { + "epoch": 1.7424657534246575, + "grad_norm": 1.920277714729309, + "learning_rate": 2.357839155158408e-06, + "loss": 2.1453, + "step": 1590 + }, + { + "epoch": 1.7424657534246575, + "eval_loss": 2.2699742317199707, + "eval_runtime": 50.0211, + "eval_samples_per_second": 19.992, + "eval_steps_per_second": 0.84, + "step": 1590 + }, + { + "epoch": 1.7435616438356165, + "grad_norm": 1.9827759265899658, + "learning_rate": 2.355808285946385e-06, + "loss": 2.1393, + "step": 1591 + }, + { + "epoch": 1.7446575342465753, + "grad_norm": 1.8602412939071655, + "learning_rate": 2.3537774167343624e-06, + "loss": 2.1317, + "step": 1592 + }, + { + "epoch": 1.7457534246575341, + "grad_norm": 1.964829444885254, + "learning_rate": 2.3517465475223396e-06, + "loss": 2.1552, + "step": 1593 + }, + { + "epoch": 1.7468493150684932, + "grad_norm": 2.0124034881591797, + "learning_rate": 2.349715678310317e-06, + "loss": 2.1322, + "step": 1594 + }, + { + "epoch": 1.7479452054794522, + "grad_norm": 1.9786088466644287, + "learning_rate": 2.347684809098294e-06, + "loss": 2.1435, + "step": 1595 + }, + { + "epoch": 1.749041095890411, + "grad_norm": 1.9575989246368408, + "learning_rate": 2.3456539398862715e-06, + "loss": 2.1032, + "step": 1596 + }, + { + "epoch": 1.7501369863013698, + "grad_norm": 2.0460586547851562, + "learning_rate": 2.3436230706742487e-06, + "loss": 2.1595, + "step": 1597 + }, + { + "epoch": 1.7512328767123289, + "grad_norm": 1.9904825687408447, + "learning_rate": 2.341592201462226e-06, + "loss": 2.1025, + "step": 1598 + }, + { + "epoch": 1.7523287671232877, + "grad_norm": 2.0797626972198486, + "learning_rate": 2.3395613322502033e-06, + "loss": 2.1221, + "step": 1599 + }, + { + "epoch": 1.7534246575342465, + "grad_norm": 2.0380358695983887, + "learning_rate": 2.3375304630381806e-06, + "loss": 2.1166, + "step": 1600 + }, + { + "epoch": 1.7534246575342465, + "eval_loss": 2.2704837322235107, + "eval_runtime": 51.9207, + "eval_samples_per_second": 19.26, + "eval_steps_per_second": 0.809, + "step": 1600 + }, + { + "epoch": 1.7545205479452055, + "grad_norm": 2.009023427963257, + "learning_rate": 2.335499593826158e-06, + "loss": 2.0922, + "step": 1601 + }, + { + "epoch": 1.7556164383561645, + "grad_norm": 2.043168783187866, + "learning_rate": 2.333468724614135e-06, + "loss": 2.0962, + "step": 1602 + }, + { + "epoch": 1.7567123287671231, + "grad_norm": 2.1130902767181396, + "learning_rate": 2.3314378554021124e-06, + "loss": 2.1318, + "step": 1603 + }, + { + "epoch": 1.7578082191780822, + "grad_norm": 1.9391744136810303, + "learning_rate": 2.3294069861900897e-06, + "loss": 2.155, + "step": 1604 + }, + { + "epoch": 1.7589041095890412, + "grad_norm": 2.0810694694519043, + "learning_rate": 2.327376116978067e-06, + "loss": 2.1066, + "step": 1605 + }, + { + "epoch": 1.76, + "grad_norm": 2.029707431793213, + "learning_rate": 2.3253452477660442e-06, + "loss": 2.0897, + "step": 1606 + }, + { + "epoch": 1.7610958904109588, + "grad_norm": 1.928080439567566, + "learning_rate": 2.3233143785540215e-06, + "loss": 2.1589, + "step": 1607 + }, + { + "epoch": 1.7621917808219179, + "grad_norm": 1.8527079820632935, + "learning_rate": 2.3212835093419988e-06, + "loss": 2.1005, + "step": 1608 + }, + { + "epoch": 1.7632876712328767, + "grad_norm": 2.142939567565918, + "learning_rate": 2.319252640129976e-06, + "loss": 2.1499, + "step": 1609 + }, + { + "epoch": 1.7643835616438355, + "grad_norm": 2.040922164916992, + "learning_rate": 2.3172217709179533e-06, + "loss": 2.0997, + "step": 1610 + }, + { + "epoch": 1.7643835616438355, + "eval_loss": 2.2691540718078613, + "eval_runtime": 42.4134, + "eval_samples_per_second": 23.577, + "eval_steps_per_second": 0.99, + "step": 1610 + }, + { + "epoch": 1.7654794520547945, + "grad_norm": 1.988370656967163, + "learning_rate": 2.3151909017059306e-06, + "loss": 2.1795, + "step": 1611 + }, + { + "epoch": 1.7665753424657535, + "grad_norm": 1.996656060218811, + "learning_rate": 2.313160032493908e-06, + "loss": 2.1536, + "step": 1612 + }, + { + "epoch": 1.7676712328767124, + "grad_norm": 2.04560923576355, + "learning_rate": 2.311129163281885e-06, + "loss": 2.0781, + "step": 1613 + }, + { + "epoch": 1.7687671232876712, + "grad_norm": 1.8977863788604736, + "learning_rate": 2.309098294069862e-06, + "loss": 2.1312, + "step": 1614 + }, + { + "epoch": 1.7698630136986302, + "grad_norm": 2.026904821395874, + "learning_rate": 2.3070674248578393e-06, + "loss": 2.0998, + "step": 1615 + }, + { + "epoch": 1.770958904109589, + "grad_norm": 1.8714247941970825, + "learning_rate": 2.3050365556458166e-06, + "loss": 2.0875, + "step": 1616 + }, + { + "epoch": 1.7720547945205478, + "grad_norm": 2.11859130859375, + "learning_rate": 2.303005686433794e-06, + "loss": 2.1361, + "step": 1617 + }, + { + "epoch": 1.7731506849315068, + "grad_norm": 2.0592799186706543, + "learning_rate": 2.300974817221771e-06, + "loss": 2.1394, + "step": 1618 + }, + { + "epoch": 1.7742465753424659, + "grad_norm": 2.0358338356018066, + "learning_rate": 2.2989439480097484e-06, + "loss": 2.104, + "step": 1619 + }, + { + "epoch": 1.7753424657534247, + "grad_norm": 2.0237109661102295, + "learning_rate": 2.2969130787977257e-06, + "loss": 2.0823, + "step": 1620 + }, + { + "epoch": 1.7753424657534247, + "eval_loss": 2.27001690864563, + "eval_runtime": 43.9168, + "eval_samples_per_second": 22.77, + "eval_steps_per_second": 0.956, + "step": 1620 + }, + { + "epoch": 1.7764383561643835, + "grad_norm": 1.8729594945907593, + "learning_rate": 2.294882209585703e-06, + "loss": 2.1243, + "step": 1621 + }, + { + "epoch": 1.7775342465753425, + "grad_norm": 1.9390734434127808, + "learning_rate": 2.2928513403736802e-06, + "loss": 2.1469, + "step": 1622 + }, + { + "epoch": 1.7786301369863013, + "grad_norm": 2.0409750938415527, + "learning_rate": 2.2908204711616575e-06, + "loss": 2.1062, + "step": 1623 + }, + { + "epoch": 1.7797260273972602, + "grad_norm": 1.9116135835647583, + "learning_rate": 2.2887896019496348e-06, + "loss": 2.1208, + "step": 1624 + }, + { + "epoch": 1.7808219178082192, + "grad_norm": 1.9144036769866943, + "learning_rate": 2.286758732737612e-06, + "loss": 2.1628, + "step": 1625 + }, + { + "epoch": 1.7819178082191782, + "grad_norm": 1.8522213697433472, + "learning_rate": 2.284727863525589e-06, + "loss": 2.1168, + "step": 1626 + }, + { + "epoch": 1.783013698630137, + "grad_norm": 2.087541103363037, + "learning_rate": 2.282696994313566e-06, + "loss": 2.1195, + "step": 1627 + }, + { + "epoch": 1.7841095890410958, + "grad_norm": 2.0399985313415527, + "learning_rate": 2.2806661251015434e-06, + "loss": 2.1557, + "step": 1628 + }, + { + "epoch": 1.7852054794520549, + "grad_norm": 1.884964108467102, + "learning_rate": 2.2786352558895207e-06, + "loss": 2.1408, + "step": 1629 + }, + { + "epoch": 1.7863013698630137, + "grad_norm": 1.901802897453308, + "learning_rate": 2.276604386677498e-06, + "loss": 2.1145, + "step": 1630 + }, + { + "epoch": 1.7863013698630137, + "eval_loss": 2.2690439224243164, + "eval_runtime": 44.8662, + "eval_samples_per_second": 22.289, + "eval_steps_per_second": 0.936, + "step": 1630 + }, + { + "epoch": 1.7873972602739725, + "grad_norm": 1.8818615674972534, + "learning_rate": 2.2745735174654753e-06, + "loss": 2.0879, + "step": 1631 + }, + { + "epoch": 1.7884931506849315, + "grad_norm": 1.9345687627792358, + "learning_rate": 2.2725426482534526e-06, + "loss": 2.1029, + "step": 1632 + }, + { + "epoch": 1.7895890410958906, + "grad_norm": 1.868314504623413, + "learning_rate": 2.27051177904143e-06, + "loss": 2.1435, + "step": 1633 + }, + { + "epoch": 1.7906849315068492, + "grad_norm": 1.8868515491485596, + "learning_rate": 2.268480909829407e-06, + "loss": 2.1294, + "step": 1634 + }, + { + "epoch": 1.7917808219178082, + "grad_norm": 1.9929358959197998, + "learning_rate": 2.2664500406173844e-06, + "loss": 2.1165, + "step": 1635 + }, + { + "epoch": 1.7928767123287672, + "grad_norm": 1.9766895771026611, + "learning_rate": 2.2644191714053617e-06, + "loss": 2.1327, + "step": 1636 + }, + { + "epoch": 1.793972602739726, + "grad_norm": 2.03336238861084, + "learning_rate": 2.262388302193339e-06, + "loss": 2.1473, + "step": 1637 + }, + { + "epoch": 1.7950684931506848, + "grad_norm": 1.8569762706756592, + "learning_rate": 2.260357432981316e-06, + "loss": 2.125, + "step": 1638 + }, + { + "epoch": 1.7961643835616439, + "grad_norm": 1.9522744417190552, + "learning_rate": 2.2583265637692935e-06, + "loss": 2.1238, + "step": 1639 + }, + { + "epoch": 1.7972602739726027, + "grad_norm": 1.9175306558609009, + "learning_rate": 2.2562956945572708e-06, + "loss": 2.1332, + "step": 1640 + }, + { + "epoch": 1.7972602739726027, + "eval_loss": 2.2702176570892334, + "eval_runtime": 44.1369, + "eval_samples_per_second": 22.657, + "eval_steps_per_second": 0.952, + "step": 1640 + }, + { + "epoch": 1.7983561643835615, + "grad_norm": 2.109614849090576, + "learning_rate": 2.254264825345248e-06, + "loss": 2.1252, + "step": 1641 + }, + { + "epoch": 1.7994520547945205, + "grad_norm": 2.017613410949707, + "learning_rate": 2.2522339561332253e-06, + "loss": 2.1093, + "step": 1642 + }, + { + "epoch": 1.8005479452054796, + "grad_norm": 2.000864028930664, + "learning_rate": 2.2502030869212026e-06, + "loss": 2.1243, + "step": 1643 + }, + { + "epoch": 1.8016438356164384, + "grad_norm": 1.9952439069747925, + "learning_rate": 2.24817221770918e-06, + "loss": 2.1213, + "step": 1644 + }, + { + "epoch": 1.8027397260273972, + "grad_norm": 2.0456748008728027, + "learning_rate": 2.246141348497157e-06, + "loss": 2.1621, + "step": 1645 + }, + { + "epoch": 1.8038356164383562, + "grad_norm": 1.9011833667755127, + "learning_rate": 2.2441104792851344e-06, + "loss": 2.1426, + "step": 1646 + }, + { + "epoch": 1.804931506849315, + "grad_norm": 1.9874658584594727, + "learning_rate": 2.2420796100731117e-06, + "loss": 2.1085, + "step": 1647 + }, + { + "epoch": 1.8060273972602738, + "grad_norm": 2.0237057209014893, + "learning_rate": 2.240048740861089e-06, + "loss": 2.1394, + "step": 1648 + }, + { + "epoch": 1.8071232876712329, + "grad_norm": 2.0127220153808594, + "learning_rate": 2.2380178716490662e-06, + "loss": 2.1474, + "step": 1649 + }, + { + "epoch": 1.808219178082192, + "grad_norm": 2.051722764968872, + "learning_rate": 2.2359870024370435e-06, + "loss": 2.0444, + "step": 1650 + }, + { + "epoch": 1.808219178082192, + "eval_loss": 2.2701034545898438, + "eval_runtime": 44.4338, + "eval_samples_per_second": 22.505, + "eval_steps_per_second": 0.945, + "step": 1650 + }, + { + "epoch": 1.8093150684931507, + "grad_norm": 2.0204219818115234, + "learning_rate": 2.2339561332250204e-06, + "loss": 2.1445, + "step": 1651 + }, + { + "epoch": 1.8104109589041095, + "grad_norm": 2.106220006942749, + "learning_rate": 2.2319252640129976e-06, + "loss": 2.153, + "step": 1652 + }, + { + "epoch": 1.8115068493150686, + "grad_norm": 1.8632844686508179, + "learning_rate": 2.229894394800975e-06, + "loss": 2.137, + "step": 1653 + }, + { + "epoch": 1.8126027397260274, + "grad_norm": 2.377120018005371, + "learning_rate": 2.227863525588952e-06, + "loss": 2.0903, + "step": 1654 + }, + { + "epoch": 1.8136986301369862, + "grad_norm": 2.1345274448394775, + "learning_rate": 2.2258326563769295e-06, + "loss": 2.066, + "step": 1655 + }, + { + "epoch": 1.8147945205479452, + "grad_norm": 1.8935933113098145, + "learning_rate": 2.2238017871649067e-06, + "loss": 2.083, + "step": 1656 + }, + { + "epoch": 1.8158904109589042, + "grad_norm": 1.8766530752182007, + "learning_rate": 2.221770917952884e-06, + "loss": 2.0792, + "step": 1657 + }, + { + "epoch": 1.816986301369863, + "grad_norm": 2.24202823638916, + "learning_rate": 2.2197400487408613e-06, + "loss": 2.1276, + "step": 1658 + }, + { + "epoch": 1.8180821917808219, + "grad_norm": 2.168358325958252, + "learning_rate": 2.2177091795288386e-06, + "loss": 2.1497, + "step": 1659 + }, + { + "epoch": 1.819178082191781, + "grad_norm": 2.052833080291748, + "learning_rate": 2.215678310316816e-06, + "loss": 2.0853, + "step": 1660 + }, + { + "epoch": 1.819178082191781, + "eval_loss": 2.270122528076172, + "eval_runtime": 44.1793, + "eval_samples_per_second": 22.635, + "eval_steps_per_second": 0.951, + "step": 1660 + }, + { + "epoch": 1.8202739726027397, + "grad_norm": 2.09708571434021, + "learning_rate": 2.213647441104793e-06, + "loss": 2.1151, + "step": 1661 + }, + { + "epoch": 1.8213698630136985, + "grad_norm": 1.928503155708313, + "learning_rate": 2.2116165718927704e-06, + "loss": 2.1234, + "step": 1662 + }, + { + "epoch": 1.8224657534246576, + "grad_norm": 1.9545074701309204, + "learning_rate": 2.2095857026807473e-06, + "loss": 2.1134, + "step": 1663 + }, + { + "epoch": 1.8235616438356166, + "grad_norm": 2.1161389350891113, + "learning_rate": 2.2075548334687245e-06, + "loss": 2.1347, + "step": 1664 + }, + { + "epoch": 1.8246575342465754, + "grad_norm": 1.9292371273040771, + "learning_rate": 2.205523964256702e-06, + "loss": 2.0911, + "step": 1665 + }, + { + "epoch": 1.8257534246575342, + "grad_norm": 1.8706191778182983, + "learning_rate": 2.203493095044679e-06, + "loss": 2.0917, + "step": 1666 + }, + { + "epoch": 1.8268493150684932, + "grad_norm": 1.9825111627578735, + "learning_rate": 2.2014622258326564e-06, + "loss": 2.125, + "step": 1667 + }, + { + "epoch": 1.827945205479452, + "grad_norm": 2.0262181758880615, + "learning_rate": 2.1994313566206336e-06, + "loss": 2.153, + "step": 1668 + }, + { + "epoch": 1.8290410958904109, + "grad_norm": 2.0071237087249756, + "learning_rate": 2.197400487408611e-06, + "loss": 2.1625, + "step": 1669 + }, + { + "epoch": 1.83013698630137, + "grad_norm": 1.940076231956482, + "learning_rate": 2.195369618196588e-06, + "loss": 2.1052, + "step": 1670 + }, + { + "epoch": 1.83013698630137, + "eval_loss": 2.2711451053619385, + "eval_runtime": 43.9311, + "eval_samples_per_second": 22.763, + "eval_steps_per_second": 0.956, + "step": 1670 + }, + { + "epoch": 1.831232876712329, + "grad_norm": 2.103487014770508, + "learning_rate": 2.1933387489845655e-06, + "loss": 2.087, + "step": 1671 + }, + { + "epoch": 1.8323287671232875, + "grad_norm": 1.935276985168457, + "learning_rate": 2.1913078797725427e-06, + "loss": 2.1196, + "step": 1672 + }, + { + "epoch": 1.8334246575342465, + "grad_norm": 1.987322449684143, + "learning_rate": 2.18927701056052e-06, + "loss": 2.1354, + "step": 1673 + }, + { + "epoch": 1.8345205479452056, + "grad_norm": 2.0403592586517334, + "learning_rate": 2.1872461413484973e-06, + "loss": 2.1715, + "step": 1674 + }, + { + "epoch": 1.8356164383561644, + "grad_norm": 2.0403592586517334, + "learning_rate": 2.1872461413484973e-06, + "loss": 2.1153, + "step": 1675 + }, + { + "epoch": 1.8367123287671232, + "grad_norm": 2.160753011703491, + "learning_rate": 2.1852152721364746e-06, + "loss": 2.1105, + "step": 1676 + }, + { + "epoch": 1.8378082191780822, + "grad_norm": 1.8320366144180298, + "learning_rate": 2.183184402924452e-06, + "loss": 2.0853, + "step": 1677 + }, + { + "epoch": 1.838904109589041, + "grad_norm": 1.9792540073394775, + "learning_rate": 2.181153533712429e-06, + "loss": 2.1478, + "step": 1678 + }, + { + "epoch": 1.8399999999999999, + "grad_norm": 2.0344042778015137, + "learning_rate": 2.1791226645004064e-06, + "loss": 2.1177, + "step": 1679 + }, + { + "epoch": 1.841095890410959, + "grad_norm": 1.8672499656677246, + "learning_rate": 2.1770917952883837e-06, + "loss": 2.1098, + "step": 1680 + }, + { + "epoch": 1.841095890410959, + "eval_loss": 2.2701213359832764, + "eval_runtime": 43.0484, + "eval_samples_per_second": 23.23, + "eval_steps_per_second": 0.976, + "step": 1680 + }, + { + "epoch": 1.842191780821918, + "grad_norm": 1.9291775226593018, + "learning_rate": 2.175060926076361e-06, + "loss": 2.1659, + "step": 1681 + }, + { + "epoch": 1.8432876712328767, + "grad_norm": 1.9650444984436035, + "learning_rate": 2.1730300568643382e-06, + "loss": 2.1764, + "step": 1682 + }, + { + "epoch": 1.8443835616438355, + "grad_norm": 1.904618501663208, + "learning_rate": 2.1709991876523155e-06, + "loss": 2.1047, + "step": 1683 + }, + { + "epoch": 1.8454794520547946, + "grad_norm": 2.0084192752838135, + "learning_rate": 2.1689683184402928e-06, + "loss": 2.0719, + "step": 1684 + }, + { + "epoch": 1.8465753424657534, + "grad_norm": 2.0233888626098633, + "learning_rate": 2.16693744922827e-06, + "loss": 2.1482, + "step": 1685 + }, + { + "epoch": 1.8476712328767122, + "grad_norm": 1.9512053728103638, + "learning_rate": 2.1649065800162473e-06, + "loss": 2.1116, + "step": 1686 + }, + { + "epoch": 1.8487671232876712, + "grad_norm": 1.9261832237243652, + "learning_rate": 2.1628757108042246e-06, + "loss": 2.1025, + "step": 1687 + }, + { + "epoch": 1.8498630136986303, + "grad_norm": 2.0704665184020996, + "learning_rate": 2.160844841592202e-06, + "loss": 2.0943, + "step": 1688 + }, + { + "epoch": 1.850958904109589, + "grad_norm": 1.8882733583450317, + "learning_rate": 2.1588139723801787e-06, + "loss": 2.09, + "step": 1689 + }, + { + "epoch": 1.8520547945205479, + "grad_norm": 1.9136085510253906, + "learning_rate": 2.156783103168156e-06, + "loss": 2.1074, + "step": 1690 + }, + { + "epoch": 1.8520547945205479, + "eval_loss": 2.2687957286834717, + "eval_runtime": 44.1571, + "eval_samples_per_second": 22.646, + "eval_steps_per_second": 0.951, + "step": 1690 + }, + { + "epoch": 1.853150684931507, + "grad_norm": 1.9338082075119019, + "learning_rate": 2.1547522339561333e-06, + "loss": 2.0945, + "step": 1691 + }, + { + "epoch": 1.8542465753424657, + "grad_norm": 1.862281084060669, + "learning_rate": 2.1527213647441106e-06, + "loss": 2.1355, + "step": 1692 + }, + { + "epoch": 1.8553424657534245, + "grad_norm": 1.8184510469436646, + "learning_rate": 2.150690495532088e-06, + "loss": 2.1628, + "step": 1693 + }, + { + "epoch": 1.8564383561643836, + "grad_norm": 1.9667037725448608, + "learning_rate": 2.148659626320065e-06, + "loss": 2.1308, + "step": 1694 + }, + { + "epoch": 1.8575342465753426, + "grad_norm": 2.0214083194732666, + "learning_rate": 2.1466287571080424e-06, + "loss": 2.1363, + "step": 1695 + }, + { + "epoch": 1.8586301369863014, + "grad_norm": 1.8136624097824097, + "learning_rate": 2.1445978878960197e-06, + "loss": 2.1074, + "step": 1696 + }, + { + "epoch": 1.8597260273972602, + "grad_norm": 2.1120195388793945, + "learning_rate": 2.142567018683997e-06, + "loss": 2.1115, + "step": 1697 + }, + { + "epoch": 1.8608219178082193, + "grad_norm": 1.9148973226547241, + "learning_rate": 2.140536149471974e-06, + "loss": 2.1554, + "step": 1698 + }, + { + "epoch": 1.861917808219178, + "grad_norm": 1.9684176445007324, + "learning_rate": 2.1385052802599515e-06, + "loss": 2.0944, + "step": 1699 + }, + { + "epoch": 1.8630136986301369, + "grad_norm": 2.0816597938537598, + "learning_rate": 2.1364744110479288e-06, + "loss": 2.1063, + "step": 1700 + }, + { + "epoch": 1.8630136986301369, + "eval_loss": 2.2695536613464355, + "eval_runtime": 48.9529, + "eval_samples_per_second": 20.428, + "eval_steps_per_second": 0.858, + "step": 1700 + } + ], + "logging_steps": 1, + "max_steps": 2736, + "num_input_tokens_seen": 0, + "num_train_epochs": 3, + "save_steps": 100, + "stateful_callbacks": { + "TrainerControl": { + "args": { + "should_epoch_stop": false, + "should_evaluate": false, + "should_log": false, + "should_save": true, + "should_training_stop": false + }, + "attributes": {} + } + }, + "total_flos": 5.299425995915264e+17, + "train_batch_size": 3, + "trial_name": null, + "trial_params": null +} diff --git a/babylm_shuffle_nondeterministic_10M_seed0/runs/checkpoint-1700/zero_to_fp32.py b/babylm_shuffle_nondeterministic_10M_seed0/runs/checkpoint-1700/zero_to_fp32.py new file mode 100644 index 0000000000000000000000000000000000000000..24cc342e78d1a006c782b3a4cd68d9ce786d8fd8 --- /dev/null +++ b/babylm_shuffle_nondeterministic_10M_seed0/runs/checkpoint-1700/zero_to_fp32.py @@ -0,0 +1,604 @@ +#!/usr/bin/env python + +# Copyright (c) Microsoft Corporation. +# SPDX-License-Identifier: Apache-2.0 + +# DeepSpeed Team + +# This script extracts fp32 consolidated weights from a zero 1, 2 and 3 DeepSpeed checkpoints. It gets +# copied into the top level checkpoint dir, so the user can easily do the conversion at any point in +# the future. Once extracted, the weights don't require DeepSpeed and can be used in any +# application. +# +# example: python zero_to_fp32.py . pytorch_model.bin + +import argparse +import torch +import glob +import math +import os +import re +from collections import OrderedDict +from dataclasses import dataclass + +# while this script doesn't use deepspeed to recover data, since the checkpoints are pickled with +# DeepSpeed data structures it has to be available in the current python environment. +from deepspeed.utils import logger +from deepspeed.checkpoint.constants import (DS_VERSION, OPTIMIZER_STATE_DICT, SINGLE_PARTITION_OF_FP32_GROUPS, + FP32_FLAT_GROUPS, ZERO_STAGE, PARTITION_COUNT, PARAM_SHAPES, BUFFER_NAMES, + FROZEN_PARAM_SHAPES, FROZEN_PARAM_FRAGMENTS) + + +@dataclass +class zero_model_state: + buffers: dict() + param_shapes: dict() + shared_params: list + ds_version: int + frozen_param_shapes: dict() + frozen_param_fragments: dict() + + +debug = 0 + +# load to cpu +device = torch.device('cpu') + + +def atoi(text): + return int(text) if text.isdigit() else text + + +def natural_keys(text): + ''' + alist.sort(key=natural_keys) sorts in human order + http://nedbatchelder.com/blog/200712/human_sorting.html + (See Toothy's implementation in the comments) + ''' + return [atoi(c) for c in re.split(r'(\d+)', text)] + + +def get_model_state_file(checkpoint_dir, zero_stage): + if not os.path.isdir(checkpoint_dir): + raise FileNotFoundError(f"Directory '{checkpoint_dir}' doesn't exist") + + # there should be only one file + if zero_stage <= 2: + file = os.path.join(checkpoint_dir, "mp_rank_00_model_states.pt") + elif zero_stage == 3: + file = os.path.join(checkpoint_dir, "zero_pp_rank_0_mp_rank_00_model_states.pt") + + if not os.path.exists(file): + raise FileNotFoundError(f"can't find model states file at '{file}'") + + return file + + +def get_checkpoint_files(checkpoint_dir, glob_pattern): + # XXX: need to test that this simple glob rule works for multi-node setup too + ckpt_files = sorted(glob.glob(os.path.join(checkpoint_dir, glob_pattern)), key=natural_keys) + + if len(ckpt_files) == 0: + raise FileNotFoundError(f"can't find {glob_pattern} files in directory '{checkpoint_dir}'") + + return ckpt_files + + +def get_optim_files(checkpoint_dir): + return get_checkpoint_files(checkpoint_dir, "*_optim_states.pt") + + +def get_model_state_files(checkpoint_dir): + return get_checkpoint_files(checkpoint_dir, "*_model_states.pt") + + +def parse_model_states(files): + zero_model_states = [] + for file in files: + state_dict = torch.load(file, map_location=device) + + if BUFFER_NAMES not in state_dict: + raise ValueError(f"{file} is not a model state checkpoint") + buffer_names = state_dict[BUFFER_NAMES] + if debug: + print("Found buffers:", buffer_names) + + # recover just the buffers while restoring them to fp32 if they were saved in fp16 + buffers = {k: v.float() for k, v in state_dict["module"].items() if k in buffer_names} + param_shapes = state_dict[PARAM_SHAPES] + + # collect parameters that are included in param_shapes + param_names = [] + for s in param_shapes: + for name in s.keys(): + param_names.append(name) + + # update with frozen parameters + frozen_param_shapes = state_dict.get(FROZEN_PARAM_SHAPES, None) + if frozen_param_shapes is not None: + if debug: + print(f"Found frozen_param_shapes: {frozen_param_shapes}") + param_names += list(frozen_param_shapes.keys()) + + # handle shared params + shared_params = [[k, v] for k, v in state_dict["shared_params"].items()] + + ds_version = state_dict.get(DS_VERSION, None) + + frozen_param_fragments = state_dict.get(FROZEN_PARAM_FRAGMENTS, None) + + z_model_state = zero_model_state(buffers=buffers, + param_shapes=param_shapes, + shared_params=shared_params, + ds_version=ds_version, + frozen_param_shapes=frozen_param_shapes, + frozen_param_fragments=frozen_param_fragments) + zero_model_states.append(z_model_state) + + return zero_model_states + + +def parse_optim_states(files, ds_checkpoint_dir): + + total_files = len(files) + state_dicts = [] + for f in files: + state_dict = torch.load(f, map_location=device) + # immediately discard the potentially huge 2 optimizer states as we only care for fp32 master weights + # and also handle the case where it was already removed by another helper script + state_dict["optimizer_state_dict"].pop("optimizer_state_dict", None) + state_dicts.append(state_dict) + + if not ZERO_STAGE in state_dicts[0][OPTIMIZER_STATE_DICT]: + raise ValueError(f"{files[0]} is not a zero checkpoint") + zero_stage = state_dicts[0][OPTIMIZER_STATE_DICT][ZERO_STAGE] + world_size = state_dicts[0][OPTIMIZER_STATE_DICT][PARTITION_COUNT] + + # For ZeRO-2 each param group can have different partition_count as data parallelism for expert + # parameters can be different from data parallelism for non-expert parameters. So we can just + # use the max of the partition_count to get the dp world_size. + + if type(world_size) is list: + world_size = max(world_size) + + if world_size != total_files: + raise ValueError( + f"Expected {world_size} of '*_optim_states.pt' under '{ds_checkpoint_dir}' but found {total_files} files. " + "Possibly due to an overwrite of an old checkpoint, or a checkpoint didn't get saved by one or more processes." + ) + + # the groups are named differently in each stage + if zero_stage <= 2: + fp32_groups_key = SINGLE_PARTITION_OF_FP32_GROUPS + elif zero_stage == 3: + fp32_groups_key = FP32_FLAT_GROUPS + else: + raise ValueError(f"unknown zero stage {zero_stage}") + + if zero_stage <= 2: + fp32_flat_groups = [state_dicts[i][OPTIMIZER_STATE_DICT][fp32_groups_key] for i in range(len(state_dicts))] + elif zero_stage == 3: + # if there is more than one param group, there will be multiple flattened tensors - one + # flattened tensor per group - for simplicity merge them into a single tensor + # + # XXX: could make the script more memory efficient for when there are multiple groups - it + # will require matching the sub-lists of param_shapes for each param group flattened tensor + + fp32_flat_groups = [ + torch.cat(state_dicts[i][OPTIMIZER_STATE_DICT][fp32_groups_key], 0) for i in range(len(state_dicts)) + ] + + return zero_stage, world_size, fp32_flat_groups + + +def _get_fp32_state_dict_from_zero_checkpoint(ds_checkpoint_dir, exclude_frozen_parameters): + """ + Returns fp32 state_dict reconstructed from ds checkpoint + + Args: + - ``ds_checkpoint_dir``: path to the deepspeed checkpoint folder (where the optimizer files are) + + """ + print(f"Processing zero checkpoint '{ds_checkpoint_dir}'") + + optim_files = get_optim_files(ds_checkpoint_dir) + zero_stage, world_size, fp32_flat_groups = parse_optim_states(optim_files, ds_checkpoint_dir) + print(f"Detected checkpoint of type zero stage {zero_stage}, world_size: {world_size}") + + model_files = get_model_state_files(ds_checkpoint_dir) + + zero_model_states = parse_model_states(model_files) + print(f'Parsing checkpoint created by deepspeed=={zero_model_states[0].ds_version}') + + if zero_stage <= 2: + return _get_fp32_state_dict_from_zero2_checkpoint(world_size, fp32_flat_groups, zero_model_states, + exclude_frozen_parameters) + elif zero_stage == 3: + return _get_fp32_state_dict_from_zero3_checkpoint(world_size, fp32_flat_groups, zero_model_states, + exclude_frozen_parameters) + + +def _zero2_merge_frozen_params(state_dict, zero_model_states): + if zero_model_states[0].frozen_param_shapes is None or len(zero_model_states[0].frozen_param_shapes) == 0: + return + + frozen_param_shapes = zero_model_states[0].frozen_param_shapes + frozen_param_fragments = zero_model_states[0].frozen_param_fragments + + if debug: + num_elem = sum(s.numel() for s in frozen_param_shapes.values()) + print(f'rank 0: {FROZEN_PARAM_SHAPES}.numel = {num_elem}') + + wanted_params = len(frozen_param_shapes) + wanted_numel = sum(s.numel() for s in frozen_param_shapes.values()) + avail_numel = sum([p.numel() for p in frozen_param_fragments.values()]) + print(f'Frozen params: Have {avail_numel} numels to process.') + print(f'Frozen params: Need {wanted_numel} numels in {wanted_params} params') + + total_params = 0 + total_numel = 0 + for name, shape in frozen_param_shapes.items(): + total_params += 1 + unpartitioned_numel = shape.numel() + total_numel += unpartitioned_numel + + state_dict[name] = frozen_param_fragments[name] + + if debug: + print(f"{name} full shape: {shape} unpartitioned numel {unpartitioned_numel} ") + + print(f"Reconstructed Frozen fp32 state dict with {total_params} params {total_numel} elements") + + +def _has_callable(obj, fn): + attr = getattr(obj, fn, None) + return callable(attr) + + +def _zero2_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states): + param_shapes = zero_model_states[0].param_shapes + + # Reconstruction protocol: + # + # XXX: document this + + if debug: + for i in range(world_size): + for j in range(len(fp32_flat_groups[0])): + print(f"{FP32_FLAT_GROUPS}[{i}][{j}].shape={fp32_flat_groups[i][j].shape}") + + # XXX: memory usage doubles here (zero2) + num_param_groups = len(fp32_flat_groups[0]) + merged_single_partition_of_fp32_groups = [] + for i in range(num_param_groups): + merged_partitions = [sd[i] for sd in fp32_flat_groups] + full_single_fp32_vector = torch.cat(merged_partitions, 0) + merged_single_partition_of_fp32_groups.append(full_single_fp32_vector) + avail_numel = sum( + [full_single_fp32_vector.numel() for full_single_fp32_vector in merged_single_partition_of_fp32_groups]) + + if debug: + wanted_params = sum([len(shapes) for shapes in param_shapes]) + wanted_numel = sum([sum(shape.numel() for shape in shapes.values()) for shapes in param_shapes]) + # not asserting if there is a mismatch due to possible padding + print(f"Have {avail_numel} numels to process.") + print(f"Need {wanted_numel} numels in {wanted_params} params.") + + # params + # XXX: for huge models that can't fit into the host's RAM we will have to recode this to support + # out-of-core computing solution + total_numel = 0 + total_params = 0 + for shapes, full_single_fp32_vector in zip(param_shapes, merged_single_partition_of_fp32_groups): + offset = 0 + avail_numel = full_single_fp32_vector.numel() + for name, shape in shapes.items(): + + unpartitioned_numel = shape.numel() if _has_callable(shape, 'numel') else math.prod(shape) + total_numel += unpartitioned_numel + total_params += 1 + + if debug: + print(f"{name} full shape: {shape} unpartitioned numel {unpartitioned_numel} ") + state_dict[name] = full_single_fp32_vector.narrow(0, offset, unpartitioned_numel).view(shape) + offset += unpartitioned_numel + + # Z2 started to align to 2*world_size to improve nccl performance. Therefore both offset and + # avail_numel can differ by anywhere between 0..2*world_size. Due to two unrelated complex + # paddings performed in the code it's almost impossible to predict the exact numbers w/o the + # live optimizer object, so we are checking that the numbers are within the right range + align_to = 2 * world_size + + def zero2_align(x): + return align_to * math.ceil(x / align_to) + + if debug: + print(f"original offset={offset}, avail_numel={avail_numel}") + + offset = zero2_align(offset) + avail_numel = zero2_align(avail_numel) + + if debug: + print(f"aligned offset={offset}, avail_numel={avail_numel}") + + # Sanity check + if offset != avail_numel: + raise ValueError(f"consumed {offset} numels out of {avail_numel} - something is wrong") + + print(f"Reconstructed fp32 state dict with {total_params} params {total_numel} elements") + + +def _get_fp32_state_dict_from_zero2_checkpoint(world_size, fp32_flat_groups, zero_model_states, + exclude_frozen_parameters): + state_dict = OrderedDict() + + # buffers + buffers = zero_model_states[0].buffers + state_dict.update(buffers) + if debug: + print(f"added {len(buffers)} buffers") + + if not exclude_frozen_parameters: + _zero2_merge_frozen_params(state_dict, zero_model_states) + + _zero2_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states) + + # recover shared parameters + for pair in zero_model_states[0].shared_params: + if pair[1] in state_dict: + state_dict[pair[0]] = state_dict[pair[1]] + + return state_dict + + +def zero3_partitioned_param_info(unpartitioned_numel, world_size): + remainder = unpartitioned_numel % world_size + padding_numel = (world_size - remainder) if remainder else 0 + partitioned_numel = math.ceil(unpartitioned_numel / world_size) + return partitioned_numel, padding_numel + + +def _zero3_merge_frozen_params(state_dict, world_size, zero_model_states): + if zero_model_states[0].frozen_param_shapes is None or len(zero_model_states[0].frozen_param_shapes) == 0: + return + + if debug: + for i in range(world_size): + num_elem = sum(s.numel() for s in zero_model_states[i].frozen_param_fragments.values()) + print(f'rank {i}: {FROZEN_PARAM_SHAPES}.numel = {num_elem}') + + frozen_param_shapes = zero_model_states[0].frozen_param_shapes + wanted_params = len(frozen_param_shapes) + wanted_numel = sum(s.numel() for s in frozen_param_shapes.values()) + avail_numel = sum([p.numel() for p in zero_model_states[0].frozen_param_fragments.values()]) * world_size + print(f'Frozen params: Have {avail_numel} numels to process.') + print(f'Frozen params: Need {wanted_numel} numels in {wanted_params} params') + + total_params = 0 + total_numel = 0 + for name, shape in zero_model_states[0].frozen_param_shapes.items(): + total_params += 1 + unpartitioned_numel = shape.numel() + total_numel += unpartitioned_numel + + param_frags = tuple(model_state.frozen_param_fragments[name] for model_state in zero_model_states) + state_dict[name] = torch.cat(param_frags, 0).narrow(0, 0, unpartitioned_numel).view(shape) + + partitioned_numel, partitioned_padding_numel = zero3_partitioned_param_info(unpartitioned_numel, world_size) + + if debug: + print( + f"Frozen params: {total_params} {name} full shape: {shape} partition0 numel={partitioned_numel} partitioned_padding_numel={partitioned_padding_numel}" + ) + + print(f"Reconstructed Frozen fp32 state dict with {total_params} params {total_numel} elements") + + +def _zero3_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states): + param_shapes = zero_model_states[0].param_shapes + avail_numel = fp32_flat_groups[0].numel() * world_size + # Reconstruction protocol: For zero3 we need to zip the partitions together at boundary of each + # param, re-consolidating each param, while dealing with padding if any + + # merge list of dicts, preserving order + param_shapes = {k: v for d in param_shapes for k, v in d.items()} + + if debug: + for i in range(world_size): + print(f"{FP32_FLAT_GROUPS}[{i}].shape={fp32_flat_groups[i].shape}") + + wanted_params = len(param_shapes) + wanted_numel = sum(shape.numel() for shape in param_shapes.values()) + # not asserting if there is a mismatch due to possible padding + avail_numel = fp32_flat_groups[0].numel() * world_size + print(f"Trainable params: Have {avail_numel} numels to process.") + print(f"Trainable params: Need {wanted_numel} numels in {wanted_params} params.") + + # params + # XXX: for huge models that can't fit into the host's RAM we will have to recode this to support + # out-of-core computing solution + offset = 0 + total_numel = 0 + total_params = 0 + for name, shape in param_shapes.items(): + + unpartitioned_numel = shape.numel() + total_numel += unpartitioned_numel + total_params += 1 + + partitioned_numel, partitioned_padding_numel = zero3_partitioned_param_info(unpartitioned_numel, world_size) + + if debug: + print( + f"Trainable params: {total_params} {name} full shape: {shape} partition0 numel={partitioned_numel} partitioned_padding_numel={partitioned_padding_numel}" + ) + + # XXX: memory usage doubles here + state_dict[name] = torch.cat( + tuple(fp32_flat_groups[i].narrow(0, offset, partitioned_numel) for i in range(world_size)), + 0).narrow(0, 0, unpartitioned_numel).view(shape) + offset += partitioned_numel + + offset *= world_size + + # Sanity check + if offset != avail_numel: + raise ValueError(f"consumed {offset} numels out of {avail_numel} - something is wrong") + + print(f"Reconstructed Trainable fp32 state dict with {total_params} params {total_numel} elements") + + +def _get_fp32_state_dict_from_zero3_checkpoint(world_size, fp32_flat_groups, zero_model_states, + exclude_frozen_parameters): + state_dict = OrderedDict() + + # buffers + buffers = zero_model_states[0].buffers + state_dict.update(buffers) + if debug: + print(f"added {len(buffers)} buffers") + + if not exclude_frozen_parameters: + _zero3_merge_frozen_params(state_dict, world_size, zero_model_states) + + _zero3_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states) + + # recover shared parameters + for pair in zero_model_states[0].shared_params: + if pair[1] in state_dict: + state_dict[pair[0]] = state_dict[pair[1]] + + return state_dict + + +def get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir, tag=None, exclude_frozen_parameters=False): + """ + Convert ZeRO 2 or 3 checkpoint into a single fp32 consolidated state_dict that can be loaded with + ``load_state_dict()`` and used for training without DeepSpeed or shared with others, for example + via a model hub. + + Args: + - ``checkpoint_dir``: path to the desired checkpoint folder + - ``tag``: checkpoint tag used as a unique identifier for checkpoint. If not provided will attempt to load tag in 'latest' file. e.g., ``global_step14`` + - ``exclude_frozen_parameters``: exclude frozen parameters + + Returns: + - pytorch ``state_dict`` + + Note: this approach may not work if your application doesn't have sufficient free CPU memory and + you may need to use the offline approach using the ``zero_to_fp32.py`` script that is saved with + the checkpoint. + + A typical usage might be :: + + from deepspeed.utils.zero_to_fp32 import get_fp32_state_dict_from_zero_checkpoint + # do the training and checkpoint saving + state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir) # already on cpu + model = model.cpu() # move to cpu + model.load_state_dict(state_dict) + # submit to model hub or save the model to share with others + + In this example the ``model`` will no longer be usable in the deepspeed context of the same + application. i.e. you will need to re-initialize the deepspeed engine, since + ``model.load_state_dict(state_dict)`` will remove all the deepspeed magic from it. + + If you want it all done for you, use ``load_state_dict_from_zero_checkpoint`` instead. + + """ + if tag is None: + latest_path = os.path.join(checkpoint_dir, 'latest') + if os.path.isfile(latest_path): + with open(latest_path, 'r') as fd: + tag = fd.read().strip() + else: + raise ValueError(f"Unable to find 'latest' file at {latest_path}") + + ds_checkpoint_dir = os.path.join(checkpoint_dir, tag) + + if not os.path.isdir(ds_checkpoint_dir): + raise FileNotFoundError(f"Directory '{ds_checkpoint_dir}' doesn't exist") + + return _get_fp32_state_dict_from_zero_checkpoint(ds_checkpoint_dir, exclude_frozen_parameters) + + +def convert_zero_checkpoint_to_fp32_state_dict(checkpoint_dir, output_file, tag=None, exclude_frozen_parameters=False): + """ + Convert ZeRO 2 or 3 checkpoint into a single fp32 consolidated ``state_dict`` file that can be + loaded with ``torch.load(file)`` + ``load_state_dict()`` and used for training without DeepSpeed. + + Args: + - ``checkpoint_dir``: path to the desired checkpoint folder. (one that contains the tag-folder, like ``global_step14``) + - ``output_file``: path to the pytorch fp32 state_dict output file (e.g. path/pytorch_model.bin) + - ``tag``: checkpoint tag used as a unique identifier for checkpoint. If not provided will attempt to load tag in the file named ``latest`` in the checkpoint folder, e.g., ``global_step14`` + - ``exclude_frozen_parameters``: exclude frozen parameters + """ + + state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir, tag, exclude_frozen_parameters) + print(f"Saving fp32 state dict to {output_file}") + torch.save(state_dict, output_file) + + +def load_state_dict_from_zero_checkpoint(model, checkpoint_dir, tag=None): + """ + 1. Put the provided model to cpu + 2. Convert ZeRO 2 or 3 checkpoint into a single fp32 consolidated ``state_dict`` + 3. Load it into the provided model + + Args: + - ``model``: the model object to update + - ``checkpoint_dir``: path to the desired checkpoint folder. (one that contains the tag-folder, like ``global_step14``) + - ``tag``: checkpoint tag used as a unique identifier for checkpoint. If not provided will attempt to load tag in the file named ``latest`` in the checkpoint folder, e.g., ``global_step14`` + + Returns: + - ``model`: modified model + + Make sure you have plenty of CPU memory available before you call this function. If you don't + have enough use the ``zero_to_fp32.py`` utility to do the conversion. You will find it + conveniently placed for you in the checkpoint folder. + + A typical usage might be :: + + from deepspeed.utils.zero_to_fp32 import load_state_dict_from_zero_checkpoint + model = load_state_dict_from_zero_checkpoint(trainer.model, checkpoint_dir) + # submit to model hub or save the model to share with others + + Note, that once this was run, the ``model`` will no longer be usable in the deepspeed context + of the same application. i.e. you will need to re-initialize the deepspeed engine, since + ``model.load_state_dict(state_dict)`` will remove all the deepspeed magic from it. + + """ + logger.info(f"Extracting fp32 weights") + state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir, tag) + + logger.info(f"Overwriting model with fp32 weights") + model = model.cpu() + model.load_state_dict(state_dict, strict=False) + + return model + + +if __name__ == "__main__": + + parser = argparse.ArgumentParser() + parser.add_argument("checkpoint_dir", + type=str, + help="path to the desired checkpoint folder, e.g., path/checkpoint-12") + parser.add_argument( + "output_file", + type=str, + help="path to the pytorch fp32 state_dict output file (e.g. path/checkpoint-12/pytorch_model.bin)") + parser.add_argument("-t", + "--tag", + type=str, + default=None, + help="checkpoint tag used as a unique identifier for checkpoint. e.g., global_step1") + parser.add_argument("--exclude_frozen_parameters", action='store_true', help="exclude frozen parameters") + parser.add_argument("-d", "--debug", action='store_true', help="enable debug") + args = parser.parse_args() + + debug = args.debug + + convert_zero_checkpoint_to_fp32_state_dict(args.checkpoint_dir, + args.output_file, + tag=args.tag, + exclude_frozen_parameters=args.exclude_frozen_parameters) diff --git a/babylm_shuffle_nondeterministic_10M_seed0/runs/checkpoint-2200/config.json b/babylm_shuffle_nondeterministic_10M_seed0/runs/checkpoint-2200/config.json new file mode 100644 index 0000000000000000000000000000000000000000..fb0f9a53a507ffcdb60410deb71d60ef801bf350 --- /dev/null +++ b/babylm_shuffle_nondeterministic_10M_seed0/runs/checkpoint-2200/config.json @@ -0,0 +1,36 @@ +{ + "_name_or_path": "meta-llama/Llama-3.2-3B", + "architectures": [ + "LlamaForCausalLM" + ], + "attention_bias": false, + "attention_dropout": 0.0, + "bos_token_id": 128000, + "eos_token_id": 128001, + "head_dim": 128, + "hidden_act": "silu", + "hidden_size": 3072, + "initializer_range": 0.02, + "intermediate_size": 8192, + "max_position_embeddings": 131072, + "mlp_bias": false, + "model_type": "llama", + "num_attention_heads": 24, + "num_hidden_layers": 28, + "num_key_value_heads": 8, + "pretraining_tp": 1, + "rms_norm_eps": 1e-05, + "rope_scaling": { + "factor": 32.0, + "high_freq_factor": 4.0, + "low_freq_factor": 1.0, + "original_max_position_embeddings": 8192, + "rope_type": "llama3" + }, + "rope_theta": 500000.0, + "tie_word_embeddings": true, + "torch_dtype": "float16", + "transformers_version": "4.45.1", + "use_cache": true, + "vocab_size": 128256 +} diff --git a/babylm_shuffle_nondeterministic_10M_seed0/runs/checkpoint-2200/generation_config.json b/babylm_shuffle_nondeterministic_10M_seed0/runs/checkpoint-2200/generation_config.json new file mode 100644 index 0000000000000000000000000000000000000000..9c389b841a9a29ddd904e02d1eb0e08dcce82ad9 --- /dev/null +++ b/babylm_shuffle_nondeterministic_10M_seed0/runs/checkpoint-2200/generation_config.json @@ -0,0 +1,9 @@ +{ + "_from_model_config": true, + "bos_token_id": 128000, + "do_sample": true, + "eos_token_id": 128001, + "temperature": 0.6, + "top_p": 0.9, + "transformers_version": "4.45.1" +} diff --git a/babylm_shuffle_nondeterministic_10M_seed0/runs/checkpoint-2200/latest b/babylm_shuffle_nondeterministic_10M_seed0/runs/checkpoint-2200/latest new file mode 100644 index 0000000000000000000000000000000000000000..adbdf3f70e1e0db88a47bff3edcb0592ed0a2e61 --- /dev/null +++ b/babylm_shuffle_nondeterministic_10M_seed0/runs/checkpoint-2200/latest @@ -0,0 +1 @@ +global_step2200 \ No newline at end of file diff --git a/babylm_shuffle_nondeterministic_10M_seed0/runs/checkpoint-2200/model.safetensors.index.json b/babylm_shuffle_nondeterministic_10M_seed0/runs/checkpoint-2200/model.safetensors.index.json new file mode 100644 index 0000000000000000000000000000000000000000..ed64de846d720b9a7859dc20575fea8e8ca51940 --- /dev/null +++ b/babylm_shuffle_nondeterministic_10M_seed0/runs/checkpoint-2200/model.safetensors.index.json @@ -0,0 +1,262 @@ +{ + "metadata": { + "total_size": 7213504512 + }, + "weight_map": { + "lm_head.weight": "model-00002-of-00002.safetensors", + "model.embed_tokens.weight": "model-00001-of-00002.safetensors", + "model.layers.0.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.0.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.0.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.0.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.0.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.0.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.0.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.0.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.0.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.1.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.1.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.1.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.1.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.1.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.1.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.1.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.1.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.1.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.10.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.10.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.10.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.10.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.10.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.10.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.10.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.10.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.10.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.11.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.11.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.11.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.11.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.11.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.11.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.11.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.11.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.11.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.12.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.12.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.12.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.12.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.12.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.12.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.12.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.12.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.12.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.13.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.13.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.13.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.13.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.13.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.13.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.13.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.13.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.13.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.14.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.14.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.14.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.14.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.14.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.14.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.14.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.14.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.14.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.15.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.15.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.15.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.15.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.15.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.15.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.15.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.15.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.15.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.16.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.16.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.16.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.16.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.16.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.16.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.16.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.16.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.16.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.17.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.17.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.17.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.17.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.17.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.17.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.17.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.17.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.17.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.18.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.18.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.18.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.18.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.18.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.18.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.18.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.18.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.18.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.19.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.19.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.19.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.19.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.19.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.19.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.19.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.19.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.19.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.2.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.2.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.2.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.2.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.2.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.2.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.2.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.2.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.2.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.20.input_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.20.mlp.down_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.20.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.20.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.20.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.20.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.20.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.20.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.20.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.21.input_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.21.mlp.down_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.21.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.21.mlp.up_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.21.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.21.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.21.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.21.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.21.self_attn.v_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.22.input_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.22.mlp.down_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.22.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.22.mlp.up_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.22.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.22.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.22.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.22.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.22.self_attn.v_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.23.input_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.23.mlp.down_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.23.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.23.mlp.up_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.23.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.23.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.23.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.23.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.23.self_attn.v_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.24.input_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.24.mlp.down_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.24.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.24.mlp.up_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.24.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.24.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.24.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.24.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.24.self_attn.v_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.25.input_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.25.mlp.down_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.25.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.25.mlp.up_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.25.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.25.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.25.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.25.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.25.self_attn.v_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.26.input_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.26.mlp.down_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.26.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.26.mlp.up_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.26.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.26.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.26.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.26.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.26.self_attn.v_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.27.input_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.27.mlp.down_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.27.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.27.mlp.up_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.27.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.27.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.27.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.27.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.27.self_attn.v_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.3.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.3.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.3.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.3.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.3.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.3.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.3.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.3.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.3.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.4.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.4.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.4.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.4.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.4.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.4.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.4.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.4.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.4.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.5.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.5.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.5.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.5.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.5.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.5.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.5.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.5.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.5.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.6.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.6.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.6.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.6.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.6.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.6.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.6.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.6.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.6.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.7.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.7.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.7.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.7.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.7.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.7.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.7.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.7.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.7.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.8.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.8.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.8.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.8.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.8.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.8.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.8.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.8.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.8.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.9.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.9.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.9.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.9.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.9.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.9.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.9.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.9.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.9.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.norm.weight": "model-00002-of-00002.safetensors" + } +} diff --git a/babylm_shuffle_nondeterministic_10M_seed0/runs/checkpoint-2200/special_tokens_map.json b/babylm_shuffle_nondeterministic_10M_seed0/runs/checkpoint-2200/special_tokens_map.json new file mode 100644 index 0000000000000000000000000000000000000000..6949c5975ee0e961ef61cf31010dce04df0a03f8 --- /dev/null +++ b/babylm_shuffle_nondeterministic_10M_seed0/runs/checkpoint-2200/special_tokens_map.json @@ -0,0 +1,23 @@ +{ + "bos_token": { + "content": "<|begin_of_text|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false + }, + "eos_token": { + "content": "<|end_of_text|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false + }, + "pad_token": { + "content": "[PAD]", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false + } +} diff --git a/babylm_shuffle_nondeterministic_10M_seed0/runs/checkpoint-2200/tokenizer_config.json b/babylm_shuffle_nondeterministic_10M_seed0/runs/checkpoint-2200/tokenizer_config.json new file mode 100644 index 0000000000000000000000000000000000000000..5684b14e2756b56aa397c1324e239b442162b34a --- /dev/null +++ b/babylm_shuffle_nondeterministic_10M_seed0/runs/checkpoint-2200/tokenizer_config.json @@ -0,0 +1,2070 @@ +{ + "added_tokens_decoder": { + "128000": { + "content": "<|begin_of_text|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128001": { + "content": "<|end_of_text|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128002": { + "content": "<|reserved_special_token_0|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128003": { + "content": "<|reserved_special_token_1|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128004": { + "content": "<|finetune_right_pad_id|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128005": { + "content": "<|reserved_special_token_2|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128006": { + "content": "<|start_header_id|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128007": { + "content": "<|end_header_id|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128008": { + "content": "<|eom_id|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128009": { + "content": "<|eot_id|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128010": { + "content": "<|python_tag|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128011": { + "content": "<|reserved_special_token_3|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128012": { + "content": "<|reserved_special_token_4|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128013": { + "content": "<|reserved_special_token_5|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128014": { + "content": "<|reserved_special_token_6|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128015": { + "content": "<|reserved_special_token_7|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128016": { + "content": "<|reserved_special_token_8|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128017": { + "content": "<|reserved_special_token_9|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128018": { + "content": "<|reserved_special_token_10|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128019": { + "content": "<|reserved_special_token_11|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128020": { + "content": "<|reserved_special_token_12|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128021": { + "content": "<|reserved_special_token_13|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128022": { + "content": "<|reserved_special_token_14|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128023": { + "content": "<|reserved_special_token_15|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128024": { + "content": "<|reserved_special_token_16|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128025": { + "content": "<|reserved_special_token_17|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128026": { + "content": "<|reserved_special_token_18|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128027": { + "content": "<|reserved_special_token_19|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128028": { + "content": "<|reserved_special_token_20|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128029": { + "content": "<|reserved_special_token_21|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128030": { + "content": "<|reserved_special_token_22|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128031": { + "content": "<|reserved_special_token_23|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128032": { + "content": "<|reserved_special_token_24|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128033": { + "content": "<|reserved_special_token_25|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128034": { + "content": "<|reserved_special_token_26|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128035": { + "content": "<|reserved_special_token_27|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128036": { + "content": "<|reserved_special_token_28|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128037": { + "content": "<|reserved_special_token_29|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128038": { + "content": "<|reserved_special_token_30|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128039": { + "content": "<|reserved_special_token_31|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128040": { + "content": "<|reserved_special_token_32|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128041": { + "content": "<|reserved_special_token_33|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128042": { + "content": "<|reserved_special_token_34|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128043": { + "content": "<|reserved_special_token_35|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128044": { + "content": "<|reserved_special_token_36|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128045": { + "content": "<|reserved_special_token_37|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128046": { + "content": "<|reserved_special_token_38|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128047": { + "content": "<|reserved_special_token_39|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128048": { + "content": "<|reserved_special_token_40|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128049": { + "content": "<|reserved_special_token_41|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128050": { + "content": "<|reserved_special_token_42|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128051": { + "content": "<|reserved_special_token_43|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128052": { + "content": "<|reserved_special_token_44|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128053": { + "content": "<|reserved_special_token_45|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128054": { + "content": "<|reserved_special_token_46|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128055": { + "content": "<|reserved_special_token_47|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128056": { + "content": "<|reserved_special_token_48|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128057": { + "content": "<|reserved_special_token_49|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128058": { + "content": "<|reserved_special_token_50|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128059": { + "content": "<|reserved_special_token_51|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128060": { + "content": "<|reserved_special_token_52|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128061": { + "content": "<|reserved_special_token_53|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128062": { + "content": "<|reserved_special_token_54|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128063": { + "content": "<|reserved_special_token_55|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128064": { + "content": "<|reserved_special_token_56|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128065": { + "content": "<|reserved_special_token_57|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128066": { + "content": "<|reserved_special_token_58|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128067": { + "content": "<|reserved_special_token_59|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128068": { + "content": "<|reserved_special_token_60|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128069": { + "content": "<|reserved_special_token_61|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128070": { + "content": "<|reserved_special_token_62|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128071": { + "content": "<|reserved_special_token_63|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128072": { + "content": "<|reserved_special_token_64|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128073": { + "content": "<|reserved_special_token_65|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128074": { + "content": "<|reserved_special_token_66|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128075": { + "content": "<|reserved_special_token_67|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128076": { + "content": "<|reserved_special_token_68|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128077": { + "content": "<|reserved_special_token_69|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128078": { + "content": "<|reserved_special_token_70|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128079": { + "content": "<|reserved_special_token_71|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128080": { + "content": "<|reserved_special_token_72|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128081": { + "content": "<|reserved_special_token_73|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128082": { + "content": "<|reserved_special_token_74|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128083": { + "content": "<|reserved_special_token_75|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128084": { + "content": "<|reserved_special_token_76|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128085": { + "content": "<|reserved_special_token_77|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128086": { + "content": "<|reserved_special_token_78|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128087": { + "content": "<|reserved_special_token_79|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128088": { + "content": "<|reserved_special_token_80|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128089": { + "content": "<|reserved_special_token_81|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128090": { + "content": "<|reserved_special_token_82|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128091": { + "content": "<|reserved_special_token_83|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128092": { + "content": "<|reserved_special_token_84|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128093": { + "content": "<|reserved_special_token_85|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128094": { + "content": "<|reserved_special_token_86|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128095": { + "content": "<|reserved_special_token_87|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128096": { + "content": "<|reserved_special_token_88|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128097": { + "content": "<|reserved_special_token_89|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128098": { + "content": "<|reserved_special_token_90|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128099": { + "content": "<|reserved_special_token_91|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128100": { + "content": "<|reserved_special_token_92|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128101": { + "content": "<|reserved_special_token_93|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128102": { + "content": "<|reserved_special_token_94|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128103": { + "content": "<|reserved_special_token_95|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128104": { + "content": "<|reserved_special_token_96|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128105": { + "content": "<|reserved_special_token_97|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128106": { + "content": "<|reserved_special_token_98|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128107": { + "content": "<|reserved_special_token_99|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128108": { + "content": "<|reserved_special_token_100|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128109": { + "content": "<|reserved_special_token_101|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128110": { + "content": "<|reserved_special_token_102|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128111": { + "content": "<|reserved_special_token_103|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128112": { + "content": "<|reserved_special_token_104|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128113": { + "content": "<|reserved_special_token_105|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128114": { + "content": "<|reserved_special_token_106|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128115": { + "content": "<|reserved_special_token_107|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128116": { + "content": "<|reserved_special_token_108|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128117": { + "content": "<|reserved_special_token_109|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128118": { + "content": "<|reserved_special_token_110|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128119": { + "content": "<|reserved_special_token_111|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128120": { + "content": "<|reserved_special_token_112|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128121": { + "content": "<|reserved_special_token_113|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128122": { + "content": "<|reserved_special_token_114|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128123": { + "content": "<|reserved_special_token_115|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128124": { + "content": "<|reserved_special_token_116|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128125": { + "content": "<|reserved_special_token_117|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128126": { + "content": "<|reserved_special_token_118|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128127": { + "content": "<|reserved_special_token_119|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128128": { + "content": "<|reserved_special_token_120|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128129": { + "content": "<|reserved_special_token_121|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128130": { + "content": "<|reserved_special_token_122|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128131": { + "content": "<|reserved_special_token_123|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128132": { + "content": "<|reserved_special_token_124|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128133": { + "content": "<|reserved_special_token_125|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128134": { + "content": "<|reserved_special_token_126|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128135": { + "content": "<|reserved_special_token_127|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128136": { + "content": "<|reserved_special_token_128|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128137": { + "content": "<|reserved_special_token_129|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128138": { + "content": "<|reserved_special_token_130|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128139": { + "content": "<|reserved_special_token_131|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128140": { + "content": "<|reserved_special_token_132|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128141": { + "content": "<|reserved_special_token_133|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128142": { + "content": "<|reserved_special_token_134|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128143": { + "content": "<|reserved_special_token_135|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128144": { + "content": "<|reserved_special_token_136|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128145": { + "content": "<|reserved_special_token_137|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128146": { + "content": "<|reserved_special_token_138|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128147": { + "content": "<|reserved_special_token_139|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128148": { + "content": "<|reserved_special_token_140|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128149": { + "content": "<|reserved_special_token_141|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128150": { + "content": "<|reserved_special_token_142|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128151": { + "content": "<|reserved_special_token_143|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128152": { + "content": "<|reserved_special_token_144|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128153": { + "content": "<|reserved_special_token_145|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128154": { + "content": "<|reserved_special_token_146|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128155": { + "content": "<|reserved_special_token_147|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128156": { + "content": "<|reserved_special_token_148|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128157": { + "content": "<|reserved_special_token_149|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128158": { + "content": "<|reserved_special_token_150|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128159": { + "content": "<|reserved_special_token_151|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128160": { + "content": "<|reserved_special_token_152|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128161": { + "content": "<|reserved_special_token_153|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128162": { + "content": "<|reserved_special_token_154|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128163": { + "content": "<|reserved_special_token_155|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128164": { + "content": "<|reserved_special_token_156|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128165": { + "content": "<|reserved_special_token_157|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128166": { + "content": "<|reserved_special_token_158|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128167": { + "content": "<|reserved_special_token_159|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128168": { + "content": "<|reserved_special_token_160|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128169": { + "content": "<|reserved_special_token_161|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128170": { + "content": "<|reserved_special_token_162|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128171": { + "content": "<|reserved_special_token_163|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128172": { + "content": "<|reserved_special_token_164|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128173": { + "content": "<|reserved_special_token_165|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128174": { + "content": "<|reserved_special_token_166|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128175": { + "content": "<|reserved_special_token_167|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128176": { + "content": "<|reserved_special_token_168|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128177": { + "content": "<|reserved_special_token_169|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128178": { + "content": "<|reserved_special_token_170|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128179": { + "content": "<|reserved_special_token_171|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128180": { + "content": "<|reserved_special_token_172|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128181": { + "content": "<|reserved_special_token_173|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128182": { + "content": "<|reserved_special_token_174|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128183": { + "content": "<|reserved_special_token_175|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128184": { + "content": "<|reserved_special_token_176|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128185": { + "content": "<|reserved_special_token_177|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128186": { + "content": "<|reserved_special_token_178|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128187": { + "content": "<|reserved_special_token_179|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128188": { + "content": "<|reserved_special_token_180|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128189": { + "content": "<|reserved_special_token_181|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128190": { + "content": "<|reserved_special_token_182|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128191": { + "content": "<|reserved_special_token_183|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128192": { + "content": "<|reserved_special_token_184|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128193": { + "content": "<|reserved_special_token_185|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128194": { + "content": "<|reserved_special_token_186|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128195": { + "content": "<|reserved_special_token_187|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128196": { + "content": "<|reserved_special_token_188|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128197": { + "content": "<|reserved_special_token_189|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128198": { + "content": "<|reserved_special_token_190|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128199": { + "content": "<|reserved_special_token_191|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128200": { + "content": "<|reserved_special_token_192|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128201": { + "content": "<|reserved_special_token_193|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128202": { + "content": "<|reserved_special_token_194|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128203": { + "content": "<|reserved_special_token_195|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128204": { + "content": "<|reserved_special_token_196|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128205": { + "content": "<|reserved_special_token_197|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128206": { + "content": "<|reserved_special_token_198|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128207": { + "content": "<|reserved_special_token_199|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128208": { + "content": "<|reserved_special_token_200|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128209": { + "content": "<|reserved_special_token_201|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128210": { + "content": "<|reserved_special_token_202|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128211": { + "content": "<|reserved_special_token_203|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128212": { + "content": "<|reserved_special_token_204|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128213": { + "content": "<|reserved_special_token_205|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128214": { + "content": "<|reserved_special_token_206|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128215": { + "content": "<|reserved_special_token_207|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128216": { + "content": "<|reserved_special_token_208|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128217": { + "content": "<|reserved_special_token_209|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128218": { + "content": "<|reserved_special_token_210|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128219": { + "content": "<|reserved_special_token_211|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128220": { + "content": "<|reserved_special_token_212|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128221": { + "content": "<|reserved_special_token_213|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128222": { + "content": "<|reserved_special_token_214|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128223": { + "content": "<|reserved_special_token_215|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128224": { + "content": "<|reserved_special_token_216|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128225": { + "content": "<|reserved_special_token_217|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128226": { + "content": "<|reserved_special_token_218|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128227": { + "content": "<|reserved_special_token_219|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128228": { + "content": "<|reserved_special_token_220|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128229": { + "content": "<|reserved_special_token_221|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128230": { + "content": "<|reserved_special_token_222|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128231": { + "content": "<|reserved_special_token_223|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128232": { + "content": "<|reserved_special_token_224|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128233": { + "content": "<|reserved_special_token_225|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128234": { + "content": "<|reserved_special_token_226|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128235": { + "content": "<|reserved_special_token_227|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128236": { + "content": "<|reserved_special_token_228|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128237": { + "content": "<|reserved_special_token_229|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128238": { + "content": "<|reserved_special_token_230|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128239": { + "content": "<|reserved_special_token_231|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128240": { + "content": "<|reserved_special_token_232|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128241": { + "content": "<|reserved_special_token_233|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128242": { + "content": "<|reserved_special_token_234|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128243": { + "content": "<|reserved_special_token_235|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128244": { + "content": "<|reserved_special_token_236|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128245": { + "content": "<|reserved_special_token_237|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128246": { + "content": "<|reserved_special_token_238|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128247": { + "content": "<|reserved_special_token_239|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128248": { + "content": "<|reserved_special_token_240|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128249": { + "content": "<|reserved_special_token_241|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128250": { + "content": "<|reserved_special_token_242|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128251": { + "content": "<|reserved_special_token_243|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128252": { + "content": "<|reserved_special_token_244|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128253": { + "content": "<|reserved_special_token_245|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128254": { + "content": "<|reserved_special_token_246|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128255": { + "content": "<|reserved_special_token_247|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128256": { + "content": "[PAD]", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + } + }, + "bos_token": "<|begin_of_text|>", + "clean_up_tokenization_spaces": true, + "eos_token": "<|end_of_text|>", + "model_input_names": [ + "input_ids", + "attention_mask" + ], + "model_max_length": 131072, + "pad_token": "[PAD]", + "tokenizer_class": "PreTrainedTokenizerFast" +} diff --git a/babylm_shuffle_nondeterministic_10M_seed0/runs/checkpoint-2200/trainer_state.json b/babylm_shuffle_nondeterministic_10M_seed0/runs/checkpoint-2200/trainer_state.json new file mode 100644 index 0000000000000000000000000000000000000000..bbd7bedb484610234f9a3ad7784ffd006b9e31bc --- /dev/null +++ b/babylm_shuffle_nondeterministic_10M_seed0/runs/checkpoint-2200/trainer_state.json @@ -0,0 +1,17193 @@ +{ + "best_metric": null, + "best_model_checkpoint": null, + "epoch": 2.410958904109589, + "eval_steps": 10, + "global_step": 2200, + "is_hyper_param_search": false, + "is_local_process_zero": true, + "is_world_process_zero": true, + "log_history": [ + { + "epoch": 0.001095890410958904, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.0998, + "step": 1 + }, + { + "epoch": 0.002191780821917808, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.0858, + "step": 2 + }, + { + "epoch": 0.0032876712328767125, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.0672, + "step": 3 + }, + { + "epoch": 0.004383561643835616, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.0802, + "step": 4 + }, + { + "epoch": 0.005479452054794521, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.0636, + "step": 5 + }, + { + "epoch": 0.006575342465753425, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.1672, + "step": 6 + }, + { + "epoch": 0.007671232876712329, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.0961, + "step": 7 + }, + { + "epoch": 0.008767123287671232, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.1258, + "step": 8 + }, + { + "epoch": 0.009863013698630137, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.0953, + "step": 9 + }, + { + "epoch": 0.010958904109589041, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.1404, + "step": 10 + }, + { + "epoch": 0.010958904109589041, + "eval_loss": 3.12593412399292, + "eval_runtime": 42.732, + "eval_samples_per_second": 23.402, + "eval_steps_per_second": 0.983, + "step": 10 + }, + { + "epoch": 0.012054794520547946, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.1187, + "step": 11 + }, + { + "epoch": 0.01315068493150685, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.0593, + "step": 12 + }, + { + "epoch": 0.014246575342465753, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.111, + "step": 13 + }, + { + "epoch": 0.015342465753424657, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.0974, + "step": 14 + }, + { + "epoch": 0.01643835616438356, + "grad_norm": 5.991453170776367, + "learning_rate": 1.8248175182481753e-08, + "loss": 3.135, + "step": 15 + }, + { + "epoch": 0.017534246575342465, + "grad_norm": 6.274210453033447, + "learning_rate": 3.6496350364963505e-08, + "loss": 3.083, + "step": 16 + }, + { + "epoch": 0.01863013698630137, + "grad_norm": 6.644786834716797, + "learning_rate": 5.4744525547445264e-08, + "loss": 3.0856, + "step": 17 + }, + { + "epoch": 0.019726027397260273, + "grad_norm": 6.769156455993652, + "learning_rate": 7.299270072992701e-08, + "loss": 3.1082, + "step": 18 + }, + { + "epoch": 0.020821917808219178, + "grad_norm": 6.771246910095215, + "learning_rate": 9.124087591240876e-08, + "loss": 3.0753, + "step": 19 + }, + { + "epoch": 0.021917808219178082, + "grad_norm": 6.468441486358643, + "learning_rate": 1.0948905109489053e-07, + "loss": 3.1478, + "step": 20 + }, + { + "epoch": 0.021917808219178082, + "eval_loss": 3.1245107650756836, + "eval_runtime": 43.6251, + "eval_samples_per_second": 22.923, + "eval_steps_per_second": 0.963, + "step": 20 + }, + { + "epoch": 0.023013698630136987, + "grad_norm": 6.83078670501709, + "learning_rate": 1.2773722627737227e-07, + "loss": 3.1213, + "step": 21 + }, + { + "epoch": 0.02410958904109589, + "grad_norm": 6.221287250518799, + "learning_rate": 1.4598540145985402e-07, + "loss": 3.0549, + "step": 22 + }, + { + "epoch": 0.025205479452054796, + "grad_norm": 5.564266204833984, + "learning_rate": 1.642335766423358e-07, + "loss": 3.0769, + "step": 23 + }, + { + "epoch": 0.0263013698630137, + "grad_norm": 5.735209941864014, + "learning_rate": 1.8248175182481753e-07, + "loss": 3.1124, + "step": 24 + }, + { + "epoch": 0.0273972602739726, + "grad_norm": 5.577925682067871, + "learning_rate": 2.0072992700729928e-07, + "loss": 3.1191, + "step": 25 + }, + { + "epoch": 0.028493150684931506, + "grad_norm": 4.814139366149902, + "learning_rate": 2.1897810218978106e-07, + "loss": 3.0947, + "step": 26 + }, + { + "epoch": 0.02958904109589041, + "grad_norm": 4.69775390625, + "learning_rate": 2.3722627737226278e-07, + "loss": 3.0997, + "step": 27 + }, + { + "epoch": 0.030684931506849315, + "grad_norm": 4.811586380004883, + "learning_rate": 2.5547445255474454e-07, + "loss": 3.0749, + "step": 28 + }, + { + "epoch": 0.03178082191780822, + "grad_norm": 4.623377323150635, + "learning_rate": 2.737226277372263e-07, + "loss": 3.0619, + "step": 29 + }, + { + "epoch": 0.03287671232876712, + "grad_norm": 4.721253395080566, + "learning_rate": 2.9197080291970804e-07, + "loss": 3.0371, + "step": 30 + }, + { + "epoch": 0.03287671232876712, + "eval_loss": 3.082857847213745, + "eval_runtime": 44.127, + "eval_samples_per_second": 22.662, + "eval_steps_per_second": 0.952, + "step": 30 + }, + { + "epoch": 0.03397260273972603, + "grad_norm": 3.968630790710449, + "learning_rate": 3.102189781021898e-07, + "loss": 3.0535, + "step": 31 + }, + { + "epoch": 0.03506849315068493, + "grad_norm": 4.4373779296875, + "learning_rate": 3.284671532846716e-07, + "loss": 3.0476, + "step": 32 + }, + { + "epoch": 0.03616438356164384, + "grad_norm": 4.479320049285889, + "learning_rate": 3.467153284671533e-07, + "loss": 3.0215, + "step": 33 + }, + { + "epoch": 0.03726027397260274, + "grad_norm": 4.122231960296631, + "learning_rate": 3.6496350364963505e-07, + "loss": 3.0106, + "step": 34 + }, + { + "epoch": 0.038356164383561646, + "grad_norm": 4.153995513916016, + "learning_rate": 3.8321167883211686e-07, + "loss": 3.0318, + "step": 35 + }, + { + "epoch": 0.03945205479452055, + "grad_norm": 3.6949076652526855, + "learning_rate": 4.0145985401459856e-07, + "loss": 3.046, + "step": 36 + }, + { + "epoch": 0.040547945205479455, + "grad_norm": 3.4610488414764404, + "learning_rate": 4.197080291970803e-07, + "loss": 3.002, + "step": 37 + }, + { + "epoch": 0.041643835616438356, + "grad_norm": 3.9523138999938965, + "learning_rate": 4.379562043795621e-07, + "loss": 2.9665, + "step": 38 + }, + { + "epoch": 0.042739726027397264, + "grad_norm": 3.68516206741333, + "learning_rate": 4.562043795620438e-07, + "loss": 3.0411, + "step": 39 + }, + { + "epoch": 0.043835616438356165, + "grad_norm": 3.692396402359009, + "learning_rate": 4.7445255474452557e-07, + "loss": 2.9666, + "step": 40 + }, + { + "epoch": 0.043835616438356165, + "eval_loss": 3.015512704849243, + "eval_runtime": 48.2438, + "eval_samples_per_second": 20.728, + "eval_steps_per_second": 0.871, + "step": 40 + }, + { + "epoch": 0.044931506849315066, + "grad_norm": 3.3765921592712402, + "learning_rate": 4.927007299270073e-07, + "loss": 2.9831, + "step": 41 + }, + { + "epoch": 0.046027397260273974, + "grad_norm": 2.9631125926971436, + "learning_rate": 5.109489051094891e-07, + "loss": 3.0126, + "step": 42 + }, + { + "epoch": 0.047123287671232875, + "grad_norm": 2.969730854034424, + "learning_rate": 5.291970802919708e-07, + "loss": 2.9803, + "step": 43 + }, + { + "epoch": 0.04821917808219178, + "grad_norm": 3.185237407684326, + "learning_rate": 5.474452554744526e-07, + "loss": 2.9888, + "step": 44 + }, + { + "epoch": 0.049315068493150684, + "grad_norm": 3.055579423904419, + "learning_rate": 5.656934306569343e-07, + "loss": 2.9715, + "step": 45 + }, + { + "epoch": 0.05041095890410959, + "grad_norm": 3.083284616470337, + "learning_rate": 5.839416058394161e-07, + "loss": 2.9293, + "step": 46 + }, + { + "epoch": 0.05150684931506849, + "grad_norm": 2.984781503677368, + "learning_rate": 6.021897810218979e-07, + "loss": 2.9228, + "step": 47 + }, + { + "epoch": 0.0526027397260274, + "grad_norm": 2.756173849105835, + "learning_rate": 6.204379562043796e-07, + "loss": 2.9406, + "step": 48 + }, + { + "epoch": 0.0536986301369863, + "grad_norm": 2.911132574081421, + "learning_rate": 6.386861313868614e-07, + "loss": 2.8829, + "step": 49 + }, + { + "epoch": 0.0547945205479452, + "grad_norm": 2.592890977859497, + "learning_rate": 6.569343065693432e-07, + "loss": 2.9839, + "step": 50 + }, + { + "epoch": 0.0547945205479452, + "eval_loss": 2.9409468173980713, + "eval_runtime": 47.1045, + "eval_samples_per_second": 21.229, + "eval_steps_per_second": 0.892, + "step": 50 + }, + { + "epoch": 0.05589041095890411, + "grad_norm": 3.0112955570220947, + "learning_rate": 6.751824817518248e-07, + "loss": 2.8847, + "step": 51 + }, + { + "epoch": 0.05698630136986301, + "grad_norm": 2.3701798915863037, + "learning_rate": 6.934306569343066e-07, + "loss": 2.9109, + "step": 52 + }, + { + "epoch": 0.05808219178082192, + "grad_norm": 2.268850088119507, + "learning_rate": 7.116788321167883e-07, + "loss": 2.8483, + "step": 53 + }, + { + "epoch": 0.05917808219178082, + "grad_norm": 2.419015645980835, + "learning_rate": 7.299270072992701e-07, + "loss": 2.8491, + "step": 54 + }, + { + "epoch": 0.06027397260273973, + "grad_norm": 2.4582653045654297, + "learning_rate": 7.48175182481752e-07, + "loss": 2.8622, + "step": 55 + }, + { + "epoch": 0.06136986301369863, + "grad_norm": 2.3471553325653076, + "learning_rate": 7.664233576642337e-07, + "loss": 2.8738, + "step": 56 + }, + { + "epoch": 0.06246575342465754, + "grad_norm": 2.479177236557007, + "learning_rate": 7.846715328467154e-07, + "loss": 2.8843, + "step": 57 + }, + { + "epoch": 0.06356164383561644, + "grad_norm": 2.1676292419433594, + "learning_rate": 8.029197080291971e-07, + "loss": 2.8674, + "step": 58 + }, + { + "epoch": 0.06465753424657535, + "grad_norm": 1.9862910509109497, + "learning_rate": 8.211678832116789e-07, + "loss": 2.8197, + "step": 59 + }, + { + "epoch": 0.06575342465753424, + "grad_norm": 2.211313486099243, + "learning_rate": 8.394160583941606e-07, + "loss": 2.7961, + "step": 60 + }, + { + "epoch": 0.06575342465753424, + "eval_loss": 2.880824327468872, + "eval_runtime": 51.3235, + "eval_samples_per_second": 19.484, + "eval_steps_per_second": 0.818, + "step": 60 + }, + { + "epoch": 0.06684931506849315, + "grad_norm": 2.3497281074523926, + "learning_rate": 8.576642335766425e-07, + "loss": 2.8233, + "step": 61 + }, + { + "epoch": 0.06794520547945206, + "grad_norm": 2.158515214920044, + "learning_rate": 8.759124087591242e-07, + "loss": 2.8136, + "step": 62 + }, + { + "epoch": 0.06904109589041096, + "grad_norm": 2.0219955444335938, + "learning_rate": 8.94160583941606e-07, + "loss": 2.8231, + "step": 63 + }, + { + "epoch": 0.07013698630136986, + "grad_norm": 1.8989133834838867, + "learning_rate": 9.124087591240876e-07, + "loss": 2.8268, + "step": 64 + }, + { + "epoch": 0.07123287671232877, + "grad_norm": 2.243851661682129, + "learning_rate": 9.306569343065694e-07, + "loss": 2.7886, + "step": 65 + }, + { + "epoch": 0.07232876712328767, + "grad_norm": 2.122464418411255, + "learning_rate": 9.489051094890511e-07, + "loss": 2.8166, + "step": 66 + }, + { + "epoch": 0.07342465753424658, + "grad_norm": 1.899857997894287, + "learning_rate": 9.67153284671533e-07, + "loss": 2.7997, + "step": 67 + }, + { + "epoch": 0.07452054794520548, + "grad_norm": 2.4424173831939697, + "learning_rate": 9.854014598540146e-07, + "loss": 2.8342, + "step": 68 + }, + { + "epoch": 0.07561643835616438, + "grad_norm": 2.5677411556243896, + "learning_rate": 1.0036496350364965e-06, + "loss": 2.7786, + "step": 69 + }, + { + "epoch": 0.07671232876712329, + "grad_norm": 1.9989523887634277, + "learning_rate": 1.0218978102189781e-06, + "loss": 2.7411, + "step": 70 + }, + { + "epoch": 0.07671232876712329, + "eval_loss": 2.8185322284698486, + "eval_runtime": 50.4998, + "eval_samples_per_second": 19.802, + "eval_steps_per_second": 0.832, + "step": 70 + }, + { + "epoch": 0.07780821917808219, + "grad_norm": 1.9376521110534668, + "learning_rate": 1.04014598540146e-06, + "loss": 2.7988, + "step": 71 + }, + { + "epoch": 0.0789041095890411, + "grad_norm": 2.5629725456237793, + "learning_rate": 1.0583941605839416e-06, + "loss": 2.7079, + "step": 72 + }, + { + "epoch": 0.08, + "grad_norm": 2.1634678840637207, + "learning_rate": 1.0766423357664233e-06, + "loss": 2.7049, + "step": 73 + }, + { + "epoch": 0.08109589041095891, + "grad_norm": 2.2143008708953857, + "learning_rate": 1.0948905109489052e-06, + "loss": 2.72, + "step": 74 + }, + { + "epoch": 0.0821917808219178, + "grad_norm": 2.2701101303100586, + "learning_rate": 1.113138686131387e-06, + "loss": 2.7418, + "step": 75 + }, + { + "epoch": 0.08328767123287671, + "grad_norm": 1.716569423675537, + "learning_rate": 1.1313868613138687e-06, + "loss": 2.7558, + "step": 76 + }, + { + "epoch": 0.08438356164383562, + "grad_norm": 2.79757022857666, + "learning_rate": 1.1496350364963505e-06, + "loss": 2.6865, + "step": 77 + }, + { + "epoch": 0.08547945205479453, + "grad_norm": 2.2189838886260986, + "learning_rate": 1.1678832116788322e-06, + "loss": 2.6684, + "step": 78 + }, + { + "epoch": 0.08657534246575342, + "grad_norm": 2.0695579051971436, + "learning_rate": 1.1861313868613138e-06, + "loss": 2.6779, + "step": 79 + }, + { + "epoch": 0.08767123287671233, + "grad_norm": 2.372326374053955, + "learning_rate": 1.2043795620437959e-06, + "loss": 2.6711, + "step": 80 + }, + { + "epoch": 0.08767123287671233, + "eval_loss": 2.7537453174591064, + "eval_runtime": 47.7274, + "eval_samples_per_second": 20.952, + "eval_steps_per_second": 0.88, + "step": 80 + }, + { + "epoch": 0.08876712328767124, + "grad_norm": 2.3398947715759277, + "learning_rate": 1.2226277372262775e-06, + "loss": 2.6776, + "step": 81 + }, + { + "epoch": 0.08986301369863013, + "grad_norm": 2.6619036197662354, + "learning_rate": 1.2408759124087592e-06, + "loss": 2.7077, + "step": 82 + }, + { + "epoch": 0.09095890410958904, + "grad_norm": 1.8245115280151367, + "learning_rate": 1.259124087591241e-06, + "loss": 2.662, + "step": 83 + }, + { + "epoch": 0.09205479452054795, + "grad_norm": 2.2627148628234863, + "learning_rate": 1.2773722627737229e-06, + "loss": 2.677, + "step": 84 + }, + { + "epoch": 0.09315068493150686, + "grad_norm": 2.4983835220336914, + "learning_rate": 1.2956204379562045e-06, + "loss": 2.6601, + "step": 85 + }, + { + "epoch": 0.09424657534246575, + "grad_norm": 2.147010087966919, + "learning_rate": 1.3138686131386864e-06, + "loss": 2.6213, + "step": 86 + }, + { + "epoch": 0.09534246575342466, + "grad_norm": 2.791682481765747, + "learning_rate": 1.332116788321168e-06, + "loss": 2.6281, + "step": 87 + }, + { + "epoch": 0.09643835616438357, + "grad_norm": 2.023970365524292, + "learning_rate": 1.3503649635036497e-06, + "loss": 2.5766, + "step": 88 + }, + { + "epoch": 0.09753424657534246, + "grad_norm": 2.8143956661224365, + "learning_rate": 1.3686131386861315e-06, + "loss": 2.6086, + "step": 89 + }, + { + "epoch": 0.09863013698630137, + "grad_norm": 2.226106882095337, + "learning_rate": 1.3868613138686132e-06, + "loss": 2.5477, + "step": 90 + }, + { + "epoch": 0.09863013698630137, + "eval_loss": 2.6880786418914795, + "eval_runtime": 47.8417, + "eval_samples_per_second": 20.902, + "eval_steps_per_second": 0.878, + "step": 90 + }, + { + "epoch": 0.09972602739726028, + "grad_norm": 2.5248148441314697, + "learning_rate": 1.405109489051095e-06, + "loss": 2.6232, + "step": 91 + }, + { + "epoch": 0.10082191780821918, + "grad_norm": 2.421330690383911, + "learning_rate": 1.4233576642335767e-06, + "loss": 2.599, + "step": 92 + }, + { + "epoch": 0.10191780821917808, + "grad_norm": 2.5562429428100586, + "learning_rate": 1.4416058394160583e-06, + "loss": 2.6197, + "step": 93 + }, + { + "epoch": 0.10301369863013699, + "grad_norm": 2.856201171875, + "learning_rate": 1.4598540145985402e-06, + "loss": 2.5825, + "step": 94 + }, + { + "epoch": 0.10410958904109589, + "grad_norm": 2.6576807498931885, + "learning_rate": 1.478102189781022e-06, + "loss": 2.5496, + "step": 95 + }, + { + "epoch": 0.1052054794520548, + "grad_norm": 3.1408190727233887, + "learning_rate": 1.496350364963504e-06, + "loss": 2.6142, + "step": 96 + }, + { + "epoch": 0.1063013698630137, + "grad_norm": 2.3270952701568604, + "learning_rate": 1.5145985401459856e-06, + "loss": 2.584, + "step": 97 + }, + { + "epoch": 0.1073972602739726, + "grad_norm": 2.3976614475250244, + "learning_rate": 1.5328467153284674e-06, + "loss": 2.5941, + "step": 98 + }, + { + "epoch": 0.10849315068493151, + "grad_norm": 2.783485174179077, + "learning_rate": 1.551094890510949e-06, + "loss": 2.5626, + "step": 99 + }, + { + "epoch": 0.1095890410958904, + "grad_norm": 2.2522788047790527, + "learning_rate": 1.5693430656934307e-06, + "loss": 2.517, + "step": 100 + }, + { + "epoch": 0.1095890410958904, + "eval_loss": 2.637315034866333, + "eval_runtime": 48.2107, + "eval_samples_per_second": 20.742, + "eval_steps_per_second": 0.871, + "step": 100 + }, + { + "epoch": 0.11068493150684931, + "grad_norm": 2.917799949645996, + "learning_rate": 1.5875912408759126e-06, + "loss": 2.5159, + "step": 101 + }, + { + "epoch": 0.11178082191780822, + "grad_norm": 2.955815076828003, + "learning_rate": 1.6058394160583942e-06, + "loss": 2.563, + "step": 102 + }, + { + "epoch": 0.11287671232876713, + "grad_norm": 2.8096001148223877, + "learning_rate": 1.624087591240876e-06, + "loss": 2.5995, + "step": 103 + }, + { + "epoch": 0.11397260273972602, + "grad_norm": 2.4967427253723145, + "learning_rate": 1.6423357664233577e-06, + "loss": 2.5336, + "step": 104 + }, + { + "epoch": 0.11506849315068493, + "grad_norm": 2.740717887878418, + "learning_rate": 1.6605839416058394e-06, + "loss": 2.4778, + "step": 105 + }, + { + "epoch": 0.11616438356164384, + "grad_norm": 2.5064632892608643, + "learning_rate": 1.6788321167883212e-06, + "loss": 2.5214, + "step": 106 + }, + { + "epoch": 0.11726027397260275, + "grad_norm": 2.2923007011413574, + "learning_rate": 1.6970802919708029e-06, + "loss": 2.5315, + "step": 107 + }, + { + "epoch": 0.11835616438356164, + "grad_norm": 3.339569568634033, + "learning_rate": 1.715328467153285e-06, + "loss": 2.4661, + "step": 108 + }, + { + "epoch": 0.11945205479452055, + "grad_norm": 2.1650891304016113, + "learning_rate": 1.7335766423357666e-06, + "loss": 2.5072, + "step": 109 + }, + { + "epoch": 0.12054794520547946, + "grad_norm": 3.1216914653778076, + "learning_rate": 1.7518248175182485e-06, + "loss": 2.5164, + "step": 110 + }, + { + "epoch": 0.12054794520547946, + "eval_loss": 2.594250202178955, + "eval_runtime": 41.8939, + "eval_samples_per_second": 23.87, + "eval_steps_per_second": 1.003, + "step": 110 + }, + { + "epoch": 0.12164383561643835, + "grad_norm": 2.191063404083252, + "learning_rate": 1.77007299270073e-06, + "loss": 2.5422, + "step": 111 + }, + { + "epoch": 0.12273972602739726, + "grad_norm": 2.922083854675293, + "learning_rate": 1.788321167883212e-06, + "loss": 2.5258, + "step": 112 + }, + { + "epoch": 0.12383561643835617, + "grad_norm": 2.518002986907959, + "learning_rate": 1.8065693430656936e-06, + "loss": 2.5588, + "step": 113 + }, + { + "epoch": 0.12493150684931507, + "grad_norm": 2.515194892883301, + "learning_rate": 1.8248175182481753e-06, + "loss": 2.5079, + "step": 114 + }, + { + "epoch": 0.12602739726027398, + "grad_norm": 2.5530874729156494, + "learning_rate": 1.8430656934306571e-06, + "loss": 2.4615, + "step": 115 + }, + { + "epoch": 0.12712328767123288, + "grad_norm": 2.3550198078155518, + "learning_rate": 1.8613138686131388e-06, + "loss": 2.4825, + "step": 116 + }, + { + "epoch": 0.12821917808219177, + "grad_norm": 2.295145034790039, + "learning_rate": 1.8795620437956206e-06, + "loss": 2.4749, + "step": 117 + }, + { + "epoch": 0.1293150684931507, + "grad_norm": 2.0158982276916504, + "learning_rate": 1.8978102189781023e-06, + "loss": 2.4756, + "step": 118 + }, + { + "epoch": 0.1304109589041096, + "grad_norm": 2.213057518005371, + "learning_rate": 1.916058394160584e-06, + "loss": 2.4744, + "step": 119 + }, + { + "epoch": 0.13150684931506848, + "grad_norm": 2.186222553253174, + "learning_rate": 1.934306569343066e-06, + "loss": 2.4957, + "step": 120 + }, + { + "epoch": 0.13150684931506848, + "eval_loss": 2.563931465148926, + "eval_runtime": 44.8714, + "eval_samples_per_second": 22.286, + "eval_steps_per_second": 0.936, + "step": 120 + }, + { + "epoch": 0.1326027397260274, + "grad_norm": 2.3143310546875, + "learning_rate": 1.952554744525548e-06, + "loss": 2.4, + "step": 121 + }, + { + "epoch": 0.1336986301369863, + "grad_norm": 2.0223965644836426, + "learning_rate": 1.9708029197080293e-06, + "loss": 2.4453, + "step": 122 + }, + { + "epoch": 0.13479452054794522, + "grad_norm": 2.2025411128997803, + "learning_rate": 1.989051094890511e-06, + "loss": 2.4463, + "step": 123 + }, + { + "epoch": 0.1358904109589041, + "grad_norm": 2.707930564880371, + "learning_rate": 2.007299270072993e-06, + "loss": 2.4589, + "step": 124 + }, + { + "epoch": 0.136986301369863, + "grad_norm": 2.146662473678589, + "learning_rate": 2.025547445255475e-06, + "loss": 2.5023, + "step": 125 + }, + { + "epoch": 0.13808219178082193, + "grad_norm": 2.520660161972046, + "learning_rate": 2.0437956204379563e-06, + "loss": 2.4464, + "step": 126 + }, + { + "epoch": 0.13917808219178082, + "grad_norm": 2.50632381439209, + "learning_rate": 2.062043795620438e-06, + "loss": 2.4556, + "step": 127 + }, + { + "epoch": 0.14027397260273972, + "grad_norm": 2.2919397354125977, + "learning_rate": 2.08029197080292e-06, + "loss": 2.4479, + "step": 128 + }, + { + "epoch": 0.14136986301369864, + "grad_norm": 2.521475315093994, + "learning_rate": 2.0985401459854014e-06, + "loss": 2.4982, + "step": 129 + }, + { + "epoch": 0.14246575342465753, + "grad_norm": 2.3812668323516846, + "learning_rate": 2.1167883211678833e-06, + "loss": 2.4163, + "step": 130 + }, + { + "epoch": 0.14246575342465753, + "eval_loss": 2.5390918254852295, + "eval_runtime": 43.8746, + "eval_samples_per_second": 22.792, + "eval_steps_per_second": 0.957, + "step": 130 + }, + { + "epoch": 0.14356164383561643, + "grad_norm": 2.302286148071289, + "learning_rate": 2.135036496350365e-06, + "loss": 2.4588, + "step": 131 + }, + { + "epoch": 0.14465753424657535, + "grad_norm": 2.4222044944763184, + "learning_rate": 2.1532846715328466e-06, + "loss": 2.492, + "step": 132 + }, + { + "epoch": 0.14575342465753424, + "grad_norm": 2.605835199356079, + "learning_rate": 2.171532846715329e-06, + "loss": 2.4055, + "step": 133 + }, + { + "epoch": 0.14684931506849316, + "grad_norm": 3.6261861324310303, + "learning_rate": 2.1897810218978103e-06, + "loss": 2.4221, + "step": 134 + }, + { + "epoch": 0.14794520547945206, + "grad_norm": 2.7647221088409424, + "learning_rate": 2.208029197080292e-06, + "loss": 2.4549, + "step": 135 + }, + { + "epoch": 0.14904109589041095, + "grad_norm": 3.2977371215820312, + "learning_rate": 2.226277372262774e-06, + "loss": 2.4331, + "step": 136 + }, + { + "epoch": 0.15013698630136987, + "grad_norm": 3.147094488143921, + "learning_rate": 2.244525547445256e-06, + "loss": 2.4268, + "step": 137 + }, + { + "epoch": 0.15123287671232877, + "grad_norm": 2.6363413333892822, + "learning_rate": 2.2627737226277373e-06, + "loss": 2.4524, + "step": 138 + }, + { + "epoch": 0.15232876712328766, + "grad_norm": 2.9288904666900635, + "learning_rate": 2.281021897810219e-06, + "loss": 2.4534, + "step": 139 + }, + { + "epoch": 0.15342465753424658, + "grad_norm": 2.5604870319366455, + "learning_rate": 2.299270072992701e-06, + "loss": 2.4138, + "step": 140 + }, + { + "epoch": 0.15342465753424658, + "eval_loss": 2.524790048599243, + "eval_runtime": 43.8851, + "eval_samples_per_second": 22.787, + "eval_steps_per_second": 0.957, + "step": 140 + }, + { + "epoch": 0.15452054794520548, + "grad_norm": 3.2746663093566895, + "learning_rate": 2.3175182481751825e-06, + "loss": 2.4094, + "step": 141 + }, + { + "epoch": 0.15561643835616437, + "grad_norm": 2.691483497619629, + "learning_rate": 2.3357664233576643e-06, + "loss": 2.4265, + "step": 142 + }, + { + "epoch": 0.1567123287671233, + "grad_norm": 2.5850117206573486, + "learning_rate": 2.354014598540146e-06, + "loss": 2.3922, + "step": 143 + }, + { + "epoch": 0.1578082191780822, + "grad_norm": 2.68204927444458, + "learning_rate": 2.3722627737226276e-06, + "loss": 2.4262, + "step": 144 + }, + { + "epoch": 0.1589041095890411, + "grad_norm": 2.952584981918335, + "learning_rate": 2.39051094890511e-06, + "loss": 2.4267, + "step": 145 + }, + { + "epoch": 0.16, + "grad_norm": 3.2354788780212402, + "learning_rate": 2.4087591240875918e-06, + "loss": 2.3836, + "step": 146 + }, + { + "epoch": 0.1610958904109589, + "grad_norm": 2.634521007537842, + "learning_rate": 2.427007299270073e-06, + "loss": 2.3953, + "step": 147 + }, + { + "epoch": 0.16219178082191782, + "grad_norm": 3.1115682125091553, + "learning_rate": 2.445255474452555e-06, + "loss": 2.409, + "step": 148 + }, + { + "epoch": 0.1632876712328767, + "grad_norm": 2.6824135780334473, + "learning_rate": 2.463503649635037e-06, + "loss": 2.4459, + "step": 149 + }, + { + "epoch": 0.1643835616438356, + "grad_norm": 2.617614507675171, + "learning_rate": 2.4817518248175183e-06, + "loss": 2.366, + "step": 150 + }, + { + "epoch": 0.1643835616438356, + "eval_loss": 2.5088160037994385, + "eval_runtime": 44.0888, + "eval_samples_per_second": 22.682, + "eval_steps_per_second": 0.953, + "step": 150 + }, + { + "epoch": 0.16547945205479453, + "grad_norm": 3.976992130279541, + "learning_rate": 2.5e-06, + "loss": 2.3517, + "step": 151 + }, + { + "epoch": 0.16657534246575342, + "grad_norm": 3.1433017253875732, + "learning_rate": 2.518248175182482e-06, + "loss": 2.4489, + "step": 152 + }, + { + "epoch": 0.16767123287671232, + "grad_norm": 4.100688934326172, + "learning_rate": 2.5364963503649635e-06, + "loss": 2.3683, + "step": 153 + }, + { + "epoch": 0.16876712328767124, + "grad_norm": 4.316102504730225, + "learning_rate": 2.5547445255474458e-06, + "loss": 2.3663, + "step": 154 + }, + { + "epoch": 0.16986301369863013, + "grad_norm": 2.3103132247924805, + "learning_rate": 2.5729927007299272e-06, + "loss": 2.4009, + "step": 155 + }, + { + "epoch": 0.17095890410958905, + "grad_norm": 3.576810836791992, + "learning_rate": 2.591240875912409e-06, + "loss": 2.3983, + "step": 156 + }, + { + "epoch": 0.17205479452054795, + "grad_norm": 2.337080955505371, + "learning_rate": 2.6094890510948905e-06, + "loss": 2.3849, + "step": 157 + }, + { + "epoch": 0.17315068493150684, + "grad_norm": 3.2523117065429688, + "learning_rate": 2.627737226277373e-06, + "loss": 2.4171, + "step": 158 + }, + { + "epoch": 0.17424657534246576, + "grad_norm": 2.529175281524658, + "learning_rate": 2.6459854014598542e-06, + "loss": 2.4303, + "step": 159 + }, + { + "epoch": 0.17534246575342466, + "grad_norm": 3.052212953567505, + "learning_rate": 2.664233576642336e-06, + "loss": 2.3895, + "step": 160 + }, + { + "epoch": 0.17534246575342466, + "eval_loss": 2.492485761642456, + "eval_runtime": 45.6738, + "eval_samples_per_second": 21.894, + "eval_steps_per_second": 0.92, + "step": 160 + }, + { + "epoch": 0.17643835616438355, + "grad_norm": 2.1544973850250244, + "learning_rate": 2.6824817518248175e-06, + "loss": 2.3649, + "step": 161 + }, + { + "epoch": 0.17753424657534247, + "grad_norm": 2.7351951599121094, + "learning_rate": 2.7007299270072994e-06, + "loss": 2.3908, + "step": 162 + }, + { + "epoch": 0.17863013698630137, + "grad_norm": 2.8431951999664307, + "learning_rate": 2.718978102189781e-06, + "loss": 2.374, + "step": 163 + }, + { + "epoch": 0.17972602739726026, + "grad_norm": 3.083190441131592, + "learning_rate": 2.737226277372263e-06, + "loss": 2.4193, + "step": 164 + }, + { + "epoch": 0.18082191780821918, + "grad_norm": 2.6523995399475098, + "learning_rate": 2.755474452554745e-06, + "loss": 2.3969, + "step": 165 + }, + { + "epoch": 0.18191780821917808, + "grad_norm": 2.626047372817993, + "learning_rate": 2.7737226277372264e-06, + "loss": 2.3742, + "step": 166 + }, + { + "epoch": 0.18301369863013697, + "grad_norm": 3.300917863845825, + "learning_rate": 2.7919708029197082e-06, + "loss": 2.4546, + "step": 167 + }, + { + "epoch": 0.1841095890410959, + "grad_norm": 2.797304391860962, + "learning_rate": 2.81021897810219e-06, + "loss": 2.4161, + "step": 168 + }, + { + "epoch": 0.1852054794520548, + "grad_norm": 2.777117967605591, + "learning_rate": 2.828467153284672e-06, + "loss": 2.3572, + "step": 169 + }, + { + "epoch": 0.1863013698630137, + "grad_norm": 2.887470006942749, + "learning_rate": 2.8467153284671534e-06, + "loss": 2.3887, + "step": 170 + }, + { + "epoch": 0.1863013698630137, + "eval_loss": 2.4854257106781006, + "eval_runtime": 44.5644, + "eval_samples_per_second": 22.439, + "eval_steps_per_second": 0.942, + "step": 170 + }, + { + "epoch": 0.1873972602739726, + "grad_norm": 3.0706076622009277, + "learning_rate": 2.8649635036496353e-06, + "loss": 2.3162, + "step": 171 + }, + { + "epoch": 0.1884931506849315, + "grad_norm": 3.6129744052886963, + "learning_rate": 2.8832116788321167e-06, + "loss": 2.3685, + "step": 172 + }, + { + "epoch": 0.18958904109589042, + "grad_norm": 2.4936370849609375, + "learning_rate": 2.901459854014599e-06, + "loss": 2.3297, + "step": 173 + }, + { + "epoch": 0.19068493150684931, + "grad_norm": 2.1041860580444336, + "learning_rate": 2.9197080291970804e-06, + "loss": 2.4085, + "step": 174 + }, + { + "epoch": 0.1917808219178082, + "grad_norm": 2.2066831588745117, + "learning_rate": 2.9379562043795623e-06, + "loss": 2.3905, + "step": 175 + }, + { + "epoch": 0.19287671232876713, + "grad_norm": 2.444652795791626, + "learning_rate": 2.956204379562044e-06, + "loss": 2.378, + "step": 176 + }, + { + "epoch": 0.19397260273972602, + "grad_norm": 2.501939058303833, + "learning_rate": 2.9744525547445256e-06, + "loss": 2.4136, + "step": 177 + }, + { + "epoch": 0.19506849315068492, + "grad_norm": 2.2167556285858154, + "learning_rate": 2.992700729927008e-06, + "loss": 2.385, + "step": 178 + }, + { + "epoch": 0.19616438356164384, + "grad_norm": 2.5298545360565186, + "learning_rate": 3.0109489051094893e-06, + "loss": 2.3508, + "step": 179 + }, + { + "epoch": 0.19726027397260273, + "grad_norm": 2.3774642944335938, + "learning_rate": 3.029197080291971e-06, + "loss": 2.3421, + "step": 180 + }, + { + "epoch": 0.19726027397260273, + "eval_loss": 2.476067304611206, + "eval_runtime": 44.3761, + "eval_samples_per_second": 22.535, + "eval_steps_per_second": 0.946, + "step": 180 + }, + { + "epoch": 0.19835616438356166, + "grad_norm": 2.9108712673187256, + "learning_rate": 3.0474452554744526e-06, + "loss": 2.3983, + "step": 181 + }, + { + "epoch": 0.19945205479452055, + "grad_norm": 2.5232033729553223, + "learning_rate": 3.065693430656935e-06, + "loss": 2.3946, + "step": 182 + }, + { + "epoch": 0.20054794520547944, + "grad_norm": 2.158168077468872, + "learning_rate": 3.0839416058394163e-06, + "loss": 2.3805, + "step": 183 + }, + { + "epoch": 0.20164383561643837, + "grad_norm": 2.3405354022979736, + "learning_rate": 3.102189781021898e-06, + "loss": 2.3974, + "step": 184 + }, + { + "epoch": 0.20273972602739726, + "grad_norm": 2.5308327674865723, + "learning_rate": 3.1204379562043796e-06, + "loss": 2.384, + "step": 185 + }, + { + "epoch": 0.20383561643835615, + "grad_norm": 2.6328237056732178, + "learning_rate": 3.1386861313868614e-06, + "loss": 2.3345, + "step": 186 + }, + { + "epoch": 0.20493150684931508, + "grad_norm": 3.0725512504577637, + "learning_rate": 3.156934306569343e-06, + "loss": 2.3191, + "step": 187 + }, + { + "epoch": 0.20602739726027397, + "grad_norm": 2.196218729019165, + "learning_rate": 3.175182481751825e-06, + "loss": 2.4073, + "step": 188 + }, + { + "epoch": 0.20712328767123286, + "grad_norm": 2.7722432613372803, + "learning_rate": 3.193430656934307e-06, + "loss": 2.3775, + "step": 189 + }, + { + "epoch": 0.20821917808219179, + "grad_norm": 2.4848241806030273, + "learning_rate": 3.2116788321167884e-06, + "loss": 2.3538, + "step": 190 + }, + { + "epoch": 0.20821917808219179, + "eval_loss": 2.462618112564087, + "eval_runtime": 43.9461, + "eval_samples_per_second": 22.755, + "eval_steps_per_second": 0.956, + "step": 190 + }, + { + "epoch": 0.20931506849315068, + "grad_norm": 2.4682741165161133, + "learning_rate": 3.2299270072992707e-06, + "loss": 2.3688, + "step": 191 + }, + { + "epoch": 0.2104109589041096, + "grad_norm": 2.518923282623291, + "learning_rate": 3.248175182481752e-06, + "loss": 2.3556, + "step": 192 + }, + { + "epoch": 0.2115068493150685, + "grad_norm": 2.4023561477661133, + "learning_rate": 3.266423357664234e-06, + "loss": 2.292, + "step": 193 + }, + { + "epoch": 0.2126027397260274, + "grad_norm": 2.5196943283081055, + "learning_rate": 3.2846715328467155e-06, + "loss": 2.3703, + "step": 194 + }, + { + "epoch": 0.2136986301369863, + "grad_norm": 2.762709617614746, + "learning_rate": 3.3029197080291973e-06, + "loss": 2.3578, + "step": 195 + }, + { + "epoch": 0.2147945205479452, + "grad_norm": 2.1287691593170166, + "learning_rate": 3.3211678832116788e-06, + "loss": 2.3687, + "step": 196 + }, + { + "epoch": 0.2158904109589041, + "grad_norm": 2.477341651916504, + "learning_rate": 3.339416058394161e-06, + "loss": 2.3545, + "step": 197 + }, + { + "epoch": 0.21698630136986302, + "grad_norm": 2.23933744430542, + "learning_rate": 3.3576642335766425e-06, + "loss": 2.3511, + "step": 198 + }, + { + "epoch": 0.21808219178082192, + "grad_norm": 2.2440133094787598, + "learning_rate": 3.3759124087591243e-06, + "loss": 2.3539, + "step": 199 + }, + { + "epoch": 0.2191780821917808, + "grad_norm": 2.5603177547454834, + "learning_rate": 3.3941605839416058e-06, + "loss": 2.36, + "step": 200 + }, + { + "epoch": 0.2191780821917808, + "eval_loss": 2.4507246017456055, + "eval_runtime": 43.7786, + "eval_samples_per_second": 22.842, + "eval_steps_per_second": 0.959, + "step": 200 + }, + { + "epoch": 0.22027397260273973, + "grad_norm": 2.4570510387420654, + "learning_rate": 3.412408759124088e-06, + "loss": 2.3702, + "step": 201 + }, + { + "epoch": 0.22136986301369863, + "grad_norm": 2.1975133419036865, + "learning_rate": 3.43065693430657e-06, + "loss": 2.3667, + "step": 202 + }, + { + "epoch": 0.22246575342465755, + "grad_norm": 2.710186004638672, + "learning_rate": 3.4489051094890513e-06, + "loss": 2.3159, + "step": 203 + }, + { + "epoch": 0.22356164383561644, + "grad_norm": 2.7799506187438965, + "learning_rate": 3.467153284671533e-06, + "loss": 2.3519, + "step": 204 + }, + { + "epoch": 0.22465753424657534, + "grad_norm": 2.2479400634765625, + "learning_rate": 3.4854014598540146e-06, + "loss": 2.3103, + "step": 205 + }, + { + "epoch": 0.22575342465753426, + "grad_norm": 3.1538405418395996, + "learning_rate": 3.503649635036497e-06, + "loss": 2.3506, + "step": 206 + }, + { + "epoch": 0.22684931506849315, + "grad_norm": 2.7607762813568115, + "learning_rate": 3.5218978102189783e-06, + "loss": 2.3386, + "step": 207 + }, + { + "epoch": 0.22794520547945205, + "grad_norm": 2.371105909347534, + "learning_rate": 3.54014598540146e-06, + "loss": 2.3583, + "step": 208 + }, + { + "epoch": 0.22904109589041097, + "grad_norm": 2.793933391571045, + "learning_rate": 3.5583941605839416e-06, + "loss": 2.3782, + "step": 209 + }, + { + "epoch": 0.23013698630136986, + "grad_norm": 2.372732400894165, + "learning_rate": 3.576642335766424e-06, + "loss": 2.417, + "step": 210 + }, + { + "epoch": 0.23013698630136986, + "eval_loss": 2.451474189758301, + "eval_runtime": 41.7197, + "eval_samples_per_second": 23.97, + "eval_steps_per_second": 1.007, + "step": 210 + }, + { + "epoch": 0.23123287671232876, + "grad_norm": 2.6989448070526123, + "learning_rate": 3.5948905109489054e-06, + "loss": 2.393, + "step": 211 + }, + { + "epoch": 0.23232876712328768, + "grad_norm": 2.4324285984039307, + "learning_rate": 3.6131386861313872e-06, + "loss": 2.2738, + "step": 212 + }, + { + "epoch": 0.23342465753424657, + "grad_norm": 3.5063016414642334, + "learning_rate": 3.6313868613138687e-06, + "loss": 2.2823, + "step": 213 + }, + { + "epoch": 0.2345205479452055, + "grad_norm": 3.953951358795166, + "learning_rate": 3.6496350364963505e-06, + "loss": 2.3132, + "step": 214 + }, + { + "epoch": 0.2356164383561644, + "grad_norm": 2.3483071327209473, + "learning_rate": 3.667883211678833e-06, + "loss": 2.3244, + "step": 215 + }, + { + "epoch": 0.23671232876712328, + "grad_norm": 2.884199380874634, + "learning_rate": 3.6861313868613142e-06, + "loss": 2.3067, + "step": 216 + }, + { + "epoch": 0.2378082191780822, + "grad_norm": 2.9514520168304443, + "learning_rate": 3.704379562043796e-06, + "loss": 2.304, + "step": 217 + }, + { + "epoch": 0.2389041095890411, + "grad_norm": 2.7530996799468994, + "learning_rate": 3.7226277372262775e-06, + "loss": 2.3691, + "step": 218 + }, + { + "epoch": 0.24, + "grad_norm": 2.8753647804260254, + "learning_rate": 3.7408759124087594e-06, + "loss": 2.3406, + "step": 219 + }, + { + "epoch": 0.2410958904109589, + "grad_norm": 2.670574903488159, + "learning_rate": 3.7591240875912412e-06, + "loss": 2.304, + "step": 220 + }, + { + "epoch": 0.2410958904109589, + "eval_loss": 2.4395809173583984, + "eval_runtime": 46.6723, + "eval_samples_per_second": 21.426, + "eval_steps_per_second": 0.9, + "step": 220 + }, + { + "epoch": 0.2421917808219178, + "grad_norm": 2.4152438640594482, + "learning_rate": 3.777372262773723e-06, + "loss": 2.3294, + "step": 221 + }, + { + "epoch": 0.2432876712328767, + "grad_norm": 2.8398540019989014, + "learning_rate": 3.7956204379562045e-06, + "loss": 2.3216, + "step": 222 + }, + { + "epoch": 0.24438356164383562, + "grad_norm": 2.6036765575408936, + "learning_rate": 3.8138686131386864e-06, + "loss": 2.3109, + "step": 223 + }, + { + "epoch": 0.24547945205479452, + "grad_norm": 2.7073473930358887, + "learning_rate": 3.832116788321168e-06, + "loss": 2.3457, + "step": 224 + }, + { + "epoch": 0.2465753424657534, + "grad_norm": 2.584404230117798, + "learning_rate": 3.85036496350365e-06, + "loss": 2.3159, + "step": 225 + }, + { + "epoch": 0.24767123287671233, + "grad_norm": 2.4220666885375977, + "learning_rate": 3.868613138686132e-06, + "loss": 2.2944, + "step": 226 + }, + { + "epoch": 0.24876712328767123, + "grad_norm": 2.5636494159698486, + "learning_rate": 3.886861313868613e-06, + "loss": 2.3219, + "step": 227 + }, + { + "epoch": 0.24986301369863015, + "grad_norm": 2.2990903854370117, + "learning_rate": 3.905109489051096e-06, + "loss": 2.3342, + "step": 228 + }, + { + "epoch": 0.25095890410958904, + "grad_norm": 2.50913143157959, + "learning_rate": 3.923357664233577e-06, + "loss": 2.3619, + "step": 229 + }, + { + "epoch": 0.25205479452054796, + "grad_norm": 2.1687769889831543, + "learning_rate": 3.9416058394160585e-06, + "loss": 2.3153, + "step": 230 + }, + { + "epoch": 0.25205479452054796, + "eval_loss": 2.4315240383148193, + "eval_runtime": 48.3542, + "eval_samples_per_second": 20.681, + "eval_steps_per_second": 0.869, + "step": 230 + }, + { + "epoch": 0.25315068493150683, + "grad_norm": 2.842125177383423, + "learning_rate": 3.95985401459854e-06, + "loss": 2.3644, + "step": 231 + }, + { + "epoch": 0.25424657534246575, + "grad_norm": 2.258049964904785, + "learning_rate": 3.978102189781022e-06, + "loss": 2.3727, + "step": 232 + }, + { + "epoch": 0.2553424657534247, + "grad_norm": 2.4630489349365234, + "learning_rate": 3.996350364963504e-06, + "loss": 2.267, + "step": 233 + }, + { + "epoch": 0.25643835616438354, + "grad_norm": 3.1469743251800537, + "learning_rate": 4.014598540145986e-06, + "loss": 2.3432, + "step": 234 + }, + { + "epoch": 0.25753424657534246, + "grad_norm": 2.6277477741241455, + "learning_rate": 4.032846715328467e-06, + "loss": 2.2545, + "step": 235 + }, + { + "epoch": 0.2586301369863014, + "grad_norm": 2.404141902923584, + "learning_rate": 4.05109489051095e-06, + "loss": 2.3109, + "step": 236 + }, + { + "epoch": 0.25972602739726025, + "grad_norm": 2.4545605182647705, + "learning_rate": 4.069343065693431e-06, + "loss": 2.3112, + "step": 237 + }, + { + "epoch": 0.2608219178082192, + "grad_norm": 2.237638235092163, + "learning_rate": 4.0875912408759126e-06, + "loss": 2.3, + "step": 238 + }, + { + "epoch": 0.2619178082191781, + "grad_norm": 2.192811965942383, + "learning_rate": 4.105839416058395e-06, + "loss": 2.3108, + "step": 239 + }, + { + "epoch": 0.26301369863013696, + "grad_norm": 2.5058720111846924, + "learning_rate": 4.124087591240876e-06, + "loss": 2.3635, + "step": 240 + }, + { + "epoch": 0.26301369863013696, + "eval_loss": 2.4219377040863037, + "eval_runtime": 45.3561, + "eval_samples_per_second": 22.048, + "eval_steps_per_second": 0.926, + "step": 240 + }, + { + "epoch": 0.2641095890410959, + "grad_norm": 2.0708935260772705, + "learning_rate": 4.1423357664233586e-06, + "loss": 2.3129, + "step": 241 + }, + { + "epoch": 0.2652054794520548, + "grad_norm": 2.6353116035461426, + "learning_rate": 4.16058394160584e-06, + "loss": 2.3011, + "step": 242 + }, + { + "epoch": 0.2663013698630137, + "grad_norm": 2.8392436504364014, + "learning_rate": 4.1788321167883214e-06, + "loss": 2.3643, + "step": 243 + }, + { + "epoch": 0.2673972602739726, + "grad_norm": 3.1773295402526855, + "learning_rate": 4.197080291970803e-06, + "loss": 2.2642, + "step": 244 + }, + { + "epoch": 0.2684931506849315, + "grad_norm": 2.4757354259490967, + "learning_rate": 4.215328467153285e-06, + "loss": 2.3211, + "step": 245 + }, + { + "epoch": 0.26958904109589044, + "grad_norm": 2.5799412727355957, + "learning_rate": 4.233576642335767e-06, + "loss": 2.2909, + "step": 246 + }, + { + "epoch": 0.2706849315068493, + "grad_norm": 2.368502378463745, + "learning_rate": 4.251824817518249e-06, + "loss": 2.2951, + "step": 247 + }, + { + "epoch": 0.2717808219178082, + "grad_norm": 2.114969253540039, + "learning_rate": 4.27007299270073e-06, + "loss": 2.3124, + "step": 248 + }, + { + "epoch": 0.27287671232876715, + "grad_norm": 3.056820869445801, + "learning_rate": 4.288321167883212e-06, + "loss": 2.2923, + "step": 249 + }, + { + "epoch": 0.273972602739726, + "grad_norm": 2.7752203941345215, + "learning_rate": 4.306569343065693e-06, + "loss": 2.2874, + "step": 250 + }, + { + "epoch": 0.273972602739726, + "eval_loss": 2.4229912757873535, + "eval_runtime": 52.5941, + "eval_samples_per_second": 19.014, + "eval_steps_per_second": 0.799, + "step": 250 + }, + { + "epoch": 0.27506849315068493, + "grad_norm": 3.045740842819214, + "learning_rate": 4.3248175182481755e-06, + "loss": 2.2686, + "step": 251 + }, + { + "epoch": 0.27616438356164386, + "grad_norm": 2.646402359008789, + "learning_rate": 4.343065693430658e-06, + "loss": 2.3247, + "step": 252 + }, + { + "epoch": 0.2772602739726027, + "grad_norm": 3.274994134902954, + "learning_rate": 4.361313868613139e-06, + "loss": 2.3093, + "step": 253 + }, + { + "epoch": 0.27835616438356164, + "grad_norm": 2.352246046066284, + "learning_rate": 4.379562043795621e-06, + "loss": 2.3124, + "step": 254 + }, + { + "epoch": 0.27945205479452057, + "grad_norm": 3.248717784881592, + "learning_rate": 4.397810218978102e-06, + "loss": 2.3417, + "step": 255 + }, + { + "epoch": 0.28054794520547943, + "grad_norm": 3.0919013023376465, + "learning_rate": 4.416058394160584e-06, + "loss": 2.3093, + "step": 256 + }, + { + "epoch": 0.28164383561643835, + "grad_norm": 2.1174702644348145, + "learning_rate": 4.434306569343066e-06, + "loss": 2.305, + "step": 257 + }, + { + "epoch": 0.2827397260273973, + "grad_norm": 2.913797378540039, + "learning_rate": 4.452554744525548e-06, + "loss": 2.2831, + "step": 258 + }, + { + "epoch": 0.28383561643835614, + "grad_norm": 2.693089723587036, + "learning_rate": 4.4708029197080295e-06, + "loss": 2.3206, + "step": 259 + }, + { + "epoch": 0.28493150684931506, + "grad_norm": 2.521411895751953, + "learning_rate": 4.489051094890512e-06, + "loss": 2.3196, + "step": 260 + }, + { + "epoch": 0.28493150684931506, + "eval_loss": 2.4178109169006348, + "eval_runtime": 50.6753, + "eval_samples_per_second": 19.733, + "eval_steps_per_second": 0.829, + "step": 260 + }, + { + "epoch": 0.286027397260274, + "grad_norm": 2.346346855163574, + "learning_rate": 4.507299270072993e-06, + "loss": 2.3433, + "step": 261 + }, + { + "epoch": 0.28712328767123285, + "grad_norm": 2.520368814468384, + "learning_rate": 4.525547445255475e-06, + "loss": 2.3174, + "step": 262 + }, + { + "epoch": 0.2882191780821918, + "grad_norm": 2.550347328186035, + "learning_rate": 4.543795620437956e-06, + "loss": 2.3072, + "step": 263 + }, + { + "epoch": 0.2893150684931507, + "grad_norm": 2.268662691116333, + "learning_rate": 4.562043795620438e-06, + "loss": 2.2763, + "step": 264 + }, + { + "epoch": 0.29041095890410956, + "grad_norm": 2.445798873901367, + "learning_rate": 4.580291970802921e-06, + "loss": 2.3395, + "step": 265 + }, + { + "epoch": 0.2915068493150685, + "grad_norm": 2.5391135215759277, + "learning_rate": 4.598540145985402e-06, + "loss": 2.3496, + "step": 266 + }, + { + "epoch": 0.2926027397260274, + "grad_norm": 2.3553097248077393, + "learning_rate": 4.6167883211678835e-06, + "loss": 2.3243, + "step": 267 + }, + { + "epoch": 0.2936986301369863, + "grad_norm": 2.448089599609375, + "learning_rate": 4.635036496350365e-06, + "loss": 2.2887, + "step": 268 + }, + { + "epoch": 0.2947945205479452, + "grad_norm": 2.1600170135498047, + "learning_rate": 4.653284671532847e-06, + "loss": 2.2816, + "step": 269 + }, + { + "epoch": 0.2958904109589041, + "grad_norm": 2.3783180713653564, + "learning_rate": 4.671532846715329e-06, + "loss": 2.3275, + "step": 270 + }, + { + "epoch": 0.2958904109589041, + "eval_loss": 2.413832902908325, + "eval_runtime": 48.7524, + "eval_samples_per_second": 20.512, + "eval_steps_per_second": 0.861, + "step": 270 + }, + { + "epoch": 0.29698630136986304, + "grad_norm": 2.468320369720459, + "learning_rate": 4.689781021897811e-06, + "loss": 2.3358, + "step": 271 + }, + { + "epoch": 0.2980821917808219, + "grad_norm": 2.401634454727173, + "learning_rate": 4.708029197080292e-06, + "loss": 2.2991, + "step": 272 + }, + { + "epoch": 0.2991780821917808, + "grad_norm": 2.4604203701019287, + "learning_rate": 4.726277372262774e-06, + "loss": 2.3496, + "step": 273 + }, + { + "epoch": 0.30027397260273975, + "grad_norm": 2.0338871479034424, + "learning_rate": 4.744525547445255e-06, + "loss": 2.3115, + "step": 274 + }, + { + "epoch": 0.3013698630136986, + "grad_norm": 2.5404436588287354, + "learning_rate": 4.7627737226277375e-06, + "loss": 2.2665, + "step": 275 + }, + { + "epoch": 0.30246575342465754, + "grad_norm": 2.5369458198547363, + "learning_rate": 4.78102189781022e-06, + "loss": 2.3072, + "step": 276 + }, + { + "epoch": 0.30356164383561646, + "grad_norm": 2.471522569656372, + "learning_rate": 4.799270072992701e-06, + "loss": 2.3113, + "step": 277 + }, + { + "epoch": 0.3046575342465753, + "grad_norm": 2.3890135288238525, + "learning_rate": 4.8175182481751835e-06, + "loss": 2.2838, + "step": 278 + }, + { + "epoch": 0.30575342465753425, + "grad_norm": 2.2485053539276123, + "learning_rate": 4.835766423357665e-06, + "loss": 2.3052, + "step": 279 + }, + { + "epoch": 0.30684931506849317, + "grad_norm": 2.1090753078460693, + "learning_rate": 4.854014598540146e-06, + "loss": 2.3054, + "step": 280 + }, + { + "epoch": 0.30684931506849317, + "eval_loss": 2.412548780441284, + "eval_runtime": 48.5034, + "eval_samples_per_second": 20.617, + "eval_steps_per_second": 0.866, + "step": 280 + }, + { + "epoch": 0.30794520547945203, + "grad_norm": 2.299133777618408, + "learning_rate": 4.872262773722628e-06, + "loss": 2.2599, + "step": 281 + }, + { + "epoch": 0.30904109589041096, + "grad_norm": 2.177905797958374, + "learning_rate": 4.89051094890511e-06, + "loss": 2.2745, + "step": 282 + }, + { + "epoch": 0.3101369863013699, + "grad_norm": 2.3178462982177734, + "learning_rate": 4.9087591240875915e-06, + "loss": 2.2562, + "step": 283 + }, + { + "epoch": 0.31123287671232874, + "grad_norm": 2.5088376998901367, + "learning_rate": 4.927007299270074e-06, + "loss": 2.2945, + "step": 284 + }, + { + "epoch": 0.31232876712328766, + "grad_norm": 2.095379590988159, + "learning_rate": 4.945255474452555e-06, + "loss": 2.3038, + "step": 285 + }, + { + "epoch": 0.3134246575342466, + "grad_norm": 2.1792995929718018, + "learning_rate": 4.963503649635037e-06, + "loss": 2.2702, + "step": 286 + }, + { + "epoch": 0.31452054794520545, + "grad_norm": 2.3001186847686768, + "learning_rate": 4.981751824817518e-06, + "loss": 2.3482, + "step": 287 + }, + { + "epoch": 0.3156164383561644, + "grad_norm": 2.1576106548309326, + "learning_rate": 5e-06, + "loss": 2.3391, + "step": 288 + }, + { + "epoch": 0.3167123287671233, + "grad_norm": 2.3715577125549316, + "learning_rate": 4.997969130787978e-06, + "loss": 2.3049, + "step": 289 + }, + { + "epoch": 0.3178082191780822, + "grad_norm": 2.0316715240478516, + "learning_rate": 4.995938261575955e-06, + "loss": 2.3111, + "step": 290 + }, + { + "epoch": 0.3178082191780822, + "eval_loss": 2.40299654006958, + "eval_runtime": 50.0873, + "eval_samples_per_second": 19.965, + "eval_steps_per_second": 0.839, + "step": 290 + }, + { + "epoch": 0.3189041095890411, + "grad_norm": 2.1135756969451904, + "learning_rate": 4.993907392363932e-06, + "loss": 2.2465, + "step": 291 + }, + { + "epoch": 0.32, + "grad_norm": 2.335655927658081, + "learning_rate": 4.9918765231519095e-06, + "loss": 2.31, + "step": 292 + }, + { + "epoch": 0.32109589041095893, + "grad_norm": 2.3255794048309326, + "learning_rate": 4.989845653939887e-06, + "loss": 2.2531, + "step": 293 + }, + { + "epoch": 0.3221917808219178, + "grad_norm": 2.111055612564087, + "learning_rate": 4.987814784727864e-06, + "loss": 2.2799, + "step": 294 + }, + { + "epoch": 0.3232876712328767, + "grad_norm": 2.1773171424865723, + "learning_rate": 4.985783915515841e-06, + "loss": 2.2884, + "step": 295 + }, + { + "epoch": 0.32438356164383564, + "grad_norm": 2.3559885025024414, + "learning_rate": 4.983753046303819e-06, + "loss": 2.2711, + "step": 296 + }, + { + "epoch": 0.3254794520547945, + "grad_norm": 2.120802879333496, + "learning_rate": 4.981722177091796e-06, + "loss": 2.2946, + "step": 297 + }, + { + "epoch": 0.3265753424657534, + "grad_norm": 2.2319388389587402, + "learning_rate": 4.979691307879773e-06, + "loss": 2.2771, + "step": 298 + }, + { + "epoch": 0.32767123287671235, + "grad_norm": 2.208740711212158, + "learning_rate": 4.9776604386677504e-06, + "loss": 2.3003, + "step": 299 + }, + { + "epoch": 0.3287671232876712, + "grad_norm": 2.543060541152954, + "learning_rate": 4.975629569455728e-06, + "loss": 2.2891, + "step": 300 + }, + { + "epoch": 0.3287671232876712, + "eval_loss": 2.398495674133301, + "eval_runtime": 51.1973, + "eval_samples_per_second": 19.532, + "eval_steps_per_second": 0.82, + "step": 300 + }, + { + "epoch": 0.32986301369863014, + "grad_norm": 2.086219072341919, + "learning_rate": 4.973598700243705e-06, + "loss": 2.3286, + "step": 301 + }, + { + "epoch": 0.33095890410958906, + "grad_norm": 2.3673319816589355, + "learning_rate": 4.971567831031681e-06, + "loss": 2.2672, + "step": 302 + }, + { + "epoch": 0.3320547945205479, + "grad_norm": 2.167593240737915, + "learning_rate": 4.9695369618196595e-06, + "loss": 2.264, + "step": 303 + }, + { + "epoch": 0.33315068493150685, + "grad_norm": 2.7218785285949707, + "learning_rate": 4.967506092607636e-06, + "loss": 2.1737, + "step": 304 + }, + { + "epoch": 0.33424657534246577, + "grad_norm": 2.673245429992676, + "learning_rate": 4.965475223395614e-06, + "loss": 2.2439, + "step": 305 + }, + { + "epoch": 0.33534246575342463, + "grad_norm": 2.0341989994049072, + "learning_rate": 4.9634443541835905e-06, + "loss": 2.2614, + "step": 306 + }, + { + "epoch": 0.33643835616438356, + "grad_norm": 1.9314247369766235, + "learning_rate": 4.961413484971569e-06, + "loss": 2.2592, + "step": 307 + }, + { + "epoch": 0.3375342465753425, + "grad_norm": 2.537794589996338, + "learning_rate": 4.959382615759545e-06, + "loss": 2.2859, + "step": 308 + }, + { + "epoch": 0.33863013698630134, + "grad_norm": 2.0811166763305664, + "learning_rate": 4.957351746547523e-06, + "loss": 2.2868, + "step": 309 + }, + { + "epoch": 0.33972602739726027, + "grad_norm": 2.783701181411743, + "learning_rate": 4.9553208773355e-06, + "loss": 2.2649, + "step": 310 + }, + { + "epoch": 0.33972602739726027, + "eval_loss": 2.400538921356201, + "eval_runtime": 43.185, + "eval_samples_per_second": 23.156, + "eval_steps_per_second": 0.973, + "step": 310 + }, + { + "epoch": 0.3408219178082192, + "grad_norm": 2.8541319370269775, + "learning_rate": 4.953290008123478e-06, + "loss": 2.3729, + "step": 311 + }, + { + "epoch": 0.3419178082191781, + "grad_norm": 2.1852617263793945, + "learning_rate": 4.951259138911454e-06, + "loss": 2.2885, + "step": 312 + }, + { + "epoch": 0.343013698630137, + "grad_norm": 2.872201919555664, + "learning_rate": 4.9492282696994315e-06, + "loss": 2.2353, + "step": 313 + }, + { + "epoch": 0.3441095890410959, + "grad_norm": 2.5140910148620605, + "learning_rate": 4.947197400487409e-06, + "loss": 2.3364, + "step": 314 + }, + { + "epoch": 0.3452054794520548, + "grad_norm": 2.619974374771118, + "learning_rate": 4.945166531275386e-06, + "loss": 2.2721, + "step": 315 + }, + { + "epoch": 0.3463013698630137, + "grad_norm": 2.4567618370056152, + "learning_rate": 4.943135662063363e-06, + "loss": 2.26, + "step": 316 + }, + { + "epoch": 0.3473972602739726, + "grad_norm": 2.9603192806243896, + "learning_rate": 4.9411047928513406e-06, + "loss": 2.2861, + "step": 317 + }, + { + "epoch": 0.34849315068493153, + "grad_norm": 2.115262985229492, + "learning_rate": 4.939073923639318e-06, + "loss": 2.3125, + "step": 318 + }, + { + "epoch": 0.3495890410958904, + "grad_norm": 2.693024158477783, + "learning_rate": 4.937043054427295e-06, + "loss": 2.3054, + "step": 319 + }, + { + "epoch": 0.3506849315068493, + "grad_norm": 2.3439996242523193, + "learning_rate": 4.935012185215272e-06, + "loss": 2.2576, + "step": 320 + }, + { + "epoch": 0.3506849315068493, + "eval_loss": 2.3920650482177734, + "eval_runtime": 47.7451, + "eval_samples_per_second": 20.945, + "eval_steps_per_second": 0.88, + "step": 320 + }, + { + "epoch": 0.35178082191780824, + "grad_norm": 2.3009634017944336, + "learning_rate": 4.93298131600325e-06, + "loss": 2.2454, + "step": 321 + }, + { + "epoch": 0.3528767123287671, + "grad_norm": 2.593203544616699, + "learning_rate": 4.930950446791227e-06, + "loss": 2.2965, + "step": 322 + }, + { + "epoch": 0.353972602739726, + "grad_norm": 2.9069650173187256, + "learning_rate": 4.928919577579204e-06, + "loss": 2.2785, + "step": 323 + }, + { + "epoch": 0.35506849315068495, + "grad_norm": 2.1021759510040283, + "learning_rate": 4.9268887083671815e-06, + "loss": 2.2857, + "step": 324 + }, + { + "epoch": 0.3561643835616438, + "grad_norm": 2.0330185890197754, + "learning_rate": 4.924857839155159e-06, + "loss": 2.2697, + "step": 325 + }, + { + "epoch": 0.35726027397260274, + "grad_norm": 2.1318063735961914, + "learning_rate": 4.922826969943136e-06, + "loss": 2.2698, + "step": 326 + }, + { + "epoch": 0.35835616438356166, + "grad_norm": 2.0157387256622314, + "learning_rate": 4.920796100731113e-06, + "loss": 2.292, + "step": 327 + }, + { + "epoch": 0.3594520547945205, + "grad_norm": 2.3266026973724365, + "learning_rate": 4.918765231519091e-06, + "loss": 2.2215, + "step": 328 + }, + { + "epoch": 0.36054794520547945, + "grad_norm": 2.066305160522461, + "learning_rate": 4.916734362307068e-06, + "loss": 2.2758, + "step": 329 + }, + { + "epoch": 0.36164383561643837, + "grad_norm": 2.0219993591308594, + "learning_rate": 4.914703493095045e-06, + "loss": 2.2381, + "step": 330 + }, + { + "epoch": 0.36164383561643837, + "eval_loss": 2.394064426422119, + "eval_runtime": 51.5482, + "eval_samples_per_second": 19.399, + "eval_steps_per_second": 0.815, + "step": 330 + }, + { + "epoch": 0.36273972602739724, + "grad_norm": 2.3964319229125977, + "learning_rate": 4.912672623883022e-06, + "loss": 2.2384, + "step": 331 + }, + { + "epoch": 0.36383561643835616, + "grad_norm": 2.5466530323028564, + "learning_rate": 4.910641754671e-06, + "loss": 2.2428, + "step": 332 + }, + { + "epoch": 0.3649315068493151, + "grad_norm": 2.236152410507202, + "learning_rate": 4.908610885458977e-06, + "loss": 2.2639, + "step": 333 + }, + { + "epoch": 0.36602739726027395, + "grad_norm": 2.7267303466796875, + "learning_rate": 4.906580016246954e-06, + "loss": 2.2608, + "step": 334 + }, + { + "epoch": 0.36712328767123287, + "grad_norm": 1.9403765201568604, + "learning_rate": 4.9045491470349315e-06, + "loss": 2.2962, + "step": 335 + }, + { + "epoch": 0.3682191780821918, + "grad_norm": 2.89070463180542, + "learning_rate": 4.902518277822909e-06, + "loss": 2.2847, + "step": 336 + }, + { + "epoch": 0.3693150684931507, + "grad_norm": 2.6427228450775146, + "learning_rate": 4.900487408610886e-06, + "loss": 2.2734, + "step": 337 + }, + { + "epoch": 0.3704109589041096, + "grad_norm": 2.1680362224578857, + "learning_rate": 4.898456539398863e-06, + "loss": 2.2869, + "step": 338 + }, + { + "epoch": 0.3715068493150685, + "grad_norm": 3.0160160064697266, + "learning_rate": 4.89642567018684e-06, + "loss": 2.3043, + "step": 339 + }, + { + "epoch": 0.3726027397260274, + "grad_norm": 1.9858418703079224, + "learning_rate": 4.894394800974818e-06, + "loss": 2.3422, + "step": 340 + }, + { + "epoch": 0.3726027397260274, + "eval_loss": 2.387559652328491, + "eval_runtime": 52.2455, + "eval_samples_per_second": 19.14, + "eval_steps_per_second": 0.804, + "step": 340 + }, + { + "epoch": 0.3736986301369863, + "grad_norm": 2.243699073791504, + "learning_rate": 4.892363931762794e-06, + "loss": 2.2394, + "step": 341 + }, + { + "epoch": 0.3747945205479452, + "grad_norm": 2.371299982070923, + "learning_rate": 4.8903330625507725e-06, + "loss": 2.2856, + "step": 342 + }, + { + "epoch": 0.37589041095890413, + "grad_norm": 2.3769736289978027, + "learning_rate": 4.888302193338749e-06, + "loss": 2.246, + "step": 343 + }, + { + "epoch": 0.376986301369863, + "grad_norm": 2.3209030628204346, + "learning_rate": 4.886271324126727e-06, + "loss": 2.2331, + "step": 344 + }, + { + "epoch": 0.3780821917808219, + "grad_norm": 2.747319221496582, + "learning_rate": 4.8842404549147034e-06, + "loss": 2.3115, + "step": 345 + }, + { + "epoch": 0.37917808219178084, + "grad_norm": 1.8599231243133545, + "learning_rate": 4.8822095857026816e-06, + "loss": 2.2994, + "step": 346 + }, + { + "epoch": 0.3802739726027397, + "grad_norm": 2.0358943939208984, + "learning_rate": 4.880178716490658e-06, + "loss": 2.271, + "step": 347 + }, + { + "epoch": 0.38136986301369863, + "grad_norm": 2.409106731414795, + "learning_rate": 4.878147847278636e-06, + "loss": 2.2281, + "step": 348 + }, + { + "epoch": 0.38246575342465755, + "grad_norm": 2.1641807556152344, + "learning_rate": 4.8761169780666125e-06, + "loss": 2.2437, + "step": 349 + }, + { + "epoch": 0.3835616438356164, + "grad_norm": 2.6917359828948975, + "learning_rate": 4.874086108854591e-06, + "loss": 2.226, + "step": 350 + }, + { + "epoch": 0.3835616438356164, + "eval_loss": 2.382991313934326, + "eval_runtime": 48.9534, + "eval_samples_per_second": 20.428, + "eval_steps_per_second": 0.858, + "step": 350 + }, + { + "epoch": 0.38465753424657534, + "grad_norm": 2.364215612411499, + "learning_rate": 4.872055239642567e-06, + "loss": 2.2319, + "step": 351 + }, + { + "epoch": 0.38575342465753426, + "grad_norm": 2.317643165588379, + "learning_rate": 4.870024370430544e-06, + "loss": 2.2096, + "step": 352 + }, + { + "epoch": 0.3868493150684931, + "grad_norm": 2.2283246517181396, + "learning_rate": 4.867993501218522e-06, + "loss": 2.2634, + "step": 353 + }, + { + "epoch": 0.38794520547945205, + "grad_norm": 2.498391628265381, + "learning_rate": 4.865962632006499e-06, + "loss": 2.2813, + "step": 354 + }, + { + "epoch": 0.38904109589041097, + "grad_norm": 2.3889403343200684, + "learning_rate": 4.863931762794476e-06, + "loss": 2.2361, + "step": 355 + }, + { + "epoch": 0.39013698630136984, + "grad_norm": 2.2049975395202637, + "learning_rate": 4.8619008935824535e-06, + "loss": 2.2583, + "step": 356 + }, + { + "epoch": 0.39123287671232876, + "grad_norm": 2.132840871810913, + "learning_rate": 4.859870024370431e-06, + "loss": 2.2823, + "step": 357 + }, + { + "epoch": 0.3923287671232877, + "grad_norm": 2.2692184448242188, + "learning_rate": 4.857839155158408e-06, + "loss": 2.2939, + "step": 358 + }, + { + "epoch": 0.3934246575342466, + "grad_norm": 2.0501933097839355, + "learning_rate": 4.855808285946385e-06, + "loss": 2.2976, + "step": 359 + }, + { + "epoch": 0.39452054794520547, + "grad_norm": 2.170236587524414, + "learning_rate": 4.8537774167343626e-06, + "loss": 2.2748, + "step": 360 + }, + { + "epoch": 0.39452054794520547, + "eval_loss": 2.3783886432647705, + "eval_runtime": 46.7324, + "eval_samples_per_second": 21.398, + "eval_steps_per_second": 0.899, + "step": 360 + }, + { + "epoch": 0.3956164383561644, + "grad_norm": 2.283137083053589, + "learning_rate": 4.85174654752234e-06, + "loss": 2.2705, + "step": 361 + }, + { + "epoch": 0.3967123287671233, + "grad_norm": 2.195984363555908, + "learning_rate": 4.849715678310317e-06, + "loss": 2.26, + "step": 362 + }, + { + "epoch": 0.3978082191780822, + "grad_norm": 2.399935007095337, + "learning_rate": 4.847684809098294e-06, + "loss": 2.2639, + "step": 363 + }, + { + "epoch": 0.3989041095890411, + "grad_norm": 2.411280393600464, + "learning_rate": 4.845653939886272e-06, + "loss": 2.2523, + "step": 364 + }, + { + "epoch": 0.4, + "grad_norm": 2.2033896446228027, + "learning_rate": 4.843623070674249e-06, + "loss": 2.2722, + "step": 365 + }, + { + "epoch": 0.4010958904109589, + "grad_norm": 2.2990238666534424, + "learning_rate": 4.841592201462226e-06, + "loss": 2.2411, + "step": 366 + }, + { + "epoch": 0.4021917808219178, + "grad_norm": 1.911453127861023, + "learning_rate": 4.8395613322502035e-06, + "loss": 2.2921, + "step": 367 + }, + { + "epoch": 0.40328767123287673, + "grad_norm": 2.453195095062256, + "learning_rate": 4.837530463038181e-06, + "loss": 2.218, + "step": 368 + }, + { + "epoch": 0.4043835616438356, + "grad_norm": 2.126992702484131, + "learning_rate": 4.835499593826158e-06, + "loss": 2.2628, + "step": 369 + }, + { + "epoch": 0.4054794520547945, + "grad_norm": 1.9305835962295532, + "learning_rate": 4.833468724614135e-06, + "loss": 2.2726, + "step": 370 + }, + { + "epoch": 0.4054794520547945, + "eval_loss": 2.3741114139556885, + "eval_runtime": 49.2956, + "eval_samples_per_second": 20.286, + "eval_steps_per_second": 0.852, + "step": 370 + }, + { + "epoch": 0.40657534246575344, + "grad_norm": 2.222024917602539, + "learning_rate": 4.831437855402113e-06, + "loss": 2.2634, + "step": 371 + }, + { + "epoch": 0.4076712328767123, + "grad_norm": 2.473958969116211, + "learning_rate": 4.82940698619009e-06, + "loss": 2.1817, + "step": 372 + }, + { + "epoch": 0.40876712328767123, + "grad_norm": 1.9208838939666748, + "learning_rate": 4.827376116978067e-06, + "loss": 2.2353, + "step": 373 + }, + { + "epoch": 0.40986301369863015, + "grad_norm": 2.7226450443267822, + "learning_rate": 4.8253452477660444e-06, + "loss": 2.2831, + "step": 374 + }, + { + "epoch": 0.410958904109589, + "grad_norm": 1.9078222513198853, + "learning_rate": 4.823314378554022e-06, + "loss": 2.264, + "step": 375 + }, + { + "epoch": 0.41205479452054794, + "grad_norm": 2.3946311473846436, + "learning_rate": 4.821283509341998e-06, + "loss": 2.251, + "step": 376 + }, + { + "epoch": 0.41315068493150686, + "grad_norm": 1.847343921661377, + "learning_rate": 4.819252640129976e-06, + "loss": 2.2441, + "step": 377 + }, + { + "epoch": 0.41424657534246573, + "grad_norm": 2.475385904312134, + "learning_rate": 4.817221770917953e-06, + "loss": 2.2639, + "step": 378 + }, + { + "epoch": 0.41534246575342465, + "grad_norm": 2.1396560668945312, + "learning_rate": 4.815190901705931e-06, + "loss": 2.2353, + "step": 379 + }, + { + "epoch": 0.41643835616438357, + "grad_norm": 2.3868885040283203, + "learning_rate": 4.813160032493907e-06, + "loss": 2.2551, + "step": 380 + }, + { + "epoch": 0.41643835616438357, + "eval_loss": 2.369168519973755, + "eval_runtime": 45.6131, + "eval_samples_per_second": 21.924, + "eval_steps_per_second": 0.921, + "step": 380 + }, + { + "epoch": 0.41753424657534244, + "grad_norm": 2.0193872451782227, + "learning_rate": 4.811129163281885e-06, + "loss": 2.2399, + "step": 381 + }, + { + "epoch": 0.41863013698630136, + "grad_norm": 2.4655935764312744, + "learning_rate": 4.809098294069862e-06, + "loss": 2.2729, + "step": 382 + }, + { + "epoch": 0.4197260273972603, + "grad_norm": 1.880974292755127, + "learning_rate": 4.80706742485784e-06, + "loss": 2.2681, + "step": 383 + }, + { + "epoch": 0.4208219178082192, + "grad_norm": 2.1912078857421875, + "learning_rate": 4.805036555645816e-06, + "loss": 2.2961, + "step": 384 + }, + { + "epoch": 0.42191780821917807, + "grad_norm": 1.985931634902954, + "learning_rate": 4.8030056864337945e-06, + "loss": 2.2407, + "step": 385 + }, + { + "epoch": 0.423013698630137, + "grad_norm": 2.4353220462799072, + "learning_rate": 4.800974817221771e-06, + "loss": 2.2867, + "step": 386 + }, + { + "epoch": 0.4241095890410959, + "grad_norm": 1.8810508251190186, + "learning_rate": 4.798943948009749e-06, + "loss": 2.2498, + "step": 387 + }, + { + "epoch": 0.4252054794520548, + "grad_norm": 2.195033073425293, + "learning_rate": 4.7969130787977254e-06, + "loss": 2.2532, + "step": 388 + }, + { + "epoch": 0.4263013698630137, + "grad_norm": 1.855372428894043, + "learning_rate": 4.794882209585703e-06, + "loss": 2.2538, + "step": 389 + }, + { + "epoch": 0.4273972602739726, + "grad_norm": 2.4465434551239014, + "learning_rate": 4.79285134037368e-06, + "loss": 2.2724, + "step": 390 + }, + { + "epoch": 0.4273972602739726, + "eval_loss": 2.364928960800171, + "eval_runtime": 48.0245, + "eval_samples_per_second": 20.823, + "eval_steps_per_second": 0.875, + "step": 390 + }, + { + "epoch": 0.4284931506849315, + "grad_norm": 1.8911314010620117, + "learning_rate": 4.790820471161657e-06, + "loss": 2.2291, + "step": 391 + }, + { + "epoch": 0.4295890410958904, + "grad_norm": 2.1975977420806885, + "learning_rate": 4.7887896019496345e-06, + "loss": 2.2373, + "step": 392 + }, + { + "epoch": 0.43068493150684933, + "grad_norm": 2.015331506729126, + "learning_rate": 4.786758732737612e-06, + "loss": 2.2483, + "step": 393 + }, + { + "epoch": 0.4317808219178082, + "grad_norm": 2.039339065551758, + "learning_rate": 4.784727863525589e-06, + "loss": 2.2746, + "step": 394 + }, + { + "epoch": 0.4328767123287671, + "grad_norm": 1.977513074874878, + "learning_rate": 4.782696994313566e-06, + "loss": 2.2275, + "step": 395 + }, + { + "epoch": 0.43397260273972604, + "grad_norm": 2.147217035293579, + "learning_rate": 4.780666125101544e-06, + "loss": 2.252, + "step": 396 + }, + { + "epoch": 0.4350684931506849, + "grad_norm": 2.1799919605255127, + "learning_rate": 4.778635255889521e-06, + "loss": 2.1975, + "step": 397 + }, + { + "epoch": 0.43616438356164383, + "grad_norm": 2.582240343093872, + "learning_rate": 4.776604386677498e-06, + "loss": 2.2294, + "step": 398 + }, + { + "epoch": 0.43726027397260275, + "grad_norm": 1.9127846956253052, + "learning_rate": 4.7745735174654755e-06, + "loss": 2.2319, + "step": 399 + }, + { + "epoch": 0.4383561643835616, + "grad_norm": 2.2444543838500977, + "learning_rate": 4.772542648253453e-06, + "loss": 2.2857, + "step": 400 + }, + { + "epoch": 0.4383561643835616, + "eval_loss": 2.3653724193573, + "eval_runtime": 43.8304, + "eval_samples_per_second": 22.815, + "eval_steps_per_second": 0.958, + "step": 400 + }, + { + "epoch": 0.43945205479452054, + "grad_norm": 1.9913676977157593, + "learning_rate": 4.77051177904143e-06, + "loss": 2.274, + "step": 401 + }, + { + "epoch": 0.44054794520547946, + "grad_norm": 2.2850067615509033, + "learning_rate": 4.768480909829407e-06, + "loss": 2.2209, + "step": 402 + }, + { + "epoch": 0.44164383561643833, + "grad_norm": 2.4984688758850098, + "learning_rate": 4.766450040617385e-06, + "loss": 2.263, + "step": 403 + }, + { + "epoch": 0.44273972602739725, + "grad_norm": 2.3244640827178955, + "learning_rate": 4.764419171405362e-06, + "loss": 2.2368, + "step": 404 + }, + { + "epoch": 0.4438356164383562, + "grad_norm": 2.7579681873321533, + "learning_rate": 4.762388302193339e-06, + "loss": 2.2762, + "step": 405 + }, + { + "epoch": 0.4449315068493151, + "grad_norm": 2.1996467113494873, + "learning_rate": 4.760357432981316e-06, + "loss": 2.2536, + "step": 406 + }, + { + "epoch": 0.44602739726027396, + "grad_norm": 2.3767571449279785, + "learning_rate": 4.758326563769294e-06, + "loss": 2.2353, + "step": 407 + }, + { + "epoch": 0.4471232876712329, + "grad_norm": 1.9148359298706055, + "learning_rate": 4.756295694557271e-06, + "loss": 2.2416, + "step": 408 + }, + { + "epoch": 0.4482191780821918, + "grad_norm": 2.6499507427215576, + "learning_rate": 4.754264825345248e-06, + "loss": 2.2641, + "step": 409 + }, + { + "epoch": 0.44931506849315067, + "grad_norm": 1.9890286922454834, + "learning_rate": 4.7522339561332255e-06, + "loss": 2.2148, + "step": 410 + }, + { + "epoch": 0.44931506849315067, + "eval_loss": 2.3670520782470703, + "eval_runtime": 41.1722, + "eval_samples_per_second": 24.288, + "eval_steps_per_second": 1.02, + "step": 410 + }, + { + "epoch": 0.4504109589041096, + "grad_norm": 2.575968027114868, + "learning_rate": 4.750203086921203e-06, + "loss": 2.2254, + "step": 411 + }, + { + "epoch": 0.4515068493150685, + "grad_norm": 2.015174388885498, + "learning_rate": 4.74817221770918e-06, + "loss": 2.1881, + "step": 412 + }, + { + "epoch": 0.4526027397260274, + "grad_norm": 2.44292950630188, + "learning_rate": 4.7461413484971565e-06, + "loss": 2.256, + "step": 413 + }, + { + "epoch": 0.4536986301369863, + "grad_norm": 1.9808441400527954, + "learning_rate": 4.744110479285135e-06, + "loss": 2.285, + "step": 414 + }, + { + "epoch": 0.4547945205479452, + "grad_norm": 2.009481906890869, + "learning_rate": 4.742079610073111e-06, + "loss": 2.2018, + "step": 415 + }, + { + "epoch": 0.4558904109589041, + "grad_norm": 2.1466031074523926, + "learning_rate": 4.740048740861089e-06, + "loss": 2.224, + "step": 416 + }, + { + "epoch": 0.456986301369863, + "grad_norm": 2.1337907314300537, + "learning_rate": 4.738017871649066e-06, + "loss": 2.2404, + "step": 417 + }, + { + "epoch": 0.45808219178082193, + "grad_norm": 1.94125235080719, + "learning_rate": 4.735987002437044e-06, + "loss": 2.2474, + "step": 418 + }, + { + "epoch": 0.4591780821917808, + "grad_norm": 2.379099130630493, + "learning_rate": 4.73395613322502e-06, + "loss": 2.2858, + "step": 419 + }, + { + "epoch": 0.4602739726027397, + "grad_norm": 1.844056248664856, + "learning_rate": 4.731925264012998e-06, + "loss": 2.2967, + "step": 420 + }, + { + "epoch": 0.4602739726027397, + "eval_loss": 2.3622241020202637, + "eval_runtime": 45.4674, + "eval_samples_per_second": 21.994, + "eval_steps_per_second": 0.924, + "step": 420 + }, + { + "epoch": 0.46136986301369864, + "grad_norm": 2.411895513534546, + "learning_rate": 4.729894394800975e-06, + "loss": 2.274, + "step": 421 + }, + { + "epoch": 0.4624657534246575, + "grad_norm": 2.014885425567627, + "learning_rate": 4.727863525588953e-06, + "loss": 2.2144, + "step": 422 + }, + { + "epoch": 0.46356164383561643, + "grad_norm": 2.112009048461914, + "learning_rate": 4.725832656376929e-06, + "loss": 2.1824, + "step": 423 + }, + { + "epoch": 0.46465753424657535, + "grad_norm": 2.111392021179199, + "learning_rate": 4.723801787164907e-06, + "loss": 2.2261, + "step": 424 + }, + { + "epoch": 0.4657534246575342, + "grad_norm": 2.0349185466766357, + "learning_rate": 4.721770917952885e-06, + "loss": 2.2622, + "step": 425 + }, + { + "epoch": 0.46684931506849314, + "grad_norm": 2.30222487449646, + "learning_rate": 4.719740048740861e-06, + "loss": 2.2206, + "step": 426 + }, + { + "epoch": 0.46794520547945206, + "grad_norm": 2.018789529800415, + "learning_rate": 4.717709179528839e-06, + "loss": 2.2196, + "step": 427 + }, + { + "epoch": 0.469041095890411, + "grad_norm": 2.046649932861328, + "learning_rate": 4.715678310316816e-06, + "loss": 2.2584, + "step": 428 + }, + { + "epoch": 0.47013698630136985, + "grad_norm": 2.0393869876861572, + "learning_rate": 4.713647441104794e-06, + "loss": 2.2537, + "step": 429 + }, + { + "epoch": 0.4712328767123288, + "grad_norm": 2.2788116931915283, + "learning_rate": 4.71161657189277e-06, + "loss": 2.2145, + "step": 430 + }, + { + "epoch": 0.4712328767123288, + "eval_loss": 2.3575785160064697, + "eval_runtime": 44.6687, + "eval_samples_per_second": 22.387, + "eval_steps_per_second": 0.94, + "step": 430 + }, + { + "epoch": 0.4723287671232877, + "grad_norm": 1.9950047731399536, + "learning_rate": 4.709585702680748e-06, + "loss": 2.2072, + "step": 431 + }, + { + "epoch": 0.47342465753424656, + "grad_norm": 1.9222451448440552, + "learning_rate": 4.707554833468725e-06, + "loss": 2.2654, + "step": 432 + }, + { + "epoch": 0.4745205479452055, + "grad_norm": 1.9735718965530396, + "learning_rate": 4.705523964256703e-06, + "loss": 2.3037, + "step": 433 + }, + { + "epoch": 0.4756164383561644, + "grad_norm": 2.3209776878356934, + "learning_rate": 4.703493095044679e-06, + "loss": 2.2083, + "step": 434 + }, + { + "epoch": 0.4767123287671233, + "grad_norm": 2.0993995666503906, + "learning_rate": 4.701462225832657e-06, + "loss": 2.2275, + "step": 435 + }, + { + "epoch": 0.4778082191780822, + "grad_norm": 2.334463119506836, + "learning_rate": 4.699431356620634e-06, + "loss": 2.1852, + "step": 436 + }, + { + "epoch": 0.4789041095890411, + "grad_norm": 1.9098213911056519, + "learning_rate": 4.697400487408612e-06, + "loss": 2.2829, + "step": 437 + }, + { + "epoch": 0.48, + "grad_norm": 2.3005056381225586, + "learning_rate": 4.695369618196588e-06, + "loss": 2.2532, + "step": 438 + }, + { + "epoch": 0.4810958904109589, + "grad_norm": 1.9069781303405762, + "learning_rate": 4.693338748984566e-06, + "loss": 2.2374, + "step": 439 + }, + { + "epoch": 0.4821917808219178, + "grad_norm": 2.050032138824463, + "learning_rate": 4.691307879772543e-06, + "loss": 2.2256, + "step": 440 + }, + { + "epoch": 0.4821917808219178, + "eval_loss": 2.3587422370910645, + "eval_runtime": 45.0941, + "eval_samples_per_second": 22.176, + "eval_steps_per_second": 0.931, + "step": 440 + }, + { + "epoch": 0.4832876712328767, + "grad_norm": 2.0563836097717285, + "learning_rate": 4.68927701056052e-06, + "loss": 2.1918, + "step": 441 + }, + { + "epoch": 0.4843835616438356, + "grad_norm": 2.1598386764526367, + "learning_rate": 4.6872461413484975e-06, + "loss": 2.2618, + "step": 442 + }, + { + "epoch": 0.48547945205479454, + "grad_norm": 1.9236539602279663, + "learning_rate": 4.685215272136475e-06, + "loss": 2.2477, + "step": 443 + }, + { + "epoch": 0.4865753424657534, + "grad_norm": 1.9923924207687378, + "learning_rate": 4.683184402924452e-06, + "loss": 2.2313, + "step": 444 + }, + { + "epoch": 0.4876712328767123, + "grad_norm": 2.162627696990967, + "learning_rate": 4.681153533712429e-06, + "loss": 2.2367, + "step": 445 + }, + { + "epoch": 0.48876712328767125, + "grad_norm": 2.083523988723755, + "learning_rate": 4.679122664500407e-06, + "loss": 2.2946, + "step": 446 + }, + { + "epoch": 0.4898630136986301, + "grad_norm": 2.504408359527588, + "learning_rate": 4.677091795288384e-06, + "loss": 2.2104, + "step": 447 + }, + { + "epoch": 0.49095890410958903, + "grad_norm": 1.8316532373428345, + "learning_rate": 4.675060926076361e-06, + "loss": 2.2027, + "step": 448 + }, + { + "epoch": 0.49205479452054796, + "grad_norm": 2.3337953090667725, + "learning_rate": 4.673030056864338e-06, + "loss": 2.278, + "step": 449 + }, + { + "epoch": 0.4931506849315068, + "grad_norm": 2.24102783203125, + "learning_rate": 4.670999187652316e-06, + "loss": 2.2018, + "step": 450 + }, + { + "epoch": 0.4931506849315068, + "eval_loss": 2.3545279502868652, + "eval_runtime": 45.1041, + "eval_samples_per_second": 22.171, + "eval_steps_per_second": 0.931, + "step": 450 + }, + { + "epoch": 0.49424657534246574, + "grad_norm": 2.07834792137146, + "learning_rate": 4.668968318440293e-06, + "loss": 2.1975, + "step": 451 + }, + { + "epoch": 0.49534246575342467, + "grad_norm": 1.7681723833084106, + "learning_rate": 4.66693744922827e-06, + "loss": 2.2033, + "step": 452 + }, + { + "epoch": 0.4964383561643836, + "grad_norm": 2.062134027481079, + "learning_rate": 4.6649065800162475e-06, + "loss": 2.2646, + "step": 453 + }, + { + "epoch": 0.49753424657534245, + "grad_norm": 1.94647216796875, + "learning_rate": 4.662875710804225e-06, + "loss": 2.1748, + "step": 454 + }, + { + "epoch": 0.4986301369863014, + "grad_norm": 1.9173872470855713, + "learning_rate": 4.660844841592202e-06, + "loss": 2.2435, + "step": 455 + }, + { + "epoch": 0.4997260273972603, + "grad_norm": 1.88880455493927, + "learning_rate": 4.658813972380179e-06, + "loss": 2.1711, + "step": 456 + }, + { + "epoch": 0.5008219178082192, + "grad_norm": 1.9628506898880005, + "learning_rate": 4.656783103168157e-06, + "loss": 2.2701, + "step": 457 + }, + { + "epoch": 0.5019178082191781, + "grad_norm": 2.0515286922454834, + "learning_rate": 4.654752233956134e-06, + "loss": 2.1656, + "step": 458 + }, + { + "epoch": 0.503013698630137, + "grad_norm": 1.978522539138794, + "learning_rate": 4.652721364744111e-06, + "loss": 2.246, + "step": 459 + }, + { + "epoch": 0.5041095890410959, + "grad_norm": 1.7993966341018677, + "learning_rate": 4.6506904955320885e-06, + "loss": 2.214, + "step": 460 + }, + { + "epoch": 0.5041095890410959, + "eval_loss": 2.3568837642669678, + "eval_runtime": 45.2086, + "eval_samples_per_second": 22.12, + "eval_steps_per_second": 0.929, + "step": 460 + }, + { + "epoch": 0.5052054794520547, + "grad_norm": 1.7809834480285645, + "learning_rate": 4.648659626320066e-06, + "loss": 2.2525, + "step": 461 + }, + { + "epoch": 0.5063013698630137, + "grad_norm": 1.9256166219711304, + "learning_rate": 4.646628757108043e-06, + "loss": 2.2611, + "step": 462 + }, + { + "epoch": 0.5073972602739726, + "grad_norm": 1.94192636013031, + "learning_rate": 4.6445978878960194e-06, + "loss": 2.2282, + "step": 463 + }, + { + "epoch": 0.5084931506849315, + "grad_norm": 1.8891818523406982, + "learning_rate": 4.6425670186839976e-06, + "loss": 2.2036, + "step": 464 + }, + { + "epoch": 0.5095890410958904, + "grad_norm": 2.352078676223755, + "learning_rate": 4.640536149471974e-06, + "loss": 2.2293, + "step": 465 + }, + { + "epoch": 0.5106849315068493, + "grad_norm": 1.8141659498214722, + "learning_rate": 4.638505280259952e-06, + "loss": 2.2551, + "step": 466 + }, + { + "epoch": 0.5117808219178083, + "grad_norm": 1.9492480754852295, + "learning_rate": 4.6364744110479285e-06, + "loss": 2.2641, + "step": 467 + }, + { + "epoch": 0.5128767123287671, + "grad_norm": 2.0210182666778564, + "learning_rate": 4.634443541835907e-06, + "loss": 2.2367, + "step": 468 + }, + { + "epoch": 0.513972602739726, + "grad_norm": 1.8936505317687988, + "learning_rate": 4.632412672623883e-06, + "loss": 2.265, + "step": 469 + }, + { + "epoch": 0.5150684931506849, + "grad_norm": 1.845888614654541, + "learning_rate": 4.630381803411861e-06, + "loss": 2.2169, + "step": 470 + }, + { + "epoch": 0.5150684931506849, + "eval_loss": 2.353489875793457, + "eval_runtime": 46.8739, + "eval_samples_per_second": 21.334, + "eval_steps_per_second": 0.896, + "step": 470 + }, + { + "epoch": 0.5161643835616438, + "grad_norm": 1.889675259590149, + "learning_rate": 4.628350934199838e-06, + "loss": 2.1383, + "step": 471 + }, + { + "epoch": 0.5172602739726028, + "grad_norm": 1.9799216985702515, + "learning_rate": 4.626320064987816e-06, + "loss": 2.2312, + "step": 472 + }, + { + "epoch": 0.5183561643835617, + "grad_norm": 2.0128607749938965, + "learning_rate": 4.624289195775792e-06, + "loss": 2.2529, + "step": 473 + }, + { + "epoch": 0.5194520547945205, + "grad_norm": 1.8791241645812988, + "learning_rate": 4.62225832656377e-06, + "loss": 2.226, + "step": 474 + }, + { + "epoch": 0.5205479452054794, + "grad_norm": 2.0497279167175293, + "learning_rate": 4.620227457351747e-06, + "loss": 2.2283, + "step": 475 + }, + { + "epoch": 0.5216438356164383, + "grad_norm": 1.8097424507141113, + "learning_rate": 4.618196588139724e-06, + "loss": 2.222, + "step": 476 + }, + { + "epoch": 0.5227397260273973, + "grad_norm": 1.8960500955581665, + "learning_rate": 4.616165718927701e-06, + "loss": 2.2342, + "step": 477 + }, + { + "epoch": 0.5238356164383562, + "grad_norm": 2.020730495452881, + "learning_rate": 4.6141348497156786e-06, + "loss": 2.2106, + "step": 478 + }, + { + "epoch": 0.5249315068493151, + "grad_norm": 1.8955297470092773, + "learning_rate": 4.612103980503656e-06, + "loss": 2.2755, + "step": 479 + }, + { + "epoch": 0.5260273972602739, + "grad_norm": 1.9767351150512695, + "learning_rate": 4.610073111291633e-06, + "loss": 2.2554, + "step": 480 + }, + { + "epoch": 0.5260273972602739, + "eval_loss": 2.347705841064453, + "eval_runtime": 46.9125, + "eval_samples_per_second": 21.316, + "eval_steps_per_second": 0.895, + "step": 480 + }, + { + "epoch": 0.5271232876712328, + "grad_norm": 1.936457633972168, + "learning_rate": 4.60804224207961e-06, + "loss": 2.2323, + "step": 481 + }, + { + "epoch": 0.5282191780821918, + "grad_norm": 1.9271862506866455, + "learning_rate": 4.606011372867588e-06, + "loss": 2.2151, + "step": 482 + }, + { + "epoch": 0.5293150684931507, + "grad_norm": 1.878153681755066, + "learning_rate": 4.603980503655565e-06, + "loss": 2.2322, + "step": 483 + }, + { + "epoch": 0.5304109589041096, + "grad_norm": 2.1009249687194824, + "learning_rate": 4.601949634443542e-06, + "loss": 2.2181, + "step": 484 + }, + { + "epoch": 0.5315068493150685, + "grad_norm": 1.9089181423187256, + "learning_rate": 4.5999187652315195e-06, + "loss": 2.1712, + "step": 485 + }, + { + "epoch": 0.5326027397260275, + "grad_norm": 2.207282781600952, + "learning_rate": 4.597887896019497e-06, + "loss": 2.1847, + "step": 486 + }, + { + "epoch": 0.5336986301369863, + "grad_norm": 2.001584053039551, + "learning_rate": 4.595857026807474e-06, + "loss": 2.214, + "step": 487 + }, + { + "epoch": 0.5347945205479452, + "grad_norm": 1.966700553894043, + "learning_rate": 4.593826157595451e-06, + "loss": 2.2237, + "step": 488 + }, + { + "epoch": 0.5358904109589041, + "grad_norm": 1.842469334602356, + "learning_rate": 4.591795288383429e-06, + "loss": 2.2089, + "step": 489 + }, + { + "epoch": 0.536986301369863, + "grad_norm": 2.0412986278533936, + "learning_rate": 4.589764419171406e-06, + "loss": 2.2734, + "step": 490 + }, + { + "epoch": 0.536986301369863, + "eval_loss": 2.3460772037506104, + "eval_runtime": 50.4048, + "eval_samples_per_second": 19.839, + "eval_steps_per_second": 0.833, + "step": 490 + }, + { + "epoch": 0.538082191780822, + "grad_norm": 1.9337031841278076, + "learning_rate": 4.587733549959383e-06, + "loss": 2.2314, + "step": 491 + }, + { + "epoch": 0.5391780821917809, + "grad_norm": 2.4448320865631104, + "learning_rate": 4.5857026807473604e-06, + "loss": 2.2098, + "step": 492 + }, + { + "epoch": 0.5402739726027397, + "grad_norm": 1.83162522315979, + "learning_rate": 4.583671811535338e-06, + "loss": 2.1902, + "step": 493 + }, + { + "epoch": 0.5413698630136986, + "grad_norm": 2.3614656925201416, + "learning_rate": 4.581640942323315e-06, + "loss": 2.1737, + "step": 494 + }, + { + "epoch": 0.5424657534246575, + "grad_norm": 2.001187801361084, + "learning_rate": 4.579610073111292e-06, + "loss": 2.2485, + "step": 495 + }, + { + "epoch": 0.5435616438356164, + "grad_norm": 2.3479933738708496, + "learning_rate": 4.5775792038992695e-06, + "loss": 2.2595, + "step": 496 + }, + { + "epoch": 0.5446575342465754, + "grad_norm": 2.135856866836548, + "learning_rate": 4.575548334687247e-06, + "loss": 2.2528, + "step": 497 + }, + { + "epoch": 0.5457534246575343, + "grad_norm": 1.983126163482666, + "learning_rate": 4.573517465475224e-06, + "loss": 2.2788, + "step": 498 + }, + { + "epoch": 0.5468493150684931, + "grad_norm": 2.18208909034729, + "learning_rate": 4.571486596263201e-06, + "loss": 2.1995, + "step": 499 + }, + { + "epoch": 0.547945205479452, + "grad_norm": 1.848747968673706, + "learning_rate": 4.569455727051178e-06, + "loss": 2.2403, + "step": 500 + }, + { + "epoch": 0.547945205479452, + "eval_loss": 2.344066619873047, + "eval_runtime": 51.1213, + "eval_samples_per_second": 19.561, + "eval_steps_per_second": 0.822, + "step": 500 + }, + { + "epoch": 0.549041095890411, + "grad_norm": 2.427860736846924, + "learning_rate": 4.567424857839156e-06, + "loss": 2.246, + "step": 501 + }, + { + "epoch": 0.5501369863013699, + "grad_norm": 1.913405418395996, + "learning_rate": 4.565393988627132e-06, + "loss": 2.2296, + "step": 502 + }, + { + "epoch": 0.5512328767123288, + "grad_norm": 2.0106029510498047, + "learning_rate": 4.5633631194151105e-06, + "loss": 2.2657, + "step": 503 + }, + { + "epoch": 0.5523287671232877, + "grad_norm": 2.2719595432281494, + "learning_rate": 4.561332250203087e-06, + "loss": 2.2341, + "step": 504 + }, + { + "epoch": 0.5534246575342465, + "grad_norm": 2.109786033630371, + "learning_rate": 4.559301380991065e-06, + "loss": 2.2181, + "step": 505 + }, + { + "epoch": 0.5545205479452054, + "grad_norm": 2.189779281616211, + "learning_rate": 4.5572705117790414e-06, + "loss": 2.1821, + "step": 506 + }, + { + "epoch": 0.5556164383561644, + "grad_norm": 2.0263590812683105, + "learning_rate": 4.5552396425670196e-06, + "loss": 2.2072, + "step": 507 + }, + { + "epoch": 0.5567123287671233, + "grad_norm": 2.215289354324341, + "learning_rate": 4.553208773354996e-06, + "loss": 2.2513, + "step": 508 + }, + { + "epoch": 0.5578082191780822, + "grad_norm": 2.3031351566314697, + "learning_rate": 4.551177904142974e-06, + "loss": 2.1988, + "step": 509 + }, + { + "epoch": 0.5589041095890411, + "grad_norm": 2.1830058097839355, + "learning_rate": 4.5491470349309505e-06, + "loss": 2.269, + "step": 510 + }, + { + "epoch": 0.5589041095890411, + "eval_loss": 2.342463731765747, + "eval_runtime": 43.1956, + "eval_samples_per_second": 23.151, + "eval_steps_per_second": 0.972, + "step": 510 + }, + { + "epoch": 0.56, + "grad_norm": 2.3210105895996094, + "learning_rate": 4.547116165718929e-06, + "loss": 2.2051, + "step": 511 + }, + { + "epoch": 0.5610958904109589, + "grad_norm": 1.9175361394882202, + "learning_rate": 4.545085296506905e-06, + "loss": 2.2304, + "step": 512 + }, + { + "epoch": 0.5621917808219178, + "grad_norm": 2.302947759628296, + "learning_rate": 4.543054427294882e-06, + "loss": 2.215, + "step": 513 + }, + { + "epoch": 0.5632876712328767, + "grad_norm": 1.8947396278381348, + "learning_rate": 4.54102355808286e-06, + "loss": 2.1926, + "step": 514 + }, + { + "epoch": 0.5643835616438356, + "grad_norm": 2.195268154144287, + "learning_rate": 4.538992688870837e-06, + "loss": 2.1937, + "step": 515 + }, + { + "epoch": 0.5654794520547946, + "grad_norm": 2.178874969482422, + "learning_rate": 4.536961819658814e-06, + "loss": 2.2248, + "step": 516 + }, + { + "epoch": 0.5665753424657535, + "grad_norm": 1.9856929779052734, + "learning_rate": 4.5349309504467915e-06, + "loss": 2.2077, + "step": 517 + }, + { + "epoch": 0.5676712328767123, + "grad_norm": 2.0686943531036377, + "learning_rate": 4.532900081234769e-06, + "loss": 2.2328, + "step": 518 + }, + { + "epoch": 0.5687671232876712, + "grad_norm": 1.9554919004440308, + "learning_rate": 4.530869212022746e-06, + "loss": 2.198, + "step": 519 + }, + { + "epoch": 0.5698630136986301, + "grad_norm": 1.9489670991897583, + "learning_rate": 4.528838342810723e-06, + "loss": 2.2287, + "step": 520 + }, + { + "epoch": 0.5698630136986301, + "eval_loss": 2.3415298461914062, + "eval_runtime": 48.7918, + "eval_samples_per_second": 20.495, + "eval_steps_per_second": 0.861, + "step": 520 + }, + { + "epoch": 0.570958904109589, + "grad_norm": 1.9868658781051636, + "learning_rate": 4.526807473598701e-06, + "loss": 2.2205, + "step": 521 + }, + { + "epoch": 0.572054794520548, + "grad_norm": 2.0965914726257324, + "learning_rate": 4.524776604386678e-06, + "loss": 2.2239, + "step": 522 + }, + { + "epoch": 0.5731506849315069, + "grad_norm": 2.041471481323242, + "learning_rate": 4.522745735174655e-06, + "loss": 2.2035, + "step": 523 + }, + { + "epoch": 0.5742465753424657, + "grad_norm": 1.885293960571289, + "learning_rate": 4.520714865962632e-06, + "loss": 2.2469, + "step": 524 + }, + { + "epoch": 0.5753424657534246, + "grad_norm": 1.9107296466827393, + "learning_rate": 4.51868399675061e-06, + "loss": 2.2386, + "step": 525 + }, + { + "epoch": 0.5764383561643835, + "grad_norm": 1.9945271015167236, + "learning_rate": 4.516653127538587e-06, + "loss": 2.235, + "step": 526 + }, + { + "epoch": 0.5775342465753425, + "grad_norm": 1.9787253141403198, + "learning_rate": 4.514622258326564e-06, + "loss": 2.2034, + "step": 527 + }, + { + "epoch": 0.5786301369863014, + "grad_norm": 2.0300636291503906, + "learning_rate": 4.5125913891145415e-06, + "loss": 2.1849, + "step": 528 + }, + { + "epoch": 0.5797260273972603, + "grad_norm": 2.0348970890045166, + "learning_rate": 4.510560519902519e-06, + "loss": 2.2174, + "step": 529 + }, + { + "epoch": 0.5808219178082191, + "grad_norm": 1.8589537143707275, + "learning_rate": 4.508529650690496e-06, + "loss": 2.2351, + "step": 530 + }, + { + "epoch": 0.5808219178082191, + "eval_loss": 2.3450186252593994, + "eval_runtime": 52.4548, + "eval_samples_per_second": 19.064, + "eval_steps_per_second": 0.801, + "step": 530 + }, + { + "epoch": 0.581917808219178, + "grad_norm": 2.0496020317077637, + "learning_rate": 4.506498781478473e-06, + "loss": 2.182, + "step": 531 + }, + { + "epoch": 0.583013698630137, + "grad_norm": 2.0649805068969727, + "learning_rate": 4.504467912266451e-06, + "loss": 2.1654, + "step": 532 + }, + { + "epoch": 0.5841095890410959, + "grad_norm": 1.9809486865997314, + "learning_rate": 4.502437043054428e-06, + "loss": 2.2182, + "step": 533 + }, + { + "epoch": 0.5852054794520548, + "grad_norm": 2.21915340423584, + "learning_rate": 4.500406173842405e-06, + "loss": 2.2697, + "step": 534 + }, + { + "epoch": 0.5863013698630137, + "grad_norm": 1.9822649955749512, + "learning_rate": 4.4983753046303824e-06, + "loss": 2.2297, + "step": 535 + }, + { + "epoch": 0.5873972602739727, + "grad_norm": 1.972575306892395, + "learning_rate": 4.49634443541836e-06, + "loss": 2.2164, + "step": 536 + }, + { + "epoch": 0.5884931506849315, + "grad_norm": 1.948667049407959, + "learning_rate": 4.494313566206336e-06, + "loss": 2.2087, + "step": 537 + }, + { + "epoch": 0.5895890410958904, + "grad_norm": 2.7486417293548584, + "learning_rate": 4.492282696994314e-06, + "loss": 2.2602, + "step": 538 + }, + { + "epoch": 0.5906849315068493, + "grad_norm": 2.3374552726745605, + "learning_rate": 4.490251827782291e-06, + "loss": 2.2054, + "step": 539 + }, + { + "epoch": 0.5917808219178082, + "grad_norm": 2.222116231918335, + "learning_rate": 4.488220958570269e-06, + "loss": 2.1868, + "step": 540 + }, + { + "epoch": 0.5917808219178082, + "eval_loss": 2.345618963241577, + "eval_runtime": 53.0302, + "eval_samples_per_second": 18.857, + "eval_steps_per_second": 0.792, + "step": 540 + }, + { + "epoch": 0.5928767123287672, + "grad_norm": 2.00752854347229, + "learning_rate": 4.486190089358245e-06, + "loss": 2.187, + "step": 541 + }, + { + "epoch": 0.5939726027397261, + "grad_norm": 2.2466132640838623, + "learning_rate": 4.484159220146223e-06, + "loss": 2.1611, + "step": 542 + }, + { + "epoch": 0.5950684931506849, + "grad_norm": 2.1279385089874268, + "learning_rate": 4.4821283509342e-06, + "loss": 2.248, + "step": 543 + }, + { + "epoch": 0.5961643835616438, + "grad_norm": 1.9140571355819702, + "learning_rate": 4.480097481722178e-06, + "loss": 2.2314, + "step": 544 + }, + { + "epoch": 0.5972602739726027, + "grad_norm": 2.120741605758667, + "learning_rate": 4.478066612510154e-06, + "loss": 2.212, + "step": 545 + }, + { + "epoch": 0.5983561643835617, + "grad_norm": 2.003713369369507, + "learning_rate": 4.4760357432981325e-06, + "loss": 2.1824, + "step": 546 + }, + { + "epoch": 0.5994520547945206, + "grad_norm": 1.9161221981048584, + "learning_rate": 4.474004874086109e-06, + "loss": 2.2015, + "step": 547 + }, + { + "epoch": 0.6005479452054795, + "grad_norm": 2.367110252380371, + "learning_rate": 4.471974004874087e-06, + "loss": 2.3122, + "step": 548 + }, + { + "epoch": 0.6016438356164383, + "grad_norm": 2.113380193710327, + "learning_rate": 4.4699431356620635e-06, + "loss": 2.1692, + "step": 549 + }, + { + "epoch": 0.6027397260273972, + "grad_norm": 1.974181890487671, + "learning_rate": 4.467912266450041e-06, + "loss": 2.2509, + "step": 550 + }, + { + "epoch": 0.6027397260273972, + "eval_loss": 2.342583656311035, + "eval_runtime": 50.7362, + "eval_samples_per_second": 19.71, + "eval_steps_per_second": 0.828, + "step": 550 + }, + { + "epoch": 0.6038356164383561, + "grad_norm": 1.889894962310791, + "learning_rate": 4.465881397238018e-06, + "loss": 2.2214, + "step": 551 + }, + { + "epoch": 0.6049315068493151, + "grad_norm": 1.8982343673706055, + "learning_rate": 4.463850528025995e-06, + "loss": 2.2135, + "step": 552 + }, + { + "epoch": 0.606027397260274, + "grad_norm": 1.9462087154388428, + "learning_rate": 4.4618196588139726e-06, + "loss": 2.2255, + "step": 553 + }, + { + "epoch": 0.6071232876712329, + "grad_norm": 1.8729931116104126, + "learning_rate": 4.45978878960195e-06, + "loss": 2.2358, + "step": 554 + }, + { + "epoch": 0.6082191780821918, + "grad_norm": 1.9230246543884277, + "learning_rate": 4.457757920389927e-06, + "loss": 2.222, + "step": 555 + }, + { + "epoch": 0.6093150684931506, + "grad_norm": 1.9495900869369507, + "learning_rate": 4.455727051177904e-06, + "loss": 2.1779, + "step": 556 + }, + { + "epoch": 0.6104109589041096, + "grad_norm": 1.983487606048584, + "learning_rate": 4.453696181965882e-06, + "loss": 2.2007, + "step": 557 + }, + { + "epoch": 0.6115068493150685, + "grad_norm": 2.1541049480438232, + "learning_rate": 4.451665312753859e-06, + "loss": 2.2513, + "step": 558 + }, + { + "epoch": 0.6126027397260274, + "grad_norm": 1.9133352041244507, + "learning_rate": 4.449634443541836e-06, + "loss": 2.2288, + "step": 559 + }, + { + "epoch": 0.6136986301369863, + "grad_norm": 2.3325741291046143, + "learning_rate": 4.4476035743298135e-06, + "loss": 2.2539, + "step": 560 + }, + { + "epoch": 0.6136986301369863, + "eval_loss": 2.3387348651885986, + "eval_runtime": 53.4301, + "eval_samples_per_second": 18.716, + "eval_steps_per_second": 0.786, + "step": 560 + }, + { + "epoch": 0.6147945205479453, + "grad_norm": 1.8099673986434937, + "learning_rate": 4.445572705117791e-06, + "loss": 2.1845, + "step": 561 + }, + { + "epoch": 0.6158904109589041, + "grad_norm": 2.201901912689209, + "learning_rate": 4.443541835905768e-06, + "loss": 2.195, + "step": 562 + }, + { + "epoch": 0.616986301369863, + "grad_norm": 1.8412714004516602, + "learning_rate": 4.441510966693745e-06, + "loss": 2.2551, + "step": 563 + }, + { + "epoch": 0.6180821917808219, + "grad_norm": 1.8761520385742188, + "learning_rate": 4.439480097481723e-06, + "loss": 2.1965, + "step": 564 + }, + { + "epoch": 0.6191780821917808, + "grad_norm": 1.9123467206954956, + "learning_rate": 4.4374492282697e-06, + "loss": 2.2502, + "step": 565 + }, + { + "epoch": 0.6202739726027398, + "grad_norm": 2.0631234645843506, + "learning_rate": 4.435418359057677e-06, + "loss": 2.1582, + "step": 566 + }, + { + "epoch": 0.6213698630136987, + "grad_norm": 2.3191616535186768, + "learning_rate": 4.433387489845654e-06, + "loss": 2.2271, + "step": 567 + }, + { + "epoch": 0.6224657534246575, + "grad_norm": 2.043057918548584, + "learning_rate": 4.431356620633632e-06, + "loss": 2.2292, + "step": 568 + }, + { + "epoch": 0.6235616438356164, + "grad_norm": 2.0716068744659424, + "learning_rate": 4.429325751421609e-06, + "loss": 2.196, + "step": 569 + }, + { + "epoch": 0.6246575342465753, + "grad_norm": 1.7669776678085327, + "learning_rate": 4.427294882209586e-06, + "loss": 2.1953, + "step": 570 + }, + { + "epoch": 0.6246575342465753, + "eval_loss": 2.336578607559204, + "eval_runtime": 50.0083, + "eval_samples_per_second": 19.997, + "eval_steps_per_second": 0.84, + "step": 570 + }, + { + "epoch": 0.6257534246575343, + "grad_norm": 1.8165771961212158, + "learning_rate": 4.4252640129975635e-06, + "loss": 2.2155, + "step": 571 + }, + { + "epoch": 0.6268493150684932, + "grad_norm": 1.8640249967575073, + "learning_rate": 4.423233143785541e-06, + "loss": 2.1648, + "step": 572 + }, + { + "epoch": 0.6279452054794521, + "grad_norm": 2.013869285583496, + "learning_rate": 4.421202274573518e-06, + "loss": 2.2131, + "step": 573 + }, + { + "epoch": 0.6290410958904109, + "grad_norm": 2.2650580406188965, + "learning_rate": 4.4191714053614945e-06, + "loss": 2.2287, + "step": 574 + }, + { + "epoch": 0.6301369863013698, + "grad_norm": 1.9232103824615479, + "learning_rate": 4.417140536149473e-06, + "loss": 2.1339, + "step": 575 + }, + { + "epoch": 0.6312328767123287, + "grad_norm": 1.910134196281433, + "learning_rate": 4.415109666937449e-06, + "loss": 2.1755, + "step": 576 + }, + { + "epoch": 0.6323287671232877, + "grad_norm": 2.2492005825042725, + "learning_rate": 4.413078797725427e-06, + "loss": 2.2362, + "step": 577 + }, + { + "epoch": 0.6334246575342466, + "grad_norm": 1.8584033250808716, + "learning_rate": 4.411047928513404e-06, + "loss": 2.1911, + "step": 578 + }, + { + "epoch": 0.6345205479452055, + "grad_norm": 2.156273126602173, + "learning_rate": 4.409017059301382e-06, + "loss": 2.2109, + "step": 579 + }, + { + "epoch": 0.6356164383561644, + "grad_norm": 1.7977535724639893, + "learning_rate": 4.406986190089358e-06, + "loss": 2.2195, + "step": 580 + }, + { + "epoch": 0.6356164383561644, + "eval_loss": 2.33555269241333, + "eval_runtime": 48.3919, + "eval_samples_per_second": 20.665, + "eval_steps_per_second": 0.868, + "step": 580 + }, + { + "epoch": 0.6367123287671232, + "grad_norm": 1.8262702226638794, + "learning_rate": 4.404955320877336e-06, + "loss": 2.2345, + "step": 581 + }, + { + "epoch": 0.6378082191780822, + "grad_norm": 1.9389383792877197, + "learning_rate": 4.402924451665313e-06, + "loss": 2.2347, + "step": 582 + }, + { + "epoch": 0.6389041095890411, + "grad_norm": 1.9548988342285156, + "learning_rate": 4.400893582453291e-06, + "loss": 2.1967, + "step": 583 + }, + { + "epoch": 0.64, + "grad_norm": 1.898092269897461, + "learning_rate": 4.398862713241267e-06, + "loss": 2.1795, + "step": 584 + }, + { + "epoch": 0.6410958904109589, + "grad_norm": 1.8183610439300537, + "learning_rate": 4.396831844029245e-06, + "loss": 2.2031, + "step": 585 + }, + { + "epoch": 0.6421917808219179, + "grad_norm": 2.1520016193389893, + "learning_rate": 4.394800974817222e-06, + "loss": 2.2051, + "step": 586 + }, + { + "epoch": 0.6432876712328767, + "grad_norm": 1.9930012226104736, + "learning_rate": 4.392770105605199e-06, + "loss": 2.2053, + "step": 587 + }, + { + "epoch": 0.6443835616438356, + "grad_norm": 2.311450958251953, + "learning_rate": 4.390739236393176e-06, + "loss": 2.2252, + "step": 588 + }, + { + "epoch": 0.6454794520547945, + "grad_norm": 2.0168697834014893, + "learning_rate": 4.388708367181154e-06, + "loss": 2.2286, + "step": 589 + }, + { + "epoch": 0.6465753424657534, + "grad_norm": 2.055964469909668, + "learning_rate": 4.386677497969131e-06, + "loss": 2.2278, + "step": 590 + }, + { + "epoch": 0.6465753424657534, + "eval_loss": 2.334367275238037, + "eval_runtime": 48.9414, + "eval_samples_per_second": 20.433, + "eval_steps_per_second": 0.858, + "step": 590 + }, + { + "epoch": 0.6476712328767124, + "grad_norm": 2.0161590576171875, + "learning_rate": 4.384646628757108e-06, + "loss": 2.2102, + "step": 591 + }, + { + "epoch": 0.6487671232876713, + "grad_norm": 2.0827667713165283, + "learning_rate": 4.3826157595450855e-06, + "loss": 2.2531, + "step": 592 + }, + { + "epoch": 0.6498630136986301, + "grad_norm": 2.077637195587158, + "learning_rate": 4.380584890333063e-06, + "loss": 2.2091, + "step": 593 + }, + { + "epoch": 0.650958904109589, + "grad_norm": 2.023892879486084, + "learning_rate": 4.37855402112104e-06, + "loss": 2.1684, + "step": 594 + }, + { + "epoch": 0.6520547945205479, + "grad_norm": 1.8509857654571533, + "learning_rate": 4.376523151909017e-06, + "loss": 2.1439, + "step": 595 + }, + { + "epoch": 0.6531506849315069, + "grad_norm": 2.1649608612060547, + "learning_rate": 4.3744922826969946e-06, + "loss": 2.2557, + "step": 596 + }, + { + "epoch": 0.6542465753424658, + "grad_norm": 2.1351773738861084, + "learning_rate": 4.372461413484972e-06, + "loss": 2.177, + "step": 597 + }, + { + "epoch": 0.6553424657534247, + "grad_norm": 2.1713974475860596, + "learning_rate": 4.370430544272949e-06, + "loss": 2.2223, + "step": 598 + }, + { + "epoch": 0.6564383561643835, + "grad_norm": 2.2905397415161133, + "learning_rate": 4.368399675060926e-06, + "loss": 2.2373, + "step": 599 + }, + { + "epoch": 0.6575342465753424, + "grad_norm": 2.007160186767578, + "learning_rate": 4.366368805848904e-06, + "loss": 2.2355, + "step": 600 + }, + { + "epoch": 0.6575342465753424, + "eval_loss": 2.3322324752807617, + "eval_runtime": 46.2769, + "eval_samples_per_second": 21.609, + "eval_steps_per_second": 0.908, + "step": 600 + }, + { + "epoch": 0.6586301369863014, + "grad_norm": 2.309783697128296, + "learning_rate": 4.364337936636881e-06, + "loss": 2.2052, + "step": 601 + }, + { + "epoch": 0.6597260273972603, + "grad_norm": 2.498034954071045, + "learning_rate": 4.362307067424858e-06, + "loss": 2.1567, + "step": 602 + }, + { + "epoch": 0.6608219178082192, + "grad_norm": 1.8302545547485352, + "learning_rate": 4.3602761982128355e-06, + "loss": 2.2117, + "step": 603 + }, + { + "epoch": 0.6619178082191781, + "grad_norm": 2.226947546005249, + "learning_rate": 4.358245329000813e-06, + "loss": 2.1881, + "step": 604 + }, + { + "epoch": 0.663013698630137, + "grad_norm": 1.8484126329421997, + "learning_rate": 4.35621445978879e-06, + "loss": 2.1955, + "step": 605 + }, + { + "epoch": 0.6641095890410958, + "grad_norm": 1.9844266176223755, + "learning_rate": 4.354183590576767e-06, + "loss": 2.2063, + "step": 606 + }, + { + "epoch": 0.6652054794520548, + "grad_norm": 1.9292365312576294, + "learning_rate": 4.352152721364745e-06, + "loss": 2.205, + "step": 607 + }, + { + "epoch": 0.6663013698630137, + "grad_norm": 1.980895757675171, + "learning_rate": 4.350121852152722e-06, + "loss": 2.1727, + "step": 608 + }, + { + "epoch": 0.6673972602739726, + "grad_norm": 1.8331713676452637, + "learning_rate": 4.348090982940699e-06, + "loss": 2.226, + "step": 609 + }, + { + "epoch": 0.6684931506849315, + "grad_norm": 1.9666537046432495, + "learning_rate": 4.3460601137286764e-06, + "loss": 2.2198, + "step": 610 + }, + { + "epoch": 0.6684931506849315, + "eval_loss": 2.3297338485717773, + "eval_runtime": 42.6274, + "eval_samples_per_second": 23.459, + "eval_steps_per_second": 0.985, + "step": 610 + }, + { + "epoch": 0.6695890410958905, + "grad_norm": 1.9741110801696777, + "learning_rate": 4.344029244516654e-06, + "loss": 2.2132, + "step": 611 + }, + { + "epoch": 0.6706849315068493, + "grad_norm": 2.0747334957122803, + "learning_rate": 4.341998375304631e-06, + "loss": 2.2081, + "step": 612 + }, + { + "epoch": 0.6717808219178082, + "grad_norm": 2.031503915786743, + "learning_rate": 4.339967506092607e-06, + "loss": 2.2642, + "step": 613 + }, + { + "epoch": 0.6728767123287671, + "grad_norm": 1.8513256311416626, + "learning_rate": 4.3379366368805855e-06, + "loss": 2.2141, + "step": 614 + }, + { + "epoch": 0.673972602739726, + "grad_norm": 1.9098315238952637, + "learning_rate": 4.335905767668562e-06, + "loss": 2.2064, + "step": 615 + }, + { + "epoch": 0.675068493150685, + "grad_norm": 1.9704065322875977, + "learning_rate": 4.33387489845654e-06, + "loss": 2.1997, + "step": 616 + }, + { + "epoch": 0.6761643835616439, + "grad_norm": 2.1520018577575684, + "learning_rate": 4.3318440292445165e-06, + "loss": 2.2278, + "step": 617 + }, + { + "epoch": 0.6772602739726027, + "grad_norm": 1.898164987564087, + "learning_rate": 4.329813160032495e-06, + "loss": 2.2147, + "step": 618 + }, + { + "epoch": 0.6783561643835616, + "grad_norm": 1.8955962657928467, + "learning_rate": 4.327782290820471e-06, + "loss": 2.1825, + "step": 619 + }, + { + "epoch": 0.6794520547945205, + "grad_norm": 1.7579313516616821, + "learning_rate": 4.325751421608449e-06, + "loss": 2.1619, + "step": 620 + }, + { + "epoch": 0.6794520547945205, + "eval_loss": 2.329692840576172, + "eval_runtime": 46.3187, + "eval_samples_per_second": 21.59, + "eval_steps_per_second": 0.907, + "step": 620 + }, + { + "epoch": 0.6805479452054795, + "grad_norm": 1.8716063499450684, + "learning_rate": 4.323720552396426e-06, + "loss": 2.2289, + "step": 621 + }, + { + "epoch": 0.6816438356164384, + "grad_norm": 1.8798290491104126, + "learning_rate": 4.321689683184404e-06, + "loss": 2.1944, + "step": 622 + }, + { + "epoch": 0.6827397260273973, + "grad_norm": 1.8438303470611572, + "learning_rate": 4.31965881397238e-06, + "loss": 2.1879, + "step": 623 + }, + { + "epoch": 0.6838356164383562, + "grad_norm": 1.7844786643981934, + "learning_rate": 4.3176279447603574e-06, + "loss": 2.1936, + "step": 624 + }, + { + "epoch": 0.684931506849315, + "grad_norm": 1.955529808998108, + "learning_rate": 4.315597075548335e-06, + "loss": 2.2258, + "step": 625 + }, + { + "epoch": 0.686027397260274, + "grad_norm": 2.3826634883880615, + "learning_rate": 4.313566206336312e-06, + "loss": 2.2135, + "step": 626 + }, + { + "epoch": 0.6871232876712329, + "grad_norm": 1.8453123569488525, + "learning_rate": 4.311535337124289e-06, + "loss": 2.2362, + "step": 627 + }, + { + "epoch": 0.6882191780821918, + "grad_norm": 2.742682933807373, + "learning_rate": 4.3095044679122665e-06, + "loss": 2.194, + "step": 628 + }, + { + "epoch": 0.6893150684931507, + "grad_norm": 2.0299901962280273, + "learning_rate": 4.307473598700244e-06, + "loss": 2.2029, + "step": 629 + }, + { + "epoch": 0.6904109589041096, + "grad_norm": 2.29683780670166, + "learning_rate": 4.305442729488221e-06, + "loss": 2.2141, + "step": 630 + }, + { + "epoch": 0.6904109589041096, + "eval_loss": 2.329368829727173, + "eval_runtime": 45.1947, + "eval_samples_per_second": 22.126, + "eval_steps_per_second": 0.929, + "step": 630 + }, + { + "epoch": 0.6915068493150685, + "grad_norm": 2.208280086517334, + "learning_rate": 4.303411860276198e-06, + "loss": 2.1637, + "step": 631 + }, + { + "epoch": 0.6926027397260274, + "grad_norm": 1.943514108657837, + "learning_rate": 4.301380991064176e-06, + "loss": 2.234, + "step": 632 + }, + { + "epoch": 0.6936986301369863, + "grad_norm": 2.486440420150757, + "learning_rate": 4.299350121852153e-06, + "loss": 2.2239, + "step": 633 + }, + { + "epoch": 0.6947945205479452, + "grad_norm": 1.7243690490722656, + "learning_rate": 4.29731925264013e-06, + "loss": 2.2428, + "step": 634 + }, + { + "epoch": 0.6958904109589041, + "grad_norm": 2.31227445602417, + "learning_rate": 4.2952883834281075e-06, + "loss": 2.2403, + "step": 635 + }, + { + "epoch": 0.6969863013698631, + "grad_norm": 1.7361263036727905, + "learning_rate": 4.293257514216085e-06, + "loss": 2.1975, + "step": 636 + }, + { + "epoch": 0.6980821917808219, + "grad_norm": 2.010831356048584, + "learning_rate": 4.291226645004062e-06, + "loss": 2.1633, + "step": 637 + }, + { + "epoch": 0.6991780821917808, + "grad_norm": 2.2796833515167236, + "learning_rate": 4.289195775792039e-06, + "loss": 2.2307, + "step": 638 + }, + { + "epoch": 0.7002739726027397, + "grad_norm": 2.1069483757019043, + "learning_rate": 4.287164906580017e-06, + "loss": 2.2389, + "step": 639 + }, + { + "epoch": 0.7013698630136986, + "grad_norm": 2.19460129737854, + "learning_rate": 4.285134037367994e-06, + "loss": 2.1992, + "step": 640 + }, + { + "epoch": 0.7013698630136986, + "eval_loss": 2.3223369121551514, + "eval_runtime": 43.1093, + "eval_samples_per_second": 23.197, + "eval_steps_per_second": 0.974, + "step": 640 + }, + { + "epoch": 0.7024657534246576, + "grad_norm": 1.875261902809143, + "learning_rate": 4.283103168155971e-06, + "loss": 2.1767, + "step": 641 + }, + { + "epoch": 0.7035616438356165, + "grad_norm": 1.906762957572937, + "learning_rate": 4.281072298943948e-06, + "loss": 2.1734, + "step": 642 + }, + { + "epoch": 0.7046575342465753, + "grad_norm": 2.1114180088043213, + "learning_rate": 4.279041429731926e-06, + "loss": 2.1734, + "step": 643 + }, + { + "epoch": 0.7057534246575342, + "grad_norm": 1.844373106956482, + "learning_rate": 4.277010560519903e-06, + "loss": 2.1967, + "step": 644 + }, + { + "epoch": 0.7068493150684931, + "grad_norm": 2.3463754653930664, + "learning_rate": 4.27497969130788e-06, + "loss": 2.1735, + "step": 645 + }, + { + "epoch": 0.707945205479452, + "grad_norm": 2.3530960083007812, + "learning_rate": 4.2729488220958575e-06, + "loss": 2.189, + "step": 646 + }, + { + "epoch": 0.709041095890411, + "grad_norm": 2.110963821411133, + "learning_rate": 4.270917952883835e-06, + "loss": 2.218, + "step": 647 + }, + { + "epoch": 0.7101369863013699, + "grad_norm": 2.625005006790161, + "learning_rate": 4.268887083671812e-06, + "loss": 2.2235, + "step": 648 + }, + { + "epoch": 0.7112328767123288, + "grad_norm": 1.8411654233932495, + "learning_rate": 4.266856214459789e-06, + "loss": 2.1757, + "step": 649 + }, + { + "epoch": 0.7123287671232876, + "grad_norm": 2.0817489624023438, + "learning_rate": 4.264825345247766e-06, + "loss": 2.2698, + "step": 650 + }, + { + "epoch": 0.7123287671232876, + "eval_loss": 2.322544574737549, + "eval_runtime": 43.0771, + "eval_samples_per_second": 23.214, + "eval_steps_per_second": 0.975, + "step": 650 + }, + { + "epoch": 0.7134246575342466, + "grad_norm": 1.8953847885131836, + "learning_rate": 4.262794476035744e-06, + "loss": 2.1745, + "step": 651 + }, + { + "epoch": 0.7145205479452055, + "grad_norm": 1.8463294506072998, + "learning_rate": 4.26076360682372e-06, + "loss": 2.2042, + "step": 652 + }, + { + "epoch": 0.7156164383561644, + "grad_norm": 1.8776047229766846, + "learning_rate": 4.2587327376116984e-06, + "loss": 2.1512, + "step": 653 + }, + { + "epoch": 0.7167123287671233, + "grad_norm": 1.8949921131134033, + "learning_rate": 4.256701868399675e-06, + "loss": 2.1895, + "step": 654 + }, + { + "epoch": 0.7178082191780822, + "grad_norm": 1.8172632455825806, + "learning_rate": 4.254670999187653e-06, + "loss": 2.2, + "step": 655 + }, + { + "epoch": 0.718904109589041, + "grad_norm": 2.105499505996704, + "learning_rate": 4.252640129975629e-06, + "loss": 2.202, + "step": 656 + }, + { + "epoch": 0.72, + "grad_norm": 1.9515924453735352, + "learning_rate": 4.2506092607636075e-06, + "loss": 2.2116, + "step": 657 + }, + { + "epoch": 0.7210958904109589, + "grad_norm": 1.761770248413086, + "learning_rate": 4.248578391551584e-06, + "loss": 2.1921, + "step": 658 + }, + { + "epoch": 0.7221917808219178, + "grad_norm": 1.9848097562789917, + "learning_rate": 4.246547522339562e-06, + "loss": 2.1553, + "step": 659 + }, + { + "epoch": 0.7232876712328767, + "grad_norm": 2.041987180709839, + "learning_rate": 4.2445166531275385e-06, + "loss": 2.1975, + "step": 660 + }, + { + "epoch": 0.7232876712328767, + "eval_loss": 2.325580358505249, + "eval_runtime": 43.6207, + "eval_samples_per_second": 22.925, + "eval_steps_per_second": 0.963, + "step": 660 + }, + { + "epoch": 0.7243835616438357, + "grad_norm": 1.9755775928497314, + "learning_rate": 4.242485783915517e-06, + "loss": 2.1955, + "step": 661 + }, + { + "epoch": 0.7254794520547945, + "grad_norm": 1.9621502161026, + "learning_rate": 4.240454914703493e-06, + "loss": 2.1554, + "step": 662 + }, + { + "epoch": 0.7265753424657534, + "grad_norm": 1.848013162612915, + "learning_rate": 4.23842404549147e-06, + "loss": 2.16, + "step": 663 + }, + { + "epoch": 0.7276712328767123, + "grad_norm": 1.7370128631591797, + "learning_rate": 4.236393176279448e-06, + "loss": 2.1696, + "step": 664 + }, + { + "epoch": 0.7287671232876712, + "grad_norm": 2.0708158016204834, + "learning_rate": 4.234362307067425e-06, + "loss": 2.1592, + "step": 665 + }, + { + "epoch": 0.7298630136986302, + "grad_norm": 1.9864391088485718, + "learning_rate": 4.232331437855402e-06, + "loss": 2.2121, + "step": 666 + }, + { + "epoch": 0.7309589041095891, + "grad_norm": 1.9988250732421875, + "learning_rate": 4.2303005686433795e-06, + "loss": 2.2306, + "step": 667 + }, + { + "epoch": 0.7320547945205479, + "grad_norm": 1.89108407497406, + "learning_rate": 4.228269699431357e-06, + "loss": 2.222, + "step": 668 + }, + { + "epoch": 0.7331506849315068, + "grad_norm": 1.7941503524780273, + "learning_rate": 4.226238830219334e-06, + "loss": 2.1975, + "step": 669 + }, + { + "epoch": 0.7342465753424657, + "grad_norm": 1.8058613538742065, + "learning_rate": 4.224207961007311e-06, + "loss": 2.2037, + "step": 670 + }, + { + "epoch": 0.7342465753424657, + "eval_loss": 2.3266475200653076, + "eval_runtime": 47.6623, + "eval_samples_per_second": 20.981, + "eval_steps_per_second": 0.881, + "step": 670 + }, + { + "epoch": 0.7353424657534247, + "grad_norm": 1.764448642730713, + "learning_rate": 4.2221770917952886e-06, + "loss": 2.1904, + "step": 671 + }, + { + "epoch": 0.7364383561643836, + "grad_norm": 1.8252112865447998, + "learning_rate": 4.220146222583266e-06, + "loss": 2.1959, + "step": 672 + }, + { + "epoch": 0.7375342465753425, + "grad_norm": 2.02898907661438, + "learning_rate": 4.218115353371243e-06, + "loss": 2.1937, + "step": 673 + }, + { + "epoch": 0.7386301369863014, + "grad_norm": 1.8766248226165771, + "learning_rate": 4.21608448415922e-06, + "loss": 2.195, + "step": 674 + }, + { + "epoch": 0.7397260273972602, + "grad_norm": 2.2350332736968994, + "learning_rate": 4.214053614947198e-06, + "loss": 2.1798, + "step": 675 + }, + { + "epoch": 0.7408219178082192, + "grad_norm": 1.820399522781372, + "learning_rate": 4.212022745735175e-06, + "loss": 2.2371, + "step": 676 + }, + { + "epoch": 0.7419178082191781, + "grad_norm": 2.3411312103271484, + "learning_rate": 4.209991876523152e-06, + "loss": 2.1603, + "step": 677 + }, + { + "epoch": 0.743013698630137, + "grad_norm": 1.8562874794006348, + "learning_rate": 4.2079610073111295e-06, + "loss": 2.2082, + "step": 678 + }, + { + "epoch": 0.7441095890410959, + "grad_norm": 2.274214506149292, + "learning_rate": 4.205930138099107e-06, + "loss": 2.1716, + "step": 679 + }, + { + "epoch": 0.7452054794520548, + "grad_norm": 1.9173530340194702, + "learning_rate": 4.203899268887084e-06, + "loss": 2.2461, + "step": 680 + }, + { + "epoch": 0.7452054794520548, + "eval_loss": 2.3253841400146484, + "eval_runtime": 46.5732, + "eval_samples_per_second": 21.472, + "eval_steps_per_second": 0.902, + "step": 680 + }, + { + "epoch": 0.7463013698630137, + "grad_norm": 1.9732033014297485, + "learning_rate": 4.201868399675061e-06, + "loss": 2.2016, + "step": 681 + }, + { + "epoch": 0.7473972602739726, + "grad_norm": 1.8850247859954834, + "learning_rate": 4.199837530463039e-06, + "loss": 2.223, + "step": 682 + }, + { + "epoch": 0.7484931506849315, + "grad_norm": 1.8936313390731812, + "learning_rate": 4.197806661251016e-06, + "loss": 2.1381, + "step": 683 + }, + { + "epoch": 0.7495890410958904, + "grad_norm": 2.11995530128479, + "learning_rate": 4.195775792038993e-06, + "loss": 2.1962, + "step": 684 + }, + { + "epoch": 0.7506849315068493, + "grad_norm": 1.8207820653915405, + "learning_rate": 4.19374492282697e-06, + "loss": 2.1835, + "step": 685 + }, + { + "epoch": 0.7517808219178083, + "grad_norm": 2.06030535697937, + "learning_rate": 4.191714053614948e-06, + "loss": 2.1347, + "step": 686 + }, + { + "epoch": 0.7528767123287671, + "grad_norm": 2.048990488052368, + "learning_rate": 4.189683184402924e-06, + "loss": 2.2189, + "step": 687 + }, + { + "epoch": 0.753972602739726, + "grad_norm": 1.976369023323059, + "learning_rate": 4.187652315190902e-06, + "loss": 2.1805, + "step": 688 + }, + { + "epoch": 0.7550684931506849, + "grad_norm": 2.1160926818847656, + "learning_rate": 4.185621445978879e-06, + "loss": 2.2322, + "step": 689 + }, + { + "epoch": 0.7561643835616438, + "grad_norm": 1.9444516897201538, + "learning_rate": 4.183590576766857e-06, + "loss": 2.1696, + "step": 690 + }, + { + "epoch": 0.7561643835616438, + "eval_loss": 2.3242011070251465, + "eval_runtime": 48.7528, + "eval_samples_per_second": 20.512, + "eval_steps_per_second": 0.861, + "step": 690 + }, + { + "epoch": 0.7572602739726028, + "grad_norm": 1.9567183256149292, + "learning_rate": 4.181559707554833e-06, + "loss": 2.1891, + "step": 691 + }, + { + "epoch": 0.7583561643835617, + "grad_norm": 1.9185020923614502, + "learning_rate": 4.179528838342811e-06, + "loss": 2.2318, + "step": 692 + }, + { + "epoch": 0.7594520547945205, + "grad_norm": 1.833487629890442, + "learning_rate": 4.177497969130788e-06, + "loss": 2.1705, + "step": 693 + }, + { + "epoch": 0.7605479452054794, + "grad_norm": 2.2949867248535156, + "learning_rate": 4.175467099918766e-06, + "loss": 2.1712, + "step": 694 + }, + { + "epoch": 0.7616438356164383, + "grad_norm": 1.8663084506988525, + "learning_rate": 4.173436230706742e-06, + "loss": 2.2262, + "step": 695 + }, + { + "epoch": 0.7627397260273973, + "grad_norm": 2.4633052349090576, + "learning_rate": 4.1714053614947205e-06, + "loss": 2.185, + "step": 696 + }, + { + "epoch": 0.7638356164383562, + "grad_norm": 2.036524772644043, + "learning_rate": 4.169374492282697e-06, + "loss": 2.1697, + "step": 697 + }, + { + "epoch": 0.7649315068493151, + "grad_norm": 2.013810634613037, + "learning_rate": 4.167343623070675e-06, + "loss": 2.1826, + "step": 698 + }, + { + "epoch": 0.766027397260274, + "grad_norm": 2.214109182357788, + "learning_rate": 4.165312753858652e-06, + "loss": 2.2162, + "step": 699 + }, + { + "epoch": 0.7671232876712328, + "grad_norm": 1.8463964462280273, + "learning_rate": 4.163281884646629e-06, + "loss": 2.1865, + "step": 700 + }, + { + "epoch": 0.7671232876712328, + "eval_loss": 2.323822498321533, + "eval_runtime": 48.3307, + "eval_samples_per_second": 20.691, + "eval_steps_per_second": 0.869, + "step": 700 + }, + { + "epoch": 0.7682191780821918, + "grad_norm": 2.321021556854248, + "learning_rate": 4.161251015434607e-06, + "loss": 2.1758, + "step": 701 + }, + { + "epoch": 0.7693150684931507, + "grad_norm": 1.8213255405426025, + "learning_rate": 4.159220146222583e-06, + "loss": 2.2002, + "step": 702 + }, + { + "epoch": 0.7704109589041096, + "grad_norm": 1.8925111293792725, + "learning_rate": 4.157189277010561e-06, + "loss": 2.1575, + "step": 703 + }, + { + "epoch": 0.7715068493150685, + "grad_norm": 1.8050175905227661, + "learning_rate": 4.155158407798538e-06, + "loss": 2.179, + "step": 704 + }, + { + "epoch": 0.7726027397260274, + "grad_norm": 1.873228907585144, + "learning_rate": 4.153127538586516e-06, + "loss": 2.2232, + "step": 705 + }, + { + "epoch": 0.7736986301369863, + "grad_norm": 1.954910397529602, + "learning_rate": 4.151096669374492e-06, + "loss": 2.1992, + "step": 706 + }, + { + "epoch": 0.7747945205479452, + "grad_norm": 1.8913410902023315, + "learning_rate": 4.1490658001624705e-06, + "loss": 2.2086, + "step": 707 + }, + { + "epoch": 0.7758904109589041, + "grad_norm": 1.8883898258209229, + "learning_rate": 4.147034930950447e-06, + "loss": 2.1801, + "step": 708 + }, + { + "epoch": 0.776986301369863, + "grad_norm": 2.054351806640625, + "learning_rate": 4.145004061738425e-06, + "loss": 2.1908, + "step": 709 + }, + { + "epoch": 0.7780821917808219, + "grad_norm": 2.029036045074463, + "learning_rate": 4.1429731925264015e-06, + "loss": 2.1891, + "step": 710 + }, + { + "epoch": 0.7780821917808219, + "eval_loss": 2.3196680545806885, + "eval_runtime": 42.8326, + "eval_samples_per_second": 23.347, + "eval_steps_per_second": 0.981, + "step": 710 + }, + { + "epoch": 0.7791780821917809, + "grad_norm": 2.27848219871521, + "learning_rate": 4.140942323314379e-06, + "loss": 2.1165, + "step": 711 + }, + { + "epoch": 0.7802739726027397, + "grad_norm": 2.134625196456909, + "learning_rate": 4.138911454102356e-06, + "loss": 2.1508, + "step": 712 + }, + { + "epoch": 0.7813698630136986, + "grad_norm": 2.103895902633667, + "learning_rate": 4.136880584890333e-06, + "loss": 2.1746, + "step": 713 + }, + { + "epoch": 0.7824657534246575, + "grad_norm": 1.713189721107483, + "learning_rate": 4.1348497156783106e-06, + "loss": 2.2023, + "step": 714 + }, + { + "epoch": 0.7835616438356164, + "grad_norm": 2.0390899181365967, + "learning_rate": 4.132818846466288e-06, + "loss": 2.2039, + "step": 715 + }, + { + "epoch": 0.7846575342465754, + "grad_norm": 1.9643651247024536, + "learning_rate": 4.130787977254265e-06, + "loss": 2.199, + "step": 716 + }, + { + "epoch": 0.7857534246575343, + "grad_norm": 1.9173564910888672, + "learning_rate": 4.128757108042242e-06, + "loss": 2.2166, + "step": 717 + }, + { + "epoch": 0.7868493150684932, + "grad_norm": 1.7923007011413574, + "learning_rate": 4.12672623883022e-06, + "loss": 2.2144, + "step": 718 + }, + { + "epoch": 0.787945205479452, + "grad_norm": 2.2235090732574463, + "learning_rate": 4.124695369618197e-06, + "loss": 2.2586, + "step": 719 + }, + { + "epoch": 0.7890410958904109, + "grad_norm": 1.780316710472107, + "learning_rate": 4.122664500406174e-06, + "loss": 2.1862, + "step": 720 + }, + { + "epoch": 0.7890410958904109, + "eval_loss": 2.318016767501831, + "eval_runtime": 50.5082, + "eval_samples_per_second": 19.799, + "eval_steps_per_second": 0.832, + "step": 720 + }, + { + "epoch": 0.7901369863013699, + "grad_norm": 2.0788493156433105, + "learning_rate": 4.1206336311941515e-06, + "loss": 2.1845, + "step": 721 + }, + { + "epoch": 0.7912328767123288, + "grad_norm": 1.9053714275360107, + "learning_rate": 4.118602761982129e-06, + "loss": 2.2005, + "step": 722 + }, + { + "epoch": 0.7923287671232877, + "grad_norm": 1.9328513145446777, + "learning_rate": 4.116571892770106e-06, + "loss": 2.1475, + "step": 723 + }, + { + "epoch": 0.7934246575342466, + "grad_norm": 2.059760570526123, + "learning_rate": 4.114541023558083e-06, + "loss": 2.1386, + "step": 724 + }, + { + "epoch": 0.7945205479452054, + "grad_norm": 2.0093464851379395, + "learning_rate": 4.112510154346061e-06, + "loss": 2.2242, + "step": 725 + }, + { + "epoch": 0.7956164383561644, + "grad_norm": 1.8513909578323364, + "learning_rate": 4.110479285134038e-06, + "loss": 2.178, + "step": 726 + }, + { + "epoch": 0.7967123287671233, + "grad_norm": 1.9481958150863647, + "learning_rate": 4.108448415922015e-06, + "loss": 2.1961, + "step": 727 + }, + { + "epoch": 0.7978082191780822, + "grad_norm": 1.9229767322540283, + "learning_rate": 4.1064175467099924e-06, + "loss": 2.2116, + "step": 728 + }, + { + "epoch": 0.7989041095890411, + "grad_norm": 1.9819737672805786, + "learning_rate": 4.10438667749797e-06, + "loss": 2.2097, + "step": 729 + }, + { + "epoch": 0.8, + "grad_norm": 1.9382416009902954, + "learning_rate": 4.102355808285947e-06, + "loss": 2.1099, + "step": 730 + }, + { + "epoch": 0.8, + "eval_loss": 2.31795334815979, + "eval_runtime": 51.8961, + "eval_samples_per_second": 19.269, + "eval_steps_per_second": 0.809, + "step": 730 + }, + { + "epoch": 0.8010958904109589, + "grad_norm": 2.006890058517456, + "learning_rate": 4.100324939073924e-06, + "loss": 2.2606, + "step": 731 + }, + { + "epoch": 0.8021917808219178, + "grad_norm": 2.164473295211792, + "learning_rate": 4.0982940698619015e-06, + "loss": 2.1869, + "step": 732 + }, + { + "epoch": 0.8032876712328767, + "grad_norm": 2.04368257522583, + "learning_rate": 4.096263200649879e-06, + "loss": 2.1846, + "step": 733 + }, + { + "epoch": 0.8043835616438356, + "grad_norm": 1.8363912105560303, + "learning_rate": 4.094232331437856e-06, + "loss": 2.1938, + "step": 734 + }, + { + "epoch": 0.8054794520547945, + "grad_norm": 1.9260189533233643, + "learning_rate": 4.092201462225833e-06, + "loss": 2.1682, + "step": 735 + }, + { + "epoch": 0.8065753424657535, + "grad_norm": 2.0688912868499756, + "learning_rate": 4.090170593013811e-06, + "loss": 2.1772, + "step": 736 + }, + { + "epoch": 0.8076712328767123, + "grad_norm": 1.859474539756775, + "learning_rate": 4.088139723801787e-06, + "loss": 2.2013, + "step": 737 + }, + { + "epoch": 0.8087671232876712, + "grad_norm": 2.0915753841400146, + "learning_rate": 4.086108854589765e-06, + "loss": 2.1519, + "step": 738 + }, + { + "epoch": 0.8098630136986301, + "grad_norm": 2.2019519805908203, + "learning_rate": 4.084077985377742e-06, + "loss": 2.1608, + "step": 739 + }, + { + "epoch": 0.810958904109589, + "grad_norm": 1.8727316856384277, + "learning_rate": 4.08204711616572e-06, + "loss": 2.1823, + "step": 740 + }, + { + "epoch": 0.810958904109589, + "eval_loss": 2.318638324737549, + "eval_runtime": 52.3669, + "eval_samples_per_second": 19.096, + "eval_steps_per_second": 0.802, + "step": 740 + }, + { + "epoch": 0.812054794520548, + "grad_norm": 2.059250831604004, + "learning_rate": 4.080016246953696e-06, + "loss": 2.2285, + "step": 741 + }, + { + "epoch": 0.8131506849315069, + "grad_norm": 1.9466749429702759, + "learning_rate": 4.077985377741674e-06, + "loss": 2.1496, + "step": 742 + }, + { + "epoch": 0.8142465753424658, + "grad_norm": 1.8967524766921997, + "learning_rate": 4.075954508529651e-06, + "loss": 2.2066, + "step": 743 + }, + { + "epoch": 0.8153424657534246, + "grad_norm": 1.976780891418457, + "learning_rate": 4.073923639317629e-06, + "loss": 2.2194, + "step": 744 + }, + { + "epoch": 0.8164383561643835, + "grad_norm": 2.077683687210083, + "learning_rate": 4.071892770105605e-06, + "loss": 2.2209, + "step": 745 + }, + { + "epoch": 0.8175342465753425, + "grad_norm": 1.9075323343276978, + "learning_rate": 4.069861900893583e-06, + "loss": 2.1727, + "step": 746 + }, + { + "epoch": 0.8186301369863014, + "grad_norm": 1.6940832138061523, + "learning_rate": 4.06783103168156e-06, + "loss": 2.1437, + "step": 747 + }, + { + "epoch": 0.8197260273972603, + "grad_norm": 1.9035768508911133, + "learning_rate": 4.065800162469538e-06, + "loss": 2.165, + "step": 748 + }, + { + "epoch": 0.8208219178082192, + "grad_norm": 1.8775745630264282, + "learning_rate": 4.063769293257514e-06, + "loss": 2.1757, + "step": 749 + }, + { + "epoch": 0.821917808219178, + "grad_norm": 1.9622418880462646, + "learning_rate": 4.061738424045492e-06, + "loss": 2.197, + "step": 750 + }, + { + "epoch": 0.821917808219178, + "eval_loss": 2.317519426345825, + "eval_runtime": 50.2699, + "eval_samples_per_second": 19.893, + "eval_steps_per_second": 0.835, + "step": 750 + }, + { + "epoch": 0.823013698630137, + "grad_norm": 1.9646214246749878, + "learning_rate": 4.059707554833469e-06, + "loss": 2.1783, + "step": 751 + }, + { + "epoch": 0.8241095890410959, + "grad_norm": 1.890342354774475, + "learning_rate": 4.057676685621446e-06, + "loss": 2.1925, + "step": 752 + }, + { + "epoch": 0.8252054794520548, + "grad_norm": 1.8977251052856445, + "learning_rate": 4.0556458164094235e-06, + "loss": 2.219, + "step": 753 + }, + { + "epoch": 0.8263013698630137, + "grad_norm": 1.8251665830612183, + "learning_rate": 4.053614947197401e-06, + "loss": 2.189, + "step": 754 + }, + { + "epoch": 0.8273972602739726, + "grad_norm": 1.7953882217407227, + "learning_rate": 4.051584077985378e-06, + "loss": 2.1237, + "step": 755 + }, + { + "epoch": 0.8284931506849315, + "grad_norm": 1.8247836828231812, + "learning_rate": 4.049553208773355e-06, + "loss": 2.1962, + "step": 756 + }, + { + "epoch": 0.8295890410958904, + "grad_norm": 1.8537967205047607, + "learning_rate": 4.047522339561333e-06, + "loss": 2.2201, + "step": 757 + }, + { + "epoch": 0.8306849315068493, + "grad_norm": 1.865315318107605, + "learning_rate": 4.04549147034931e-06, + "loss": 2.1819, + "step": 758 + }, + { + "epoch": 0.8317808219178082, + "grad_norm": 1.8620522022247314, + "learning_rate": 4.043460601137287e-06, + "loss": 2.1717, + "step": 759 + }, + { + "epoch": 0.8328767123287671, + "grad_norm": 1.9475384950637817, + "learning_rate": 4.041429731925264e-06, + "loss": 2.2718, + "step": 760 + }, + { + "epoch": 0.8328767123287671, + "eval_loss": 2.314429759979248, + "eval_runtime": 54.2324, + "eval_samples_per_second": 18.439, + "eval_steps_per_second": 0.774, + "step": 760 + }, + { + "epoch": 0.8339726027397261, + "grad_norm": 1.7170753479003906, + "learning_rate": 4.039398862713242e-06, + "loss": 2.1785, + "step": 761 + }, + { + "epoch": 0.8350684931506849, + "grad_norm": 1.968220591545105, + "learning_rate": 4.037367993501219e-06, + "loss": 2.1798, + "step": 762 + }, + { + "epoch": 0.8361643835616438, + "grad_norm": 1.7927111387252808, + "learning_rate": 4.035337124289196e-06, + "loss": 2.2372, + "step": 763 + }, + { + "epoch": 0.8372602739726027, + "grad_norm": 1.7871185541152954, + "learning_rate": 4.0333062550771735e-06, + "loss": 2.182, + "step": 764 + }, + { + "epoch": 0.8383561643835616, + "grad_norm": 1.9415812492370605, + "learning_rate": 4.031275385865151e-06, + "loss": 2.1527, + "step": 765 + }, + { + "epoch": 0.8394520547945206, + "grad_norm": 1.7327690124511719, + "learning_rate": 4.029244516653128e-06, + "loss": 2.1744, + "step": 766 + }, + { + "epoch": 0.8405479452054795, + "grad_norm": 1.9123411178588867, + "learning_rate": 4.027213647441105e-06, + "loss": 2.2049, + "step": 767 + }, + { + "epoch": 0.8416438356164384, + "grad_norm": 2.011591911315918, + "learning_rate": 4.025182778229083e-06, + "loss": 2.1391, + "step": 768 + }, + { + "epoch": 0.8427397260273972, + "grad_norm": 1.9169542789459229, + "learning_rate": 4.02315190901706e-06, + "loss": 2.1562, + "step": 769 + }, + { + "epoch": 0.8438356164383561, + "grad_norm": 1.8020051717758179, + "learning_rate": 4.021121039805037e-06, + "loss": 2.1795, + "step": 770 + }, + { + "epoch": 0.8438356164383561, + "eval_loss": 2.3199150562286377, + "eval_runtime": 52.468, + "eval_samples_per_second": 19.059, + "eval_steps_per_second": 0.8, + "step": 770 + }, + { + "epoch": 0.8449315068493151, + "grad_norm": 1.927666425704956, + "learning_rate": 4.0190901705930144e-06, + "loss": 2.1834, + "step": 771 + }, + { + "epoch": 0.846027397260274, + "grad_norm": 2.0998382568359375, + "learning_rate": 4.017059301380992e-06, + "loss": 2.2541, + "step": 772 + }, + { + "epoch": 0.8471232876712329, + "grad_norm": 1.88541579246521, + "learning_rate": 4.015028432168969e-06, + "loss": 2.1631, + "step": 773 + }, + { + "epoch": 0.8482191780821918, + "grad_norm": 1.9546253681182861, + "learning_rate": 4.012997562956945e-06, + "loss": 2.2266, + "step": 774 + }, + { + "epoch": 0.8493150684931506, + "grad_norm": 1.7969225645065308, + "learning_rate": 4.0109666937449235e-06, + "loss": 2.1526, + "step": 775 + }, + { + "epoch": 0.8504109589041096, + "grad_norm": 1.7237955331802368, + "learning_rate": 4.0089358245329e-06, + "loss": 2.1618, + "step": 776 + }, + { + "epoch": 0.8515068493150685, + "grad_norm": 1.8355857133865356, + "learning_rate": 4.006904955320878e-06, + "loss": 2.1673, + "step": 777 + }, + { + "epoch": 0.8526027397260274, + "grad_norm": 1.8049591779708862, + "learning_rate": 4.0048740861088545e-06, + "loss": 2.1676, + "step": 778 + }, + { + "epoch": 0.8536986301369863, + "grad_norm": 1.8143601417541504, + "learning_rate": 4.002843216896833e-06, + "loss": 2.214, + "step": 779 + }, + { + "epoch": 0.8547945205479452, + "grad_norm": 1.8344221115112305, + "learning_rate": 4.000812347684809e-06, + "loss": 2.1958, + "step": 780 + }, + { + "epoch": 0.8547945205479452, + "eval_loss": 2.314359426498413, + "eval_runtime": 51.8717, + "eval_samples_per_second": 19.278, + "eval_steps_per_second": 0.81, + "step": 780 + }, + { + "epoch": 0.8558904109589041, + "grad_norm": 1.944402813911438, + "learning_rate": 3.998781478472787e-06, + "loss": 2.1497, + "step": 781 + }, + { + "epoch": 0.856986301369863, + "grad_norm": 1.93898606300354, + "learning_rate": 3.996750609260764e-06, + "loss": 2.1737, + "step": 782 + }, + { + "epoch": 0.8580821917808219, + "grad_norm": 1.9499506950378418, + "learning_rate": 3.994719740048742e-06, + "loss": 2.1545, + "step": 783 + }, + { + "epoch": 0.8591780821917808, + "grad_norm": 1.9311052560806274, + "learning_rate": 3.992688870836718e-06, + "loss": 2.2559, + "step": 784 + }, + { + "epoch": 0.8602739726027397, + "grad_norm": 1.864862084388733, + "learning_rate": 3.990658001624696e-06, + "loss": 2.1996, + "step": 785 + }, + { + "epoch": 0.8613698630136987, + "grad_norm": 1.8248569965362549, + "learning_rate": 3.988627132412673e-06, + "loss": 2.1591, + "step": 786 + }, + { + "epoch": 0.8624657534246576, + "grad_norm": 2.2370970249176025, + "learning_rate": 3.98659626320065e-06, + "loss": 2.2095, + "step": 787 + }, + { + "epoch": 0.8635616438356164, + "grad_norm": 1.8877084255218506, + "learning_rate": 3.984565393988627e-06, + "loss": 2.1813, + "step": 788 + }, + { + "epoch": 0.8646575342465753, + "grad_norm": 2.1722517013549805, + "learning_rate": 3.9825345247766046e-06, + "loss": 2.1866, + "step": 789 + }, + { + "epoch": 0.8657534246575342, + "grad_norm": 1.9640655517578125, + "learning_rate": 3.980503655564582e-06, + "loss": 2.1818, + "step": 790 + }, + { + "epoch": 0.8657534246575342, + "eval_loss": 2.314087390899658, + "eval_runtime": 51.9778, + "eval_samples_per_second": 19.239, + "eval_steps_per_second": 0.808, + "step": 790 + }, + { + "epoch": 0.8668493150684932, + "grad_norm": 2.0583789348602295, + "learning_rate": 3.978472786352559e-06, + "loss": 2.1956, + "step": 791 + }, + { + "epoch": 0.8679452054794521, + "grad_norm": 2.2128682136535645, + "learning_rate": 3.976441917140536e-06, + "loss": 2.204, + "step": 792 + }, + { + "epoch": 0.869041095890411, + "grad_norm": 1.917763113975525, + "learning_rate": 3.974411047928514e-06, + "loss": 2.1841, + "step": 793 + }, + { + "epoch": 0.8701369863013698, + "grad_norm": 2.0327019691467285, + "learning_rate": 3.972380178716491e-06, + "loss": 2.2351, + "step": 794 + }, + { + "epoch": 0.8712328767123287, + "grad_norm": 1.9498991966247559, + "learning_rate": 3.970349309504468e-06, + "loss": 2.2074, + "step": 795 + }, + { + "epoch": 0.8723287671232877, + "grad_norm": 1.8105775117874146, + "learning_rate": 3.9683184402924455e-06, + "loss": 2.2088, + "step": 796 + }, + { + "epoch": 0.8734246575342466, + "grad_norm": 1.9299930334091187, + "learning_rate": 3.966287571080423e-06, + "loss": 2.1469, + "step": 797 + }, + { + "epoch": 0.8745205479452055, + "grad_norm": 1.9346121549606323, + "learning_rate": 3.9642567018684e-06, + "loss": 2.1434, + "step": 798 + }, + { + "epoch": 0.8756164383561644, + "grad_norm": 1.8202977180480957, + "learning_rate": 3.962225832656377e-06, + "loss": 2.194, + "step": 799 + }, + { + "epoch": 0.8767123287671232, + "grad_norm": 1.6922450065612793, + "learning_rate": 3.960194963444355e-06, + "loss": 2.1821, + "step": 800 + }, + { + "epoch": 0.8767123287671232, + "eval_loss": 2.310209035873413, + "eval_runtime": 46.1102, + "eval_samples_per_second": 21.687, + "eval_steps_per_second": 0.911, + "step": 800 + }, + { + "epoch": 0.8778082191780822, + "grad_norm": 1.9258904457092285, + "learning_rate": 3.958164094232332e-06, + "loss": 2.1326, + "step": 801 + }, + { + "epoch": 0.8789041095890411, + "grad_norm": 1.8485561609268188, + "learning_rate": 3.956133225020309e-06, + "loss": 2.1888, + "step": 802 + }, + { + "epoch": 0.88, + "grad_norm": 1.9802418947219849, + "learning_rate": 3.954102355808286e-06, + "loss": 2.1876, + "step": 803 + }, + { + "epoch": 0.8810958904109589, + "grad_norm": 1.8365259170532227, + "learning_rate": 3.952071486596264e-06, + "loss": 2.1804, + "step": 804 + }, + { + "epoch": 0.8821917808219178, + "grad_norm": 2.023909568786621, + "learning_rate": 3.950040617384241e-06, + "loss": 2.1801, + "step": 805 + }, + { + "epoch": 0.8832876712328767, + "grad_norm": 1.8889018297195435, + "learning_rate": 3.948009748172218e-06, + "loss": 2.1657, + "step": 806 + }, + { + "epoch": 0.8843835616438356, + "grad_norm": 1.7983578443527222, + "learning_rate": 3.9459788789601955e-06, + "loss": 2.1589, + "step": 807 + }, + { + "epoch": 0.8854794520547945, + "grad_norm": 2.157043218612671, + "learning_rate": 3.943948009748173e-06, + "loss": 2.2258, + "step": 808 + }, + { + "epoch": 0.8865753424657534, + "grad_norm": 1.7387373447418213, + "learning_rate": 3.94191714053615e-06, + "loss": 2.1878, + "step": 809 + }, + { + "epoch": 0.8876712328767123, + "grad_norm": 1.963233232498169, + "learning_rate": 3.939886271324127e-06, + "loss": 2.186, + "step": 810 + }, + { + "epoch": 0.8876712328767123, + "eval_loss": 2.306187152862549, + "eval_runtime": 42.563, + "eval_samples_per_second": 23.495, + "eval_steps_per_second": 0.987, + "step": 810 + }, + { + "epoch": 0.8887671232876713, + "grad_norm": 1.8932961225509644, + "learning_rate": 3.937855402112104e-06, + "loss": 2.198, + "step": 811 + }, + { + "epoch": 0.8898630136986302, + "grad_norm": 2.0431783199310303, + "learning_rate": 3.935824532900082e-06, + "loss": 2.1505, + "step": 812 + }, + { + "epoch": 0.890958904109589, + "grad_norm": 1.8954466581344604, + "learning_rate": 3.933793663688058e-06, + "loss": 2.1494, + "step": 813 + }, + { + "epoch": 0.8920547945205479, + "grad_norm": 1.841184139251709, + "learning_rate": 3.9317627944760365e-06, + "loss": 2.1932, + "step": 814 + }, + { + "epoch": 0.8931506849315068, + "grad_norm": 1.9083219766616821, + "learning_rate": 3.929731925264013e-06, + "loss": 2.2081, + "step": 815 + }, + { + "epoch": 0.8942465753424658, + "grad_norm": 1.9737226963043213, + "learning_rate": 3.927701056051991e-06, + "loss": 2.1405, + "step": 816 + }, + { + "epoch": 0.8953424657534247, + "grad_norm": 1.869194746017456, + "learning_rate": 3.9256701868399674e-06, + "loss": 2.1642, + "step": 817 + }, + { + "epoch": 0.8964383561643836, + "grad_norm": 1.9553205966949463, + "learning_rate": 3.9236393176279456e-06, + "loss": 2.1761, + "step": 818 + }, + { + "epoch": 0.8975342465753424, + "grad_norm": 1.8355534076690674, + "learning_rate": 3.921608448415922e-06, + "loss": 2.1551, + "step": 819 + }, + { + "epoch": 0.8986301369863013, + "grad_norm": 1.9770171642303467, + "learning_rate": 3.9195775792039e-06, + "loss": 2.1424, + "step": 820 + }, + { + "epoch": 0.8986301369863013, + "eval_loss": 2.307065010070801, + "eval_runtime": 44.5816, + "eval_samples_per_second": 22.431, + "eval_steps_per_second": 0.942, + "step": 820 + }, + { + "epoch": 0.8997260273972603, + "grad_norm": 1.7375415563583374, + "learning_rate": 3.9175467099918765e-06, + "loss": 2.1895, + "step": 821 + }, + { + "epoch": 0.9008219178082192, + "grad_norm": 1.8910484313964844, + "learning_rate": 3.915515840779855e-06, + "loss": 2.1685, + "step": 822 + }, + { + "epoch": 0.9019178082191781, + "grad_norm": 1.9191482067108154, + "learning_rate": 3.913484971567831e-06, + "loss": 2.2074, + "step": 823 + }, + { + "epoch": 0.903013698630137, + "grad_norm": 1.7864760160446167, + "learning_rate": 3.911454102355808e-06, + "loss": 2.1972, + "step": 824 + }, + { + "epoch": 0.9041095890410958, + "grad_norm": 1.9044650793075562, + "learning_rate": 3.909423233143786e-06, + "loss": 2.1237, + "step": 825 + }, + { + "epoch": 0.9052054794520548, + "grad_norm": 1.8378455638885498, + "learning_rate": 3.907392363931763e-06, + "loss": 2.1907, + "step": 826 + }, + { + "epoch": 0.9063013698630137, + "grad_norm": 1.8708094358444214, + "learning_rate": 3.90536149471974e-06, + "loss": 2.1883, + "step": 827 + }, + { + "epoch": 0.9073972602739726, + "grad_norm": 1.7802201509475708, + "learning_rate": 3.9033306255077175e-06, + "loss": 2.2356, + "step": 828 + }, + { + "epoch": 0.9084931506849315, + "grad_norm": 1.9374496936798096, + "learning_rate": 3.901299756295695e-06, + "loss": 2.2546, + "step": 829 + }, + { + "epoch": 0.9095890410958904, + "grad_norm": 1.9246270656585693, + "learning_rate": 3.899268887083672e-06, + "loss": 2.2684, + "step": 830 + }, + { + "epoch": 0.9095890410958904, + "eval_loss": 2.3065874576568604, + "eval_runtime": 45.18, + "eval_samples_per_second": 22.134, + "eval_steps_per_second": 0.93, + "step": 830 + }, + { + "epoch": 0.9106849315068493, + "grad_norm": 1.8681883811950684, + "learning_rate": 3.897238017871649e-06, + "loss": 2.1737, + "step": 831 + }, + { + "epoch": 0.9117808219178082, + "grad_norm": 1.9293930530548096, + "learning_rate": 3.8952071486596266e-06, + "loss": 2.1813, + "step": 832 + }, + { + "epoch": 0.9128767123287671, + "grad_norm": 1.844586968421936, + "learning_rate": 3.893176279447604e-06, + "loss": 2.1788, + "step": 833 + }, + { + "epoch": 0.913972602739726, + "grad_norm": 1.7658660411834717, + "learning_rate": 3.891145410235581e-06, + "loss": 2.1494, + "step": 834 + }, + { + "epoch": 0.915068493150685, + "grad_norm": 2.05153751373291, + "learning_rate": 3.889114541023558e-06, + "loss": 2.241, + "step": 835 + }, + { + "epoch": 0.9161643835616439, + "grad_norm": 1.7869952917099, + "learning_rate": 3.887083671811536e-06, + "loss": 2.1866, + "step": 836 + }, + { + "epoch": 0.9172602739726028, + "grad_norm": 1.99534273147583, + "learning_rate": 3.885052802599513e-06, + "loss": 2.1682, + "step": 837 + }, + { + "epoch": 0.9183561643835616, + "grad_norm": 1.9593979120254517, + "learning_rate": 3.88302193338749e-06, + "loss": 2.1357, + "step": 838 + }, + { + "epoch": 0.9194520547945205, + "grad_norm": 1.8441778421401978, + "learning_rate": 3.8809910641754675e-06, + "loss": 2.2283, + "step": 839 + }, + { + "epoch": 0.9205479452054794, + "grad_norm": 1.9153512716293335, + "learning_rate": 3.878960194963445e-06, + "loss": 2.1899, + "step": 840 + }, + { + "epoch": 0.9205479452054794, + "eval_loss": 2.307375907897949, + "eval_runtime": 48.0216, + "eval_samples_per_second": 20.824, + "eval_steps_per_second": 0.875, + "step": 840 + }, + { + "epoch": 0.9216438356164384, + "grad_norm": 2.004232883453369, + "learning_rate": 3.876929325751422e-06, + "loss": 2.173, + "step": 841 + }, + { + "epoch": 0.9227397260273973, + "grad_norm": 2.12992262840271, + "learning_rate": 3.874898456539399e-06, + "loss": 2.2164, + "step": 842 + }, + { + "epoch": 0.9238356164383562, + "grad_norm": 2.0505411624908447, + "learning_rate": 3.872867587327377e-06, + "loss": 2.1432, + "step": 843 + }, + { + "epoch": 0.924931506849315, + "grad_norm": 1.767324686050415, + "learning_rate": 3.870836718115354e-06, + "loss": 2.1838, + "step": 844 + }, + { + "epoch": 0.9260273972602739, + "grad_norm": 2.189107894897461, + "learning_rate": 3.868805848903331e-06, + "loss": 2.1623, + "step": 845 + }, + { + "epoch": 0.9271232876712329, + "grad_norm": 1.9781970977783203, + "learning_rate": 3.8667749796913084e-06, + "loss": 2.1738, + "step": 846 + }, + { + "epoch": 0.9282191780821918, + "grad_norm": 1.8734674453735352, + "learning_rate": 3.864744110479286e-06, + "loss": 2.1829, + "step": 847 + }, + { + "epoch": 0.9293150684931507, + "grad_norm": 1.7699538469314575, + "learning_rate": 3.862713241267262e-06, + "loss": 2.186, + "step": 848 + }, + { + "epoch": 0.9304109589041096, + "grad_norm": 1.8354437351226807, + "learning_rate": 3.86068237205524e-06, + "loss": 2.2149, + "step": 849 + }, + { + "epoch": 0.9315068493150684, + "grad_norm": 1.913249135017395, + "learning_rate": 3.858651502843217e-06, + "loss": 2.2208, + "step": 850 + }, + { + "epoch": 0.9315068493150684, + "eval_loss": 2.307549238204956, + "eval_runtime": 43.1706, + "eval_samples_per_second": 23.164, + "eval_steps_per_second": 0.973, + "step": 850 + }, + { + "epoch": 0.9326027397260274, + "grad_norm": 1.92095148563385, + "learning_rate": 3.856620633631195e-06, + "loss": 2.1839, + "step": 851 + }, + { + "epoch": 0.9336986301369863, + "grad_norm": 1.8833959102630615, + "learning_rate": 3.854589764419171e-06, + "loss": 2.1372, + "step": 852 + }, + { + "epoch": 0.9347945205479452, + "grad_norm": 1.9410737752914429, + "learning_rate": 3.852558895207149e-06, + "loss": 2.1726, + "step": 853 + }, + { + "epoch": 0.9358904109589041, + "grad_norm": 1.8328828811645508, + "learning_rate": 3.850528025995126e-06, + "loss": 2.2425, + "step": 854 + }, + { + "epoch": 0.936986301369863, + "grad_norm": 1.89652419090271, + "learning_rate": 3.848497156783104e-06, + "loss": 2.1956, + "step": 855 + }, + { + "epoch": 0.938082191780822, + "grad_norm": 1.9026265144348145, + "learning_rate": 3.84646628757108e-06, + "loss": 2.167, + "step": 856 + }, + { + "epoch": 0.9391780821917808, + "grad_norm": 2.010864734649658, + "learning_rate": 3.8444354183590585e-06, + "loss": 2.1725, + "step": 857 + }, + { + "epoch": 0.9402739726027397, + "grad_norm": 2.293717384338379, + "learning_rate": 3.842404549147035e-06, + "loss": 2.1668, + "step": 858 + }, + { + "epoch": 0.9413698630136986, + "grad_norm": 1.8209116458892822, + "learning_rate": 3.840373679935013e-06, + "loss": 2.1649, + "step": 859 + }, + { + "epoch": 0.9424657534246575, + "grad_norm": 2.061277389526367, + "learning_rate": 3.8383428107229894e-06, + "loss": 2.1479, + "step": 860 + }, + { + "epoch": 0.9424657534246575, + "eval_loss": 2.3073670864105225, + "eval_runtime": 44.3781, + "eval_samples_per_second": 22.534, + "eval_steps_per_second": 0.946, + "step": 860 + }, + { + "epoch": 0.9435616438356165, + "grad_norm": 1.8032523393630981, + "learning_rate": 3.836311941510967e-06, + "loss": 2.1665, + "step": 861 + }, + { + "epoch": 0.9446575342465754, + "grad_norm": 1.989533543586731, + "learning_rate": 3.834281072298944e-06, + "loss": 2.1612, + "step": 862 + }, + { + "epoch": 0.9457534246575342, + "grad_norm": 2.017179012298584, + "learning_rate": 3.832250203086921e-06, + "loss": 2.201, + "step": 863 + }, + { + "epoch": 0.9468493150684931, + "grad_norm": 2.0291004180908203, + "learning_rate": 3.8302193338748985e-06, + "loss": 2.1508, + "step": 864 + }, + { + "epoch": 0.947945205479452, + "grad_norm": 1.8693255186080933, + "learning_rate": 3.828188464662876e-06, + "loss": 2.2106, + "step": 865 + }, + { + "epoch": 0.949041095890411, + "grad_norm": 1.8890169858932495, + "learning_rate": 3.826157595450853e-06, + "loss": 2.2467, + "step": 866 + }, + { + "epoch": 0.9501369863013699, + "grad_norm": 2.0550642013549805, + "learning_rate": 3.82412672623883e-06, + "loss": 2.1899, + "step": 867 + }, + { + "epoch": 0.9512328767123288, + "grad_norm": 1.8681174516677856, + "learning_rate": 3.822095857026808e-06, + "loss": 2.1428, + "step": 868 + }, + { + "epoch": 0.9523287671232876, + "grad_norm": 1.9133951663970947, + "learning_rate": 3.820064987814785e-06, + "loss": 2.1947, + "step": 869 + }, + { + "epoch": 0.9534246575342465, + "grad_norm": 1.8717621564865112, + "learning_rate": 3.818034118602762e-06, + "loss": 2.1542, + "step": 870 + }, + { + "epoch": 0.9534246575342465, + "eval_loss": 2.307584047317505, + "eval_runtime": 43.718, + "eval_samples_per_second": 22.874, + "eval_steps_per_second": 0.961, + "step": 870 + }, + { + "epoch": 0.9545205479452055, + "grad_norm": 1.9843461513519287, + "learning_rate": 3.8160032493907395e-06, + "loss": 2.1547, + "step": 871 + }, + { + "epoch": 0.9556164383561644, + "grad_norm": 1.725468397140503, + "learning_rate": 3.813972380178717e-06, + "loss": 2.1679, + "step": 872 + }, + { + "epoch": 0.9567123287671233, + "grad_norm": 1.8076086044311523, + "learning_rate": 3.811941510966694e-06, + "loss": 2.1654, + "step": 873 + }, + { + "epoch": 0.9578082191780822, + "grad_norm": 1.8879774808883667, + "learning_rate": 3.809910641754671e-06, + "loss": 2.191, + "step": 874 + }, + { + "epoch": 0.958904109589041, + "grad_norm": 1.943313717842102, + "learning_rate": 3.8078797725426486e-06, + "loss": 2.1448, + "step": 875 + }, + { + "epoch": 0.96, + "grad_norm": 1.8877806663513184, + "learning_rate": 3.8058489033306254e-06, + "loss": 2.1129, + "step": 876 + }, + { + "epoch": 0.9610958904109589, + "grad_norm": 1.787083625793457, + "learning_rate": 3.803818034118603e-06, + "loss": 2.1854, + "step": 877 + }, + { + "epoch": 0.9621917808219178, + "grad_norm": 2.0925140380859375, + "learning_rate": 3.80178716490658e-06, + "loss": 2.1614, + "step": 878 + }, + { + "epoch": 0.9632876712328767, + "grad_norm": 1.9043725728988647, + "learning_rate": 3.7997562956945577e-06, + "loss": 2.2161, + "step": 879 + }, + { + "epoch": 0.9643835616438357, + "grad_norm": 1.8173545598983765, + "learning_rate": 3.7977254264825345e-06, + "loss": 2.2067, + "step": 880 + }, + { + "epoch": 0.9643835616438357, + "eval_loss": 2.302727460861206, + "eval_runtime": 44.2015, + "eval_samples_per_second": 22.624, + "eval_steps_per_second": 0.95, + "step": 880 + }, + { + "epoch": 0.9654794520547946, + "grad_norm": 1.9871562719345093, + "learning_rate": 3.7956945572705122e-06, + "loss": 2.1658, + "step": 881 + }, + { + "epoch": 0.9665753424657534, + "grad_norm": 2.103553295135498, + "learning_rate": 3.793663688058489e-06, + "loss": 2.1468, + "step": 882 + }, + { + "epoch": 0.9676712328767123, + "grad_norm": 1.9089381694793701, + "learning_rate": 3.791632818846467e-06, + "loss": 2.2289, + "step": 883 + }, + { + "epoch": 0.9687671232876712, + "grad_norm": 2.3238179683685303, + "learning_rate": 3.7896019496344436e-06, + "loss": 2.2179, + "step": 884 + }, + { + "epoch": 0.9698630136986301, + "grad_norm": 1.902161955833435, + "learning_rate": 3.787571080422421e-06, + "loss": 2.2086, + "step": 885 + }, + { + "epoch": 0.9709589041095891, + "grad_norm": 1.8587497472763062, + "learning_rate": 3.785540211210398e-06, + "loss": 2.2175, + "step": 886 + }, + { + "epoch": 0.972054794520548, + "grad_norm": 2.029719829559326, + "learning_rate": 3.7835093419983755e-06, + "loss": 2.207, + "step": 887 + }, + { + "epoch": 0.9731506849315068, + "grad_norm": 2.0228402614593506, + "learning_rate": 3.7814784727863527e-06, + "loss": 2.1851, + "step": 888 + }, + { + "epoch": 0.9742465753424657, + "grad_norm": 1.9885436296463013, + "learning_rate": 3.77944760357433e-06, + "loss": 2.1667, + "step": 889 + }, + { + "epoch": 0.9753424657534246, + "grad_norm": 1.8621459007263184, + "learning_rate": 3.7774167343623073e-06, + "loss": 2.2529, + "step": 890 + }, + { + "epoch": 0.9753424657534246, + "eval_loss": 2.303697347640991, + "eval_runtime": 44.9459, + "eval_samples_per_second": 22.249, + "eval_steps_per_second": 0.934, + "step": 890 + }, + { + "epoch": 0.9764383561643836, + "grad_norm": 2.0311460494995117, + "learning_rate": 3.7753858651502846e-06, + "loss": 2.2185, + "step": 891 + }, + { + "epoch": 0.9775342465753425, + "grad_norm": 2.2017149925231934, + "learning_rate": 3.773354995938262e-06, + "loss": 2.164, + "step": 892 + }, + { + "epoch": 0.9786301369863014, + "grad_norm": 1.949684977531433, + "learning_rate": 3.771324126726239e-06, + "loss": 2.2128, + "step": 893 + }, + { + "epoch": 0.9797260273972602, + "grad_norm": 1.9395498037338257, + "learning_rate": 3.7692932575142164e-06, + "loss": 2.1957, + "step": 894 + }, + { + "epoch": 0.9808219178082191, + "grad_norm": 2.096432685852051, + "learning_rate": 3.7672623883021937e-06, + "loss": 2.219, + "step": 895 + }, + { + "epoch": 0.9819178082191781, + "grad_norm": 2.0942952632904053, + "learning_rate": 3.765231519090171e-06, + "loss": 2.1546, + "step": 896 + }, + { + "epoch": 0.983013698630137, + "grad_norm": 2.0097291469573975, + "learning_rate": 3.7632006498781482e-06, + "loss": 2.1021, + "step": 897 + }, + { + "epoch": 0.9841095890410959, + "grad_norm": 2.0909335613250732, + "learning_rate": 3.761169780666125e-06, + "loss": 2.1862, + "step": 898 + }, + { + "epoch": 0.9852054794520548, + "grad_norm": 2.187309980392456, + "learning_rate": 3.7591389114541028e-06, + "loss": 2.1511, + "step": 899 + }, + { + "epoch": 0.9863013698630136, + "grad_norm": 2.194276809692383, + "learning_rate": 3.7571080422420796e-06, + "loss": 2.196, + "step": 900 + }, + { + "epoch": 0.9863013698630136, + "eval_loss": 2.3047611713409424, + "eval_runtime": 45.609, + "eval_samples_per_second": 21.925, + "eval_steps_per_second": 0.921, + "step": 900 + }, + { + "epoch": 0.9873972602739726, + "grad_norm": 2.1132829189300537, + "learning_rate": 3.7550771730300573e-06, + "loss": 2.2046, + "step": 901 + }, + { + "epoch": 0.9884931506849315, + "grad_norm": 2.1946468353271484, + "learning_rate": 3.753046303818034e-06, + "loss": 2.1656, + "step": 902 + }, + { + "epoch": 0.9895890410958904, + "grad_norm": 2.2210536003112793, + "learning_rate": 3.751015434606012e-06, + "loss": 2.2186, + "step": 903 + }, + { + "epoch": 0.9906849315068493, + "grad_norm": 1.913695216178894, + "learning_rate": 3.7489845653939887e-06, + "loss": 2.1933, + "step": 904 + }, + { + "epoch": 0.9917808219178083, + "grad_norm": 2.1088368892669678, + "learning_rate": 3.7469536961819664e-06, + "loss": 2.2406, + "step": 905 + }, + { + "epoch": 0.9928767123287672, + "grad_norm": 2.04239821434021, + "learning_rate": 3.7449228269699433e-06, + "loss": 2.142, + "step": 906 + }, + { + "epoch": 0.993972602739726, + "grad_norm": 1.7852507829666138, + "learning_rate": 3.742891957757921e-06, + "loss": 2.1701, + "step": 907 + }, + { + "epoch": 0.9950684931506849, + "grad_norm": 2.0275776386260986, + "learning_rate": 3.740861088545898e-06, + "loss": 2.2004, + "step": 908 + }, + { + "epoch": 0.9961643835616438, + "grad_norm": 1.9028904438018799, + "learning_rate": 3.7388302193338755e-06, + "loss": 2.177, + "step": 909 + }, + { + "epoch": 0.9972602739726028, + "grad_norm": 1.7815115451812744, + "learning_rate": 3.7367993501218524e-06, + "loss": 2.207, + "step": 910 + }, + { + "epoch": 0.9972602739726028, + "eval_loss": 2.302318811416626, + "eval_runtime": 42.9296, + "eval_samples_per_second": 23.294, + "eval_steps_per_second": 0.978, + "step": 910 + }, + { + "epoch": 0.9983561643835617, + "grad_norm": 1.913689374923706, + "learning_rate": 3.7347684809098297e-06, + "loss": 2.1594, + "step": 911 + }, + { + "epoch": 0.9994520547945206, + "grad_norm": 2.171222686767578, + "learning_rate": 3.732737611697807e-06, + "loss": 2.1689, + "step": 912 + }, + { + "epoch": 1.0005479452054795, + "grad_norm": 1.831642985343933, + "learning_rate": 3.7307067424857842e-06, + "loss": 2.1367, + "step": 913 + }, + { + "epoch": 1.0016438356164383, + "grad_norm": 2.1831955909729004, + "learning_rate": 3.7286758732737615e-06, + "loss": 2.1035, + "step": 914 + }, + { + "epoch": 1.0027397260273974, + "grad_norm": 2.038025379180908, + "learning_rate": 3.7266450040617388e-06, + "loss": 2.1461, + "step": 915 + }, + { + "epoch": 1.0038356164383562, + "grad_norm": 1.866502046585083, + "learning_rate": 3.724614134849716e-06, + "loss": 2.1322, + "step": 916 + }, + { + "epoch": 1.004931506849315, + "grad_norm": 2.1359357833862305, + "learning_rate": 3.7225832656376933e-06, + "loss": 2.2333, + "step": 917 + }, + { + "epoch": 1.006027397260274, + "grad_norm": 1.8689136505126953, + "learning_rate": 3.7205523964256706e-06, + "loss": 2.1176, + "step": 918 + }, + { + "epoch": 1.0071232876712328, + "grad_norm": 1.8869425058364868, + "learning_rate": 3.718521527213648e-06, + "loss": 2.2127, + "step": 919 + }, + { + "epoch": 1.0082191780821919, + "grad_norm": 1.9148075580596924, + "learning_rate": 3.716490658001625e-06, + "loss": 2.1846, + "step": 920 + }, + { + "epoch": 1.0082191780821919, + "eval_loss": 2.30368709564209, + "eval_runtime": 52.1845, + "eval_samples_per_second": 19.163, + "eval_steps_per_second": 0.805, + "step": 920 + }, + { + "epoch": 1.0093150684931507, + "grad_norm": 1.9502761363983154, + "learning_rate": 3.7144597887896024e-06, + "loss": 2.1731, + "step": 921 + }, + { + "epoch": 1.0104109589041095, + "grad_norm": 1.9093869924545288, + "learning_rate": 3.7124289195775797e-06, + "loss": 2.0864, + "step": 922 + }, + { + "epoch": 1.0115068493150685, + "grad_norm": 2.0871293544769287, + "learning_rate": 3.710398050365557e-06, + "loss": 2.1838, + "step": 923 + }, + { + "epoch": 1.0126027397260273, + "grad_norm": 2.0559985637664795, + "learning_rate": 3.708367181153534e-06, + "loss": 2.1764, + "step": 924 + }, + { + "epoch": 1.0136986301369864, + "grad_norm": 1.8968807458877563, + "learning_rate": 3.7063363119415115e-06, + "loss": 2.1515, + "step": 925 + }, + { + "epoch": 1.0147945205479452, + "grad_norm": 1.8613216876983643, + "learning_rate": 3.7043054427294884e-06, + "loss": 2.1694, + "step": 926 + }, + { + "epoch": 1.0158904109589042, + "grad_norm": 2.002044439315796, + "learning_rate": 3.702274573517466e-06, + "loss": 2.148, + "step": 927 + }, + { + "epoch": 1.016986301369863, + "grad_norm": 2.1138675212860107, + "learning_rate": 3.700243704305443e-06, + "loss": 2.1401, + "step": 928 + }, + { + "epoch": 1.0180821917808218, + "grad_norm": 1.7620841264724731, + "learning_rate": 3.6982128350934206e-06, + "loss": 2.1399, + "step": 929 + }, + { + "epoch": 1.0191780821917809, + "grad_norm": 1.8970746994018555, + "learning_rate": 3.6961819658813975e-06, + "loss": 2.1404, + "step": 930 + }, + { + "epoch": 1.0191780821917809, + "eval_loss": 2.3068811893463135, + "eval_runtime": 53.0188, + "eval_samples_per_second": 18.861, + "eval_steps_per_second": 0.792, + "step": 930 + }, + { + "epoch": 1.0202739726027397, + "grad_norm": 1.9405478239059448, + "learning_rate": 3.694151096669375e-06, + "loss": 2.1286, + "step": 931 + }, + { + "epoch": 1.0213698630136987, + "grad_norm": 2.112222909927368, + "learning_rate": 3.692120227457352e-06, + "loss": 2.2246, + "step": 932 + }, + { + "epoch": 1.0224657534246575, + "grad_norm": 1.8792580366134644, + "learning_rate": 3.6900893582453297e-06, + "loss": 2.148, + "step": 933 + }, + { + "epoch": 1.0235616438356165, + "grad_norm": 1.9530014991760254, + "learning_rate": 3.6880584890333066e-06, + "loss": 2.1707, + "step": 934 + }, + { + "epoch": 1.0246575342465754, + "grad_norm": 2.1778736114501953, + "learning_rate": 3.6860276198212834e-06, + "loss": 2.0879, + "step": 935 + }, + { + "epoch": 1.0257534246575342, + "grad_norm": 1.8960413932800293, + "learning_rate": 3.683996750609261e-06, + "loss": 2.1659, + "step": 936 + }, + { + "epoch": 1.0268493150684932, + "grad_norm": 1.9805221557617188, + "learning_rate": 3.681965881397238e-06, + "loss": 2.1674, + "step": 937 + }, + { + "epoch": 1.027945205479452, + "grad_norm": 1.9999068975448608, + "learning_rate": 3.6799350121852157e-06, + "loss": 2.1677, + "step": 938 + }, + { + "epoch": 1.029041095890411, + "grad_norm": 2.016911029815674, + "learning_rate": 3.6779041429731925e-06, + "loss": 2.1841, + "step": 939 + }, + { + "epoch": 1.0301369863013699, + "grad_norm": 1.7200374603271484, + "learning_rate": 3.6758732737611702e-06, + "loss": 2.1391, + "step": 940 + }, + { + "epoch": 1.0301369863013699, + "eval_loss": 2.3022048473358154, + "eval_runtime": 51.2908, + "eval_samples_per_second": 19.497, + "eval_steps_per_second": 0.819, + "step": 940 + }, + { + "epoch": 1.0312328767123287, + "grad_norm": 2.249377489089966, + "learning_rate": 3.673842404549147e-06, + "loss": 2.1618, + "step": 941 + }, + { + "epoch": 1.0323287671232877, + "grad_norm": 1.838922381401062, + "learning_rate": 3.671811535337125e-06, + "loss": 2.1527, + "step": 942 + }, + { + "epoch": 1.0334246575342465, + "grad_norm": 1.9571045637130737, + "learning_rate": 3.6697806661251016e-06, + "loss": 2.149, + "step": 943 + }, + { + "epoch": 1.0345205479452055, + "grad_norm": 1.9929102659225464, + "learning_rate": 3.6677497969130793e-06, + "loss": 2.159, + "step": 944 + }, + { + "epoch": 1.0356164383561643, + "grad_norm": 2.004830837249756, + "learning_rate": 3.665718927701056e-06, + "loss": 2.2369, + "step": 945 + }, + { + "epoch": 1.0367123287671234, + "grad_norm": 1.9605739116668701, + "learning_rate": 3.663688058489034e-06, + "loss": 2.1795, + "step": 946 + }, + { + "epoch": 1.0378082191780822, + "grad_norm": 1.9499164819717407, + "learning_rate": 3.6616571892770107e-06, + "loss": 2.1979, + "step": 947 + }, + { + "epoch": 1.038904109589041, + "grad_norm": 1.867347240447998, + "learning_rate": 3.659626320064988e-06, + "loss": 2.1616, + "step": 948 + }, + { + "epoch": 1.04, + "grad_norm": 1.9342801570892334, + "learning_rate": 3.6575954508529653e-06, + "loss": 2.1044, + "step": 949 + }, + { + "epoch": 1.0410958904109588, + "grad_norm": 2.0009584426879883, + "learning_rate": 3.6555645816409426e-06, + "loss": 2.1543, + "step": 950 + }, + { + "epoch": 1.0410958904109588, + "eval_loss": 2.3028438091278076, + "eval_runtime": 46.7426, + "eval_samples_per_second": 21.394, + "eval_steps_per_second": 0.899, + "step": 950 + }, + { + "epoch": 1.0421917808219179, + "grad_norm": 2.0411503314971924, + "learning_rate": 3.65353371242892e-06, + "loss": 2.1465, + "step": 951 + }, + { + "epoch": 1.0432876712328767, + "grad_norm": 2.0512070655822754, + "learning_rate": 3.651502843216897e-06, + "loss": 2.1828, + "step": 952 + }, + { + "epoch": 1.0443835616438357, + "grad_norm": 2.0948643684387207, + "learning_rate": 3.6494719740048744e-06, + "loss": 2.0671, + "step": 953 + }, + { + "epoch": 1.0454794520547945, + "grad_norm": 1.9103025197982788, + "learning_rate": 3.6474411047928517e-06, + "loss": 2.1651, + "step": 954 + }, + { + "epoch": 1.0465753424657533, + "grad_norm": 2.1500349044799805, + "learning_rate": 3.645410235580829e-06, + "loss": 2.137, + "step": 955 + }, + { + "epoch": 1.0476712328767124, + "grad_norm": 1.7760887145996094, + "learning_rate": 3.6433793663688062e-06, + "loss": 2.1738, + "step": 956 + }, + { + "epoch": 1.0487671232876712, + "grad_norm": 2.0111730098724365, + "learning_rate": 3.6413484971567835e-06, + "loss": 2.1569, + "step": 957 + }, + { + "epoch": 1.0498630136986302, + "grad_norm": 2.0040462017059326, + "learning_rate": 3.6393176279447608e-06, + "loss": 2.1483, + "step": 958 + }, + { + "epoch": 1.050958904109589, + "grad_norm": 1.9583499431610107, + "learning_rate": 3.637286758732738e-06, + "loss": 2.1575, + "step": 959 + }, + { + "epoch": 1.0520547945205478, + "grad_norm": 1.8842004537582397, + "learning_rate": 3.6352558895207153e-06, + "loss": 2.1797, + "step": 960 + }, + { + "epoch": 1.0520547945205478, + "eval_loss": 2.301715850830078, + "eval_runtime": 52.7345, + "eval_samples_per_second": 18.963, + "eval_steps_per_second": 0.796, + "step": 960 + }, + { + "epoch": 1.0531506849315069, + "grad_norm": 2.020909309387207, + "learning_rate": 3.633225020308692e-06, + "loss": 2.1939, + "step": 961 + }, + { + "epoch": 1.0542465753424657, + "grad_norm": 2.09222149848938, + "learning_rate": 3.63119415109667e-06, + "loss": 2.1777, + "step": 962 + }, + { + "epoch": 1.0553424657534247, + "grad_norm": 1.8582347631454468, + "learning_rate": 3.6291632818846467e-06, + "loss": 2.1505, + "step": 963 + }, + { + "epoch": 1.0564383561643835, + "grad_norm": 1.9468824863433838, + "learning_rate": 3.6271324126726244e-06, + "loss": 2.1762, + "step": 964 + }, + { + "epoch": 1.0575342465753426, + "grad_norm": 1.9484895467758179, + "learning_rate": 3.6251015434606013e-06, + "loss": 2.1102, + "step": 965 + }, + { + "epoch": 1.0586301369863014, + "grad_norm": 2.11462140083313, + "learning_rate": 3.623070674248579e-06, + "loss": 2.1114, + "step": 966 + }, + { + "epoch": 1.0597260273972602, + "grad_norm": 2.2038373947143555, + "learning_rate": 3.621039805036556e-06, + "loss": 2.1513, + "step": 967 + }, + { + "epoch": 1.0608219178082192, + "grad_norm": 2.1271684169769287, + "learning_rate": 3.6190089358245335e-06, + "loss": 2.137, + "step": 968 + }, + { + "epoch": 1.061917808219178, + "grad_norm": 1.8624194860458374, + "learning_rate": 3.6169780666125104e-06, + "loss": 2.1701, + "step": 969 + }, + { + "epoch": 1.063013698630137, + "grad_norm": 1.935445785522461, + "learning_rate": 3.614947197400488e-06, + "loss": 2.1726, + "step": 970 + }, + { + "epoch": 1.063013698630137, + "eval_loss": 2.302633047103882, + "eval_runtime": 51.8284, + "eval_samples_per_second": 19.294, + "eval_steps_per_second": 0.81, + "step": 970 + }, + { + "epoch": 1.0641095890410959, + "grad_norm": 2.048907518386841, + "learning_rate": 3.612916328188465e-06, + "loss": 2.1712, + "step": 971 + }, + { + "epoch": 1.0652054794520547, + "grad_norm": 1.850943684577942, + "learning_rate": 3.610885458976442e-06, + "loss": 2.102, + "step": 972 + }, + { + "epoch": 1.0663013698630137, + "grad_norm": 1.8364315032958984, + "learning_rate": 3.6088545897644195e-06, + "loss": 2.1903, + "step": 973 + }, + { + "epoch": 1.0673972602739725, + "grad_norm": 1.8227031230926514, + "learning_rate": 3.6068237205523963e-06, + "loss": 2.1787, + "step": 974 + }, + { + "epoch": 1.0684931506849316, + "grad_norm": 1.9136372804641724, + "learning_rate": 3.604792851340374e-06, + "loss": 2.1244, + "step": 975 + }, + { + "epoch": 1.0695890410958904, + "grad_norm": 1.8585968017578125, + "learning_rate": 3.602761982128351e-06, + "loss": 2.1287, + "step": 976 + }, + { + "epoch": 1.0706849315068494, + "grad_norm": 2.139573812484741, + "learning_rate": 3.6007311129163286e-06, + "loss": 2.1938, + "step": 977 + }, + { + "epoch": 1.0717808219178082, + "grad_norm": 2.1584925651550293, + "learning_rate": 3.5987002437043054e-06, + "loss": 2.1626, + "step": 978 + }, + { + "epoch": 1.072876712328767, + "grad_norm": 1.8905949592590332, + "learning_rate": 3.596669374492283e-06, + "loss": 2.1219, + "step": 979 + }, + { + "epoch": 1.073972602739726, + "grad_norm": 2.099651575088501, + "learning_rate": 3.59463850528026e-06, + "loss": 2.1385, + "step": 980 + }, + { + "epoch": 1.073972602739726, + "eval_loss": 2.3004090785980225, + "eval_runtime": 52.7406, + "eval_samples_per_second": 18.961, + "eval_steps_per_second": 0.796, + "step": 980 + }, + { + "epoch": 1.0750684931506849, + "grad_norm": 2.1456780433654785, + "learning_rate": 3.5926076360682377e-06, + "loss": 2.1973, + "step": 981 + }, + { + "epoch": 1.076164383561644, + "grad_norm": 1.929372787475586, + "learning_rate": 3.5905767668562145e-06, + "loss": 2.1115, + "step": 982 + }, + { + "epoch": 1.0772602739726027, + "grad_norm": 2.08174991607666, + "learning_rate": 3.5885458976441922e-06, + "loss": 2.1785, + "step": 983 + }, + { + "epoch": 1.0783561643835617, + "grad_norm": 1.9355952739715576, + "learning_rate": 3.586515028432169e-06, + "loss": 2.1567, + "step": 984 + }, + { + "epoch": 1.0794520547945206, + "grad_norm": 1.9983466863632202, + "learning_rate": 3.5844841592201464e-06, + "loss": 2.1913, + "step": 985 + }, + { + "epoch": 1.0805479452054794, + "grad_norm": 1.8714051246643066, + "learning_rate": 3.5824532900081237e-06, + "loss": 2.1938, + "step": 986 + }, + { + "epoch": 1.0816438356164384, + "grad_norm": 2.007049798965454, + "learning_rate": 3.580422420796101e-06, + "loss": 2.1093, + "step": 987 + }, + { + "epoch": 1.0827397260273972, + "grad_norm": 1.965572476387024, + "learning_rate": 3.578391551584078e-06, + "loss": 2.1426, + "step": 988 + }, + { + "epoch": 1.0838356164383562, + "grad_norm": 1.9040261507034302, + "learning_rate": 3.5763606823720555e-06, + "loss": 2.1627, + "step": 989 + }, + { + "epoch": 1.084931506849315, + "grad_norm": 2.001929759979248, + "learning_rate": 3.5743298131600328e-06, + "loss": 2.0983, + "step": 990 + }, + { + "epoch": 1.084931506849315, + "eval_loss": 2.2979862689971924, + "eval_runtime": 50.6325, + "eval_samples_per_second": 19.75, + "eval_steps_per_second": 0.83, + "step": 990 + }, + { + "epoch": 1.086027397260274, + "grad_norm": 1.9204787015914917, + "learning_rate": 3.57229894394801e-06, + "loss": 2.1571, + "step": 991 + }, + { + "epoch": 1.087123287671233, + "grad_norm": 1.9628651142120361, + "learning_rate": 3.5702680747359873e-06, + "loss": 2.1736, + "step": 992 + }, + { + "epoch": 1.0882191780821917, + "grad_norm": 1.9888029098510742, + "learning_rate": 3.5682372055239646e-06, + "loss": 2.17, + "step": 993 + }, + { + "epoch": 1.0893150684931507, + "grad_norm": 2.049769401550293, + "learning_rate": 3.566206336311942e-06, + "loss": 2.1352, + "step": 994 + }, + { + "epoch": 1.0904109589041096, + "grad_norm": 2.0208375453948975, + "learning_rate": 3.564175467099919e-06, + "loss": 2.1564, + "step": 995 + }, + { + "epoch": 1.0915068493150686, + "grad_norm": 1.976775050163269, + "learning_rate": 3.5621445978878964e-06, + "loss": 2.1292, + "step": 996 + }, + { + "epoch": 1.0926027397260274, + "grad_norm": 2.0502822399139404, + "learning_rate": 3.5601137286758737e-06, + "loss": 2.1115, + "step": 997 + }, + { + "epoch": 1.0936986301369862, + "grad_norm": 2.1184189319610596, + "learning_rate": 3.5580828594638505e-06, + "loss": 2.1641, + "step": 998 + }, + { + "epoch": 1.0947945205479452, + "grad_norm": 1.9662559032440186, + "learning_rate": 3.5560519902518282e-06, + "loss": 2.1338, + "step": 999 + }, + { + "epoch": 1.095890410958904, + "grad_norm": 2.13386869430542, + "learning_rate": 3.554021121039805e-06, + "loss": 2.158, + "step": 1000 + }, + { + "epoch": 1.095890410958904, + "eval_loss": 2.3001608848571777, + "eval_runtime": 45.9219, + "eval_samples_per_second": 21.776, + "eval_steps_per_second": 0.915, + "step": 1000 + }, + { + "epoch": 1.096986301369863, + "grad_norm": 1.8791636228561401, + "learning_rate": 3.551990251827783e-06, + "loss": 2.1752, + "step": 1001 + }, + { + "epoch": 1.098082191780822, + "grad_norm": 1.9105546474456787, + "learning_rate": 3.5499593826157596e-06, + "loss": 2.1543, + "step": 1002 + }, + { + "epoch": 1.099178082191781, + "grad_norm": 1.887668251991272, + "learning_rate": 3.5479285134037373e-06, + "loss": 2.1226, + "step": 1003 + }, + { + "epoch": 1.1002739726027397, + "grad_norm": 2.2854599952697754, + "learning_rate": 3.545897644191714e-06, + "loss": 2.1425, + "step": 1004 + }, + { + "epoch": 1.1013698630136985, + "grad_norm": 2.03892183303833, + "learning_rate": 3.543866774979692e-06, + "loss": 2.135, + "step": 1005 + }, + { + "epoch": 1.1024657534246576, + "grad_norm": 1.9144858121871948, + "learning_rate": 3.5418359057676687e-06, + "loss": 2.1323, + "step": 1006 + }, + { + "epoch": 1.1035616438356164, + "grad_norm": 2.1402337551116943, + "learning_rate": 3.5398050365556464e-06, + "loss": 2.1668, + "step": 1007 + }, + { + "epoch": 1.1046575342465754, + "grad_norm": 2.084592819213867, + "learning_rate": 3.5377741673436233e-06, + "loss": 2.1331, + "step": 1008 + }, + { + "epoch": 1.1057534246575342, + "grad_norm": 2.0012383460998535, + "learning_rate": 3.535743298131601e-06, + "loss": 2.1677, + "step": 1009 + }, + { + "epoch": 1.106849315068493, + "grad_norm": 2.272204637527466, + "learning_rate": 3.533712428919578e-06, + "loss": 2.13, + "step": 1010 + }, + { + "epoch": 1.106849315068493, + "eval_loss": 2.3001229763031006, + "eval_runtime": 42.0444, + "eval_samples_per_second": 23.784, + "eval_steps_per_second": 0.999, + "step": 1010 + }, + { + "epoch": 1.107945205479452, + "grad_norm": 2.1061534881591797, + "learning_rate": 3.5316815597075547e-06, + "loss": 2.1581, + "step": 1011 + }, + { + "epoch": 1.1090410958904109, + "grad_norm": 2.159306764602661, + "learning_rate": 3.5296506904955324e-06, + "loss": 2.1868, + "step": 1012 + }, + { + "epoch": 1.11013698630137, + "grad_norm": 2.1429545879364014, + "learning_rate": 3.5276198212835093e-06, + "loss": 2.1274, + "step": 1013 + }, + { + "epoch": 1.1112328767123287, + "grad_norm": 1.8494454622268677, + "learning_rate": 3.525588952071487e-06, + "loss": 2.1513, + "step": 1014 + }, + { + "epoch": 1.1123287671232878, + "grad_norm": 1.8494454622268677, + "learning_rate": 3.525588952071487e-06, + "loss": 2.1883, + "step": 1015 + }, + { + "epoch": 1.1134246575342466, + "grad_norm": 2.2244861125946045, + "learning_rate": 3.523558082859464e-06, + "loss": 2.1896, + "step": 1016 + }, + { + "epoch": 1.1145205479452054, + "grad_norm": 2.181950569152832, + "learning_rate": 3.5215272136474415e-06, + "loss": 2.1561, + "step": 1017 + }, + { + "epoch": 1.1156164383561644, + "grad_norm": 1.7740849256515503, + "learning_rate": 3.5194963444354184e-06, + "loss": 2.1358, + "step": 1018 + }, + { + "epoch": 1.1167123287671232, + "grad_norm": 2.3540139198303223, + "learning_rate": 3.517465475223396e-06, + "loss": 2.1241, + "step": 1019 + }, + { + "epoch": 1.1178082191780823, + "grad_norm": 2.2737820148468018, + "learning_rate": 3.515434606011373e-06, + "loss": 2.1117, + "step": 1020 + }, + { + "epoch": 1.1178082191780823, + "eval_loss": 2.2998275756835938, + "eval_runtime": 46.2142, + "eval_samples_per_second": 21.638, + "eval_steps_per_second": 0.909, + "step": 1020 + }, + { + "epoch": 1.118904109589041, + "grad_norm": 2.0705313682556152, + "learning_rate": 3.5134037367993506e-06, + "loss": 2.1743, + "step": 1021 + }, + { + "epoch": 1.12, + "grad_norm": 2.2605583667755127, + "learning_rate": 3.5113728675873275e-06, + "loss": 2.2082, + "step": 1022 + }, + { + "epoch": 1.121095890410959, + "grad_norm": 2.028203248977661, + "learning_rate": 3.5093419983753047e-06, + "loss": 2.1722, + "step": 1023 + }, + { + "epoch": 1.1221917808219177, + "grad_norm": 1.9794270992279053, + "learning_rate": 3.507311129163282e-06, + "loss": 2.1581, + "step": 1024 + }, + { + "epoch": 1.1232876712328768, + "grad_norm": 2.1830344200134277, + "learning_rate": 3.5052802599512593e-06, + "loss": 2.1142, + "step": 1025 + }, + { + "epoch": 1.1243835616438356, + "grad_norm": 1.926081895828247, + "learning_rate": 3.5032493907392366e-06, + "loss": 2.1532, + "step": 1026 + }, + { + "epoch": 1.1254794520547946, + "grad_norm": 1.9909394979476929, + "learning_rate": 3.501218521527214e-06, + "loss": 2.153, + "step": 1027 + }, + { + "epoch": 1.1265753424657534, + "grad_norm": 1.9685133695602417, + "learning_rate": 3.499187652315191e-06, + "loss": 2.0983, + "step": 1028 + }, + { + "epoch": 1.1276712328767124, + "grad_norm": 1.9185279607772827, + "learning_rate": 3.4971567831031684e-06, + "loss": 2.1495, + "step": 1029 + }, + { + "epoch": 1.1287671232876713, + "grad_norm": 2.0214717388153076, + "learning_rate": 3.4951259138911457e-06, + "loss": 2.176, + "step": 1030 + }, + { + "epoch": 1.1287671232876713, + "eval_loss": 2.2963690757751465, + "eval_runtime": 44.2655, + "eval_samples_per_second": 22.591, + "eval_steps_per_second": 0.949, + "step": 1030 + }, + { + "epoch": 1.12986301369863, + "grad_norm": 1.9134252071380615, + "learning_rate": 3.493095044679123e-06, + "loss": 2.1034, + "step": 1031 + }, + { + "epoch": 1.130958904109589, + "grad_norm": 1.8448772430419922, + "learning_rate": 3.4910641754671002e-06, + "loss": 2.1947, + "step": 1032 + }, + { + "epoch": 1.132054794520548, + "grad_norm": 1.856341004371643, + "learning_rate": 3.4890333062550775e-06, + "loss": 2.0676, + "step": 1033 + }, + { + "epoch": 1.133150684931507, + "grad_norm": 2.039062261581421, + "learning_rate": 3.4870024370430548e-06, + "loss": 2.1618, + "step": 1034 + }, + { + "epoch": 1.1342465753424658, + "grad_norm": 1.8440991640090942, + "learning_rate": 3.484971567831032e-06, + "loss": 2.0956, + "step": 1035 + }, + { + "epoch": 1.1353424657534246, + "grad_norm": 1.8483991622924805, + "learning_rate": 3.482940698619009e-06, + "loss": 2.1364, + "step": 1036 + }, + { + "epoch": 1.1364383561643836, + "grad_norm": 1.9045069217681885, + "learning_rate": 3.4809098294069866e-06, + "loss": 2.1786, + "step": 1037 + }, + { + "epoch": 1.1375342465753424, + "grad_norm": 1.8203203678131104, + "learning_rate": 3.4788789601949634e-06, + "loss": 2.1438, + "step": 1038 + }, + { + "epoch": 1.1386301369863014, + "grad_norm": 2.1084606647491455, + "learning_rate": 3.476848090982941e-06, + "loss": 2.1553, + "step": 1039 + }, + { + "epoch": 1.1397260273972603, + "grad_norm": 2.1041061878204346, + "learning_rate": 3.474817221770918e-06, + "loss": 2.1435, + "step": 1040 + }, + { + "epoch": 1.1397260273972603, + "eval_loss": 2.2970690727233887, + "eval_runtime": 43.2304, + "eval_samples_per_second": 23.132, + "eval_steps_per_second": 0.972, + "step": 1040 + }, + { + "epoch": 1.140821917808219, + "grad_norm": 1.9701104164123535, + "learning_rate": 3.4727863525588957e-06, + "loss": 2.2201, + "step": 1041 + }, + { + "epoch": 1.141917808219178, + "grad_norm": 1.8816475868225098, + "learning_rate": 3.4707554833468725e-06, + "loss": 2.1643, + "step": 1042 + }, + { + "epoch": 1.143013698630137, + "grad_norm": 2.096355438232422, + "learning_rate": 3.4687246141348502e-06, + "loss": 2.1514, + "step": 1043 + }, + { + "epoch": 1.144109589041096, + "grad_norm": 1.9975471496582031, + "learning_rate": 3.466693744922827e-06, + "loss": 2.1246, + "step": 1044 + }, + { + "epoch": 1.1452054794520548, + "grad_norm": 2.0587127208709717, + "learning_rate": 3.464662875710805e-06, + "loss": 2.1371, + "step": 1045 + }, + { + "epoch": 1.1463013698630138, + "grad_norm": 1.9275381565093994, + "learning_rate": 3.4626320064987817e-06, + "loss": 2.1518, + "step": 1046 + }, + { + "epoch": 1.1473972602739726, + "grad_norm": 2.1690948009490967, + "learning_rate": 3.4606011372867594e-06, + "loss": 2.1528, + "step": 1047 + }, + { + "epoch": 1.1484931506849314, + "grad_norm": 1.970973014831543, + "learning_rate": 3.458570268074736e-06, + "loss": 2.1569, + "step": 1048 + }, + { + "epoch": 1.1495890410958904, + "grad_norm": 1.8897690773010254, + "learning_rate": 3.4565393988627135e-06, + "loss": 2.1401, + "step": 1049 + }, + { + "epoch": 1.1506849315068493, + "grad_norm": 1.9245911836624146, + "learning_rate": 3.4545085296506908e-06, + "loss": 2.1311, + "step": 1050 + }, + { + "epoch": 1.1506849315068493, + "eval_loss": 2.2956528663635254, + "eval_runtime": 44.1778, + "eval_samples_per_second": 22.636, + "eval_steps_per_second": 0.951, + "step": 1050 + }, + { + "epoch": 1.1517808219178083, + "grad_norm": 2.142209529876709, + "learning_rate": 3.452477660438668e-06, + "loss": 2.1144, + "step": 1051 + }, + { + "epoch": 1.152876712328767, + "grad_norm": 1.932369351387024, + "learning_rate": 3.4504467912266453e-06, + "loss": 2.1739, + "step": 1052 + }, + { + "epoch": 1.1539726027397261, + "grad_norm": 2.259359359741211, + "learning_rate": 3.4484159220146226e-06, + "loss": 2.1036, + "step": 1053 + }, + { + "epoch": 1.155068493150685, + "grad_norm": 2.112485647201538, + "learning_rate": 3.4463850528026e-06, + "loss": 2.1429, + "step": 1054 + }, + { + "epoch": 1.1561643835616437, + "grad_norm": 2.6211049556732178, + "learning_rate": 3.444354183590577e-06, + "loss": 2.1421, + "step": 1055 + }, + { + "epoch": 1.1572602739726028, + "grad_norm": 1.8724851608276367, + "learning_rate": 3.4423233143785544e-06, + "loss": 2.1622, + "step": 1056 + }, + { + "epoch": 1.1583561643835616, + "grad_norm": 1.9292882680892944, + "learning_rate": 3.4402924451665317e-06, + "loss": 2.1071, + "step": 1057 + }, + { + "epoch": 1.1594520547945206, + "grad_norm": 1.9832521677017212, + "learning_rate": 3.438261575954509e-06, + "loss": 2.0879, + "step": 1058 + }, + { + "epoch": 1.1605479452054794, + "grad_norm": 1.9720780849456787, + "learning_rate": 3.4362307067424862e-06, + "loss": 2.1603, + "step": 1059 + }, + { + "epoch": 1.1616438356164385, + "grad_norm": 1.8179587125778198, + "learning_rate": 3.434199837530463e-06, + "loss": 2.1851, + "step": 1060 + }, + { + "epoch": 1.1616438356164385, + "eval_loss": 2.2950079441070557, + "eval_runtime": 42.9625, + "eval_samples_per_second": 23.276, + "eval_steps_per_second": 0.978, + "step": 1060 + }, + { + "epoch": 1.1627397260273973, + "grad_norm": 1.8762129545211792, + "learning_rate": 3.432168968318441e-06, + "loss": 2.1527, + "step": 1061 + }, + { + "epoch": 1.163835616438356, + "grad_norm": 2.173297643661499, + "learning_rate": 3.4301380991064176e-06, + "loss": 2.1023, + "step": 1062 + }, + { + "epoch": 1.1649315068493151, + "grad_norm": 1.9358222484588623, + "learning_rate": 3.4281072298943953e-06, + "loss": 2.1638, + "step": 1063 + }, + { + "epoch": 1.166027397260274, + "grad_norm": 1.9421939849853516, + "learning_rate": 3.426076360682372e-06, + "loss": 2.1794, + "step": 1064 + }, + { + "epoch": 1.167123287671233, + "grad_norm": 2.212996006011963, + "learning_rate": 3.42404549147035e-06, + "loss": 2.1877, + "step": 1065 + }, + { + "epoch": 1.1682191780821918, + "grad_norm": 2.036180257797241, + "learning_rate": 3.4220146222583267e-06, + "loss": 2.1534, + "step": 1066 + }, + { + "epoch": 1.1693150684931506, + "grad_norm": 1.8819142580032349, + "learning_rate": 3.4199837530463044e-06, + "loss": 2.099, + "step": 1067 + }, + { + "epoch": 1.1704109589041096, + "grad_norm": 2.0781915187835693, + "learning_rate": 3.4179528838342813e-06, + "loss": 2.1326, + "step": 1068 + }, + { + "epoch": 1.1715068493150684, + "grad_norm": 2.034895181655884, + "learning_rate": 3.415922014622259e-06, + "loss": 2.2107, + "step": 1069 + }, + { + "epoch": 1.1726027397260275, + "grad_norm": 1.92604660987854, + "learning_rate": 3.413891145410236e-06, + "loss": 2.1773, + "step": 1070 + }, + { + "epoch": 1.1726027397260275, + "eval_loss": 2.2959954738616943, + "eval_runtime": 47.2123, + "eval_samples_per_second": 21.181, + "eval_steps_per_second": 0.89, + "step": 1070 + }, + { + "epoch": 1.1736986301369863, + "grad_norm": 1.762730598449707, + "learning_rate": 3.4118602761982135e-06, + "loss": 2.1803, + "step": 1071 + }, + { + "epoch": 1.174794520547945, + "grad_norm": 2.0661215782165527, + "learning_rate": 3.4098294069861904e-06, + "loss": 2.1919, + "step": 1072 + }, + { + "epoch": 1.1758904109589041, + "grad_norm": 2.079982280731201, + "learning_rate": 3.4077985377741673e-06, + "loss": 2.1536, + "step": 1073 + }, + { + "epoch": 1.176986301369863, + "grad_norm": 1.9044359922409058, + "learning_rate": 3.405767668562145e-06, + "loss": 2.1403, + "step": 1074 + }, + { + "epoch": 1.178082191780822, + "grad_norm": 1.8870306015014648, + "learning_rate": 3.403736799350122e-06, + "loss": 2.1845, + "step": 1075 + }, + { + "epoch": 1.1791780821917808, + "grad_norm": 1.9305400848388672, + "learning_rate": 3.4017059301380995e-06, + "loss": 2.1786, + "step": 1076 + }, + { + "epoch": 1.1802739726027398, + "grad_norm": 1.987713098526001, + "learning_rate": 3.3996750609260764e-06, + "loss": 2.1214, + "step": 1077 + }, + { + "epoch": 1.1813698630136986, + "grad_norm": 1.8673466444015503, + "learning_rate": 3.397644191714054e-06, + "loss": 2.2027, + "step": 1078 + }, + { + "epoch": 1.1824657534246574, + "grad_norm": 2.009316921234131, + "learning_rate": 3.395613322502031e-06, + "loss": 2.139, + "step": 1079 + }, + { + "epoch": 1.1835616438356165, + "grad_norm": 2.1024956703186035, + "learning_rate": 3.3935824532900086e-06, + "loss": 2.1438, + "step": 1080 + }, + { + "epoch": 1.1835616438356165, + "eval_loss": 2.2966299057006836, + "eval_runtime": 44.8946, + "eval_samples_per_second": 22.274, + "eval_steps_per_second": 0.936, + "step": 1080 + }, + { + "epoch": 1.1846575342465753, + "grad_norm": 1.8911967277526855, + "learning_rate": 3.3915515840779855e-06, + "loss": 2.1238, + "step": 1081 + }, + { + "epoch": 1.1857534246575343, + "grad_norm": 2.0649192333221436, + "learning_rate": 3.389520714865963e-06, + "loss": 2.1534, + "step": 1082 + }, + { + "epoch": 1.1868493150684931, + "grad_norm": 1.9236847162246704, + "learning_rate": 3.38748984565394e-06, + "loss": 2.1427, + "step": 1083 + }, + { + "epoch": 1.1879452054794521, + "grad_norm": 1.8662182092666626, + "learning_rate": 3.3854589764419177e-06, + "loss": 2.1699, + "step": 1084 + }, + { + "epoch": 1.189041095890411, + "grad_norm": 1.9226020574569702, + "learning_rate": 3.3834281072298946e-06, + "loss": 2.1192, + "step": 1085 + }, + { + "epoch": 1.1901369863013698, + "grad_norm": 1.9659301042556763, + "learning_rate": 3.381397238017872e-06, + "loss": 2.1529, + "step": 1086 + }, + { + "epoch": 1.1912328767123288, + "grad_norm": 2.2714693546295166, + "learning_rate": 3.379366368805849e-06, + "loss": 2.1273, + "step": 1087 + }, + { + "epoch": 1.1923287671232876, + "grad_norm": 2.1853270530700684, + "learning_rate": 3.3773354995938264e-06, + "loss": 2.2029, + "step": 1088 + }, + { + "epoch": 1.1934246575342466, + "grad_norm": 1.8302830457687378, + "learning_rate": 3.3753046303818037e-06, + "loss": 2.1599, + "step": 1089 + }, + { + "epoch": 1.1945205479452055, + "grad_norm": 2.005591869354248, + "learning_rate": 3.373273761169781e-06, + "loss": 2.139, + "step": 1090 + }, + { + "epoch": 1.1945205479452055, + "eval_loss": 2.295994520187378, + "eval_runtime": 45.4802, + "eval_samples_per_second": 21.988, + "eval_steps_per_second": 0.923, + "step": 1090 + }, + { + "epoch": 1.1956164383561645, + "grad_norm": 2.139267683029175, + "learning_rate": 3.3712428919577582e-06, + "loss": 2.151, + "step": 1091 + }, + { + "epoch": 1.1967123287671233, + "grad_norm": 1.8547207117080688, + "learning_rate": 3.3692120227457355e-06, + "loss": 2.1249, + "step": 1092 + }, + { + "epoch": 1.1978082191780821, + "grad_norm": 2.214583396911621, + "learning_rate": 3.3671811535337128e-06, + "loss": 2.1656, + "step": 1093 + }, + { + "epoch": 1.1989041095890411, + "grad_norm": 2.1608800888061523, + "learning_rate": 3.36515028432169e-06, + "loss": 2.1056, + "step": 1094 + }, + { + "epoch": 1.2, + "grad_norm": 2.0850605964660645, + "learning_rate": 3.3631194151096673e-06, + "loss": 2.1215, + "step": 1095 + }, + { + "epoch": 1.201095890410959, + "grad_norm": 1.9199103116989136, + "learning_rate": 3.3610885458976446e-06, + "loss": 2.1584, + "step": 1096 + }, + { + "epoch": 1.2021917808219178, + "grad_norm": 1.9766334295272827, + "learning_rate": 3.359057676685622e-06, + "loss": 2.1821, + "step": 1097 + }, + { + "epoch": 1.2032876712328768, + "grad_norm": 1.948988676071167, + "learning_rate": 3.357026807473599e-06, + "loss": 2.182, + "step": 1098 + }, + { + "epoch": 1.2043835616438356, + "grad_norm": 2.0419397354125977, + "learning_rate": 3.354995938261576e-06, + "loss": 2.1051, + "step": 1099 + }, + { + "epoch": 1.2054794520547945, + "grad_norm": 1.9751213788986206, + "learning_rate": 3.3529650690495537e-06, + "loss": 2.1255, + "step": 1100 + }, + { + "epoch": 1.2054794520547945, + "eval_loss": 2.293811798095703, + "eval_runtime": 46.1182, + "eval_samples_per_second": 21.683, + "eval_steps_per_second": 0.911, + "step": 1100 + }, + { + "epoch": 1.2065753424657535, + "grad_norm": 1.9806510210037231, + "learning_rate": 3.3509341998375305e-06, + "loss": 2.1021, + "step": 1101 + }, + { + "epoch": 1.2076712328767123, + "grad_norm": 2.0570414066314697, + "learning_rate": 3.3489033306255082e-06, + "loss": 2.1417, + "step": 1102 + }, + { + "epoch": 1.208767123287671, + "grad_norm": 1.868134617805481, + "learning_rate": 3.346872461413485e-06, + "loss": 2.1494, + "step": 1103 + }, + { + "epoch": 1.2098630136986301, + "grad_norm": 2.0156848430633545, + "learning_rate": 3.344841592201463e-06, + "loss": 2.1238, + "step": 1104 + }, + { + "epoch": 1.210958904109589, + "grad_norm": 2.082301616668701, + "learning_rate": 3.3428107229894397e-06, + "loss": 2.1105, + "step": 1105 + }, + { + "epoch": 1.212054794520548, + "grad_norm": 1.8853904008865356, + "learning_rate": 3.3407798537774174e-06, + "loss": 2.1406, + "step": 1106 + }, + { + "epoch": 1.2131506849315068, + "grad_norm": 2.0410518646240234, + "learning_rate": 3.338748984565394e-06, + "loss": 2.1567, + "step": 1107 + }, + { + "epoch": 1.2142465753424658, + "grad_norm": 1.9901232719421387, + "learning_rate": 3.336718115353372e-06, + "loss": 2.1778, + "step": 1108 + }, + { + "epoch": 1.2153424657534246, + "grad_norm": 1.9840792417526245, + "learning_rate": 3.3346872461413488e-06, + "loss": 2.1935, + "step": 1109 + }, + { + "epoch": 1.2164383561643834, + "grad_norm": 1.8854445219039917, + "learning_rate": 3.3326563769293256e-06, + "loss": 2.137, + "step": 1110 + }, + { + "epoch": 1.2164383561643834, + "eval_loss": 2.294447660446167, + "eval_runtime": 42.6423, + "eval_samples_per_second": 23.451, + "eval_steps_per_second": 0.985, + "step": 1110 + }, + { + "epoch": 1.2175342465753425, + "grad_norm": 2.0948665142059326, + "learning_rate": 3.3306255077173033e-06, + "loss": 2.1288, + "step": 1111 + }, + { + "epoch": 1.2186301369863013, + "grad_norm": 1.859448790550232, + "learning_rate": 3.32859463850528e-06, + "loss": 2.1403, + "step": 1112 + }, + { + "epoch": 1.2197260273972603, + "grad_norm": 1.92074453830719, + "learning_rate": 3.326563769293258e-06, + "loss": 2.141, + "step": 1113 + }, + { + "epoch": 1.2208219178082191, + "grad_norm": 1.8377320766448975, + "learning_rate": 3.3245329000812347e-06, + "loss": 2.0889, + "step": 1114 + }, + { + "epoch": 1.2219178082191782, + "grad_norm": 1.9320917129516602, + "learning_rate": 3.3225020308692124e-06, + "loss": 2.1254, + "step": 1115 + }, + { + "epoch": 1.223013698630137, + "grad_norm": 1.9691251516342163, + "learning_rate": 3.3204711616571893e-06, + "loss": 2.1409, + "step": 1116 + }, + { + "epoch": 1.2241095890410958, + "grad_norm": 1.8214130401611328, + "learning_rate": 3.318440292445167e-06, + "loss": 2.1046, + "step": 1117 + }, + { + "epoch": 1.2252054794520548, + "grad_norm": 1.959455966949463, + "learning_rate": 3.316409423233144e-06, + "loss": 2.1418, + "step": 1118 + }, + { + "epoch": 1.2263013698630136, + "grad_norm": 1.9920449256896973, + "learning_rate": 3.3143785540211215e-06, + "loss": 2.1192, + "step": 1119 + }, + { + "epoch": 1.2273972602739727, + "grad_norm": 2.0276854038238525, + "learning_rate": 3.3123476848090984e-06, + "loss": 2.1939, + "step": 1120 + }, + { + "epoch": 1.2273972602739727, + "eval_loss": 2.295330047607422, + "eval_runtime": 49.8691, + "eval_samples_per_second": 20.052, + "eval_steps_per_second": 0.842, + "step": 1120 + }, + { + "epoch": 1.2284931506849315, + "grad_norm": 1.9765129089355469, + "learning_rate": 3.310316815597076e-06, + "loss": 2.1545, + "step": 1121 + }, + { + "epoch": 1.2295890410958905, + "grad_norm": 1.8449276685714722, + "learning_rate": 3.308285946385053e-06, + "loss": 2.1459, + "step": 1122 + }, + { + "epoch": 1.2306849315068493, + "grad_norm": 1.953072190284729, + "learning_rate": 3.30625507717303e-06, + "loss": 2.2046, + "step": 1123 + }, + { + "epoch": 1.2317808219178081, + "grad_norm": 1.9570913314819336, + "learning_rate": 3.3042242079610075e-06, + "loss": 2.1015, + "step": 1124 + }, + { + "epoch": 1.2328767123287672, + "grad_norm": 2.00569224357605, + "learning_rate": 3.3021933387489847e-06, + "loss": 2.1681, + "step": 1125 + }, + { + "epoch": 1.233972602739726, + "grad_norm": 1.9131590127944946, + "learning_rate": 3.300162469536962e-06, + "loss": 2.1765, + "step": 1126 + }, + { + "epoch": 1.235068493150685, + "grad_norm": 1.8375864028930664, + "learning_rate": 3.2981316003249393e-06, + "loss": 2.1175, + "step": 1127 + }, + { + "epoch": 1.2361643835616438, + "grad_norm": 1.852847933769226, + "learning_rate": 3.2961007311129166e-06, + "loss": 2.1838, + "step": 1128 + }, + { + "epoch": 1.2372602739726029, + "grad_norm": 1.8772693872451782, + "learning_rate": 3.294069861900894e-06, + "loss": 2.1524, + "step": 1129 + }, + { + "epoch": 1.2383561643835617, + "grad_norm": 1.8593205213546753, + "learning_rate": 3.292038992688871e-06, + "loss": 2.2059, + "step": 1130 + }, + { + "epoch": 1.2383561643835617, + "eval_loss": 2.294188976287842, + "eval_runtime": 50.4549, + "eval_samples_per_second": 19.82, + "eval_steps_per_second": 0.832, + "step": 1130 + }, + { + "epoch": 1.2394520547945205, + "grad_norm": 1.8967972993850708, + "learning_rate": 3.2900081234768484e-06, + "loss": 2.1551, + "step": 1131 + }, + { + "epoch": 1.2405479452054795, + "grad_norm": 1.9799107313156128, + "learning_rate": 3.2879772542648257e-06, + "loss": 2.1644, + "step": 1132 + }, + { + "epoch": 1.2416438356164383, + "grad_norm": 1.978883147239685, + "learning_rate": 3.285946385052803e-06, + "loss": 2.182, + "step": 1133 + }, + { + "epoch": 1.2427397260273974, + "grad_norm": 1.8705304861068726, + "learning_rate": 3.2839155158407802e-06, + "loss": 2.0981, + "step": 1134 + }, + { + "epoch": 1.2438356164383562, + "grad_norm": 2.0146889686584473, + "learning_rate": 3.2818846466287575e-06, + "loss": 2.1686, + "step": 1135 + }, + { + "epoch": 1.244931506849315, + "grad_norm": 1.94461989402771, + "learning_rate": 3.2798537774167344e-06, + "loss": 2.093, + "step": 1136 + }, + { + "epoch": 1.246027397260274, + "grad_norm": 1.7823240756988525, + "learning_rate": 3.277822908204712e-06, + "loss": 2.198, + "step": 1137 + }, + { + "epoch": 1.2471232876712328, + "grad_norm": 1.89297354221344, + "learning_rate": 3.275792038992689e-06, + "loss": 2.1359, + "step": 1138 + }, + { + "epoch": 1.2482191780821918, + "grad_norm": 1.8670459985733032, + "learning_rate": 3.2737611697806666e-06, + "loss": 2.0956, + "step": 1139 + }, + { + "epoch": 1.2493150684931507, + "grad_norm": 1.813568353652954, + "learning_rate": 3.2717303005686435e-06, + "loss": 2.1448, + "step": 1140 + }, + { + "epoch": 1.2493150684931507, + "eval_loss": 2.295804023742676, + "eval_runtime": 53.286, + "eval_samples_per_second": 18.767, + "eval_steps_per_second": 0.788, + "step": 1140 + }, + { + "epoch": 1.2504109589041095, + "grad_norm": 1.912916898727417, + "learning_rate": 3.269699431356621e-06, + "loss": 2.1159, + "step": 1141 + }, + { + "epoch": 1.2515068493150685, + "grad_norm": 1.7939883470535278, + "learning_rate": 3.267668562144598e-06, + "loss": 2.1566, + "step": 1142 + }, + { + "epoch": 1.2526027397260273, + "grad_norm": 1.8494806289672852, + "learning_rate": 3.2656376929325757e-06, + "loss": 2.1496, + "step": 1143 + }, + { + "epoch": 1.2536986301369863, + "grad_norm": 1.8468523025512695, + "learning_rate": 3.2636068237205526e-06, + "loss": 2.1159, + "step": 1144 + }, + { + "epoch": 1.2547945205479452, + "grad_norm": 1.8962055444717407, + "learning_rate": 3.2615759545085303e-06, + "loss": 2.1712, + "step": 1145 + }, + { + "epoch": 1.2558904109589042, + "grad_norm": 1.800353765487671, + "learning_rate": 3.259545085296507e-06, + "loss": 2.1486, + "step": 1146 + }, + { + "epoch": 1.256986301369863, + "grad_norm": 2.124359607696533, + "learning_rate": 3.257514216084484e-06, + "loss": 2.1913, + "step": 1147 + }, + { + "epoch": 1.2580821917808218, + "grad_norm": 1.9248145818710327, + "learning_rate": 3.2554833468724617e-06, + "loss": 2.1128, + "step": 1148 + }, + { + "epoch": 1.2591780821917808, + "grad_norm": 1.9990876913070679, + "learning_rate": 3.2534524776604385e-06, + "loss": 2.1854, + "step": 1149 + }, + { + "epoch": 1.2602739726027397, + "grad_norm": 1.8600568771362305, + "learning_rate": 3.2514216084484162e-06, + "loss": 2.1565, + "step": 1150 + }, + { + "epoch": 1.2602739726027397, + "eval_loss": 2.2942006587982178, + "eval_runtime": 51.4021, + "eval_samples_per_second": 19.454, + "eval_steps_per_second": 0.817, + "step": 1150 + }, + { + "epoch": 1.2613698630136987, + "grad_norm": 1.8621101379394531, + "learning_rate": 3.249390739236393e-06, + "loss": 2.1438, + "step": 1151 + }, + { + "epoch": 1.2624657534246575, + "grad_norm": 1.9290356636047363, + "learning_rate": 3.2473598700243708e-06, + "loss": 2.1577, + "step": 1152 + }, + { + "epoch": 1.2635616438356165, + "grad_norm": 1.9007307291030884, + "learning_rate": 3.2453290008123476e-06, + "loss": 2.1295, + "step": 1153 + }, + { + "epoch": 1.2646575342465753, + "grad_norm": 1.7647960186004639, + "learning_rate": 3.2432981316003253e-06, + "loss": 2.1472, + "step": 1154 + }, + { + "epoch": 1.2657534246575342, + "grad_norm": 1.9796825647354126, + "learning_rate": 3.241267262388302e-06, + "loss": 2.1542, + "step": 1155 + }, + { + "epoch": 1.2668493150684932, + "grad_norm": 1.9505796432495117, + "learning_rate": 3.23923639317628e-06, + "loss": 2.1093, + "step": 1156 + }, + { + "epoch": 1.267945205479452, + "grad_norm": 1.9152203798294067, + "learning_rate": 3.2372055239642567e-06, + "loss": 2.1456, + "step": 1157 + }, + { + "epoch": 1.269041095890411, + "grad_norm": 1.9094963073730469, + "learning_rate": 3.2351746547522344e-06, + "loss": 2.1149, + "step": 1158 + }, + { + "epoch": 1.2701369863013698, + "grad_norm": 1.8872679471969604, + "learning_rate": 3.2331437855402113e-06, + "loss": 2.1545, + "step": 1159 + }, + { + "epoch": 1.2712328767123289, + "grad_norm": 2.171703338623047, + "learning_rate": 3.2311129163281885e-06, + "loss": 2.2108, + "step": 1160 + }, + { + "epoch": 1.2712328767123289, + "eval_loss": 2.2910356521606445, + "eval_runtime": 52.5176, + "eval_samples_per_second": 19.041, + "eval_steps_per_second": 0.8, + "step": 1160 + }, + { + "epoch": 1.2723287671232877, + "grad_norm": 1.8578304052352905, + "learning_rate": 3.229082047116166e-06, + "loss": 2.1158, + "step": 1161 + }, + { + "epoch": 1.2734246575342465, + "grad_norm": 1.8873049020767212, + "learning_rate": 3.227051177904143e-06, + "loss": 2.1618, + "step": 1162 + }, + { + "epoch": 1.2745205479452055, + "grad_norm": 1.8442409038543701, + "learning_rate": 3.2250203086921204e-06, + "loss": 2.1454, + "step": 1163 + }, + { + "epoch": 1.2756164383561643, + "grad_norm": 1.9609211683273315, + "learning_rate": 3.2229894394800977e-06, + "loss": 2.1519, + "step": 1164 + }, + { + "epoch": 1.2767123287671232, + "grad_norm": 1.9093680381774902, + "learning_rate": 3.220958570268075e-06, + "loss": 2.1775, + "step": 1165 + }, + { + "epoch": 1.2778082191780822, + "grad_norm": 1.8355298042297363, + "learning_rate": 3.218927701056052e-06, + "loss": 2.0871, + "step": 1166 + }, + { + "epoch": 1.2789041095890412, + "grad_norm": 1.9497056007385254, + "learning_rate": 3.2168968318440295e-06, + "loss": 2.1441, + "step": 1167 + }, + { + "epoch": 1.28, + "grad_norm": 1.8635729551315308, + "learning_rate": 3.2148659626320068e-06, + "loss": 2.0748, + "step": 1168 + }, + { + "epoch": 1.2810958904109588, + "grad_norm": 2.0972132682800293, + "learning_rate": 3.212835093419984e-06, + "loss": 2.1843, + "step": 1169 + }, + { + "epoch": 1.2821917808219179, + "grad_norm": 1.9235926866531372, + "learning_rate": 3.2108042242079613e-06, + "loss": 2.1194, + "step": 1170 + }, + { + "epoch": 1.2821917808219179, + "eval_loss": 2.291309118270874, + "eval_runtime": 55.6408, + "eval_samples_per_second": 17.972, + "eval_steps_per_second": 0.755, + "step": 1170 + }, + { + "epoch": 1.2832876712328767, + "grad_norm": 1.8674591779708862, + "learning_rate": 3.2087733549959386e-06, + "loss": 2.1264, + "step": 1171 + }, + { + "epoch": 1.2843835616438355, + "grad_norm": 1.810589075088501, + "learning_rate": 3.206742485783916e-06, + "loss": 2.1781, + "step": 1172 + }, + { + "epoch": 1.2854794520547945, + "grad_norm": 1.9794505834579468, + "learning_rate": 3.2047116165718927e-06, + "loss": 2.1089, + "step": 1173 + }, + { + "epoch": 1.2865753424657536, + "grad_norm": 1.8487070798873901, + "learning_rate": 3.2026807473598704e-06, + "loss": 2.097, + "step": 1174 + }, + { + "epoch": 1.2876712328767124, + "grad_norm": 1.9668854475021362, + "learning_rate": 3.2006498781478473e-06, + "loss": 2.1516, + "step": 1175 + }, + { + "epoch": 1.2887671232876712, + "grad_norm": 1.855817198753357, + "learning_rate": 3.198619008935825e-06, + "loss": 2.1678, + "step": 1176 + }, + { + "epoch": 1.2898630136986302, + "grad_norm": 2.381211757659912, + "learning_rate": 3.196588139723802e-06, + "loss": 2.099, + "step": 1177 + }, + { + "epoch": 1.290958904109589, + "grad_norm": 2.1524250507354736, + "learning_rate": 3.1945572705117795e-06, + "loss": 2.1345, + "step": 1178 + }, + { + "epoch": 1.2920547945205478, + "grad_norm": 1.9120781421661377, + "learning_rate": 3.1925264012997564e-06, + "loss": 2.1586, + "step": 1179 + }, + { + "epoch": 1.2931506849315069, + "grad_norm": 2.153817653656006, + "learning_rate": 3.190495532087734e-06, + "loss": 2.1597, + "step": 1180 + }, + { + "epoch": 1.2931506849315069, + "eval_loss": 2.2936320304870605, + "eval_runtime": 53.6426, + "eval_samples_per_second": 18.642, + "eval_steps_per_second": 0.783, + "step": 1180 + }, + { + "epoch": 1.2942465753424657, + "grad_norm": 2.074873447418213, + "learning_rate": 3.188464662875711e-06, + "loss": 2.1572, + "step": 1181 + }, + { + "epoch": 1.2953424657534247, + "grad_norm": 1.9313570261001587, + "learning_rate": 3.1864337936636886e-06, + "loss": 2.1986, + "step": 1182 + }, + { + "epoch": 1.2964383561643835, + "grad_norm": 1.9836641550064087, + "learning_rate": 3.1844029244516655e-06, + "loss": 2.13, + "step": 1183 + }, + { + "epoch": 1.2975342465753426, + "grad_norm": 2.0893137454986572, + "learning_rate": 3.182372055239643e-06, + "loss": 2.1138, + "step": 1184 + }, + { + "epoch": 1.2986301369863014, + "grad_norm": 2.1667087078094482, + "learning_rate": 3.18034118602762e-06, + "loss": 2.1119, + "step": 1185 + }, + { + "epoch": 1.2997260273972602, + "grad_norm": 2.093961000442505, + "learning_rate": 3.1783103168155973e-06, + "loss": 2.1728, + "step": 1186 + }, + { + "epoch": 1.3008219178082192, + "grad_norm": 1.919267177581787, + "learning_rate": 3.1762794476035746e-06, + "loss": 2.1469, + "step": 1187 + }, + { + "epoch": 1.301917808219178, + "grad_norm": 2.017967939376831, + "learning_rate": 3.174248578391552e-06, + "loss": 2.1307, + "step": 1188 + }, + { + "epoch": 1.303013698630137, + "grad_norm": 1.921632170677185, + "learning_rate": 3.172217709179529e-06, + "loss": 2.13, + "step": 1189 + }, + { + "epoch": 1.3041095890410959, + "grad_norm": 2.0171291828155518, + "learning_rate": 3.1701868399675064e-06, + "loss": 2.135, + "step": 1190 + }, + { + "epoch": 1.3041095890410959, + "eval_loss": 2.2934045791625977, + "eval_runtime": 48.0821, + "eval_samples_per_second": 20.798, + "eval_steps_per_second": 0.874, + "step": 1190 + }, + { + "epoch": 1.305205479452055, + "grad_norm": 1.8685601949691772, + "learning_rate": 3.1681559707554837e-06, + "loss": 2.1329, + "step": 1191 + }, + { + "epoch": 1.3063013698630137, + "grad_norm": 1.8909096717834473, + "learning_rate": 3.166125101543461e-06, + "loss": 2.1004, + "step": 1192 + }, + { + "epoch": 1.3073972602739725, + "grad_norm": 1.798614740371704, + "learning_rate": 3.1640942323314382e-06, + "loss": 2.1555, + "step": 1193 + }, + { + "epoch": 1.3084931506849315, + "grad_norm": 1.8296692371368408, + "learning_rate": 3.1620633631194155e-06, + "loss": 2.1549, + "step": 1194 + }, + { + "epoch": 1.3095890410958904, + "grad_norm": 1.8676908016204834, + "learning_rate": 3.1600324939073928e-06, + "loss": 2.1081, + "step": 1195 + }, + { + "epoch": 1.3106849315068494, + "grad_norm": 1.8682451248168945, + "learning_rate": 3.15800162469537e-06, + "loss": 2.138, + "step": 1196 + }, + { + "epoch": 1.3117808219178082, + "grad_norm": 2.0049314498901367, + "learning_rate": 3.155970755483347e-06, + "loss": 2.144, + "step": 1197 + }, + { + "epoch": 1.3128767123287672, + "grad_norm": 2.00691556930542, + "learning_rate": 3.1539398862713246e-06, + "loss": 2.1539, + "step": 1198 + }, + { + "epoch": 1.313972602739726, + "grad_norm": 1.8521753549575806, + "learning_rate": 3.1519090170593015e-06, + "loss": 2.1597, + "step": 1199 + }, + { + "epoch": 1.3150684931506849, + "grad_norm": 2.0974042415618896, + "learning_rate": 3.149878147847279e-06, + "loss": 2.147, + "step": 1200 + }, + { + "epoch": 1.3150684931506849, + "eval_loss": 2.2891385555267334, + "eval_runtime": 48.1618, + "eval_samples_per_second": 20.763, + "eval_steps_per_second": 0.872, + "step": 1200 + }, + { + "epoch": 1.316164383561644, + "grad_norm": 1.9671052694320679, + "learning_rate": 3.147847278635256e-06, + "loss": 2.1838, + "step": 1201 + }, + { + "epoch": 1.3172602739726027, + "grad_norm": 1.9421573877334595, + "learning_rate": 3.1458164094232337e-06, + "loss": 2.2093, + "step": 1202 + }, + { + "epoch": 1.3183561643835615, + "grad_norm": 1.917198657989502, + "learning_rate": 3.1437855402112106e-06, + "loss": 2.136, + "step": 1203 + }, + { + "epoch": 1.3194520547945205, + "grad_norm": 1.8664982318878174, + "learning_rate": 3.1417546709991883e-06, + "loss": 2.089, + "step": 1204 + }, + { + "epoch": 1.3205479452054796, + "grad_norm": 2.043558359146118, + "learning_rate": 3.139723801787165e-06, + "loss": 2.159, + "step": 1205 + }, + { + "epoch": 1.3216438356164384, + "grad_norm": 1.8314157724380493, + "learning_rate": 3.137692932575143e-06, + "loss": 2.13, + "step": 1206 + }, + { + "epoch": 1.3227397260273972, + "grad_norm": 1.907884120941162, + "learning_rate": 3.1356620633631197e-06, + "loss": 2.1746, + "step": 1207 + }, + { + "epoch": 1.3238356164383562, + "grad_norm": 2.0570952892303467, + "learning_rate": 3.1336311941510974e-06, + "loss": 2.1187, + "step": 1208 + }, + { + "epoch": 1.324931506849315, + "grad_norm": 1.832292914390564, + "learning_rate": 3.1316003249390742e-06, + "loss": 2.1425, + "step": 1209 + }, + { + "epoch": 1.3260273972602739, + "grad_norm": 1.8307697772979736, + "learning_rate": 3.129569455727051e-06, + "loss": 2.1749, + "step": 1210 + }, + { + "epoch": 1.3260273972602739, + "eval_loss": 2.2865941524505615, + "eval_runtime": 42.2877, + "eval_samples_per_second": 23.648, + "eval_steps_per_second": 0.993, + "step": 1210 + }, + { + "epoch": 1.3271232876712329, + "grad_norm": 1.8292433023452759, + "learning_rate": 3.1275385865150288e-06, + "loss": 2.1702, + "step": 1211 + }, + { + "epoch": 1.3282191780821917, + "grad_norm": 1.9146367311477661, + "learning_rate": 3.1255077173030056e-06, + "loss": 2.16, + "step": 1212 + }, + { + "epoch": 1.3293150684931507, + "grad_norm": 1.809275507926941, + "learning_rate": 3.1234768480909833e-06, + "loss": 2.157, + "step": 1213 + }, + { + "epoch": 1.3304109589041095, + "grad_norm": 1.939081072807312, + "learning_rate": 3.12144597887896e-06, + "loss": 2.1496, + "step": 1214 + }, + { + "epoch": 1.3315068493150686, + "grad_norm": 1.8665997982025146, + "learning_rate": 3.119415109666938e-06, + "loss": 2.1535, + "step": 1215 + }, + { + "epoch": 1.3326027397260274, + "grad_norm": 1.8455616235733032, + "learning_rate": 3.1173842404549147e-06, + "loss": 2.1845, + "step": 1216 + }, + { + "epoch": 1.3336986301369862, + "grad_norm": 2.050560235977173, + "learning_rate": 3.1153533712428924e-06, + "loss": 2.1461, + "step": 1217 + }, + { + "epoch": 1.3347945205479452, + "grad_norm": 1.8440574407577515, + "learning_rate": 3.1133225020308693e-06, + "loss": 2.1476, + "step": 1218 + }, + { + "epoch": 1.335890410958904, + "grad_norm": 1.8801573514938354, + "learning_rate": 3.111291632818847e-06, + "loss": 2.1721, + "step": 1219 + }, + { + "epoch": 1.336986301369863, + "grad_norm": 1.9227471351623535, + "learning_rate": 3.109260763606824e-06, + "loss": 2.178, + "step": 1220 + }, + { + "epoch": 1.336986301369863, + "eval_loss": 2.2896478176116943, + "eval_runtime": 43.157, + "eval_samples_per_second": 23.171, + "eval_steps_per_second": 0.973, + "step": 1220 + }, + { + "epoch": 1.3380821917808219, + "grad_norm": 1.9979732036590576, + "learning_rate": 3.1072298943948015e-06, + "loss": 2.148, + "step": 1221 + }, + { + "epoch": 1.339178082191781, + "grad_norm": 1.9481284618377686, + "learning_rate": 3.1051990251827784e-06, + "loss": 2.165, + "step": 1222 + }, + { + "epoch": 1.3402739726027397, + "grad_norm": 1.8830820322036743, + "learning_rate": 3.1031681559707557e-06, + "loss": 2.1239, + "step": 1223 + }, + { + "epoch": 1.3413698630136985, + "grad_norm": 1.890512466430664, + "learning_rate": 3.101137286758733e-06, + "loss": 2.1761, + "step": 1224 + }, + { + "epoch": 1.3424657534246576, + "grad_norm": 2.0024659633636475, + "learning_rate": 3.09910641754671e-06, + "loss": 2.179, + "step": 1225 + }, + { + "epoch": 1.3435616438356164, + "grad_norm": 1.8659477233886719, + "learning_rate": 3.0970755483346875e-06, + "loss": 2.1722, + "step": 1226 + }, + { + "epoch": 1.3446575342465754, + "grad_norm": 1.93405020236969, + "learning_rate": 3.0950446791226648e-06, + "loss": 2.1545, + "step": 1227 + }, + { + "epoch": 1.3457534246575342, + "grad_norm": 1.9514124393463135, + "learning_rate": 3.093013809910642e-06, + "loss": 2.187, + "step": 1228 + }, + { + "epoch": 1.3468493150684933, + "grad_norm": 1.9571877717971802, + "learning_rate": 3.0909829406986193e-06, + "loss": 2.1507, + "step": 1229 + }, + { + "epoch": 1.347945205479452, + "grad_norm": 1.9434577226638794, + "learning_rate": 3.0889520714865966e-06, + "loss": 2.1953, + "step": 1230 + }, + { + "epoch": 1.347945205479452, + "eval_loss": 2.289128065109253, + "eval_runtime": 46.7151, + "eval_samples_per_second": 21.406, + "eval_steps_per_second": 0.899, + "step": 1230 + }, + { + "epoch": 1.3490410958904109, + "grad_norm": 1.8294577598571777, + "learning_rate": 3.086921202274574e-06, + "loss": 2.1684, + "step": 1231 + }, + { + "epoch": 1.35013698630137, + "grad_norm": 1.9432605504989624, + "learning_rate": 3.084890333062551e-06, + "loss": 2.1393, + "step": 1232 + }, + { + "epoch": 1.3512328767123287, + "grad_norm": 1.9030417203903198, + "learning_rate": 3.0828594638505284e-06, + "loss": 2.1181, + "step": 1233 + }, + { + "epoch": 1.3523287671232875, + "grad_norm": 1.9233827590942383, + "learning_rate": 3.0808285946385057e-06, + "loss": 2.1479, + "step": 1234 + }, + { + "epoch": 1.3534246575342466, + "grad_norm": 1.9549237489700317, + "learning_rate": 3.078797725426483e-06, + "loss": 2.1253, + "step": 1235 + }, + { + "epoch": 1.3545205479452056, + "grad_norm": 2.0584867000579834, + "learning_rate": 3.07676685621446e-06, + "loss": 2.0846, + "step": 1236 + }, + { + "epoch": 1.3556164383561644, + "grad_norm": 1.7543913125991821, + "learning_rate": 3.0747359870024375e-06, + "loss": 2.1495, + "step": 1237 + }, + { + "epoch": 1.3567123287671232, + "grad_norm": 1.750698447227478, + "learning_rate": 3.0727051177904144e-06, + "loss": 2.1179, + "step": 1238 + }, + { + "epoch": 1.3578082191780823, + "grad_norm": 1.9867119789123535, + "learning_rate": 3.070674248578392e-06, + "loss": 2.1313, + "step": 1239 + }, + { + "epoch": 1.358904109589041, + "grad_norm": 1.8228135108947754, + "learning_rate": 3.068643379366369e-06, + "loss": 2.158, + "step": 1240 + }, + { + "epoch": 1.358904109589041, + "eval_loss": 2.2861993312835693, + "eval_runtime": 44.081, + "eval_samples_per_second": 22.686, + "eval_steps_per_second": 0.953, + "step": 1240 + }, + { + "epoch": 1.3599999999999999, + "grad_norm": 1.8551791906356812, + "learning_rate": 3.0666125101543466e-06, + "loss": 2.129, + "step": 1241 + }, + { + "epoch": 1.361095890410959, + "grad_norm": 1.8729921579360962, + "learning_rate": 3.0645816409423235e-06, + "loss": 2.1069, + "step": 1242 + }, + { + "epoch": 1.362191780821918, + "grad_norm": 1.8203959465026855, + "learning_rate": 3.062550771730301e-06, + "loss": 2.1509, + "step": 1243 + }, + { + "epoch": 1.3632876712328768, + "grad_norm": 1.9498761892318726, + "learning_rate": 3.060519902518278e-06, + "loss": 2.1336, + "step": 1244 + }, + { + "epoch": 1.3643835616438356, + "grad_norm": 2.134521245956421, + "learning_rate": 3.0584890333062557e-06, + "loss": 2.1294, + "step": 1245 + }, + { + "epoch": 1.3654794520547946, + "grad_norm": 1.9321913719177246, + "learning_rate": 3.0564581640942326e-06, + "loss": 2.0823, + "step": 1246 + }, + { + "epoch": 1.3665753424657534, + "grad_norm": 1.8948324918746948, + "learning_rate": 3.0544272948822094e-06, + "loss": 2.1425, + "step": 1247 + }, + { + "epoch": 1.3676712328767122, + "grad_norm": 2.2063205242156982, + "learning_rate": 3.052396425670187e-06, + "loss": 2.1242, + "step": 1248 + }, + { + "epoch": 1.3687671232876713, + "grad_norm": 2.0060935020446777, + "learning_rate": 3.050365556458164e-06, + "loss": 2.1586, + "step": 1249 + }, + { + "epoch": 1.36986301369863, + "grad_norm": 1.8441938161849976, + "learning_rate": 3.0483346872461417e-06, + "loss": 2.1305, + "step": 1250 + }, + { + "epoch": 1.36986301369863, + "eval_loss": 2.2872142791748047, + "eval_runtime": 43.4998, + "eval_samples_per_second": 22.989, + "eval_steps_per_second": 0.966, + "step": 1250 + }, + { + "epoch": 1.370958904109589, + "grad_norm": 2.019545793533325, + "learning_rate": 3.0463038180341185e-06, + "loss": 2.1257, + "step": 1251 + }, + { + "epoch": 1.372054794520548, + "grad_norm": 1.981020212173462, + "learning_rate": 3.0442729488220962e-06, + "loss": 2.155, + "step": 1252 + }, + { + "epoch": 1.373150684931507, + "grad_norm": 1.9707001447677612, + "learning_rate": 3.042242079610073e-06, + "loss": 2.1671, + "step": 1253 + }, + { + "epoch": 1.3742465753424657, + "grad_norm": 1.8645458221435547, + "learning_rate": 3.0402112103980508e-06, + "loss": 2.1655, + "step": 1254 + }, + { + "epoch": 1.3753424657534246, + "grad_norm": 1.863911747932434, + "learning_rate": 3.0381803411860276e-06, + "loss": 2.0917, + "step": 1255 + }, + { + "epoch": 1.3764383561643836, + "grad_norm": 2.1079251766204834, + "learning_rate": 3.0361494719740053e-06, + "loss": 2.119, + "step": 1256 + }, + { + "epoch": 1.3775342465753424, + "grad_norm": 1.8242024183273315, + "learning_rate": 3.034118602761982e-06, + "loss": 2.1398, + "step": 1257 + }, + { + "epoch": 1.3786301369863014, + "grad_norm": 1.8663142919540405, + "learning_rate": 3.03208773354996e-06, + "loss": 2.1405, + "step": 1258 + }, + { + "epoch": 1.3797260273972602, + "grad_norm": 1.9437193870544434, + "learning_rate": 3.0300568643379367e-06, + "loss": 2.1156, + "step": 1259 + }, + { + "epoch": 1.3808219178082193, + "grad_norm": 2.0837693214416504, + "learning_rate": 3.028025995125914e-06, + "loss": 2.1731, + "step": 1260 + }, + { + "epoch": 1.3808219178082193, + "eval_loss": 2.283296823501587, + "eval_runtime": 45.341, + "eval_samples_per_second": 22.055, + "eval_steps_per_second": 0.926, + "step": 1260 + }, + { + "epoch": 1.381917808219178, + "grad_norm": 1.9430493116378784, + "learning_rate": 3.0259951259138913e-06, + "loss": 2.1623, + "step": 1261 + }, + { + "epoch": 1.383013698630137, + "grad_norm": 1.825675368309021, + "learning_rate": 3.0239642567018686e-06, + "loss": 2.1022, + "step": 1262 + }, + { + "epoch": 1.384109589041096, + "grad_norm": 2.0050806999206543, + "learning_rate": 3.021933387489846e-06, + "loss": 2.1443, + "step": 1263 + }, + { + "epoch": 1.3852054794520547, + "grad_norm": 1.9548894166946411, + "learning_rate": 3.019902518277823e-06, + "loss": 2.0983, + "step": 1264 + }, + { + "epoch": 1.3863013698630138, + "grad_norm": 1.9097814559936523, + "learning_rate": 3.0178716490658004e-06, + "loss": 2.1488, + "step": 1265 + }, + { + "epoch": 1.3873972602739726, + "grad_norm": 1.8977514505386353, + "learning_rate": 3.0158407798537777e-06, + "loss": 2.1641, + "step": 1266 + }, + { + "epoch": 1.3884931506849316, + "grad_norm": 1.9222278594970703, + "learning_rate": 3.013809910641755e-06, + "loss": 2.1043, + "step": 1267 + }, + { + "epoch": 1.3895890410958904, + "grad_norm": 1.9333430528640747, + "learning_rate": 3.0117790414297322e-06, + "loss": 2.1152, + "step": 1268 + }, + { + "epoch": 1.3906849315068492, + "grad_norm": 1.8405439853668213, + "learning_rate": 3.0097481722177095e-06, + "loss": 2.1046, + "step": 1269 + }, + { + "epoch": 1.3917808219178083, + "grad_norm": 2.0583181381225586, + "learning_rate": 3.0077173030056868e-06, + "loss": 2.1378, + "step": 1270 + }, + { + "epoch": 1.3917808219178083, + "eval_loss": 2.286198616027832, + "eval_runtime": 46.6914, + "eval_samples_per_second": 21.417, + "eval_steps_per_second": 0.9, + "step": 1270 + }, + { + "epoch": 1.392876712328767, + "grad_norm": 1.8485472202301025, + "learning_rate": 3.005686433793664e-06, + "loss": 2.1184, + "step": 1271 + }, + { + "epoch": 1.393972602739726, + "grad_norm": 2.0430188179016113, + "learning_rate": 3.0036555645816413e-06, + "loss": 2.1518, + "step": 1272 + }, + { + "epoch": 1.395068493150685, + "grad_norm": 2.307474374771118, + "learning_rate": 3.001624695369618e-06, + "loss": 2.1288, + "step": 1273 + }, + { + "epoch": 1.396164383561644, + "grad_norm": 1.85003662109375, + "learning_rate": 2.999593826157596e-06, + "loss": 2.1573, + "step": 1274 + }, + { + "epoch": 1.3972602739726028, + "grad_norm": 2.047619581222534, + "learning_rate": 2.9975629569455727e-06, + "loss": 2.1285, + "step": 1275 + }, + { + "epoch": 1.3983561643835616, + "grad_norm": 2.013644218444824, + "learning_rate": 2.9955320877335504e-06, + "loss": 2.1603, + "step": 1276 + }, + { + "epoch": 1.3994520547945206, + "grad_norm": 1.8814456462860107, + "learning_rate": 2.9935012185215273e-06, + "loss": 2.1333, + "step": 1277 + }, + { + "epoch": 1.4005479452054794, + "grad_norm": 2.0338361263275146, + "learning_rate": 2.991470349309505e-06, + "loss": 2.1564, + "step": 1278 + }, + { + "epoch": 1.4016438356164382, + "grad_norm": 1.972992181777954, + "learning_rate": 2.989439480097482e-06, + "loss": 2.1305, + "step": 1279 + }, + { + "epoch": 1.4027397260273973, + "grad_norm": 1.845928430557251, + "learning_rate": 2.9874086108854595e-06, + "loss": 2.1384, + "step": 1280 + }, + { + "epoch": 1.4027397260273973, + "eval_loss": 2.28485107421875, + "eval_runtime": 46.9757, + "eval_samples_per_second": 21.288, + "eval_steps_per_second": 0.894, + "step": 1280 + }, + { + "epoch": 1.403835616438356, + "grad_norm": 1.7601349353790283, + "learning_rate": 2.9853777416734364e-06, + "loss": 2.1219, + "step": 1281 + }, + { + "epoch": 1.4049315068493151, + "grad_norm": 2.2385566234588623, + "learning_rate": 2.983346872461414e-06, + "loss": 2.1581, + "step": 1282 + }, + { + "epoch": 1.406027397260274, + "grad_norm": 2.019174098968506, + "learning_rate": 2.981316003249391e-06, + "loss": 2.1753, + "step": 1283 + }, + { + "epoch": 1.407123287671233, + "grad_norm": 1.9927541017532349, + "learning_rate": 2.9792851340373678e-06, + "loss": 2.147, + "step": 1284 + }, + { + "epoch": 1.4082191780821918, + "grad_norm": 2.5143957138061523, + "learning_rate": 2.9772542648253455e-06, + "loss": 2.138, + "step": 1285 + }, + { + "epoch": 1.4093150684931506, + "grad_norm": 1.9925073385238647, + "learning_rate": 2.9752233956133223e-06, + "loss": 2.1642, + "step": 1286 + }, + { + "epoch": 1.4104109589041096, + "grad_norm": 2.1739184856414795, + "learning_rate": 2.9731925264013e-06, + "loss": 2.1258, + "step": 1287 + }, + { + "epoch": 1.4115068493150684, + "grad_norm": 2.1719982624053955, + "learning_rate": 2.971161657189277e-06, + "loss": 2.1323, + "step": 1288 + }, + { + "epoch": 1.4126027397260275, + "grad_norm": 1.9897956848144531, + "learning_rate": 2.9691307879772546e-06, + "loss": 2.1286, + "step": 1289 + }, + { + "epoch": 1.4136986301369863, + "grad_norm": 1.997477412223816, + "learning_rate": 2.9670999187652314e-06, + "loss": 2.125, + "step": 1290 + }, + { + "epoch": 1.4136986301369863, + "eval_loss": 2.2839529514312744, + "eval_runtime": 44.5005, + "eval_samples_per_second": 22.472, + "eval_steps_per_second": 0.944, + "step": 1290 + }, + { + "epoch": 1.4147945205479453, + "grad_norm": 1.9529412984848022, + "learning_rate": 2.965069049553209e-06, + "loss": 2.1074, + "step": 1291 + }, + { + "epoch": 1.415890410958904, + "grad_norm": 1.900692343711853, + "learning_rate": 2.963038180341186e-06, + "loss": 2.1796, + "step": 1292 + }, + { + "epoch": 1.416986301369863, + "grad_norm": 2.0269863605499268, + "learning_rate": 2.9610073111291637e-06, + "loss": 2.0998, + "step": 1293 + }, + { + "epoch": 1.418082191780822, + "grad_norm": 2.0290050506591797, + "learning_rate": 2.9589764419171405e-06, + "loss": 2.1638, + "step": 1294 + }, + { + "epoch": 1.4191780821917808, + "grad_norm": 1.8717535734176636, + "learning_rate": 2.9569455727051182e-06, + "loss": 2.1756, + "step": 1295 + }, + { + "epoch": 1.4202739726027398, + "grad_norm": 1.9470241069793701, + "learning_rate": 2.954914703493095e-06, + "loss": 2.116, + "step": 1296 + }, + { + "epoch": 1.4213698630136986, + "grad_norm": 2.1749250888824463, + "learning_rate": 2.9528838342810724e-06, + "loss": 2.1381, + "step": 1297 + }, + { + "epoch": 1.4224657534246576, + "grad_norm": 1.905742883682251, + "learning_rate": 2.9508529650690496e-06, + "loss": 2.118, + "step": 1298 + }, + { + "epoch": 1.4235616438356165, + "grad_norm": 2.1214921474456787, + "learning_rate": 2.948822095857027e-06, + "loss": 2.1223, + "step": 1299 + }, + { + "epoch": 1.4246575342465753, + "grad_norm": 1.894509196281433, + "learning_rate": 2.946791226645004e-06, + "loss": 2.1457, + "step": 1300 + }, + { + "epoch": 1.4246575342465753, + "eval_loss": 2.2824268341064453, + "eval_runtime": 47.8031, + "eval_samples_per_second": 20.919, + "eval_steps_per_second": 0.879, + "step": 1300 + }, + { + "epoch": 1.4257534246575343, + "grad_norm": 1.9588708877563477, + "learning_rate": 2.9447603574329815e-06, + "loss": 2.1113, + "step": 1301 + }, + { + "epoch": 1.426849315068493, + "grad_norm": 1.9709562063217163, + "learning_rate": 2.9427294882209587e-06, + "loss": 2.1086, + "step": 1302 + }, + { + "epoch": 1.427945205479452, + "grad_norm": 1.82259202003479, + "learning_rate": 2.940698619008936e-06, + "loss": 2.118, + "step": 1303 + }, + { + "epoch": 1.429041095890411, + "grad_norm": 1.9776451587677002, + "learning_rate": 2.9386677497969133e-06, + "loss": 2.1386, + "step": 1304 + }, + { + "epoch": 1.43013698630137, + "grad_norm": 2.147463321685791, + "learning_rate": 2.9366368805848906e-06, + "loss": 2.1309, + "step": 1305 + }, + { + "epoch": 1.4312328767123288, + "grad_norm": 1.9845961332321167, + "learning_rate": 2.934606011372868e-06, + "loss": 2.1586, + "step": 1306 + }, + { + "epoch": 1.4323287671232876, + "grad_norm": 1.9153180122375488, + "learning_rate": 2.932575142160845e-06, + "loss": 2.1588, + "step": 1307 + }, + { + "epoch": 1.4334246575342466, + "grad_norm": 1.9372929334640503, + "learning_rate": 2.9305442729488224e-06, + "loss": 2.1351, + "step": 1308 + }, + { + "epoch": 1.4345205479452054, + "grad_norm": 1.852669358253479, + "learning_rate": 2.9285134037367997e-06, + "loss": 2.1394, + "step": 1309 + }, + { + "epoch": 1.4356164383561643, + "grad_norm": 1.934251308441162, + "learning_rate": 2.9264825345247765e-06, + "loss": 2.1364, + "step": 1310 + }, + { + "epoch": 1.4356164383561643, + "eval_loss": 2.280508279800415, + "eval_runtime": 43.1734, + "eval_samples_per_second": 23.162, + "eval_steps_per_second": 0.973, + "step": 1310 + }, + { + "epoch": 1.4367123287671233, + "grad_norm": 1.8886998891830444, + "learning_rate": 2.9244516653127542e-06, + "loss": 2.1452, + "step": 1311 + }, + { + "epoch": 1.4378082191780823, + "grad_norm": 2.275813102722168, + "learning_rate": 2.922420796100731e-06, + "loss": 2.1465, + "step": 1312 + }, + { + "epoch": 1.4389041095890411, + "grad_norm": 1.9017852544784546, + "learning_rate": 2.9203899268887088e-06, + "loss": 2.0785, + "step": 1313 + }, + { + "epoch": 1.44, + "grad_norm": 1.8482435941696167, + "learning_rate": 2.9183590576766856e-06, + "loss": 2.1892, + "step": 1314 + }, + { + "epoch": 1.441095890410959, + "grad_norm": 2.2122418880462646, + "learning_rate": 2.9163281884646633e-06, + "loss": 2.1058, + "step": 1315 + }, + { + "epoch": 1.4421917808219178, + "grad_norm": 1.9721672534942627, + "learning_rate": 2.91429731925264e-06, + "loss": 2.1391, + "step": 1316 + }, + { + "epoch": 1.4432876712328766, + "grad_norm": 1.9381605386734009, + "learning_rate": 2.912266450040618e-06, + "loss": 2.1651, + "step": 1317 + }, + { + "epoch": 1.4443835616438356, + "grad_norm": 2.274630069732666, + "learning_rate": 2.9102355808285947e-06, + "loss": 2.1434, + "step": 1318 + }, + { + "epoch": 1.4454794520547944, + "grad_norm": 1.91957688331604, + "learning_rate": 2.9082047116165724e-06, + "loss": 2.1218, + "step": 1319 + }, + { + "epoch": 1.4465753424657535, + "grad_norm": 2.042171001434326, + "learning_rate": 2.9061738424045493e-06, + "loss": 2.1233, + "step": 1320 + }, + { + "epoch": 1.4465753424657535, + "eval_loss": 2.280571699142456, + "eval_runtime": 48.425, + "eval_samples_per_second": 20.65, + "eval_steps_per_second": 0.867, + "step": 1320 + }, + { + "epoch": 1.4476712328767123, + "grad_norm": 2.0493276119232178, + "learning_rate": 2.904142973192527e-06, + "loss": 2.1241, + "step": 1321 + }, + { + "epoch": 1.4487671232876713, + "grad_norm": 1.8865023851394653, + "learning_rate": 2.902112103980504e-06, + "loss": 2.1826, + "step": 1322 + }, + { + "epoch": 1.4498630136986301, + "grad_norm": 1.9723960161209106, + "learning_rate": 2.900081234768481e-06, + "loss": 2.1508, + "step": 1323 + }, + { + "epoch": 1.450958904109589, + "grad_norm": 1.8321057558059692, + "learning_rate": 2.8980503655564584e-06, + "loss": 2.1319, + "step": 1324 + }, + { + "epoch": 1.452054794520548, + "grad_norm": 1.97796630859375, + "learning_rate": 2.8960194963444357e-06, + "loss": 2.1289, + "step": 1325 + }, + { + "epoch": 1.4531506849315068, + "grad_norm": 1.7793136835098267, + "learning_rate": 2.893988627132413e-06, + "loss": 2.1059, + "step": 1326 + }, + { + "epoch": 1.4542465753424658, + "grad_norm": 1.878381609916687, + "learning_rate": 2.8919577579203902e-06, + "loss": 2.125, + "step": 1327 + }, + { + "epoch": 1.4553424657534246, + "grad_norm": 1.8854790925979614, + "learning_rate": 2.8899268887083675e-06, + "loss": 2.0908, + "step": 1328 + }, + { + "epoch": 1.4564383561643837, + "grad_norm": 1.940931797027588, + "learning_rate": 2.8878960194963448e-06, + "loss": 2.1338, + "step": 1329 + }, + { + "epoch": 1.4575342465753425, + "grad_norm": 1.8014280796051025, + "learning_rate": 2.885865150284322e-06, + "loss": 2.1384, + "step": 1330 + }, + { + "epoch": 1.4575342465753425, + "eval_loss": 2.2815349102020264, + "eval_runtime": 48.7646, + "eval_samples_per_second": 20.507, + "eval_steps_per_second": 0.861, + "step": 1330 + }, + { + "epoch": 1.4586301369863013, + "grad_norm": 1.9306361675262451, + "learning_rate": 2.8838342810722993e-06, + "loss": 2.1562, + "step": 1331 + }, + { + "epoch": 1.4597260273972603, + "grad_norm": 1.902762770652771, + "learning_rate": 2.8818034118602766e-06, + "loss": 2.1183, + "step": 1332 + }, + { + "epoch": 1.4608219178082191, + "grad_norm": 1.980449914932251, + "learning_rate": 2.879772542648254e-06, + "loss": 2.1066, + "step": 1333 + }, + { + "epoch": 1.4619178082191782, + "grad_norm": 2.0774567127227783, + "learning_rate": 2.8777416734362307e-06, + "loss": 2.157, + "step": 1334 + }, + { + "epoch": 1.463013698630137, + "grad_norm": 1.8203723430633545, + "learning_rate": 2.8757108042242084e-06, + "loss": 2.0792, + "step": 1335 + }, + { + "epoch": 1.464109589041096, + "grad_norm": 1.8849077224731445, + "learning_rate": 2.8736799350121853e-06, + "loss": 2.1917, + "step": 1336 + }, + { + "epoch": 1.4652054794520548, + "grad_norm": 1.8782836198806763, + "learning_rate": 2.871649065800163e-06, + "loss": 2.1744, + "step": 1337 + }, + { + "epoch": 1.4663013698630136, + "grad_norm": 1.8628056049346924, + "learning_rate": 2.86961819658814e-06, + "loss": 2.1788, + "step": 1338 + }, + { + "epoch": 1.4673972602739727, + "grad_norm": 1.865370750427246, + "learning_rate": 2.8675873273761175e-06, + "loss": 2.1159, + "step": 1339 + }, + { + "epoch": 1.4684931506849315, + "grad_norm": 1.8673044443130493, + "learning_rate": 2.8655564581640944e-06, + "loss": 2.1447, + "step": 1340 + }, + { + "epoch": 1.4684931506849315, + "eval_loss": 2.282686471939087, + "eval_runtime": 49.85, + "eval_samples_per_second": 20.06, + "eval_steps_per_second": 0.843, + "step": 1340 + }, + { + "epoch": 1.4695890410958903, + "grad_norm": 1.8928618431091309, + "learning_rate": 2.863525588952072e-06, + "loss": 2.1362, + "step": 1341 + }, + { + "epoch": 1.4706849315068493, + "grad_norm": 1.8964238166809082, + "learning_rate": 2.861494719740049e-06, + "loss": 2.1628, + "step": 1342 + }, + { + "epoch": 1.4717808219178083, + "grad_norm": 1.9889185428619385, + "learning_rate": 2.8594638505280266e-06, + "loss": 2.0922, + "step": 1343 + }, + { + "epoch": 1.4728767123287672, + "grad_norm": 1.9503910541534424, + "learning_rate": 2.8574329813160035e-06, + "loss": 2.0979, + "step": 1344 + }, + { + "epoch": 1.473972602739726, + "grad_norm": 1.8889869451522827, + "learning_rate": 2.855402112103981e-06, + "loss": 2.1329, + "step": 1345 + }, + { + "epoch": 1.475068493150685, + "grad_norm": 1.7875893115997314, + "learning_rate": 2.853371242891958e-06, + "loss": 2.123, + "step": 1346 + }, + { + "epoch": 1.4761643835616438, + "grad_norm": 1.9358372688293457, + "learning_rate": 2.851340373679935e-06, + "loss": 2.1284, + "step": 1347 + }, + { + "epoch": 1.4772602739726026, + "grad_norm": 2.125243663787842, + "learning_rate": 2.8493095044679126e-06, + "loss": 2.1386, + "step": 1348 + }, + { + "epoch": 1.4783561643835617, + "grad_norm": 1.927788257598877, + "learning_rate": 2.8472786352558894e-06, + "loss": 2.1835, + "step": 1349 + }, + { + "epoch": 1.4794520547945205, + "grad_norm": 1.9361577033996582, + "learning_rate": 2.845247766043867e-06, + "loss": 2.1411, + "step": 1350 + }, + { + "epoch": 1.4794520547945205, + "eval_loss": 2.281172275543213, + "eval_runtime": 47.4205, + "eval_samples_per_second": 21.088, + "eval_steps_per_second": 0.886, + "step": 1350 + }, + { + "epoch": 1.4805479452054795, + "grad_norm": 1.847364068031311, + "learning_rate": 2.843216896831844e-06, + "loss": 2.111, + "step": 1351 + }, + { + "epoch": 1.4816438356164383, + "grad_norm": 1.9927674531936646, + "learning_rate": 2.8411860276198217e-06, + "loss": 2.107, + "step": 1352 + }, + { + "epoch": 1.4827397260273973, + "grad_norm": 1.8565406799316406, + "learning_rate": 2.8391551584077985e-06, + "loss": 2.12, + "step": 1353 + }, + { + "epoch": 1.4838356164383562, + "grad_norm": 1.9477596282958984, + "learning_rate": 2.8371242891957762e-06, + "loss": 2.1661, + "step": 1354 + }, + { + "epoch": 1.484931506849315, + "grad_norm": 1.9354814291000366, + "learning_rate": 2.835093419983753e-06, + "loss": 2.168, + "step": 1355 + }, + { + "epoch": 1.486027397260274, + "grad_norm": 2.0603487491607666, + "learning_rate": 2.833062550771731e-06, + "loss": 2.1338, + "step": 1356 + }, + { + "epoch": 1.4871232876712328, + "grad_norm": 1.987990379333496, + "learning_rate": 2.8310316815597076e-06, + "loss": 2.1282, + "step": 1357 + }, + { + "epoch": 1.4882191780821918, + "grad_norm": 1.895529866218567, + "learning_rate": 2.8290008123476853e-06, + "loss": 2.1032, + "step": 1358 + }, + { + "epoch": 1.4893150684931507, + "grad_norm": 1.917994737625122, + "learning_rate": 2.826969943135662e-06, + "loss": 2.149, + "step": 1359 + }, + { + "epoch": 1.4904109589041097, + "grad_norm": 1.9238080978393555, + "learning_rate": 2.8249390739236395e-06, + "loss": 2.1275, + "step": 1360 + }, + { + "epoch": 1.4904109589041097, + "eval_loss": 2.280546188354492, + "eval_runtime": 46.9985, + "eval_samples_per_second": 21.277, + "eval_steps_per_second": 0.894, + "step": 1360 + }, + { + "epoch": 1.4915068493150685, + "grad_norm": 1.947768211364746, + "learning_rate": 2.8229082047116167e-06, + "loss": 2.1282, + "step": 1361 + }, + { + "epoch": 1.4926027397260273, + "grad_norm": 1.7888189554214478, + "learning_rate": 2.820877335499594e-06, + "loss": 2.1134, + "step": 1362 + }, + { + "epoch": 1.4936986301369863, + "grad_norm": 1.8480347394943237, + "learning_rate": 2.8188464662875713e-06, + "loss": 2.1194, + "step": 1363 + }, + { + "epoch": 1.4947945205479451, + "grad_norm": 1.8146322965621948, + "learning_rate": 2.8168155970755486e-06, + "loss": 2.0781, + "step": 1364 + }, + { + "epoch": 1.4958904109589042, + "grad_norm": 1.7994698286056519, + "learning_rate": 2.814784727863526e-06, + "loss": 2.1027, + "step": 1365 + }, + { + "epoch": 1.496986301369863, + "grad_norm": 1.9375026226043701, + "learning_rate": 2.812753858651503e-06, + "loss": 2.1273, + "step": 1366 + }, + { + "epoch": 1.498082191780822, + "grad_norm": 1.9606287479400635, + "learning_rate": 2.8107229894394804e-06, + "loss": 2.1438, + "step": 1367 + }, + { + "epoch": 1.4991780821917808, + "grad_norm": 2.1815521717071533, + "learning_rate": 2.8086921202274577e-06, + "loss": 2.1176, + "step": 1368 + }, + { + "epoch": 1.5002739726027396, + "grad_norm": 1.8796950578689575, + "learning_rate": 2.806661251015435e-06, + "loss": 2.096, + "step": 1369 + }, + { + "epoch": 1.5013698630136987, + "grad_norm": 2.016091823577881, + "learning_rate": 2.8046303818034122e-06, + "loss": 2.1451, + "step": 1370 + }, + { + "epoch": 1.5013698630136987, + "eval_loss": 2.280881643295288, + "eval_runtime": 53.6471, + "eval_samples_per_second": 18.64, + "eval_steps_per_second": 0.783, + "step": 1370 + }, + { + "epoch": 1.5024657534246575, + "grad_norm": 1.9303081035614014, + "learning_rate": 2.802599512591389e-06, + "loss": 2.1736, + "step": 1371 + }, + { + "epoch": 1.5035616438356163, + "grad_norm": 1.9831101894378662, + "learning_rate": 2.8005686433793668e-06, + "loss": 2.2147, + "step": 1372 + }, + { + "epoch": 1.5046575342465753, + "grad_norm": 1.7864863872528076, + "learning_rate": 2.7985377741673436e-06, + "loss": 2.1231, + "step": 1373 + }, + { + "epoch": 1.5057534246575344, + "grad_norm": 1.8945375680923462, + "learning_rate": 2.7965069049553213e-06, + "loss": 2.0866, + "step": 1374 + }, + { + "epoch": 1.5068493150684932, + "grad_norm": 1.9881761074066162, + "learning_rate": 2.794476035743298e-06, + "loss": 2.1398, + "step": 1375 + }, + { + "epoch": 1.507945205479452, + "grad_norm": 1.9888391494750977, + "learning_rate": 2.792445166531276e-06, + "loss": 2.172, + "step": 1376 + }, + { + "epoch": 1.509041095890411, + "grad_norm": 1.9845142364501953, + "learning_rate": 2.7904142973192527e-06, + "loss": 2.1378, + "step": 1377 + }, + { + "epoch": 1.5101369863013698, + "grad_norm": 2.0640218257904053, + "learning_rate": 2.7883834281072304e-06, + "loss": 2.16, + "step": 1378 + }, + { + "epoch": 1.5112328767123286, + "grad_norm": 1.9453262090682983, + "learning_rate": 2.7863525588952073e-06, + "loss": 2.1112, + "step": 1379 + }, + { + "epoch": 1.5123287671232877, + "grad_norm": 2.0385119915008545, + "learning_rate": 2.784321689683185e-06, + "loss": 2.1068, + "step": 1380 + }, + { + "epoch": 1.5123287671232877, + "eval_loss": 2.2804832458496094, + "eval_runtime": 50.1317, + "eval_samples_per_second": 19.947, + "eval_steps_per_second": 0.838, + "step": 1380 + }, + { + "epoch": 1.5134246575342467, + "grad_norm": 1.9497289657592773, + "learning_rate": 2.782290820471162e-06, + "loss": 2.1017, + "step": 1381 + }, + { + "epoch": 1.5145205479452055, + "grad_norm": 1.9402867555618286, + "learning_rate": 2.7802599512591395e-06, + "loss": 2.1141, + "step": 1382 + }, + { + "epoch": 1.5156164383561643, + "grad_norm": 2.008141279220581, + "learning_rate": 2.7782290820471164e-06, + "loss": 2.1601, + "step": 1383 + }, + { + "epoch": 1.5167123287671234, + "grad_norm": 1.9685825109481812, + "learning_rate": 2.7761982128350932e-06, + "loss": 2.1494, + "step": 1384 + }, + { + "epoch": 1.5178082191780822, + "grad_norm": 1.9075162410736084, + "learning_rate": 2.774167343623071e-06, + "loss": 2.147, + "step": 1385 + }, + { + "epoch": 1.518904109589041, + "grad_norm": 1.8381123542785645, + "learning_rate": 2.772136474411048e-06, + "loss": 2.1772, + "step": 1386 + }, + { + "epoch": 1.52, + "grad_norm": 1.806057095527649, + "learning_rate": 2.7701056051990255e-06, + "loss": 2.1077, + "step": 1387 + }, + { + "epoch": 1.521095890410959, + "grad_norm": 1.9030957221984863, + "learning_rate": 2.7680747359870023e-06, + "loss": 2.0735, + "step": 1388 + }, + { + "epoch": 1.5221917808219179, + "grad_norm": 1.9967273473739624, + "learning_rate": 2.76604386677498e-06, + "loss": 2.155, + "step": 1389 + }, + { + "epoch": 1.5232876712328767, + "grad_norm": 1.864998698234558, + "learning_rate": 2.764012997562957e-06, + "loss": 2.1319, + "step": 1390 + }, + { + "epoch": 1.5232876712328767, + "eval_loss": 2.279405355453491, + "eval_runtime": 49.6092, + "eval_samples_per_second": 20.158, + "eval_steps_per_second": 0.847, + "step": 1390 + }, + { + "epoch": 1.5243835616438357, + "grad_norm": 1.831916093826294, + "learning_rate": 2.7619821283509346e-06, + "loss": 2.1518, + "step": 1391 + }, + { + "epoch": 1.5254794520547945, + "grad_norm": 2.1456778049468994, + "learning_rate": 2.7599512591389114e-06, + "loss": 2.0955, + "step": 1392 + }, + { + "epoch": 1.5265753424657533, + "grad_norm": 1.9264456033706665, + "learning_rate": 2.757920389926889e-06, + "loss": 2.1594, + "step": 1393 + }, + { + "epoch": 1.5276712328767124, + "grad_norm": 1.9881576299667358, + "learning_rate": 2.755889520714866e-06, + "loss": 2.115, + "step": 1394 + }, + { + "epoch": 1.5287671232876714, + "grad_norm": 1.9104706048965454, + "learning_rate": 2.7538586515028437e-06, + "loss": 2.1074, + "step": 1395 + }, + { + "epoch": 1.52986301369863, + "grad_norm": 1.8815401792526245, + "learning_rate": 2.7518277822908205e-06, + "loss": 2.0779, + "step": 1396 + }, + { + "epoch": 1.530958904109589, + "grad_norm": 1.999870777130127, + "learning_rate": 2.749796913078798e-06, + "loss": 2.1409, + "step": 1397 + }, + { + "epoch": 1.532054794520548, + "grad_norm": 2.037266969680786, + "learning_rate": 2.747766043866775e-06, + "loss": 2.1565, + "step": 1398 + }, + { + "epoch": 1.5331506849315069, + "grad_norm": 1.865891695022583, + "learning_rate": 2.7457351746547524e-06, + "loss": 2.1269, + "step": 1399 + }, + { + "epoch": 1.5342465753424657, + "grad_norm": 1.959797739982605, + "learning_rate": 2.7437043054427297e-06, + "loss": 2.1674, + "step": 1400 + }, + { + "epoch": 1.5342465753424657, + "eval_loss": 2.279064893722534, + "eval_runtime": 50.0471, + "eval_samples_per_second": 19.981, + "eval_steps_per_second": 0.839, + "step": 1400 + }, + { + "epoch": 1.5353424657534247, + "grad_norm": 2.021819829940796, + "learning_rate": 2.741673436230707e-06, + "loss": 2.2088, + "step": 1401 + }, + { + "epoch": 1.5364383561643835, + "grad_norm": 1.8549301624298096, + "learning_rate": 2.739642567018684e-06, + "loss": 2.1384, + "step": 1402 + }, + { + "epoch": 1.5375342465753423, + "grad_norm": 2.0776336193084717, + "learning_rate": 2.7376116978066615e-06, + "loss": 2.1068, + "step": 1403 + }, + { + "epoch": 1.5386301369863014, + "grad_norm": 1.9860707521438599, + "learning_rate": 2.7355808285946388e-06, + "loss": 2.1377, + "step": 1404 + }, + { + "epoch": 1.5397260273972604, + "grad_norm": 2.009176015853882, + "learning_rate": 2.733549959382616e-06, + "loss": 2.136, + "step": 1405 + }, + { + "epoch": 1.5408219178082192, + "grad_norm": 1.8771592378616333, + "learning_rate": 2.7315190901705933e-06, + "loss": 2.1104, + "step": 1406 + }, + { + "epoch": 1.541917808219178, + "grad_norm": 1.9801877737045288, + "learning_rate": 2.7294882209585706e-06, + "loss": 2.1803, + "step": 1407 + }, + { + "epoch": 1.543013698630137, + "grad_norm": 1.8967032432556152, + "learning_rate": 2.727457351746548e-06, + "loss": 2.141, + "step": 1408 + }, + { + "epoch": 1.5441095890410959, + "grad_norm": 1.8221750259399414, + "learning_rate": 2.725426482534525e-06, + "loss": 2.1181, + "step": 1409 + }, + { + "epoch": 1.5452054794520547, + "grad_norm": 1.972421407699585, + "learning_rate": 2.723395613322502e-06, + "loss": 2.1056, + "step": 1410 + }, + { + "epoch": 1.5452054794520547, + "eval_loss": 2.2803397178649902, + "eval_runtime": 42.0131, + "eval_samples_per_second": 23.802, + "eval_steps_per_second": 1.0, + "step": 1410 + }, + { + "epoch": 1.5463013698630137, + "grad_norm": 1.9438509941101074, + "learning_rate": 2.7213647441104797e-06, + "loss": 2.1378, + "step": 1411 + }, + { + "epoch": 1.5473972602739727, + "grad_norm": 1.8502674102783203, + "learning_rate": 2.7193338748984565e-06, + "loss": 2.1244, + "step": 1412 + }, + { + "epoch": 1.5484931506849315, + "grad_norm": 1.8806159496307373, + "learning_rate": 2.7173030056864342e-06, + "loss": 2.1371, + "step": 1413 + }, + { + "epoch": 1.5495890410958904, + "grad_norm": 1.823891282081604, + "learning_rate": 2.715272136474411e-06, + "loss": 2.2526, + "step": 1414 + }, + { + "epoch": 1.5506849315068494, + "grad_norm": 1.9969522953033447, + "learning_rate": 2.713241267262389e-06, + "loss": 2.0843, + "step": 1415 + }, + { + "epoch": 1.5517808219178082, + "grad_norm": 1.9167888164520264, + "learning_rate": 2.7112103980503656e-06, + "loss": 2.139, + "step": 1416 + }, + { + "epoch": 1.552876712328767, + "grad_norm": 1.8950753211975098, + "learning_rate": 2.7091795288383433e-06, + "loss": 2.1184, + "step": 1417 + }, + { + "epoch": 1.553972602739726, + "grad_norm": 1.8647392988204956, + "learning_rate": 2.70714865962632e-06, + "loss": 2.1466, + "step": 1418 + }, + { + "epoch": 1.555068493150685, + "grad_norm": 2.074019432067871, + "learning_rate": 2.705117790414298e-06, + "loss": 2.1154, + "step": 1419 + }, + { + "epoch": 1.5561643835616439, + "grad_norm": 1.883457064628601, + "learning_rate": 2.7030869212022747e-06, + "loss": 2.1251, + "step": 1420 + }, + { + "epoch": 1.5561643835616439, + "eval_loss": 2.279707908630371, + "eval_runtime": 43.6933, + "eval_samples_per_second": 22.887, + "eval_steps_per_second": 0.961, + "step": 1420 + }, + { + "epoch": 1.5572602739726027, + "grad_norm": 1.88644540309906, + "learning_rate": 2.7010560519902516e-06, + "loss": 2.1554, + "step": 1421 + }, + { + "epoch": 1.5583561643835617, + "grad_norm": 1.960942029953003, + "learning_rate": 2.6990251827782293e-06, + "loss": 2.1134, + "step": 1422 + }, + { + "epoch": 1.5594520547945205, + "grad_norm": 1.9701318740844727, + "learning_rate": 2.696994313566206e-06, + "loss": 2.1212, + "step": 1423 + }, + { + "epoch": 1.5605479452054793, + "grad_norm": 1.9110147953033447, + "learning_rate": 2.694963444354184e-06, + "loss": 2.1245, + "step": 1424 + }, + { + "epoch": 1.5616438356164384, + "grad_norm": 2.1578176021575928, + "learning_rate": 2.6929325751421607e-06, + "loss": 2.1809, + "step": 1425 + }, + { + "epoch": 1.5627397260273974, + "grad_norm": 1.9082913398742676, + "learning_rate": 2.6909017059301384e-06, + "loss": 2.1186, + "step": 1426 + }, + { + "epoch": 1.563835616438356, + "grad_norm": 1.9049410820007324, + "learning_rate": 2.6888708367181153e-06, + "loss": 2.1531, + "step": 1427 + }, + { + "epoch": 1.564931506849315, + "grad_norm": 2.041297197341919, + "learning_rate": 2.686839967506093e-06, + "loss": 2.0996, + "step": 1428 + }, + { + "epoch": 1.566027397260274, + "grad_norm": 1.9282728433609009, + "learning_rate": 2.68480909829407e-06, + "loss": 2.1492, + "step": 1429 + }, + { + "epoch": 1.5671232876712329, + "grad_norm": 1.908298134803772, + "learning_rate": 2.6827782290820475e-06, + "loss": 2.0726, + "step": 1430 + }, + { + "epoch": 1.5671232876712329, + "eval_loss": 2.278843402862549, + "eval_runtime": 43.3233, + "eval_samples_per_second": 23.082, + "eval_steps_per_second": 0.969, + "step": 1430 + }, + { + "epoch": 1.5682191780821917, + "grad_norm": 2.0996971130371094, + "learning_rate": 2.6807473598700244e-06, + "loss": 2.0829, + "step": 1431 + }, + { + "epoch": 1.5693150684931507, + "grad_norm": 1.8816159963607788, + "learning_rate": 2.678716490658002e-06, + "loss": 2.1659, + "step": 1432 + }, + { + "epoch": 1.5704109589041098, + "grad_norm": 1.91497004032135, + "learning_rate": 2.676685621445979e-06, + "loss": 2.1437, + "step": 1433 + }, + { + "epoch": 1.5715068493150683, + "grad_norm": 1.8993713855743408, + "learning_rate": 2.674654752233956e-06, + "loss": 2.1324, + "step": 1434 + }, + { + "epoch": 1.5726027397260274, + "grad_norm": 1.9361183643341064, + "learning_rate": 2.6726238830219335e-06, + "loss": 2.138, + "step": 1435 + }, + { + "epoch": 1.5736986301369864, + "grad_norm": 1.8992613554000854, + "learning_rate": 2.6705930138099107e-06, + "loss": 2.1309, + "step": 1436 + }, + { + "epoch": 1.5747945205479452, + "grad_norm": 1.9301854372024536, + "learning_rate": 2.668562144597888e-06, + "loss": 2.1645, + "step": 1437 + }, + { + "epoch": 1.575890410958904, + "grad_norm": 1.8373525142669678, + "learning_rate": 2.6665312753858653e-06, + "loss": 2.0699, + "step": 1438 + }, + { + "epoch": 1.576986301369863, + "grad_norm": 2.0120513439178467, + "learning_rate": 2.6645004061738426e-06, + "loss": 2.1472, + "step": 1439 + }, + { + "epoch": 1.5780821917808219, + "grad_norm": 2.007274627685547, + "learning_rate": 2.66246953696182e-06, + "loss": 2.1406, + "step": 1440 + }, + { + "epoch": 1.5780821917808219, + "eval_loss": 2.278693914413452, + "eval_runtime": 43.8767, + "eval_samples_per_second": 22.791, + "eval_steps_per_second": 0.957, + "step": 1440 + }, + { + "epoch": 1.5791780821917807, + "grad_norm": 1.9184056520462036, + "learning_rate": 2.660438667749797e-06, + "loss": 2.1076, + "step": 1441 + }, + { + "epoch": 1.5802739726027397, + "grad_norm": 1.9140634536743164, + "learning_rate": 2.6584077985377744e-06, + "loss": 2.0868, + "step": 1442 + }, + { + "epoch": 1.5813698630136988, + "grad_norm": 1.9146252870559692, + "learning_rate": 2.6563769293257517e-06, + "loss": 2.1205, + "step": 1443 + }, + { + "epoch": 1.5824657534246576, + "grad_norm": 1.8893204927444458, + "learning_rate": 2.654346060113729e-06, + "loss": 2.1249, + "step": 1444 + }, + { + "epoch": 1.5835616438356164, + "grad_norm": 2.0452592372894287, + "learning_rate": 2.6523151909017062e-06, + "loss": 2.1069, + "step": 1445 + }, + { + "epoch": 1.5846575342465754, + "grad_norm": 1.8728071451187134, + "learning_rate": 2.6502843216896835e-06, + "loss": 2.1161, + "step": 1446 + }, + { + "epoch": 1.5857534246575342, + "grad_norm": 2.1464476585388184, + "learning_rate": 2.6482534524776603e-06, + "loss": 2.1032, + "step": 1447 + }, + { + "epoch": 1.586849315068493, + "grad_norm": 1.9760676622390747, + "learning_rate": 2.646222583265638e-06, + "loss": 2.1754, + "step": 1448 + }, + { + "epoch": 1.587945205479452, + "grad_norm": 1.9876340627670288, + "learning_rate": 2.644191714053615e-06, + "loss": 2.1512, + "step": 1449 + }, + { + "epoch": 1.589041095890411, + "grad_norm": 2.039198160171509, + "learning_rate": 2.6421608448415926e-06, + "loss": 2.1416, + "step": 1450 + }, + { + "epoch": 1.589041095890411, + "eval_loss": 2.2811355590820312, + "eval_runtime": 44.209, + "eval_samples_per_second": 22.62, + "eval_steps_per_second": 0.95, + "step": 1450 + }, + { + "epoch": 1.59013698630137, + "grad_norm": 1.882146954536438, + "learning_rate": 2.6401299756295694e-06, + "loss": 2.1106, + "step": 1451 + }, + { + "epoch": 1.5912328767123287, + "grad_norm": 2.0288634300231934, + "learning_rate": 2.638099106417547e-06, + "loss": 2.1011, + "step": 1452 + }, + { + "epoch": 1.5923287671232877, + "grad_norm": 2.0438144207000732, + "learning_rate": 2.636068237205524e-06, + "loss": 2.1352, + "step": 1453 + }, + { + "epoch": 1.5934246575342466, + "grad_norm": 1.9043279886245728, + "learning_rate": 2.6340373679935017e-06, + "loss": 2.1044, + "step": 1454 + }, + { + "epoch": 1.5945205479452054, + "grad_norm": 2.112670660018921, + "learning_rate": 2.6320064987814785e-06, + "loss": 2.1724, + "step": 1455 + }, + { + "epoch": 1.5956164383561644, + "grad_norm": 1.8258006572723389, + "learning_rate": 2.6299756295694562e-06, + "loss": 2.0825, + "step": 1456 + }, + { + "epoch": 1.5967123287671234, + "grad_norm": 1.970802664756775, + "learning_rate": 2.627944760357433e-06, + "loss": 2.1431, + "step": 1457 + }, + { + "epoch": 1.5978082191780822, + "grad_norm": 1.9067145586013794, + "learning_rate": 2.6259138911454104e-06, + "loss": 2.1729, + "step": 1458 + }, + { + "epoch": 1.598904109589041, + "grad_norm": 1.9728924036026, + "learning_rate": 2.6238830219333877e-06, + "loss": 2.1773, + "step": 1459 + }, + { + "epoch": 1.6, + "grad_norm": 2.0142018795013428, + "learning_rate": 2.621852152721365e-06, + "loss": 2.1626, + "step": 1460 + }, + { + "epoch": 1.6, + "eval_loss": 2.2811660766601562, + "eval_runtime": 44.1735, + "eval_samples_per_second": 22.638, + "eval_steps_per_second": 0.951, + "step": 1460 + }, + { + "epoch": 1.601095890410959, + "grad_norm": 2.064955472946167, + "learning_rate": 2.619821283509342e-06, + "loss": 2.0833, + "step": 1461 + }, + { + "epoch": 1.6021917808219177, + "grad_norm": 1.9600436687469482, + "learning_rate": 2.6177904142973195e-06, + "loss": 2.1596, + "step": 1462 + }, + { + "epoch": 1.6032876712328767, + "grad_norm": 1.8413004875183105, + "learning_rate": 2.6157595450852968e-06, + "loss": 2.0957, + "step": 1463 + }, + { + "epoch": 1.6043835616438358, + "grad_norm": 2.1802690029144287, + "learning_rate": 2.613728675873274e-06, + "loss": 2.1432, + "step": 1464 + }, + { + "epoch": 1.6054794520547944, + "grad_norm": 1.992746353149414, + "learning_rate": 2.6116978066612513e-06, + "loss": 2.1631, + "step": 1465 + }, + { + "epoch": 1.6065753424657534, + "grad_norm": 1.9906671047210693, + "learning_rate": 2.6096669374492286e-06, + "loss": 2.1587, + "step": 1466 + }, + { + "epoch": 1.6076712328767124, + "grad_norm": 2.1134135723114014, + "learning_rate": 2.607636068237206e-06, + "loss": 2.1385, + "step": 1467 + }, + { + "epoch": 1.6087671232876712, + "grad_norm": 2.259915590286255, + "learning_rate": 2.605605199025183e-06, + "loss": 2.1042, + "step": 1468 + }, + { + "epoch": 1.60986301369863, + "grad_norm": 2.0705254077911377, + "learning_rate": 2.6035743298131604e-06, + "loss": 2.145, + "step": 1469 + }, + { + "epoch": 1.610958904109589, + "grad_norm": 2.0326738357543945, + "learning_rate": 2.6015434606011377e-06, + "loss": 2.133, + "step": 1470 + }, + { + "epoch": 1.610958904109589, + "eval_loss": 2.2812676429748535, + "eval_runtime": 43.4091, + "eval_samples_per_second": 23.037, + "eval_steps_per_second": 0.968, + "step": 1470 + }, + { + "epoch": 1.612054794520548, + "grad_norm": 2.034182071685791, + "learning_rate": 2.5995125913891145e-06, + "loss": 2.1507, + "step": 1471 + }, + { + "epoch": 1.6131506849315067, + "grad_norm": 2.069373846054077, + "learning_rate": 2.5974817221770922e-06, + "loss": 2.152, + "step": 1472 + }, + { + "epoch": 1.6142465753424657, + "grad_norm": 2.2003285884857178, + "learning_rate": 2.595450852965069e-06, + "loss": 2.1347, + "step": 1473 + }, + { + "epoch": 1.6153424657534248, + "grad_norm": 1.919945478439331, + "learning_rate": 2.593419983753047e-06, + "loss": 2.1476, + "step": 1474 + }, + { + "epoch": 1.6164383561643836, + "grad_norm": 1.825148344039917, + "learning_rate": 2.5913891145410236e-06, + "loss": 2.1757, + "step": 1475 + }, + { + "epoch": 1.6175342465753424, + "grad_norm": 2.0014286041259766, + "learning_rate": 2.5893582453290013e-06, + "loss": 2.1208, + "step": 1476 + }, + { + "epoch": 1.6186301369863014, + "grad_norm": 2.0642764568328857, + "learning_rate": 2.587327376116978e-06, + "loss": 2.1619, + "step": 1477 + }, + { + "epoch": 1.6197260273972602, + "grad_norm": 2.002497911453247, + "learning_rate": 2.585296506904956e-06, + "loss": 2.1198, + "step": 1478 + }, + { + "epoch": 1.620821917808219, + "grad_norm": 1.7852193117141724, + "learning_rate": 2.5832656376929327e-06, + "loss": 2.1247, + "step": 1479 + }, + { + "epoch": 1.621917808219178, + "grad_norm": 1.9539082050323486, + "learning_rate": 2.5812347684809104e-06, + "loss": 2.1046, + "step": 1480 + }, + { + "epoch": 1.621917808219178, + "eval_loss": 2.2807998657226562, + "eval_runtime": 44.2896, + "eval_samples_per_second": 22.579, + "eval_steps_per_second": 0.948, + "step": 1480 + }, + { + "epoch": 1.6230136986301371, + "grad_norm": 2.095280647277832, + "learning_rate": 2.5792038992688873e-06, + "loss": 2.0685, + "step": 1481 + }, + { + "epoch": 1.624109589041096, + "grad_norm": 2.230388879776001, + "learning_rate": 2.577173030056865e-06, + "loss": 2.1976, + "step": 1482 + }, + { + "epoch": 1.6252054794520547, + "grad_norm": 1.9304745197296143, + "learning_rate": 2.575142160844842e-06, + "loss": 2.0647, + "step": 1483 + }, + { + "epoch": 1.6263013698630138, + "grad_norm": 1.856342077255249, + "learning_rate": 2.5731112916328187e-06, + "loss": 2.1444, + "step": 1484 + }, + { + "epoch": 1.6273972602739726, + "grad_norm": 2.117694854736328, + "learning_rate": 2.5710804224207964e-06, + "loss": 2.1536, + "step": 1485 + }, + { + "epoch": 1.6284931506849314, + "grad_norm": 2.043900489807129, + "learning_rate": 2.5690495532087733e-06, + "loss": 2.106, + "step": 1486 + }, + { + "epoch": 1.6295890410958904, + "grad_norm": 1.9283581972122192, + "learning_rate": 2.567018683996751e-06, + "loss": 2.0667, + "step": 1487 + }, + { + "epoch": 1.6306849315068495, + "grad_norm": 1.8138651847839355, + "learning_rate": 2.564987814784728e-06, + "loss": 2.1386, + "step": 1488 + }, + { + "epoch": 1.6317808219178083, + "grad_norm": 2.0534632205963135, + "learning_rate": 2.5629569455727055e-06, + "loss": 2.105, + "step": 1489 + }, + { + "epoch": 1.632876712328767, + "grad_norm": 1.9466804265975952, + "learning_rate": 2.5609260763606824e-06, + "loss": 2.1211, + "step": 1490 + }, + { + "epoch": 1.632876712328767, + "eval_loss": 2.275447368621826, + "eval_runtime": 44.2319, + "eval_samples_per_second": 22.608, + "eval_steps_per_second": 0.95, + "step": 1490 + }, + { + "epoch": 1.633972602739726, + "grad_norm": 1.8767902851104736, + "learning_rate": 2.55889520714866e-06, + "loss": 2.116, + "step": 1491 + }, + { + "epoch": 1.635068493150685, + "grad_norm": 1.9194284677505493, + "learning_rate": 2.556864337936637e-06, + "loss": 2.1928, + "step": 1492 + }, + { + "epoch": 1.6361643835616437, + "grad_norm": 2.0496408939361572, + "learning_rate": 2.5548334687246146e-06, + "loss": 2.1182, + "step": 1493 + }, + { + "epoch": 1.6372602739726028, + "grad_norm": 1.8577388525009155, + "learning_rate": 2.5528025995125915e-06, + "loss": 2.1313, + "step": 1494 + }, + { + "epoch": 1.6383561643835618, + "grad_norm": 2.020623207092285, + "learning_rate": 2.550771730300569e-06, + "loss": 2.1627, + "step": 1495 + }, + { + "epoch": 1.6394520547945204, + "grad_norm": 1.8472096920013428, + "learning_rate": 2.548740861088546e-06, + "loss": 2.1514, + "step": 1496 + }, + { + "epoch": 1.6405479452054794, + "grad_norm": 1.9105209112167358, + "learning_rate": 2.5467099918765233e-06, + "loss": 2.1558, + "step": 1497 + }, + { + "epoch": 1.6416438356164385, + "grad_norm": 1.9635803699493408, + "learning_rate": 2.5446791226645006e-06, + "loss": 2.08, + "step": 1498 + }, + { + "epoch": 1.6427397260273973, + "grad_norm": 1.8911043405532837, + "learning_rate": 2.542648253452478e-06, + "loss": 2.0648, + "step": 1499 + }, + { + "epoch": 1.643835616438356, + "grad_norm": 1.9337847232818604, + "learning_rate": 2.540617384240455e-06, + "loss": 2.1526, + "step": 1500 + }, + { + "epoch": 1.643835616438356, + "eval_loss": 2.2771341800689697, + "eval_runtime": 44.2753, + "eval_samples_per_second": 22.586, + "eval_steps_per_second": 0.949, + "step": 1500 + }, + { + "epoch": 1.644931506849315, + "grad_norm": 2.152970552444458, + "learning_rate": 2.5385865150284324e-06, + "loss": 2.1328, + "step": 1501 + }, + { + "epoch": 1.6460273972602741, + "grad_norm": 1.8749687671661377, + "learning_rate": 2.5365556458164097e-06, + "loss": 2.1093, + "step": 1502 + }, + { + "epoch": 1.6471232876712327, + "grad_norm": 1.8745299577713013, + "learning_rate": 2.534524776604387e-06, + "loss": 2.1172, + "step": 1503 + }, + { + "epoch": 1.6482191780821918, + "grad_norm": 2.006110906600952, + "learning_rate": 2.5324939073923642e-06, + "loss": 2.1318, + "step": 1504 + }, + { + "epoch": 1.6493150684931508, + "grad_norm": 2.055590867996216, + "learning_rate": 2.5304630381803415e-06, + "loss": 2.0463, + "step": 1505 + }, + { + "epoch": 1.6504109589041096, + "grad_norm": 1.9270331859588623, + "learning_rate": 2.5284321689683188e-06, + "loss": 2.1195, + "step": 1506 + }, + { + "epoch": 1.6515068493150684, + "grad_norm": 1.9981428384780884, + "learning_rate": 2.526401299756296e-06, + "loss": 2.1282, + "step": 1507 + }, + { + "epoch": 1.6526027397260274, + "grad_norm": 2.051969528198242, + "learning_rate": 2.524370430544273e-06, + "loss": 2.1013, + "step": 1508 + }, + { + "epoch": 1.6536986301369863, + "grad_norm": 2.067718744277954, + "learning_rate": 2.5223395613322506e-06, + "loss": 2.1265, + "step": 1509 + }, + { + "epoch": 1.654794520547945, + "grad_norm": 2.0843262672424316, + "learning_rate": 2.5203086921202274e-06, + "loss": 2.0994, + "step": 1510 + }, + { + "epoch": 1.654794520547945, + "eval_loss": 2.27781081199646, + "eval_runtime": 42.8213, + "eval_samples_per_second": 23.353, + "eval_steps_per_second": 0.981, + "step": 1510 + }, + { + "epoch": 1.655890410958904, + "grad_norm": 2.146921396255493, + "learning_rate": 2.518277822908205e-06, + "loss": 2.1456, + "step": 1511 + }, + { + "epoch": 1.6569863013698631, + "grad_norm": 1.9604636430740356, + "learning_rate": 2.516246953696182e-06, + "loss": 2.1118, + "step": 1512 + }, + { + "epoch": 1.658082191780822, + "grad_norm": 1.9613317251205444, + "learning_rate": 2.5142160844841597e-06, + "loss": 2.1401, + "step": 1513 + }, + { + "epoch": 1.6591780821917808, + "grad_norm": 2.0410237312316895, + "learning_rate": 2.5121852152721365e-06, + "loss": 2.1532, + "step": 1514 + }, + { + "epoch": 1.6602739726027398, + "grad_norm": 1.8603837490081787, + "learning_rate": 2.5101543460601142e-06, + "loss": 2.1437, + "step": 1515 + }, + { + "epoch": 1.6613698630136986, + "grad_norm": 1.9498400688171387, + "learning_rate": 2.508123476848091e-06, + "loss": 2.1384, + "step": 1516 + }, + { + "epoch": 1.6624657534246574, + "grad_norm": 1.8603938817977905, + "learning_rate": 2.506092607636069e-06, + "loss": 2.0903, + "step": 1517 + }, + { + "epoch": 1.6635616438356164, + "grad_norm": 1.8238614797592163, + "learning_rate": 2.5040617384240457e-06, + "loss": 2.1644, + "step": 1518 + }, + { + "epoch": 1.6646575342465755, + "grad_norm": 1.9574072360992432, + "learning_rate": 2.5020308692120234e-06, + "loss": 2.1072, + "step": 1519 + }, + { + "epoch": 1.6657534246575343, + "grad_norm": 2.100829601287842, + "learning_rate": 2.5e-06, + "loss": 2.111, + "step": 1520 + }, + { + "epoch": 1.6657534246575343, + "eval_loss": 2.278552293777466, + "eval_runtime": 46.587, + "eval_samples_per_second": 21.465, + "eval_steps_per_second": 0.902, + "step": 1520 + }, + { + "epoch": 1.666849315068493, + "grad_norm": 1.921262264251709, + "learning_rate": 2.4979691307879775e-06, + "loss": 2.1667, + "step": 1521 + }, + { + "epoch": 1.6679452054794521, + "grad_norm": 1.9328664541244507, + "learning_rate": 2.4959382615759548e-06, + "loss": 2.1175, + "step": 1522 + }, + { + "epoch": 1.669041095890411, + "grad_norm": 1.909544587135315, + "learning_rate": 2.493907392363932e-06, + "loss": 2.0905, + "step": 1523 + }, + { + "epoch": 1.6701369863013698, + "grad_norm": 1.928046464920044, + "learning_rate": 2.4918765231519093e-06, + "loss": 2.1757, + "step": 1524 + }, + { + "epoch": 1.6712328767123288, + "grad_norm": 1.8839048147201538, + "learning_rate": 2.4898456539398866e-06, + "loss": 2.1424, + "step": 1525 + }, + { + "epoch": 1.6723287671232878, + "grad_norm": 1.913812518119812, + "learning_rate": 2.487814784727864e-06, + "loss": 2.1178, + "step": 1526 + }, + { + "epoch": 1.6734246575342466, + "grad_norm": 1.7485435009002686, + "learning_rate": 2.4857839155158407e-06, + "loss": 2.1794, + "step": 1527 + }, + { + "epoch": 1.6745205479452054, + "grad_norm": 1.8693492412567139, + "learning_rate": 2.483753046303818e-06, + "loss": 2.1265, + "step": 1528 + }, + { + "epoch": 1.6756164383561645, + "grad_norm": 1.8982772827148438, + "learning_rate": 2.4817221770917953e-06, + "loss": 2.1254, + "step": 1529 + }, + { + "epoch": 1.6767123287671233, + "grad_norm": 1.9917978048324585, + "learning_rate": 2.4796913078797725e-06, + "loss": 2.13, + "step": 1530 + }, + { + "epoch": 1.6767123287671233, + "eval_loss": 2.277986764907837, + "eval_runtime": 52.1464, + "eval_samples_per_second": 19.177, + "eval_steps_per_second": 0.805, + "step": 1530 + }, + { + "epoch": 1.677808219178082, + "grad_norm": 1.8703829050064087, + "learning_rate": 2.47766043866775e-06, + "loss": 2.1527, + "step": 1531 + }, + { + "epoch": 1.6789041095890411, + "grad_norm": 1.950713872909546, + "learning_rate": 2.475629569455727e-06, + "loss": 2.1481, + "step": 1532 + }, + { + "epoch": 1.6800000000000002, + "grad_norm": 1.8339334726333618, + "learning_rate": 2.4735987002437044e-06, + "loss": 2.1422, + "step": 1533 + }, + { + "epoch": 1.6810958904109587, + "grad_norm": 1.8423672914505005, + "learning_rate": 2.4715678310316816e-06, + "loss": 2.1026, + "step": 1534 + }, + { + "epoch": 1.6821917808219178, + "grad_norm": 1.9338176250457764, + "learning_rate": 2.469536961819659e-06, + "loss": 2.1095, + "step": 1535 + }, + { + "epoch": 1.6832876712328768, + "grad_norm": 1.8666706085205078, + "learning_rate": 2.467506092607636e-06, + "loss": 2.1176, + "step": 1536 + }, + { + "epoch": 1.6843835616438356, + "grad_norm": 2.0762805938720703, + "learning_rate": 2.4654752233956135e-06, + "loss": 2.127, + "step": 1537 + }, + { + "epoch": 1.6854794520547944, + "grad_norm": 1.9729993343353271, + "learning_rate": 2.4634443541835907e-06, + "loss": 2.1194, + "step": 1538 + }, + { + "epoch": 1.6865753424657535, + "grad_norm": 1.910555362701416, + "learning_rate": 2.461413484971568e-06, + "loss": 2.1745, + "step": 1539 + }, + { + "epoch": 1.6876712328767123, + "grad_norm": 1.953911304473877, + "learning_rate": 2.4593826157595453e-06, + "loss": 2.1236, + "step": 1540 + }, + { + "epoch": 1.6876712328767123, + "eval_loss": 2.277575731277466, + "eval_runtime": 48.0126, + "eval_samples_per_second": 20.828, + "eval_steps_per_second": 0.875, + "step": 1540 + }, + { + "epoch": 1.688767123287671, + "grad_norm": 2.0469322204589844, + "learning_rate": 2.4573517465475226e-06, + "loss": 2.1999, + "step": 1541 + }, + { + "epoch": 1.6898630136986301, + "grad_norm": 1.9962514638900757, + "learning_rate": 2.4553208773355e-06, + "loss": 2.0865, + "step": 1542 + }, + { + "epoch": 1.6909589041095892, + "grad_norm": 1.911471962928772, + "learning_rate": 2.453290008123477e-06, + "loss": 2.1107, + "step": 1543 + }, + { + "epoch": 1.692054794520548, + "grad_norm": 1.9354206323623657, + "learning_rate": 2.4512591389114544e-06, + "loss": 2.1215, + "step": 1544 + }, + { + "epoch": 1.6931506849315068, + "grad_norm": 1.938178539276123, + "learning_rate": 2.4492282696994317e-06, + "loss": 2.1378, + "step": 1545 + }, + { + "epoch": 1.6942465753424658, + "grad_norm": 2.003552198410034, + "learning_rate": 2.447197400487409e-06, + "loss": 2.1361, + "step": 1546 + }, + { + "epoch": 1.6953424657534246, + "grad_norm": 1.9754462242126465, + "learning_rate": 2.4451665312753862e-06, + "loss": 2.134, + "step": 1547 + }, + { + "epoch": 1.6964383561643834, + "grad_norm": 1.924299955368042, + "learning_rate": 2.4431356620633635e-06, + "loss": 2.1497, + "step": 1548 + }, + { + "epoch": 1.6975342465753425, + "grad_norm": 2.1239371299743652, + "learning_rate": 2.4411047928513408e-06, + "loss": 2.1427, + "step": 1549 + }, + { + "epoch": 1.6986301369863015, + "grad_norm": 1.984377384185791, + "learning_rate": 2.439073923639318e-06, + "loss": 2.1268, + "step": 1550 + }, + { + "epoch": 1.6986301369863015, + "eval_loss": 2.2747738361358643, + "eval_runtime": 53.2249, + "eval_samples_per_second": 18.788, + "eval_steps_per_second": 0.789, + "step": 1550 + }, + { + "epoch": 1.6997260273972603, + "grad_norm": 1.9375758171081543, + "learning_rate": 2.4370430544272953e-06, + "loss": 2.1369, + "step": 1551 + }, + { + "epoch": 1.7008219178082191, + "grad_norm": 1.9191703796386719, + "learning_rate": 2.435012185215272e-06, + "loss": 2.1061, + "step": 1552 + }, + { + "epoch": 1.7019178082191782, + "grad_norm": 2.1356418132781982, + "learning_rate": 2.4329813160032495e-06, + "loss": 2.0923, + "step": 1553 + }, + { + "epoch": 1.703013698630137, + "grad_norm": 2.02870774269104, + "learning_rate": 2.4309504467912267e-06, + "loss": 2.105, + "step": 1554 + }, + { + "epoch": 1.7041095890410958, + "grad_norm": 1.9436805248260498, + "learning_rate": 2.428919577579204e-06, + "loss": 2.1194, + "step": 1555 + }, + { + "epoch": 1.7052054794520548, + "grad_norm": 1.997019648551941, + "learning_rate": 2.4268887083671813e-06, + "loss": 2.1114, + "step": 1556 + }, + { + "epoch": 1.7063013698630138, + "grad_norm": 2.0759427547454834, + "learning_rate": 2.4248578391551586e-06, + "loss": 2.1955, + "step": 1557 + }, + { + "epoch": 1.7073972602739727, + "grad_norm": 2.0526654720306396, + "learning_rate": 2.422826969943136e-06, + "loss": 2.1103, + "step": 1558 + }, + { + "epoch": 1.7084931506849315, + "grad_norm": 1.8765089511871338, + "learning_rate": 2.420796100731113e-06, + "loss": 2.1068, + "step": 1559 + }, + { + "epoch": 1.7095890410958905, + "grad_norm": 1.9164950847625732, + "learning_rate": 2.4187652315190904e-06, + "loss": 2.082, + "step": 1560 + }, + { + "epoch": 1.7095890410958905, + "eval_loss": 2.272998332977295, + "eval_runtime": 49.0487, + "eval_samples_per_second": 20.388, + "eval_steps_per_second": 0.856, + "step": 1560 + }, + { + "epoch": 1.7106849315068493, + "grad_norm": 1.9921822547912598, + "learning_rate": 2.4167343623070677e-06, + "loss": 2.1285, + "step": 1561 + }, + { + "epoch": 1.7117808219178081, + "grad_norm": 1.8890291452407837, + "learning_rate": 2.414703493095045e-06, + "loss": 2.1385, + "step": 1562 + }, + { + "epoch": 1.7128767123287671, + "grad_norm": 1.9233667850494385, + "learning_rate": 2.4126726238830222e-06, + "loss": 2.1282, + "step": 1563 + }, + { + "epoch": 1.7139726027397262, + "grad_norm": 2.0368800163269043, + "learning_rate": 2.410641754670999e-06, + "loss": 2.1366, + "step": 1564 + }, + { + "epoch": 1.7150684931506848, + "grad_norm": 2.0321929454803467, + "learning_rate": 2.4086108854589763e-06, + "loss": 2.1163, + "step": 1565 + }, + { + "epoch": 1.7161643835616438, + "grad_norm": 2.042773485183716, + "learning_rate": 2.4065800162469536e-06, + "loss": 2.1529, + "step": 1566 + }, + { + "epoch": 1.7172602739726028, + "grad_norm": 1.937940239906311, + "learning_rate": 2.404549147034931e-06, + "loss": 2.106, + "step": 1567 + }, + { + "epoch": 1.7183561643835616, + "grad_norm": 1.878649353981018, + "learning_rate": 2.402518277822908e-06, + "loss": 2.1179, + "step": 1568 + }, + { + "epoch": 1.7194520547945205, + "grad_norm": 1.997246265411377, + "learning_rate": 2.4004874086108854e-06, + "loss": 2.1672, + "step": 1569 + }, + { + "epoch": 1.7205479452054795, + "grad_norm": 1.9837015867233276, + "learning_rate": 2.3984565393988627e-06, + "loss": 2.0816, + "step": 1570 + }, + { + "epoch": 1.7205479452054795, + "eval_loss": 2.2730956077575684, + "eval_runtime": 52.1793, + "eval_samples_per_second": 19.165, + "eval_steps_per_second": 0.805, + "step": 1570 + }, + { + "epoch": 1.7216438356164385, + "grad_norm": 1.8919317722320557, + "learning_rate": 2.39642567018684e-06, + "loss": 2.1541, + "step": 1571 + }, + { + "epoch": 1.7227397260273971, + "grad_norm": 1.8129284381866455, + "learning_rate": 2.3943948009748173e-06, + "loss": 2.1033, + "step": 1572 + }, + { + "epoch": 1.7238356164383561, + "grad_norm": 1.79408860206604, + "learning_rate": 2.3923639317627945e-06, + "loss": 2.1163, + "step": 1573 + }, + { + "epoch": 1.7249315068493152, + "grad_norm": 1.9178274869918823, + "learning_rate": 2.390333062550772e-06, + "loss": 2.1383, + "step": 1574 + }, + { + "epoch": 1.726027397260274, + "grad_norm": 1.9113222360610962, + "learning_rate": 2.388302193338749e-06, + "loss": 2.0987, + "step": 1575 + }, + { + "epoch": 1.7271232876712328, + "grad_norm": 1.915170431137085, + "learning_rate": 2.3862713241267264e-06, + "loss": 2.0235, + "step": 1576 + }, + { + "epoch": 1.7282191780821918, + "grad_norm": 1.9789308309555054, + "learning_rate": 2.3842404549147037e-06, + "loss": 2.1942, + "step": 1577 + }, + { + "epoch": 1.7293150684931506, + "grad_norm": 1.8410224914550781, + "learning_rate": 2.382209585702681e-06, + "loss": 2.1105, + "step": 1578 + }, + { + "epoch": 1.7304109589041095, + "grad_norm": 1.9388853311538696, + "learning_rate": 2.380178716490658e-06, + "loss": 2.1021, + "step": 1579 + }, + { + "epoch": 1.7315068493150685, + "grad_norm": 1.9328628778457642, + "learning_rate": 2.3781478472786355e-06, + "loss": 2.1584, + "step": 1580 + }, + { + "epoch": 1.7315068493150685, + "eval_loss": 2.271240472793579, + "eval_runtime": 47.9206, + "eval_samples_per_second": 20.868, + "eval_steps_per_second": 0.876, + "step": 1580 + }, + { + "epoch": 1.7326027397260275, + "grad_norm": 1.8846843242645264, + "learning_rate": 2.3761169780666128e-06, + "loss": 2.1269, + "step": 1581 + }, + { + "epoch": 1.7336986301369863, + "grad_norm": 1.8232983350753784, + "learning_rate": 2.37408610885459e-06, + "loss": 2.1522, + "step": 1582 + }, + { + "epoch": 1.7347945205479451, + "grad_norm": 2.023982286453247, + "learning_rate": 2.3720552396425673e-06, + "loss": 2.1571, + "step": 1583 + }, + { + "epoch": 1.7358904109589042, + "grad_norm": 1.861229658126831, + "learning_rate": 2.3700243704305446e-06, + "loss": 2.103, + "step": 1584 + }, + { + "epoch": 1.736986301369863, + "grad_norm": 1.8577842712402344, + "learning_rate": 2.367993501218522e-06, + "loss": 2.1766, + "step": 1585 + }, + { + "epoch": 1.7380821917808218, + "grad_norm": 1.915722131729126, + "learning_rate": 2.365962632006499e-06, + "loss": 2.1306, + "step": 1586 + }, + { + "epoch": 1.7391780821917808, + "grad_norm": 1.8876861333847046, + "learning_rate": 2.3639317627944764e-06, + "loss": 2.1303, + "step": 1587 + }, + { + "epoch": 1.7402739726027399, + "grad_norm": 1.9599688053131104, + "learning_rate": 2.3619008935824537e-06, + "loss": 2.1513, + "step": 1588 + }, + { + "epoch": 1.7413698630136987, + "grad_norm": 1.90068519115448, + "learning_rate": 2.3598700243704305e-06, + "loss": 2.1459, + "step": 1589 + }, + { + "epoch": 1.7424657534246575, + "grad_norm": 1.920277714729309, + "learning_rate": 2.357839155158408e-06, + "loss": 2.1453, + "step": 1590 + }, + { + "epoch": 1.7424657534246575, + "eval_loss": 2.2699742317199707, + "eval_runtime": 50.0211, + "eval_samples_per_second": 19.992, + "eval_steps_per_second": 0.84, + "step": 1590 + }, + { + "epoch": 1.7435616438356165, + "grad_norm": 1.9827759265899658, + "learning_rate": 2.355808285946385e-06, + "loss": 2.1393, + "step": 1591 + }, + { + "epoch": 1.7446575342465753, + "grad_norm": 1.8602412939071655, + "learning_rate": 2.3537774167343624e-06, + "loss": 2.1317, + "step": 1592 + }, + { + "epoch": 1.7457534246575341, + "grad_norm": 1.964829444885254, + "learning_rate": 2.3517465475223396e-06, + "loss": 2.1552, + "step": 1593 + }, + { + "epoch": 1.7468493150684932, + "grad_norm": 2.0124034881591797, + "learning_rate": 2.349715678310317e-06, + "loss": 2.1322, + "step": 1594 + }, + { + "epoch": 1.7479452054794522, + "grad_norm": 1.9786088466644287, + "learning_rate": 2.347684809098294e-06, + "loss": 2.1435, + "step": 1595 + }, + { + "epoch": 1.749041095890411, + "grad_norm": 1.9575989246368408, + "learning_rate": 2.3456539398862715e-06, + "loss": 2.1032, + "step": 1596 + }, + { + "epoch": 1.7501369863013698, + "grad_norm": 2.0460586547851562, + "learning_rate": 2.3436230706742487e-06, + "loss": 2.1595, + "step": 1597 + }, + { + "epoch": 1.7512328767123289, + "grad_norm": 1.9904825687408447, + "learning_rate": 2.341592201462226e-06, + "loss": 2.1025, + "step": 1598 + }, + { + "epoch": 1.7523287671232877, + "grad_norm": 2.0797626972198486, + "learning_rate": 2.3395613322502033e-06, + "loss": 2.1221, + "step": 1599 + }, + { + "epoch": 1.7534246575342465, + "grad_norm": 2.0380358695983887, + "learning_rate": 2.3375304630381806e-06, + "loss": 2.1166, + "step": 1600 + }, + { + "epoch": 1.7534246575342465, + "eval_loss": 2.2704837322235107, + "eval_runtime": 51.9207, + "eval_samples_per_second": 19.26, + "eval_steps_per_second": 0.809, + "step": 1600 + }, + { + "epoch": 1.7545205479452055, + "grad_norm": 2.009023427963257, + "learning_rate": 2.335499593826158e-06, + "loss": 2.0922, + "step": 1601 + }, + { + "epoch": 1.7556164383561645, + "grad_norm": 2.043168783187866, + "learning_rate": 2.333468724614135e-06, + "loss": 2.0962, + "step": 1602 + }, + { + "epoch": 1.7567123287671231, + "grad_norm": 2.1130902767181396, + "learning_rate": 2.3314378554021124e-06, + "loss": 2.1318, + "step": 1603 + }, + { + "epoch": 1.7578082191780822, + "grad_norm": 1.9391744136810303, + "learning_rate": 2.3294069861900897e-06, + "loss": 2.155, + "step": 1604 + }, + { + "epoch": 1.7589041095890412, + "grad_norm": 2.0810694694519043, + "learning_rate": 2.327376116978067e-06, + "loss": 2.1066, + "step": 1605 + }, + { + "epoch": 1.76, + "grad_norm": 2.029707431793213, + "learning_rate": 2.3253452477660442e-06, + "loss": 2.0897, + "step": 1606 + }, + { + "epoch": 1.7610958904109588, + "grad_norm": 1.928080439567566, + "learning_rate": 2.3233143785540215e-06, + "loss": 2.1589, + "step": 1607 + }, + { + "epoch": 1.7621917808219179, + "grad_norm": 1.8527079820632935, + "learning_rate": 2.3212835093419988e-06, + "loss": 2.1005, + "step": 1608 + }, + { + "epoch": 1.7632876712328767, + "grad_norm": 2.142939567565918, + "learning_rate": 2.319252640129976e-06, + "loss": 2.1499, + "step": 1609 + }, + { + "epoch": 1.7643835616438355, + "grad_norm": 2.040922164916992, + "learning_rate": 2.3172217709179533e-06, + "loss": 2.0997, + "step": 1610 + }, + { + "epoch": 1.7643835616438355, + "eval_loss": 2.2691540718078613, + "eval_runtime": 42.4134, + "eval_samples_per_second": 23.577, + "eval_steps_per_second": 0.99, + "step": 1610 + }, + { + "epoch": 1.7654794520547945, + "grad_norm": 1.988370656967163, + "learning_rate": 2.3151909017059306e-06, + "loss": 2.1795, + "step": 1611 + }, + { + "epoch": 1.7665753424657535, + "grad_norm": 1.996656060218811, + "learning_rate": 2.313160032493908e-06, + "loss": 2.1536, + "step": 1612 + }, + { + "epoch": 1.7676712328767124, + "grad_norm": 2.04560923576355, + "learning_rate": 2.311129163281885e-06, + "loss": 2.0781, + "step": 1613 + }, + { + "epoch": 1.7687671232876712, + "grad_norm": 1.8977863788604736, + "learning_rate": 2.309098294069862e-06, + "loss": 2.1312, + "step": 1614 + }, + { + "epoch": 1.7698630136986302, + "grad_norm": 2.026904821395874, + "learning_rate": 2.3070674248578393e-06, + "loss": 2.0998, + "step": 1615 + }, + { + "epoch": 1.770958904109589, + "grad_norm": 1.8714247941970825, + "learning_rate": 2.3050365556458166e-06, + "loss": 2.0875, + "step": 1616 + }, + { + "epoch": 1.7720547945205478, + "grad_norm": 2.11859130859375, + "learning_rate": 2.303005686433794e-06, + "loss": 2.1361, + "step": 1617 + }, + { + "epoch": 1.7731506849315068, + "grad_norm": 2.0592799186706543, + "learning_rate": 2.300974817221771e-06, + "loss": 2.1394, + "step": 1618 + }, + { + "epoch": 1.7742465753424659, + "grad_norm": 2.0358338356018066, + "learning_rate": 2.2989439480097484e-06, + "loss": 2.104, + "step": 1619 + }, + { + "epoch": 1.7753424657534247, + "grad_norm": 2.0237109661102295, + "learning_rate": 2.2969130787977257e-06, + "loss": 2.0823, + "step": 1620 + }, + { + "epoch": 1.7753424657534247, + "eval_loss": 2.27001690864563, + "eval_runtime": 43.9168, + "eval_samples_per_second": 22.77, + "eval_steps_per_second": 0.956, + "step": 1620 + }, + { + "epoch": 1.7764383561643835, + "grad_norm": 1.8729594945907593, + "learning_rate": 2.294882209585703e-06, + "loss": 2.1243, + "step": 1621 + }, + { + "epoch": 1.7775342465753425, + "grad_norm": 1.9390734434127808, + "learning_rate": 2.2928513403736802e-06, + "loss": 2.1469, + "step": 1622 + }, + { + "epoch": 1.7786301369863013, + "grad_norm": 2.0409750938415527, + "learning_rate": 2.2908204711616575e-06, + "loss": 2.1062, + "step": 1623 + }, + { + "epoch": 1.7797260273972602, + "grad_norm": 1.9116135835647583, + "learning_rate": 2.2887896019496348e-06, + "loss": 2.1208, + "step": 1624 + }, + { + "epoch": 1.7808219178082192, + "grad_norm": 1.9144036769866943, + "learning_rate": 2.286758732737612e-06, + "loss": 2.1628, + "step": 1625 + }, + { + "epoch": 1.7819178082191782, + "grad_norm": 1.8522213697433472, + "learning_rate": 2.284727863525589e-06, + "loss": 2.1168, + "step": 1626 + }, + { + "epoch": 1.783013698630137, + "grad_norm": 2.087541103363037, + "learning_rate": 2.282696994313566e-06, + "loss": 2.1195, + "step": 1627 + }, + { + "epoch": 1.7841095890410958, + "grad_norm": 2.0399985313415527, + "learning_rate": 2.2806661251015434e-06, + "loss": 2.1557, + "step": 1628 + }, + { + "epoch": 1.7852054794520549, + "grad_norm": 1.884964108467102, + "learning_rate": 2.2786352558895207e-06, + "loss": 2.1408, + "step": 1629 + }, + { + "epoch": 1.7863013698630137, + "grad_norm": 1.901802897453308, + "learning_rate": 2.276604386677498e-06, + "loss": 2.1145, + "step": 1630 + }, + { + "epoch": 1.7863013698630137, + "eval_loss": 2.2690439224243164, + "eval_runtime": 44.8662, + "eval_samples_per_second": 22.289, + "eval_steps_per_second": 0.936, + "step": 1630 + }, + { + "epoch": 1.7873972602739725, + "grad_norm": 1.8818615674972534, + "learning_rate": 2.2745735174654753e-06, + "loss": 2.0879, + "step": 1631 + }, + { + "epoch": 1.7884931506849315, + "grad_norm": 1.9345687627792358, + "learning_rate": 2.2725426482534526e-06, + "loss": 2.1029, + "step": 1632 + }, + { + "epoch": 1.7895890410958906, + "grad_norm": 1.868314504623413, + "learning_rate": 2.27051177904143e-06, + "loss": 2.1435, + "step": 1633 + }, + { + "epoch": 1.7906849315068492, + "grad_norm": 1.8868515491485596, + "learning_rate": 2.268480909829407e-06, + "loss": 2.1294, + "step": 1634 + }, + { + "epoch": 1.7917808219178082, + "grad_norm": 1.9929358959197998, + "learning_rate": 2.2664500406173844e-06, + "loss": 2.1165, + "step": 1635 + }, + { + "epoch": 1.7928767123287672, + "grad_norm": 1.9766895771026611, + "learning_rate": 2.2644191714053617e-06, + "loss": 2.1327, + "step": 1636 + }, + { + "epoch": 1.793972602739726, + "grad_norm": 2.03336238861084, + "learning_rate": 2.262388302193339e-06, + "loss": 2.1473, + "step": 1637 + }, + { + "epoch": 1.7950684931506848, + "grad_norm": 1.8569762706756592, + "learning_rate": 2.260357432981316e-06, + "loss": 2.125, + "step": 1638 + }, + { + "epoch": 1.7961643835616439, + "grad_norm": 1.9522744417190552, + "learning_rate": 2.2583265637692935e-06, + "loss": 2.1238, + "step": 1639 + }, + { + "epoch": 1.7972602739726027, + "grad_norm": 1.9175306558609009, + "learning_rate": 2.2562956945572708e-06, + "loss": 2.1332, + "step": 1640 + }, + { + "epoch": 1.7972602739726027, + "eval_loss": 2.2702176570892334, + "eval_runtime": 44.1369, + "eval_samples_per_second": 22.657, + "eval_steps_per_second": 0.952, + "step": 1640 + }, + { + "epoch": 1.7983561643835615, + "grad_norm": 2.109614849090576, + "learning_rate": 2.254264825345248e-06, + "loss": 2.1252, + "step": 1641 + }, + { + "epoch": 1.7994520547945205, + "grad_norm": 2.017613410949707, + "learning_rate": 2.2522339561332253e-06, + "loss": 2.1093, + "step": 1642 + }, + { + "epoch": 1.8005479452054796, + "grad_norm": 2.000864028930664, + "learning_rate": 2.2502030869212026e-06, + "loss": 2.1243, + "step": 1643 + }, + { + "epoch": 1.8016438356164384, + "grad_norm": 1.9952439069747925, + "learning_rate": 2.24817221770918e-06, + "loss": 2.1213, + "step": 1644 + }, + { + "epoch": 1.8027397260273972, + "grad_norm": 2.0456748008728027, + "learning_rate": 2.246141348497157e-06, + "loss": 2.1621, + "step": 1645 + }, + { + "epoch": 1.8038356164383562, + "grad_norm": 1.9011833667755127, + "learning_rate": 2.2441104792851344e-06, + "loss": 2.1426, + "step": 1646 + }, + { + "epoch": 1.804931506849315, + "grad_norm": 1.9874658584594727, + "learning_rate": 2.2420796100731117e-06, + "loss": 2.1085, + "step": 1647 + }, + { + "epoch": 1.8060273972602738, + "grad_norm": 2.0237057209014893, + "learning_rate": 2.240048740861089e-06, + "loss": 2.1394, + "step": 1648 + }, + { + "epoch": 1.8071232876712329, + "grad_norm": 2.0127220153808594, + "learning_rate": 2.2380178716490662e-06, + "loss": 2.1474, + "step": 1649 + }, + { + "epoch": 1.808219178082192, + "grad_norm": 2.051722764968872, + "learning_rate": 2.2359870024370435e-06, + "loss": 2.0444, + "step": 1650 + }, + { + "epoch": 1.808219178082192, + "eval_loss": 2.2701034545898438, + "eval_runtime": 44.4338, + "eval_samples_per_second": 22.505, + "eval_steps_per_second": 0.945, + "step": 1650 + }, + { + "epoch": 1.8093150684931507, + "grad_norm": 2.0204219818115234, + "learning_rate": 2.2339561332250204e-06, + "loss": 2.1445, + "step": 1651 + }, + { + "epoch": 1.8104109589041095, + "grad_norm": 2.106220006942749, + "learning_rate": 2.2319252640129976e-06, + "loss": 2.153, + "step": 1652 + }, + { + "epoch": 1.8115068493150686, + "grad_norm": 1.8632844686508179, + "learning_rate": 2.229894394800975e-06, + "loss": 2.137, + "step": 1653 + }, + { + "epoch": 1.8126027397260274, + "grad_norm": 2.377120018005371, + "learning_rate": 2.227863525588952e-06, + "loss": 2.0903, + "step": 1654 + }, + { + "epoch": 1.8136986301369862, + "grad_norm": 2.1345274448394775, + "learning_rate": 2.2258326563769295e-06, + "loss": 2.066, + "step": 1655 + }, + { + "epoch": 1.8147945205479452, + "grad_norm": 1.8935933113098145, + "learning_rate": 2.2238017871649067e-06, + "loss": 2.083, + "step": 1656 + }, + { + "epoch": 1.8158904109589042, + "grad_norm": 1.8766530752182007, + "learning_rate": 2.221770917952884e-06, + "loss": 2.0792, + "step": 1657 + }, + { + "epoch": 1.816986301369863, + "grad_norm": 2.24202823638916, + "learning_rate": 2.2197400487408613e-06, + "loss": 2.1276, + "step": 1658 + }, + { + "epoch": 1.8180821917808219, + "grad_norm": 2.168358325958252, + "learning_rate": 2.2177091795288386e-06, + "loss": 2.1497, + "step": 1659 + }, + { + "epoch": 1.819178082191781, + "grad_norm": 2.052833080291748, + "learning_rate": 2.215678310316816e-06, + "loss": 2.0853, + "step": 1660 + }, + { + "epoch": 1.819178082191781, + "eval_loss": 2.270122528076172, + "eval_runtime": 44.1793, + "eval_samples_per_second": 22.635, + "eval_steps_per_second": 0.951, + "step": 1660 + }, + { + "epoch": 1.8202739726027397, + "grad_norm": 2.09708571434021, + "learning_rate": 2.213647441104793e-06, + "loss": 2.1151, + "step": 1661 + }, + { + "epoch": 1.8213698630136985, + "grad_norm": 1.928503155708313, + "learning_rate": 2.2116165718927704e-06, + "loss": 2.1234, + "step": 1662 + }, + { + "epoch": 1.8224657534246576, + "grad_norm": 1.9545074701309204, + "learning_rate": 2.2095857026807473e-06, + "loss": 2.1134, + "step": 1663 + }, + { + "epoch": 1.8235616438356166, + "grad_norm": 2.1161389350891113, + "learning_rate": 2.2075548334687245e-06, + "loss": 2.1347, + "step": 1664 + }, + { + "epoch": 1.8246575342465754, + "grad_norm": 1.9292371273040771, + "learning_rate": 2.205523964256702e-06, + "loss": 2.0911, + "step": 1665 + }, + { + "epoch": 1.8257534246575342, + "grad_norm": 1.8706191778182983, + "learning_rate": 2.203493095044679e-06, + "loss": 2.0917, + "step": 1666 + }, + { + "epoch": 1.8268493150684932, + "grad_norm": 1.9825111627578735, + "learning_rate": 2.2014622258326564e-06, + "loss": 2.125, + "step": 1667 + }, + { + "epoch": 1.827945205479452, + "grad_norm": 2.0262181758880615, + "learning_rate": 2.1994313566206336e-06, + "loss": 2.153, + "step": 1668 + }, + { + "epoch": 1.8290410958904109, + "grad_norm": 2.0071237087249756, + "learning_rate": 2.197400487408611e-06, + "loss": 2.1625, + "step": 1669 + }, + { + "epoch": 1.83013698630137, + "grad_norm": 1.940076231956482, + "learning_rate": 2.195369618196588e-06, + "loss": 2.1052, + "step": 1670 + }, + { + "epoch": 1.83013698630137, + "eval_loss": 2.2711451053619385, + "eval_runtime": 43.9311, + "eval_samples_per_second": 22.763, + "eval_steps_per_second": 0.956, + "step": 1670 + }, + { + "epoch": 1.831232876712329, + "grad_norm": 2.103487014770508, + "learning_rate": 2.1933387489845655e-06, + "loss": 2.087, + "step": 1671 + }, + { + "epoch": 1.8323287671232875, + "grad_norm": 1.935276985168457, + "learning_rate": 2.1913078797725427e-06, + "loss": 2.1196, + "step": 1672 + }, + { + "epoch": 1.8334246575342465, + "grad_norm": 1.987322449684143, + "learning_rate": 2.18927701056052e-06, + "loss": 2.1354, + "step": 1673 + }, + { + "epoch": 1.8345205479452056, + "grad_norm": 2.0403592586517334, + "learning_rate": 2.1872461413484973e-06, + "loss": 2.1715, + "step": 1674 + }, + { + "epoch": 1.8356164383561644, + "grad_norm": 2.0403592586517334, + "learning_rate": 2.1872461413484973e-06, + "loss": 2.1153, + "step": 1675 + }, + { + "epoch": 1.8367123287671232, + "grad_norm": 2.160753011703491, + "learning_rate": 2.1852152721364746e-06, + "loss": 2.1105, + "step": 1676 + }, + { + "epoch": 1.8378082191780822, + "grad_norm": 1.8320366144180298, + "learning_rate": 2.183184402924452e-06, + "loss": 2.0853, + "step": 1677 + }, + { + "epoch": 1.838904109589041, + "grad_norm": 1.9792540073394775, + "learning_rate": 2.181153533712429e-06, + "loss": 2.1478, + "step": 1678 + }, + { + "epoch": 1.8399999999999999, + "grad_norm": 2.0344042778015137, + "learning_rate": 2.1791226645004064e-06, + "loss": 2.1177, + "step": 1679 + }, + { + "epoch": 1.841095890410959, + "grad_norm": 1.8672499656677246, + "learning_rate": 2.1770917952883837e-06, + "loss": 2.1098, + "step": 1680 + }, + { + "epoch": 1.841095890410959, + "eval_loss": 2.2701213359832764, + "eval_runtime": 43.0484, + "eval_samples_per_second": 23.23, + "eval_steps_per_second": 0.976, + "step": 1680 + }, + { + "epoch": 1.842191780821918, + "grad_norm": 1.9291775226593018, + "learning_rate": 2.175060926076361e-06, + "loss": 2.1659, + "step": 1681 + }, + { + "epoch": 1.8432876712328767, + "grad_norm": 1.9650444984436035, + "learning_rate": 2.1730300568643382e-06, + "loss": 2.1764, + "step": 1682 + }, + { + "epoch": 1.8443835616438355, + "grad_norm": 1.904618501663208, + "learning_rate": 2.1709991876523155e-06, + "loss": 2.1047, + "step": 1683 + }, + { + "epoch": 1.8454794520547946, + "grad_norm": 2.0084192752838135, + "learning_rate": 2.1689683184402928e-06, + "loss": 2.0719, + "step": 1684 + }, + { + "epoch": 1.8465753424657534, + "grad_norm": 2.0233888626098633, + "learning_rate": 2.16693744922827e-06, + "loss": 2.1482, + "step": 1685 + }, + { + "epoch": 1.8476712328767122, + "grad_norm": 1.9512053728103638, + "learning_rate": 2.1649065800162473e-06, + "loss": 2.1116, + "step": 1686 + }, + { + "epoch": 1.8487671232876712, + "grad_norm": 1.9261832237243652, + "learning_rate": 2.1628757108042246e-06, + "loss": 2.1025, + "step": 1687 + }, + { + "epoch": 1.8498630136986303, + "grad_norm": 2.0704665184020996, + "learning_rate": 2.160844841592202e-06, + "loss": 2.0943, + "step": 1688 + }, + { + "epoch": 1.850958904109589, + "grad_norm": 1.8882733583450317, + "learning_rate": 2.1588139723801787e-06, + "loss": 2.09, + "step": 1689 + }, + { + "epoch": 1.8520547945205479, + "grad_norm": 1.9136085510253906, + "learning_rate": 2.156783103168156e-06, + "loss": 2.1074, + "step": 1690 + }, + { + "epoch": 1.8520547945205479, + "eval_loss": 2.2687957286834717, + "eval_runtime": 44.1571, + "eval_samples_per_second": 22.646, + "eval_steps_per_second": 0.951, + "step": 1690 + }, + { + "epoch": 1.853150684931507, + "grad_norm": 1.9338082075119019, + "learning_rate": 2.1547522339561333e-06, + "loss": 2.0945, + "step": 1691 + }, + { + "epoch": 1.8542465753424657, + "grad_norm": 1.862281084060669, + "learning_rate": 2.1527213647441106e-06, + "loss": 2.1355, + "step": 1692 + }, + { + "epoch": 1.8553424657534245, + "grad_norm": 1.8184510469436646, + "learning_rate": 2.150690495532088e-06, + "loss": 2.1628, + "step": 1693 + }, + { + "epoch": 1.8564383561643836, + "grad_norm": 1.9667037725448608, + "learning_rate": 2.148659626320065e-06, + "loss": 2.1308, + "step": 1694 + }, + { + "epoch": 1.8575342465753426, + "grad_norm": 2.0214083194732666, + "learning_rate": 2.1466287571080424e-06, + "loss": 2.1363, + "step": 1695 + }, + { + "epoch": 1.8586301369863014, + "grad_norm": 1.8136624097824097, + "learning_rate": 2.1445978878960197e-06, + "loss": 2.1074, + "step": 1696 + }, + { + "epoch": 1.8597260273972602, + "grad_norm": 2.1120195388793945, + "learning_rate": 2.142567018683997e-06, + "loss": 2.1115, + "step": 1697 + }, + { + "epoch": 1.8608219178082193, + "grad_norm": 1.9148973226547241, + "learning_rate": 2.140536149471974e-06, + "loss": 2.1554, + "step": 1698 + }, + { + "epoch": 1.861917808219178, + "grad_norm": 1.9684176445007324, + "learning_rate": 2.1385052802599515e-06, + "loss": 2.0944, + "step": 1699 + }, + { + "epoch": 1.8630136986301369, + "grad_norm": 2.0816597938537598, + "learning_rate": 2.1364744110479288e-06, + "loss": 2.1063, + "step": 1700 + }, + { + "epoch": 1.8630136986301369, + "eval_loss": 2.2695536613464355, + "eval_runtime": 48.9529, + "eval_samples_per_second": 20.428, + "eval_steps_per_second": 0.858, + "step": 1700 + }, + { + "epoch": 1.864109589041096, + "grad_norm": 1.9451358318328857, + "learning_rate": 2.134443541835906e-06, + "loss": 2.1521, + "step": 1701 + }, + { + "epoch": 1.865205479452055, + "grad_norm": 1.9210634231567383, + "learning_rate": 2.132412672623883e-06, + "loss": 2.1216, + "step": 1702 + }, + { + "epoch": 1.8663013698630135, + "grad_norm": 2.0976240634918213, + "learning_rate": 2.13038180341186e-06, + "loss": 2.0947, + "step": 1703 + }, + { + "epoch": 1.8673972602739726, + "grad_norm": 1.9917279481887817, + "learning_rate": 2.1283509341998374e-06, + "loss": 2.1267, + "step": 1704 + }, + { + "epoch": 1.8684931506849316, + "grad_norm": 2.1310553550720215, + "learning_rate": 2.1263200649878147e-06, + "loss": 2.1566, + "step": 1705 + }, + { + "epoch": 1.8695890410958904, + "grad_norm": 2.0364415645599365, + "learning_rate": 2.124289195775792e-06, + "loss": 2.1077, + "step": 1706 + }, + { + "epoch": 1.8706849315068492, + "grad_norm": 1.8810032606124878, + "learning_rate": 2.1222583265637693e-06, + "loss": 2.1061, + "step": 1707 + }, + { + "epoch": 1.8717808219178083, + "grad_norm": 2.034438371658325, + "learning_rate": 2.1202274573517465e-06, + "loss": 2.0523, + "step": 1708 + }, + { + "epoch": 1.872876712328767, + "grad_norm": 2.0231120586395264, + "learning_rate": 2.118196588139724e-06, + "loss": 2.0829, + "step": 1709 + }, + { + "epoch": 1.8739726027397259, + "grad_norm": 2.0092573165893555, + "learning_rate": 2.116165718927701e-06, + "loss": 2.1101, + "step": 1710 + }, + { + "epoch": 1.8739726027397259, + "eval_loss": 2.2675492763519287, + "eval_runtime": 43.0128, + "eval_samples_per_second": 23.249, + "eval_steps_per_second": 0.976, + "step": 1710 + }, + { + "epoch": 1.875068493150685, + "grad_norm": 1.884216070175171, + "learning_rate": 2.1141348497156784e-06, + "loss": 2.1249, + "step": 1711 + }, + { + "epoch": 1.876164383561644, + "grad_norm": 1.954170823097229, + "learning_rate": 2.1121039805036556e-06, + "loss": 2.0915, + "step": 1712 + }, + { + "epoch": 1.8772602739726028, + "grad_norm": 2.0208632946014404, + "learning_rate": 2.110073111291633e-06, + "loss": 2.0824, + "step": 1713 + }, + { + "epoch": 1.8783561643835616, + "grad_norm": 1.9104353189468384, + "learning_rate": 2.10804224207961e-06, + "loss": 2.1265, + "step": 1714 + }, + { + "epoch": 1.8794520547945206, + "grad_norm": 1.945338249206543, + "learning_rate": 2.1060113728675875e-06, + "loss": 2.1403, + "step": 1715 + }, + { + "epoch": 1.8805479452054794, + "grad_norm": 2.006586790084839, + "learning_rate": 2.1039805036555647e-06, + "loss": 2.1439, + "step": 1716 + }, + { + "epoch": 1.8816438356164382, + "grad_norm": 1.941493034362793, + "learning_rate": 2.101949634443542e-06, + "loss": 2.0415, + "step": 1717 + }, + { + "epoch": 1.8827397260273973, + "grad_norm": 2.146116018295288, + "learning_rate": 2.0999187652315193e-06, + "loss": 2.1118, + "step": 1718 + }, + { + "epoch": 1.8838356164383563, + "grad_norm": 2.0794291496276855, + "learning_rate": 2.0978878960194966e-06, + "loss": 2.0731, + "step": 1719 + }, + { + "epoch": 1.884931506849315, + "grad_norm": 2.1100215911865234, + "learning_rate": 2.095857026807474e-06, + "loss": 2.1303, + "step": 1720 + }, + { + "epoch": 1.884931506849315, + "eval_loss": 2.267274856567383, + "eval_runtime": 48.0078, + "eval_samples_per_second": 20.83, + "eval_steps_per_second": 0.875, + "step": 1720 + }, + { + "epoch": 1.886027397260274, + "grad_norm": 1.9509919881820679, + "learning_rate": 2.093826157595451e-06, + "loss": 2.1191, + "step": 1721 + }, + { + "epoch": 1.887123287671233, + "grad_norm": 2.0161266326904297, + "learning_rate": 2.0917952883834284e-06, + "loss": 2.104, + "step": 1722 + }, + { + "epoch": 1.8882191780821918, + "grad_norm": 2.07547664642334, + "learning_rate": 2.0897644191714057e-06, + "loss": 2.1299, + "step": 1723 + }, + { + "epoch": 1.8893150684931506, + "grad_norm": 2.1406030654907227, + "learning_rate": 2.087733549959383e-06, + "loss": 2.1385, + "step": 1724 + }, + { + "epoch": 1.8904109589041096, + "grad_norm": 1.9793925285339355, + "learning_rate": 2.0857026807473602e-06, + "loss": 2.0814, + "step": 1725 + }, + { + "epoch": 1.8915068493150686, + "grad_norm": 1.9256374835968018, + "learning_rate": 2.0836718115353375e-06, + "loss": 2.0708, + "step": 1726 + }, + { + "epoch": 1.8926027397260274, + "grad_norm": 1.960571527481079, + "learning_rate": 2.0816409423233144e-06, + "loss": 2.156, + "step": 1727 + }, + { + "epoch": 1.8936986301369862, + "grad_norm": 2.1507158279418945, + "learning_rate": 2.0796100731112916e-06, + "loss": 2.0958, + "step": 1728 + }, + { + "epoch": 1.8947945205479453, + "grad_norm": 2.2358651161193848, + "learning_rate": 2.077579203899269e-06, + "loss": 2.11, + "step": 1729 + }, + { + "epoch": 1.895890410958904, + "grad_norm": 2.0311756134033203, + "learning_rate": 2.075548334687246e-06, + "loss": 2.072, + "step": 1730 + }, + { + "epoch": 1.895890410958904, + "eval_loss": 2.2663629055023193, + "eval_runtime": 50.5422, + "eval_samples_per_second": 19.785, + "eval_steps_per_second": 0.831, + "step": 1730 + }, + { + "epoch": 1.896986301369863, + "grad_norm": 1.98939049243927, + "learning_rate": 2.0735174654752235e-06, + "loss": 2.1327, + "step": 1731 + }, + { + "epoch": 1.898082191780822, + "grad_norm": 2.009493350982666, + "learning_rate": 2.0714865962632007e-06, + "loss": 2.0798, + "step": 1732 + }, + { + "epoch": 1.899178082191781, + "grad_norm": 2.1011271476745605, + "learning_rate": 2.069455727051178e-06, + "loss": 2.1194, + "step": 1733 + }, + { + "epoch": 1.9002739726027398, + "grad_norm": 1.9027225971221924, + "learning_rate": 2.0674248578391553e-06, + "loss": 2.0955, + "step": 1734 + }, + { + "epoch": 1.9013698630136986, + "grad_norm": 1.9784867763519287, + "learning_rate": 2.0653939886271326e-06, + "loss": 2.1012, + "step": 1735 + }, + { + "epoch": 1.9024657534246576, + "grad_norm": 2.0302555561065674, + "learning_rate": 2.06336311941511e-06, + "loss": 2.1472, + "step": 1736 + }, + { + "epoch": 1.9035616438356164, + "grad_norm": 1.81975257396698, + "learning_rate": 2.061332250203087e-06, + "loss": 2.1062, + "step": 1737 + }, + { + "epoch": 1.9046575342465752, + "grad_norm": 1.9868931770324707, + "learning_rate": 2.0593013809910644e-06, + "loss": 2.0829, + "step": 1738 + }, + { + "epoch": 1.9057534246575343, + "grad_norm": 2.001765012741089, + "learning_rate": 2.0572705117790417e-06, + "loss": 2.1113, + "step": 1739 + }, + { + "epoch": 1.9068493150684933, + "grad_norm": 1.891752004623413, + "learning_rate": 2.055239642567019e-06, + "loss": 2.1311, + "step": 1740 + }, + { + "epoch": 1.9068493150684933, + "eval_loss": 2.2680604457855225, + "eval_runtime": 50.0795, + "eval_samples_per_second": 19.968, + "eval_steps_per_second": 0.839, + "step": 1740 + }, + { + "epoch": 1.907945205479452, + "grad_norm": 1.978259563446045, + "learning_rate": 2.0532087733549962e-06, + "loss": 2.1187, + "step": 1741 + }, + { + "epoch": 1.909041095890411, + "grad_norm": 1.8946164846420288, + "learning_rate": 2.0511779041429735e-06, + "loss": 2.1189, + "step": 1742 + }, + { + "epoch": 1.91013698630137, + "grad_norm": 1.8812536001205444, + "learning_rate": 2.0491470349309508e-06, + "loss": 2.0895, + "step": 1743 + }, + { + "epoch": 1.9112328767123288, + "grad_norm": 1.990494728088379, + "learning_rate": 2.047116165718928e-06, + "loss": 2.1192, + "step": 1744 + }, + { + "epoch": 1.9123287671232876, + "grad_norm": 2.0288472175598145, + "learning_rate": 2.0450852965069053e-06, + "loss": 2.1182, + "step": 1745 + }, + { + "epoch": 1.9134246575342466, + "grad_norm": 2.0876214504241943, + "learning_rate": 2.0430544272948826e-06, + "loss": 2.1546, + "step": 1746 + }, + { + "epoch": 1.9145205479452054, + "grad_norm": 1.9842756986618042, + "learning_rate": 2.04102355808286e-06, + "loss": 2.1126, + "step": 1747 + }, + { + "epoch": 1.9156164383561642, + "grad_norm": 2.0303354263305664, + "learning_rate": 2.038992688870837e-06, + "loss": 2.0771, + "step": 1748 + }, + { + "epoch": 1.9167123287671233, + "grad_norm": 2.0054662227630615, + "learning_rate": 2.0369618196588144e-06, + "loss": 2.0895, + "step": 1749 + }, + { + "epoch": 1.9178082191780823, + "grad_norm": 1.9232923984527588, + "learning_rate": 2.0349309504467917e-06, + "loss": 2.0952, + "step": 1750 + }, + { + "epoch": 1.9178082191780823, + "eval_loss": 2.266294240951538, + "eval_runtime": 53.7831, + "eval_samples_per_second": 18.593, + "eval_steps_per_second": 0.781, + "step": 1750 + }, + { + "epoch": 1.9189041095890411, + "grad_norm": 1.9103394746780396, + "learning_rate": 2.032900081234769e-06, + "loss": 2.1324, + "step": 1751 + }, + { + "epoch": 1.92, + "grad_norm": 2.0642080307006836, + "learning_rate": 2.030869212022746e-06, + "loss": 2.1495, + "step": 1752 + }, + { + "epoch": 1.921095890410959, + "grad_norm": 1.9586466550827026, + "learning_rate": 2.028838342810723e-06, + "loss": 2.1036, + "step": 1753 + }, + { + "epoch": 1.9221917808219178, + "grad_norm": 2.024435043334961, + "learning_rate": 2.0268074735987004e-06, + "loss": 2.1049, + "step": 1754 + }, + { + "epoch": 1.9232876712328766, + "grad_norm": 1.9115983247756958, + "learning_rate": 2.0247766043866777e-06, + "loss": 2.0889, + "step": 1755 + }, + { + "epoch": 1.9243835616438356, + "grad_norm": 1.9067517518997192, + "learning_rate": 2.022745735174655e-06, + "loss": 2.1432, + "step": 1756 + }, + { + "epoch": 1.9254794520547946, + "grad_norm": 2.0182650089263916, + "learning_rate": 2.020714865962632e-06, + "loss": 2.1011, + "step": 1757 + }, + { + "epoch": 1.9265753424657535, + "grad_norm": 2.0648891925811768, + "learning_rate": 2.0186839967506095e-06, + "loss": 2.1266, + "step": 1758 + }, + { + "epoch": 1.9276712328767123, + "grad_norm": 2.0217037200927734, + "learning_rate": 2.0166531275385868e-06, + "loss": 2.1206, + "step": 1759 + }, + { + "epoch": 1.9287671232876713, + "grad_norm": 1.9545989036560059, + "learning_rate": 2.014622258326564e-06, + "loss": 2.0822, + "step": 1760 + }, + { + "epoch": 1.9287671232876713, + "eval_loss": 2.2664599418640137, + "eval_runtime": 53.4817, + "eval_samples_per_second": 18.698, + "eval_steps_per_second": 0.785, + "step": 1760 + }, + { + "epoch": 1.9298630136986301, + "grad_norm": 2.047677516937256, + "learning_rate": 2.0125913891145413e-06, + "loss": 2.1169, + "step": 1761 + }, + { + "epoch": 1.930958904109589, + "grad_norm": 2.170466184616089, + "learning_rate": 2.0105605199025186e-06, + "loss": 2.1211, + "step": 1762 + }, + { + "epoch": 1.932054794520548, + "grad_norm": 1.9144518375396729, + "learning_rate": 2.008529650690496e-06, + "loss": 2.0912, + "step": 1763 + }, + { + "epoch": 1.933150684931507, + "grad_norm": 2.004936933517456, + "learning_rate": 2.0064987814784727e-06, + "loss": 2.136, + "step": 1764 + }, + { + "epoch": 1.9342465753424658, + "grad_norm": 1.8312169313430786, + "learning_rate": 2.00446791226645e-06, + "loss": 2.1151, + "step": 1765 + }, + { + "epoch": 1.9353424657534246, + "grad_norm": 1.9144090414047241, + "learning_rate": 2.0024370430544273e-06, + "loss": 2.0835, + "step": 1766 + }, + { + "epoch": 1.9364383561643836, + "grad_norm": 2.020298719406128, + "learning_rate": 2.0004061738424045e-06, + "loss": 2.1203, + "step": 1767 + }, + { + "epoch": 1.9375342465753425, + "grad_norm": 1.9231114387512207, + "learning_rate": 1.998375304630382e-06, + "loss": 2.1425, + "step": 1768 + }, + { + "epoch": 1.9386301369863013, + "grad_norm": 1.8474186658859253, + "learning_rate": 1.996344435418359e-06, + "loss": 2.1071, + "step": 1769 + }, + { + "epoch": 1.9397260273972603, + "grad_norm": 2.171400785446167, + "learning_rate": 1.9943135662063364e-06, + "loss": 2.1344, + "step": 1770 + }, + { + "epoch": 1.9397260273972603, + "eval_loss": 2.266789436340332, + "eval_runtime": 50.4218, + "eval_samples_per_second": 19.833, + "eval_steps_per_second": 0.833, + "step": 1770 + }, + { + "epoch": 1.9408219178082193, + "grad_norm": 1.929115891456604, + "learning_rate": 1.9922826969943136e-06, + "loss": 2.1205, + "step": 1771 + }, + { + "epoch": 1.941917808219178, + "grad_norm": 1.961046814918518, + "learning_rate": 1.990251827782291e-06, + "loss": 2.1396, + "step": 1772 + }, + { + "epoch": 1.943013698630137, + "grad_norm": 2.056424617767334, + "learning_rate": 1.988220958570268e-06, + "loss": 2.125, + "step": 1773 + }, + { + "epoch": 1.944109589041096, + "grad_norm": 1.9814075231552124, + "learning_rate": 1.9861900893582455e-06, + "loss": 2.1297, + "step": 1774 + }, + { + "epoch": 1.9452054794520548, + "grad_norm": 1.937506079673767, + "learning_rate": 1.9841592201462227e-06, + "loss": 2.1288, + "step": 1775 + }, + { + "epoch": 1.9463013698630136, + "grad_norm": 1.902376413345337, + "learning_rate": 1.9821283509342e-06, + "loss": 2.0782, + "step": 1776 + }, + { + "epoch": 1.9473972602739726, + "grad_norm": 2.0806174278259277, + "learning_rate": 1.9800974817221773e-06, + "loss": 2.0778, + "step": 1777 + }, + { + "epoch": 1.9484931506849315, + "grad_norm": 2.0370235443115234, + "learning_rate": 1.9780666125101546e-06, + "loss": 2.1304, + "step": 1778 + }, + { + "epoch": 1.9495890410958903, + "grad_norm": 1.980531930923462, + "learning_rate": 1.976035743298132e-06, + "loss": 2.0989, + "step": 1779 + }, + { + "epoch": 1.9506849315068493, + "grad_norm": 1.96549654006958, + "learning_rate": 1.974004874086109e-06, + "loss": 2.1556, + "step": 1780 + }, + { + "epoch": 1.9506849315068493, + "eval_loss": 2.266958475112915, + "eval_runtime": 52.3331, + "eval_samples_per_second": 19.108, + "eval_steps_per_second": 0.803, + "step": 1780 + }, + { + "epoch": 1.9517808219178083, + "grad_norm": 2.1921818256378174, + "learning_rate": 1.9719740048740864e-06, + "loss": 2.0666, + "step": 1781 + }, + { + "epoch": 1.9528767123287671, + "grad_norm": 1.9451124668121338, + "learning_rate": 1.9699431356620637e-06, + "loss": 2.1078, + "step": 1782 + }, + { + "epoch": 1.953972602739726, + "grad_norm": 1.9163907766342163, + "learning_rate": 1.967912266450041e-06, + "loss": 2.0955, + "step": 1783 + }, + { + "epoch": 1.955068493150685, + "grad_norm": 2.1186892986297607, + "learning_rate": 1.9658813972380182e-06, + "loss": 2.0826, + "step": 1784 + }, + { + "epoch": 1.9561643835616438, + "grad_norm": 2.1072819232940674, + "learning_rate": 1.9638505280259955e-06, + "loss": 2.1869, + "step": 1785 + }, + { + "epoch": 1.9572602739726026, + "grad_norm": 1.9530322551727295, + "learning_rate": 1.9618196588139728e-06, + "loss": 2.0983, + "step": 1786 + }, + { + "epoch": 1.9583561643835616, + "grad_norm": 2.0034027099609375, + "learning_rate": 1.95978878960195e-06, + "loss": 2.1121, + "step": 1787 + }, + { + "epoch": 1.9594520547945207, + "grad_norm": 2.1277832984924316, + "learning_rate": 1.9577579203899273e-06, + "loss": 2.1769, + "step": 1788 + }, + { + "epoch": 1.9605479452054795, + "grad_norm": 1.9420982599258423, + "learning_rate": 1.955727051177904e-06, + "loss": 2.1424, + "step": 1789 + }, + { + "epoch": 1.9616438356164383, + "grad_norm": 1.9175841808319092, + "learning_rate": 1.9536961819658815e-06, + "loss": 2.0957, + "step": 1790 + }, + { + "epoch": 1.9616438356164383, + "eval_loss": 2.267153739929199, + "eval_runtime": 47.5802, + "eval_samples_per_second": 21.017, + "eval_steps_per_second": 0.883, + "step": 1790 + }, + { + "epoch": 1.9627397260273973, + "grad_norm": 1.9897676706314087, + "learning_rate": 1.9516653127538587e-06, + "loss": 2.0809, + "step": 1791 + }, + { + "epoch": 1.9638356164383561, + "grad_norm": 2.0855610370635986, + "learning_rate": 1.949634443541836e-06, + "loss": 2.0924, + "step": 1792 + }, + { + "epoch": 1.964931506849315, + "grad_norm": 2.0065109729766846, + "learning_rate": 1.9476035743298133e-06, + "loss": 2.0819, + "step": 1793 + }, + { + "epoch": 1.966027397260274, + "grad_norm": 1.9256092309951782, + "learning_rate": 1.9455727051177906e-06, + "loss": 2.1389, + "step": 1794 + }, + { + "epoch": 1.967123287671233, + "grad_norm": 1.9086920022964478, + "learning_rate": 1.943541835905768e-06, + "loss": 2.1106, + "step": 1795 + }, + { + "epoch": 1.9682191780821918, + "grad_norm": 2.0217244625091553, + "learning_rate": 1.941510966693745e-06, + "loss": 2.1624, + "step": 1796 + }, + { + "epoch": 1.9693150684931506, + "grad_norm": 2.005068063735962, + "learning_rate": 1.9394800974817224e-06, + "loss": 2.0751, + "step": 1797 + }, + { + "epoch": 1.9704109589041097, + "grad_norm": 2.0949530601501465, + "learning_rate": 1.9374492282696997e-06, + "loss": 2.1447, + "step": 1798 + }, + { + "epoch": 1.9715068493150685, + "grad_norm": 1.8868833780288696, + "learning_rate": 1.935418359057677e-06, + "loss": 2.1568, + "step": 1799 + }, + { + "epoch": 1.9726027397260273, + "grad_norm": 1.889626145362854, + "learning_rate": 1.9333874898456542e-06, + "loss": 2.1152, + "step": 1800 + }, + { + "epoch": 1.9726027397260273, + "eval_loss": 2.268470287322998, + "eval_runtime": 44.7042, + "eval_samples_per_second": 22.369, + "eval_steps_per_second": 0.94, + "step": 1800 + }, + { + "epoch": 1.9736986301369863, + "grad_norm": 1.9561820030212402, + "learning_rate": 1.931356620633631e-06, + "loss": 2.1584, + "step": 1801 + }, + { + "epoch": 1.9747945205479454, + "grad_norm": 1.9774333238601685, + "learning_rate": 1.9293257514216083e-06, + "loss": 2.0726, + "step": 1802 + }, + { + "epoch": 1.9758904109589042, + "grad_norm": 1.9008616209030151, + "learning_rate": 1.9272948822095856e-06, + "loss": 2.1222, + "step": 1803 + }, + { + "epoch": 1.976986301369863, + "grad_norm": 1.9981689453125, + "learning_rate": 1.925264012997563e-06, + "loss": 2.0761, + "step": 1804 + }, + { + "epoch": 1.978082191780822, + "grad_norm": 1.926156997680664, + "learning_rate": 1.92323314378554e-06, + "loss": 2.0962, + "step": 1805 + }, + { + "epoch": 1.9791780821917808, + "grad_norm": 1.9634530544281006, + "learning_rate": 1.9212022745735174e-06, + "loss": 2.1349, + "step": 1806 + }, + { + "epoch": 1.9802739726027396, + "grad_norm": 1.9713245630264282, + "learning_rate": 1.9191714053614947e-06, + "loss": 2.0881, + "step": 1807 + }, + { + "epoch": 1.9813698630136987, + "grad_norm": 1.9258323907852173, + "learning_rate": 1.917140536149472e-06, + "loss": 2.0744, + "step": 1808 + }, + { + "epoch": 1.9824657534246577, + "grad_norm": 1.9645394086837769, + "learning_rate": 1.9151096669374493e-06, + "loss": 2.0701, + "step": 1809 + }, + { + "epoch": 1.9835616438356163, + "grad_norm": 2.145430088043213, + "learning_rate": 1.9130787977254266e-06, + "loss": 2.1372, + "step": 1810 + }, + { + "epoch": 1.9835616438356163, + "eval_loss": 2.269702196121216, + "eval_runtime": 41.513, + "eval_samples_per_second": 24.089, + "eval_steps_per_second": 1.012, + "step": 1810 + }, + { + "epoch": 1.9846575342465753, + "grad_norm": 2.1812174320220947, + "learning_rate": 1.911047928513404e-06, + "loss": 2.1209, + "step": 1811 + }, + { + "epoch": 1.9857534246575343, + "grad_norm": 1.933090329170227, + "learning_rate": 1.909017059301381e-06, + "loss": 2.0905, + "step": 1812 + }, + { + "epoch": 1.9868493150684932, + "grad_norm": 2.0478832721710205, + "learning_rate": 1.9069861900893586e-06, + "loss": 2.0921, + "step": 1813 + }, + { + "epoch": 1.987945205479452, + "grad_norm": 1.9921154975891113, + "learning_rate": 1.9049553208773354e-06, + "loss": 2.1232, + "step": 1814 + }, + { + "epoch": 1.989041095890411, + "grad_norm": 1.8685261011123657, + "learning_rate": 1.9029244516653127e-06, + "loss": 2.0978, + "step": 1815 + }, + { + "epoch": 1.9901369863013698, + "grad_norm": 2.045020818710327, + "learning_rate": 1.90089358245329e-06, + "loss": 2.0614, + "step": 1816 + }, + { + "epoch": 1.9912328767123286, + "grad_norm": 2.059267520904541, + "learning_rate": 1.8988627132412673e-06, + "loss": 2.1098, + "step": 1817 + }, + { + "epoch": 1.9923287671232877, + "grad_norm": 2.0033164024353027, + "learning_rate": 1.8968318440292445e-06, + "loss": 2.135, + "step": 1818 + }, + { + "epoch": 1.9934246575342467, + "grad_norm": 1.8544939756393433, + "learning_rate": 1.8948009748172218e-06, + "loss": 2.0644, + "step": 1819 + }, + { + "epoch": 1.9945205479452055, + "grad_norm": 1.8415113687515259, + "learning_rate": 1.892770105605199e-06, + "loss": 2.1283, + "step": 1820 + }, + { + "epoch": 1.9945205479452055, + "eval_loss": 2.267810344696045, + "eval_runtime": 43.9738, + "eval_samples_per_second": 22.741, + "eval_steps_per_second": 0.955, + "step": 1820 + }, + { + "epoch": 1.9956164383561643, + "grad_norm": 1.90078866481781, + "learning_rate": 1.8907392363931764e-06, + "loss": 2.1274, + "step": 1821 + }, + { + "epoch": 1.9967123287671233, + "grad_norm": 1.8798222541809082, + "learning_rate": 1.8887083671811536e-06, + "loss": 2.1014, + "step": 1822 + }, + { + "epoch": 1.9978082191780822, + "grad_norm": 1.9333341121673584, + "learning_rate": 1.886677497969131e-06, + "loss": 2.1165, + "step": 1823 + }, + { + "epoch": 1.998904109589041, + "grad_norm": 1.9861575365066528, + "learning_rate": 1.8846466287571082e-06, + "loss": 2.036, + "step": 1824 + }, + { + "epoch": 2.0, + "grad_norm": 2.0310938358306885, + "learning_rate": 1.8826157595450855e-06, + "loss": 2.1324, + "step": 1825 + }, + { + "epoch": 2.001095890410959, + "grad_norm": 1.9574612379074097, + "learning_rate": 1.8805848903330625e-06, + "loss": 2.0485, + "step": 1826 + }, + { + "epoch": 2.0021917808219176, + "grad_norm": 2.0283970832824707, + "learning_rate": 1.8785540211210398e-06, + "loss": 2.1768, + "step": 1827 + }, + { + "epoch": 2.0032876712328767, + "grad_norm": 2.062384843826294, + "learning_rate": 1.876523151909017e-06, + "loss": 2.1199, + "step": 1828 + }, + { + "epoch": 2.0043835616438357, + "grad_norm": 2.0543339252471924, + "learning_rate": 1.8744922826969944e-06, + "loss": 2.0887, + "step": 1829 + }, + { + "epoch": 2.0054794520547947, + "grad_norm": 2.1769917011260986, + "learning_rate": 1.8724614134849716e-06, + "loss": 2.0821, + "step": 1830 + }, + { + "epoch": 2.0054794520547947, + "eval_loss": 2.2689950466156006, + "eval_runtime": 43.9448, + "eval_samples_per_second": 22.756, + "eval_steps_per_second": 0.956, + "step": 1830 + }, + { + "epoch": 2.0065753424657533, + "grad_norm": 2.0331335067749023, + "learning_rate": 1.870430544272949e-06, + "loss": 2.0916, + "step": 1831 + }, + { + "epoch": 2.0076712328767123, + "grad_norm": 2.1848011016845703, + "learning_rate": 1.8683996750609262e-06, + "loss": 2.1525, + "step": 1832 + }, + { + "epoch": 2.0087671232876714, + "grad_norm": 2.086441993713379, + "learning_rate": 1.8663688058489035e-06, + "loss": 2.0733, + "step": 1833 + }, + { + "epoch": 2.00986301369863, + "grad_norm": 2.1663239002227783, + "learning_rate": 1.8643379366368807e-06, + "loss": 2.1026, + "step": 1834 + }, + { + "epoch": 2.010958904109589, + "grad_norm": 2.1658689975738525, + "learning_rate": 1.862307067424858e-06, + "loss": 2.1178, + "step": 1835 + }, + { + "epoch": 2.012054794520548, + "grad_norm": 2.043203592300415, + "learning_rate": 1.8602761982128353e-06, + "loss": 2.1025, + "step": 1836 + }, + { + "epoch": 2.013150684931507, + "grad_norm": 2.1104307174682617, + "learning_rate": 1.8582453290008126e-06, + "loss": 2.1218, + "step": 1837 + }, + { + "epoch": 2.0142465753424657, + "grad_norm": 2.078554630279541, + "learning_rate": 1.8562144597887898e-06, + "loss": 2.082, + "step": 1838 + }, + { + "epoch": 2.0153424657534247, + "grad_norm": 2.1440582275390625, + "learning_rate": 1.854183590576767e-06, + "loss": 2.101, + "step": 1839 + }, + { + "epoch": 2.0164383561643837, + "grad_norm": 2.0824856758117676, + "learning_rate": 1.8521527213647442e-06, + "loss": 2.1377, + "step": 1840 + }, + { + "epoch": 2.0164383561643837, + "eval_loss": 2.268453359603882, + "eval_runtime": 43.8166, + "eval_samples_per_second": 22.822, + "eval_steps_per_second": 0.959, + "step": 1840 + }, + { + "epoch": 2.0175342465753423, + "grad_norm": 2.1607978343963623, + "learning_rate": 1.8501218521527215e-06, + "loss": 2.0734, + "step": 1841 + }, + { + "epoch": 2.0186301369863013, + "grad_norm": 2.044123649597168, + "learning_rate": 1.8480909829406987e-06, + "loss": 2.04, + "step": 1842 + }, + { + "epoch": 2.0197260273972604, + "grad_norm": 2.1130807399749756, + "learning_rate": 1.846060113728676e-06, + "loss": 2.069, + "step": 1843 + }, + { + "epoch": 2.020821917808219, + "grad_norm": 2.0879838466644287, + "learning_rate": 1.8440292445166533e-06, + "loss": 2.1006, + "step": 1844 + }, + { + "epoch": 2.021917808219178, + "grad_norm": 2.311849355697632, + "learning_rate": 1.8419983753046306e-06, + "loss": 2.0691, + "step": 1845 + }, + { + "epoch": 2.023013698630137, + "grad_norm": 2.122504711151123, + "learning_rate": 1.8399675060926078e-06, + "loss": 2.1297, + "step": 1846 + }, + { + "epoch": 2.024109589041096, + "grad_norm": 2.135181427001953, + "learning_rate": 1.8379366368805851e-06, + "loss": 2.0951, + "step": 1847 + }, + { + "epoch": 2.0252054794520546, + "grad_norm": 2.0710697174072266, + "learning_rate": 1.8359057676685624e-06, + "loss": 2.0754, + "step": 1848 + }, + { + "epoch": 2.0263013698630137, + "grad_norm": 2.0922486782073975, + "learning_rate": 1.8338748984565397e-06, + "loss": 2.068, + "step": 1849 + }, + { + "epoch": 2.0273972602739727, + "grad_norm": 2.0101380348205566, + "learning_rate": 1.831844029244517e-06, + "loss": 2.0831, + "step": 1850 + }, + { + "epoch": 2.0273972602739727, + "eval_loss": 2.267380952835083, + "eval_runtime": 44.4219, + "eval_samples_per_second": 22.511, + "eval_steps_per_second": 0.945, + "step": 1850 + }, + { + "epoch": 2.0284931506849313, + "grad_norm": 2.012316942214966, + "learning_rate": 1.829813160032494e-06, + "loss": 2.0571, + "step": 1851 + }, + { + "epoch": 2.0295890410958903, + "grad_norm": 2.039604663848877, + "learning_rate": 1.8277822908204713e-06, + "loss": 2.0691, + "step": 1852 + }, + { + "epoch": 2.0306849315068494, + "grad_norm": 2.242504119873047, + "learning_rate": 1.8257514216084486e-06, + "loss": 2.1025, + "step": 1853 + }, + { + "epoch": 2.0317808219178084, + "grad_norm": 2.0629303455352783, + "learning_rate": 1.8237205523964258e-06, + "loss": 2.0803, + "step": 1854 + }, + { + "epoch": 2.032876712328767, + "grad_norm": 2.0935115814208984, + "learning_rate": 1.8216896831844031e-06, + "loss": 2.101, + "step": 1855 + }, + { + "epoch": 2.033972602739726, + "grad_norm": 2.164381504058838, + "learning_rate": 1.8196588139723804e-06, + "loss": 2.1227, + "step": 1856 + }, + { + "epoch": 2.035068493150685, + "grad_norm": 2.240391492843628, + "learning_rate": 1.8176279447603577e-06, + "loss": 2.0549, + "step": 1857 + }, + { + "epoch": 2.0361643835616436, + "grad_norm": 2.1460423469543457, + "learning_rate": 1.815597075548335e-06, + "loss": 2.0858, + "step": 1858 + }, + { + "epoch": 2.0372602739726027, + "grad_norm": 2.0199930667877197, + "learning_rate": 1.8135662063363122e-06, + "loss": 2.1058, + "step": 1859 + }, + { + "epoch": 2.0383561643835617, + "grad_norm": 2.074615955352783, + "learning_rate": 1.8115353371242895e-06, + "loss": 2.113, + "step": 1860 + }, + { + "epoch": 2.0383561643835617, + "eval_loss": 2.2699100971221924, + "eval_runtime": 42.9781, + "eval_samples_per_second": 23.268, + "eval_steps_per_second": 0.977, + "step": 1860 + }, + { + "epoch": 2.0394520547945207, + "grad_norm": 2.091417074203491, + "learning_rate": 1.8095044679122668e-06, + "loss": 2.045, + "step": 1861 + }, + { + "epoch": 2.0405479452054793, + "grad_norm": 2.2032699584960938, + "learning_rate": 1.807473598700244e-06, + "loss": 2.081, + "step": 1862 + }, + { + "epoch": 2.0416438356164384, + "grad_norm": 2.0829155445098877, + "learning_rate": 1.805442729488221e-06, + "loss": 2.0798, + "step": 1863 + }, + { + "epoch": 2.0427397260273974, + "grad_norm": 1.97968590259552, + "learning_rate": 1.8034118602761982e-06, + "loss": 2.0851, + "step": 1864 + }, + { + "epoch": 2.043835616438356, + "grad_norm": 1.9853475093841553, + "learning_rate": 1.8013809910641754e-06, + "loss": 2.0826, + "step": 1865 + }, + { + "epoch": 2.044931506849315, + "grad_norm": 2.0638294219970703, + "learning_rate": 1.7993501218521527e-06, + "loss": 2.0806, + "step": 1866 + }, + { + "epoch": 2.046027397260274, + "grad_norm": 2.0647425651550293, + "learning_rate": 1.79731925264013e-06, + "loss": 2.0613, + "step": 1867 + }, + { + "epoch": 2.047123287671233, + "grad_norm": 2.1473264694213867, + "learning_rate": 1.7952883834281073e-06, + "loss": 2.1211, + "step": 1868 + }, + { + "epoch": 2.0482191780821917, + "grad_norm": 2.037551164627075, + "learning_rate": 1.7932575142160846e-06, + "loss": 2.0727, + "step": 1869 + }, + { + "epoch": 2.0493150684931507, + "grad_norm": 2.010626792907715, + "learning_rate": 1.7912266450040618e-06, + "loss": 2.1013, + "step": 1870 + }, + { + "epoch": 2.0493150684931507, + "eval_loss": 2.266191005706787, + "eval_runtime": 45.1806, + "eval_samples_per_second": 22.133, + "eval_steps_per_second": 0.93, + "step": 1870 + }, + { + "epoch": 2.0504109589041097, + "grad_norm": 1.955633282661438, + "learning_rate": 1.789195775792039e-06, + "loss": 2.114, + "step": 1871 + }, + { + "epoch": 2.0515068493150683, + "grad_norm": 2.1419057846069336, + "learning_rate": 1.7871649065800164e-06, + "loss": 2.0296, + "step": 1872 + }, + { + "epoch": 2.0526027397260274, + "grad_norm": 2.167240858078003, + "learning_rate": 1.7851340373679937e-06, + "loss": 2.1391, + "step": 1873 + }, + { + "epoch": 2.0536986301369864, + "grad_norm": 2.0724527835845947, + "learning_rate": 1.783103168155971e-06, + "loss": 2.0571, + "step": 1874 + }, + { + "epoch": 2.0547945205479454, + "grad_norm": 2.104583263397217, + "learning_rate": 1.7810722989439482e-06, + "loss": 2.0917, + "step": 1875 + }, + { + "epoch": 2.055890410958904, + "grad_norm": 2.083120584487915, + "learning_rate": 1.7790414297319253e-06, + "loss": 2.093, + "step": 1876 + }, + { + "epoch": 2.056986301369863, + "grad_norm": 2.136396646499634, + "learning_rate": 1.7770105605199025e-06, + "loss": 2.0654, + "step": 1877 + }, + { + "epoch": 2.058082191780822, + "grad_norm": 2.1515700817108154, + "learning_rate": 1.7749796913078798e-06, + "loss": 2.05, + "step": 1878 + }, + { + "epoch": 2.0591780821917807, + "grad_norm": 2.1882944107055664, + "learning_rate": 1.772948822095857e-06, + "loss": 2.0974, + "step": 1879 + }, + { + "epoch": 2.0602739726027397, + "grad_norm": 2.261822462081909, + "learning_rate": 1.7709179528838344e-06, + "loss": 2.1375, + "step": 1880 + }, + { + "epoch": 2.0602739726027397, + "eval_loss": 2.267698287963867, + "eval_runtime": 45.8285, + "eval_samples_per_second": 21.82, + "eval_steps_per_second": 0.916, + "step": 1880 + }, + { + "epoch": 2.0613698630136987, + "grad_norm": 2.121238946914673, + "learning_rate": 1.7688870836718116e-06, + "loss": 2.1102, + "step": 1881 + }, + { + "epoch": 2.0624657534246573, + "grad_norm": 2.183793544769287, + "learning_rate": 1.766856214459789e-06, + "loss": 2.0862, + "step": 1882 + }, + { + "epoch": 2.0635616438356164, + "grad_norm": 2.209182024002075, + "learning_rate": 1.7648253452477662e-06, + "loss": 2.088, + "step": 1883 + }, + { + "epoch": 2.0646575342465754, + "grad_norm": 1.9593517780303955, + "learning_rate": 1.7627944760357435e-06, + "loss": 2.0644, + "step": 1884 + }, + { + "epoch": 2.0657534246575344, + "grad_norm": 1.9904463291168213, + "learning_rate": 1.7607636068237208e-06, + "loss": 2.1313, + "step": 1885 + }, + { + "epoch": 2.066849315068493, + "grad_norm": 2.2765681743621826, + "learning_rate": 1.758732737611698e-06, + "loss": 2.0979, + "step": 1886 + }, + { + "epoch": 2.067945205479452, + "grad_norm": 2.066204309463501, + "learning_rate": 1.7567018683996753e-06, + "loss": 2.0934, + "step": 1887 + }, + { + "epoch": 2.069041095890411, + "grad_norm": 2.104768753051758, + "learning_rate": 1.7546709991876524e-06, + "loss": 2.0887, + "step": 1888 + }, + { + "epoch": 2.0701369863013697, + "grad_norm": 2.0848052501678467, + "learning_rate": 1.7526401299756296e-06, + "loss": 2.0635, + "step": 1889 + }, + { + "epoch": 2.0712328767123287, + "grad_norm": 2.052314519882202, + "learning_rate": 1.750609260763607e-06, + "loss": 2.0728, + "step": 1890 + }, + { + "epoch": 2.0712328767123287, + "eval_loss": 2.266915798187256, + "eval_runtime": 46.3568, + "eval_samples_per_second": 21.572, + "eval_steps_per_second": 0.906, + "step": 1890 + }, + { + "epoch": 2.0723287671232877, + "grad_norm": 2.209855318069458, + "learning_rate": 1.7485783915515842e-06, + "loss": 2.0951, + "step": 1891 + }, + { + "epoch": 2.0734246575342468, + "grad_norm": 2.1558358669281006, + "learning_rate": 1.7465475223395615e-06, + "loss": 2.0872, + "step": 1892 + }, + { + "epoch": 2.0745205479452054, + "grad_norm": 2.426797389984131, + "learning_rate": 1.7445166531275387e-06, + "loss": 2.0578, + "step": 1893 + }, + { + "epoch": 2.0756164383561644, + "grad_norm": 2.1906144618988037, + "learning_rate": 1.742485783915516e-06, + "loss": 2.1664, + "step": 1894 + }, + { + "epoch": 2.0767123287671234, + "grad_norm": 2.2041025161743164, + "learning_rate": 1.7404549147034933e-06, + "loss": 2.1532, + "step": 1895 + }, + { + "epoch": 2.077808219178082, + "grad_norm": 2.179513931274414, + "learning_rate": 1.7384240454914706e-06, + "loss": 2.0349, + "step": 1896 + }, + { + "epoch": 2.078904109589041, + "grad_norm": 2.239232301712036, + "learning_rate": 1.7363931762794478e-06, + "loss": 2.0926, + "step": 1897 + }, + { + "epoch": 2.08, + "grad_norm": 2.2452921867370605, + "learning_rate": 1.7343623070674251e-06, + "loss": 2.1082, + "step": 1898 + }, + { + "epoch": 2.081095890410959, + "grad_norm": 2.145571708679199, + "learning_rate": 1.7323314378554024e-06, + "loss": 2.0886, + "step": 1899 + }, + { + "epoch": 2.0821917808219177, + "grad_norm": 2.201815366744995, + "learning_rate": 1.7303005686433797e-06, + "loss": 2.0947, + "step": 1900 + }, + { + "epoch": 2.0821917808219177, + "eval_loss": 2.2664403915405273, + "eval_runtime": 46.4641, + "eval_samples_per_second": 21.522, + "eval_steps_per_second": 0.904, + "step": 1900 + }, + { + "epoch": 2.0832876712328767, + "grad_norm": 2.095641851425171, + "learning_rate": 1.7282696994313567e-06, + "loss": 2.0925, + "step": 1901 + }, + { + "epoch": 2.0843835616438358, + "grad_norm": 2.186495304107666, + "learning_rate": 1.726238830219334e-06, + "loss": 2.0971, + "step": 1902 + }, + { + "epoch": 2.0854794520547943, + "grad_norm": 2.135972023010254, + "learning_rate": 1.7242079610073113e-06, + "loss": 2.076, + "step": 1903 + }, + { + "epoch": 2.0865753424657534, + "grad_norm": 2.1793875694274902, + "learning_rate": 1.7221770917952886e-06, + "loss": 2.1336, + "step": 1904 + }, + { + "epoch": 2.0876712328767124, + "grad_norm": 2.178579330444336, + "learning_rate": 1.7201462225832658e-06, + "loss": 2.0959, + "step": 1905 + }, + { + "epoch": 2.0887671232876714, + "grad_norm": 2.1725568771362305, + "learning_rate": 1.7181153533712431e-06, + "loss": 2.0965, + "step": 1906 + }, + { + "epoch": 2.08986301369863, + "grad_norm": 2.126034736633301, + "learning_rate": 1.7160844841592204e-06, + "loss": 2.0225, + "step": 1907 + }, + { + "epoch": 2.090958904109589, + "grad_norm": 2.0938918590545654, + "learning_rate": 1.7140536149471977e-06, + "loss": 2.0447, + "step": 1908 + }, + { + "epoch": 2.092054794520548, + "grad_norm": 2.1260006427764893, + "learning_rate": 1.712022745735175e-06, + "loss": 2.0739, + "step": 1909 + }, + { + "epoch": 2.0931506849315067, + "grad_norm": 2.2151126861572266, + "learning_rate": 1.7099918765231522e-06, + "loss": 2.0786, + "step": 1910 + }, + { + "epoch": 2.0931506849315067, + "eval_loss": 2.2675349712371826, + "eval_runtime": 43.2795, + "eval_samples_per_second": 23.106, + "eval_steps_per_second": 0.97, + "step": 1910 + }, + { + "epoch": 2.0942465753424657, + "grad_norm": 2.146458625793457, + "learning_rate": 1.7079610073111295e-06, + "loss": 2.0928, + "step": 1911 + }, + { + "epoch": 2.0953424657534248, + "grad_norm": 2.1360816955566406, + "learning_rate": 1.7059301380991068e-06, + "loss": 2.0404, + "step": 1912 + }, + { + "epoch": 2.0964383561643833, + "grad_norm": 2.130659341812134, + "learning_rate": 1.7038992688870836e-06, + "loss": 2.0858, + "step": 1913 + }, + { + "epoch": 2.0975342465753424, + "grad_norm": 2.0875484943389893, + "learning_rate": 1.701868399675061e-06, + "loss": 2.0613, + "step": 1914 + }, + { + "epoch": 2.0986301369863014, + "grad_norm": 2.1744940280914307, + "learning_rate": 1.6998375304630382e-06, + "loss": 2.1012, + "step": 1915 + }, + { + "epoch": 2.0997260273972604, + "grad_norm": 2.195559024810791, + "learning_rate": 1.6978066612510155e-06, + "loss": 2.1595, + "step": 1916 + }, + { + "epoch": 2.100821917808219, + "grad_norm": 2.3818888664245605, + "learning_rate": 1.6957757920389927e-06, + "loss": 2.1246, + "step": 1917 + }, + { + "epoch": 2.101917808219178, + "grad_norm": 2.296870708465576, + "learning_rate": 1.69374492282697e-06, + "loss": 2.0893, + "step": 1918 + }, + { + "epoch": 2.103013698630137, + "grad_norm": 2.3288252353668213, + "learning_rate": 1.6917140536149473e-06, + "loss": 2.0764, + "step": 1919 + }, + { + "epoch": 2.1041095890410957, + "grad_norm": 2.1017708778381348, + "learning_rate": 1.6896831844029246e-06, + "loss": 2.1312, + "step": 1920 + }, + { + "epoch": 2.1041095890410957, + "eval_loss": 2.2681941986083984, + "eval_runtime": 51.4964, + "eval_samples_per_second": 19.419, + "eval_steps_per_second": 0.816, + "step": 1920 + }, + { + "epoch": 2.1052054794520547, + "grad_norm": 2.2251839637756348, + "learning_rate": 1.6876523151909018e-06, + "loss": 2.0967, + "step": 1921 + }, + { + "epoch": 2.1063013698630138, + "grad_norm": 2.1989998817443848, + "learning_rate": 1.6856214459788791e-06, + "loss": 2.1064, + "step": 1922 + }, + { + "epoch": 2.107397260273973, + "grad_norm": 2.2629430294036865, + "learning_rate": 1.6835905767668564e-06, + "loss": 2.0574, + "step": 1923 + }, + { + "epoch": 2.1084931506849314, + "grad_norm": 2.2153608798980713, + "learning_rate": 1.6815597075548337e-06, + "loss": 2.0736, + "step": 1924 + }, + { + "epoch": 2.1095890410958904, + "grad_norm": 2.1380763053894043, + "learning_rate": 1.679528838342811e-06, + "loss": 2.1437, + "step": 1925 + }, + { + "epoch": 2.1106849315068494, + "grad_norm": 2.198996067047119, + "learning_rate": 1.677497969130788e-06, + "loss": 2.0623, + "step": 1926 + }, + { + "epoch": 2.111780821917808, + "grad_norm": 2.1622812747955322, + "learning_rate": 1.6754670999187653e-06, + "loss": 2.072, + "step": 1927 + }, + { + "epoch": 2.112876712328767, + "grad_norm": 2.185896873474121, + "learning_rate": 1.6734362307067426e-06, + "loss": 2.0931, + "step": 1928 + }, + { + "epoch": 2.113972602739726, + "grad_norm": 2.118260383605957, + "learning_rate": 1.6714053614947198e-06, + "loss": 2.0555, + "step": 1929 + }, + { + "epoch": 2.115068493150685, + "grad_norm": 2.070236921310425, + "learning_rate": 1.669374492282697e-06, + "loss": 2.0503, + "step": 1930 + }, + { + "epoch": 2.115068493150685, + "eval_loss": 2.267728328704834, + "eval_runtime": 48.9617, + "eval_samples_per_second": 20.424, + "eval_steps_per_second": 0.858, + "step": 1930 + }, + { + "epoch": 2.1161643835616437, + "grad_norm": 2.2161452770233154, + "learning_rate": 1.6673436230706744e-06, + "loss": 2.1211, + "step": 1931 + }, + { + "epoch": 2.1172602739726027, + "grad_norm": 2.0978891849517822, + "learning_rate": 1.6653127538586517e-06, + "loss": 2.1055, + "step": 1932 + }, + { + "epoch": 2.118356164383562, + "grad_norm": 2.208021879196167, + "learning_rate": 1.663281884646629e-06, + "loss": 2.0754, + "step": 1933 + }, + { + "epoch": 2.1194520547945204, + "grad_norm": 2.0411078929901123, + "learning_rate": 1.6612510154346062e-06, + "loss": 2.0687, + "step": 1934 + }, + { + "epoch": 2.1205479452054794, + "grad_norm": 2.275355339050293, + "learning_rate": 1.6592201462225835e-06, + "loss": 2.1354, + "step": 1935 + }, + { + "epoch": 2.1216438356164384, + "grad_norm": 2.102735996246338, + "learning_rate": 1.6571892770105608e-06, + "loss": 2.105, + "step": 1936 + }, + { + "epoch": 2.1227397260273975, + "grad_norm": 2.1679234504699707, + "learning_rate": 1.655158407798538e-06, + "loss": 2.1314, + "step": 1937 + }, + { + "epoch": 2.123835616438356, + "grad_norm": 2.244574785232544, + "learning_rate": 1.653127538586515e-06, + "loss": 2.0923, + "step": 1938 + }, + { + "epoch": 2.124931506849315, + "grad_norm": 2.0957608222961426, + "learning_rate": 1.6510966693744924e-06, + "loss": 2.1115, + "step": 1939 + }, + { + "epoch": 2.126027397260274, + "grad_norm": 2.0705184936523438, + "learning_rate": 1.6490658001624696e-06, + "loss": 2.1125, + "step": 1940 + }, + { + "epoch": 2.126027397260274, + "eval_loss": 2.2668750286102295, + "eval_runtime": 48.4476, + "eval_samples_per_second": 20.641, + "eval_steps_per_second": 0.867, + "step": 1940 + }, + { + "epoch": 2.1271232876712327, + "grad_norm": 2.185532569885254, + "learning_rate": 1.647034930950447e-06, + "loss": 2.1148, + "step": 1941 + }, + { + "epoch": 2.1282191780821917, + "grad_norm": 2.1876089572906494, + "learning_rate": 1.6450040617384242e-06, + "loss": 2.0646, + "step": 1942 + }, + { + "epoch": 2.1293150684931508, + "grad_norm": 2.065310478210449, + "learning_rate": 1.6429731925264015e-06, + "loss": 2.0843, + "step": 1943 + }, + { + "epoch": 2.1304109589041094, + "grad_norm": 2.1671154499053955, + "learning_rate": 1.6409423233143788e-06, + "loss": 2.112, + "step": 1944 + }, + { + "epoch": 2.1315068493150684, + "grad_norm": 2.1573469638824463, + "learning_rate": 1.638911454102356e-06, + "loss": 2.0781, + "step": 1945 + }, + { + "epoch": 2.1326027397260274, + "grad_norm": 2.0912411212921143, + "learning_rate": 1.6368805848903333e-06, + "loss": 2.0621, + "step": 1946 + }, + { + "epoch": 2.1336986301369865, + "grad_norm": 2.1280713081359863, + "learning_rate": 1.6348497156783106e-06, + "loss": 2.1174, + "step": 1947 + }, + { + "epoch": 2.134794520547945, + "grad_norm": 2.1215052604675293, + "learning_rate": 1.6328188464662879e-06, + "loss": 2.0725, + "step": 1948 + }, + { + "epoch": 2.135890410958904, + "grad_norm": 2.0699117183685303, + "learning_rate": 1.6307879772542651e-06, + "loss": 2.0783, + "step": 1949 + }, + { + "epoch": 2.136986301369863, + "grad_norm": 2.0162057876586914, + "learning_rate": 1.628757108042242e-06, + "loss": 2.0868, + "step": 1950 + }, + { + "epoch": 2.136986301369863, + "eval_loss": 2.2651138305664062, + "eval_runtime": 53.3744, + "eval_samples_per_second": 18.736, + "eval_steps_per_second": 0.787, + "step": 1950 + }, + { + "epoch": 2.138082191780822, + "grad_norm": 2.1494126319885254, + "learning_rate": 1.6267262388302193e-06, + "loss": 2.1147, + "step": 1951 + }, + { + "epoch": 2.1391780821917807, + "grad_norm": 2.1003870964050293, + "learning_rate": 1.6246953696181965e-06, + "loss": 2.082, + "step": 1952 + }, + { + "epoch": 2.1402739726027398, + "grad_norm": 2.1687192916870117, + "learning_rate": 1.6226645004061738e-06, + "loss": 2.1084, + "step": 1953 + }, + { + "epoch": 2.141369863013699, + "grad_norm": 2.010176181793213, + "learning_rate": 1.620633631194151e-06, + "loss": 2.1568, + "step": 1954 + }, + { + "epoch": 2.1424657534246574, + "grad_norm": 2.13127064704895, + "learning_rate": 1.6186027619821284e-06, + "loss": 2.116, + "step": 1955 + }, + { + "epoch": 2.1435616438356164, + "grad_norm": 2.2229974269866943, + "learning_rate": 1.6165718927701056e-06, + "loss": 2.0596, + "step": 1956 + }, + { + "epoch": 2.1446575342465755, + "grad_norm": 2.155855894088745, + "learning_rate": 1.614541023558083e-06, + "loss": 2.1052, + "step": 1957 + }, + { + "epoch": 2.145753424657534, + "grad_norm": 2.1598658561706543, + "learning_rate": 1.6125101543460602e-06, + "loss": 2.097, + "step": 1958 + }, + { + "epoch": 2.146849315068493, + "grad_norm": 2.3428308963775635, + "learning_rate": 1.6104792851340375e-06, + "loss": 2.0991, + "step": 1959 + }, + { + "epoch": 2.147945205479452, + "grad_norm": 2.293128728866577, + "learning_rate": 1.6084484159220147e-06, + "loss": 2.1146, + "step": 1960 + }, + { + "epoch": 2.147945205479452, + "eval_loss": 2.265352964401245, + "eval_runtime": 49.6478, + "eval_samples_per_second": 20.142, + "eval_steps_per_second": 0.846, + "step": 1960 + }, + { + "epoch": 2.149041095890411, + "grad_norm": 2.1581852436065674, + "learning_rate": 1.606417546709992e-06, + "loss": 2.0901, + "step": 1961 + }, + { + "epoch": 2.1501369863013697, + "grad_norm": 2.177950382232666, + "learning_rate": 1.6043866774979693e-06, + "loss": 2.0765, + "step": 1962 + }, + { + "epoch": 2.1512328767123288, + "grad_norm": 2.176633596420288, + "learning_rate": 1.6023558082859464e-06, + "loss": 2.0917, + "step": 1963 + }, + { + "epoch": 2.152328767123288, + "grad_norm": 2.132002592086792, + "learning_rate": 1.6003249390739236e-06, + "loss": 2.0712, + "step": 1964 + }, + { + "epoch": 2.1534246575342464, + "grad_norm": 2.1170928478240967, + "learning_rate": 1.598294069861901e-06, + "loss": 2.0835, + "step": 1965 + }, + { + "epoch": 2.1545205479452054, + "grad_norm": 2.2532341480255127, + "learning_rate": 1.5962632006498782e-06, + "loss": 2.0638, + "step": 1966 + }, + { + "epoch": 2.1556164383561645, + "grad_norm": 2.1742208003997803, + "learning_rate": 1.5942323314378555e-06, + "loss": 2.1164, + "step": 1967 + }, + { + "epoch": 2.1567123287671235, + "grad_norm": 2.1353843212127686, + "learning_rate": 1.5922014622258327e-06, + "loss": 2.0447, + "step": 1968 + }, + { + "epoch": 2.157808219178082, + "grad_norm": 2.1103644371032715, + "learning_rate": 1.59017059301381e-06, + "loss": 2.0631, + "step": 1969 + }, + { + "epoch": 2.158904109589041, + "grad_norm": 2.129606246948242, + "learning_rate": 1.5881397238017873e-06, + "loss": 2.1184, + "step": 1970 + }, + { + "epoch": 2.158904109589041, + "eval_loss": 2.26619553565979, + "eval_runtime": 52.3189, + "eval_samples_per_second": 19.114, + "eval_steps_per_second": 0.803, + "step": 1970 + }, + { + "epoch": 2.16, + "grad_norm": 2.2155823707580566, + "learning_rate": 1.5861088545897646e-06, + "loss": 2.0885, + "step": 1971 + }, + { + "epoch": 2.1610958904109587, + "grad_norm": 2.1816463470458984, + "learning_rate": 1.5840779853777418e-06, + "loss": 2.0517, + "step": 1972 + }, + { + "epoch": 2.1621917808219178, + "grad_norm": 2.2179040908813477, + "learning_rate": 1.5820471161657191e-06, + "loss": 2.0884, + "step": 1973 + }, + { + "epoch": 2.163287671232877, + "grad_norm": 2.3558337688446045, + "learning_rate": 1.5800162469536964e-06, + "loss": 2.0682, + "step": 1974 + }, + { + "epoch": 2.1643835616438354, + "grad_norm": 2.406665086746216, + "learning_rate": 1.5779853777416735e-06, + "loss": 2.1172, + "step": 1975 + }, + { + "epoch": 2.1654794520547944, + "grad_norm": 2.3815410137176514, + "learning_rate": 1.5759545085296507e-06, + "loss": 2.115, + "step": 1976 + }, + { + "epoch": 2.1665753424657535, + "grad_norm": 2.2329678535461426, + "learning_rate": 1.573923639317628e-06, + "loss": 2.087, + "step": 1977 + }, + { + "epoch": 2.1676712328767125, + "grad_norm": 2.331451177597046, + "learning_rate": 1.5718927701056053e-06, + "loss": 2.0996, + "step": 1978 + }, + { + "epoch": 2.168767123287671, + "grad_norm": 2.158210039138794, + "learning_rate": 1.5698619008935826e-06, + "loss": 2.0905, + "step": 1979 + }, + { + "epoch": 2.16986301369863, + "grad_norm": 2.1256191730499268, + "learning_rate": 1.5678310316815598e-06, + "loss": 2.1099, + "step": 1980 + }, + { + "epoch": 2.16986301369863, + "eval_loss": 2.2658803462982178, + "eval_runtime": 51.5278, + "eval_samples_per_second": 19.407, + "eval_steps_per_second": 0.815, + "step": 1980 + }, + { + "epoch": 2.170958904109589, + "grad_norm": 2.2626984119415283, + "learning_rate": 1.5658001624695371e-06, + "loss": 2.0651, + "step": 1981 + }, + { + "epoch": 2.172054794520548, + "grad_norm": 2.132742404937744, + "learning_rate": 1.5637692932575144e-06, + "loss": 2.1246, + "step": 1982 + }, + { + "epoch": 2.1731506849315068, + "grad_norm": 2.0621824264526367, + "learning_rate": 1.5617384240454917e-06, + "loss": 2.0696, + "step": 1983 + }, + { + "epoch": 2.174246575342466, + "grad_norm": 2.0715548992156982, + "learning_rate": 1.559707554833469e-06, + "loss": 2.0894, + "step": 1984 + }, + { + "epoch": 2.175342465753425, + "grad_norm": 2.1211655139923096, + "learning_rate": 1.5576766856214462e-06, + "loss": 2.1122, + "step": 1985 + }, + { + "epoch": 2.1764383561643834, + "grad_norm": 2.14603328704834, + "learning_rate": 1.5556458164094235e-06, + "loss": 2.0239, + "step": 1986 + }, + { + "epoch": 2.1775342465753424, + "grad_norm": 2.092494249343872, + "learning_rate": 1.5536149471974008e-06, + "loss": 2.0795, + "step": 1987 + }, + { + "epoch": 2.1786301369863015, + "grad_norm": 2.14874267578125, + "learning_rate": 1.5515840779853778e-06, + "loss": 2.0476, + "step": 1988 + }, + { + "epoch": 2.17972602739726, + "grad_norm": 2.024716854095459, + "learning_rate": 1.549553208773355e-06, + "loss": 2.0904, + "step": 1989 + }, + { + "epoch": 2.180821917808219, + "grad_norm": 2.1700267791748047, + "learning_rate": 1.5475223395613324e-06, + "loss": 2.0806, + "step": 1990 + }, + { + "epoch": 2.180821917808219, + "eval_loss": 2.2657630443573, + "eval_runtime": 49.7954, + "eval_samples_per_second": 20.082, + "eval_steps_per_second": 0.843, + "step": 1990 + }, + { + "epoch": 2.181917808219178, + "grad_norm": 2.1046273708343506, + "learning_rate": 1.5454914703493097e-06, + "loss": 2.0738, + "step": 1991 + }, + { + "epoch": 2.183013698630137, + "grad_norm": 2.2489562034606934, + "learning_rate": 1.543460601137287e-06, + "loss": 2.1468, + "step": 1992 + }, + { + "epoch": 2.1841095890410958, + "grad_norm": 2.176658868789673, + "learning_rate": 1.5414297319252642e-06, + "loss": 2.0977, + "step": 1993 + }, + { + "epoch": 2.185205479452055, + "grad_norm": 2.2471182346343994, + "learning_rate": 1.5393988627132415e-06, + "loss": 2.1023, + "step": 1994 + }, + { + "epoch": 2.186301369863014, + "grad_norm": 2.306699514389038, + "learning_rate": 1.5373679935012188e-06, + "loss": 2.084, + "step": 1995 + }, + { + "epoch": 2.1873972602739724, + "grad_norm": 2.2053849697113037, + "learning_rate": 1.535337124289196e-06, + "loss": 2.0419, + "step": 1996 + }, + { + "epoch": 2.1884931506849314, + "grad_norm": 2.2459211349487305, + "learning_rate": 1.5333062550771733e-06, + "loss": 2.0882, + "step": 1997 + }, + { + "epoch": 2.1895890410958905, + "grad_norm": 2.574877977371216, + "learning_rate": 1.5312753858651506e-06, + "loss": 2.1627, + "step": 1998 + }, + { + "epoch": 2.1906849315068495, + "grad_norm": 2.111438274383545, + "learning_rate": 1.5292445166531279e-06, + "loss": 2.0762, + "step": 1999 + }, + { + "epoch": 2.191780821917808, + "grad_norm": 2.1831490993499756, + "learning_rate": 1.5272136474411047e-06, + "loss": 2.0954, + "step": 2000 + }, + { + "epoch": 2.191780821917808, + "eval_loss": 2.2667267322540283, + "eval_runtime": 46.6703, + "eval_samples_per_second": 21.427, + "eval_steps_per_second": 0.9, + "step": 2000 + }, + { + "epoch": 2.192876712328767, + "grad_norm": 2.161336660385132, + "learning_rate": 1.525182778229082e-06, + "loss": 2.1393, + "step": 2001 + }, + { + "epoch": 2.193972602739726, + "grad_norm": 2.210535764694214, + "learning_rate": 1.5231519090170593e-06, + "loss": 2.0925, + "step": 2002 + }, + { + "epoch": 2.1950684931506848, + "grad_norm": 2.123384714126587, + "learning_rate": 1.5211210398050365e-06, + "loss": 2.0623, + "step": 2003 + }, + { + "epoch": 2.196164383561644, + "grad_norm": 2.147090196609497, + "learning_rate": 1.5190901705930138e-06, + "loss": 2.0839, + "step": 2004 + }, + { + "epoch": 2.197260273972603, + "grad_norm": 2.261474609375, + "learning_rate": 1.517059301380991e-06, + "loss": 2.0987, + "step": 2005 + }, + { + "epoch": 2.198356164383562, + "grad_norm": 2.295656204223633, + "learning_rate": 1.5150284321689684e-06, + "loss": 2.0786, + "step": 2006 + }, + { + "epoch": 2.1994520547945204, + "grad_norm": 2.055387020111084, + "learning_rate": 1.5129975629569456e-06, + "loss": 2.0781, + "step": 2007 + }, + { + "epoch": 2.2005479452054795, + "grad_norm": 2.1994800567626953, + "learning_rate": 1.510966693744923e-06, + "loss": 2.088, + "step": 2008 + }, + { + "epoch": 2.2016438356164385, + "grad_norm": 2.3267500400543213, + "learning_rate": 1.5089358245329002e-06, + "loss": 2.0945, + "step": 2009 + }, + { + "epoch": 2.202739726027397, + "grad_norm": 2.1972532272338867, + "learning_rate": 1.5069049553208775e-06, + "loss": 2.0822, + "step": 2010 + }, + { + "epoch": 2.202739726027397, + "eval_loss": 2.26534104347229, + "eval_runtime": 41.6482, + "eval_samples_per_second": 24.011, + "eval_steps_per_second": 1.008, + "step": 2010 + }, + { + "epoch": 2.203835616438356, + "grad_norm": 2.161071538925171, + "learning_rate": 1.5048740861088547e-06, + "loss": 2.0908, + "step": 2011 + }, + { + "epoch": 2.204931506849315, + "grad_norm": 2.1437389850616455, + "learning_rate": 1.502843216896832e-06, + "loss": 2.1041, + "step": 2012 + }, + { + "epoch": 2.206027397260274, + "grad_norm": 2.16727614402771, + "learning_rate": 1.500812347684809e-06, + "loss": 2.0209, + "step": 2013 + }, + { + "epoch": 2.207123287671233, + "grad_norm": 2.1556594371795654, + "learning_rate": 1.4987814784727864e-06, + "loss": 2.09, + "step": 2014 + }, + { + "epoch": 2.208219178082192, + "grad_norm": 2.0891411304473877, + "learning_rate": 1.4967506092607636e-06, + "loss": 2.0864, + "step": 2015 + }, + { + "epoch": 2.209315068493151, + "grad_norm": 2.031249523162842, + "learning_rate": 1.494719740048741e-06, + "loss": 2.0363, + "step": 2016 + }, + { + "epoch": 2.2104109589041094, + "grad_norm": 2.2452943325042725, + "learning_rate": 1.4926888708367182e-06, + "loss": 2.114, + "step": 2017 + }, + { + "epoch": 2.2115068493150685, + "grad_norm": 2.1752209663391113, + "learning_rate": 1.4906580016246955e-06, + "loss": 2.1123, + "step": 2018 + }, + { + "epoch": 2.2126027397260275, + "grad_norm": 2.2148444652557373, + "learning_rate": 1.4886271324126727e-06, + "loss": 2.1132, + "step": 2019 + }, + { + "epoch": 2.213698630136986, + "grad_norm": 2.1530041694641113, + "learning_rate": 1.48659626320065e-06, + "loss": 2.1278, + "step": 2020 + }, + { + "epoch": 2.213698630136986, + "eval_loss": 2.264028310775757, + "eval_runtime": 43.6346, + "eval_samples_per_second": 22.918, + "eval_steps_per_second": 0.963, + "step": 2020 + }, + { + "epoch": 2.214794520547945, + "grad_norm": 2.223219871520996, + "learning_rate": 1.4845653939886273e-06, + "loss": 2.0293, + "step": 2021 + }, + { + "epoch": 2.215890410958904, + "grad_norm": 2.122913360595703, + "learning_rate": 1.4825345247766046e-06, + "loss": 2.0411, + "step": 2022 + }, + { + "epoch": 2.216986301369863, + "grad_norm": 2.1176650524139404, + "learning_rate": 1.4805036555645818e-06, + "loss": 2.0326, + "step": 2023 + }, + { + "epoch": 2.2180821917808218, + "grad_norm": 2.1717002391815186, + "learning_rate": 1.4784727863525591e-06, + "loss": 2.0835, + "step": 2024 + }, + { + "epoch": 2.219178082191781, + "grad_norm": 2.1869051456451416, + "learning_rate": 1.4764419171405362e-06, + "loss": 2.0571, + "step": 2025 + }, + { + "epoch": 2.22027397260274, + "grad_norm": 2.2699813842773438, + "learning_rate": 1.4744110479285135e-06, + "loss": 2.1156, + "step": 2026 + }, + { + "epoch": 2.2213698630136984, + "grad_norm": 2.243837594985962, + "learning_rate": 1.4723801787164907e-06, + "loss": 2.1231, + "step": 2027 + }, + { + "epoch": 2.2224657534246575, + "grad_norm": 2.1143088340759277, + "learning_rate": 1.470349309504468e-06, + "loss": 2.11, + "step": 2028 + }, + { + "epoch": 2.2235616438356165, + "grad_norm": 2.0839781761169434, + "learning_rate": 1.4683184402924453e-06, + "loss": 2.0971, + "step": 2029 + }, + { + "epoch": 2.2246575342465755, + "grad_norm": 2.182480812072754, + "learning_rate": 1.4662875710804226e-06, + "loss": 2.1094, + "step": 2030 + }, + { + "epoch": 2.2246575342465755, + "eval_loss": 2.2636313438415527, + "eval_runtime": 43.184, + "eval_samples_per_second": 23.157, + "eval_steps_per_second": 0.973, + "step": 2030 + }, + { + "epoch": 2.225753424657534, + "grad_norm": 2.2751855850219727, + "learning_rate": 1.4642567018683998e-06, + "loss": 2.1533, + "step": 2031 + }, + { + "epoch": 2.226849315068493, + "grad_norm": 2.049236536026001, + "learning_rate": 1.4622258326563771e-06, + "loss": 2.0847, + "step": 2032 + }, + { + "epoch": 2.227945205479452, + "grad_norm": 2.0883119106292725, + "learning_rate": 1.4601949634443544e-06, + "loss": 2.0687, + "step": 2033 + }, + { + "epoch": 2.2290410958904108, + "grad_norm": 2.152400016784668, + "learning_rate": 1.4581640942323317e-06, + "loss": 2.0979, + "step": 2034 + }, + { + "epoch": 2.23013698630137, + "grad_norm": 2.221442699432373, + "learning_rate": 1.456133225020309e-06, + "loss": 2.1601, + "step": 2035 + }, + { + "epoch": 2.231232876712329, + "grad_norm": 2.1799275875091553, + "learning_rate": 1.4541023558082862e-06, + "loss": 2.1133, + "step": 2036 + }, + { + "epoch": 2.232328767123288, + "grad_norm": 2.182882070541382, + "learning_rate": 1.4520714865962635e-06, + "loss": 2.1038, + "step": 2037 + }, + { + "epoch": 2.2334246575342465, + "grad_norm": 2.1438207626342773, + "learning_rate": 1.4500406173842406e-06, + "loss": 2.1176, + "step": 2038 + }, + { + "epoch": 2.2345205479452055, + "grad_norm": 2.1394078731536865, + "learning_rate": 1.4480097481722178e-06, + "loss": 2.0928, + "step": 2039 + }, + { + "epoch": 2.2356164383561645, + "grad_norm": 2.2296125888824463, + "learning_rate": 1.4459788789601951e-06, + "loss": 2.0938, + "step": 2040 + }, + { + "epoch": 2.2356164383561645, + "eval_loss": 2.2618958950042725, + "eval_runtime": 43.253, + "eval_samples_per_second": 23.12, + "eval_steps_per_second": 0.971, + "step": 2040 + }, + { + "epoch": 2.236712328767123, + "grad_norm": 2.248995542526245, + "learning_rate": 1.4439480097481724e-06, + "loss": 2.0666, + "step": 2041 + }, + { + "epoch": 2.237808219178082, + "grad_norm": 2.0892133712768555, + "learning_rate": 1.4419171405361497e-06, + "loss": 2.0437, + "step": 2042 + }, + { + "epoch": 2.238904109589041, + "grad_norm": 2.191180944442749, + "learning_rate": 1.439886271324127e-06, + "loss": 2.1038, + "step": 2043 + }, + { + "epoch": 2.24, + "grad_norm": 2.1049792766571045, + "learning_rate": 1.4378554021121042e-06, + "loss": 2.096, + "step": 2044 + }, + { + "epoch": 2.241095890410959, + "grad_norm": 2.166987419128418, + "learning_rate": 1.4358245329000815e-06, + "loss": 2.0704, + "step": 2045 + }, + { + "epoch": 2.242191780821918, + "grad_norm": 2.3071136474609375, + "learning_rate": 1.4337936636880588e-06, + "loss": 2.1118, + "step": 2046 + }, + { + "epoch": 2.243287671232877, + "grad_norm": 2.185621976852417, + "learning_rate": 1.431762794476036e-06, + "loss": 2.1032, + "step": 2047 + }, + { + "epoch": 2.2443835616438355, + "grad_norm": 2.2100017070770264, + "learning_rate": 1.4297319252640133e-06, + "loss": 2.0542, + "step": 2048 + }, + { + "epoch": 2.2454794520547945, + "grad_norm": 2.2313308715820312, + "learning_rate": 1.4277010560519906e-06, + "loss": 2.0812, + "step": 2049 + }, + { + "epoch": 2.2465753424657535, + "grad_norm": 2.1170578002929688, + "learning_rate": 1.4256701868399674e-06, + "loss": 2.0542, + "step": 2050 + }, + { + "epoch": 2.2465753424657535, + "eval_loss": 2.2627694606781006, + "eval_runtime": 44.8509, + "eval_samples_per_second": 22.296, + "eval_steps_per_second": 0.936, + "step": 2050 + }, + { + "epoch": 2.247671232876712, + "grad_norm": 2.207075357437134, + "learning_rate": 1.4236393176279447e-06, + "loss": 2.1467, + "step": 2051 + }, + { + "epoch": 2.248767123287671, + "grad_norm": 2.1787123680114746, + "learning_rate": 1.421608448415922e-06, + "loss": 2.0824, + "step": 2052 + }, + { + "epoch": 2.24986301369863, + "grad_norm": 2.179396867752075, + "learning_rate": 1.4195775792038993e-06, + "loss": 2.1394, + "step": 2053 + }, + { + "epoch": 2.250958904109589, + "grad_norm": 2.1963279247283936, + "learning_rate": 1.4175467099918765e-06, + "loss": 2.0594, + "step": 2054 + }, + { + "epoch": 2.252054794520548, + "grad_norm": 2.165109872817993, + "learning_rate": 1.4155158407798538e-06, + "loss": 2.0812, + "step": 2055 + }, + { + "epoch": 2.253150684931507, + "grad_norm": 2.033921957015991, + "learning_rate": 1.413484971567831e-06, + "loss": 2.1124, + "step": 2056 + }, + { + "epoch": 2.254246575342466, + "grad_norm": 2.12203311920166, + "learning_rate": 1.4114541023558084e-06, + "loss": 2.0361, + "step": 2057 + }, + { + "epoch": 2.255342465753425, + "grad_norm": 2.268758535385132, + "learning_rate": 1.4094232331437856e-06, + "loss": 2.0766, + "step": 2058 + }, + { + "epoch": 2.2564383561643835, + "grad_norm": 2.1977181434631348, + "learning_rate": 1.407392363931763e-06, + "loss": 2.124, + "step": 2059 + }, + { + "epoch": 2.2575342465753425, + "grad_norm": 2.100123882293701, + "learning_rate": 1.4053614947197402e-06, + "loss": 2.1554, + "step": 2060 + }, + { + "epoch": 2.2575342465753425, + "eval_loss": 2.263918399810791, + "eval_runtime": 44.5109, + "eval_samples_per_second": 22.466, + "eval_steps_per_second": 0.944, + "step": 2060 + }, + { + "epoch": 2.2586301369863016, + "grad_norm": 2.1824963092803955, + "learning_rate": 1.4033306255077175e-06, + "loss": 2.0998, + "step": 2061 + }, + { + "epoch": 2.25972602739726, + "grad_norm": 2.211451530456543, + "learning_rate": 1.4012997562956945e-06, + "loss": 2.0266, + "step": 2062 + }, + { + "epoch": 2.260821917808219, + "grad_norm": 2.2530057430267334, + "learning_rate": 1.3992688870836718e-06, + "loss": 2.1366, + "step": 2063 + }, + { + "epoch": 2.261917808219178, + "grad_norm": 2.121558904647827, + "learning_rate": 1.397238017871649e-06, + "loss": 2.0927, + "step": 2064 + }, + { + "epoch": 2.263013698630137, + "grad_norm": 2.1322062015533447, + "learning_rate": 1.3952071486596264e-06, + "loss": 2.0929, + "step": 2065 + }, + { + "epoch": 2.264109589041096, + "grad_norm": 2.1645631790161133, + "learning_rate": 1.3931762794476036e-06, + "loss": 2.057, + "step": 2066 + }, + { + "epoch": 2.265205479452055, + "grad_norm": 2.1870594024658203, + "learning_rate": 1.391145410235581e-06, + "loss": 2.1274, + "step": 2067 + }, + { + "epoch": 2.266301369863014, + "grad_norm": 2.1355020999908447, + "learning_rate": 1.3891145410235582e-06, + "loss": 2.0784, + "step": 2068 + }, + { + "epoch": 2.2673972602739725, + "grad_norm": 2.138497829437256, + "learning_rate": 1.3870836718115355e-06, + "loss": 2.0265, + "step": 2069 + }, + { + "epoch": 2.2684931506849315, + "grad_norm": 2.3041086196899414, + "learning_rate": 1.3850528025995127e-06, + "loss": 2.0716, + "step": 2070 + }, + { + "epoch": 2.2684931506849315, + "eval_loss": 2.263742685317993, + "eval_runtime": 45.8378, + "eval_samples_per_second": 21.816, + "eval_steps_per_second": 0.916, + "step": 2070 + }, + { + "epoch": 2.2695890410958905, + "grad_norm": 2.2722716331481934, + "learning_rate": 1.38302193338749e-06, + "loss": 2.0744, + "step": 2071 + }, + { + "epoch": 2.270684931506849, + "grad_norm": 2.267909288406372, + "learning_rate": 1.3809910641754673e-06, + "loss": 2.0269, + "step": 2072 + }, + { + "epoch": 2.271780821917808, + "grad_norm": 2.271430015563965, + "learning_rate": 1.3789601949634446e-06, + "loss": 2.1165, + "step": 2073 + }, + { + "epoch": 2.272876712328767, + "grad_norm": 2.3091483116149902, + "learning_rate": 1.3769293257514218e-06, + "loss": 2.0336, + "step": 2074 + }, + { + "epoch": 2.2739726027397262, + "grad_norm": 2.2311320304870605, + "learning_rate": 1.374898456539399e-06, + "loss": 2.0964, + "step": 2075 + }, + { + "epoch": 2.275068493150685, + "grad_norm": 2.2187438011169434, + "learning_rate": 1.3728675873273762e-06, + "loss": 2.0672, + "step": 2076 + }, + { + "epoch": 2.276164383561644, + "grad_norm": 2.0933640003204346, + "learning_rate": 1.3708367181153535e-06, + "loss": 2.1113, + "step": 2077 + }, + { + "epoch": 2.277260273972603, + "grad_norm": 2.2611358165740967, + "learning_rate": 1.3688058489033307e-06, + "loss": 2.101, + "step": 2078 + }, + { + "epoch": 2.2783561643835615, + "grad_norm": 2.269864797592163, + "learning_rate": 1.366774979691308e-06, + "loss": 2.0993, + "step": 2079 + }, + { + "epoch": 2.2794520547945205, + "grad_norm": 2.242854356765747, + "learning_rate": 1.3647441104792853e-06, + "loss": 2.107, + "step": 2080 + }, + { + "epoch": 2.2794520547945205, + "eval_loss": 2.2640912532806396, + "eval_runtime": 46.7194, + "eval_samples_per_second": 21.404, + "eval_steps_per_second": 0.899, + "step": 2080 + }, + { + "epoch": 2.2805479452054795, + "grad_norm": 2.2348811626434326, + "learning_rate": 1.3627132412672626e-06, + "loss": 2.0941, + "step": 2081 + }, + { + "epoch": 2.281643835616438, + "grad_norm": 2.214179277420044, + "learning_rate": 1.3606823720552398e-06, + "loss": 2.0548, + "step": 2082 + }, + { + "epoch": 2.282739726027397, + "grad_norm": 2.216902256011963, + "learning_rate": 1.3586515028432171e-06, + "loss": 2.0955, + "step": 2083 + }, + { + "epoch": 2.283835616438356, + "grad_norm": 2.1791675090789795, + "learning_rate": 1.3566206336311944e-06, + "loss": 2.0395, + "step": 2084 + }, + { + "epoch": 2.2849315068493152, + "grad_norm": 2.135376214981079, + "learning_rate": 1.3545897644191717e-06, + "loss": 2.0837, + "step": 2085 + }, + { + "epoch": 2.286027397260274, + "grad_norm": 2.244448184967041, + "learning_rate": 1.352558895207149e-06, + "loss": 2.0805, + "step": 2086 + }, + { + "epoch": 2.287123287671233, + "grad_norm": 2.28576922416687, + "learning_rate": 1.3505280259951258e-06, + "loss": 2.1014, + "step": 2087 + }, + { + "epoch": 2.288219178082192, + "grad_norm": 2.247220516204834, + "learning_rate": 1.348497156783103e-06, + "loss": 2.0773, + "step": 2088 + }, + { + "epoch": 2.289315068493151, + "grad_norm": 2.107109308242798, + "learning_rate": 1.3464662875710804e-06, + "loss": 2.1236, + "step": 2089 + }, + { + "epoch": 2.2904109589041095, + "grad_norm": 2.2805957794189453, + "learning_rate": 1.3444354183590576e-06, + "loss": 2.0516, + "step": 2090 + }, + { + "epoch": 2.2904109589041095, + "eval_loss": 2.262258291244507, + "eval_runtime": 46.3329, + "eval_samples_per_second": 21.583, + "eval_steps_per_second": 0.906, + "step": 2090 + }, + { + "epoch": 2.2915068493150685, + "grad_norm": 2.2838428020477295, + "learning_rate": 1.342404549147035e-06, + "loss": 2.1008, + "step": 2091 + }, + { + "epoch": 2.2926027397260276, + "grad_norm": 2.1949639320373535, + "learning_rate": 1.3403736799350122e-06, + "loss": 2.0634, + "step": 2092 + }, + { + "epoch": 2.293698630136986, + "grad_norm": 2.144423246383667, + "learning_rate": 1.3383428107229895e-06, + "loss": 2.0996, + "step": 2093 + }, + { + "epoch": 2.294794520547945, + "grad_norm": 2.293369770050049, + "learning_rate": 1.3363119415109667e-06, + "loss": 2.1206, + "step": 2094 + }, + { + "epoch": 2.2958904109589042, + "grad_norm": 2.2426724433898926, + "learning_rate": 1.334281072298944e-06, + "loss": 2.1026, + "step": 2095 + }, + { + "epoch": 2.296986301369863, + "grad_norm": 2.159067392349243, + "learning_rate": 1.3322502030869213e-06, + "loss": 2.0993, + "step": 2096 + }, + { + "epoch": 2.298082191780822, + "grad_norm": 2.2819197177886963, + "learning_rate": 1.3302193338748986e-06, + "loss": 2.0776, + "step": 2097 + }, + { + "epoch": 2.299178082191781, + "grad_norm": 2.253448963165283, + "learning_rate": 1.3281884646628758e-06, + "loss": 2.1211, + "step": 2098 + }, + { + "epoch": 2.30027397260274, + "grad_norm": 2.2652339935302734, + "learning_rate": 1.3261575954508531e-06, + "loss": 2.1175, + "step": 2099 + }, + { + "epoch": 2.3013698630136985, + "grad_norm": 2.1319172382354736, + "learning_rate": 1.3241267262388302e-06, + "loss": 2.1137, + "step": 2100 + }, + { + "epoch": 2.3013698630136985, + "eval_loss": 2.263416051864624, + "eval_runtime": 47.0463, + "eval_samples_per_second": 21.256, + "eval_steps_per_second": 0.893, + "step": 2100 + }, + { + "epoch": 2.3024657534246575, + "grad_norm": 2.1784651279449463, + "learning_rate": 1.3220958570268074e-06, + "loss": 2.0951, + "step": 2101 + }, + { + "epoch": 2.3035616438356166, + "grad_norm": 2.1439764499664307, + "learning_rate": 1.3200649878147847e-06, + "loss": 2.0483, + "step": 2102 + }, + { + "epoch": 2.304657534246575, + "grad_norm": 2.1723854541778564, + "learning_rate": 1.318034118602762e-06, + "loss": 2.0851, + "step": 2103 + }, + { + "epoch": 2.305753424657534, + "grad_norm": 2.098573923110962, + "learning_rate": 1.3160032493907393e-06, + "loss": 2.1204, + "step": 2104 + }, + { + "epoch": 2.3068493150684932, + "grad_norm": 2.1380534172058105, + "learning_rate": 1.3139723801787166e-06, + "loss": 2.1094, + "step": 2105 + }, + { + "epoch": 2.3079452054794523, + "grad_norm": 2.1696887016296387, + "learning_rate": 1.3119415109666938e-06, + "loss": 2.1411, + "step": 2106 + }, + { + "epoch": 2.309041095890411, + "grad_norm": 2.1761717796325684, + "learning_rate": 1.309910641754671e-06, + "loss": 2.0893, + "step": 2107 + }, + { + "epoch": 2.31013698630137, + "grad_norm": 2.2763733863830566, + "learning_rate": 1.3078797725426484e-06, + "loss": 2.0441, + "step": 2108 + }, + { + "epoch": 2.311232876712329, + "grad_norm": 2.1061692237854004, + "learning_rate": 1.3058489033306257e-06, + "loss": 2.0652, + "step": 2109 + }, + { + "epoch": 2.3123287671232875, + "grad_norm": 2.242868423461914, + "learning_rate": 1.303818034118603e-06, + "loss": 2.0636, + "step": 2110 + }, + { + "epoch": 2.3123287671232875, + "eval_loss": 2.2643768787384033, + "eval_runtime": 43.9201, + "eval_samples_per_second": 22.769, + "eval_steps_per_second": 0.956, + "step": 2110 + }, + { + "epoch": 2.3134246575342465, + "grad_norm": 2.4198052883148193, + "learning_rate": 1.3017871649065802e-06, + "loss": 2.0939, + "step": 2111 + }, + { + "epoch": 2.3145205479452056, + "grad_norm": 2.3702943325042725, + "learning_rate": 1.2997562956945573e-06, + "loss": 2.0663, + "step": 2112 + }, + { + "epoch": 2.315616438356164, + "grad_norm": 2.1991565227508545, + "learning_rate": 1.2977254264825345e-06, + "loss": 2.0594, + "step": 2113 + }, + { + "epoch": 2.316712328767123, + "grad_norm": 2.255929946899414, + "learning_rate": 1.2956945572705118e-06, + "loss": 2.0973, + "step": 2114 + }, + { + "epoch": 2.317808219178082, + "grad_norm": 2.307612419128418, + "learning_rate": 1.293663688058489e-06, + "loss": 2.1085, + "step": 2115 + }, + { + "epoch": 2.3189041095890413, + "grad_norm": 2.237325429916382, + "learning_rate": 1.2916328188464664e-06, + "loss": 2.0485, + "step": 2116 + }, + { + "epoch": 2.32, + "grad_norm": 2.3728206157684326, + "learning_rate": 1.2896019496344436e-06, + "loss": 2.0086, + "step": 2117 + }, + { + "epoch": 2.321095890410959, + "grad_norm": 2.138505697250366, + "learning_rate": 1.287571080422421e-06, + "loss": 2.0628, + "step": 2118 + }, + { + "epoch": 2.322191780821918, + "grad_norm": 2.1808021068573, + "learning_rate": 1.2855402112103982e-06, + "loss": 2.1146, + "step": 2119 + }, + { + "epoch": 2.323287671232877, + "grad_norm": 2.1783456802368164, + "learning_rate": 1.2835093419983755e-06, + "loss": 2.0589, + "step": 2120 + }, + { + "epoch": 2.323287671232877, + "eval_loss": 2.262885332107544, + "eval_runtime": 48.7798, + "eval_samples_per_second": 20.5, + "eval_steps_per_second": 0.861, + "step": 2120 + }, + { + "epoch": 2.3243835616438355, + "grad_norm": 2.3852620124816895, + "learning_rate": 1.2814784727863528e-06, + "loss": 2.0862, + "step": 2121 + }, + { + "epoch": 2.3254794520547946, + "grad_norm": 2.223008155822754, + "learning_rate": 1.27944760357433e-06, + "loss": 2.072, + "step": 2122 + }, + { + "epoch": 2.3265753424657536, + "grad_norm": 2.1848905086517334, + "learning_rate": 1.2774167343623073e-06, + "loss": 2.1234, + "step": 2123 + }, + { + "epoch": 2.327671232876712, + "grad_norm": 2.1176791191101074, + "learning_rate": 1.2753858651502846e-06, + "loss": 2.1188, + "step": 2124 + }, + { + "epoch": 2.328767123287671, + "grad_norm": 2.2725491523742676, + "learning_rate": 1.2733549959382616e-06, + "loss": 2.0892, + "step": 2125 + }, + { + "epoch": 2.3298630136986302, + "grad_norm": 2.218189001083374, + "learning_rate": 1.271324126726239e-06, + "loss": 2.0827, + "step": 2126 + }, + { + "epoch": 2.330958904109589, + "grad_norm": 2.2110207080841064, + "learning_rate": 1.2692932575142162e-06, + "loss": 2.0857, + "step": 2127 + }, + { + "epoch": 2.332054794520548, + "grad_norm": 2.175096273422241, + "learning_rate": 1.2672623883021935e-06, + "loss": 2.0461, + "step": 2128 + }, + { + "epoch": 2.333150684931507, + "grad_norm": 2.1493241786956787, + "learning_rate": 1.2652315190901707e-06, + "loss": 2.0361, + "step": 2129 + }, + { + "epoch": 2.334246575342466, + "grad_norm": 2.276071310043335, + "learning_rate": 1.263200649878148e-06, + "loss": 2.1336, + "step": 2130 + }, + { + "epoch": 2.334246575342466, + "eval_loss": 2.262963056564331, + "eval_runtime": 49.2242, + "eval_samples_per_second": 20.315, + "eval_steps_per_second": 0.853, + "step": 2130 + }, + { + "epoch": 2.3353424657534245, + "grad_norm": 2.1708502769470215, + "learning_rate": 1.2611697806661253e-06, + "loss": 2.081, + "step": 2131 + }, + { + "epoch": 2.3364383561643836, + "grad_norm": 2.219454526901245, + "learning_rate": 1.2591389114541026e-06, + "loss": 2.0941, + "step": 2132 + }, + { + "epoch": 2.3375342465753426, + "grad_norm": 2.110680103302002, + "learning_rate": 1.2571080422420798e-06, + "loss": 2.0692, + "step": 2133 + }, + { + "epoch": 2.338630136986301, + "grad_norm": 2.186999559402466, + "learning_rate": 1.2550771730300571e-06, + "loss": 2.0652, + "step": 2134 + }, + { + "epoch": 2.33972602739726, + "grad_norm": 2.182494878768921, + "learning_rate": 1.2530463038180344e-06, + "loss": 2.1238, + "step": 2135 + }, + { + "epoch": 2.3408219178082192, + "grad_norm": 2.163588047027588, + "learning_rate": 1.2510154346060117e-06, + "loss": 2.0788, + "step": 2136 + }, + { + "epoch": 2.3419178082191783, + "grad_norm": 2.1190285682678223, + "learning_rate": 1.2489845653939887e-06, + "loss": 2.0861, + "step": 2137 + }, + { + "epoch": 2.343013698630137, + "grad_norm": 2.187049388885498, + "learning_rate": 1.246953696181966e-06, + "loss": 2.1098, + "step": 2138 + }, + { + "epoch": 2.344109589041096, + "grad_norm": 2.3307504653930664, + "learning_rate": 1.2449228269699433e-06, + "loss": 2.1443, + "step": 2139 + }, + { + "epoch": 2.345205479452055, + "grad_norm": 2.149183511734009, + "learning_rate": 1.2428919577579204e-06, + "loss": 2.0855, + "step": 2140 + }, + { + "epoch": 2.345205479452055, + "eval_loss": 2.263383388519287, + "eval_runtime": 49.7579, + "eval_samples_per_second": 20.097, + "eval_steps_per_second": 0.844, + "step": 2140 + }, + { + "epoch": 2.3463013698630135, + "grad_norm": 2.190133571624756, + "learning_rate": 1.2408610885458976e-06, + "loss": 2.059, + "step": 2141 + }, + { + "epoch": 2.3473972602739726, + "grad_norm": 2.1775476932525635, + "learning_rate": 1.238830219333875e-06, + "loss": 2.0795, + "step": 2142 + }, + { + "epoch": 2.3484931506849316, + "grad_norm": 2.148383617401123, + "learning_rate": 1.2367993501218522e-06, + "loss": 2.0614, + "step": 2143 + }, + { + "epoch": 2.34958904109589, + "grad_norm": 2.184270143508911, + "learning_rate": 1.2347684809098295e-06, + "loss": 2.1111, + "step": 2144 + }, + { + "epoch": 2.350684931506849, + "grad_norm": 2.136432647705078, + "learning_rate": 1.2327376116978067e-06, + "loss": 2.1162, + "step": 2145 + }, + { + "epoch": 2.3517808219178082, + "grad_norm": 2.1389224529266357, + "learning_rate": 1.230706742485784e-06, + "loss": 2.0369, + "step": 2146 + }, + { + "epoch": 2.3528767123287673, + "grad_norm": 2.184678316116333, + "learning_rate": 1.2286758732737613e-06, + "loss": 2.0485, + "step": 2147 + }, + { + "epoch": 2.353972602739726, + "grad_norm": 2.1494438648223877, + "learning_rate": 1.2266450040617386e-06, + "loss": 2.1399, + "step": 2148 + }, + { + "epoch": 2.355068493150685, + "grad_norm": 2.197528839111328, + "learning_rate": 1.2246141348497158e-06, + "loss": 2.1553, + "step": 2149 + }, + { + "epoch": 2.356164383561644, + "grad_norm": 2.222090005874634, + "learning_rate": 1.2225832656376931e-06, + "loss": 2.0896, + "step": 2150 + }, + { + "epoch": 2.356164383561644, + "eval_loss": 2.261875867843628, + "eval_runtime": 50.0806, + "eval_samples_per_second": 19.968, + "eval_steps_per_second": 0.839, + "step": 2150 + }, + { + "epoch": 2.357260273972603, + "grad_norm": 2.157257318496704, + "learning_rate": 1.2205523964256704e-06, + "loss": 2.0824, + "step": 2151 + }, + { + "epoch": 2.3583561643835615, + "grad_norm": 2.1270229816436768, + "learning_rate": 1.2185215272136477e-06, + "loss": 2.08, + "step": 2152 + }, + { + "epoch": 2.3594520547945206, + "grad_norm": 2.1364707946777344, + "learning_rate": 1.2164906580016247e-06, + "loss": 2.091, + "step": 2153 + }, + { + "epoch": 2.3605479452054796, + "grad_norm": 2.1324431896209717, + "learning_rate": 1.214459788789602e-06, + "loss": 2.0935, + "step": 2154 + }, + { + "epoch": 2.361643835616438, + "grad_norm": 2.1576826572418213, + "learning_rate": 1.2124289195775793e-06, + "loss": 2.0128, + "step": 2155 + }, + { + "epoch": 2.3627397260273972, + "grad_norm": 2.145265579223633, + "learning_rate": 1.2103980503655566e-06, + "loss": 2.0418, + "step": 2156 + }, + { + "epoch": 2.3638356164383563, + "grad_norm": 2.1475625038146973, + "learning_rate": 1.2083671811535338e-06, + "loss": 2.0726, + "step": 2157 + }, + { + "epoch": 2.364931506849315, + "grad_norm": 2.2365241050720215, + "learning_rate": 1.2063363119415111e-06, + "loss": 2.0598, + "step": 2158 + }, + { + "epoch": 2.366027397260274, + "grad_norm": 2.208568572998047, + "learning_rate": 1.2043054427294882e-06, + "loss": 2.0614, + "step": 2159 + }, + { + "epoch": 2.367123287671233, + "grad_norm": 2.1937685012817383, + "learning_rate": 1.2022745735174654e-06, + "loss": 2.0799, + "step": 2160 + }, + { + "epoch": 2.367123287671233, + "eval_loss": 2.2633299827575684, + "eval_runtime": 49.4985, + "eval_samples_per_second": 20.203, + "eval_steps_per_second": 0.849, + "step": 2160 + }, + { + "epoch": 2.368219178082192, + "grad_norm": 2.1548843383789062, + "learning_rate": 1.2002437043054427e-06, + "loss": 2.092, + "step": 2161 + }, + { + "epoch": 2.3693150684931505, + "grad_norm": 2.139727830886841, + "learning_rate": 1.19821283509342e-06, + "loss": 2.1167, + "step": 2162 + }, + { + "epoch": 2.3704109589041096, + "grad_norm": 2.1580495834350586, + "learning_rate": 1.1961819658813973e-06, + "loss": 2.1222, + "step": 2163 + }, + { + "epoch": 2.3715068493150686, + "grad_norm": 2.1156060695648193, + "learning_rate": 1.1941510966693746e-06, + "loss": 2.0996, + "step": 2164 + }, + { + "epoch": 2.3726027397260276, + "grad_norm": 2.280191659927368, + "learning_rate": 1.1921202274573518e-06, + "loss": 2.0737, + "step": 2165 + }, + { + "epoch": 2.3736986301369862, + "grad_norm": 2.1379661560058594, + "learning_rate": 1.190089358245329e-06, + "loss": 2.0803, + "step": 2166 + }, + { + "epoch": 2.3747945205479453, + "grad_norm": 2.261583089828491, + "learning_rate": 1.1880584890333064e-06, + "loss": 2.1225, + "step": 2167 + }, + { + "epoch": 2.3758904109589043, + "grad_norm": 2.233469009399414, + "learning_rate": 1.1860276198212837e-06, + "loss": 2.0774, + "step": 2168 + }, + { + "epoch": 2.376986301369863, + "grad_norm": 2.2570717334747314, + "learning_rate": 1.183996750609261e-06, + "loss": 2.1057, + "step": 2169 + }, + { + "epoch": 2.378082191780822, + "grad_norm": 2.2743802070617676, + "learning_rate": 1.1819658813972382e-06, + "loss": 2.093, + "step": 2170 + }, + { + "epoch": 2.378082191780822, + "eval_loss": 2.2612462043762207, + "eval_runtime": 50.375, + "eval_samples_per_second": 19.851, + "eval_steps_per_second": 0.834, + "step": 2170 + }, + { + "epoch": 2.379178082191781, + "grad_norm": 2.2214903831481934, + "learning_rate": 1.1799350121852153e-06, + "loss": 2.084, + "step": 2171 + }, + { + "epoch": 2.3802739726027395, + "grad_norm": 2.143113136291504, + "learning_rate": 1.1779041429731925e-06, + "loss": 2.1153, + "step": 2172 + }, + { + "epoch": 2.3813698630136986, + "grad_norm": 2.243147134780884, + "learning_rate": 1.1758732737611698e-06, + "loss": 2.1186, + "step": 2173 + }, + { + "epoch": 2.3824657534246576, + "grad_norm": 2.2865614891052246, + "learning_rate": 1.173842404549147e-06, + "loss": 2.0712, + "step": 2174 + }, + { + "epoch": 2.383561643835616, + "grad_norm": 2.274543523788452, + "learning_rate": 1.1718115353371244e-06, + "loss": 2.1214, + "step": 2175 + }, + { + "epoch": 2.3846575342465752, + "grad_norm": 2.249037981033325, + "learning_rate": 1.1697806661251016e-06, + "loss": 2.0703, + "step": 2176 + }, + { + "epoch": 2.3857534246575343, + "grad_norm": 2.2552239894866943, + "learning_rate": 1.167749796913079e-06, + "loss": 2.1198, + "step": 2177 + }, + { + "epoch": 2.3868493150684933, + "grad_norm": 2.2714426517486572, + "learning_rate": 1.1657189277010562e-06, + "loss": 2.1037, + "step": 2178 + }, + { + "epoch": 2.387945205479452, + "grad_norm": 2.2843732833862305, + "learning_rate": 1.1636880584890335e-06, + "loss": 2.0833, + "step": 2179 + }, + { + "epoch": 2.389041095890411, + "grad_norm": 2.3192999362945557, + "learning_rate": 1.1616571892770108e-06, + "loss": 2.0807, + "step": 2180 + }, + { + "epoch": 2.389041095890411, + "eval_loss": 2.2609643936157227, + "eval_runtime": 45.7294, + "eval_samples_per_second": 21.868, + "eval_steps_per_second": 0.918, + "step": 2180 + }, + { + "epoch": 2.39013698630137, + "grad_norm": 2.2852554321289062, + "learning_rate": 1.159626320064988e-06, + "loss": 2.1198, + "step": 2181 + }, + { + "epoch": 2.391232876712329, + "grad_norm": 2.3606770038604736, + "learning_rate": 1.1575954508529653e-06, + "loss": 2.0938, + "step": 2182 + }, + { + "epoch": 2.3923287671232876, + "grad_norm": 2.2159781455993652, + "learning_rate": 1.1555645816409426e-06, + "loss": 2.1089, + "step": 2183 + }, + { + "epoch": 2.3934246575342466, + "grad_norm": 2.2246878147125244, + "learning_rate": 1.1535337124289196e-06, + "loss": 2.0974, + "step": 2184 + }, + { + "epoch": 2.3945205479452056, + "grad_norm": 2.180250406265259, + "learning_rate": 1.151502843216897e-06, + "loss": 2.0527, + "step": 2185 + }, + { + "epoch": 2.3956164383561642, + "grad_norm": 2.1486611366271973, + "learning_rate": 1.1494719740048742e-06, + "loss": 2.0251, + "step": 2186 + }, + { + "epoch": 2.3967123287671233, + "grad_norm": 2.242518424987793, + "learning_rate": 1.1474411047928515e-06, + "loss": 2.0885, + "step": 2187 + }, + { + "epoch": 2.3978082191780823, + "grad_norm": 2.1474642753601074, + "learning_rate": 1.1454102355808287e-06, + "loss": 2.0946, + "step": 2188 + }, + { + "epoch": 2.398904109589041, + "grad_norm": 2.230221748352051, + "learning_rate": 1.143379366368806e-06, + "loss": 2.0639, + "step": 2189 + }, + { + "epoch": 2.4, + "grad_norm": 2.1985819339752197, + "learning_rate": 1.141348497156783e-06, + "loss": 2.0578, + "step": 2190 + }, + { + "epoch": 2.4, + "eval_loss": 2.264763832092285, + "eval_runtime": 48.4178, + "eval_samples_per_second": 20.654, + "eval_steps_per_second": 0.867, + "step": 2190 + }, + { + "epoch": 2.401095890410959, + "grad_norm": 2.2248053550720215, + "learning_rate": 1.1393176279447604e-06, + "loss": 2.149, + "step": 2191 + }, + { + "epoch": 2.402191780821918, + "grad_norm": 2.197953939437866, + "learning_rate": 1.1372867587327376e-06, + "loss": 2.138, + "step": 2192 + }, + { + "epoch": 2.4032876712328766, + "grad_norm": 2.2678914070129395, + "learning_rate": 1.135255889520715e-06, + "loss": 2.0202, + "step": 2193 + }, + { + "epoch": 2.4043835616438356, + "grad_norm": 2.3155181407928467, + "learning_rate": 1.1332250203086922e-06, + "loss": 2.1037, + "step": 2194 + }, + { + "epoch": 2.4054794520547946, + "grad_norm": 2.191927909851074, + "learning_rate": 1.1311941510966695e-06, + "loss": 2.0933, + "step": 2195 + }, + { + "epoch": 2.4065753424657537, + "grad_norm": 2.28680419921875, + "learning_rate": 1.1291632818846467e-06, + "loss": 2.0694, + "step": 2196 + }, + { + "epoch": 2.4076712328767123, + "grad_norm": 2.1030619144439697, + "learning_rate": 1.127132412672624e-06, + "loss": 2.0599, + "step": 2197 + }, + { + "epoch": 2.4087671232876713, + "grad_norm": 2.170182228088379, + "learning_rate": 1.1251015434606013e-06, + "loss": 2.0769, + "step": 2198 + }, + { + "epoch": 2.4098630136986303, + "grad_norm": 2.372509241104126, + "learning_rate": 1.1230706742485786e-06, + "loss": 2.0848, + "step": 2199 + }, + { + "epoch": 2.410958904109589, + "grad_norm": 2.3988277912139893, + "learning_rate": 1.1210398050365558e-06, + "loss": 2.1207, + "step": 2200 + }, + { + "epoch": 2.410958904109589, + "eval_loss": 2.261070966720581, + "eval_runtime": 47.1736, + "eval_samples_per_second": 21.198, + "eval_steps_per_second": 0.89, + "step": 2200 + } + ], + "logging_steps": 1, + "max_steps": 2736, + "num_input_tokens_seen": 0, + "num_train_epochs": 3, + "save_steps": 100, + "stateful_callbacks": { + "TrainerControl": { + "args": { + "should_epoch_stop": false, + "should_evaluate": false, + "should_log": false, + "should_save": true, + "should_training_stop": false + }, + "attributes": {} + } + }, + "total_flos": 6.858080700596224e+17, + "train_batch_size": 3, + "trial_name": null, + "trial_params": null +} diff --git a/babylm_shuffle_nondeterministic_10M_seed0/runs/checkpoint-2200/zero_to_fp32.py b/babylm_shuffle_nondeterministic_10M_seed0/runs/checkpoint-2200/zero_to_fp32.py new file mode 100644 index 0000000000000000000000000000000000000000..24cc342e78d1a006c782b3a4cd68d9ce786d8fd8 --- /dev/null +++ b/babylm_shuffle_nondeterministic_10M_seed0/runs/checkpoint-2200/zero_to_fp32.py @@ -0,0 +1,604 @@ +#!/usr/bin/env python + +# Copyright (c) Microsoft Corporation. +# SPDX-License-Identifier: Apache-2.0 + +# DeepSpeed Team + +# This script extracts fp32 consolidated weights from a zero 1, 2 and 3 DeepSpeed checkpoints. It gets +# copied into the top level checkpoint dir, so the user can easily do the conversion at any point in +# the future. Once extracted, the weights don't require DeepSpeed and can be used in any +# application. +# +# example: python zero_to_fp32.py . pytorch_model.bin + +import argparse +import torch +import glob +import math +import os +import re +from collections import OrderedDict +from dataclasses import dataclass + +# while this script doesn't use deepspeed to recover data, since the checkpoints are pickled with +# DeepSpeed data structures it has to be available in the current python environment. +from deepspeed.utils import logger +from deepspeed.checkpoint.constants import (DS_VERSION, OPTIMIZER_STATE_DICT, SINGLE_PARTITION_OF_FP32_GROUPS, + FP32_FLAT_GROUPS, ZERO_STAGE, PARTITION_COUNT, PARAM_SHAPES, BUFFER_NAMES, + FROZEN_PARAM_SHAPES, FROZEN_PARAM_FRAGMENTS) + + +@dataclass +class zero_model_state: + buffers: dict() + param_shapes: dict() + shared_params: list + ds_version: int + frozen_param_shapes: dict() + frozen_param_fragments: dict() + + +debug = 0 + +# load to cpu +device = torch.device('cpu') + + +def atoi(text): + return int(text) if text.isdigit() else text + + +def natural_keys(text): + ''' + alist.sort(key=natural_keys) sorts in human order + http://nedbatchelder.com/blog/200712/human_sorting.html + (See Toothy's implementation in the comments) + ''' + return [atoi(c) for c in re.split(r'(\d+)', text)] + + +def get_model_state_file(checkpoint_dir, zero_stage): + if not os.path.isdir(checkpoint_dir): + raise FileNotFoundError(f"Directory '{checkpoint_dir}' doesn't exist") + + # there should be only one file + if zero_stage <= 2: + file = os.path.join(checkpoint_dir, "mp_rank_00_model_states.pt") + elif zero_stage == 3: + file = os.path.join(checkpoint_dir, "zero_pp_rank_0_mp_rank_00_model_states.pt") + + if not os.path.exists(file): + raise FileNotFoundError(f"can't find model states file at '{file}'") + + return file + + +def get_checkpoint_files(checkpoint_dir, glob_pattern): + # XXX: need to test that this simple glob rule works for multi-node setup too + ckpt_files = sorted(glob.glob(os.path.join(checkpoint_dir, glob_pattern)), key=natural_keys) + + if len(ckpt_files) == 0: + raise FileNotFoundError(f"can't find {glob_pattern} files in directory '{checkpoint_dir}'") + + return ckpt_files + + +def get_optim_files(checkpoint_dir): + return get_checkpoint_files(checkpoint_dir, "*_optim_states.pt") + + +def get_model_state_files(checkpoint_dir): + return get_checkpoint_files(checkpoint_dir, "*_model_states.pt") + + +def parse_model_states(files): + zero_model_states = [] + for file in files: + state_dict = torch.load(file, map_location=device) + + if BUFFER_NAMES not in state_dict: + raise ValueError(f"{file} is not a model state checkpoint") + buffer_names = state_dict[BUFFER_NAMES] + if debug: + print("Found buffers:", buffer_names) + + # recover just the buffers while restoring them to fp32 if they were saved in fp16 + buffers = {k: v.float() for k, v in state_dict["module"].items() if k in buffer_names} + param_shapes = state_dict[PARAM_SHAPES] + + # collect parameters that are included in param_shapes + param_names = [] + for s in param_shapes: + for name in s.keys(): + param_names.append(name) + + # update with frozen parameters + frozen_param_shapes = state_dict.get(FROZEN_PARAM_SHAPES, None) + if frozen_param_shapes is not None: + if debug: + print(f"Found frozen_param_shapes: {frozen_param_shapes}") + param_names += list(frozen_param_shapes.keys()) + + # handle shared params + shared_params = [[k, v] for k, v in state_dict["shared_params"].items()] + + ds_version = state_dict.get(DS_VERSION, None) + + frozen_param_fragments = state_dict.get(FROZEN_PARAM_FRAGMENTS, None) + + z_model_state = zero_model_state(buffers=buffers, + param_shapes=param_shapes, + shared_params=shared_params, + ds_version=ds_version, + frozen_param_shapes=frozen_param_shapes, + frozen_param_fragments=frozen_param_fragments) + zero_model_states.append(z_model_state) + + return zero_model_states + + +def parse_optim_states(files, ds_checkpoint_dir): + + total_files = len(files) + state_dicts = [] + for f in files: + state_dict = torch.load(f, map_location=device) + # immediately discard the potentially huge 2 optimizer states as we only care for fp32 master weights + # and also handle the case where it was already removed by another helper script + state_dict["optimizer_state_dict"].pop("optimizer_state_dict", None) + state_dicts.append(state_dict) + + if not ZERO_STAGE in state_dicts[0][OPTIMIZER_STATE_DICT]: + raise ValueError(f"{files[0]} is not a zero checkpoint") + zero_stage = state_dicts[0][OPTIMIZER_STATE_DICT][ZERO_STAGE] + world_size = state_dicts[0][OPTIMIZER_STATE_DICT][PARTITION_COUNT] + + # For ZeRO-2 each param group can have different partition_count as data parallelism for expert + # parameters can be different from data parallelism for non-expert parameters. So we can just + # use the max of the partition_count to get the dp world_size. + + if type(world_size) is list: + world_size = max(world_size) + + if world_size != total_files: + raise ValueError( + f"Expected {world_size} of '*_optim_states.pt' under '{ds_checkpoint_dir}' but found {total_files} files. " + "Possibly due to an overwrite of an old checkpoint, or a checkpoint didn't get saved by one or more processes." + ) + + # the groups are named differently in each stage + if zero_stage <= 2: + fp32_groups_key = SINGLE_PARTITION_OF_FP32_GROUPS + elif zero_stage == 3: + fp32_groups_key = FP32_FLAT_GROUPS + else: + raise ValueError(f"unknown zero stage {zero_stage}") + + if zero_stage <= 2: + fp32_flat_groups = [state_dicts[i][OPTIMIZER_STATE_DICT][fp32_groups_key] for i in range(len(state_dicts))] + elif zero_stage == 3: + # if there is more than one param group, there will be multiple flattened tensors - one + # flattened tensor per group - for simplicity merge them into a single tensor + # + # XXX: could make the script more memory efficient for when there are multiple groups - it + # will require matching the sub-lists of param_shapes for each param group flattened tensor + + fp32_flat_groups = [ + torch.cat(state_dicts[i][OPTIMIZER_STATE_DICT][fp32_groups_key], 0) for i in range(len(state_dicts)) + ] + + return zero_stage, world_size, fp32_flat_groups + + +def _get_fp32_state_dict_from_zero_checkpoint(ds_checkpoint_dir, exclude_frozen_parameters): + """ + Returns fp32 state_dict reconstructed from ds checkpoint + + Args: + - ``ds_checkpoint_dir``: path to the deepspeed checkpoint folder (where the optimizer files are) + + """ + print(f"Processing zero checkpoint '{ds_checkpoint_dir}'") + + optim_files = get_optim_files(ds_checkpoint_dir) + zero_stage, world_size, fp32_flat_groups = parse_optim_states(optim_files, ds_checkpoint_dir) + print(f"Detected checkpoint of type zero stage {zero_stage}, world_size: {world_size}") + + model_files = get_model_state_files(ds_checkpoint_dir) + + zero_model_states = parse_model_states(model_files) + print(f'Parsing checkpoint created by deepspeed=={zero_model_states[0].ds_version}') + + if zero_stage <= 2: + return _get_fp32_state_dict_from_zero2_checkpoint(world_size, fp32_flat_groups, zero_model_states, + exclude_frozen_parameters) + elif zero_stage == 3: + return _get_fp32_state_dict_from_zero3_checkpoint(world_size, fp32_flat_groups, zero_model_states, + exclude_frozen_parameters) + + +def _zero2_merge_frozen_params(state_dict, zero_model_states): + if zero_model_states[0].frozen_param_shapes is None or len(zero_model_states[0].frozen_param_shapes) == 0: + return + + frozen_param_shapes = zero_model_states[0].frozen_param_shapes + frozen_param_fragments = zero_model_states[0].frozen_param_fragments + + if debug: + num_elem = sum(s.numel() for s in frozen_param_shapes.values()) + print(f'rank 0: {FROZEN_PARAM_SHAPES}.numel = {num_elem}') + + wanted_params = len(frozen_param_shapes) + wanted_numel = sum(s.numel() for s in frozen_param_shapes.values()) + avail_numel = sum([p.numel() for p in frozen_param_fragments.values()]) + print(f'Frozen params: Have {avail_numel} numels to process.') + print(f'Frozen params: Need {wanted_numel} numels in {wanted_params} params') + + total_params = 0 + total_numel = 0 + for name, shape in frozen_param_shapes.items(): + total_params += 1 + unpartitioned_numel = shape.numel() + total_numel += unpartitioned_numel + + state_dict[name] = frozen_param_fragments[name] + + if debug: + print(f"{name} full shape: {shape} unpartitioned numel {unpartitioned_numel} ") + + print(f"Reconstructed Frozen fp32 state dict with {total_params} params {total_numel} elements") + + +def _has_callable(obj, fn): + attr = getattr(obj, fn, None) + return callable(attr) + + +def _zero2_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states): + param_shapes = zero_model_states[0].param_shapes + + # Reconstruction protocol: + # + # XXX: document this + + if debug: + for i in range(world_size): + for j in range(len(fp32_flat_groups[0])): + print(f"{FP32_FLAT_GROUPS}[{i}][{j}].shape={fp32_flat_groups[i][j].shape}") + + # XXX: memory usage doubles here (zero2) + num_param_groups = len(fp32_flat_groups[0]) + merged_single_partition_of_fp32_groups = [] + for i in range(num_param_groups): + merged_partitions = [sd[i] for sd in fp32_flat_groups] + full_single_fp32_vector = torch.cat(merged_partitions, 0) + merged_single_partition_of_fp32_groups.append(full_single_fp32_vector) + avail_numel = sum( + [full_single_fp32_vector.numel() for full_single_fp32_vector in merged_single_partition_of_fp32_groups]) + + if debug: + wanted_params = sum([len(shapes) for shapes in param_shapes]) + wanted_numel = sum([sum(shape.numel() for shape in shapes.values()) for shapes in param_shapes]) + # not asserting if there is a mismatch due to possible padding + print(f"Have {avail_numel} numels to process.") + print(f"Need {wanted_numel} numels in {wanted_params} params.") + + # params + # XXX: for huge models that can't fit into the host's RAM we will have to recode this to support + # out-of-core computing solution + total_numel = 0 + total_params = 0 + for shapes, full_single_fp32_vector in zip(param_shapes, merged_single_partition_of_fp32_groups): + offset = 0 + avail_numel = full_single_fp32_vector.numel() + for name, shape in shapes.items(): + + unpartitioned_numel = shape.numel() if _has_callable(shape, 'numel') else math.prod(shape) + total_numel += unpartitioned_numel + total_params += 1 + + if debug: + print(f"{name} full shape: {shape} unpartitioned numel {unpartitioned_numel} ") + state_dict[name] = full_single_fp32_vector.narrow(0, offset, unpartitioned_numel).view(shape) + offset += unpartitioned_numel + + # Z2 started to align to 2*world_size to improve nccl performance. Therefore both offset and + # avail_numel can differ by anywhere between 0..2*world_size. Due to two unrelated complex + # paddings performed in the code it's almost impossible to predict the exact numbers w/o the + # live optimizer object, so we are checking that the numbers are within the right range + align_to = 2 * world_size + + def zero2_align(x): + return align_to * math.ceil(x / align_to) + + if debug: + print(f"original offset={offset}, avail_numel={avail_numel}") + + offset = zero2_align(offset) + avail_numel = zero2_align(avail_numel) + + if debug: + print(f"aligned offset={offset}, avail_numel={avail_numel}") + + # Sanity check + if offset != avail_numel: + raise ValueError(f"consumed {offset} numels out of {avail_numel} - something is wrong") + + print(f"Reconstructed fp32 state dict with {total_params} params {total_numel} elements") + + +def _get_fp32_state_dict_from_zero2_checkpoint(world_size, fp32_flat_groups, zero_model_states, + exclude_frozen_parameters): + state_dict = OrderedDict() + + # buffers + buffers = zero_model_states[0].buffers + state_dict.update(buffers) + if debug: + print(f"added {len(buffers)} buffers") + + if not exclude_frozen_parameters: + _zero2_merge_frozen_params(state_dict, zero_model_states) + + _zero2_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states) + + # recover shared parameters + for pair in zero_model_states[0].shared_params: + if pair[1] in state_dict: + state_dict[pair[0]] = state_dict[pair[1]] + + return state_dict + + +def zero3_partitioned_param_info(unpartitioned_numel, world_size): + remainder = unpartitioned_numel % world_size + padding_numel = (world_size - remainder) if remainder else 0 + partitioned_numel = math.ceil(unpartitioned_numel / world_size) + return partitioned_numel, padding_numel + + +def _zero3_merge_frozen_params(state_dict, world_size, zero_model_states): + if zero_model_states[0].frozen_param_shapes is None or len(zero_model_states[0].frozen_param_shapes) == 0: + return + + if debug: + for i in range(world_size): + num_elem = sum(s.numel() for s in zero_model_states[i].frozen_param_fragments.values()) + print(f'rank {i}: {FROZEN_PARAM_SHAPES}.numel = {num_elem}') + + frozen_param_shapes = zero_model_states[0].frozen_param_shapes + wanted_params = len(frozen_param_shapes) + wanted_numel = sum(s.numel() for s in frozen_param_shapes.values()) + avail_numel = sum([p.numel() for p in zero_model_states[0].frozen_param_fragments.values()]) * world_size + print(f'Frozen params: Have {avail_numel} numels to process.') + print(f'Frozen params: Need {wanted_numel} numels in {wanted_params} params') + + total_params = 0 + total_numel = 0 + for name, shape in zero_model_states[0].frozen_param_shapes.items(): + total_params += 1 + unpartitioned_numel = shape.numel() + total_numel += unpartitioned_numel + + param_frags = tuple(model_state.frozen_param_fragments[name] for model_state in zero_model_states) + state_dict[name] = torch.cat(param_frags, 0).narrow(0, 0, unpartitioned_numel).view(shape) + + partitioned_numel, partitioned_padding_numel = zero3_partitioned_param_info(unpartitioned_numel, world_size) + + if debug: + print( + f"Frozen params: {total_params} {name} full shape: {shape} partition0 numel={partitioned_numel} partitioned_padding_numel={partitioned_padding_numel}" + ) + + print(f"Reconstructed Frozen fp32 state dict with {total_params} params {total_numel} elements") + + +def _zero3_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states): + param_shapes = zero_model_states[0].param_shapes + avail_numel = fp32_flat_groups[0].numel() * world_size + # Reconstruction protocol: For zero3 we need to zip the partitions together at boundary of each + # param, re-consolidating each param, while dealing with padding if any + + # merge list of dicts, preserving order + param_shapes = {k: v for d in param_shapes for k, v in d.items()} + + if debug: + for i in range(world_size): + print(f"{FP32_FLAT_GROUPS}[{i}].shape={fp32_flat_groups[i].shape}") + + wanted_params = len(param_shapes) + wanted_numel = sum(shape.numel() for shape in param_shapes.values()) + # not asserting if there is a mismatch due to possible padding + avail_numel = fp32_flat_groups[0].numel() * world_size + print(f"Trainable params: Have {avail_numel} numels to process.") + print(f"Trainable params: Need {wanted_numel} numels in {wanted_params} params.") + + # params + # XXX: for huge models that can't fit into the host's RAM we will have to recode this to support + # out-of-core computing solution + offset = 0 + total_numel = 0 + total_params = 0 + for name, shape in param_shapes.items(): + + unpartitioned_numel = shape.numel() + total_numel += unpartitioned_numel + total_params += 1 + + partitioned_numel, partitioned_padding_numel = zero3_partitioned_param_info(unpartitioned_numel, world_size) + + if debug: + print( + f"Trainable params: {total_params} {name} full shape: {shape} partition0 numel={partitioned_numel} partitioned_padding_numel={partitioned_padding_numel}" + ) + + # XXX: memory usage doubles here + state_dict[name] = torch.cat( + tuple(fp32_flat_groups[i].narrow(0, offset, partitioned_numel) for i in range(world_size)), + 0).narrow(0, 0, unpartitioned_numel).view(shape) + offset += partitioned_numel + + offset *= world_size + + # Sanity check + if offset != avail_numel: + raise ValueError(f"consumed {offset} numels out of {avail_numel} - something is wrong") + + print(f"Reconstructed Trainable fp32 state dict with {total_params} params {total_numel} elements") + + +def _get_fp32_state_dict_from_zero3_checkpoint(world_size, fp32_flat_groups, zero_model_states, + exclude_frozen_parameters): + state_dict = OrderedDict() + + # buffers + buffers = zero_model_states[0].buffers + state_dict.update(buffers) + if debug: + print(f"added {len(buffers)} buffers") + + if not exclude_frozen_parameters: + _zero3_merge_frozen_params(state_dict, world_size, zero_model_states) + + _zero3_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states) + + # recover shared parameters + for pair in zero_model_states[0].shared_params: + if pair[1] in state_dict: + state_dict[pair[0]] = state_dict[pair[1]] + + return state_dict + + +def get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir, tag=None, exclude_frozen_parameters=False): + """ + Convert ZeRO 2 or 3 checkpoint into a single fp32 consolidated state_dict that can be loaded with + ``load_state_dict()`` and used for training without DeepSpeed or shared with others, for example + via a model hub. + + Args: + - ``checkpoint_dir``: path to the desired checkpoint folder + - ``tag``: checkpoint tag used as a unique identifier for checkpoint. If not provided will attempt to load tag in 'latest' file. e.g., ``global_step14`` + - ``exclude_frozen_parameters``: exclude frozen parameters + + Returns: + - pytorch ``state_dict`` + + Note: this approach may not work if your application doesn't have sufficient free CPU memory and + you may need to use the offline approach using the ``zero_to_fp32.py`` script that is saved with + the checkpoint. + + A typical usage might be :: + + from deepspeed.utils.zero_to_fp32 import get_fp32_state_dict_from_zero_checkpoint + # do the training and checkpoint saving + state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir) # already on cpu + model = model.cpu() # move to cpu + model.load_state_dict(state_dict) + # submit to model hub or save the model to share with others + + In this example the ``model`` will no longer be usable in the deepspeed context of the same + application. i.e. you will need to re-initialize the deepspeed engine, since + ``model.load_state_dict(state_dict)`` will remove all the deepspeed magic from it. + + If you want it all done for you, use ``load_state_dict_from_zero_checkpoint`` instead. + + """ + if tag is None: + latest_path = os.path.join(checkpoint_dir, 'latest') + if os.path.isfile(latest_path): + with open(latest_path, 'r') as fd: + tag = fd.read().strip() + else: + raise ValueError(f"Unable to find 'latest' file at {latest_path}") + + ds_checkpoint_dir = os.path.join(checkpoint_dir, tag) + + if not os.path.isdir(ds_checkpoint_dir): + raise FileNotFoundError(f"Directory '{ds_checkpoint_dir}' doesn't exist") + + return _get_fp32_state_dict_from_zero_checkpoint(ds_checkpoint_dir, exclude_frozen_parameters) + + +def convert_zero_checkpoint_to_fp32_state_dict(checkpoint_dir, output_file, tag=None, exclude_frozen_parameters=False): + """ + Convert ZeRO 2 or 3 checkpoint into a single fp32 consolidated ``state_dict`` file that can be + loaded with ``torch.load(file)`` + ``load_state_dict()`` and used for training without DeepSpeed. + + Args: + - ``checkpoint_dir``: path to the desired checkpoint folder. (one that contains the tag-folder, like ``global_step14``) + - ``output_file``: path to the pytorch fp32 state_dict output file (e.g. path/pytorch_model.bin) + - ``tag``: checkpoint tag used as a unique identifier for checkpoint. If not provided will attempt to load tag in the file named ``latest`` in the checkpoint folder, e.g., ``global_step14`` + - ``exclude_frozen_parameters``: exclude frozen parameters + """ + + state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir, tag, exclude_frozen_parameters) + print(f"Saving fp32 state dict to {output_file}") + torch.save(state_dict, output_file) + + +def load_state_dict_from_zero_checkpoint(model, checkpoint_dir, tag=None): + """ + 1. Put the provided model to cpu + 2. Convert ZeRO 2 or 3 checkpoint into a single fp32 consolidated ``state_dict`` + 3. Load it into the provided model + + Args: + - ``model``: the model object to update + - ``checkpoint_dir``: path to the desired checkpoint folder. (one that contains the tag-folder, like ``global_step14``) + - ``tag``: checkpoint tag used as a unique identifier for checkpoint. If not provided will attempt to load tag in the file named ``latest`` in the checkpoint folder, e.g., ``global_step14`` + + Returns: + - ``model`: modified model + + Make sure you have plenty of CPU memory available before you call this function. If you don't + have enough use the ``zero_to_fp32.py`` utility to do the conversion. You will find it + conveniently placed for you in the checkpoint folder. + + A typical usage might be :: + + from deepspeed.utils.zero_to_fp32 import load_state_dict_from_zero_checkpoint + model = load_state_dict_from_zero_checkpoint(trainer.model, checkpoint_dir) + # submit to model hub or save the model to share with others + + Note, that once this was run, the ``model`` will no longer be usable in the deepspeed context + of the same application. i.e. you will need to re-initialize the deepspeed engine, since + ``model.load_state_dict(state_dict)`` will remove all the deepspeed magic from it. + + """ + logger.info(f"Extracting fp32 weights") + state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir, tag) + + logger.info(f"Overwriting model with fp32 weights") + model = model.cpu() + model.load_state_dict(state_dict, strict=False) + + return model + + +if __name__ == "__main__": + + parser = argparse.ArgumentParser() + parser.add_argument("checkpoint_dir", + type=str, + help="path to the desired checkpoint folder, e.g., path/checkpoint-12") + parser.add_argument( + "output_file", + type=str, + help="path to the pytorch fp32 state_dict output file (e.g. path/checkpoint-12/pytorch_model.bin)") + parser.add_argument("-t", + "--tag", + type=str, + default=None, + help="checkpoint tag used as a unique identifier for checkpoint. e.g., global_step1") + parser.add_argument("--exclude_frozen_parameters", action='store_true', help="exclude frozen parameters") + parser.add_argument("-d", "--debug", action='store_true', help="enable debug") + args = parser.parse_args() + + debug = args.debug + + convert_zero_checkpoint_to_fp32_state_dict(args.checkpoint_dir, + args.output_file, + tag=args.tag, + exclude_frozen_parameters=args.exclude_frozen_parameters) diff --git a/babylm_shuffle_nondeterministic_10M_seed0/runs/checkpoint-2300/config.json b/babylm_shuffle_nondeterministic_10M_seed0/runs/checkpoint-2300/config.json new file mode 100644 index 0000000000000000000000000000000000000000..fb0f9a53a507ffcdb60410deb71d60ef801bf350 --- /dev/null +++ b/babylm_shuffle_nondeterministic_10M_seed0/runs/checkpoint-2300/config.json @@ -0,0 +1,36 @@ +{ + "_name_or_path": "meta-llama/Llama-3.2-3B", + "architectures": [ + "LlamaForCausalLM" + ], + "attention_bias": false, + "attention_dropout": 0.0, + "bos_token_id": 128000, + "eos_token_id": 128001, + "head_dim": 128, + "hidden_act": "silu", + "hidden_size": 3072, + "initializer_range": 0.02, + "intermediate_size": 8192, + "max_position_embeddings": 131072, + "mlp_bias": false, + "model_type": "llama", + "num_attention_heads": 24, + "num_hidden_layers": 28, + "num_key_value_heads": 8, + "pretraining_tp": 1, + "rms_norm_eps": 1e-05, + "rope_scaling": { + "factor": 32.0, + "high_freq_factor": 4.0, + "low_freq_factor": 1.0, + "original_max_position_embeddings": 8192, + "rope_type": "llama3" + }, + "rope_theta": 500000.0, + "tie_word_embeddings": true, + "torch_dtype": "float16", + "transformers_version": "4.45.1", + "use_cache": true, + "vocab_size": 128256 +} diff --git a/babylm_shuffle_nondeterministic_10M_seed0/runs/checkpoint-2300/generation_config.json b/babylm_shuffle_nondeterministic_10M_seed0/runs/checkpoint-2300/generation_config.json new file mode 100644 index 0000000000000000000000000000000000000000..9c389b841a9a29ddd904e02d1eb0e08dcce82ad9 --- /dev/null +++ b/babylm_shuffle_nondeterministic_10M_seed0/runs/checkpoint-2300/generation_config.json @@ -0,0 +1,9 @@ +{ + "_from_model_config": true, + "bos_token_id": 128000, + "do_sample": true, + "eos_token_id": 128001, + "temperature": 0.6, + "top_p": 0.9, + "transformers_version": "4.45.1" +} diff --git a/babylm_shuffle_nondeterministic_10M_seed0/runs/checkpoint-2300/latest b/babylm_shuffle_nondeterministic_10M_seed0/runs/checkpoint-2300/latest new file mode 100644 index 0000000000000000000000000000000000000000..78dc01ce4ce1498b753887dfb7af92409705a69b --- /dev/null +++ b/babylm_shuffle_nondeterministic_10M_seed0/runs/checkpoint-2300/latest @@ -0,0 +1 @@ +global_step2300 \ No newline at end of file diff --git a/babylm_shuffle_nondeterministic_10M_seed0/runs/checkpoint-2300/model.safetensors.index.json b/babylm_shuffle_nondeterministic_10M_seed0/runs/checkpoint-2300/model.safetensors.index.json new file mode 100644 index 0000000000000000000000000000000000000000..ed64de846d720b9a7859dc20575fea8e8ca51940 --- /dev/null +++ b/babylm_shuffle_nondeterministic_10M_seed0/runs/checkpoint-2300/model.safetensors.index.json @@ -0,0 +1,262 @@ +{ + "metadata": { + "total_size": 7213504512 + }, + "weight_map": { + "lm_head.weight": "model-00002-of-00002.safetensors", + "model.embed_tokens.weight": "model-00001-of-00002.safetensors", + "model.layers.0.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.0.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.0.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.0.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.0.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.0.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.0.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.0.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.0.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.1.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.1.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.1.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.1.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.1.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.1.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.1.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.1.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.1.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.10.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.10.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.10.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.10.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.10.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.10.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.10.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.10.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.10.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.11.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.11.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.11.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.11.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.11.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.11.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.11.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.11.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.11.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.12.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.12.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.12.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.12.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.12.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.12.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.12.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.12.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.12.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.13.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.13.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.13.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.13.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.13.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.13.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.13.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.13.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.13.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.14.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.14.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.14.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.14.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.14.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.14.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.14.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.14.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.14.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.15.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.15.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.15.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.15.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.15.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.15.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.15.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.15.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.15.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.16.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.16.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.16.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.16.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.16.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.16.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.16.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.16.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.16.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.17.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.17.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.17.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.17.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.17.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.17.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.17.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.17.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.17.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.18.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.18.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.18.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.18.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.18.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.18.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.18.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.18.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.18.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.19.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.19.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.19.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.19.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.19.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.19.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.19.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.19.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.19.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.2.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.2.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.2.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.2.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.2.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.2.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.2.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.2.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.2.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.20.input_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.20.mlp.down_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.20.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.20.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.20.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.20.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.20.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.20.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.20.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.21.input_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.21.mlp.down_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.21.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.21.mlp.up_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.21.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.21.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.21.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.21.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.21.self_attn.v_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.22.input_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.22.mlp.down_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.22.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.22.mlp.up_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.22.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.22.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.22.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.22.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.22.self_attn.v_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.23.input_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.23.mlp.down_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.23.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.23.mlp.up_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.23.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.23.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.23.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.23.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.23.self_attn.v_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.24.input_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.24.mlp.down_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.24.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.24.mlp.up_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.24.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.24.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.24.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.24.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.24.self_attn.v_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.25.input_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.25.mlp.down_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.25.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.25.mlp.up_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.25.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.25.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.25.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.25.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.25.self_attn.v_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.26.input_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.26.mlp.down_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.26.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.26.mlp.up_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.26.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.26.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.26.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.26.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.26.self_attn.v_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.27.input_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.27.mlp.down_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.27.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.27.mlp.up_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.27.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.27.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.27.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.27.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.27.self_attn.v_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.3.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.3.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.3.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.3.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.3.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.3.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.3.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.3.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.3.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.4.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.4.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.4.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.4.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.4.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.4.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.4.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.4.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.4.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.5.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.5.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.5.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.5.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.5.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.5.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.5.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.5.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.5.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.6.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.6.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.6.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.6.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.6.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.6.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.6.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.6.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.6.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.7.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.7.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.7.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.7.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.7.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.7.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.7.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.7.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.7.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.8.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.8.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.8.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.8.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.8.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.8.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.8.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.8.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.8.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.9.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.9.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.9.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.9.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.9.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.9.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.9.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.9.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.9.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.norm.weight": "model-00002-of-00002.safetensors" + } +} diff --git a/babylm_shuffle_nondeterministic_10M_seed0/runs/checkpoint-2300/special_tokens_map.json b/babylm_shuffle_nondeterministic_10M_seed0/runs/checkpoint-2300/special_tokens_map.json new file mode 100644 index 0000000000000000000000000000000000000000..6949c5975ee0e961ef61cf31010dce04df0a03f8 --- /dev/null +++ b/babylm_shuffle_nondeterministic_10M_seed0/runs/checkpoint-2300/special_tokens_map.json @@ -0,0 +1,23 @@ +{ + "bos_token": { + "content": "<|begin_of_text|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false + }, + "eos_token": { + "content": "<|end_of_text|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false + }, + "pad_token": { + "content": "[PAD]", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false + } +} diff --git a/babylm_shuffle_nondeterministic_10M_seed0/runs/checkpoint-2300/tokenizer_config.json b/babylm_shuffle_nondeterministic_10M_seed0/runs/checkpoint-2300/tokenizer_config.json new file mode 100644 index 0000000000000000000000000000000000000000..5684b14e2756b56aa397c1324e239b442162b34a --- /dev/null +++ b/babylm_shuffle_nondeterministic_10M_seed0/runs/checkpoint-2300/tokenizer_config.json @@ -0,0 +1,2070 @@ +{ + "added_tokens_decoder": { + "128000": { + "content": "<|begin_of_text|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128001": { + "content": "<|end_of_text|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128002": { + "content": "<|reserved_special_token_0|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128003": { + "content": "<|reserved_special_token_1|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128004": { + "content": "<|finetune_right_pad_id|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128005": { + "content": "<|reserved_special_token_2|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128006": { + "content": "<|start_header_id|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128007": { + "content": "<|end_header_id|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128008": { + "content": "<|eom_id|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128009": { + "content": "<|eot_id|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128010": { + "content": "<|python_tag|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128011": { + "content": "<|reserved_special_token_3|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128012": { + "content": "<|reserved_special_token_4|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128013": { + "content": "<|reserved_special_token_5|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128014": { + "content": "<|reserved_special_token_6|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128015": { + "content": "<|reserved_special_token_7|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128016": { + "content": "<|reserved_special_token_8|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128017": { + "content": "<|reserved_special_token_9|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128018": { + "content": "<|reserved_special_token_10|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128019": { + "content": "<|reserved_special_token_11|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128020": { + "content": "<|reserved_special_token_12|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128021": { + "content": "<|reserved_special_token_13|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128022": { + "content": "<|reserved_special_token_14|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128023": { + "content": "<|reserved_special_token_15|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128024": { + "content": "<|reserved_special_token_16|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128025": { + "content": "<|reserved_special_token_17|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128026": { + "content": "<|reserved_special_token_18|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128027": { + "content": "<|reserved_special_token_19|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128028": { + "content": "<|reserved_special_token_20|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128029": { + "content": "<|reserved_special_token_21|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128030": { + "content": "<|reserved_special_token_22|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128031": { + "content": "<|reserved_special_token_23|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128032": { + "content": "<|reserved_special_token_24|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128033": { + "content": "<|reserved_special_token_25|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128034": { + "content": "<|reserved_special_token_26|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128035": { + "content": "<|reserved_special_token_27|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128036": { + "content": "<|reserved_special_token_28|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128037": { + "content": "<|reserved_special_token_29|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128038": { + "content": "<|reserved_special_token_30|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128039": { + "content": "<|reserved_special_token_31|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128040": { + "content": "<|reserved_special_token_32|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128041": { + "content": "<|reserved_special_token_33|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128042": { + "content": "<|reserved_special_token_34|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128043": { + "content": "<|reserved_special_token_35|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128044": { + "content": "<|reserved_special_token_36|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128045": { + "content": "<|reserved_special_token_37|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128046": { + "content": "<|reserved_special_token_38|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128047": { + "content": "<|reserved_special_token_39|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128048": { + "content": "<|reserved_special_token_40|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128049": { + "content": "<|reserved_special_token_41|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128050": { + "content": "<|reserved_special_token_42|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128051": { + "content": "<|reserved_special_token_43|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128052": { + "content": "<|reserved_special_token_44|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128053": { + "content": "<|reserved_special_token_45|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128054": { + "content": "<|reserved_special_token_46|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128055": { + "content": "<|reserved_special_token_47|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128056": { + "content": "<|reserved_special_token_48|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128057": { + "content": "<|reserved_special_token_49|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128058": { + "content": "<|reserved_special_token_50|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128059": { + "content": "<|reserved_special_token_51|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128060": { + "content": "<|reserved_special_token_52|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128061": { + "content": "<|reserved_special_token_53|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128062": { + "content": "<|reserved_special_token_54|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128063": { + "content": "<|reserved_special_token_55|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128064": { + "content": "<|reserved_special_token_56|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128065": { + "content": "<|reserved_special_token_57|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128066": { + "content": "<|reserved_special_token_58|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128067": { + "content": "<|reserved_special_token_59|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128068": { + "content": "<|reserved_special_token_60|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128069": { + "content": "<|reserved_special_token_61|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128070": { + "content": "<|reserved_special_token_62|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128071": { + "content": "<|reserved_special_token_63|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128072": { + "content": "<|reserved_special_token_64|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128073": { + "content": "<|reserved_special_token_65|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128074": { + "content": "<|reserved_special_token_66|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128075": { + "content": "<|reserved_special_token_67|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128076": { + "content": "<|reserved_special_token_68|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128077": { + "content": "<|reserved_special_token_69|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128078": { + "content": "<|reserved_special_token_70|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128079": { + "content": "<|reserved_special_token_71|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128080": { + "content": "<|reserved_special_token_72|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128081": { + "content": "<|reserved_special_token_73|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128082": { + "content": "<|reserved_special_token_74|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128083": { + "content": "<|reserved_special_token_75|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128084": { + "content": "<|reserved_special_token_76|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128085": { + "content": "<|reserved_special_token_77|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128086": { + "content": "<|reserved_special_token_78|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128087": { + "content": "<|reserved_special_token_79|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128088": { + "content": "<|reserved_special_token_80|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128089": { + "content": "<|reserved_special_token_81|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128090": { + "content": "<|reserved_special_token_82|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128091": { + "content": "<|reserved_special_token_83|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128092": { + "content": "<|reserved_special_token_84|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128093": { + "content": "<|reserved_special_token_85|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128094": { + "content": "<|reserved_special_token_86|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128095": { + "content": "<|reserved_special_token_87|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128096": { + "content": "<|reserved_special_token_88|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128097": { + "content": "<|reserved_special_token_89|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128098": { + "content": "<|reserved_special_token_90|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128099": { + "content": "<|reserved_special_token_91|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128100": { + "content": "<|reserved_special_token_92|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128101": { + "content": "<|reserved_special_token_93|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128102": { + "content": "<|reserved_special_token_94|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128103": { + "content": "<|reserved_special_token_95|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128104": { + "content": "<|reserved_special_token_96|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128105": { + "content": "<|reserved_special_token_97|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128106": { + "content": "<|reserved_special_token_98|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128107": { + "content": "<|reserved_special_token_99|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128108": { + "content": "<|reserved_special_token_100|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128109": { + "content": "<|reserved_special_token_101|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128110": { + "content": "<|reserved_special_token_102|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128111": { + "content": "<|reserved_special_token_103|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128112": { + "content": "<|reserved_special_token_104|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128113": { + "content": "<|reserved_special_token_105|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128114": { + "content": "<|reserved_special_token_106|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128115": { + "content": "<|reserved_special_token_107|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128116": { + "content": "<|reserved_special_token_108|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128117": { + "content": "<|reserved_special_token_109|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128118": { + "content": "<|reserved_special_token_110|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128119": { + "content": "<|reserved_special_token_111|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128120": { + "content": "<|reserved_special_token_112|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128121": { + "content": "<|reserved_special_token_113|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128122": { + "content": "<|reserved_special_token_114|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128123": { + "content": "<|reserved_special_token_115|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128124": { + "content": "<|reserved_special_token_116|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128125": { + "content": "<|reserved_special_token_117|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128126": { + "content": "<|reserved_special_token_118|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128127": { + "content": "<|reserved_special_token_119|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128128": { + "content": "<|reserved_special_token_120|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128129": { + "content": "<|reserved_special_token_121|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128130": { + "content": "<|reserved_special_token_122|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128131": { + "content": "<|reserved_special_token_123|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128132": { + "content": "<|reserved_special_token_124|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128133": { + "content": "<|reserved_special_token_125|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128134": { + "content": "<|reserved_special_token_126|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128135": { + "content": "<|reserved_special_token_127|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128136": { + "content": "<|reserved_special_token_128|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128137": { + "content": "<|reserved_special_token_129|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128138": { + "content": "<|reserved_special_token_130|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128139": { + "content": "<|reserved_special_token_131|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128140": { + "content": "<|reserved_special_token_132|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128141": { + "content": "<|reserved_special_token_133|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128142": { + "content": "<|reserved_special_token_134|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128143": { + "content": "<|reserved_special_token_135|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128144": { + "content": "<|reserved_special_token_136|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128145": { + "content": "<|reserved_special_token_137|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128146": { + "content": "<|reserved_special_token_138|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128147": { + "content": "<|reserved_special_token_139|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128148": { + "content": "<|reserved_special_token_140|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128149": { + "content": "<|reserved_special_token_141|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128150": { + "content": "<|reserved_special_token_142|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128151": { + "content": "<|reserved_special_token_143|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128152": { + "content": "<|reserved_special_token_144|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128153": { + "content": "<|reserved_special_token_145|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128154": { + "content": "<|reserved_special_token_146|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128155": { + "content": "<|reserved_special_token_147|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128156": { + "content": "<|reserved_special_token_148|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128157": { + "content": "<|reserved_special_token_149|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128158": { + "content": "<|reserved_special_token_150|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128159": { + "content": "<|reserved_special_token_151|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128160": { + "content": "<|reserved_special_token_152|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128161": { + "content": "<|reserved_special_token_153|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128162": { + "content": "<|reserved_special_token_154|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128163": { + "content": "<|reserved_special_token_155|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128164": { + "content": "<|reserved_special_token_156|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128165": { + "content": "<|reserved_special_token_157|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128166": { + "content": "<|reserved_special_token_158|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128167": { + "content": "<|reserved_special_token_159|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128168": { + "content": "<|reserved_special_token_160|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128169": { + "content": "<|reserved_special_token_161|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128170": { + "content": "<|reserved_special_token_162|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128171": { + "content": "<|reserved_special_token_163|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128172": { + "content": "<|reserved_special_token_164|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128173": { + "content": "<|reserved_special_token_165|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128174": { + "content": "<|reserved_special_token_166|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128175": { + "content": "<|reserved_special_token_167|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128176": { + "content": "<|reserved_special_token_168|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128177": { + "content": "<|reserved_special_token_169|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128178": { + "content": "<|reserved_special_token_170|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128179": { + "content": "<|reserved_special_token_171|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128180": { + "content": "<|reserved_special_token_172|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128181": { + "content": "<|reserved_special_token_173|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128182": { + "content": "<|reserved_special_token_174|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128183": { + "content": "<|reserved_special_token_175|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128184": { + "content": "<|reserved_special_token_176|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128185": { + "content": "<|reserved_special_token_177|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128186": { + "content": "<|reserved_special_token_178|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128187": { + "content": "<|reserved_special_token_179|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128188": { + "content": "<|reserved_special_token_180|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128189": { + "content": "<|reserved_special_token_181|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128190": { + "content": "<|reserved_special_token_182|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128191": { + "content": "<|reserved_special_token_183|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128192": { + "content": "<|reserved_special_token_184|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128193": { + "content": "<|reserved_special_token_185|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128194": { + "content": "<|reserved_special_token_186|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128195": { + "content": "<|reserved_special_token_187|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128196": { + "content": "<|reserved_special_token_188|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128197": { + "content": "<|reserved_special_token_189|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128198": { + "content": "<|reserved_special_token_190|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128199": { + "content": "<|reserved_special_token_191|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128200": { + "content": "<|reserved_special_token_192|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128201": { + "content": "<|reserved_special_token_193|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128202": { + "content": "<|reserved_special_token_194|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128203": { + "content": "<|reserved_special_token_195|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128204": { + "content": "<|reserved_special_token_196|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128205": { + "content": "<|reserved_special_token_197|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128206": { + "content": "<|reserved_special_token_198|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128207": { + "content": "<|reserved_special_token_199|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128208": { + "content": "<|reserved_special_token_200|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128209": { + "content": "<|reserved_special_token_201|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128210": { + "content": "<|reserved_special_token_202|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128211": { + "content": "<|reserved_special_token_203|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128212": { + "content": "<|reserved_special_token_204|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128213": { + "content": "<|reserved_special_token_205|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128214": { + "content": "<|reserved_special_token_206|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128215": { + "content": "<|reserved_special_token_207|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128216": { + "content": "<|reserved_special_token_208|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128217": { + "content": "<|reserved_special_token_209|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128218": { + "content": "<|reserved_special_token_210|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128219": { + "content": "<|reserved_special_token_211|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128220": { + "content": "<|reserved_special_token_212|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128221": { + "content": "<|reserved_special_token_213|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128222": { + "content": "<|reserved_special_token_214|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128223": { + "content": "<|reserved_special_token_215|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128224": { + "content": "<|reserved_special_token_216|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128225": { + "content": "<|reserved_special_token_217|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128226": { + "content": "<|reserved_special_token_218|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128227": { + "content": "<|reserved_special_token_219|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128228": { + "content": "<|reserved_special_token_220|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128229": { + "content": "<|reserved_special_token_221|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128230": { + "content": "<|reserved_special_token_222|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128231": { + "content": "<|reserved_special_token_223|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128232": { + "content": "<|reserved_special_token_224|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128233": { + "content": "<|reserved_special_token_225|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128234": { + "content": "<|reserved_special_token_226|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128235": { + "content": "<|reserved_special_token_227|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128236": { + "content": "<|reserved_special_token_228|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128237": { + "content": "<|reserved_special_token_229|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128238": { + "content": "<|reserved_special_token_230|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128239": { + "content": "<|reserved_special_token_231|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128240": { + "content": "<|reserved_special_token_232|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128241": { + "content": "<|reserved_special_token_233|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128242": { + "content": "<|reserved_special_token_234|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128243": { + "content": "<|reserved_special_token_235|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128244": { + "content": "<|reserved_special_token_236|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128245": { + "content": "<|reserved_special_token_237|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128246": { + "content": "<|reserved_special_token_238|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128247": { + "content": "<|reserved_special_token_239|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128248": { + "content": "<|reserved_special_token_240|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128249": { + "content": "<|reserved_special_token_241|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128250": { + "content": "<|reserved_special_token_242|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128251": { + "content": "<|reserved_special_token_243|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128252": { + "content": "<|reserved_special_token_244|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128253": { + "content": "<|reserved_special_token_245|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128254": { + "content": "<|reserved_special_token_246|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128255": { + "content": "<|reserved_special_token_247|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128256": { + "content": "[PAD]", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + } + }, + "bos_token": "<|begin_of_text|>", + "clean_up_tokenization_spaces": true, + "eos_token": "<|end_of_text|>", + "model_input_names": [ + "input_ids", + "attention_mask" + ], + "model_max_length": 131072, + "pad_token": "[PAD]", + "tokenizer_class": "PreTrainedTokenizerFast" +} diff --git a/babylm_shuffle_nondeterministic_10M_seed0/runs/checkpoint-2300/trainer_state.json b/babylm_shuffle_nondeterministic_10M_seed0/runs/checkpoint-2300/trainer_state.json new file mode 100644 index 0000000000000000000000000000000000000000..0694da1e840009210f9efbfe9fb4ce59b0952ab6 --- /dev/null +++ b/babylm_shuffle_nondeterministic_10M_seed0/runs/checkpoint-2300/trainer_state.json @@ -0,0 +1,17973 @@ +{ + "best_metric": null, + "best_model_checkpoint": null, + "epoch": 2.5205479452054793, + "eval_steps": 10, + "global_step": 2300, + "is_hyper_param_search": false, + "is_local_process_zero": true, + "is_world_process_zero": true, + "log_history": [ + { + "epoch": 0.001095890410958904, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.0998, + "step": 1 + }, + { + "epoch": 0.002191780821917808, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.0858, + "step": 2 + }, + { + "epoch": 0.0032876712328767125, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.0672, + "step": 3 + }, + { + "epoch": 0.004383561643835616, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.0802, + "step": 4 + }, + { + "epoch": 0.005479452054794521, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.0636, + "step": 5 + }, + { + "epoch": 0.006575342465753425, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.1672, + "step": 6 + }, + { + "epoch": 0.007671232876712329, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.0961, + "step": 7 + }, + { + "epoch": 0.008767123287671232, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.1258, + "step": 8 + }, + { + "epoch": 0.009863013698630137, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.0953, + "step": 9 + }, + { + "epoch": 0.010958904109589041, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.1404, + "step": 10 + }, + { + "epoch": 0.010958904109589041, + "eval_loss": 3.12593412399292, + "eval_runtime": 42.732, + "eval_samples_per_second": 23.402, + "eval_steps_per_second": 0.983, + "step": 10 + }, + { + "epoch": 0.012054794520547946, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.1187, + "step": 11 + }, + { + "epoch": 0.01315068493150685, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.0593, + "step": 12 + }, + { + "epoch": 0.014246575342465753, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.111, + "step": 13 + }, + { + "epoch": 0.015342465753424657, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.0974, + "step": 14 + }, + { + "epoch": 0.01643835616438356, + "grad_norm": 5.991453170776367, + "learning_rate": 1.8248175182481753e-08, + "loss": 3.135, + "step": 15 + }, + { + "epoch": 0.017534246575342465, + "grad_norm": 6.274210453033447, + "learning_rate": 3.6496350364963505e-08, + "loss": 3.083, + "step": 16 + }, + { + "epoch": 0.01863013698630137, + "grad_norm": 6.644786834716797, + "learning_rate": 5.4744525547445264e-08, + "loss": 3.0856, + "step": 17 + }, + { + "epoch": 0.019726027397260273, + "grad_norm": 6.769156455993652, + "learning_rate": 7.299270072992701e-08, + "loss": 3.1082, + "step": 18 + }, + { + "epoch": 0.020821917808219178, + "grad_norm": 6.771246910095215, + "learning_rate": 9.124087591240876e-08, + "loss": 3.0753, + "step": 19 + }, + { + "epoch": 0.021917808219178082, + "grad_norm": 6.468441486358643, + "learning_rate": 1.0948905109489053e-07, + "loss": 3.1478, + "step": 20 + }, + { + "epoch": 0.021917808219178082, + "eval_loss": 3.1245107650756836, + "eval_runtime": 43.6251, + "eval_samples_per_second": 22.923, + "eval_steps_per_second": 0.963, + "step": 20 + }, + { + "epoch": 0.023013698630136987, + "grad_norm": 6.83078670501709, + "learning_rate": 1.2773722627737227e-07, + "loss": 3.1213, + "step": 21 + }, + { + "epoch": 0.02410958904109589, + "grad_norm": 6.221287250518799, + "learning_rate": 1.4598540145985402e-07, + "loss": 3.0549, + "step": 22 + }, + { + "epoch": 0.025205479452054796, + "grad_norm": 5.564266204833984, + "learning_rate": 1.642335766423358e-07, + "loss": 3.0769, + "step": 23 + }, + { + "epoch": 0.0263013698630137, + "grad_norm": 5.735209941864014, + "learning_rate": 1.8248175182481753e-07, + "loss": 3.1124, + "step": 24 + }, + { + "epoch": 0.0273972602739726, + "grad_norm": 5.577925682067871, + "learning_rate": 2.0072992700729928e-07, + "loss": 3.1191, + "step": 25 + }, + { + "epoch": 0.028493150684931506, + "grad_norm": 4.814139366149902, + "learning_rate": 2.1897810218978106e-07, + "loss": 3.0947, + "step": 26 + }, + { + "epoch": 0.02958904109589041, + "grad_norm": 4.69775390625, + "learning_rate": 2.3722627737226278e-07, + "loss": 3.0997, + "step": 27 + }, + { + "epoch": 0.030684931506849315, + "grad_norm": 4.811586380004883, + "learning_rate": 2.5547445255474454e-07, + "loss": 3.0749, + "step": 28 + }, + { + "epoch": 0.03178082191780822, + "grad_norm": 4.623377323150635, + "learning_rate": 2.737226277372263e-07, + "loss": 3.0619, + "step": 29 + }, + { + "epoch": 0.03287671232876712, + "grad_norm": 4.721253395080566, + "learning_rate": 2.9197080291970804e-07, + "loss": 3.0371, + "step": 30 + }, + { + "epoch": 0.03287671232876712, + "eval_loss": 3.082857847213745, + "eval_runtime": 44.127, + "eval_samples_per_second": 22.662, + "eval_steps_per_second": 0.952, + "step": 30 + }, + { + "epoch": 0.03397260273972603, + "grad_norm": 3.968630790710449, + "learning_rate": 3.102189781021898e-07, + "loss": 3.0535, + "step": 31 + }, + { + "epoch": 0.03506849315068493, + "grad_norm": 4.4373779296875, + "learning_rate": 3.284671532846716e-07, + "loss": 3.0476, + "step": 32 + }, + { + "epoch": 0.03616438356164384, + "grad_norm": 4.479320049285889, + "learning_rate": 3.467153284671533e-07, + "loss": 3.0215, + "step": 33 + }, + { + "epoch": 0.03726027397260274, + "grad_norm": 4.122231960296631, + "learning_rate": 3.6496350364963505e-07, + "loss": 3.0106, + "step": 34 + }, + { + "epoch": 0.038356164383561646, + "grad_norm": 4.153995513916016, + "learning_rate": 3.8321167883211686e-07, + "loss": 3.0318, + "step": 35 + }, + { + "epoch": 0.03945205479452055, + "grad_norm": 3.6949076652526855, + "learning_rate": 4.0145985401459856e-07, + "loss": 3.046, + "step": 36 + }, + { + "epoch": 0.040547945205479455, + "grad_norm": 3.4610488414764404, + "learning_rate": 4.197080291970803e-07, + "loss": 3.002, + "step": 37 + }, + { + "epoch": 0.041643835616438356, + "grad_norm": 3.9523138999938965, + "learning_rate": 4.379562043795621e-07, + "loss": 2.9665, + "step": 38 + }, + { + "epoch": 0.042739726027397264, + "grad_norm": 3.68516206741333, + "learning_rate": 4.562043795620438e-07, + "loss": 3.0411, + "step": 39 + }, + { + "epoch": 0.043835616438356165, + "grad_norm": 3.692396402359009, + "learning_rate": 4.7445255474452557e-07, + "loss": 2.9666, + "step": 40 + }, + { + "epoch": 0.043835616438356165, + "eval_loss": 3.015512704849243, + "eval_runtime": 48.2438, + "eval_samples_per_second": 20.728, + "eval_steps_per_second": 0.871, + "step": 40 + }, + { + "epoch": 0.044931506849315066, + "grad_norm": 3.3765921592712402, + "learning_rate": 4.927007299270073e-07, + "loss": 2.9831, + "step": 41 + }, + { + "epoch": 0.046027397260273974, + "grad_norm": 2.9631125926971436, + "learning_rate": 5.109489051094891e-07, + "loss": 3.0126, + "step": 42 + }, + { + "epoch": 0.047123287671232875, + "grad_norm": 2.969730854034424, + "learning_rate": 5.291970802919708e-07, + "loss": 2.9803, + "step": 43 + }, + { + "epoch": 0.04821917808219178, + "grad_norm": 3.185237407684326, + "learning_rate": 5.474452554744526e-07, + "loss": 2.9888, + "step": 44 + }, + { + "epoch": 0.049315068493150684, + "grad_norm": 3.055579423904419, + "learning_rate": 5.656934306569343e-07, + "loss": 2.9715, + "step": 45 + }, + { + "epoch": 0.05041095890410959, + "grad_norm": 3.083284616470337, + "learning_rate": 5.839416058394161e-07, + "loss": 2.9293, + "step": 46 + }, + { + "epoch": 0.05150684931506849, + "grad_norm": 2.984781503677368, + "learning_rate": 6.021897810218979e-07, + "loss": 2.9228, + "step": 47 + }, + { + "epoch": 0.0526027397260274, + "grad_norm": 2.756173849105835, + "learning_rate": 6.204379562043796e-07, + "loss": 2.9406, + "step": 48 + }, + { + "epoch": 0.0536986301369863, + "grad_norm": 2.911132574081421, + "learning_rate": 6.386861313868614e-07, + "loss": 2.8829, + "step": 49 + }, + { + "epoch": 0.0547945205479452, + "grad_norm": 2.592890977859497, + "learning_rate": 6.569343065693432e-07, + "loss": 2.9839, + "step": 50 + }, + { + "epoch": 0.0547945205479452, + "eval_loss": 2.9409468173980713, + "eval_runtime": 47.1045, + "eval_samples_per_second": 21.229, + "eval_steps_per_second": 0.892, + "step": 50 + }, + { + "epoch": 0.05589041095890411, + "grad_norm": 3.0112955570220947, + "learning_rate": 6.751824817518248e-07, + "loss": 2.8847, + "step": 51 + }, + { + "epoch": 0.05698630136986301, + "grad_norm": 2.3701798915863037, + "learning_rate": 6.934306569343066e-07, + "loss": 2.9109, + "step": 52 + }, + { + "epoch": 0.05808219178082192, + "grad_norm": 2.268850088119507, + "learning_rate": 7.116788321167883e-07, + "loss": 2.8483, + "step": 53 + }, + { + "epoch": 0.05917808219178082, + "grad_norm": 2.419015645980835, + "learning_rate": 7.299270072992701e-07, + "loss": 2.8491, + "step": 54 + }, + { + "epoch": 0.06027397260273973, + "grad_norm": 2.4582653045654297, + "learning_rate": 7.48175182481752e-07, + "loss": 2.8622, + "step": 55 + }, + { + "epoch": 0.06136986301369863, + "grad_norm": 2.3471553325653076, + "learning_rate": 7.664233576642337e-07, + "loss": 2.8738, + "step": 56 + }, + { + "epoch": 0.06246575342465754, + "grad_norm": 2.479177236557007, + "learning_rate": 7.846715328467154e-07, + "loss": 2.8843, + "step": 57 + }, + { + "epoch": 0.06356164383561644, + "grad_norm": 2.1676292419433594, + "learning_rate": 8.029197080291971e-07, + "loss": 2.8674, + "step": 58 + }, + { + "epoch": 0.06465753424657535, + "grad_norm": 1.9862910509109497, + "learning_rate": 8.211678832116789e-07, + "loss": 2.8197, + "step": 59 + }, + { + "epoch": 0.06575342465753424, + "grad_norm": 2.211313486099243, + "learning_rate": 8.394160583941606e-07, + "loss": 2.7961, + "step": 60 + }, + { + "epoch": 0.06575342465753424, + "eval_loss": 2.880824327468872, + "eval_runtime": 51.3235, + "eval_samples_per_second": 19.484, + "eval_steps_per_second": 0.818, + "step": 60 + }, + { + "epoch": 0.06684931506849315, + "grad_norm": 2.3497281074523926, + "learning_rate": 8.576642335766425e-07, + "loss": 2.8233, + "step": 61 + }, + { + "epoch": 0.06794520547945206, + "grad_norm": 2.158515214920044, + "learning_rate": 8.759124087591242e-07, + "loss": 2.8136, + "step": 62 + }, + { + "epoch": 0.06904109589041096, + "grad_norm": 2.0219955444335938, + "learning_rate": 8.94160583941606e-07, + "loss": 2.8231, + "step": 63 + }, + { + "epoch": 0.07013698630136986, + "grad_norm": 1.8989133834838867, + "learning_rate": 9.124087591240876e-07, + "loss": 2.8268, + "step": 64 + }, + { + "epoch": 0.07123287671232877, + "grad_norm": 2.243851661682129, + "learning_rate": 9.306569343065694e-07, + "loss": 2.7886, + "step": 65 + }, + { + "epoch": 0.07232876712328767, + "grad_norm": 2.122464418411255, + "learning_rate": 9.489051094890511e-07, + "loss": 2.8166, + "step": 66 + }, + { + "epoch": 0.07342465753424658, + "grad_norm": 1.899857997894287, + "learning_rate": 9.67153284671533e-07, + "loss": 2.7997, + "step": 67 + }, + { + "epoch": 0.07452054794520548, + "grad_norm": 2.4424173831939697, + "learning_rate": 9.854014598540146e-07, + "loss": 2.8342, + "step": 68 + }, + { + "epoch": 0.07561643835616438, + "grad_norm": 2.5677411556243896, + "learning_rate": 1.0036496350364965e-06, + "loss": 2.7786, + "step": 69 + }, + { + "epoch": 0.07671232876712329, + "grad_norm": 1.9989523887634277, + "learning_rate": 1.0218978102189781e-06, + "loss": 2.7411, + "step": 70 + }, + { + "epoch": 0.07671232876712329, + "eval_loss": 2.8185322284698486, + "eval_runtime": 50.4998, + "eval_samples_per_second": 19.802, + "eval_steps_per_second": 0.832, + "step": 70 + }, + { + "epoch": 0.07780821917808219, + "grad_norm": 1.9376521110534668, + "learning_rate": 1.04014598540146e-06, + "loss": 2.7988, + "step": 71 + }, + { + "epoch": 0.0789041095890411, + "grad_norm": 2.5629725456237793, + "learning_rate": 1.0583941605839416e-06, + "loss": 2.7079, + "step": 72 + }, + { + "epoch": 0.08, + "grad_norm": 2.1634678840637207, + "learning_rate": 1.0766423357664233e-06, + "loss": 2.7049, + "step": 73 + }, + { + "epoch": 0.08109589041095891, + "grad_norm": 2.2143008708953857, + "learning_rate": 1.0948905109489052e-06, + "loss": 2.72, + "step": 74 + }, + { + "epoch": 0.0821917808219178, + "grad_norm": 2.2701101303100586, + "learning_rate": 1.113138686131387e-06, + "loss": 2.7418, + "step": 75 + }, + { + "epoch": 0.08328767123287671, + "grad_norm": 1.716569423675537, + "learning_rate": 1.1313868613138687e-06, + "loss": 2.7558, + "step": 76 + }, + { + "epoch": 0.08438356164383562, + "grad_norm": 2.79757022857666, + "learning_rate": 1.1496350364963505e-06, + "loss": 2.6865, + "step": 77 + }, + { + "epoch": 0.08547945205479453, + "grad_norm": 2.2189838886260986, + "learning_rate": 1.1678832116788322e-06, + "loss": 2.6684, + "step": 78 + }, + { + "epoch": 0.08657534246575342, + "grad_norm": 2.0695579051971436, + "learning_rate": 1.1861313868613138e-06, + "loss": 2.6779, + "step": 79 + }, + { + "epoch": 0.08767123287671233, + "grad_norm": 2.372326374053955, + "learning_rate": 1.2043795620437959e-06, + "loss": 2.6711, + "step": 80 + }, + { + "epoch": 0.08767123287671233, + "eval_loss": 2.7537453174591064, + "eval_runtime": 47.7274, + "eval_samples_per_second": 20.952, + "eval_steps_per_second": 0.88, + "step": 80 + }, + { + "epoch": 0.08876712328767124, + "grad_norm": 2.3398947715759277, + "learning_rate": 1.2226277372262775e-06, + "loss": 2.6776, + "step": 81 + }, + { + "epoch": 0.08986301369863013, + "grad_norm": 2.6619036197662354, + "learning_rate": 1.2408759124087592e-06, + "loss": 2.7077, + "step": 82 + }, + { + "epoch": 0.09095890410958904, + "grad_norm": 1.8245115280151367, + "learning_rate": 1.259124087591241e-06, + "loss": 2.662, + "step": 83 + }, + { + "epoch": 0.09205479452054795, + "grad_norm": 2.2627148628234863, + "learning_rate": 1.2773722627737229e-06, + "loss": 2.677, + "step": 84 + }, + { + "epoch": 0.09315068493150686, + "grad_norm": 2.4983835220336914, + "learning_rate": 1.2956204379562045e-06, + "loss": 2.6601, + "step": 85 + }, + { + "epoch": 0.09424657534246575, + "grad_norm": 2.147010087966919, + "learning_rate": 1.3138686131386864e-06, + "loss": 2.6213, + "step": 86 + }, + { + "epoch": 0.09534246575342466, + "grad_norm": 2.791682481765747, + "learning_rate": 1.332116788321168e-06, + "loss": 2.6281, + "step": 87 + }, + { + "epoch": 0.09643835616438357, + "grad_norm": 2.023970365524292, + "learning_rate": 1.3503649635036497e-06, + "loss": 2.5766, + "step": 88 + }, + { + "epoch": 0.09753424657534246, + "grad_norm": 2.8143956661224365, + "learning_rate": 1.3686131386861315e-06, + "loss": 2.6086, + "step": 89 + }, + { + "epoch": 0.09863013698630137, + "grad_norm": 2.226106882095337, + "learning_rate": 1.3868613138686132e-06, + "loss": 2.5477, + "step": 90 + }, + { + "epoch": 0.09863013698630137, + "eval_loss": 2.6880786418914795, + "eval_runtime": 47.8417, + "eval_samples_per_second": 20.902, + "eval_steps_per_second": 0.878, + "step": 90 + }, + { + "epoch": 0.09972602739726028, + "grad_norm": 2.5248148441314697, + "learning_rate": 1.405109489051095e-06, + "loss": 2.6232, + "step": 91 + }, + { + "epoch": 0.10082191780821918, + "grad_norm": 2.421330690383911, + "learning_rate": 1.4233576642335767e-06, + "loss": 2.599, + "step": 92 + }, + { + "epoch": 0.10191780821917808, + "grad_norm": 2.5562429428100586, + "learning_rate": 1.4416058394160583e-06, + "loss": 2.6197, + "step": 93 + }, + { + "epoch": 0.10301369863013699, + "grad_norm": 2.856201171875, + "learning_rate": 1.4598540145985402e-06, + "loss": 2.5825, + "step": 94 + }, + { + "epoch": 0.10410958904109589, + "grad_norm": 2.6576807498931885, + "learning_rate": 1.478102189781022e-06, + "loss": 2.5496, + "step": 95 + }, + { + "epoch": 0.1052054794520548, + "grad_norm": 3.1408190727233887, + "learning_rate": 1.496350364963504e-06, + "loss": 2.6142, + "step": 96 + }, + { + "epoch": 0.1063013698630137, + "grad_norm": 2.3270952701568604, + "learning_rate": 1.5145985401459856e-06, + "loss": 2.584, + "step": 97 + }, + { + "epoch": 0.1073972602739726, + "grad_norm": 2.3976614475250244, + "learning_rate": 1.5328467153284674e-06, + "loss": 2.5941, + "step": 98 + }, + { + "epoch": 0.10849315068493151, + "grad_norm": 2.783485174179077, + "learning_rate": 1.551094890510949e-06, + "loss": 2.5626, + "step": 99 + }, + { + "epoch": 0.1095890410958904, + "grad_norm": 2.2522788047790527, + "learning_rate": 1.5693430656934307e-06, + "loss": 2.517, + "step": 100 + }, + { + "epoch": 0.1095890410958904, + "eval_loss": 2.637315034866333, + "eval_runtime": 48.2107, + "eval_samples_per_second": 20.742, + "eval_steps_per_second": 0.871, + "step": 100 + }, + { + "epoch": 0.11068493150684931, + "grad_norm": 2.917799949645996, + "learning_rate": 1.5875912408759126e-06, + "loss": 2.5159, + "step": 101 + }, + { + "epoch": 0.11178082191780822, + "grad_norm": 2.955815076828003, + "learning_rate": 1.6058394160583942e-06, + "loss": 2.563, + "step": 102 + }, + { + "epoch": 0.11287671232876713, + "grad_norm": 2.8096001148223877, + "learning_rate": 1.624087591240876e-06, + "loss": 2.5995, + "step": 103 + }, + { + "epoch": 0.11397260273972602, + "grad_norm": 2.4967427253723145, + "learning_rate": 1.6423357664233577e-06, + "loss": 2.5336, + "step": 104 + }, + { + "epoch": 0.11506849315068493, + "grad_norm": 2.740717887878418, + "learning_rate": 1.6605839416058394e-06, + "loss": 2.4778, + "step": 105 + }, + { + "epoch": 0.11616438356164384, + "grad_norm": 2.5064632892608643, + "learning_rate": 1.6788321167883212e-06, + "loss": 2.5214, + "step": 106 + }, + { + "epoch": 0.11726027397260275, + "grad_norm": 2.2923007011413574, + "learning_rate": 1.6970802919708029e-06, + "loss": 2.5315, + "step": 107 + }, + { + "epoch": 0.11835616438356164, + "grad_norm": 3.339569568634033, + "learning_rate": 1.715328467153285e-06, + "loss": 2.4661, + "step": 108 + }, + { + "epoch": 0.11945205479452055, + "grad_norm": 2.1650891304016113, + "learning_rate": 1.7335766423357666e-06, + "loss": 2.5072, + "step": 109 + }, + { + "epoch": 0.12054794520547946, + "grad_norm": 3.1216914653778076, + "learning_rate": 1.7518248175182485e-06, + "loss": 2.5164, + "step": 110 + }, + { + "epoch": 0.12054794520547946, + "eval_loss": 2.594250202178955, + "eval_runtime": 41.8939, + "eval_samples_per_second": 23.87, + "eval_steps_per_second": 1.003, + "step": 110 + }, + { + "epoch": 0.12164383561643835, + "grad_norm": 2.191063404083252, + "learning_rate": 1.77007299270073e-06, + "loss": 2.5422, + "step": 111 + }, + { + "epoch": 0.12273972602739726, + "grad_norm": 2.922083854675293, + "learning_rate": 1.788321167883212e-06, + "loss": 2.5258, + "step": 112 + }, + { + "epoch": 0.12383561643835617, + "grad_norm": 2.518002986907959, + "learning_rate": 1.8065693430656936e-06, + "loss": 2.5588, + "step": 113 + }, + { + "epoch": 0.12493150684931507, + "grad_norm": 2.515194892883301, + "learning_rate": 1.8248175182481753e-06, + "loss": 2.5079, + "step": 114 + }, + { + "epoch": 0.12602739726027398, + "grad_norm": 2.5530874729156494, + "learning_rate": 1.8430656934306571e-06, + "loss": 2.4615, + "step": 115 + }, + { + "epoch": 0.12712328767123288, + "grad_norm": 2.3550198078155518, + "learning_rate": 1.8613138686131388e-06, + "loss": 2.4825, + "step": 116 + }, + { + "epoch": 0.12821917808219177, + "grad_norm": 2.295145034790039, + "learning_rate": 1.8795620437956206e-06, + "loss": 2.4749, + "step": 117 + }, + { + "epoch": 0.1293150684931507, + "grad_norm": 2.0158982276916504, + "learning_rate": 1.8978102189781023e-06, + "loss": 2.4756, + "step": 118 + }, + { + "epoch": 0.1304109589041096, + "grad_norm": 2.213057518005371, + "learning_rate": 1.916058394160584e-06, + "loss": 2.4744, + "step": 119 + }, + { + "epoch": 0.13150684931506848, + "grad_norm": 2.186222553253174, + "learning_rate": 1.934306569343066e-06, + "loss": 2.4957, + "step": 120 + }, + { + "epoch": 0.13150684931506848, + "eval_loss": 2.563931465148926, + "eval_runtime": 44.8714, + "eval_samples_per_second": 22.286, + "eval_steps_per_second": 0.936, + "step": 120 + }, + { + "epoch": 0.1326027397260274, + "grad_norm": 2.3143310546875, + "learning_rate": 1.952554744525548e-06, + "loss": 2.4, + "step": 121 + }, + { + "epoch": 0.1336986301369863, + "grad_norm": 2.0223965644836426, + "learning_rate": 1.9708029197080293e-06, + "loss": 2.4453, + "step": 122 + }, + { + "epoch": 0.13479452054794522, + "grad_norm": 2.2025411128997803, + "learning_rate": 1.989051094890511e-06, + "loss": 2.4463, + "step": 123 + }, + { + "epoch": 0.1358904109589041, + "grad_norm": 2.707930564880371, + "learning_rate": 2.007299270072993e-06, + "loss": 2.4589, + "step": 124 + }, + { + "epoch": 0.136986301369863, + "grad_norm": 2.146662473678589, + "learning_rate": 2.025547445255475e-06, + "loss": 2.5023, + "step": 125 + }, + { + "epoch": 0.13808219178082193, + "grad_norm": 2.520660161972046, + "learning_rate": 2.0437956204379563e-06, + "loss": 2.4464, + "step": 126 + }, + { + "epoch": 0.13917808219178082, + "grad_norm": 2.50632381439209, + "learning_rate": 2.062043795620438e-06, + "loss": 2.4556, + "step": 127 + }, + { + "epoch": 0.14027397260273972, + "grad_norm": 2.2919397354125977, + "learning_rate": 2.08029197080292e-06, + "loss": 2.4479, + "step": 128 + }, + { + "epoch": 0.14136986301369864, + "grad_norm": 2.521475315093994, + "learning_rate": 2.0985401459854014e-06, + "loss": 2.4982, + "step": 129 + }, + { + "epoch": 0.14246575342465753, + "grad_norm": 2.3812668323516846, + "learning_rate": 2.1167883211678833e-06, + "loss": 2.4163, + "step": 130 + }, + { + "epoch": 0.14246575342465753, + "eval_loss": 2.5390918254852295, + "eval_runtime": 43.8746, + "eval_samples_per_second": 22.792, + "eval_steps_per_second": 0.957, + "step": 130 + }, + { + "epoch": 0.14356164383561643, + "grad_norm": 2.302286148071289, + "learning_rate": 2.135036496350365e-06, + "loss": 2.4588, + "step": 131 + }, + { + "epoch": 0.14465753424657535, + "grad_norm": 2.4222044944763184, + "learning_rate": 2.1532846715328466e-06, + "loss": 2.492, + "step": 132 + }, + { + "epoch": 0.14575342465753424, + "grad_norm": 2.605835199356079, + "learning_rate": 2.171532846715329e-06, + "loss": 2.4055, + "step": 133 + }, + { + "epoch": 0.14684931506849316, + "grad_norm": 3.6261861324310303, + "learning_rate": 2.1897810218978103e-06, + "loss": 2.4221, + "step": 134 + }, + { + "epoch": 0.14794520547945206, + "grad_norm": 2.7647221088409424, + "learning_rate": 2.208029197080292e-06, + "loss": 2.4549, + "step": 135 + }, + { + "epoch": 0.14904109589041095, + "grad_norm": 3.2977371215820312, + "learning_rate": 2.226277372262774e-06, + "loss": 2.4331, + "step": 136 + }, + { + "epoch": 0.15013698630136987, + "grad_norm": 3.147094488143921, + "learning_rate": 2.244525547445256e-06, + "loss": 2.4268, + "step": 137 + }, + { + "epoch": 0.15123287671232877, + "grad_norm": 2.6363413333892822, + "learning_rate": 2.2627737226277373e-06, + "loss": 2.4524, + "step": 138 + }, + { + "epoch": 0.15232876712328766, + "grad_norm": 2.9288904666900635, + "learning_rate": 2.281021897810219e-06, + "loss": 2.4534, + "step": 139 + }, + { + "epoch": 0.15342465753424658, + "grad_norm": 2.5604870319366455, + "learning_rate": 2.299270072992701e-06, + "loss": 2.4138, + "step": 140 + }, + { + "epoch": 0.15342465753424658, + "eval_loss": 2.524790048599243, + "eval_runtime": 43.8851, + "eval_samples_per_second": 22.787, + "eval_steps_per_second": 0.957, + "step": 140 + }, + { + "epoch": 0.15452054794520548, + "grad_norm": 3.2746663093566895, + "learning_rate": 2.3175182481751825e-06, + "loss": 2.4094, + "step": 141 + }, + { + "epoch": 0.15561643835616437, + "grad_norm": 2.691483497619629, + "learning_rate": 2.3357664233576643e-06, + "loss": 2.4265, + "step": 142 + }, + { + "epoch": 0.1567123287671233, + "grad_norm": 2.5850117206573486, + "learning_rate": 2.354014598540146e-06, + "loss": 2.3922, + "step": 143 + }, + { + "epoch": 0.1578082191780822, + "grad_norm": 2.68204927444458, + "learning_rate": 2.3722627737226276e-06, + "loss": 2.4262, + "step": 144 + }, + { + "epoch": 0.1589041095890411, + "grad_norm": 2.952584981918335, + "learning_rate": 2.39051094890511e-06, + "loss": 2.4267, + "step": 145 + }, + { + "epoch": 0.16, + "grad_norm": 3.2354788780212402, + "learning_rate": 2.4087591240875918e-06, + "loss": 2.3836, + "step": 146 + }, + { + "epoch": 0.1610958904109589, + "grad_norm": 2.634521007537842, + "learning_rate": 2.427007299270073e-06, + "loss": 2.3953, + "step": 147 + }, + { + "epoch": 0.16219178082191782, + "grad_norm": 3.1115682125091553, + "learning_rate": 2.445255474452555e-06, + "loss": 2.409, + "step": 148 + }, + { + "epoch": 0.1632876712328767, + "grad_norm": 2.6824135780334473, + "learning_rate": 2.463503649635037e-06, + "loss": 2.4459, + "step": 149 + }, + { + "epoch": 0.1643835616438356, + "grad_norm": 2.617614507675171, + "learning_rate": 2.4817518248175183e-06, + "loss": 2.366, + "step": 150 + }, + { + "epoch": 0.1643835616438356, + "eval_loss": 2.5088160037994385, + "eval_runtime": 44.0888, + "eval_samples_per_second": 22.682, + "eval_steps_per_second": 0.953, + "step": 150 + }, + { + "epoch": 0.16547945205479453, + "grad_norm": 3.976992130279541, + "learning_rate": 2.5e-06, + "loss": 2.3517, + "step": 151 + }, + { + "epoch": 0.16657534246575342, + "grad_norm": 3.1433017253875732, + "learning_rate": 2.518248175182482e-06, + "loss": 2.4489, + "step": 152 + }, + { + "epoch": 0.16767123287671232, + "grad_norm": 4.100688934326172, + "learning_rate": 2.5364963503649635e-06, + "loss": 2.3683, + "step": 153 + }, + { + "epoch": 0.16876712328767124, + "grad_norm": 4.316102504730225, + "learning_rate": 2.5547445255474458e-06, + "loss": 2.3663, + "step": 154 + }, + { + "epoch": 0.16986301369863013, + "grad_norm": 2.3103132247924805, + "learning_rate": 2.5729927007299272e-06, + "loss": 2.4009, + "step": 155 + }, + { + "epoch": 0.17095890410958905, + "grad_norm": 3.576810836791992, + "learning_rate": 2.591240875912409e-06, + "loss": 2.3983, + "step": 156 + }, + { + "epoch": 0.17205479452054795, + "grad_norm": 2.337080955505371, + "learning_rate": 2.6094890510948905e-06, + "loss": 2.3849, + "step": 157 + }, + { + "epoch": 0.17315068493150684, + "grad_norm": 3.2523117065429688, + "learning_rate": 2.627737226277373e-06, + "loss": 2.4171, + "step": 158 + }, + { + "epoch": 0.17424657534246576, + "grad_norm": 2.529175281524658, + "learning_rate": 2.6459854014598542e-06, + "loss": 2.4303, + "step": 159 + }, + { + "epoch": 0.17534246575342466, + "grad_norm": 3.052212953567505, + "learning_rate": 2.664233576642336e-06, + "loss": 2.3895, + "step": 160 + }, + { + "epoch": 0.17534246575342466, + "eval_loss": 2.492485761642456, + "eval_runtime": 45.6738, + "eval_samples_per_second": 21.894, + "eval_steps_per_second": 0.92, + "step": 160 + }, + { + "epoch": 0.17643835616438355, + "grad_norm": 2.1544973850250244, + "learning_rate": 2.6824817518248175e-06, + "loss": 2.3649, + "step": 161 + }, + { + "epoch": 0.17753424657534247, + "grad_norm": 2.7351951599121094, + "learning_rate": 2.7007299270072994e-06, + "loss": 2.3908, + "step": 162 + }, + { + "epoch": 0.17863013698630137, + "grad_norm": 2.8431951999664307, + "learning_rate": 2.718978102189781e-06, + "loss": 2.374, + "step": 163 + }, + { + "epoch": 0.17972602739726026, + "grad_norm": 3.083190441131592, + "learning_rate": 2.737226277372263e-06, + "loss": 2.4193, + "step": 164 + }, + { + "epoch": 0.18082191780821918, + "grad_norm": 2.6523995399475098, + "learning_rate": 2.755474452554745e-06, + "loss": 2.3969, + "step": 165 + }, + { + "epoch": 0.18191780821917808, + "grad_norm": 2.626047372817993, + "learning_rate": 2.7737226277372264e-06, + "loss": 2.3742, + "step": 166 + }, + { + "epoch": 0.18301369863013697, + "grad_norm": 3.300917863845825, + "learning_rate": 2.7919708029197082e-06, + "loss": 2.4546, + "step": 167 + }, + { + "epoch": 0.1841095890410959, + "grad_norm": 2.797304391860962, + "learning_rate": 2.81021897810219e-06, + "loss": 2.4161, + "step": 168 + }, + { + "epoch": 0.1852054794520548, + "grad_norm": 2.777117967605591, + "learning_rate": 2.828467153284672e-06, + "loss": 2.3572, + "step": 169 + }, + { + "epoch": 0.1863013698630137, + "grad_norm": 2.887470006942749, + "learning_rate": 2.8467153284671534e-06, + "loss": 2.3887, + "step": 170 + }, + { + "epoch": 0.1863013698630137, + "eval_loss": 2.4854257106781006, + "eval_runtime": 44.5644, + "eval_samples_per_second": 22.439, + "eval_steps_per_second": 0.942, + "step": 170 + }, + { + "epoch": 0.1873972602739726, + "grad_norm": 3.0706076622009277, + "learning_rate": 2.8649635036496353e-06, + "loss": 2.3162, + "step": 171 + }, + { + "epoch": 0.1884931506849315, + "grad_norm": 3.6129744052886963, + "learning_rate": 2.8832116788321167e-06, + "loss": 2.3685, + "step": 172 + }, + { + "epoch": 0.18958904109589042, + "grad_norm": 2.4936370849609375, + "learning_rate": 2.901459854014599e-06, + "loss": 2.3297, + "step": 173 + }, + { + "epoch": 0.19068493150684931, + "grad_norm": 2.1041860580444336, + "learning_rate": 2.9197080291970804e-06, + "loss": 2.4085, + "step": 174 + }, + { + "epoch": 0.1917808219178082, + "grad_norm": 2.2066831588745117, + "learning_rate": 2.9379562043795623e-06, + "loss": 2.3905, + "step": 175 + }, + { + "epoch": 0.19287671232876713, + "grad_norm": 2.444652795791626, + "learning_rate": 2.956204379562044e-06, + "loss": 2.378, + "step": 176 + }, + { + "epoch": 0.19397260273972602, + "grad_norm": 2.501939058303833, + "learning_rate": 2.9744525547445256e-06, + "loss": 2.4136, + "step": 177 + }, + { + "epoch": 0.19506849315068492, + "grad_norm": 2.2167556285858154, + "learning_rate": 2.992700729927008e-06, + "loss": 2.385, + "step": 178 + }, + { + "epoch": 0.19616438356164384, + "grad_norm": 2.5298545360565186, + "learning_rate": 3.0109489051094893e-06, + "loss": 2.3508, + "step": 179 + }, + { + "epoch": 0.19726027397260273, + "grad_norm": 2.3774642944335938, + "learning_rate": 3.029197080291971e-06, + "loss": 2.3421, + "step": 180 + }, + { + "epoch": 0.19726027397260273, + "eval_loss": 2.476067304611206, + "eval_runtime": 44.3761, + "eval_samples_per_second": 22.535, + "eval_steps_per_second": 0.946, + "step": 180 + }, + { + "epoch": 0.19835616438356166, + "grad_norm": 2.9108712673187256, + "learning_rate": 3.0474452554744526e-06, + "loss": 2.3983, + "step": 181 + }, + { + "epoch": 0.19945205479452055, + "grad_norm": 2.5232033729553223, + "learning_rate": 3.065693430656935e-06, + "loss": 2.3946, + "step": 182 + }, + { + "epoch": 0.20054794520547944, + "grad_norm": 2.158168077468872, + "learning_rate": 3.0839416058394163e-06, + "loss": 2.3805, + "step": 183 + }, + { + "epoch": 0.20164383561643837, + "grad_norm": 2.3405354022979736, + "learning_rate": 3.102189781021898e-06, + "loss": 2.3974, + "step": 184 + }, + { + "epoch": 0.20273972602739726, + "grad_norm": 2.5308327674865723, + "learning_rate": 3.1204379562043796e-06, + "loss": 2.384, + "step": 185 + }, + { + "epoch": 0.20383561643835615, + "grad_norm": 2.6328237056732178, + "learning_rate": 3.1386861313868614e-06, + "loss": 2.3345, + "step": 186 + }, + { + "epoch": 0.20493150684931508, + "grad_norm": 3.0725512504577637, + "learning_rate": 3.156934306569343e-06, + "loss": 2.3191, + "step": 187 + }, + { + "epoch": 0.20602739726027397, + "grad_norm": 2.196218729019165, + "learning_rate": 3.175182481751825e-06, + "loss": 2.4073, + "step": 188 + }, + { + "epoch": 0.20712328767123286, + "grad_norm": 2.7722432613372803, + "learning_rate": 3.193430656934307e-06, + "loss": 2.3775, + "step": 189 + }, + { + "epoch": 0.20821917808219179, + "grad_norm": 2.4848241806030273, + "learning_rate": 3.2116788321167884e-06, + "loss": 2.3538, + "step": 190 + }, + { + "epoch": 0.20821917808219179, + "eval_loss": 2.462618112564087, + "eval_runtime": 43.9461, + "eval_samples_per_second": 22.755, + "eval_steps_per_second": 0.956, + "step": 190 + }, + { + "epoch": 0.20931506849315068, + "grad_norm": 2.4682741165161133, + "learning_rate": 3.2299270072992707e-06, + "loss": 2.3688, + "step": 191 + }, + { + "epoch": 0.2104109589041096, + "grad_norm": 2.518923282623291, + "learning_rate": 3.248175182481752e-06, + "loss": 2.3556, + "step": 192 + }, + { + "epoch": 0.2115068493150685, + "grad_norm": 2.4023561477661133, + "learning_rate": 3.266423357664234e-06, + "loss": 2.292, + "step": 193 + }, + { + "epoch": 0.2126027397260274, + "grad_norm": 2.5196943283081055, + "learning_rate": 3.2846715328467155e-06, + "loss": 2.3703, + "step": 194 + }, + { + "epoch": 0.2136986301369863, + "grad_norm": 2.762709617614746, + "learning_rate": 3.3029197080291973e-06, + "loss": 2.3578, + "step": 195 + }, + { + "epoch": 0.2147945205479452, + "grad_norm": 2.1287691593170166, + "learning_rate": 3.3211678832116788e-06, + "loss": 2.3687, + "step": 196 + }, + { + "epoch": 0.2158904109589041, + "grad_norm": 2.477341651916504, + "learning_rate": 3.339416058394161e-06, + "loss": 2.3545, + "step": 197 + }, + { + "epoch": 0.21698630136986302, + "grad_norm": 2.23933744430542, + "learning_rate": 3.3576642335766425e-06, + "loss": 2.3511, + "step": 198 + }, + { + "epoch": 0.21808219178082192, + "grad_norm": 2.2440133094787598, + "learning_rate": 3.3759124087591243e-06, + "loss": 2.3539, + "step": 199 + }, + { + "epoch": 0.2191780821917808, + "grad_norm": 2.5603177547454834, + "learning_rate": 3.3941605839416058e-06, + "loss": 2.36, + "step": 200 + }, + { + "epoch": 0.2191780821917808, + "eval_loss": 2.4507246017456055, + "eval_runtime": 43.7786, + "eval_samples_per_second": 22.842, + "eval_steps_per_second": 0.959, + "step": 200 + }, + { + "epoch": 0.22027397260273973, + "grad_norm": 2.4570510387420654, + "learning_rate": 3.412408759124088e-06, + "loss": 2.3702, + "step": 201 + }, + { + "epoch": 0.22136986301369863, + "grad_norm": 2.1975133419036865, + "learning_rate": 3.43065693430657e-06, + "loss": 2.3667, + "step": 202 + }, + { + "epoch": 0.22246575342465755, + "grad_norm": 2.710186004638672, + "learning_rate": 3.4489051094890513e-06, + "loss": 2.3159, + "step": 203 + }, + { + "epoch": 0.22356164383561644, + "grad_norm": 2.7799506187438965, + "learning_rate": 3.467153284671533e-06, + "loss": 2.3519, + "step": 204 + }, + { + "epoch": 0.22465753424657534, + "grad_norm": 2.2479400634765625, + "learning_rate": 3.4854014598540146e-06, + "loss": 2.3103, + "step": 205 + }, + { + "epoch": 0.22575342465753426, + "grad_norm": 3.1538405418395996, + "learning_rate": 3.503649635036497e-06, + "loss": 2.3506, + "step": 206 + }, + { + "epoch": 0.22684931506849315, + "grad_norm": 2.7607762813568115, + "learning_rate": 3.5218978102189783e-06, + "loss": 2.3386, + "step": 207 + }, + { + "epoch": 0.22794520547945205, + "grad_norm": 2.371105909347534, + "learning_rate": 3.54014598540146e-06, + "loss": 2.3583, + "step": 208 + }, + { + "epoch": 0.22904109589041097, + "grad_norm": 2.793933391571045, + "learning_rate": 3.5583941605839416e-06, + "loss": 2.3782, + "step": 209 + }, + { + "epoch": 0.23013698630136986, + "grad_norm": 2.372732400894165, + "learning_rate": 3.576642335766424e-06, + "loss": 2.417, + "step": 210 + }, + { + "epoch": 0.23013698630136986, + "eval_loss": 2.451474189758301, + "eval_runtime": 41.7197, + "eval_samples_per_second": 23.97, + "eval_steps_per_second": 1.007, + "step": 210 + }, + { + "epoch": 0.23123287671232876, + "grad_norm": 2.6989448070526123, + "learning_rate": 3.5948905109489054e-06, + "loss": 2.393, + "step": 211 + }, + { + "epoch": 0.23232876712328768, + "grad_norm": 2.4324285984039307, + "learning_rate": 3.6131386861313872e-06, + "loss": 2.2738, + "step": 212 + }, + { + "epoch": 0.23342465753424657, + "grad_norm": 3.5063016414642334, + "learning_rate": 3.6313868613138687e-06, + "loss": 2.2823, + "step": 213 + }, + { + "epoch": 0.2345205479452055, + "grad_norm": 3.953951358795166, + "learning_rate": 3.6496350364963505e-06, + "loss": 2.3132, + "step": 214 + }, + { + "epoch": 0.2356164383561644, + "grad_norm": 2.3483071327209473, + "learning_rate": 3.667883211678833e-06, + "loss": 2.3244, + "step": 215 + }, + { + "epoch": 0.23671232876712328, + "grad_norm": 2.884199380874634, + "learning_rate": 3.6861313868613142e-06, + "loss": 2.3067, + "step": 216 + }, + { + "epoch": 0.2378082191780822, + "grad_norm": 2.9514520168304443, + "learning_rate": 3.704379562043796e-06, + "loss": 2.304, + "step": 217 + }, + { + "epoch": 0.2389041095890411, + "grad_norm": 2.7530996799468994, + "learning_rate": 3.7226277372262775e-06, + "loss": 2.3691, + "step": 218 + }, + { + "epoch": 0.24, + "grad_norm": 2.8753647804260254, + "learning_rate": 3.7408759124087594e-06, + "loss": 2.3406, + "step": 219 + }, + { + "epoch": 0.2410958904109589, + "grad_norm": 2.670574903488159, + "learning_rate": 3.7591240875912412e-06, + "loss": 2.304, + "step": 220 + }, + { + "epoch": 0.2410958904109589, + "eval_loss": 2.4395809173583984, + "eval_runtime": 46.6723, + "eval_samples_per_second": 21.426, + "eval_steps_per_second": 0.9, + "step": 220 + }, + { + "epoch": 0.2421917808219178, + "grad_norm": 2.4152438640594482, + "learning_rate": 3.777372262773723e-06, + "loss": 2.3294, + "step": 221 + }, + { + "epoch": 0.2432876712328767, + "grad_norm": 2.8398540019989014, + "learning_rate": 3.7956204379562045e-06, + "loss": 2.3216, + "step": 222 + }, + { + "epoch": 0.24438356164383562, + "grad_norm": 2.6036765575408936, + "learning_rate": 3.8138686131386864e-06, + "loss": 2.3109, + "step": 223 + }, + { + "epoch": 0.24547945205479452, + "grad_norm": 2.7073473930358887, + "learning_rate": 3.832116788321168e-06, + "loss": 2.3457, + "step": 224 + }, + { + "epoch": 0.2465753424657534, + "grad_norm": 2.584404230117798, + "learning_rate": 3.85036496350365e-06, + "loss": 2.3159, + "step": 225 + }, + { + "epoch": 0.24767123287671233, + "grad_norm": 2.4220666885375977, + "learning_rate": 3.868613138686132e-06, + "loss": 2.2944, + "step": 226 + }, + { + "epoch": 0.24876712328767123, + "grad_norm": 2.5636494159698486, + "learning_rate": 3.886861313868613e-06, + "loss": 2.3219, + "step": 227 + }, + { + "epoch": 0.24986301369863015, + "grad_norm": 2.2990903854370117, + "learning_rate": 3.905109489051096e-06, + "loss": 2.3342, + "step": 228 + }, + { + "epoch": 0.25095890410958904, + "grad_norm": 2.50913143157959, + "learning_rate": 3.923357664233577e-06, + "loss": 2.3619, + "step": 229 + }, + { + "epoch": 0.25205479452054796, + "grad_norm": 2.1687769889831543, + "learning_rate": 3.9416058394160585e-06, + "loss": 2.3153, + "step": 230 + }, + { + "epoch": 0.25205479452054796, + "eval_loss": 2.4315240383148193, + "eval_runtime": 48.3542, + "eval_samples_per_second": 20.681, + "eval_steps_per_second": 0.869, + "step": 230 + }, + { + "epoch": 0.25315068493150683, + "grad_norm": 2.842125177383423, + "learning_rate": 3.95985401459854e-06, + "loss": 2.3644, + "step": 231 + }, + { + "epoch": 0.25424657534246575, + "grad_norm": 2.258049964904785, + "learning_rate": 3.978102189781022e-06, + "loss": 2.3727, + "step": 232 + }, + { + "epoch": 0.2553424657534247, + "grad_norm": 2.4630489349365234, + "learning_rate": 3.996350364963504e-06, + "loss": 2.267, + "step": 233 + }, + { + "epoch": 0.25643835616438354, + "grad_norm": 3.1469743251800537, + "learning_rate": 4.014598540145986e-06, + "loss": 2.3432, + "step": 234 + }, + { + "epoch": 0.25753424657534246, + "grad_norm": 2.6277477741241455, + "learning_rate": 4.032846715328467e-06, + "loss": 2.2545, + "step": 235 + }, + { + "epoch": 0.2586301369863014, + "grad_norm": 2.404141902923584, + "learning_rate": 4.05109489051095e-06, + "loss": 2.3109, + "step": 236 + }, + { + "epoch": 0.25972602739726025, + "grad_norm": 2.4545605182647705, + "learning_rate": 4.069343065693431e-06, + "loss": 2.3112, + "step": 237 + }, + { + "epoch": 0.2608219178082192, + "grad_norm": 2.237638235092163, + "learning_rate": 4.0875912408759126e-06, + "loss": 2.3, + "step": 238 + }, + { + "epoch": 0.2619178082191781, + "grad_norm": 2.192811965942383, + "learning_rate": 4.105839416058395e-06, + "loss": 2.3108, + "step": 239 + }, + { + "epoch": 0.26301369863013696, + "grad_norm": 2.5058720111846924, + "learning_rate": 4.124087591240876e-06, + "loss": 2.3635, + "step": 240 + }, + { + "epoch": 0.26301369863013696, + "eval_loss": 2.4219377040863037, + "eval_runtime": 45.3561, + "eval_samples_per_second": 22.048, + "eval_steps_per_second": 0.926, + "step": 240 + }, + { + "epoch": 0.2641095890410959, + "grad_norm": 2.0708935260772705, + "learning_rate": 4.1423357664233586e-06, + "loss": 2.3129, + "step": 241 + }, + { + "epoch": 0.2652054794520548, + "grad_norm": 2.6353116035461426, + "learning_rate": 4.16058394160584e-06, + "loss": 2.3011, + "step": 242 + }, + { + "epoch": 0.2663013698630137, + "grad_norm": 2.8392436504364014, + "learning_rate": 4.1788321167883214e-06, + "loss": 2.3643, + "step": 243 + }, + { + "epoch": 0.2673972602739726, + "grad_norm": 3.1773295402526855, + "learning_rate": 4.197080291970803e-06, + "loss": 2.2642, + "step": 244 + }, + { + "epoch": 0.2684931506849315, + "grad_norm": 2.4757354259490967, + "learning_rate": 4.215328467153285e-06, + "loss": 2.3211, + "step": 245 + }, + { + "epoch": 0.26958904109589044, + "grad_norm": 2.5799412727355957, + "learning_rate": 4.233576642335767e-06, + "loss": 2.2909, + "step": 246 + }, + { + "epoch": 0.2706849315068493, + "grad_norm": 2.368502378463745, + "learning_rate": 4.251824817518249e-06, + "loss": 2.2951, + "step": 247 + }, + { + "epoch": 0.2717808219178082, + "grad_norm": 2.114969253540039, + "learning_rate": 4.27007299270073e-06, + "loss": 2.3124, + "step": 248 + }, + { + "epoch": 0.27287671232876715, + "grad_norm": 3.056820869445801, + "learning_rate": 4.288321167883212e-06, + "loss": 2.2923, + "step": 249 + }, + { + "epoch": 0.273972602739726, + "grad_norm": 2.7752203941345215, + "learning_rate": 4.306569343065693e-06, + "loss": 2.2874, + "step": 250 + }, + { + "epoch": 0.273972602739726, + "eval_loss": 2.4229912757873535, + "eval_runtime": 52.5941, + "eval_samples_per_second": 19.014, + "eval_steps_per_second": 0.799, + "step": 250 + }, + { + "epoch": 0.27506849315068493, + "grad_norm": 3.045740842819214, + "learning_rate": 4.3248175182481755e-06, + "loss": 2.2686, + "step": 251 + }, + { + "epoch": 0.27616438356164386, + "grad_norm": 2.646402359008789, + "learning_rate": 4.343065693430658e-06, + "loss": 2.3247, + "step": 252 + }, + { + "epoch": 0.2772602739726027, + "grad_norm": 3.274994134902954, + "learning_rate": 4.361313868613139e-06, + "loss": 2.3093, + "step": 253 + }, + { + "epoch": 0.27835616438356164, + "grad_norm": 2.352246046066284, + "learning_rate": 4.379562043795621e-06, + "loss": 2.3124, + "step": 254 + }, + { + "epoch": 0.27945205479452057, + "grad_norm": 3.248717784881592, + "learning_rate": 4.397810218978102e-06, + "loss": 2.3417, + "step": 255 + }, + { + "epoch": 0.28054794520547943, + "grad_norm": 3.0919013023376465, + "learning_rate": 4.416058394160584e-06, + "loss": 2.3093, + "step": 256 + }, + { + "epoch": 0.28164383561643835, + "grad_norm": 2.1174702644348145, + "learning_rate": 4.434306569343066e-06, + "loss": 2.305, + "step": 257 + }, + { + "epoch": 0.2827397260273973, + "grad_norm": 2.913797378540039, + "learning_rate": 4.452554744525548e-06, + "loss": 2.2831, + "step": 258 + }, + { + "epoch": 0.28383561643835614, + "grad_norm": 2.693089723587036, + "learning_rate": 4.4708029197080295e-06, + "loss": 2.3206, + "step": 259 + }, + { + "epoch": 0.28493150684931506, + "grad_norm": 2.521411895751953, + "learning_rate": 4.489051094890512e-06, + "loss": 2.3196, + "step": 260 + }, + { + "epoch": 0.28493150684931506, + "eval_loss": 2.4178109169006348, + "eval_runtime": 50.6753, + "eval_samples_per_second": 19.733, + "eval_steps_per_second": 0.829, + "step": 260 + }, + { + "epoch": 0.286027397260274, + "grad_norm": 2.346346855163574, + "learning_rate": 4.507299270072993e-06, + "loss": 2.3433, + "step": 261 + }, + { + "epoch": 0.28712328767123285, + "grad_norm": 2.520368814468384, + "learning_rate": 4.525547445255475e-06, + "loss": 2.3174, + "step": 262 + }, + { + "epoch": 0.2882191780821918, + "grad_norm": 2.550347328186035, + "learning_rate": 4.543795620437956e-06, + "loss": 2.3072, + "step": 263 + }, + { + "epoch": 0.2893150684931507, + "grad_norm": 2.268662691116333, + "learning_rate": 4.562043795620438e-06, + "loss": 2.2763, + "step": 264 + }, + { + "epoch": 0.29041095890410956, + "grad_norm": 2.445798873901367, + "learning_rate": 4.580291970802921e-06, + "loss": 2.3395, + "step": 265 + }, + { + "epoch": 0.2915068493150685, + "grad_norm": 2.5391135215759277, + "learning_rate": 4.598540145985402e-06, + "loss": 2.3496, + "step": 266 + }, + { + "epoch": 0.2926027397260274, + "grad_norm": 2.3553097248077393, + "learning_rate": 4.6167883211678835e-06, + "loss": 2.3243, + "step": 267 + }, + { + "epoch": 0.2936986301369863, + "grad_norm": 2.448089599609375, + "learning_rate": 4.635036496350365e-06, + "loss": 2.2887, + "step": 268 + }, + { + "epoch": 0.2947945205479452, + "grad_norm": 2.1600170135498047, + "learning_rate": 4.653284671532847e-06, + "loss": 2.2816, + "step": 269 + }, + { + "epoch": 0.2958904109589041, + "grad_norm": 2.3783180713653564, + "learning_rate": 4.671532846715329e-06, + "loss": 2.3275, + "step": 270 + }, + { + "epoch": 0.2958904109589041, + "eval_loss": 2.413832902908325, + "eval_runtime": 48.7524, + "eval_samples_per_second": 20.512, + "eval_steps_per_second": 0.861, + "step": 270 + }, + { + "epoch": 0.29698630136986304, + "grad_norm": 2.468320369720459, + "learning_rate": 4.689781021897811e-06, + "loss": 2.3358, + "step": 271 + }, + { + "epoch": 0.2980821917808219, + "grad_norm": 2.401634454727173, + "learning_rate": 4.708029197080292e-06, + "loss": 2.2991, + "step": 272 + }, + { + "epoch": 0.2991780821917808, + "grad_norm": 2.4604203701019287, + "learning_rate": 4.726277372262774e-06, + "loss": 2.3496, + "step": 273 + }, + { + "epoch": 0.30027397260273975, + "grad_norm": 2.0338871479034424, + "learning_rate": 4.744525547445255e-06, + "loss": 2.3115, + "step": 274 + }, + { + "epoch": 0.3013698630136986, + "grad_norm": 2.5404436588287354, + "learning_rate": 4.7627737226277375e-06, + "loss": 2.2665, + "step": 275 + }, + { + "epoch": 0.30246575342465754, + "grad_norm": 2.5369458198547363, + "learning_rate": 4.78102189781022e-06, + "loss": 2.3072, + "step": 276 + }, + { + "epoch": 0.30356164383561646, + "grad_norm": 2.471522569656372, + "learning_rate": 4.799270072992701e-06, + "loss": 2.3113, + "step": 277 + }, + { + "epoch": 0.3046575342465753, + "grad_norm": 2.3890135288238525, + "learning_rate": 4.8175182481751835e-06, + "loss": 2.2838, + "step": 278 + }, + { + "epoch": 0.30575342465753425, + "grad_norm": 2.2485053539276123, + "learning_rate": 4.835766423357665e-06, + "loss": 2.3052, + "step": 279 + }, + { + "epoch": 0.30684931506849317, + "grad_norm": 2.1090753078460693, + "learning_rate": 4.854014598540146e-06, + "loss": 2.3054, + "step": 280 + }, + { + "epoch": 0.30684931506849317, + "eval_loss": 2.412548780441284, + "eval_runtime": 48.5034, + "eval_samples_per_second": 20.617, + "eval_steps_per_second": 0.866, + "step": 280 + }, + { + "epoch": 0.30794520547945203, + "grad_norm": 2.299133777618408, + "learning_rate": 4.872262773722628e-06, + "loss": 2.2599, + "step": 281 + }, + { + "epoch": 0.30904109589041096, + "grad_norm": 2.177905797958374, + "learning_rate": 4.89051094890511e-06, + "loss": 2.2745, + "step": 282 + }, + { + "epoch": 0.3101369863013699, + "grad_norm": 2.3178462982177734, + "learning_rate": 4.9087591240875915e-06, + "loss": 2.2562, + "step": 283 + }, + { + "epoch": 0.31123287671232874, + "grad_norm": 2.5088376998901367, + "learning_rate": 4.927007299270074e-06, + "loss": 2.2945, + "step": 284 + }, + { + "epoch": 0.31232876712328766, + "grad_norm": 2.095379590988159, + "learning_rate": 4.945255474452555e-06, + "loss": 2.3038, + "step": 285 + }, + { + "epoch": 0.3134246575342466, + "grad_norm": 2.1792995929718018, + "learning_rate": 4.963503649635037e-06, + "loss": 2.2702, + "step": 286 + }, + { + "epoch": 0.31452054794520545, + "grad_norm": 2.3001186847686768, + "learning_rate": 4.981751824817518e-06, + "loss": 2.3482, + "step": 287 + }, + { + "epoch": 0.3156164383561644, + "grad_norm": 2.1576106548309326, + "learning_rate": 5e-06, + "loss": 2.3391, + "step": 288 + }, + { + "epoch": 0.3167123287671233, + "grad_norm": 2.3715577125549316, + "learning_rate": 4.997969130787978e-06, + "loss": 2.3049, + "step": 289 + }, + { + "epoch": 0.3178082191780822, + "grad_norm": 2.0316715240478516, + "learning_rate": 4.995938261575955e-06, + "loss": 2.3111, + "step": 290 + }, + { + "epoch": 0.3178082191780822, + "eval_loss": 2.40299654006958, + "eval_runtime": 50.0873, + "eval_samples_per_second": 19.965, + "eval_steps_per_second": 0.839, + "step": 290 + }, + { + "epoch": 0.3189041095890411, + "grad_norm": 2.1135756969451904, + "learning_rate": 4.993907392363932e-06, + "loss": 2.2465, + "step": 291 + }, + { + "epoch": 0.32, + "grad_norm": 2.335655927658081, + "learning_rate": 4.9918765231519095e-06, + "loss": 2.31, + "step": 292 + }, + { + "epoch": 0.32109589041095893, + "grad_norm": 2.3255794048309326, + "learning_rate": 4.989845653939887e-06, + "loss": 2.2531, + "step": 293 + }, + { + "epoch": 0.3221917808219178, + "grad_norm": 2.111055612564087, + "learning_rate": 4.987814784727864e-06, + "loss": 2.2799, + "step": 294 + }, + { + "epoch": 0.3232876712328767, + "grad_norm": 2.1773171424865723, + "learning_rate": 4.985783915515841e-06, + "loss": 2.2884, + "step": 295 + }, + { + "epoch": 0.32438356164383564, + "grad_norm": 2.3559885025024414, + "learning_rate": 4.983753046303819e-06, + "loss": 2.2711, + "step": 296 + }, + { + "epoch": 0.3254794520547945, + "grad_norm": 2.120802879333496, + "learning_rate": 4.981722177091796e-06, + "loss": 2.2946, + "step": 297 + }, + { + "epoch": 0.3265753424657534, + "grad_norm": 2.2319388389587402, + "learning_rate": 4.979691307879773e-06, + "loss": 2.2771, + "step": 298 + }, + { + "epoch": 0.32767123287671235, + "grad_norm": 2.208740711212158, + "learning_rate": 4.9776604386677504e-06, + "loss": 2.3003, + "step": 299 + }, + { + "epoch": 0.3287671232876712, + "grad_norm": 2.543060541152954, + "learning_rate": 4.975629569455728e-06, + "loss": 2.2891, + "step": 300 + }, + { + "epoch": 0.3287671232876712, + "eval_loss": 2.398495674133301, + "eval_runtime": 51.1973, + "eval_samples_per_second": 19.532, + "eval_steps_per_second": 0.82, + "step": 300 + }, + { + "epoch": 0.32986301369863014, + "grad_norm": 2.086219072341919, + "learning_rate": 4.973598700243705e-06, + "loss": 2.3286, + "step": 301 + }, + { + "epoch": 0.33095890410958906, + "grad_norm": 2.3673319816589355, + "learning_rate": 4.971567831031681e-06, + "loss": 2.2672, + "step": 302 + }, + { + "epoch": 0.3320547945205479, + "grad_norm": 2.167593240737915, + "learning_rate": 4.9695369618196595e-06, + "loss": 2.264, + "step": 303 + }, + { + "epoch": 0.33315068493150685, + "grad_norm": 2.7218785285949707, + "learning_rate": 4.967506092607636e-06, + "loss": 2.1737, + "step": 304 + }, + { + "epoch": 0.33424657534246577, + "grad_norm": 2.673245429992676, + "learning_rate": 4.965475223395614e-06, + "loss": 2.2439, + "step": 305 + }, + { + "epoch": 0.33534246575342463, + "grad_norm": 2.0341989994049072, + "learning_rate": 4.9634443541835905e-06, + "loss": 2.2614, + "step": 306 + }, + { + "epoch": 0.33643835616438356, + "grad_norm": 1.9314247369766235, + "learning_rate": 4.961413484971569e-06, + "loss": 2.2592, + "step": 307 + }, + { + "epoch": 0.3375342465753425, + "grad_norm": 2.537794589996338, + "learning_rate": 4.959382615759545e-06, + "loss": 2.2859, + "step": 308 + }, + { + "epoch": 0.33863013698630134, + "grad_norm": 2.0811166763305664, + "learning_rate": 4.957351746547523e-06, + "loss": 2.2868, + "step": 309 + }, + { + "epoch": 0.33972602739726027, + "grad_norm": 2.783701181411743, + "learning_rate": 4.9553208773355e-06, + "loss": 2.2649, + "step": 310 + }, + { + "epoch": 0.33972602739726027, + "eval_loss": 2.400538921356201, + "eval_runtime": 43.185, + "eval_samples_per_second": 23.156, + "eval_steps_per_second": 0.973, + "step": 310 + }, + { + "epoch": 0.3408219178082192, + "grad_norm": 2.8541319370269775, + "learning_rate": 4.953290008123478e-06, + "loss": 2.3729, + "step": 311 + }, + { + "epoch": 0.3419178082191781, + "grad_norm": 2.1852617263793945, + "learning_rate": 4.951259138911454e-06, + "loss": 2.2885, + "step": 312 + }, + { + "epoch": 0.343013698630137, + "grad_norm": 2.872201919555664, + "learning_rate": 4.9492282696994315e-06, + "loss": 2.2353, + "step": 313 + }, + { + "epoch": 0.3441095890410959, + "grad_norm": 2.5140910148620605, + "learning_rate": 4.947197400487409e-06, + "loss": 2.3364, + "step": 314 + }, + { + "epoch": 0.3452054794520548, + "grad_norm": 2.619974374771118, + "learning_rate": 4.945166531275386e-06, + "loss": 2.2721, + "step": 315 + }, + { + "epoch": 0.3463013698630137, + "grad_norm": 2.4567618370056152, + "learning_rate": 4.943135662063363e-06, + "loss": 2.26, + "step": 316 + }, + { + "epoch": 0.3473972602739726, + "grad_norm": 2.9603192806243896, + "learning_rate": 4.9411047928513406e-06, + "loss": 2.2861, + "step": 317 + }, + { + "epoch": 0.34849315068493153, + "grad_norm": 2.115262985229492, + "learning_rate": 4.939073923639318e-06, + "loss": 2.3125, + "step": 318 + }, + { + "epoch": 0.3495890410958904, + "grad_norm": 2.693024158477783, + "learning_rate": 4.937043054427295e-06, + "loss": 2.3054, + "step": 319 + }, + { + "epoch": 0.3506849315068493, + "grad_norm": 2.3439996242523193, + "learning_rate": 4.935012185215272e-06, + "loss": 2.2576, + "step": 320 + }, + { + "epoch": 0.3506849315068493, + "eval_loss": 2.3920650482177734, + "eval_runtime": 47.7451, + "eval_samples_per_second": 20.945, + "eval_steps_per_second": 0.88, + "step": 320 + }, + { + "epoch": 0.35178082191780824, + "grad_norm": 2.3009634017944336, + "learning_rate": 4.93298131600325e-06, + "loss": 2.2454, + "step": 321 + }, + { + "epoch": 0.3528767123287671, + "grad_norm": 2.593203544616699, + "learning_rate": 4.930950446791227e-06, + "loss": 2.2965, + "step": 322 + }, + { + "epoch": 0.353972602739726, + "grad_norm": 2.9069650173187256, + "learning_rate": 4.928919577579204e-06, + "loss": 2.2785, + "step": 323 + }, + { + "epoch": 0.35506849315068495, + "grad_norm": 2.1021759510040283, + "learning_rate": 4.9268887083671815e-06, + "loss": 2.2857, + "step": 324 + }, + { + "epoch": 0.3561643835616438, + "grad_norm": 2.0330185890197754, + "learning_rate": 4.924857839155159e-06, + "loss": 2.2697, + "step": 325 + }, + { + "epoch": 0.35726027397260274, + "grad_norm": 2.1318063735961914, + "learning_rate": 4.922826969943136e-06, + "loss": 2.2698, + "step": 326 + }, + { + "epoch": 0.35835616438356166, + "grad_norm": 2.0157387256622314, + "learning_rate": 4.920796100731113e-06, + "loss": 2.292, + "step": 327 + }, + { + "epoch": 0.3594520547945205, + "grad_norm": 2.3266026973724365, + "learning_rate": 4.918765231519091e-06, + "loss": 2.2215, + "step": 328 + }, + { + "epoch": 0.36054794520547945, + "grad_norm": 2.066305160522461, + "learning_rate": 4.916734362307068e-06, + "loss": 2.2758, + "step": 329 + }, + { + "epoch": 0.36164383561643837, + "grad_norm": 2.0219993591308594, + "learning_rate": 4.914703493095045e-06, + "loss": 2.2381, + "step": 330 + }, + { + "epoch": 0.36164383561643837, + "eval_loss": 2.394064426422119, + "eval_runtime": 51.5482, + "eval_samples_per_second": 19.399, + "eval_steps_per_second": 0.815, + "step": 330 + }, + { + "epoch": 0.36273972602739724, + "grad_norm": 2.3964319229125977, + "learning_rate": 4.912672623883022e-06, + "loss": 2.2384, + "step": 331 + }, + { + "epoch": 0.36383561643835616, + "grad_norm": 2.5466530323028564, + "learning_rate": 4.910641754671e-06, + "loss": 2.2428, + "step": 332 + }, + { + "epoch": 0.3649315068493151, + "grad_norm": 2.236152410507202, + "learning_rate": 4.908610885458977e-06, + "loss": 2.2639, + "step": 333 + }, + { + "epoch": 0.36602739726027395, + "grad_norm": 2.7267303466796875, + "learning_rate": 4.906580016246954e-06, + "loss": 2.2608, + "step": 334 + }, + { + "epoch": 0.36712328767123287, + "grad_norm": 1.9403765201568604, + "learning_rate": 4.9045491470349315e-06, + "loss": 2.2962, + "step": 335 + }, + { + "epoch": 0.3682191780821918, + "grad_norm": 2.89070463180542, + "learning_rate": 4.902518277822909e-06, + "loss": 2.2847, + "step": 336 + }, + { + "epoch": 0.3693150684931507, + "grad_norm": 2.6427228450775146, + "learning_rate": 4.900487408610886e-06, + "loss": 2.2734, + "step": 337 + }, + { + "epoch": 0.3704109589041096, + "grad_norm": 2.1680362224578857, + "learning_rate": 4.898456539398863e-06, + "loss": 2.2869, + "step": 338 + }, + { + "epoch": 0.3715068493150685, + "grad_norm": 3.0160160064697266, + "learning_rate": 4.89642567018684e-06, + "loss": 2.3043, + "step": 339 + }, + { + "epoch": 0.3726027397260274, + "grad_norm": 1.9858418703079224, + "learning_rate": 4.894394800974818e-06, + "loss": 2.3422, + "step": 340 + }, + { + "epoch": 0.3726027397260274, + "eval_loss": 2.387559652328491, + "eval_runtime": 52.2455, + "eval_samples_per_second": 19.14, + "eval_steps_per_second": 0.804, + "step": 340 + }, + { + "epoch": 0.3736986301369863, + "grad_norm": 2.243699073791504, + "learning_rate": 4.892363931762794e-06, + "loss": 2.2394, + "step": 341 + }, + { + "epoch": 0.3747945205479452, + "grad_norm": 2.371299982070923, + "learning_rate": 4.8903330625507725e-06, + "loss": 2.2856, + "step": 342 + }, + { + "epoch": 0.37589041095890413, + "grad_norm": 2.3769736289978027, + "learning_rate": 4.888302193338749e-06, + "loss": 2.246, + "step": 343 + }, + { + "epoch": 0.376986301369863, + "grad_norm": 2.3209030628204346, + "learning_rate": 4.886271324126727e-06, + "loss": 2.2331, + "step": 344 + }, + { + "epoch": 0.3780821917808219, + "grad_norm": 2.747319221496582, + "learning_rate": 4.8842404549147034e-06, + "loss": 2.3115, + "step": 345 + }, + { + "epoch": 0.37917808219178084, + "grad_norm": 1.8599231243133545, + "learning_rate": 4.8822095857026816e-06, + "loss": 2.2994, + "step": 346 + }, + { + "epoch": 0.3802739726027397, + "grad_norm": 2.0358943939208984, + "learning_rate": 4.880178716490658e-06, + "loss": 2.271, + "step": 347 + }, + { + "epoch": 0.38136986301369863, + "grad_norm": 2.409106731414795, + "learning_rate": 4.878147847278636e-06, + "loss": 2.2281, + "step": 348 + }, + { + "epoch": 0.38246575342465755, + "grad_norm": 2.1641807556152344, + "learning_rate": 4.8761169780666125e-06, + "loss": 2.2437, + "step": 349 + }, + { + "epoch": 0.3835616438356164, + "grad_norm": 2.6917359828948975, + "learning_rate": 4.874086108854591e-06, + "loss": 2.226, + "step": 350 + }, + { + "epoch": 0.3835616438356164, + "eval_loss": 2.382991313934326, + "eval_runtime": 48.9534, + "eval_samples_per_second": 20.428, + "eval_steps_per_second": 0.858, + "step": 350 + }, + { + "epoch": 0.38465753424657534, + "grad_norm": 2.364215612411499, + "learning_rate": 4.872055239642567e-06, + "loss": 2.2319, + "step": 351 + }, + { + "epoch": 0.38575342465753426, + "grad_norm": 2.317643165588379, + "learning_rate": 4.870024370430544e-06, + "loss": 2.2096, + "step": 352 + }, + { + "epoch": 0.3868493150684931, + "grad_norm": 2.2283246517181396, + "learning_rate": 4.867993501218522e-06, + "loss": 2.2634, + "step": 353 + }, + { + "epoch": 0.38794520547945205, + "grad_norm": 2.498391628265381, + "learning_rate": 4.865962632006499e-06, + "loss": 2.2813, + "step": 354 + }, + { + "epoch": 0.38904109589041097, + "grad_norm": 2.3889403343200684, + "learning_rate": 4.863931762794476e-06, + "loss": 2.2361, + "step": 355 + }, + { + "epoch": 0.39013698630136984, + "grad_norm": 2.2049975395202637, + "learning_rate": 4.8619008935824535e-06, + "loss": 2.2583, + "step": 356 + }, + { + "epoch": 0.39123287671232876, + "grad_norm": 2.132840871810913, + "learning_rate": 4.859870024370431e-06, + "loss": 2.2823, + "step": 357 + }, + { + "epoch": 0.3923287671232877, + "grad_norm": 2.2692184448242188, + "learning_rate": 4.857839155158408e-06, + "loss": 2.2939, + "step": 358 + }, + { + "epoch": 0.3934246575342466, + "grad_norm": 2.0501933097839355, + "learning_rate": 4.855808285946385e-06, + "loss": 2.2976, + "step": 359 + }, + { + "epoch": 0.39452054794520547, + "grad_norm": 2.170236587524414, + "learning_rate": 4.8537774167343626e-06, + "loss": 2.2748, + "step": 360 + }, + { + "epoch": 0.39452054794520547, + "eval_loss": 2.3783886432647705, + "eval_runtime": 46.7324, + "eval_samples_per_second": 21.398, + "eval_steps_per_second": 0.899, + "step": 360 + }, + { + "epoch": 0.3956164383561644, + "grad_norm": 2.283137083053589, + "learning_rate": 4.85174654752234e-06, + "loss": 2.2705, + "step": 361 + }, + { + "epoch": 0.3967123287671233, + "grad_norm": 2.195984363555908, + "learning_rate": 4.849715678310317e-06, + "loss": 2.26, + "step": 362 + }, + { + "epoch": 0.3978082191780822, + "grad_norm": 2.399935007095337, + "learning_rate": 4.847684809098294e-06, + "loss": 2.2639, + "step": 363 + }, + { + "epoch": 0.3989041095890411, + "grad_norm": 2.411280393600464, + "learning_rate": 4.845653939886272e-06, + "loss": 2.2523, + "step": 364 + }, + { + "epoch": 0.4, + "grad_norm": 2.2033896446228027, + "learning_rate": 4.843623070674249e-06, + "loss": 2.2722, + "step": 365 + }, + { + "epoch": 0.4010958904109589, + "grad_norm": 2.2990238666534424, + "learning_rate": 4.841592201462226e-06, + "loss": 2.2411, + "step": 366 + }, + { + "epoch": 0.4021917808219178, + "grad_norm": 1.911453127861023, + "learning_rate": 4.8395613322502035e-06, + "loss": 2.2921, + "step": 367 + }, + { + "epoch": 0.40328767123287673, + "grad_norm": 2.453195095062256, + "learning_rate": 4.837530463038181e-06, + "loss": 2.218, + "step": 368 + }, + { + "epoch": 0.4043835616438356, + "grad_norm": 2.126992702484131, + "learning_rate": 4.835499593826158e-06, + "loss": 2.2628, + "step": 369 + }, + { + "epoch": 0.4054794520547945, + "grad_norm": 1.9305835962295532, + "learning_rate": 4.833468724614135e-06, + "loss": 2.2726, + "step": 370 + }, + { + "epoch": 0.4054794520547945, + "eval_loss": 2.3741114139556885, + "eval_runtime": 49.2956, + "eval_samples_per_second": 20.286, + "eval_steps_per_second": 0.852, + "step": 370 + }, + { + "epoch": 0.40657534246575344, + "grad_norm": 2.222024917602539, + "learning_rate": 4.831437855402113e-06, + "loss": 2.2634, + "step": 371 + }, + { + "epoch": 0.4076712328767123, + "grad_norm": 2.473958969116211, + "learning_rate": 4.82940698619009e-06, + "loss": 2.1817, + "step": 372 + }, + { + "epoch": 0.40876712328767123, + "grad_norm": 1.9208838939666748, + "learning_rate": 4.827376116978067e-06, + "loss": 2.2353, + "step": 373 + }, + { + "epoch": 0.40986301369863015, + "grad_norm": 2.7226450443267822, + "learning_rate": 4.8253452477660444e-06, + "loss": 2.2831, + "step": 374 + }, + { + "epoch": 0.410958904109589, + "grad_norm": 1.9078222513198853, + "learning_rate": 4.823314378554022e-06, + "loss": 2.264, + "step": 375 + }, + { + "epoch": 0.41205479452054794, + "grad_norm": 2.3946311473846436, + "learning_rate": 4.821283509341998e-06, + "loss": 2.251, + "step": 376 + }, + { + "epoch": 0.41315068493150686, + "grad_norm": 1.847343921661377, + "learning_rate": 4.819252640129976e-06, + "loss": 2.2441, + "step": 377 + }, + { + "epoch": 0.41424657534246573, + "grad_norm": 2.475385904312134, + "learning_rate": 4.817221770917953e-06, + "loss": 2.2639, + "step": 378 + }, + { + "epoch": 0.41534246575342465, + "grad_norm": 2.1396560668945312, + "learning_rate": 4.815190901705931e-06, + "loss": 2.2353, + "step": 379 + }, + { + "epoch": 0.41643835616438357, + "grad_norm": 2.3868885040283203, + "learning_rate": 4.813160032493907e-06, + "loss": 2.2551, + "step": 380 + }, + { + "epoch": 0.41643835616438357, + "eval_loss": 2.369168519973755, + "eval_runtime": 45.6131, + "eval_samples_per_second": 21.924, + "eval_steps_per_second": 0.921, + "step": 380 + }, + { + "epoch": 0.41753424657534244, + "grad_norm": 2.0193872451782227, + "learning_rate": 4.811129163281885e-06, + "loss": 2.2399, + "step": 381 + }, + { + "epoch": 0.41863013698630136, + "grad_norm": 2.4655935764312744, + "learning_rate": 4.809098294069862e-06, + "loss": 2.2729, + "step": 382 + }, + { + "epoch": 0.4197260273972603, + "grad_norm": 1.880974292755127, + "learning_rate": 4.80706742485784e-06, + "loss": 2.2681, + "step": 383 + }, + { + "epoch": 0.4208219178082192, + "grad_norm": 2.1912078857421875, + "learning_rate": 4.805036555645816e-06, + "loss": 2.2961, + "step": 384 + }, + { + "epoch": 0.42191780821917807, + "grad_norm": 1.985931634902954, + "learning_rate": 4.8030056864337945e-06, + "loss": 2.2407, + "step": 385 + }, + { + "epoch": 0.423013698630137, + "grad_norm": 2.4353220462799072, + "learning_rate": 4.800974817221771e-06, + "loss": 2.2867, + "step": 386 + }, + { + "epoch": 0.4241095890410959, + "grad_norm": 1.8810508251190186, + "learning_rate": 4.798943948009749e-06, + "loss": 2.2498, + "step": 387 + }, + { + "epoch": 0.4252054794520548, + "grad_norm": 2.195033073425293, + "learning_rate": 4.7969130787977254e-06, + "loss": 2.2532, + "step": 388 + }, + { + "epoch": 0.4263013698630137, + "grad_norm": 1.855372428894043, + "learning_rate": 4.794882209585703e-06, + "loss": 2.2538, + "step": 389 + }, + { + "epoch": 0.4273972602739726, + "grad_norm": 2.4465434551239014, + "learning_rate": 4.79285134037368e-06, + "loss": 2.2724, + "step": 390 + }, + { + "epoch": 0.4273972602739726, + "eval_loss": 2.364928960800171, + "eval_runtime": 48.0245, + "eval_samples_per_second": 20.823, + "eval_steps_per_second": 0.875, + "step": 390 + }, + { + "epoch": 0.4284931506849315, + "grad_norm": 1.8911314010620117, + "learning_rate": 4.790820471161657e-06, + "loss": 2.2291, + "step": 391 + }, + { + "epoch": 0.4295890410958904, + "grad_norm": 2.1975977420806885, + "learning_rate": 4.7887896019496345e-06, + "loss": 2.2373, + "step": 392 + }, + { + "epoch": 0.43068493150684933, + "grad_norm": 2.015331506729126, + "learning_rate": 4.786758732737612e-06, + "loss": 2.2483, + "step": 393 + }, + { + "epoch": 0.4317808219178082, + "grad_norm": 2.039339065551758, + "learning_rate": 4.784727863525589e-06, + "loss": 2.2746, + "step": 394 + }, + { + "epoch": 0.4328767123287671, + "grad_norm": 1.977513074874878, + "learning_rate": 4.782696994313566e-06, + "loss": 2.2275, + "step": 395 + }, + { + "epoch": 0.43397260273972604, + "grad_norm": 2.147217035293579, + "learning_rate": 4.780666125101544e-06, + "loss": 2.252, + "step": 396 + }, + { + "epoch": 0.4350684931506849, + "grad_norm": 2.1799919605255127, + "learning_rate": 4.778635255889521e-06, + "loss": 2.1975, + "step": 397 + }, + { + "epoch": 0.43616438356164383, + "grad_norm": 2.582240343093872, + "learning_rate": 4.776604386677498e-06, + "loss": 2.2294, + "step": 398 + }, + { + "epoch": 0.43726027397260275, + "grad_norm": 1.9127846956253052, + "learning_rate": 4.7745735174654755e-06, + "loss": 2.2319, + "step": 399 + }, + { + "epoch": 0.4383561643835616, + "grad_norm": 2.2444543838500977, + "learning_rate": 4.772542648253453e-06, + "loss": 2.2857, + "step": 400 + }, + { + "epoch": 0.4383561643835616, + "eval_loss": 2.3653724193573, + "eval_runtime": 43.8304, + "eval_samples_per_second": 22.815, + "eval_steps_per_second": 0.958, + "step": 400 + }, + { + "epoch": 0.43945205479452054, + "grad_norm": 1.9913676977157593, + "learning_rate": 4.77051177904143e-06, + "loss": 2.274, + "step": 401 + }, + { + "epoch": 0.44054794520547946, + "grad_norm": 2.2850067615509033, + "learning_rate": 4.768480909829407e-06, + "loss": 2.2209, + "step": 402 + }, + { + "epoch": 0.44164383561643833, + "grad_norm": 2.4984688758850098, + "learning_rate": 4.766450040617385e-06, + "loss": 2.263, + "step": 403 + }, + { + "epoch": 0.44273972602739725, + "grad_norm": 2.3244640827178955, + "learning_rate": 4.764419171405362e-06, + "loss": 2.2368, + "step": 404 + }, + { + "epoch": 0.4438356164383562, + "grad_norm": 2.7579681873321533, + "learning_rate": 4.762388302193339e-06, + "loss": 2.2762, + "step": 405 + }, + { + "epoch": 0.4449315068493151, + "grad_norm": 2.1996467113494873, + "learning_rate": 4.760357432981316e-06, + "loss": 2.2536, + "step": 406 + }, + { + "epoch": 0.44602739726027396, + "grad_norm": 2.3767571449279785, + "learning_rate": 4.758326563769294e-06, + "loss": 2.2353, + "step": 407 + }, + { + "epoch": 0.4471232876712329, + "grad_norm": 1.9148359298706055, + "learning_rate": 4.756295694557271e-06, + "loss": 2.2416, + "step": 408 + }, + { + "epoch": 0.4482191780821918, + "grad_norm": 2.6499507427215576, + "learning_rate": 4.754264825345248e-06, + "loss": 2.2641, + "step": 409 + }, + { + "epoch": 0.44931506849315067, + "grad_norm": 1.9890286922454834, + "learning_rate": 4.7522339561332255e-06, + "loss": 2.2148, + "step": 410 + }, + { + "epoch": 0.44931506849315067, + "eval_loss": 2.3670520782470703, + "eval_runtime": 41.1722, + "eval_samples_per_second": 24.288, + "eval_steps_per_second": 1.02, + "step": 410 + }, + { + "epoch": 0.4504109589041096, + "grad_norm": 2.575968027114868, + "learning_rate": 4.750203086921203e-06, + "loss": 2.2254, + "step": 411 + }, + { + "epoch": 0.4515068493150685, + "grad_norm": 2.015174388885498, + "learning_rate": 4.74817221770918e-06, + "loss": 2.1881, + "step": 412 + }, + { + "epoch": 0.4526027397260274, + "grad_norm": 2.44292950630188, + "learning_rate": 4.7461413484971565e-06, + "loss": 2.256, + "step": 413 + }, + { + "epoch": 0.4536986301369863, + "grad_norm": 1.9808441400527954, + "learning_rate": 4.744110479285135e-06, + "loss": 2.285, + "step": 414 + }, + { + "epoch": 0.4547945205479452, + "grad_norm": 2.009481906890869, + "learning_rate": 4.742079610073111e-06, + "loss": 2.2018, + "step": 415 + }, + { + "epoch": 0.4558904109589041, + "grad_norm": 2.1466031074523926, + "learning_rate": 4.740048740861089e-06, + "loss": 2.224, + "step": 416 + }, + { + "epoch": 0.456986301369863, + "grad_norm": 2.1337907314300537, + "learning_rate": 4.738017871649066e-06, + "loss": 2.2404, + "step": 417 + }, + { + "epoch": 0.45808219178082193, + "grad_norm": 1.94125235080719, + "learning_rate": 4.735987002437044e-06, + "loss": 2.2474, + "step": 418 + }, + { + "epoch": 0.4591780821917808, + "grad_norm": 2.379099130630493, + "learning_rate": 4.73395613322502e-06, + "loss": 2.2858, + "step": 419 + }, + { + "epoch": 0.4602739726027397, + "grad_norm": 1.844056248664856, + "learning_rate": 4.731925264012998e-06, + "loss": 2.2967, + "step": 420 + }, + { + "epoch": 0.4602739726027397, + "eval_loss": 2.3622241020202637, + "eval_runtime": 45.4674, + "eval_samples_per_second": 21.994, + "eval_steps_per_second": 0.924, + "step": 420 + }, + { + "epoch": 0.46136986301369864, + "grad_norm": 2.411895513534546, + "learning_rate": 4.729894394800975e-06, + "loss": 2.274, + "step": 421 + }, + { + "epoch": 0.4624657534246575, + "grad_norm": 2.014885425567627, + "learning_rate": 4.727863525588953e-06, + "loss": 2.2144, + "step": 422 + }, + { + "epoch": 0.46356164383561643, + "grad_norm": 2.112009048461914, + "learning_rate": 4.725832656376929e-06, + "loss": 2.1824, + "step": 423 + }, + { + "epoch": 0.46465753424657535, + "grad_norm": 2.111392021179199, + "learning_rate": 4.723801787164907e-06, + "loss": 2.2261, + "step": 424 + }, + { + "epoch": 0.4657534246575342, + "grad_norm": 2.0349185466766357, + "learning_rate": 4.721770917952885e-06, + "loss": 2.2622, + "step": 425 + }, + { + "epoch": 0.46684931506849314, + "grad_norm": 2.30222487449646, + "learning_rate": 4.719740048740861e-06, + "loss": 2.2206, + "step": 426 + }, + { + "epoch": 0.46794520547945206, + "grad_norm": 2.018789529800415, + "learning_rate": 4.717709179528839e-06, + "loss": 2.2196, + "step": 427 + }, + { + "epoch": 0.469041095890411, + "grad_norm": 2.046649932861328, + "learning_rate": 4.715678310316816e-06, + "loss": 2.2584, + "step": 428 + }, + { + "epoch": 0.47013698630136985, + "grad_norm": 2.0393869876861572, + "learning_rate": 4.713647441104794e-06, + "loss": 2.2537, + "step": 429 + }, + { + "epoch": 0.4712328767123288, + "grad_norm": 2.2788116931915283, + "learning_rate": 4.71161657189277e-06, + "loss": 2.2145, + "step": 430 + }, + { + "epoch": 0.4712328767123288, + "eval_loss": 2.3575785160064697, + "eval_runtime": 44.6687, + "eval_samples_per_second": 22.387, + "eval_steps_per_second": 0.94, + "step": 430 + }, + { + "epoch": 0.4723287671232877, + "grad_norm": 1.9950047731399536, + "learning_rate": 4.709585702680748e-06, + "loss": 2.2072, + "step": 431 + }, + { + "epoch": 0.47342465753424656, + "grad_norm": 1.9222451448440552, + "learning_rate": 4.707554833468725e-06, + "loss": 2.2654, + "step": 432 + }, + { + "epoch": 0.4745205479452055, + "grad_norm": 1.9735718965530396, + "learning_rate": 4.705523964256703e-06, + "loss": 2.3037, + "step": 433 + }, + { + "epoch": 0.4756164383561644, + "grad_norm": 2.3209776878356934, + "learning_rate": 4.703493095044679e-06, + "loss": 2.2083, + "step": 434 + }, + { + "epoch": 0.4767123287671233, + "grad_norm": 2.0993995666503906, + "learning_rate": 4.701462225832657e-06, + "loss": 2.2275, + "step": 435 + }, + { + "epoch": 0.4778082191780822, + "grad_norm": 2.334463119506836, + "learning_rate": 4.699431356620634e-06, + "loss": 2.1852, + "step": 436 + }, + { + "epoch": 0.4789041095890411, + "grad_norm": 1.9098213911056519, + "learning_rate": 4.697400487408612e-06, + "loss": 2.2829, + "step": 437 + }, + { + "epoch": 0.48, + "grad_norm": 2.3005056381225586, + "learning_rate": 4.695369618196588e-06, + "loss": 2.2532, + "step": 438 + }, + { + "epoch": 0.4810958904109589, + "grad_norm": 1.9069781303405762, + "learning_rate": 4.693338748984566e-06, + "loss": 2.2374, + "step": 439 + }, + { + "epoch": 0.4821917808219178, + "grad_norm": 2.050032138824463, + "learning_rate": 4.691307879772543e-06, + "loss": 2.2256, + "step": 440 + }, + { + "epoch": 0.4821917808219178, + "eval_loss": 2.3587422370910645, + "eval_runtime": 45.0941, + "eval_samples_per_second": 22.176, + "eval_steps_per_second": 0.931, + "step": 440 + }, + { + "epoch": 0.4832876712328767, + "grad_norm": 2.0563836097717285, + "learning_rate": 4.68927701056052e-06, + "loss": 2.1918, + "step": 441 + }, + { + "epoch": 0.4843835616438356, + "grad_norm": 2.1598386764526367, + "learning_rate": 4.6872461413484975e-06, + "loss": 2.2618, + "step": 442 + }, + { + "epoch": 0.48547945205479454, + "grad_norm": 1.9236539602279663, + "learning_rate": 4.685215272136475e-06, + "loss": 2.2477, + "step": 443 + }, + { + "epoch": 0.4865753424657534, + "grad_norm": 1.9923924207687378, + "learning_rate": 4.683184402924452e-06, + "loss": 2.2313, + "step": 444 + }, + { + "epoch": 0.4876712328767123, + "grad_norm": 2.162627696990967, + "learning_rate": 4.681153533712429e-06, + "loss": 2.2367, + "step": 445 + }, + { + "epoch": 0.48876712328767125, + "grad_norm": 2.083523988723755, + "learning_rate": 4.679122664500407e-06, + "loss": 2.2946, + "step": 446 + }, + { + "epoch": 0.4898630136986301, + "grad_norm": 2.504408359527588, + "learning_rate": 4.677091795288384e-06, + "loss": 2.2104, + "step": 447 + }, + { + "epoch": 0.49095890410958903, + "grad_norm": 1.8316532373428345, + "learning_rate": 4.675060926076361e-06, + "loss": 2.2027, + "step": 448 + }, + { + "epoch": 0.49205479452054796, + "grad_norm": 2.3337953090667725, + "learning_rate": 4.673030056864338e-06, + "loss": 2.278, + "step": 449 + }, + { + "epoch": 0.4931506849315068, + "grad_norm": 2.24102783203125, + "learning_rate": 4.670999187652316e-06, + "loss": 2.2018, + "step": 450 + }, + { + "epoch": 0.4931506849315068, + "eval_loss": 2.3545279502868652, + "eval_runtime": 45.1041, + "eval_samples_per_second": 22.171, + "eval_steps_per_second": 0.931, + "step": 450 + }, + { + "epoch": 0.49424657534246574, + "grad_norm": 2.07834792137146, + "learning_rate": 4.668968318440293e-06, + "loss": 2.1975, + "step": 451 + }, + { + "epoch": 0.49534246575342467, + "grad_norm": 1.7681723833084106, + "learning_rate": 4.66693744922827e-06, + "loss": 2.2033, + "step": 452 + }, + { + "epoch": 0.4964383561643836, + "grad_norm": 2.062134027481079, + "learning_rate": 4.6649065800162475e-06, + "loss": 2.2646, + "step": 453 + }, + { + "epoch": 0.49753424657534245, + "grad_norm": 1.94647216796875, + "learning_rate": 4.662875710804225e-06, + "loss": 2.1748, + "step": 454 + }, + { + "epoch": 0.4986301369863014, + "grad_norm": 1.9173872470855713, + "learning_rate": 4.660844841592202e-06, + "loss": 2.2435, + "step": 455 + }, + { + "epoch": 0.4997260273972603, + "grad_norm": 1.88880455493927, + "learning_rate": 4.658813972380179e-06, + "loss": 2.1711, + "step": 456 + }, + { + "epoch": 0.5008219178082192, + "grad_norm": 1.9628506898880005, + "learning_rate": 4.656783103168157e-06, + "loss": 2.2701, + "step": 457 + }, + { + "epoch": 0.5019178082191781, + "grad_norm": 2.0515286922454834, + "learning_rate": 4.654752233956134e-06, + "loss": 2.1656, + "step": 458 + }, + { + "epoch": 0.503013698630137, + "grad_norm": 1.978522539138794, + "learning_rate": 4.652721364744111e-06, + "loss": 2.246, + "step": 459 + }, + { + "epoch": 0.5041095890410959, + "grad_norm": 1.7993966341018677, + "learning_rate": 4.6506904955320885e-06, + "loss": 2.214, + "step": 460 + }, + { + "epoch": 0.5041095890410959, + "eval_loss": 2.3568837642669678, + "eval_runtime": 45.2086, + "eval_samples_per_second": 22.12, + "eval_steps_per_second": 0.929, + "step": 460 + }, + { + "epoch": 0.5052054794520547, + "grad_norm": 1.7809834480285645, + "learning_rate": 4.648659626320066e-06, + "loss": 2.2525, + "step": 461 + }, + { + "epoch": 0.5063013698630137, + "grad_norm": 1.9256166219711304, + "learning_rate": 4.646628757108043e-06, + "loss": 2.2611, + "step": 462 + }, + { + "epoch": 0.5073972602739726, + "grad_norm": 1.94192636013031, + "learning_rate": 4.6445978878960194e-06, + "loss": 2.2282, + "step": 463 + }, + { + "epoch": 0.5084931506849315, + "grad_norm": 1.8891818523406982, + "learning_rate": 4.6425670186839976e-06, + "loss": 2.2036, + "step": 464 + }, + { + "epoch": 0.5095890410958904, + "grad_norm": 2.352078676223755, + "learning_rate": 4.640536149471974e-06, + "loss": 2.2293, + "step": 465 + }, + { + "epoch": 0.5106849315068493, + "grad_norm": 1.8141659498214722, + "learning_rate": 4.638505280259952e-06, + "loss": 2.2551, + "step": 466 + }, + { + "epoch": 0.5117808219178083, + "grad_norm": 1.9492480754852295, + "learning_rate": 4.6364744110479285e-06, + "loss": 2.2641, + "step": 467 + }, + { + "epoch": 0.5128767123287671, + "grad_norm": 2.0210182666778564, + "learning_rate": 4.634443541835907e-06, + "loss": 2.2367, + "step": 468 + }, + { + "epoch": 0.513972602739726, + "grad_norm": 1.8936505317687988, + "learning_rate": 4.632412672623883e-06, + "loss": 2.265, + "step": 469 + }, + { + "epoch": 0.5150684931506849, + "grad_norm": 1.845888614654541, + "learning_rate": 4.630381803411861e-06, + "loss": 2.2169, + "step": 470 + }, + { + "epoch": 0.5150684931506849, + "eval_loss": 2.353489875793457, + "eval_runtime": 46.8739, + "eval_samples_per_second": 21.334, + "eval_steps_per_second": 0.896, + "step": 470 + }, + { + "epoch": 0.5161643835616438, + "grad_norm": 1.889675259590149, + "learning_rate": 4.628350934199838e-06, + "loss": 2.1383, + "step": 471 + }, + { + "epoch": 0.5172602739726028, + "grad_norm": 1.9799216985702515, + "learning_rate": 4.626320064987816e-06, + "loss": 2.2312, + "step": 472 + }, + { + "epoch": 0.5183561643835617, + "grad_norm": 2.0128607749938965, + "learning_rate": 4.624289195775792e-06, + "loss": 2.2529, + "step": 473 + }, + { + "epoch": 0.5194520547945205, + "grad_norm": 1.8791241645812988, + "learning_rate": 4.62225832656377e-06, + "loss": 2.226, + "step": 474 + }, + { + "epoch": 0.5205479452054794, + "grad_norm": 2.0497279167175293, + "learning_rate": 4.620227457351747e-06, + "loss": 2.2283, + "step": 475 + }, + { + "epoch": 0.5216438356164383, + "grad_norm": 1.8097424507141113, + "learning_rate": 4.618196588139724e-06, + "loss": 2.222, + "step": 476 + }, + { + "epoch": 0.5227397260273973, + "grad_norm": 1.8960500955581665, + "learning_rate": 4.616165718927701e-06, + "loss": 2.2342, + "step": 477 + }, + { + "epoch": 0.5238356164383562, + "grad_norm": 2.020730495452881, + "learning_rate": 4.6141348497156786e-06, + "loss": 2.2106, + "step": 478 + }, + { + "epoch": 0.5249315068493151, + "grad_norm": 1.8955297470092773, + "learning_rate": 4.612103980503656e-06, + "loss": 2.2755, + "step": 479 + }, + { + "epoch": 0.5260273972602739, + "grad_norm": 1.9767351150512695, + "learning_rate": 4.610073111291633e-06, + "loss": 2.2554, + "step": 480 + }, + { + "epoch": 0.5260273972602739, + "eval_loss": 2.347705841064453, + "eval_runtime": 46.9125, + "eval_samples_per_second": 21.316, + "eval_steps_per_second": 0.895, + "step": 480 + }, + { + "epoch": 0.5271232876712328, + "grad_norm": 1.936457633972168, + "learning_rate": 4.60804224207961e-06, + "loss": 2.2323, + "step": 481 + }, + { + "epoch": 0.5282191780821918, + "grad_norm": 1.9271862506866455, + "learning_rate": 4.606011372867588e-06, + "loss": 2.2151, + "step": 482 + }, + { + "epoch": 0.5293150684931507, + "grad_norm": 1.878153681755066, + "learning_rate": 4.603980503655565e-06, + "loss": 2.2322, + "step": 483 + }, + { + "epoch": 0.5304109589041096, + "grad_norm": 2.1009249687194824, + "learning_rate": 4.601949634443542e-06, + "loss": 2.2181, + "step": 484 + }, + { + "epoch": 0.5315068493150685, + "grad_norm": 1.9089181423187256, + "learning_rate": 4.5999187652315195e-06, + "loss": 2.1712, + "step": 485 + }, + { + "epoch": 0.5326027397260275, + "grad_norm": 2.207282781600952, + "learning_rate": 4.597887896019497e-06, + "loss": 2.1847, + "step": 486 + }, + { + "epoch": 0.5336986301369863, + "grad_norm": 2.001584053039551, + "learning_rate": 4.595857026807474e-06, + "loss": 2.214, + "step": 487 + }, + { + "epoch": 0.5347945205479452, + "grad_norm": 1.966700553894043, + "learning_rate": 4.593826157595451e-06, + "loss": 2.2237, + "step": 488 + }, + { + "epoch": 0.5358904109589041, + "grad_norm": 1.842469334602356, + "learning_rate": 4.591795288383429e-06, + "loss": 2.2089, + "step": 489 + }, + { + "epoch": 0.536986301369863, + "grad_norm": 2.0412986278533936, + "learning_rate": 4.589764419171406e-06, + "loss": 2.2734, + "step": 490 + }, + { + "epoch": 0.536986301369863, + "eval_loss": 2.3460772037506104, + "eval_runtime": 50.4048, + "eval_samples_per_second": 19.839, + "eval_steps_per_second": 0.833, + "step": 490 + }, + { + "epoch": 0.538082191780822, + "grad_norm": 1.9337031841278076, + "learning_rate": 4.587733549959383e-06, + "loss": 2.2314, + "step": 491 + }, + { + "epoch": 0.5391780821917809, + "grad_norm": 2.4448320865631104, + "learning_rate": 4.5857026807473604e-06, + "loss": 2.2098, + "step": 492 + }, + { + "epoch": 0.5402739726027397, + "grad_norm": 1.83162522315979, + "learning_rate": 4.583671811535338e-06, + "loss": 2.1902, + "step": 493 + }, + { + "epoch": 0.5413698630136986, + "grad_norm": 2.3614656925201416, + "learning_rate": 4.581640942323315e-06, + "loss": 2.1737, + "step": 494 + }, + { + "epoch": 0.5424657534246575, + "grad_norm": 2.001187801361084, + "learning_rate": 4.579610073111292e-06, + "loss": 2.2485, + "step": 495 + }, + { + "epoch": 0.5435616438356164, + "grad_norm": 2.3479933738708496, + "learning_rate": 4.5775792038992695e-06, + "loss": 2.2595, + "step": 496 + }, + { + "epoch": 0.5446575342465754, + "grad_norm": 2.135856866836548, + "learning_rate": 4.575548334687247e-06, + "loss": 2.2528, + "step": 497 + }, + { + "epoch": 0.5457534246575343, + "grad_norm": 1.983126163482666, + "learning_rate": 4.573517465475224e-06, + "loss": 2.2788, + "step": 498 + }, + { + "epoch": 0.5468493150684931, + "grad_norm": 2.18208909034729, + "learning_rate": 4.571486596263201e-06, + "loss": 2.1995, + "step": 499 + }, + { + "epoch": 0.547945205479452, + "grad_norm": 1.848747968673706, + "learning_rate": 4.569455727051178e-06, + "loss": 2.2403, + "step": 500 + }, + { + "epoch": 0.547945205479452, + "eval_loss": 2.344066619873047, + "eval_runtime": 51.1213, + "eval_samples_per_second": 19.561, + "eval_steps_per_second": 0.822, + "step": 500 + }, + { + "epoch": 0.549041095890411, + "grad_norm": 2.427860736846924, + "learning_rate": 4.567424857839156e-06, + "loss": 2.246, + "step": 501 + }, + { + "epoch": 0.5501369863013699, + "grad_norm": 1.913405418395996, + "learning_rate": 4.565393988627132e-06, + "loss": 2.2296, + "step": 502 + }, + { + "epoch": 0.5512328767123288, + "grad_norm": 2.0106029510498047, + "learning_rate": 4.5633631194151105e-06, + "loss": 2.2657, + "step": 503 + }, + { + "epoch": 0.5523287671232877, + "grad_norm": 2.2719595432281494, + "learning_rate": 4.561332250203087e-06, + "loss": 2.2341, + "step": 504 + }, + { + "epoch": 0.5534246575342465, + "grad_norm": 2.109786033630371, + "learning_rate": 4.559301380991065e-06, + "loss": 2.2181, + "step": 505 + }, + { + "epoch": 0.5545205479452054, + "grad_norm": 2.189779281616211, + "learning_rate": 4.5572705117790414e-06, + "loss": 2.1821, + "step": 506 + }, + { + "epoch": 0.5556164383561644, + "grad_norm": 2.0263590812683105, + "learning_rate": 4.5552396425670196e-06, + "loss": 2.2072, + "step": 507 + }, + { + "epoch": 0.5567123287671233, + "grad_norm": 2.215289354324341, + "learning_rate": 4.553208773354996e-06, + "loss": 2.2513, + "step": 508 + }, + { + "epoch": 0.5578082191780822, + "grad_norm": 2.3031351566314697, + "learning_rate": 4.551177904142974e-06, + "loss": 2.1988, + "step": 509 + }, + { + "epoch": 0.5589041095890411, + "grad_norm": 2.1830058097839355, + "learning_rate": 4.5491470349309505e-06, + "loss": 2.269, + "step": 510 + }, + { + "epoch": 0.5589041095890411, + "eval_loss": 2.342463731765747, + "eval_runtime": 43.1956, + "eval_samples_per_second": 23.151, + "eval_steps_per_second": 0.972, + "step": 510 + }, + { + "epoch": 0.56, + "grad_norm": 2.3210105895996094, + "learning_rate": 4.547116165718929e-06, + "loss": 2.2051, + "step": 511 + }, + { + "epoch": 0.5610958904109589, + "grad_norm": 1.9175361394882202, + "learning_rate": 4.545085296506905e-06, + "loss": 2.2304, + "step": 512 + }, + { + "epoch": 0.5621917808219178, + "grad_norm": 2.302947759628296, + "learning_rate": 4.543054427294882e-06, + "loss": 2.215, + "step": 513 + }, + { + "epoch": 0.5632876712328767, + "grad_norm": 1.8947396278381348, + "learning_rate": 4.54102355808286e-06, + "loss": 2.1926, + "step": 514 + }, + { + "epoch": 0.5643835616438356, + "grad_norm": 2.195268154144287, + "learning_rate": 4.538992688870837e-06, + "loss": 2.1937, + "step": 515 + }, + { + "epoch": 0.5654794520547946, + "grad_norm": 2.178874969482422, + "learning_rate": 4.536961819658814e-06, + "loss": 2.2248, + "step": 516 + }, + { + "epoch": 0.5665753424657535, + "grad_norm": 1.9856929779052734, + "learning_rate": 4.5349309504467915e-06, + "loss": 2.2077, + "step": 517 + }, + { + "epoch": 0.5676712328767123, + "grad_norm": 2.0686943531036377, + "learning_rate": 4.532900081234769e-06, + "loss": 2.2328, + "step": 518 + }, + { + "epoch": 0.5687671232876712, + "grad_norm": 1.9554919004440308, + "learning_rate": 4.530869212022746e-06, + "loss": 2.198, + "step": 519 + }, + { + "epoch": 0.5698630136986301, + "grad_norm": 1.9489670991897583, + "learning_rate": 4.528838342810723e-06, + "loss": 2.2287, + "step": 520 + }, + { + "epoch": 0.5698630136986301, + "eval_loss": 2.3415298461914062, + "eval_runtime": 48.7918, + "eval_samples_per_second": 20.495, + "eval_steps_per_second": 0.861, + "step": 520 + }, + { + "epoch": 0.570958904109589, + "grad_norm": 1.9868658781051636, + "learning_rate": 4.526807473598701e-06, + "loss": 2.2205, + "step": 521 + }, + { + "epoch": 0.572054794520548, + "grad_norm": 2.0965914726257324, + "learning_rate": 4.524776604386678e-06, + "loss": 2.2239, + "step": 522 + }, + { + "epoch": 0.5731506849315069, + "grad_norm": 2.041471481323242, + "learning_rate": 4.522745735174655e-06, + "loss": 2.2035, + "step": 523 + }, + { + "epoch": 0.5742465753424657, + "grad_norm": 1.885293960571289, + "learning_rate": 4.520714865962632e-06, + "loss": 2.2469, + "step": 524 + }, + { + "epoch": 0.5753424657534246, + "grad_norm": 1.9107296466827393, + "learning_rate": 4.51868399675061e-06, + "loss": 2.2386, + "step": 525 + }, + { + "epoch": 0.5764383561643835, + "grad_norm": 1.9945271015167236, + "learning_rate": 4.516653127538587e-06, + "loss": 2.235, + "step": 526 + }, + { + "epoch": 0.5775342465753425, + "grad_norm": 1.9787253141403198, + "learning_rate": 4.514622258326564e-06, + "loss": 2.2034, + "step": 527 + }, + { + "epoch": 0.5786301369863014, + "grad_norm": 2.0300636291503906, + "learning_rate": 4.5125913891145415e-06, + "loss": 2.1849, + "step": 528 + }, + { + "epoch": 0.5797260273972603, + "grad_norm": 2.0348970890045166, + "learning_rate": 4.510560519902519e-06, + "loss": 2.2174, + "step": 529 + }, + { + "epoch": 0.5808219178082191, + "grad_norm": 1.8589537143707275, + "learning_rate": 4.508529650690496e-06, + "loss": 2.2351, + "step": 530 + }, + { + "epoch": 0.5808219178082191, + "eval_loss": 2.3450186252593994, + "eval_runtime": 52.4548, + "eval_samples_per_second": 19.064, + "eval_steps_per_second": 0.801, + "step": 530 + }, + { + "epoch": 0.581917808219178, + "grad_norm": 2.0496020317077637, + "learning_rate": 4.506498781478473e-06, + "loss": 2.182, + "step": 531 + }, + { + "epoch": 0.583013698630137, + "grad_norm": 2.0649805068969727, + "learning_rate": 4.504467912266451e-06, + "loss": 2.1654, + "step": 532 + }, + { + "epoch": 0.5841095890410959, + "grad_norm": 1.9809486865997314, + "learning_rate": 4.502437043054428e-06, + "loss": 2.2182, + "step": 533 + }, + { + "epoch": 0.5852054794520548, + "grad_norm": 2.21915340423584, + "learning_rate": 4.500406173842405e-06, + "loss": 2.2697, + "step": 534 + }, + { + "epoch": 0.5863013698630137, + "grad_norm": 1.9822649955749512, + "learning_rate": 4.4983753046303824e-06, + "loss": 2.2297, + "step": 535 + }, + { + "epoch": 0.5873972602739727, + "grad_norm": 1.972575306892395, + "learning_rate": 4.49634443541836e-06, + "loss": 2.2164, + "step": 536 + }, + { + "epoch": 0.5884931506849315, + "grad_norm": 1.948667049407959, + "learning_rate": 4.494313566206336e-06, + "loss": 2.2087, + "step": 537 + }, + { + "epoch": 0.5895890410958904, + "grad_norm": 2.7486417293548584, + "learning_rate": 4.492282696994314e-06, + "loss": 2.2602, + "step": 538 + }, + { + "epoch": 0.5906849315068493, + "grad_norm": 2.3374552726745605, + "learning_rate": 4.490251827782291e-06, + "loss": 2.2054, + "step": 539 + }, + { + "epoch": 0.5917808219178082, + "grad_norm": 2.222116231918335, + "learning_rate": 4.488220958570269e-06, + "loss": 2.1868, + "step": 540 + }, + { + "epoch": 0.5917808219178082, + "eval_loss": 2.345618963241577, + "eval_runtime": 53.0302, + "eval_samples_per_second": 18.857, + "eval_steps_per_second": 0.792, + "step": 540 + }, + { + "epoch": 0.5928767123287672, + "grad_norm": 2.00752854347229, + "learning_rate": 4.486190089358245e-06, + "loss": 2.187, + "step": 541 + }, + { + "epoch": 0.5939726027397261, + "grad_norm": 2.2466132640838623, + "learning_rate": 4.484159220146223e-06, + "loss": 2.1611, + "step": 542 + }, + { + "epoch": 0.5950684931506849, + "grad_norm": 2.1279385089874268, + "learning_rate": 4.4821283509342e-06, + "loss": 2.248, + "step": 543 + }, + { + "epoch": 0.5961643835616438, + "grad_norm": 1.9140571355819702, + "learning_rate": 4.480097481722178e-06, + "loss": 2.2314, + "step": 544 + }, + { + "epoch": 0.5972602739726027, + "grad_norm": 2.120741605758667, + "learning_rate": 4.478066612510154e-06, + "loss": 2.212, + "step": 545 + }, + { + "epoch": 0.5983561643835617, + "grad_norm": 2.003713369369507, + "learning_rate": 4.4760357432981325e-06, + "loss": 2.1824, + "step": 546 + }, + { + "epoch": 0.5994520547945206, + "grad_norm": 1.9161221981048584, + "learning_rate": 4.474004874086109e-06, + "loss": 2.2015, + "step": 547 + }, + { + "epoch": 0.6005479452054795, + "grad_norm": 2.367110252380371, + "learning_rate": 4.471974004874087e-06, + "loss": 2.3122, + "step": 548 + }, + { + "epoch": 0.6016438356164383, + "grad_norm": 2.113380193710327, + "learning_rate": 4.4699431356620635e-06, + "loss": 2.1692, + "step": 549 + }, + { + "epoch": 0.6027397260273972, + "grad_norm": 1.974181890487671, + "learning_rate": 4.467912266450041e-06, + "loss": 2.2509, + "step": 550 + }, + { + "epoch": 0.6027397260273972, + "eval_loss": 2.342583656311035, + "eval_runtime": 50.7362, + "eval_samples_per_second": 19.71, + "eval_steps_per_second": 0.828, + "step": 550 + }, + { + "epoch": 0.6038356164383561, + "grad_norm": 1.889894962310791, + "learning_rate": 4.465881397238018e-06, + "loss": 2.2214, + "step": 551 + }, + { + "epoch": 0.6049315068493151, + "grad_norm": 1.8982343673706055, + "learning_rate": 4.463850528025995e-06, + "loss": 2.2135, + "step": 552 + }, + { + "epoch": 0.606027397260274, + "grad_norm": 1.9462087154388428, + "learning_rate": 4.4618196588139726e-06, + "loss": 2.2255, + "step": 553 + }, + { + "epoch": 0.6071232876712329, + "grad_norm": 1.8729931116104126, + "learning_rate": 4.45978878960195e-06, + "loss": 2.2358, + "step": 554 + }, + { + "epoch": 0.6082191780821918, + "grad_norm": 1.9230246543884277, + "learning_rate": 4.457757920389927e-06, + "loss": 2.222, + "step": 555 + }, + { + "epoch": 0.6093150684931506, + "grad_norm": 1.9495900869369507, + "learning_rate": 4.455727051177904e-06, + "loss": 2.1779, + "step": 556 + }, + { + "epoch": 0.6104109589041096, + "grad_norm": 1.983487606048584, + "learning_rate": 4.453696181965882e-06, + "loss": 2.2007, + "step": 557 + }, + { + "epoch": 0.6115068493150685, + "grad_norm": 2.1541049480438232, + "learning_rate": 4.451665312753859e-06, + "loss": 2.2513, + "step": 558 + }, + { + "epoch": 0.6126027397260274, + "grad_norm": 1.9133352041244507, + "learning_rate": 4.449634443541836e-06, + "loss": 2.2288, + "step": 559 + }, + { + "epoch": 0.6136986301369863, + "grad_norm": 2.3325741291046143, + "learning_rate": 4.4476035743298135e-06, + "loss": 2.2539, + "step": 560 + }, + { + "epoch": 0.6136986301369863, + "eval_loss": 2.3387348651885986, + "eval_runtime": 53.4301, + "eval_samples_per_second": 18.716, + "eval_steps_per_second": 0.786, + "step": 560 + }, + { + "epoch": 0.6147945205479453, + "grad_norm": 1.8099673986434937, + "learning_rate": 4.445572705117791e-06, + "loss": 2.1845, + "step": 561 + }, + { + "epoch": 0.6158904109589041, + "grad_norm": 2.201901912689209, + "learning_rate": 4.443541835905768e-06, + "loss": 2.195, + "step": 562 + }, + { + "epoch": 0.616986301369863, + "grad_norm": 1.8412714004516602, + "learning_rate": 4.441510966693745e-06, + "loss": 2.2551, + "step": 563 + }, + { + "epoch": 0.6180821917808219, + "grad_norm": 1.8761520385742188, + "learning_rate": 4.439480097481723e-06, + "loss": 2.1965, + "step": 564 + }, + { + "epoch": 0.6191780821917808, + "grad_norm": 1.9123467206954956, + "learning_rate": 4.4374492282697e-06, + "loss": 2.2502, + "step": 565 + }, + { + "epoch": 0.6202739726027398, + "grad_norm": 2.0631234645843506, + "learning_rate": 4.435418359057677e-06, + "loss": 2.1582, + "step": 566 + }, + { + "epoch": 0.6213698630136987, + "grad_norm": 2.3191616535186768, + "learning_rate": 4.433387489845654e-06, + "loss": 2.2271, + "step": 567 + }, + { + "epoch": 0.6224657534246575, + "grad_norm": 2.043057918548584, + "learning_rate": 4.431356620633632e-06, + "loss": 2.2292, + "step": 568 + }, + { + "epoch": 0.6235616438356164, + "grad_norm": 2.0716068744659424, + "learning_rate": 4.429325751421609e-06, + "loss": 2.196, + "step": 569 + }, + { + "epoch": 0.6246575342465753, + "grad_norm": 1.7669776678085327, + "learning_rate": 4.427294882209586e-06, + "loss": 2.1953, + "step": 570 + }, + { + "epoch": 0.6246575342465753, + "eval_loss": 2.336578607559204, + "eval_runtime": 50.0083, + "eval_samples_per_second": 19.997, + "eval_steps_per_second": 0.84, + "step": 570 + }, + { + "epoch": 0.6257534246575343, + "grad_norm": 1.8165771961212158, + "learning_rate": 4.4252640129975635e-06, + "loss": 2.2155, + "step": 571 + }, + { + "epoch": 0.6268493150684932, + "grad_norm": 1.8640249967575073, + "learning_rate": 4.423233143785541e-06, + "loss": 2.1648, + "step": 572 + }, + { + "epoch": 0.6279452054794521, + "grad_norm": 2.013869285583496, + "learning_rate": 4.421202274573518e-06, + "loss": 2.2131, + "step": 573 + }, + { + "epoch": 0.6290410958904109, + "grad_norm": 2.2650580406188965, + "learning_rate": 4.4191714053614945e-06, + "loss": 2.2287, + "step": 574 + }, + { + "epoch": 0.6301369863013698, + "grad_norm": 1.9232103824615479, + "learning_rate": 4.417140536149473e-06, + "loss": 2.1339, + "step": 575 + }, + { + "epoch": 0.6312328767123287, + "grad_norm": 1.910134196281433, + "learning_rate": 4.415109666937449e-06, + "loss": 2.1755, + "step": 576 + }, + { + "epoch": 0.6323287671232877, + "grad_norm": 2.2492005825042725, + "learning_rate": 4.413078797725427e-06, + "loss": 2.2362, + "step": 577 + }, + { + "epoch": 0.6334246575342466, + "grad_norm": 1.8584033250808716, + "learning_rate": 4.411047928513404e-06, + "loss": 2.1911, + "step": 578 + }, + { + "epoch": 0.6345205479452055, + "grad_norm": 2.156273126602173, + "learning_rate": 4.409017059301382e-06, + "loss": 2.2109, + "step": 579 + }, + { + "epoch": 0.6356164383561644, + "grad_norm": 1.7977535724639893, + "learning_rate": 4.406986190089358e-06, + "loss": 2.2195, + "step": 580 + }, + { + "epoch": 0.6356164383561644, + "eval_loss": 2.33555269241333, + "eval_runtime": 48.3919, + "eval_samples_per_second": 20.665, + "eval_steps_per_second": 0.868, + "step": 580 + }, + { + "epoch": 0.6367123287671232, + "grad_norm": 1.8262702226638794, + "learning_rate": 4.404955320877336e-06, + "loss": 2.2345, + "step": 581 + }, + { + "epoch": 0.6378082191780822, + "grad_norm": 1.9389383792877197, + "learning_rate": 4.402924451665313e-06, + "loss": 2.2347, + "step": 582 + }, + { + "epoch": 0.6389041095890411, + "grad_norm": 1.9548988342285156, + "learning_rate": 4.400893582453291e-06, + "loss": 2.1967, + "step": 583 + }, + { + "epoch": 0.64, + "grad_norm": 1.898092269897461, + "learning_rate": 4.398862713241267e-06, + "loss": 2.1795, + "step": 584 + }, + { + "epoch": 0.6410958904109589, + "grad_norm": 1.8183610439300537, + "learning_rate": 4.396831844029245e-06, + "loss": 2.2031, + "step": 585 + }, + { + "epoch": 0.6421917808219179, + "grad_norm": 2.1520016193389893, + "learning_rate": 4.394800974817222e-06, + "loss": 2.2051, + "step": 586 + }, + { + "epoch": 0.6432876712328767, + "grad_norm": 1.9930012226104736, + "learning_rate": 4.392770105605199e-06, + "loss": 2.2053, + "step": 587 + }, + { + "epoch": 0.6443835616438356, + "grad_norm": 2.311450958251953, + "learning_rate": 4.390739236393176e-06, + "loss": 2.2252, + "step": 588 + }, + { + "epoch": 0.6454794520547945, + "grad_norm": 2.0168697834014893, + "learning_rate": 4.388708367181154e-06, + "loss": 2.2286, + "step": 589 + }, + { + "epoch": 0.6465753424657534, + "grad_norm": 2.055964469909668, + "learning_rate": 4.386677497969131e-06, + "loss": 2.2278, + "step": 590 + }, + { + "epoch": 0.6465753424657534, + "eval_loss": 2.334367275238037, + "eval_runtime": 48.9414, + "eval_samples_per_second": 20.433, + "eval_steps_per_second": 0.858, + "step": 590 + }, + { + "epoch": 0.6476712328767124, + "grad_norm": 2.0161590576171875, + "learning_rate": 4.384646628757108e-06, + "loss": 2.2102, + "step": 591 + }, + { + "epoch": 0.6487671232876713, + "grad_norm": 2.0827667713165283, + "learning_rate": 4.3826157595450855e-06, + "loss": 2.2531, + "step": 592 + }, + { + "epoch": 0.6498630136986301, + "grad_norm": 2.077637195587158, + "learning_rate": 4.380584890333063e-06, + "loss": 2.2091, + "step": 593 + }, + { + "epoch": 0.650958904109589, + "grad_norm": 2.023892879486084, + "learning_rate": 4.37855402112104e-06, + "loss": 2.1684, + "step": 594 + }, + { + "epoch": 0.6520547945205479, + "grad_norm": 1.8509857654571533, + "learning_rate": 4.376523151909017e-06, + "loss": 2.1439, + "step": 595 + }, + { + "epoch": 0.6531506849315069, + "grad_norm": 2.1649608612060547, + "learning_rate": 4.3744922826969946e-06, + "loss": 2.2557, + "step": 596 + }, + { + "epoch": 0.6542465753424658, + "grad_norm": 2.1351773738861084, + "learning_rate": 4.372461413484972e-06, + "loss": 2.177, + "step": 597 + }, + { + "epoch": 0.6553424657534247, + "grad_norm": 2.1713974475860596, + "learning_rate": 4.370430544272949e-06, + "loss": 2.2223, + "step": 598 + }, + { + "epoch": 0.6564383561643835, + "grad_norm": 2.2905397415161133, + "learning_rate": 4.368399675060926e-06, + "loss": 2.2373, + "step": 599 + }, + { + "epoch": 0.6575342465753424, + "grad_norm": 2.007160186767578, + "learning_rate": 4.366368805848904e-06, + "loss": 2.2355, + "step": 600 + }, + { + "epoch": 0.6575342465753424, + "eval_loss": 2.3322324752807617, + "eval_runtime": 46.2769, + "eval_samples_per_second": 21.609, + "eval_steps_per_second": 0.908, + "step": 600 + }, + { + "epoch": 0.6586301369863014, + "grad_norm": 2.309783697128296, + "learning_rate": 4.364337936636881e-06, + "loss": 2.2052, + "step": 601 + }, + { + "epoch": 0.6597260273972603, + "grad_norm": 2.498034954071045, + "learning_rate": 4.362307067424858e-06, + "loss": 2.1567, + "step": 602 + }, + { + "epoch": 0.6608219178082192, + "grad_norm": 1.8302545547485352, + "learning_rate": 4.3602761982128355e-06, + "loss": 2.2117, + "step": 603 + }, + { + "epoch": 0.6619178082191781, + "grad_norm": 2.226947546005249, + "learning_rate": 4.358245329000813e-06, + "loss": 2.1881, + "step": 604 + }, + { + "epoch": 0.663013698630137, + "grad_norm": 1.8484126329421997, + "learning_rate": 4.35621445978879e-06, + "loss": 2.1955, + "step": 605 + }, + { + "epoch": 0.6641095890410958, + "grad_norm": 1.9844266176223755, + "learning_rate": 4.354183590576767e-06, + "loss": 2.2063, + "step": 606 + }, + { + "epoch": 0.6652054794520548, + "grad_norm": 1.9292365312576294, + "learning_rate": 4.352152721364745e-06, + "loss": 2.205, + "step": 607 + }, + { + "epoch": 0.6663013698630137, + "grad_norm": 1.980895757675171, + "learning_rate": 4.350121852152722e-06, + "loss": 2.1727, + "step": 608 + }, + { + "epoch": 0.6673972602739726, + "grad_norm": 1.8331713676452637, + "learning_rate": 4.348090982940699e-06, + "loss": 2.226, + "step": 609 + }, + { + "epoch": 0.6684931506849315, + "grad_norm": 1.9666537046432495, + "learning_rate": 4.3460601137286764e-06, + "loss": 2.2198, + "step": 610 + }, + { + "epoch": 0.6684931506849315, + "eval_loss": 2.3297338485717773, + "eval_runtime": 42.6274, + "eval_samples_per_second": 23.459, + "eval_steps_per_second": 0.985, + "step": 610 + }, + { + "epoch": 0.6695890410958905, + "grad_norm": 1.9741110801696777, + "learning_rate": 4.344029244516654e-06, + "loss": 2.2132, + "step": 611 + }, + { + "epoch": 0.6706849315068493, + "grad_norm": 2.0747334957122803, + "learning_rate": 4.341998375304631e-06, + "loss": 2.2081, + "step": 612 + }, + { + "epoch": 0.6717808219178082, + "grad_norm": 2.031503915786743, + "learning_rate": 4.339967506092607e-06, + "loss": 2.2642, + "step": 613 + }, + { + "epoch": 0.6728767123287671, + "grad_norm": 1.8513256311416626, + "learning_rate": 4.3379366368805855e-06, + "loss": 2.2141, + "step": 614 + }, + { + "epoch": 0.673972602739726, + "grad_norm": 1.9098315238952637, + "learning_rate": 4.335905767668562e-06, + "loss": 2.2064, + "step": 615 + }, + { + "epoch": 0.675068493150685, + "grad_norm": 1.9704065322875977, + "learning_rate": 4.33387489845654e-06, + "loss": 2.1997, + "step": 616 + }, + { + "epoch": 0.6761643835616439, + "grad_norm": 2.1520018577575684, + "learning_rate": 4.3318440292445165e-06, + "loss": 2.2278, + "step": 617 + }, + { + "epoch": 0.6772602739726027, + "grad_norm": 1.898164987564087, + "learning_rate": 4.329813160032495e-06, + "loss": 2.2147, + "step": 618 + }, + { + "epoch": 0.6783561643835616, + "grad_norm": 1.8955962657928467, + "learning_rate": 4.327782290820471e-06, + "loss": 2.1825, + "step": 619 + }, + { + "epoch": 0.6794520547945205, + "grad_norm": 1.7579313516616821, + "learning_rate": 4.325751421608449e-06, + "loss": 2.1619, + "step": 620 + }, + { + "epoch": 0.6794520547945205, + "eval_loss": 2.329692840576172, + "eval_runtime": 46.3187, + "eval_samples_per_second": 21.59, + "eval_steps_per_second": 0.907, + "step": 620 + }, + { + "epoch": 0.6805479452054795, + "grad_norm": 1.8716063499450684, + "learning_rate": 4.323720552396426e-06, + "loss": 2.2289, + "step": 621 + }, + { + "epoch": 0.6816438356164384, + "grad_norm": 1.8798290491104126, + "learning_rate": 4.321689683184404e-06, + "loss": 2.1944, + "step": 622 + }, + { + "epoch": 0.6827397260273973, + "grad_norm": 1.8438303470611572, + "learning_rate": 4.31965881397238e-06, + "loss": 2.1879, + "step": 623 + }, + { + "epoch": 0.6838356164383562, + "grad_norm": 1.7844786643981934, + "learning_rate": 4.3176279447603574e-06, + "loss": 2.1936, + "step": 624 + }, + { + "epoch": 0.684931506849315, + "grad_norm": 1.955529808998108, + "learning_rate": 4.315597075548335e-06, + "loss": 2.2258, + "step": 625 + }, + { + "epoch": 0.686027397260274, + "grad_norm": 2.3826634883880615, + "learning_rate": 4.313566206336312e-06, + "loss": 2.2135, + "step": 626 + }, + { + "epoch": 0.6871232876712329, + "grad_norm": 1.8453123569488525, + "learning_rate": 4.311535337124289e-06, + "loss": 2.2362, + "step": 627 + }, + { + "epoch": 0.6882191780821918, + "grad_norm": 2.742682933807373, + "learning_rate": 4.3095044679122665e-06, + "loss": 2.194, + "step": 628 + }, + { + "epoch": 0.6893150684931507, + "grad_norm": 2.0299901962280273, + "learning_rate": 4.307473598700244e-06, + "loss": 2.2029, + "step": 629 + }, + { + "epoch": 0.6904109589041096, + "grad_norm": 2.29683780670166, + "learning_rate": 4.305442729488221e-06, + "loss": 2.2141, + "step": 630 + }, + { + "epoch": 0.6904109589041096, + "eval_loss": 2.329368829727173, + "eval_runtime": 45.1947, + "eval_samples_per_second": 22.126, + "eval_steps_per_second": 0.929, + "step": 630 + }, + { + "epoch": 0.6915068493150685, + "grad_norm": 2.208280086517334, + "learning_rate": 4.303411860276198e-06, + "loss": 2.1637, + "step": 631 + }, + { + "epoch": 0.6926027397260274, + "grad_norm": 1.943514108657837, + "learning_rate": 4.301380991064176e-06, + "loss": 2.234, + "step": 632 + }, + { + "epoch": 0.6936986301369863, + "grad_norm": 2.486440420150757, + "learning_rate": 4.299350121852153e-06, + "loss": 2.2239, + "step": 633 + }, + { + "epoch": 0.6947945205479452, + "grad_norm": 1.7243690490722656, + "learning_rate": 4.29731925264013e-06, + "loss": 2.2428, + "step": 634 + }, + { + "epoch": 0.6958904109589041, + "grad_norm": 2.31227445602417, + "learning_rate": 4.2952883834281075e-06, + "loss": 2.2403, + "step": 635 + }, + { + "epoch": 0.6969863013698631, + "grad_norm": 1.7361263036727905, + "learning_rate": 4.293257514216085e-06, + "loss": 2.1975, + "step": 636 + }, + { + "epoch": 0.6980821917808219, + "grad_norm": 2.010831356048584, + "learning_rate": 4.291226645004062e-06, + "loss": 2.1633, + "step": 637 + }, + { + "epoch": 0.6991780821917808, + "grad_norm": 2.2796833515167236, + "learning_rate": 4.289195775792039e-06, + "loss": 2.2307, + "step": 638 + }, + { + "epoch": 0.7002739726027397, + "grad_norm": 2.1069483757019043, + "learning_rate": 4.287164906580017e-06, + "loss": 2.2389, + "step": 639 + }, + { + "epoch": 0.7013698630136986, + "grad_norm": 2.19460129737854, + "learning_rate": 4.285134037367994e-06, + "loss": 2.1992, + "step": 640 + }, + { + "epoch": 0.7013698630136986, + "eval_loss": 2.3223369121551514, + "eval_runtime": 43.1093, + "eval_samples_per_second": 23.197, + "eval_steps_per_second": 0.974, + "step": 640 + }, + { + "epoch": 0.7024657534246576, + "grad_norm": 1.875261902809143, + "learning_rate": 4.283103168155971e-06, + "loss": 2.1767, + "step": 641 + }, + { + "epoch": 0.7035616438356165, + "grad_norm": 1.906762957572937, + "learning_rate": 4.281072298943948e-06, + "loss": 2.1734, + "step": 642 + }, + { + "epoch": 0.7046575342465753, + "grad_norm": 2.1114180088043213, + "learning_rate": 4.279041429731926e-06, + "loss": 2.1734, + "step": 643 + }, + { + "epoch": 0.7057534246575342, + "grad_norm": 1.844373106956482, + "learning_rate": 4.277010560519903e-06, + "loss": 2.1967, + "step": 644 + }, + { + "epoch": 0.7068493150684931, + "grad_norm": 2.3463754653930664, + "learning_rate": 4.27497969130788e-06, + "loss": 2.1735, + "step": 645 + }, + { + "epoch": 0.707945205479452, + "grad_norm": 2.3530960083007812, + "learning_rate": 4.2729488220958575e-06, + "loss": 2.189, + "step": 646 + }, + { + "epoch": 0.709041095890411, + "grad_norm": 2.110963821411133, + "learning_rate": 4.270917952883835e-06, + "loss": 2.218, + "step": 647 + }, + { + "epoch": 0.7101369863013699, + "grad_norm": 2.625005006790161, + "learning_rate": 4.268887083671812e-06, + "loss": 2.2235, + "step": 648 + }, + { + "epoch": 0.7112328767123288, + "grad_norm": 1.8411654233932495, + "learning_rate": 4.266856214459789e-06, + "loss": 2.1757, + "step": 649 + }, + { + "epoch": 0.7123287671232876, + "grad_norm": 2.0817489624023438, + "learning_rate": 4.264825345247766e-06, + "loss": 2.2698, + "step": 650 + }, + { + "epoch": 0.7123287671232876, + "eval_loss": 2.322544574737549, + "eval_runtime": 43.0771, + "eval_samples_per_second": 23.214, + "eval_steps_per_second": 0.975, + "step": 650 + }, + { + "epoch": 0.7134246575342466, + "grad_norm": 1.8953847885131836, + "learning_rate": 4.262794476035744e-06, + "loss": 2.1745, + "step": 651 + }, + { + "epoch": 0.7145205479452055, + "grad_norm": 1.8463294506072998, + "learning_rate": 4.26076360682372e-06, + "loss": 2.2042, + "step": 652 + }, + { + "epoch": 0.7156164383561644, + "grad_norm": 1.8776047229766846, + "learning_rate": 4.2587327376116984e-06, + "loss": 2.1512, + "step": 653 + }, + { + "epoch": 0.7167123287671233, + "grad_norm": 1.8949921131134033, + "learning_rate": 4.256701868399675e-06, + "loss": 2.1895, + "step": 654 + }, + { + "epoch": 0.7178082191780822, + "grad_norm": 1.8172632455825806, + "learning_rate": 4.254670999187653e-06, + "loss": 2.2, + "step": 655 + }, + { + "epoch": 0.718904109589041, + "grad_norm": 2.105499505996704, + "learning_rate": 4.252640129975629e-06, + "loss": 2.202, + "step": 656 + }, + { + "epoch": 0.72, + "grad_norm": 1.9515924453735352, + "learning_rate": 4.2506092607636075e-06, + "loss": 2.2116, + "step": 657 + }, + { + "epoch": 0.7210958904109589, + "grad_norm": 1.761770248413086, + "learning_rate": 4.248578391551584e-06, + "loss": 2.1921, + "step": 658 + }, + { + "epoch": 0.7221917808219178, + "grad_norm": 1.9848097562789917, + "learning_rate": 4.246547522339562e-06, + "loss": 2.1553, + "step": 659 + }, + { + "epoch": 0.7232876712328767, + "grad_norm": 2.041987180709839, + "learning_rate": 4.2445166531275385e-06, + "loss": 2.1975, + "step": 660 + }, + { + "epoch": 0.7232876712328767, + "eval_loss": 2.325580358505249, + "eval_runtime": 43.6207, + "eval_samples_per_second": 22.925, + "eval_steps_per_second": 0.963, + "step": 660 + }, + { + "epoch": 0.7243835616438357, + "grad_norm": 1.9755775928497314, + "learning_rate": 4.242485783915517e-06, + "loss": 2.1955, + "step": 661 + }, + { + "epoch": 0.7254794520547945, + "grad_norm": 1.9621502161026, + "learning_rate": 4.240454914703493e-06, + "loss": 2.1554, + "step": 662 + }, + { + "epoch": 0.7265753424657534, + "grad_norm": 1.848013162612915, + "learning_rate": 4.23842404549147e-06, + "loss": 2.16, + "step": 663 + }, + { + "epoch": 0.7276712328767123, + "grad_norm": 1.7370128631591797, + "learning_rate": 4.236393176279448e-06, + "loss": 2.1696, + "step": 664 + }, + { + "epoch": 0.7287671232876712, + "grad_norm": 2.0708158016204834, + "learning_rate": 4.234362307067425e-06, + "loss": 2.1592, + "step": 665 + }, + { + "epoch": 0.7298630136986302, + "grad_norm": 1.9864391088485718, + "learning_rate": 4.232331437855402e-06, + "loss": 2.2121, + "step": 666 + }, + { + "epoch": 0.7309589041095891, + "grad_norm": 1.9988250732421875, + "learning_rate": 4.2303005686433795e-06, + "loss": 2.2306, + "step": 667 + }, + { + "epoch": 0.7320547945205479, + "grad_norm": 1.89108407497406, + "learning_rate": 4.228269699431357e-06, + "loss": 2.222, + "step": 668 + }, + { + "epoch": 0.7331506849315068, + "grad_norm": 1.7941503524780273, + "learning_rate": 4.226238830219334e-06, + "loss": 2.1975, + "step": 669 + }, + { + "epoch": 0.7342465753424657, + "grad_norm": 1.8058613538742065, + "learning_rate": 4.224207961007311e-06, + "loss": 2.2037, + "step": 670 + }, + { + "epoch": 0.7342465753424657, + "eval_loss": 2.3266475200653076, + "eval_runtime": 47.6623, + "eval_samples_per_second": 20.981, + "eval_steps_per_second": 0.881, + "step": 670 + }, + { + "epoch": 0.7353424657534247, + "grad_norm": 1.764448642730713, + "learning_rate": 4.2221770917952886e-06, + "loss": 2.1904, + "step": 671 + }, + { + "epoch": 0.7364383561643836, + "grad_norm": 1.8252112865447998, + "learning_rate": 4.220146222583266e-06, + "loss": 2.1959, + "step": 672 + }, + { + "epoch": 0.7375342465753425, + "grad_norm": 2.02898907661438, + "learning_rate": 4.218115353371243e-06, + "loss": 2.1937, + "step": 673 + }, + { + "epoch": 0.7386301369863014, + "grad_norm": 1.8766248226165771, + "learning_rate": 4.21608448415922e-06, + "loss": 2.195, + "step": 674 + }, + { + "epoch": 0.7397260273972602, + "grad_norm": 2.2350332736968994, + "learning_rate": 4.214053614947198e-06, + "loss": 2.1798, + "step": 675 + }, + { + "epoch": 0.7408219178082192, + "grad_norm": 1.820399522781372, + "learning_rate": 4.212022745735175e-06, + "loss": 2.2371, + "step": 676 + }, + { + "epoch": 0.7419178082191781, + "grad_norm": 2.3411312103271484, + "learning_rate": 4.209991876523152e-06, + "loss": 2.1603, + "step": 677 + }, + { + "epoch": 0.743013698630137, + "grad_norm": 1.8562874794006348, + "learning_rate": 4.2079610073111295e-06, + "loss": 2.2082, + "step": 678 + }, + { + "epoch": 0.7441095890410959, + "grad_norm": 2.274214506149292, + "learning_rate": 4.205930138099107e-06, + "loss": 2.1716, + "step": 679 + }, + { + "epoch": 0.7452054794520548, + "grad_norm": 1.9173530340194702, + "learning_rate": 4.203899268887084e-06, + "loss": 2.2461, + "step": 680 + }, + { + "epoch": 0.7452054794520548, + "eval_loss": 2.3253841400146484, + "eval_runtime": 46.5732, + "eval_samples_per_second": 21.472, + "eval_steps_per_second": 0.902, + "step": 680 + }, + { + "epoch": 0.7463013698630137, + "grad_norm": 1.9732033014297485, + "learning_rate": 4.201868399675061e-06, + "loss": 2.2016, + "step": 681 + }, + { + "epoch": 0.7473972602739726, + "grad_norm": 1.8850247859954834, + "learning_rate": 4.199837530463039e-06, + "loss": 2.223, + "step": 682 + }, + { + "epoch": 0.7484931506849315, + "grad_norm": 1.8936313390731812, + "learning_rate": 4.197806661251016e-06, + "loss": 2.1381, + "step": 683 + }, + { + "epoch": 0.7495890410958904, + "grad_norm": 2.11995530128479, + "learning_rate": 4.195775792038993e-06, + "loss": 2.1962, + "step": 684 + }, + { + "epoch": 0.7506849315068493, + "grad_norm": 1.8207820653915405, + "learning_rate": 4.19374492282697e-06, + "loss": 2.1835, + "step": 685 + }, + { + "epoch": 0.7517808219178083, + "grad_norm": 2.06030535697937, + "learning_rate": 4.191714053614948e-06, + "loss": 2.1347, + "step": 686 + }, + { + "epoch": 0.7528767123287671, + "grad_norm": 2.048990488052368, + "learning_rate": 4.189683184402924e-06, + "loss": 2.2189, + "step": 687 + }, + { + "epoch": 0.753972602739726, + "grad_norm": 1.976369023323059, + "learning_rate": 4.187652315190902e-06, + "loss": 2.1805, + "step": 688 + }, + { + "epoch": 0.7550684931506849, + "grad_norm": 2.1160926818847656, + "learning_rate": 4.185621445978879e-06, + "loss": 2.2322, + "step": 689 + }, + { + "epoch": 0.7561643835616438, + "grad_norm": 1.9444516897201538, + "learning_rate": 4.183590576766857e-06, + "loss": 2.1696, + "step": 690 + }, + { + "epoch": 0.7561643835616438, + "eval_loss": 2.3242011070251465, + "eval_runtime": 48.7528, + "eval_samples_per_second": 20.512, + "eval_steps_per_second": 0.861, + "step": 690 + }, + { + "epoch": 0.7572602739726028, + "grad_norm": 1.9567183256149292, + "learning_rate": 4.181559707554833e-06, + "loss": 2.1891, + "step": 691 + }, + { + "epoch": 0.7583561643835617, + "grad_norm": 1.9185020923614502, + "learning_rate": 4.179528838342811e-06, + "loss": 2.2318, + "step": 692 + }, + { + "epoch": 0.7594520547945205, + "grad_norm": 1.833487629890442, + "learning_rate": 4.177497969130788e-06, + "loss": 2.1705, + "step": 693 + }, + { + "epoch": 0.7605479452054794, + "grad_norm": 2.2949867248535156, + "learning_rate": 4.175467099918766e-06, + "loss": 2.1712, + "step": 694 + }, + { + "epoch": 0.7616438356164383, + "grad_norm": 1.8663084506988525, + "learning_rate": 4.173436230706742e-06, + "loss": 2.2262, + "step": 695 + }, + { + "epoch": 0.7627397260273973, + "grad_norm": 2.4633052349090576, + "learning_rate": 4.1714053614947205e-06, + "loss": 2.185, + "step": 696 + }, + { + "epoch": 0.7638356164383562, + "grad_norm": 2.036524772644043, + "learning_rate": 4.169374492282697e-06, + "loss": 2.1697, + "step": 697 + }, + { + "epoch": 0.7649315068493151, + "grad_norm": 2.013810634613037, + "learning_rate": 4.167343623070675e-06, + "loss": 2.1826, + "step": 698 + }, + { + "epoch": 0.766027397260274, + "grad_norm": 2.214109182357788, + "learning_rate": 4.165312753858652e-06, + "loss": 2.2162, + "step": 699 + }, + { + "epoch": 0.7671232876712328, + "grad_norm": 1.8463964462280273, + "learning_rate": 4.163281884646629e-06, + "loss": 2.1865, + "step": 700 + }, + { + "epoch": 0.7671232876712328, + "eval_loss": 2.323822498321533, + "eval_runtime": 48.3307, + "eval_samples_per_second": 20.691, + "eval_steps_per_second": 0.869, + "step": 700 + }, + { + "epoch": 0.7682191780821918, + "grad_norm": 2.321021556854248, + "learning_rate": 4.161251015434607e-06, + "loss": 2.1758, + "step": 701 + }, + { + "epoch": 0.7693150684931507, + "grad_norm": 1.8213255405426025, + "learning_rate": 4.159220146222583e-06, + "loss": 2.2002, + "step": 702 + }, + { + "epoch": 0.7704109589041096, + "grad_norm": 1.8925111293792725, + "learning_rate": 4.157189277010561e-06, + "loss": 2.1575, + "step": 703 + }, + { + "epoch": 0.7715068493150685, + "grad_norm": 1.8050175905227661, + "learning_rate": 4.155158407798538e-06, + "loss": 2.179, + "step": 704 + }, + { + "epoch": 0.7726027397260274, + "grad_norm": 1.873228907585144, + "learning_rate": 4.153127538586516e-06, + "loss": 2.2232, + "step": 705 + }, + { + "epoch": 0.7736986301369863, + "grad_norm": 1.954910397529602, + "learning_rate": 4.151096669374492e-06, + "loss": 2.1992, + "step": 706 + }, + { + "epoch": 0.7747945205479452, + "grad_norm": 1.8913410902023315, + "learning_rate": 4.1490658001624705e-06, + "loss": 2.2086, + "step": 707 + }, + { + "epoch": 0.7758904109589041, + "grad_norm": 1.8883898258209229, + "learning_rate": 4.147034930950447e-06, + "loss": 2.1801, + "step": 708 + }, + { + "epoch": 0.776986301369863, + "grad_norm": 2.054351806640625, + "learning_rate": 4.145004061738425e-06, + "loss": 2.1908, + "step": 709 + }, + { + "epoch": 0.7780821917808219, + "grad_norm": 2.029036045074463, + "learning_rate": 4.1429731925264015e-06, + "loss": 2.1891, + "step": 710 + }, + { + "epoch": 0.7780821917808219, + "eval_loss": 2.3196680545806885, + "eval_runtime": 42.8326, + "eval_samples_per_second": 23.347, + "eval_steps_per_second": 0.981, + "step": 710 + }, + { + "epoch": 0.7791780821917809, + "grad_norm": 2.27848219871521, + "learning_rate": 4.140942323314379e-06, + "loss": 2.1165, + "step": 711 + }, + { + "epoch": 0.7802739726027397, + "grad_norm": 2.134625196456909, + "learning_rate": 4.138911454102356e-06, + "loss": 2.1508, + "step": 712 + }, + { + "epoch": 0.7813698630136986, + "grad_norm": 2.103895902633667, + "learning_rate": 4.136880584890333e-06, + "loss": 2.1746, + "step": 713 + }, + { + "epoch": 0.7824657534246575, + "grad_norm": 1.713189721107483, + "learning_rate": 4.1348497156783106e-06, + "loss": 2.2023, + "step": 714 + }, + { + "epoch": 0.7835616438356164, + "grad_norm": 2.0390899181365967, + "learning_rate": 4.132818846466288e-06, + "loss": 2.2039, + "step": 715 + }, + { + "epoch": 0.7846575342465754, + "grad_norm": 1.9643651247024536, + "learning_rate": 4.130787977254265e-06, + "loss": 2.199, + "step": 716 + }, + { + "epoch": 0.7857534246575343, + "grad_norm": 1.9173564910888672, + "learning_rate": 4.128757108042242e-06, + "loss": 2.2166, + "step": 717 + }, + { + "epoch": 0.7868493150684932, + "grad_norm": 1.7923007011413574, + "learning_rate": 4.12672623883022e-06, + "loss": 2.2144, + "step": 718 + }, + { + "epoch": 0.787945205479452, + "grad_norm": 2.2235090732574463, + "learning_rate": 4.124695369618197e-06, + "loss": 2.2586, + "step": 719 + }, + { + "epoch": 0.7890410958904109, + "grad_norm": 1.780316710472107, + "learning_rate": 4.122664500406174e-06, + "loss": 2.1862, + "step": 720 + }, + { + "epoch": 0.7890410958904109, + "eval_loss": 2.318016767501831, + "eval_runtime": 50.5082, + "eval_samples_per_second": 19.799, + "eval_steps_per_second": 0.832, + "step": 720 + }, + { + "epoch": 0.7901369863013699, + "grad_norm": 2.0788493156433105, + "learning_rate": 4.1206336311941515e-06, + "loss": 2.1845, + "step": 721 + }, + { + "epoch": 0.7912328767123288, + "grad_norm": 1.9053714275360107, + "learning_rate": 4.118602761982129e-06, + "loss": 2.2005, + "step": 722 + }, + { + "epoch": 0.7923287671232877, + "grad_norm": 1.9328513145446777, + "learning_rate": 4.116571892770106e-06, + "loss": 2.1475, + "step": 723 + }, + { + "epoch": 0.7934246575342466, + "grad_norm": 2.059760570526123, + "learning_rate": 4.114541023558083e-06, + "loss": 2.1386, + "step": 724 + }, + { + "epoch": 0.7945205479452054, + "grad_norm": 2.0093464851379395, + "learning_rate": 4.112510154346061e-06, + "loss": 2.2242, + "step": 725 + }, + { + "epoch": 0.7956164383561644, + "grad_norm": 1.8513909578323364, + "learning_rate": 4.110479285134038e-06, + "loss": 2.178, + "step": 726 + }, + { + "epoch": 0.7967123287671233, + "grad_norm": 1.9481958150863647, + "learning_rate": 4.108448415922015e-06, + "loss": 2.1961, + "step": 727 + }, + { + "epoch": 0.7978082191780822, + "grad_norm": 1.9229767322540283, + "learning_rate": 4.1064175467099924e-06, + "loss": 2.2116, + "step": 728 + }, + { + "epoch": 0.7989041095890411, + "grad_norm": 1.9819737672805786, + "learning_rate": 4.10438667749797e-06, + "loss": 2.2097, + "step": 729 + }, + { + "epoch": 0.8, + "grad_norm": 1.9382416009902954, + "learning_rate": 4.102355808285947e-06, + "loss": 2.1099, + "step": 730 + }, + { + "epoch": 0.8, + "eval_loss": 2.31795334815979, + "eval_runtime": 51.8961, + "eval_samples_per_second": 19.269, + "eval_steps_per_second": 0.809, + "step": 730 + }, + { + "epoch": 0.8010958904109589, + "grad_norm": 2.006890058517456, + "learning_rate": 4.100324939073924e-06, + "loss": 2.2606, + "step": 731 + }, + { + "epoch": 0.8021917808219178, + "grad_norm": 2.164473295211792, + "learning_rate": 4.0982940698619015e-06, + "loss": 2.1869, + "step": 732 + }, + { + "epoch": 0.8032876712328767, + "grad_norm": 2.04368257522583, + "learning_rate": 4.096263200649879e-06, + "loss": 2.1846, + "step": 733 + }, + { + "epoch": 0.8043835616438356, + "grad_norm": 1.8363912105560303, + "learning_rate": 4.094232331437856e-06, + "loss": 2.1938, + "step": 734 + }, + { + "epoch": 0.8054794520547945, + "grad_norm": 1.9260189533233643, + "learning_rate": 4.092201462225833e-06, + "loss": 2.1682, + "step": 735 + }, + { + "epoch": 0.8065753424657535, + "grad_norm": 2.0688912868499756, + "learning_rate": 4.090170593013811e-06, + "loss": 2.1772, + "step": 736 + }, + { + "epoch": 0.8076712328767123, + "grad_norm": 1.859474539756775, + "learning_rate": 4.088139723801787e-06, + "loss": 2.2013, + "step": 737 + }, + { + "epoch": 0.8087671232876712, + "grad_norm": 2.0915753841400146, + "learning_rate": 4.086108854589765e-06, + "loss": 2.1519, + "step": 738 + }, + { + "epoch": 0.8098630136986301, + "grad_norm": 2.2019519805908203, + "learning_rate": 4.084077985377742e-06, + "loss": 2.1608, + "step": 739 + }, + { + "epoch": 0.810958904109589, + "grad_norm": 1.8727316856384277, + "learning_rate": 4.08204711616572e-06, + "loss": 2.1823, + "step": 740 + }, + { + "epoch": 0.810958904109589, + "eval_loss": 2.318638324737549, + "eval_runtime": 52.3669, + "eval_samples_per_second": 19.096, + "eval_steps_per_second": 0.802, + "step": 740 + }, + { + "epoch": 0.812054794520548, + "grad_norm": 2.059250831604004, + "learning_rate": 4.080016246953696e-06, + "loss": 2.2285, + "step": 741 + }, + { + "epoch": 0.8131506849315069, + "grad_norm": 1.9466749429702759, + "learning_rate": 4.077985377741674e-06, + "loss": 2.1496, + "step": 742 + }, + { + "epoch": 0.8142465753424658, + "grad_norm": 1.8967524766921997, + "learning_rate": 4.075954508529651e-06, + "loss": 2.2066, + "step": 743 + }, + { + "epoch": 0.8153424657534246, + "grad_norm": 1.976780891418457, + "learning_rate": 4.073923639317629e-06, + "loss": 2.2194, + "step": 744 + }, + { + "epoch": 0.8164383561643835, + "grad_norm": 2.077683687210083, + "learning_rate": 4.071892770105605e-06, + "loss": 2.2209, + "step": 745 + }, + { + "epoch": 0.8175342465753425, + "grad_norm": 1.9075323343276978, + "learning_rate": 4.069861900893583e-06, + "loss": 2.1727, + "step": 746 + }, + { + "epoch": 0.8186301369863014, + "grad_norm": 1.6940832138061523, + "learning_rate": 4.06783103168156e-06, + "loss": 2.1437, + "step": 747 + }, + { + "epoch": 0.8197260273972603, + "grad_norm": 1.9035768508911133, + "learning_rate": 4.065800162469538e-06, + "loss": 2.165, + "step": 748 + }, + { + "epoch": 0.8208219178082192, + "grad_norm": 1.8775745630264282, + "learning_rate": 4.063769293257514e-06, + "loss": 2.1757, + "step": 749 + }, + { + "epoch": 0.821917808219178, + "grad_norm": 1.9622418880462646, + "learning_rate": 4.061738424045492e-06, + "loss": 2.197, + "step": 750 + }, + { + "epoch": 0.821917808219178, + "eval_loss": 2.317519426345825, + "eval_runtime": 50.2699, + "eval_samples_per_second": 19.893, + "eval_steps_per_second": 0.835, + "step": 750 + }, + { + "epoch": 0.823013698630137, + "grad_norm": 1.9646214246749878, + "learning_rate": 4.059707554833469e-06, + "loss": 2.1783, + "step": 751 + }, + { + "epoch": 0.8241095890410959, + "grad_norm": 1.890342354774475, + "learning_rate": 4.057676685621446e-06, + "loss": 2.1925, + "step": 752 + }, + { + "epoch": 0.8252054794520548, + "grad_norm": 1.8977251052856445, + "learning_rate": 4.0556458164094235e-06, + "loss": 2.219, + "step": 753 + }, + { + "epoch": 0.8263013698630137, + "grad_norm": 1.8251665830612183, + "learning_rate": 4.053614947197401e-06, + "loss": 2.189, + "step": 754 + }, + { + "epoch": 0.8273972602739726, + "grad_norm": 1.7953882217407227, + "learning_rate": 4.051584077985378e-06, + "loss": 2.1237, + "step": 755 + }, + { + "epoch": 0.8284931506849315, + "grad_norm": 1.8247836828231812, + "learning_rate": 4.049553208773355e-06, + "loss": 2.1962, + "step": 756 + }, + { + "epoch": 0.8295890410958904, + "grad_norm": 1.8537967205047607, + "learning_rate": 4.047522339561333e-06, + "loss": 2.2201, + "step": 757 + }, + { + "epoch": 0.8306849315068493, + "grad_norm": 1.865315318107605, + "learning_rate": 4.04549147034931e-06, + "loss": 2.1819, + "step": 758 + }, + { + "epoch": 0.8317808219178082, + "grad_norm": 1.8620522022247314, + "learning_rate": 4.043460601137287e-06, + "loss": 2.1717, + "step": 759 + }, + { + "epoch": 0.8328767123287671, + "grad_norm": 1.9475384950637817, + "learning_rate": 4.041429731925264e-06, + "loss": 2.2718, + "step": 760 + }, + { + "epoch": 0.8328767123287671, + "eval_loss": 2.314429759979248, + "eval_runtime": 54.2324, + "eval_samples_per_second": 18.439, + "eval_steps_per_second": 0.774, + "step": 760 + }, + { + "epoch": 0.8339726027397261, + "grad_norm": 1.7170753479003906, + "learning_rate": 4.039398862713242e-06, + "loss": 2.1785, + "step": 761 + }, + { + "epoch": 0.8350684931506849, + "grad_norm": 1.968220591545105, + "learning_rate": 4.037367993501219e-06, + "loss": 2.1798, + "step": 762 + }, + { + "epoch": 0.8361643835616438, + "grad_norm": 1.7927111387252808, + "learning_rate": 4.035337124289196e-06, + "loss": 2.2372, + "step": 763 + }, + { + "epoch": 0.8372602739726027, + "grad_norm": 1.7871185541152954, + "learning_rate": 4.0333062550771735e-06, + "loss": 2.182, + "step": 764 + }, + { + "epoch": 0.8383561643835616, + "grad_norm": 1.9415812492370605, + "learning_rate": 4.031275385865151e-06, + "loss": 2.1527, + "step": 765 + }, + { + "epoch": 0.8394520547945206, + "grad_norm": 1.7327690124511719, + "learning_rate": 4.029244516653128e-06, + "loss": 2.1744, + "step": 766 + }, + { + "epoch": 0.8405479452054795, + "grad_norm": 1.9123411178588867, + "learning_rate": 4.027213647441105e-06, + "loss": 2.2049, + "step": 767 + }, + { + "epoch": 0.8416438356164384, + "grad_norm": 2.011591911315918, + "learning_rate": 4.025182778229083e-06, + "loss": 2.1391, + "step": 768 + }, + { + "epoch": 0.8427397260273972, + "grad_norm": 1.9169542789459229, + "learning_rate": 4.02315190901706e-06, + "loss": 2.1562, + "step": 769 + }, + { + "epoch": 0.8438356164383561, + "grad_norm": 1.8020051717758179, + "learning_rate": 4.021121039805037e-06, + "loss": 2.1795, + "step": 770 + }, + { + "epoch": 0.8438356164383561, + "eval_loss": 2.3199150562286377, + "eval_runtime": 52.468, + "eval_samples_per_second": 19.059, + "eval_steps_per_second": 0.8, + "step": 770 + }, + { + "epoch": 0.8449315068493151, + "grad_norm": 1.927666425704956, + "learning_rate": 4.0190901705930144e-06, + "loss": 2.1834, + "step": 771 + }, + { + "epoch": 0.846027397260274, + "grad_norm": 2.0998382568359375, + "learning_rate": 4.017059301380992e-06, + "loss": 2.2541, + "step": 772 + }, + { + "epoch": 0.8471232876712329, + "grad_norm": 1.88541579246521, + "learning_rate": 4.015028432168969e-06, + "loss": 2.1631, + "step": 773 + }, + { + "epoch": 0.8482191780821918, + "grad_norm": 1.9546253681182861, + "learning_rate": 4.012997562956945e-06, + "loss": 2.2266, + "step": 774 + }, + { + "epoch": 0.8493150684931506, + "grad_norm": 1.7969225645065308, + "learning_rate": 4.0109666937449235e-06, + "loss": 2.1526, + "step": 775 + }, + { + "epoch": 0.8504109589041096, + "grad_norm": 1.7237955331802368, + "learning_rate": 4.0089358245329e-06, + "loss": 2.1618, + "step": 776 + }, + { + "epoch": 0.8515068493150685, + "grad_norm": 1.8355857133865356, + "learning_rate": 4.006904955320878e-06, + "loss": 2.1673, + "step": 777 + }, + { + "epoch": 0.8526027397260274, + "grad_norm": 1.8049591779708862, + "learning_rate": 4.0048740861088545e-06, + "loss": 2.1676, + "step": 778 + }, + { + "epoch": 0.8536986301369863, + "grad_norm": 1.8143601417541504, + "learning_rate": 4.002843216896833e-06, + "loss": 2.214, + "step": 779 + }, + { + "epoch": 0.8547945205479452, + "grad_norm": 1.8344221115112305, + "learning_rate": 4.000812347684809e-06, + "loss": 2.1958, + "step": 780 + }, + { + "epoch": 0.8547945205479452, + "eval_loss": 2.314359426498413, + "eval_runtime": 51.8717, + "eval_samples_per_second": 19.278, + "eval_steps_per_second": 0.81, + "step": 780 + }, + { + "epoch": 0.8558904109589041, + "grad_norm": 1.944402813911438, + "learning_rate": 3.998781478472787e-06, + "loss": 2.1497, + "step": 781 + }, + { + "epoch": 0.856986301369863, + "grad_norm": 1.93898606300354, + "learning_rate": 3.996750609260764e-06, + "loss": 2.1737, + "step": 782 + }, + { + "epoch": 0.8580821917808219, + "grad_norm": 1.9499506950378418, + "learning_rate": 3.994719740048742e-06, + "loss": 2.1545, + "step": 783 + }, + { + "epoch": 0.8591780821917808, + "grad_norm": 1.9311052560806274, + "learning_rate": 3.992688870836718e-06, + "loss": 2.2559, + "step": 784 + }, + { + "epoch": 0.8602739726027397, + "grad_norm": 1.864862084388733, + "learning_rate": 3.990658001624696e-06, + "loss": 2.1996, + "step": 785 + }, + { + "epoch": 0.8613698630136987, + "grad_norm": 1.8248569965362549, + "learning_rate": 3.988627132412673e-06, + "loss": 2.1591, + "step": 786 + }, + { + "epoch": 0.8624657534246576, + "grad_norm": 2.2370970249176025, + "learning_rate": 3.98659626320065e-06, + "loss": 2.2095, + "step": 787 + }, + { + "epoch": 0.8635616438356164, + "grad_norm": 1.8877084255218506, + "learning_rate": 3.984565393988627e-06, + "loss": 2.1813, + "step": 788 + }, + { + "epoch": 0.8646575342465753, + "grad_norm": 2.1722517013549805, + "learning_rate": 3.9825345247766046e-06, + "loss": 2.1866, + "step": 789 + }, + { + "epoch": 0.8657534246575342, + "grad_norm": 1.9640655517578125, + "learning_rate": 3.980503655564582e-06, + "loss": 2.1818, + "step": 790 + }, + { + "epoch": 0.8657534246575342, + "eval_loss": 2.314087390899658, + "eval_runtime": 51.9778, + "eval_samples_per_second": 19.239, + "eval_steps_per_second": 0.808, + "step": 790 + }, + { + "epoch": 0.8668493150684932, + "grad_norm": 2.0583789348602295, + "learning_rate": 3.978472786352559e-06, + "loss": 2.1956, + "step": 791 + }, + { + "epoch": 0.8679452054794521, + "grad_norm": 2.2128682136535645, + "learning_rate": 3.976441917140536e-06, + "loss": 2.204, + "step": 792 + }, + { + "epoch": 0.869041095890411, + "grad_norm": 1.917763113975525, + "learning_rate": 3.974411047928514e-06, + "loss": 2.1841, + "step": 793 + }, + { + "epoch": 0.8701369863013698, + "grad_norm": 2.0327019691467285, + "learning_rate": 3.972380178716491e-06, + "loss": 2.2351, + "step": 794 + }, + { + "epoch": 0.8712328767123287, + "grad_norm": 1.9498991966247559, + "learning_rate": 3.970349309504468e-06, + "loss": 2.2074, + "step": 795 + }, + { + "epoch": 0.8723287671232877, + "grad_norm": 1.8105775117874146, + "learning_rate": 3.9683184402924455e-06, + "loss": 2.2088, + "step": 796 + }, + { + "epoch": 0.8734246575342466, + "grad_norm": 1.9299930334091187, + "learning_rate": 3.966287571080423e-06, + "loss": 2.1469, + "step": 797 + }, + { + "epoch": 0.8745205479452055, + "grad_norm": 1.9346121549606323, + "learning_rate": 3.9642567018684e-06, + "loss": 2.1434, + "step": 798 + }, + { + "epoch": 0.8756164383561644, + "grad_norm": 1.8202977180480957, + "learning_rate": 3.962225832656377e-06, + "loss": 2.194, + "step": 799 + }, + { + "epoch": 0.8767123287671232, + "grad_norm": 1.6922450065612793, + "learning_rate": 3.960194963444355e-06, + "loss": 2.1821, + "step": 800 + }, + { + "epoch": 0.8767123287671232, + "eval_loss": 2.310209035873413, + "eval_runtime": 46.1102, + "eval_samples_per_second": 21.687, + "eval_steps_per_second": 0.911, + "step": 800 + }, + { + "epoch": 0.8778082191780822, + "grad_norm": 1.9258904457092285, + "learning_rate": 3.958164094232332e-06, + "loss": 2.1326, + "step": 801 + }, + { + "epoch": 0.8789041095890411, + "grad_norm": 1.8485561609268188, + "learning_rate": 3.956133225020309e-06, + "loss": 2.1888, + "step": 802 + }, + { + "epoch": 0.88, + "grad_norm": 1.9802418947219849, + "learning_rate": 3.954102355808286e-06, + "loss": 2.1876, + "step": 803 + }, + { + "epoch": 0.8810958904109589, + "grad_norm": 1.8365259170532227, + "learning_rate": 3.952071486596264e-06, + "loss": 2.1804, + "step": 804 + }, + { + "epoch": 0.8821917808219178, + "grad_norm": 2.023909568786621, + "learning_rate": 3.950040617384241e-06, + "loss": 2.1801, + "step": 805 + }, + { + "epoch": 0.8832876712328767, + "grad_norm": 1.8889018297195435, + "learning_rate": 3.948009748172218e-06, + "loss": 2.1657, + "step": 806 + }, + { + "epoch": 0.8843835616438356, + "grad_norm": 1.7983578443527222, + "learning_rate": 3.9459788789601955e-06, + "loss": 2.1589, + "step": 807 + }, + { + "epoch": 0.8854794520547945, + "grad_norm": 2.157043218612671, + "learning_rate": 3.943948009748173e-06, + "loss": 2.2258, + "step": 808 + }, + { + "epoch": 0.8865753424657534, + "grad_norm": 1.7387373447418213, + "learning_rate": 3.94191714053615e-06, + "loss": 2.1878, + "step": 809 + }, + { + "epoch": 0.8876712328767123, + "grad_norm": 1.963233232498169, + "learning_rate": 3.939886271324127e-06, + "loss": 2.186, + "step": 810 + }, + { + "epoch": 0.8876712328767123, + "eval_loss": 2.306187152862549, + "eval_runtime": 42.563, + "eval_samples_per_second": 23.495, + "eval_steps_per_second": 0.987, + "step": 810 + }, + { + "epoch": 0.8887671232876713, + "grad_norm": 1.8932961225509644, + "learning_rate": 3.937855402112104e-06, + "loss": 2.198, + "step": 811 + }, + { + "epoch": 0.8898630136986302, + "grad_norm": 2.0431783199310303, + "learning_rate": 3.935824532900082e-06, + "loss": 2.1505, + "step": 812 + }, + { + "epoch": 0.890958904109589, + "grad_norm": 1.8954466581344604, + "learning_rate": 3.933793663688058e-06, + "loss": 2.1494, + "step": 813 + }, + { + "epoch": 0.8920547945205479, + "grad_norm": 1.841184139251709, + "learning_rate": 3.9317627944760365e-06, + "loss": 2.1932, + "step": 814 + }, + { + "epoch": 0.8931506849315068, + "grad_norm": 1.9083219766616821, + "learning_rate": 3.929731925264013e-06, + "loss": 2.2081, + "step": 815 + }, + { + "epoch": 0.8942465753424658, + "grad_norm": 1.9737226963043213, + "learning_rate": 3.927701056051991e-06, + "loss": 2.1405, + "step": 816 + }, + { + "epoch": 0.8953424657534247, + "grad_norm": 1.869194746017456, + "learning_rate": 3.9256701868399674e-06, + "loss": 2.1642, + "step": 817 + }, + { + "epoch": 0.8964383561643836, + "grad_norm": 1.9553205966949463, + "learning_rate": 3.9236393176279456e-06, + "loss": 2.1761, + "step": 818 + }, + { + "epoch": 0.8975342465753424, + "grad_norm": 1.8355534076690674, + "learning_rate": 3.921608448415922e-06, + "loss": 2.1551, + "step": 819 + }, + { + "epoch": 0.8986301369863013, + "grad_norm": 1.9770171642303467, + "learning_rate": 3.9195775792039e-06, + "loss": 2.1424, + "step": 820 + }, + { + "epoch": 0.8986301369863013, + "eval_loss": 2.307065010070801, + "eval_runtime": 44.5816, + "eval_samples_per_second": 22.431, + "eval_steps_per_second": 0.942, + "step": 820 + }, + { + "epoch": 0.8997260273972603, + "grad_norm": 1.7375415563583374, + "learning_rate": 3.9175467099918765e-06, + "loss": 2.1895, + "step": 821 + }, + { + "epoch": 0.9008219178082192, + "grad_norm": 1.8910484313964844, + "learning_rate": 3.915515840779855e-06, + "loss": 2.1685, + "step": 822 + }, + { + "epoch": 0.9019178082191781, + "grad_norm": 1.9191482067108154, + "learning_rate": 3.913484971567831e-06, + "loss": 2.2074, + "step": 823 + }, + { + "epoch": 0.903013698630137, + "grad_norm": 1.7864760160446167, + "learning_rate": 3.911454102355808e-06, + "loss": 2.1972, + "step": 824 + }, + { + "epoch": 0.9041095890410958, + "grad_norm": 1.9044650793075562, + "learning_rate": 3.909423233143786e-06, + "loss": 2.1237, + "step": 825 + }, + { + "epoch": 0.9052054794520548, + "grad_norm": 1.8378455638885498, + "learning_rate": 3.907392363931763e-06, + "loss": 2.1907, + "step": 826 + }, + { + "epoch": 0.9063013698630137, + "grad_norm": 1.8708094358444214, + "learning_rate": 3.90536149471974e-06, + "loss": 2.1883, + "step": 827 + }, + { + "epoch": 0.9073972602739726, + "grad_norm": 1.7802201509475708, + "learning_rate": 3.9033306255077175e-06, + "loss": 2.2356, + "step": 828 + }, + { + "epoch": 0.9084931506849315, + "grad_norm": 1.9374496936798096, + "learning_rate": 3.901299756295695e-06, + "loss": 2.2546, + "step": 829 + }, + { + "epoch": 0.9095890410958904, + "grad_norm": 1.9246270656585693, + "learning_rate": 3.899268887083672e-06, + "loss": 2.2684, + "step": 830 + }, + { + "epoch": 0.9095890410958904, + "eval_loss": 2.3065874576568604, + "eval_runtime": 45.18, + "eval_samples_per_second": 22.134, + "eval_steps_per_second": 0.93, + "step": 830 + }, + { + "epoch": 0.9106849315068493, + "grad_norm": 1.8681883811950684, + "learning_rate": 3.897238017871649e-06, + "loss": 2.1737, + "step": 831 + }, + { + "epoch": 0.9117808219178082, + "grad_norm": 1.9293930530548096, + "learning_rate": 3.8952071486596266e-06, + "loss": 2.1813, + "step": 832 + }, + { + "epoch": 0.9128767123287671, + "grad_norm": 1.844586968421936, + "learning_rate": 3.893176279447604e-06, + "loss": 2.1788, + "step": 833 + }, + { + "epoch": 0.913972602739726, + "grad_norm": 1.7658660411834717, + "learning_rate": 3.891145410235581e-06, + "loss": 2.1494, + "step": 834 + }, + { + "epoch": 0.915068493150685, + "grad_norm": 2.05153751373291, + "learning_rate": 3.889114541023558e-06, + "loss": 2.241, + "step": 835 + }, + { + "epoch": 0.9161643835616439, + "grad_norm": 1.7869952917099, + "learning_rate": 3.887083671811536e-06, + "loss": 2.1866, + "step": 836 + }, + { + "epoch": 0.9172602739726028, + "grad_norm": 1.99534273147583, + "learning_rate": 3.885052802599513e-06, + "loss": 2.1682, + "step": 837 + }, + { + "epoch": 0.9183561643835616, + "grad_norm": 1.9593979120254517, + "learning_rate": 3.88302193338749e-06, + "loss": 2.1357, + "step": 838 + }, + { + "epoch": 0.9194520547945205, + "grad_norm": 1.8441778421401978, + "learning_rate": 3.8809910641754675e-06, + "loss": 2.2283, + "step": 839 + }, + { + "epoch": 0.9205479452054794, + "grad_norm": 1.9153512716293335, + "learning_rate": 3.878960194963445e-06, + "loss": 2.1899, + "step": 840 + }, + { + "epoch": 0.9205479452054794, + "eval_loss": 2.307375907897949, + "eval_runtime": 48.0216, + "eval_samples_per_second": 20.824, + "eval_steps_per_second": 0.875, + "step": 840 + }, + { + "epoch": 0.9216438356164384, + "grad_norm": 2.004232883453369, + "learning_rate": 3.876929325751422e-06, + "loss": 2.173, + "step": 841 + }, + { + "epoch": 0.9227397260273973, + "grad_norm": 2.12992262840271, + "learning_rate": 3.874898456539399e-06, + "loss": 2.2164, + "step": 842 + }, + { + "epoch": 0.9238356164383562, + "grad_norm": 2.0505411624908447, + "learning_rate": 3.872867587327377e-06, + "loss": 2.1432, + "step": 843 + }, + { + "epoch": 0.924931506849315, + "grad_norm": 1.767324686050415, + "learning_rate": 3.870836718115354e-06, + "loss": 2.1838, + "step": 844 + }, + { + "epoch": 0.9260273972602739, + "grad_norm": 2.189107894897461, + "learning_rate": 3.868805848903331e-06, + "loss": 2.1623, + "step": 845 + }, + { + "epoch": 0.9271232876712329, + "grad_norm": 1.9781970977783203, + "learning_rate": 3.8667749796913084e-06, + "loss": 2.1738, + "step": 846 + }, + { + "epoch": 0.9282191780821918, + "grad_norm": 1.8734674453735352, + "learning_rate": 3.864744110479286e-06, + "loss": 2.1829, + "step": 847 + }, + { + "epoch": 0.9293150684931507, + "grad_norm": 1.7699538469314575, + "learning_rate": 3.862713241267262e-06, + "loss": 2.186, + "step": 848 + }, + { + "epoch": 0.9304109589041096, + "grad_norm": 1.8354437351226807, + "learning_rate": 3.86068237205524e-06, + "loss": 2.2149, + "step": 849 + }, + { + "epoch": 0.9315068493150684, + "grad_norm": 1.913249135017395, + "learning_rate": 3.858651502843217e-06, + "loss": 2.2208, + "step": 850 + }, + { + "epoch": 0.9315068493150684, + "eval_loss": 2.307549238204956, + "eval_runtime": 43.1706, + "eval_samples_per_second": 23.164, + "eval_steps_per_second": 0.973, + "step": 850 + }, + { + "epoch": 0.9326027397260274, + "grad_norm": 1.92095148563385, + "learning_rate": 3.856620633631195e-06, + "loss": 2.1839, + "step": 851 + }, + { + "epoch": 0.9336986301369863, + "grad_norm": 1.8833959102630615, + "learning_rate": 3.854589764419171e-06, + "loss": 2.1372, + "step": 852 + }, + { + "epoch": 0.9347945205479452, + "grad_norm": 1.9410737752914429, + "learning_rate": 3.852558895207149e-06, + "loss": 2.1726, + "step": 853 + }, + { + "epoch": 0.9358904109589041, + "grad_norm": 1.8328828811645508, + "learning_rate": 3.850528025995126e-06, + "loss": 2.2425, + "step": 854 + }, + { + "epoch": 0.936986301369863, + "grad_norm": 1.89652419090271, + "learning_rate": 3.848497156783104e-06, + "loss": 2.1956, + "step": 855 + }, + { + "epoch": 0.938082191780822, + "grad_norm": 1.9026265144348145, + "learning_rate": 3.84646628757108e-06, + "loss": 2.167, + "step": 856 + }, + { + "epoch": 0.9391780821917808, + "grad_norm": 2.010864734649658, + "learning_rate": 3.8444354183590585e-06, + "loss": 2.1725, + "step": 857 + }, + { + "epoch": 0.9402739726027397, + "grad_norm": 2.293717384338379, + "learning_rate": 3.842404549147035e-06, + "loss": 2.1668, + "step": 858 + }, + { + "epoch": 0.9413698630136986, + "grad_norm": 1.8209116458892822, + "learning_rate": 3.840373679935013e-06, + "loss": 2.1649, + "step": 859 + }, + { + "epoch": 0.9424657534246575, + "grad_norm": 2.061277389526367, + "learning_rate": 3.8383428107229894e-06, + "loss": 2.1479, + "step": 860 + }, + { + "epoch": 0.9424657534246575, + "eval_loss": 2.3073670864105225, + "eval_runtime": 44.3781, + "eval_samples_per_second": 22.534, + "eval_steps_per_second": 0.946, + "step": 860 + }, + { + "epoch": 0.9435616438356165, + "grad_norm": 1.8032523393630981, + "learning_rate": 3.836311941510967e-06, + "loss": 2.1665, + "step": 861 + }, + { + "epoch": 0.9446575342465754, + "grad_norm": 1.989533543586731, + "learning_rate": 3.834281072298944e-06, + "loss": 2.1612, + "step": 862 + }, + { + "epoch": 0.9457534246575342, + "grad_norm": 2.017179012298584, + "learning_rate": 3.832250203086921e-06, + "loss": 2.201, + "step": 863 + }, + { + "epoch": 0.9468493150684931, + "grad_norm": 2.0291004180908203, + "learning_rate": 3.8302193338748985e-06, + "loss": 2.1508, + "step": 864 + }, + { + "epoch": 0.947945205479452, + "grad_norm": 1.8693255186080933, + "learning_rate": 3.828188464662876e-06, + "loss": 2.2106, + "step": 865 + }, + { + "epoch": 0.949041095890411, + "grad_norm": 1.8890169858932495, + "learning_rate": 3.826157595450853e-06, + "loss": 2.2467, + "step": 866 + }, + { + "epoch": 0.9501369863013699, + "grad_norm": 2.0550642013549805, + "learning_rate": 3.82412672623883e-06, + "loss": 2.1899, + "step": 867 + }, + { + "epoch": 0.9512328767123288, + "grad_norm": 1.8681174516677856, + "learning_rate": 3.822095857026808e-06, + "loss": 2.1428, + "step": 868 + }, + { + "epoch": 0.9523287671232876, + "grad_norm": 1.9133951663970947, + "learning_rate": 3.820064987814785e-06, + "loss": 2.1947, + "step": 869 + }, + { + "epoch": 0.9534246575342465, + "grad_norm": 1.8717621564865112, + "learning_rate": 3.818034118602762e-06, + "loss": 2.1542, + "step": 870 + }, + { + "epoch": 0.9534246575342465, + "eval_loss": 2.307584047317505, + "eval_runtime": 43.718, + "eval_samples_per_second": 22.874, + "eval_steps_per_second": 0.961, + "step": 870 + }, + { + "epoch": 0.9545205479452055, + "grad_norm": 1.9843461513519287, + "learning_rate": 3.8160032493907395e-06, + "loss": 2.1547, + "step": 871 + }, + { + "epoch": 0.9556164383561644, + "grad_norm": 1.725468397140503, + "learning_rate": 3.813972380178717e-06, + "loss": 2.1679, + "step": 872 + }, + { + "epoch": 0.9567123287671233, + "grad_norm": 1.8076086044311523, + "learning_rate": 3.811941510966694e-06, + "loss": 2.1654, + "step": 873 + }, + { + "epoch": 0.9578082191780822, + "grad_norm": 1.8879774808883667, + "learning_rate": 3.809910641754671e-06, + "loss": 2.191, + "step": 874 + }, + { + "epoch": 0.958904109589041, + "grad_norm": 1.943313717842102, + "learning_rate": 3.8078797725426486e-06, + "loss": 2.1448, + "step": 875 + }, + { + "epoch": 0.96, + "grad_norm": 1.8877806663513184, + "learning_rate": 3.8058489033306254e-06, + "loss": 2.1129, + "step": 876 + }, + { + "epoch": 0.9610958904109589, + "grad_norm": 1.787083625793457, + "learning_rate": 3.803818034118603e-06, + "loss": 2.1854, + "step": 877 + }, + { + "epoch": 0.9621917808219178, + "grad_norm": 2.0925140380859375, + "learning_rate": 3.80178716490658e-06, + "loss": 2.1614, + "step": 878 + }, + { + "epoch": 0.9632876712328767, + "grad_norm": 1.9043725728988647, + "learning_rate": 3.7997562956945577e-06, + "loss": 2.2161, + "step": 879 + }, + { + "epoch": 0.9643835616438357, + "grad_norm": 1.8173545598983765, + "learning_rate": 3.7977254264825345e-06, + "loss": 2.2067, + "step": 880 + }, + { + "epoch": 0.9643835616438357, + "eval_loss": 2.302727460861206, + "eval_runtime": 44.2015, + "eval_samples_per_second": 22.624, + "eval_steps_per_second": 0.95, + "step": 880 + }, + { + "epoch": 0.9654794520547946, + "grad_norm": 1.9871562719345093, + "learning_rate": 3.7956945572705122e-06, + "loss": 2.1658, + "step": 881 + }, + { + "epoch": 0.9665753424657534, + "grad_norm": 2.103553295135498, + "learning_rate": 3.793663688058489e-06, + "loss": 2.1468, + "step": 882 + }, + { + "epoch": 0.9676712328767123, + "grad_norm": 1.9089381694793701, + "learning_rate": 3.791632818846467e-06, + "loss": 2.2289, + "step": 883 + }, + { + "epoch": 0.9687671232876712, + "grad_norm": 2.3238179683685303, + "learning_rate": 3.7896019496344436e-06, + "loss": 2.2179, + "step": 884 + }, + { + "epoch": 0.9698630136986301, + "grad_norm": 1.902161955833435, + "learning_rate": 3.787571080422421e-06, + "loss": 2.2086, + "step": 885 + }, + { + "epoch": 0.9709589041095891, + "grad_norm": 1.8587497472763062, + "learning_rate": 3.785540211210398e-06, + "loss": 2.2175, + "step": 886 + }, + { + "epoch": 0.972054794520548, + "grad_norm": 2.029719829559326, + "learning_rate": 3.7835093419983755e-06, + "loss": 2.207, + "step": 887 + }, + { + "epoch": 0.9731506849315068, + "grad_norm": 2.0228402614593506, + "learning_rate": 3.7814784727863527e-06, + "loss": 2.1851, + "step": 888 + }, + { + "epoch": 0.9742465753424657, + "grad_norm": 1.9885436296463013, + "learning_rate": 3.77944760357433e-06, + "loss": 2.1667, + "step": 889 + }, + { + "epoch": 0.9753424657534246, + "grad_norm": 1.8621459007263184, + "learning_rate": 3.7774167343623073e-06, + "loss": 2.2529, + "step": 890 + }, + { + "epoch": 0.9753424657534246, + "eval_loss": 2.303697347640991, + "eval_runtime": 44.9459, + "eval_samples_per_second": 22.249, + "eval_steps_per_second": 0.934, + "step": 890 + }, + { + "epoch": 0.9764383561643836, + "grad_norm": 2.0311460494995117, + "learning_rate": 3.7753858651502846e-06, + "loss": 2.2185, + "step": 891 + }, + { + "epoch": 0.9775342465753425, + "grad_norm": 2.2017149925231934, + "learning_rate": 3.773354995938262e-06, + "loss": 2.164, + "step": 892 + }, + { + "epoch": 0.9786301369863014, + "grad_norm": 1.949684977531433, + "learning_rate": 3.771324126726239e-06, + "loss": 2.2128, + "step": 893 + }, + { + "epoch": 0.9797260273972602, + "grad_norm": 1.9395498037338257, + "learning_rate": 3.7692932575142164e-06, + "loss": 2.1957, + "step": 894 + }, + { + "epoch": 0.9808219178082191, + "grad_norm": 2.096432685852051, + "learning_rate": 3.7672623883021937e-06, + "loss": 2.219, + "step": 895 + }, + { + "epoch": 0.9819178082191781, + "grad_norm": 2.0942952632904053, + "learning_rate": 3.765231519090171e-06, + "loss": 2.1546, + "step": 896 + }, + { + "epoch": 0.983013698630137, + "grad_norm": 2.0097291469573975, + "learning_rate": 3.7632006498781482e-06, + "loss": 2.1021, + "step": 897 + }, + { + "epoch": 0.9841095890410959, + "grad_norm": 2.0909335613250732, + "learning_rate": 3.761169780666125e-06, + "loss": 2.1862, + "step": 898 + }, + { + "epoch": 0.9852054794520548, + "grad_norm": 2.187309980392456, + "learning_rate": 3.7591389114541028e-06, + "loss": 2.1511, + "step": 899 + }, + { + "epoch": 0.9863013698630136, + "grad_norm": 2.194276809692383, + "learning_rate": 3.7571080422420796e-06, + "loss": 2.196, + "step": 900 + }, + { + "epoch": 0.9863013698630136, + "eval_loss": 2.3047611713409424, + "eval_runtime": 45.609, + "eval_samples_per_second": 21.925, + "eval_steps_per_second": 0.921, + "step": 900 + }, + { + "epoch": 0.9873972602739726, + "grad_norm": 2.1132829189300537, + "learning_rate": 3.7550771730300573e-06, + "loss": 2.2046, + "step": 901 + }, + { + "epoch": 0.9884931506849315, + "grad_norm": 2.1946468353271484, + "learning_rate": 3.753046303818034e-06, + "loss": 2.1656, + "step": 902 + }, + { + "epoch": 0.9895890410958904, + "grad_norm": 2.2210536003112793, + "learning_rate": 3.751015434606012e-06, + "loss": 2.2186, + "step": 903 + }, + { + "epoch": 0.9906849315068493, + "grad_norm": 1.913695216178894, + "learning_rate": 3.7489845653939887e-06, + "loss": 2.1933, + "step": 904 + }, + { + "epoch": 0.9917808219178083, + "grad_norm": 2.1088368892669678, + "learning_rate": 3.7469536961819664e-06, + "loss": 2.2406, + "step": 905 + }, + { + "epoch": 0.9928767123287672, + "grad_norm": 2.04239821434021, + "learning_rate": 3.7449228269699433e-06, + "loss": 2.142, + "step": 906 + }, + { + "epoch": 0.993972602739726, + "grad_norm": 1.7852507829666138, + "learning_rate": 3.742891957757921e-06, + "loss": 2.1701, + "step": 907 + }, + { + "epoch": 0.9950684931506849, + "grad_norm": 2.0275776386260986, + "learning_rate": 3.740861088545898e-06, + "loss": 2.2004, + "step": 908 + }, + { + "epoch": 0.9961643835616438, + "grad_norm": 1.9028904438018799, + "learning_rate": 3.7388302193338755e-06, + "loss": 2.177, + "step": 909 + }, + { + "epoch": 0.9972602739726028, + "grad_norm": 1.7815115451812744, + "learning_rate": 3.7367993501218524e-06, + "loss": 2.207, + "step": 910 + }, + { + "epoch": 0.9972602739726028, + "eval_loss": 2.302318811416626, + "eval_runtime": 42.9296, + "eval_samples_per_second": 23.294, + "eval_steps_per_second": 0.978, + "step": 910 + }, + { + "epoch": 0.9983561643835617, + "grad_norm": 1.913689374923706, + "learning_rate": 3.7347684809098297e-06, + "loss": 2.1594, + "step": 911 + }, + { + "epoch": 0.9994520547945206, + "grad_norm": 2.171222686767578, + "learning_rate": 3.732737611697807e-06, + "loss": 2.1689, + "step": 912 + }, + { + "epoch": 1.0005479452054795, + "grad_norm": 1.831642985343933, + "learning_rate": 3.7307067424857842e-06, + "loss": 2.1367, + "step": 913 + }, + { + "epoch": 1.0016438356164383, + "grad_norm": 2.1831955909729004, + "learning_rate": 3.7286758732737615e-06, + "loss": 2.1035, + "step": 914 + }, + { + "epoch": 1.0027397260273974, + "grad_norm": 2.038025379180908, + "learning_rate": 3.7266450040617388e-06, + "loss": 2.1461, + "step": 915 + }, + { + "epoch": 1.0038356164383562, + "grad_norm": 1.866502046585083, + "learning_rate": 3.724614134849716e-06, + "loss": 2.1322, + "step": 916 + }, + { + "epoch": 1.004931506849315, + "grad_norm": 2.1359357833862305, + "learning_rate": 3.7225832656376933e-06, + "loss": 2.2333, + "step": 917 + }, + { + "epoch": 1.006027397260274, + "grad_norm": 1.8689136505126953, + "learning_rate": 3.7205523964256706e-06, + "loss": 2.1176, + "step": 918 + }, + { + "epoch": 1.0071232876712328, + "grad_norm": 1.8869425058364868, + "learning_rate": 3.718521527213648e-06, + "loss": 2.2127, + "step": 919 + }, + { + "epoch": 1.0082191780821919, + "grad_norm": 1.9148075580596924, + "learning_rate": 3.716490658001625e-06, + "loss": 2.1846, + "step": 920 + }, + { + "epoch": 1.0082191780821919, + "eval_loss": 2.30368709564209, + "eval_runtime": 52.1845, + "eval_samples_per_second": 19.163, + "eval_steps_per_second": 0.805, + "step": 920 + }, + { + "epoch": 1.0093150684931507, + "grad_norm": 1.9502761363983154, + "learning_rate": 3.7144597887896024e-06, + "loss": 2.1731, + "step": 921 + }, + { + "epoch": 1.0104109589041095, + "grad_norm": 1.9093869924545288, + "learning_rate": 3.7124289195775797e-06, + "loss": 2.0864, + "step": 922 + }, + { + "epoch": 1.0115068493150685, + "grad_norm": 2.0871293544769287, + "learning_rate": 3.710398050365557e-06, + "loss": 2.1838, + "step": 923 + }, + { + "epoch": 1.0126027397260273, + "grad_norm": 2.0559985637664795, + "learning_rate": 3.708367181153534e-06, + "loss": 2.1764, + "step": 924 + }, + { + "epoch": 1.0136986301369864, + "grad_norm": 1.8968807458877563, + "learning_rate": 3.7063363119415115e-06, + "loss": 2.1515, + "step": 925 + }, + { + "epoch": 1.0147945205479452, + "grad_norm": 1.8613216876983643, + "learning_rate": 3.7043054427294884e-06, + "loss": 2.1694, + "step": 926 + }, + { + "epoch": 1.0158904109589042, + "grad_norm": 2.002044439315796, + "learning_rate": 3.702274573517466e-06, + "loss": 2.148, + "step": 927 + }, + { + "epoch": 1.016986301369863, + "grad_norm": 2.1138675212860107, + "learning_rate": 3.700243704305443e-06, + "loss": 2.1401, + "step": 928 + }, + { + "epoch": 1.0180821917808218, + "grad_norm": 1.7620841264724731, + "learning_rate": 3.6982128350934206e-06, + "loss": 2.1399, + "step": 929 + }, + { + "epoch": 1.0191780821917809, + "grad_norm": 1.8970746994018555, + "learning_rate": 3.6961819658813975e-06, + "loss": 2.1404, + "step": 930 + }, + { + "epoch": 1.0191780821917809, + "eval_loss": 2.3068811893463135, + "eval_runtime": 53.0188, + "eval_samples_per_second": 18.861, + "eval_steps_per_second": 0.792, + "step": 930 + }, + { + "epoch": 1.0202739726027397, + "grad_norm": 1.9405478239059448, + "learning_rate": 3.694151096669375e-06, + "loss": 2.1286, + "step": 931 + }, + { + "epoch": 1.0213698630136987, + "grad_norm": 2.112222909927368, + "learning_rate": 3.692120227457352e-06, + "loss": 2.2246, + "step": 932 + }, + { + "epoch": 1.0224657534246575, + "grad_norm": 1.8792580366134644, + "learning_rate": 3.6900893582453297e-06, + "loss": 2.148, + "step": 933 + }, + { + "epoch": 1.0235616438356165, + "grad_norm": 1.9530014991760254, + "learning_rate": 3.6880584890333066e-06, + "loss": 2.1707, + "step": 934 + }, + { + "epoch": 1.0246575342465754, + "grad_norm": 2.1778736114501953, + "learning_rate": 3.6860276198212834e-06, + "loss": 2.0879, + "step": 935 + }, + { + "epoch": 1.0257534246575342, + "grad_norm": 1.8960413932800293, + "learning_rate": 3.683996750609261e-06, + "loss": 2.1659, + "step": 936 + }, + { + "epoch": 1.0268493150684932, + "grad_norm": 1.9805221557617188, + "learning_rate": 3.681965881397238e-06, + "loss": 2.1674, + "step": 937 + }, + { + "epoch": 1.027945205479452, + "grad_norm": 1.9999068975448608, + "learning_rate": 3.6799350121852157e-06, + "loss": 2.1677, + "step": 938 + }, + { + "epoch": 1.029041095890411, + "grad_norm": 2.016911029815674, + "learning_rate": 3.6779041429731925e-06, + "loss": 2.1841, + "step": 939 + }, + { + "epoch": 1.0301369863013699, + "grad_norm": 1.7200374603271484, + "learning_rate": 3.6758732737611702e-06, + "loss": 2.1391, + "step": 940 + }, + { + "epoch": 1.0301369863013699, + "eval_loss": 2.3022048473358154, + "eval_runtime": 51.2908, + "eval_samples_per_second": 19.497, + "eval_steps_per_second": 0.819, + "step": 940 + }, + { + "epoch": 1.0312328767123287, + "grad_norm": 2.249377489089966, + "learning_rate": 3.673842404549147e-06, + "loss": 2.1618, + "step": 941 + }, + { + "epoch": 1.0323287671232877, + "grad_norm": 1.838922381401062, + "learning_rate": 3.671811535337125e-06, + "loss": 2.1527, + "step": 942 + }, + { + "epoch": 1.0334246575342465, + "grad_norm": 1.9571045637130737, + "learning_rate": 3.6697806661251016e-06, + "loss": 2.149, + "step": 943 + }, + { + "epoch": 1.0345205479452055, + "grad_norm": 1.9929102659225464, + "learning_rate": 3.6677497969130793e-06, + "loss": 2.159, + "step": 944 + }, + { + "epoch": 1.0356164383561643, + "grad_norm": 2.004830837249756, + "learning_rate": 3.665718927701056e-06, + "loss": 2.2369, + "step": 945 + }, + { + "epoch": 1.0367123287671234, + "grad_norm": 1.9605739116668701, + "learning_rate": 3.663688058489034e-06, + "loss": 2.1795, + "step": 946 + }, + { + "epoch": 1.0378082191780822, + "grad_norm": 1.9499164819717407, + "learning_rate": 3.6616571892770107e-06, + "loss": 2.1979, + "step": 947 + }, + { + "epoch": 1.038904109589041, + "grad_norm": 1.867347240447998, + "learning_rate": 3.659626320064988e-06, + "loss": 2.1616, + "step": 948 + }, + { + "epoch": 1.04, + "grad_norm": 1.9342801570892334, + "learning_rate": 3.6575954508529653e-06, + "loss": 2.1044, + "step": 949 + }, + { + "epoch": 1.0410958904109588, + "grad_norm": 2.0009584426879883, + "learning_rate": 3.6555645816409426e-06, + "loss": 2.1543, + "step": 950 + }, + { + "epoch": 1.0410958904109588, + "eval_loss": 2.3028438091278076, + "eval_runtime": 46.7426, + "eval_samples_per_second": 21.394, + "eval_steps_per_second": 0.899, + "step": 950 + }, + { + "epoch": 1.0421917808219179, + "grad_norm": 2.0411503314971924, + "learning_rate": 3.65353371242892e-06, + "loss": 2.1465, + "step": 951 + }, + { + "epoch": 1.0432876712328767, + "grad_norm": 2.0512070655822754, + "learning_rate": 3.651502843216897e-06, + "loss": 2.1828, + "step": 952 + }, + { + "epoch": 1.0443835616438357, + "grad_norm": 2.0948643684387207, + "learning_rate": 3.6494719740048744e-06, + "loss": 2.0671, + "step": 953 + }, + { + "epoch": 1.0454794520547945, + "grad_norm": 1.9103025197982788, + "learning_rate": 3.6474411047928517e-06, + "loss": 2.1651, + "step": 954 + }, + { + "epoch": 1.0465753424657533, + "grad_norm": 2.1500349044799805, + "learning_rate": 3.645410235580829e-06, + "loss": 2.137, + "step": 955 + }, + { + "epoch": 1.0476712328767124, + "grad_norm": 1.7760887145996094, + "learning_rate": 3.6433793663688062e-06, + "loss": 2.1738, + "step": 956 + }, + { + "epoch": 1.0487671232876712, + "grad_norm": 2.0111730098724365, + "learning_rate": 3.6413484971567835e-06, + "loss": 2.1569, + "step": 957 + }, + { + "epoch": 1.0498630136986302, + "grad_norm": 2.0040462017059326, + "learning_rate": 3.6393176279447608e-06, + "loss": 2.1483, + "step": 958 + }, + { + "epoch": 1.050958904109589, + "grad_norm": 1.9583499431610107, + "learning_rate": 3.637286758732738e-06, + "loss": 2.1575, + "step": 959 + }, + { + "epoch": 1.0520547945205478, + "grad_norm": 1.8842004537582397, + "learning_rate": 3.6352558895207153e-06, + "loss": 2.1797, + "step": 960 + }, + { + "epoch": 1.0520547945205478, + "eval_loss": 2.301715850830078, + "eval_runtime": 52.7345, + "eval_samples_per_second": 18.963, + "eval_steps_per_second": 0.796, + "step": 960 + }, + { + "epoch": 1.0531506849315069, + "grad_norm": 2.020909309387207, + "learning_rate": 3.633225020308692e-06, + "loss": 2.1939, + "step": 961 + }, + { + "epoch": 1.0542465753424657, + "grad_norm": 2.09222149848938, + "learning_rate": 3.63119415109667e-06, + "loss": 2.1777, + "step": 962 + }, + { + "epoch": 1.0553424657534247, + "grad_norm": 1.8582347631454468, + "learning_rate": 3.6291632818846467e-06, + "loss": 2.1505, + "step": 963 + }, + { + "epoch": 1.0564383561643835, + "grad_norm": 1.9468824863433838, + "learning_rate": 3.6271324126726244e-06, + "loss": 2.1762, + "step": 964 + }, + { + "epoch": 1.0575342465753426, + "grad_norm": 1.9484895467758179, + "learning_rate": 3.6251015434606013e-06, + "loss": 2.1102, + "step": 965 + }, + { + "epoch": 1.0586301369863014, + "grad_norm": 2.11462140083313, + "learning_rate": 3.623070674248579e-06, + "loss": 2.1114, + "step": 966 + }, + { + "epoch": 1.0597260273972602, + "grad_norm": 2.2038373947143555, + "learning_rate": 3.621039805036556e-06, + "loss": 2.1513, + "step": 967 + }, + { + "epoch": 1.0608219178082192, + "grad_norm": 2.1271684169769287, + "learning_rate": 3.6190089358245335e-06, + "loss": 2.137, + "step": 968 + }, + { + "epoch": 1.061917808219178, + "grad_norm": 1.8624194860458374, + "learning_rate": 3.6169780666125104e-06, + "loss": 2.1701, + "step": 969 + }, + { + "epoch": 1.063013698630137, + "grad_norm": 1.935445785522461, + "learning_rate": 3.614947197400488e-06, + "loss": 2.1726, + "step": 970 + }, + { + "epoch": 1.063013698630137, + "eval_loss": 2.302633047103882, + "eval_runtime": 51.8284, + "eval_samples_per_second": 19.294, + "eval_steps_per_second": 0.81, + "step": 970 + }, + { + "epoch": 1.0641095890410959, + "grad_norm": 2.048907518386841, + "learning_rate": 3.612916328188465e-06, + "loss": 2.1712, + "step": 971 + }, + { + "epoch": 1.0652054794520547, + "grad_norm": 1.850943684577942, + "learning_rate": 3.610885458976442e-06, + "loss": 2.102, + "step": 972 + }, + { + "epoch": 1.0663013698630137, + "grad_norm": 1.8364315032958984, + "learning_rate": 3.6088545897644195e-06, + "loss": 2.1903, + "step": 973 + }, + { + "epoch": 1.0673972602739725, + "grad_norm": 1.8227031230926514, + "learning_rate": 3.6068237205523963e-06, + "loss": 2.1787, + "step": 974 + }, + { + "epoch": 1.0684931506849316, + "grad_norm": 1.9136372804641724, + "learning_rate": 3.604792851340374e-06, + "loss": 2.1244, + "step": 975 + }, + { + "epoch": 1.0695890410958904, + "grad_norm": 1.8585968017578125, + "learning_rate": 3.602761982128351e-06, + "loss": 2.1287, + "step": 976 + }, + { + "epoch": 1.0706849315068494, + "grad_norm": 2.139573812484741, + "learning_rate": 3.6007311129163286e-06, + "loss": 2.1938, + "step": 977 + }, + { + "epoch": 1.0717808219178082, + "grad_norm": 2.1584925651550293, + "learning_rate": 3.5987002437043054e-06, + "loss": 2.1626, + "step": 978 + }, + { + "epoch": 1.072876712328767, + "grad_norm": 1.8905949592590332, + "learning_rate": 3.596669374492283e-06, + "loss": 2.1219, + "step": 979 + }, + { + "epoch": 1.073972602739726, + "grad_norm": 2.099651575088501, + "learning_rate": 3.59463850528026e-06, + "loss": 2.1385, + "step": 980 + }, + { + "epoch": 1.073972602739726, + "eval_loss": 2.3004090785980225, + "eval_runtime": 52.7406, + "eval_samples_per_second": 18.961, + "eval_steps_per_second": 0.796, + "step": 980 + }, + { + "epoch": 1.0750684931506849, + "grad_norm": 2.1456780433654785, + "learning_rate": 3.5926076360682377e-06, + "loss": 2.1973, + "step": 981 + }, + { + "epoch": 1.076164383561644, + "grad_norm": 1.929372787475586, + "learning_rate": 3.5905767668562145e-06, + "loss": 2.1115, + "step": 982 + }, + { + "epoch": 1.0772602739726027, + "grad_norm": 2.08174991607666, + "learning_rate": 3.5885458976441922e-06, + "loss": 2.1785, + "step": 983 + }, + { + "epoch": 1.0783561643835617, + "grad_norm": 1.9355952739715576, + "learning_rate": 3.586515028432169e-06, + "loss": 2.1567, + "step": 984 + }, + { + "epoch": 1.0794520547945206, + "grad_norm": 1.9983466863632202, + "learning_rate": 3.5844841592201464e-06, + "loss": 2.1913, + "step": 985 + }, + { + "epoch": 1.0805479452054794, + "grad_norm": 1.8714051246643066, + "learning_rate": 3.5824532900081237e-06, + "loss": 2.1938, + "step": 986 + }, + { + "epoch": 1.0816438356164384, + "grad_norm": 2.007049798965454, + "learning_rate": 3.580422420796101e-06, + "loss": 2.1093, + "step": 987 + }, + { + "epoch": 1.0827397260273972, + "grad_norm": 1.965572476387024, + "learning_rate": 3.578391551584078e-06, + "loss": 2.1426, + "step": 988 + }, + { + "epoch": 1.0838356164383562, + "grad_norm": 1.9040261507034302, + "learning_rate": 3.5763606823720555e-06, + "loss": 2.1627, + "step": 989 + }, + { + "epoch": 1.084931506849315, + "grad_norm": 2.001929759979248, + "learning_rate": 3.5743298131600328e-06, + "loss": 2.0983, + "step": 990 + }, + { + "epoch": 1.084931506849315, + "eval_loss": 2.2979862689971924, + "eval_runtime": 50.6325, + "eval_samples_per_second": 19.75, + "eval_steps_per_second": 0.83, + "step": 990 + }, + { + "epoch": 1.086027397260274, + "grad_norm": 1.9204787015914917, + "learning_rate": 3.57229894394801e-06, + "loss": 2.1571, + "step": 991 + }, + { + "epoch": 1.087123287671233, + "grad_norm": 1.9628651142120361, + "learning_rate": 3.5702680747359873e-06, + "loss": 2.1736, + "step": 992 + }, + { + "epoch": 1.0882191780821917, + "grad_norm": 1.9888029098510742, + "learning_rate": 3.5682372055239646e-06, + "loss": 2.17, + "step": 993 + }, + { + "epoch": 1.0893150684931507, + "grad_norm": 2.049769401550293, + "learning_rate": 3.566206336311942e-06, + "loss": 2.1352, + "step": 994 + }, + { + "epoch": 1.0904109589041096, + "grad_norm": 2.0208375453948975, + "learning_rate": 3.564175467099919e-06, + "loss": 2.1564, + "step": 995 + }, + { + "epoch": 1.0915068493150686, + "grad_norm": 1.976775050163269, + "learning_rate": 3.5621445978878964e-06, + "loss": 2.1292, + "step": 996 + }, + { + "epoch": 1.0926027397260274, + "grad_norm": 2.0502822399139404, + "learning_rate": 3.5601137286758737e-06, + "loss": 2.1115, + "step": 997 + }, + { + "epoch": 1.0936986301369862, + "grad_norm": 2.1184189319610596, + "learning_rate": 3.5580828594638505e-06, + "loss": 2.1641, + "step": 998 + }, + { + "epoch": 1.0947945205479452, + "grad_norm": 1.9662559032440186, + "learning_rate": 3.5560519902518282e-06, + "loss": 2.1338, + "step": 999 + }, + { + "epoch": 1.095890410958904, + "grad_norm": 2.13386869430542, + "learning_rate": 3.554021121039805e-06, + "loss": 2.158, + "step": 1000 + }, + { + "epoch": 1.095890410958904, + "eval_loss": 2.3001608848571777, + "eval_runtime": 45.9219, + "eval_samples_per_second": 21.776, + "eval_steps_per_second": 0.915, + "step": 1000 + }, + { + "epoch": 1.096986301369863, + "grad_norm": 1.8791636228561401, + "learning_rate": 3.551990251827783e-06, + "loss": 2.1752, + "step": 1001 + }, + { + "epoch": 1.098082191780822, + "grad_norm": 1.9105546474456787, + "learning_rate": 3.5499593826157596e-06, + "loss": 2.1543, + "step": 1002 + }, + { + "epoch": 1.099178082191781, + "grad_norm": 1.887668251991272, + "learning_rate": 3.5479285134037373e-06, + "loss": 2.1226, + "step": 1003 + }, + { + "epoch": 1.1002739726027397, + "grad_norm": 2.2854599952697754, + "learning_rate": 3.545897644191714e-06, + "loss": 2.1425, + "step": 1004 + }, + { + "epoch": 1.1013698630136985, + "grad_norm": 2.03892183303833, + "learning_rate": 3.543866774979692e-06, + "loss": 2.135, + "step": 1005 + }, + { + "epoch": 1.1024657534246576, + "grad_norm": 1.9144858121871948, + "learning_rate": 3.5418359057676687e-06, + "loss": 2.1323, + "step": 1006 + }, + { + "epoch": 1.1035616438356164, + "grad_norm": 2.1402337551116943, + "learning_rate": 3.5398050365556464e-06, + "loss": 2.1668, + "step": 1007 + }, + { + "epoch": 1.1046575342465754, + "grad_norm": 2.084592819213867, + "learning_rate": 3.5377741673436233e-06, + "loss": 2.1331, + "step": 1008 + }, + { + "epoch": 1.1057534246575342, + "grad_norm": 2.0012383460998535, + "learning_rate": 3.535743298131601e-06, + "loss": 2.1677, + "step": 1009 + }, + { + "epoch": 1.106849315068493, + "grad_norm": 2.272204637527466, + "learning_rate": 3.533712428919578e-06, + "loss": 2.13, + "step": 1010 + }, + { + "epoch": 1.106849315068493, + "eval_loss": 2.3001229763031006, + "eval_runtime": 42.0444, + "eval_samples_per_second": 23.784, + "eval_steps_per_second": 0.999, + "step": 1010 + }, + { + "epoch": 1.107945205479452, + "grad_norm": 2.1061534881591797, + "learning_rate": 3.5316815597075547e-06, + "loss": 2.1581, + "step": 1011 + }, + { + "epoch": 1.1090410958904109, + "grad_norm": 2.159306764602661, + "learning_rate": 3.5296506904955324e-06, + "loss": 2.1868, + "step": 1012 + }, + { + "epoch": 1.11013698630137, + "grad_norm": 2.1429545879364014, + "learning_rate": 3.5276198212835093e-06, + "loss": 2.1274, + "step": 1013 + }, + { + "epoch": 1.1112328767123287, + "grad_norm": 1.8494454622268677, + "learning_rate": 3.525588952071487e-06, + "loss": 2.1513, + "step": 1014 + }, + { + "epoch": 1.1123287671232878, + "grad_norm": 1.8494454622268677, + "learning_rate": 3.525588952071487e-06, + "loss": 2.1883, + "step": 1015 + }, + { + "epoch": 1.1134246575342466, + "grad_norm": 2.2244861125946045, + "learning_rate": 3.523558082859464e-06, + "loss": 2.1896, + "step": 1016 + }, + { + "epoch": 1.1145205479452054, + "grad_norm": 2.181950569152832, + "learning_rate": 3.5215272136474415e-06, + "loss": 2.1561, + "step": 1017 + }, + { + "epoch": 1.1156164383561644, + "grad_norm": 1.7740849256515503, + "learning_rate": 3.5194963444354184e-06, + "loss": 2.1358, + "step": 1018 + }, + { + "epoch": 1.1167123287671232, + "grad_norm": 2.3540139198303223, + "learning_rate": 3.517465475223396e-06, + "loss": 2.1241, + "step": 1019 + }, + { + "epoch": 1.1178082191780823, + "grad_norm": 2.2737820148468018, + "learning_rate": 3.515434606011373e-06, + "loss": 2.1117, + "step": 1020 + }, + { + "epoch": 1.1178082191780823, + "eval_loss": 2.2998275756835938, + "eval_runtime": 46.2142, + "eval_samples_per_second": 21.638, + "eval_steps_per_second": 0.909, + "step": 1020 + }, + { + "epoch": 1.118904109589041, + "grad_norm": 2.0705313682556152, + "learning_rate": 3.5134037367993506e-06, + "loss": 2.1743, + "step": 1021 + }, + { + "epoch": 1.12, + "grad_norm": 2.2605583667755127, + "learning_rate": 3.5113728675873275e-06, + "loss": 2.2082, + "step": 1022 + }, + { + "epoch": 1.121095890410959, + "grad_norm": 2.028203248977661, + "learning_rate": 3.5093419983753047e-06, + "loss": 2.1722, + "step": 1023 + }, + { + "epoch": 1.1221917808219177, + "grad_norm": 1.9794270992279053, + "learning_rate": 3.507311129163282e-06, + "loss": 2.1581, + "step": 1024 + }, + { + "epoch": 1.1232876712328768, + "grad_norm": 2.1830344200134277, + "learning_rate": 3.5052802599512593e-06, + "loss": 2.1142, + "step": 1025 + }, + { + "epoch": 1.1243835616438356, + "grad_norm": 1.926081895828247, + "learning_rate": 3.5032493907392366e-06, + "loss": 2.1532, + "step": 1026 + }, + { + "epoch": 1.1254794520547946, + "grad_norm": 1.9909394979476929, + "learning_rate": 3.501218521527214e-06, + "loss": 2.153, + "step": 1027 + }, + { + "epoch": 1.1265753424657534, + "grad_norm": 1.9685133695602417, + "learning_rate": 3.499187652315191e-06, + "loss": 2.0983, + "step": 1028 + }, + { + "epoch": 1.1276712328767124, + "grad_norm": 1.9185279607772827, + "learning_rate": 3.4971567831031684e-06, + "loss": 2.1495, + "step": 1029 + }, + { + "epoch": 1.1287671232876713, + "grad_norm": 2.0214717388153076, + "learning_rate": 3.4951259138911457e-06, + "loss": 2.176, + "step": 1030 + }, + { + "epoch": 1.1287671232876713, + "eval_loss": 2.2963690757751465, + "eval_runtime": 44.2655, + "eval_samples_per_second": 22.591, + "eval_steps_per_second": 0.949, + "step": 1030 + }, + { + "epoch": 1.12986301369863, + "grad_norm": 1.9134252071380615, + "learning_rate": 3.493095044679123e-06, + "loss": 2.1034, + "step": 1031 + }, + { + "epoch": 1.130958904109589, + "grad_norm": 1.8448772430419922, + "learning_rate": 3.4910641754671002e-06, + "loss": 2.1947, + "step": 1032 + }, + { + "epoch": 1.132054794520548, + "grad_norm": 1.856341004371643, + "learning_rate": 3.4890333062550775e-06, + "loss": 2.0676, + "step": 1033 + }, + { + "epoch": 1.133150684931507, + "grad_norm": 2.039062261581421, + "learning_rate": 3.4870024370430548e-06, + "loss": 2.1618, + "step": 1034 + }, + { + "epoch": 1.1342465753424658, + "grad_norm": 1.8440991640090942, + "learning_rate": 3.484971567831032e-06, + "loss": 2.0956, + "step": 1035 + }, + { + "epoch": 1.1353424657534246, + "grad_norm": 1.8483991622924805, + "learning_rate": 3.482940698619009e-06, + "loss": 2.1364, + "step": 1036 + }, + { + "epoch": 1.1364383561643836, + "grad_norm": 1.9045069217681885, + "learning_rate": 3.4809098294069866e-06, + "loss": 2.1786, + "step": 1037 + }, + { + "epoch": 1.1375342465753424, + "grad_norm": 1.8203203678131104, + "learning_rate": 3.4788789601949634e-06, + "loss": 2.1438, + "step": 1038 + }, + { + "epoch": 1.1386301369863014, + "grad_norm": 2.1084606647491455, + "learning_rate": 3.476848090982941e-06, + "loss": 2.1553, + "step": 1039 + }, + { + "epoch": 1.1397260273972603, + "grad_norm": 2.1041061878204346, + "learning_rate": 3.474817221770918e-06, + "loss": 2.1435, + "step": 1040 + }, + { + "epoch": 1.1397260273972603, + "eval_loss": 2.2970690727233887, + "eval_runtime": 43.2304, + "eval_samples_per_second": 23.132, + "eval_steps_per_second": 0.972, + "step": 1040 + }, + { + "epoch": 1.140821917808219, + "grad_norm": 1.9701104164123535, + "learning_rate": 3.4727863525588957e-06, + "loss": 2.2201, + "step": 1041 + }, + { + "epoch": 1.141917808219178, + "grad_norm": 1.8816475868225098, + "learning_rate": 3.4707554833468725e-06, + "loss": 2.1643, + "step": 1042 + }, + { + "epoch": 1.143013698630137, + "grad_norm": 2.096355438232422, + "learning_rate": 3.4687246141348502e-06, + "loss": 2.1514, + "step": 1043 + }, + { + "epoch": 1.144109589041096, + "grad_norm": 1.9975471496582031, + "learning_rate": 3.466693744922827e-06, + "loss": 2.1246, + "step": 1044 + }, + { + "epoch": 1.1452054794520548, + "grad_norm": 2.0587127208709717, + "learning_rate": 3.464662875710805e-06, + "loss": 2.1371, + "step": 1045 + }, + { + "epoch": 1.1463013698630138, + "grad_norm": 1.9275381565093994, + "learning_rate": 3.4626320064987817e-06, + "loss": 2.1518, + "step": 1046 + }, + { + "epoch": 1.1473972602739726, + "grad_norm": 2.1690948009490967, + "learning_rate": 3.4606011372867594e-06, + "loss": 2.1528, + "step": 1047 + }, + { + "epoch": 1.1484931506849314, + "grad_norm": 1.970973014831543, + "learning_rate": 3.458570268074736e-06, + "loss": 2.1569, + "step": 1048 + }, + { + "epoch": 1.1495890410958904, + "grad_norm": 1.8897690773010254, + "learning_rate": 3.4565393988627135e-06, + "loss": 2.1401, + "step": 1049 + }, + { + "epoch": 1.1506849315068493, + "grad_norm": 1.9245911836624146, + "learning_rate": 3.4545085296506908e-06, + "loss": 2.1311, + "step": 1050 + }, + { + "epoch": 1.1506849315068493, + "eval_loss": 2.2956528663635254, + "eval_runtime": 44.1778, + "eval_samples_per_second": 22.636, + "eval_steps_per_second": 0.951, + "step": 1050 + }, + { + "epoch": 1.1517808219178083, + "grad_norm": 2.142209529876709, + "learning_rate": 3.452477660438668e-06, + "loss": 2.1144, + "step": 1051 + }, + { + "epoch": 1.152876712328767, + "grad_norm": 1.932369351387024, + "learning_rate": 3.4504467912266453e-06, + "loss": 2.1739, + "step": 1052 + }, + { + "epoch": 1.1539726027397261, + "grad_norm": 2.259359359741211, + "learning_rate": 3.4484159220146226e-06, + "loss": 2.1036, + "step": 1053 + }, + { + "epoch": 1.155068493150685, + "grad_norm": 2.112485647201538, + "learning_rate": 3.4463850528026e-06, + "loss": 2.1429, + "step": 1054 + }, + { + "epoch": 1.1561643835616437, + "grad_norm": 2.6211049556732178, + "learning_rate": 3.444354183590577e-06, + "loss": 2.1421, + "step": 1055 + }, + { + "epoch": 1.1572602739726028, + "grad_norm": 1.8724851608276367, + "learning_rate": 3.4423233143785544e-06, + "loss": 2.1622, + "step": 1056 + }, + { + "epoch": 1.1583561643835616, + "grad_norm": 1.9292882680892944, + "learning_rate": 3.4402924451665317e-06, + "loss": 2.1071, + "step": 1057 + }, + { + "epoch": 1.1594520547945206, + "grad_norm": 1.9832521677017212, + "learning_rate": 3.438261575954509e-06, + "loss": 2.0879, + "step": 1058 + }, + { + "epoch": 1.1605479452054794, + "grad_norm": 1.9720780849456787, + "learning_rate": 3.4362307067424862e-06, + "loss": 2.1603, + "step": 1059 + }, + { + "epoch": 1.1616438356164385, + "grad_norm": 1.8179587125778198, + "learning_rate": 3.434199837530463e-06, + "loss": 2.1851, + "step": 1060 + }, + { + "epoch": 1.1616438356164385, + "eval_loss": 2.2950079441070557, + "eval_runtime": 42.9625, + "eval_samples_per_second": 23.276, + "eval_steps_per_second": 0.978, + "step": 1060 + }, + { + "epoch": 1.1627397260273973, + "grad_norm": 1.8762129545211792, + "learning_rate": 3.432168968318441e-06, + "loss": 2.1527, + "step": 1061 + }, + { + "epoch": 1.163835616438356, + "grad_norm": 2.173297643661499, + "learning_rate": 3.4301380991064176e-06, + "loss": 2.1023, + "step": 1062 + }, + { + "epoch": 1.1649315068493151, + "grad_norm": 1.9358222484588623, + "learning_rate": 3.4281072298943953e-06, + "loss": 2.1638, + "step": 1063 + }, + { + "epoch": 1.166027397260274, + "grad_norm": 1.9421939849853516, + "learning_rate": 3.426076360682372e-06, + "loss": 2.1794, + "step": 1064 + }, + { + "epoch": 1.167123287671233, + "grad_norm": 2.212996006011963, + "learning_rate": 3.42404549147035e-06, + "loss": 2.1877, + "step": 1065 + }, + { + "epoch": 1.1682191780821918, + "grad_norm": 2.036180257797241, + "learning_rate": 3.4220146222583267e-06, + "loss": 2.1534, + "step": 1066 + }, + { + "epoch": 1.1693150684931506, + "grad_norm": 1.8819142580032349, + "learning_rate": 3.4199837530463044e-06, + "loss": 2.099, + "step": 1067 + }, + { + "epoch": 1.1704109589041096, + "grad_norm": 2.0781915187835693, + "learning_rate": 3.4179528838342813e-06, + "loss": 2.1326, + "step": 1068 + }, + { + "epoch": 1.1715068493150684, + "grad_norm": 2.034895181655884, + "learning_rate": 3.415922014622259e-06, + "loss": 2.2107, + "step": 1069 + }, + { + "epoch": 1.1726027397260275, + "grad_norm": 1.92604660987854, + "learning_rate": 3.413891145410236e-06, + "loss": 2.1773, + "step": 1070 + }, + { + "epoch": 1.1726027397260275, + "eval_loss": 2.2959954738616943, + "eval_runtime": 47.2123, + "eval_samples_per_second": 21.181, + "eval_steps_per_second": 0.89, + "step": 1070 + }, + { + "epoch": 1.1736986301369863, + "grad_norm": 1.762730598449707, + "learning_rate": 3.4118602761982135e-06, + "loss": 2.1803, + "step": 1071 + }, + { + "epoch": 1.174794520547945, + "grad_norm": 2.0661215782165527, + "learning_rate": 3.4098294069861904e-06, + "loss": 2.1919, + "step": 1072 + }, + { + "epoch": 1.1758904109589041, + "grad_norm": 2.079982280731201, + "learning_rate": 3.4077985377741673e-06, + "loss": 2.1536, + "step": 1073 + }, + { + "epoch": 1.176986301369863, + "grad_norm": 1.9044359922409058, + "learning_rate": 3.405767668562145e-06, + "loss": 2.1403, + "step": 1074 + }, + { + "epoch": 1.178082191780822, + "grad_norm": 1.8870306015014648, + "learning_rate": 3.403736799350122e-06, + "loss": 2.1845, + "step": 1075 + }, + { + "epoch": 1.1791780821917808, + "grad_norm": 1.9305400848388672, + "learning_rate": 3.4017059301380995e-06, + "loss": 2.1786, + "step": 1076 + }, + { + "epoch": 1.1802739726027398, + "grad_norm": 1.987713098526001, + "learning_rate": 3.3996750609260764e-06, + "loss": 2.1214, + "step": 1077 + }, + { + "epoch": 1.1813698630136986, + "grad_norm": 1.8673466444015503, + "learning_rate": 3.397644191714054e-06, + "loss": 2.2027, + "step": 1078 + }, + { + "epoch": 1.1824657534246574, + "grad_norm": 2.009316921234131, + "learning_rate": 3.395613322502031e-06, + "loss": 2.139, + "step": 1079 + }, + { + "epoch": 1.1835616438356165, + "grad_norm": 2.1024956703186035, + "learning_rate": 3.3935824532900086e-06, + "loss": 2.1438, + "step": 1080 + }, + { + "epoch": 1.1835616438356165, + "eval_loss": 2.2966299057006836, + "eval_runtime": 44.8946, + "eval_samples_per_second": 22.274, + "eval_steps_per_second": 0.936, + "step": 1080 + }, + { + "epoch": 1.1846575342465753, + "grad_norm": 1.8911967277526855, + "learning_rate": 3.3915515840779855e-06, + "loss": 2.1238, + "step": 1081 + }, + { + "epoch": 1.1857534246575343, + "grad_norm": 2.0649192333221436, + "learning_rate": 3.389520714865963e-06, + "loss": 2.1534, + "step": 1082 + }, + { + "epoch": 1.1868493150684931, + "grad_norm": 1.9236847162246704, + "learning_rate": 3.38748984565394e-06, + "loss": 2.1427, + "step": 1083 + }, + { + "epoch": 1.1879452054794521, + "grad_norm": 1.8662182092666626, + "learning_rate": 3.3854589764419177e-06, + "loss": 2.1699, + "step": 1084 + }, + { + "epoch": 1.189041095890411, + "grad_norm": 1.9226020574569702, + "learning_rate": 3.3834281072298946e-06, + "loss": 2.1192, + "step": 1085 + }, + { + "epoch": 1.1901369863013698, + "grad_norm": 1.9659301042556763, + "learning_rate": 3.381397238017872e-06, + "loss": 2.1529, + "step": 1086 + }, + { + "epoch": 1.1912328767123288, + "grad_norm": 2.2714693546295166, + "learning_rate": 3.379366368805849e-06, + "loss": 2.1273, + "step": 1087 + }, + { + "epoch": 1.1923287671232876, + "grad_norm": 2.1853270530700684, + "learning_rate": 3.3773354995938264e-06, + "loss": 2.2029, + "step": 1088 + }, + { + "epoch": 1.1934246575342466, + "grad_norm": 1.8302830457687378, + "learning_rate": 3.3753046303818037e-06, + "loss": 2.1599, + "step": 1089 + }, + { + "epoch": 1.1945205479452055, + "grad_norm": 2.005591869354248, + "learning_rate": 3.373273761169781e-06, + "loss": 2.139, + "step": 1090 + }, + { + "epoch": 1.1945205479452055, + "eval_loss": 2.295994520187378, + "eval_runtime": 45.4802, + "eval_samples_per_second": 21.988, + "eval_steps_per_second": 0.923, + "step": 1090 + }, + { + "epoch": 1.1956164383561645, + "grad_norm": 2.139267683029175, + "learning_rate": 3.3712428919577582e-06, + "loss": 2.151, + "step": 1091 + }, + { + "epoch": 1.1967123287671233, + "grad_norm": 1.8547207117080688, + "learning_rate": 3.3692120227457355e-06, + "loss": 2.1249, + "step": 1092 + }, + { + "epoch": 1.1978082191780821, + "grad_norm": 2.214583396911621, + "learning_rate": 3.3671811535337128e-06, + "loss": 2.1656, + "step": 1093 + }, + { + "epoch": 1.1989041095890411, + "grad_norm": 2.1608800888061523, + "learning_rate": 3.36515028432169e-06, + "loss": 2.1056, + "step": 1094 + }, + { + "epoch": 1.2, + "grad_norm": 2.0850605964660645, + "learning_rate": 3.3631194151096673e-06, + "loss": 2.1215, + "step": 1095 + }, + { + "epoch": 1.201095890410959, + "grad_norm": 1.9199103116989136, + "learning_rate": 3.3610885458976446e-06, + "loss": 2.1584, + "step": 1096 + }, + { + "epoch": 1.2021917808219178, + "grad_norm": 1.9766334295272827, + "learning_rate": 3.359057676685622e-06, + "loss": 2.1821, + "step": 1097 + }, + { + "epoch": 1.2032876712328768, + "grad_norm": 1.948988676071167, + "learning_rate": 3.357026807473599e-06, + "loss": 2.182, + "step": 1098 + }, + { + "epoch": 1.2043835616438356, + "grad_norm": 2.0419397354125977, + "learning_rate": 3.354995938261576e-06, + "loss": 2.1051, + "step": 1099 + }, + { + "epoch": 1.2054794520547945, + "grad_norm": 1.9751213788986206, + "learning_rate": 3.3529650690495537e-06, + "loss": 2.1255, + "step": 1100 + }, + { + "epoch": 1.2054794520547945, + "eval_loss": 2.293811798095703, + "eval_runtime": 46.1182, + "eval_samples_per_second": 21.683, + "eval_steps_per_second": 0.911, + "step": 1100 + }, + { + "epoch": 1.2065753424657535, + "grad_norm": 1.9806510210037231, + "learning_rate": 3.3509341998375305e-06, + "loss": 2.1021, + "step": 1101 + }, + { + "epoch": 1.2076712328767123, + "grad_norm": 2.0570414066314697, + "learning_rate": 3.3489033306255082e-06, + "loss": 2.1417, + "step": 1102 + }, + { + "epoch": 1.208767123287671, + "grad_norm": 1.868134617805481, + "learning_rate": 3.346872461413485e-06, + "loss": 2.1494, + "step": 1103 + }, + { + "epoch": 1.2098630136986301, + "grad_norm": 2.0156848430633545, + "learning_rate": 3.344841592201463e-06, + "loss": 2.1238, + "step": 1104 + }, + { + "epoch": 1.210958904109589, + "grad_norm": 2.082301616668701, + "learning_rate": 3.3428107229894397e-06, + "loss": 2.1105, + "step": 1105 + }, + { + "epoch": 1.212054794520548, + "grad_norm": 1.8853904008865356, + "learning_rate": 3.3407798537774174e-06, + "loss": 2.1406, + "step": 1106 + }, + { + "epoch": 1.2131506849315068, + "grad_norm": 2.0410518646240234, + "learning_rate": 3.338748984565394e-06, + "loss": 2.1567, + "step": 1107 + }, + { + "epoch": 1.2142465753424658, + "grad_norm": 1.9901232719421387, + "learning_rate": 3.336718115353372e-06, + "loss": 2.1778, + "step": 1108 + }, + { + "epoch": 1.2153424657534246, + "grad_norm": 1.9840792417526245, + "learning_rate": 3.3346872461413488e-06, + "loss": 2.1935, + "step": 1109 + }, + { + "epoch": 1.2164383561643834, + "grad_norm": 1.8854445219039917, + "learning_rate": 3.3326563769293256e-06, + "loss": 2.137, + "step": 1110 + }, + { + "epoch": 1.2164383561643834, + "eval_loss": 2.294447660446167, + "eval_runtime": 42.6423, + "eval_samples_per_second": 23.451, + "eval_steps_per_second": 0.985, + "step": 1110 + }, + { + "epoch": 1.2175342465753425, + "grad_norm": 2.0948665142059326, + "learning_rate": 3.3306255077173033e-06, + "loss": 2.1288, + "step": 1111 + }, + { + "epoch": 1.2186301369863013, + "grad_norm": 1.859448790550232, + "learning_rate": 3.32859463850528e-06, + "loss": 2.1403, + "step": 1112 + }, + { + "epoch": 1.2197260273972603, + "grad_norm": 1.92074453830719, + "learning_rate": 3.326563769293258e-06, + "loss": 2.141, + "step": 1113 + }, + { + "epoch": 1.2208219178082191, + "grad_norm": 1.8377320766448975, + "learning_rate": 3.3245329000812347e-06, + "loss": 2.0889, + "step": 1114 + }, + { + "epoch": 1.2219178082191782, + "grad_norm": 1.9320917129516602, + "learning_rate": 3.3225020308692124e-06, + "loss": 2.1254, + "step": 1115 + }, + { + "epoch": 1.223013698630137, + "grad_norm": 1.9691251516342163, + "learning_rate": 3.3204711616571893e-06, + "loss": 2.1409, + "step": 1116 + }, + { + "epoch": 1.2241095890410958, + "grad_norm": 1.8214130401611328, + "learning_rate": 3.318440292445167e-06, + "loss": 2.1046, + "step": 1117 + }, + { + "epoch": 1.2252054794520548, + "grad_norm": 1.959455966949463, + "learning_rate": 3.316409423233144e-06, + "loss": 2.1418, + "step": 1118 + }, + { + "epoch": 1.2263013698630136, + "grad_norm": 1.9920449256896973, + "learning_rate": 3.3143785540211215e-06, + "loss": 2.1192, + "step": 1119 + }, + { + "epoch": 1.2273972602739727, + "grad_norm": 2.0276854038238525, + "learning_rate": 3.3123476848090984e-06, + "loss": 2.1939, + "step": 1120 + }, + { + "epoch": 1.2273972602739727, + "eval_loss": 2.295330047607422, + "eval_runtime": 49.8691, + "eval_samples_per_second": 20.052, + "eval_steps_per_second": 0.842, + "step": 1120 + }, + { + "epoch": 1.2284931506849315, + "grad_norm": 1.9765129089355469, + "learning_rate": 3.310316815597076e-06, + "loss": 2.1545, + "step": 1121 + }, + { + "epoch": 1.2295890410958905, + "grad_norm": 1.8449276685714722, + "learning_rate": 3.308285946385053e-06, + "loss": 2.1459, + "step": 1122 + }, + { + "epoch": 1.2306849315068493, + "grad_norm": 1.953072190284729, + "learning_rate": 3.30625507717303e-06, + "loss": 2.2046, + "step": 1123 + }, + { + "epoch": 1.2317808219178081, + "grad_norm": 1.9570913314819336, + "learning_rate": 3.3042242079610075e-06, + "loss": 2.1015, + "step": 1124 + }, + { + "epoch": 1.2328767123287672, + "grad_norm": 2.00569224357605, + "learning_rate": 3.3021933387489847e-06, + "loss": 2.1681, + "step": 1125 + }, + { + "epoch": 1.233972602739726, + "grad_norm": 1.9131590127944946, + "learning_rate": 3.300162469536962e-06, + "loss": 2.1765, + "step": 1126 + }, + { + "epoch": 1.235068493150685, + "grad_norm": 1.8375864028930664, + "learning_rate": 3.2981316003249393e-06, + "loss": 2.1175, + "step": 1127 + }, + { + "epoch": 1.2361643835616438, + "grad_norm": 1.852847933769226, + "learning_rate": 3.2961007311129166e-06, + "loss": 2.1838, + "step": 1128 + }, + { + "epoch": 1.2372602739726029, + "grad_norm": 1.8772693872451782, + "learning_rate": 3.294069861900894e-06, + "loss": 2.1524, + "step": 1129 + }, + { + "epoch": 1.2383561643835617, + "grad_norm": 1.8593205213546753, + "learning_rate": 3.292038992688871e-06, + "loss": 2.2059, + "step": 1130 + }, + { + "epoch": 1.2383561643835617, + "eval_loss": 2.294188976287842, + "eval_runtime": 50.4549, + "eval_samples_per_second": 19.82, + "eval_steps_per_second": 0.832, + "step": 1130 + }, + { + "epoch": 1.2394520547945205, + "grad_norm": 1.8967972993850708, + "learning_rate": 3.2900081234768484e-06, + "loss": 2.1551, + "step": 1131 + }, + { + "epoch": 1.2405479452054795, + "grad_norm": 1.9799107313156128, + "learning_rate": 3.2879772542648257e-06, + "loss": 2.1644, + "step": 1132 + }, + { + "epoch": 1.2416438356164383, + "grad_norm": 1.978883147239685, + "learning_rate": 3.285946385052803e-06, + "loss": 2.182, + "step": 1133 + }, + { + "epoch": 1.2427397260273974, + "grad_norm": 1.8705304861068726, + "learning_rate": 3.2839155158407802e-06, + "loss": 2.0981, + "step": 1134 + }, + { + "epoch": 1.2438356164383562, + "grad_norm": 2.0146889686584473, + "learning_rate": 3.2818846466287575e-06, + "loss": 2.1686, + "step": 1135 + }, + { + "epoch": 1.244931506849315, + "grad_norm": 1.94461989402771, + "learning_rate": 3.2798537774167344e-06, + "loss": 2.093, + "step": 1136 + }, + { + "epoch": 1.246027397260274, + "grad_norm": 1.7823240756988525, + "learning_rate": 3.277822908204712e-06, + "loss": 2.198, + "step": 1137 + }, + { + "epoch": 1.2471232876712328, + "grad_norm": 1.89297354221344, + "learning_rate": 3.275792038992689e-06, + "loss": 2.1359, + "step": 1138 + }, + { + "epoch": 1.2482191780821918, + "grad_norm": 1.8670459985733032, + "learning_rate": 3.2737611697806666e-06, + "loss": 2.0956, + "step": 1139 + }, + { + "epoch": 1.2493150684931507, + "grad_norm": 1.813568353652954, + "learning_rate": 3.2717303005686435e-06, + "loss": 2.1448, + "step": 1140 + }, + { + "epoch": 1.2493150684931507, + "eval_loss": 2.295804023742676, + "eval_runtime": 53.286, + "eval_samples_per_second": 18.767, + "eval_steps_per_second": 0.788, + "step": 1140 + }, + { + "epoch": 1.2504109589041095, + "grad_norm": 1.912916898727417, + "learning_rate": 3.269699431356621e-06, + "loss": 2.1159, + "step": 1141 + }, + { + "epoch": 1.2515068493150685, + "grad_norm": 1.7939883470535278, + "learning_rate": 3.267668562144598e-06, + "loss": 2.1566, + "step": 1142 + }, + { + "epoch": 1.2526027397260273, + "grad_norm": 1.8494806289672852, + "learning_rate": 3.2656376929325757e-06, + "loss": 2.1496, + "step": 1143 + }, + { + "epoch": 1.2536986301369863, + "grad_norm": 1.8468523025512695, + "learning_rate": 3.2636068237205526e-06, + "loss": 2.1159, + "step": 1144 + }, + { + "epoch": 1.2547945205479452, + "grad_norm": 1.8962055444717407, + "learning_rate": 3.2615759545085303e-06, + "loss": 2.1712, + "step": 1145 + }, + { + "epoch": 1.2558904109589042, + "grad_norm": 1.800353765487671, + "learning_rate": 3.259545085296507e-06, + "loss": 2.1486, + "step": 1146 + }, + { + "epoch": 1.256986301369863, + "grad_norm": 2.124359607696533, + "learning_rate": 3.257514216084484e-06, + "loss": 2.1913, + "step": 1147 + }, + { + "epoch": 1.2580821917808218, + "grad_norm": 1.9248145818710327, + "learning_rate": 3.2554833468724617e-06, + "loss": 2.1128, + "step": 1148 + }, + { + "epoch": 1.2591780821917808, + "grad_norm": 1.9990876913070679, + "learning_rate": 3.2534524776604385e-06, + "loss": 2.1854, + "step": 1149 + }, + { + "epoch": 1.2602739726027397, + "grad_norm": 1.8600568771362305, + "learning_rate": 3.2514216084484162e-06, + "loss": 2.1565, + "step": 1150 + }, + { + "epoch": 1.2602739726027397, + "eval_loss": 2.2942006587982178, + "eval_runtime": 51.4021, + "eval_samples_per_second": 19.454, + "eval_steps_per_second": 0.817, + "step": 1150 + }, + { + "epoch": 1.2613698630136987, + "grad_norm": 1.8621101379394531, + "learning_rate": 3.249390739236393e-06, + "loss": 2.1438, + "step": 1151 + }, + { + "epoch": 1.2624657534246575, + "grad_norm": 1.9290356636047363, + "learning_rate": 3.2473598700243708e-06, + "loss": 2.1577, + "step": 1152 + }, + { + "epoch": 1.2635616438356165, + "grad_norm": 1.9007307291030884, + "learning_rate": 3.2453290008123476e-06, + "loss": 2.1295, + "step": 1153 + }, + { + "epoch": 1.2646575342465753, + "grad_norm": 1.7647960186004639, + "learning_rate": 3.2432981316003253e-06, + "loss": 2.1472, + "step": 1154 + }, + { + "epoch": 1.2657534246575342, + "grad_norm": 1.9796825647354126, + "learning_rate": 3.241267262388302e-06, + "loss": 2.1542, + "step": 1155 + }, + { + "epoch": 1.2668493150684932, + "grad_norm": 1.9505796432495117, + "learning_rate": 3.23923639317628e-06, + "loss": 2.1093, + "step": 1156 + }, + { + "epoch": 1.267945205479452, + "grad_norm": 1.9152203798294067, + "learning_rate": 3.2372055239642567e-06, + "loss": 2.1456, + "step": 1157 + }, + { + "epoch": 1.269041095890411, + "grad_norm": 1.9094963073730469, + "learning_rate": 3.2351746547522344e-06, + "loss": 2.1149, + "step": 1158 + }, + { + "epoch": 1.2701369863013698, + "grad_norm": 1.8872679471969604, + "learning_rate": 3.2331437855402113e-06, + "loss": 2.1545, + "step": 1159 + }, + { + "epoch": 1.2712328767123289, + "grad_norm": 2.171703338623047, + "learning_rate": 3.2311129163281885e-06, + "loss": 2.2108, + "step": 1160 + }, + { + "epoch": 1.2712328767123289, + "eval_loss": 2.2910356521606445, + "eval_runtime": 52.5176, + "eval_samples_per_second": 19.041, + "eval_steps_per_second": 0.8, + "step": 1160 + }, + { + "epoch": 1.2723287671232877, + "grad_norm": 1.8578304052352905, + "learning_rate": 3.229082047116166e-06, + "loss": 2.1158, + "step": 1161 + }, + { + "epoch": 1.2734246575342465, + "grad_norm": 1.8873049020767212, + "learning_rate": 3.227051177904143e-06, + "loss": 2.1618, + "step": 1162 + }, + { + "epoch": 1.2745205479452055, + "grad_norm": 1.8442409038543701, + "learning_rate": 3.2250203086921204e-06, + "loss": 2.1454, + "step": 1163 + }, + { + "epoch": 1.2756164383561643, + "grad_norm": 1.9609211683273315, + "learning_rate": 3.2229894394800977e-06, + "loss": 2.1519, + "step": 1164 + }, + { + "epoch": 1.2767123287671232, + "grad_norm": 1.9093680381774902, + "learning_rate": 3.220958570268075e-06, + "loss": 2.1775, + "step": 1165 + }, + { + "epoch": 1.2778082191780822, + "grad_norm": 1.8355298042297363, + "learning_rate": 3.218927701056052e-06, + "loss": 2.0871, + "step": 1166 + }, + { + "epoch": 1.2789041095890412, + "grad_norm": 1.9497056007385254, + "learning_rate": 3.2168968318440295e-06, + "loss": 2.1441, + "step": 1167 + }, + { + "epoch": 1.28, + "grad_norm": 1.8635729551315308, + "learning_rate": 3.2148659626320068e-06, + "loss": 2.0748, + "step": 1168 + }, + { + "epoch": 1.2810958904109588, + "grad_norm": 2.0972132682800293, + "learning_rate": 3.212835093419984e-06, + "loss": 2.1843, + "step": 1169 + }, + { + "epoch": 1.2821917808219179, + "grad_norm": 1.9235926866531372, + "learning_rate": 3.2108042242079613e-06, + "loss": 2.1194, + "step": 1170 + }, + { + "epoch": 1.2821917808219179, + "eval_loss": 2.291309118270874, + "eval_runtime": 55.6408, + "eval_samples_per_second": 17.972, + "eval_steps_per_second": 0.755, + "step": 1170 + }, + { + "epoch": 1.2832876712328767, + "grad_norm": 1.8674591779708862, + "learning_rate": 3.2087733549959386e-06, + "loss": 2.1264, + "step": 1171 + }, + { + "epoch": 1.2843835616438355, + "grad_norm": 1.810589075088501, + "learning_rate": 3.206742485783916e-06, + "loss": 2.1781, + "step": 1172 + }, + { + "epoch": 1.2854794520547945, + "grad_norm": 1.9794505834579468, + "learning_rate": 3.2047116165718927e-06, + "loss": 2.1089, + "step": 1173 + }, + { + "epoch": 1.2865753424657536, + "grad_norm": 1.8487070798873901, + "learning_rate": 3.2026807473598704e-06, + "loss": 2.097, + "step": 1174 + }, + { + "epoch": 1.2876712328767124, + "grad_norm": 1.9668854475021362, + "learning_rate": 3.2006498781478473e-06, + "loss": 2.1516, + "step": 1175 + }, + { + "epoch": 1.2887671232876712, + "grad_norm": 1.855817198753357, + "learning_rate": 3.198619008935825e-06, + "loss": 2.1678, + "step": 1176 + }, + { + "epoch": 1.2898630136986302, + "grad_norm": 2.381211757659912, + "learning_rate": 3.196588139723802e-06, + "loss": 2.099, + "step": 1177 + }, + { + "epoch": 1.290958904109589, + "grad_norm": 2.1524250507354736, + "learning_rate": 3.1945572705117795e-06, + "loss": 2.1345, + "step": 1178 + }, + { + "epoch": 1.2920547945205478, + "grad_norm": 1.9120781421661377, + "learning_rate": 3.1925264012997564e-06, + "loss": 2.1586, + "step": 1179 + }, + { + "epoch": 1.2931506849315069, + "grad_norm": 2.153817653656006, + "learning_rate": 3.190495532087734e-06, + "loss": 2.1597, + "step": 1180 + }, + { + "epoch": 1.2931506849315069, + "eval_loss": 2.2936320304870605, + "eval_runtime": 53.6426, + "eval_samples_per_second": 18.642, + "eval_steps_per_second": 0.783, + "step": 1180 + }, + { + "epoch": 1.2942465753424657, + "grad_norm": 2.074873447418213, + "learning_rate": 3.188464662875711e-06, + "loss": 2.1572, + "step": 1181 + }, + { + "epoch": 1.2953424657534247, + "grad_norm": 1.9313570261001587, + "learning_rate": 3.1864337936636886e-06, + "loss": 2.1986, + "step": 1182 + }, + { + "epoch": 1.2964383561643835, + "grad_norm": 1.9836641550064087, + "learning_rate": 3.1844029244516655e-06, + "loss": 2.13, + "step": 1183 + }, + { + "epoch": 1.2975342465753426, + "grad_norm": 2.0893137454986572, + "learning_rate": 3.182372055239643e-06, + "loss": 2.1138, + "step": 1184 + }, + { + "epoch": 1.2986301369863014, + "grad_norm": 2.1667087078094482, + "learning_rate": 3.18034118602762e-06, + "loss": 2.1119, + "step": 1185 + }, + { + "epoch": 1.2997260273972602, + "grad_norm": 2.093961000442505, + "learning_rate": 3.1783103168155973e-06, + "loss": 2.1728, + "step": 1186 + }, + { + "epoch": 1.3008219178082192, + "grad_norm": 1.919267177581787, + "learning_rate": 3.1762794476035746e-06, + "loss": 2.1469, + "step": 1187 + }, + { + "epoch": 1.301917808219178, + "grad_norm": 2.017967939376831, + "learning_rate": 3.174248578391552e-06, + "loss": 2.1307, + "step": 1188 + }, + { + "epoch": 1.303013698630137, + "grad_norm": 1.921632170677185, + "learning_rate": 3.172217709179529e-06, + "loss": 2.13, + "step": 1189 + }, + { + "epoch": 1.3041095890410959, + "grad_norm": 2.0171291828155518, + "learning_rate": 3.1701868399675064e-06, + "loss": 2.135, + "step": 1190 + }, + { + "epoch": 1.3041095890410959, + "eval_loss": 2.2934045791625977, + "eval_runtime": 48.0821, + "eval_samples_per_second": 20.798, + "eval_steps_per_second": 0.874, + "step": 1190 + }, + { + "epoch": 1.305205479452055, + "grad_norm": 1.8685601949691772, + "learning_rate": 3.1681559707554837e-06, + "loss": 2.1329, + "step": 1191 + }, + { + "epoch": 1.3063013698630137, + "grad_norm": 1.8909096717834473, + "learning_rate": 3.166125101543461e-06, + "loss": 2.1004, + "step": 1192 + }, + { + "epoch": 1.3073972602739725, + "grad_norm": 1.798614740371704, + "learning_rate": 3.1640942323314382e-06, + "loss": 2.1555, + "step": 1193 + }, + { + "epoch": 1.3084931506849315, + "grad_norm": 1.8296692371368408, + "learning_rate": 3.1620633631194155e-06, + "loss": 2.1549, + "step": 1194 + }, + { + "epoch": 1.3095890410958904, + "grad_norm": 1.8676908016204834, + "learning_rate": 3.1600324939073928e-06, + "loss": 2.1081, + "step": 1195 + }, + { + "epoch": 1.3106849315068494, + "grad_norm": 1.8682451248168945, + "learning_rate": 3.15800162469537e-06, + "loss": 2.138, + "step": 1196 + }, + { + "epoch": 1.3117808219178082, + "grad_norm": 2.0049314498901367, + "learning_rate": 3.155970755483347e-06, + "loss": 2.144, + "step": 1197 + }, + { + "epoch": 1.3128767123287672, + "grad_norm": 2.00691556930542, + "learning_rate": 3.1539398862713246e-06, + "loss": 2.1539, + "step": 1198 + }, + { + "epoch": 1.313972602739726, + "grad_norm": 1.8521753549575806, + "learning_rate": 3.1519090170593015e-06, + "loss": 2.1597, + "step": 1199 + }, + { + "epoch": 1.3150684931506849, + "grad_norm": 2.0974042415618896, + "learning_rate": 3.149878147847279e-06, + "loss": 2.147, + "step": 1200 + }, + { + "epoch": 1.3150684931506849, + "eval_loss": 2.2891385555267334, + "eval_runtime": 48.1618, + "eval_samples_per_second": 20.763, + "eval_steps_per_second": 0.872, + "step": 1200 + }, + { + "epoch": 1.316164383561644, + "grad_norm": 1.9671052694320679, + "learning_rate": 3.147847278635256e-06, + "loss": 2.1838, + "step": 1201 + }, + { + "epoch": 1.3172602739726027, + "grad_norm": 1.9421573877334595, + "learning_rate": 3.1458164094232337e-06, + "loss": 2.2093, + "step": 1202 + }, + { + "epoch": 1.3183561643835615, + "grad_norm": 1.917198657989502, + "learning_rate": 3.1437855402112106e-06, + "loss": 2.136, + "step": 1203 + }, + { + "epoch": 1.3194520547945205, + "grad_norm": 1.8664982318878174, + "learning_rate": 3.1417546709991883e-06, + "loss": 2.089, + "step": 1204 + }, + { + "epoch": 1.3205479452054796, + "grad_norm": 2.043558359146118, + "learning_rate": 3.139723801787165e-06, + "loss": 2.159, + "step": 1205 + }, + { + "epoch": 1.3216438356164384, + "grad_norm": 1.8314157724380493, + "learning_rate": 3.137692932575143e-06, + "loss": 2.13, + "step": 1206 + }, + { + "epoch": 1.3227397260273972, + "grad_norm": 1.907884120941162, + "learning_rate": 3.1356620633631197e-06, + "loss": 2.1746, + "step": 1207 + }, + { + "epoch": 1.3238356164383562, + "grad_norm": 2.0570952892303467, + "learning_rate": 3.1336311941510974e-06, + "loss": 2.1187, + "step": 1208 + }, + { + "epoch": 1.324931506849315, + "grad_norm": 1.832292914390564, + "learning_rate": 3.1316003249390742e-06, + "loss": 2.1425, + "step": 1209 + }, + { + "epoch": 1.3260273972602739, + "grad_norm": 1.8307697772979736, + "learning_rate": 3.129569455727051e-06, + "loss": 2.1749, + "step": 1210 + }, + { + "epoch": 1.3260273972602739, + "eval_loss": 2.2865941524505615, + "eval_runtime": 42.2877, + "eval_samples_per_second": 23.648, + "eval_steps_per_second": 0.993, + "step": 1210 + }, + { + "epoch": 1.3271232876712329, + "grad_norm": 1.8292433023452759, + "learning_rate": 3.1275385865150288e-06, + "loss": 2.1702, + "step": 1211 + }, + { + "epoch": 1.3282191780821917, + "grad_norm": 1.9146367311477661, + "learning_rate": 3.1255077173030056e-06, + "loss": 2.16, + "step": 1212 + }, + { + "epoch": 1.3293150684931507, + "grad_norm": 1.809275507926941, + "learning_rate": 3.1234768480909833e-06, + "loss": 2.157, + "step": 1213 + }, + { + "epoch": 1.3304109589041095, + "grad_norm": 1.939081072807312, + "learning_rate": 3.12144597887896e-06, + "loss": 2.1496, + "step": 1214 + }, + { + "epoch": 1.3315068493150686, + "grad_norm": 1.8665997982025146, + "learning_rate": 3.119415109666938e-06, + "loss": 2.1535, + "step": 1215 + }, + { + "epoch": 1.3326027397260274, + "grad_norm": 1.8455616235733032, + "learning_rate": 3.1173842404549147e-06, + "loss": 2.1845, + "step": 1216 + }, + { + "epoch": 1.3336986301369862, + "grad_norm": 2.050560235977173, + "learning_rate": 3.1153533712428924e-06, + "loss": 2.1461, + "step": 1217 + }, + { + "epoch": 1.3347945205479452, + "grad_norm": 1.8440574407577515, + "learning_rate": 3.1133225020308693e-06, + "loss": 2.1476, + "step": 1218 + }, + { + "epoch": 1.335890410958904, + "grad_norm": 1.8801573514938354, + "learning_rate": 3.111291632818847e-06, + "loss": 2.1721, + "step": 1219 + }, + { + "epoch": 1.336986301369863, + "grad_norm": 1.9227471351623535, + "learning_rate": 3.109260763606824e-06, + "loss": 2.178, + "step": 1220 + }, + { + "epoch": 1.336986301369863, + "eval_loss": 2.2896478176116943, + "eval_runtime": 43.157, + "eval_samples_per_second": 23.171, + "eval_steps_per_second": 0.973, + "step": 1220 + }, + { + "epoch": 1.3380821917808219, + "grad_norm": 1.9979732036590576, + "learning_rate": 3.1072298943948015e-06, + "loss": 2.148, + "step": 1221 + }, + { + "epoch": 1.339178082191781, + "grad_norm": 1.9481284618377686, + "learning_rate": 3.1051990251827784e-06, + "loss": 2.165, + "step": 1222 + }, + { + "epoch": 1.3402739726027397, + "grad_norm": 1.8830820322036743, + "learning_rate": 3.1031681559707557e-06, + "loss": 2.1239, + "step": 1223 + }, + { + "epoch": 1.3413698630136985, + "grad_norm": 1.890512466430664, + "learning_rate": 3.101137286758733e-06, + "loss": 2.1761, + "step": 1224 + }, + { + "epoch": 1.3424657534246576, + "grad_norm": 2.0024659633636475, + "learning_rate": 3.09910641754671e-06, + "loss": 2.179, + "step": 1225 + }, + { + "epoch": 1.3435616438356164, + "grad_norm": 1.8659477233886719, + "learning_rate": 3.0970755483346875e-06, + "loss": 2.1722, + "step": 1226 + }, + { + "epoch": 1.3446575342465754, + "grad_norm": 1.93405020236969, + "learning_rate": 3.0950446791226648e-06, + "loss": 2.1545, + "step": 1227 + }, + { + "epoch": 1.3457534246575342, + "grad_norm": 1.9514124393463135, + "learning_rate": 3.093013809910642e-06, + "loss": 2.187, + "step": 1228 + }, + { + "epoch": 1.3468493150684933, + "grad_norm": 1.9571877717971802, + "learning_rate": 3.0909829406986193e-06, + "loss": 2.1507, + "step": 1229 + }, + { + "epoch": 1.347945205479452, + "grad_norm": 1.9434577226638794, + "learning_rate": 3.0889520714865966e-06, + "loss": 2.1953, + "step": 1230 + }, + { + "epoch": 1.347945205479452, + "eval_loss": 2.289128065109253, + "eval_runtime": 46.7151, + "eval_samples_per_second": 21.406, + "eval_steps_per_second": 0.899, + "step": 1230 + }, + { + "epoch": 1.3490410958904109, + "grad_norm": 1.8294577598571777, + "learning_rate": 3.086921202274574e-06, + "loss": 2.1684, + "step": 1231 + }, + { + "epoch": 1.35013698630137, + "grad_norm": 1.9432605504989624, + "learning_rate": 3.084890333062551e-06, + "loss": 2.1393, + "step": 1232 + }, + { + "epoch": 1.3512328767123287, + "grad_norm": 1.9030417203903198, + "learning_rate": 3.0828594638505284e-06, + "loss": 2.1181, + "step": 1233 + }, + { + "epoch": 1.3523287671232875, + "grad_norm": 1.9233827590942383, + "learning_rate": 3.0808285946385057e-06, + "loss": 2.1479, + "step": 1234 + }, + { + "epoch": 1.3534246575342466, + "grad_norm": 1.9549237489700317, + "learning_rate": 3.078797725426483e-06, + "loss": 2.1253, + "step": 1235 + }, + { + "epoch": 1.3545205479452056, + "grad_norm": 2.0584867000579834, + "learning_rate": 3.07676685621446e-06, + "loss": 2.0846, + "step": 1236 + }, + { + "epoch": 1.3556164383561644, + "grad_norm": 1.7543913125991821, + "learning_rate": 3.0747359870024375e-06, + "loss": 2.1495, + "step": 1237 + }, + { + "epoch": 1.3567123287671232, + "grad_norm": 1.750698447227478, + "learning_rate": 3.0727051177904144e-06, + "loss": 2.1179, + "step": 1238 + }, + { + "epoch": 1.3578082191780823, + "grad_norm": 1.9867119789123535, + "learning_rate": 3.070674248578392e-06, + "loss": 2.1313, + "step": 1239 + }, + { + "epoch": 1.358904109589041, + "grad_norm": 1.8228135108947754, + "learning_rate": 3.068643379366369e-06, + "loss": 2.158, + "step": 1240 + }, + { + "epoch": 1.358904109589041, + "eval_loss": 2.2861993312835693, + "eval_runtime": 44.081, + "eval_samples_per_second": 22.686, + "eval_steps_per_second": 0.953, + "step": 1240 + }, + { + "epoch": 1.3599999999999999, + "grad_norm": 1.8551791906356812, + "learning_rate": 3.0666125101543466e-06, + "loss": 2.129, + "step": 1241 + }, + { + "epoch": 1.361095890410959, + "grad_norm": 1.8729921579360962, + "learning_rate": 3.0645816409423235e-06, + "loss": 2.1069, + "step": 1242 + }, + { + "epoch": 1.362191780821918, + "grad_norm": 1.8203959465026855, + "learning_rate": 3.062550771730301e-06, + "loss": 2.1509, + "step": 1243 + }, + { + "epoch": 1.3632876712328768, + "grad_norm": 1.9498761892318726, + "learning_rate": 3.060519902518278e-06, + "loss": 2.1336, + "step": 1244 + }, + { + "epoch": 1.3643835616438356, + "grad_norm": 2.134521245956421, + "learning_rate": 3.0584890333062557e-06, + "loss": 2.1294, + "step": 1245 + }, + { + "epoch": 1.3654794520547946, + "grad_norm": 1.9321913719177246, + "learning_rate": 3.0564581640942326e-06, + "loss": 2.0823, + "step": 1246 + }, + { + "epoch": 1.3665753424657534, + "grad_norm": 1.8948324918746948, + "learning_rate": 3.0544272948822094e-06, + "loss": 2.1425, + "step": 1247 + }, + { + "epoch": 1.3676712328767122, + "grad_norm": 2.2063205242156982, + "learning_rate": 3.052396425670187e-06, + "loss": 2.1242, + "step": 1248 + }, + { + "epoch": 1.3687671232876713, + "grad_norm": 2.0060935020446777, + "learning_rate": 3.050365556458164e-06, + "loss": 2.1586, + "step": 1249 + }, + { + "epoch": 1.36986301369863, + "grad_norm": 1.8441938161849976, + "learning_rate": 3.0483346872461417e-06, + "loss": 2.1305, + "step": 1250 + }, + { + "epoch": 1.36986301369863, + "eval_loss": 2.2872142791748047, + "eval_runtime": 43.4998, + "eval_samples_per_second": 22.989, + "eval_steps_per_second": 0.966, + "step": 1250 + }, + { + "epoch": 1.370958904109589, + "grad_norm": 2.019545793533325, + "learning_rate": 3.0463038180341185e-06, + "loss": 2.1257, + "step": 1251 + }, + { + "epoch": 1.372054794520548, + "grad_norm": 1.981020212173462, + "learning_rate": 3.0442729488220962e-06, + "loss": 2.155, + "step": 1252 + }, + { + "epoch": 1.373150684931507, + "grad_norm": 1.9707001447677612, + "learning_rate": 3.042242079610073e-06, + "loss": 2.1671, + "step": 1253 + }, + { + "epoch": 1.3742465753424657, + "grad_norm": 1.8645458221435547, + "learning_rate": 3.0402112103980508e-06, + "loss": 2.1655, + "step": 1254 + }, + { + "epoch": 1.3753424657534246, + "grad_norm": 1.863911747932434, + "learning_rate": 3.0381803411860276e-06, + "loss": 2.0917, + "step": 1255 + }, + { + "epoch": 1.3764383561643836, + "grad_norm": 2.1079251766204834, + "learning_rate": 3.0361494719740053e-06, + "loss": 2.119, + "step": 1256 + }, + { + "epoch": 1.3775342465753424, + "grad_norm": 1.8242024183273315, + "learning_rate": 3.034118602761982e-06, + "loss": 2.1398, + "step": 1257 + }, + { + "epoch": 1.3786301369863014, + "grad_norm": 1.8663142919540405, + "learning_rate": 3.03208773354996e-06, + "loss": 2.1405, + "step": 1258 + }, + { + "epoch": 1.3797260273972602, + "grad_norm": 1.9437193870544434, + "learning_rate": 3.0300568643379367e-06, + "loss": 2.1156, + "step": 1259 + }, + { + "epoch": 1.3808219178082193, + "grad_norm": 2.0837693214416504, + "learning_rate": 3.028025995125914e-06, + "loss": 2.1731, + "step": 1260 + }, + { + "epoch": 1.3808219178082193, + "eval_loss": 2.283296823501587, + "eval_runtime": 45.341, + "eval_samples_per_second": 22.055, + "eval_steps_per_second": 0.926, + "step": 1260 + }, + { + "epoch": 1.381917808219178, + "grad_norm": 1.9430493116378784, + "learning_rate": 3.0259951259138913e-06, + "loss": 2.1623, + "step": 1261 + }, + { + "epoch": 1.383013698630137, + "grad_norm": 1.825675368309021, + "learning_rate": 3.0239642567018686e-06, + "loss": 2.1022, + "step": 1262 + }, + { + "epoch": 1.384109589041096, + "grad_norm": 2.0050806999206543, + "learning_rate": 3.021933387489846e-06, + "loss": 2.1443, + "step": 1263 + }, + { + "epoch": 1.3852054794520547, + "grad_norm": 1.9548894166946411, + "learning_rate": 3.019902518277823e-06, + "loss": 2.0983, + "step": 1264 + }, + { + "epoch": 1.3863013698630138, + "grad_norm": 1.9097814559936523, + "learning_rate": 3.0178716490658004e-06, + "loss": 2.1488, + "step": 1265 + }, + { + "epoch": 1.3873972602739726, + "grad_norm": 1.8977514505386353, + "learning_rate": 3.0158407798537777e-06, + "loss": 2.1641, + "step": 1266 + }, + { + "epoch": 1.3884931506849316, + "grad_norm": 1.9222278594970703, + "learning_rate": 3.013809910641755e-06, + "loss": 2.1043, + "step": 1267 + }, + { + "epoch": 1.3895890410958904, + "grad_norm": 1.9333430528640747, + "learning_rate": 3.0117790414297322e-06, + "loss": 2.1152, + "step": 1268 + }, + { + "epoch": 1.3906849315068492, + "grad_norm": 1.8405439853668213, + "learning_rate": 3.0097481722177095e-06, + "loss": 2.1046, + "step": 1269 + }, + { + "epoch": 1.3917808219178083, + "grad_norm": 2.0583181381225586, + "learning_rate": 3.0077173030056868e-06, + "loss": 2.1378, + "step": 1270 + }, + { + "epoch": 1.3917808219178083, + "eval_loss": 2.286198616027832, + "eval_runtime": 46.6914, + "eval_samples_per_second": 21.417, + "eval_steps_per_second": 0.9, + "step": 1270 + }, + { + "epoch": 1.392876712328767, + "grad_norm": 1.8485472202301025, + "learning_rate": 3.005686433793664e-06, + "loss": 2.1184, + "step": 1271 + }, + { + "epoch": 1.393972602739726, + "grad_norm": 2.0430188179016113, + "learning_rate": 3.0036555645816413e-06, + "loss": 2.1518, + "step": 1272 + }, + { + "epoch": 1.395068493150685, + "grad_norm": 2.307474374771118, + "learning_rate": 3.001624695369618e-06, + "loss": 2.1288, + "step": 1273 + }, + { + "epoch": 1.396164383561644, + "grad_norm": 1.85003662109375, + "learning_rate": 2.999593826157596e-06, + "loss": 2.1573, + "step": 1274 + }, + { + "epoch": 1.3972602739726028, + "grad_norm": 2.047619581222534, + "learning_rate": 2.9975629569455727e-06, + "loss": 2.1285, + "step": 1275 + }, + { + "epoch": 1.3983561643835616, + "grad_norm": 2.013644218444824, + "learning_rate": 2.9955320877335504e-06, + "loss": 2.1603, + "step": 1276 + }, + { + "epoch": 1.3994520547945206, + "grad_norm": 1.8814456462860107, + "learning_rate": 2.9935012185215273e-06, + "loss": 2.1333, + "step": 1277 + }, + { + "epoch": 1.4005479452054794, + "grad_norm": 2.0338361263275146, + "learning_rate": 2.991470349309505e-06, + "loss": 2.1564, + "step": 1278 + }, + { + "epoch": 1.4016438356164382, + "grad_norm": 1.972992181777954, + "learning_rate": 2.989439480097482e-06, + "loss": 2.1305, + "step": 1279 + }, + { + "epoch": 1.4027397260273973, + "grad_norm": 1.845928430557251, + "learning_rate": 2.9874086108854595e-06, + "loss": 2.1384, + "step": 1280 + }, + { + "epoch": 1.4027397260273973, + "eval_loss": 2.28485107421875, + "eval_runtime": 46.9757, + "eval_samples_per_second": 21.288, + "eval_steps_per_second": 0.894, + "step": 1280 + }, + { + "epoch": 1.403835616438356, + "grad_norm": 1.7601349353790283, + "learning_rate": 2.9853777416734364e-06, + "loss": 2.1219, + "step": 1281 + }, + { + "epoch": 1.4049315068493151, + "grad_norm": 2.2385566234588623, + "learning_rate": 2.983346872461414e-06, + "loss": 2.1581, + "step": 1282 + }, + { + "epoch": 1.406027397260274, + "grad_norm": 2.019174098968506, + "learning_rate": 2.981316003249391e-06, + "loss": 2.1753, + "step": 1283 + }, + { + "epoch": 1.407123287671233, + "grad_norm": 1.9927541017532349, + "learning_rate": 2.9792851340373678e-06, + "loss": 2.147, + "step": 1284 + }, + { + "epoch": 1.4082191780821918, + "grad_norm": 2.5143957138061523, + "learning_rate": 2.9772542648253455e-06, + "loss": 2.138, + "step": 1285 + }, + { + "epoch": 1.4093150684931506, + "grad_norm": 1.9925073385238647, + "learning_rate": 2.9752233956133223e-06, + "loss": 2.1642, + "step": 1286 + }, + { + "epoch": 1.4104109589041096, + "grad_norm": 2.1739184856414795, + "learning_rate": 2.9731925264013e-06, + "loss": 2.1258, + "step": 1287 + }, + { + "epoch": 1.4115068493150684, + "grad_norm": 2.1719982624053955, + "learning_rate": 2.971161657189277e-06, + "loss": 2.1323, + "step": 1288 + }, + { + "epoch": 1.4126027397260275, + "grad_norm": 1.9897956848144531, + "learning_rate": 2.9691307879772546e-06, + "loss": 2.1286, + "step": 1289 + }, + { + "epoch": 1.4136986301369863, + "grad_norm": 1.997477412223816, + "learning_rate": 2.9670999187652314e-06, + "loss": 2.125, + "step": 1290 + }, + { + "epoch": 1.4136986301369863, + "eval_loss": 2.2839529514312744, + "eval_runtime": 44.5005, + "eval_samples_per_second": 22.472, + "eval_steps_per_second": 0.944, + "step": 1290 + }, + { + "epoch": 1.4147945205479453, + "grad_norm": 1.9529412984848022, + "learning_rate": 2.965069049553209e-06, + "loss": 2.1074, + "step": 1291 + }, + { + "epoch": 1.415890410958904, + "grad_norm": 1.900692343711853, + "learning_rate": 2.963038180341186e-06, + "loss": 2.1796, + "step": 1292 + }, + { + "epoch": 1.416986301369863, + "grad_norm": 2.0269863605499268, + "learning_rate": 2.9610073111291637e-06, + "loss": 2.0998, + "step": 1293 + }, + { + "epoch": 1.418082191780822, + "grad_norm": 2.0290050506591797, + "learning_rate": 2.9589764419171405e-06, + "loss": 2.1638, + "step": 1294 + }, + { + "epoch": 1.4191780821917808, + "grad_norm": 1.8717535734176636, + "learning_rate": 2.9569455727051182e-06, + "loss": 2.1756, + "step": 1295 + }, + { + "epoch": 1.4202739726027398, + "grad_norm": 1.9470241069793701, + "learning_rate": 2.954914703493095e-06, + "loss": 2.116, + "step": 1296 + }, + { + "epoch": 1.4213698630136986, + "grad_norm": 2.1749250888824463, + "learning_rate": 2.9528838342810724e-06, + "loss": 2.1381, + "step": 1297 + }, + { + "epoch": 1.4224657534246576, + "grad_norm": 1.905742883682251, + "learning_rate": 2.9508529650690496e-06, + "loss": 2.118, + "step": 1298 + }, + { + "epoch": 1.4235616438356165, + "grad_norm": 2.1214921474456787, + "learning_rate": 2.948822095857027e-06, + "loss": 2.1223, + "step": 1299 + }, + { + "epoch": 1.4246575342465753, + "grad_norm": 1.894509196281433, + "learning_rate": 2.946791226645004e-06, + "loss": 2.1457, + "step": 1300 + }, + { + "epoch": 1.4246575342465753, + "eval_loss": 2.2824268341064453, + "eval_runtime": 47.8031, + "eval_samples_per_second": 20.919, + "eval_steps_per_second": 0.879, + "step": 1300 + }, + { + "epoch": 1.4257534246575343, + "grad_norm": 1.9588708877563477, + "learning_rate": 2.9447603574329815e-06, + "loss": 2.1113, + "step": 1301 + }, + { + "epoch": 1.426849315068493, + "grad_norm": 1.9709562063217163, + "learning_rate": 2.9427294882209587e-06, + "loss": 2.1086, + "step": 1302 + }, + { + "epoch": 1.427945205479452, + "grad_norm": 1.82259202003479, + "learning_rate": 2.940698619008936e-06, + "loss": 2.118, + "step": 1303 + }, + { + "epoch": 1.429041095890411, + "grad_norm": 1.9776451587677002, + "learning_rate": 2.9386677497969133e-06, + "loss": 2.1386, + "step": 1304 + }, + { + "epoch": 1.43013698630137, + "grad_norm": 2.147463321685791, + "learning_rate": 2.9366368805848906e-06, + "loss": 2.1309, + "step": 1305 + }, + { + "epoch": 1.4312328767123288, + "grad_norm": 1.9845961332321167, + "learning_rate": 2.934606011372868e-06, + "loss": 2.1586, + "step": 1306 + }, + { + "epoch": 1.4323287671232876, + "grad_norm": 1.9153180122375488, + "learning_rate": 2.932575142160845e-06, + "loss": 2.1588, + "step": 1307 + }, + { + "epoch": 1.4334246575342466, + "grad_norm": 1.9372929334640503, + "learning_rate": 2.9305442729488224e-06, + "loss": 2.1351, + "step": 1308 + }, + { + "epoch": 1.4345205479452054, + "grad_norm": 1.852669358253479, + "learning_rate": 2.9285134037367997e-06, + "loss": 2.1394, + "step": 1309 + }, + { + "epoch": 1.4356164383561643, + "grad_norm": 1.934251308441162, + "learning_rate": 2.9264825345247765e-06, + "loss": 2.1364, + "step": 1310 + }, + { + "epoch": 1.4356164383561643, + "eval_loss": 2.280508279800415, + "eval_runtime": 43.1734, + "eval_samples_per_second": 23.162, + "eval_steps_per_second": 0.973, + "step": 1310 + }, + { + "epoch": 1.4367123287671233, + "grad_norm": 1.8886998891830444, + "learning_rate": 2.9244516653127542e-06, + "loss": 2.1452, + "step": 1311 + }, + { + "epoch": 1.4378082191780823, + "grad_norm": 2.275813102722168, + "learning_rate": 2.922420796100731e-06, + "loss": 2.1465, + "step": 1312 + }, + { + "epoch": 1.4389041095890411, + "grad_norm": 1.9017852544784546, + "learning_rate": 2.9203899268887088e-06, + "loss": 2.0785, + "step": 1313 + }, + { + "epoch": 1.44, + "grad_norm": 1.8482435941696167, + "learning_rate": 2.9183590576766856e-06, + "loss": 2.1892, + "step": 1314 + }, + { + "epoch": 1.441095890410959, + "grad_norm": 2.2122418880462646, + "learning_rate": 2.9163281884646633e-06, + "loss": 2.1058, + "step": 1315 + }, + { + "epoch": 1.4421917808219178, + "grad_norm": 1.9721672534942627, + "learning_rate": 2.91429731925264e-06, + "loss": 2.1391, + "step": 1316 + }, + { + "epoch": 1.4432876712328766, + "grad_norm": 1.9381605386734009, + "learning_rate": 2.912266450040618e-06, + "loss": 2.1651, + "step": 1317 + }, + { + "epoch": 1.4443835616438356, + "grad_norm": 2.274630069732666, + "learning_rate": 2.9102355808285947e-06, + "loss": 2.1434, + "step": 1318 + }, + { + "epoch": 1.4454794520547944, + "grad_norm": 1.91957688331604, + "learning_rate": 2.9082047116165724e-06, + "loss": 2.1218, + "step": 1319 + }, + { + "epoch": 1.4465753424657535, + "grad_norm": 2.042171001434326, + "learning_rate": 2.9061738424045493e-06, + "loss": 2.1233, + "step": 1320 + }, + { + "epoch": 1.4465753424657535, + "eval_loss": 2.280571699142456, + "eval_runtime": 48.425, + "eval_samples_per_second": 20.65, + "eval_steps_per_second": 0.867, + "step": 1320 + }, + { + "epoch": 1.4476712328767123, + "grad_norm": 2.0493276119232178, + "learning_rate": 2.904142973192527e-06, + "loss": 2.1241, + "step": 1321 + }, + { + "epoch": 1.4487671232876713, + "grad_norm": 1.8865023851394653, + "learning_rate": 2.902112103980504e-06, + "loss": 2.1826, + "step": 1322 + }, + { + "epoch": 1.4498630136986301, + "grad_norm": 1.9723960161209106, + "learning_rate": 2.900081234768481e-06, + "loss": 2.1508, + "step": 1323 + }, + { + "epoch": 1.450958904109589, + "grad_norm": 1.8321057558059692, + "learning_rate": 2.8980503655564584e-06, + "loss": 2.1319, + "step": 1324 + }, + { + "epoch": 1.452054794520548, + "grad_norm": 1.97796630859375, + "learning_rate": 2.8960194963444357e-06, + "loss": 2.1289, + "step": 1325 + }, + { + "epoch": 1.4531506849315068, + "grad_norm": 1.7793136835098267, + "learning_rate": 2.893988627132413e-06, + "loss": 2.1059, + "step": 1326 + }, + { + "epoch": 1.4542465753424658, + "grad_norm": 1.878381609916687, + "learning_rate": 2.8919577579203902e-06, + "loss": 2.125, + "step": 1327 + }, + { + "epoch": 1.4553424657534246, + "grad_norm": 1.8854790925979614, + "learning_rate": 2.8899268887083675e-06, + "loss": 2.0908, + "step": 1328 + }, + { + "epoch": 1.4564383561643837, + "grad_norm": 1.940931797027588, + "learning_rate": 2.8878960194963448e-06, + "loss": 2.1338, + "step": 1329 + }, + { + "epoch": 1.4575342465753425, + "grad_norm": 1.8014280796051025, + "learning_rate": 2.885865150284322e-06, + "loss": 2.1384, + "step": 1330 + }, + { + "epoch": 1.4575342465753425, + "eval_loss": 2.2815349102020264, + "eval_runtime": 48.7646, + "eval_samples_per_second": 20.507, + "eval_steps_per_second": 0.861, + "step": 1330 + }, + { + "epoch": 1.4586301369863013, + "grad_norm": 1.9306361675262451, + "learning_rate": 2.8838342810722993e-06, + "loss": 2.1562, + "step": 1331 + }, + { + "epoch": 1.4597260273972603, + "grad_norm": 1.902762770652771, + "learning_rate": 2.8818034118602766e-06, + "loss": 2.1183, + "step": 1332 + }, + { + "epoch": 1.4608219178082191, + "grad_norm": 1.980449914932251, + "learning_rate": 2.879772542648254e-06, + "loss": 2.1066, + "step": 1333 + }, + { + "epoch": 1.4619178082191782, + "grad_norm": 2.0774567127227783, + "learning_rate": 2.8777416734362307e-06, + "loss": 2.157, + "step": 1334 + }, + { + "epoch": 1.463013698630137, + "grad_norm": 1.8203723430633545, + "learning_rate": 2.8757108042242084e-06, + "loss": 2.0792, + "step": 1335 + }, + { + "epoch": 1.464109589041096, + "grad_norm": 1.8849077224731445, + "learning_rate": 2.8736799350121853e-06, + "loss": 2.1917, + "step": 1336 + }, + { + "epoch": 1.4652054794520548, + "grad_norm": 1.8782836198806763, + "learning_rate": 2.871649065800163e-06, + "loss": 2.1744, + "step": 1337 + }, + { + "epoch": 1.4663013698630136, + "grad_norm": 1.8628056049346924, + "learning_rate": 2.86961819658814e-06, + "loss": 2.1788, + "step": 1338 + }, + { + "epoch": 1.4673972602739727, + "grad_norm": 1.865370750427246, + "learning_rate": 2.8675873273761175e-06, + "loss": 2.1159, + "step": 1339 + }, + { + "epoch": 1.4684931506849315, + "grad_norm": 1.8673044443130493, + "learning_rate": 2.8655564581640944e-06, + "loss": 2.1447, + "step": 1340 + }, + { + "epoch": 1.4684931506849315, + "eval_loss": 2.282686471939087, + "eval_runtime": 49.85, + "eval_samples_per_second": 20.06, + "eval_steps_per_second": 0.843, + "step": 1340 + }, + { + "epoch": 1.4695890410958903, + "grad_norm": 1.8928618431091309, + "learning_rate": 2.863525588952072e-06, + "loss": 2.1362, + "step": 1341 + }, + { + "epoch": 1.4706849315068493, + "grad_norm": 1.8964238166809082, + "learning_rate": 2.861494719740049e-06, + "loss": 2.1628, + "step": 1342 + }, + { + "epoch": 1.4717808219178083, + "grad_norm": 1.9889185428619385, + "learning_rate": 2.8594638505280266e-06, + "loss": 2.0922, + "step": 1343 + }, + { + "epoch": 1.4728767123287672, + "grad_norm": 1.9503910541534424, + "learning_rate": 2.8574329813160035e-06, + "loss": 2.0979, + "step": 1344 + }, + { + "epoch": 1.473972602739726, + "grad_norm": 1.8889869451522827, + "learning_rate": 2.855402112103981e-06, + "loss": 2.1329, + "step": 1345 + }, + { + "epoch": 1.475068493150685, + "grad_norm": 1.7875893115997314, + "learning_rate": 2.853371242891958e-06, + "loss": 2.123, + "step": 1346 + }, + { + "epoch": 1.4761643835616438, + "grad_norm": 1.9358372688293457, + "learning_rate": 2.851340373679935e-06, + "loss": 2.1284, + "step": 1347 + }, + { + "epoch": 1.4772602739726026, + "grad_norm": 2.125243663787842, + "learning_rate": 2.8493095044679126e-06, + "loss": 2.1386, + "step": 1348 + }, + { + "epoch": 1.4783561643835617, + "grad_norm": 1.927788257598877, + "learning_rate": 2.8472786352558894e-06, + "loss": 2.1835, + "step": 1349 + }, + { + "epoch": 1.4794520547945205, + "grad_norm": 1.9361577033996582, + "learning_rate": 2.845247766043867e-06, + "loss": 2.1411, + "step": 1350 + }, + { + "epoch": 1.4794520547945205, + "eval_loss": 2.281172275543213, + "eval_runtime": 47.4205, + "eval_samples_per_second": 21.088, + "eval_steps_per_second": 0.886, + "step": 1350 + }, + { + "epoch": 1.4805479452054795, + "grad_norm": 1.847364068031311, + "learning_rate": 2.843216896831844e-06, + "loss": 2.111, + "step": 1351 + }, + { + "epoch": 1.4816438356164383, + "grad_norm": 1.9927674531936646, + "learning_rate": 2.8411860276198217e-06, + "loss": 2.107, + "step": 1352 + }, + { + "epoch": 1.4827397260273973, + "grad_norm": 1.8565406799316406, + "learning_rate": 2.8391551584077985e-06, + "loss": 2.12, + "step": 1353 + }, + { + "epoch": 1.4838356164383562, + "grad_norm": 1.9477596282958984, + "learning_rate": 2.8371242891957762e-06, + "loss": 2.1661, + "step": 1354 + }, + { + "epoch": 1.484931506849315, + "grad_norm": 1.9354814291000366, + "learning_rate": 2.835093419983753e-06, + "loss": 2.168, + "step": 1355 + }, + { + "epoch": 1.486027397260274, + "grad_norm": 2.0603487491607666, + "learning_rate": 2.833062550771731e-06, + "loss": 2.1338, + "step": 1356 + }, + { + "epoch": 1.4871232876712328, + "grad_norm": 1.987990379333496, + "learning_rate": 2.8310316815597076e-06, + "loss": 2.1282, + "step": 1357 + }, + { + "epoch": 1.4882191780821918, + "grad_norm": 1.895529866218567, + "learning_rate": 2.8290008123476853e-06, + "loss": 2.1032, + "step": 1358 + }, + { + "epoch": 1.4893150684931507, + "grad_norm": 1.917994737625122, + "learning_rate": 2.826969943135662e-06, + "loss": 2.149, + "step": 1359 + }, + { + "epoch": 1.4904109589041097, + "grad_norm": 1.9238080978393555, + "learning_rate": 2.8249390739236395e-06, + "loss": 2.1275, + "step": 1360 + }, + { + "epoch": 1.4904109589041097, + "eval_loss": 2.280546188354492, + "eval_runtime": 46.9985, + "eval_samples_per_second": 21.277, + "eval_steps_per_second": 0.894, + "step": 1360 + }, + { + "epoch": 1.4915068493150685, + "grad_norm": 1.947768211364746, + "learning_rate": 2.8229082047116167e-06, + "loss": 2.1282, + "step": 1361 + }, + { + "epoch": 1.4926027397260273, + "grad_norm": 1.7888189554214478, + "learning_rate": 2.820877335499594e-06, + "loss": 2.1134, + "step": 1362 + }, + { + "epoch": 1.4936986301369863, + "grad_norm": 1.8480347394943237, + "learning_rate": 2.8188464662875713e-06, + "loss": 2.1194, + "step": 1363 + }, + { + "epoch": 1.4947945205479451, + "grad_norm": 1.8146322965621948, + "learning_rate": 2.8168155970755486e-06, + "loss": 2.0781, + "step": 1364 + }, + { + "epoch": 1.4958904109589042, + "grad_norm": 1.7994698286056519, + "learning_rate": 2.814784727863526e-06, + "loss": 2.1027, + "step": 1365 + }, + { + "epoch": 1.496986301369863, + "grad_norm": 1.9375026226043701, + "learning_rate": 2.812753858651503e-06, + "loss": 2.1273, + "step": 1366 + }, + { + "epoch": 1.498082191780822, + "grad_norm": 1.9606287479400635, + "learning_rate": 2.8107229894394804e-06, + "loss": 2.1438, + "step": 1367 + }, + { + "epoch": 1.4991780821917808, + "grad_norm": 2.1815521717071533, + "learning_rate": 2.8086921202274577e-06, + "loss": 2.1176, + "step": 1368 + }, + { + "epoch": 1.5002739726027396, + "grad_norm": 1.8796950578689575, + "learning_rate": 2.806661251015435e-06, + "loss": 2.096, + "step": 1369 + }, + { + "epoch": 1.5013698630136987, + "grad_norm": 2.016091823577881, + "learning_rate": 2.8046303818034122e-06, + "loss": 2.1451, + "step": 1370 + }, + { + "epoch": 1.5013698630136987, + "eval_loss": 2.280881643295288, + "eval_runtime": 53.6471, + "eval_samples_per_second": 18.64, + "eval_steps_per_second": 0.783, + "step": 1370 + }, + { + "epoch": 1.5024657534246575, + "grad_norm": 1.9303081035614014, + "learning_rate": 2.802599512591389e-06, + "loss": 2.1736, + "step": 1371 + }, + { + "epoch": 1.5035616438356163, + "grad_norm": 1.9831101894378662, + "learning_rate": 2.8005686433793668e-06, + "loss": 2.2147, + "step": 1372 + }, + { + "epoch": 1.5046575342465753, + "grad_norm": 1.7864863872528076, + "learning_rate": 2.7985377741673436e-06, + "loss": 2.1231, + "step": 1373 + }, + { + "epoch": 1.5057534246575344, + "grad_norm": 1.8945375680923462, + "learning_rate": 2.7965069049553213e-06, + "loss": 2.0866, + "step": 1374 + }, + { + "epoch": 1.5068493150684932, + "grad_norm": 1.9881761074066162, + "learning_rate": 2.794476035743298e-06, + "loss": 2.1398, + "step": 1375 + }, + { + "epoch": 1.507945205479452, + "grad_norm": 1.9888391494750977, + "learning_rate": 2.792445166531276e-06, + "loss": 2.172, + "step": 1376 + }, + { + "epoch": 1.509041095890411, + "grad_norm": 1.9845142364501953, + "learning_rate": 2.7904142973192527e-06, + "loss": 2.1378, + "step": 1377 + }, + { + "epoch": 1.5101369863013698, + "grad_norm": 2.0640218257904053, + "learning_rate": 2.7883834281072304e-06, + "loss": 2.16, + "step": 1378 + }, + { + "epoch": 1.5112328767123286, + "grad_norm": 1.9453262090682983, + "learning_rate": 2.7863525588952073e-06, + "loss": 2.1112, + "step": 1379 + }, + { + "epoch": 1.5123287671232877, + "grad_norm": 2.0385119915008545, + "learning_rate": 2.784321689683185e-06, + "loss": 2.1068, + "step": 1380 + }, + { + "epoch": 1.5123287671232877, + "eval_loss": 2.2804832458496094, + "eval_runtime": 50.1317, + "eval_samples_per_second": 19.947, + "eval_steps_per_second": 0.838, + "step": 1380 + }, + { + "epoch": 1.5134246575342467, + "grad_norm": 1.9497289657592773, + "learning_rate": 2.782290820471162e-06, + "loss": 2.1017, + "step": 1381 + }, + { + "epoch": 1.5145205479452055, + "grad_norm": 1.9402867555618286, + "learning_rate": 2.7802599512591395e-06, + "loss": 2.1141, + "step": 1382 + }, + { + "epoch": 1.5156164383561643, + "grad_norm": 2.008141279220581, + "learning_rate": 2.7782290820471164e-06, + "loss": 2.1601, + "step": 1383 + }, + { + "epoch": 1.5167123287671234, + "grad_norm": 1.9685825109481812, + "learning_rate": 2.7761982128350932e-06, + "loss": 2.1494, + "step": 1384 + }, + { + "epoch": 1.5178082191780822, + "grad_norm": 1.9075162410736084, + "learning_rate": 2.774167343623071e-06, + "loss": 2.147, + "step": 1385 + }, + { + "epoch": 1.518904109589041, + "grad_norm": 1.8381123542785645, + "learning_rate": 2.772136474411048e-06, + "loss": 2.1772, + "step": 1386 + }, + { + "epoch": 1.52, + "grad_norm": 1.806057095527649, + "learning_rate": 2.7701056051990255e-06, + "loss": 2.1077, + "step": 1387 + }, + { + "epoch": 1.521095890410959, + "grad_norm": 1.9030957221984863, + "learning_rate": 2.7680747359870023e-06, + "loss": 2.0735, + "step": 1388 + }, + { + "epoch": 1.5221917808219179, + "grad_norm": 1.9967273473739624, + "learning_rate": 2.76604386677498e-06, + "loss": 2.155, + "step": 1389 + }, + { + "epoch": 1.5232876712328767, + "grad_norm": 1.864998698234558, + "learning_rate": 2.764012997562957e-06, + "loss": 2.1319, + "step": 1390 + }, + { + "epoch": 1.5232876712328767, + "eval_loss": 2.279405355453491, + "eval_runtime": 49.6092, + "eval_samples_per_second": 20.158, + "eval_steps_per_second": 0.847, + "step": 1390 + }, + { + "epoch": 1.5243835616438357, + "grad_norm": 1.831916093826294, + "learning_rate": 2.7619821283509346e-06, + "loss": 2.1518, + "step": 1391 + }, + { + "epoch": 1.5254794520547945, + "grad_norm": 2.1456778049468994, + "learning_rate": 2.7599512591389114e-06, + "loss": 2.0955, + "step": 1392 + }, + { + "epoch": 1.5265753424657533, + "grad_norm": 1.9264456033706665, + "learning_rate": 2.757920389926889e-06, + "loss": 2.1594, + "step": 1393 + }, + { + "epoch": 1.5276712328767124, + "grad_norm": 1.9881576299667358, + "learning_rate": 2.755889520714866e-06, + "loss": 2.115, + "step": 1394 + }, + { + "epoch": 1.5287671232876714, + "grad_norm": 1.9104706048965454, + "learning_rate": 2.7538586515028437e-06, + "loss": 2.1074, + "step": 1395 + }, + { + "epoch": 1.52986301369863, + "grad_norm": 1.8815401792526245, + "learning_rate": 2.7518277822908205e-06, + "loss": 2.0779, + "step": 1396 + }, + { + "epoch": 1.530958904109589, + "grad_norm": 1.999870777130127, + "learning_rate": 2.749796913078798e-06, + "loss": 2.1409, + "step": 1397 + }, + { + "epoch": 1.532054794520548, + "grad_norm": 2.037266969680786, + "learning_rate": 2.747766043866775e-06, + "loss": 2.1565, + "step": 1398 + }, + { + "epoch": 1.5331506849315069, + "grad_norm": 1.865891695022583, + "learning_rate": 2.7457351746547524e-06, + "loss": 2.1269, + "step": 1399 + }, + { + "epoch": 1.5342465753424657, + "grad_norm": 1.959797739982605, + "learning_rate": 2.7437043054427297e-06, + "loss": 2.1674, + "step": 1400 + }, + { + "epoch": 1.5342465753424657, + "eval_loss": 2.279064893722534, + "eval_runtime": 50.0471, + "eval_samples_per_second": 19.981, + "eval_steps_per_second": 0.839, + "step": 1400 + }, + { + "epoch": 1.5353424657534247, + "grad_norm": 2.021819829940796, + "learning_rate": 2.741673436230707e-06, + "loss": 2.2088, + "step": 1401 + }, + { + "epoch": 1.5364383561643835, + "grad_norm": 1.8549301624298096, + "learning_rate": 2.739642567018684e-06, + "loss": 2.1384, + "step": 1402 + }, + { + "epoch": 1.5375342465753423, + "grad_norm": 2.0776336193084717, + "learning_rate": 2.7376116978066615e-06, + "loss": 2.1068, + "step": 1403 + }, + { + "epoch": 1.5386301369863014, + "grad_norm": 1.9860707521438599, + "learning_rate": 2.7355808285946388e-06, + "loss": 2.1377, + "step": 1404 + }, + { + "epoch": 1.5397260273972604, + "grad_norm": 2.009176015853882, + "learning_rate": 2.733549959382616e-06, + "loss": 2.136, + "step": 1405 + }, + { + "epoch": 1.5408219178082192, + "grad_norm": 1.8771592378616333, + "learning_rate": 2.7315190901705933e-06, + "loss": 2.1104, + "step": 1406 + }, + { + "epoch": 1.541917808219178, + "grad_norm": 1.9801877737045288, + "learning_rate": 2.7294882209585706e-06, + "loss": 2.1803, + "step": 1407 + }, + { + "epoch": 1.543013698630137, + "grad_norm": 1.8967032432556152, + "learning_rate": 2.727457351746548e-06, + "loss": 2.141, + "step": 1408 + }, + { + "epoch": 1.5441095890410959, + "grad_norm": 1.8221750259399414, + "learning_rate": 2.725426482534525e-06, + "loss": 2.1181, + "step": 1409 + }, + { + "epoch": 1.5452054794520547, + "grad_norm": 1.972421407699585, + "learning_rate": 2.723395613322502e-06, + "loss": 2.1056, + "step": 1410 + }, + { + "epoch": 1.5452054794520547, + "eval_loss": 2.2803397178649902, + "eval_runtime": 42.0131, + "eval_samples_per_second": 23.802, + "eval_steps_per_second": 1.0, + "step": 1410 + }, + { + "epoch": 1.5463013698630137, + "grad_norm": 1.9438509941101074, + "learning_rate": 2.7213647441104797e-06, + "loss": 2.1378, + "step": 1411 + }, + { + "epoch": 1.5473972602739727, + "grad_norm": 1.8502674102783203, + "learning_rate": 2.7193338748984565e-06, + "loss": 2.1244, + "step": 1412 + }, + { + "epoch": 1.5484931506849315, + "grad_norm": 1.8806159496307373, + "learning_rate": 2.7173030056864342e-06, + "loss": 2.1371, + "step": 1413 + }, + { + "epoch": 1.5495890410958904, + "grad_norm": 1.823891282081604, + "learning_rate": 2.715272136474411e-06, + "loss": 2.2526, + "step": 1414 + }, + { + "epoch": 1.5506849315068494, + "grad_norm": 1.9969522953033447, + "learning_rate": 2.713241267262389e-06, + "loss": 2.0843, + "step": 1415 + }, + { + "epoch": 1.5517808219178082, + "grad_norm": 1.9167888164520264, + "learning_rate": 2.7112103980503656e-06, + "loss": 2.139, + "step": 1416 + }, + { + "epoch": 1.552876712328767, + "grad_norm": 1.8950753211975098, + "learning_rate": 2.7091795288383433e-06, + "loss": 2.1184, + "step": 1417 + }, + { + "epoch": 1.553972602739726, + "grad_norm": 1.8647392988204956, + "learning_rate": 2.70714865962632e-06, + "loss": 2.1466, + "step": 1418 + }, + { + "epoch": 1.555068493150685, + "grad_norm": 2.074019432067871, + "learning_rate": 2.705117790414298e-06, + "loss": 2.1154, + "step": 1419 + }, + { + "epoch": 1.5561643835616439, + "grad_norm": 1.883457064628601, + "learning_rate": 2.7030869212022747e-06, + "loss": 2.1251, + "step": 1420 + }, + { + "epoch": 1.5561643835616439, + "eval_loss": 2.279707908630371, + "eval_runtime": 43.6933, + "eval_samples_per_second": 22.887, + "eval_steps_per_second": 0.961, + "step": 1420 + }, + { + "epoch": 1.5572602739726027, + "grad_norm": 1.88644540309906, + "learning_rate": 2.7010560519902516e-06, + "loss": 2.1554, + "step": 1421 + }, + { + "epoch": 1.5583561643835617, + "grad_norm": 1.960942029953003, + "learning_rate": 2.6990251827782293e-06, + "loss": 2.1134, + "step": 1422 + }, + { + "epoch": 1.5594520547945205, + "grad_norm": 1.9701318740844727, + "learning_rate": 2.696994313566206e-06, + "loss": 2.1212, + "step": 1423 + }, + { + "epoch": 1.5605479452054793, + "grad_norm": 1.9110147953033447, + "learning_rate": 2.694963444354184e-06, + "loss": 2.1245, + "step": 1424 + }, + { + "epoch": 1.5616438356164384, + "grad_norm": 2.1578176021575928, + "learning_rate": 2.6929325751421607e-06, + "loss": 2.1809, + "step": 1425 + }, + { + "epoch": 1.5627397260273974, + "grad_norm": 1.9082913398742676, + "learning_rate": 2.6909017059301384e-06, + "loss": 2.1186, + "step": 1426 + }, + { + "epoch": 1.563835616438356, + "grad_norm": 1.9049410820007324, + "learning_rate": 2.6888708367181153e-06, + "loss": 2.1531, + "step": 1427 + }, + { + "epoch": 1.564931506849315, + "grad_norm": 2.041297197341919, + "learning_rate": 2.686839967506093e-06, + "loss": 2.0996, + "step": 1428 + }, + { + "epoch": 1.566027397260274, + "grad_norm": 1.9282728433609009, + "learning_rate": 2.68480909829407e-06, + "loss": 2.1492, + "step": 1429 + }, + { + "epoch": 1.5671232876712329, + "grad_norm": 1.908298134803772, + "learning_rate": 2.6827782290820475e-06, + "loss": 2.0726, + "step": 1430 + }, + { + "epoch": 1.5671232876712329, + "eval_loss": 2.278843402862549, + "eval_runtime": 43.3233, + "eval_samples_per_second": 23.082, + "eval_steps_per_second": 0.969, + "step": 1430 + }, + { + "epoch": 1.5682191780821917, + "grad_norm": 2.0996971130371094, + "learning_rate": 2.6807473598700244e-06, + "loss": 2.0829, + "step": 1431 + }, + { + "epoch": 1.5693150684931507, + "grad_norm": 1.8816159963607788, + "learning_rate": 2.678716490658002e-06, + "loss": 2.1659, + "step": 1432 + }, + { + "epoch": 1.5704109589041098, + "grad_norm": 1.91497004032135, + "learning_rate": 2.676685621445979e-06, + "loss": 2.1437, + "step": 1433 + }, + { + "epoch": 1.5715068493150683, + "grad_norm": 1.8993713855743408, + "learning_rate": 2.674654752233956e-06, + "loss": 2.1324, + "step": 1434 + }, + { + "epoch": 1.5726027397260274, + "grad_norm": 1.9361183643341064, + "learning_rate": 2.6726238830219335e-06, + "loss": 2.138, + "step": 1435 + }, + { + "epoch": 1.5736986301369864, + "grad_norm": 1.8992613554000854, + "learning_rate": 2.6705930138099107e-06, + "loss": 2.1309, + "step": 1436 + }, + { + "epoch": 1.5747945205479452, + "grad_norm": 1.9301854372024536, + "learning_rate": 2.668562144597888e-06, + "loss": 2.1645, + "step": 1437 + }, + { + "epoch": 1.575890410958904, + "grad_norm": 1.8373525142669678, + "learning_rate": 2.6665312753858653e-06, + "loss": 2.0699, + "step": 1438 + }, + { + "epoch": 1.576986301369863, + "grad_norm": 2.0120513439178467, + "learning_rate": 2.6645004061738426e-06, + "loss": 2.1472, + "step": 1439 + }, + { + "epoch": 1.5780821917808219, + "grad_norm": 2.007274627685547, + "learning_rate": 2.66246953696182e-06, + "loss": 2.1406, + "step": 1440 + }, + { + "epoch": 1.5780821917808219, + "eval_loss": 2.278693914413452, + "eval_runtime": 43.8767, + "eval_samples_per_second": 22.791, + "eval_steps_per_second": 0.957, + "step": 1440 + }, + { + "epoch": 1.5791780821917807, + "grad_norm": 1.9184056520462036, + "learning_rate": 2.660438667749797e-06, + "loss": 2.1076, + "step": 1441 + }, + { + "epoch": 1.5802739726027397, + "grad_norm": 1.9140634536743164, + "learning_rate": 2.6584077985377744e-06, + "loss": 2.0868, + "step": 1442 + }, + { + "epoch": 1.5813698630136988, + "grad_norm": 1.9146252870559692, + "learning_rate": 2.6563769293257517e-06, + "loss": 2.1205, + "step": 1443 + }, + { + "epoch": 1.5824657534246576, + "grad_norm": 1.8893204927444458, + "learning_rate": 2.654346060113729e-06, + "loss": 2.1249, + "step": 1444 + }, + { + "epoch": 1.5835616438356164, + "grad_norm": 2.0452592372894287, + "learning_rate": 2.6523151909017062e-06, + "loss": 2.1069, + "step": 1445 + }, + { + "epoch": 1.5846575342465754, + "grad_norm": 1.8728071451187134, + "learning_rate": 2.6502843216896835e-06, + "loss": 2.1161, + "step": 1446 + }, + { + "epoch": 1.5857534246575342, + "grad_norm": 2.1464476585388184, + "learning_rate": 2.6482534524776603e-06, + "loss": 2.1032, + "step": 1447 + }, + { + "epoch": 1.586849315068493, + "grad_norm": 1.9760676622390747, + "learning_rate": 2.646222583265638e-06, + "loss": 2.1754, + "step": 1448 + }, + { + "epoch": 1.587945205479452, + "grad_norm": 1.9876340627670288, + "learning_rate": 2.644191714053615e-06, + "loss": 2.1512, + "step": 1449 + }, + { + "epoch": 1.589041095890411, + "grad_norm": 2.039198160171509, + "learning_rate": 2.6421608448415926e-06, + "loss": 2.1416, + "step": 1450 + }, + { + "epoch": 1.589041095890411, + "eval_loss": 2.2811355590820312, + "eval_runtime": 44.209, + "eval_samples_per_second": 22.62, + "eval_steps_per_second": 0.95, + "step": 1450 + }, + { + "epoch": 1.59013698630137, + "grad_norm": 1.882146954536438, + "learning_rate": 2.6401299756295694e-06, + "loss": 2.1106, + "step": 1451 + }, + { + "epoch": 1.5912328767123287, + "grad_norm": 2.0288634300231934, + "learning_rate": 2.638099106417547e-06, + "loss": 2.1011, + "step": 1452 + }, + { + "epoch": 1.5923287671232877, + "grad_norm": 2.0438144207000732, + "learning_rate": 2.636068237205524e-06, + "loss": 2.1352, + "step": 1453 + }, + { + "epoch": 1.5934246575342466, + "grad_norm": 1.9043279886245728, + "learning_rate": 2.6340373679935017e-06, + "loss": 2.1044, + "step": 1454 + }, + { + "epoch": 1.5945205479452054, + "grad_norm": 2.112670660018921, + "learning_rate": 2.6320064987814785e-06, + "loss": 2.1724, + "step": 1455 + }, + { + "epoch": 1.5956164383561644, + "grad_norm": 1.8258006572723389, + "learning_rate": 2.6299756295694562e-06, + "loss": 2.0825, + "step": 1456 + }, + { + "epoch": 1.5967123287671234, + "grad_norm": 1.970802664756775, + "learning_rate": 2.627944760357433e-06, + "loss": 2.1431, + "step": 1457 + }, + { + "epoch": 1.5978082191780822, + "grad_norm": 1.9067145586013794, + "learning_rate": 2.6259138911454104e-06, + "loss": 2.1729, + "step": 1458 + }, + { + "epoch": 1.598904109589041, + "grad_norm": 1.9728924036026, + "learning_rate": 2.6238830219333877e-06, + "loss": 2.1773, + "step": 1459 + }, + { + "epoch": 1.6, + "grad_norm": 2.0142018795013428, + "learning_rate": 2.621852152721365e-06, + "loss": 2.1626, + "step": 1460 + }, + { + "epoch": 1.6, + "eval_loss": 2.2811660766601562, + "eval_runtime": 44.1735, + "eval_samples_per_second": 22.638, + "eval_steps_per_second": 0.951, + "step": 1460 + }, + { + "epoch": 1.601095890410959, + "grad_norm": 2.064955472946167, + "learning_rate": 2.619821283509342e-06, + "loss": 2.0833, + "step": 1461 + }, + { + "epoch": 1.6021917808219177, + "grad_norm": 1.9600436687469482, + "learning_rate": 2.6177904142973195e-06, + "loss": 2.1596, + "step": 1462 + }, + { + "epoch": 1.6032876712328767, + "grad_norm": 1.8413004875183105, + "learning_rate": 2.6157595450852968e-06, + "loss": 2.0957, + "step": 1463 + }, + { + "epoch": 1.6043835616438358, + "grad_norm": 2.1802690029144287, + "learning_rate": 2.613728675873274e-06, + "loss": 2.1432, + "step": 1464 + }, + { + "epoch": 1.6054794520547944, + "grad_norm": 1.992746353149414, + "learning_rate": 2.6116978066612513e-06, + "loss": 2.1631, + "step": 1465 + }, + { + "epoch": 1.6065753424657534, + "grad_norm": 1.9906671047210693, + "learning_rate": 2.6096669374492286e-06, + "loss": 2.1587, + "step": 1466 + }, + { + "epoch": 1.6076712328767124, + "grad_norm": 2.1134135723114014, + "learning_rate": 2.607636068237206e-06, + "loss": 2.1385, + "step": 1467 + }, + { + "epoch": 1.6087671232876712, + "grad_norm": 2.259915590286255, + "learning_rate": 2.605605199025183e-06, + "loss": 2.1042, + "step": 1468 + }, + { + "epoch": 1.60986301369863, + "grad_norm": 2.0705254077911377, + "learning_rate": 2.6035743298131604e-06, + "loss": 2.145, + "step": 1469 + }, + { + "epoch": 1.610958904109589, + "grad_norm": 2.0326738357543945, + "learning_rate": 2.6015434606011377e-06, + "loss": 2.133, + "step": 1470 + }, + { + "epoch": 1.610958904109589, + "eval_loss": 2.2812676429748535, + "eval_runtime": 43.4091, + "eval_samples_per_second": 23.037, + "eval_steps_per_second": 0.968, + "step": 1470 + }, + { + "epoch": 1.612054794520548, + "grad_norm": 2.034182071685791, + "learning_rate": 2.5995125913891145e-06, + "loss": 2.1507, + "step": 1471 + }, + { + "epoch": 1.6131506849315067, + "grad_norm": 2.069373846054077, + "learning_rate": 2.5974817221770922e-06, + "loss": 2.152, + "step": 1472 + }, + { + "epoch": 1.6142465753424657, + "grad_norm": 2.2003285884857178, + "learning_rate": 2.595450852965069e-06, + "loss": 2.1347, + "step": 1473 + }, + { + "epoch": 1.6153424657534248, + "grad_norm": 1.919945478439331, + "learning_rate": 2.593419983753047e-06, + "loss": 2.1476, + "step": 1474 + }, + { + "epoch": 1.6164383561643836, + "grad_norm": 1.825148344039917, + "learning_rate": 2.5913891145410236e-06, + "loss": 2.1757, + "step": 1475 + }, + { + "epoch": 1.6175342465753424, + "grad_norm": 2.0014286041259766, + "learning_rate": 2.5893582453290013e-06, + "loss": 2.1208, + "step": 1476 + }, + { + "epoch": 1.6186301369863014, + "grad_norm": 2.0642764568328857, + "learning_rate": 2.587327376116978e-06, + "loss": 2.1619, + "step": 1477 + }, + { + "epoch": 1.6197260273972602, + "grad_norm": 2.002497911453247, + "learning_rate": 2.585296506904956e-06, + "loss": 2.1198, + "step": 1478 + }, + { + "epoch": 1.620821917808219, + "grad_norm": 1.7852193117141724, + "learning_rate": 2.5832656376929327e-06, + "loss": 2.1247, + "step": 1479 + }, + { + "epoch": 1.621917808219178, + "grad_norm": 1.9539082050323486, + "learning_rate": 2.5812347684809104e-06, + "loss": 2.1046, + "step": 1480 + }, + { + "epoch": 1.621917808219178, + "eval_loss": 2.2807998657226562, + "eval_runtime": 44.2896, + "eval_samples_per_second": 22.579, + "eval_steps_per_second": 0.948, + "step": 1480 + }, + { + "epoch": 1.6230136986301371, + "grad_norm": 2.095280647277832, + "learning_rate": 2.5792038992688873e-06, + "loss": 2.0685, + "step": 1481 + }, + { + "epoch": 1.624109589041096, + "grad_norm": 2.230388879776001, + "learning_rate": 2.577173030056865e-06, + "loss": 2.1976, + "step": 1482 + }, + { + "epoch": 1.6252054794520547, + "grad_norm": 1.9304745197296143, + "learning_rate": 2.575142160844842e-06, + "loss": 2.0647, + "step": 1483 + }, + { + "epoch": 1.6263013698630138, + "grad_norm": 1.856342077255249, + "learning_rate": 2.5731112916328187e-06, + "loss": 2.1444, + "step": 1484 + }, + { + "epoch": 1.6273972602739726, + "grad_norm": 2.117694854736328, + "learning_rate": 2.5710804224207964e-06, + "loss": 2.1536, + "step": 1485 + }, + { + "epoch": 1.6284931506849314, + "grad_norm": 2.043900489807129, + "learning_rate": 2.5690495532087733e-06, + "loss": 2.106, + "step": 1486 + }, + { + "epoch": 1.6295890410958904, + "grad_norm": 1.9283581972122192, + "learning_rate": 2.567018683996751e-06, + "loss": 2.0667, + "step": 1487 + }, + { + "epoch": 1.6306849315068495, + "grad_norm": 1.8138651847839355, + "learning_rate": 2.564987814784728e-06, + "loss": 2.1386, + "step": 1488 + }, + { + "epoch": 1.6317808219178083, + "grad_norm": 2.0534632205963135, + "learning_rate": 2.5629569455727055e-06, + "loss": 2.105, + "step": 1489 + }, + { + "epoch": 1.632876712328767, + "grad_norm": 1.9466804265975952, + "learning_rate": 2.5609260763606824e-06, + "loss": 2.1211, + "step": 1490 + }, + { + "epoch": 1.632876712328767, + "eval_loss": 2.275447368621826, + "eval_runtime": 44.2319, + "eval_samples_per_second": 22.608, + "eval_steps_per_second": 0.95, + "step": 1490 + }, + { + "epoch": 1.633972602739726, + "grad_norm": 1.8767902851104736, + "learning_rate": 2.55889520714866e-06, + "loss": 2.116, + "step": 1491 + }, + { + "epoch": 1.635068493150685, + "grad_norm": 1.9194284677505493, + "learning_rate": 2.556864337936637e-06, + "loss": 2.1928, + "step": 1492 + }, + { + "epoch": 1.6361643835616437, + "grad_norm": 2.0496408939361572, + "learning_rate": 2.5548334687246146e-06, + "loss": 2.1182, + "step": 1493 + }, + { + "epoch": 1.6372602739726028, + "grad_norm": 1.8577388525009155, + "learning_rate": 2.5528025995125915e-06, + "loss": 2.1313, + "step": 1494 + }, + { + "epoch": 1.6383561643835618, + "grad_norm": 2.020623207092285, + "learning_rate": 2.550771730300569e-06, + "loss": 2.1627, + "step": 1495 + }, + { + "epoch": 1.6394520547945204, + "grad_norm": 1.8472096920013428, + "learning_rate": 2.548740861088546e-06, + "loss": 2.1514, + "step": 1496 + }, + { + "epoch": 1.6405479452054794, + "grad_norm": 1.9105209112167358, + "learning_rate": 2.5467099918765233e-06, + "loss": 2.1558, + "step": 1497 + }, + { + "epoch": 1.6416438356164385, + "grad_norm": 1.9635803699493408, + "learning_rate": 2.5446791226645006e-06, + "loss": 2.08, + "step": 1498 + }, + { + "epoch": 1.6427397260273973, + "grad_norm": 1.8911043405532837, + "learning_rate": 2.542648253452478e-06, + "loss": 2.0648, + "step": 1499 + }, + { + "epoch": 1.643835616438356, + "grad_norm": 1.9337847232818604, + "learning_rate": 2.540617384240455e-06, + "loss": 2.1526, + "step": 1500 + }, + { + "epoch": 1.643835616438356, + "eval_loss": 2.2771341800689697, + "eval_runtime": 44.2753, + "eval_samples_per_second": 22.586, + "eval_steps_per_second": 0.949, + "step": 1500 + }, + { + "epoch": 1.644931506849315, + "grad_norm": 2.152970552444458, + "learning_rate": 2.5385865150284324e-06, + "loss": 2.1328, + "step": 1501 + }, + { + "epoch": 1.6460273972602741, + "grad_norm": 1.8749687671661377, + "learning_rate": 2.5365556458164097e-06, + "loss": 2.1093, + "step": 1502 + }, + { + "epoch": 1.6471232876712327, + "grad_norm": 1.8745299577713013, + "learning_rate": 2.534524776604387e-06, + "loss": 2.1172, + "step": 1503 + }, + { + "epoch": 1.6482191780821918, + "grad_norm": 2.006110906600952, + "learning_rate": 2.5324939073923642e-06, + "loss": 2.1318, + "step": 1504 + }, + { + "epoch": 1.6493150684931508, + "grad_norm": 2.055590867996216, + "learning_rate": 2.5304630381803415e-06, + "loss": 2.0463, + "step": 1505 + }, + { + "epoch": 1.6504109589041096, + "grad_norm": 1.9270331859588623, + "learning_rate": 2.5284321689683188e-06, + "loss": 2.1195, + "step": 1506 + }, + { + "epoch": 1.6515068493150684, + "grad_norm": 1.9981428384780884, + "learning_rate": 2.526401299756296e-06, + "loss": 2.1282, + "step": 1507 + }, + { + "epoch": 1.6526027397260274, + "grad_norm": 2.051969528198242, + "learning_rate": 2.524370430544273e-06, + "loss": 2.1013, + "step": 1508 + }, + { + "epoch": 1.6536986301369863, + "grad_norm": 2.067718744277954, + "learning_rate": 2.5223395613322506e-06, + "loss": 2.1265, + "step": 1509 + }, + { + "epoch": 1.654794520547945, + "grad_norm": 2.0843262672424316, + "learning_rate": 2.5203086921202274e-06, + "loss": 2.0994, + "step": 1510 + }, + { + "epoch": 1.654794520547945, + "eval_loss": 2.27781081199646, + "eval_runtime": 42.8213, + "eval_samples_per_second": 23.353, + "eval_steps_per_second": 0.981, + "step": 1510 + }, + { + "epoch": 1.655890410958904, + "grad_norm": 2.146921396255493, + "learning_rate": 2.518277822908205e-06, + "loss": 2.1456, + "step": 1511 + }, + { + "epoch": 1.6569863013698631, + "grad_norm": 1.9604636430740356, + "learning_rate": 2.516246953696182e-06, + "loss": 2.1118, + "step": 1512 + }, + { + "epoch": 1.658082191780822, + "grad_norm": 1.9613317251205444, + "learning_rate": 2.5142160844841597e-06, + "loss": 2.1401, + "step": 1513 + }, + { + "epoch": 1.6591780821917808, + "grad_norm": 2.0410237312316895, + "learning_rate": 2.5121852152721365e-06, + "loss": 2.1532, + "step": 1514 + }, + { + "epoch": 1.6602739726027398, + "grad_norm": 1.8603837490081787, + "learning_rate": 2.5101543460601142e-06, + "loss": 2.1437, + "step": 1515 + }, + { + "epoch": 1.6613698630136986, + "grad_norm": 1.9498400688171387, + "learning_rate": 2.508123476848091e-06, + "loss": 2.1384, + "step": 1516 + }, + { + "epoch": 1.6624657534246574, + "grad_norm": 1.8603938817977905, + "learning_rate": 2.506092607636069e-06, + "loss": 2.0903, + "step": 1517 + }, + { + "epoch": 1.6635616438356164, + "grad_norm": 1.8238614797592163, + "learning_rate": 2.5040617384240457e-06, + "loss": 2.1644, + "step": 1518 + }, + { + "epoch": 1.6646575342465755, + "grad_norm": 1.9574072360992432, + "learning_rate": 2.5020308692120234e-06, + "loss": 2.1072, + "step": 1519 + }, + { + "epoch": 1.6657534246575343, + "grad_norm": 2.100829601287842, + "learning_rate": 2.5e-06, + "loss": 2.111, + "step": 1520 + }, + { + "epoch": 1.6657534246575343, + "eval_loss": 2.278552293777466, + "eval_runtime": 46.587, + "eval_samples_per_second": 21.465, + "eval_steps_per_second": 0.902, + "step": 1520 + }, + { + "epoch": 1.666849315068493, + "grad_norm": 1.921262264251709, + "learning_rate": 2.4979691307879775e-06, + "loss": 2.1667, + "step": 1521 + }, + { + "epoch": 1.6679452054794521, + "grad_norm": 1.9328664541244507, + "learning_rate": 2.4959382615759548e-06, + "loss": 2.1175, + "step": 1522 + }, + { + "epoch": 1.669041095890411, + "grad_norm": 1.909544587135315, + "learning_rate": 2.493907392363932e-06, + "loss": 2.0905, + "step": 1523 + }, + { + "epoch": 1.6701369863013698, + "grad_norm": 1.928046464920044, + "learning_rate": 2.4918765231519093e-06, + "loss": 2.1757, + "step": 1524 + }, + { + "epoch": 1.6712328767123288, + "grad_norm": 1.8839048147201538, + "learning_rate": 2.4898456539398866e-06, + "loss": 2.1424, + "step": 1525 + }, + { + "epoch": 1.6723287671232878, + "grad_norm": 1.913812518119812, + "learning_rate": 2.487814784727864e-06, + "loss": 2.1178, + "step": 1526 + }, + { + "epoch": 1.6734246575342466, + "grad_norm": 1.7485435009002686, + "learning_rate": 2.4857839155158407e-06, + "loss": 2.1794, + "step": 1527 + }, + { + "epoch": 1.6745205479452054, + "grad_norm": 1.8693492412567139, + "learning_rate": 2.483753046303818e-06, + "loss": 2.1265, + "step": 1528 + }, + { + "epoch": 1.6756164383561645, + "grad_norm": 1.8982772827148438, + "learning_rate": 2.4817221770917953e-06, + "loss": 2.1254, + "step": 1529 + }, + { + "epoch": 1.6767123287671233, + "grad_norm": 1.9917978048324585, + "learning_rate": 2.4796913078797725e-06, + "loss": 2.13, + "step": 1530 + }, + { + "epoch": 1.6767123287671233, + "eval_loss": 2.277986764907837, + "eval_runtime": 52.1464, + "eval_samples_per_second": 19.177, + "eval_steps_per_second": 0.805, + "step": 1530 + }, + { + "epoch": 1.677808219178082, + "grad_norm": 1.8703829050064087, + "learning_rate": 2.47766043866775e-06, + "loss": 2.1527, + "step": 1531 + }, + { + "epoch": 1.6789041095890411, + "grad_norm": 1.950713872909546, + "learning_rate": 2.475629569455727e-06, + "loss": 2.1481, + "step": 1532 + }, + { + "epoch": 1.6800000000000002, + "grad_norm": 1.8339334726333618, + "learning_rate": 2.4735987002437044e-06, + "loss": 2.1422, + "step": 1533 + }, + { + "epoch": 1.6810958904109587, + "grad_norm": 1.8423672914505005, + "learning_rate": 2.4715678310316816e-06, + "loss": 2.1026, + "step": 1534 + }, + { + "epoch": 1.6821917808219178, + "grad_norm": 1.9338176250457764, + "learning_rate": 2.469536961819659e-06, + "loss": 2.1095, + "step": 1535 + }, + { + "epoch": 1.6832876712328768, + "grad_norm": 1.8666706085205078, + "learning_rate": 2.467506092607636e-06, + "loss": 2.1176, + "step": 1536 + }, + { + "epoch": 1.6843835616438356, + "grad_norm": 2.0762805938720703, + "learning_rate": 2.4654752233956135e-06, + "loss": 2.127, + "step": 1537 + }, + { + "epoch": 1.6854794520547944, + "grad_norm": 1.9729993343353271, + "learning_rate": 2.4634443541835907e-06, + "loss": 2.1194, + "step": 1538 + }, + { + "epoch": 1.6865753424657535, + "grad_norm": 1.910555362701416, + "learning_rate": 2.461413484971568e-06, + "loss": 2.1745, + "step": 1539 + }, + { + "epoch": 1.6876712328767123, + "grad_norm": 1.953911304473877, + "learning_rate": 2.4593826157595453e-06, + "loss": 2.1236, + "step": 1540 + }, + { + "epoch": 1.6876712328767123, + "eval_loss": 2.277575731277466, + "eval_runtime": 48.0126, + "eval_samples_per_second": 20.828, + "eval_steps_per_second": 0.875, + "step": 1540 + }, + { + "epoch": 1.688767123287671, + "grad_norm": 2.0469322204589844, + "learning_rate": 2.4573517465475226e-06, + "loss": 2.1999, + "step": 1541 + }, + { + "epoch": 1.6898630136986301, + "grad_norm": 1.9962514638900757, + "learning_rate": 2.4553208773355e-06, + "loss": 2.0865, + "step": 1542 + }, + { + "epoch": 1.6909589041095892, + "grad_norm": 1.911471962928772, + "learning_rate": 2.453290008123477e-06, + "loss": 2.1107, + "step": 1543 + }, + { + "epoch": 1.692054794520548, + "grad_norm": 1.9354206323623657, + "learning_rate": 2.4512591389114544e-06, + "loss": 2.1215, + "step": 1544 + }, + { + "epoch": 1.6931506849315068, + "grad_norm": 1.938178539276123, + "learning_rate": 2.4492282696994317e-06, + "loss": 2.1378, + "step": 1545 + }, + { + "epoch": 1.6942465753424658, + "grad_norm": 2.003552198410034, + "learning_rate": 2.447197400487409e-06, + "loss": 2.1361, + "step": 1546 + }, + { + "epoch": 1.6953424657534246, + "grad_norm": 1.9754462242126465, + "learning_rate": 2.4451665312753862e-06, + "loss": 2.134, + "step": 1547 + }, + { + "epoch": 1.6964383561643834, + "grad_norm": 1.924299955368042, + "learning_rate": 2.4431356620633635e-06, + "loss": 2.1497, + "step": 1548 + }, + { + "epoch": 1.6975342465753425, + "grad_norm": 2.1239371299743652, + "learning_rate": 2.4411047928513408e-06, + "loss": 2.1427, + "step": 1549 + }, + { + "epoch": 1.6986301369863015, + "grad_norm": 1.984377384185791, + "learning_rate": 2.439073923639318e-06, + "loss": 2.1268, + "step": 1550 + }, + { + "epoch": 1.6986301369863015, + "eval_loss": 2.2747738361358643, + "eval_runtime": 53.2249, + "eval_samples_per_second": 18.788, + "eval_steps_per_second": 0.789, + "step": 1550 + }, + { + "epoch": 1.6997260273972603, + "grad_norm": 1.9375758171081543, + "learning_rate": 2.4370430544272953e-06, + "loss": 2.1369, + "step": 1551 + }, + { + "epoch": 1.7008219178082191, + "grad_norm": 1.9191703796386719, + "learning_rate": 2.435012185215272e-06, + "loss": 2.1061, + "step": 1552 + }, + { + "epoch": 1.7019178082191782, + "grad_norm": 2.1356418132781982, + "learning_rate": 2.4329813160032495e-06, + "loss": 2.0923, + "step": 1553 + }, + { + "epoch": 1.703013698630137, + "grad_norm": 2.02870774269104, + "learning_rate": 2.4309504467912267e-06, + "loss": 2.105, + "step": 1554 + }, + { + "epoch": 1.7041095890410958, + "grad_norm": 1.9436805248260498, + "learning_rate": 2.428919577579204e-06, + "loss": 2.1194, + "step": 1555 + }, + { + "epoch": 1.7052054794520548, + "grad_norm": 1.997019648551941, + "learning_rate": 2.4268887083671813e-06, + "loss": 2.1114, + "step": 1556 + }, + { + "epoch": 1.7063013698630138, + "grad_norm": 2.0759427547454834, + "learning_rate": 2.4248578391551586e-06, + "loss": 2.1955, + "step": 1557 + }, + { + "epoch": 1.7073972602739727, + "grad_norm": 2.0526654720306396, + "learning_rate": 2.422826969943136e-06, + "loss": 2.1103, + "step": 1558 + }, + { + "epoch": 1.7084931506849315, + "grad_norm": 1.8765089511871338, + "learning_rate": 2.420796100731113e-06, + "loss": 2.1068, + "step": 1559 + }, + { + "epoch": 1.7095890410958905, + "grad_norm": 1.9164950847625732, + "learning_rate": 2.4187652315190904e-06, + "loss": 2.082, + "step": 1560 + }, + { + "epoch": 1.7095890410958905, + "eval_loss": 2.272998332977295, + "eval_runtime": 49.0487, + "eval_samples_per_second": 20.388, + "eval_steps_per_second": 0.856, + "step": 1560 + }, + { + "epoch": 1.7106849315068493, + "grad_norm": 1.9921822547912598, + "learning_rate": 2.4167343623070677e-06, + "loss": 2.1285, + "step": 1561 + }, + { + "epoch": 1.7117808219178081, + "grad_norm": 1.8890291452407837, + "learning_rate": 2.414703493095045e-06, + "loss": 2.1385, + "step": 1562 + }, + { + "epoch": 1.7128767123287671, + "grad_norm": 1.9233667850494385, + "learning_rate": 2.4126726238830222e-06, + "loss": 2.1282, + "step": 1563 + }, + { + "epoch": 1.7139726027397262, + "grad_norm": 2.0368800163269043, + "learning_rate": 2.410641754670999e-06, + "loss": 2.1366, + "step": 1564 + }, + { + "epoch": 1.7150684931506848, + "grad_norm": 2.0321929454803467, + "learning_rate": 2.4086108854589763e-06, + "loss": 2.1163, + "step": 1565 + }, + { + "epoch": 1.7161643835616438, + "grad_norm": 2.042773485183716, + "learning_rate": 2.4065800162469536e-06, + "loss": 2.1529, + "step": 1566 + }, + { + "epoch": 1.7172602739726028, + "grad_norm": 1.937940239906311, + "learning_rate": 2.404549147034931e-06, + "loss": 2.106, + "step": 1567 + }, + { + "epoch": 1.7183561643835616, + "grad_norm": 1.878649353981018, + "learning_rate": 2.402518277822908e-06, + "loss": 2.1179, + "step": 1568 + }, + { + "epoch": 1.7194520547945205, + "grad_norm": 1.997246265411377, + "learning_rate": 2.4004874086108854e-06, + "loss": 2.1672, + "step": 1569 + }, + { + "epoch": 1.7205479452054795, + "grad_norm": 1.9837015867233276, + "learning_rate": 2.3984565393988627e-06, + "loss": 2.0816, + "step": 1570 + }, + { + "epoch": 1.7205479452054795, + "eval_loss": 2.2730956077575684, + "eval_runtime": 52.1793, + "eval_samples_per_second": 19.165, + "eval_steps_per_second": 0.805, + "step": 1570 + }, + { + "epoch": 1.7216438356164385, + "grad_norm": 1.8919317722320557, + "learning_rate": 2.39642567018684e-06, + "loss": 2.1541, + "step": 1571 + }, + { + "epoch": 1.7227397260273971, + "grad_norm": 1.8129284381866455, + "learning_rate": 2.3943948009748173e-06, + "loss": 2.1033, + "step": 1572 + }, + { + "epoch": 1.7238356164383561, + "grad_norm": 1.79408860206604, + "learning_rate": 2.3923639317627945e-06, + "loss": 2.1163, + "step": 1573 + }, + { + "epoch": 1.7249315068493152, + "grad_norm": 1.9178274869918823, + "learning_rate": 2.390333062550772e-06, + "loss": 2.1383, + "step": 1574 + }, + { + "epoch": 1.726027397260274, + "grad_norm": 1.9113222360610962, + "learning_rate": 2.388302193338749e-06, + "loss": 2.0987, + "step": 1575 + }, + { + "epoch": 1.7271232876712328, + "grad_norm": 1.915170431137085, + "learning_rate": 2.3862713241267264e-06, + "loss": 2.0235, + "step": 1576 + }, + { + "epoch": 1.7282191780821918, + "grad_norm": 1.9789308309555054, + "learning_rate": 2.3842404549147037e-06, + "loss": 2.1942, + "step": 1577 + }, + { + "epoch": 1.7293150684931506, + "grad_norm": 1.8410224914550781, + "learning_rate": 2.382209585702681e-06, + "loss": 2.1105, + "step": 1578 + }, + { + "epoch": 1.7304109589041095, + "grad_norm": 1.9388853311538696, + "learning_rate": 2.380178716490658e-06, + "loss": 2.1021, + "step": 1579 + }, + { + "epoch": 1.7315068493150685, + "grad_norm": 1.9328628778457642, + "learning_rate": 2.3781478472786355e-06, + "loss": 2.1584, + "step": 1580 + }, + { + "epoch": 1.7315068493150685, + "eval_loss": 2.271240472793579, + "eval_runtime": 47.9206, + "eval_samples_per_second": 20.868, + "eval_steps_per_second": 0.876, + "step": 1580 + }, + { + "epoch": 1.7326027397260275, + "grad_norm": 1.8846843242645264, + "learning_rate": 2.3761169780666128e-06, + "loss": 2.1269, + "step": 1581 + }, + { + "epoch": 1.7336986301369863, + "grad_norm": 1.8232983350753784, + "learning_rate": 2.37408610885459e-06, + "loss": 2.1522, + "step": 1582 + }, + { + "epoch": 1.7347945205479451, + "grad_norm": 2.023982286453247, + "learning_rate": 2.3720552396425673e-06, + "loss": 2.1571, + "step": 1583 + }, + { + "epoch": 1.7358904109589042, + "grad_norm": 1.861229658126831, + "learning_rate": 2.3700243704305446e-06, + "loss": 2.103, + "step": 1584 + }, + { + "epoch": 1.736986301369863, + "grad_norm": 1.8577842712402344, + "learning_rate": 2.367993501218522e-06, + "loss": 2.1766, + "step": 1585 + }, + { + "epoch": 1.7380821917808218, + "grad_norm": 1.915722131729126, + "learning_rate": 2.365962632006499e-06, + "loss": 2.1306, + "step": 1586 + }, + { + "epoch": 1.7391780821917808, + "grad_norm": 1.8876861333847046, + "learning_rate": 2.3639317627944764e-06, + "loss": 2.1303, + "step": 1587 + }, + { + "epoch": 1.7402739726027399, + "grad_norm": 1.9599688053131104, + "learning_rate": 2.3619008935824537e-06, + "loss": 2.1513, + "step": 1588 + }, + { + "epoch": 1.7413698630136987, + "grad_norm": 1.90068519115448, + "learning_rate": 2.3598700243704305e-06, + "loss": 2.1459, + "step": 1589 + }, + { + "epoch": 1.7424657534246575, + "grad_norm": 1.920277714729309, + "learning_rate": 2.357839155158408e-06, + "loss": 2.1453, + "step": 1590 + }, + { + "epoch": 1.7424657534246575, + "eval_loss": 2.2699742317199707, + "eval_runtime": 50.0211, + "eval_samples_per_second": 19.992, + "eval_steps_per_second": 0.84, + "step": 1590 + }, + { + "epoch": 1.7435616438356165, + "grad_norm": 1.9827759265899658, + "learning_rate": 2.355808285946385e-06, + "loss": 2.1393, + "step": 1591 + }, + { + "epoch": 1.7446575342465753, + "grad_norm": 1.8602412939071655, + "learning_rate": 2.3537774167343624e-06, + "loss": 2.1317, + "step": 1592 + }, + { + "epoch": 1.7457534246575341, + "grad_norm": 1.964829444885254, + "learning_rate": 2.3517465475223396e-06, + "loss": 2.1552, + "step": 1593 + }, + { + "epoch": 1.7468493150684932, + "grad_norm": 2.0124034881591797, + "learning_rate": 2.349715678310317e-06, + "loss": 2.1322, + "step": 1594 + }, + { + "epoch": 1.7479452054794522, + "grad_norm": 1.9786088466644287, + "learning_rate": 2.347684809098294e-06, + "loss": 2.1435, + "step": 1595 + }, + { + "epoch": 1.749041095890411, + "grad_norm": 1.9575989246368408, + "learning_rate": 2.3456539398862715e-06, + "loss": 2.1032, + "step": 1596 + }, + { + "epoch": 1.7501369863013698, + "grad_norm": 2.0460586547851562, + "learning_rate": 2.3436230706742487e-06, + "loss": 2.1595, + "step": 1597 + }, + { + "epoch": 1.7512328767123289, + "grad_norm": 1.9904825687408447, + "learning_rate": 2.341592201462226e-06, + "loss": 2.1025, + "step": 1598 + }, + { + "epoch": 1.7523287671232877, + "grad_norm": 2.0797626972198486, + "learning_rate": 2.3395613322502033e-06, + "loss": 2.1221, + "step": 1599 + }, + { + "epoch": 1.7534246575342465, + "grad_norm": 2.0380358695983887, + "learning_rate": 2.3375304630381806e-06, + "loss": 2.1166, + "step": 1600 + }, + { + "epoch": 1.7534246575342465, + "eval_loss": 2.2704837322235107, + "eval_runtime": 51.9207, + "eval_samples_per_second": 19.26, + "eval_steps_per_second": 0.809, + "step": 1600 + }, + { + "epoch": 1.7545205479452055, + "grad_norm": 2.009023427963257, + "learning_rate": 2.335499593826158e-06, + "loss": 2.0922, + "step": 1601 + }, + { + "epoch": 1.7556164383561645, + "grad_norm": 2.043168783187866, + "learning_rate": 2.333468724614135e-06, + "loss": 2.0962, + "step": 1602 + }, + { + "epoch": 1.7567123287671231, + "grad_norm": 2.1130902767181396, + "learning_rate": 2.3314378554021124e-06, + "loss": 2.1318, + "step": 1603 + }, + { + "epoch": 1.7578082191780822, + "grad_norm": 1.9391744136810303, + "learning_rate": 2.3294069861900897e-06, + "loss": 2.155, + "step": 1604 + }, + { + "epoch": 1.7589041095890412, + "grad_norm": 2.0810694694519043, + "learning_rate": 2.327376116978067e-06, + "loss": 2.1066, + "step": 1605 + }, + { + "epoch": 1.76, + "grad_norm": 2.029707431793213, + "learning_rate": 2.3253452477660442e-06, + "loss": 2.0897, + "step": 1606 + }, + { + "epoch": 1.7610958904109588, + "grad_norm": 1.928080439567566, + "learning_rate": 2.3233143785540215e-06, + "loss": 2.1589, + "step": 1607 + }, + { + "epoch": 1.7621917808219179, + "grad_norm": 1.8527079820632935, + "learning_rate": 2.3212835093419988e-06, + "loss": 2.1005, + "step": 1608 + }, + { + "epoch": 1.7632876712328767, + "grad_norm": 2.142939567565918, + "learning_rate": 2.319252640129976e-06, + "loss": 2.1499, + "step": 1609 + }, + { + "epoch": 1.7643835616438355, + "grad_norm": 2.040922164916992, + "learning_rate": 2.3172217709179533e-06, + "loss": 2.0997, + "step": 1610 + }, + { + "epoch": 1.7643835616438355, + "eval_loss": 2.2691540718078613, + "eval_runtime": 42.4134, + "eval_samples_per_second": 23.577, + "eval_steps_per_second": 0.99, + "step": 1610 + }, + { + "epoch": 1.7654794520547945, + "grad_norm": 1.988370656967163, + "learning_rate": 2.3151909017059306e-06, + "loss": 2.1795, + "step": 1611 + }, + { + "epoch": 1.7665753424657535, + "grad_norm": 1.996656060218811, + "learning_rate": 2.313160032493908e-06, + "loss": 2.1536, + "step": 1612 + }, + { + "epoch": 1.7676712328767124, + "grad_norm": 2.04560923576355, + "learning_rate": 2.311129163281885e-06, + "loss": 2.0781, + "step": 1613 + }, + { + "epoch": 1.7687671232876712, + "grad_norm": 1.8977863788604736, + "learning_rate": 2.309098294069862e-06, + "loss": 2.1312, + "step": 1614 + }, + { + "epoch": 1.7698630136986302, + "grad_norm": 2.026904821395874, + "learning_rate": 2.3070674248578393e-06, + "loss": 2.0998, + "step": 1615 + }, + { + "epoch": 1.770958904109589, + "grad_norm": 1.8714247941970825, + "learning_rate": 2.3050365556458166e-06, + "loss": 2.0875, + "step": 1616 + }, + { + "epoch": 1.7720547945205478, + "grad_norm": 2.11859130859375, + "learning_rate": 2.303005686433794e-06, + "loss": 2.1361, + "step": 1617 + }, + { + "epoch": 1.7731506849315068, + "grad_norm": 2.0592799186706543, + "learning_rate": 2.300974817221771e-06, + "loss": 2.1394, + "step": 1618 + }, + { + "epoch": 1.7742465753424659, + "grad_norm": 2.0358338356018066, + "learning_rate": 2.2989439480097484e-06, + "loss": 2.104, + "step": 1619 + }, + { + "epoch": 1.7753424657534247, + "grad_norm": 2.0237109661102295, + "learning_rate": 2.2969130787977257e-06, + "loss": 2.0823, + "step": 1620 + }, + { + "epoch": 1.7753424657534247, + "eval_loss": 2.27001690864563, + "eval_runtime": 43.9168, + "eval_samples_per_second": 22.77, + "eval_steps_per_second": 0.956, + "step": 1620 + }, + { + "epoch": 1.7764383561643835, + "grad_norm": 1.8729594945907593, + "learning_rate": 2.294882209585703e-06, + "loss": 2.1243, + "step": 1621 + }, + { + "epoch": 1.7775342465753425, + "grad_norm": 1.9390734434127808, + "learning_rate": 2.2928513403736802e-06, + "loss": 2.1469, + "step": 1622 + }, + { + "epoch": 1.7786301369863013, + "grad_norm": 2.0409750938415527, + "learning_rate": 2.2908204711616575e-06, + "loss": 2.1062, + "step": 1623 + }, + { + "epoch": 1.7797260273972602, + "grad_norm": 1.9116135835647583, + "learning_rate": 2.2887896019496348e-06, + "loss": 2.1208, + "step": 1624 + }, + { + "epoch": 1.7808219178082192, + "grad_norm": 1.9144036769866943, + "learning_rate": 2.286758732737612e-06, + "loss": 2.1628, + "step": 1625 + }, + { + "epoch": 1.7819178082191782, + "grad_norm": 1.8522213697433472, + "learning_rate": 2.284727863525589e-06, + "loss": 2.1168, + "step": 1626 + }, + { + "epoch": 1.783013698630137, + "grad_norm": 2.087541103363037, + "learning_rate": 2.282696994313566e-06, + "loss": 2.1195, + "step": 1627 + }, + { + "epoch": 1.7841095890410958, + "grad_norm": 2.0399985313415527, + "learning_rate": 2.2806661251015434e-06, + "loss": 2.1557, + "step": 1628 + }, + { + "epoch": 1.7852054794520549, + "grad_norm": 1.884964108467102, + "learning_rate": 2.2786352558895207e-06, + "loss": 2.1408, + "step": 1629 + }, + { + "epoch": 1.7863013698630137, + "grad_norm": 1.901802897453308, + "learning_rate": 2.276604386677498e-06, + "loss": 2.1145, + "step": 1630 + }, + { + "epoch": 1.7863013698630137, + "eval_loss": 2.2690439224243164, + "eval_runtime": 44.8662, + "eval_samples_per_second": 22.289, + "eval_steps_per_second": 0.936, + "step": 1630 + }, + { + "epoch": 1.7873972602739725, + "grad_norm": 1.8818615674972534, + "learning_rate": 2.2745735174654753e-06, + "loss": 2.0879, + "step": 1631 + }, + { + "epoch": 1.7884931506849315, + "grad_norm": 1.9345687627792358, + "learning_rate": 2.2725426482534526e-06, + "loss": 2.1029, + "step": 1632 + }, + { + "epoch": 1.7895890410958906, + "grad_norm": 1.868314504623413, + "learning_rate": 2.27051177904143e-06, + "loss": 2.1435, + "step": 1633 + }, + { + "epoch": 1.7906849315068492, + "grad_norm": 1.8868515491485596, + "learning_rate": 2.268480909829407e-06, + "loss": 2.1294, + "step": 1634 + }, + { + "epoch": 1.7917808219178082, + "grad_norm": 1.9929358959197998, + "learning_rate": 2.2664500406173844e-06, + "loss": 2.1165, + "step": 1635 + }, + { + "epoch": 1.7928767123287672, + "grad_norm": 1.9766895771026611, + "learning_rate": 2.2644191714053617e-06, + "loss": 2.1327, + "step": 1636 + }, + { + "epoch": 1.793972602739726, + "grad_norm": 2.03336238861084, + "learning_rate": 2.262388302193339e-06, + "loss": 2.1473, + "step": 1637 + }, + { + "epoch": 1.7950684931506848, + "grad_norm": 1.8569762706756592, + "learning_rate": 2.260357432981316e-06, + "loss": 2.125, + "step": 1638 + }, + { + "epoch": 1.7961643835616439, + "grad_norm": 1.9522744417190552, + "learning_rate": 2.2583265637692935e-06, + "loss": 2.1238, + "step": 1639 + }, + { + "epoch": 1.7972602739726027, + "grad_norm": 1.9175306558609009, + "learning_rate": 2.2562956945572708e-06, + "loss": 2.1332, + "step": 1640 + }, + { + "epoch": 1.7972602739726027, + "eval_loss": 2.2702176570892334, + "eval_runtime": 44.1369, + "eval_samples_per_second": 22.657, + "eval_steps_per_second": 0.952, + "step": 1640 + }, + { + "epoch": 1.7983561643835615, + "grad_norm": 2.109614849090576, + "learning_rate": 2.254264825345248e-06, + "loss": 2.1252, + "step": 1641 + }, + { + "epoch": 1.7994520547945205, + "grad_norm": 2.017613410949707, + "learning_rate": 2.2522339561332253e-06, + "loss": 2.1093, + "step": 1642 + }, + { + "epoch": 1.8005479452054796, + "grad_norm": 2.000864028930664, + "learning_rate": 2.2502030869212026e-06, + "loss": 2.1243, + "step": 1643 + }, + { + "epoch": 1.8016438356164384, + "grad_norm": 1.9952439069747925, + "learning_rate": 2.24817221770918e-06, + "loss": 2.1213, + "step": 1644 + }, + { + "epoch": 1.8027397260273972, + "grad_norm": 2.0456748008728027, + "learning_rate": 2.246141348497157e-06, + "loss": 2.1621, + "step": 1645 + }, + { + "epoch": 1.8038356164383562, + "grad_norm": 1.9011833667755127, + "learning_rate": 2.2441104792851344e-06, + "loss": 2.1426, + "step": 1646 + }, + { + "epoch": 1.804931506849315, + "grad_norm": 1.9874658584594727, + "learning_rate": 2.2420796100731117e-06, + "loss": 2.1085, + "step": 1647 + }, + { + "epoch": 1.8060273972602738, + "grad_norm": 2.0237057209014893, + "learning_rate": 2.240048740861089e-06, + "loss": 2.1394, + "step": 1648 + }, + { + "epoch": 1.8071232876712329, + "grad_norm": 2.0127220153808594, + "learning_rate": 2.2380178716490662e-06, + "loss": 2.1474, + "step": 1649 + }, + { + "epoch": 1.808219178082192, + "grad_norm": 2.051722764968872, + "learning_rate": 2.2359870024370435e-06, + "loss": 2.0444, + "step": 1650 + }, + { + "epoch": 1.808219178082192, + "eval_loss": 2.2701034545898438, + "eval_runtime": 44.4338, + "eval_samples_per_second": 22.505, + "eval_steps_per_second": 0.945, + "step": 1650 + }, + { + "epoch": 1.8093150684931507, + "grad_norm": 2.0204219818115234, + "learning_rate": 2.2339561332250204e-06, + "loss": 2.1445, + "step": 1651 + }, + { + "epoch": 1.8104109589041095, + "grad_norm": 2.106220006942749, + "learning_rate": 2.2319252640129976e-06, + "loss": 2.153, + "step": 1652 + }, + { + "epoch": 1.8115068493150686, + "grad_norm": 1.8632844686508179, + "learning_rate": 2.229894394800975e-06, + "loss": 2.137, + "step": 1653 + }, + { + "epoch": 1.8126027397260274, + "grad_norm": 2.377120018005371, + "learning_rate": 2.227863525588952e-06, + "loss": 2.0903, + "step": 1654 + }, + { + "epoch": 1.8136986301369862, + "grad_norm": 2.1345274448394775, + "learning_rate": 2.2258326563769295e-06, + "loss": 2.066, + "step": 1655 + }, + { + "epoch": 1.8147945205479452, + "grad_norm": 1.8935933113098145, + "learning_rate": 2.2238017871649067e-06, + "loss": 2.083, + "step": 1656 + }, + { + "epoch": 1.8158904109589042, + "grad_norm": 1.8766530752182007, + "learning_rate": 2.221770917952884e-06, + "loss": 2.0792, + "step": 1657 + }, + { + "epoch": 1.816986301369863, + "grad_norm": 2.24202823638916, + "learning_rate": 2.2197400487408613e-06, + "loss": 2.1276, + "step": 1658 + }, + { + "epoch": 1.8180821917808219, + "grad_norm": 2.168358325958252, + "learning_rate": 2.2177091795288386e-06, + "loss": 2.1497, + "step": 1659 + }, + { + "epoch": 1.819178082191781, + "grad_norm": 2.052833080291748, + "learning_rate": 2.215678310316816e-06, + "loss": 2.0853, + "step": 1660 + }, + { + "epoch": 1.819178082191781, + "eval_loss": 2.270122528076172, + "eval_runtime": 44.1793, + "eval_samples_per_second": 22.635, + "eval_steps_per_second": 0.951, + "step": 1660 + }, + { + "epoch": 1.8202739726027397, + "grad_norm": 2.09708571434021, + "learning_rate": 2.213647441104793e-06, + "loss": 2.1151, + "step": 1661 + }, + { + "epoch": 1.8213698630136985, + "grad_norm": 1.928503155708313, + "learning_rate": 2.2116165718927704e-06, + "loss": 2.1234, + "step": 1662 + }, + { + "epoch": 1.8224657534246576, + "grad_norm": 1.9545074701309204, + "learning_rate": 2.2095857026807473e-06, + "loss": 2.1134, + "step": 1663 + }, + { + "epoch": 1.8235616438356166, + "grad_norm": 2.1161389350891113, + "learning_rate": 2.2075548334687245e-06, + "loss": 2.1347, + "step": 1664 + }, + { + "epoch": 1.8246575342465754, + "grad_norm": 1.9292371273040771, + "learning_rate": 2.205523964256702e-06, + "loss": 2.0911, + "step": 1665 + }, + { + "epoch": 1.8257534246575342, + "grad_norm": 1.8706191778182983, + "learning_rate": 2.203493095044679e-06, + "loss": 2.0917, + "step": 1666 + }, + { + "epoch": 1.8268493150684932, + "grad_norm": 1.9825111627578735, + "learning_rate": 2.2014622258326564e-06, + "loss": 2.125, + "step": 1667 + }, + { + "epoch": 1.827945205479452, + "grad_norm": 2.0262181758880615, + "learning_rate": 2.1994313566206336e-06, + "loss": 2.153, + "step": 1668 + }, + { + "epoch": 1.8290410958904109, + "grad_norm": 2.0071237087249756, + "learning_rate": 2.197400487408611e-06, + "loss": 2.1625, + "step": 1669 + }, + { + "epoch": 1.83013698630137, + "grad_norm": 1.940076231956482, + "learning_rate": 2.195369618196588e-06, + "loss": 2.1052, + "step": 1670 + }, + { + "epoch": 1.83013698630137, + "eval_loss": 2.2711451053619385, + "eval_runtime": 43.9311, + "eval_samples_per_second": 22.763, + "eval_steps_per_second": 0.956, + "step": 1670 + }, + { + "epoch": 1.831232876712329, + "grad_norm": 2.103487014770508, + "learning_rate": 2.1933387489845655e-06, + "loss": 2.087, + "step": 1671 + }, + { + "epoch": 1.8323287671232875, + "grad_norm": 1.935276985168457, + "learning_rate": 2.1913078797725427e-06, + "loss": 2.1196, + "step": 1672 + }, + { + "epoch": 1.8334246575342465, + "grad_norm": 1.987322449684143, + "learning_rate": 2.18927701056052e-06, + "loss": 2.1354, + "step": 1673 + }, + { + "epoch": 1.8345205479452056, + "grad_norm": 2.0403592586517334, + "learning_rate": 2.1872461413484973e-06, + "loss": 2.1715, + "step": 1674 + }, + { + "epoch": 1.8356164383561644, + "grad_norm": 2.0403592586517334, + "learning_rate": 2.1872461413484973e-06, + "loss": 2.1153, + "step": 1675 + }, + { + "epoch": 1.8367123287671232, + "grad_norm": 2.160753011703491, + "learning_rate": 2.1852152721364746e-06, + "loss": 2.1105, + "step": 1676 + }, + { + "epoch": 1.8378082191780822, + "grad_norm": 1.8320366144180298, + "learning_rate": 2.183184402924452e-06, + "loss": 2.0853, + "step": 1677 + }, + { + "epoch": 1.838904109589041, + "grad_norm": 1.9792540073394775, + "learning_rate": 2.181153533712429e-06, + "loss": 2.1478, + "step": 1678 + }, + { + "epoch": 1.8399999999999999, + "grad_norm": 2.0344042778015137, + "learning_rate": 2.1791226645004064e-06, + "loss": 2.1177, + "step": 1679 + }, + { + "epoch": 1.841095890410959, + "grad_norm": 1.8672499656677246, + "learning_rate": 2.1770917952883837e-06, + "loss": 2.1098, + "step": 1680 + }, + { + "epoch": 1.841095890410959, + "eval_loss": 2.2701213359832764, + "eval_runtime": 43.0484, + "eval_samples_per_second": 23.23, + "eval_steps_per_second": 0.976, + "step": 1680 + }, + { + "epoch": 1.842191780821918, + "grad_norm": 1.9291775226593018, + "learning_rate": 2.175060926076361e-06, + "loss": 2.1659, + "step": 1681 + }, + { + "epoch": 1.8432876712328767, + "grad_norm": 1.9650444984436035, + "learning_rate": 2.1730300568643382e-06, + "loss": 2.1764, + "step": 1682 + }, + { + "epoch": 1.8443835616438355, + "grad_norm": 1.904618501663208, + "learning_rate": 2.1709991876523155e-06, + "loss": 2.1047, + "step": 1683 + }, + { + "epoch": 1.8454794520547946, + "grad_norm": 2.0084192752838135, + "learning_rate": 2.1689683184402928e-06, + "loss": 2.0719, + "step": 1684 + }, + { + "epoch": 1.8465753424657534, + "grad_norm": 2.0233888626098633, + "learning_rate": 2.16693744922827e-06, + "loss": 2.1482, + "step": 1685 + }, + { + "epoch": 1.8476712328767122, + "grad_norm": 1.9512053728103638, + "learning_rate": 2.1649065800162473e-06, + "loss": 2.1116, + "step": 1686 + }, + { + "epoch": 1.8487671232876712, + "grad_norm": 1.9261832237243652, + "learning_rate": 2.1628757108042246e-06, + "loss": 2.1025, + "step": 1687 + }, + { + "epoch": 1.8498630136986303, + "grad_norm": 2.0704665184020996, + "learning_rate": 2.160844841592202e-06, + "loss": 2.0943, + "step": 1688 + }, + { + "epoch": 1.850958904109589, + "grad_norm": 1.8882733583450317, + "learning_rate": 2.1588139723801787e-06, + "loss": 2.09, + "step": 1689 + }, + { + "epoch": 1.8520547945205479, + "grad_norm": 1.9136085510253906, + "learning_rate": 2.156783103168156e-06, + "loss": 2.1074, + "step": 1690 + }, + { + "epoch": 1.8520547945205479, + "eval_loss": 2.2687957286834717, + "eval_runtime": 44.1571, + "eval_samples_per_second": 22.646, + "eval_steps_per_second": 0.951, + "step": 1690 + }, + { + "epoch": 1.853150684931507, + "grad_norm": 1.9338082075119019, + "learning_rate": 2.1547522339561333e-06, + "loss": 2.0945, + "step": 1691 + }, + { + "epoch": 1.8542465753424657, + "grad_norm": 1.862281084060669, + "learning_rate": 2.1527213647441106e-06, + "loss": 2.1355, + "step": 1692 + }, + { + "epoch": 1.8553424657534245, + "grad_norm": 1.8184510469436646, + "learning_rate": 2.150690495532088e-06, + "loss": 2.1628, + "step": 1693 + }, + { + "epoch": 1.8564383561643836, + "grad_norm": 1.9667037725448608, + "learning_rate": 2.148659626320065e-06, + "loss": 2.1308, + "step": 1694 + }, + { + "epoch": 1.8575342465753426, + "grad_norm": 2.0214083194732666, + "learning_rate": 2.1466287571080424e-06, + "loss": 2.1363, + "step": 1695 + }, + { + "epoch": 1.8586301369863014, + "grad_norm": 1.8136624097824097, + "learning_rate": 2.1445978878960197e-06, + "loss": 2.1074, + "step": 1696 + }, + { + "epoch": 1.8597260273972602, + "grad_norm": 2.1120195388793945, + "learning_rate": 2.142567018683997e-06, + "loss": 2.1115, + "step": 1697 + }, + { + "epoch": 1.8608219178082193, + "grad_norm": 1.9148973226547241, + "learning_rate": 2.140536149471974e-06, + "loss": 2.1554, + "step": 1698 + }, + { + "epoch": 1.861917808219178, + "grad_norm": 1.9684176445007324, + "learning_rate": 2.1385052802599515e-06, + "loss": 2.0944, + "step": 1699 + }, + { + "epoch": 1.8630136986301369, + "grad_norm": 2.0816597938537598, + "learning_rate": 2.1364744110479288e-06, + "loss": 2.1063, + "step": 1700 + }, + { + "epoch": 1.8630136986301369, + "eval_loss": 2.2695536613464355, + "eval_runtime": 48.9529, + "eval_samples_per_second": 20.428, + "eval_steps_per_second": 0.858, + "step": 1700 + }, + { + "epoch": 1.864109589041096, + "grad_norm": 1.9451358318328857, + "learning_rate": 2.134443541835906e-06, + "loss": 2.1521, + "step": 1701 + }, + { + "epoch": 1.865205479452055, + "grad_norm": 1.9210634231567383, + "learning_rate": 2.132412672623883e-06, + "loss": 2.1216, + "step": 1702 + }, + { + "epoch": 1.8663013698630135, + "grad_norm": 2.0976240634918213, + "learning_rate": 2.13038180341186e-06, + "loss": 2.0947, + "step": 1703 + }, + { + "epoch": 1.8673972602739726, + "grad_norm": 1.9917279481887817, + "learning_rate": 2.1283509341998374e-06, + "loss": 2.1267, + "step": 1704 + }, + { + "epoch": 1.8684931506849316, + "grad_norm": 2.1310553550720215, + "learning_rate": 2.1263200649878147e-06, + "loss": 2.1566, + "step": 1705 + }, + { + "epoch": 1.8695890410958904, + "grad_norm": 2.0364415645599365, + "learning_rate": 2.124289195775792e-06, + "loss": 2.1077, + "step": 1706 + }, + { + "epoch": 1.8706849315068492, + "grad_norm": 1.8810032606124878, + "learning_rate": 2.1222583265637693e-06, + "loss": 2.1061, + "step": 1707 + }, + { + "epoch": 1.8717808219178083, + "grad_norm": 2.034438371658325, + "learning_rate": 2.1202274573517465e-06, + "loss": 2.0523, + "step": 1708 + }, + { + "epoch": 1.872876712328767, + "grad_norm": 2.0231120586395264, + "learning_rate": 2.118196588139724e-06, + "loss": 2.0829, + "step": 1709 + }, + { + "epoch": 1.8739726027397259, + "grad_norm": 2.0092573165893555, + "learning_rate": 2.116165718927701e-06, + "loss": 2.1101, + "step": 1710 + }, + { + "epoch": 1.8739726027397259, + "eval_loss": 2.2675492763519287, + "eval_runtime": 43.0128, + "eval_samples_per_second": 23.249, + "eval_steps_per_second": 0.976, + "step": 1710 + }, + { + "epoch": 1.875068493150685, + "grad_norm": 1.884216070175171, + "learning_rate": 2.1141348497156784e-06, + "loss": 2.1249, + "step": 1711 + }, + { + "epoch": 1.876164383561644, + "grad_norm": 1.954170823097229, + "learning_rate": 2.1121039805036556e-06, + "loss": 2.0915, + "step": 1712 + }, + { + "epoch": 1.8772602739726028, + "grad_norm": 2.0208632946014404, + "learning_rate": 2.110073111291633e-06, + "loss": 2.0824, + "step": 1713 + }, + { + "epoch": 1.8783561643835616, + "grad_norm": 1.9104353189468384, + "learning_rate": 2.10804224207961e-06, + "loss": 2.1265, + "step": 1714 + }, + { + "epoch": 1.8794520547945206, + "grad_norm": 1.945338249206543, + "learning_rate": 2.1060113728675875e-06, + "loss": 2.1403, + "step": 1715 + }, + { + "epoch": 1.8805479452054794, + "grad_norm": 2.006586790084839, + "learning_rate": 2.1039805036555647e-06, + "loss": 2.1439, + "step": 1716 + }, + { + "epoch": 1.8816438356164382, + "grad_norm": 1.941493034362793, + "learning_rate": 2.101949634443542e-06, + "loss": 2.0415, + "step": 1717 + }, + { + "epoch": 1.8827397260273973, + "grad_norm": 2.146116018295288, + "learning_rate": 2.0999187652315193e-06, + "loss": 2.1118, + "step": 1718 + }, + { + "epoch": 1.8838356164383563, + "grad_norm": 2.0794291496276855, + "learning_rate": 2.0978878960194966e-06, + "loss": 2.0731, + "step": 1719 + }, + { + "epoch": 1.884931506849315, + "grad_norm": 2.1100215911865234, + "learning_rate": 2.095857026807474e-06, + "loss": 2.1303, + "step": 1720 + }, + { + "epoch": 1.884931506849315, + "eval_loss": 2.267274856567383, + "eval_runtime": 48.0078, + "eval_samples_per_second": 20.83, + "eval_steps_per_second": 0.875, + "step": 1720 + }, + { + "epoch": 1.886027397260274, + "grad_norm": 1.9509919881820679, + "learning_rate": 2.093826157595451e-06, + "loss": 2.1191, + "step": 1721 + }, + { + "epoch": 1.887123287671233, + "grad_norm": 2.0161266326904297, + "learning_rate": 2.0917952883834284e-06, + "loss": 2.104, + "step": 1722 + }, + { + "epoch": 1.8882191780821918, + "grad_norm": 2.07547664642334, + "learning_rate": 2.0897644191714057e-06, + "loss": 2.1299, + "step": 1723 + }, + { + "epoch": 1.8893150684931506, + "grad_norm": 2.1406030654907227, + "learning_rate": 2.087733549959383e-06, + "loss": 2.1385, + "step": 1724 + }, + { + "epoch": 1.8904109589041096, + "grad_norm": 1.9793925285339355, + "learning_rate": 2.0857026807473602e-06, + "loss": 2.0814, + "step": 1725 + }, + { + "epoch": 1.8915068493150686, + "grad_norm": 1.9256374835968018, + "learning_rate": 2.0836718115353375e-06, + "loss": 2.0708, + "step": 1726 + }, + { + "epoch": 1.8926027397260274, + "grad_norm": 1.960571527481079, + "learning_rate": 2.0816409423233144e-06, + "loss": 2.156, + "step": 1727 + }, + { + "epoch": 1.8936986301369862, + "grad_norm": 2.1507158279418945, + "learning_rate": 2.0796100731112916e-06, + "loss": 2.0958, + "step": 1728 + }, + { + "epoch": 1.8947945205479453, + "grad_norm": 2.2358651161193848, + "learning_rate": 2.077579203899269e-06, + "loss": 2.11, + "step": 1729 + }, + { + "epoch": 1.895890410958904, + "grad_norm": 2.0311756134033203, + "learning_rate": 2.075548334687246e-06, + "loss": 2.072, + "step": 1730 + }, + { + "epoch": 1.895890410958904, + "eval_loss": 2.2663629055023193, + "eval_runtime": 50.5422, + "eval_samples_per_second": 19.785, + "eval_steps_per_second": 0.831, + "step": 1730 + }, + { + "epoch": 1.896986301369863, + "grad_norm": 1.98939049243927, + "learning_rate": 2.0735174654752235e-06, + "loss": 2.1327, + "step": 1731 + }, + { + "epoch": 1.898082191780822, + "grad_norm": 2.009493350982666, + "learning_rate": 2.0714865962632007e-06, + "loss": 2.0798, + "step": 1732 + }, + { + "epoch": 1.899178082191781, + "grad_norm": 2.1011271476745605, + "learning_rate": 2.069455727051178e-06, + "loss": 2.1194, + "step": 1733 + }, + { + "epoch": 1.9002739726027398, + "grad_norm": 1.9027225971221924, + "learning_rate": 2.0674248578391553e-06, + "loss": 2.0955, + "step": 1734 + }, + { + "epoch": 1.9013698630136986, + "grad_norm": 1.9784867763519287, + "learning_rate": 2.0653939886271326e-06, + "loss": 2.1012, + "step": 1735 + }, + { + "epoch": 1.9024657534246576, + "grad_norm": 2.0302555561065674, + "learning_rate": 2.06336311941511e-06, + "loss": 2.1472, + "step": 1736 + }, + { + "epoch": 1.9035616438356164, + "grad_norm": 1.81975257396698, + "learning_rate": 2.061332250203087e-06, + "loss": 2.1062, + "step": 1737 + }, + { + "epoch": 1.9046575342465752, + "grad_norm": 1.9868931770324707, + "learning_rate": 2.0593013809910644e-06, + "loss": 2.0829, + "step": 1738 + }, + { + "epoch": 1.9057534246575343, + "grad_norm": 2.001765012741089, + "learning_rate": 2.0572705117790417e-06, + "loss": 2.1113, + "step": 1739 + }, + { + "epoch": 1.9068493150684933, + "grad_norm": 1.891752004623413, + "learning_rate": 2.055239642567019e-06, + "loss": 2.1311, + "step": 1740 + }, + { + "epoch": 1.9068493150684933, + "eval_loss": 2.2680604457855225, + "eval_runtime": 50.0795, + "eval_samples_per_second": 19.968, + "eval_steps_per_second": 0.839, + "step": 1740 + }, + { + "epoch": 1.907945205479452, + "grad_norm": 1.978259563446045, + "learning_rate": 2.0532087733549962e-06, + "loss": 2.1187, + "step": 1741 + }, + { + "epoch": 1.909041095890411, + "grad_norm": 1.8946164846420288, + "learning_rate": 2.0511779041429735e-06, + "loss": 2.1189, + "step": 1742 + }, + { + "epoch": 1.91013698630137, + "grad_norm": 1.8812536001205444, + "learning_rate": 2.0491470349309508e-06, + "loss": 2.0895, + "step": 1743 + }, + { + "epoch": 1.9112328767123288, + "grad_norm": 1.990494728088379, + "learning_rate": 2.047116165718928e-06, + "loss": 2.1192, + "step": 1744 + }, + { + "epoch": 1.9123287671232876, + "grad_norm": 2.0288472175598145, + "learning_rate": 2.0450852965069053e-06, + "loss": 2.1182, + "step": 1745 + }, + { + "epoch": 1.9134246575342466, + "grad_norm": 2.0876214504241943, + "learning_rate": 2.0430544272948826e-06, + "loss": 2.1546, + "step": 1746 + }, + { + "epoch": 1.9145205479452054, + "grad_norm": 1.9842756986618042, + "learning_rate": 2.04102355808286e-06, + "loss": 2.1126, + "step": 1747 + }, + { + "epoch": 1.9156164383561642, + "grad_norm": 2.0303354263305664, + "learning_rate": 2.038992688870837e-06, + "loss": 2.0771, + "step": 1748 + }, + { + "epoch": 1.9167123287671233, + "grad_norm": 2.0054662227630615, + "learning_rate": 2.0369618196588144e-06, + "loss": 2.0895, + "step": 1749 + }, + { + "epoch": 1.9178082191780823, + "grad_norm": 1.9232923984527588, + "learning_rate": 2.0349309504467917e-06, + "loss": 2.0952, + "step": 1750 + }, + { + "epoch": 1.9178082191780823, + "eval_loss": 2.266294240951538, + "eval_runtime": 53.7831, + "eval_samples_per_second": 18.593, + "eval_steps_per_second": 0.781, + "step": 1750 + }, + { + "epoch": 1.9189041095890411, + "grad_norm": 1.9103394746780396, + "learning_rate": 2.032900081234769e-06, + "loss": 2.1324, + "step": 1751 + }, + { + "epoch": 1.92, + "grad_norm": 2.0642080307006836, + "learning_rate": 2.030869212022746e-06, + "loss": 2.1495, + "step": 1752 + }, + { + "epoch": 1.921095890410959, + "grad_norm": 1.9586466550827026, + "learning_rate": 2.028838342810723e-06, + "loss": 2.1036, + "step": 1753 + }, + { + "epoch": 1.9221917808219178, + "grad_norm": 2.024435043334961, + "learning_rate": 2.0268074735987004e-06, + "loss": 2.1049, + "step": 1754 + }, + { + "epoch": 1.9232876712328766, + "grad_norm": 1.9115983247756958, + "learning_rate": 2.0247766043866777e-06, + "loss": 2.0889, + "step": 1755 + }, + { + "epoch": 1.9243835616438356, + "grad_norm": 1.9067517518997192, + "learning_rate": 2.022745735174655e-06, + "loss": 2.1432, + "step": 1756 + }, + { + "epoch": 1.9254794520547946, + "grad_norm": 2.0182650089263916, + "learning_rate": 2.020714865962632e-06, + "loss": 2.1011, + "step": 1757 + }, + { + "epoch": 1.9265753424657535, + "grad_norm": 2.0648891925811768, + "learning_rate": 2.0186839967506095e-06, + "loss": 2.1266, + "step": 1758 + }, + { + "epoch": 1.9276712328767123, + "grad_norm": 2.0217037200927734, + "learning_rate": 2.0166531275385868e-06, + "loss": 2.1206, + "step": 1759 + }, + { + "epoch": 1.9287671232876713, + "grad_norm": 1.9545989036560059, + "learning_rate": 2.014622258326564e-06, + "loss": 2.0822, + "step": 1760 + }, + { + "epoch": 1.9287671232876713, + "eval_loss": 2.2664599418640137, + "eval_runtime": 53.4817, + "eval_samples_per_second": 18.698, + "eval_steps_per_second": 0.785, + "step": 1760 + }, + { + "epoch": 1.9298630136986301, + "grad_norm": 2.047677516937256, + "learning_rate": 2.0125913891145413e-06, + "loss": 2.1169, + "step": 1761 + }, + { + "epoch": 1.930958904109589, + "grad_norm": 2.170466184616089, + "learning_rate": 2.0105605199025186e-06, + "loss": 2.1211, + "step": 1762 + }, + { + "epoch": 1.932054794520548, + "grad_norm": 1.9144518375396729, + "learning_rate": 2.008529650690496e-06, + "loss": 2.0912, + "step": 1763 + }, + { + "epoch": 1.933150684931507, + "grad_norm": 2.004936933517456, + "learning_rate": 2.0064987814784727e-06, + "loss": 2.136, + "step": 1764 + }, + { + "epoch": 1.9342465753424658, + "grad_norm": 1.8312169313430786, + "learning_rate": 2.00446791226645e-06, + "loss": 2.1151, + "step": 1765 + }, + { + "epoch": 1.9353424657534246, + "grad_norm": 1.9144090414047241, + "learning_rate": 2.0024370430544273e-06, + "loss": 2.0835, + "step": 1766 + }, + { + "epoch": 1.9364383561643836, + "grad_norm": 2.020298719406128, + "learning_rate": 2.0004061738424045e-06, + "loss": 2.1203, + "step": 1767 + }, + { + "epoch": 1.9375342465753425, + "grad_norm": 1.9231114387512207, + "learning_rate": 1.998375304630382e-06, + "loss": 2.1425, + "step": 1768 + }, + { + "epoch": 1.9386301369863013, + "grad_norm": 1.8474186658859253, + "learning_rate": 1.996344435418359e-06, + "loss": 2.1071, + "step": 1769 + }, + { + "epoch": 1.9397260273972603, + "grad_norm": 2.171400785446167, + "learning_rate": 1.9943135662063364e-06, + "loss": 2.1344, + "step": 1770 + }, + { + "epoch": 1.9397260273972603, + "eval_loss": 2.266789436340332, + "eval_runtime": 50.4218, + "eval_samples_per_second": 19.833, + "eval_steps_per_second": 0.833, + "step": 1770 + }, + { + "epoch": 1.9408219178082193, + "grad_norm": 1.929115891456604, + "learning_rate": 1.9922826969943136e-06, + "loss": 2.1205, + "step": 1771 + }, + { + "epoch": 1.941917808219178, + "grad_norm": 1.961046814918518, + "learning_rate": 1.990251827782291e-06, + "loss": 2.1396, + "step": 1772 + }, + { + "epoch": 1.943013698630137, + "grad_norm": 2.056424617767334, + "learning_rate": 1.988220958570268e-06, + "loss": 2.125, + "step": 1773 + }, + { + "epoch": 1.944109589041096, + "grad_norm": 1.9814075231552124, + "learning_rate": 1.9861900893582455e-06, + "loss": 2.1297, + "step": 1774 + }, + { + "epoch": 1.9452054794520548, + "grad_norm": 1.937506079673767, + "learning_rate": 1.9841592201462227e-06, + "loss": 2.1288, + "step": 1775 + }, + { + "epoch": 1.9463013698630136, + "grad_norm": 1.902376413345337, + "learning_rate": 1.9821283509342e-06, + "loss": 2.0782, + "step": 1776 + }, + { + "epoch": 1.9473972602739726, + "grad_norm": 2.0806174278259277, + "learning_rate": 1.9800974817221773e-06, + "loss": 2.0778, + "step": 1777 + }, + { + "epoch": 1.9484931506849315, + "grad_norm": 2.0370235443115234, + "learning_rate": 1.9780666125101546e-06, + "loss": 2.1304, + "step": 1778 + }, + { + "epoch": 1.9495890410958903, + "grad_norm": 1.980531930923462, + "learning_rate": 1.976035743298132e-06, + "loss": 2.0989, + "step": 1779 + }, + { + "epoch": 1.9506849315068493, + "grad_norm": 1.96549654006958, + "learning_rate": 1.974004874086109e-06, + "loss": 2.1556, + "step": 1780 + }, + { + "epoch": 1.9506849315068493, + "eval_loss": 2.266958475112915, + "eval_runtime": 52.3331, + "eval_samples_per_second": 19.108, + "eval_steps_per_second": 0.803, + "step": 1780 + }, + { + "epoch": 1.9517808219178083, + "grad_norm": 2.1921818256378174, + "learning_rate": 1.9719740048740864e-06, + "loss": 2.0666, + "step": 1781 + }, + { + "epoch": 1.9528767123287671, + "grad_norm": 1.9451124668121338, + "learning_rate": 1.9699431356620637e-06, + "loss": 2.1078, + "step": 1782 + }, + { + "epoch": 1.953972602739726, + "grad_norm": 1.9163907766342163, + "learning_rate": 1.967912266450041e-06, + "loss": 2.0955, + "step": 1783 + }, + { + "epoch": 1.955068493150685, + "grad_norm": 2.1186892986297607, + "learning_rate": 1.9658813972380182e-06, + "loss": 2.0826, + "step": 1784 + }, + { + "epoch": 1.9561643835616438, + "grad_norm": 2.1072819232940674, + "learning_rate": 1.9638505280259955e-06, + "loss": 2.1869, + "step": 1785 + }, + { + "epoch": 1.9572602739726026, + "grad_norm": 1.9530322551727295, + "learning_rate": 1.9618196588139728e-06, + "loss": 2.0983, + "step": 1786 + }, + { + "epoch": 1.9583561643835616, + "grad_norm": 2.0034027099609375, + "learning_rate": 1.95978878960195e-06, + "loss": 2.1121, + "step": 1787 + }, + { + "epoch": 1.9594520547945207, + "grad_norm": 2.1277832984924316, + "learning_rate": 1.9577579203899273e-06, + "loss": 2.1769, + "step": 1788 + }, + { + "epoch": 1.9605479452054795, + "grad_norm": 1.9420982599258423, + "learning_rate": 1.955727051177904e-06, + "loss": 2.1424, + "step": 1789 + }, + { + "epoch": 1.9616438356164383, + "grad_norm": 1.9175841808319092, + "learning_rate": 1.9536961819658815e-06, + "loss": 2.0957, + "step": 1790 + }, + { + "epoch": 1.9616438356164383, + "eval_loss": 2.267153739929199, + "eval_runtime": 47.5802, + "eval_samples_per_second": 21.017, + "eval_steps_per_second": 0.883, + "step": 1790 + }, + { + "epoch": 1.9627397260273973, + "grad_norm": 1.9897676706314087, + "learning_rate": 1.9516653127538587e-06, + "loss": 2.0809, + "step": 1791 + }, + { + "epoch": 1.9638356164383561, + "grad_norm": 2.0855610370635986, + "learning_rate": 1.949634443541836e-06, + "loss": 2.0924, + "step": 1792 + }, + { + "epoch": 1.964931506849315, + "grad_norm": 2.0065109729766846, + "learning_rate": 1.9476035743298133e-06, + "loss": 2.0819, + "step": 1793 + }, + { + "epoch": 1.966027397260274, + "grad_norm": 1.9256092309951782, + "learning_rate": 1.9455727051177906e-06, + "loss": 2.1389, + "step": 1794 + }, + { + "epoch": 1.967123287671233, + "grad_norm": 1.9086920022964478, + "learning_rate": 1.943541835905768e-06, + "loss": 2.1106, + "step": 1795 + }, + { + "epoch": 1.9682191780821918, + "grad_norm": 2.0217244625091553, + "learning_rate": 1.941510966693745e-06, + "loss": 2.1624, + "step": 1796 + }, + { + "epoch": 1.9693150684931506, + "grad_norm": 2.005068063735962, + "learning_rate": 1.9394800974817224e-06, + "loss": 2.0751, + "step": 1797 + }, + { + "epoch": 1.9704109589041097, + "grad_norm": 2.0949530601501465, + "learning_rate": 1.9374492282696997e-06, + "loss": 2.1447, + "step": 1798 + }, + { + "epoch": 1.9715068493150685, + "grad_norm": 1.8868833780288696, + "learning_rate": 1.935418359057677e-06, + "loss": 2.1568, + "step": 1799 + }, + { + "epoch": 1.9726027397260273, + "grad_norm": 1.889626145362854, + "learning_rate": 1.9333874898456542e-06, + "loss": 2.1152, + "step": 1800 + }, + { + "epoch": 1.9726027397260273, + "eval_loss": 2.268470287322998, + "eval_runtime": 44.7042, + "eval_samples_per_second": 22.369, + "eval_steps_per_second": 0.94, + "step": 1800 + }, + { + "epoch": 1.9736986301369863, + "grad_norm": 1.9561820030212402, + "learning_rate": 1.931356620633631e-06, + "loss": 2.1584, + "step": 1801 + }, + { + "epoch": 1.9747945205479454, + "grad_norm": 1.9774333238601685, + "learning_rate": 1.9293257514216083e-06, + "loss": 2.0726, + "step": 1802 + }, + { + "epoch": 1.9758904109589042, + "grad_norm": 1.9008616209030151, + "learning_rate": 1.9272948822095856e-06, + "loss": 2.1222, + "step": 1803 + }, + { + "epoch": 1.976986301369863, + "grad_norm": 1.9981689453125, + "learning_rate": 1.925264012997563e-06, + "loss": 2.0761, + "step": 1804 + }, + { + "epoch": 1.978082191780822, + "grad_norm": 1.926156997680664, + "learning_rate": 1.92323314378554e-06, + "loss": 2.0962, + "step": 1805 + }, + { + "epoch": 1.9791780821917808, + "grad_norm": 1.9634530544281006, + "learning_rate": 1.9212022745735174e-06, + "loss": 2.1349, + "step": 1806 + }, + { + "epoch": 1.9802739726027396, + "grad_norm": 1.9713245630264282, + "learning_rate": 1.9191714053614947e-06, + "loss": 2.0881, + "step": 1807 + }, + { + "epoch": 1.9813698630136987, + "grad_norm": 1.9258323907852173, + "learning_rate": 1.917140536149472e-06, + "loss": 2.0744, + "step": 1808 + }, + { + "epoch": 1.9824657534246577, + "grad_norm": 1.9645394086837769, + "learning_rate": 1.9151096669374493e-06, + "loss": 2.0701, + "step": 1809 + }, + { + "epoch": 1.9835616438356163, + "grad_norm": 2.145430088043213, + "learning_rate": 1.9130787977254266e-06, + "loss": 2.1372, + "step": 1810 + }, + { + "epoch": 1.9835616438356163, + "eval_loss": 2.269702196121216, + "eval_runtime": 41.513, + "eval_samples_per_second": 24.089, + "eval_steps_per_second": 1.012, + "step": 1810 + }, + { + "epoch": 1.9846575342465753, + "grad_norm": 2.1812174320220947, + "learning_rate": 1.911047928513404e-06, + "loss": 2.1209, + "step": 1811 + }, + { + "epoch": 1.9857534246575343, + "grad_norm": 1.933090329170227, + "learning_rate": 1.909017059301381e-06, + "loss": 2.0905, + "step": 1812 + }, + { + "epoch": 1.9868493150684932, + "grad_norm": 2.0478832721710205, + "learning_rate": 1.9069861900893586e-06, + "loss": 2.0921, + "step": 1813 + }, + { + "epoch": 1.987945205479452, + "grad_norm": 1.9921154975891113, + "learning_rate": 1.9049553208773354e-06, + "loss": 2.1232, + "step": 1814 + }, + { + "epoch": 1.989041095890411, + "grad_norm": 1.8685261011123657, + "learning_rate": 1.9029244516653127e-06, + "loss": 2.0978, + "step": 1815 + }, + { + "epoch": 1.9901369863013698, + "grad_norm": 2.045020818710327, + "learning_rate": 1.90089358245329e-06, + "loss": 2.0614, + "step": 1816 + }, + { + "epoch": 1.9912328767123286, + "grad_norm": 2.059267520904541, + "learning_rate": 1.8988627132412673e-06, + "loss": 2.1098, + "step": 1817 + }, + { + "epoch": 1.9923287671232877, + "grad_norm": 2.0033164024353027, + "learning_rate": 1.8968318440292445e-06, + "loss": 2.135, + "step": 1818 + }, + { + "epoch": 1.9934246575342467, + "grad_norm": 1.8544939756393433, + "learning_rate": 1.8948009748172218e-06, + "loss": 2.0644, + "step": 1819 + }, + { + "epoch": 1.9945205479452055, + "grad_norm": 1.8415113687515259, + "learning_rate": 1.892770105605199e-06, + "loss": 2.1283, + "step": 1820 + }, + { + "epoch": 1.9945205479452055, + "eval_loss": 2.267810344696045, + "eval_runtime": 43.9738, + "eval_samples_per_second": 22.741, + "eval_steps_per_second": 0.955, + "step": 1820 + }, + { + "epoch": 1.9956164383561643, + "grad_norm": 1.90078866481781, + "learning_rate": 1.8907392363931764e-06, + "loss": 2.1274, + "step": 1821 + }, + { + "epoch": 1.9967123287671233, + "grad_norm": 1.8798222541809082, + "learning_rate": 1.8887083671811536e-06, + "loss": 2.1014, + "step": 1822 + }, + { + "epoch": 1.9978082191780822, + "grad_norm": 1.9333341121673584, + "learning_rate": 1.886677497969131e-06, + "loss": 2.1165, + "step": 1823 + }, + { + "epoch": 1.998904109589041, + "grad_norm": 1.9861575365066528, + "learning_rate": 1.8846466287571082e-06, + "loss": 2.036, + "step": 1824 + }, + { + "epoch": 2.0, + "grad_norm": 2.0310938358306885, + "learning_rate": 1.8826157595450855e-06, + "loss": 2.1324, + "step": 1825 + }, + { + "epoch": 2.001095890410959, + "grad_norm": 1.9574612379074097, + "learning_rate": 1.8805848903330625e-06, + "loss": 2.0485, + "step": 1826 + }, + { + "epoch": 2.0021917808219176, + "grad_norm": 2.0283970832824707, + "learning_rate": 1.8785540211210398e-06, + "loss": 2.1768, + "step": 1827 + }, + { + "epoch": 2.0032876712328767, + "grad_norm": 2.062384843826294, + "learning_rate": 1.876523151909017e-06, + "loss": 2.1199, + "step": 1828 + }, + { + "epoch": 2.0043835616438357, + "grad_norm": 2.0543339252471924, + "learning_rate": 1.8744922826969944e-06, + "loss": 2.0887, + "step": 1829 + }, + { + "epoch": 2.0054794520547947, + "grad_norm": 2.1769917011260986, + "learning_rate": 1.8724614134849716e-06, + "loss": 2.0821, + "step": 1830 + }, + { + "epoch": 2.0054794520547947, + "eval_loss": 2.2689950466156006, + "eval_runtime": 43.9448, + "eval_samples_per_second": 22.756, + "eval_steps_per_second": 0.956, + "step": 1830 + }, + { + "epoch": 2.0065753424657533, + "grad_norm": 2.0331335067749023, + "learning_rate": 1.870430544272949e-06, + "loss": 2.0916, + "step": 1831 + }, + { + "epoch": 2.0076712328767123, + "grad_norm": 2.1848011016845703, + "learning_rate": 1.8683996750609262e-06, + "loss": 2.1525, + "step": 1832 + }, + { + "epoch": 2.0087671232876714, + "grad_norm": 2.086441993713379, + "learning_rate": 1.8663688058489035e-06, + "loss": 2.0733, + "step": 1833 + }, + { + "epoch": 2.00986301369863, + "grad_norm": 2.1663239002227783, + "learning_rate": 1.8643379366368807e-06, + "loss": 2.1026, + "step": 1834 + }, + { + "epoch": 2.010958904109589, + "grad_norm": 2.1658689975738525, + "learning_rate": 1.862307067424858e-06, + "loss": 2.1178, + "step": 1835 + }, + { + "epoch": 2.012054794520548, + "grad_norm": 2.043203592300415, + "learning_rate": 1.8602761982128353e-06, + "loss": 2.1025, + "step": 1836 + }, + { + "epoch": 2.013150684931507, + "grad_norm": 2.1104307174682617, + "learning_rate": 1.8582453290008126e-06, + "loss": 2.1218, + "step": 1837 + }, + { + "epoch": 2.0142465753424657, + "grad_norm": 2.078554630279541, + "learning_rate": 1.8562144597887898e-06, + "loss": 2.082, + "step": 1838 + }, + { + "epoch": 2.0153424657534247, + "grad_norm": 2.1440582275390625, + "learning_rate": 1.854183590576767e-06, + "loss": 2.101, + "step": 1839 + }, + { + "epoch": 2.0164383561643837, + "grad_norm": 2.0824856758117676, + "learning_rate": 1.8521527213647442e-06, + "loss": 2.1377, + "step": 1840 + }, + { + "epoch": 2.0164383561643837, + "eval_loss": 2.268453359603882, + "eval_runtime": 43.8166, + "eval_samples_per_second": 22.822, + "eval_steps_per_second": 0.959, + "step": 1840 + }, + { + "epoch": 2.0175342465753423, + "grad_norm": 2.1607978343963623, + "learning_rate": 1.8501218521527215e-06, + "loss": 2.0734, + "step": 1841 + }, + { + "epoch": 2.0186301369863013, + "grad_norm": 2.044123649597168, + "learning_rate": 1.8480909829406987e-06, + "loss": 2.04, + "step": 1842 + }, + { + "epoch": 2.0197260273972604, + "grad_norm": 2.1130807399749756, + "learning_rate": 1.846060113728676e-06, + "loss": 2.069, + "step": 1843 + }, + { + "epoch": 2.020821917808219, + "grad_norm": 2.0879838466644287, + "learning_rate": 1.8440292445166533e-06, + "loss": 2.1006, + "step": 1844 + }, + { + "epoch": 2.021917808219178, + "grad_norm": 2.311849355697632, + "learning_rate": 1.8419983753046306e-06, + "loss": 2.0691, + "step": 1845 + }, + { + "epoch": 2.023013698630137, + "grad_norm": 2.122504711151123, + "learning_rate": 1.8399675060926078e-06, + "loss": 2.1297, + "step": 1846 + }, + { + "epoch": 2.024109589041096, + "grad_norm": 2.135181427001953, + "learning_rate": 1.8379366368805851e-06, + "loss": 2.0951, + "step": 1847 + }, + { + "epoch": 2.0252054794520546, + "grad_norm": 2.0710697174072266, + "learning_rate": 1.8359057676685624e-06, + "loss": 2.0754, + "step": 1848 + }, + { + "epoch": 2.0263013698630137, + "grad_norm": 2.0922486782073975, + "learning_rate": 1.8338748984565397e-06, + "loss": 2.068, + "step": 1849 + }, + { + "epoch": 2.0273972602739727, + "grad_norm": 2.0101380348205566, + "learning_rate": 1.831844029244517e-06, + "loss": 2.0831, + "step": 1850 + }, + { + "epoch": 2.0273972602739727, + "eval_loss": 2.267380952835083, + "eval_runtime": 44.4219, + "eval_samples_per_second": 22.511, + "eval_steps_per_second": 0.945, + "step": 1850 + }, + { + "epoch": 2.0284931506849313, + "grad_norm": 2.012316942214966, + "learning_rate": 1.829813160032494e-06, + "loss": 2.0571, + "step": 1851 + }, + { + "epoch": 2.0295890410958903, + "grad_norm": 2.039604663848877, + "learning_rate": 1.8277822908204713e-06, + "loss": 2.0691, + "step": 1852 + }, + { + "epoch": 2.0306849315068494, + "grad_norm": 2.242504119873047, + "learning_rate": 1.8257514216084486e-06, + "loss": 2.1025, + "step": 1853 + }, + { + "epoch": 2.0317808219178084, + "grad_norm": 2.0629303455352783, + "learning_rate": 1.8237205523964258e-06, + "loss": 2.0803, + "step": 1854 + }, + { + "epoch": 2.032876712328767, + "grad_norm": 2.0935115814208984, + "learning_rate": 1.8216896831844031e-06, + "loss": 2.101, + "step": 1855 + }, + { + "epoch": 2.033972602739726, + "grad_norm": 2.164381504058838, + "learning_rate": 1.8196588139723804e-06, + "loss": 2.1227, + "step": 1856 + }, + { + "epoch": 2.035068493150685, + "grad_norm": 2.240391492843628, + "learning_rate": 1.8176279447603577e-06, + "loss": 2.0549, + "step": 1857 + }, + { + "epoch": 2.0361643835616436, + "grad_norm": 2.1460423469543457, + "learning_rate": 1.815597075548335e-06, + "loss": 2.0858, + "step": 1858 + }, + { + "epoch": 2.0372602739726027, + "grad_norm": 2.0199930667877197, + "learning_rate": 1.8135662063363122e-06, + "loss": 2.1058, + "step": 1859 + }, + { + "epoch": 2.0383561643835617, + "grad_norm": 2.074615955352783, + "learning_rate": 1.8115353371242895e-06, + "loss": 2.113, + "step": 1860 + }, + { + "epoch": 2.0383561643835617, + "eval_loss": 2.2699100971221924, + "eval_runtime": 42.9781, + "eval_samples_per_second": 23.268, + "eval_steps_per_second": 0.977, + "step": 1860 + }, + { + "epoch": 2.0394520547945207, + "grad_norm": 2.091417074203491, + "learning_rate": 1.8095044679122668e-06, + "loss": 2.045, + "step": 1861 + }, + { + "epoch": 2.0405479452054793, + "grad_norm": 2.2032699584960938, + "learning_rate": 1.807473598700244e-06, + "loss": 2.081, + "step": 1862 + }, + { + "epoch": 2.0416438356164384, + "grad_norm": 2.0829155445098877, + "learning_rate": 1.805442729488221e-06, + "loss": 2.0798, + "step": 1863 + }, + { + "epoch": 2.0427397260273974, + "grad_norm": 1.97968590259552, + "learning_rate": 1.8034118602761982e-06, + "loss": 2.0851, + "step": 1864 + }, + { + "epoch": 2.043835616438356, + "grad_norm": 1.9853475093841553, + "learning_rate": 1.8013809910641754e-06, + "loss": 2.0826, + "step": 1865 + }, + { + "epoch": 2.044931506849315, + "grad_norm": 2.0638294219970703, + "learning_rate": 1.7993501218521527e-06, + "loss": 2.0806, + "step": 1866 + }, + { + "epoch": 2.046027397260274, + "grad_norm": 2.0647425651550293, + "learning_rate": 1.79731925264013e-06, + "loss": 2.0613, + "step": 1867 + }, + { + "epoch": 2.047123287671233, + "grad_norm": 2.1473264694213867, + "learning_rate": 1.7952883834281073e-06, + "loss": 2.1211, + "step": 1868 + }, + { + "epoch": 2.0482191780821917, + "grad_norm": 2.037551164627075, + "learning_rate": 1.7932575142160846e-06, + "loss": 2.0727, + "step": 1869 + }, + { + "epoch": 2.0493150684931507, + "grad_norm": 2.010626792907715, + "learning_rate": 1.7912266450040618e-06, + "loss": 2.1013, + "step": 1870 + }, + { + "epoch": 2.0493150684931507, + "eval_loss": 2.266191005706787, + "eval_runtime": 45.1806, + "eval_samples_per_second": 22.133, + "eval_steps_per_second": 0.93, + "step": 1870 + }, + { + "epoch": 2.0504109589041097, + "grad_norm": 1.955633282661438, + "learning_rate": 1.789195775792039e-06, + "loss": 2.114, + "step": 1871 + }, + { + "epoch": 2.0515068493150683, + "grad_norm": 2.1419057846069336, + "learning_rate": 1.7871649065800164e-06, + "loss": 2.0296, + "step": 1872 + }, + { + "epoch": 2.0526027397260274, + "grad_norm": 2.167240858078003, + "learning_rate": 1.7851340373679937e-06, + "loss": 2.1391, + "step": 1873 + }, + { + "epoch": 2.0536986301369864, + "grad_norm": 2.0724527835845947, + "learning_rate": 1.783103168155971e-06, + "loss": 2.0571, + "step": 1874 + }, + { + "epoch": 2.0547945205479454, + "grad_norm": 2.104583263397217, + "learning_rate": 1.7810722989439482e-06, + "loss": 2.0917, + "step": 1875 + }, + { + "epoch": 2.055890410958904, + "grad_norm": 2.083120584487915, + "learning_rate": 1.7790414297319253e-06, + "loss": 2.093, + "step": 1876 + }, + { + "epoch": 2.056986301369863, + "grad_norm": 2.136396646499634, + "learning_rate": 1.7770105605199025e-06, + "loss": 2.0654, + "step": 1877 + }, + { + "epoch": 2.058082191780822, + "grad_norm": 2.1515700817108154, + "learning_rate": 1.7749796913078798e-06, + "loss": 2.05, + "step": 1878 + }, + { + "epoch": 2.0591780821917807, + "grad_norm": 2.1882944107055664, + "learning_rate": 1.772948822095857e-06, + "loss": 2.0974, + "step": 1879 + }, + { + "epoch": 2.0602739726027397, + "grad_norm": 2.261822462081909, + "learning_rate": 1.7709179528838344e-06, + "loss": 2.1375, + "step": 1880 + }, + { + "epoch": 2.0602739726027397, + "eval_loss": 2.267698287963867, + "eval_runtime": 45.8285, + "eval_samples_per_second": 21.82, + "eval_steps_per_second": 0.916, + "step": 1880 + }, + { + "epoch": 2.0613698630136987, + "grad_norm": 2.121238946914673, + "learning_rate": 1.7688870836718116e-06, + "loss": 2.1102, + "step": 1881 + }, + { + "epoch": 2.0624657534246573, + "grad_norm": 2.183793544769287, + "learning_rate": 1.766856214459789e-06, + "loss": 2.0862, + "step": 1882 + }, + { + "epoch": 2.0635616438356164, + "grad_norm": 2.209182024002075, + "learning_rate": 1.7648253452477662e-06, + "loss": 2.088, + "step": 1883 + }, + { + "epoch": 2.0646575342465754, + "grad_norm": 1.9593517780303955, + "learning_rate": 1.7627944760357435e-06, + "loss": 2.0644, + "step": 1884 + }, + { + "epoch": 2.0657534246575344, + "grad_norm": 1.9904463291168213, + "learning_rate": 1.7607636068237208e-06, + "loss": 2.1313, + "step": 1885 + }, + { + "epoch": 2.066849315068493, + "grad_norm": 2.2765681743621826, + "learning_rate": 1.758732737611698e-06, + "loss": 2.0979, + "step": 1886 + }, + { + "epoch": 2.067945205479452, + "grad_norm": 2.066204309463501, + "learning_rate": 1.7567018683996753e-06, + "loss": 2.0934, + "step": 1887 + }, + { + "epoch": 2.069041095890411, + "grad_norm": 2.104768753051758, + "learning_rate": 1.7546709991876524e-06, + "loss": 2.0887, + "step": 1888 + }, + { + "epoch": 2.0701369863013697, + "grad_norm": 2.0848052501678467, + "learning_rate": 1.7526401299756296e-06, + "loss": 2.0635, + "step": 1889 + }, + { + "epoch": 2.0712328767123287, + "grad_norm": 2.052314519882202, + "learning_rate": 1.750609260763607e-06, + "loss": 2.0728, + "step": 1890 + }, + { + "epoch": 2.0712328767123287, + "eval_loss": 2.266915798187256, + "eval_runtime": 46.3568, + "eval_samples_per_second": 21.572, + "eval_steps_per_second": 0.906, + "step": 1890 + }, + { + "epoch": 2.0723287671232877, + "grad_norm": 2.209855318069458, + "learning_rate": 1.7485783915515842e-06, + "loss": 2.0951, + "step": 1891 + }, + { + "epoch": 2.0734246575342468, + "grad_norm": 2.1558358669281006, + "learning_rate": 1.7465475223395615e-06, + "loss": 2.0872, + "step": 1892 + }, + { + "epoch": 2.0745205479452054, + "grad_norm": 2.426797389984131, + "learning_rate": 1.7445166531275387e-06, + "loss": 2.0578, + "step": 1893 + }, + { + "epoch": 2.0756164383561644, + "grad_norm": 2.1906144618988037, + "learning_rate": 1.742485783915516e-06, + "loss": 2.1664, + "step": 1894 + }, + { + "epoch": 2.0767123287671234, + "grad_norm": 2.2041025161743164, + "learning_rate": 1.7404549147034933e-06, + "loss": 2.1532, + "step": 1895 + }, + { + "epoch": 2.077808219178082, + "grad_norm": 2.179513931274414, + "learning_rate": 1.7384240454914706e-06, + "loss": 2.0349, + "step": 1896 + }, + { + "epoch": 2.078904109589041, + "grad_norm": 2.239232301712036, + "learning_rate": 1.7363931762794478e-06, + "loss": 2.0926, + "step": 1897 + }, + { + "epoch": 2.08, + "grad_norm": 2.2452921867370605, + "learning_rate": 1.7343623070674251e-06, + "loss": 2.1082, + "step": 1898 + }, + { + "epoch": 2.081095890410959, + "grad_norm": 2.145571708679199, + "learning_rate": 1.7323314378554024e-06, + "loss": 2.0886, + "step": 1899 + }, + { + "epoch": 2.0821917808219177, + "grad_norm": 2.201815366744995, + "learning_rate": 1.7303005686433797e-06, + "loss": 2.0947, + "step": 1900 + }, + { + "epoch": 2.0821917808219177, + "eval_loss": 2.2664403915405273, + "eval_runtime": 46.4641, + "eval_samples_per_second": 21.522, + "eval_steps_per_second": 0.904, + "step": 1900 + }, + { + "epoch": 2.0832876712328767, + "grad_norm": 2.095641851425171, + "learning_rate": 1.7282696994313567e-06, + "loss": 2.0925, + "step": 1901 + }, + { + "epoch": 2.0843835616438358, + "grad_norm": 2.186495304107666, + "learning_rate": 1.726238830219334e-06, + "loss": 2.0971, + "step": 1902 + }, + { + "epoch": 2.0854794520547943, + "grad_norm": 2.135972023010254, + "learning_rate": 1.7242079610073113e-06, + "loss": 2.076, + "step": 1903 + }, + { + "epoch": 2.0865753424657534, + "grad_norm": 2.1793875694274902, + "learning_rate": 1.7221770917952886e-06, + "loss": 2.1336, + "step": 1904 + }, + { + "epoch": 2.0876712328767124, + "grad_norm": 2.178579330444336, + "learning_rate": 1.7201462225832658e-06, + "loss": 2.0959, + "step": 1905 + }, + { + "epoch": 2.0887671232876714, + "grad_norm": 2.1725568771362305, + "learning_rate": 1.7181153533712431e-06, + "loss": 2.0965, + "step": 1906 + }, + { + "epoch": 2.08986301369863, + "grad_norm": 2.126034736633301, + "learning_rate": 1.7160844841592204e-06, + "loss": 2.0225, + "step": 1907 + }, + { + "epoch": 2.090958904109589, + "grad_norm": 2.0938918590545654, + "learning_rate": 1.7140536149471977e-06, + "loss": 2.0447, + "step": 1908 + }, + { + "epoch": 2.092054794520548, + "grad_norm": 2.1260006427764893, + "learning_rate": 1.712022745735175e-06, + "loss": 2.0739, + "step": 1909 + }, + { + "epoch": 2.0931506849315067, + "grad_norm": 2.2151126861572266, + "learning_rate": 1.7099918765231522e-06, + "loss": 2.0786, + "step": 1910 + }, + { + "epoch": 2.0931506849315067, + "eval_loss": 2.2675349712371826, + "eval_runtime": 43.2795, + "eval_samples_per_second": 23.106, + "eval_steps_per_second": 0.97, + "step": 1910 + }, + { + "epoch": 2.0942465753424657, + "grad_norm": 2.146458625793457, + "learning_rate": 1.7079610073111295e-06, + "loss": 2.0928, + "step": 1911 + }, + { + "epoch": 2.0953424657534248, + "grad_norm": 2.1360816955566406, + "learning_rate": 1.7059301380991068e-06, + "loss": 2.0404, + "step": 1912 + }, + { + "epoch": 2.0964383561643833, + "grad_norm": 2.130659341812134, + "learning_rate": 1.7038992688870836e-06, + "loss": 2.0858, + "step": 1913 + }, + { + "epoch": 2.0975342465753424, + "grad_norm": 2.0875484943389893, + "learning_rate": 1.701868399675061e-06, + "loss": 2.0613, + "step": 1914 + }, + { + "epoch": 2.0986301369863014, + "grad_norm": 2.1744940280914307, + "learning_rate": 1.6998375304630382e-06, + "loss": 2.1012, + "step": 1915 + }, + { + "epoch": 2.0997260273972604, + "grad_norm": 2.195559024810791, + "learning_rate": 1.6978066612510155e-06, + "loss": 2.1595, + "step": 1916 + }, + { + "epoch": 2.100821917808219, + "grad_norm": 2.3818888664245605, + "learning_rate": 1.6957757920389927e-06, + "loss": 2.1246, + "step": 1917 + }, + { + "epoch": 2.101917808219178, + "grad_norm": 2.296870708465576, + "learning_rate": 1.69374492282697e-06, + "loss": 2.0893, + "step": 1918 + }, + { + "epoch": 2.103013698630137, + "grad_norm": 2.3288252353668213, + "learning_rate": 1.6917140536149473e-06, + "loss": 2.0764, + "step": 1919 + }, + { + "epoch": 2.1041095890410957, + "grad_norm": 2.1017708778381348, + "learning_rate": 1.6896831844029246e-06, + "loss": 2.1312, + "step": 1920 + }, + { + "epoch": 2.1041095890410957, + "eval_loss": 2.2681941986083984, + "eval_runtime": 51.4964, + "eval_samples_per_second": 19.419, + "eval_steps_per_second": 0.816, + "step": 1920 + }, + { + "epoch": 2.1052054794520547, + "grad_norm": 2.2251839637756348, + "learning_rate": 1.6876523151909018e-06, + "loss": 2.0967, + "step": 1921 + }, + { + "epoch": 2.1063013698630138, + "grad_norm": 2.1989998817443848, + "learning_rate": 1.6856214459788791e-06, + "loss": 2.1064, + "step": 1922 + }, + { + "epoch": 2.107397260273973, + "grad_norm": 2.2629430294036865, + "learning_rate": 1.6835905767668564e-06, + "loss": 2.0574, + "step": 1923 + }, + { + "epoch": 2.1084931506849314, + "grad_norm": 2.2153608798980713, + "learning_rate": 1.6815597075548337e-06, + "loss": 2.0736, + "step": 1924 + }, + { + "epoch": 2.1095890410958904, + "grad_norm": 2.1380763053894043, + "learning_rate": 1.679528838342811e-06, + "loss": 2.1437, + "step": 1925 + }, + { + "epoch": 2.1106849315068494, + "grad_norm": 2.198996067047119, + "learning_rate": 1.677497969130788e-06, + "loss": 2.0623, + "step": 1926 + }, + { + "epoch": 2.111780821917808, + "grad_norm": 2.1622812747955322, + "learning_rate": 1.6754670999187653e-06, + "loss": 2.072, + "step": 1927 + }, + { + "epoch": 2.112876712328767, + "grad_norm": 2.185896873474121, + "learning_rate": 1.6734362307067426e-06, + "loss": 2.0931, + "step": 1928 + }, + { + "epoch": 2.113972602739726, + "grad_norm": 2.118260383605957, + "learning_rate": 1.6714053614947198e-06, + "loss": 2.0555, + "step": 1929 + }, + { + "epoch": 2.115068493150685, + "grad_norm": 2.070236921310425, + "learning_rate": 1.669374492282697e-06, + "loss": 2.0503, + "step": 1930 + }, + { + "epoch": 2.115068493150685, + "eval_loss": 2.267728328704834, + "eval_runtime": 48.9617, + "eval_samples_per_second": 20.424, + "eval_steps_per_second": 0.858, + "step": 1930 + }, + { + "epoch": 2.1161643835616437, + "grad_norm": 2.2161452770233154, + "learning_rate": 1.6673436230706744e-06, + "loss": 2.1211, + "step": 1931 + }, + { + "epoch": 2.1172602739726027, + "grad_norm": 2.0978891849517822, + "learning_rate": 1.6653127538586517e-06, + "loss": 2.1055, + "step": 1932 + }, + { + "epoch": 2.118356164383562, + "grad_norm": 2.208021879196167, + "learning_rate": 1.663281884646629e-06, + "loss": 2.0754, + "step": 1933 + }, + { + "epoch": 2.1194520547945204, + "grad_norm": 2.0411078929901123, + "learning_rate": 1.6612510154346062e-06, + "loss": 2.0687, + "step": 1934 + }, + { + "epoch": 2.1205479452054794, + "grad_norm": 2.275355339050293, + "learning_rate": 1.6592201462225835e-06, + "loss": 2.1354, + "step": 1935 + }, + { + "epoch": 2.1216438356164384, + "grad_norm": 2.102735996246338, + "learning_rate": 1.6571892770105608e-06, + "loss": 2.105, + "step": 1936 + }, + { + "epoch": 2.1227397260273975, + "grad_norm": 2.1679234504699707, + "learning_rate": 1.655158407798538e-06, + "loss": 2.1314, + "step": 1937 + }, + { + "epoch": 2.123835616438356, + "grad_norm": 2.244574785232544, + "learning_rate": 1.653127538586515e-06, + "loss": 2.0923, + "step": 1938 + }, + { + "epoch": 2.124931506849315, + "grad_norm": 2.0957608222961426, + "learning_rate": 1.6510966693744924e-06, + "loss": 2.1115, + "step": 1939 + }, + { + "epoch": 2.126027397260274, + "grad_norm": 2.0705184936523438, + "learning_rate": 1.6490658001624696e-06, + "loss": 2.1125, + "step": 1940 + }, + { + "epoch": 2.126027397260274, + "eval_loss": 2.2668750286102295, + "eval_runtime": 48.4476, + "eval_samples_per_second": 20.641, + "eval_steps_per_second": 0.867, + "step": 1940 + }, + { + "epoch": 2.1271232876712327, + "grad_norm": 2.185532569885254, + "learning_rate": 1.647034930950447e-06, + "loss": 2.1148, + "step": 1941 + }, + { + "epoch": 2.1282191780821917, + "grad_norm": 2.1876089572906494, + "learning_rate": 1.6450040617384242e-06, + "loss": 2.0646, + "step": 1942 + }, + { + "epoch": 2.1293150684931508, + "grad_norm": 2.065310478210449, + "learning_rate": 1.6429731925264015e-06, + "loss": 2.0843, + "step": 1943 + }, + { + "epoch": 2.1304109589041094, + "grad_norm": 2.1671154499053955, + "learning_rate": 1.6409423233143788e-06, + "loss": 2.112, + "step": 1944 + }, + { + "epoch": 2.1315068493150684, + "grad_norm": 2.1573469638824463, + "learning_rate": 1.638911454102356e-06, + "loss": 2.0781, + "step": 1945 + }, + { + "epoch": 2.1326027397260274, + "grad_norm": 2.0912411212921143, + "learning_rate": 1.6368805848903333e-06, + "loss": 2.0621, + "step": 1946 + }, + { + "epoch": 2.1336986301369865, + "grad_norm": 2.1280713081359863, + "learning_rate": 1.6348497156783106e-06, + "loss": 2.1174, + "step": 1947 + }, + { + "epoch": 2.134794520547945, + "grad_norm": 2.1215052604675293, + "learning_rate": 1.6328188464662879e-06, + "loss": 2.0725, + "step": 1948 + }, + { + "epoch": 2.135890410958904, + "grad_norm": 2.0699117183685303, + "learning_rate": 1.6307879772542651e-06, + "loss": 2.0783, + "step": 1949 + }, + { + "epoch": 2.136986301369863, + "grad_norm": 2.0162057876586914, + "learning_rate": 1.628757108042242e-06, + "loss": 2.0868, + "step": 1950 + }, + { + "epoch": 2.136986301369863, + "eval_loss": 2.2651138305664062, + "eval_runtime": 53.3744, + "eval_samples_per_second": 18.736, + "eval_steps_per_second": 0.787, + "step": 1950 + }, + { + "epoch": 2.138082191780822, + "grad_norm": 2.1494126319885254, + "learning_rate": 1.6267262388302193e-06, + "loss": 2.1147, + "step": 1951 + }, + { + "epoch": 2.1391780821917807, + "grad_norm": 2.1003870964050293, + "learning_rate": 1.6246953696181965e-06, + "loss": 2.082, + "step": 1952 + }, + { + "epoch": 2.1402739726027398, + "grad_norm": 2.1687192916870117, + "learning_rate": 1.6226645004061738e-06, + "loss": 2.1084, + "step": 1953 + }, + { + "epoch": 2.141369863013699, + "grad_norm": 2.010176181793213, + "learning_rate": 1.620633631194151e-06, + "loss": 2.1568, + "step": 1954 + }, + { + "epoch": 2.1424657534246574, + "grad_norm": 2.13127064704895, + "learning_rate": 1.6186027619821284e-06, + "loss": 2.116, + "step": 1955 + }, + { + "epoch": 2.1435616438356164, + "grad_norm": 2.2229974269866943, + "learning_rate": 1.6165718927701056e-06, + "loss": 2.0596, + "step": 1956 + }, + { + "epoch": 2.1446575342465755, + "grad_norm": 2.155855894088745, + "learning_rate": 1.614541023558083e-06, + "loss": 2.1052, + "step": 1957 + }, + { + "epoch": 2.145753424657534, + "grad_norm": 2.1598658561706543, + "learning_rate": 1.6125101543460602e-06, + "loss": 2.097, + "step": 1958 + }, + { + "epoch": 2.146849315068493, + "grad_norm": 2.3428308963775635, + "learning_rate": 1.6104792851340375e-06, + "loss": 2.0991, + "step": 1959 + }, + { + "epoch": 2.147945205479452, + "grad_norm": 2.293128728866577, + "learning_rate": 1.6084484159220147e-06, + "loss": 2.1146, + "step": 1960 + }, + { + "epoch": 2.147945205479452, + "eval_loss": 2.265352964401245, + "eval_runtime": 49.6478, + "eval_samples_per_second": 20.142, + "eval_steps_per_second": 0.846, + "step": 1960 + }, + { + "epoch": 2.149041095890411, + "grad_norm": 2.1581852436065674, + "learning_rate": 1.606417546709992e-06, + "loss": 2.0901, + "step": 1961 + }, + { + "epoch": 2.1501369863013697, + "grad_norm": 2.177950382232666, + "learning_rate": 1.6043866774979693e-06, + "loss": 2.0765, + "step": 1962 + }, + { + "epoch": 2.1512328767123288, + "grad_norm": 2.176633596420288, + "learning_rate": 1.6023558082859464e-06, + "loss": 2.0917, + "step": 1963 + }, + { + "epoch": 2.152328767123288, + "grad_norm": 2.132002592086792, + "learning_rate": 1.6003249390739236e-06, + "loss": 2.0712, + "step": 1964 + }, + { + "epoch": 2.1534246575342464, + "grad_norm": 2.1170928478240967, + "learning_rate": 1.598294069861901e-06, + "loss": 2.0835, + "step": 1965 + }, + { + "epoch": 2.1545205479452054, + "grad_norm": 2.2532341480255127, + "learning_rate": 1.5962632006498782e-06, + "loss": 2.0638, + "step": 1966 + }, + { + "epoch": 2.1556164383561645, + "grad_norm": 2.1742208003997803, + "learning_rate": 1.5942323314378555e-06, + "loss": 2.1164, + "step": 1967 + }, + { + "epoch": 2.1567123287671235, + "grad_norm": 2.1353843212127686, + "learning_rate": 1.5922014622258327e-06, + "loss": 2.0447, + "step": 1968 + }, + { + "epoch": 2.157808219178082, + "grad_norm": 2.1103644371032715, + "learning_rate": 1.59017059301381e-06, + "loss": 2.0631, + "step": 1969 + }, + { + "epoch": 2.158904109589041, + "grad_norm": 2.129606246948242, + "learning_rate": 1.5881397238017873e-06, + "loss": 2.1184, + "step": 1970 + }, + { + "epoch": 2.158904109589041, + "eval_loss": 2.26619553565979, + "eval_runtime": 52.3189, + "eval_samples_per_second": 19.114, + "eval_steps_per_second": 0.803, + "step": 1970 + }, + { + "epoch": 2.16, + "grad_norm": 2.2155823707580566, + "learning_rate": 1.5861088545897646e-06, + "loss": 2.0885, + "step": 1971 + }, + { + "epoch": 2.1610958904109587, + "grad_norm": 2.1816463470458984, + "learning_rate": 1.5840779853777418e-06, + "loss": 2.0517, + "step": 1972 + }, + { + "epoch": 2.1621917808219178, + "grad_norm": 2.2179040908813477, + "learning_rate": 1.5820471161657191e-06, + "loss": 2.0884, + "step": 1973 + }, + { + "epoch": 2.163287671232877, + "grad_norm": 2.3558337688446045, + "learning_rate": 1.5800162469536964e-06, + "loss": 2.0682, + "step": 1974 + }, + { + "epoch": 2.1643835616438354, + "grad_norm": 2.406665086746216, + "learning_rate": 1.5779853777416735e-06, + "loss": 2.1172, + "step": 1975 + }, + { + "epoch": 2.1654794520547944, + "grad_norm": 2.3815410137176514, + "learning_rate": 1.5759545085296507e-06, + "loss": 2.115, + "step": 1976 + }, + { + "epoch": 2.1665753424657535, + "grad_norm": 2.2329678535461426, + "learning_rate": 1.573923639317628e-06, + "loss": 2.087, + "step": 1977 + }, + { + "epoch": 2.1676712328767125, + "grad_norm": 2.331451177597046, + "learning_rate": 1.5718927701056053e-06, + "loss": 2.0996, + "step": 1978 + }, + { + "epoch": 2.168767123287671, + "grad_norm": 2.158210039138794, + "learning_rate": 1.5698619008935826e-06, + "loss": 2.0905, + "step": 1979 + }, + { + "epoch": 2.16986301369863, + "grad_norm": 2.1256191730499268, + "learning_rate": 1.5678310316815598e-06, + "loss": 2.1099, + "step": 1980 + }, + { + "epoch": 2.16986301369863, + "eval_loss": 2.2658803462982178, + "eval_runtime": 51.5278, + "eval_samples_per_second": 19.407, + "eval_steps_per_second": 0.815, + "step": 1980 + }, + { + "epoch": 2.170958904109589, + "grad_norm": 2.2626984119415283, + "learning_rate": 1.5658001624695371e-06, + "loss": 2.0651, + "step": 1981 + }, + { + "epoch": 2.172054794520548, + "grad_norm": 2.132742404937744, + "learning_rate": 1.5637692932575144e-06, + "loss": 2.1246, + "step": 1982 + }, + { + "epoch": 2.1731506849315068, + "grad_norm": 2.0621824264526367, + "learning_rate": 1.5617384240454917e-06, + "loss": 2.0696, + "step": 1983 + }, + { + "epoch": 2.174246575342466, + "grad_norm": 2.0715548992156982, + "learning_rate": 1.559707554833469e-06, + "loss": 2.0894, + "step": 1984 + }, + { + "epoch": 2.175342465753425, + "grad_norm": 2.1211655139923096, + "learning_rate": 1.5576766856214462e-06, + "loss": 2.1122, + "step": 1985 + }, + { + "epoch": 2.1764383561643834, + "grad_norm": 2.14603328704834, + "learning_rate": 1.5556458164094235e-06, + "loss": 2.0239, + "step": 1986 + }, + { + "epoch": 2.1775342465753424, + "grad_norm": 2.092494249343872, + "learning_rate": 1.5536149471974008e-06, + "loss": 2.0795, + "step": 1987 + }, + { + "epoch": 2.1786301369863015, + "grad_norm": 2.14874267578125, + "learning_rate": 1.5515840779853778e-06, + "loss": 2.0476, + "step": 1988 + }, + { + "epoch": 2.17972602739726, + "grad_norm": 2.024716854095459, + "learning_rate": 1.549553208773355e-06, + "loss": 2.0904, + "step": 1989 + }, + { + "epoch": 2.180821917808219, + "grad_norm": 2.1700267791748047, + "learning_rate": 1.5475223395613324e-06, + "loss": 2.0806, + "step": 1990 + }, + { + "epoch": 2.180821917808219, + "eval_loss": 2.2657630443573, + "eval_runtime": 49.7954, + "eval_samples_per_second": 20.082, + "eval_steps_per_second": 0.843, + "step": 1990 + }, + { + "epoch": 2.181917808219178, + "grad_norm": 2.1046273708343506, + "learning_rate": 1.5454914703493097e-06, + "loss": 2.0738, + "step": 1991 + }, + { + "epoch": 2.183013698630137, + "grad_norm": 2.2489562034606934, + "learning_rate": 1.543460601137287e-06, + "loss": 2.1468, + "step": 1992 + }, + { + "epoch": 2.1841095890410958, + "grad_norm": 2.176658868789673, + "learning_rate": 1.5414297319252642e-06, + "loss": 2.0977, + "step": 1993 + }, + { + "epoch": 2.185205479452055, + "grad_norm": 2.2471182346343994, + "learning_rate": 1.5393988627132415e-06, + "loss": 2.1023, + "step": 1994 + }, + { + "epoch": 2.186301369863014, + "grad_norm": 2.306699514389038, + "learning_rate": 1.5373679935012188e-06, + "loss": 2.084, + "step": 1995 + }, + { + "epoch": 2.1873972602739724, + "grad_norm": 2.2053849697113037, + "learning_rate": 1.535337124289196e-06, + "loss": 2.0419, + "step": 1996 + }, + { + "epoch": 2.1884931506849314, + "grad_norm": 2.2459211349487305, + "learning_rate": 1.5333062550771733e-06, + "loss": 2.0882, + "step": 1997 + }, + { + "epoch": 2.1895890410958905, + "grad_norm": 2.574877977371216, + "learning_rate": 1.5312753858651506e-06, + "loss": 2.1627, + "step": 1998 + }, + { + "epoch": 2.1906849315068495, + "grad_norm": 2.111438274383545, + "learning_rate": 1.5292445166531279e-06, + "loss": 2.0762, + "step": 1999 + }, + { + "epoch": 2.191780821917808, + "grad_norm": 2.1831490993499756, + "learning_rate": 1.5272136474411047e-06, + "loss": 2.0954, + "step": 2000 + }, + { + "epoch": 2.191780821917808, + "eval_loss": 2.2667267322540283, + "eval_runtime": 46.6703, + "eval_samples_per_second": 21.427, + "eval_steps_per_second": 0.9, + "step": 2000 + }, + { + "epoch": 2.192876712328767, + "grad_norm": 2.161336660385132, + "learning_rate": 1.525182778229082e-06, + "loss": 2.1393, + "step": 2001 + }, + { + "epoch": 2.193972602739726, + "grad_norm": 2.210535764694214, + "learning_rate": 1.5231519090170593e-06, + "loss": 2.0925, + "step": 2002 + }, + { + "epoch": 2.1950684931506848, + "grad_norm": 2.123384714126587, + "learning_rate": 1.5211210398050365e-06, + "loss": 2.0623, + "step": 2003 + }, + { + "epoch": 2.196164383561644, + "grad_norm": 2.147090196609497, + "learning_rate": 1.5190901705930138e-06, + "loss": 2.0839, + "step": 2004 + }, + { + "epoch": 2.197260273972603, + "grad_norm": 2.261474609375, + "learning_rate": 1.517059301380991e-06, + "loss": 2.0987, + "step": 2005 + }, + { + "epoch": 2.198356164383562, + "grad_norm": 2.295656204223633, + "learning_rate": 1.5150284321689684e-06, + "loss": 2.0786, + "step": 2006 + }, + { + "epoch": 2.1994520547945204, + "grad_norm": 2.055387020111084, + "learning_rate": 1.5129975629569456e-06, + "loss": 2.0781, + "step": 2007 + }, + { + "epoch": 2.2005479452054795, + "grad_norm": 2.1994800567626953, + "learning_rate": 1.510966693744923e-06, + "loss": 2.088, + "step": 2008 + }, + { + "epoch": 2.2016438356164385, + "grad_norm": 2.3267500400543213, + "learning_rate": 1.5089358245329002e-06, + "loss": 2.0945, + "step": 2009 + }, + { + "epoch": 2.202739726027397, + "grad_norm": 2.1972532272338867, + "learning_rate": 1.5069049553208775e-06, + "loss": 2.0822, + "step": 2010 + }, + { + "epoch": 2.202739726027397, + "eval_loss": 2.26534104347229, + "eval_runtime": 41.6482, + "eval_samples_per_second": 24.011, + "eval_steps_per_second": 1.008, + "step": 2010 + }, + { + "epoch": 2.203835616438356, + "grad_norm": 2.161071538925171, + "learning_rate": 1.5048740861088547e-06, + "loss": 2.0908, + "step": 2011 + }, + { + "epoch": 2.204931506849315, + "grad_norm": 2.1437389850616455, + "learning_rate": 1.502843216896832e-06, + "loss": 2.1041, + "step": 2012 + }, + { + "epoch": 2.206027397260274, + "grad_norm": 2.16727614402771, + "learning_rate": 1.500812347684809e-06, + "loss": 2.0209, + "step": 2013 + }, + { + "epoch": 2.207123287671233, + "grad_norm": 2.1556594371795654, + "learning_rate": 1.4987814784727864e-06, + "loss": 2.09, + "step": 2014 + }, + { + "epoch": 2.208219178082192, + "grad_norm": 2.0891411304473877, + "learning_rate": 1.4967506092607636e-06, + "loss": 2.0864, + "step": 2015 + }, + { + "epoch": 2.209315068493151, + "grad_norm": 2.031249523162842, + "learning_rate": 1.494719740048741e-06, + "loss": 2.0363, + "step": 2016 + }, + { + "epoch": 2.2104109589041094, + "grad_norm": 2.2452943325042725, + "learning_rate": 1.4926888708367182e-06, + "loss": 2.114, + "step": 2017 + }, + { + "epoch": 2.2115068493150685, + "grad_norm": 2.1752209663391113, + "learning_rate": 1.4906580016246955e-06, + "loss": 2.1123, + "step": 2018 + }, + { + "epoch": 2.2126027397260275, + "grad_norm": 2.2148444652557373, + "learning_rate": 1.4886271324126727e-06, + "loss": 2.1132, + "step": 2019 + }, + { + "epoch": 2.213698630136986, + "grad_norm": 2.1530041694641113, + "learning_rate": 1.48659626320065e-06, + "loss": 2.1278, + "step": 2020 + }, + { + "epoch": 2.213698630136986, + "eval_loss": 2.264028310775757, + "eval_runtime": 43.6346, + "eval_samples_per_second": 22.918, + "eval_steps_per_second": 0.963, + "step": 2020 + }, + { + "epoch": 2.214794520547945, + "grad_norm": 2.223219871520996, + "learning_rate": 1.4845653939886273e-06, + "loss": 2.0293, + "step": 2021 + }, + { + "epoch": 2.215890410958904, + "grad_norm": 2.122913360595703, + "learning_rate": 1.4825345247766046e-06, + "loss": 2.0411, + "step": 2022 + }, + { + "epoch": 2.216986301369863, + "grad_norm": 2.1176650524139404, + "learning_rate": 1.4805036555645818e-06, + "loss": 2.0326, + "step": 2023 + }, + { + "epoch": 2.2180821917808218, + "grad_norm": 2.1717002391815186, + "learning_rate": 1.4784727863525591e-06, + "loss": 2.0835, + "step": 2024 + }, + { + "epoch": 2.219178082191781, + "grad_norm": 2.1869051456451416, + "learning_rate": 1.4764419171405362e-06, + "loss": 2.0571, + "step": 2025 + }, + { + "epoch": 2.22027397260274, + "grad_norm": 2.2699813842773438, + "learning_rate": 1.4744110479285135e-06, + "loss": 2.1156, + "step": 2026 + }, + { + "epoch": 2.2213698630136984, + "grad_norm": 2.243837594985962, + "learning_rate": 1.4723801787164907e-06, + "loss": 2.1231, + "step": 2027 + }, + { + "epoch": 2.2224657534246575, + "grad_norm": 2.1143088340759277, + "learning_rate": 1.470349309504468e-06, + "loss": 2.11, + "step": 2028 + }, + { + "epoch": 2.2235616438356165, + "grad_norm": 2.0839781761169434, + "learning_rate": 1.4683184402924453e-06, + "loss": 2.0971, + "step": 2029 + }, + { + "epoch": 2.2246575342465755, + "grad_norm": 2.182480812072754, + "learning_rate": 1.4662875710804226e-06, + "loss": 2.1094, + "step": 2030 + }, + { + "epoch": 2.2246575342465755, + "eval_loss": 2.2636313438415527, + "eval_runtime": 43.184, + "eval_samples_per_second": 23.157, + "eval_steps_per_second": 0.973, + "step": 2030 + }, + { + "epoch": 2.225753424657534, + "grad_norm": 2.2751855850219727, + "learning_rate": 1.4642567018683998e-06, + "loss": 2.1533, + "step": 2031 + }, + { + "epoch": 2.226849315068493, + "grad_norm": 2.049236536026001, + "learning_rate": 1.4622258326563771e-06, + "loss": 2.0847, + "step": 2032 + }, + { + "epoch": 2.227945205479452, + "grad_norm": 2.0883119106292725, + "learning_rate": 1.4601949634443544e-06, + "loss": 2.0687, + "step": 2033 + }, + { + "epoch": 2.2290410958904108, + "grad_norm": 2.152400016784668, + "learning_rate": 1.4581640942323317e-06, + "loss": 2.0979, + "step": 2034 + }, + { + "epoch": 2.23013698630137, + "grad_norm": 2.221442699432373, + "learning_rate": 1.456133225020309e-06, + "loss": 2.1601, + "step": 2035 + }, + { + "epoch": 2.231232876712329, + "grad_norm": 2.1799275875091553, + "learning_rate": 1.4541023558082862e-06, + "loss": 2.1133, + "step": 2036 + }, + { + "epoch": 2.232328767123288, + "grad_norm": 2.182882070541382, + "learning_rate": 1.4520714865962635e-06, + "loss": 2.1038, + "step": 2037 + }, + { + "epoch": 2.2334246575342465, + "grad_norm": 2.1438207626342773, + "learning_rate": 1.4500406173842406e-06, + "loss": 2.1176, + "step": 2038 + }, + { + "epoch": 2.2345205479452055, + "grad_norm": 2.1394078731536865, + "learning_rate": 1.4480097481722178e-06, + "loss": 2.0928, + "step": 2039 + }, + { + "epoch": 2.2356164383561645, + "grad_norm": 2.2296125888824463, + "learning_rate": 1.4459788789601951e-06, + "loss": 2.0938, + "step": 2040 + }, + { + "epoch": 2.2356164383561645, + "eval_loss": 2.2618958950042725, + "eval_runtime": 43.253, + "eval_samples_per_second": 23.12, + "eval_steps_per_second": 0.971, + "step": 2040 + }, + { + "epoch": 2.236712328767123, + "grad_norm": 2.248995542526245, + "learning_rate": 1.4439480097481724e-06, + "loss": 2.0666, + "step": 2041 + }, + { + "epoch": 2.237808219178082, + "grad_norm": 2.0892133712768555, + "learning_rate": 1.4419171405361497e-06, + "loss": 2.0437, + "step": 2042 + }, + { + "epoch": 2.238904109589041, + "grad_norm": 2.191180944442749, + "learning_rate": 1.439886271324127e-06, + "loss": 2.1038, + "step": 2043 + }, + { + "epoch": 2.24, + "grad_norm": 2.1049792766571045, + "learning_rate": 1.4378554021121042e-06, + "loss": 2.096, + "step": 2044 + }, + { + "epoch": 2.241095890410959, + "grad_norm": 2.166987419128418, + "learning_rate": 1.4358245329000815e-06, + "loss": 2.0704, + "step": 2045 + }, + { + "epoch": 2.242191780821918, + "grad_norm": 2.3071136474609375, + "learning_rate": 1.4337936636880588e-06, + "loss": 2.1118, + "step": 2046 + }, + { + "epoch": 2.243287671232877, + "grad_norm": 2.185621976852417, + "learning_rate": 1.431762794476036e-06, + "loss": 2.1032, + "step": 2047 + }, + { + "epoch": 2.2443835616438355, + "grad_norm": 2.2100017070770264, + "learning_rate": 1.4297319252640133e-06, + "loss": 2.0542, + "step": 2048 + }, + { + "epoch": 2.2454794520547945, + "grad_norm": 2.2313308715820312, + "learning_rate": 1.4277010560519906e-06, + "loss": 2.0812, + "step": 2049 + }, + { + "epoch": 2.2465753424657535, + "grad_norm": 2.1170578002929688, + "learning_rate": 1.4256701868399674e-06, + "loss": 2.0542, + "step": 2050 + }, + { + "epoch": 2.2465753424657535, + "eval_loss": 2.2627694606781006, + "eval_runtime": 44.8509, + "eval_samples_per_second": 22.296, + "eval_steps_per_second": 0.936, + "step": 2050 + }, + { + "epoch": 2.247671232876712, + "grad_norm": 2.207075357437134, + "learning_rate": 1.4236393176279447e-06, + "loss": 2.1467, + "step": 2051 + }, + { + "epoch": 2.248767123287671, + "grad_norm": 2.1787123680114746, + "learning_rate": 1.421608448415922e-06, + "loss": 2.0824, + "step": 2052 + }, + { + "epoch": 2.24986301369863, + "grad_norm": 2.179396867752075, + "learning_rate": 1.4195775792038993e-06, + "loss": 2.1394, + "step": 2053 + }, + { + "epoch": 2.250958904109589, + "grad_norm": 2.1963279247283936, + "learning_rate": 1.4175467099918765e-06, + "loss": 2.0594, + "step": 2054 + }, + { + "epoch": 2.252054794520548, + "grad_norm": 2.165109872817993, + "learning_rate": 1.4155158407798538e-06, + "loss": 2.0812, + "step": 2055 + }, + { + "epoch": 2.253150684931507, + "grad_norm": 2.033921957015991, + "learning_rate": 1.413484971567831e-06, + "loss": 2.1124, + "step": 2056 + }, + { + "epoch": 2.254246575342466, + "grad_norm": 2.12203311920166, + "learning_rate": 1.4114541023558084e-06, + "loss": 2.0361, + "step": 2057 + }, + { + "epoch": 2.255342465753425, + "grad_norm": 2.268758535385132, + "learning_rate": 1.4094232331437856e-06, + "loss": 2.0766, + "step": 2058 + }, + { + "epoch": 2.2564383561643835, + "grad_norm": 2.1977181434631348, + "learning_rate": 1.407392363931763e-06, + "loss": 2.124, + "step": 2059 + }, + { + "epoch": 2.2575342465753425, + "grad_norm": 2.100123882293701, + "learning_rate": 1.4053614947197402e-06, + "loss": 2.1554, + "step": 2060 + }, + { + "epoch": 2.2575342465753425, + "eval_loss": 2.263918399810791, + "eval_runtime": 44.5109, + "eval_samples_per_second": 22.466, + "eval_steps_per_second": 0.944, + "step": 2060 + }, + { + "epoch": 2.2586301369863016, + "grad_norm": 2.1824963092803955, + "learning_rate": 1.4033306255077175e-06, + "loss": 2.0998, + "step": 2061 + }, + { + "epoch": 2.25972602739726, + "grad_norm": 2.211451530456543, + "learning_rate": 1.4012997562956945e-06, + "loss": 2.0266, + "step": 2062 + }, + { + "epoch": 2.260821917808219, + "grad_norm": 2.2530057430267334, + "learning_rate": 1.3992688870836718e-06, + "loss": 2.1366, + "step": 2063 + }, + { + "epoch": 2.261917808219178, + "grad_norm": 2.121558904647827, + "learning_rate": 1.397238017871649e-06, + "loss": 2.0927, + "step": 2064 + }, + { + "epoch": 2.263013698630137, + "grad_norm": 2.1322062015533447, + "learning_rate": 1.3952071486596264e-06, + "loss": 2.0929, + "step": 2065 + }, + { + "epoch": 2.264109589041096, + "grad_norm": 2.1645631790161133, + "learning_rate": 1.3931762794476036e-06, + "loss": 2.057, + "step": 2066 + }, + { + "epoch": 2.265205479452055, + "grad_norm": 2.1870594024658203, + "learning_rate": 1.391145410235581e-06, + "loss": 2.1274, + "step": 2067 + }, + { + "epoch": 2.266301369863014, + "grad_norm": 2.1355020999908447, + "learning_rate": 1.3891145410235582e-06, + "loss": 2.0784, + "step": 2068 + }, + { + "epoch": 2.2673972602739725, + "grad_norm": 2.138497829437256, + "learning_rate": 1.3870836718115355e-06, + "loss": 2.0265, + "step": 2069 + }, + { + "epoch": 2.2684931506849315, + "grad_norm": 2.3041086196899414, + "learning_rate": 1.3850528025995127e-06, + "loss": 2.0716, + "step": 2070 + }, + { + "epoch": 2.2684931506849315, + "eval_loss": 2.263742685317993, + "eval_runtime": 45.8378, + "eval_samples_per_second": 21.816, + "eval_steps_per_second": 0.916, + "step": 2070 + }, + { + "epoch": 2.2695890410958905, + "grad_norm": 2.2722716331481934, + "learning_rate": 1.38302193338749e-06, + "loss": 2.0744, + "step": 2071 + }, + { + "epoch": 2.270684931506849, + "grad_norm": 2.267909288406372, + "learning_rate": 1.3809910641754673e-06, + "loss": 2.0269, + "step": 2072 + }, + { + "epoch": 2.271780821917808, + "grad_norm": 2.271430015563965, + "learning_rate": 1.3789601949634446e-06, + "loss": 2.1165, + "step": 2073 + }, + { + "epoch": 2.272876712328767, + "grad_norm": 2.3091483116149902, + "learning_rate": 1.3769293257514218e-06, + "loss": 2.0336, + "step": 2074 + }, + { + "epoch": 2.2739726027397262, + "grad_norm": 2.2311320304870605, + "learning_rate": 1.374898456539399e-06, + "loss": 2.0964, + "step": 2075 + }, + { + "epoch": 2.275068493150685, + "grad_norm": 2.2187438011169434, + "learning_rate": 1.3728675873273762e-06, + "loss": 2.0672, + "step": 2076 + }, + { + "epoch": 2.276164383561644, + "grad_norm": 2.0933640003204346, + "learning_rate": 1.3708367181153535e-06, + "loss": 2.1113, + "step": 2077 + }, + { + "epoch": 2.277260273972603, + "grad_norm": 2.2611358165740967, + "learning_rate": 1.3688058489033307e-06, + "loss": 2.101, + "step": 2078 + }, + { + "epoch": 2.2783561643835615, + "grad_norm": 2.269864797592163, + "learning_rate": 1.366774979691308e-06, + "loss": 2.0993, + "step": 2079 + }, + { + "epoch": 2.2794520547945205, + "grad_norm": 2.242854356765747, + "learning_rate": 1.3647441104792853e-06, + "loss": 2.107, + "step": 2080 + }, + { + "epoch": 2.2794520547945205, + "eval_loss": 2.2640912532806396, + "eval_runtime": 46.7194, + "eval_samples_per_second": 21.404, + "eval_steps_per_second": 0.899, + "step": 2080 + }, + { + "epoch": 2.2805479452054795, + "grad_norm": 2.2348811626434326, + "learning_rate": 1.3627132412672626e-06, + "loss": 2.0941, + "step": 2081 + }, + { + "epoch": 2.281643835616438, + "grad_norm": 2.214179277420044, + "learning_rate": 1.3606823720552398e-06, + "loss": 2.0548, + "step": 2082 + }, + { + "epoch": 2.282739726027397, + "grad_norm": 2.216902256011963, + "learning_rate": 1.3586515028432171e-06, + "loss": 2.0955, + "step": 2083 + }, + { + "epoch": 2.283835616438356, + "grad_norm": 2.1791675090789795, + "learning_rate": 1.3566206336311944e-06, + "loss": 2.0395, + "step": 2084 + }, + { + "epoch": 2.2849315068493152, + "grad_norm": 2.135376214981079, + "learning_rate": 1.3545897644191717e-06, + "loss": 2.0837, + "step": 2085 + }, + { + "epoch": 2.286027397260274, + "grad_norm": 2.244448184967041, + "learning_rate": 1.352558895207149e-06, + "loss": 2.0805, + "step": 2086 + }, + { + "epoch": 2.287123287671233, + "grad_norm": 2.28576922416687, + "learning_rate": 1.3505280259951258e-06, + "loss": 2.1014, + "step": 2087 + }, + { + "epoch": 2.288219178082192, + "grad_norm": 2.247220516204834, + "learning_rate": 1.348497156783103e-06, + "loss": 2.0773, + "step": 2088 + }, + { + "epoch": 2.289315068493151, + "grad_norm": 2.107109308242798, + "learning_rate": 1.3464662875710804e-06, + "loss": 2.1236, + "step": 2089 + }, + { + "epoch": 2.2904109589041095, + "grad_norm": 2.2805957794189453, + "learning_rate": 1.3444354183590576e-06, + "loss": 2.0516, + "step": 2090 + }, + { + "epoch": 2.2904109589041095, + "eval_loss": 2.262258291244507, + "eval_runtime": 46.3329, + "eval_samples_per_second": 21.583, + "eval_steps_per_second": 0.906, + "step": 2090 + }, + { + "epoch": 2.2915068493150685, + "grad_norm": 2.2838428020477295, + "learning_rate": 1.342404549147035e-06, + "loss": 2.1008, + "step": 2091 + }, + { + "epoch": 2.2926027397260276, + "grad_norm": 2.1949639320373535, + "learning_rate": 1.3403736799350122e-06, + "loss": 2.0634, + "step": 2092 + }, + { + "epoch": 2.293698630136986, + "grad_norm": 2.144423246383667, + "learning_rate": 1.3383428107229895e-06, + "loss": 2.0996, + "step": 2093 + }, + { + "epoch": 2.294794520547945, + "grad_norm": 2.293369770050049, + "learning_rate": 1.3363119415109667e-06, + "loss": 2.1206, + "step": 2094 + }, + { + "epoch": 2.2958904109589042, + "grad_norm": 2.2426724433898926, + "learning_rate": 1.334281072298944e-06, + "loss": 2.1026, + "step": 2095 + }, + { + "epoch": 2.296986301369863, + "grad_norm": 2.159067392349243, + "learning_rate": 1.3322502030869213e-06, + "loss": 2.0993, + "step": 2096 + }, + { + "epoch": 2.298082191780822, + "grad_norm": 2.2819197177886963, + "learning_rate": 1.3302193338748986e-06, + "loss": 2.0776, + "step": 2097 + }, + { + "epoch": 2.299178082191781, + "grad_norm": 2.253448963165283, + "learning_rate": 1.3281884646628758e-06, + "loss": 2.1211, + "step": 2098 + }, + { + "epoch": 2.30027397260274, + "grad_norm": 2.2652339935302734, + "learning_rate": 1.3261575954508531e-06, + "loss": 2.1175, + "step": 2099 + }, + { + "epoch": 2.3013698630136985, + "grad_norm": 2.1319172382354736, + "learning_rate": 1.3241267262388302e-06, + "loss": 2.1137, + "step": 2100 + }, + { + "epoch": 2.3013698630136985, + "eval_loss": 2.263416051864624, + "eval_runtime": 47.0463, + "eval_samples_per_second": 21.256, + "eval_steps_per_second": 0.893, + "step": 2100 + }, + { + "epoch": 2.3024657534246575, + "grad_norm": 2.1784651279449463, + "learning_rate": 1.3220958570268074e-06, + "loss": 2.0951, + "step": 2101 + }, + { + "epoch": 2.3035616438356166, + "grad_norm": 2.1439764499664307, + "learning_rate": 1.3200649878147847e-06, + "loss": 2.0483, + "step": 2102 + }, + { + "epoch": 2.304657534246575, + "grad_norm": 2.1723854541778564, + "learning_rate": 1.318034118602762e-06, + "loss": 2.0851, + "step": 2103 + }, + { + "epoch": 2.305753424657534, + "grad_norm": 2.098573923110962, + "learning_rate": 1.3160032493907393e-06, + "loss": 2.1204, + "step": 2104 + }, + { + "epoch": 2.3068493150684932, + "grad_norm": 2.1380534172058105, + "learning_rate": 1.3139723801787166e-06, + "loss": 2.1094, + "step": 2105 + }, + { + "epoch": 2.3079452054794523, + "grad_norm": 2.1696887016296387, + "learning_rate": 1.3119415109666938e-06, + "loss": 2.1411, + "step": 2106 + }, + { + "epoch": 2.309041095890411, + "grad_norm": 2.1761717796325684, + "learning_rate": 1.309910641754671e-06, + "loss": 2.0893, + "step": 2107 + }, + { + "epoch": 2.31013698630137, + "grad_norm": 2.2763733863830566, + "learning_rate": 1.3078797725426484e-06, + "loss": 2.0441, + "step": 2108 + }, + { + "epoch": 2.311232876712329, + "grad_norm": 2.1061692237854004, + "learning_rate": 1.3058489033306257e-06, + "loss": 2.0652, + "step": 2109 + }, + { + "epoch": 2.3123287671232875, + "grad_norm": 2.242868423461914, + "learning_rate": 1.303818034118603e-06, + "loss": 2.0636, + "step": 2110 + }, + { + "epoch": 2.3123287671232875, + "eval_loss": 2.2643768787384033, + "eval_runtime": 43.9201, + "eval_samples_per_second": 22.769, + "eval_steps_per_second": 0.956, + "step": 2110 + }, + { + "epoch": 2.3134246575342465, + "grad_norm": 2.4198052883148193, + "learning_rate": 1.3017871649065802e-06, + "loss": 2.0939, + "step": 2111 + }, + { + "epoch": 2.3145205479452056, + "grad_norm": 2.3702943325042725, + "learning_rate": 1.2997562956945573e-06, + "loss": 2.0663, + "step": 2112 + }, + { + "epoch": 2.315616438356164, + "grad_norm": 2.1991565227508545, + "learning_rate": 1.2977254264825345e-06, + "loss": 2.0594, + "step": 2113 + }, + { + "epoch": 2.316712328767123, + "grad_norm": 2.255929946899414, + "learning_rate": 1.2956945572705118e-06, + "loss": 2.0973, + "step": 2114 + }, + { + "epoch": 2.317808219178082, + "grad_norm": 2.307612419128418, + "learning_rate": 1.293663688058489e-06, + "loss": 2.1085, + "step": 2115 + }, + { + "epoch": 2.3189041095890413, + "grad_norm": 2.237325429916382, + "learning_rate": 1.2916328188464664e-06, + "loss": 2.0485, + "step": 2116 + }, + { + "epoch": 2.32, + "grad_norm": 2.3728206157684326, + "learning_rate": 1.2896019496344436e-06, + "loss": 2.0086, + "step": 2117 + }, + { + "epoch": 2.321095890410959, + "grad_norm": 2.138505697250366, + "learning_rate": 1.287571080422421e-06, + "loss": 2.0628, + "step": 2118 + }, + { + "epoch": 2.322191780821918, + "grad_norm": 2.1808021068573, + "learning_rate": 1.2855402112103982e-06, + "loss": 2.1146, + "step": 2119 + }, + { + "epoch": 2.323287671232877, + "grad_norm": 2.1783456802368164, + "learning_rate": 1.2835093419983755e-06, + "loss": 2.0589, + "step": 2120 + }, + { + "epoch": 2.323287671232877, + "eval_loss": 2.262885332107544, + "eval_runtime": 48.7798, + "eval_samples_per_second": 20.5, + "eval_steps_per_second": 0.861, + "step": 2120 + }, + { + "epoch": 2.3243835616438355, + "grad_norm": 2.3852620124816895, + "learning_rate": 1.2814784727863528e-06, + "loss": 2.0862, + "step": 2121 + }, + { + "epoch": 2.3254794520547946, + "grad_norm": 2.223008155822754, + "learning_rate": 1.27944760357433e-06, + "loss": 2.072, + "step": 2122 + }, + { + "epoch": 2.3265753424657536, + "grad_norm": 2.1848905086517334, + "learning_rate": 1.2774167343623073e-06, + "loss": 2.1234, + "step": 2123 + }, + { + "epoch": 2.327671232876712, + "grad_norm": 2.1176791191101074, + "learning_rate": 1.2753858651502846e-06, + "loss": 2.1188, + "step": 2124 + }, + { + "epoch": 2.328767123287671, + "grad_norm": 2.2725491523742676, + "learning_rate": 1.2733549959382616e-06, + "loss": 2.0892, + "step": 2125 + }, + { + "epoch": 2.3298630136986302, + "grad_norm": 2.218189001083374, + "learning_rate": 1.271324126726239e-06, + "loss": 2.0827, + "step": 2126 + }, + { + "epoch": 2.330958904109589, + "grad_norm": 2.2110207080841064, + "learning_rate": 1.2692932575142162e-06, + "loss": 2.0857, + "step": 2127 + }, + { + "epoch": 2.332054794520548, + "grad_norm": 2.175096273422241, + "learning_rate": 1.2672623883021935e-06, + "loss": 2.0461, + "step": 2128 + }, + { + "epoch": 2.333150684931507, + "grad_norm": 2.1493241786956787, + "learning_rate": 1.2652315190901707e-06, + "loss": 2.0361, + "step": 2129 + }, + { + "epoch": 2.334246575342466, + "grad_norm": 2.276071310043335, + "learning_rate": 1.263200649878148e-06, + "loss": 2.1336, + "step": 2130 + }, + { + "epoch": 2.334246575342466, + "eval_loss": 2.262963056564331, + "eval_runtime": 49.2242, + "eval_samples_per_second": 20.315, + "eval_steps_per_second": 0.853, + "step": 2130 + }, + { + "epoch": 2.3353424657534245, + "grad_norm": 2.1708502769470215, + "learning_rate": 1.2611697806661253e-06, + "loss": 2.081, + "step": 2131 + }, + { + "epoch": 2.3364383561643836, + "grad_norm": 2.219454526901245, + "learning_rate": 1.2591389114541026e-06, + "loss": 2.0941, + "step": 2132 + }, + { + "epoch": 2.3375342465753426, + "grad_norm": 2.110680103302002, + "learning_rate": 1.2571080422420798e-06, + "loss": 2.0692, + "step": 2133 + }, + { + "epoch": 2.338630136986301, + "grad_norm": 2.186999559402466, + "learning_rate": 1.2550771730300571e-06, + "loss": 2.0652, + "step": 2134 + }, + { + "epoch": 2.33972602739726, + "grad_norm": 2.182494878768921, + "learning_rate": 1.2530463038180344e-06, + "loss": 2.1238, + "step": 2135 + }, + { + "epoch": 2.3408219178082192, + "grad_norm": 2.163588047027588, + "learning_rate": 1.2510154346060117e-06, + "loss": 2.0788, + "step": 2136 + }, + { + "epoch": 2.3419178082191783, + "grad_norm": 2.1190285682678223, + "learning_rate": 1.2489845653939887e-06, + "loss": 2.0861, + "step": 2137 + }, + { + "epoch": 2.343013698630137, + "grad_norm": 2.187049388885498, + "learning_rate": 1.246953696181966e-06, + "loss": 2.1098, + "step": 2138 + }, + { + "epoch": 2.344109589041096, + "grad_norm": 2.3307504653930664, + "learning_rate": 1.2449228269699433e-06, + "loss": 2.1443, + "step": 2139 + }, + { + "epoch": 2.345205479452055, + "grad_norm": 2.149183511734009, + "learning_rate": 1.2428919577579204e-06, + "loss": 2.0855, + "step": 2140 + }, + { + "epoch": 2.345205479452055, + "eval_loss": 2.263383388519287, + "eval_runtime": 49.7579, + "eval_samples_per_second": 20.097, + "eval_steps_per_second": 0.844, + "step": 2140 + }, + { + "epoch": 2.3463013698630135, + "grad_norm": 2.190133571624756, + "learning_rate": 1.2408610885458976e-06, + "loss": 2.059, + "step": 2141 + }, + { + "epoch": 2.3473972602739726, + "grad_norm": 2.1775476932525635, + "learning_rate": 1.238830219333875e-06, + "loss": 2.0795, + "step": 2142 + }, + { + "epoch": 2.3484931506849316, + "grad_norm": 2.148383617401123, + "learning_rate": 1.2367993501218522e-06, + "loss": 2.0614, + "step": 2143 + }, + { + "epoch": 2.34958904109589, + "grad_norm": 2.184270143508911, + "learning_rate": 1.2347684809098295e-06, + "loss": 2.1111, + "step": 2144 + }, + { + "epoch": 2.350684931506849, + "grad_norm": 2.136432647705078, + "learning_rate": 1.2327376116978067e-06, + "loss": 2.1162, + "step": 2145 + }, + { + "epoch": 2.3517808219178082, + "grad_norm": 2.1389224529266357, + "learning_rate": 1.230706742485784e-06, + "loss": 2.0369, + "step": 2146 + }, + { + "epoch": 2.3528767123287673, + "grad_norm": 2.184678316116333, + "learning_rate": 1.2286758732737613e-06, + "loss": 2.0485, + "step": 2147 + }, + { + "epoch": 2.353972602739726, + "grad_norm": 2.1494438648223877, + "learning_rate": 1.2266450040617386e-06, + "loss": 2.1399, + "step": 2148 + }, + { + "epoch": 2.355068493150685, + "grad_norm": 2.197528839111328, + "learning_rate": 1.2246141348497158e-06, + "loss": 2.1553, + "step": 2149 + }, + { + "epoch": 2.356164383561644, + "grad_norm": 2.222090005874634, + "learning_rate": 1.2225832656376931e-06, + "loss": 2.0896, + "step": 2150 + }, + { + "epoch": 2.356164383561644, + "eval_loss": 2.261875867843628, + "eval_runtime": 50.0806, + "eval_samples_per_second": 19.968, + "eval_steps_per_second": 0.839, + "step": 2150 + }, + { + "epoch": 2.357260273972603, + "grad_norm": 2.157257318496704, + "learning_rate": 1.2205523964256704e-06, + "loss": 2.0824, + "step": 2151 + }, + { + "epoch": 2.3583561643835615, + "grad_norm": 2.1270229816436768, + "learning_rate": 1.2185215272136477e-06, + "loss": 2.08, + "step": 2152 + }, + { + "epoch": 2.3594520547945206, + "grad_norm": 2.1364707946777344, + "learning_rate": 1.2164906580016247e-06, + "loss": 2.091, + "step": 2153 + }, + { + "epoch": 2.3605479452054796, + "grad_norm": 2.1324431896209717, + "learning_rate": 1.214459788789602e-06, + "loss": 2.0935, + "step": 2154 + }, + { + "epoch": 2.361643835616438, + "grad_norm": 2.1576826572418213, + "learning_rate": 1.2124289195775793e-06, + "loss": 2.0128, + "step": 2155 + }, + { + "epoch": 2.3627397260273972, + "grad_norm": 2.145265579223633, + "learning_rate": 1.2103980503655566e-06, + "loss": 2.0418, + "step": 2156 + }, + { + "epoch": 2.3638356164383563, + "grad_norm": 2.1475625038146973, + "learning_rate": 1.2083671811535338e-06, + "loss": 2.0726, + "step": 2157 + }, + { + "epoch": 2.364931506849315, + "grad_norm": 2.2365241050720215, + "learning_rate": 1.2063363119415111e-06, + "loss": 2.0598, + "step": 2158 + }, + { + "epoch": 2.366027397260274, + "grad_norm": 2.208568572998047, + "learning_rate": 1.2043054427294882e-06, + "loss": 2.0614, + "step": 2159 + }, + { + "epoch": 2.367123287671233, + "grad_norm": 2.1937685012817383, + "learning_rate": 1.2022745735174654e-06, + "loss": 2.0799, + "step": 2160 + }, + { + "epoch": 2.367123287671233, + "eval_loss": 2.2633299827575684, + "eval_runtime": 49.4985, + "eval_samples_per_second": 20.203, + "eval_steps_per_second": 0.849, + "step": 2160 + }, + { + "epoch": 2.368219178082192, + "grad_norm": 2.1548843383789062, + "learning_rate": 1.2002437043054427e-06, + "loss": 2.092, + "step": 2161 + }, + { + "epoch": 2.3693150684931505, + "grad_norm": 2.139727830886841, + "learning_rate": 1.19821283509342e-06, + "loss": 2.1167, + "step": 2162 + }, + { + "epoch": 2.3704109589041096, + "grad_norm": 2.1580495834350586, + "learning_rate": 1.1961819658813973e-06, + "loss": 2.1222, + "step": 2163 + }, + { + "epoch": 2.3715068493150686, + "grad_norm": 2.1156060695648193, + "learning_rate": 1.1941510966693746e-06, + "loss": 2.0996, + "step": 2164 + }, + { + "epoch": 2.3726027397260276, + "grad_norm": 2.280191659927368, + "learning_rate": 1.1921202274573518e-06, + "loss": 2.0737, + "step": 2165 + }, + { + "epoch": 2.3736986301369862, + "grad_norm": 2.1379661560058594, + "learning_rate": 1.190089358245329e-06, + "loss": 2.0803, + "step": 2166 + }, + { + "epoch": 2.3747945205479453, + "grad_norm": 2.261583089828491, + "learning_rate": 1.1880584890333064e-06, + "loss": 2.1225, + "step": 2167 + }, + { + "epoch": 2.3758904109589043, + "grad_norm": 2.233469009399414, + "learning_rate": 1.1860276198212837e-06, + "loss": 2.0774, + "step": 2168 + }, + { + "epoch": 2.376986301369863, + "grad_norm": 2.2570717334747314, + "learning_rate": 1.183996750609261e-06, + "loss": 2.1057, + "step": 2169 + }, + { + "epoch": 2.378082191780822, + "grad_norm": 2.2743802070617676, + "learning_rate": 1.1819658813972382e-06, + "loss": 2.093, + "step": 2170 + }, + { + "epoch": 2.378082191780822, + "eval_loss": 2.2612462043762207, + "eval_runtime": 50.375, + "eval_samples_per_second": 19.851, + "eval_steps_per_second": 0.834, + "step": 2170 + }, + { + "epoch": 2.379178082191781, + "grad_norm": 2.2214903831481934, + "learning_rate": 1.1799350121852153e-06, + "loss": 2.084, + "step": 2171 + }, + { + "epoch": 2.3802739726027395, + "grad_norm": 2.143113136291504, + "learning_rate": 1.1779041429731925e-06, + "loss": 2.1153, + "step": 2172 + }, + { + "epoch": 2.3813698630136986, + "grad_norm": 2.243147134780884, + "learning_rate": 1.1758732737611698e-06, + "loss": 2.1186, + "step": 2173 + }, + { + "epoch": 2.3824657534246576, + "grad_norm": 2.2865614891052246, + "learning_rate": 1.173842404549147e-06, + "loss": 2.0712, + "step": 2174 + }, + { + "epoch": 2.383561643835616, + "grad_norm": 2.274543523788452, + "learning_rate": 1.1718115353371244e-06, + "loss": 2.1214, + "step": 2175 + }, + { + "epoch": 2.3846575342465752, + "grad_norm": 2.249037981033325, + "learning_rate": 1.1697806661251016e-06, + "loss": 2.0703, + "step": 2176 + }, + { + "epoch": 2.3857534246575343, + "grad_norm": 2.2552239894866943, + "learning_rate": 1.167749796913079e-06, + "loss": 2.1198, + "step": 2177 + }, + { + "epoch": 2.3868493150684933, + "grad_norm": 2.2714426517486572, + "learning_rate": 1.1657189277010562e-06, + "loss": 2.1037, + "step": 2178 + }, + { + "epoch": 2.387945205479452, + "grad_norm": 2.2843732833862305, + "learning_rate": 1.1636880584890335e-06, + "loss": 2.0833, + "step": 2179 + }, + { + "epoch": 2.389041095890411, + "grad_norm": 2.3192999362945557, + "learning_rate": 1.1616571892770108e-06, + "loss": 2.0807, + "step": 2180 + }, + { + "epoch": 2.389041095890411, + "eval_loss": 2.2609643936157227, + "eval_runtime": 45.7294, + "eval_samples_per_second": 21.868, + "eval_steps_per_second": 0.918, + "step": 2180 + }, + { + "epoch": 2.39013698630137, + "grad_norm": 2.2852554321289062, + "learning_rate": 1.159626320064988e-06, + "loss": 2.1198, + "step": 2181 + }, + { + "epoch": 2.391232876712329, + "grad_norm": 2.3606770038604736, + "learning_rate": 1.1575954508529653e-06, + "loss": 2.0938, + "step": 2182 + }, + { + "epoch": 2.3923287671232876, + "grad_norm": 2.2159781455993652, + "learning_rate": 1.1555645816409426e-06, + "loss": 2.1089, + "step": 2183 + }, + { + "epoch": 2.3934246575342466, + "grad_norm": 2.2246878147125244, + "learning_rate": 1.1535337124289196e-06, + "loss": 2.0974, + "step": 2184 + }, + { + "epoch": 2.3945205479452056, + "grad_norm": 2.180250406265259, + "learning_rate": 1.151502843216897e-06, + "loss": 2.0527, + "step": 2185 + }, + { + "epoch": 2.3956164383561642, + "grad_norm": 2.1486611366271973, + "learning_rate": 1.1494719740048742e-06, + "loss": 2.0251, + "step": 2186 + }, + { + "epoch": 2.3967123287671233, + "grad_norm": 2.242518424987793, + "learning_rate": 1.1474411047928515e-06, + "loss": 2.0885, + "step": 2187 + }, + { + "epoch": 2.3978082191780823, + "grad_norm": 2.1474642753601074, + "learning_rate": 1.1454102355808287e-06, + "loss": 2.0946, + "step": 2188 + }, + { + "epoch": 2.398904109589041, + "grad_norm": 2.230221748352051, + "learning_rate": 1.143379366368806e-06, + "loss": 2.0639, + "step": 2189 + }, + { + "epoch": 2.4, + "grad_norm": 2.1985819339752197, + "learning_rate": 1.141348497156783e-06, + "loss": 2.0578, + "step": 2190 + }, + { + "epoch": 2.4, + "eval_loss": 2.264763832092285, + "eval_runtime": 48.4178, + "eval_samples_per_second": 20.654, + "eval_steps_per_second": 0.867, + "step": 2190 + }, + { + "epoch": 2.401095890410959, + "grad_norm": 2.2248053550720215, + "learning_rate": 1.1393176279447604e-06, + "loss": 2.149, + "step": 2191 + }, + { + "epoch": 2.402191780821918, + "grad_norm": 2.197953939437866, + "learning_rate": 1.1372867587327376e-06, + "loss": 2.138, + "step": 2192 + }, + { + "epoch": 2.4032876712328766, + "grad_norm": 2.2678914070129395, + "learning_rate": 1.135255889520715e-06, + "loss": 2.0202, + "step": 2193 + }, + { + "epoch": 2.4043835616438356, + "grad_norm": 2.3155181407928467, + "learning_rate": 1.1332250203086922e-06, + "loss": 2.1037, + "step": 2194 + }, + { + "epoch": 2.4054794520547946, + "grad_norm": 2.191927909851074, + "learning_rate": 1.1311941510966695e-06, + "loss": 2.0933, + "step": 2195 + }, + { + "epoch": 2.4065753424657537, + "grad_norm": 2.28680419921875, + "learning_rate": 1.1291632818846467e-06, + "loss": 2.0694, + "step": 2196 + }, + { + "epoch": 2.4076712328767123, + "grad_norm": 2.1030619144439697, + "learning_rate": 1.127132412672624e-06, + "loss": 2.0599, + "step": 2197 + }, + { + "epoch": 2.4087671232876713, + "grad_norm": 2.170182228088379, + "learning_rate": 1.1251015434606013e-06, + "loss": 2.0769, + "step": 2198 + }, + { + "epoch": 2.4098630136986303, + "grad_norm": 2.372509241104126, + "learning_rate": 1.1230706742485786e-06, + "loss": 2.0848, + "step": 2199 + }, + { + "epoch": 2.410958904109589, + "grad_norm": 2.3988277912139893, + "learning_rate": 1.1210398050365558e-06, + "loss": 2.1207, + "step": 2200 + }, + { + "epoch": 2.410958904109589, + "eval_loss": 2.261070966720581, + "eval_runtime": 47.1736, + "eval_samples_per_second": 21.198, + "eval_steps_per_second": 0.89, + "step": 2200 + }, + { + "epoch": 2.412054794520548, + "grad_norm": 2.240713357925415, + "learning_rate": 1.1190089358245331e-06, + "loss": 2.0871, + "step": 2201 + }, + { + "epoch": 2.413150684931507, + "grad_norm": 2.161547899246216, + "learning_rate": 1.1169780666125102e-06, + "loss": 2.0417, + "step": 2202 + }, + { + "epoch": 2.4142465753424656, + "grad_norm": 2.179996967315674, + "learning_rate": 1.1149471974004875e-06, + "loss": 2.1289, + "step": 2203 + }, + { + "epoch": 2.4153424657534246, + "grad_norm": 2.2374696731567383, + "learning_rate": 1.1129163281884647e-06, + "loss": 2.073, + "step": 2204 + }, + { + "epoch": 2.4164383561643836, + "grad_norm": 2.237311363220215, + "learning_rate": 1.110885458976442e-06, + "loss": 2.1315, + "step": 2205 + }, + { + "epoch": 2.417534246575342, + "grad_norm": 2.150914192199707, + "learning_rate": 1.1088545897644193e-06, + "loss": 2.0558, + "step": 2206 + }, + { + "epoch": 2.4186301369863012, + "grad_norm": 2.122668504714966, + "learning_rate": 1.1068237205523966e-06, + "loss": 2.0954, + "step": 2207 + }, + { + "epoch": 2.4197260273972603, + "grad_norm": 2.133937358856201, + "learning_rate": 1.1047928513403736e-06, + "loss": 2.0999, + "step": 2208 + }, + { + "epoch": 2.4208219178082193, + "grad_norm": 2.394454002380371, + "learning_rate": 1.102761982128351e-06, + "loss": 2.1137, + "step": 2209 + }, + { + "epoch": 2.421917808219178, + "grad_norm": 2.22770094871521, + "learning_rate": 1.1007311129163282e-06, + "loss": 2.0568, + "step": 2210 + }, + { + "epoch": 2.421917808219178, + "eval_loss": 2.2609522342681885, + "eval_runtime": 42.8266, + "eval_samples_per_second": 23.35, + "eval_steps_per_second": 0.981, + "step": 2210 + }, + { + "epoch": 2.423013698630137, + "grad_norm": 2.177499294281006, + "learning_rate": 1.0987002437043055e-06, + "loss": 2.0884, + "step": 2211 + }, + { + "epoch": 2.424109589041096, + "grad_norm": 2.113409996032715, + "learning_rate": 1.0966693744922827e-06, + "loss": 2.0629, + "step": 2212 + }, + { + "epoch": 2.425205479452055, + "grad_norm": 2.185673952102661, + "learning_rate": 1.09463850528026e-06, + "loss": 2.0363, + "step": 2213 + }, + { + "epoch": 2.4263013698630136, + "grad_norm": 2.180579423904419, + "learning_rate": 1.0926076360682373e-06, + "loss": 2.0713, + "step": 2214 + }, + { + "epoch": 2.4273972602739726, + "grad_norm": 2.1550121307373047, + "learning_rate": 1.0905767668562146e-06, + "loss": 2.059, + "step": 2215 + }, + { + "epoch": 2.4284931506849317, + "grad_norm": 2.084537982940674, + "learning_rate": 1.0885458976441918e-06, + "loss": 2.0733, + "step": 2216 + }, + { + "epoch": 2.4295890410958902, + "grad_norm": 2.1601779460906982, + "learning_rate": 1.0865150284321691e-06, + "loss": 2.1492, + "step": 2217 + }, + { + "epoch": 2.4306849315068493, + "grad_norm": 2.2117674350738525, + "learning_rate": 1.0844841592201464e-06, + "loss": 2.0873, + "step": 2218 + }, + { + "epoch": 2.4317808219178083, + "grad_norm": 2.283747434616089, + "learning_rate": 1.0824532900081237e-06, + "loss": 2.104, + "step": 2219 + }, + { + "epoch": 2.432876712328767, + "grad_norm": 2.23386287689209, + "learning_rate": 1.080422420796101e-06, + "loss": 2.0922, + "step": 2220 + }, + { + "epoch": 2.432876712328767, + "eval_loss": 2.261997699737549, + "eval_runtime": 43.1121, + "eval_samples_per_second": 23.195, + "eval_steps_per_second": 0.974, + "step": 2220 + }, + { + "epoch": 2.433972602739726, + "grad_norm": 2.2691948413848877, + "learning_rate": 1.078391551584078e-06, + "loss": 2.0831, + "step": 2221 + }, + { + "epoch": 2.435068493150685, + "grad_norm": 2.2326037883758545, + "learning_rate": 1.0763606823720553e-06, + "loss": 2.1057, + "step": 2222 + }, + { + "epoch": 2.436164383561644, + "grad_norm": 2.159956455230713, + "learning_rate": 1.0743298131600326e-06, + "loss": 2.1423, + "step": 2223 + }, + { + "epoch": 2.4372602739726026, + "grad_norm": 2.1688168048858643, + "learning_rate": 1.0722989439480098e-06, + "loss": 2.1128, + "step": 2224 + }, + { + "epoch": 2.4383561643835616, + "grad_norm": 2.167076587677002, + "learning_rate": 1.070268074735987e-06, + "loss": 2.0763, + "step": 2225 + }, + { + "epoch": 2.4394520547945207, + "grad_norm": 2.1673460006713867, + "learning_rate": 1.0682372055239644e-06, + "loss": 2.0812, + "step": 2226 + }, + { + "epoch": 2.4405479452054797, + "grad_norm": 2.1561999320983887, + "learning_rate": 1.0662063363119414e-06, + "loss": 2.068, + "step": 2227 + }, + { + "epoch": 2.4416438356164383, + "grad_norm": 2.2114264965057373, + "learning_rate": 1.0641754670999187e-06, + "loss": 2.0561, + "step": 2228 + }, + { + "epoch": 2.4427397260273973, + "grad_norm": 2.3172457218170166, + "learning_rate": 1.062144597887896e-06, + "loss": 2.082, + "step": 2229 + }, + { + "epoch": 2.4438356164383563, + "grad_norm": 2.1823253631591797, + "learning_rate": 1.0601137286758733e-06, + "loss": 2.0863, + "step": 2230 + }, + { + "epoch": 2.4438356164383563, + "eval_loss": 2.261075019836426, + "eval_runtime": 42.9325, + "eval_samples_per_second": 23.292, + "eval_steps_per_second": 0.978, + "step": 2230 + }, + { + "epoch": 2.444931506849315, + "grad_norm": 2.210676908493042, + "learning_rate": 1.0580828594638505e-06, + "loss": 2.1205, + "step": 2231 + }, + { + "epoch": 2.446027397260274, + "grad_norm": 2.1634883880615234, + "learning_rate": 1.0560519902518278e-06, + "loss": 2.0283, + "step": 2232 + }, + { + "epoch": 2.447123287671233, + "grad_norm": 2.261366844177246, + "learning_rate": 1.054021121039805e-06, + "loss": 2.1171, + "step": 2233 + }, + { + "epoch": 2.4482191780821916, + "grad_norm": 2.2115914821624756, + "learning_rate": 1.0519902518277824e-06, + "loss": 2.0596, + "step": 2234 + }, + { + "epoch": 2.4493150684931506, + "grad_norm": 2.115755558013916, + "learning_rate": 1.0499593826157596e-06, + "loss": 2.0325, + "step": 2235 + }, + { + "epoch": 2.4504109589041096, + "grad_norm": 2.189357280731201, + "learning_rate": 1.047928513403737e-06, + "loss": 2.1108, + "step": 2236 + }, + { + "epoch": 2.4515068493150687, + "grad_norm": 2.2712552547454834, + "learning_rate": 1.0458976441917142e-06, + "loss": 2.0796, + "step": 2237 + }, + { + "epoch": 2.4526027397260273, + "grad_norm": 2.1391139030456543, + "learning_rate": 1.0438667749796915e-06, + "loss": 2.0221, + "step": 2238 + }, + { + "epoch": 2.4536986301369863, + "grad_norm": 2.1458547115325928, + "learning_rate": 1.0418359057676688e-06, + "loss": 2.0283, + "step": 2239 + }, + { + "epoch": 2.4547945205479453, + "grad_norm": 2.277369737625122, + "learning_rate": 1.0398050365556458e-06, + "loss": 2.0602, + "step": 2240 + }, + { + "epoch": 2.4547945205479453, + "eval_loss": 2.2614047527313232, + "eval_runtime": 43.3285, + "eval_samples_per_second": 23.079, + "eval_steps_per_second": 0.969, + "step": 2240 + }, + { + "epoch": 2.455890410958904, + "grad_norm": 2.2398059368133545, + "learning_rate": 1.037774167343623e-06, + "loss": 2.1035, + "step": 2241 + }, + { + "epoch": 2.456986301369863, + "grad_norm": 2.065532922744751, + "learning_rate": 1.0357432981316004e-06, + "loss": 2.0609, + "step": 2242 + }, + { + "epoch": 2.458082191780822, + "grad_norm": 2.128404140472412, + "learning_rate": 1.0337124289195776e-06, + "loss": 2.0687, + "step": 2243 + }, + { + "epoch": 2.459178082191781, + "grad_norm": 2.3261191844940186, + "learning_rate": 1.031681559707555e-06, + "loss": 2.1186, + "step": 2244 + }, + { + "epoch": 2.4602739726027396, + "grad_norm": 2.1382572650909424, + "learning_rate": 1.0296506904955322e-06, + "loss": 2.11, + "step": 2245 + }, + { + "epoch": 2.4613698630136986, + "grad_norm": 2.1222827434539795, + "learning_rate": 1.0276198212835095e-06, + "loss": 2.0832, + "step": 2246 + }, + { + "epoch": 2.4624657534246577, + "grad_norm": 2.1852123737335205, + "learning_rate": 1.0255889520714867e-06, + "loss": 2.1013, + "step": 2247 + }, + { + "epoch": 2.4635616438356163, + "grad_norm": 2.2199220657348633, + "learning_rate": 1.023558082859464e-06, + "loss": 2.0696, + "step": 2248 + }, + { + "epoch": 2.4646575342465753, + "grad_norm": 2.3017048835754395, + "learning_rate": 1.0215272136474413e-06, + "loss": 2.0473, + "step": 2249 + }, + { + "epoch": 2.4657534246575343, + "grad_norm": 2.2449405193328857, + "learning_rate": 1.0194963444354186e-06, + "loss": 2.1242, + "step": 2250 + }, + { + "epoch": 2.4657534246575343, + "eval_loss": 2.2596442699432373, + "eval_runtime": 43.6954, + "eval_samples_per_second": 22.886, + "eval_steps_per_second": 0.961, + "step": 2250 + }, + { + "epoch": 2.466849315068493, + "grad_norm": 2.2314534187316895, + "learning_rate": 1.0174654752233958e-06, + "loss": 2.0427, + "step": 2251 + }, + { + "epoch": 2.467945205479452, + "grad_norm": 2.1703758239746094, + "learning_rate": 1.015434606011373e-06, + "loss": 2.1062, + "step": 2252 + }, + { + "epoch": 2.469041095890411, + "grad_norm": 2.2300264835357666, + "learning_rate": 1.0134037367993502e-06, + "loss": 2.0712, + "step": 2253 + }, + { + "epoch": 2.47013698630137, + "grad_norm": 2.234168767929077, + "learning_rate": 1.0113728675873275e-06, + "loss": 2.0909, + "step": 2254 + }, + { + "epoch": 2.4712328767123286, + "grad_norm": 2.3168294429779053, + "learning_rate": 1.0093419983753047e-06, + "loss": 2.0677, + "step": 2255 + }, + { + "epoch": 2.4723287671232876, + "grad_norm": 2.3184220790863037, + "learning_rate": 1.007311129163282e-06, + "loss": 2.0826, + "step": 2256 + }, + { + "epoch": 2.4734246575342467, + "grad_norm": 2.1959195137023926, + "learning_rate": 1.0052802599512593e-06, + "loss": 2.0322, + "step": 2257 + }, + { + "epoch": 2.4745205479452057, + "grad_norm": 2.149315595626831, + "learning_rate": 1.0032493907392364e-06, + "loss": 2.0481, + "step": 2258 + }, + { + "epoch": 2.4756164383561643, + "grad_norm": 2.296046733856201, + "learning_rate": 1.0012185215272136e-06, + "loss": 2.0772, + "step": 2259 + }, + { + "epoch": 2.4767123287671233, + "grad_norm": 2.27614688873291, + "learning_rate": 9.99187652315191e-07, + "loss": 2.058, + "step": 2260 + }, + { + "epoch": 2.4767123287671233, + "eval_loss": 2.2602272033691406, + "eval_runtime": 44.8225, + "eval_samples_per_second": 22.31, + "eval_steps_per_second": 0.937, + "step": 2260 + }, + { + "epoch": 2.4778082191780824, + "grad_norm": 2.2012453079223633, + "learning_rate": 9.971567831031682e-07, + "loss": 2.1203, + "step": 2261 + }, + { + "epoch": 2.478904109589041, + "grad_norm": 2.163407802581787, + "learning_rate": 9.951259138911455e-07, + "loss": 2.1094, + "step": 2262 + }, + { + "epoch": 2.48, + "grad_norm": 2.3561294078826904, + "learning_rate": 9.930950446791227e-07, + "loss": 2.0397, + "step": 2263 + }, + { + "epoch": 2.481095890410959, + "grad_norm": 2.1073873043060303, + "learning_rate": 9.910641754671e-07, + "loss": 2.1113, + "step": 2264 + }, + { + "epoch": 2.4821917808219176, + "grad_norm": 2.164715528488159, + "learning_rate": 9.890333062550773e-07, + "loss": 2.1396, + "step": 2265 + }, + { + "epoch": 2.4832876712328766, + "grad_norm": 2.1889424324035645, + "learning_rate": 9.870024370430546e-07, + "loss": 2.0828, + "step": 2266 + }, + { + "epoch": 2.4843835616438357, + "grad_norm": 2.2024834156036377, + "learning_rate": 9.849715678310318e-07, + "loss": 2.0857, + "step": 2267 + }, + { + "epoch": 2.4854794520547947, + "grad_norm": 2.2106359004974365, + "learning_rate": 9.829406986190091e-07, + "loss": 2.088, + "step": 2268 + }, + { + "epoch": 2.4865753424657533, + "grad_norm": 2.1396069526672363, + "learning_rate": 9.809098294069864e-07, + "loss": 2.0961, + "step": 2269 + }, + { + "epoch": 2.4876712328767123, + "grad_norm": 2.240629196166992, + "learning_rate": 9.788789601949637e-07, + "loss": 2.0289, + "step": 2270 + }, + { + "epoch": 2.4876712328767123, + "eval_loss": 2.2604970932006836, + "eval_runtime": 44.21, + "eval_samples_per_second": 22.619, + "eval_steps_per_second": 0.95, + "step": 2270 + }, + { + "epoch": 2.4887671232876714, + "grad_norm": 2.154015064239502, + "learning_rate": 9.768480909829407e-07, + "loss": 2.0948, + "step": 2271 + }, + { + "epoch": 2.48986301369863, + "grad_norm": 2.1784112453460693, + "learning_rate": 9.74817221770918e-07, + "loss": 2.137, + "step": 2272 + }, + { + "epoch": 2.490958904109589, + "grad_norm": 2.2804994583129883, + "learning_rate": 9.727863525588953e-07, + "loss": 2.081, + "step": 2273 + }, + { + "epoch": 2.492054794520548, + "grad_norm": 2.2236075401306152, + "learning_rate": 9.707554833468726e-07, + "loss": 2.1024, + "step": 2274 + }, + { + "epoch": 2.493150684931507, + "grad_norm": 2.1746160984039307, + "learning_rate": 9.687246141348498e-07, + "loss": 2.0705, + "step": 2275 + }, + { + "epoch": 2.4942465753424656, + "grad_norm": 2.2176294326782227, + "learning_rate": 9.666937449228271e-07, + "loss": 2.0681, + "step": 2276 + }, + { + "epoch": 2.4953424657534247, + "grad_norm": 2.1879889965057373, + "learning_rate": 9.646628757108042e-07, + "loss": 2.0762, + "step": 2277 + }, + { + "epoch": 2.4964383561643837, + "grad_norm": 2.391286611557007, + "learning_rate": 9.626320064987814e-07, + "loss": 2.0389, + "step": 2278 + }, + { + "epoch": 2.4975342465753423, + "grad_norm": 2.1714260578155518, + "learning_rate": 9.606011372867587e-07, + "loss": 2.098, + "step": 2279 + }, + { + "epoch": 2.4986301369863013, + "grad_norm": 2.0793519020080566, + "learning_rate": 9.58570268074736e-07, + "loss": 2.0494, + "step": 2280 + }, + { + "epoch": 2.4986301369863013, + "eval_loss": 2.260951280593872, + "eval_runtime": 46.0038, + "eval_samples_per_second": 21.737, + "eval_steps_per_second": 0.913, + "step": 2280 + }, + { + "epoch": 2.4997260273972604, + "grad_norm": 2.296259880065918, + "learning_rate": 9.565393988627133e-07, + "loss": 2.0811, + "step": 2281 + }, + { + "epoch": 2.500821917808219, + "grad_norm": 2.1998987197875977, + "learning_rate": 9.545085296506906e-07, + "loss": 2.0685, + "step": 2282 + }, + { + "epoch": 2.501917808219178, + "grad_norm": 2.187220573425293, + "learning_rate": 9.524776604386677e-07, + "loss": 2.0941, + "step": 2283 + }, + { + "epoch": 2.503013698630137, + "grad_norm": 2.2390315532684326, + "learning_rate": 9.50446791226645e-07, + "loss": 2.0918, + "step": 2284 + }, + { + "epoch": 2.504109589041096, + "grad_norm": 2.225874423980713, + "learning_rate": 9.484159220146223e-07, + "loss": 2.0376, + "step": 2285 + }, + { + "epoch": 2.5052054794520546, + "grad_norm": 2.198423147201538, + "learning_rate": 9.463850528025995e-07, + "loss": 2.1163, + "step": 2286 + }, + { + "epoch": 2.5063013698630137, + "grad_norm": 2.277414560317993, + "learning_rate": 9.443541835905768e-07, + "loss": 2.0969, + "step": 2287 + }, + { + "epoch": 2.5073972602739727, + "grad_norm": 2.134843111038208, + "learning_rate": 9.423233143785541e-07, + "loss": 2.1279, + "step": 2288 + }, + { + "epoch": 2.5084931506849317, + "grad_norm": 2.2586472034454346, + "learning_rate": 9.402924451665313e-07, + "loss": 2.1381, + "step": 2289 + }, + { + "epoch": 2.5095890410958903, + "grad_norm": 2.1865246295928955, + "learning_rate": 9.382615759545085e-07, + "loss": 2.0601, + "step": 2290 + }, + { + "epoch": 2.5095890410958903, + "eval_loss": 2.2607827186584473, + "eval_runtime": 45.5132, + "eval_samples_per_second": 21.972, + "eval_steps_per_second": 0.923, + "step": 2290 + }, + { + "epoch": 2.5106849315068493, + "grad_norm": 2.169753074645996, + "learning_rate": 9.362307067424858e-07, + "loss": 2.092, + "step": 2291 + }, + { + "epoch": 2.5117808219178084, + "grad_norm": 2.221168041229248, + "learning_rate": 9.341998375304631e-07, + "loss": 2.0689, + "step": 2292 + }, + { + "epoch": 2.512876712328767, + "grad_norm": 2.2252960205078125, + "learning_rate": 9.321689683184404e-07, + "loss": 2.1182, + "step": 2293 + }, + { + "epoch": 2.513972602739726, + "grad_norm": 2.1615395545959473, + "learning_rate": 9.301380991064176e-07, + "loss": 2.0489, + "step": 2294 + }, + { + "epoch": 2.515068493150685, + "grad_norm": 2.184530258178711, + "learning_rate": 9.281072298943949e-07, + "loss": 2.0474, + "step": 2295 + }, + { + "epoch": 2.5161643835616436, + "grad_norm": 2.248020887374878, + "learning_rate": 9.260763606823721e-07, + "loss": 2.0676, + "step": 2296 + }, + { + "epoch": 2.5172602739726027, + "grad_norm": 2.3340086936950684, + "learning_rate": 9.240454914703494e-07, + "loss": 2.0891, + "step": 2297 + }, + { + "epoch": 2.5183561643835617, + "grad_norm": 2.283479690551758, + "learning_rate": 9.220146222583266e-07, + "loss": 2.0921, + "step": 2298 + }, + { + "epoch": 2.5194520547945203, + "grad_norm": 2.1698408126831055, + "learning_rate": 9.199837530463039e-07, + "loss": 2.0567, + "step": 2299 + }, + { + "epoch": 2.5205479452054793, + "grad_norm": 2.1792118549346924, + "learning_rate": 9.179528838342812e-07, + "loss": 2.0857, + "step": 2300 + }, + { + "epoch": 2.5205479452054793, + "eval_loss": 2.2594637870788574, + "eval_runtime": 46.1455, + "eval_samples_per_second": 21.671, + "eval_steps_per_second": 0.91, + "step": 2300 + } + ], + "logging_steps": 1, + "max_steps": 2736, + "num_input_tokens_seen": 0, + "num_train_epochs": 3, + "save_steps": 100, + "stateful_callbacks": { + "TrainerControl": { + "args": { + "should_epoch_stop": false, + "should_evaluate": false, + "should_log": false, + "should_save": true, + "should_training_stop": false + }, + "attributes": {} + } + }, + "total_flos": 7.169811641532416e+17, + "train_batch_size": 3, + "trial_name": null, + "trial_params": null +} diff --git a/babylm_shuffle_nondeterministic_10M_seed0/runs/checkpoint-2300/zero_to_fp32.py b/babylm_shuffle_nondeterministic_10M_seed0/runs/checkpoint-2300/zero_to_fp32.py new file mode 100644 index 0000000000000000000000000000000000000000..24cc342e78d1a006c782b3a4cd68d9ce786d8fd8 --- /dev/null +++ b/babylm_shuffle_nondeterministic_10M_seed0/runs/checkpoint-2300/zero_to_fp32.py @@ -0,0 +1,604 @@ +#!/usr/bin/env python + +# Copyright (c) Microsoft Corporation. +# SPDX-License-Identifier: Apache-2.0 + +# DeepSpeed Team + +# This script extracts fp32 consolidated weights from a zero 1, 2 and 3 DeepSpeed checkpoints. It gets +# copied into the top level checkpoint dir, so the user can easily do the conversion at any point in +# the future. Once extracted, the weights don't require DeepSpeed and can be used in any +# application. +# +# example: python zero_to_fp32.py . pytorch_model.bin + +import argparse +import torch +import glob +import math +import os +import re +from collections import OrderedDict +from dataclasses import dataclass + +# while this script doesn't use deepspeed to recover data, since the checkpoints are pickled with +# DeepSpeed data structures it has to be available in the current python environment. +from deepspeed.utils import logger +from deepspeed.checkpoint.constants import (DS_VERSION, OPTIMIZER_STATE_DICT, SINGLE_PARTITION_OF_FP32_GROUPS, + FP32_FLAT_GROUPS, ZERO_STAGE, PARTITION_COUNT, PARAM_SHAPES, BUFFER_NAMES, + FROZEN_PARAM_SHAPES, FROZEN_PARAM_FRAGMENTS) + + +@dataclass +class zero_model_state: + buffers: dict() + param_shapes: dict() + shared_params: list + ds_version: int + frozen_param_shapes: dict() + frozen_param_fragments: dict() + + +debug = 0 + +# load to cpu +device = torch.device('cpu') + + +def atoi(text): + return int(text) if text.isdigit() else text + + +def natural_keys(text): + ''' + alist.sort(key=natural_keys) sorts in human order + http://nedbatchelder.com/blog/200712/human_sorting.html + (See Toothy's implementation in the comments) + ''' + return [atoi(c) for c in re.split(r'(\d+)', text)] + + +def get_model_state_file(checkpoint_dir, zero_stage): + if not os.path.isdir(checkpoint_dir): + raise FileNotFoundError(f"Directory '{checkpoint_dir}' doesn't exist") + + # there should be only one file + if zero_stage <= 2: + file = os.path.join(checkpoint_dir, "mp_rank_00_model_states.pt") + elif zero_stage == 3: + file = os.path.join(checkpoint_dir, "zero_pp_rank_0_mp_rank_00_model_states.pt") + + if not os.path.exists(file): + raise FileNotFoundError(f"can't find model states file at '{file}'") + + return file + + +def get_checkpoint_files(checkpoint_dir, glob_pattern): + # XXX: need to test that this simple glob rule works for multi-node setup too + ckpt_files = sorted(glob.glob(os.path.join(checkpoint_dir, glob_pattern)), key=natural_keys) + + if len(ckpt_files) == 0: + raise FileNotFoundError(f"can't find {glob_pattern} files in directory '{checkpoint_dir}'") + + return ckpt_files + + +def get_optim_files(checkpoint_dir): + return get_checkpoint_files(checkpoint_dir, "*_optim_states.pt") + + +def get_model_state_files(checkpoint_dir): + return get_checkpoint_files(checkpoint_dir, "*_model_states.pt") + + +def parse_model_states(files): + zero_model_states = [] + for file in files: + state_dict = torch.load(file, map_location=device) + + if BUFFER_NAMES not in state_dict: + raise ValueError(f"{file} is not a model state checkpoint") + buffer_names = state_dict[BUFFER_NAMES] + if debug: + print("Found buffers:", buffer_names) + + # recover just the buffers while restoring them to fp32 if they were saved in fp16 + buffers = {k: v.float() for k, v in state_dict["module"].items() if k in buffer_names} + param_shapes = state_dict[PARAM_SHAPES] + + # collect parameters that are included in param_shapes + param_names = [] + for s in param_shapes: + for name in s.keys(): + param_names.append(name) + + # update with frozen parameters + frozen_param_shapes = state_dict.get(FROZEN_PARAM_SHAPES, None) + if frozen_param_shapes is not None: + if debug: + print(f"Found frozen_param_shapes: {frozen_param_shapes}") + param_names += list(frozen_param_shapes.keys()) + + # handle shared params + shared_params = [[k, v] for k, v in state_dict["shared_params"].items()] + + ds_version = state_dict.get(DS_VERSION, None) + + frozen_param_fragments = state_dict.get(FROZEN_PARAM_FRAGMENTS, None) + + z_model_state = zero_model_state(buffers=buffers, + param_shapes=param_shapes, + shared_params=shared_params, + ds_version=ds_version, + frozen_param_shapes=frozen_param_shapes, + frozen_param_fragments=frozen_param_fragments) + zero_model_states.append(z_model_state) + + return zero_model_states + + +def parse_optim_states(files, ds_checkpoint_dir): + + total_files = len(files) + state_dicts = [] + for f in files: + state_dict = torch.load(f, map_location=device) + # immediately discard the potentially huge 2 optimizer states as we only care for fp32 master weights + # and also handle the case where it was already removed by another helper script + state_dict["optimizer_state_dict"].pop("optimizer_state_dict", None) + state_dicts.append(state_dict) + + if not ZERO_STAGE in state_dicts[0][OPTIMIZER_STATE_DICT]: + raise ValueError(f"{files[0]} is not a zero checkpoint") + zero_stage = state_dicts[0][OPTIMIZER_STATE_DICT][ZERO_STAGE] + world_size = state_dicts[0][OPTIMIZER_STATE_DICT][PARTITION_COUNT] + + # For ZeRO-2 each param group can have different partition_count as data parallelism for expert + # parameters can be different from data parallelism for non-expert parameters. So we can just + # use the max of the partition_count to get the dp world_size. + + if type(world_size) is list: + world_size = max(world_size) + + if world_size != total_files: + raise ValueError( + f"Expected {world_size} of '*_optim_states.pt' under '{ds_checkpoint_dir}' but found {total_files} files. " + "Possibly due to an overwrite of an old checkpoint, or a checkpoint didn't get saved by one or more processes." + ) + + # the groups are named differently in each stage + if zero_stage <= 2: + fp32_groups_key = SINGLE_PARTITION_OF_FP32_GROUPS + elif zero_stage == 3: + fp32_groups_key = FP32_FLAT_GROUPS + else: + raise ValueError(f"unknown zero stage {zero_stage}") + + if zero_stage <= 2: + fp32_flat_groups = [state_dicts[i][OPTIMIZER_STATE_DICT][fp32_groups_key] for i in range(len(state_dicts))] + elif zero_stage == 3: + # if there is more than one param group, there will be multiple flattened tensors - one + # flattened tensor per group - for simplicity merge them into a single tensor + # + # XXX: could make the script more memory efficient for when there are multiple groups - it + # will require matching the sub-lists of param_shapes for each param group flattened tensor + + fp32_flat_groups = [ + torch.cat(state_dicts[i][OPTIMIZER_STATE_DICT][fp32_groups_key], 0) for i in range(len(state_dicts)) + ] + + return zero_stage, world_size, fp32_flat_groups + + +def _get_fp32_state_dict_from_zero_checkpoint(ds_checkpoint_dir, exclude_frozen_parameters): + """ + Returns fp32 state_dict reconstructed from ds checkpoint + + Args: + - ``ds_checkpoint_dir``: path to the deepspeed checkpoint folder (where the optimizer files are) + + """ + print(f"Processing zero checkpoint '{ds_checkpoint_dir}'") + + optim_files = get_optim_files(ds_checkpoint_dir) + zero_stage, world_size, fp32_flat_groups = parse_optim_states(optim_files, ds_checkpoint_dir) + print(f"Detected checkpoint of type zero stage {zero_stage}, world_size: {world_size}") + + model_files = get_model_state_files(ds_checkpoint_dir) + + zero_model_states = parse_model_states(model_files) + print(f'Parsing checkpoint created by deepspeed=={zero_model_states[0].ds_version}') + + if zero_stage <= 2: + return _get_fp32_state_dict_from_zero2_checkpoint(world_size, fp32_flat_groups, zero_model_states, + exclude_frozen_parameters) + elif zero_stage == 3: + return _get_fp32_state_dict_from_zero3_checkpoint(world_size, fp32_flat_groups, zero_model_states, + exclude_frozen_parameters) + + +def _zero2_merge_frozen_params(state_dict, zero_model_states): + if zero_model_states[0].frozen_param_shapes is None or len(zero_model_states[0].frozen_param_shapes) == 0: + return + + frozen_param_shapes = zero_model_states[0].frozen_param_shapes + frozen_param_fragments = zero_model_states[0].frozen_param_fragments + + if debug: + num_elem = sum(s.numel() for s in frozen_param_shapes.values()) + print(f'rank 0: {FROZEN_PARAM_SHAPES}.numel = {num_elem}') + + wanted_params = len(frozen_param_shapes) + wanted_numel = sum(s.numel() for s in frozen_param_shapes.values()) + avail_numel = sum([p.numel() for p in frozen_param_fragments.values()]) + print(f'Frozen params: Have {avail_numel} numels to process.') + print(f'Frozen params: Need {wanted_numel} numels in {wanted_params} params') + + total_params = 0 + total_numel = 0 + for name, shape in frozen_param_shapes.items(): + total_params += 1 + unpartitioned_numel = shape.numel() + total_numel += unpartitioned_numel + + state_dict[name] = frozen_param_fragments[name] + + if debug: + print(f"{name} full shape: {shape} unpartitioned numel {unpartitioned_numel} ") + + print(f"Reconstructed Frozen fp32 state dict with {total_params} params {total_numel} elements") + + +def _has_callable(obj, fn): + attr = getattr(obj, fn, None) + return callable(attr) + + +def _zero2_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states): + param_shapes = zero_model_states[0].param_shapes + + # Reconstruction protocol: + # + # XXX: document this + + if debug: + for i in range(world_size): + for j in range(len(fp32_flat_groups[0])): + print(f"{FP32_FLAT_GROUPS}[{i}][{j}].shape={fp32_flat_groups[i][j].shape}") + + # XXX: memory usage doubles here (zero2) + num_param_groups = len(fp32_flat_groups[0]) + merged_single_partition_of_fp32_groups = [] + for i in range(num_param_groups): + merged_partitions = [sd[i] for sd in fp32_flat_groups] + full_single_fp32_vector = torch.cat(merged_partitions, 0) + merged_single_partition_of_fp32_groups.append(full_single_fp32_vector) + avail_numel = sum( + [full_single_fp32_vector.numel() for full_single_fp32_vector in merged_single_partition_of_fp32_groups]) + + if debug: + wanted_params = sum([len(shapes) for shapes in param_shapes]) + wanted_numel = sum([sum(shape.numel() for shape in shapes.values()) for shapes in param_shapes]) + # not asserting if there is a mismatch due to possible padding + print(f"Have {avail_numel} numels to process.") + print(f"Need {wanted_numel} numels in {wanted_params} params.") + + # params + # XXX: for huge models that can't fit into the host's RAM we will have to recode this to support + # out-of-core computing solution + total_numel = 0 + total_params = 0 + for shapes, full_single_fp32_vector in zip(param_shapes, merged_single_partition_of_fp32_groups): + offset = 0 + avail_numel = full_single_fp32_vector.numel() + for name, shape in shapes.items(): + + unpartitioned_numel = shape.numel() if _has_callable(shape, 'numel') else math.prod(shape) + total_numel += unpartitioned_numel + total_params += 1 + + if debug: + print(f"{name} full shape: {shape} unpartitioned numel {unpartitioned_numel} ") + state_dict[name] = full_single_fp32_vector.narrow(0, offset, unpartitioned_numel).view(shape) + offset += unpartitioned_numel + + # Z2 started to align to 2*world_size to improve nccl performance. Therefore both offset and + # avail_numel can differ by anywhere between 0..2*world_size. Due to two unrelated complex + # paddings performed in the code it's almost impossible to predict the exact numbers w/o the + # live optimizer object, so we are checking that the numbers are within the right range + align_to = 2 * world_size + + def zero2_align(x): + return align_to * math.ceil(x / align_to) + + if debug: + print(f"original offset={offset}, avail_numel={avail_numel}") + + offset = zero2_align(offset) + avail_numel = zero2_align(avail_numel) + + if debug: + print(f"aligned offset={offset}, avail_numel={avail_numel}") + + # Sanity check + if offset != avail_numel: + raise ValueError(f"consumed {offset} numels out of {avail_numel} - something is wrong") + + print(f"Reconstructed fp32 state dict with {total_params} params {total_numel} elements") + + +def _get_fp32_state_dict_from_zero2_checkpoint(world_size, fp32_flat_groups, zero_model_states, + exclude_frozen_parameters): + state_dict = OrderedDict() + + # buffers + buffers = zero_model_states[0].buffers + state_dict.update(buffers) + if debug: + print(f"added {len(buffers)} buffers") + + if not exclude_frozen_parameters: + _zero2_merge_frozen_params(state_dict, zero_model_states) + + _zero2_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states) + + # recover shared parameters + for pair in zero_model_states[0].shared_params: + if pair[1] in state_dict: + state_dict[pair[0]] = state_dict[pair[1]] + + return state_dict + + +def zero3_partitioned_param_info(unpartitioned_numel, world_size): + remainder = unpartitioned_numel % world_size + padding_numel = (world_size - remainder) if remainder else 0 + partitioned_numel = math.ceil(unpartitioned_numel / world_size) + return partitioned_numel, padding_numel + + +def _zero3_merge_frozen_params(state_dict, world_size, zero_model_states): + if zero_model_states[0].frozen_param_shapes is None or len(zero_model_states[0].frozen_param_shapes) == 0: + return + + if debug: + for i in range(world_size): + num_elem = sum(s.numel() for s in zero_model_states[i].frozen_param_fragments.values()) + print(f'rank {i}: {FROZEN_PARAM_SHAPES}.numel = {num_elem}') + + frozen_param_shapes = zero_model_states[0].frozen_param_shapes + wanted_params = len(frozen_param_shapes) + wanted_numel = sum(s.numel() for s in frozen_param_shapes.values()) + avail_numel = sum([p.numel() for p in zero_model_states[0].frozen_param_fragments.values()]) * world_size + print(f'Frozen params: Have {avail_numel} numels to process.') + print(f'Frozen params: Need {wanted_numel} numels in {wanted_params} params') + + total_params = 0 + total_numel = 0 + for name, shape in zero_model_states[0].frozen_param_shapes.items(): + total_params += 1 + unpartitioned_numel = shape.numel() + total_numel += unpartitioned_numel + + param_frags = tuple(model_state.frozen_param_fragments[name] for model_state in zero_model_states) + state_dict[name] = torch.cat(param_frags, 0).narrow(0, 0, unpartitioned_numel).view(shape) + + partitioned_numel, partitioned_padding_numel = zero3_partitioned_param_info(unpartitioned_numel, world_size) + + if debug: + print( + f"Frozen params: {total_params} {name} full shape: {shape} partition0 numel={partitioned_numel} partitioned_padding_numel={partitioned_padding_numel}" + ) + + print(f"Reconstructed Frozen fp32 state dict with {total_params} params {total_numel} elements") + + +def _zero3_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states): + param_shapes = zero_model_states[0].param_shapes + avail_numel = fp32_flat_groups[0].numel() * world_size + # Reconstruction protocol: For zero3 we need to zip the partitions together at boundary of each + # param, re-consolidating each param, while dealing with padding if any + + # merge list of dicts, preserving order + param_shapes = {k: v for d in param_shapes for k, v in d.items()} + + if debug: + for i in range(world_size): + print(f"{FP32_FLAT_GROUPS}[{i}].shape={fp32_flat_groups[i].shape}") + + wanted_params = len(param_shapes) + wanted_numel = sum(shape.numel() for shape in param_shapes.values()) + # not asserting if there is a mismatch due to possible padding + avail_numel = fp32_flat_groups[0].numel() * world_size + print(f"Trainable params: Have {avail_numel} numels to process.") + print(f"Trainable params: Need {wanted_numel} numels in {wanted_params} params.") + + # params + # XXX: for huge models that can't fit into the host's RAM we will have to recode this to support + # out-of-core computing solution + offset = 0 + total_numel = 0 + total_params = 0 + for name, shape in param_shapes.items(): + + unpartitioned_numel = shape.numel() + total_numel += unpartitioned_numel + total_params += 1 + + partitioned_numel, partitioned_padding_numel = zero3_partitioned_param_info(unpartitioned_numel, world_size) + + if debug: + print( + f"Trainable params: {total_params} {name} full shape: {shape} partition0 numel={partitioned_numel} partitioned_padding_numel={partitioned_padding_numel}" + ) + + # XXX: memory usage doubles here + state_dict[name] = torch.cat( + tuple(fp32_flat_groups[i].narrow(0, offset, partitioned_numel) for i in range(world_size)), + 0).narrow(0, 0, unpartitioned_numel).view(shape) + offset += partitioned_numel + + offset *= world_size + + # Sanity check + if offset != avail_numel: + raise ValueError(f"consumed {offset} numels out of {avail_numel} - something is wrong") + + print(f"Reconstructed Trainable fp32 state dict with {total_params} params {total_numel} elements") + + +def _get_fp32_state_dict_from_zero3_checkpoint(world_size, fp32_flat_groups, zero_model_states, + exclude_frozen_parameters): + state_dict = OrderedDict() + + # buffers + buffers = zero_model_states[0].buffers + state_dict.update(buffers) + if debug: + print(f"added {len(buffers)} buffers") + + if not exclude_frozen_parameters: + _zero3_merge_frozen_params(state_dict, world_size, zero_model_states) + + _zero3_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states) + + # recover shared parameters + for pair in zero_model_states[0].shared_params: + if pair[1] in state_dict: + state_dict[pair[0]] = state_dict[pair[1]] + + return state_dict + + +def get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir, tag=None, exclude_frozen_parameters=False): + """ + Convert ZeRO 2 or 3 checkpoint into a single fp32 consolidated state_dict that can be loaded with + ``load_state_dict()`` and used for training without DeepSpeed or shared with others, for example + via a model hub. + + Args: + - ``checkpoint_dir``: path to the desired checkpoint folder + - ``tag``: checkpoint tag used as a unique identifier for checkpoint. If not provided will attempt to load tag in 'latest' file. e.g., ``global_step14`` + - ``exclude_frozen_parameters``: exclude frozen parameters + + Returns: + - pytorch ``state_dict`` + + Note: this approach may not work if your application doesn't have sufficient free CPU memory and + you may need to use the offline approach using the ``zero_to_fp32.py`` script that is saved with + the checkpoint. + + A typical usage might be :: + + from deepspeed.utils.zero_to_fp32 import get_fp32_state_dict_from_zero_checkpoint + # do the training and checkpoint saving + state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir) # already on cpu + model = model.cpu() # move to cpu + model.load_state_dict(state_dict) + # submit to model hub or save the model to share with others + + In this example the ``model`` will no longer be usable in the deepspeed context of the same + application. i.e. you will need to re-initialize the deepspeed engine, since + ``model.load_state_dict(state_dict)`` will remove all the deepspeed magic from it. + + If you want it all done for you, use ``load_state_dict_from_zero_checkpoint`` instead. + + """ + if tag is None: + latest_path = os.path.join(checkpoint_dir, 'latest') + if os.path.isfile(latest_path): + with open(latest_path, 'r') as fd: + tag = fd.read().strip() + else: + raise ValueError(f"Unable to find 'latest' file at {latest_path}") + + ds_checkpoint_dir = os.path.join(checkpoint_dir, tag) + + if not os.path.isdir(ds_checkpoint_dir): + raise FileNotFoundError(f"Directory '{ds_checkpoint_dir}' doesn't exist") + + return _get_fp32_state_dict_from_zero_checkpoint(ds_checkpoint_dir, exclude_frozen_parameters) + + +def convert_zero_checkpoint_to_fp32_state_dict(checkpoint_dir, output_file, tag=None, exclude_frozen_parameters=False): + """ + Convert ZeRO 2 or 3 checkpoint into a single fp32 consolidated ``state_dict`` file that can be + loaded with ``torch.load(file)`` + ``load_state_dict()`` and used for training without DeepSpeed. + + Args: + - ``checkpoint_dir``: path to the desired checkpoint folder. (one that contains the tag-folder, like ``global_step14``) + - ``output_file``: path to the pytorch fp32 state_dict output file (e.g. path/pytorch_model.bin) + - ``tag``: checkpoint tag used as a unique identifier for checkpoint. If not provided will attempt to load tag in the file named ``latest`` in the checkpoint folder, e.g., ``global_step14`` + - ``exclude_frozen_parameters``: exclude frozen parameters + """ + + state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir, tag, exclude_frozen_parameters) + print(f"Saving fp32 state dict to {output_file}") + torch.save(state_dict, output_file) + + +def load_state_dict_from_zero_checkpoint(model, checkpoint_dir, tag=None): + """ + 1. Put the provided model to cpu + 2. Convert ZeRO 2 or 3 checkpoint into a single fp32 consolidated ``state_dict`` + 3. Load it into the provided model + + Args: + - ``model``: the model object to update + - ``checkpoint_dir``: path to the desired checkpoint folder. (one that contains the tag-folder, like ``global_step14``) + - ``tag``: checkpoint tag used as a unique identifier for checkpoint. If not provided will attempt to load tag in the file named ``latest`` in the checkpoint folder, e.g., ``global_step14`` + + Returns: + - ``model`: modified model + + Make sure you have plenty of CPU memory available before you call this function. If you don't + have enough use the ``zero_to_fp32.py`` utility to do the conversion. You will find it + conveniently placed for you in the checkpoint folder. + + A typical usage might be :: + + from deepspeed.utils.zero_to_fp32 import load_state_dict_from_zero_checkpoint + model = load_state_dict_from_zero_checkpoint(trainer.model, checkpoint_dir) + # submit to model hub or save the model to share with others + + Note, that once this was run, the ``model`` will no longer be usable in the deepspeed context + of the same application. i.e. you will need to re-initialize the deepspeed engine, since + ``model.load_state_dict(state_dict)`` will remove all the deepspeed magic from it. + + """ + logger.info(f"Extracting fp32 weights") + state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir, tag) + + logger.info(f"Overwriting model with fp32 weights") + model = model.cpu() + model.load_state_dict(state_dict, strict=False) + + return model + + +if __name__ == "__main__": + + parser = argparse.ArgumentParser() + parser.add_argument("checkpoint_dir", + type=str, + help="path to the desired checkpoint folder, e.g., path/checkpoint-12") + parser.add_argument( + "output_file", + type=str, + help="path to the pytorch fp32 state_dict output file (e.g. path/checkpoint-12/pytorch_model.bin)") + parser.add_argument("-t", + "--tag", + type=str, + default=None, + help="checkpoint tag used as a unique identifier for checkpoint. e.g., global_step1") + parser.add_argument("--exclude_frozen_parameters", action='store_true', help="exclude frozen parameters") + parser.add_argument("-d", "--debug", action='store_true', help="enable debug") + args = parser.parse_args() + + debug = args.debug + + convert_zero_checkpoint_to_fp32_state_dict(args.checkpoint_dir, + args.output_file, + tag=args.tag, + exclude_frozen_parameters=args.exclude_frozen_parameters) diff --git a/babylm_shuffle_nondeterministic_10M_seed0/runs/checkpoint-2736/config.json b/babylm_shuffle_nondeterministic_10M_seed0/runs/checkpoint-2736/config.json new file mode 100644 index 0000000000000000000000000000000000000000..fb0f9a53a507ffcdb60410deb71d60ef801bf350 --- /dev/null +++ b/babylm_shuffle_nondeterministic_10M_seed0/runs/checkpoint-2736/config.json @@ -0,0 +1,36 @@ +{ + "_name_or_path": "meta-llama/Llama-3.2-3B", + "architectures": [ + "LlamaForCausalLM" + ], + "attention_bias": false, + "attention_dropout": 0.0, + "bos_token_id": 128000, + "eos_token_id": 128001, + "head_dim": 128, + "hidden_act": "silu", + "hidden_size": 3072, + "initializer_range": 0.02, + "intermediate_size": 8192, + "max_position_embeddings": 131072, + "mlp_bias": false, + "model_type": "llama", + "num_attention_heads": 24, + "num_hidden_layers": 28, + "num_key_value_heads": 8, + "pretraining_tp": 1, + "rms_norm_eps": 1e-05, + "rope_scaling": { + "factor": 32.0, + "high_freq_factor": 4.0, + "low_freq_factor": 1.0, + "original_max_position_embeddings": 8192, + "rope_type": "llama3" + }, + "rope_theta": 500000.0, + "tie_word_embeddings": true, + "torch_dtype": "float16", + "transformers_version": "4.45.1", + "use_cache": true, + "vocab_size": 128256 +} diff --git a/babylm_shuffle_nondeterministic_10M_seed0/runs/checkpoint-2736/generation_config.json b/babylm_shuffle_nondeterministic_10M_seed0/runs/checkpoint-2736/generation_config.json new file mode 100644 index 0000000000000000000000000000000000000000..9c389b841a9a29ddd904e02d1eb0e08dcce82ad9 --- /dev/null +++ b/babylm_shuffle_nondeterministic_10M_seed0/runs/checkpoint-2736/generation_config.json @@ -0,0 +1,9 @@ +{ + "_from_model_config": true, + "bos_token_id": 128000, + "do_sample": true, + "eos_token_id": 128001, + "temperature": 0.6, + "top_p": 0.9, + "transformers_version": "4.45.1" +} diff --git a/babylm_shuffle_nondeterministic_10M_seed0/runs/checkpoint-2736/latest b/babylm_shuffle_nondeterministic_10M_seed0/runs/checkpoint-2736/latest new file mode 100644 index 0000000000000000000000000000000000000000..bb5cfcd38ec9615f610a0b181977e932b7a8cdd6 --- /dev/null +++ b/babylm_shuffle_nondeterministic_10M_seed0/runs/checkpoint-2736/latest @@ -0,0 +1 @@ +global_step2736 \ No newline at end of file diff --git a/babylm_shuffle_nondeterministic_10M_seed0/runs/checkpoint-2736/special_tokens_map.json b/babylm_shuffle_nondeterministic_10M_seed0/runs/checkpoint-2736/special_tokens_map.json new file mode 100644 index 0000000000000000000000000000000000000000..6949c5975ee0e961ef61cf31010dce04df0a03f8 --- /dev/null +++ b/babylm_shuffle_nondeterministic_10M_seed0/runs/checkpoint-2736/special_tokens_map.json @@ -0,0 +1,23 @@ +{ + "bos_token": { + "content": "<|begin_of_text|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false + }, + "eos_token": { + "content": "<|end_of_text|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false + }, + "pad_token": { + "content": "[PAD]", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false + } +} diff --git a/babylm_shuffle_nondeterministic_10M_seed0/runs/checkpoint-2736/trainer_state.json b/babylm_shuffle_nondeterministic_10M_seed0/runs/checkpoint-2736/trainer_state.json new file mode 100644 index 0000000000000000000000000000000000000000..5e455bb19469116e0fd759728ea842405d9f591d --- /dev/null +++ b/babylm_shuffle_nondeterministic_10M_seed0/runs/checkpoint-2736/trainer_state.json @@ -0,0 +1,21369 @@ +{ + "best_metric": null, + "best_model_checkpoint": null, + "epoch": 2.9983561643835617, + "eval_steps": 10, + "global_step": 2736, + "is_hyper_param_search": false, + "is_local_process_zero": true, + "is_world_process_zero": true, + "log_history": [ + { + "epoch": 0.001095890410958904, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.0998, + "step": 1 + }, + { + "epoch": 0.002191780821917808, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.0858, + "step": 2 + }, + { + "epoch": 0.0032876712328767125, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.0672, + "step": 3 + }, + { + "epoch": 0.004383561643835616, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.0802, + "step": 4 + }, + { + "epoch": 0.005479452054794521, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.0636, + "step": 5 + }, + { + "epoch": 0.006575342465753425, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.1672, + "step": 6 + }, + { + "epoch": 0.007671232876712329, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.0961, + "step": 7 + }, + { + "epoch": 0.008767123287671232, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.1258, + "step": 8 + }, + { + "epoch": 0.009863013698630137, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.0953, + "step": 9 + }, + { + "epoch": 0.010958904109589041, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.1404, + "step": 10 + }, + { + "epoch": 0.010958904109589041, + "eval_loss": 3.12593412399292, + "eval_runtime": 42.732, + "eval_samples_per_second": 23.402, + "eval_steps_per_second": 0.983, + "step": 10 + }, + { + "epoch": 0.012054794520547946, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.1187, + "step": 11 + }, + { + "epoch": 0.01315068493150685, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.0593, + "step": 12 + }, + { + "epoch": 0.014246575342465753, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.111, + "step": 13 + }, + { + "epoch": 0.015342465753424657, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.0974, + "step": 14 + }, + { + "epoch": 0.01643835616438356, + "grad_norm": 5.991453170776367, + "learning_rate": 1.8248175182481753e-08, + "loss": 3.135, + "step": 15 + }, + { + "epoch": 0.017534246575342465, + "grad_norm": 6.274210453033447, + "learning_rate": 3.6496350364963505e-08, + "loss": 3.083, + "step": 16 + }, + { + "epoch": 0.01863013698630137, + "grad_norm": 6.644786834716797, + "learning_rate": 5.4744525547445264e-08, + "loss": 3.0856, + "step": 17 + }, + { + "epoch": 0.019726027397260273, + "grad_norm": 6.769156455993652, + "learning_rate": 7.299270072992701e-08, + "loss": 3.1082, + "step": 18 + }, + { + "epoch": 0.020821917808219178, + "grad_norm": 6.771246910095215, + "learning_rate": 9.124087591240876e-08, + "loss": 3.0753, + "step": 19 + }, + { + "epoch": 0.021917808219178082, + "grad_norm": 6.468441486358643, + "learning_rate": 1.0948905109489053e-07, + "loss": 3.1478, + "step": 20 + }, + { + "epoch": 0.021917808219178082, + "eval_loss": 3.1245107650756836, + "eval_runtime": 43.6251, + "eval_samples_per_second": 22.923, + "eval_steps_per_second": 0.963, + "step": 20 + }, + { + "epoch": 0.023013698630136987, + "grad_norm": 6.83078670501709, + "learning_rate": 1.2773722627737227e-07, + "loss": 3.1213, + "step": 21 + }, + { + "epoch": 0.02410958904109589, + "grad_norm": 6.221287250518799, + "learning_rate": 1.4598540145985402e-07, + "loss": 3.0549, + "step": 22 + }, + { + "epoch": 0.025205479452054796, + "grad_norm": 5.564266204833984, + "learning_rate": 1.642335766423358e-07, + "loss": 3.0769, + "step": 23 + }, + { + "epoch": 0.0263013698630137, + "grad_norm": 5.735209941864014, + "learning_rate": 1.8248175182481753e-07, + "loss": 3.1124, + "step": 24 + }, + { + "epoch": 0.0273972602739726, + "grad_norm": 5.577925682067871, + "learning_rate": 2.0072992700729928e-07, + "loss": 3.1191, + "step": 25 + }, + { + "epoch": 0.028493150684931506, + "grad_norm": 4.814139366149902, + "learning_rate": 2.1897810218978106e-07, + "loss": 3.0947, + "step": 26 + }, + { + "epoch": 0.02958904109589041, + "grad_norm": 4.69775390625, + "learning_rate": 2.3722627737226278e-07, + "loss": 3.0997, + "step": 27 + }, + { + "epoch": 0.030684931506849315, + "grad_norm": 4.811586380004883, + "learning_rate": 2.5547445255474454e-07, + "loss": 3.0749, + "step": 28 + }, + { + "epoch": 0.03178082191780822, + "grad_norm": 4.623377323150635, + "learning_rate": 2.737226277372263e-07, + "loss": 3.0619, + "step": 29 + }, + { + "epoch": 0.03287671232876712, + "grad_norm": 4.721253395080566, + "learning_rate": 2.9197080291970804e-07, + "loss": 3.0371, + "step": 30 + }, + { + "epoch": 0.03287671232876712, + "eval_loss": 3.082857847213745, + "eval_runtime": 44.127, + "eval_samples_per_second": 22.662, + "eval_steps_per_second": 0.952, + "step": 30 + }, + { + "epoch": 0.03397260273972603, + "grad_norm": 3.968630790710449, + "learning_rate": 3.102189781021898e-07, + "loss": 3.0535, + "step": 31 + }, + { + "epoch": 0.03506849315068493, + "grad_norm": 4.4373779296875, + "learning_rate": 3.284671532846716e-07, + "loss": 3.0476, + "step": 32 + }, + { + "epoch": 0.03616438356164384, + "grad_norm": 4.479320049285889, + "learning_rate": 3.467153284671533e-07, + "loss": 3.0215, + "step": 33 + }, + { + "epoch": 0.03726027397260274, + "grad_norm": 4.122231960296631, + "learning_rate": 3.6496350364963505e-07, + "loss": 3.0106, + "step": 34 + }, + { + "epoch": 0.038356164383561646, + "grad_norm": 4.153995513916016, + "learning_rate": 3.8321167883211686e-07, + "loss": 3.0318, + "step": 35 + }, + { + "epoch": 0.03945205479452055, + "grad_norm": 3.6949076652526855, + "learning_rate": 4.0145985401459856e-07, + "loss": 3.046, + "step": 36 + }, + { + "epoch": 0.040547945205479455, + "grad_norm": 3.4610488414764404, + "learning_rate": 4.197080291970803e-07, + "loss": 3.002, + "step": 37 + }, + { + "epoch": 0.041643835616438356, + "grad_norm": 3.9523138999938965, + "learning_rate": 4.379562043795621e-07, + "loss": 2.9665, + "step": 38 + }, + { + "epoch": 0.042739726027397264, + "grad_norm": 3.68516206741333, + "learning_rate": 4.562043795620438e-07, + "loss": 3.0411, + "step": 39 + }, + { + "epoch": 0.043835616438356165, + "grad_norm": 3.692396402359009, + "learning_rate": 4.7445255474452557e-07, + "loss": 2.9666, + "step": 40 + }, + { + "epoch": 0.043835616438356165, + "eval_loss": 3.015512704849243, + "eval_runtime": 48.2438, + "eval_samples_per_second": 20.728, + "eval_steps_per_second": 0.871, + "step": 40 + }, + { + "epoch": 0.044931506849315066, + "grad_norm": 3.3765921592712402, + "learning_rate": 4.927007299270073e-07, + "loss": 2.9831, + "step": 41 + }, + { + "epoch": 0.046027397260273974, + "grad_norm": 2.9631125926971436, + "learning_rate": 5.109489051094891e-07, + "loss": 3.0126, + "step": 42 + }, + { + "epoch": 0.047123287671232875, + "grad_norm": 2.969730854034424, + "learning_rate": 5.291970802919708e-07, + "loss": 2.9803, + "step": 43 + }, + { + "epoch": 0.04821917808219178, + "grad_norm": 3.185237407684326, + "learning_rate": 5.474452554744526e-07, + "loss": 2.9888, + "step": 44 + }, + { + "epoch": 0.049315068493150684, + "grad_norm": 3.055579423904419, + "learning_rate": 5.656934306569343e-07, + "loss": 2.9715, + "step": 45 + }, + { + "epoch": 0.05041095890410959, + "grad_norm": 3.083284616470337, + "learning_rate": 5.839416058394161e-07, + "loss": 2.9293, + "step": 46 + }, + { + "epoch": 0.05150684931506849, + "grad_norm": 2.984781503677368, + "learning_rate": 6.021897810218979e-07, + "loss": 2.9228, + "step": 47 + }, + { + "epoch": 0.0526027397260274, + "grad_norm": 2.756173849105835, + "learning_rate": 6.204379562043796e-07, + "loss": 2.9406, + "step": 48 + }, + { + "epoch": 0.0536986301369863, + "grad_norm": 2.911132574081421, + "learning_rate": 6.386861313868614e-07, + "loss": 2.8829, + "step": 49 + }, + { + "epoch": 0.0547945205479452, + "grad_norm": 2.592890977859497, + "learning_rate": 6.569343065693432e-07, + "loss": 2.9839, + "step": 50 + }, + { + "epoch": 0.0547945205479452, + "eval_loss": 2.9409468173980713, + "eval_runtime": 47.1045, + "eval_samples_per_second": 21.229, + "eval_steps_per_second": 0.892, + "step": 50 + }, + { + "epoch": 0.05589041095890411, + "grad_norm": 3.0112955570220947, + "learning_rate": 6.751824817518248e-07, + "loss": 2.8847, + "step": 51 + }, + { + "epoch": 0.05698630136986301, + "grad_norm": 2.3701798915863037, + "learning_rate": 6.934306569343066e-07, + "loss": 2.9109, + "step": 52 + }, + { + "epoch": 0.05808219178082192, + "grad_norm": 2.268850088119507, + "learning_rate": 7.116788321167883e-07, + "loss": 2.8483, + "step": 53 + }, + { + "epoch": 0.05917808219178082, + "grad_norm": 2.419015645980835, + "learning_rate": 7.299270072992701e-07, + "loss": 2.8491, + "step": 54 + }, + { + "epoch": 0.06027397260273973, + "grad_norm": 2.4582653045654297, + "learning_rate": 7.48175182481752e-07, + "loss": 2.8622, + "step": 55 + }, + { + "epoch": 0.06136986301369863, + "grad_norm": 2.3471553325653076, + "learning_rate": 7.664233576642337e-07, + "loss": 2.8738, + "step": 56 + }, + { + "epoch": 0.06246575342465754, + "grad_norm": 2.479177236557007, + "learning_rate": 7.846715328467154e-07, + "loss": 2.8843, + "step": 57 + }, + { + "epoch": 0.06356164383561644, + "grad_norm": 2.1676292419433594, + "learning_rate": 8.029197080291971e-07, + "loss": 2.8674, + "step": 58 + }, + { + "epoch": 0.06465753424657535, + "grad_norm": 1.9862910509109497, + "learning_rate": 8.211678832116789e-07, + "loss": 2.8197, + "step": 59 + }, + { + "epoch": 0.06575342465753424, + "grad_norm": 2.211313486099243, + "learning_rate": 8.394160583941606e-07, + "loss": 2.7961, + "step": 60 + }, + { + "epoch": 0.06575342465753424, + "eval_loss": 2.880824327468872, + "eval_runtime": 51.3235, + "eval_samples_per_second": 19.484, + "eval_steps_per_second": 0.818, + "step": 60 + }, + { + "epoch": 0.06684931506849315, + "grad_norm": 2.3497281074523926, + "learning_rate": 8.576642335766425e-07, + "loss": 2.8233, + "step": 61 + }, + { + "epoch": 0.06794520547945206, + "grad_norm": 2.158515214920044, + "learning_rate": 8.759124087591242e-07, + "loss": 2.8136, + "step": 62 + }, + { + "epoch": 0.06904109589041096, + "grad_norm": 2.0219955444335938, + "learning_rate": 8.94160583941606e-07, + "loss": 2.8231, + "step": 63 + }, + { + "epoch": 0.07013698630136986, + "grad_norm": 1.8989133834838867, + "learning_rate": 9.124087591240876e-07, + "loss": 2.8268, + "step": 64 + }, + { + "epoch": 0.07123287671232877, + "grad_norm": 2.243851661682129, + "learning_rate": 9.306569343065694e-07, + "loss": 2.7886, + "step": 65 + }, + { + "epoch": 0.07232876712328767, + "grad_norm": 2.122464418411255, + "learning_rate": 9.489051094890511e-07, + "loss": 2.8166, + "step": 66 + }, + { + "epoch": 0.07342465753424658, + "grad_norm": 1.899857997894287, + "learning_rate": 9.67153284671533e-07, + "loss": 2.7997, + "step": 67 + }, + { + "epoch": 0.07452054794520548, + "grad_norm": 2.4424173831939697, + "learning_rate": 9.854014598540146e-07, + "loss": 2.8342, + "step": 68 + }, + { + "epoch": 0.07561643835616438, + "grad_norm": 2.5677411556243896, + "learning_rate": 1.0036496350364965e-06, + "loss": 2.7786, + "step": 69 + }, + { + "epoch": 0.07671232876712329, + "grad_norm": 1.9989523887634277, + "learning_rate": 1.0218978102189781e-06, + "loss": 2.7411, + "step": 70 + }, + { + "epoch": 0.07671232876712329, + "eval_loss": 2.8185322284698486, + "eval_runtime": 50.4998, + "eval_samples_per_second": 19.802, + "eval_steps_per_second": 0.832, + "step": 70 + }, + { + "epoch": 0.07780821917808219, + "grad_norm": 1.9376521110534668, + "learning_rate": 1.04014598540146e-06, + "loss": 2.7988, + "step": 71 + }, + { + "epoch": 0.0789041095890411, + "grad_norm": 2.5629725456237793, + "learning_rate": 1.0583941605839416e-06, + "loss": 2.7079, + "step": 72 + }, + { + "epoch": 0.08, + "grad_norm": 2.1634678840637207, + "learning_rate": 1.0766423357664233e-06, + "loss": 2.7049, + "step": 73 + }, + { + "epoch": 0.08109589041095891, + "grad_norm": 2.2143008708953857, + "learning_rate": 1.0948905109489052e-06, + "loss": 2.72, + "step": 74 + }, + { + "epoch": 0.0821917808219178, + "grad_norm": 2.2701101303100586, + "learning_rate": 1.113138686131387e-06, + "loss": 2.7418, + "step": 75 + }, + { + "epoch": 0.08328767123287671, + "grad_norm": 1.716569423675537, + "learning_rate": 1.1313868613138687e-06, + "loss": 2.7558, + "step": 76 + }, + { + "epoch": 0.08438356164383562, + "grad_norm": 2.79757022857666, + "learning_rate": 1.1496350364963505e-06, + "loss": 2.6865, + "step": 77 + }, + { + "epoch": 0.08547945205479453, + "grad_norm": 2.2189838886260986, + "learning_rate": 1.1678832116788322e-06, + "loss": 2.6684, + "step": 78 + }, + { + "epoch": 0.08657534246575342, + "grad_norm": 2.0695579051971436, + "learning_rate": 1.1861313868613138e-06, + "loss": 2.6779, + "step": 79 + }, + { + "epoch": 0.08767123287671233, + "grad_norm": 2.372326374053955, + "learning_rate": 1.2043795620437959e-06, + "loss": 2.6711, + "step": 80 + }, + { + "epoch": 0.08767123287671233, + "eval_loss": 2.7537453174591064, + "eval_runtime": 47.7274, + "eval_samples_per_second": 20.952, + "eval_steps_per_second": 0.88, + "step": 80 + }, + { + "epoch": 0.08876712328767124, + "grad_norm": 2.3398947715759277, + "learning_rate": 1.2226277372262775e-06, + "loss": 2.6776, + "step": 81 + }, + { + "epoch": 0.08986301369863013, + "grad_norm": 2.6619036197662354, + "learning_rate": 1.2408759124087592e-06, + "loss": 2.7077, + "step": 82 + }, + { + "epoch": 0.09095890410958904, + "grad_norm": 1.8245115280151367, + "learning_rate": 1.259124087591241e-06, + "loss": 2.662, + "step": 83 + }, + { + "epoch": 0.09205479452054795, + "grad_norm": 2.2627148628234863, + "learning_rate": 1.2773722627737229e-06, + "loss": 2.677, + "step": 84 + }, + { + "epoch": 0.09315068493150686, + "grad_norm": 2.4983835220336914, + "learning_rate": 1.2956204379562045e-06, + "loss": 2.6601, + "step": 85 + }, + { + "epoch": 0.09424657534246575, + "grad_norm": 2.147010087966919, + "learning_rate": 1.3138686131386864e-06, + "loss": 2.6213, + "step": 86 + }, + { + "epoch": 0.09534246575342466, + "grad_norm": 2.791682481765747, + "learning_rate": 1.332116788321168e-06, + "loss": 2.6281, + "step": 87 + }, + { + "epoch": 0.09643835616438357, + "grad_norm": 2.023970365524292, + "learning_rate": 1.3503649635036497e-06, + "loss": 2.5766, + "step": 88 + }, + { + "epoch": 0.09753424657534246, + "grad_norm": 2.8143956661224365, + "learning_rate": 1.3686131386861315e-06, + "loss": 2.6086, + "step": 89 + }, + { + "epoch": 0.09863013698630137, + "grad_norm": 2.226106882095337, + "learning_rate": 1.3868613138686132e-06, + "loss": 2.5477, + "step": 90 + }, + { + "epoch": 0.09863013698630137, + "eval_loss": 2.6880786418914795, + "eval_runtime": 47.8417, + "eval_samples_per_second": 20.902, + "eval_steps_per_second": 0.878, + "step": 90 + }, + { + "epoch": 0.09972602739726028, + "grad_norm": 2.5248148441314697, + "learning_rate": 1.405109489051095e-06, + "loss": 2.6232, + "step": 91 + }, + { + "epoch": 0.10082191780821918, + "grad_norm": 2.421330690383911, + "learning_rate": 1.4233576642335767e-06, + "loss": 2.599, + "step": 92 + }, + { + "epoch": 0.10191780821917808, + "grad_norm": 2.5562429428100586, + "learning_rate": 1.4416058394160583e-06, + "loss": 2.6197, + "step": 93 + }, + { + "epoch": 0.10301369863013699, + "grad_norm": 2.856201171875, + "learning_rate": 1.4598540145985402e-06, + "loss": 2.5825, + "step": 94 + }, + { + "epoch": 0.10410958904109589, + "grad_norm": 2.6576807498931885, + "learning_rate": 1.478102189781022e-06, + "loss": 2.5496, + "step": 95 + }, + { + "epoch": 0.1052054794520548, + "grad_norm": 3.1408190727233887, + "learning_rate": 1.496350364963504e-06, + "loss": 2.6142, + "step": 96 + }, + { + "epoch": 0.1063013698630137, + "grad_norm": 2.3270952701568604, + "learning_rate": 1.5145985401459856e-06, + "loss": 2.584, + "step": 97 + }, + { + "epoch": 0.1073972602739726, + "grad_norm": 2.3976614475250244, + "learning_rate": 1.5328467153284674e-06, + "loss": 2.5941, + "step": 98 + }, + { + "epoch": 0.10849315068493151, + "grad_norm": 2.783485174179077, + "learning_rate": 1.551094890510949e-06, + "loss": 2.5626, + "step": 99 + }, + { + "epoch": 0.1095890410958904, + "grad_norm": 2.2522788047790527, + "learning_rate": 1.5693430656934307e-06, + "loss": 2.517, + "step": 100 + }, + { + "epoch": 0.1095890410958904, + "eval_loss": 2.637315034866333, + "eval_runtime": 48.2107, + "eval_samples_per_second": 20.742, + "eval_steps_per_second": 0.871, + "step": 100 + }, + { + "epoch": 0.11068493150684931, + "grad_norm": 2.917799949645996, + "learning_rate": 1.5875912408759126e-06, + "loss": 2.5159, + "step": 101 + }, + { + "epoch": 0.11178082191780822, + "grad_norm": 2.955815076828003, + "learning_rate": 1.6058394160583942e-06, + "loss": 2.563, + "step": 102 + }, + { + "epoch": 0.11287671232876713, + "grad_norm": 2.8096001148223877, + "learning_rate": 1.624087591240876e-06, + "loss": 2.5995, + "step": 103 + }, + { + "epoch": 0.11397260273972602, + "grad_norm": 2.4967427253723145, + "learning_rate": 1.6423357664233577e-06, + "loss": 2.5336, + "step": 104 + }, + { + "epoch": 0.11506849315068493, + "grad_norm": 2.740717887878418, + "learning_rate": 1.6605839416058394e-06, + "loss": 2.4778, + "step": 105 + }, + { + "epoch": 0.11616438356164384, + "grad_norm": 2.5064632892608643, + "learning_rate": 1.6788321167883212e-06, + "loss": 2.5214, + "step": 106 + }, + { + "epoch": 0.11726027397260275, + "grad_norm": 2.2923007011413574, + "learning_rate": 1.6970802919708029e-06, + "loss": 2.5315, + "step": 107 + }, + { + "epoch": 0.11835616438356164, + "grad_norm": 3.339569568634033, + "learning_rate": 1.715328467153285e-06, + "loss": 2.4661, + "step": 108 + }, + { + "epoch": 0.11945205479452055, + "grad_norm": 2.1650891304016113, + "learning_rate": 1.7335766423357666e-06, + "loss": 2.5072, + "step": 109 + }, + { + "epoch": 0.12054794520547946, + "grad_norm": 3.1216914653778076, + "learning_rate": 1.7518248175182485e-06, + "loss": 2.5164, + "step": 110 + }, + { + "epoch": 0.12054794520547946, + "eval_loss": 2.594250202178955, + "eval_runtime": 41.8939, + "eval_samples_per_second": 23.87, + "eval_steps_per_second": 1.003, + "step": 110 + }, + { + "epoch": 0.12164383561643835, + "grad_norm": 2.191063404083252, + "learning_rate": 1.77007299270073e-06, + "loss": 2.5422, + "step": 111 + }, + { + "epoch": 0.12273972602739726, + "grad_norm": 2.922083854675293, + "learning_rate": 1.788321167883212e-06, + "loss": 2.5258, + "step": 112 + }, + { + "epoch": 0.12383561643835617, + "grad_norm": 2.518002986907959, + "learning_rate": 1.8065693430656936e-06, + "loss": 2.5588, + "step": 113 + }, + { + "epoch": 0.12493150684931507, + "grad_norm": 2.515194892883301, + "learning_rate": 1.8248175182481753e-06, + "loss": 2.5079, + "step": 114 + }, + { + "epoch": 0.12602739726027398, + "grad_norm": 2.5530874729156494, + "learning_rate": 1.8430656934306571e-06, + "loss": 2.4615, + "step": 115 + }, + { + "epoch": 0.12712328767123288, + "grad_norm": 2.3550198078155518, + "learning_rate": 1.8613138686131388e-06, + "loss": 2.4825, + "step": 116 + }, + { + "epoch": 0.12821917808219177, + "grad_norm": 2.295145034790039, + "learning_rate": 1.8795620437956206e-06, + "loss": 2.4749, + "step": 117 + }, + { + "epoch": 0.1293150684931507, + "grad_norm": 2.0158982276916504, + "learning_rate": 1.8978102189781023e-06, + "loss": 2.4756, + "step": 118 + }, + { + "epoch": 0.1304109589041096, + "grad_norm": 2.213057518005371, + "learning_rate": 1.916058394160584e-06, + "loss": 2.4744, + "step": 119 + }, + { + "epoch": 0.13150684931506848, + "grad_norm": 2.186222553253174, + "learning_rate": 1.934306569343066e-06, + "loss": 2.4957, + "step": 120 + }, + { + "epoch": 0.13150684931506848, + "eval_loss": 2.563931465148926, + "eval_runtime": 44.8714, + "eval_samples_per_second": 22.286, + "eval_steps_per_second": 0.936, + "step": 120 + }, + { + "epoch": 0.1326027397260274, + "grad_norm": 2.3143310546875, + "learning_rate": 1.952554744525548e-06, + "loss": 2.4, + "step": 121 + }, + { + "epoch": 0.1336986301369863, + "grad_norm": 2.0223965644836426, + "learning_rate": 1.9708029197080293e-06, + "loss": 2.4453, + "step": 122 + }, + { + "epoch": 0.13479452054794522, + "grad_norm": 2.2025411128997803, + "learning_rate": 1.989051094890511e-06, + "loss": 2.4463, + "step": 123 + }, + { + "epoch": 0.1358904109589041, + "grad_norm": 2.707930564880371, + "learning_rate": 2.007299270072993e-06, + "loss": 2.4589, + "step": 124 + }, + { + "epoch": 0.136986301369863, + "grad_norm": 2.146662473678589, + "learning_rate": 2.025547445255475e-06, + "loss": 2.5023, + "step": 125 + }, + { + "epoch": 0.13808219178082193, + "grad_norm": 2.520660161972046, + "learning_rate": 2.0437956204379563e-06, + "loss": 2.4464, + "step": 126 + }, + { + "epoch": 0.13917808219178082, + "grad_norm": 2.50632381439209, + "learning_rate": 2.062043795620438e-06, + "loss": 2.4556, + "step": 127 + }, + { + "epoch": 0.14027397260273972, + "grad_norm": 2.2919397354125977, + "learning_rate": 2.08029197080292e-06, + "loss": 2.4479, + "step": 128 + }, + { + "epoch": 0.14136986301369864, + "grad_norm": 2.521475315093994, + "learning_rate": 2.0985401459854014e-06, + "loss": 2.4982, + "step": 129 + }, + { + "epoch": 0.14246575342465753, + "grad_norm": 2.3812668323516846, + "learning_rate": 2.1167883211678833e-06, + "loss": 2.4163, + "step": 130 + }, + { + "epoch": 0.14246575342465753, + "eval_loss": 2.5390918254852295, + "eval_runtime": 43.8746, + "eval_samples_per_second": 22.792, + "eval_steps_per_second": 0.957, + "step": 130 + }, + { + "epoch": 0.14356164383561643, + "grad_norm": 2.302286148071289, + "learning_rate": 2.135036496350365e-06, + "loss": 2.4588, + "step": 131 + }, + { + "epoch": 0.14465753424657535, + "grad_norm": 2.4222044944763184, + "learning_rate": 2.1532846715328466e-06, + "loss": 2.492, + "step": 132 + }, + { + "epoch": 0.14575342465753424, + "grad_norm": 2.605835199356079, + "learning_rate": 2.171532846715329e-06, + "loss": 2.4055, + "step": 133 + }, + { + "epoch": 0.14684931506849316, + "grad_norm": 3.6261861324310303, + "learning_rate": 2.1897810218978103e-06, + "loss": 2.4221, + "step": 134 + }, + { + "epoch": 0.14794520547945206, + "grad_norm": 2.7647221088409424, + "learning_rate": 2.208029197080292e-06, + "loss": 2.4549, + "step": 135 + }, + { + "epoch": 0.14904109589041095, + "grad_norm": 3.2977371215820312, + "learning_rate": 2.226277372262774e-06, + "loss": 2.4331, + "step": 136 + }, + { + "epoch": 0.15013698630136987, + "grad_norm": 3.147094488143921, + "learning_rate": 2.244525547445256e-06, + "loss": 2.4268, + "step": 137 + }, + { + "epoch": 0.15123287671232877, + "grad_norm": 2.6363413333892822, + "learning_rate": 2.2627737226277373e-06, + "loss": 2.4524, + "step": 138 + }, + { + "epoch": 0.15232876712328766, + "grad_norm": 2.9288904666900635, + "learning_rate": 2.281021897810219e-06, + "loss": 2.4534, + "step": 139 + }, + { + "epoch": 0.15342465753424658, + "grad_norm": 2.5604870319366455, + "learning_rate": 2.299270072992701e-06, + "loss": 2.4138, + "step": 140 + }, + { + "epoch": 0.15342465753424658, + "eval_loss": 2.524790048599243, + "eval_runtime": 43.8851, + "eval_samples_per_second": 22.787, + "eval_steps_per_second": 0.957, + "step": 140 + }, + { + "epoch": 0.15452054794520548, + "grad_norm": 3.2746663093566895, + "learning_rate": 2.3175182481751825e-06, + "loss": 2.4094, + "step": 141 + }, + { + "epoch": 0.15561643835616437, + "grad_norm": 2.691483497619629, + "learning_rate": 2.3357664233576643e-06, + "loss": 2.4265, + "step": 142 + }, + { + "epoch": 0.1567123287671233, + "grad_norm": 2.5850117206573486, + "learning_rate": 2.354014598540146e-06, + "loss": 2.3922, + "step": 143 + }, + { + "epoch": 0.1578082191780822, + "grad_norm": 2.68204927444458, + "learning_rate": 2.3722627737226276e-06, + "loss": 2.4262, + "step": 144 + }, + { + "epoch": 0.1589041095890411, + "grad_norm": 2.952584981918335, + "learning_rate": 2.39051094890511e-06, + "loss": 2.4267, + "step": 145 + }, + { + "epoch": 0.16, + "grad_norm": 3.2354788780212402, + "learning_rate": 2.4087591240875918e-06, + "loss": 2.3836, + "step": 146 + }, + { + "epoch": 0.1610958904109589, + "grad_norm": 2.634521007537842, + "learning_rate": 2.427007299270073e-06, + "loss": 2.3953, + "step": 147 + }, + { + "epoch": 0.16219178082191782, + "grad_norm": 3.1115682125091553, + "learning_rate": 2.445255474452555e-06, + "loss": 2.409, + "step": 148 + }, + { + "epoch": 0.1632876712328767, + "grad_norm": 2.6824135780334473, + "learning_rate": 2.463503649635037e-06, + "loss": 2.4459, + "step": 149 + }, + { + "epoch": 0.1643835616438356, + "grad_norm": 2.617614507675171, + "learning_rate": 2.4817518248175183e-06, + "loss": 2.366, + "step": 150 + }, + { + "epoch": 0.1643835616438356, + "eval_loss": 2.5088160037994385, + "eval_runtime": 44.0888, + "eval_samples_per_second": 22.682, + "eval_steps_per_second": 0.953, + "step": 150 + }, + { + "epoch": 0.16547945205479453, + "grad_norm": 3.976992130279541, + "learning_rate": 2.5e-06, + "loss": 2.3517, + "step": 151 + }, + { + "epoch": 0.16657534246575342, + "grad_norm": 3.1433017253875732, + "learning_rate": 2.518248175182482e-06, + "loss": 2.4489, + "step": 152 + }, + { + "epoch": 0.16767123287671232, + "grad_norm": 4.100688934326172, + "learning_rate": 2.5364963503649635e-06, + "loss": 2.3683, + "step": 153 + }, + { + "epoch": 0.16876712328767124, + "grad_norm": 4.316102504730225, + "learning_rate": 2.5547445255474458e-06, + "loss": 2.3663, + "step": 154 + }, + { + "epoch": 0.16986301369863013, + "grad_norm": 2.3103132247924805, + "learning_rate": 2.5729927007299272e-06, + "loss": 2.4009, + "step": 155 + }, + { + "epoch": 0.17095890410958905, + "grad_norm": 3.576810836791992, + "learning_rate": 2.591240875912409e-06, + "loss": 2.3983, + "step": 156 + }, + { + "epoch": 0.17205479452054795, + "grad_norm": 2.337080955505371, + "learning_rate": 2.6094890510948905e-06, + "loss": 2.3849, + "step": 157 + }, + { + "epoch": 0.17315068493150684, + "grad_norm": 3.2523117065429688, + "learning_rate": 2.627737226277373e-06, + "loss": 2.4171, + "step": 158 + }, + { + "epoch": 0.17424657534246576, + "grad_norm": 2.529175281524658, + "learning_rate": 2.6459854014598542e-06, + "loss": 2.4303, + "step": 159 + }, + { + "epoch": 0.17534246575342466, + "grad_norm": 3.052212953567505, + "learning_rate": 2.664233576642336e-06, + "loss": 2.3895, + "step": 160 + }, + { + "epoch": 0.17534246575342466, + "eval_loss": 2.492485761642456, + "eval_runtime": 45.6738, + "eval_samples_per_second": 21.894, + "eval_steps_per_second": 0.92, + "step": 160 + }, + { + "epoch": 0.17643835616438355, + "grad_norm": 2.1544973850250244, + "learning_rate": 2.6824817518248175e-06, + "loss": 2.3649, + "step": 161 + }, + { + "epoch": 0.17753424657534247, + "grad_norm": 2.7351951599121094, + "learning_rate": 2.7007299270072994e-06, + "loss": 2.3908, + "step": 162 + }, + { + "epoch": 0.17863013698630137, + "grad_norm": 2.8431951999664307, + "learning_rate": 2.718978102189781e-06, + "loss": 2.374, + "step": 163 + }, + { + "epoch": 0.17972602739726026, + "grad_norm": 3.083190441131592, + "learning_rate": 2.737226277372263e-06, + "loss": 2.4193, + "step": 164 + }, + { + "epoch": 0.18082191780821918, + "grad_norm": 2.6523995399475098, + "learning_rate": 2.755474452554745e-06, + "loss": 2.3969, + "step": 165 + }, + { + "epoch": 0.18191780821917808, + "grad_norm": 2.626047372817993, + "learning_rate": 2.7737226277372264e-06, + "loss": 2.3742, + "step": 166 + }, + { + "epoch": 0.18301369863013697, + "grad_norm": 3.300917863845825, + "learning_rate": 2.7919708029197082e-06, + "loss": 2.4546, + "step": 167 + }, + { + "epoch": 0.1841095890410959, + "grad_norm": 2.797304391860962, + "learning_rate": 2.81021897810219e-06, + "loss": 2.4161, + "step": 168 + }, + { + "epoch": 0.1852054794520548, + "grad_norm": 2.777117967605591, + "learning_rate": 2.828467153284672e-06, + "loss": 2.3572, + "step": 169 + }, + { + "epoch": 0.1863013698630137, + "grad_norm": 2.887470006942749, + "learning_rate": 2.8467153284671534e-06, + "loss": 2.3887, + "step": 170 + }, + { + "epoch": 0.1863013698630137, + "eval_loss": 2.4854257106781006, + "eval_runtime": 44.5644, + "eval_samples_per_second": 22.439, + "eval_steps_per_second": 0.942, + "step": 170 + }, + { + "epoch": 0.1873972602739726, + "grad_norm": 3.0706076622009277, + "learning_rate": 2.8649635036496353e-06, + "loss": 2.3162, + "step": 171 + }, + { + "epoch": 0.1884931506849315, + "grad_norm": 3.6129744052886963, + "learning_rate": 2.8832116788321167e-06, + "loss": 2.3685, + "step": 172 + }, + { + "epoch": 0.18958904109589042, + "grad_norm": 2.4936370849609375, + "learning_rate": 2.901459854014599e-06, + "loss": 2.3297, + "step": 173 + }, + { + "epoch": 0.19068493150684931, + "grad_norm": 2.1041860580444336, + "learning_rate": 2.9197080291970804e-06, + "loss": 2.4085, + "step": 174 + }, + { + "epoch": 0.1917808219178082, + "grad_norm": 2.2066831588745117, + "learning_rate": 2.9379562043795623e-06, + "loss": 2.3905, + "step": 175 + }, + { + "epoch": 0.19287671232876713, + "grad_norm": 2.444652795791626, + "learning_rate": 2.956204379562044e-06, + "loss": 2.378, + "step": 176 + }, + { + "epoch": 0.19397260273972602, + "grad_norm": 2.501939058303833, + "learning_rate": 2.9744525547445256e-06, + "loss": 2.4136, + "step": 177 + }, + { + "epoch": 0.19506849315068492, + "grad_norm": 2.2167556285858154, + "learning_rate": 2.992700729927008e-06, + "loss": 2.385, + "step": 178 + }, + { + "epoch": 0.19616438356164384, + "grad_norm": 2.5298545360565186, + "learning_rate": 3.0109489051094893e-06, + "loss": 2.3508, + "step": 179 + }, + { + "epoch": 0.19726027397260273, + "grad_norm": 2.3774642944335938, + "learning_rate": 3.029197080291971e-06, + "loss": 2.3421, + "step": 180 + }, + { + "epoch": 0.19726027397260273, + "eval_loss": 2.476067304611206, + "eval_runtime": 44.3761, + "eval_samples_per_second": 22.535, + "eval_steps_per_second": 0.946, + "step": 180 + }, + { + "epoch": 0.19835616438356166, + "grad_norm": 2.9108712673187256, + "learning_rate": 3.0474452554744526e-06, + "loss": 2.3983, + "step": 181 + }, + { + "epoch": 0.19945205479452055, + "grad_norm": 2.5232033729553223, + "learning_rate": 3.065693430656935e-06, + "loss": 2.3946, + "step": 182 + }, + { + "epoch": 0.20054794520547944, + "grad_norm": 2.158168077468872, + "learning_rate": 3.0839416058394163e-06, + "loss": 2.3805, + "step": 183 + }, + { + "epoch": 0.20164383561643837, + "grad_norm": 2.3405354022979736, + "learning_rate": 3.102189781021898e-06, + "loss": 2.3974, + "step": 184 + }, + { + "epoch": 0.20273972602739726, + "grad_norm": 2.5308327674865723, + "learning_rate": 3.1204379562043796e-06, + "loss": 2.384, + "step": 185 + }, + { + "epoch": 0.20383561643835615, + "grad_norm": 2.6328237056732178, + "learning_rate": 3.1386861313868614e-06, + "loss": 2.3345, + "step": 186 + }, + { + "epoch": 0.20493150684931508, + "grad_norm": 3.0725512504577637, + "learning_rate": 3.156934306569343e-06, + "loss": 2.3191, + "step": 187 + }, + { + "epoch": 0.20602739726027397, + "grad_norm": 2.196218729019165, + "learning_rate": 3.175182481751825e-06, + "loss": 2.4073, + "step": 188 + }, + { + "epoch": 0.20712328767123286, + "grad_norm": 2.7722432613372803, + "learning_rate": 3.193430656934307e-06, + "loss": 2.3775, + "step": 189 + }, + { + "epoch": 0.20821917808219179, + "grad_norm": 2.4848241806030273, + "learning_rate": 3.2116788321167884e-06, + "loss": 2.3538, + "step": 190 + }, + { + "epoch": 0.20821917808219179, + "eval_loss": 2.462618112564087, + "eval_runtime": 43.9461, + "eval_samples_per_second": 22.755, + "eval_steps_per_second": 0.956, + "step": 190 + }, + { + "epoch": 0.20931506849315068, + "grad_norm": 2.4682741165161133, + "learning_rate": 3.2299270072992707e-06, + "loss": 2.3688, + "step": 191 + }, + { + "epoch": 0.2104109589041096, + "grad_norm": 2.518923282623291, + "learning_rate": 3.248175182481752e-06, + "loss": 2.3556, + "step": 192 + }, + { + "epoch": 0.2115068493150685, + "grad_norm": 2.4023561477661133, + "learning_rate": 3.266423357664234e-06, + "loss": 2.292, + "step": 193 + }, + { + "epoch": 0.2126027397260274, + "grad_norm": 2.5196943283081055, + "learning_rate": 3.2846715328467155e-06, + "loss": 2.3703, + "step": 194 + }, + { + "epoch": 0.2136986301369863, + "grad_norm": 2.762709617614746, + "learning_rate": 3.3029197080291973e-06, + "loss": 2.3578, + "step": 195 + }, + { + "epoch": 0.2147945205479452, + "grad_norm": 2.1287691593170166, + "learning_rate": 3.3211678832116788e-06, + "loss": 2.3687, + "step": 196 + }, + { + "epoch": 0.2158904109589041, + "grad_norm": 2.477341651916504, + "learning_rate": 3.339416058394161e-06, + "loss": 2.3545, + "step": 197 + }, + { + "epoch": 0.21698630136986302, + "grad_norm": 2.23933744430542, + "learning_rate": 3.3576642335766425e-06, + "loss": 2.3511, + "step": 198 + }, + { + "epoch": 0.21808219178082192, + "grad_norm": 2.2440133094787598, + "learning_rate": 3.3759124087591243e-06, + "loss": 2.3539, + "step": 199 + }, + { + "epoch": 0.2191780821917808, + "grad_norm": 2.5603177547454834, + "learning_rate": 3.3941605839416058e-06, + "loss": 2.36, + "step": 200 + }, + { + "epoch": 0.2191780821917808, + "eval_loss": 2.4507246017456055, + "eval_runtime": 43.7786, + "eval_samples_per_second": 22.842, + "eval_steps_per_second": 0.959, + "step": 200 + }, + { + "epoch": 0.22027397260273973, + "grad_norm": 2.4570510387420654, + "learning_rate": 3.412408759124088e-06, + "loss": 2.3702, + "step": 201 + }, + { + "epoch": 0.22136986301369863, + "grad_norm": 2.1975133419036865, + "learning_rate": 3.43065693430657e-06, + "loss": 2.3667, + "step": 202 + }, + { + "epoch": 0.22246575342465755, + "grad_norm": 2.710186004638672, + "learning_rate": 3.4489051094890513e-06, + "loss": 2.3159, + "step": 203 + }, + { + "epoch": 0.22356164383561644, + "grad_norm": 2.7799506187438965, + "learning_rate": 3.467153284671533e-06, + "loss": 2.3519, + "step": 204 + }, + { + "epoch": 0.22465753424657534, + "grad_norm": 2.2479400634765625, + "learning_rate": 3.4854014598540146e-06, + "loss": 2.3103, + "step": 205 + }, + { + "epoch": 0.22575342465753426, + "grad_norm": 3.1538405418395996, + "learning_rate": 3.503649635036497e-06, + "loss": 2.3506, + "step": 206 + }, + { + "epoch": 0.22684931506849315, + "grad_norm": 2.7607762813568115, + "learning_rate": 3.5218978102189783e-06, + "loss": 2.3386, + "step": 207 + }, + { + "epoch": 0.22794520547945205, + "grad_norm": 2.371105909347534, + "learning_rate": 3.54014598540146e-06, + "loss": 2.3583, + "step": 208 + }, + { + "epoch": 0.22904109589041097, + "grad_norm": 2.793933391571045, + "learning_rate": 3.5583941605839416e-06, + "loss": 2.3782, + "step": 209 + }, + { + "epoch": 0.23013698630136986, + "grad_norm": 2.372732400894165, + "learning_rate": 3.576642335766424e-06, + "loss": 2.417, + "step": 210 + }, + { + "epoch": 0.23013698630136986, + "eval_loss": 2.451474189758301, + "eval_runtime": 41.7197, + "eval_samples_per_second": 23.97, + "eval_steps_per_second": 1.007, + "step": 210 + }, + { + "epoch": 0.23123287671232876, + "grad_norm": 2.6989448070526123, + "learning_rate": 3.5948905109489054e-06, + "loss": 2.393, + "step": 211 + }, + { + "epoch": 0.23232876712328768, + "grad_norm": 2.4324285984039307, + "learning_rate": 3.6131386861313872e-06, + "loss": 2.2738, + "step": 212 + }, + { + "epoch": 0.23342465753424657, + "grad_norm": 3.5063016414642334, + "learning_rate": 3.6313868613138687e-06, + "loss": 2.2823, + "step": 213 + }, + { + "epoch": 0.2345205479452055, + "grad_norm": 3.953951358795166, + "learning_rate": 3.6496350364963505e-06, + "loss": 2.3132, + "step": 214 + }, + { + "epoch": 0.2356164383561644, + "grad_norm": 2.3483071327209473, + "learning_rate": 3.667883211678833e-06, + "loss": 2.3244, + "step": 215 + }, + { + "epoch": 0.23671232876712328, + "grad_norm": 2.884199380874634, + "learning_rate": 3.6861313868613142e-06, + "loss": 2.3067, + "step": 216 + }, + { + "epoch": 0.2378082191780822, + "grad_norm": 2.9514520168304443, + "learning_rate": 3.704379562043796e-06, + "loss": 2.304, + "step": 217 + }, + { + "epoch": 0.2389041095890411, + "grad_norm": 2.7530996799468994, + "learning_rate": 3.7226277372262775e-06, + "loss": 2.3691, + "step": 218 + }, + { + "epoch": 0.24, + "grad_norm": 2.8753647804260254, + "learning_rate": 3.7408759124087594e-06, + "loss": 2.3406, + "step": 219 + }, + { + "epoch": 0.2410958904109589, + "grad_norm": 2.670574903488159, + "learning_rate": 3.7591240875912412e-06, + "loss": 2.304, + "step": 220 + }, + { + "epoch": 0.2410958904109589, + "eval_loss": 2.4395809173583984, + "eval_runtime": 46.6723, + "eval_samples_per_second": 21.426, + "eval_steps_per_second": 0.9, + "step": 220 + }, + { + "epoch": 0.2421917808219178, + "grad_norm": 2.4152438640594482, + "learning_rate": 3.777372262773723e-06, + "loss": 2.3294, + "step": 221 + }, + { + "epoch": 0.2432876712328767, + "grad_norm": 2.8398540019989014, + "learning_rate": 3.7956204379562045e-06, + "loss": 2.3216, + "step": 222 + }, + { + "epoch": 0.24438356164383562, + "grad_norm": 2.6036765575408936, + "learning_rate": 3.8138686131386864e-06, + "loss": 2.3109, + "step": 223 + }, + { + "epoch": 0.24547945205479452, + "grad_norm": 2.7073473930358887, + "learning_rate": 3.832116788321168e-06, + "loss": 2.3457, + "step": 224 + }, + { + "epoch": 0.2465753424657534, + "grad_norm": 2.584404230117798, + "learning_rate": 3.85036496350365e-06, + "loss": 2.3159, + "step": 225 + }, + { + "epoch": 0.24767123287671233, + "grad_norm": 2.4220666885375977, + "learning_rate": 3.868613138686132e-06, + "loss": 2.2944, + "step": 226 + }, + { + "epoch": 0.24876712328767123, + "grad_norm": 2.5636494159698486, + "learning_rate": 3.886861313868613e-06, + "loss": 2.3219, + "step": 227 + }, + { + "epoch": 0.24986301369863015, + "grad_norm": 2.2990903854370117, + "learning_rate": 3.905109489051096e-06, + "loss": 2.3342, + "step": 228 + }, + { + "epoch": 0.25095890410958904, + "grad_norm": 2.50913143157959, + "learning_rate": 3.923357664233577e-06, + "loss": 2.3619, + "step": 229 + }, + { + "epoch": 0.25205479452054796, + "grad_norm": 2.1687769889831543, + "learning_rate": 3.9416058394160585e-06, + "loss": 2.3153, + "step": 230 + }, + { + "epoch": 0.25205479452054796, + "eval_loss": 2.4315240383148193, + "eval_runtime": 48.3542, + "eval_samples_per_second": 20.681, + "eval_steps_per_second": 0.869, + "step": 230 + }, + { + "epoch": 0.25315068493150683, + "grad_norm": 2.842125177383423, + "learning_rate": 3.95985401459854e-06, + "loss": 2.3644, + "step": 231 + }, + { + "epoch": 0.25424657534246575, + "grad_norm": 2.258049964904785, + "learning_rate": 3.978102189781022e-06, + "loss": 2.3727, + "step": 232 + }, + { + "epoch": 0.2553424657534247, + "grad_norm": 2.4630489349365234, + "learning_rate": 3.996350364963504e-06, + "loss": 2.267, + "step": 233 + }, + { + "epoch": 0.25643835616438354, + "grad_norm": 3.1469743251800537, + "learning_rate": 4.014598540145986e-06, + "loss": 2.3432, + "step": 234 + }, + { + "epoch": 0.25753424657534246, + "grad_norm": 2.6277477741241455, + "learning_rate": 4.032846715328467e-06, + "loss": 2.2545, + "step": 235 + }, + { + "epoch": 0.2586301369863014, + "grad_norm": 2.404141902923584, + "learning_rate": 4.05109489051095e-06, + "loss": 2.3109, + "step": 236 + }, + { + "epoch": 0.25972602739726025, + "grad_norm": 2.4545605182647705, + "learning_rate": 4.069343065693431e-06, + "loss": 2.3112, + "step": 237 + }, + { + "epoch": 0.2608219178082192, + "grad_norm": 2.237638235092163, + "learning_rate": 4.0875912408759126e-06, + "loss": 2.3, + "step": 238 + }, + { + "epoch": 0.2619178082191781, + "grad_norm": 2.192811965942383, + "learning_rate": 4.105839416058395e-06, + "loss": 2.3108, + "step": 239 + }, + { + "epoch": 0.26301369863013696, + "grad_norm": 2.5058720111846924, + "learning_rate": 4.124087591240876e-06, + "loss": 2.3635, + "step": 240 + }, + { + "epoch": 0.26301369863013696, + "eval_loss": 2.4219377040863037, + "eval_runtime": 45.3561, + "eval_samples_per_second": 22.048, + "eval_steps_per_second": 0.926, + "step": 240 + }, + { + "epoch": 0.2641095890410959, + "grad_norm": 2.0708935260772705, + "learning_rate": 4.1423357664233586e-06, + "loss": 2.3129, + "step": 241 + }, + { + "epoch": 0.2652054794520548, + "grad_norm": 2.6353116035461426, + "learning_rate": 4.16058394160584e-06, + "loss": 2.3011, + "step": 242 + }, + { + "epoch": 0.2663013698630137, + "grad_norm": 2.8392436504364014, + "learning_rate": 4.1788321167883214e-06, + "loss": 2.3643, + "step": 243 + }, + { + "epoch": 0.2673972602739726, + "grad_norm": 3.1773295402526855, + "learning_rate": 4.197080291970803e-06, + "loss": 2.2642, + "step": 244 + }, + { + "epoch": 0.2684931506849315, + "grad_norm": 2.4757354259490967, + "learning_rate": 4.215328467153285e-06, + "loss": 2.3211, + "step": 245 + }, + { + "epoch": 0.26958904109589044, + "grad_norm": 2.5799412727355957, + "learning_rate": 4.233576642335767e-06, + "loss": 2.2909, + "step": 246 + }, + { + "epoch": 0.2706849315068493, + "grad_norm": 2.368502378463745, + "learning_rate": 4.251824817518249e-06, + "loss": 2.2951, + "step": 247 + }, + { + "epoch": 0.2717808219178082, + "grad_norm": 2.114969253540039, + "learning_rate": 4.27007299270073e-06, + "loss": 2.3124, + "step": 248 + }, + { + "epoch": 0.27287671232876715, + "grad_norm": 3.056820869445801, + "learning_rate": 4.288321167883212e-06, + "loss": 2.2923, + "step": 249 + }, + { + "epoch": 0.273972602739726, + "grad_norm": 2.7752203941345215, + "learning_rate": 4.306569343065693e-06, + "loss": 2.2874, + "step": 250 + }, + { + "epoch": 0.273972602739726, + "eval_loss": 2.4229912757873535, + "eval_runtime": 52.5941, + "eval_samples_per_second": 19.014, + "eval_steps_per_second": 0.799, + "step": 250 + }, + { + "epoch": 0.27506849315068493, + "grad_norm": 3.045740842819214, + "learning_rate": 4.3248175182481755e-06, + "loss": 2.2686, + "step": 251 + }, + { + "epoch": 0.27616438356164386, + "grad_norm": 2.646402359008789, + "learning_rate": 4.343065693430658e-06, + "loss": 2.3247, + "step": 252 + }, + { + "epoch": 0.2772602739726027, + "grad_norm": 3.274994134902954, + "learning_rate": 4.361313868613139e-06, + "loss": 2.3093, + "step": 253 + }, + { + "epoch": 0.27835616438356164, + "grad_norm": 2.352246046066284, + "learning_rate": 4.379562043795621e-06, + "loss": 2.3124, + "step": 254 + }, + { + "epoch": 0.27945205479452057, + "grad_norm": 3.248717784881592, + "learning_rate": 4.397810218978102e-06, + "loss": 2.3417, + "step": 255 + }, + { + "epoch": 0.28054794520547943, + "grad_norm": 3.0919013023376465, + "learning_rate": 4.416058394160584e-06, + "loss": 2.3093, + "step": 256 + }, + { + "epoch": 0.28164383561643835, + "grad_norm": 2.1174702644348145, + "learning_rate": 4.434306569343066e-06, + "loss": 2.305, + "step": 257 + }, + { + "epoch": 0.2827397260273973, + "grad_norm": 2.913797378540039, + "learning_rate": 4.452554744525548e-06, + "loss": 2.2831, + "step": 258 + }, + { + "epoch": 0.28383561643835614, + "grad_norm": 2.693089723587036, + "learning_rate": 4.4708029197080295e-06, + "loss": 2.3206, + "step": 259 + }, + { + "epoch": 0.28493150684931506, + "grad_norm": 2.521411895751953, + "learning_rate": 4.489051094890512e-06, + "loss": 2.3196, + "step": 260 + }, + { + "epoch": 0.28493150684931506, + "eval_loss": 2.4178109169006348, + "eval_runtime": 50.6753, + "eval_samples_per_second": 19.733, + "eval_steps_per_second": 0.829, + "step": 260 + }, + { + "epoch": 0.286027397260274, + "grad_norm": 2.346346855163574, + "learning_rate": 4.507299270072993e-06, + "loss": 2.3433, + "step": 261 + }, + { + "epoch": 0.28712328767123285, + "grad_norm": 2.520368814468384, + "learning_rate": 4.525547445255475e-06, + "loss": 2.3174, + "step": 262 + }, + { + "epoch": 0.2882191780821918, + "grad_norm": 2.550347328186035, + "learning_rate": 4.543795620437956e-06, + "loss": 2.3072, + "step": 263 + }, + { + "epoch": 0.2893150684931507, + "grad_norm": 2.268662691116333, + "learning_rate": 4.562043795620438e-06, + "loss": 2.2763, + "step": 264 + }, + { + "epoch": 0.29041095890410956, + "grad_norm": 2.445798873901367, + "learning_rate": 4.580291970802921e-06, + "loss": 2.3395, + "step": 265 + }, + { + "epoch": 0.2915068493150685, + "grad_norm": 2.5391135215759277, + "learning_rate": 4.598540145985402e-06, + "loss": 2.3496, + "step": 266 + }, + { + "epoch": 0.2926027397260274, + "grad_norm": 2.3553097248077393, + "learning_rate": 4.6167883211678835e-06, + "loss": 2.3243, + "step": 267 + }, + { + "epoch": 0.2936986301369863, + "grad_norm": 2.448089599609375, + "learning_rate": 4.635036496350365e-06, + "loss": 2.2887, + "step": 268 + }, + { + "epoch": 0.2947945205479452, + "grad_norm": 2.1600170135498047, + "learning_rate": 4.653284671532847e-06, + "loss": 2.2816, + "step": 269 + }, + { + "epoch": 0.2958904109589041, + "grad_norm": 2.3783180713653564, + "learning_rate": 4.671532846715329e-06, + "loss": 2.3275, + "step": 270 + }, + { + "epoch": 0.2958904109589041, + "eval_loss": 2.413832902908325, + "eval_runtime": 48.7524, + "eval_samples_per_second": 20.512, + "eval_steps_per_second": 0.861, + "step": 270 + }, + { + "epoch": 0.29698630136986304, + "grad_norm": 2.468320369720459, + "learning_rate": 4.689781021897811e-06, + "loss": 2.3358, + "step": 271 + }, + { + "epoch": 0.2980821917808219, + "grad_norm": 2.401634454727173, + "learning_rate": 4.708029197080292e-06, + "loss": 2.2991, + "step": 272 + }, + { + "epoch": 0.2991780821917808, + "grad_norm": 2.4604203701019287, + "learning_rate": 4.726277372262774e-06, + "loss": 2.3496, + "step": 273 + }, + { + "epoch": 0.30027397260273975, + "grad_norm": 2.0338871479034424, + "learning_rate": 4.744525547445255e-06, + "loss": 2.3115, + "step": 274 + }, + { + "epoch": 0.3013698630136986, + "grad_norm": 2.5404436588287354, + "learning_rate": 4.7627737226277375e-06, + "loss": 2.2665, + "step": 275 + }, + { + "epoch": 0.30246575342465754, + "grad_norm": 2.5369458198547363, + "learning_rate": 4.78102189781022e-06, + "loss": 2.3072, + "step": 276 + }, + { + "epoch": 0.30356164383561646, + "grad_norm": 2.471522569656372, + "learning_rate": 4.799270072992701e-06, + "loss": 2.3113, + "step": 277 + }, + { + "epoch": 0.3046575342465753, + "grad_norm": 2.3890135288238525, + "learning_rate": 4.8175182481751835e-06, + "loss": 2.2838, + "step": 278 + }, + { + "epoch": 0.30575342465753425, + "grad_norm": 2.2485053539276123, + "learning_rate": 4.835766423357665e-06, + "loss": 2.3052, + "step": 279 + }, + { + "epoch": 0.30684931506849317, + "grad_norm": 2.1090753078460693, + "learning_rate": 4.854014598540146e-06, + "loss": 2.3054, + "step": 280 + }, + { + "epoch": 0.30684931506849317, + "eval_loss": 2.412548780441284, + "eval_runtime": 48.5034, + "eval_samples_per_second": 20.617, + "eval_steps_per_second": 0.866, + "step": 280 + }, + { + "epoch": 0.30794520547945203, + "grad_norm": 2.299133777618408, + "learning_rate": 4.872262773722628e-06, + "loss": 2.2599, + "step": 281 + }, + { + "epoch": 0.30904109589041096, + "grad_norm": 2.177905797958374, + "learning_rate": 4.89051094890511e-06, + "loss": 2.2745, + "step": 282 + }, + { + "epoch": 0.3101369863013699, + "grad_norm": 2.3178462982177734, + "learning_rate": 4.9087591240875915e-06, + "loss": 2.2562, + "step": 283 + }, + { + "epoch": 0.31123287671232874, + "grad_norm": 2.5088376998901367, + "learning_rate": 4.927007299270074e-06, + "loss": 2.2945, + "step": 284 + }, + { + "epoch": 0.31232876712328766, + "grad_norm": 2.095379590988159, + "learning_rate": 4.945255474452555e-06, + "loss": 2.3038, + "step": 285 + }, + { + "epoch": 0.3134246575342466, + "grad_norm": 2.1792995929718018, + "learning_rate": 4.963503649635037e-06, + "loss": 2.2702, + "step": 286 + }, + { + "epoch": 0.31452054794520545, + "grad_norm": 2.3001186847686768, + "learning_rate": 4.981751824817518e-06, + "loss": 2.3482, + "step": 287 + }, + { + "epoch": 0.3156164383561644, + "grad_norm": 2.1576106548309326, + "learning_rate": 5e-06, + "loss": 2.3391, + "step": 288 + }, + { + "epoch": 0.3167123287671233, + "grad_norm": 2.3715577125549316, + "learning_rate": 4.997969130787978e-06, + "loss": 2.3049, + "step": 289 + }, + { + "epoch": 0.3178082191780822, + "grad_norm": 2.0316715240478516, + "learning_rate": 4.995938261575955e-06, + "loss": 2.3111, + "step": 290 + }, + { + "epoch": 0.3178082191780822, + "eval_loss": 2.40299654006958, + "eval_runtime": 50.0873, + "eval_samples_per_second": 19.965, + "eval_steps_per_second": 0.839, + "step": 290 + }, + { + "epoch": 0.3189041095890411, + "grad_norm": 2.1135756969451904, + "learning_rate": 4.993907392363932e-06, + "loss": 2.2465, + "step": 291 + }, + { + "epoch": 0.32, + "grad_norm": 2.335655927658081, + "learning_rate": 4.9918765231519095e-06, + "loss": 2.31, + "step": 292 + }, + { + "epoch": 0.32109589041095893, + "grad_norm": 2.3255794048309326, + "learning_rate": 4.989845653939887e-06, + "loss": 2.2531, + "step": 293 + }, + { + "epoch": 0.3221917808219178, + "grad_norm": 2.111055612564087, + "learning_rate": 4.987814784727864e-06, + "loss": 2.2799, + "step": 294 + }, + { + "epoch": 0.3232876712328767, + "grad_norm": 2.1773171424865723, + "learning_rate": 4.985783915515841e-06, + "loss": 2.2884, + "step": 295 + }, + { + "epoch": 0.32438356164383564, + "grad_norm": 2.3559885025024414, + "learning_rate": 4.983753046303819e-06, + "loss": 2.2711, + "step": 296 + }, + { + "epoch": 0.3254794520547945, + "grad_norm": 2.120802879333496, + "learning_rate": 4.981722177091796e-06, + "loss": 2.2946, + "step": 297 + }, + { + "epoch": 0.3265753424657534, + "grad_norm": 2.2319388389587402, + "learning_rate": 4.979691307879773e-06, + "loss": 2.2771, + "step": 298 + }, + { + "epoch": 0.32767123287671235, + "grad_norm": 2.208740711212158, + "learning_rate": 4.9776604386677504e-06, + "loss": 2.3003, + "step": 299 + }, + { + "epoch": 0.3287671232876712, + "grad_norm": 2.543060541152954, + "learning_rate": 4.975629569455728e-06, + "loss": 2.2891, + "step": 300 + }, + { + "epoch": 0.3287671232876712, + "eval_loss": 2.398495674133301, + "eval_runtime": 51.1973, + "eval_samples_per_second": 19.532, + "eval_steps_per_second": 0.82, + "step": 300 + }, + { + "epoch": 0.32986301369863014, + "grad_norm": 2.086219072341919, + "learning_rate": 4.973598700243705e-06, + "loss": 2.3286, + "step": 301 + }, + { + "epoch": 0.33095890410958906, + "grad_norm": 2.3673319816589355, + "learning_rate": 4.971567831031681e-06, + "loss": 2.2672, + "step": 302 + }, + { + "epoch": 0.3320547945205479, + "grad_norm": 2.167593240737915, + "learning_rate": 4.9695369618196595e-06, + "loss": 2.264, + "step": 303 + }, + { + "epoch": 0.33315068493150685, + "grad_norm": 2.7218785285949707, + "learning_rate": 4.967506092607636e-06, + "loss": 2.1737, + "step": 304 + }, + { + "epoch": 0.33424657534246577, + "grad_norm": 2.673245429992676, + "learning_rate": 4.965475223395614e-06, + "loss": 2.2439, + "step": 305 + }, + { + "epoch": 0.33534246575342463, + "grad_norm": 2.0341989994049072, + "learning_rate": 4.9634443541835905e-06, + "loss": 2.2614, + "step": 306 + }, + { + "epoch": 0.33643835616438356, + "grad_norm": 1.9314247369766235, + "learning_rate": 4.961413484971569e-06, + "loss": 2.2592, + "step": 307 + }, + { + "epoch": 0.3375342465753425, + "grad_norm": 2.537794589996338, + "learning_rate": 4.959382615759545e-06, + "loss": 2.2859, + "step": 308 + }, + { + "epoch": 0.33863013698630134, + "grad_norm": 2.0811166763305664, + "learning_rate": 4.957351746547523e-06, + "loss": 2.2868, + "step": 309 + }, + { + "epoch": 0.33972602739726027, + "grad_norm": 2.783701181411743, + "learning_rate": 4.9553208773355e-06, + "loss": 2.2649, + "step": 310 + }, + { + "epoch": 0.33972602739726027, + "eval_loss": 2.400538921356201, + "eval_runtime": 43.185, + "eval_samples_per_second": 23.156, + "eval_steps_per_second": 0.973, + "step": 310 + }, + { + "epoch": 0.3408219178082192, + "grad_norm": 2.8541319370269775, + "learning_rate": 4.953290008123478e-06, + "loss": 2.3729, + "step": 311 + }, + { + "epoch": 0.3419178082191781, + "grad_norm": 2.1852617263793945, + "learning_rate": 4.951259138911454e-06, + "loss": 2.2885, + "step": 312 + }, + { + "epoch": 0.343013698630137, + "grad_norm": 2.872201919555664, + "learning_rate": 4.9492282696994315e-06, + "loss": 2.2353, + "step": 313 + }, + { + "epoch": 0.3441095890410959, + "grad_norm": 2.5140910148620605, + "learning_rate": 4.947197400487409e-06, + "loss": 2.3364, + "step": 314 + }, + { + "epoch": 0.3452054794520548, + "grad_norm": 2.619974374771118, + "learning_rate": 4.945166531275386e-06, + "loss": 2.2721, + "step": 315 + }, + { + "epoch": 0.3463013698630137, + "grad_norm": 2.4567618370056152, + "learning_rate": 4.943135662063363e-06, + "loss": 2.26, + "step": 316 + }, + { + "epoch": 0.3473972602739726, + "grad_norm": 2.9603192806243896, + "learning_rate": 4.9411047928513406e-06, + "loss": 2.2861, + "step": 317 + }, + { + "epoch": 0.34849315068493153, + "grad_norm": 2.115262985229492, + "learning_rate": 4.939073923639318e-06, + "loss": 2.3125, + "step": 318 + }, + { + "epoch": 0.3495890410958904, + "grad_norm": 2.693024158477783, + "learning_rate": 4.937043054427295e-06, + "loss": 2.3054, + "step": 319 + }, + { + "epoch": 0.3506849315068493, + "grad_norm": 2.3439996242523193, + "learning_rate": 4.935012185215272e-06, + "loss": 2.2576, + "step": 320 + }, + { + "epoch": 0.3506849315068493, + "eval_loss": 2.3920650482177734, + "eval_runtime": 47.7451, + "eval_samples_per_second": 20.945, + "eval_steps_per_second": 0.88, + "step": 320 + }, + { + "epoch": 0.35178082191780824, + "grad_norm": 2.3009634017944336, + "learning_rate": 4.93298131600325e-06, + "loss": 2.2454, + "step": 321 + }, + { + "epoch": 0.3528767123287671, + "grad_norm": 2.593203544616699, + "learning_rate": 4.930950446791227e-06, + "loss": 2.2965, + "step": 322 + }, + { + "epoch": 0.353972602739726, + "grad_norm": 2.9069650173187256, + "learning_rate": 4.928919577579204e-06, + "loss": 2.2785, + "step": 323 + }, + { + "epoch": 0.35506849315068495, + "grad_norm": 2.1021759510040283, + "learning_rate": 4.9268887083671815e-06, + "loss": 2.2857, + "step": 324 + }, + { + "epoch": 0.3561643835616438, + "grad_norm": 2.0330185890197754, + "learning_rate": 4.924857839155159e-06, + "loss": 2.2697, + "step": 325 + }, + { + "epoch": 0.35726027397260274, + "grad_norm": 2.1318063735961914, + "learning_rate": 4.922826969943136e-06, + "loss": 2.2698, + "step": 326 + }, + { + "epoch": 0.35835616438356166, + "grad_norm": 2.0157387256622314, + "learning_rate": 4.920796100731113e-06, + "loss": 2.292, + "step": 327 + }, + { + "epoch": 0.3594520547945205, + "grad_norm": 2.3266026973724365, + "learning_rate": 4.918765231519091e-06, + "loss": 2.2215, + "step": 328 + }, + { + "epoch": 0.36054794520547945, + "grad_norm": 2.066305160522461, + "learning_rate": 4.916734362307068e-06, + "loss": 2.2758, + "step": 329 + }, + { + "epoch": 0.36164383561643837, + "grad_norm": 2.0219993591308594, + "learning_rate": 4.914703493095045e-06, + "loss": 2.2381, + "step": 330 + }, + { + "epoch": 0.36164383561643837, + "eval_loss": 2.394064426422119, + "eval_runtime": 51.5482, + "eval_samples_per_second": 19.399, + "eval_steps_per_second": 0.815, + "step": 330 + }, + { + "epoch": 0.36273972602739724, + "grad_norm": 2.3964319229125977, + "learning_rate": 4.912672623883022e-06, + "loss": 2.2384, + "step": 331 + }, + { + "epoch": 0.36383561643835616, + "grad_norm": 2.5466530323028564, + "learning_rate": 4.910641754671e-06, + "loss": 2.2428, + "step": 332 + }, + { + "epoch": 0.3649315068493151, + "grad_norm": 2.236152410507202, + "learning_rate": 4.908610885458977e-06, + "loss": 2.2639, + "step": 333 + }, + { + "epoch": 0.36602739726027395, + "grad_norm": 2.7267303466796875, + "learning_rate": 4.906580016246954e-06, + "loss": 2.2608, + "step": 334 + }, + { + "epoch": 0.36712328767123287, + "grad_norm": 1.9403765201568604, + "learning_rate": 4.9045491470349315e-06, + "loss": 2.2962, + "step": 335 + }, + { + "epoch": 0.3682191780821918, + "grad_norm": 2.89070463180542, + "learning_rate": 4.902518277822909e-06, + "loss": 2.2847, + "step": 336 + }, + { + "epoch": 0.3693150684931507, + "grad_norm": 2.6427228450775146, + "learning_rate": 4.900487408610886e-06, + "loss": 2.2734, + "step": 337 + }, + { + "epoch": 0.3704109589041096, + "grad_norm": 2.1680362224578857, + "learning_rate": 4.898456539398863e-06, + "loss": 2.2869, + "step": 338 + }, + { + "epoch": 0.3715068493150685, + "grad_norm": 3.0160160064697266, + "learning_rate": 4.89642567018684e-06, + "loss": 2.3043, + "step": 339 + }, + { + "epoch": 0.3726027397260274, + "grad_norm": 1.9858418703079224, + "learning_rate": 4.894394800974818e-06, + "loss": 2.3422, + "step": 340 + }, + { + "epoch": 0.3726027397260274, + "eval_loss": 2.387559652328491, + "eval_runtime": 52.2455, + "eval_samples_per_second": 19.14, + "eval_steps_per_second": 0.804, + "step": 340 + }, + { + "epoch": 0.3736986301369863, + "grad_norm": 2.243699073791504, + "learning_rate": 4.892363931762794e-06, + "loss": 2.2394, + "step": 341 + }, + { + "epoch": 0.3747945205479452, + "grad_norm": 2.371299982070923, + "learning_rate": 4.8903330625507725e-06, + "loss": 2.2856, + "step": 342 + }, + { + "epoch": 0.37589041095890413, + "grad_norm": 2.3769736289978027, + "learning_rate": 4.888302193338749e-06, + "loss": 2.246, + "step": 343 + }, + { + "epoch": 0.376986301369863, + "grad_norm": 2.3209030628204346, + "learning_rate": 4.886271324126727e-06, + "loss": 2.2331, + "step": 344 + }, + { + "epoch": 0.3780821917808219, + "grad_norm": 2.747319221496582, + "learning_rate": 4.8842404549147034e-06, + "loss": 2.3115, + "step": 345 + }, + { + "epoch": 0.37917808219178084, + "grad_norm": 1.8599231243133545, + "learning_rate": 4.8822095857026816e-06, + "loss": 2.2994, + "step": 346 + }, + { + "epoch": 0.3802739726027397, + "grad_norm": 2.0358943939208984, + "learning_rate": 4.880178716490658e-06, + "loss": 2.271, + "step": 347 + }, + { + "epoch": 0.38136986301369863, + "grad_norm": 2.409106731414795, + "learning_rate": 4.878147847278636e-06, + "loss": 2.2281, + "step": 348 + }, + { + "epoch": 0.38246575342465755, + "grad_norm": 2.1641807556152344, + "learning_rate": 4.8761169780666125e-06, + "loss": 2.2437, + "step": 349 + }, + { + "epoch": 0.3835616438356164, + "grad_norm": 2.6917359828948975, + "learning_rate": 4.874086108854591e-06, + "loss": 2.226, + "step": 350 + }, + { + "epoch": 0.3835616438356164, + "eval_loss": 2.382991313934326, + "eval_runtime": 48.9534, + "eval_samples_per_second": 20.428, + "eval_steps_per_second": 0.858, + "step": 350 + }, + { + "epoch": 0.38465753424657534, + "grad_norm": 2.364215612411499, + "learning_rate": 4.872055239642567e-06, + "loss": 2.2319, + "step": 351 + }, + { + "epoch": 0.38575342465753426, + "grad_norm": 2.317643165588379, + "learning_rate": 4.870024370430544e-06, + "loss": 2.2096, + "step": 352 + }, + { + "epoch": 0.3868493150684931, + "grad_norm": 2.2283246517181396, + "learning_rate": 4.867993501218522e-06, + "loss": 2.2634, + "step": 353 + }, + { + "epoch": 0.38794520547945205, + "grad_norm": 2.498391628265381, + "learning_rate": 4.865962632006499e-06, + "loss": 2.2813, + "step": 354 + }, + { + "epoch": 0.38904109589041097, + "grad_norm": 2.3889403343200684, + "learning_rate": 4.863931762794476e-06, + "loss": 2.2361, + "step": 355 + }, + { + "epoch": 0.39013698630136984, + "grad_norm": 2.2049975395202637, + "learning_rate": 4.8619008935824535e-06, + "loss": 2.2583, + "step": 356 + }, + { + "epoch": 0.39123287671232876, + "grad_norm": 2.132840871810913, + "learning_rate": 4.859870024370431e-06, + "loss": 2.2823, + "step": 357 + }, + { + "epoch": 0.3923287671232877, + "grad_norm": 2.2692184448242188, + "learning_rate": 4.857839155158408e-06, + "loss": 2.2939, + "step": 358 + }, + { + "epoch": 0.3934246575342466, + "grad_norm": 2.0501933097839355, + "learning_rate": 4.855808285946385e-06, + "loss": 2.2976, + "step": 359 + }, + { + "epoch": 0.39452054794520547, + "grad_norm": 2.170236587524414, + "learning_rate": 4.8537774167343626e-06, + "loss": 2.2748, + "step": 360 + }, + { + "epoch": 0.39452054794520547, + "eval_loss": 2.3783886432647705, + "eval_runtime": 46.7324, + "eval_samples_per_second": 21.398, + "eval_steps_per_second": 0.899, + "step": 360 + }, + { + "epoch": 0.3956164383561644, + "grad_norm": 2.283137083053589, + "learning_rate": 4.85174654752234e-06, + "loss": 2.2705, + "step": 361 + }, + { + "epoch": 0.3967123287671233, + "grad_norm": 2.195984363555908, + "learning_rate": 4.849715678310317e-06, + "loss": 2.26, + "step": 362 + }, + { + "epoch": 0.3978082191780822, + "grad_norm": 2.399935007095337, + "learning_rate": 4.847684809098294e-06, + "loss": 2.2639, + "step": 363 + }, + { + "epoch": 0.3989041095890411, + "grad_norm": 2.411280393600464, + "learning_rate": 4.845653939886272e-06, + "loss": 2.2523, + "step": 364 + }, + { + "epoch": 0.4, + "grad_norm": 2.2033896446228027, + "learning_rate": 4.843623070674249e-06, + "loss": 2.2722, + "step": 365 + }, + { + "epoch": 0.4010958904109589, + "grad_norm": 2.2990238666534424, + "learning_rate": 4.841592201462226e-06, + "loss": 2.2411, + "step": 366 + }, + { + "epoch": 0.4021917808219178, + "grad_norm": 1.911453127861023, + "learning_rate": 4.8395613322502035e-06, + "loss": 2.2921, + "step": 367 + }, + { + "epoch": 0.40328767123287673, + "grad_norm": 2.453195095062256, + "learning_rate": 4.837530463038181e-06, + "loss": 2.218, + "step": 368 + }, + { + "epoch": 0.4043835616438356, + "grad_norm": 2.126992702484131, + "learning_rate": 4.835499593826158e-06, + "loss": 2.2628, + "step": 369 + }, + { + "epoch": 0.4054794520547945, + "grad_norm": 1.9305835962295532, + "learning_rate": 4.833468724614135e-06, + "loss": 2.2726, + "step": 370 + }, + { + "epoch": 0.4054794520547945, + "eval_loss": 2.3741114139556885, + "eval_runtime": 49.2956, + "eval_samples_per_second": 20.286, + "eval_steps_per_second": 0.852, + "step": 370 + }, + { + "epoch": 0.40657534246575344, + "grad_norm": 2.222024917602539, + "learning_rate": 4.831437855402113e-06, + "loss": 2.2634, + "step": 371 + }, + { + "epoch": 0.4076712328767123, + "grad_norm": 2.473958969116211, + "learning_rate": 4.82940698619009e-06, + "loss": 2.1817, + "step": 372 + }, + { + "epoch": 0.40876712328767123, + "grad_norm": 1.9208838939666748, + "learning_rate": 4.827376116978067e-06, + "loss": 2.2353, + "step": 373 + }, + { + "epoch": 0.40986301369863015, + "grad_norm": 2.7226450443267822, + "learning_rate": 4.8253452477660444e-06, + "loss": 2.2831, + "step": 374 + }, + { + "epoch": 0.410958904109589, + "grad_norm": 1.9078222513198853, + "learning_rate": 4.823314378554022e-06, + "loss": 2.264, + "step": 375 + }, + { + "epoch": 0.41205479452054794, + "grad_norm": 2.3946311473846436, + "learning_rate": 4.821283509341998e-06, + "loss": 2.251, + "step": 376 + }, + { + "epoch": 0.41315068493150686, + "grad_norm": 1.847343921661377, + "learning_rate": 4.819252640129976e-06, + "loss": 2.2441, + "step": 377 + }, + { + "epoch": 0.41424657534246573, + "grad_norm": 2.475385904312134, + "learning_rate": 4.817221770917953e-06, + "loss": 2.2639, + "step": 378 + }, + { + "epoch": 0.41534246575342465, + "grad_norm": 2.1396560668945312, + "learning_rate": 4.815190901705931e-06, + "loss": 2.2353, + "step": 379 + }, + { + "epoch": 0.41643835616438357, + "grad_norm": 2.3868885040283203, + "learning_rate": 4.813160032493907e-06, + "loss": 2.2551, + "step": 380 + }, + { + "epoch": 0.41643835616438357, + "eval_loss": 2.369168519973755, + "eval_runtime": 45.6131, + "eval_samples_per_second": 21.924, + "eval_steps_per_second": 0.921, + "step": 380 + }, + { + "epoch": 0.41753424657534244, + "grad_norm": 2.0193872451782227, + "learning_rate": 4.811129163281885e-06, + "loss": 2.2399, + "step": 381 + }, + { + "epoch": 0.41863013698630136, + "grad_norm": 2.4655935764312744, + "learning_rate": 4.809098294069862e-06, + "loss": 2.2729, + "step": 382 + }, + { + "epoch": 0.4197260273972603, + "grad_norm": 1.880974292755127, + "learning_rate": 4.80706742485784e-06, + "loss": 2.2681, + "step": 383 + }, + { + "epoch": 0.4208219178082192, + "grad_norm": 2.1912078857421875, + "learning_rate": 4.805036555645816e-06, + "loss": 2.2961, + "step": 384 + }, + { + "epoch": 0.42191780821917807, + "grad_norm": 1.985931634902954, + "learning_rate": 4.8030056864337945e-06, + "loss": 2.2407, + "step": 385 + }, + { + "epoch": 0.423013698630137, + "grad_norm": 2.4353220462799072, + "learning_rate": 4.800974817221771e-06, + "loss": 2.2867, + "step": 386 + }, + { + "epoch": 0.4241095890410959, + "grad_norm": 1.8810508251190186, + "learning_rate": 4.798943948009749e-06, + "loss": 2.2498, + "step": 387 + }, + { + "epoch": 0.4252054794520548, + "grad_norm": 2.195033073425293, + "learning_rate": 4.7969130787977254e-06, + "loss": 2.2532, + "step": 388 + }, + { + "epoch": 0.4263013698630137, + "grad_norm": 1.855372428894043, + "learning_rate": 4.794882209585703e-06, + "loss": 2.2538, + "step": 389 + }, + { + "epoch": 0.4273972602739726, + "grad_norm": 2.4465434551239014, + "learning_rate": 4.79285134037368e-06, + "loss": 2.2724, + "step": 390 + }, + { + "epoch": 0.4273972602739726, + "eval_loss": 2.364928960800171, + "eval_runtime": 48.0245, + "eval_samples_per_second": 20.823, + "eval_steps_per_second": 0.875, + "step": 390 + }, + { + "epoch": 0.4284931506849315, + "grad_norm": 1.8911314010620117, + "learning_rate": 4.790820471161657e-06, + "loss": 2.2291, + "step": 391 + }, + { + "epoch": 0.4295890410958904, + "grad_norm": 2.1975977420806885, + "learning_rate": 4.7887896019496345e-06, + "loss": 2.2373, + "step": 392 + }, + { + "epoch": 0.43068493150684933, + "grad_norm": 2.015331506729126, + "learning_rate": 4.786758732737612e-06, + "loss": 2.2483, + "step": 393 + }, + { + "epoch": 0.4317808219178082, + "grad_norm": 2.039339065551758, + "learning_rate": 4.784727863525589e-06, + "loss": 2.2746, + "step": 394 + }, + { + "epoch": 0.4328767123287671, + "grad_norm": 1.977513074874878, + "learning_rate": 4.782696994313566e-06, + "loss": 2.2275, + "step": 395 + }, + { + "epoch": 0.43397260273972604, + "grad_norm": 2.147217035293579, + "learning_rate": 4.780666125101544e-06, + "loss": 2.252, + "step": 396 + }, + { + "epoch": 0.4350684931506849, + "grad_norm": 2.1799919605255127, + "learning_rate": 4.778635255889521e-06, + "loss": 2.1975, + "step": 397 + }, + { + "epoch": 0.43616438356164383, + "grad_norm": 2.582240343093872, + "learning_rate": 4.776604386677498e-06, + "loss": 2.2294, + "step": 398 + }, + { + "epoch": 0.43726027397260275, + "grad_norm": 1.9127846956253052, + "learning_rate": 4.7745735174654755e-06, + "loss": 2.2319, + "step": 399 + }, + { + "epoch": 0.4383561643835616, + "grad_norm": 2.2444543838500977, + "learning_rate": 4.772542648253453e-06, + "loss": 2.2857, + "step": 400 + }, + { + "epoch": 0.4383561643835616, + "eval_loss": 2.3653724193573, + "eval_runtime": 43.8304, + "eval_samples_per_second": 22.815, + "eval_steps_per_second": 0.958, + "step": 400 + }, + { + "epoch": 0.43945205479452054, + "grad_norm": 1.9913676977157593, + "learning_rate": 4.77051177904143e-06, + "loss": 2.274, + "step": 401 + }, + { + "epoch": 0.44054794520547946, + "grad_norm": 2.2850067615509033, + "learning_rate": 4.768480909829407e-06, + "loss": 2.2209, + "step": 402 + }, + { + "epoch": 0.44164383561643833, + "grad_norm": 2.4984688758850098, + "learning_rate": 4.766450040617385e-06, + "loss": 2.263, + "step": 403 + }, + { + "epoch": 0.44273972602739725, + "grad_norm": 2.3244640827178955, + "learning_rate": 4.764419171405362e-06, + "loss": 2.2368, + "step": 404 + }, + { + "epoch": 0.4438356164383562, + "grad_norm": 2.7579681873321533, + "learning_rate": 4.762388302193339e-06, + "loss": 2.2762, + "step": 405 + }, + { + "epoch": 0.4449315068493151, + "grad_norm": 2.1996467113494873, + "learning_rate": 4.760357432981316e-06, + "loss": 2.2536, + "step": 406 + }, + { + "epoch": 0.44602739726027396, + "grad_norm": 2.3767571449279785, + "learning_rate": 4.758326563769294e-06, + "loss": 2.2353, + "step": 407 + }, + { + "epoch": 0.4471232876712329, + "grad_norm": 1.9148359298706055, + "learning_rate": 4.756295694557271e-06, + "loss": 2.2416, + "step": 408 + }, + { + "epoch": 0.4482191780821918, + "grad_norm": 2.6499507427215576, + "learning_rate": 4.754264825345248e-06, + "loss": 2.2641, + "step": 409 + }, + { + "epoch": 0.44931506849315067, + "grad_norm": 1.9890286922454834, + "learning_rate": 4.7522339561332255e-06, + "loss": 2.2148, + "step": 410 + }, + { + "epoch": 0.44931506849315067, + "eval_loss": 2.3670520782470703, + "eval_runtime": 41.1722, + "eval_samples_per_second": 24.288, + "eval_steps_per_second": 1.02, + "step": 410 + }, + { + "epoch": 0.4504109589041096, + "grad_norm": 2.575968027114868, + "learning_rate": 4.750203086921203e-06, + "loss": 2.2254, + "step": 411 + }, + { + "epoch": 0.4515068493150685, + "grad_norm": 2.015174388885498, + "learning_rate": 4.74817221770918e-06, + "loss": 2.1881, + "step": 412 + }, + { + "epoch": 0.4526027397260274, + "grad_norm": 2.44292950630188, + "learning_rate": 4.7461413484971565e-06, + "loss": 2.256, + "step": 413 + }, + { + "epoch": 0.4536986301369863, + "grad_norm": 1.9808441400527954, + "learning_rate": 4.744110479285135e-06, + "loss": 2.285, + "step": 414 + }, + { + "epoch": 0.4547945205479452, + "grad_norm": 2.009481906890869, + "learning_rate": 4.742079610073111e-06, + "loss": 2.2018, + "step": 415 + }, + { + "epoch": 0.4558904109589041, + "grad_norm": 2.1466031074523926, + "learning_rate": 4.740048740861089e-06, + "loss": 2.224, + "step": 416 + }, + { + "epoch": 0.456986301369863, + "grad_norm": 2.1337907314300537, + "learning_rate": 4.738017871649066e-06, + "loss": 2.2404, + "step": 417 + }, + { + "epoch": 0.45808219178082193, + "grad_norm": 1.94125235080719, + "learning_rate": 4.735987002437044e-06, + "loss": 2.2474, + "step": 418 + }, + { + "epoch": 0.4591780821917808, + "grad_norm": 2.379099130630493, + "learning_rate": 4.73395613322502e-06, + "loss": 2.2858, + "step": 419 + }, + { + "epoch": 0.4602739726027397, + "grad_norm": 1.844056248664856, + "learning_rate": 4.731925264012998e-06, + "loss": 2.2967, + "step": 420 + }, + { + "epoch": 0.4602739726027397, + "eval_loss": 2.3622241020202637, + "eval_runtime": 45.4674, + "eval_samples_per_second": 21.994, + "eval_steps_per_second": 0.924, + "step": 420 + }, + { + "epoch": 0.46136986301369864, + "grad_norm": 2.411895513534546, + "learning_rate": 4.729894394800975e-06, + "loss": 2.274, + "step": 421 + }, + { + "epoch": 0.4624657534246575, + "grad_norm": 2.014885425567627, + "learning_rate": 4.727863525588953e-06, + "loss": 2.2144, + "step": 422 + }, + { + "epoch": 0.46356164383561643, + "grad_norm": 2.112009048461914, + "learning_rate": 4.725832656376929e-06, + "loss": 2.1824, + "step": 423 + }, + { + "epoch": 0.46465753424657535, + "grad_norm": 2.111392021179199, + "learning_rate": 4.723801787164907e-06, + "loss": 2.2261, + "step": 424 + }, + { + "epoch": 0.4657534246575342, + "grad_norm": 2.0349185466766357, + "learning_rate": 4.721770917952885e-06, + "loss": 2.2622, + "step": 425 + }, + { + "epoch": 0.46684931506849314, + "grad_norm": 2.30222487449646, + "learning_rate": 4.719740048740861e-06, + "loss": 2.2206, + "step": 426 + }, + { + "epoch": 0.46794520547945206, + "grad_norm": 2.018789529800415, + "learning_rate": 4.717709179528839e-06, + "loss": 2.2196, + "step": 427 + }, + { + "epoch": 0.469041095890411, + "grad_norm": 2.046649932861328, + "learning_rate": 4.715678310316816e-06, + "loss": 2.2584, + "step": 428 + }, + { + "epoch": 0.47013698630136985, + "grad_norm": 2.0393869876861572, + "learning_rate": 4.713647441104794e-06, + "loss": 2.2537, + "step": 429 + }, + { + "epoch": 0.4712328767123288, + "grad_norm": 2.2788116931915283, + "learning_rate": 4.71161657189277e-06, + "loss": 2.2145, + "step": 430 + }, + { + "epoch": 0.4712328767123288, + "eval_loss": 2.3575785160064697, + "eval_runtime": 44.6687, + "eval_samples_per_second": 22.387, + "eval_steps_per_second": 0.94, + "step": 430 + }, + { + "epoch": 0.4723287671232877, + "grad_norm": 1.9950047731399536, + "learning_rate": 4.709585702680748e-06, + "loss": 2.2072, + "step": 431 + }, + { + "epoch": 0.47342465753424656, + "grad_norm": 1.9222451448440552, + "learning_rate": 4.707554833468725e-06, + "loss": 2.2654, + "step": 432 + }, + { + "epoch": 0.4745205479452055, + "grad_norm": 1.9735718965530396, + "learning_rate": 4.705523964256703e-06, + "loss": 2.3037, + "step": 433 + }, + { + "epoch": 0.4756164383561644, + "grad_norm": 2.3209776878356934, + "learning_rate": 4.703493095044679e-06, + "loss": 2.2083, + "step": 434 + }, + { + "epoch": 0.4767123287671233, + "grad_norm": 2.0993995666503906, + "learning_rate": 4.701462225832657e-06, + "loss": 2.2275, + "step": 435 + }, + { + "epoch": 0.4778082191780822, + "grad_norm": 2.334463119506836, + "learning_rate": 4.699431356620634e-06, + "loss": 2.1852, + "step": 436 + }, + { + "epoch": 0.4789041095890411, + "grad_norm": 1.9098213911056519, + "learning_rate": 4.697400487408612e-06, + "loss": 2.2829, + "step": 437 + }, + { + "epoch": 0.48, + "grad_norm": 2.3005056381225586, + "learning_rate": 4.695369618196588e-06, + "loss": 2.2532, + "step": 438 + }, + { + "epoch": 0.4810958904109589, + "grad_norm": 1.9069781303405762, + "learning_rate": 4.693338748984566e-06, + "loss": 2.2374, + "step": 439 + }, + { + "epoch": 0.4821917808219178, + "grad_norm": 2.050032138824463, + "learning_rate": 4.691307879772543e-06, + "loss": 2.2256, + "step": 440 + }, + { + "epoch": 0.4821917808219178, + "eval_loss": 2.3587422370910645, + "eval_runtime": 45.0941, + "eval_samples_per_second": 22.176, + "eval_steps_per_second": 0.931, + "step": 440 + }, + { + "epoch": 0.4832876712328767, + "grad_norm": 2.0563836097717285, + "learning_rate": 4.68927701056052e-06, + "loss": 2.1918, + "step": 441 + }, + { + "epoch": 0.4843835616438356, + "grad_norm": 2.1598386764526367, + "learning_rate": 4.6872461413484975e-06, + "loss": 2.2618, + "step": 442 + }, + { + "epoch": 0.48547945205479454, + "grad_norm": 1.9236539602279663, + "learning_rate": 4.685215272136475e-06, + "loss": 2.2477, + "step": 443 + }, + { + "epoch": 0.4865753424657534, + "grad_norm": 1.9923924207687378, + "learning_rate": 4.683184402924452e-06, + "loss": 2.2313, + "step": 444 + }, + { + "epoch": 0.4876712328767123, + "grad_norm": 2.162627696990967, + "learning_rate": 4.681153533712429e-06, + "loss": 2.2367, + "step": 445 + }, + { + "epoch": 0.48876712328767125, + "grad_norm": 2.083523988723755, + "learning_rate": 4.679122664500407e-06, + "loss": 2.2946, + "step": 446 + }, + { + "epoch": 0.4898630136986301, + "grad_norm": 2.504408359527588, + "learning_rate": 4.677091795288384e-06, + "loss": 2.2104, + "step": 447 + }, + { + "epoch": 0.49095890410958903, + "grad_norm": 1.8316532373428345, + "learning_rate": 4.675060926076361e-06, + "loss": 2.2027, + "step": 448 + }, + { + "epoch": 0.49205479452054796, + "grad_norm": 2.3337953090667725, + "learning_rate": 4.673030056864338e-06, + "loss": 2.278, + "step": 449 + }, + { + "epoch": 0.4931506849315068, + "grad_norm": 2.24102783203125, + "learning_rate": 4.670999187652316e-06, + "loss": 2.2018, + "step": 450 + }, + { + "epoch": 0.4931506849315068, + "eval_loss": 2.3545279502868652, + "eval_runtime": 45.1041, + "eval_samples_per_second": 22.171, + "eval_steps_per_second": 0.931, + "step": 450 + }, + { + "epoch": 0.49424657534246574, + "grad_norm": 2.07834792137146, + "learning_rate": 4.668968318440293e-06, + "loss": 2.1975, + "step": 451 + }, + { + "epoch": 0.49534246575342467, + "grad_norm": 1.7681723833084106, + "learning_rate": 4.66693744922827e-06, + "loss": 2.2033, + "step": 452 + }, + { + "epoch": 0.4964383561643836, + "grad_norm": 2.062134027481079, + "learning_rate": 4.6649065800162475e-06, + "loss": 2.2646, + "step": 453 + }, + { + "epoch": 0.49753424657534245, + "grad_norm": 1.94647216796875, + "learning_rate": 4.662875710804225e-06, + "loss": 2.1748, + "step": 454 + }, + { + "epoch": 0.4986301369863014, + "grad_norm": 1.9173872470855713, + "learning_rate": 4.660844841592202e-06, + "loss": 2.2435, + "step": 455 + }, + { + "epoch": 0.4997260273972603, + "grad_norm": 1.88880455493927, + "learning_rate": 4.658813972380179e-06, + "loss": 2.1711, + "step": 456 + }, + { + "epoch": 0.5008219178082192, + "grad_norm": 1.9628506898880005, + "learning_rate": 4.656783103168157e-06, + "loss": 2.2701, + "step": 457 + }, + { + "epoch": 0.5019178082191781, + "grad_norm": 2.0515286922454834, + "learning_rate": 4.654752233956134e-06, + "loss": 2.1656, + "step": 458 + }, + { + "epoch": 0.503013698630137, + "grad_norm": 1.978522539138794, + "learning_rate": 4.652721364744111e-06, + "loss": 2.246, + "step": 459 + }, + { + "epoch": 0.5041095890410959, + "grad_norm": 1.7993966341018677, + "learning_rate": 4.6506904955320885e-06, + "loss": 2.214, + "step": 460 + }, + { + "epoch": 0.5041095890410959, + "eval_loss": 2.3568837642669678, + "eval_runtime": 45.2086, + "eval_samples_per_second": 22.12, + "eval_steps_per_second": 0.929, + "step": 460 + }, + { + "epoch": 0.5052054794520547, + "grad_norm": 1.7809834480285645, + "learning_rate": 4.648659626320066e-06, + "loss": 2.2525, + "step": 461 + }, + { + "epoch": 0.5063013698630137, + "grad_norm": 1.9256166219711304, + "learning_rate": 4.646628757108043e-06, + "loss": 2.2611, + "step": 462 + }, + { + "epoch": 0.5073972602739726, + "grad_norm": 1.94192636013031, + "learning_rate": 4.6445978878960194e-06, + "loss": 2.2282, + "step": 463 + }, + { + "epoch": 0.5084931506849315, + "grad_norm": 1.8891818523406982, + "learning_rate": 4.6425670186839976e-06, + "loss": 2.2036, + "step": 464 + }, + { + "epoch": 0.5095890410958904, + "grad_norm": 2.352078676223755, + "learning_rate": 4.640536149471974e-06, + "loss": 2.2293, + "step": 465 + }, + { + "epoch": 0.5106849315068493, + "grad_norm": 1.8141659498214722, + "learning_rate": 4.638505280259952e-06, + "loss": 2.2551, + "step": 466 + }, + { + "epoch": 0.5117808219178083, + "grad_norm": 1.9492480754852295, + "learning_rate": 4.6364744110479285e-06, + "loss": 2.2641, + "step": 467 + }, + { + "epoch": 0.5128767123287671, + "grad_norm": 2.0210182666778564, + "learning_rate": 4.634443541835907e-06, + "loss": 2.2367, + "step": 468 + }, + { + "epoch": 0.513972602739726, + "grad_norm": 1.8936505317687988, + "learning_rate": 4.632412672623883e-06, + "loss": 2.265, + "step": 469 + }, + { + "epoch": 0.5150684931506849, + "grad_norm": 1.845888614654541, + "learning_rate": 4.630381803411861e-06, + "loss": 2.2169, + "step": 470 + }, + { + "epoch": 0.5150684931506849, + "eval_loss": 2.353489875793457, + "eval_runtime": 46.8739, + "eval_samples_per_second": 21.334, + "eval_steps_per_second": 0.896, + "step": 470 + }, + { + "epoch": 0.5161643835616438, + "grad_norm": 1.889675259590149, + "learning_rate": 4.628350934199838e-06, + "loss": 2.1383, + "step": 471 + }, + { + "epoch": 0.5172602739726028, + "grad_norm": 1.9799216985702515, + "learning_rate": 4.626320064987816e-06, + "loss": 2.2312, + "step": 472 + }, + { + "epoch": 0.5183561643835617, + "grad_norm": 2.0128607749938965, + "learning_rate": 4.624289195775792e-06, + "loss": 2.2529, + "step": 473 + }, + { + "epoch": 0.5194520547945205, + "grad_norm": 1.8791241645812988, + "learning_rate": 4.62225832656377e-06, + "loss": 2.226, + "step": 474 + }, + { + "epoch": 0.5205479452054794, + "grad_norm": 2.0497279167175293, + "learning_rate": 4.620227457351747e-06, + "loss": 2.2283, + "step": 475 + }, + { + "epoch": 0.5216438356164383, + "grad_norm": 1.8097424507141113, + "learning_rate": 4.618196588139724e-06, + "loss": 2.222, + "step": 476 + }, + { + "epoch": 0.5227397260273973, + "grad_norm": 1.8960500955581665, + "learning_rate": 4.616165718927701e-06, + "loss": 2.2342, + "step": 477 + }, + { + "epoch": 0.5238356164383562, + "grad_norm": 2.020730495452881, + "learning_rate": 4.6141348497156786e-06, + "loss": 2.2106, + "step": 478 + }, + { + "epoch": 0.5249315068493151, + "grad_norm": 1.8955297470092773, + "learning_rate": 4.612103980503656e-06, + "loss": 2.2755, + "step": 479 + }, + { + "epoch": 0.5260273972602739, + "grad_norm": 1.9767351150512695, + "learning_rate": 4.610073111291633e-06, + "loss": 2.2554, + "step": 480 + }, + { + "epoch": 0.5260273972602739, + "eval_loss": 2.347705841064453, + "eval_runtime": 46.9125, + "eval_samples_per_second": 21.316, + "eval_steps_per_second": 0.895, + "step": 480 + }, + { + "epoch": 0.5271232876712328, + "grad_norm": 1.936457633972168, + "learning_rate": 4.60804224207961e-06, + "loss": 2.2323, + "step": 481 + }, + { + "epoch": 0.5282191780821918, + "grad_norm": 1.9271862506866455, + "learning_rate": 4.606011372867588e-06, + "loss": 2.2151, + "step": 482 + }, + { + "epoch": 0.5293150684931507, + "grad_norm": 1.878153681755066, + "learning_rate": 4.603980503655565e-06, + "loss": 2.2322, + "step": 483 + }, + { + "epoch": 0.5304109589041096, + "grad_norm": 2.1009249687194824, + "learning_rate": 4.601949634443542e-06, + "loss": 2.2181, + "step": 484 + }, + { + "epoch": 0.5315068493150685, + "grad_norm": 1.9089181423187256, + "learning_rate": 4.5999187652315195e-06, + "loss": 2.1712, + "step": 485 + }, + { + "epoch": 0.5326027397260275, + "grad_norm": 2.207282781600952, + "learning_rate": 4.597887896019497e-06, + "loss": 2.1847, + "step": 486 + }, + { + "epoch": 0.5336986301369863, + "grad_norm": 2.001584053039551, + "learning_rate": 4.595857026807474e-06, + "loss": 2.214, + "step": 487 + }, + { + "epoch": 0.5347945205479452, + "grad_norm": 1.966700553894043, + "learning_rate": 4.593826157595451e-06, + "loss": 2.2237, + "step": 488 + }, + { + "epoch": 0.5358904109589041, + "grad_norm": 1.842469334602356, + "learning_rate": 4.591795288383429e-06, + "loss": 2.2089, + "step": 489 + }, + { + "epoch": 0.536986301369863, + "grad_norm": 2.0412986278533936, + "learning_rate": 4.589764419171406e-06, + "loss": 2.2734, + "step": 490 + }, + { + "epoch": 0.536986301369863, + "eval_loss": 2.3460772037506104, + "eval_runtime": 50.4048, + "eval_samples_per_second": 19.839, + "eval_steps_per_second": 0.833, + "step": 490 + }, + { + "epoch": 0.538082191780822, + "grad_norm": 1.9337031841278076, + "learning_rate": 4.587733549959383e-06, + "loss": 2.2314, + "step": 491 + }, + { + "epoch": 0.5391780821917809, + "grad_norm": 2.4448320865631104, + "learning_rate": 4.5857026807473604e-06, + "loss": 2.2098, + "step": 492 + }, + { + "epoch": 0.5402739726027397, + "grad_norm": 1.83162522315979, + "learning_rate": 4.583671811535338e-06, + "loss": 2.1902, + "step": 493 + }, + { + "epoch": 0.5413698630136986, + "grad_norm": 2.3614656925201416, + "learning_rate": 4.581640942323315e-06, + "loss": 2.1737, + "step": 494 + }, + { + "epoch": 0.5424657534246575, + "grad_norm": 2.001187801361084, + "learning_rate": 4.579610073111292e-06, + "loss": 2.2485, + "step": 495 + }, + { + "epoch": 0.5435616438356164, + "grad_norm": 2.3479933738708496, + "learning_rate": 4.5775792038992695e-06, + "loss": 2.2595, + "step": 496 + }, + { + "epoch": 0.5446575342465754, + "grad_norm": 2.135856866836548, + "learning_rate": 4.575548334687247e-06, + "loss": 2.2528, + "step": 497 + }, + { + "epoch": 0.5457534246575343, + "grad_norm": 1.983126163482666, + "learning_rate": 4.573517465475224e-06, + "loss": 2.2788, + "step": 498 + }, + { + "epoch": 0.5468493150684931, + "grad_norm": 2.18208909034729, + "learning_rate": 4.571486596263201e-06, + "loss": 2.1995, + "step": 499 + }, + { + "epoch": 0.547945205479452, + "grad_norm": 1.848747968673706, + "learning_rate": 4.569455727051178e-06, + "loss": 2.2403, + "step": 500 + }, + { + "epoch": 0.547945205479452, + "eval_loss": 2.344066619873047, + "eval_runtime": 51.1213, + "eval_samples_per_second": 19.561, + "eval_steps_per_second": 0.822, + "step": 500 + }, + { + "epoch": 0.549041095890411, + "grad_norm": 2.427860736846924, + "learning_rate": 4.567424857839156e-06, + "loss": 2.246, + "step": 501 + }, + { + "epoch": 0.5501369863013699, + "grad_norm": 1.913405418395996, + "learning_rate": 4.565393988627132e-06, + "loss": 2.2296, + "step": 502 + }, + { + "epoch": 0.5512328767123288, + "grad_norm": 2.0106029510498047, + "learning_rate": 4.5633631194151105e-06, + "loss": 2.2657, + "step": 503 + }, + { + "epoch": 0.5523287671232877, + "grad_norm": 2.2719595432281494, + "learning_rate": 4.561332250203087e-06, + "loss": 2.2341, + "step": 504 + }, + { + "epoch": 0.5534246575342465, + "grad_norm": 2.109786033630371, + "learning_rate": 4.559301380991065e-06, + "loss": 2.2181, + "step": 505 + }, + { + "epoch": 0.5545205479452054, + "grad_norm": 2.189779281616211, + "learning_rate": 4.5572705117790414e-06, + "loss": 2.1821, + "step": 506 + }, + { + "epoch": 0.5556164383561644, + "grad_norm": 2.0263590812683105, + "learning_rate": 4.5552396425670196e-06, + "loss": 2.2072, + "step": 507 + }, + { + "epoch": 0.5567123287671233, + "grad_norm": 2.215289354324341, + "learning_rate": 4.553208773354996e-06, + "loss": 2.2513, + "step": 508 + }, + { + "epoch": 0.5578082191780822, + "grad_norm": 2.3031351566314697, + "learning_rate": 4.551177904142974e-06, + "loss": 2.1988, + "step": 509 + }, + { + "epoch": 0.5589041095890411, + "grad_norm": 2.1830058097839355, + "learning_rate": 4.5491470349309505e-06, + "loss": 2.269, + "step": 510 + }, + { + "epoch": 0.5589041095890411, + "eval_loss": 2.342463731765747, + "eval_runtime": 43.1956, + "eval_samples_per_second": 23.151, + "eval_steps_per_second": 0.972, + "step": 510 + }, + { + "epoch": 0.56, + "grad_norm": 2.3210105895996094, + "learning_rate": 4.547116165718929e-06, + "loss": 2.2051, + "step": 511 + }, + { + "epoch": 0.5610958904109589, + "grad_norm": 1.9175361394882202, + "learning_rate": 4.545085296506905e-06, + "loss": 2.2304, + "step": 512 + }, + { + "epoch": 0.5621917808219178, + "grad_norm": 2.302947759628296, + "learning_rate": 4.543054427294882e-06, + "loss": 2.215, + "step": 513 + }, + { + "epoch": 0.5632876712328767, + "grad_norm": 1.8947396278381348, + "learning_rate": 4.54102355808286e-06, + "loss": 2.1926, + "step": 514 + }, + { + "epoch": 0.5643835616438356, + "grad_norm": 2.195268154144287, + "learning_rate": 4.538992688870837e-06, + "loss": 2.1937, + "step": 515 + }, + { + "epoch": 0.5654794520547946, + "grad_norm": 2.178874969482422, + "learning_rate": 4.536961819658814e-06, + "loss": 2.2248, + "step": 516 + }, + { + "epoch": 0.5665753424657535, + "grad_norm": 1.9856929779052734, + "learning_rate": 4.5349309504467915e-06, + "loss": 2.2077, + "step": 517 + }, + { + "epoch": 0.5676712328767123, + "grad_norm": 2.0686943531036377, + "learning_rate": 4.532900081234769e-06, + "loss": 2.2328, + "step": 518 + }, + { + "epoch": 0.5687671232876712, + "grad_norm": 1.9554919004440308, + "learning_rate": 4.530869212022746e-06, + "loss": 2.198, + "step": 519 + }, + { + "epoch": 0.5698630136986301, + "grad_norm": 1.9489670991897583, + "learning_rate": 4.528838342810723e-06, + "loss": 2.2287, + "step": 520 + }, + { + "epoch": 0.5698630136986301, + "eval_loss": 2.3415298461914062, + "eval_runtime": 48.7918, + "eval_samples_per_second": 20.495, + "eval_steps_per_second": 0.861, + "step": 520 + }, + { + "epoch": 0.570958904109589, + "grad_norm": 1.9868658781051636, + "learning_rate": 4.526807473598701e-06, + "loss": 2.2205, + "step": 521 + }, + { + "epoch": 0.572054794520548, + "grad_norm": 2.0965914726257324, + "learning_rate": 4.524776604386678e-06, + "loss": 2.2239, + "step": 522 + }, + { + "epoch": 0.5731506849315069, + "grad_norm": 2.041471481323242, + "learning_rate": 4.522745735174655e-06, + "loss": 2.2035, + "step": 523 + }, + { + "epoch": 0.5742465753424657, + "grad_norm": 1.885293960571289, + "learning_rate": 4.520714865962632e-06, + "loss": 2.2469, + "step": 524 + }, + { + "epoch": 0.5753424657534246, + "grad_norm": 1.9107296466827393, + "learning_rate": 4.51868399675061e-06, + "loss": 2.2386, + "step": 525 + }, + { + "epoch": 0.5764383561643835, + "grad_norm": 1.9945271015167236, + "learning_rate": 4.516653127538587e-06, + "loss": 2.235, + "step": 526 + }, + { + "epoch": 0.5775342465753425, + "grad_norm": 1.9787253141403198, + "learning_rate": 4.514622258326564e-06, + "loss": 2.2034, + "step": 527 + }, + { + "epoch": 0.5786301369863014, + "grad_norm": 2.0300636291503906, + "learning_rate": 4.5125913891145415e-06, + "loss": 2.1849, + "step": 528 + }, + { + "epoch": 0.5797260273972603, + "grad_norm": 2.0348970890045166, + "learning_rate": 4.510560519902519e-06, + "loss": 2.2174, + "step": 529 + }, + { + "epoch": 0.5808219178082191, + "grad_norm": 1.8589537143707275, + "learning_rate": 4.508529650690496e-06, + "loss": 2.2351, + "step": 530 + }, + { + "epoch": 0.5808219178082191, + "eval_loss": 2.3450186252593994, + "eval_runtime": 52.4548, + "eval_samples_per_second": 19.064, + "eval_steps_per_second": 0.801, + "step": 530 + }, + { + "epoch": 0.581917808219178, + "grad_norm": 2.0496020317077637, + "learning_rate": 4.506498781478473e-06, + "loss": 2.182, + "step": 531 + }, + { + "epoch": 0.583013698630137, + "grad_norm": 2.0649805068969727, + "learning_rate": 4.504467912266451e-06, + "loss": 2.1654, + "step": 532 + }, + { + "epoch": 0.5841095890410959, + "grad_norm": 1.9809486865997314, + "learning_rate": 4.502437043054428e-06, + "loss": 2.2182, + "step": 533 + }, + { + "epoch": 0.5852054794520548, + "grad_norm": 2.21915340423584, + "learning_rate": 4.500406173842405e-06, + "loss": 2.2697, + "step": 534 + }, + { + "epoch": 0.5863013698630137, + "grad_norm": 1.9822649955749512, + "learning_rate": 4.4983753046303824e-06, + "loss": 2.2297, + "step": 535 + }, + { + "epoch": 0.5873972602739727, + "grad_norm": 1.972575306892395, + "learning_rate": 4.49634443541836e-06, + "loss": 2.2164, + "step": 536 + }, + { + "epoch": 0.5884931506849315, + "grad_norm": 1.948667049407959, + "learning_rate": 4.494313566206336e-06, + "loss": 2.2087, + "step": 537 + }, + { + "epoch": 0.5895890410958904, + "grad_norm": 2.7486417293548584, + "learning_rate": 4.492282696994314e-06, + "loss": 2.2602, + "step": 538 + }, + { + "epoch": 0.5906849315068493, + "grad_norm": 2.3374552726745605, + "learning_rate": 4.490251827782291e-06, + "loss": 2.2054, + "step": 539 + }, + { + "epoch": 0.5917808219178082, + "grad_norm": 2.222116231918335, + "learning_rate": 4.488220958570269e-06, + "loss": 2.1868, + "step": 540 + }, + { + "epoch": 0.5917808219178082, + "eval_loss": 2.345618963241577, + "eval_runtime": 53.0302, + "eval_samples_per_second": 18.857, + "eval_steps_per_second": 0.792, + "step": 540 + }, + { + "epoch": 0.5928767123287672, + "grad_norm": 2.00752854347229, + "learning_rate": 4.486190089358245e-06, + "loss": 2.187, + "step": 541 + }, + { + "epoch": 0.5939726027397261, + "grad_norm": 2.2466132640838623, + "learning_rate": 4.484159220146223e-06, + "loss": 2.1611, + "step": 542 + }, + { + "epoch": 0.5950684931506849, + "grad_norm": 2.1279385089874268, + "learning_rate": 4.4821283509342e-06, + "loss": 2.248, + "step": 543 + }, + { + "epoch": 0.5961643835616438, + "grad_norm": 1.9140571355819702, + "learning_rate": 4.480097481722178e-06, + "loss": 2.2314, + "step": 544 + }, + { + "epoch": 0.5972602739726027, + "grad_norm": 2.120741605758667, + "learning_rate": 4.478066612510154e-06, + "loss": 2.212, + "step": 545 + }, + { + "epoch": 0.5983561643835617, + "grad_norm": 2.003713369369507, + "learning_rate": 4.4760357432981325e-06, + "loss": 2.1824, + "step": 546 + }, + { + "epoch": 0.5994520547945206, + "grad_norm": 1.9161221981048584, + "learning_rate": 4.474004874086109e-06, + "loss": 2.2015, + "step": 547 + }, + { + "epoch": 0.6005479452054795, + "grad_norm": 2.367110252380371, + "learning_rate": 4.471974004874087e-06, + "loss": 2.3122, + "step": 548 + }, + { + "epoch": 0.6016438356164383, + "grad_norm": 2.113380193710327, + "learning_rate": 4.4699431356620635e-06, + "loss": 2.1692, + "step": 549 + }, + { + "epoch": 0.6027397260273972, + "grad_norm": 1.974181890487671, + "learning_rate": 4.467912266450041e-06, + "loss": 2.2509, + "step": 550 + }, + { + "epoch": 0.6027397260273972, + "eval_loss": 2.342583656311035, + "eval_runtime": 50.7362, + "eval_samples_per_second": 19.71, + "eval_steps_per_second": 0.828, + "step": 550 + }, + { + "epoch": 0.6038356164383561, + "grad_norm": 1.889894962310791, + "learning_rate": 4.465881397238018e-06, + "loss": 2.2214, + "step": 551 + }, + { + "epoch": 0.6049315068493151, + "grad_norm": 1.8982343673706055, + "learning_rate": 4.463850528025995e-06, + "loss": 2.2135, + "step": 552 + }, + { + "epoch": 0.606027397260274, + "grad_norm": 1.9462087154388428, + "learning_rate": 4.4618196588139726e-06, + "loss": 2.2255, + "step": 553 + }, + { + "epoch": 0.6071232876712329, + "grad_norm": 1.8729931116104126, + "learning_rate": 4.45978878960195e-06, + "loss": 2.2358, + "step": 554 + }, + { + "epoch": 0.6082191780821918, + "grad_norm": 1.9230246543884277, + "learning_rate": 4.457757920389927e-06, + "loss": 2.222, + "step": 555 + }, + { + "epoch": 0.6093150684931506, + "grad_norm": 1.9495900869369507, + "learning_rate": 4.455727051177904e-06, + "loss": 2.1779, + "step": 556 + }, + { + "epoch": 0.6104109589041096, + "grad_norm": 1.983487606048584, + "learning_rate": 4.453696181965882e-06, + "loss": 2.2007, + "step": 557 + }, + { + "epoch": 0.6115068493150685, + "grad_norm": 2.1541049480438232, + "learning_rate": 4.451665312753859e-06, + "loss": 2.2513, + "step": 558 + }, + { + "epoch": 0.6126027397260274, + "grad_norm": 1.9133352041244507, + "learning_rate": 4.449634443541836e-06, + "loss": 2.2288, + "step": 559 + }, + { + "epoch": 0.6136986301369863, + "grad_norm": 2.3325741291046143, + "learning_rate": 4.4476035743298135e-06, + "loss": 2.2539, + "step": 560 + }, + { + "epoch": 0.6136986301369863, + "eval_loss": 2.3387348651885986, + "eval_runtime": 53.4301, + "eval_samples_per_second": 18.716, + "eval_steps_per_second": 0.786, + "step": 560 + }, + { + "epoch": 0.6147945205479453, + "grad_norm": 1.8099673986434937, + "learning_rate": 4.445572705117791e-06, + "loss": 2.1845, + "step": 561 + }, + { + "epoch": 0.6158904109589041, + "grad_norm": 2.201901912689209, + "learning_rate": 4.443541835905768e-06, + "loss": 2.195, + "step": 562 + }, + { + "epoch": 0.616986301369863, + "grad_norm": 1.8412714004516602, + "learning_rate": 4.441510966693745e-06, + "loss": 2.2551, + "step": 563 + }, + { + "epoch": 0.6180821917808219, + "grad_norm": 1.8761520385742188, + "learning_rate": 4.439480097481723e-06, + "loss": 2.1965, + "step": 564 + }, + { + "epoch": 0.6191780821917808, + "grad_norm": 1.9123467206954956, + "learning_rate": 4.4374492282697e-06, + "loss": 2.2502, + "step": 565 + }, + { + "epoch": 0.6202739726027398, + "grad_norm": 2.0631234645843506, + "learning_rate": 4.435418359057677e-06, + "loss": 2.1582, + "step": 566 + }, + { + "epoch": 0.6213698630136987, + "grad_norm": 2.3191616535186768, + "learning_rate": 4.433387489845654e-06, + "loss": 2.2271, + "step": 567 + }, + { + "epoch": 0.6224657534246575, + "grad_norm": 2.043057918548584, + "learning_rate": 4.431356620633632e-06, + "loss": 2.2292, + "step": 568 + }, + { + "epoch": 0.6235616438356164, + "grad_norm": 2.0716068744659424, + "learning_rate": 4.429325751421609e-06, + "loss": 2.196, + "step": 569 + }, + { + "epoch": 0.6246575342465753, + "grad_norm": 1.7669776678085327, + "learning_rate": 4.427294882209586e-06, + "loss": 2.1953, + "step": 570 + }, + { + "epoch": 0.6246575342465753, + "eval_loss": 2.336578607559204, + "eval_runtime": 50.0083, + "eval_samples_per_second": 19.997, + "eval_steps_per_second": 0.84, + "step": 570 + }, + { + "epoch": 0.6257534246575343, + "grad_norm": 1.8165771961212158, + "learning_rate": 4.4252640129975635e-06, + "loss": 2.2155, + "step": 571 + }, + { + "epoch": 0.6268493150684932, + "grad_norm": 1.8640249967575073, + "learning_rate": 4.423233143785541e-06, + "loss": 2.1648, + "step": 572 + }, + { + "epoch": 0.6279452054794521, + "grad_norm": 2.013869285583496, + "learning_rate": 4.421202274573518e-06, + "loss": 2.2131, + "step": 573 + }, + { + "epoch": 0.6290410958904109, + "grad_norm": 2.2650580406188965, + "learning_rate": 4.4191714053614945e-06, + "loss": 2.2287, + "step": 574 + }, + { + "epoch": 0.6301369863013698, + "grad_norm": 1.9232103824615479, + "learning_rate": 4.417140536149473e-06, + "loss": 2.1339, + "step": 575 + }, + { + "epoch": 0.6312328767123287, + "grad_norm": 1.910134196281433, + "learning_rate": 4.415109666937449e-06, + "loss": 2.1755, + "step": 576 + }, + { + "epoch": 0.6323287671232877, + "grad_norm": 2.2492005825042725, + "learning_rate": 4.413078797725427e-06, + "loss": 2.2362, + "step": 577 + }, + { + "epoch": 0.6334246575342466, + "grad_norm": 1.8584033250808716, + "learning_rate": 4.411047928513404e-06, + "loss": 2.1911, + "step": 578 + }, + { + "epoch": 0.6345205479452055, + "grad_norm": 2.156273126602173, + "learning_rate": 4.409017059301382e-06, + "loss": 2.2109, + "step": 579 + }, + { + "epoch": 0.6356164383561644, + "grad_norm": 1.7977535724639893, + "learning_rate": 4.406986190089358e-06, + "loss": 2.2195, + "step": 580 + }, + { + "epoch": 0.6356164383561644, + "eval_loss": 2.33555269241333, + "eval_runtime": 48.3919, + "eval_samples_per_second": 20.665, + "eval_steps_per_second": 0.868, + "step": 580 + }, + { + "epoch": 0.6367123287671232, + "grad_norm": 1.8262702226638794, + "learning_rate": 4.404955320877336e-06, + "loss": 2.2345, + "step": 581 + }, + { + "epoch": 0.6378082191780822, + "grad_norm": 1.9389383792877197, + "learning_rate": 4.402924451665313e-06, + "loss": 2.2347, + "step": 582 + }, + { + "epoch": 0.6389041095890411, + "grad_norm": 1.9548988342285156, + "learning_rate": 4.400893582453291e-06, + "loss": 2.1967, + "step": 583 + }, + { + "epoch": 0.64, + "grad_norm": 1.898092269897461, + "learning_rate": 4.398862713241267e-06, + "loss": 2.1795, + "step": 584 + }, + { + "epoch": 0.6410958904109589, + "grad_norm": 1.8183610439300537, + "learning_rate": 4.396831844029245e-06, + "loss": 2.2031, + "step": 585 + }, + { + "epoch": 0.6421917808219179, + "grad_norm": 2.1520016193389893, + "learning_rate": 4.394800974817222e-06, + "loss": 2.2051, + "step": 586 + }, + { + "epoch": 0.6432876712328767, + "grad_norm": 1.9930012226104736, + "learning_rate": 4.392770105605199e-06, + "loss": 2.2053, + "step": 587 + }, + { + "epoch": 0.6443835616438356, + "grad_norm": 2.311450958251953, + "learning_rate": 4.390739236393176e-06, + "loss": 2.2252, + "step": 588 + }, + { + "epoch": 0.6454794520547945, + "grad_norm": 2.0168697834014893, + "learning_rate": 4.388708367181154e-06, + "loss": 2.2286, + "step": 589 + }, + { + "epoch": 0.6465753424657534, + "grad_norm": 2.055964469909668, + "learning_rate": 4.386677497969131e-06, + "loss": 2.2278, + "step": 590 + }, + { + "epoch": 0.6465753424657534, + "eval_loss": 2.334367275238037, + "eval_runtime": 48.9414, + "eval_samples_per_second": 20.433, + "eval_steps_per_second": 0.858, + "step": 590 + }, + { + "epoch": 0.6476712328767124, + "grad_norm": 2.0161590576171875, + "learning_rate": 4.384646628757108e-06, + "loss": 2.2102, + "step": 591 + }, + { + "epoch": 0.6487671232876713, + "grad_norm": 2.0827667713165283, + "learning_rate": 4.3826157595450855e-06, + "loss": 2.2531, + "step": 592 + }, + { + "epoch": 0.6498630136986301, + "grad_norm": 2.077637195587158, + "learning_rate": 4.380584890333063e-06, + "loss": 2.2091, + "step": 593 + }, + { + "epoch": 0.650958904109589, + "grad_norm": 2.023892879486084, + "learning_rate": 4.37855402112104e-06, + "loss": 2.1684, + "step": 594 + }, + { + "epoch": 0.6520547945205479, + "grad_norm": 1.8509857654571533, + "learning_rate": 4.376523151909017e-06, + "loss": 2.1439, + "step": 595 + }, + { + "epoch": 0.6531506849315069, + "grad_norm": 2.1649608612060547, + "learning_rate": 4.3744922826969946e-06, + "loss": 2.2557, + "step": 596 + }, + { + "epoch": 0.6542465753424658, + "grad_norm": 2.1351773738861084, + "learning_rate": 4.372461413484972e-06, + "loss": 2.177, + "step": 597 + }, + { + "epoch": 0.6553424657534247, + "grad_norm": 2.1713974475860596, + "learning_rate": 4.370430544272949e-06, + "loss": 2.2223, + "step": 598 + }, + { + "epoch": 0.6564383561643835, + "grad_norm": 2.2905397415161133, + "learning_rate": 4.368399675060926e-06, + "loss": 2.2373, + "step": 599 + }, + { + "epoch": 0.6575342465753424, + "grad_norm": 2.007160186767578, + "learning_rate": 4.366368805848904e-06, + "loss": 2.2355, + "step": 600 + }, + { + "epoch": 0.6575342465753424, + "eval_loss": 2.3322324752807617, + "eval_runtime": 46.2769, + "eval_samples_per_second": 21.609, + "eval_steps_per_second": 0.908, + "step": 600 + }, + { + "epoch": 0.6586301369863014, + "grad_norm": 2.309783697128296, + "learning_rate": 4.364337936636881e-06, + "loss": 2.2052, + "step": 601 + }, + { + "epoch": 0.6597260273972603, + "grad_norm": 2.498034954071045, + "learning_rate": 4.362307067424858e-06, + "loss": 2.1567, + "step": 602 + }, + { + "epoch": 0.6608219178082192, + "grad_norm": 1.8302545547485352, + "learning_rate": 4.3602761982128355e-06, + "loss": 2.2117, + "step": 603 + }, + { + "epoch": 0.6619178082191781, + "grad_norm": 2.226947546005249, + "learning_rate": 4.358245329000813e-06, + "loss": 2.1881, + "step": 604 + }, + { + "epoch": 0.663013698630137, + "grad_norm": 1.8484126329421997, + "learning_rate": 4.35621445978879e-06, + "loss": 2.1955, + "step": 605 + }, + { + "epoch": 0.6641095890410958, + "grad_norm": 1.9844266176223755, + "learning_rate": 4.354183590576767e-06, + "loss": 2.2063, + "step": 606 + }, + { + "epoch": 0.6652054794520548, + "grad_norm": 1.9292365312576294, + "learning_rate": 4.352152721364745e-06, + "loss": 2.205, + "step": 607 + }, + { + "epoch": 0.6663013698630137, + "grad_norm": 1.980895757675171, + "learning_rate": 4.350121852152722e-06, + "loss": 2.1727, + "step": 608 + }, + { + "epoch": 0.6673972602739726, + "grad_norm": 1.8331713676452637, + "learning_rate": 4.348090982940699e-06, + "loss": 2.226, + "step": 609 + }, + { + "epoch": 0.6684931506849315, + "grad_norm": 1.9666537046432495, + "learning_rate": 4.3460601137286764e-06, + "loss": 2.2198, + "step": 610 + }, + { + "epoch": 0.6684931506849315, + "eval_loss": 2.3297338485717773, + "eval_runtime": 42.6274, + "eval_samples_per_second": 23.459, + "eval_steps_per_second": 0.985, + "step": 610 + }, + { + "epoch": 0.6695890410958905, + "grad_norm": 1.9741110801696777, + "learning_rate": 4.344029244516654e-06, + "loss": 2.2132, + "step": 611 + }, + { + "epoch": 0.6706849315068493, + "grad_norm": 2.0747334957122803, + "learning_rate": 4.341998375304631e-06, + "loss": 2.2081, + "step": 612 + }, + { + "epoch": 0.6717808219178082, + "grad_norm": 2.031503915786743, + "learning_rate": 4.339967506092607e-06, + "loss": 2.2642, + "step": 613 + }, + { + "epoch": 0.6728767123287671, + "grad_norm": 1.8513256311416626, + "learning_rate": 4.3379366368805855e-06, + "loss": 2.2141, + "step": 614 + }, + { + "epoch": 0.673972602739726, + "grad_norm": 1.9098315238952637, + "learning_rate": 4.335905767668562e-06, + "loss": 2.2064, + "step": 615 + }, + { + "epoch": 0.675068493150685, + "grad_norm": 1.9704065322875977, + "learning_rate": 4.33387489845654e-06, + "loss": 2.1997, + "step": 616 + }, + { + "epoch": 0.6761643835616439, + "grad_norm": 2.1520018577575684, + "learning_rate": 4.3318440292445165e-06, + "loss": 2.2278, + "step": 617 + }, + { + "epoch": 0.6772602739726027, + "grad_norm": 1.898164987564087, + "learning_rate": 4.329813160032495e-06, + "loss": 2.2147, + "step": 618 + }, + { + "epoch": 0.6783561643835616, + "grad_norm": 1.8955962657928467, + "learning_rate": 4.327782290820471e-06, + "loss": 2.1825, + "step": 619 + }, + { + "epoch": 0.6794520547945205, + "grad_norm": 1.7579313516616821, + "learning_rate": 4.325751421608449e-06, + "loss": 2.1619, + "step": 620 + }, + { + "epoch": 0.6794520547945205, + "eval_loss": 2.329692840576172, + "eval_runtime": 46.3187, + "eval_samples_per_second": 21.59, + "eval_steps_per_second": 0.907, + "step": 620 + }, + { + "epoch": 0.6805479452054795, + "grad_norm": 1.8716063499450684, + "learning_rate": 4.323720552396426e-06, + "loss": 2.2289, + "step": 621 + }, + { + "epoch": 0.6816438356164384, + "grad_norm": 1.8798290491104126, + "learning_rate": 4.321689683184404e-06, + "loss": 2.1944, + "step": 622 + }, + { + "epoch": 0.6827397260273973, + "grad_norm": 1.8438303470611572, + "learning_rate": 4.31965881397238e-06, + "loss": 2.1879, + "step": 623 + }, + { + "epoch": 0.6838356164383562, + "grad_norm": 1.7844786643981934, + "learning_rate": 4.3176279447603574e-06, + "loss": 2.1936, + "step": 624 + }, + { + "epoch": 0.684931506849315, + "grad_norm": 1.955529808998108, + "learning_rate": 4.315597075548335e-06, + "loss": 2.2258, + "step": 625 + }, + { + "epoch": 0.686027397260274, + "grad_norm": 2.3826634883880615, + "learning_rate": 4.313566206336312e-06, + "loss": 2.2135, + "step": 626 + }, + { + "epoch": 0.6871232876712329, + "grad_norm": 1.8453123569488525, + "learning_rate": 4.311535337124289e-06, + "loss": 2.2362, + "step": 627 + }, + { + "epoch": 0.6882191780821918, + "grad_norm": 2.742682933807373, + "learning_rate": 4.3095044679122665e-06, + "loss": 2.194, + "step": 628 + }, + { + "epoch": 0.6893150684931507, + "grad_norm": 2.0299901962280273, + "learning_rate": 4.307473598700244e-06, + "loss": 2.2029, + "step": 629 + }, + { + "epoch": 0.6904109589041096, + "grad_norm": 2.29683780670166, + "learning_rate": 4.305442729488221e-06, + "loss": 2.2141, + "step": 630 + }, + { + "epoch": 0.6904109589041096, + "eval_loss": 2.329368829727173, + "eval_runtime": 45.1947, + "eval_samples_per_second": 22.126, + "eval_steps_per_second": 0.929, + "step": 630 + }, + { + "epoch": 0.6915068493150685, + "grad_norm": 2.208280086517334, + "learning_rate": 4.303411860276198e-06, + "loss": 2.1637, + "step": 631 + }, + { + "epoch": 0.6926027397260274, + "grad_norm": 1.943514108657837, + "learning_rate": 4.301380991064176e-06, + "loss": 2.234, + "step": 632 + }, + { + "epoch": 0.6936986301369863, + "grad_norm": 2.486440420150757, + "learning_rate": 4.299350121852153e-06, + "loss": 2.2239, + "step": 633 + }, + { + "epoch": 0.6947945205479452, + "grad_norm": 1.7243690490722656, + "learning_rate": 4.29731925264013e-06, + "loss": 2.2428, + "step": 634 + }, + { + "epoch": 0.6958904109589041, + "grad_norm": 2.31227445602417, + "learning_rate": 4.2952883834281075e-06, + "loss": 2.2403, + "step": 635 + }, + { + "epoch": 0.6969863013698631, + "grad_norm": 1.7361263036727905, + "learning_rate": 4.293257514216085e-06, + "loss": 2.1975, + "step": 636 + }, + { + "epoch": 0.6980821917808219, + "grad_norm": 2.010831356048584, + "learning_rate": 4.291226645004062e-06, + "loss": 2.1633, + "step": 637 + }, + { + "epoch": 0.6991780821917808, + "grad_norm": 2.2796833515167236, + "learning_rate": 4.289195775792039e-06, + "loss": 2.2307, + "step": 638 + }, + { + "epoch": 0.7002739726027397, + "grad_norm": 2.1069483757019043, + "learning_rate": 4.287164906580017e-06, + "loss": 2.2389, + "step": 639 + }, + { + "epoch": 0.7013698630136986, + "grad_norm": 2.19460129737854, + "learning_rate": 4.285134037367994e-06, + "loss": 2.1992, + "step": 640 + }, + { + "epoch": 0.7013698630136986, + "eval_loss": 2.3223369121551514, + "eval_runtime": 43.1093, + "eval_samples_per_second": 23.197, + "eval_steps_per_second": 0.974, + "step": 640 + }, + { + "epoch": 0.7024657534246576, + "grad_norm": 1.875261902809143, + "learning_rate": 4.283103168155971e-06, + "loss": 2.1767, + "step": 641 + }, + { + "epoch": 0.7035616438356165, + "grad_norm": 1.906762957572937, + "learning_rate": 4.281072298943948e-06, + "loss": 2.1734, + "step": 642 + }, + { + "epoch": 0.7046575342465753, + "grad_norm": 2.1114180088043213, + "learning_rate": 4.279041429731926e-06, + "loss": 2.1734, + "step": 643 + }, + { + "epoch": 0.7057534246575342, + "grad_norm": 1.844373106956482, + "learning_rate": 4.277010560519903e-06, + "loss": 2.1967, + "step": 644 + }, + { + "epoch": 0.7068493150684931, + "grad_norm": 2.3463754653930664, + "learning_rate": 4.27497969130788e-06, + "loss": 2.1735, + "step": 645 + }, + { + "epoch": 0.707945205479452, + "grad_norm": 2.3530960083007812, + "learning_rate": 4.2729488220958575e-06, + "loss": 2.189, + "step": 646 + }, + { + "epoch": 0.709041095890411, + "grad_norm": 2.110963821411133, + "learning_rate": 4.270917952883835e-06, + "loss": 2.218, + "step": 647 + }, + { + "epoch": 0.7101369863013699, + "grad_norm": 2.625005006790161, + "learning_rate": 4.268887083671812e-06, + "loss": 2.2235, + "step": 648 + }, + { + "epoch": 0.7112328767123288, + "grad_norm": 1.8411654233932495, + "learning_rate": 4.266856214459789e-06, + "loss": 2.1757, + "step": 649 + }, + { + "epoch": 0.7123287671232876, + "grad_norm": 2.0817489624023438, + "learning_rate": 4.264825345247766e-06, + "loss": 2.2698, + "step": 650 + }, + { + "epoch": 0.7123287671232876, + "eval_loss": 2.322544574737549, + "eval_runtime": 43.0771, + "eval_samples_per_second": 23.214, + "eval_steps_per_second": 0.975, + "step": 650 + }, + { + "epoch": 0.7134246575342466, + "grad_norm": 1.8953847885131836, + "learning_rate": 4.262794476035744e-06, + "loss": 2.1745, + "step": 651 + }, + { + "epoch": 0.7145205479452055, + "grad_norm": 1.8463294506072998, + "learning_rate": 4.26076360682372e-06, + "loss": 2.2042, + "step": 652 + }, + { + "epoch": 0.7156164383561644, + "grad_norm": 1.8776047229766846, + "learning_rate": 4.2587327376116984e-06, + "loss": 2.1512, + "step": 653 + }, + { + "epoch": 0.7167123287671233, + "grad_norm": 1.8949921131134033, + "learning_rate": 4.256701868399675e-06, + "loss": 2.1895, + "step": 654 + }, + { + "epoch": 0.7178082191780822, + "grad_norm": 1.8172632455825806, + "learning_rate": 4.254670999187653e-06, + "loss": 2.2, + "step": 655 + }, + { + "epoch": 0.718904109589041, + "grad_norm": 2.105499505996704, + "learning_rate": 4.252640129975629e-06, + "loss": 2.202, + "step": 656 + }, + { + "epoch": 0.72, + "grad_norm": 1.9515924453735352, + "learning_rate": 4.2506092607636075e-06, + "loss": 2.2116, + "step": 657 + }, + { + "epoch": 0.7210958904109589, + "grad_norm": 1.761770248413086, + "learning_rate": 4.248578391551584e-06, + "loss": 2.1921, + "step": 658 + }, + { + "epoch": 0.7221917808219178, + "grad_norm": 1.9848097562789917, + "learning_rate": 4.246547522339562e-06, + "loss": 2.1553, + "step": 659 + }, + { + "epoch": 0.7232876712328767, + "grad_norm": 2.041987180709839, + "learning_rate": 4.2445166531275385e-06, + "loss": 2.1975, + "step": 660 + }, + { + "epoch": 0.7232876712328767, + "eval_loss": 2.325580358505249, + "eval_runtime": 43.6207, + "eval_samples_per_second": 22.925, + "eval_steps_per_second": 0.963, + "step": 660 + }, + { + "epoch": 0.7243835616438357, + "grad_norm": 1.9755775928497314, + "learning_rate": 4.242485783915517e-06, + "loss": 2.1955, + "step": 661 + }, + { + "epoch": 0.7254794520547945, + "grad_norm": 1.9621502161026, + "learning_rate": 4.240454914703493e-06, + "loss": 2.1554, + "step": 662 + }, + { + "epoch": 0.7265753424657534, + "grad_norm": 1.848013162612915, + "learning_rate": 4.23842404549147e-06, + "loss": 2.16, + "step": 663 + }, + { + "epoch": 0.7276712328767123, + "grad_norm": 1.7370128631591797, + "learning_rate": 4.236393176279448e-06, + "loss": 2.1696, + "step": 664 + }, + { + "epoch": 0.7287671232876712, + "grad_norm": 2.0708158016204834, + "learning_rate": 4.234362307067425e-06, + "loss": 2.1592, + "step": 665 + }, + { + "epoch": 0.7298630136986302, + "grad_norm": 1.9864391088485718, + "learning_rate": 4.232331437855402e-06, + "loss": 2.2121, + "step": 666 + }, + { + "epoch": 0.7309589041095891, + "grad_norm": 1.9988250732421875, + "learning_rate": 4.2303005686433795e-06, + "loss": 2.2306, + "step": 667 + }, + { + "epoch": 0.7320547945205479, + "grad_norm": 1.89108407497406, + "learning_rate": 4.228269699431357e-06, + "loss": 2.222, + "step": 668 + }, + { + "epoch": 0.7331506849315068, + "grad_norm": 1.7941503524780273, + "learning_rate": 4.226238830219334e-06, + "loss": 2.1975, + "step": 669 + }, + { + "epoch": 0.7342465753424657, + "grad_norm": 1.8058613538742065, + "learning_rate": 4.224207961007311e-06, + "loss": 2.2037, + "step": 670 + }, + { + "epoch": 0.7342465753424657, + "eval_loss": 2.3266475200653076, + "eval_runtime": 47.6623, + "eval_samples_per_second": 20.981, + "eval_steps_per_second": 0.881, + "step": 670 + }, + { + "epoch": 0.7353424657534247, + "grad_norm": 1.764448642730713, + "learning_rate": 4.2221770917952886e-06, + "loss": 2.1904, + "step": 671 + }, + { + "epoch": 0.7364383561643836, + "grad_norm": 1.8252112865447998, + "learning_rate": 4.220146222583266e-06, + "loss": 2.1959, + "step": 672 + }, + { + "epoch": 0.7375342465753425, + "grad_norm": 2.02898907661438, + "learning_rate": 4.218115353371243e-06, + "loss": 2.1937, + "step": 673 + }, + { + "epoch": 0.7386301369863014, + "grad_norm": 1.8766248226165771, + "learning_rate": 4.21608448415922e-06, + "loss": 2.195, + "step": 674 + }, + { + "epoch": 0.7397260273972602, + "grad_norm": 2.2350332736968994, + "learning_rate": 4.214053614947198e-06, + "loss": 2.1798, + "step": 675 + }, + { + "epoch": 0.7408219178082192, + "grad_norm": 1.820399522781372, + "learning_rate": 4.212022745735175e-06, + "loss": 2.2371, + "step": 676 + }, + { + "epoch": 0.7419178082191781, + "grad_norm": 2.3411312103271484, + "learning_rate": 4.209991876523152e-06, + "loss": 2.1603, + "step": 677 + }, + { + "epoch": 0.743013698630137, + "grad_norm": 1.8562874794006348, + "learning_rate": 4.2079610073111295e-06, + "loss": 2.2082, + "step": 678 + }, + { + "epoch": 0.7441095890410959, + "grad_norm": 2.274214506149292, + "learning_rate": 4.205930138099107e-06, + "loss": 2.1716, + "step": 679 + }, + { + "epoch": 0.7452054794520548, + "grad_norm": 1.9173530340194702, + "learning_rate": 4.203899268887084e-06, + "loss": 2.2461, + "step": 680 + }, + { + "epoch": 0.7452054794520548, + "eval_loss": 2.3253841400146484, + "eval_runtime": 46.5732, + "eval_samples_per_second": 21.472, + "eval_steps_per_second": 0.902, + "step": 680 + }, + { + "epoch": 0.7463013698630137, + "grad_norm": 1.9732033014297485, + "learning_rate": 4.201868399675061e-06, + "loss": 2.2016, + "step": 681 + }, + { + "epoch": 0.7473972602739726, + "grad_norm": 1.8850247859954834, + "learning_rate": 4.199837530463039e-06, + "loss": 2.223, + "step": 682 + }, + { + "epoch": 0.7484931506849315, + "grad_norm": 1.8936313390731812, + "learning_rate": 4.197806661251016e-06, + "loss": 2.1381, + "step": 683 + }, + { + "epoch": 0.7495890410958904, + "grad_norm": 2.11995530128479, + "learning_rate": 4.195775792038993e-06, + "loss": 2.1962, + "step": 684 + }, + { + "epoch": 0.7506849315068493, + "grad_norm": 1.8207820653915405, + "learning_rate": 4.19374492282697e-06, + "loss": 2.1835, + "step": 685 + }, + { + "epoch": 0.7517808219178083, + "grad_norm": 2.06030535697937, + "learning_rate": 4.191714053614948e-06, + "loss": 2.1347, + "step": 686 + }, + { + "epoch": 0.7528767123287671, + "grad_norm": 2.048990488052368, + "learning_rate": 4.189683184402924e-06, + "loss": 2.2189, + "step": 687 + }, + { + "epoch": 0.753972602739726, + "grad_norm": 1.976369023323059, + "learning_rate": 4.187652315190902e-06, + "loss": 2.1805, + "step": 688 + }, + { + "epoch": 0.7550684931506849, + "grad_norm": 2.1160926818847656, + "learning_rate": 4.185621445978879e-06, + "loss": 2.2322, + "step": 689 + }, + { + "epoch": 0.7561643835616438, + "grad_norm": 1.9444516897201538, + "learning_rate": 4.183590576766857e-06, + "loss": 2.1696, + "step": 690 + }, + { + "epoch": 0.7561643835616438, + "eval_loss": 2.3242011070251465, + "eval_runtime": 48.7528, + "eval_samples_per_second": 20.512, + "eval_steps_per_second": 0.861, + "step": 690 + }, + { + "epoch": 0.7572602739726028, + "grad_norm": 1.9567183256149292, + "learning_rate": 4.181559707554833e-06, + "loss": 2.1891, + "step": 691 + }, + { + "epoch": 0.7583561643835617, + "grad_norm": 1.9185020923614502, + "learning_rate": 4.179528838342811e-06, + "loss": 2.2318, + "step": 692 + }, + { + "epoch": 0.7594520547945205, + "grad_norm": 1.833487629890442, + "learning_rate": 4.177497969130788e-06, + "loss": 2.1705, + "step": 693 + }, + { + "epoch": 0.7605479452054794, + "grad_norm": 2.2949867248535156, + "learning_rate": 4.175467099918766e-06, + "loss": 2.1712, + "step": 694 + }, + { + "epoch": 0.7616438356164383, + "grad_norm": 1.8663084506988525, + "learning_rate": 4.173436230706742e-06, + "loss": 2.2262, + "step": 695 + }, + { + "epoch": 0.7627397260273973, + "grad_norm": 2.4633052349090576, + "learning_rate": 4.1714053614947205e-06, + "loss": 2.185, + "step": 696 + }, + { + "epoch": 0.7638356164383562, + "grad_norm": 2.036524772644043, + "learning_rate": 4.169374492282697e-06, + "loss": 2.1697, + "step": 697 + }, + { + "epoch": 0.7649315068493151, + "grad_norm": 2.013810634613037, + "learning_rate": 4.167343623070675e-06, + "loss": 2.1826, + "step": 698 + }, + { + "epoch": 0.766027397260274, + "grad_norm": 2.214109182357788, + "learning_rate": 4.165312753858652e-06, + "loss": 2.2162, + "step": 699 + }, + { + "epoch": 0.7671232876712328, + "grad_norm": 1.8463964462280273, + "learning_rate": 4.163281884646629e-06, + "loss": 2.1865, + "step": 700 + }, + { + "epoch": 0.7671232876712328, + "eval_loss": 2.323822498321533, + "eval_runtime": 48.3307, + "eval_samples_per_second": 20.691, + "eval_steps_per_second": 0.869, + "step": 700 + }, + { + "epoch": 0.7682191780821918, + "grad_norm": 2.321021556854248, + "learning_rate": 4.161251015434607e-06, + "loss": 2.1758, + "step": 701 + }, + { + "epoch": 0.7693150684931507, + "grad_norm": 1.8213255405426025, + "learning_rate": 4.159220146222583e-06, + "loss": 2.2002, + "step": 702 + }, + { + "epoch": 0.7704109589041096, + "grad_norm": 1.8925111293792725, + "learning_rate": 4.157189277010561e-06, + "loss": 2.1575, + "step": 703 + }, + { + "epoch": 0.7715068493150685, + "grad_norm": 1.8050175905227661, + "learning_rate": 4.155158407798538e-06, + "loss": 2.179, + "step": 704 + }, + { + "epoch": 0.7726027397260274, + "grad_norm": 1.873228907585144, + "learning_rate": 4.153127538586516e-06, + "loss": 2.2232, + "step": 705 + }, + { + "epoch": 0.7736986301369863, + "grad_norm": 1.954910397529602, + "learning_rate": 4.151096669374492e-06, + "loss": 2.1992, + "step": 706 + }, + { + "epoch": 0.7747945205479452, + "grad_norm": 1.8913410902023315, + "learning_rate": 4.1490658001624705e-06, + "loss": 2.2086, + "step": 707 + }, + { + "epoch": 0.7758904109589041, + "grad_norm": 1.8883898258209229, + "learning_rate": 4.147034930950447e-06, + "loss": 2.1801, + "step": 708 + }, + { + "epoch": 0.776986301369863, + "grad_norm": 2.054351806640625, + "learning_rate": 4.145004061738425e-06, + "loss": 2.1908, + "step": 709 + }, + { + "epoch": 0.7780821917808219, + "grad_norm": 2.029036045074463, + "learning_rate": 4.1429731925264015e-06, + "loss": 2.1891, + "step": 710 + }, + { + "epoch": 0.7780821917808219, + "eval_loss": 2.3196680545806885, + "eval_runtime": 42.8326, + "eval_samples_per_second": 23.347, + "eval_steps_per_second": 0.981, + "step": 710 + }, + { + "epoch": 0.7791780821917809, + "grad_norm": 2.27848219871521, + "learning_rate": 4.140942323314379e-06, + "loss": 2.1165, + "step": 711 + }, + { + "epoch": 0.7802739726027397, + "grad_norm": 2.134625196456909, + "learning_rate": 4.138911454102356e-06, + "loss": 2.1508, + "step": 712 + }, + { + "epoch": 0.7813698630136986, + "grad_norm": 2.103895902633667, + "learning_rate": 4.136880584890333e-06, + "loss": 2.1746, + "step": 713 + }, + { + "epoch": 0.7824657534246575, + "grad_norm": 1.713189721107483, + "learning_rate": 4.1348497156783106e-06, + "loss": 2.2023, + "step": 714 + }, + { + "epoch": 0.7835616438356164, + "grad_norm": 2.0390899181365967, + "learning_rate": 4.132818846466288e-06, + "loss": 2.2039, + "step": 715 + }, + { + "epoch": 0.7846575342465754, + "grad_norm": 1.9643651247024536, + "learning_rate": 4.130787977254265e-06, + "loss": 2.199, + "step": 716 + }, + { + "epoch": 0.7857534246575343, + "grad_norm": 1.9173564910888672, + "learning_rate": 4.128757108042242e-06, + "loss": 2.2166, + "step": 717 + }, + { + "epoch": 0.7868493150684932, + "grad_norm": 1.7923007011413574, + "learning_rate": 4.12672623883022e-06, + "loss": 2.2144, + "step": 718 + }, + { + "epoch": 0.787945205479452, + "grad_norm": 2.2235090732574463, + "learning_rate": 4.124695369618197e-06, + "loss": 2.2586, + "step": 719 + }, + { + "epoch": 0.7890410958904109, + "grad_norm": 1.780316710472107, + "learning_rate": 4.122664500406174e-06, + "loss": 2.1862, + "step": 720 + }, + { + "epoch": 0.7890410958904109, + "eval_loss": 2.318016767501831, + "eval_runtime": 50.5082, + "eval_samples_per_second": 19.799, + "eval_steps_per_second": 0.832, + "step": 720 + }, + { + "epoch": 0.7901369863013699, + "grad_norm": 2.0788493156433105, + "learning_rate": 4.1206336311941515e-06, + "loss": 2.1845, + "step": 721 + }, + { + "epoch": 0.7912328767123288, + "grad_norm": 1.9053714275360107, + "learning_rate": 4.118602761982129e-06, + "loss": 2.2005, + "step": 722 + }, + { + "epoch": 0.7923287671232877, + "grad_norm": 1.9328513145446777, + "learning_rate": 4.116571892770106e-06, + "loss": 2.1475, + "step": 723 + }, + { + "epoch": 0.7934246575342466, + "grad_norm": 2.059760570526123, + "learning_rate": 4.114541023558083e-06, + "loss": 2.1386, + "step": 724 + }, + { + "epoch": 0.7945205479452054, + "grad_norm": 2.0093464851379395, + "learning_rate": 4.112510154346061e-06, + "loss": 2.2242, + "step": 725 + }, + { + "epoch": 0.7956164383561644, + "grad_norm": 1.8513909578323364, + "learning_rate": 4.110479285134038e-06, + "loss": 2.178, + "step": 726 + }, + { + "epoch": 0.7967123287671233, + "grad_norm": 1.9481958150863647, + "learning_rate": 4.108448415922015e-06, + "loss": 2.1961, + "step": 727 + }, + { + "epoch": 0.7978082191780822, + "grad_norm": 1.9229767322540283, + "learning_rate": 4.1064175467099924e-06, + "loss": 2.2116, + "step": 728 + }, + { + "epoch": 0.7989041095890411, + "grad_norm": 1.9819737672805786, + "learning_rate": 4.10438667749797e-06, + "loss": 2.2097, + "step": 729 + }, + { + "epoch": 0.8, + "grad_norm": 1.9382416009902954, + "learning_rate": 4.102355808285947e-06, + "loss": 2.1099, + "step": 730 + }, + { + "epoch": 0.8, + "eval_loss": 2.31795334815979, + "eval_runtime": 51.8961, + "eval_samples_per_second": 19.269, + "eval_steps_per_second": 0.809, + "step": 730 + }, + { + "epoch": 0.8010958904109589, + "grad_norm": 2.006890058517456, + "learning_rate": 4.100324939073924e-06, + "loss": 2.2606, + "step": 731 + }, + { + "epoch": 0.8021917808219178, + "grad_norm": 2.164473295211792, + "learning_rate": 4.0982940698619015e-06, + "loss": 2.1869, + "step": 732 + }, + { + "epoch": 0.8032876712328767, + "grad_norm": 2.04368257522583, + "learning_rate": 4.096263200649879e-06, + "loss": 2.1846, + "step": 733 + }, + { + "epoch": 0.8043835616438356, + "grad_norm": 1.8363912105560303, + "learning_rate": 4.094232331437856e-06, + "loss": 2.1938, + "step": 734 + }, + { + "epoch": 0.8054794520547945, + "grad_norm": 1.9260189533233643, + "learning_rate": 4.092201462225833e-06, + "loss": 2.1682, + "step": 735 + }, + { + "epoch": 0.8065753424657535, + "grad_norm": 2.0688912868499756, + "learning_rate": 4.090170593013811e-06, + "loss": 2.1772, + "step": 736 + }, + { + "epoch": 0.8076712328767123, + "grad_norm": 1.859474539756775, + "learning_rate": 4.088139723801787e-06, + "loss": 2.2013, + "step": 737 + }, + { + "epoch": 0.8087671232876712, + "grad_norm": 2.0915753841400146, + "learning_rate": 4.086108854589765e-06, + "loss": 2.1519, + "step": 738 + }, + { + "epoch": 0.8098630136986301, + "grad_norm": 2.2019519805908203, + "learning_rate": 4.084077985377742e-06, + "loss": 2.1608, + "step": 739 + }, + { + "epoch": 0.810958904109589, + "grad_norm": 1.8727316856384277, + "learning_rate": 4.08204711616572e-06, + "loss": 2.1823, + "step": 740 + }, + { + "epoch": 0.810958904109589, + "eval_loss": 2.318638324737549, + "eval_runtime": 52.3669, + "eval_samples_per_second": 19.096, + "eval_steps_per_second": 0.802, + "step": 740 + }, + { + "epoch": 0.812054794520548, + "grad_norm": 2.059250831604004, + "learning_rate": 4.080016246953696e-06, + "loss": 2.2285, + "step": 741 + }, + { + "epoch": 0.8131506849315069, + "grad_norm": 1.9466749429702759, + "learning_rate": 4.077985377741674e-06, + "loss": 2.1496, + "step": 742 + }, + { + "epoch": 0.8142465753424658, + "grad_norm": 1.8967524766921997, + "learning_rate": 4.075954508529651e-06, + "loss": 2.2066, + "step": 743 + }, + { + "epoch": 0.8153424657534246, + "grad_norm": 1.976780891418457, + "learning_rate": 4.073923639317629e-06, + "loss": 2.2194, + "step": 744 + }, + { + "epoch": 0.8164383561643835, + "grad_norm": 2.077683687210083, + "learning_rate": 4.071892770105605e-06, + "loss": 2.2209, + "step": 745 + }, + { + "epoch": 0.8175342465753425, + "grad_norm": 1.9075323343276978, + "learning_rate": 4.069861900893583e-06, + "loss": 2.1727, + "step": 746 + }, + { + "epoch": 0.8186301369863014, + "grad_norm": 1.6940832138061523, + "learning_rate": 4.06783103168156e-06, + "loss": 2.1437, + "step": 747 + }, + { + "epoch": 0.8197260273972603, + "grad_norm": 1.9035768508911133, + "learning_rate": 4.065800162469538e-06, + "loss": 2.165, + "step": 748 + }, + { + "epoch": 0.8208219178082192, + "grad_norm": 1.8775745630264282, + "learning_rate": 4.063769293257514e-06, + "loss": 2.1757, + "step": 749 + }, + { + "epoch": 0.821917808219178, + "grad_norm": 1.9622418880462646, + "learning_rate": 4.061738424045492e-06, + "loss": 2.197, + "step": 750 + }, + { + "epoch": 0.821917808219178, + "eval_loss": 2.317519426345825, + "eval_runtime": 50.2699, + "eval_samples_per_second": 19.893, + "eval_steps_per_second": 0.835, + "step": 750 + }, + { + "epoch": 0.823013698630137, + "grad_norm": 1.9646214246749878, + "learning_rate": 4.059707554833469e-06, + "loss": 2.1783, + "step": 751 + }, + { + "epoch": 0.8241095890410959, + "grad_norm": 1.890342354774475, + "learning_rate": 4.057676685621446e-06, + "loss": 2.1925, + "step": 752 + }, + { + "epoch": 0.8252054794520548, + "grad_norm": 1.8977251052856445, + "learning_rate": 4.0556458164094235e-06, + "loss": 2.219, + "step": 753 + }, + { + "epoch": 0.8263013698630137, + "grad_norm": 1.8251665830612183, + "learning_rate": 4.053614947197401e-06, + "loss": 2.189, + "step": 754 + }, + { + "epoch": 0.8273972602739726, + "grad_norm": 1.7953882217407227, + "learning_rate": 4.051584077985378e-06, + "loss": 2.1237, + "step": 755 + }, + { + "epoch": 0.8284931506849315, + "grad_norm": 1.8247836828231812, + "learning_rate": 4.049553208773355e-06, + "loss": 2.1962, + "step": 756 + }, + { + "epoch": 0.8295890410958904, + "grad_norm": 1.8537967205047607, + "learning_rate": 4.047522339561333e-06, + "loss": 2.2201, + "step": 757 + }, + { + "epoch": 0.8306849315068493, + "grad_norm": 1.865315318107605, + "learning_rate": 4.04549147034931e-06, + "loss": 2.1819, + "step": 758 + }, + { + "epoch": 0.8317808219178082, + "grad_norm": 1.8620522022247314, + "learning_rate": 4.043460601137287e-06, + "loss": 2.1717, + "step": 759 + }, + { + "epoch": 0.8328767123287671, + "grad_norm": 1.9475384950637817, + "learning_rate": 4.041429731925264e-06, + "loss": 2.2718, + "step": 760 + }, + { + "epoch": 0.8328767123287671, + "eval_loss": 2.314429759979248, + "eval_runtime": 54.2324, + "eval_samples_per_second": 18.439, + "eval_steps_per_second": 0.774, + "step": 760 + }, + { + "epoch": 0.8339726027397261, + "grad_norm": 1.7170753479003906, + "learning_rate": 4.039398862713242e-06, + "loss": 2.1785, + "step": 761 + }, + { + "epoch": 0.8350684931506849, + "grad_norm": 1.968220591545105, + "learning_rate": 4.037367993501219e-06, + "loss": 2.1798, + "step": 762 + }, + { + "epoch": 0.8361643835616438, + "grad_norm": 1.7927111387252808, + "learning_rate": 4.035337124289196e-06, + "loss": 2.2372, + "step": 763 + }, + { + "epoch": 0.8372602739726027, + "grad_norm": 1.7871185541152954, + "learning_rate": 4.0333062550771735e-06, + "loss": 2.182, + "step": 764 + }, + { + "epoch": 0.8383561643835616, + "grad_norm": 1.9415812492370605, + "learning_rate": 4.031275385865151e-06, + "loss": 2.1527, + "step": 765 + }, + { + "epoch": 0.8394520547945206, + "grad_norm": 1.7327690124511719, + "learning_rate": 4.029244516653128e-06, + "loss": 2.1744, + "step": 766 + }, + { + "epoch": 0.8405479452054795, + "grad_norm": 1.9123411178588867, + "learning_rate": 4.027213647441105e-06, + "loss": 2.2049, + "step": 767 + }, + { + "epoch": 0.8416438356164384, + "grad_norm": 2.011591911315918, + "learning_rate": 4.025182778229083e-06, + "loss": 2.1391, + "step": 768 + }, + { + "epoch": 0.8427397260273972, + "grad_norm": 1.9169542789459229, + "learning_rate": 4.02315190901706e-06, + "loss": 2.1562, + "step": 769 + }, + { + "epoch": 0.8438356164383561, + "grad_norm": 1.8020051717758179, + "learning_rate": 4.021121039805037e-06, + "loss": 2.1795, + "step": 770 + }, + { + "epoch": 0.8438356164383561, + "eval_loss": 2.3199150562286377, + "eval_runtime": 52.468, + "eval_samples_per_second": 19.059, + "eval_steps_per_second": 0.8, + "step": 770 + }, + { + "epoch": 0.8449315068493151, + "grad_norm": 1.927666425704956, + "learning_rate": 4.0190901705930144e-06, + "loss": 2.1834, + "step": 771 + }, + { + "epoch": 0.846027397260274, + "grad_norm": 2.0998382568359375, + "learning_rate": 4.017059301380992e-06, + "loss": 2.2541, + "step": 772 + }, + { + "epoch": 0.8471232876712329, + "grad_norm": 1.88541579246521, + "learning_rate": 4.015028432168969e-06, + "loss": 2.1631, + "step": 773 + }, + { + "epoch": 0.8482191780821918, + "grad_norm": 1.9546253681182861, + "learning_rate": 4.012997562956945e-06, + "loss": 2.2266, + "step": 774 + }, + { + "epoch": 0.8493150684931506, + "grad_norm": 1.7969225645065308, + "learning_rate": 4.0109666937449235e-06, + "loss": 2.1526, + "step": 775 + }, + { + "epoch": 0.8504109589041096, + "grad_norm": 1.7237955331802368, + "learning_rate": 4.0089358245329e-06, + "loss": 2.1618, + "step": 776 + }, + { + "epoch": 0.8515068493150685, + "grad_norm": 1.8355857133865356, + "learning_rate": 4.006904955320878e-06, + "loss": 2.1673, + "step": 777 + }, + { + "epoch": 0.8526027397260274, + "grad_norm": 1.8049591779708862, + "learning_rate": 4.0048740861088545e-06, + "loss": 2.1676, + "step": 778 + }, + { + "epoch": 0.8536986301369863, + "grad_norm": 1.8143601417541504, + "learning_rate": 4.002843216896833e-06, + "loss": 2.214, + "step": 779 + }, + { + "epoch": 0.8547945205479452, + "grad_norm": 1.8344221115112305, + "learning_rate": 4.000812347684809e-06, + "loss": 2.1958, + "step": 780 + }, + { + "epoch": 0.8547945205479452, + "eval_loss": 2.314359426498413, + "eval_runtime": 51.8717, + "eval_samples_per_second": 19.278, + "eval_steps_per_second": 0.81, + "step": 780 + }, + { + "epoch": 0.8558904109589041, + "grad_norm": 1.944402813911438, + "learning_rate": 3.998781478472787e-06, + "loss": 2.1497, + "step": 781 + }, + { + "epoch": 0.856986301369863, + "grad_norm": 1.93898606300354, + "learning_rate": 3.996750609260764e-06, + "loss": 2.1737, + "step": 782 + }, + { + "epoch": 0.8580821917808219, + "grad_norm": 1.9499506950378418, + "learning_rate": 3.994719740048742e-06, + "loss": 2.1545, + "step": 783 + }, + { + "epoch": 0.8591780821917808, + "grad_norm": 1.9311052560806274, + "learning_rate": 3.992688870836718e-06, + "loss": 2.2559, + "step": 784 + }, + { + "epoch": 0.8602739726027397, + "grad_norm": 1.864862084388733, + "learning_rate": 3.990658001624696e-06, + "loss": 2.1996, + "step": 785 + }, + { + "epoch": 0.8613698630136987, + "grad_norm": 1.8248569965362549, + "learning_rate": 3.988627132412673e-06, + "loss": 2.1591, + "step": 786 + }, + { + "epoch": 0.8624657534246576, + "grad_norm": 2.2370970249176025, + "learning_rate": 3.98659626320065e-06, + "loss": 2.2095, + "step": 787 + }, + { + "epoch": 0.8635616438356164, + "grad_norm": 1.8877084255218506, + "learning_rate": 3.984565393988627e-06, + "loss": 2.1813, + "step": 788 + }, + { + "epoch": 0.8646575342465753, + "grad_norm": 2.1722517013549805, + "learning_rate": 3.9825345247766046e-06, + "loss": 2.1866, + "step": 789 + }, + { + "epoch": 0.8657534246575342, + "grad_norm": 1.9640655517578125, + "learning_rate": 3.980503655564582e-06, + "loss": 2.1818, + "step": 790 + }, + { + "epoch": 0.8657534246575342, + "eval_loss": 2.314087390899658, + "eval_runtime": 51.9778, + "eval_samples_per_second": 19.239, + "eval_steps_per_second": 0.808, + "step": 790 + }, + { + "epoch": 0.8668493150684932, + "grad_norm": 2.0583789348602295, + "learning_rate": 3.978472786352559e-06, + "loss": 2.1956, + "step": 791 + }, + { + "epoch": 0.8679452054794521, + "grad_norm": 2.2128682136535645, + "learning_rate": 3.976441917140536e-06, + "loss": 2.204, + "step": 792 + }, + { + "epoch": 0.869041095890411, + "grad_norm": 1.917763113975525, + "learning_rate": 3.974411047928514e-06, + "loss": 2.1841, + "step": 793 + }, + { + "epoch": 0.8701369863013698, + "grad_norm": 2.0327019691467285, + "learning_rate": 3.972380178716491e-06, + "loss": 2.2351, + "step": 794 + }, + { + "epoch": 0.8712328767123287, + "grad_norm": 1.9498991966247559, + "learning_rate": 3.970349309504468e-06, + "loss": 2.2074, + "step": 795 + }, + { + "epoch": 0.8723287671232877, + "grad_norm": 1.8105775117874146, + "learning_rate": 3.9683184402924455e-06, + "loss": 2.2088, + "step": 796 + }, + { + "epoch": 0.8734246575342466, + "grad_norm": 1.9299930334091187, + "learning_rate": 3.966287571080423e-06, + "loss": 2.1469, + "step": 797 + }, + { + "epoch": 0.8745205479452055, + "grad_norm": 1.9346121549606323, + "learning_rate": 3.9642567018684e-06, + "loss": 2.1434, + "step": 798 + }, + { + "epoch": 0.8756164383561644, + "grad_norm": 1.8202977180480957, + "learning_rate": 3.962225832656377e-06, + "loss": 2.194, + "step": 799 + }, + { + "epoch": 0.8767123287671232, + "grad_norm": 1.6922450065612793, + "learning_rate": 3.960194963444355e-06, + "loss": 2.1821, + "step": 800 + }, + { + "epoch": 0.8767123287671232, + "eval_loss": 2.310209035873413, + "eval_runtime": 46.1102, + "eval_samples_per_second": 21.687, + "eval_steps_per_second": 0.911, + "step": 800 + }, + { + "epoch": 0.8778082191780822, + "grad_norm": 1.9258904457092285, + "learning_rate": 3.958164094232332e-06, + "loss": 2.1326, + "step": 801 + }, + { + "epoch": 0.8789041095890411, + "grad_norm": 1.8485561609268188, + "learning_rate": 3.956133225020309e-06, + "loss": 2.1888, + "step": 802 + }, + { + "epoch": 0.88, + "grad_norm": 1.9802418947219849, + "learning_rate": 3.954102355808286e-06, + "loss": 2.1876, + "step": 803 + }, + { + "epoch": 0.8810958904109589, + "grad_norm": 1.8365259170532227, + "learning_rate": 3.952071486596264e-06, + "loss": 2.1804, + "step": 804 + }, + { + "epoch": 0.8821917808219178, + "grad_norm": 2.023909568786621, + "learning_rate": 3.950040617384241e-06, + "loss": 2.1801, + "step": 805 + }, + { + "epoch": 0.8832876712328767, + "grad_norm": 1.8889018297195435, + "learning_rate": 3.948009748172218e-06, + "loss": 2.1657, + "step": 806 + }, + { + "epoch": 0.8843835616438356, + "grad_norm": 1.7983578443527222, + "learning_rate": 3.9459788789601955e-06, + "loss": 2.1589, + "step": 807 + }, + { + "epoch": 0.8854794520547945, + "grad_norm": 2.157043218612671, + "learning_rate": 3.943948009748173e-06, + "loss": 2.2258, + "step": 808 + }, + { + "epoch": 0.8865753424657534, + "grad_norm": 1.7387373447418213, + "learning_rate": 3.94191714053615e-06, + "loss": 2.1878, + "step": 809 + }, + { + "epoch": 0.8876712328767123, + "grad_norm": 1.963233232498169, + "learning_rate": 3.939886271324127e-06, + "loss": 2.186, + "step": 810 + }, + { + "epoch": 0.8876712328767123, + "eval_loss": 2.306187152862549, + "eval_runtime": 42.563, + "eval_samples_per_second": 23.495, + "eval_steps_per_second": 0.987, + "step": 810 + }, + { + "epoch": 0.8887671232876713, + "grad_norm": 1.8932961225509644, + "learning_rate": 3.937855402112104e-06, + "loss": 2.198, + "step": 811 + }, + { + "epoch": 0.8898630136986302, + "grad_norm": 2.0431783199310303, + "learning_rate": 3.935824532900082e-06, + "loss": 2.1505, + "step": 812 + }, + { + "epoch": 0.890958904109589, + "grad_norm": 1.8954466581344604, + "learning_rate": 3.933793663688058e-06, + "loss": 2.1494, + "step": 813 + }, + { + "epoch": 0.8920547945205479, + "grad_norm": 1.841184139251709, + "learning_rate": 3.9317627944760365e-06, + "loss": 2.1932, + "step": 814 + }, + { + "epoch": 0.8931506849315068, + "grad_norm": 1.9083219766616821, + "learning_rate": 3.929731925264013e-06, + "loss": 2.2081, + "step": 815 + }, + { + "epoch": 0.8942465753424658, + "grad_norm": 1.9737226963043213, + "learning_rate": 3.927701056051991e-06, + "loss": 2.1405, + "step": 816 + }, + { + "epoch": 0.8953424657534247, + "grad_norm": 1.869194746017456, + "learning_rate": 3.9256701868399674e-06, + "loss": 2.1642, + "step": 817 + }, + { + "epoch": 0.8964383561643836, + "grad_norm": 1.9553205966949463, + "learning_rate": 3.9236393176279456e-06, + "loss": 2.1761, + "step": 818 + }, + { + "epoch": 0.8975342465753424, + "grad_norm": 1.8355534076690674, + "learning_rate": 3.921608448415922e-06, + "loss": 2.1551, + "step": 819 + }, + { + "epoch": 0.8986301369863013, + "grad_norm": 1.9770171642303467, + "learning_rate": 3.9195775792039e-06, + "loss": 2.1424, + "step": 820 + }, + { + "epoch": 0.8986301369863013, + "eval_loss": 2.307065010070801, + "eval_runtime": 44.5816, + "eval_samples_per_second": 22.431, + "eval_steps_per_second": 0.942, + "step": 820 + }, + { + "epoch": 0.8997260273972603, + "grad_norm": 1.7375415563583374, + "learning_rate": 3.9175467099918765e-06, + "loss": 2.1895, + "step": 821 + }, + { + "epoch": 0.9008219178082192, + "grad_norm": 1.8910484313964844, + "learning_rate": 3.915515840779855e-06, + "loss": 2.1685, + "step": 822 + }, + { + "epoch": 0.9019178082191781, + "grad_norm": 1.9191482067108154, + "learning_rate": 3.913484971567831e-06, + "loss": 2.2074, + "step": 823 + }, + { + "epoch": 0.903013698630137, + "grad_norm": 1.7864760160446167, + "learning_rate": 3.911454102355808e-06, + "loss": 2.1972, + "step": 824 + }, + { + "epoch": 0.9041095890410958, + "grad_norm": 1.9044650793075562, + "learning_rate": 3.909423233143786e-06, + "loss": 2.1237, + "step": 825 + }, + { + "epoch": 0.9052054794520548, + "grad_norm": 1.8378455638885498, + "learning_rate": 3.907392363931763e-06, + "loss": 2.1907, + "step": 826 + }, + { + "epoch": 0.9063013698630137, + "grad_norm": 1.8708094358444214, + "learning_rate": 3.90536149471974e-06, + "loss": 2.1883, + "step": 827 + }, + { + "epoch": 0.9073972602739726, + "grad_norm": 1.7802201509475708, + "learning_rate": 3.9033306255077175e-06, + "loss": 2.2356, + "step": 828 + }, + { + "epoch": 0.9084931506849315, + "grad_norm": 1.9374496936798096, + "learning_rate": 3.901299756295695e-06, + "loss": 2.2546, + "step": 829 + }, + { + "epoch": 0.9095890410958904, + "grad_norm": 1.9246270656585693, + "learning_rate": 3.899268887083672e-06, + "loss": 2.2684, + "step": 830 + }, + { + "epoch": 0.9095890410958904, + "eval_loss": 2.3065874576568604, + "eval_runtime": 45.18, + "eval_samples_per_second": 22.134, + "eval_steps_per_second": 0.93, + "step": 830 + }, + { + "epoch": 0.9106849315068493, + "grad_norm": 1.8681883811950684, + "learning_rate": 3.897238017871649e-06, + "loss": 2.1737, + "step": 831 + }, + { + "epoch": 0.9117808219178082, + "grad_norm": 1.9293930530548096, + "learning_rate": 3.8952071486596266e-06, + "loss": 2.1813, + "step": 832 + }, + { + "epoch": 0.9128767123287671, + "grad_norm": 1.844586968421936, + "learning_rate": 3.893176279447604e-06, + "loss": 2.1788, + "step": 833 + }, + { + "epoch": 0.913972602739726, + "grad_norm": 1.7658660411834717, + "learning_rate": 3.891145410235581e-06, + "loss": 2.1494, + "step": 834 + }, + { + "epoch": 0.915068493150685, + "grad_norm": 2.05153751373291, + "learning_rate": 3.889114541023558e-06, + "loss": 2.241, + "step": 835 + }, + { + "epoch": 0.9161643835616439, + "grad_norm": 1.7869952917099, + "learning_rate": 3.887083671811536e-06, + "loss": 2.1866, + "step": 836 + }, + { + "epoch": 0.9172602739726028, + "grad_norm": 1.99534273147583, + "learning_rate": 3.885052802599513e-06, + "loss": 2.1682, + "step": 837 + }, + { + "epoch": 0.9183561643835616, + "grad_norm": 1.9593979120254517, + "learning_rate": 3.88302193338749e-06, + "loss": 2.1357, + "step": 838 + }, + { + "epoch": 0.9194520547945205, + "grad_norm": 1.8441778421401978, + "learning_rate": 3.8809910641754675e-06, + "loss": 2.2283, + "step": 839 + }, + { + "epoch": 0.9205479452054794, + "grad_norm": 1.9153512716293335, + "learning_rate": 3.878960194963445e-06, + "loss": 2.1899, + "step": 840 + }, + { + "epoch": 0.9205479452054794, + "eval_loss": 2.307375907897949, + "eval_runtime": 48.0216, + "eval_samples_per_second": 20.824, + "eval_steps_per_second": 0.875, + "step": 840 + }, + { + "epoch": 0.9216438356164384, + "grad_norm": 2.004232883453369, + "learning_rate": 3.876929325751422e-06, + "loss": 2.173, + "step": 841 + }, + { + "epoch": 0.9227397260273973, + "grad_norm": 2.12992262840271, + "learning_rate": 3.874898456539399e-06, + "loss": 2.2164, + "step": 842 + }, + { + "epoch": 0.9238356164383562, + "grad_norm": 2.0505411624908447, + "learning_rate": 3.872867587327377e-06, + "loss": 2.1432, + "step": 843 + }, + { + "epoch": 0.924931506849315, + "grad_norm": 1.767324686050415, + "learning_rate": 3.870836718115354e-06, + "loss": 2.1838, + "step": 844 + }, + { + "epoch": 0.9260273972602739, + "grad_norm": 2.189107894897461, + "learning_rate": 3.868805848903331e-06, + "loss": 2.1623, + "step": 845 + }, + { + "epoch": 0.9271232876712329, + "grad_norm": 1.9781970977783203, + "learning_rate": 3.8667749796913084e-06, + "loss": 2.1738, + "step": 846 + }, + { + "epoch": 0.9282191780821918, + "grad_norm": 1.8734674453735352, + "learning_rate": 3.864744110479286e-06, + "loss": 2.1829, + "step": 847 + }, + { + "epoch": 0.9293150684931507, + "grad_norm": 1.7699538469314575, + "learning_rate": 3.862713241267262e-06, + "loss": 2.186, + "step": 848 + }, + { + "epoch": 0.9304109589041096, + "grad_norm": 1.8354437351226807, + "learning_rate": 3.86068237205524e-06, + "loss": 2.2149, + "step": 849 + }, + { + "epoch": 0.9315068493150684, + "grad_norm": 1.913249135017395, + "learning_rate": 3.858651502843217e-06, + "loss": 2.2208, + "step": 850 + }, + { + "epoch": 0.9315068493150684, + "eval_loss": 2.307549238204956, + "eval_runtime": 43.1706, + "eval_samples_per_second": 23.164, + "eval_steps_per_second": 0.973, + "step": 850 + }, + { + "epoch": 0.9326027397260274, + "grad_norm": 1.92095148563385, + "learning_rate": 3.856620633631195e-06, + "loss": 2.1839, + "step": 851 + }, + { + "epoch": 0.9336986301369863, + "grad_norm": 1.8833959102630615, + "learning_rate": 3.854589764419171e-06, + "loss": 2.1372, + "step": 852 + }, + { + "epoch": 0.9347945205479452, + "grad_norm": 1.9410737752914429, + "learning_rate": 3.852558895207149e-06, + "loss": 2.1726, + "step": 853 + }, + { + "epoch": 0.9358904109589041, + "grad_norm": 1.8328828811645508, + "learning_rate": 3.850528025995126e-06, + "loss": 2.2425, + "step": 854 + }, + { + "epoch": 0.936986301369863, + "grad_norm": 1.89652419090271, + "learning_rate": 3.848497156783104e-06, + "loss": 2.1956, + "step": 855 + }, + { + "epoch": 0.938082191780822, + "grad_norm": 1.9026265144348145, + "learning_rate": 3.84646628757108e-06, + "loss": 2.167, + "step": 856 + }, + { + "epoch": 0.9391780821917808, + "grad_norm": 2.010864734649658, + "learning_rate": 3.8444354183590585e-06, + "loss": 2.1725, + "step": 857 + }, + { + "epoch": 0.9402739726027397, + "grad_norm": 2.293717384338379, + "learning_rate": 3.842404549147035e-06, + "loss": 2.1668, + "step": 858 + }, + { + "epoch": 0.9413698630136986, + "grad_norm": 1.8209116458892822, + "learning_rate": 3.840373679935013e-06, + "loss": 2.1649, + "step": 859 + }, + { + "epoch": 0.9424657534246575, + "grad_norm": 2.061277389526367, + "learning_rate": 3.8383428107229894e-06, + "loss": 2.1479, + "step": 860 + }, + { + "epoch": 0.9424657534246575, + "eval_loss": 2.3073670864105225, + "eval_runtime": 44.3781, + "eval_samples_per_second": 22.534, + "eval_steps_per_second": 0.946, + "step": 860 + }, + { + "epoch": 0.9435616438356165, + "grad_norm": 1.8032523393630981, + "learning_rate": 3.836311941510967e-06, + "loss": 2.1665, + "step": 861 + }, + { + "epoch": 0.9446575342465754, + "grad_norm": 1.989533543586731, + "learning_rate": 3.834281072298944e-06, + "loss": 2.1612, + "step": 862 + }, + { + "epoch": 0.9457534246575342, + "grad_norm": 2.017179012298584, + "learning_rate": 3.832250203086921e-06, + "loss": 2.201, + "step": 863 + }, + { + "epoch": 0.9468493150684931, + "grad_norm": 2.0291004180908203, + "learning_rate": 3.8302193338748985e-06, + "loss": 2.1508, + "step": 864 + }, + { + "epoch": 0.947945205479452, + "grad_norm": 1.8693255186080933, + "learning_rate": 3.828188464662876e-06, + "loss": 2.2106, + "step": 865 + }, + { + "epoch": 0.949041095890411, + "grad_norm": 1.8890169858932495, + "learning_rate": 3.826157595450853e-06, + "loss": 2.2467, + "step": 866 + }, + { + "epoch": 0.9501369863013699, + "grad_norm": 2.0550642013549805, + "learning_rate": 3.82412672623883e-06, + "loss": 2.1899, + "step": 867 + }, + { + "epoch": 0.9512328767123288, + "grad_norm": 1.8681174516677856, + "learning_rate": 3.822095857026808e-06, + "loss": 2.1428, + "step": 868 + }, + { + "epoch": 0.9523287671232876, + "grad_norm": 1.9133951663970947, + "learning_rate": 3.820064987814785e-06, + "loss": 2.1947, + "step": 869 + }, + { + "epoch": 0.9534246575342465, + "grad_norm": 1.8717621564865112, + "learning_rate": 3.818034118602762e-06, + "loss": 2.1542, + "step": 870 + }, + { + "epoch": 0.9534246575342465, + "eval_loss": 2.307584047317505, + "eval_runtime": 43.718, + "eval_samples_per_second": 22.874, + "eval_steps_per_second": 0.961, + "step": 870 + }, + { + "epoch": 0.9545205479452055, + "grad_norm": 1.9843461513519287, + "learning_rate": 3.8160032493907395e-06, + "loss": 2.1547, + "step": 871 + }, + { + "epoch": 0.9556164383561644, + "grad_norm": 1.725468397140503, + "learning_rate": 3.813972380178717e-06, + "loss": 2.1679, + "step": 872 + }, + { + "epoch": 0.9567123287671233, + "grad_norm": 1.8076086044311523, + "learning_rate": 3.811941510966694e-06, + "loss": 2.1654, + "step": 873 + }, + { + "epoch": 0.9578082191780822, + "grad_norm": 1.8879774808883667, + "learning_rate": 3.809910641754671e-06, + "loss": 2.191, + "step": 874 + }, + { + "epoch": 0.958904109589041, + "grad_norm": 1.943313717842102, + "learning_rate": 3.8078797725426486e-06, + "loss": 2.1448, + "step": 875 + }, + { + "epoch": 0.96, + "grad_norm": 1.8877806663513184, + "learning_rate": 3.8058489033306254e-06, + "loss": 2.1129, + "step": 876 + }, + { + "epoch": 0.9610958904109589, + "grad_norm": 1.787083625793457, + "learning_rate": 3.803818034118603e-06, + "loss": 2.1854, + "step": 877 + }, + { + "epoch": 0.9621917808219178, + "grad_norm": 2.0925140380859375, + "learning_rate": 3.80178716490658e-06, + "loss": 2.1614, + "step": 878 + }, + { + "epoch": 0.9632876712328767, + "grad_norm": 1.9043725728988647, + "learning_rate": 3.7997562956945577e-06, + "loss": 2.2161, + "step": 879 + }, + { + "epoch": 0.9643835616438357, + "grad_norm": 1.8173545598983765, + "learning_rate": 3.7977254264825345e-06, + "loss": 2.2067, + "step": 880 + }, + { + "epoch": 0.9643835616438357, + "eval_loss": 2.302727460861206, + "eval_runtime": 44.2015, + "eval_samples_per_second": 22.624, + "eval_steps_per_second": 0.95, + "step": 880 + }, + { + "epoch": 0.9654794520547946, + "grad_norm": 1.9871562719345093, + "learning_rate": 3.7956945572705122e-06, + "loss": 2.1658, + "step": 881 + }, + { + "epoch": 0.9665753424657534, + "grad_norm": 2.103553295135498, + "learning_rate": 3.793663688058489e-06, + "loss": 2.1468, + "step": 882 + }, + { + "epoch": 0.9676712328767123, + "grad_norm": 1.9089381694793701, + "learning_rate": 3.791632818846467e-06, + "loss": 2.2289, + "step": 883 + }, + { + "epoch": 0.9687671232876712, + "grad_norm": 2.3238179683685303, + "learning_rate": 3.7896019496344436e-06, + "loss": 2.2179, + "step": 884 + }, + { + "epoch": 0.9698630136986301, + "grad_norm": 1.902161955833435, + "learning_rate": 3.787571080422421e-06, + "loss": 2.2086, + "step": 885 + }, + { + "epoch": 0.9709589041095891, + "grad_norm": 1.8587497472763062, + "learning_rate": 3.785540211210398e-06, + "loss": 2.2175, + "step": 886 + }, + { + "epoch": 0.972054794520548, + "grad_norm": 2.029719829559326, + "learning_rate": 3.7835093419983755e-06, + "loss": 2.207, + "step": 887 + }, + { + "epoch": 0.9731506849315068, + "grad_norm": 2.0228402614593506, + "learning_rate": 3.7814784727863527e-06, + "loss": 2.1851, + "step": 888 + }, + { + "epoch": 0.9742465753424657, + "grad_norm": 1.9885436296463013, + "learning_rate": 3.77944760357433e-06, + "loss": 2.1667, + "step": 889 + }, + { + "epoch": 0.9753424657534246, + "grad_norm": 1.8621459007263184, + "learning_rate": 3.7774167343623073e-06, + "loss": 2.2529, + "step": 890 + }, + { + "epoch": 0.9753424657534246, + "eval_loss": 2.303697347640991, + "eval_runtime": 44.9459, + "eval_samples_per_second": 22.249, + "eval_steps_per_second": 0.934, + "step": 890 + }, + { + "epoch": 0.9764383561643836, + "grad_norm": 2.0311460494995117, + "learning_rate": 3.7753858651502846e-06, + "loss": 2.2185, + "step": 891 + }, + { + "epoch": 0.9775342465753425, + "grad_norm": 2.2017149925231934, + "learning_rate": 3.773354995938262e-06, + "loss": 2.164, + "step": 892 + }, + { + "epoch": 0.9786301369863014, + "grad_norm": 1.949684977531433, + "learning_rate": 3.771324126726239e-06, + "loss": 2.2128, + "step": 893 + }, + { + "epoch": 0.9797260273972602, + "grad_norm": 1.9395498037338257, + "learning_rate": 3.7692932575142164e-06, + "loss": 2.1957, + "step": 894 + }, + { + "epoch": 0.9808219178082191, + "grad_norm": 2.096432685852051, + "learning_rate": 3.7672623883021937e-06, + "loss": 2.219, + "step": 895 + }, + { + "epoch": 0.9819178082191781, + "grad_norm": 2.0942952632904053, + "learning_rate": 3.765231519090171e-06, + "loss": 2.1546, + "step": 896 + }, + { + "epoch": 0.983013698630137, + "grad_norm": 2.0097291469573975, + "learning_rate": 3.7632006498781482e-06, + "loss": 2.1021, + "step": 897 + }, + { + "epoch": 0.9841095890410959, + "grad_norm": 2.0909335613250732, + "learning_rate": 3.761169780666125e-06, + "loss": 2.1862, + "step": 898 + }, + { + "epoch": 0.9852054794520548, + "grad_norm": 2.187309980392456, + "learning_rate": 3.7591389114541028e-06, + "loss": 2.1511, + "step": 899 + }, + { + "epoch": 0.9863013698630136, + "grad_norm": 2.194276809692383, + "learning_rate": 3.7571080422420796e-06, + "loss": 2.196, + "step": 900 + }, + { + "epoch": 0.9863013698630136, + "eval_loss": 2.3047611713409424, + "eval_runtime": 45.609, + "eval_samples_per_second": 21.925, + "eval_steps_per_second": 0.921, + "step": 900 + }, + { + "epoch": 0.9873972602739726, + "grad_norm": 2.1132829189300537, + "learning_rate": 3.7550771730300573e-06, + "loss": 2.2046, + "step": 901 + }, + { + "epoch": 0.9884931506849315, + "grad_norm": 2.1946468353271484, + "learning_rate": 3.753046303818034e-06, + "loss": 2.1656, + "step": 902 + }, + { + "epoch": 0.9895890410958904, + "grad_norm": 2.2210536003112793, + "learning_rate": 3.751015434606012e-06, + "loss": 2.2186, + "step": 903 + }, + { + "epoch": 0.9906849315068493, + "grad_norm": 1.913695216178894, + "learning_rate": 3.7489845653939887e-06, + "loss": 2.1933, + "step": 904 + }, + { + "epoch": 0.9917808219178083, + "grad_norm": 2.1088368892669678, + "learning_rate": 3.7469536961819664e-06, + "loss": 2.2406, + "step": 905 + }, + { + "epoch": 0.9928767123287672, + "grad_norm": 2.04239821434021, + "learning_rate": 3.7449228269699433e-06, + "loss": 2.142, + "step": 906 + }, + { + "epoch": 0.993972602739726, + "grad_norm": 1.7852507829666138, + "learning_rate": 3.742891957757921e-06, + "loss": 2.1701, + "step": 907 + }, + { + "epoch": 0.9950684931506849, + "grad_norm": 2.0275776386260986, + "learning_rate": 3.740861088545898e-06, + "loss": 2.2004, + "step": 908 + }, + { + "epoch": 0.9961643835616438, + "grad_norm": 1.9028904438018799, + "learning_rate": 3.7388302193338755e-06, + "loss": 2.177, + "step": 909 + }, + { + "epoch": 0.9972602739726028, + "grad_norm": 1.7815115451812744, + "learning_rate": 3.7367993501218524e-06, + "loss": 2.207, + "step": 910 + }, + { + "epoch": 0.9972602739726028, + "eval_loss": 2.302318811416626, + "eval_runtime": 42.9296, + "eval_samples_per_second": 23.294, + "eval_steps_per_second": 0.978, + "step": 910 + }, + { + "epoch": 0.9983561643835617, + "grad_norm": 1.913689374923706, + "learning_rate": 3.7347684809098297e-06, + "loss": 2.1594, + "step": 911 + }, + { + "epoch": 0.9994520547945206, + "grad_norm": 2.171222686767578, + "learning_rate": 3.732737611697807e-06, + "loss": 2.1689, + "step": 912 + }, + { + "epoch": 1.0005479452054795, + "grad_norm": 1.831642985343933, + "learning_rate": 3.7307067424857842e-06, + "loss": 2.1367, + "step": 913 + }, + { + "epoch": 1.0016438356164383, + "grad_norm": 2.1831955909729004, + "learning_rate": 3.7286758732737615e-06, + "loss": 2.1035, + "step": 914 + }, + { + "epoch": 1.0027397260273974, + "grad_norm": 2.038025379180908, + "learning_rate": 3.7266450040617388e-06, + "loss": 2.1461, + "step": 915 + }, + { + "epoch": 1.0038356164383562, + "grad_norm": 1.866502046585083, + "learning_rate": 3.724614134849716e-06, + "loss": 2.1322, + "step": 916 + }, + { + "epoch": 1.004931506849315, + "grad_norm": 2.1359357833862305, + "learning_rate": 3.7225832656376933e-06, + "loss": 2.2333, + "step": 917 + }, + { + "epoch": 1.006027397260274, + "grad_norm": 1.8689136505126953, + "learning_rate": 3.7205523964256706e-06, + "loss": 2.1176, + "step": 918 + }, + { + "epoch": 1.0071232876712328, + "grad_norm": 1.8869425058364868, + "learning_rate": 3.718521527213648e-06, + "loss": 2.2127, + "step": 919 + }, + { + "epoch": 1.0082191780821919, + "grad_norm": 1.9148075580596924, + "learning_rate": 3.716490658001625e-06, + "loss": 2.1846, + "step": 920 + }, + { + "epoch": 1.0082191780821919, + "eval_loss": 2.30368709564209, + "eval_runtime": 52.1845, + "eval_samples_per_second": 19.163, + "eval_steps_per_second": 0.805, + "step": 920 + }, + { + "epoch": 1.0093150684931507, + "grad_norm": 1.9502761363983154, + "learning_rate": 3.7144597887896024e-06, + "loss": 2.1731, + "step": 921 + }, + { + "epoch": 1.0104109589041095, + "grad_norm": 1.9093869924545288, + "learning_rate": 3.7124289195775797e-06, + "loss": 2.0864, + "step": 922 + }, + { + "epoch": 1.0115068493150685, + "grad_norm": 2.0871293544769287, + "learning_rate": 3.710398050365557e-06, + "loss": 2.1838, + "step": 923 + }, + { + "epoch": 1.0126027397260273, + "grad_norm": 2.0559985637664795, + "learning_rate": 3.708367181153534e-06, + "loss": 2.1764, + "step": 924 + }, + { + "epoch": 1.0136986301369864, + "grad_norm": 1.8968807458877563, + "learning_rate": 3.7063363119415115e-06, + "loss": 2.1515, + "step": 925 + }, + { + "epoch": 1.0147945205479452, + "grad_norm": 1.8613216876983643, + "learning_rate": 3.7043054427294884e-06, + "loss": 2.1694, + "step": 926 + }, + { + "epoch": 1.0158904109589042, + "grad_norm": 2.002044439315796, + "learning_rate": 3.702274573517466e-06, + "loss": 2.148, + "step": 927 + }, + { + "epoch": 1.016986301369863, + "grad_norm": 2.1138675212860107, + "learning_rate": 3.700243704305443e-06, + "loss": 2.1401, + "step": 928 + }, + { + "epoch": 1.0180821917808218, + "grad_norm": 1.7620841264724731, + "learning_rate": 3.6982128350934206e-06, + "loss": 2.1399, + "step": 929 + }, + { + "epoch": 1.0191780821917809, + "grad_norm": 1.8970746994018555, + "learning_rate": 3.6961819658813975e-06, + "loss": 2.1404, + "step": 930 + }, + { + "epoch": 1.0191780821917809, + "eval_loss": 2.3068811893463135, + "eval_runtime": 53.0188, + "eval_samples_per_second": 18.861, + "eval_steps_per_second": 0.792, + "step": 930 + }, + { + "epoch": 1.0202739726027397, + "grad_norm": 1.9405478239059448, + "learning_rate": 3.694151096669375e-06, + "loss": 2.1286, + "step": 931 + }, + { + "epoch": 1.0213698630136987, + "grad_norm": 2.112222909927368, + "learning_rate": 3.692120227457352e-06, + "loss": 2.2246, + "step": 932 + }, + { + "epoch": 1.0224657534246575, + "grad_norm": 1.8792580366134644, + "learning_rate": 3.6900893582453297e-06, + "loss": 2.148, + "step": 933 + }, + { + "epoch": 1.0235616438356165, + "grad_norm": 1.9530014991760254, + "learning_rate": 3.6880584890333066e-06, + "loss": 2.1707, + "step": 934 + }, + { + "epoch": 1.0246575342465754, + "grad_norm": 2.1778736114501953, + "learning_rate": 3.6860276198212834e-06, + "loss": 2.0879, + "step": 935 + }, + { + "epoch": 1.0257534246575342, + "grad_norm": 1.8960413932800293, + "learning_rate": 3.683996750609261e-06, + "loss": 2.1659, + "step": 936 + }, + { + "epoch": 1.0268493150684932, + "grad_norm": 1.9805221557617188, + "learning_rate": 3.681965881397238e-06, + "loss": 2.1674, + "step": 937 + }, + { + "epoch": 1.027945205479452, + "grad_norm": 1.9999068975448608, + "learning_rate": 3.6799350121852157e-06, + "loss": 2.1677, + "step": 938 + }, + { + "epoch": 1.029041095890411, + "grad_norm": 2.016911029815674, + "learning_rate": 3.6779041429731925e-06, + "loss": 2.1841, + "step": 939 + }, + { + "epoch": 1.0301369863013699, + "grad_norm": 1.7200374603271484, + "learning_rate": 3.6758732737611702e-06, + "loss": 2.1391, + "step": 940 + }, + { + "epoch": 1.0301369863013699, + "eval_loss": 2.3022048473358154, + "eval_runtime": 51.2908, + "eval_samples_per_second": 19.497, + "eval_steps_per_second": 0.819, + "step": 940 + }, + { + "epoch": 1.0312328767123287, + "grad_norm": 2.249377489089966, + "learning_rate": 3.673842404549147e-06, + "loss": 2.1618, + "step": 941 + }, + { + "epoch": 1.0323287671232877, + "grad_norm": 1.838922381401062, + "learning_rate": 3.671811535337125e-06, + "loss": 2.1527, + "step": 942 + }, + { + "epoch": 1.0334246575342465, + "grad_norm": 1.9571045637130737, + "learning_rate": 3.6697806661251016e-06, + "loss": 2.149, + "step": 943 + }, + { + "epoch": 1.0345205479452055, + "grad_norm": 1.9929102659225464, + "learning_rate": 3.6677497969130793e-06, + "loss": 2.159, + "step": 944 + }, + { + "epoch": 1.0356164383561643, + "grad_norm": 2.004830837249756, + "learning_rate": 3.665718927701056e-06, + "loss": 2.2369, + "step": 945 + }, + { + "epoch": 1.0367123287671234, + "grad_norm": 1.9605739116668701, + "learning_rate": 3.663688058489034e-06, + "loss": 2.1795, + "step": 946 + }, + { + "epoch": 1.0378082191780822, + "grad_norm": 1.9499164819717407, + "learning_rate": 3.6616571892770107e-06, + "loss": 2.1979, + "step": 947 + }, + { + "epoch": 1.038904109589041, + "grad_norm": 1.867347240447998, + "learning_rate": 3.659626320064988e-06, + "loss": 2.1616, + "step": 948 + }, + { + "epoch": 1.04, + "grad_norm": 1.9342801570892334, + "learning_rate": 3.6575954508529653e-06, + "loss": 2.1044, + "step": 949 + }, + { + "epoch": 1.0410958904109588, + "grad_norm": 2.0009584426879883, + "learning_rate": 3.6555645816409426e-06, + "loss": 2.1543, + "step": 950 + }, + { + "epoch": 1.0410958904109588, + "eval_loss": 2.3028438091278076, + "eval_runtime": 46.7426, + "eval_samples_per_second": 21.394, + "eval_steps_per_second": 0.899, + "step": 950 + }, + { + "epoch": 1.0421917808219179, + "grad_norm": 2.0411503314971924, + "learning_rate": 3.65353371242892e-06, + "loss": 2.1465, + "step": 951 + }, + { + "epoch": 1.0432876712328767, + "grad_norm": 2.0512070655822754, + "learning_rate": 3.651502843216897e-06, + "loss": 2.1828, + "step": 952 + }, + { + "epoch": 1.0443835616438357, + "grad_norm": 2.0948643684387207, + "learning_rate": 3.6494719740048744e-06, + "loss": 2.0671, + "step": 953 + }, + { + "epoch": 1.0454794520547945, + "grad_norm": 1.9103025197982788, + "learning_rate": 3.6474411047928517e-06, + "loss": 2.1651, + "step": 954 + }, + { + "epoch": 1.0465753424657533, + "grad_norm": 2.1500349044799805, + "learning_rate": 3.645410235580829e-06, + "loss": 2.137, + "step": 955 + }, + { + "epoch": 1.0476712328767124, + "grad_norm": 1.7760887145996094, + "learning_rate": 3.6433793663688062e-06, + "loss": 2.1738, + "step": 956 + }, + { + "epoch": 1.0487671232876712, + "grad_norm": 2.0111730098724365, + "learning_rate": 3.6413484971567835e-06, + "loss": 2.1569, + "step": 957 + }, + { + "epoch": 1.0498630136986302, + "grad_norm": 2.0040462017059326, + "learning_rate": 3.6393176279447608e-06, + "loss": 2.1483, + "step": 958 + }, + { + "epoch": 1.050958904109589, + "grad_norm": 1.9583499431610107, + "learning_rate": 3.637286758732738e-06, + "loss": 2.1575, + "step": 959 + }, + { + "epoch": 1.0520547945205478, + "grad_norm": 1.8842004537582397, + "learning_rate": 3.6352558895207153e-06, + "loss": 2.1797, + "step": 960 + }, + { + "epoch": 1.0520547945205478, + "eval_loss": 2.301715850830078, + "eval_runtime": 52.7345, + "eval_samples_per_second": 18.963, + "eval_steps_per_second": 0.796, + "step": 960 + }, + { + "epoch": 1.0531506849315069, + "grad_norm": 2.020909309387207, + "learning_rate": 3.633225020308692e-06, + "loss": 2.1939, + "step": 961 + }, + { + "epoch": 1.0542465753424657, + "grad_norm": 2.09222149848938, + "learning_rate": 3.63119415109667e-06, + "loss": 2.1777, + "step": 962 + }, + { + "epoch": 1.0553424657534247, + "grad_norm": 1.8582347631454468, + "learning_rate": 3.6291632818846467e-06, + "loss": 2.1505, + "step": 963 + }, + { + "epoch": 1.0564383561643835, + "grad_norm": 1.9468824863433838, + "learning_rate": 3.6271324126726244e-06, + "loss": 2.1762, + "step": 964 + }, + { + "epoch": 1.0575342465753426, + "grad_norm": 1.9484895467758179, + "learning_rate": 3.6251015434606013e-06, + "loss": 2.1102, + "step": 965 + }, + { + "epoch": 1.0586301369863014, + "grad_norm": 2.11462140083313, + "learning_rate": 3.623070674248579e-06, + "loss": 2.1114, + "step": 966 + }, + { + "epoch": 1.0597260273972602, + "grad_norm": 2.2038373947143555, + "learning_rate": 3.621039805036556e-06, + "loss": 2.1513, + "step": 967 + }, + { + "epoch": 1.0608219178082192, + "grad_norm": 2.1271684169769287, + "learning_rate": 3.6190089358245335e-06, + "loss": 2.137, + "step": 968 + }, + { + "epoch": 1.061917808219178, + "grad_norm": 1.8624194860458374, + "learning_rate": 3.6169780666125104e-06, + "loss": 2.1701, + "step": 969 + }, + { + "epoch": 1.063013698630137, + "grad_norm": 1.935445785522461, + "learning_rate": 3.614947197400488e-06, + "loss": 2.1726, + "step": 970 + }, + { + "epoch": 1.063013698630137, + "eval_loss": 2.302633047103882, + "eval_runtime": 51.8284, + "eval_samples_per_second": 19.294, + "eval_steps_per_second": 0.81, + "step": 970 + }, + { + "epoch": 1.0641095890410959, + "grad_norm": 2.048907518386841, + "learning_rate": 3.612916328188465e-06, + "loss": 2.1712, + "step": 971 + }, + { + "epoch": 1.0652054794520547, + "grad_norm": 1.850943684577942, + "learning_rate": 3.610885458976442e-06, + "loss": 2.102, + "step": 972 + }, + { + "epoch": 1.0663013698630137, + "grad_norm": 1.8364315032958984, + "learning_rate": 3.6088545897644195e-06, + "loss": 2.1903, + "step": 973 + }, + { + "epoch": 1.0673972602739725, + "grad_norm": 1.8227031230926514, + "learning_rate": 3.6068237205523963e-06, + "loss": 2.1787, + "step": 974 + }, + { + "epoch": 1.0684931506849316, + "grad_norm": 1.9136372804641724, + "learning_rate": 3.604792851340374e-06, + "loss": 2.1244, + "step": 975 + }, + { + "epoch": 1.0695890410958904, + "grad_norm": 1.8585968017578125, + "learning_rate": 3.602761982128351e-06, + "loss": 2.1287, + "step": 976 + }, + { + "epoch": 1.0706849315068494, + "grad_norm": 2.139573812484741, + "learning_rate": 3.6007311129163286e-06, + "loss": 2.1938, + "step": 977 + }, + { + "epoch": 1.0717808219178082, + "grad_norm": 2.1584925651550293, + "learning_rate": 3.5987002437043054e-06, + "loss": 2.1626, + "step": 978 + }, + { + "epoch": 1.072876712328767, + "grad_norm": 1.8905949592590332, + "learning_rate": 3.596669374492283e-06, + "loss": 2.1219, + "step": 979 + }, + { + "epoch": 1.073972602739726, + "grad_norm": 2.099651575088501, + "learning_rate": 3.59463850528026e-06, + "loss": 2.1385, + "step": 980 + }, + { + "epoch": 1.073972602739726, + "eval_loss": 2.3004090785980225, + "eval_runtime": 52.7406, + "eval_samples_per_second": 18.961, + "eval_steps_per_second": 0.796, + "step": 980 + }, + { + "epoch": 1.0750684931506849, + "grad_norm": 2.1456780433654785, + "learning_rate": 3.5926076360682377e-06, + "loss": 2.1973, + "step": 981 + }, + { + "epoch": 1.076164383561644, + "grad_norm": 1.929372787475586, + "learning_rate": 3.5905767668562145e-06, + "loss": 2.1115, + "step": 982 + }, + { + "epoch": 1.0772602739726027, + "grad_norm": 2.08174991607666, + "learning_rate": 3.5885458976441922e-06, + "loss": 2.1785, + "step": 983 + }, + { + "epoch": 1.0783561643835617, + "grad_norm": 1.9355952739715576, + "learning_rate": 3.586515028432169e-06, + "loss": 2.1567, + "step": 984 + }, + { + "epoch": 1.0794520547945206, + "grad_norm": 1.9983466863632202, + "learning_rate": 3.5844841592201464e-06, + "loss": 2.1913, + "step": 985 + }, + { + "epoch": 1.0805479452054794, + "grad_norm": 1.8714051246643066, + "learning_rate": 3.5824532900081237e-06, + "loss": 2.1938, + "step": 986 + }, + { + "epoch": 1.0816438356164384, + "grad_norm": 2.007049798965454, + "learning_rate": 3.580422420796101e-06, + "loss": 2.1093, + "step": 987 + }, + { + "epoch": 1.0827397260273972, + "grad_norm": 1.965572476387024, + "learning_rate": 3.578391551584078e-06, + "loss": 2.1426, + "step": 988 + }, + { + "epoch": 1.0838356164383562, + "grad_norm": 1.9040261507034302, + "learning_rate": 3.5763606823720555e-06, + "loss": 2.1627, + "step": 989 + }, + { + "epoch": 1.084931506849315, + "grad_norm": 2.001929759979248, + "learning_rate": 3.5743298131600328e-06, + "loss": 2.0983, + "step": 990 + }, + { + "epoch": 1.084931506849315, + "eval_loss": 2.2979862689971924, + "eval_runtime": 50.6325, + "eval_samples_per_second": 19.75, + "eval_steps_per_second": 0.83, + "step": 990 + }, + { + "epoch": 1.086027397260274, + "grad_norm": 1.9204787015914917, + "learning_rate": 3.57229894394801e-06, + "loss": 2.1571, + "step": 991 + }, + { + "epoch": 1.087123287671233, + "grad_norm": 1.9628651142120361, + "learning_rate": 3.5702680747359873e-06, + "loss": 2.1736, + "step": 992 + }, + { + "epoch": 1.0882191780821917, + "grad_norm": 1.9888029098510742, + "learning_rate": 3.5682372055239646e-06, + "loss": 2.17, + "step": 993 + }, + { + "epoch": 1.0893150684931507, + "grad_norm": 2.049769401550293, + "learning_rate": 3.566206336311942e-06, + "loss": 2.1352, + "step": 994 + }, + { + "epoch": 1.0904109589041096, + "grad_norm": 2.0208375453948975, + "learning_rate": 3.564175467099919e-06, + "loss": 2.1564, + "step": 995 + }, + { + "epoch": 1.0915068493150686, + "grad_norm": 1.976775050163269, + "learning_rate": 3.5621445978878964e-06, + "loss": 2.1292, + "step": 996 + }, + { + "epoch": 1.0926027397260274, + "grad_norm": 2.0502822399139404, + "learning_rate": 3.5601137286758737e-06, + "loss": 2.1115, + "step": 997 + }, + { + "epoch": 1.0936986301369862, + "grad_norm": 2.1184189319610596, + "learning_rate": 3.5580828594638505e-06, + "loss": 2.1641, + "step": 998 + }, + { + "epoch": 1.0947945205479452, + "grad_norm": 1.9662559032440186, + "learning_rate": 3.5560519902518282e-06, + "loss": 2.1338, + "step": 999 + }, + { + "epoch": 1.095890410958904, + "grad_norm": 2.13386869430542, + "learning_rate": 3.554021121039805e-06, + "loss": 2.158, + "step": 1000 + }, + { + "epoch": 1.095890410958904, + "eval_loss": 2.3001608848571777, + "eval_runtime": 45.9219, + "eval_samples_per_second": 21.776, + "eval_steps_per_second": 0.915, + "step": 1000 + }, + { + "epoch": 1.096986301369863, + "grad_norm": 1.8791636228561401, + "learning_rate": 3.551990251827783e-06, + "loss": 2.1752, + "step": 1001 + }, + { + "epoch": 1.098082191780822, + "grad_norm": 1.9105546474456787, + "learning_rate": 3.5499593826157596e-06, + "loss": 2.1543, + "step": 1002 + }, + { + "epoch": 1.099178082191781, + "grad_norm": 1.887668251991272, + "learning_rate": 3.5479285134037373e-06, + "loss": 2.1226, + "step": 1003 + }, + { + "epoch": 1.1002739726027397, + "grad_norm": 2.2854599952697754, + "learning_rate": 3.545897644191714e-06, + "loss": 2.1425, + "step": 1004 + }, + { + "epoch": 1.1013698630136985, + "grad_norm": 2.03892183303833, + "learning_rate": 3.543866774979692e-06, + "loss": 2.135, + "step": 1005 + }, + { + "epoch": 1.1024657534246576, + "grad_norm": 1.9144858121871948, + "learning_rate": 3.5418359057676687e-06, + "loss": 2.1323, + "step": 1006 + }, + { + "epoch": 1.1035616438356164, + "grad_norm": 2.1402337551116943, + "learning_rate": 3.5398050365556464e-06, + "loss": 2.1668, + "step": 1007 + }, + { + "epoch": 1.1046575342465754, + "grad_norm": 2.084592819213867, + "learning_rate": 3.5377741673436233e-06, + "loss": 2.1331, + "step": 1008 + }, + { + "epoch": 1.1057534246575342, + "grad_norm": 2.0012383460998535, + "learning_rate": 3.535743298131601e-06, + "loss": 2.1677, + "step": 1009 + }, + { + "epoch": 1.106849315068493, + "grad_norm": 2.272204637527466, + "learning_rate": 3.533712428919578e-06, + "loss": 2.13, + "step": 1010 + }, + { + "epoch": 1.106849315068493, + "eval_loss": 2.3001229763031006, + "eval_runtime": 42.0444, + "eval_samples_per_second": 23.784, + "eval_steps_per_second": 0.999, + "step": 1010 + }, + { + "epoch": 1.107945205479452, + "grad_norm": 2.1061534881591797, + "learning_rate": 3.5316815597075547e-06, + "loss": 2.1581, + "step": 1011 + }, + { + "epoch": 1.1090410958904109, + "grad_norm": 2.159306764602661, + "learning_rate": 3.5296506904955324e-06, + "loss": 2.1868, + "step": 1012 + }, + { + "epoch": 1.11013698630137, + "grad_norm": 2.1429545879364014, + "learning_rate": 3.5276198212835093e-06, + "loss": 2.1274, + "step": 1013 + }, + { + "epoch": 1.1112328767123287, + "grad_norm": 1.8494454622268677, + "learning_rate": 3.525588952071487e-06, + "loss": 2.1513, + "step": 1014 + }, + { + "epoch": 1.1123287671232878, + "grad_norm": 1.8494454622268677, + "learning_rate": 3.525588952071487e-06, + "loss": 2.1883, + "step": 1015 + }, + { + "epoch": 1.1134246575342466, + "grad_norm": 2.2244861125946045, + "learning_rate": 3.523558082859464e-06, + "loss": 2.1896, + "step": 1016 + }, + { + "epoch": 1.1145205479452054, + "grad_norm": 2.181950569152832, + "learning_rate": 3.5215272136474415e-06, + "loss": 2.1561, + "step": 1017 + }, + { + "epoch": 1.1156164383561644, + "grad_norm": 1.7740849256515503, + "learning_rate": 3.5194963444354184e-06, + "loss": 2.1358, + "step": 1018 + }, + { + "epoch": 1.1167123287671232, + "grad_norm": 2.3540139198303223, + "learning_rate": 3.517465475223396e-06, + "loss": 2.1241, + "step": 1019 + }, + { + "epoch": 1.1178082191780823, + "grad_norm": 2.2737820148468018, + "learning_rate": 3.515434606011373e-06, + "loss": 2.1117, + "step": 1020 + }, + { + "epoch": 1.1178082191780823, + "eval_loss": 2.2998275756835938, + "eval_runtime": 46.2142, + "eval_samples_per_second": 21.638, + "eval_steps_per_second": 0.909, + "step": 1020 + }, + { + "epoch": 1.118904109589041, + "grad_norm": 2.0705313682556152, + "learning_rate": 3.5134037367993506e-06, + "loss": 2.1743, + "step": 1021 + }, + { + "epoch": 1.12, + "grad_norm": 2.2605583667755127, + "learning_rate": 3.5113728675873275e-06, + "loss": 2.2082, + "step": 1022 + }, + { + "epoch": 1.121095890410959, + "grad_norm": 2.028203248977661, + "learning_rate": 3.5093419983753047e-06, + "loss": 2.1722, + "step": 1023 + }, + { + "epoch": 1.1221917808219177, + "grad_norm": 1.9794270992279053, + "learning_rate": 3.507311129163282e-06, + "loss": 2.1581, + "step": 1024 + }, + { + "epoch": 1.1232876712328768, + "grad_norm": 2.1830344200134277, + "learning_rate": 3.5052802599512593e-06, + "loss": 2.1142, + "step": 1025 + }, + { + "epoch": 1.1243835616438356, + "grad_norm": 1.926081895828247, + "learning_rate": 3.5032493907392366e-06, + "loss": 2.1532, + "step": 1026 + }, + { + "epoch": 1.1254794520547946, + "grad_norm": 1.9909394979476929, + "learning_rate": 3.501218521527214e-06, + "loss": 2.153, + "step": 1027 + }, + { + "epoch": 1.1265753424657534, + "grad_norm": 1.9685133695602417, + "learning_rate": 3.499187652315191e-06, + "loss": 2.0983, + "step": 1028 + }, + { + "epoch": 1.1276712328767124, + "grad_norm": 1.9185279607772827, + "learning_rate": 3.4971567831031684e-06, + "loss": 2.1495, + "step": 1029 + }, + { + "epoch": 1.1287671232876713, + "grad_norm": 2.0214717388153076, + "learning_rate": 3.4951259138911457e-06, + "loss": 2.176, + "step": 1030 + }, + { + "epoch": 1.1287671232876713, + "eval_loss": 2.2963690757751465, + "eval_runtime": 44.2655, + "eval_samples_per_second": 22.591, + "eval_steps_per_second": 0.949, + "step": 1030 + }, + { + "epoch": 1.12986301369863, + "grad_norm": 1.9134252071380615, + "learning_rate": 3.493095044679123e-06, + "loss": 2.1034, + "step": 1031 + }, + { + "epoch": 1.130958904109589, + "grad_norm": 1.8448772430419922, + "learning_rate": 3.4910641754671002e-06, + "loss": 2.1947, + "step": 1032 + }, + { + "epoch": 1.132054794520548, + "grad_norm": 1.856341004371643, + "learning_rate": 3.4890333062550775e-06, + "loss": 2.0676, + "step": 1033 + }, + { + "epoch": 1.133150684931507, + "grad_norm": 2.039062261581421, + "learning_rate": 3.4870024370430548e-06, + "loss": 2.1618, + "step": 1034 + }, + { + "epoch": 1.1342465753424658, + "grad_norm": 1.8440991640090942, + "learning_rate": 3.484971567831032e-06, + "loss": 2.0956, + "step": 1035 + }, + { + "epoch": 1.1353424657534246, + "grad_norm": 1.8483991622924805, + "learning_rate": 3.482940698619009e-06, + "loss": 2.1364, + "step": 1036 + }, + { + "epoch": 1.1364383561643836, + "grad_norm": 1.9045069217681885, + "learning_rate": 3.4809098294069866e-06, + "loss": 2.1786, + "step": 1037 + }, + { + "epoch": 1.1375342465753424, + "grad_norm": 1.8203203678131104, + "learning_rate": 3.4788789601949634e-06, + "loss": 2.1438, + "step": 1038 + }, + { + "epoch": 1.1386301369863014, + "grad_norm": 2.1084606647491455, + "learning_rate": 3.476848090982941e-06, + "loss": 2.1553, + "step": 1039 + }, + { + "epoch": 1.1397260273972603, + "grad_norm": 2.1041061878204346, + "learning_rate": 3.474817221770918e-06, + "loss": 2.1435, + "step": 1040 + }, + { + "epoch": 1.1397260273972603, + "eval_loss": 2.2970690727233887, + "eval_runtime": 43.2304, + "eval_samples_per_second": 23.132, + "eval_steps_per_second": 0.972, + "step": 1040 + }, + { + "epoch": 1.140821917808219, + "grad_norm": 1.9701104164123535, + "learning_rate": 3.4727863525588957e-06, + "loss": 2.2201, + "step": 1041 + }, + { + "epoch": 1.141917808219178, + "grad_norm": 1.8816475868225098, + "learning_rate": 3.4707554833468725e-06, + "loss": 2.1643, + "step": 1042 + }, + { + "epoch": 1.143013698630137, + "grad_norm": 2.096355438232422, + "learning_rate": 3.4687246141348502e-06, + "loss": 2.1514, + "step": 1043 + }, + { + "epoch": 1.144109589041096, + "grad_norm": 1.9975471496582031, + "learning_rate": 3.466693744922827e-06, + "loss": 2.1246, + "step": 1044 + }, + { + "epoch": 1.1452054794520548, + "grad_norm": 2.0587127208709717, + "learning_rate": 3.464662875710805e-06, + "loss": 2.1371, + "step": 1045 + }, + { + "epoch": 1.1463013698630138, + "grad_norm": 1.9275381565093994, + "learning_rate": 3.4626320064987817e-06, + "loss": 2.1518, + "step": 1046 + }, + { + "epoch": 1.1473972602739726, + "grad_norm": 2.1690948009490967, + "learning_rate": 3.4606011372867594e-06, + "loss": 2.1528, + "step": 1047 + }, + { + "epoch": 1.1484931506849314, + "grad_norm": 1.970973014831543, + "learning_rate": 3.458570268074736e-06, + "loss": 2.1569, + "step": 1048 + }, + { + "epoch": 1.1495890410958904, + "grad_norm": 1.8897690773010254, + "learning_rate": 3.4565393988627135e-06, + "loss": 2.1401, + "step": 1049 + }, + { + "epoch": 1.1506849315068493, + "grad_norm": 1.9245911836624146, + "learning_rate": 3.4545085296506908e-06, + "loss": 2.1311, + "step": 1050 + }, + { + "epoch": 1.1506849315068493, + "eval_loss": 2.2956528663635254, + "eval_runtime": 44.1778, + "eval_samples_per_second": 22.636, + "eval_steps_per_second": 0.951, + "step": 1050 + }, + { + "epoch": 1.1517808219178083, + "grad_norm": 2.142209529876709, + "learning_rate": 3.452477660438668e-06, + "loss": 2.1144, + "step": 1051 + }, + { + "epoch": 1.152876712328767, + "grad_norm": 1.932369351387024, + "learning_rate": 3.4504467912266453e-06, + "loss": 2.1739, + "step": 1052 + }, + { + "epoch": 1.1539726027397261, + "grad_norm": 2.259359359741211, + "learning_rate": 3.4484159220146226e-06, + "loss": 2.1036, + "step": 1053 + }, + { + "epoch": 1.155068493150685, + "grad_norm": 2.112485647201538, + "learning_rate": 3.4463850528026e-06, + "loss": 2.1429, + "step": 1054 + }, + { + "epoch": 1.1561643835616437, + "grad_norm": 2.6211049556732178, + "learning_rate": 3.444354183590577e-06, + "loss": 2.1421, + "step": 1055 + }, + { + "epoch": 1.1572602739726028, + "grad_norm": 1.8724851608276367, + "learning_rate": 3.4423233143785544e-06, + "loss": 2.1622, + "step": 1056 + }, + { + "epoch": 1.1583561643835616, + "grad_norm": 1.9292882680892944, + "learning_rate": 3.4402924451665317e-06, + "loss": 2.1071, + "step": 1057 + }, + { + "epoch": 1.1594520547945206, + "grad_norm": 1.9832521677017212, + "learning_rate": 3.438261575954509e-06, + "loss": 2.0879, + "step": 1058 + }, + { + "epoch": 1.1605479452054794, + "grad_norm": 1.9720780849456787, + "learning_rate": 3.4362307067424862e-06, + "loss": 2.1603, + "step": 1059 + }, + { + "epoch": 1.1616438356164385, + "grad_norm": 1.8179587125778198, + "learning_rate": 3.434199837530463e-06, + "loss": 2.1851, + "step": 1060 + }, + { + "epoch": 1.1616438356164385, + "eval_loss": 2.2950079441070557, + "eval_runtime": 42.9625, + "eval_samples_per_second": 23.276, + "eval_steps_per_second": 0.978, + "step": 1060 + }, + { + "epoch": 1.1627397260273973, + "grad_norm": 1.8762129545211792, + "learning_rate": 3.432168968318441e-06, + "loss": 2.1527, + "step": 1061 + }, + { + "epoch": 1.163835616438356, + "grad_norm": 2.173297643661499, + "learning_rate": 3.4301380991064176e-06, + "loss": 2.1023, + "step": 1062 + }, + { + "epoch": 1.1649315068493151, + "grad_norm": 1.9358222484588623, + "learning_rate": 3.4281072298943953e-06, + "loss": 2.1638, + "step": 1063 + }, + { + "epoch": 1.166027397260274, + "grad_norm": 1.9421939849853516, + "learning_rate": 3.426076360682372e-06, + "loss": 2.1794, + "step": 1064 + }, + { + "epoch": 1.167123287671233, + "grad_norm": 2.212996006011963, + "learning_rate": 3.42404549147035e-06, + "loss": 2.1877, + "step": 1065 + }, + { + "epoch": 1.1682191780821918, + "grad_norm": 2.036180257797241, + "learning_rate": 3.4220146222583267e-06, + "loss": 2.1534, + "step": 1066 + }, + { + "epoch": 1.1693150684931506, + "grad_norm": 1.8819142580032349, + "learning_rate": 3.4199837530463044e-06, + "loss": 2.099, + "step": 1067 + }, + { + "epoch": 1.1704109589041096, + "grad_norm": 2.0781915187835693, + "learning_rate": 3.4179528838342813e-06, + "loss": 2.1326, + "step": 1068 + }, + { + "epoch": 1.1715068493150684, + "grad_norm": 2.034895181655884, + "learning_rate": 3.415922014622259e-06, + "loss": 2.2107, + "step": 1069 + }, + { + "epoch": 1.1726027397260275, + "grad_norm": 1.92604660987854, + "learning_rate": 3.413891145410236e-06, + "loss": 2.1773, + "step": 1070 + }, + { + "epoch": 1.1726027397260275, + "eval_loss": 2.2959954738616943, + "eval_runtime": 47.2123, + "eval_samples_per_second": 21.181, + "eval_steps_per_second": 0.89, + "step": 1070 + }, + { + "epoch": 1.1736986301369863, + "grad_norm": 1.762730598449707, + "learning_rate": 3.4118602761982135e-06, + "loss": 2.1803, + "step": 1071 + }, + { + "epoch": 1.174794520547945, + "grad_norm": 2.0661215782165527, + "learning_rate": 3.4098294069861904e-06, + "loss": 2.1919, + "step": 1072 + }, + { + "epoch": 1.1758904109589041, + "grad_norm": 2.079982280731201, + "learning_rate": 3.4077985377741673e-06, + "loss": 2.1536, + "step": 1073 + }, + { + "epoch": 1.176986301369863, + "grad_norm": 1.9044359922409058, + "learning_rate": 3.405767668562145e-06, + "loss": 2.1403, + "step": 1074 + }, + { + "epoch": 1.178082191780822, + "grad_norm": 1.8870306015014648, + "learning_rate": 3.403736799350122e-06, + "loss": 2.1845, + "step": 1075 + }, + { + "epoch": 1.1791780821917808, + "grad_norm": 1.9305400848388672, + "learning_rate": 3.4017059301380995e-06, + "loss": 2.1786, + "step": 1076 + }, + { + "epoch": 1.1802739726027398, + "grad_norm": 1.987713098526001, + "learning_rate": 3.3996750609260764e-06, + "loss": 2.1214, + "step": 1077 + }, + { + "epoch": 1.1813698630136986, + "grad_norm": 1.8673466444015503, + "learning_rate": 3.397644191714054e-06, + "loss": 2.2027, + "step": 1078 + }, + { + "epoch": 1.1824657534246574, + "grad_norm": 2.009316921234131, + "learning_rate": 3.395613322502031e-06, + "loss": 2.139, + "step": 1079 + }, + { + "epoch": 1.1835616438356165, + "grad_norm": 2.1024956703186035, + "learning_rate": 3.3935824532900086e-06, + "loss": 2.1438, + "step": 1080 + }, + { + "epoch": 1.1835616438356165, + "eval_loss": 2.2966299057006836, + "eval_runtime": 44.8946, + "eval_samples_per_second": 22.274, + "eval_steps_per_second": 0.936, + "step": 1080 + }, + { + "epoch": 1.1846575342465753, + "grad_norm": 1.8911967277526855, + "learning_rate": 3.3915515840779855e-06, + "loss": 2.1238, + "step": 1081 + }, + { + "epoch": 1.1857534246575343, + "grad_norm": 2.0649192333221436, + "learning_rate": 3.389520714865963e-06, + "loss": 2.1534, + "step": 1082 + }, + { + "epoch": 1.1868493150684931, + "grad_norm": 1.9236847162246704, + "learning_rate": 3.38748984565394e-06, + "loss": 2.1427, + "step": 1083 + }, + { + "epoch": 1.1879452054794521, + "grad_norm": 1.8662182092666626, + "learning_rate": 3.3854589764419177e-06, + "loss": 2.1699, + "step": 1084 + }, + { + "epoch": 1.189041095890411, + "grad_norm": 1.9226020574569702, + "learning_rate": 3.3834281072298946e-06, + "loss": 2.1192, + "step": 1085 + }, + { + "epoch": 1.1901369863013698, + "grad_norm": 1.9659301042556763, + "learning_rate": 3.381397238017872e-06, + "loss": 2.1529, + "step": 1086 + }, + { + "epoch": 1.1912328767123288, + "grad_norm": 2.2714693546295166, + "learning_rate": 3.379366368805849e-06, + "loss": 2.1273, + "step": 1087 + }, + { + "epoch": 1.1923287671232876, + "grad_norm": 2.1853270530700684, + "learning_rate": 3.3773354995938264e-06, + "loss": 2.2029, + "step": 1088 + }, + { + "epoch": 1.1934246575342466, + "grad_norm": 1.8302830457687378, + "learning_rate": 3.3753046303818037e-06, + "loss": 2.1599, + "step": 1089 + }, + { + "epoch": 1.1945205479452055, + "grad_norm": 2.005591869354248, + "learning_rate": 3.373273761169781e-06, + "loss": 2.139, + "step": 1090 + }, + { + "epoch": 1.1945205479452055, + "eval_loss": 2.295994520187378, + "eval_runtime": 45.4802, + "eval_samples_per_second": 21.988, + "eval_steps_per_second": 0.923, + "step": 1090 + }, + { + "epoch": 1.1956164383561645, + "grad_norm": 2.139267683029175, + "learning_rate": 3.3712428919577582e-06, + "loss": 2.151, + "step": 1091 + }, + { + "epoch": 1.1967123287671233, + "grad_norm": 1.8547207117080688, + "learning_rate": 3.3692120227457355e-06, + "loss": 2.1249, + "step": 1092 + }, + { + "epoch": 1.1978082191780821, + "grad_norm": 2.214583396911621, + "learning_rate": 3.3671811535337128e-06, + "loss": 2.1656, + "step": 1093 + }, + { + "epoch": 1.1989041095890411, + "grad_norm": 2.1608800888061523, + "learning_rate": 3.36515028432169e-06, + "loss": 2.1056, + "step": 1094 + }, + { + "epoch": 1.2, + "grad_norm": 2.0850605964660645, + "learning_rate": 3.3631194151096673e-06, + "loss": 2.1215, + "step": 1095 + }, + { + "epoch": 1.201095890410959, + "grad_norm": 1.9199103116989136, + "learning_rate": 3.3610885458976446e-06, + "loss": 2.1584, + "step": 1096 + }, + { + "epoch": 1.2021917808219178, + "grad_norm": 1.9766334295272827, + "learning_rate": 3.359057676685622e-06, + "loss": 2.1821, + "step": 1097 + }, + { + "epoch": 1.2032876712328768, + "grad_norm": 1.948988676071167, + "learning_rate": 3.357026807473599e-06, + "loss": 2.182, + "step": 1098 + }, + { + "epoch": 1.2043835616438356, + "grad_norm": 2.0419397354125977, + "learning_rate": 3.354995938261576e-06, + "loss": 2.1051, + "step": 1099 + }, + { + "epoch": 1.2054794520547945, + "grad_norm": 1.9751213788986206, + "learning_rate": 3.3529650690495537e-06, + "loss": 2.1255, + "step": 1100 + }, + { + "epoch": 1.2054794520547945, + "eval_loss": 2.293811798095703, + "eval_runtime": 46.1182, + "eval_samples_per_second": 21.683, + "eval_steps_per_second": 0.911, + "step": 1100 + }, + { + "epoch": 1.2065753424657535, + "grad_norm": 1.9806510210037231, + "learning_rate": 3.3509341998375305e-06, + "loss": 2.1021, + "step": 1101 + }, + { + "epoch": 1.2076712328767123, + "grad_norm": 2.0570414066314697, + "learning_rate": 3.3489033306255082e-06, + "loss": 2.1417, + "step": 1102 + }, + { + "epoch": 1.208767123287671, + "grad_norm": 1.868134617805481, + "learning_rate": 3.346872461413485e-06, + "loss": 2.1494, + "step": 1103 + }, + { + "epoch": 1.2098630136986301, + "grad_norm": 2.0156848430633545, + "learning_rate": 3.344841592201463e-06, + "loss": 2.1238, + "step": 1104 + }, + { + "epoch": 1.210958904109589, + "grad_norm": 2.082301616668701, + "learning_rate": 3.3428107229894397e-06, + "loss": 2.1105, + "step": 1105 + }, + { + "epoch": 1.212054794520548, + "grad_norm": 1.8853904008865356, + "learning_rate": 3.3407798537774174e-06, + "loss": 2.1406, + "step": 1106 + }, + { + "epoch": 1.2131506849315068, + "grad_norm": 2.0410518646240234, + "learning_rate": 3.338748984565394e-06, + "loss": 2.1567, + "step": 1107 + }, + { + "epoch": 1.2142465753424658, + "grad_norm": 1.9901232719421387, + "learning_rate": 3.336718115353372e-06, + "loss": 2.1778, + "step": 1108 + }, + { + "epoch": 1.2153424657534246, + "grad_norm": 1.9840792417526245, + "learning_rate": 3.3346872461413488e-06, + "loss": 2.1935, + "step": 1109 + }, + { + "epoch": 1.2164383561643834, + "grad_norm": 1.8854445219039917, + "learning_rate": 3.3326563769293256e-06, + "loss": 2.137, + "step": 1110 + }, + { + "epoch": 1.2164383561643834, + "eval_loss": 2.294447660446167, + "eval_runtime": 42.6423, + "eval_samples_per_second": 23.451, + "eval_steps_per_second": 0.985, + "step": 1110 + }, + { + "epoch": 1.2175342465753425, + "grad_norm": 2.0948665142059326, + "learning_rate": 3.3306255077173033e-06, + "loss": 2.1288, + "step": 1111 + }, + { + "epoch": 1.2186301369863013, + "grad_norm": 1.859448790550232, + "learning_rate": 3.32859463850528e-06, + "loss": 2.1403, + "step": 1112 + }, + { + "epoch": 1.2197260273972603, + "grad_norm": 1.92074453830719, + "learning_rate": 3.326563769293258e-06, + "loss": 2.141, + "step": 1113 + }, + { + "epoch": 1.2208219178082191, + "grad_norm": 1.8377320766448975, + "learning_rate": 3.3245329000812347e-06, + "loss": 2.0889, + "step": 1114 + }, + { + "epoch": 1.2219178082191782, + "grad_norm": 1.9320917129516602, + "learning_rate": 3.3225020308692124e-06, + "loss": 2.1254, + "step": 1115 + }, + { + "epoch": 1.223013698630137, + "grad_norm": 1.9691251516342163, + "learning_rate": 3.3204711616571893e-06, + "loss": 2.1409, + "step": 1116 + }, + { + "epoch": 1.2241095890410958, + "grad_norm": 1.8214130401611328, + "learning_rate": 3.318440292445167e-06, + "loss": 2.1046, + "step": 1117 + }, + { + "epoch": 1.2252054794520548, + "grad_norm": 1.959455966949463, + "learning_rate": 3.316409423233144e-06, + "loss": 2.1418, + "step": 1118 + }, + { + "epoch": 1.2263013698630136, + "grad_norm": 1.9920449256896973, + "learning_rate": 3.3143785540211215e-06, + "loss": 2.1192, + "step": 1119 + }, + { + "epoch": 1.2273972602739727, + "grad_norm": 2.0276854038238525, + "learning_rate": 3.3123476848090984e-06, + "loss": 2.1939, + "step": 1120 + }, + { + "epoch": 1.2273972602739727, + "eval_loss": 2.295330047607422, + "eval_runtime": 49.8691, + "eval_samples_per_second": 20.052, + "eval_steps_per_second": 0.842, + "step": 1120 + }, + { + "epoch": 1.2284931506849315, + "grad_norm": 1.9765129089355469, + "learning_rate": 3.310316815597076e-06, + "loss": 2.1545, + "step": 1121 + }, + { + "epoch": 1.2295890410958905, + "grad_norm": 1.8449276685714722, + "learning_rate": 3.308285946385053e-06, + "loss": 2.1459, + "step": 1122 + }, + { + "epoch": 1.2306849315068493, + "grad_norm": 1.953072190284729, + "learning_rate": 3.30625507717303e-06, + "loss": 2.2046, + "step": 1123 + }, + { + "epoch": 1.2317808219178081, + "grad_norm": 1.9570913314819336, + "learning_rate": 3.3042242079610075e-06, + "loss": 2.1015, + "step": 1124 + }, + { + "epoch": 1.2328767123287672, + "grad_norm": 2.00569224357605, + "learning_rate": 3.3021933387489847e-06, + "loss": 2.1681, + "step": 1125 + }, + { + "epoch": 1.233972602739726, + "grad_norm": 1.9131590127944946, + "learning_rate": 3.300162469536962e-06, + "loss": 2.1765, + "step": 1126 + }, + { + "epoch": 1.235068493150685, + "grad_norm": 1.8375864028930664, + "learning_rate": 3.2981316003249393e-06, + "loss": 2.1175, + "step": 1127 + }, + { + "epoch": 1.2361643835616438, + "grad_norm": 1.852847933769226, + "learning_rate": 3.2961007311129166e-06, + "loss": 2.1838, + "step": 1128 + }, + { + "epoch": 1.2372602739726029, + "grad_norm": 1.8772693872451782, + "learning_rate": 3.294069861900894e-06, + "loss": 2.1524, + "step": 1129 + }, + { + "epoch": 1.2383561643835617, + "grad_norm": 1.8593205213546753, + "learning_rate": 3.292038992688871e-06, + "loss": 2.2059, + "step": 1130 + }, + { + "epoch": 1.2383561643835617, + "eval_loss": 2.294188976287842, + "eval_runtime": 50.4549, + "eval_samples_per_second": 19.82, + "eval_steps_per_second": 0.832, + "step": 1130 + }, + { + "epoch": 1.2394520547945205, + "grad_norm": 1.8967972993850708, + "learning_rate": 3.2900081234768484e-06, + "loss": 2.1551, + "step": 1131 + }, + { + "epoch": 1.2405479452054795, + "grad_norm": 1.9799107313156128, + "learning_rate": 3.2879772542648257e-06, + "loss": 2.1644, + "step": 1132 + }, + { + "epoch": 1.2416438356164383, + "grad_norm": 1.978883147239685, + "learning_rate": 3.285946385052803e-06, + "loss": 2.182, + "step": 1133 + }, + { + "epoch": 1.2427397260273974, + "grad_norm": 1.8705304861068726, + "learning_rate": 3.2839155158407802e-06, + "loss": 2.0981, + "step": 1134 + }, + { + "epoch": 1.2438356164383562, + "grad_norm": 2.0146889686584473, + "learning_rate": 3.2818846466287575e-06, + "loss": 2.1686, + "step": 1135 + }, + { + "epoch": 1.244931506849315, + "grad_norm": 1.94461989402771, + "learning_rate": 3.2798537774167344e-06, + "loss": 2.093, + "step": 1136 + }, + { + "epoch": 1.246027397260274, + "grad_norm": 1.7823240756988525, + "learning_rate": 3.277822908204712e-06, + "loss": 2.198, + "step": 1137 + }, + { + "epoch": 1.2471232876712328, + "grad_norm": 1.89297354221344, + "learning_rate": 3.275792038992689e-06, + "loss": 2.1359, + "step": 1138 + }, + { + "epoch": 1.2482191780821918, + "grad_norm": 1.8670459985733032, + "learning_rate": 3.2737611697806666e-06, + "loss": 2.0956, + "step": 1139 + }, + { + "epoch": 1.2493150684931507, + "grad_norm": 1.813568353652954, + "learning_rate": 3.2717303005686435e-06, + "loss": 2.1448, + "step": 1140 + }, + { + "epoch": 1.2493150684931507, + "eval_loss": 2.295804023742676, + "eval_runtime": 53.286, + "eval_samples_per_second": 18.767, + "eval_steps_per_second": 0.788, + "step": 1140 + }, + { + "epoch": 1.2504109589041095, + "grad_norm": 1.912916898727417, + "learning_rate": 3.269699431356621e-06, + "loss": 2.1159, + "step": 1141 + }, + { + "epoch": 1.2515068493150685, + "grad_norm": 1.7939883470535278, + "learning_rate": 3.267668562144598e-06, + "loss": 2.1566, + "step": 1142 + }, + { + "epoch": 1.2526027397260273, + "grad_norm": 1.8494806289672852, + "learning_rate": 3.2656376929325757e-06, + "loss": 2.1496, + "step": 1143 + }, + { + "epoch": 1.2536986301369863, + "grad_norm": 1.8468523025512695, + "learning_rate": 3.2636068237205526e-06, + "loss": 2.1159, + "step": 1144 + }, + { + "epoch": 1.2547945205479452, + "grad_norm": 1.8962055444717407, + "learning_rate": 3.2615759545085303e-06, + "loss": 2.1712, + "step": 1145 + }, + { + "epoch": 1.2558904109589042, + "grad_norm": 1.800353765487671, + "learning_rate": 3.259545085296507e-06, + "loss": 2.1486, + "step": 1146 + }, + { + "epoch": 1.256986301369863, + "grad_norm": 2.124359607696533, + "learning_rate": 3.257514216084484e-06, + "loss": 2.1913, + "step": 1147 + }, + { + "epoch": 1.2580821917808218, + "grad_norm": 1.9248145818710327, + "learning_rate": 3.2554833468724617e-06, + "loss": 2.1128, + "step": 1148 + }, + { + "epoch": 1.2591780821917808, + "grad_norm": 1.9990876913070679, + "learning_rate": 3.2534524776604385e-06, + "loss": 2.1854, + "step": 1149 + }, + { + "epoch": 1.2602739726027397, + "grad_norm": 1.8600568771362305, + "learning_rate": 3.2514216084484162e-06, + "loss": 2.1565, + "step": 1150 + }, + { + "epoch": 1.2602739726027397, + "eval_loss": 2.2942006587982178, + "eval_runtime": 51.4021, + "eval_samples_per_second": 19.454, + "eval_steps_per_second": 0.817, + "step": 1150 + }, + { + "epoch": 1.2613698630136987, + "grad_norm": 1.8621101379394531, + "learning_rate": 3.249390739236393e-06, + "loss": 2.1438, + "step": 1151 + }, + { + "epoch": 1.2624657534246575, + "grad_norm": 1.9290356636047363, + "learning_rate": 3.2473598700243708e-06, + "loss": 2.1577, + "step": 1152 + }, + { + "epoch": 1.2635616438356165, + "grad_norm": 1.9007307291030884, + "learning_rate": 3.2453290008123476e-06, + "loss": 2.1295, + "step": 1153 + }, + { + "epoch": 1.2646575342465753, + "grad_norm": 1.7647960186004639, + "learning_rate": 3.2432981316003253e-06, + "loss": 2.1472, + "step": 1154 + }, + { + "epoch": 1.2657534246575342, + "grad_norm": 1.9796825647354126, + "learning_rate": 3.241267262388302e-06, + "loss": 2.1542, + "step": 1155 + }, + { + "epoch": 1.2668493150684932, + "grad_norm": 1.9505796432495117, + "learning_rate": 3.23923639317628e-06, + "loss": 2.1093, + "step": 1156 + }, + { + "epoch": 1.267945205479452, + "grad_norm": 1.9152203798294067, + "learning_rate": 3.2372055239642567e-06, + "loss": 2.1456, + "step": 1157 + }, + { + "epoch": 1.269041095890411, + "grad_norm": 1.9094963073730469, + "learning_rate": 3.2351746547522344e-06, + "loss": 2.1149, + "step": 1158 + }, + { + "epoch": 1.2701369863013698, + "grad_norm": 1.8872679471969604, + "learning_rate": 3.2331437855402113e-06, + "loss": 2.1545, + "step": 1159 + }, + { + "epoch": 1.2712328767123289, + "grad_norm": 2.171703338623047, + "learning_rate": 3.2311129163281885e-06, + "loss": 2.2108, + "step": 1160 + }, + { + "epoch": 1.2712328767123289, + "eval_loss": 2.2910356521606445, + "eval_runtime": 52.5176, + "eval_samples_per_second": 19.041, + "eval_steps_per_second": 0.8, + "step": 1160 + }, + { + "epoch": 1.2723287671232877, + "grad_norm": 1.8578304052352905, + "learning_rate": 3.229082047116166e-06, + "loss": 2.1158, + "step": 1161 + }, + { + "epoch": 1.2734246575342465, + "grad_norm": 1.8873049020767212, + "learning_rate": 3.227051177904143e-06, + "loss": 2.1618, + "step": 1162 + }, + { + "epoch": 1.2745205479452055, + "grad_norm": 1.8442409038543701, + "learning_rate": 3.2250203086921204e-06, + "loss": 2.1454, + "step": 1163 + }, + { + "epoch": 1.2756164383561643, + "grad_norm": 1.9609211683273315, + "learning_rate": 3.2229894394800977e-06, + "loss": 2.1519, + "step": 1164 + }, + { + "epoch": 1.2767123287671232, + "grad_norm": 1.9093680381774902, + "learning_rate": 3.220958570268075e-06, + "loss": 2.1775, + "step": 1165 + }, + { + "epoch": 1.2778082191780822, + "grad_norm": 1.8355298042297363, + "learning_rate": 3.218927701056052e-06, + "loss": 2.0871, + "step": 1166 + }, + { + "epoch": 1.2789041095890412, + "grad_norm": 1.9497056007385254, + "learning_rate": 3.2168968318440295e-06, + "loss": 2.1441, + "step": 1167 + }, + { + "epoch": 1.28, + "grad_norm": 1.8635729551315308, + "learning_rate": 3.2148659626320068e-06, + "loss": 2.0748, + "step": 1168 + }, + { + "epoch": 1.2810958904109588, + "grad_norm": 2.0972132682800293, + "learning_rate": 3.212835093419984e-06, + "loss": 2.1843, + "step": 1169 + }, + { + "epoch": 1.2821917808219179, + "grad_norm": 1.9235926866531372, + "learning_rate": 3.2108042242079613e-06, + "loss": 2.1194, + "step": 1170 + }, + { + "epoch": 1.2821917808219179, + "eval_loss": 2.291309118270874, + "eval_runtime": 55.6408, + "eval_samples_per_second": 17.972, + "eval_steps_per_second": 0.755, + "step": 1170 + }, + { + "epoch": 1.2832876712328767, + "grad_norm": 1.8674591779708862, + "learning_rate": 3.2087733549959386e-06, + "loss": 2.1264, + "step": 1171 + }, + { + "epoch": 1.2843835616438355, + "grad_norm": 1.810589075088501, + "learning_rate": 3.206742485783916e-06, + "loss": 2.1781, + "step": 1172 + }, + { + "epoch": 1.2854794520547945, + "grad_norm": 1.9794505834579468, + "learning_rate": 3.2047116165718927e-06, + "loss": 2.1089, + "step": 1173 + }, + { + "epoch": 1.2865753424657536, + "grad_norm": 1.8487070798873901, + "learning_rate": 3.2026807473598704e-06, + "loss": 2.097, + "step": 1174 + }, + { + "epoch": 1.2876712328767124, + "grad_norm": 1.9668854475021362, + "learning_rate": 3.2006498781478473e-06, + "loss": 2.1516, + "step": 1175 + }, + { + "epoch": 1.2887671232876712, + "grad_norm": 1.855817198753357, + "learning_rate": 3.198619008935825e-06, + "loss": 2.1678, + "step": 1176 + }, + { + "epoch": 1.2898630136986302, + "grad_norm": 2.381211757659912, + "learning_rate": 3.196588139723802e-06, + "loss": 2.099, + "step": 1177 + }, + { + "epoch": 1.290958904109589, + "grad_norm": 2.1524250507354736, + "learning_rate": 3.1945572705117795e-06, + "loss": 2.1345, + "step": 1178 + }, + { + "epoch": 1.2920547945205478, + "grad_norm": 1.9120781421661377, + "learning_rate": 3.1925264012997564e-06, + "loss": 2.1586, + "step": 1179 + }, + { + "epoch": 1.2931506849315069, + "grad_norm": 2.153817653656006, + "learning_rate": 3.190495532087734e-06, + "loss": 2.1597, + "step": 1180 + }, + { + "epoch": 1.2931506849315069, + "eval_loss": 2.2936320304870605, + "eval_runtime": 53.6426, + "eval_samples_per_second": 18.642, + "eval_steps_per_second": 0.783, + "step": 1180 + }, + { + "epoch": 1.2942465753424657, + "grad_norm": 2.074873447418213, + "learning_rate": 3.188464662875711e-06, + "loss": 2.1572, + "step": 1181 + }, + { + "epoch": 1.2953424657534247, + "grad_norm": 1.9313570261001587, + "learning_rate": 3.1864337936636886e-06, + "loss": 2.1986, + "step": 1182 + }, + { + "epoch": 1.2964383561643835, + "grad_norm": 1.9836641550064087, + "learning_rate": 3.1844029244516655e-06, + "loss": 2.13, + "step": 1183 + }, + { + "epoch": 1.2975342465753426, + "grad_norm": 2.0893137454986572, + "learning_rate": 3.182372055239643e-06, + "loss": 2.1138, + "step": 1184 + }, + { + "epoch": 1.2986301369863014, + "grad_norm": 2.1667087078094482, + "learning_rate": 3.18034118602762e-06, + "loss": 2.1119, + "step": 1185 + }, + { + "epoch": 1.2997260273972602, + "grad_norm": 2.093961000442505, + "learning_rate": 3.1783103168155973e-06, + "loss": 2.1728, + "step": 1186 + }, + { + "epoch": 1.3008219178082192, + "grad_norm": 1.919267177581787, + "learning_rate": 3.1762794476035746e-06, + "loss": 2.1469, + "step": 1187 + }, + { + "epoch": 1.301917808219178, + "grad_norm": 2.017967939376831, + "learning_rate": 3.174248578391552e-06, + "loss": 2.1307, + "step": 1188 + }, + { + "epoch": 1.303013698630137, + "grad_norm": 1.921632170677185, + "learning_rate": 3.172217709179529e-06, + "loss": 2.13, + "step": 1189 + }, + { + "epoch": 1.3041095890410959, + "grad_norm": 2.0171291828155518, + "learning_rate": 3.1701868399675064e-06, + "loss": 2.135, + "step": 1190 + }, + { + "epoch": 1.3041095890410959, + "eval_loss": 2.2934045791625977, + "eval_runtime": 48.0821, + "eval_samples_per_second": 20.798, + "eval_steps_per_second": 0.874, + "step": 1190 + }, + { + "epoch": 1.305205479452055, + "grad_norm": 1.8685601949691772, + "learning_rate": 3.1681559707554837e-06, + "loss": 2.1329, + "step": 1191 + }, + { + "epoch": 1.3063013698630137, + "grad_norm": 1.8909096717834473, + "learning_rate": 3.166125101543461e-06, + "loss": 2.1004, + "step": 1192 + }, + { + "epoch": 1.3073972602739725, + "grad_norm": 1.798614740371704, + "learning_rate": 3.1640942323314382e-06, + "loss": 2.1555, + "step": 1193 + }, + { + "epoch": 1.3084931506849315, + "grad_norm": 1.8296692371368408, + "learning_rate": 3.1620633631194155e-06, + "loss": 2.1549, + "step": 1194 + }, + { + "epoch": 1.3095890410958904, + "grad_norm": 1.8676908016204834, + "learning_rate": 3.1600324939073928e-06, + "loss": 2.1081, + "step": 1195 + }, + { + "epoch": 1.3106849315068494, + "grad_norm": 1.8682451248168945, + "learning_rate": 3.15800162469537e-06, + "loss": 2.138, + "step": 1196 + }, + { + "epoch": 1.3117808219178082, + "grad_norm": 2.0049314498901367, + "learning_rate": 3.155970755483347e-06, + "loss": 2.144, + "step": 1197 + }, + { + "epoch": 1.3128767123287672, + "grad_norm": 2.00691556930542, + "learning_rate": 3.1539398862713246e-06, + "loss": 2.1539, + "step": 1198 + }, + { + "epoch": 1.313972602739726, + "grad_norm": 1.8521753549575806, + "learning_rate": 3.1519090170593015e-06, + "loss": 2.1597, + "step": 1199 + }, + { + "epoch": 1.3150684931506849, + "grad_norm": 2.0974042415618896, + "learning_rate": 3.149878147847279e-06, + "loss": 2.147, + "step": 1200 + }, + { + "epoch": 1.3150684931506849, + "eval_loss": 2.2891385555267334, + "eval_runtime": 48.1618, + "eval_samples_per_second": 20.763, + "eval_steps_per_second": 0.872, + "step": 1200 + }, + { + "epoch": 1.316164383561644, + "grad_norm": 1.9671052694320679, + "learning_rate": 3.147847278635256e-06, + "loss": 2.1838, + "step": 1201 + }, + { + "epoch": 1.3172602739726027, + "grad_norm": 1.9421573877334595, + "learning_rate": 3.1458164094232337e-06, + "loss": 2.2093, + "step": 1202 + }, + { + "epoch": 1.3183561643835615, + "grad_norm": 1.917198657989502, + "learning_rate": 3.1437855402112106e-06, + "loss": 2.136, + "step": 1203 + }, + { + "epoch": 1.3194520547945205, + "grad_norm": 1.8664982318878174, + "learning_rate": 3.1417546709991883e-06, + "loss": 2.089, + "step": 1204 + }, + { + "epoch": 1.3205479452054796, + "grad_norm": 2.043558359146118, + "learning_rate": 3.139723801787165e-06, + "loss": 2.159, + "step": 1205 + }, + { + "epoch": 1.3216438356164384, + "grad_norm": 1.8314157724380493, + "learning_rate": 3.137692932575143e-06, + "loss": 2.13, + "step": 1206 + }, + { + "epoch": 1.3227397260273972, + "grad_norm": 1.907884120941162, + "learning_rate": 3.1356620633631197e-06, + "loss": 2.1746, + "step": 1207 + }, + { + "epoch": 1.3238356164383562, + "grad_norm": 2.0570952892303467, + "learning_rate": 3.1336311941510974e-06, + "loss": 2.1187, + "step": 1208 + }, + { + "epoch": 1.324931506849315, + "grad_norm": 1.832292914390564, + "learning_rate": 3.1316003249390742e-06, + "loss": 2.1425, + "step": 1209 + }, + { + "epoch": 1.3260273972602739, + "grad_norm": 1.8307697772979736, + "learning_rate": 3.129569455727051e-06, + "loss": 2.1749, + "step": 1210 + }, + { + "epoch": 1.3260273972602739, + "eval_loss": 2.2865941524505615, + "eval_runtime": 42.2877, + "eval_samples_per_second": 23.648, + "eval_steps_per_second": 0.993, + "step": 1210 + }, + { + "epoch": 1.3271232876712329, + "grad_norm": 1.8292433023452759, + "learning_rate": 3.1275385865150288e-06, + "loss": 2.1702, + "step": 1211 + }, + { + "epoch": 1.3282191780821917, + "grad_norm": 1.9146367311477661, + "learning_rate": 3.1255077173030056e-06, + "loss": 2.16, + "step": 1212 + }, + { + "epoch": 1.3293150684931507, + "grad_norm": 1.809275507926941, + "learning_rate": 3.1234768480909833e-06, + "loss": 2.157, + "step": 1213 + }, + { + "epoch": 1.3304109589041095, + "grad_norm": 1.939081072807312, + "learning_rate": 3.12144597887896e-06, + "loss": 2.1496, + "step": 1214 + }, + { + "epoch": 1.3315068493150686, + "grad_norm": 1.8665997982025146, + "learning_rate": 3.119415109666938e-06, + "loss": 2.1535, + "step": 1215 + }, + { + "epoch": 1.3326027397260274, + "grad_norm": 1.8455616235733032, + "learning_rate": 3.1173842404549147e-06, + "loss": 2.1845, + "step": 1216 + }, + { + "epoch": 1.3336986301369862, + "grad_norm": 2.050560235977173, + "learning_rate": 3.1153533712428924e-06, + "loss": 2.1461, + "step": 1217 + }, + { + "epoch": 1.3347945205479452, + "grad_norm": 1.8440574407577515, + "learning_rate": 3.1133225020308693e-06, + "loss": 2.1476, + "step": 1218 + }, + { + "epoch": 1.335890410958904, + "grad_norm": 1.8801573514938354, + "learning_rate": 3.111291632818847e-06, + "loss": 2.1721, + "step": 1219 + }, + { + "epoch": 1.336986301369863, + "grad_norm": 1.9227471351623535, + "learning_rate": 3.109260763606824e-06, + "loss": 2.178, + "step": 1220 + }, + { + "epoch": 1.336986301369863, + "eval_loss": 2.2896478176116943, + "eval_runtime": 43.157, + "eval_samples_per_second": 23.171, + "eval_steps_per_second": 0.973, + "step": 1220 + }, + { + "epoch": 1.3380821917808219, + "grad_norm": 1.9979732036590576, + "learning_rate": 3.1072298943948015e-06, + "loss": 2.148, + "step": 1221 + }, + { + "epoch": 1.339178082191781, + "grad_norm": 1.9481284618377686, + "learning_rate": 3.1051990251827784e-06, + "loss": 2.165, + "step": 1222 + }, + { + "epoch": 1.3402739726027397, + "grad_norm": 1.8830820322036743, + "learning_rate": 3.1031681559707557e-06, + "loss": 2.1239, + "step": 1223 + }, + { + "epoch": 1.3413698630136985, + "grad_norm": 1.890512466430664, + "learning_rate": 3.101137286758733e-06, + "loss": 2.1761, + "step": 1224 + }, + { + "epoch": 1.3424657534246576, + "grad_norm": 2.0024659633636475, + "learning_rate": 3.09910641754671e-06, + "loss": 2.179, + "step": 1225 + }, + { + "epoch": 1.3435616438356164, + "grad_norm": 1.8659477233886719, + "learning_rate": 3.0970755483346875e-06, + "loss": 2.1722, + "step": 1226 + }, + { + "epoch": 1.3446575342465754, + "grad_norm": 1.93405020236969, + "learning_rate": 3.0950446791226648e-06, + "loss": 2.1545, + "step": 1227 + }, + { + "epoch": 1.3457534246575342, + "grad_norm": 1.9514124393463135, + "learning_rate": 3.093013809910642e-06, + "loss": 2.187, + "step": 1228 + }, + { + "epoch": 1.3468493150684933, + "grad_norm": 1.9571877717971802, + "learning_rate": 3.0909829406986193e-06, + "loss": 2.1507, + "step": 1229 + }, + { + "epoch": 1.347945205479452, + "grad_norm": 1.9434577226638794, + "learning_rate": 3.0889520714865966e-06, + "loss": 2.1953, + "step": 1230 + }, + { + "epoch": 1.347945205479452, + "eval_loss": 2.289128065109253, + "eval_runtime": 46.7151, + "eval_samples_per_second": 21.406, + "eval_steps_per_second": 0.899, + "step": 1230 + }, + { + "epoch": 1.3490410958904109, + "grad_norm": 1.8294577598571777, + "learning_rate": 3.086921202274574e-06, + "loss": 2.1684, + "step": 1231 + }, + { + "epoch": 1.35013698630137, + "grad_norm": 1.9432605504989624, + "learning_rate": 3.084890333062551e-06, + "loss": 2.1393, + "step": 1232 + }, + { + "epoch": 1.3512328767123287, + "grad_norm": 1.9030417203903198, + "learning_rate": 3.0828594638505284e-06, + "loss": 2.1181, + "step": 1233 + }, + { + "epoch": 1.3523287671232875, + "grad_norm": 1.9233827590942383, + "learning_rate": 3.0808285946385057e-06, + "loss": 2.1479, + "step": 1234 + }, + { + "epoch": 1.3534246575342466, + "grad_norm": 1.9549237489700317, + "learning_rate": 3.078797725426483e-06, + "loss": 2.1253, + "step": 1235 + }, + { + "epoch": 1.3545205479452056, + "grad_norm": 2.0584867000579834, + "learning_rate": 3.07676685621446e-06, + "loss": 2.0846, + "step": 1236 + }, + { + "epoch": 1.3556164383561644, + "grad_norm": 1.7543913125991821, + "learning_rate": 3.0747359870024375e-06, + "loss": 2.1495, + "step": 1237 + }, + { + "epoch": 1.3567123287671232, + "grad_norm": 1.750698447227478, + "learning_rate": 3.0727051177904144e-06, + "loss": 2.1179, + "step": 1238 + }, + { + "epoch": 1.3578082191780823, + "grad_norm": 1.9867119789123535, + "learning_rate": 3.070674248578392e-06, + "loss": 2.1313, + "step": 1239 + }, + { + "epoch": 1.358904109589041, + "grad_norm": 1.8228135108947754, + "learning_rate": 3.068643379366369e-06, + "loss": 2.158, + "step": 1240 + }, + { + "epoch": 1.358904109589041, + "eval_loss": 2.2861993312835693, + "eval_runtime": 44.081, + "eval_samples_per_second": 22.686, + "eval_steps_per_second": 0.953, + "step": 1240 + }, + { + "epoch": 1.3599999999999999, + "grad_norm": 1.8551791906356812, + "learning_rate": 3.0666125101543466e-06, + "loss": 2.129, + "step": 1241 + }, + { + "epoch": 1.361095890410959, + "grad_norm": 1.8729921579360962, + "learning_rate": 3.0645816409423235e-06, + "loss": 2.1069, + "step": 1242 + }, + { + "epoch": 1.362191780821918, + "grad_norm": 1.8203959465026855, + "learning_rate": 3.062550771730301e-06, + "loss": 2.1509, + "step": 1243 + }, + { + "epoch": 1.3632876712328768, + "grad_norm": 1.9498761892318726, + "learning_rate": 3.060519902518278e-06, + "loss": 2.1336, + "step": 1244 + }, + { + "epoch": 1.3643835616438356, + "grad_norm": 2.134521245956421, + "learning_rate": 3.0584890333062557e-06, + "loss": 2.1294, + "step": 1245 + }, + { + "epoch": 1.3654794520547946, + "grad_norm": 1.9321913719177246, + "learning_rate": 3.0564581640942326e-06, + "loss": 2.0823, + "step": 1246 + }, + { + "epoch": 1.3665753424657534, + "grad_norm": 1.8948324918746948, + "learning_rate": 3.0544272948822094e-06, + "loss": 2.1425, + "step": 1247 + }, + { + "epoch": 1.3676712328767122, + "grad_norm": 2.2063205242156982, + "learning_rate": 3.052396425670187e-06, + "loss": 2.1242, + "step": 1248 + }, + { + "epoch": 1.3687671232876713, + "grad_norm": 2.0060935020446777, + "learning_rate": 3.050365556458164e-06, + "loss": 2.1586, + "step": 1249 + }, + { + "epoch": 1.36986301369863, + "grad_norm": 1.8441938161849976, + "learning_rate": 3.0483346872461417e-06, + "loss": 2.1305, + "step": 1250 + }, + { + "epoch": 1.36986301369863, + "eval_loss": 2.2872142791748047, + "eval_runtime": 43.4998, + "eval_samples_per_second": 22.989, + "eval_steps_per_second": 0.966, + "step": 1250 + }, + { + "epoch": 1.370958904109589, + "grad_norm": 2.019545793533325, + "learning_rate": 3.0463038180341185e-06, + "loss": 2.1257, + "step": 1251 + }, + { + "epoch": 1.372054794520548, + "grad_norm": 1.981020212173462, + "learning_rate": 3.0442729488220962e-06, + "loss": 2.155, + "step": 1252 + }, + { + "epoch": 1.373150684931507, + "grad_norm": 1.9707001447677612, + "learning_rate": 3.042242079610073e-06, + "loss": 2.1671, + "step": 1253 + }, + { + "epoch": 1.3742465753424657, + "grad_norm": 1.8645458221435547, + "learning_rate": 3.0402112103980508e-06, + "loss": 2.1655, + "step": 1254 + }, + { + "epoch": 1.3753424657534246, + "grad_norm": 1.863911747932434, + "learning_rate": 3.0381803411860276e-06, + "loss": 2.0917, + "step": 1255 + }, + { + "epoch": 1.3764383561643836, + "grad_norm": 2.1079251766204834, + "learning_rate": 3.0361494719740053e-06, + "loss": 2.119, + "step": 1256 + }, + { + "epoch": 1.3775342465753424, + "grad_norm": 1.8242024183273315, + "learning_rate": 3.034118602761982e-06, + "loss": 2.1398, + "step": 1257 + }, + { + "epoch": 1.3786301369863014, + "grad_norm": 1.8663142919540405, + "learning_rate": 3.03208773354996e-06, + "loss": 2.1405, + "step": 1258 + }, + { + "epoch": 1.3797260273972602, + "grad_norm": 1.9437193870544434, + "learning_rate": 3.0300568643379367e-06, + "loss": 2.1156, + "step": 1259 + }, + { + "epoch": 1.3808219178082193, + "grad_norm": 2.0837693214416504, + "learning_rate": 3.028025995125914e-06, + "loss": 2.1731, + "step": 1260 + }, + { + "epoch": 1.3808219178082193, + "eval_loss": 2.283296823501587, + "eval_runtime": 45.341, + "eval_samples_per_second": 22.055, + "eval_steps_per_second": 0.926, + "step": 1260 + }, + { + "epoch": 1.381917808219178, + "grad_norm": 1.9430493116378784, + "learning_rate": 3.0259951259138913e-06, + "loss": 2.1623, + "step": 1261 + }, + { + "epoch": 1.383013698630137, + "grad_norm": 1.825675368309021, + "learning_rate": 3.0239642567018686e-06, + "loss": 2.1022, + "step": 1262 + }, + { + "epoch": 1.384109589041096, + "grad_norm": 2.0050806999206543, + "learning_rate": 3.021933387489846e-06, + "loss": 2.1443, + "step": 1263 + }, + { + "epoch": 1.3852054794520547, + "grad_norm": 1.9548894166946411, + "learning_rate": 3.019902518277823e-06, + "loss": 2.0983, + "step": 1264 + }, + { + "epoch": 1.3863013698630138, + "grad_norm": 1.9097814559936523, + "learning_rate": 3.0178716490658004e-06, + "loss": 2.1488, + "step": 1265 + }, + { + "epoch": 1.3873972602739726, + "grad_norm": 1.8977514505386353, + "learning_rate": 3.0158407798537777e-06, + "loss": 2.1641, + "step": 1266 + }, + { + "epoch": 1.3884931506849316, + "grad_norm": 1.9222278594970703, + "learning_rate": 3.013809910641755e-06, + "loss": 2.1043, + "step": 1267 + }, + { + "epoch": 1.3895890410958904, + "grad_norm": 1.9333430528640747, + "learning_rate": 3.0117790414297322e-06, + "loss": 2.1152, + "step": 1268 + }, + { + "epoch": 1.3906849315068492, + "grad_norm": 1.8405439853668213, + "learning_rate": 3.0097481722177095e-06, + "loss": 2.1046, + "step": 1269 + }, + { + "epoch": 1.3917808219178083, + "grad_norm": 2.0583181381225586, + "learning_rate": 3.0077173030056868e-06, + "loss": 2.1378, + "step": 1270 + }, + { + "epoch": 1.3917808219178083, + "eval_loss": 2.286198616027832, + "eval_runtime": 46.6914, + "eval_samples_per_second": 21.417, + "eval_steps_per_second": 0.9, + "step": 1270 + }, + { + "epoch": 1.392876712328767, + "grad_norm": 1.8485472202301025, + "learning_rate": 3.005686433793664e-06, + "loss": 2.1184, + "step": 1271 + }, + { + "epoch": 1.393972602739726, + "grad_norm": 2.0430188179016113, + "learning_rate": 3.0036555645816413e-06, + "loss": 2.1518, + "step": 1272 + }, + { + "epoch": 1.395068493150685, + "grad_norm": 2.307474374771118, + "learning_rate": 3.001624695369618e-06, + "loss": 2.1288, + "step": 1273 + }, + { + "epoch": 1.396164383561644, + "grad_norm": 1.85003662109375, + "learning_rate": 2.999593826157596e-06, + "loss": 2.1573, + "step": 1274 + }, + { + "epoch": 1.3972602739726028, + "grad_norm": 2.047619581222534, + "learning_rate": 2.9975629569455727e-06, + "loss": 2.1285, + "step": 1275 + }, + { + "epoch": 1.3983561643835616, + "grad_norm": 2.013644218444824, + "learning_rate": 2.9955320877335504e-06, + "loss": 2.1603, + "step": 1276 + }, + { + "epoch": 1.3994520547945206, + "grad_norm": 1.8814456462860107, + "learning_rate": 2.9935012185215273e-06, + "loss": 2.1333, + "step": 1277 + }, + { + "epoch": 1.4005479452054794, + "grad_norm": 2.0338361263275146, + "learning_rate": 2.991470349309505e-06, + "loss": 2.1564, + "step": 1278 + }, + { + "epoch": 1.4016438356164382, + "grad_norm": 1.972992181777954, + "learning_rate": 2.989439480097482e-06, + "loss": 2.1305, + "step": 1279 + }, + { + "epoch": 1.4027397260273973, + "grad_norm": 1.845928430557251, + "learning_rate": 2.9874086108854595e-06, + "loss": 2.1384, + "step": 1280 + }, + { + "epoch": 1.4027397260273973, + "eval_loss": 2.28485107421875, + "eval_runtime": 46.9757, + "eval_samples_per_second": 21.288, + "eval_steps_per_second": 0.894, + "step": 1280 + }, + { + "epoch": 1.403835616438356, + "grad_norm": 1.7601349353790283, + "learning_rate": 2.9853777416734364e-06, + "loss": 2.1219, + "step": 1281 + }, + { + "epoch": 1.4049315068493151, + "grad_norm": 2.2385566234588623, + "learning_rate": 2.983346872461414e-06, + "loss": 2.1581, + "step": 1282 + }, + { + "epoch": 1.406027397260274, + "grad_norm": 2.019174098968506, + "learning_rate": 2.981316003249391e-06, + "loss": 2.1753, + "step": 1283 + }, + { + "epoch": 1.407123287671233, + "grad_norm": 1.9927541017532349, + "learning_rate": 2.9792851340373678e-06, + "loss": 2.147, + "step": 1284 + }, + { + "epoch": 1.4082191780821918, + "grad_norm": 2.5143957138061523, + "learning_rate": 2.9772542648253455e-06, + "loss": 2.138, + "step": 1285 + }, + { + "epoch": 1.4093150684931506, + "grad_norm": 1.9925073385238647, + "learning_rate": 2.9752233956133223e-06, + "loss": 2.1642, + "step": 1286 + }, + { + "epoch": 1.4104109589041096, + "grad_norm": 2.1739184856414795, + "learning_rate": 2.9731925264013e-06, + "loss": 2.1258, + "step": 1287 + }, + { + "epoch": 1.4115068493150684, + "grad_norm": 2.1719982624053955, + "learning_rate": 2.971161657189277e-06, + "loss": 2.1323, + "step": 1288 + }, + { + "epoch": 1.4126027397260275, + "grad_norm": 1.9897956848144531, + "learning_rate": 2.9691307879772546e-06, + "loss": 2.1286, + "step": 1289 + }, + { + "epoch": 1.4136986301369863, + "grad_norm": 1.997477412223816, + "learning_rate": 2.9670999187652314e-06, + "loss": 2.125, + "step": 1290 + }, + { + "epoch": 1.4136986301369863, + "eval_loss": 2.2839529514312744, + "eval_runtime": 44.5005, + "eval_samples_per_second": 22.472, + "eval_steps_per_second": 0.944, + "step": 1290 + }, + { + "epoch": 1.4147945205479453, + "grad_norm": 1.9529412984848022, + "learning_rate": 2.965069049553209e-06, + "loss": 2.1074, + "step": 1291 + }, + { + "epoch": 1.415890410958904, + "grad_norm": 1.900692343711853, + "learning_rate": 2.963038180341186e-06, + "loss": 2.1796, + "step": 1292 + }, + { + "epoch": 1.416986301369863, + "grad_norm": 2.0269863605499268, + "learning_rate": 2.9610073111291637e-06, + "loss": 2.0998, + "step": 1293 + }, + { + "epoch": 1.418082191780822, + "grad_norm": 2.0290050506591797, + "learning_rate": 2.9589764419171405e-06, + "loss": 2.1638, + "step": 1294 + }, + { + "epoch": 1.4191780821917808, + "grad_norm": 1.8717535734176636, + "learning_rate": 2.9569455727051182e-06, + "loss": 2.1756, + "step": 1295 + }, + { + "epoch": 1.4202739726027398, + "grad_norm": 1.9470241069793701, + "learning_rate": 2.954914703493095e-06, + "loss": 2.116, + "step": 1296 + }, + { + "epoch": 1.4213698630136986, + "grad_norm": 2.1749250888824463, + "learning_rate": 2.9528838342810724e-06, + "loss": 2.1381, + "step": 1297 + }, + { + "epoch": 1.4224657534246576, + "grad_norm": 1.905742883682251, + "learning_rate": 2.9508529650690496e-06, + "loss": 2.118, + "step": 1298 + }, + { + "epoch": 1.4235616438356165, + "grad_norm": 2.1214921474456787, + "learning_rate": 2.948822095857027e-06, + "loss": 2.1223, + "step": 1299 + }, + { + "epoch": 1.4246575342465753, + "grad_norm": 1.894509196281433, + "learning_rate": 2.946791226645004e-06, + "loss": 2.1457, + "step": 1300 + }, + { + "epoch": 1.4246575342465753, + "eval_loss": 2.2824268341064453, + "eval_runtime": 47.8031, + "eval_samples_per_second": 20.919, + "eval_steps_per_second": 0.879, + "step": 1300 + }, + { + "epoch": 1.4257534246575343, + "grad_norm": 1.9588708877563477, + "learning_rate": 2.9447603574329815e-06, + "loss": 2.1113, + "step": 1301 + }, + { + "epoch": 1.426849315068493, + "grad_norm": 1.9709562063217163, + "learning_rate": 2.9427294882209587e-06, + "loss": 2.1086, + "step": 1302 + }, + { + "epoch": 1.427945205479452, + "grad_norm": 1.82259202003479, + "learning_rate": 2.940698619008936e-06, + "loss": 2.118, + "step": 1303 + }, + { + "epoch": 1.429041095890411, + "grad_norm": 1.9776451587677002, + "learning_rate": 2.9386677497969133e-06, + "loss": 2.1386, + "step": 1304 + }, + { + "epoch": 1.43013698630137, + "grad_norm": 2.147463321685791, + "learning_rate": 2.9366368805848906e-06, + "loss": 2.1309, + "step": 1305 + }, + { + "epoch": 1.4312328767123288, + "grad_norm": 1.9845961332321167, + "learning_rate": 2.934606011372868e-06, + "loss": 2.1586, + "step": 1306 + }, + { + "epoch": 1.4323287671232876, + "grad_norm": 1.9153180122375488, + "learning_rate": 2.932575142160845e-06, + "loss": 2.1588, + "step": 1307 + }, + { + "epoch": 1.4334246575342466, + "grad_norm": 1.9372929334640503, + "learning_rate": 2.9305442729488224e-06, + "loss": 2.1351, + "step": 1308 + }, + { + "epoch": 1.4345205479452054, + "grad_norm": 1.852669358253479, + "learning_rate": 2.9285134037367997e-06, + "loss": 2.1394, + "step": 1309 + }, + { + "epoch": 1.4356164383561643, + "grad_norm": 1.934251308441162, + "learning_rate": 2.9264825345247765e-06, + "loss": 2.1364, + "step": 1310 + }, + { + "epoch": 1.4356164383561643, + "eval_loss": 2.280508279800415, + "eval_runtime": 43.1734, + "eval_samples_per_second": 23.162, + "eval_steps_per_second": 0.973, + "step": 1310 + }, + { + "epoch": 1.4367123287671233, + "grad_norm": 1.8886998891830444, + "learning_rate": 2.9244516653127542e-06, + "loss": 2.1452, + "step": 1311 + }, + { + "epoch": 1.4378082191780823, + "grad_norm": 2.275813102722168, + "learning_rate": 2.922420796100731e-06, + "loss": 2.1465, + "step": 1312 + }, + { + "epoch": 1.4389041095890411, + "grad_norm": 1.9017852544784546, + "learning_rate": 2.9203899268887088e-06, + "loss": 2.0785, + "step": 1313 + }, + { + "epoch": 1.44, + "grad_norm": 1.8482435941696167, + "learning_rate": 2.9183590576766856e-06, + "loss": 2.1892, + "step": 1314 + }, + { + "epoch": 1.441095890410959, + "grad_norm": 2.2122418880462646, + "learning_rate": 2.9163281884646633e-06, + "loss": 2.1058, + "step": 1315 + }, + { + "epoch": 1.4421917808219178, + "grad_norm": 1.9721672534942627, + "learning_rate": 2.91429731925264e-06, + "loss": 2.1391, + "step": 1316 + }, + { + "epoch": 1.4432876712328766, + "grad_norm": 1.9381605386734009, + "learning_rate": 2.912266450040618e-06, + "loss": 2.1651, + "step": 1317 + }, + { + "epoch": 1.4443835616438356, + "grad_norm": 2.274630069732666, + "learning_rate": 2.9102355808285947e-06, + "loss": 2.1434, + "step": 1318 + }, + { + "epoch": 1.4454794520547944, + "grad_norm": 1.91957688331604, + "learning_rate": 2.9082047116165724e-06, + "loss": 2.1218, + "step": 1319 + }, + { + "epoch": 1.4465753424657535, + "grad_norm": 2.042171001434326, + "learning_rate": 2.9061738424045493e-06, + "loss": 2.1233, + "step": 1320 + }, + { + "epoch": 1.4465753424657535, + "eval_loss": 2.280571699142456, + "eval_runtime": 48.425, + "eval_samples_per_second": 20.65, + "eval_steps_per_second": 0.867, + "step": 1320 + }, + { + "epoch": 1.4476712328767123, + "grad_norm": 2.0493276119232178, + "learning_rate": 2.904142973192527e-06, + "loss": 2.1241, + "step": 1321 + }, + { + "epoch": 1.4487671232876713, + "grad_norm": 1.8865023851394653, + "learning_rate": 2.902112103980504e-06, + "loss": 2.1826, + "step": 1322 + }, + { + "epoch": 1.4498630136986301, + "grad_norm": 1.9723960161209106, + "learning_rate": 2.900081234768481e-06, + "loss": 2.1508, + "step": 1323 + }, + { + "epoch": 1.450958904109589, + "grad_norm": 1.8321057558059692, + "learning_rate": 2.8980503655564584e-06, + "loss": 2.1319, + "step": 1324 + }, + { + "epoch": 1.452054794520548, + "grad_norm": 1.97796630859375, + "learning_rate": 2.8960194963444357e-06, + "loss": 2.1289, + "step": 1325 + }, + { + "epoch": 1.4531506849315068, + "grad_norm": 1.7793136835098267, + "learning_rate": 2.893988627132413e-06, + "loss": 2.1059, + "step": 1326 + }, + { + "epoch": 1.4542465753424658, + "grad_norm": 1.878381609916687, + "learning_rate": 2.8919577579203902e-06, + "loss": 2.125, + "step": 1327 + }, + { + "epoch": 1.4553424657534246, + "grad_norm": 1.8854790925979614, + "learning_rate": 2.8899268887083675e-06, + "loss": 2.0908, + "step": 1328 + }, + { + "epoch": 1.4564383561643837, + "grad_norm": 1.940931797027588, + "learning_rate": 2.8878960194963448e-06, + "loss": 2.1338, + "step": 1329 + }, + { + "epoch": 1.4575342465753425, + "grad_norm": 1.8014280796051025, + "learning_rate": 2.885865150284322e-06, + "loss": 2.1384, + "step": 1330 + }, + { + "epoch": 1.4575342465753425, + "eval_loss": 2.2815349102020264, + "eval_runtime": 48.7646, + "eval_samples_per_second": 20.507, + "eval_steps_per_second": 0.861, + "step": 1330 + }, + { + "epoch": 1.4586301369863013, + "grad_norm": 1.9306361675262451, + "learning_rate": 2.8838342810722993e-06, + "loss": 2.1562, + "step": 1331 + }, + { + "epoch": 1.4597260273972603, + "grad_norm": 1.902762770652771, + "learning_rate": 2.8818034118602766e-06, + "loss": 2.1183, + "step": 1332 + }, + { + "epoch": 1.4608219178082191, + "grad_norm": 1.980449914932251, + "learning_rate": 2.879772542648254e-06, + "loss": 2.1066, + "step": 1333 + }, + { + "epoch": 1.4619178082191782, + "grad_norm": 2.0774567127227783, + "learning_rate": 2.8777416734362307e-06, + "loss": 2.157, + "step": 1334 + }, + { + "epoch": 1.463013698630137, + "grad_norm": 1.8203723430633545, + "learning_rate": 2.8757108042242084e-06, + "loss": 2.0792, + "step": 1335 + }, + { + "epoch": 1.464109589041096, + "grad_norm": 1.8849077224731445, + "learning_rate": 2.8736799350121853e-06, + "loss": 2.1917, + "step": 1336 + }, + { + "epoch": 1.4652054794520548, + "grad_norm": 1.8782836198806763, + "learning_rate": 2.871649065800163e-06, + "loss": 2.1744, + "step": 1337 + }, + { + "epoch": 1.4663013698630136, + "grad_norm": 1.8628056049346924, + "learning_rate": 2.86961819658814e-06, + "loss": 2.1788, + "step": 1338 + }, + { + "epoch": 1.4673972602739727, + "grad_norm": 1.865370750427246, + "learning_rate": 2.8675873273761175e-06, + "loss": 2.1159, + "step": 1339 + }, + { + "epoch": 1.4684931506849315, + "grad_norm": 1.8673044443130493, + "learning_rate": 2.8655564581640944e-06, + "loss": 2.1447, + "step": 1340 + }, + { + "epoch": 1.4684931506849315, + "eval_loss": 2.282686471939087, + "eval_runtime": 49.85, + "eval_samples_per_second": 20.06, + "eval_steps_per_second": 0.843, + "step": 1340 + }, + { + "epoch": 1.4695890410958903, + "grad_norm": 1.8928618431091309, + "learning_rate": 2.863525588952072e-06, + "loss": 2.1362, + "step": 1341 + }, + { + "epoch": 1.4706849315068493, + "grad_norm": 1.8964238166809082, + "learning_rate": 2.861494719740049e-06, + "loss": 2.1628, + "step": 1342 + }, + { + "epoch": 1.4717808219178083, + "grad_norm": 1.9889185428619385, + "learning_rate": 2.8594638505280266e-06, + "loss": 2.0922, + "step": 1343 + }, + { + "epoch": 1.4728767123287672, + "grad_norm": 1.9503910541534424, + "learning_rate": 2.8574329813160035e-06, + "loss": 2.0979, + "step": 1344 + }, + { + "epoch": 1.473972602739726, + "grad_norm": 1.8889869451522827, + "learning_rate": 2.855402112103981e-06, + "loss": 2.1329, + "step": 1345 + }, + { + "epoch": 1.475068493150685, + "grad_norm": 1.7875893115997314, + "learning_rate": 2.853371242891958e-06, + "loss": 2.123, + "step": 1346 + }, + { + "epoch": 1.4761643835616438, + "grad_norm": 1.9358372688293457, + "learning_rate": 2.851340373679935e-06, + "loss": 2.1284, + "step": 1347 + }, + { + "epoch": 1.4772602739726026, + "grad_norm": 2.125243663787842, + "learning_rate": 2.8493095044679126e-06, + "loss": 2.1386, + "step": 1348 + }, + { + "epoch": 1.4783561643835617, + "grad_norm": 1.927788257598877, + "learning_rate": 2.8472786352558894e-06, + "loss": 2.1835, + "step": 1349 + }, + { + "epoch": 1.4794520547945205, + "grad_norm": 1.9361577033996582, + "learning_rate": 2.845247766043867e-06, + "loss": 2.1411, + "step": 1350 + }, + { + "epoch": 1.4794520547945205, + "eval_loss": 2.281172275543213, + "eval_runtime": 47.4205, + "eval_samples_per_second": 21.088, + "eval_steps_per_second": 0.886, + "step": 1350 + }, + { + "epoch": 1.4805479452054795, + "grad_norm": 1.847364068031311, + "learning_rate": 2.843216896831844e-06, + "loss": 2.111, + "step": 1351 + }, + { + "epoch": 1.4816438356164383, + "grad_norm": 1.9927674531936646, + "learning_rate": 2.8411860276198217e-06, + "loss": 2.107, + "step": 1352 + }, + { + "epoch": 1.4827397260273973, + "grad_norm": 1.8565406799316406, + "learning_rate": 2.8391551584077985e-06, + "loss": 2.12, + "step": 1353 + }, + { + "epoch": 1.4838356164383562, + "grad_norm": 1.9477596282958984, + "learning_rate": 2.8371242891957762e-06, + "loss": 2.1661, + "step": 1354 + }, + { + "epoch": 1.484931506849315, + "grad_norm": 1.9354814291000366, + "learning_rate": 2.835093419983753e-06, + "loss": 2.168, + "step": 1355 + }, + { + "epoch": 1.486027397260274, + "grad_norm": 2.0603487491607666, + "learning_rate": 2.833062550771731e-06, + "loss": 2.1338, + "step": 1356 + }, + { + "epoch": 1.4871232876712328, + "grad_norm": 1.987990379333496, + "learning_rate": 2.8310316815597076e-06, + "loss": 2.1282, + "step": 1357 + }, + { + "epoch": 1.4882191780821918, + "grad_norm": 1.895529866218567, + "learning_rate": 2.8290008123476853e-06, + "loss": 2.1032, + "step": 1358 + }, + { + "epoch": 1.4893150684931507, + "grad_norm": 1.917994737625122, + "learning_rate": 2.826969943135662e-06, + "loss": 2.149, + "step": 1359 + }, + { + "epoch": 1.4904109589041097, + "grad_norm": 1.9238080978393555, + "learning_rate": 2.8249390739236395e-06, + "loss": 2.1275, + "step": 1360 + }, + { + "epoch": 1.4904109589041097, + "eval_loss": 2.280546188354492, + "eval_runtime": 46.9985, + "eval_samples_per_second": 21.277, + "eval_steps_per_second": 0.894, + "step": 1360 + }, + { + "epoch": 1.4915068493150685, + "grad_norm": 1.947768211364746, + "learning_rate": 2.8229082047116167e-06, + "loss": 2.1282, + "step": 1361 + }, + { + "epoch": 1.4926027397260273, + "grad_norm": 1.7888189554214478, + "learning_rate": 2.820877335499594e-06, + "loss": 2.1134, + "step": 1362 + }, + { + "epoch": 1.4936986301369863, + "grad_norm": 1.8480347394943237, + "learning_rate": 2.8188464662875713e-06, + "loss": 2.1194, + "step": 1363 + }, + { + "epoch": 1.4947945205479451, + "grad_norm": 1.8146322965621948, + "learning_rate": 2.8168155970755486e-06, + "loss": 2.0781, + "step": 1364 + }, + { + "epoch": 1.4958904109589042, + "grad_norm": 1.7994698286056519, + "learning_rate": 2.814784727863526e-06, + "loss": 2.1027, + "step": 1365 + }, + { + "epoch": 1.496986301369863, + "grad_norm": 1.9375026226043701, + "learning_rate": 2.812753858651503e-06, + "loss": 2.1273, + "step": 1366 + }, + { + "epoch": 1.498082191780822, + "grad_norm": 1.9606287479400635, + "learning_rate": 2.8107229894394804e-06, + "loss": 2.1438, + "step": 1367 + }, + { + "epoch": 1.4991780821917808, + "grad_norm": 2.1815521717071533, + "learning_rate": 2.8086921202274577e-06, + "loss": 2.1176, + "step": 1368 + }, + { + "epoch": 1.5002739726027396, + "grad_norm": 1.8796950578689575, + "learning_rate": 2.806661251015435e-06, + "loss": 2.096, + "step": 1369 + }, + { + "epoch": 1.5013698630136987, + "grad_norm": 2.016091823577881, + "learning_rate": 2.8046303818034122e-06, + "loss": 2.1451, + "step": 1370 + }, + { + "epoch": 1.5013698630136987, + "eval_loss": 2.280881643295288, + "eval_runtime": 53.6471, + "eval_samples_per_second": 18.64, + "eval_steps_per_second": 0.783, + "step": 1370 + }, + { + "epoch": 1.5024657534246575, + "grad_norm": 1.9303081035614014, + "learning_rate": 2.802599512591389e-06, + "loss": 2.1736, + "step": 1371 + }, + { + "epoch": 1.5035616438356163, + "grad_norm": 1.9831101894378662, + "learning_rate": 2.8005686433793668e-06, + "loss": 2.2147, + "step": 1372 + }, + { + "epoch": 1.5046575342465753, + "grad_norm": 1.7864863872528076, + "learning_rate": 2.7985377741673436e-06, + "loss": 2.1231, + "step": 1373 + }, + { + "epoch": 1.5057534246575344, + "grad_norm": 1.8945375680923462, + "learning_rate": 2.7965069049553213e-06, + "loss": 2.0866, + "step": 1374 + }, + { + "epoch": 1.5068493150684932, + "grad_norm": 1.9881761074066162, + "learning_rate": 2.794476035743298e-06, + "loss": 2.1398, + "step": 1375 + }, + { + "epoch": 1.507945205479452, + "grad_norm": 1.9888391494750977, + "learning_rate": 2.792445166531276e-06, + "loss": 2.172, + "step": 1376 + }, + { + "epoch": 1.509041095890411, + "grad_norm": 1.9845142364501953, + "learning_rate": 2.7904142973192527e-06, + "loss": 2.1378, + "step": 1377 + }, + { + "epoch": 1.5101369863013698, + "grad_norm": 2.0640218257904053, + "learning_rate": 2.7883834281072304e-06, + "loss": 2.16, + "step": 1378 + }, + { + "epoch": 1.5112328767123286, + "grad_norm": 1.9453262090682983, + "learning_rate": 2.7863525588952073e-06, + "loss": 2.1112, + "step": 1379 + }, + { + "epoch": 1.5123287671232877, + "grad_norm": 2.0385119915008545, + "learning_rate": 2.784321689683185e-06, + "loss": 2.1068, + "step": 1380 + }, + { + "epoch": 1.5123287671232877, + "eval_loss": 2.2804832458496094, + "eval_runtime": 50.1317, + "eval_samples_per_second": 19.947, + "eval_steps_per_second": 0.838, + "step": 1380 + }, + { + "epoch": 1.5134246575342467, + "grad_norm": 1.9497289657592773, + "learning_rate": 2.782290820471162e-06, + "loss": 2.1017, + "step": 1381 + }, + { + "epoch": 1.5145205479452055, + "grad_norm": 1.9402867555618286, + "learning_rate": 2.7802599512591395e-06, + "loss": 2.1141, + "step": 1382 + }, + { + "epoch": 1.5156164383561643, + "grad_norm": 2.008141279220581, + "learning_rate": 2.7782290820471164e-06, + "loss": 2.1601, + "step": 1383 + }, + { + "epoch": 1.5167123287671234, + "grad_norm": 1.9685825109481812, + "learning_rate": 2.7761982128350932e-06, + "loss": 2.1494, + "step": 1384 + }, + { + "epoch": 1.5178082191780822, + "grad_norm": 1.9075162410736084, + "learning_rate": 2.774167343623071e-06, + "loss": 2.147, + "step": 1385 + }, + { + "epoch": 1.518904109589041, + "grad_norm": 1.8381123542785645, + "learning_rate": 2.772136474411048e-06, + "loss": 2.1772, + "step": 1386 + }, + { + "epoch": 1.52, + "grad_norm": 1.806057095527649, + "learning_rate": 2.7701056051990255e-06, + "loss": 2.1077, + "step": 1387 + }, + { + "epoch": 1.521095890410959, + "grad_norm": 1.9030957221984863, + "learning_rate": 2.7680747359870023e-06, + "loss": 2.0735, + "step": 1388 + }, + { + "epoch": 1.5221917808219179, + "grad_norm": 1.9967273473739624, + "learning_rate": 2.76604386677498e-06, + "loss": 2.155, + "step": 1389 + }, + { + "epoch": 1.5232876712328767, + "grad_norm": 1.864998698234558, + "learning_rate": 2.764012997562957e-06, + "loss": 2.1319, + "step": 1390 + }, + { + "epoch": 1.5232876712328767, + "eval_loss": 2.279405355453491, + "eval_runtime": 49.6092, + "eval_samples_per_second": 20.158, + "eval_steps_per_second": 0.847, + "step": 1390 + }, + { + "epoch": 1.5243835616438357, + "grad_norm": 1.831916093826294, + "learning_rate": 2.7619821283509346e-06, + "loss": 2.1518, + "step": 1391 + }, + { + "epoch": 1.5254794520547945, + "grad_norm": 2.1456778049468994, + "learning_rate": 2.7599512591389114e-06, + "loss": 2.0955, + "step": 1392 + }, + { + "epoch": 1.5265753424657533, + "grad_norm": 1.9264456033706665, + "learning_rate": 2.757920389926889e-06, + "loss": 2.1594, + "step": 1393 + }, + { + "epoch": 1.5276712328767124, + "grad_norm": 1.9881576299667358, + "learning_rate": 2.755889520714866e-06, + "loss": 2.115, + "step": 1394 + }, + { + "epoch": 1.5287671232876714, + "grad_norm": 1.9104706048965454, + "learning_rate": 2.7538586515028437e-06, + "loss": 2.1074, + "step": 1395 + }, + { + "epoch": 1.52986301369863, + "grad_norm": 1.8815401792526245, + "learning_rate": 2.7518277822908205e-06, + "loss": 2.0779, + "step": 1396 + }, + { + "epoch": 1.530958904109589, + "grad_norm": 1.999870777130127, + "learning_rate": 2.749796913078798e-06, + "loss": 2.1409, + "step": 1397 + }, + { + "epoch": 1.532054794520548, + "grad_norm": 2.037266969680786, + "learning_rate": 2.747766043866775e-06, + "loss": 2.1565, + "step": 1398 + }, + { + "epoch": 1.5331506849315069, + "grad_norm": 1.865891695022583, + "learning_rate": 2.7457351746547524e-06, + "loss": 2.1269, + "step": 1399 + }, + { + "epoch": 1.5342465753424657, + "grad_norm": 1.959797739982605, + "learning_rate": 2.7437043054427297e-06, + "loss": 2.1674, + "step": 1400 + }, + { + "epoch": 1.5342465753424657, + "eval_loss": 2.279064893722534, + "eval_runtime": 50.0471, + "eval_samples_per_second": 19.981, + "eval_steps_per_second": 0.839, + "step": 1400 + }, + { + "epoch": 1.5353424657534247, + "grad_norm": 2.021819829940796, + "learning_rate": 2.741673436230707e-06, + "loss": 2.2088, + "step": 1401 + }, + { + "epoch": 1.5364383561643835, + "grad_norm": 1.8549301624298096, + "learning_rate": 2.739642567018684e-06, + "loss": 2.1384, + "step": 1402 + }, + { + "epoch": 1.5375342465753423, + "grad_norm": 2.0776336193084717, + "learning_rate": 2.7376116978066615e-06, + "loss": 2.1068, + "step": 1403 + }, + { + "epoch": 1.5386301369863014, + "grad_norm": 1.9860707521438599, + "learning_rate": 2.7355808285946388e-06, + "loss": 2.1377, + "step": 1404 + }, + { + "epoch": 1.5397260273972604, + "grad_norm": 2.009176015853882, + "learning_rate": 2.733549959382616e-06, + "loss": 2.136, + "step": 1405 + }, + { + "epoch": 1.5408219178082192, + "grad_norm": 1.8771592378616333, + "learning_rate": 2.7315190901705933e-06, + "loss": 2.1104, + "step": 1406 + }, + { + "epoch": 1.541917808219178, + "grad_norm": 1.9801877737045288, + "learning_rate": 2.7294882209585706e-06, + "loss": 2.1803, + "step": 1407 + }, + { + "epoch": 1.543013698630137, + "grad_norm": 1.8967032432556152, + "learning_rate": 2.727457351746548e-06, + "loss": 2.141, + "step": 1408 + }, + { + "epoch": 1.5441095890410959, + "grad_norm": 1.8221750259399414, + "learning_rate": 2.725426482534525e-06, + "loss": 2.1181, + "step": 1409 + }, + { + "epoch": 1.5452054794520547, + "grad_norm": 1.972421407699585, + "learning_rate": 2.723395613322502e-06, + "loss": 2.1056, + "step": 1410 + }, + { + "epoch": 1.5452054794520547, + "eval_loss": 2.2803397178649902, + "eval_runtime": 42.0131, + "eval_samples_per_second": 23.802, + "eval_steps_per_second": 1.0, + "step": 1410 + }, + { + "epoch": 1.5463013698630137, + "grad_norm": 1.9438509941101074, + "learning_rate": 2.7213647441104797e-06, + "loss": 2.1378, + "step": 1411 + }, + { + "epoch": 1.5473972602739727, + "grad_norm": 1.8502674102783203, + "learning_rate": 2.7193338748984565e-06, + "loss": 2.1244, + "step": 1412 + }, + { + "epoch": 1.5484931506849315, + "grad_norm": 1.8806159496307373, + "learning_rate": 2.7173030056864342e-06, + "loss": 2.1371, + "step": 1413 + }, + { + "epoch": 1.5495890410958904, + "grad_norm": 1.823891282081604, + "learning_rate": 2.715272136474411e-06, + "loss": 2.2526, + "step": 1414 + }, + { + "epoch": 1.5506849315068494, + "grad_norm": 1.9969522953033447, + "learning_rate": 2.713241267262389e-06, + "loss": 2.0843, + "step": 1415 + }, + { + "epoch": 1.5517808219178082, + "grad_norm": 1.9167888164520264, + "learning_rate": 2.7112103980503656e-06, + "loss": 2.139, + "step": 1416 + }, + { + "epoch": 1.552876712328767, + "grad_norm": 1.8950753211975098, + "learning_rate": 2.7091795288383433e-06, + "loss": 2.1184, + "step": 1417 + }, + { + "epoch": 1.553972602739726, + "grad_norm": 1.8647392988204956, + "learning_rate": 2.70714865962632e-06, + "loss": 2.1466, + "step": 1418 + }, + { + "epoch": 1.555068493150685, + "grad_norm": 2.074019432067871, + "learning_rate": 2.705117790414298e-06, + "loss": 2.1154, + "step": 1419 + }, + { + "epoch": 1.5561643835616439, + "grad_norm": 1.883457064628601, + "learning_rate": 2.7030869212022747e-06, + "loss": 2.1251, + "step": 1420 + }, + { + "epoch": 1.5561643835616439, + "eval_loss": 2.279707908630371, + "eval_runtime": 43.6933, + "eval_samples_per_second": 22.887, + "eval_steps_per_second": 0.961, + "step": 1420 + }, + { + "epoch": 1.5572602739726027, + "grad_norm": 1.88644540309906, + "learning_rate": 2.7010560519902516e-06, + "loss": 2.1554, + "step": 1421 + }, + { + "epoch": 1.5583561643835617, + "grad_norm": 1.960942029953003, + "learning_rate": 2.6990251827782293e-06, + "loss": 2.1134, + "step": 1422 + }, + { + "epoch": 1.5594520547945205, + "grad_norm": 1.9701318740844727, + "learning_rate": 2.696994313566206e-06, + "loss": 2.1212, + "step": 1423 + }, + { + "epoch": 1.5605479452054793, + "grad_norm": 1.9110147953033447, + "learning_rate": 2.694963444354184e-06, + "loss": 2.1245, + "step": 1424 + }, + { + "epoch": 1.5616438356164384, + "grad_norm": 2.1578176021575928, + "learning_rate": 2.6929325751421607e-06, + "loss": 2.1809, + "step": 1425 + }, + { + "epoch": 1.5627397260273974, + "grad_norm": 1.9082913398742676, + "learning_rate": 2.6909017059301384e-06, + "loss": 2.1186, + "step": 1426 + }, + { + "epoch": 1.563835616438356, + "grad_norm": 1.9049410820007324, + "learning_rate": 2.6888708367181153e-06, + "loss": 2.1531, + "step": 1427 + }, + { + "epoch": 1.564931506849315, + "grad_norm": 2.041297197341919, + "learning_rate": 2.686839967506093e-06, + "loss": 2.0996, + "step": 1428 + }, + { + "epoch": 1.566027397260274, + "grad_norm": 1.9282728433609009, + "learning_rate": 2.68480909829407e-06, + "loss": 2.1492, + "step": 1429 + }, + { + "epoch": 1.5671232876712329, + "grad_norm": 1.908298134803772, + "learning_rate": 2.6827782290820475e-06, + "loss": 2.0726, + "step": 1430 + }, + { + "epoch": 1.5671232876712329, + "eval_loss": 2.278843402862549, + "eval_runtime": 43.3233, + "eval_samples_per_second": 23.082, + "eval_steps_per_second": 0.969, + "step": 1430 + }, + { + "epoch": 1.5682191780821917, + "grad_norm": 2.0996971130371094, + "learning_rate": 2.6807473598700244e-06, + "loss": 2.0829, + "step": 1431 + }, + { + "epoch": 1.5693150684931507, + "grad_norm": 1.8816159963607788, + "learning_rate": 2.678716490658002e-06, + "loss": 2.1659, + "step": 1432 + }, + { + "epoch": 1.5704109589041098, + "grad_norm": 1.91497004032135, + "learning_rate": 2.676685621445979e-06, + "loss": 2.1437, + "step": 1433 + }, + { + "epoch": 1.5715068493150683, + "grad_norm": 1.8993713855743408, + "learning_rate": 2.674654752233956e-06, + "loss": 2.1324, + "step": 1434 + }, + { + "epoch": 1.5726027397260274, + "grad_norm": 1.9361183643341064, + "learning_rate": 2.6726238830219335e-06, + "loss": 2.138, + "step": 1435 + }, + { + "epoch": 1.5736986301369864, + "grad_norm": 1.8992613554000854, + "learning_rate": 2.6705930138099107e-06, + "loss": 2.1309, + "step": 1436 + }, + { + "epoch": 1.5747945205479452, + "grad_norm": 1.9301854372024536, + "learning_rate": 2.668562144597888e-06, + "loss": 2.1645, + "step": 1437 + }, + { + "epoch": 1.575890410958904, + "grad_norm": 1.8373525142669678, + "learning_rate": 2.6665312753858653e-06, + "loss": 2.0699, + "step": 1438 + }, + { + "epoch": 1.576986301369863, + "grad_norm": 2.0120513439178467, + "learning_rate": 2.6645004061738426e-06, + "loss": 2.1472, + "step": 1439 + }, + { + "epoch": 1.5780821917808219, + "grad_norm": 2.007274627685547, + "learning_rate": 2.66246953696182e-06, + "loss": 2.1406, + "step": 1440 + }, + { + "epoch": 1.5780821917808219, + "eval_loss": 2.278693914413452, + "eval_runtime": 43.8767, + "eval_samples_per_second": 22.791, + "eval_steps_per_second": 0.957, + "step": 1440 + }, + { + "epoch": 1.5791780821917807, + "grad_norm": 1.9184056520462036, + "learning_rate": 2.660438667749797e-06, + "loss": 2.1076, + "step": 1441 + }, + { + "epoch": 1.5802739726027397, + "grad_norm": 1.9140634536743164, + "learning_rate": 2.6584077985377744e-06, + "loss": 2.0868, + "step": 1442 + }, + { + "epoch": 1.5813698630136988, + "grad_norm": 1.9146252870559692, + "learning_rate": 2.6563769293257517e-06, + "loss": 2.1205, + "step": 1443 + }, + { + "epoch": 1.5824657534246576, + "grad_norm": 1.8893204927444458, + "learning_rate": 2.654346060113729e-06, + "loss": 2.1249, + "step": 1444 + }, + { + "epoch": 1.5835616438356164, + "grad_norm": 2.0452592372894287, + "learning_rate": 2.6523151909017062e-06, + "loss": 2.1069, + "step": 1445 + }, + { + "epoch": 1.5846575342465754, + "grad_norm": 1.8728071451187134, + "learning_rate": 2.6502843216896835e-06, + "loss": 2.1161, + "step": 1446 + }, + { + "epoch": 1.5857534246575342, + "grad_norm": 2.1464476585388184, + "learning_rate": 2.6482534524776603e-06, + "loss": 2.1032, + "step": 1447 + }, + { + "epoch": 1.586849315068493, + "grad_norm": 1.9760676622390747, + "learning_rate": 2.646222583265638e-06, + "loss": 2.1754, + "step": 1448 + }, + { + "epoch": 1.587945205479452, + "grad_norm": 1.9876340627670288, + "learning_rate": 2.644191714053615e-06, + "loss": 2.1512, + "step": 1449 + }, + { + "epoch": 1.589041095890411, + "grad_norm": 2.039198160171509, + "learning_rate": 2.6421608448415926e-06, + "loss": 2.1416, + "step": 1450 + }, + { + "epoch": 1.589041095890411, + "eval_loss": 2.2811355590820312, + "eval_runtime": 44.209, + "eval_samples_per_second": 22.62, + "eval_steps_per_second": 0.95, + "step": 1450 + }, + { + "epoch": 1.59013698630137, + "grad_norm": 1.882146954536438, + "learning_rate": 2.6401299756295694e-06, + "loss": 2.1106, + "step": 1451 + }, + { + "epoch": 1.5912328767123287, + "grad_norm": 2.0288634300231934, + "learning_rate": 2.638099106417547e-06, + "loss": 2.1011, + "step": 1452 + }, + { + "epoch": 1.5923287671232877, + "grad_norm": 2.0438144207000732, + "learning_rate": 2.636068237205524e-06, + "loss": 2.1352, + "step": 1453 + }, + { + "epoch": 1.5934246575342466, + "grad_norm": 1.9043279886245728, + "learning_rate": 2.6340373679935017e-06, + "loss": 2.1044, + "step": 1454 + }, + { + "epoch": 1.5945205479452054, + "grad_norm": 2.112670660018921, + "learning_rate": 2.6320064987814785e-06, + "loss": 2.1724, + "step": 1455 + }, + { + "epoch": 1.5956164383561644, + "grad_norm": 1.8258006572723389, + "learning_rate": 2.6299756295694562e-06, + "loss": 2.0825, + "step": 1456 + }, + { + "epoch": 1.5967123287671234, + "grad_norm": 1.970802664756775, + "learning_rate": 2.627944760357433e-06, + "loss": 2.1431, + "step": 1457 + }, + { + "epoch": 1.5978082191780822, + "grad_norm": 1.9067145586013794, + "learning_rate": 2.6259138911454104e-06, + "loss": 2.1729, + "step": 1458 + }, + { + "epoch": 1.598904109589041, + "grad_norm": 1.9728924036026, + "learning_rate": 2.6238830219333877e-06, + "loss": 2.1773, + "step": 1459 + }, + { + "epoch": 1.6, + "grad_norm": 2.0142018795013428, + "learning_rate": 2.621852152721365e-06, + "loss": 2.1626, + "step": 1460 + }, + { + "epoch": 1.6, + "eval_loss": 2.2811660766601562, + "eval_runtime": 44.1735, + "eval_samples_per_second": 22.638, + "eval_steps_per_second": 0.951, + "step": 1460 + }, + { + "epoch": 1.601095890410959, + "grad_norm": 2.064955472946167, + "learning_rate": 2.619821283509342e-06, + "loss": 2.0833, + "step": 1461 + }, + { + "epoch": 1.6021917808219177, + "grad_norm": 1.9600436687469482, + "learning_rate": 2.6177904142973195e-06, + "loss": 2.1596, + "step": 1462 + }, + { + "epoch": 1.6032876712328767, + "grad_norm": 1.8413004875183105, + "learning_rate": 2.6157595450852968e-06, + "loss": 2.0957, + "step": 1463 + }, + { + "epoch": 1.6043835616438358, + "grad_norm": 2.1802690029144287, + "learning_rate": 2.613728675873274e-06, + "loss": 2.1432, + "step": 1464 + }, + { + "epoch": 1.6054794520547944, + "grad_norm": 1.992746353149414, + "learning_rate": 2.6116978066612513e-06, + "loss": 2.1631, + "step": 1465 + }, + { + "epoch": 1.6065753424657534, + "grad_norm": 1.9906671047210693, + "learning_rate": 2.6096669374492286e-06, + "loss": 2.1587, + "step": 1466 + }, + { + "epoch": 1.6076712328767124, + "grad_norm": 2.1134135723114014, + "learning_rate": 2.607636068237206e-06, + "loss": 2.1385, + "step": 1467 + }, + { + "epoch": 1.6087671232876712, + "grad_norm": 2.259915590286255, + "learning_rate": 2.605605199025183e-06, + "loss": 2.1042, + "step": 1468 + }, + { + "epoch": 1.60986301369863, + "grad_norm": 2.0705254077911377, + "learning_rate": 2.6035743298131604e-06, + "loss": 2.145, + "step": 1469 + }, + { + "epoch": 1.610958904109589, + "grad_norm": 2.0326738357543945, + "learning_rate": 2.6015434606011377e-06, + "loss": 2.133, + "step": 1470 + }, + { + "epoch": 1.610958904109589, + "eval_loss": 2.2812676429748535, + "eval_runtime": 43.4091, + "eval_samples_per_second": 23.037, + "eval_steps_per_second": 0.968, + "step": 1470 + }, + { + "epoch": 1.612054794520548, + "grad_norm": 2.034182071685791, + "learning_rate": 2.5995125913891145e-06, + "loss": 2.1507, + "step": 1471 + }, + { + "epoch": 1.6131506849315067, + "grad_norm": 2.069373846054077, + "learning_rate": 2.5974817221770922e-06, + "loss": 2.152, + "step": 1472 + }, + { + "epoch": 1.6142465753424657, + "grad_norm": 2.2003285884857178, + "learning_rate": 2.595450852965069e-06, + "loss": 2.1347, + "step": 1473 + }, + { + "epoch": 1.6153424657534248, + "grad_norm": 1.919945478439331, + "learning_rate": 2.593419983753047e-06, + "loss": 2.1476, + "step": 1474 + }, + { + "epoch": 1.6164383561643836, + "grad_norm": 1.825148344039917, + "learning_rate": 2.5913891145410236e-06, + "loss": 2.1757, + "step": 1475 + }, + { + "epoch": 1.6175342465753424, + "grad_norm": 2.0014286041259766, + "learning_rate": 2.5893582453290013e-06, + "loss": 2.1208, + "step": 1476 + }, + { + "epoch": 1.6186301369863014, + "grad_norm": 2.0642764568328857, + "learning_rate": 2.587327376116978e-06, + "loss": 2.1619, + "step": 1477 + }, + { + "epoch": 1.6197260273972602, + "grad_norm": 2.002497911453247, + "learning_rate": 2.585296506904956e-06, + "loss": 2.1198, + "step": 1478 + }, + { + "epoch": 1.620821917808219, + "grad_norm": 1.7852193117141724, + "learning_rate": 2.5832656376929327e-06, + "loss": 2.1247, + "step": 1479 + }, + { + "epoch": 1.621917808219178, + "grad_norm": 1.9539082050323486, + "learning_rate": 2.5812347684809104e-06, + "loss": 2.1046, + "step": 1480 + }, + { + "epoch": 1.621917808219178, + "eval_loss": 2.2807998657226562, + "eval_runtime": 44.2896, + "eval_samples_per_second": 22.579, + "eval_steps_per_second": 0.948, + "step": 1480 + }, + { + "epoch": 1.6230136986301371, + "grad_norm": 2.095280647277832, + "learning_rate": 2.5792038992688873e-06, + "loss": 2.0685, + "step": 1481 + }, + { + "epoch": 1.624109589041096, + "grad_norm": 2.230388879776001, + "learning_rate": 2.577173030056865e-06, + "loss": 2.1976, + "step": 1482 + }, + { + "epoch": 1.6252054794520547, + "grad_norm": 1.9304745197296143, + "learning_rate": 2.575142160844842e-06, + "loss": 2.0647, + "step": 1483 + }, + { + "epoch": 1.6263013698630138, + "grad_norm": 1.856342077255249, + "learning_rate": 2.5731112916328187e-06, + "loss": 2.1444, + "step": 1484 + }, + { + "epoch": 1.6273972602739726, + "grad_norm": 2.117694854736328, + "learning_rate": 2.5710804224207964e-06, + "loss": 2.1536, + "step": 1485 + }, + { + "epoch": 1.6284931506849314, + "grad_norm": 2.043900489807129, + "learning_rate": 2.5690495532087733e-06, + "loss": 2.106, + "step": 1486 + }, + { + "epoch": 1.6295890410958904, + "grad_norm": 1.9283581972122192, + "learning_rate": 2.567018683996751e-06, + "loss": 2.0667, + "step": 1487 + }, + { + "epoch": 1.6306849315068495, + "grad_norm": 1.8138651847839355, + "learning_rate": 2.564987814784728e-06, + "loss": 2.1386, + "step": 1488 + }, + { + "epoch": 1.6317808219178083, + "grad_norm": 2.0534632205963135, + "learning_rate": 2.5629569455727055e-06, + "loss": 2.105, + "step": 1489 + }, + { + "epoch": 1.632876712328767, + "grad_norm": 1.9466804265975952, + "learning_rate": 2.5609260763606824e-06, + "loss": 2.1211, + "step": 1490 + }, + { + "epoch": 1.632876712328767, + "eval_loss": 2.275447368621826, + "eval_runtime": 44.2319, + "eval_samples_per_second": 22.608, + "eval_steps_per_second": 0.95, + "step": 1490 + }, + { + "epoch": 1.633972602739726, + "grad_norm": 1.8767902851104736, + "learning_rate": 2.55889520714866e-06, + "loss": 2.116, + "step": 1491 + }, + { + "epoch": 1.635068493150685, + "grad_norm": 1.9194284677505493, + "learning_rate": 2.556864337936637e-06, + "loss": 2.1928, + "step": 1492 + }, + { + "epoch": 1.6361643835616437, + "grad_norm": 2.0496408939361572, + "learning_rate": 2.5548334687246146e-06, + "loss": 2.1182, + "step": 1493 + }, + { + "epoch": 1.6372602739726028, + "grad_norm": 1.8577388525009155, + "learning_rate": 2.5528025995125915e-06, + "loss": 2.1313, + "step": 1494 + }, + { + "epoch": 1.6383561643835618, + "grad_norm": 2.020623207092285, + "learning_rate": 2.550771730300569e-06, + "loss": 2.1627, + "step": 1495 + }, + { + "epoch": 1.6394520547945204, + "grad_norm": 1.8472096920013428, + "learning_rate": 2.548740861088546e-06, + "loss": 2.1514, + "step": 1496 + }, + { + "epoch": 1.6405479452054794, + "grad_norm": 1.9105209112167358, + "learning_rate": 2.5467099918765233e-06, + "loss": 2.1558, + "step": 1497 + }, + { + "epoch": 1.6416438356164385, + "grad_norm": 1.9635803699493408, + "learning_rate": 2.5446791226645006e-06, + "loss": 2.08, + "step": 1498 + }, + { + "epoch": 1.6427397260273973, + "grad_norm": 1.8911043405532837, + "learning_rate": 2.542648253452478e-06, + "loss": 2.0648, + "step": 1499 + }, + { + "epoch": 1.643835616438356, + "grad_norm": 1.9337847232818604, + "learning_rate": 2.540617384240455e-06, + "loss": 2.1526, + "step": 1500 + }, + { + "epoch": 1.643835616438356, + "eval_loss": 2.2771341800689697, + "eval_runtime": 44.2753, + "eval_samples_per_second": 22.586, + "eval_steps_per_second": 0.949, + "step": 1500 + }, + { + "epoch": 1.644931506849315, + "grad_norm": 2.152970552444458, + "learning_rate": 2.5385865150284324e-06, + "loss": 2.1328, + "step": 1501 + }, + { + "epoch": 1.6460273972602741, + "grad_norm": 1.8749687671661377, + "learning_rate": 2.5365556458164097e-06, + "loss": 2.1093, + "step": 1502 + }, + { + "epoch": 1.6471232876712327, + "grad_norm": 1.8745299577713013, + "learning_rate": 2.534524776604387e-06, + "loss": 2.1172, + "step": 1503 + }, + { + "epoch": 1.6482191780821918, + "grad_norm": 2.006110906600952, + "learning_rate": 2.5324939073923642e-06, + "loss": 2.1318, + "step": 1504 + }, + { + "epoch": 1.6493150684931508, + "grad_norm": 2.055590867996216, + "learning_rate": 2.5304630381803415e-06, + "loss": 2.0463, + "step": 1505 + }, + { + "epoch": 1.6504109589041096, + "grad_norm": 1.9270331859588623, + "learning_rate": 2.5284321689683188e-06, + "loss": 2.1195, + "step": 1506 + }, + { + "epoch": 1.6515068493150684, + "grad_norm": 1.9981428384780884, + "learning_rate": 2.526401299756296e-06, + "loss": 2.1282, + "step": 1507 + }, + { + "epoch": 1.6526027397260274, + "grad_norm": 2.051969528198242, + "learning_rate": 2.524370430544273e-06, + "loss": 2.1013, + "step": 1508 + }, + { + "epoch": 1.6536986301369863, + "grad_norm": 2.067718744277954, + "learning_rate": 2.5223395613322506e-06, + "loss": 2.1265, + "step": 1509 + }, + { + "epoch": 1.654794520547945, + "grad_norm": 2.0843262672424316, + "learning_rate": 2.5203086921202274e-06, + "loss": 2.0994, + "step": 1510 + }, + { + "epoch": 1.654794520547945, + "eval_loss": 2.27781081199646, + "eval_runtime": 42.8213, + "eval_samples_per_second": 23.353, + "eval_steps_per_second": 0.981, + "step": 1510 + }, + { + "epoch": 1.655890410958904, + "grad_norm": 2.146921396255493, + "learning_rate": 2.518277822908205e-06, + "loss": 2.1456, + "step": 1511 + }, + { + "epoch": 1.6569863013698631, + "grad_norm": 1.9604636430740356, + "learning_rate": 2.516246953696182e-06, + "loss": 2.1118, + "step": 1512 + }, + { + "epoch": 1.658082191780822, + "grad_norm": 1.9613317251205444, + "learning_rate": 2.5142160844841597e-06, + "loss": 2.1401, + "step": 1513 + }, + { + "epoch": 1.6591780821917808, + "grad_norm": 2.0410237312316895, + "learning_rate": 2.5121852152721365e-06, + "loss": 2.1532, + "step": 1514 + }, + { + "epoch": 1.6602739726027398, + "grad_norm": 1.8603837490081787, + "learning_rate": 2.5101543460601142e-06, + "loss": 2.1437, + "step": 1515 + }, + { + "epoch": 1.6613698630136986, + "grad_norm": 1.9498400688171387, + "learning_rate": 2.508123476848091e-06, + "loss": 2.1384, + "step": 1516 + }, + { + "epoch": 1.6624657534246574, + "grad_norm": 1.8603938817977905, + "learning_rate": 2.506092607636069e-06, + "loss": 2.0903, + "step": 1517 + }, + { + "epoch": 1.6635616438356164, + "grad_norm": 1.8238614797592163, + "learning_rate": 2.5040617384240457e-06, + "loss": 2.1644, + "step": 1518 + }, + { + "epoch": 1.6646575342465755, + "grad_norm": 1.9574072360992432, + "learning_rate": 2.5020308692120234e-06, + "loss": 2.1072, + "step": 1519 + }, + { + "epoch": 1.6657534246575343, + "grad_norm": 2.100829601287842, + "learning_rate": 2.5e-06, + "loss": 2.111, + "step": 1520 + }, + { + "epoch": 1.6657534246575343, + "eval_loss": 2.278552293777466, + "eval_runtime": 46.587, + "eval_samples_per_second": 21.465, + "eval_steps_per_second": 0.902, + "step": 1520 + }, + { + "epoch": 1.666849315068493, + "grad_norm": 1.921262264251709, + "learning_rate": 2.4979691307879775e-06, + "loss": 2.1667, + "step": 1521 + }, + { + "epoch": 1.6679452054794521, + "grad_norm": 1.9328664541244507, + "learning_rate": 2.4959382615759548e-06, + "loss": 2.1175, + "step": 1522 + }, + { + "epoch": 1.669041095890411, + "grad_norm": 1.909544587135315, + "learning_rate": 2.493907392363932e-06, + "loss": 2.0905, + "step": 1523 + }, + { + "epoch": 1.6701369863013698, + "grad_norm": 1.928046464920044, + "learning_rate": 2.4918765231519093e-06, + "loss": 2.1757, + "step": 1524 + }, + { + "epoch": 1.6712328767123288, + "grad_norm": 1.8839048147201538, + "learning_rate": 2.4898456539398866e-06, + "loss": 2.1424, + "step": 1525 + }, + { + "epoch": 1.6723287671232878, + "grad_norm": 1.913812518119812, + "learning_rate": 2.487814784727864e-06, + "loss": 2.1178, + "step": 1526 + }, + { + "epoch": 1.6734246575342466, + "grad_norm": 1.7485435009002686, + "learning_rate": 2.4857839155158407e-06, + "loss": 2.1794, + "step": 1527 + }, + { + "epoch": 1.6745205479452054, + "grad_norm": 1.8693492412567139, + "learning_rate": 2.483753046303818e-06, + "loss": 2.1265, + "step": 1528 + }, + { + "epoch": 1.6756164383561645, + "grad_norm": 1.8982772827148438, + "learning_rate": 2.4817221770917953e-06, + "loss": 2.1254, + "step": 1529 + }, + { + "epoch": 1.6767123287671233, + "grad_norm": 1.9917978048324585, + "learning_rate": 2.4796913078797725e-06, + "loss": 2.13, + "step": 1530 + }, + { + "epoch": 1.6767123287671233, + "eval_loss": 2.277986764907837, + "eval_runtime": 52.1464, + "eval_samples_per_second": 19.177, + "eval_steps_per_second": 0.805, + "step": 1530 + }, + { + "epoch": 1.677808219178082, + "grad_norm": 1.8703829050064087, + "learning_rate": 2.47766043866775e-06, + "loss": 2.1527, + "step": 1531 + }, + { + "epoch": 1.6789041095890411, + "grad_norm": 1.950713872909546, + "learning_rate": 2.475629569455727e-06, + "loss": 2.1481, + "step": 1532 + }, + { + "epoch": 1.6800000000000002, + "grad_norm": 1.8339334726333618, + "learning_rate": 2.4735987002437044e-06, + "loss": 2.1422, + "step": 1533 + }, + { + "epoch": 1.6810958904109587, + "grad_norm": 1.8423672914505005, + "learning_rate": 2.4715678310316816e-06, + "loss": 2.1026, + "step": 1534 + }, + { + "epoch": 1.6821917808219178, + "grad_norm": 1.9338176250457764, + "learning_rate": 2.469536961819659e-06, + "loss": 2.1095, + "step": 1535 + }, + { + "epoch": 1.6832876712328768, + "grad_norm": 1.8666706085205078, + "learning_rate": 2.467506092607636e-06, + "loss": 2.1176, + "step": 1536 + }, + { + "epoch": 1.6843835616438356, + "grad_norm": 2.0762805938720703, + "learning_rate": 2.4654752233956135e-06, + "loss": 2.127, + "step": 1537 + }, + { + "epoch": 1.6854794520547944, + "grad_norm": 1.9729993343353271, + "learning_rate": 2.4634443541835907e-06, + "loss": 2.1194, + "step": 1538 + }, + { + "epoch": 1.6865753424657535, + "grad_norm": 1.910555362701416, + "learning_rate": 2.461413484971568e-06, + "loss": 2.1745, + "step": 1539 + }, + { + "epoch": 1.6876712328767123, + "grad_norm": 1.953911304473877, + "learning_rate": 2.4593826157595453e-06, + "loss": 2.1236, + "step": 1540 + }, + { + "epoch": 1.6876712328767123, + "eval_loss": 2.277575731277466, + "eval_runtime": 48.0126, + "eval_samples_per_second": 20.828, + "eval_steps_per_second": 0.875, + "step": 1540 + }, + { + "epoch": 1.688767123287671, + "grad_norm": 2.0469322204589844, + "learning_rate": 2.4573517465475226e-06, + "loss": 2.1999, + "step": 1541 + }, + { + "epoch": 1.6898630136986301, + "grad_norm": 1.9962514638900757, + "learning_rate": 2.4553208773355e-06, + "loss": 2.0865, + "step": 1542 + }, + { + "epoch": 1.6909589041095892, + "grad_norm": 1.911471962928772, + "learning_rate": 2.453290008123477e-06, + "loss": 2.1107, + "step": 1543 + }, + { + "epoch": 1.692054794520548, + "grad_norm": 1.9354206323623657, + "learning_rate": 2.4512591389114544e-06, + "loss": 2.1215, + "step": 1544 + }, + { + "epoch": 1.6931506849315068, + "grad_norm": 1.938178539276123, + "learning_rate": 2.4492282696994317e-06, + "loss": 2.1378, + "step": 1545 + }, + { + "epoch": 1.6942465753424658, + "grad_norm": 2.003552198410034, + "learning_rate": 2.447197400487409e-06, + "loss": 2.1361, + "step": 1546 + }, + { + "epoch": 1.6953424657534246, + "grad_norm": 1.9754462242126465, + "learning_rate": 2.4451665312753862e-06, + "loss": 2.134, + "step": 1547 + }, + { + "epoch": 1.6964383561643834, + "grad_norm": 1.924299955368042, + "learning_rate": 2.4431356620633635e-06, + "loss": 2.1497, + "step": 1548 + }, + { + "epoch": 1.6975342465753425, + "grad_norm": 2.1239371299743652, + "learning_rate": 2.4411047928513408e-06, + "loss": 2.1427, + "step": 1549 + }, + { + "epoch": 1.6986301369863015, + "grad_norm": 1.984377384185791, + "learning_rate": 2.439073923639318e-06, + "loss": 2.1268, + "step": 1550 + }, + { + "epoch": 1.6986301369863015, + "eval_loss": 2.2747738361358643, + "eval_runtime": 53.2249, + "eval_samples_per_second": 18.788, + "eval_steps_per_second": 0.789, + "step": 1550 + }, + { + "epoch": 1.6997260273972603, + "grad_norm": 1.9375758171081543, + "learning_rate": 2.4370430544272953e-06, + "loss": 2.1369, + "step": 1551 + }, + { + "epoch": 1.7008219178082191, + "grad_norm": 1.9191703796386719, + "learning_rate": 2.435012185215272e-06, + "loss": 2.1061, + "step": 1552 + }, + { + "epoch": 1.7019178082191782, + "grad_norm": 2.1356418132781982, + "learning_rate": 2.4329813160032495e-06, + "loss": 2.0923, + "step": 1553 + }, + { + "epoch": 1.703013698630137, + "grad_norm": 2.02870774269104, + "learning_rate": 2.4309504467912267e-06, + "loss": 2.105, + "step": 1554 + }, + { + "epoch": 1.7041095890410958, + "grad_norm": 1.9436805248260498, + "learning_rate": 2.428919577579204e-06, + "loss": 2.1194, + "step": 1555 + }, + { + "epoch": 1.7052054794520548, + "grad_norm": 1.997019648551941, + "learning_rate": 2.4268887083671813e-06, + "loss": 2.1114, + "step": 1556 + }, + { + "epoch": 1.7063013698630138, + "grad_norm": 2.0759427547454834, + "learning_rate": 2.4248578391551586e-06, + "loss": 2.1955, + "step": 1557 + }, + { + "epoch": 1.7073972602739727, + "grad_norm": 2.0526654720306396, + "learning_rate": 2.422826969943136e-06, + "loss": 2.1103, + "step": 1558 + }, + { + "epoch": 1.7084931506849315, + "grad_norm": 1.8765089511871338, + "learning_rate": 2.420796100731113e-06, + "loss": 2.1068, + "step": 1559 + }, + { + "epoch": 1.7095890410958905, + "grad_norm": 1.9164950847625732, + "learning_rate": 2.4187652315190904e-06, + "loss": 2.082, + "step": 1560 + }, + { + "epoch": 1.7095890410958905, + "eval_loss": 2.272998332977295, + "eval_runtime": 49.0487, + "eval_samples_per_second": 20.388, + "eval_steps_per_second": 0.856, + "step": 1560 + }, + { + "epoch": 1.7106849315068493, + "grad_norm": 1.9921822547912598, + "learning_rate": 2.4167343623070677e-06, + "loss": 2.1285, + "step": 1561 + }, + { + "epoch": 1.7117808219178081, + "grad_norm": 1.8890291452407837, + "learning_rate": 2.414703493095045e-06, + "loss": 2.1385, + "step": 1562 + }, + { + "epoch": 1.7128767123287671, + "grad_norm": 1.9233667850494385, + "learning_rate": 2.4126726238830222e-06, + "loss": 2.1282, + "step": 1563 + }, + { + "epoch": 1.7139726027397262, + "grad_norm": 2.0368800163269043, + "learning_rate": 2.410641754670999e-06, + "loss": 2.1366, + "step": 1564 + }, + { + "epoch": 1.7150684931506848, + "grad_norm": 2.0321929454803467, + "learning_rate": 2.4086108854589763e-06, + "loss": 2.1163, + "step": 1565 + }, + { + "epoch": 1.7161643835616438, + "grad_norm": 2.042773485183716, + "learning_rate": 2.4065800162469536e-06, + "loss": 2.1529, + "step": 1566 + }, + { + "epoch": 1.7172602739726028, + "grad_norm": 1.937940239906311, + "learning_rate": 2.404549147034931e-06, + "loss": 2.106, + "step": 1567 + }, + { + "epoch": 1.7183561643835616, + "grad_norm": 1.878649353981018, + "learning_rate": 2.402518277822908e-06, + "loss": 2.1179, + "step": 1568 + }, + { + "epoch": 1.7194520547945205, + "grad_norm": 1.997246265411377, + "learning_rate": 2.4004874086108854e-06, + "loss": 2.1672, + "step": 1569 + }, + { + "epoch": 1.7205479452054795, + "grad_norm": 1.9837015867233276, + "learning_rate": 2.3984565393988627e-06, + "loss": 2.0816, + "step": 1570 + }, + { + "epoch": 1.7205479452054795, + "eval_loss": 2.2730956077575684, + "eval_runtime": 52.1793, + "eval_samples_per_second": 19.165, + "eval_steps_per_second": 0.805, + "step": 1570 + }, + { + "epoch": 1.7216438356164385, + "grad_norm": 1.8919317722320557, + "learning_rate": 2.39642567018684e-06, + "loss": 2.1541, + "step": 1571 + }, + { + "epoch": 1.7227397260273971, + "grad_norm": 1.8129284381866455, + "learning_rate": 2.3943948009748173e-06, + "loss": 2.1033, + "step": 1572 + }, + { + "epoch": 1.7238356164383561, + "grad_norm": 1.79408860206604, + "learning_rate": 2.3923639317627945e-06, + "loss": 2.1163, + "step": 1573 + }, + { + "epoch": 1.7249315068493152, + "grad_norm": 1.9178274869918823, + "learning_rate": 2.390333062550772e-06, + "loss": 2.1383, + "step": 1574 + }, + { + "epoch": 1.726027397260274, + "grad_norm": 1.9113222360610962, + "learning_rate": 2.388302193338749e-06, + "loss": 2.0987, + "step": 1575 + }, + { + "epoch": 1.7271232876712328, + "grad_norm": 1.915170431137085, + "learning_rate": 2.3862713241267264e-06, + "loss": 2.0235, + "step": 1576 + }, + { + "epoch": 1.7282191780821918, + "grad_norm": 1.9789308309555054, + "learning_rate": 2.3842404549147037e-06, + "loss": 2.1942, + "step": 1577 + }, + { + "epoch": 1.7293150684931506, + "grad_norm": 1.8410224914550781, + "learning_rate": 2.382209585702681e-06, + "loss": 2.1105, + "step": 1578 + }, + { + "epoch": 1.7304109589041095, + "grad_norm": 1.9388853311538696, + "learning_rate": 2.380178716490658e-06, + "loss": 2.1021, + "step": 1579 + }, + { + "epoch": 1.7315068493150685, + "grad_norm": 1.9328628778457642, + "learning_rate": 2.3781478472786355e-06, + "loss": 2.1584, + "step": 1580 + }, + { + "epoch": 1.7315068493150685, + "eval_loss": 2.271240472793579, + "eval_runtime": 47.9206, + "eval_samples_per_second": 20.868, + "eval_steps_per_second": 0.876, + "step": 1580 + }, + { + "epoch": 1.7326027397260275, + "grad_norm": 1.8846843242645264, + "learning_rate": 2.3761169780666128e-06, + "loss": 2.1269, + "step": 1581 + }, + { + "epoch": 1.7336986301369863, + "grad_norm": 1.8232983350753784, + "learning_rate": 2.37408610885459e-06, + "loss": 2.1522, + "step": 1582 + }, + { + "epoch": 1.7347945205479451, + "grad_norm": 2.023982286453247, + "learning_rate": 2.3720552396425673e-06, + "loss": 2.1571, + "step": 1583 + }, + { + "epoch": 1.7358904109589042, + "grad_norm": 1.861229658126831, + "learning_rate": 2.3700243704305446e-06, + "loss": 2.103, + "step": 1584 + }, + { + "epoch": 1.736986301369863, + "grad_norm": 1.8577842712402344, + "learning_rate": 2.367993501218522e-06, + "loss": 2.1766, + "step": 1585 + }, + { + "epoch": 1.7380821917808218, + "grad_norm": 1.915722131729126, + "learning_rate": 2.365962632006499e-06, + "loss": 2.1306, + "step": 1586 + }, + { + "epoch": 1.7391780821917808, + "grad_norm": 1.8876861333847046, + "learning_rate": 2.3639317627944764e-06, + "loss": 2.1303, + "step": 1587 + }, + { + "epoch": 1.7402739726027399, + "grad_norm": 1.9599688053131104, + "learning_rate": 2.3619008935824537e-06, + "loss": 2.1513, + "step": 1588 + }, + { + "epoch": 1.7413698630136987, + "grad_norm": 1.90068519115448, + "learning_rate": 2.3598700243704305e-06, + "loss": 2.1459, + "step": 1589 + }, + { + "epoch": 1.7424657534246575, + "grad_norm": 1.920277714729309, + "learning_rate": 2.357839155158408e-06, + "loss": 2.1453, + "step": 1590 + }, + { + "epoch": 1.7424657534246575, + "eval_loss": 2.2699742317199707, + "eval_runtime": 50.0211, + "eval_samples_per_second": 19.992, + "eval_steps_per_second": 0.84, + "step": 1590 + }, + { + "epoch": 1.7435616438356165, + "grad_norm": 1.9827759265899658, + "learning_rate": 2.355808285946385e-06, + "loss": 2.1393, + "step": 1591 + }, + { + "epoch": 1.7446575342465753, + "grad_norm": 1.8602412939071655, + "learning_rate": 2.3537774167343624e-06, + "loss": 2.1317, + "step": 1592 + }, + { + "epoch": 1.7457534246575341, + "grad_norm": 1.964829444885254, + "learning_rate": 2.3517465475223396e-06, + "loss": 2.1552, + "step": 1593 + }, + { + "epoch": 1.7468493150684932, + "grad_norm": 2.0124034881591797, + "learning_rate": 2.349715678310317e-06, + "loss": 2.1322, + "step": 1594 + }, + { + "epoch": 1.7479452054794522, + "grad_norm": 1.9786088466644287, + "learning_rate": 2.347684809098294e-06, + "loss": 2.1435, + "step": 1595 + }, + { + "epoch": 1.749041095890411, + "grad_norm": 1.9575989246368408, + "learning_rate": 2.3456539398862715e-06, + "loss": 2.1032, + "step": 1596 + }, + { + "epoch": 1.7501369863013698, + "grad_norm": 2.0460586547851562, + "learning_rate": 2.3436230706742487e-06, + "loss": 2.1595, + "step": 1597 + }, + { + "epoch": 1.7512328767123289, + "grad_norm": 1.9904825687408447, + "learning_rate": 2.341592201462226e-06, + "loss": 2.1025, + "step": 1598 + }, + { + "epoch": 1.7523287671232877, + "grad_norm": 2.0797626972198486, + "learning_rate": 2.3395613322502033e-06, + "loss": 2.1221, + "step": 1599 + }, + { + "epoch": 1.7534246575342465, + "grad_norm": 2.0380358695983887, + "learning_rate": 2.3375304630381806e-06, + "loss": 2.1166, + "step": 1600 + }, + { + "epoch": 1.7534246575342465, + "eval_loss": 2.2704837322235107, + "eval_runtime": 51.9207, + "eval_samples_per_second": 19.26, + "eval_steps_per_second": 0.809, + "step": 1600 + }, + { + "epoch": 1.7545205479452055, + "grad_norm": 2.009023427963257, + "learning_rate": 2.335499593826158e-06, + "loss": 2.0922, + "step": 1601 + }, + { + "epoch": 1.7556164383561645, + "grad_norm": 2.043168783187866, + "learning_rate": 2.333468724614135e-06, + "loss": 2.0962, + "step": 1602 + }, + { + "epoch": 1.7567123287671231, + "grad_norm": 2.1130902767181396, + "learning_rate": 2.3314378554021124e-06, + "loss": 2.1318, + "step": 1603 + }, + { + "epoch": 1.7578082191780822, + "grad_norm": 1.9391744136810303, + "learning_rate": 2.3294069861900897e-06, + "loss": 2.155, + "step": 1604 + }, + { + "epoch": 1.7589041095890412, + "grad_norm": 2.0810694694519043, + "learning_rate": 2.327376116978067e-06, + "loss": 2.1066, + "step": 1605 + }, + { + "epoch": 1.76, + "grad_norm": 2.029707431793213, + "learning_rate": 2.3253452477660442e-06, + "loss": 2.0897, + "step": 1606 + }, + { + "epoch": 1.7610958904109588, + "grad_norm": 1.928080439567566, + "learning_rate": 2.3233143785540215e-06, + "loss": 2.1589, + "step": 1607 + }, + { + "epoch": 1.7621917808219179, + "grad_norm": 1.8527079820632935, + "learning_rate": 2.3212835093419988e-06, + "loss": 2.1005, + "step": 1608 + }, + { + "epoch": 1.7632876712328767, + "grad_norm": 2.142939567565918, + "learning_rate": 2.319252640129976e-06, + "loss": 2.1499, + "step": 1609 + }, + { + "epoch": 1.7643835616438355, + "grad_norm": 2.040922164916992, + "learning_rate": 2.3172217709179533e-06, + "loss": 2.0997, + "step": 1610 + }, + { + "epoch": 1.7643835616438355, + "eval_loss": 2.2691540718078613, + "eval_runtime": 42.4134, + "eval_samples_per_second": 23.577, + "eval_steps_per_second": 0.99, + "step": 1610 + }, + { + "epoch": 1.7654794520547945, + "grad_norm": 1.988370656967163, + "learning_rate": 2.3151909017059306e-06, + "loss": 2.1795, + "step": 1611 + }, + { + "epoch": 1.7665753424657535, + "grad_norm": 1.996656060218811, + "learning_rate": 2.313160032493908e-06, + "loss": 2.1536, + "step": 1612 + }, + { + "epoch": 1.7676712328767124, + "grad_norm": 2.04560923576355, + "learning_rate": 2.311129163281885e-06, + "loss": 2.0781, + "step": 1613 + }, + { + "epoch": 1.7687671232876712, + "grad_norm": 1.8977863788604736, + "learning_rate": 2.309098294069862e-06, + "loss": 2.1312, + "step": 1614 + }, + { + "epoch": 1.7698630136986302, + "grad_norm": 2.026904821395874, + "learning_rate": 2.3070674248578393e-06, + "loss": 2.0998, + "step": 1615 + }, + { + "epoch": 1.770958904109589, + "grad_norm": 1.8714247941970825, + "learning_rate": 2.3050365556458166e-06, + "loss": 2.0875, + "step": 1616 + }, + { + "epoch": 1.7720547945205478, + "grad_norm": 2.11859130859375, + "learning_rate": 2.303005686433794e-06, + "loss": 2.1361, + "step": 1617 + }, + { + "epoch": 1.7731506849315068, + "grad_norm": 2.0592799186706543, + "learning_rate": 2.300974817221771e-06, + "loss": 2.1394, + "step": 1618 + }, + { + "epoch": 1.7742465753424659, + "grad_norm": 2.0358338356018066, + "learning_rate": 2.2989439480097484e-06, + "loss": 2.104, + "step": 1619 + }, + { + "epoch": 1.7753424657534247, + "grad_norm": 2.0237109661102295, + "learning_rate": 2.2969130787977257e-06, + "loss": 2.0823, + "step": 1620 + }, + { + "epoch": 1.7753424657534247, + "eval_loss": 2.27001690864563, + "eval_runtime": 43.9168, + "eval_samples_per_second": 22.77, + "eval_steps_per_second": 0.956, + "step": 1620 + }, + { + "epoch": 1.7764383561643835, + "grad_norm": 1.8729594945907593, + "learning_rate": 2.294882209585703e-06, + "loss": 2.1243, + "step": 1621 + }, + { + "epoch": 1.7775342465753425, + "grad_norm": 1.9390734434127808, + "learning_rate": 2.2928513403736802e-06, + "loss": 2.1469, + "step": 1622 + }, + { + "epoch": 1.7786301369863013, + "grad_norm": 2.0409750938415527, + "learning_rate": 2.2908204711616575e-06, + "loss": 2.1062, + "step": 1623 + }, + { + "epoch": 1.7797260273972602, + "grad_norm": 1.9116135835647583, + "learning_rate": 2.2887896019496348e-06, + "loss": 2.1208, + "step": 1624 + }, + { + "epoch": 1.7808219178082192, + "grad_norm": 1.9144036769866943, + "learning_rate": 2.286758732737612e-06, + "loss": 2.1628, + "step": 1625 + }, + { + "epoch": 1.7819178082191782, + "grad_norm": 1.8522213697433472, + "learning_rate": 2.284727863525589e-06, + "loss": 2.1168, + "step": 1626 + }, + { + "epoch": 1.783013698630137, + "grad_norm": 2.087541103363037, + "learning_rate": 2.282696994313566e-06, + "loss": 2.1195, + "step": 1627 + }, + { + "epoch": 1.7841095890410958, + "grad_norm": 2.0399985313415527, + "learning_rate": 2.2806661251015434e-06, + "loss": 2.1557, + "step": 1628 + }, + { + "epoch": 1.7852054794520549, + "grad_norm": 1.884964108467102, + "learning_rate": 2.2786352558895207e-06, + "loss": 2.1408, + "step": 1629 + }, + { + "epoch": 1.7863013698630137, + "grad_norm": 1.901802897453308, + "learning_rate": 2.276604386677498e-06, + "loss": 2.1145, + "step": 1630 + }, + { + "epoch": 1.7863013698630137, + "eval_loss": 2.2690439224243164, + "eval_runtime": 44.8662, + "eval_samples_per_second": 22.289, + "eval_steps_per_second": 0.936, + "step": 1630 + }, + { + "epoch": 1.7873972602739725, + "grad_norm": 1.8818615674972534, + "learning_rate": 2.2745735174654753e-06, + "loss": 2.0879, + "step": 1631 + }, + { + "epoch": 1.7884931506849315, + "grad_norm": 1.9345687627792358, + "learning_rate": 2.2725426482534526e-06, + "loss": 2.1029, + "step": 1632 + }, + { + "epoch": 1.7895890410958906, + "grad_norm": 1.868314504623413, + "learning_rate": 2.27051177904143e-06, + "loss": 2.1435, + "step": 1633 + }, + { + "epoch": 1.7906849315068492, + "grad_norm": 1.8868515491485596, + "learning_rate": 2.268480909829407e-06, + "loss": 2.1294, + "step": 1634 + }, + { + "epoch": 1.7917808219178082, + "grad_norm": 1.9929358959197998, + "learning_rate": 2.2664500406173844e-06, + "loss": 2.1165, + "step": 1635 + }, + { + "epoch": 1.7928767123287672, + "grad_norm": 1.9766895771026611, + "learning_rate": 2.2644191714053617e-06, + "loss": 2.1327, + "step": 1636 + }, + { + "epoch": 1.793972602739726, + "grad_norm": 2.03336238861084, + "learning_rate": 2.262388302193339e-06, + "loss": 2.1473, + "step": 1637 + }, + { + "epoch": 1.7950684931506848, + "grad_norm": 1.8569762706756592, + "learning_rate": 2.260357432981316e-06, + "loss": 2.125, + "step": 1638 + }, + { + "epoch": 1.7961643835616439, + "grad_norm": 1.9522744417190552, + "learning_rate": 2.2583265637692935e-06, + "loss": 2.1238, + "step": 1639 + }, + { + "epoch": 1.7972602739726027, + "grad_norm": 1.9175306558609009, + "learning_rate": 2.2562956945572708e-06, + "loss": 2.1332, + "step": 1640 + }, + { + "epoch": 1.7972602739726027, + "eval_loss": 2.2702176570892334, + "eval_runtime": 44.1369, + "eval_samples_per_second": 22.657, + "eval_steps_per_second": 0.952, + "step": 1640 + }, + { + "epoch": 1.7983561643835615, + "grad_norm": 2.109614849090576, + "learning_rate": 2.254264825345248e-06, + "loss": 2.1252, + "step": 1641 + }, + { + "epoch": 1.7994520547945205, + "grad_norm": 2.017613410949707, + "learning_rate": 2.2522339561332253e-06, + "loss": 2.1093, + "step": 1642 + }, + { + "epoch": 1.8005479452054796, + "grad_norm": 2.000864028930664, + "learning_rate": 2.2502030869212026e-06, + "loss": 2.1243, + "step": 1643 + }, + { + "epoch": 1.8016438356164384, + "grad_norm": 1.9952439069747925, + "learning_rate": 2.24817221770918e-06, + "loss": 2.1213, + "step": 1644 + }, + { + "epoch": 1.8027397260273972, + "grad_norm": 2.0456748008728027, + "learning_rate": 2.246141348497157e-06, + "loss": 2.1621, + "step": 1645 + }, + { + "epoch": 1.8038356164383562, + "grad_norm": 1.9011833667755127, + "learning_rate": 2.2441104792851344e-06, + "loss": 2.1426, + "step": 1646 + }, + { + "epoch": 1.804931506849315, + "grad_norm": 1.9874658584594727, + "learning_rate": 2.2420796100731117e-06, + "loss": 2.1085, + "step": 1647 + }, + { + "epoch": 1.8060273972602738, + "grad_norm": 2.0237057209014893, + "learning_rate": 2.240048740861089e-06, + "loss": 2.1394, + "step": 1648 + }, + { + "epoch": 1.8071232876712329, + "grad_norm": 2.0127220153808594, + "learning_rate": 2.2380178716490662e-06, + "loss": 2.1474, + "step": 1649 + }, + { + "epoch": 1.808219178082192, + "grad_norm": 2.051722764968872, + "learning_rate": 2.2359870024370435e-06, + "loss": 2.0444, + "step": 1650 + }, + { + "epoch": 1.808219178082192, + "eval_loss": 2.2701034545898438, + "eval_runtime": 44.4338, + "eval_samples_per_second": 22.505, + "eval_steps_per_second": 0.945, + "step": 1650 + }, + { + "epoch": 1.8093150684931507, + "grad_norm": 2.0204219818115234, + "learning_rate": 2.2339561332250204e-06, + "loss": 2.1445, + "step": 1651 + }, + { + "epoch": 1.8104109589041095, + "grad_norm": 2.106220006942749, + "learning_rate": 2.2319252640129976e-06, + "loss": 2.153, + "step": 1652 + }, + { + "epoch": 1.8115068493150686, + "grad_norm": 1.8632844686508179, + "learning_rate": 2.229894394800975e-06, + "loss": 2.137, + "step": 1653 + }, + { + "epoch": 1.8126027397260274, + "grad_norm": 2.377120018005371, + "learning_rate": 2.227863525588952e-06, + "loss": 2.0903, + "step": 1654 + }, + { + "epoch": 1.8136986301369862, + "grad_norm": 2.1345274448394775, + "learning_rate": 2.2258326563769295e-06, + "loss": 2.066, + "step": 1655 + }, + { + "epoch": 1.8147945205479452, + "grad_norm": 1.8935933113098145, + "learning_rate": 2.2238017871649067e-06, + "loss": 2.083, + "step": 1656 + }, + { + "epoch": 1.8158904109589042, + "grad_norm": 1.8766530752182007, + "learning_rate": 2.221770917952884e-06, + "loss": 2.0792, + "step": 1657 + }, + { + "epoch": 1.816986301369863, + "grad_norm": 2.24202823638916, + "learning_rate": 2.2197400487408613e-06, + "loss": 2.1276, + "step": 1658 + }, + { + "epoch": 1.8180821917808219, + "grad_norm": 2.168358325958252, + "learning_rate": 2.2177091795288386e-06, + "loss": 2.1497, + "step": 1659 + }, + { + "epoch": 1.819178082191781, + "grad_norm": 2.052833080291748, + "learning_rate": 2.215678310316816e-06, + "loss": 2.0853, + "step": 1660 + }, + { + "epoch": 1.819178082191781, + "eval_loss": 2.270122528076172, + "eval_runtime": 44.1793, + "eval_samples_per_second": 22.635, + "eval_steps_per_second": 0.951, + "step": 1660 + }, + { + "epoch": 1.8202739726027397, + "grad_norm": 2.09708571434021, + "learning_rate": 2.213647441104793e-06, + "loss": 2.1151, + "step": 1661 + }, + { + "epoch": 1.8213698630136985, + "grad_norm": 1.928503155708313, + "learning_rate": 2.2116165718927704e-06, + "loss": 2.1234, + "step": 1662 + }, + { + "epoch": 1.8224657534246576, + "grad_norm": 1.9545074701309204, + "learning_rate": 2.2095857026807473e-06, + "loss": 2.1134, + "step": 1663 + }, + { + "epoch": 1.8235616438356166, + "grad_norm": 2.1161389350891113, + "learning_rate": 2.2075548334687245e-06, + "loss": 2.1347, + "step": 1664 + }, + { + "epoch": 1.8246575342465754, + "grad_norm": 1.9292371273040771, + "learning_rate": 2.205523964256702e-06, + "loss": 2.0911, + "step": 1665 + }, + { + "epoch": 1.8257534246575342, + "grad_norm": 1.8706191778182983, + "learning_rate": 2.203493095044679e-06, + "loss": 2.0917, + "step": 1666 + }, + { + "epoch": 1.8268493150684932, + "grad_norm": 1.9825111627578735, + "learning_rate": 2.2014622258326564e-06, + "loss": 2.125, + "step": 1667 + }, + { + "epoch": 1.827945205479452, + "grad_norm": 2.0262181758880615, + "learning_rate": 2.1994313566206336e-06, + "loss": 2.153, + "step": 1668 + }, + { + "epoch": 1.8290410958904109, + "grad_norm": 2.0071237087249756, + "learning_rate": 2.197400487408611e-06, + "loss": 2.1625, + "step": 1669 + }, + { + "epoch": 1.83013698630137, + "grad_norm": 1.940076231956482, + "learning_rate": 2.195369618196588e-06, + "loss": 2.1052, + "step": 1670 + }, + { + "epoch": 1.83013698630137, + "eval_loss": 2.2711451053619385, + "eval_runtime": 43.9311, + "eval_samples_per_second": 22.763, + "eval_steps_per_second": 0.956, + "step": 1670 + }, + { + "epoch": 1.831232876712329, + "grad_norm": 2.103487014770508, + "learning_rate": 2.1933387489845655e-06, + "loss": 2.087, + "step": 1671 + }, + { + "epoch": 1.8323287671232875, + "grad_norm": 1.935276985168457, + "learning_rate": 2.1913078797725427e-06, + "loss": 2.1196, + "step": 1672 + }, + { + "epoch": 1.8334246575342465, + "grad_norm": 1.987322449684143, + "learning_rate": 2.18927701056052e-06, + "loss": 2.1354, + "step": 1673 + }, + { + "epoch": 1.8345205479452056, + "grad_norm": 2.0403592586517334, + "learning_rate": 2.1872461413484973e-06, + "loss": 2.1715, + "step": 1674 + }, + { + "epoch": 1.8356164383561644, + "grad_norm": 2.0403592586517334, + "learning_rate": 2.1872461413484973e-06, + "loss": 2.1153, + "step": 1675 + }, + { + "epoch": 1.8367123287671232, + "grad_norm": 2.160753011703491, + "learning_rate": 2.1852152721364746e-06, + "loss": 2.1105, + "step": 1676 + }, + { + "epoch": 1.8378082191780822, + "grad_norm": 1.8320366144180298, + "learning_rate": 2.183184402924452e-06, + "loss": 2.0853, + "step": 1677 + }, + { + "epoch": 1.838904109589041, + "grad_norm": 1.9792540073394775, + "learning_rate": 2.181153533712429e-06, + "loss": 2.1478, + "step": 1678 + }, + { + "epoch": 1.8399999999999999, + "grad_norm": 2.0344042778015137, + "learning_rate": 2.1791226645004064e-06, + "loss": 2.1177, + "step": 1679 + }, + { + "epoch": 1.841095890410959, + "grad_norm": 1.8672499656677246, + "learning_rate": 2.1770917952883837e-06, + "loss": 2.1098, + "step": 1680 + }, + { + "epoch": 1.841095890410959, + "eval_loss": 2.2701213359832764, + "eval_runtime": 43.0484, + "eval_samples_per_second": 23.23, + "eval_steps_per_second": 0.976, + "step": 1680 + }, + { + "epoch": 1.842191780821918, + "grad_norm": 1.9291775226593018, + "learning_rate": 2.175060926076361e-06, + "loss": 2.1659, + "step": 1681 + }, + { + "epoch": 1.8432876712328767, + "grad_norm": 1.9650444984436035, + "learning_rate": 2.1730300568643382e-06, + "loss": 2.1764, + "step": 1682 + }, + { + "epoch": 1.8443835616438355, + "grad_norm": 1.904618501663208, + "learning_rate": 2.1709991876523155e-06, + "loss": 2.1047, + "step": 1683 + }, + { + "epoch": 1.8454794520547946, + "grad_norm": 2.0084192752838135, + "learning_rate": 2.1689683184402928e-06, + "loss": 2.0719, + "step": 1684 + }, + { + "epoch": 1.8465753424657534, + "grad_norm": 2.0233888626098633, + "learning_rate": 2.16693744922827e-06, + "loss": 2.1482, + "step": 1685 + }, + { + "epoch": 1.8476712328767122, + "grad_norm": 1.9512053728103638, + "learning_rate": 2.1649065800162473e-06, + "loss": 2.1116, + "step": 1686 + }, + { + "epoch": 1.8487671232876712, + "grad_norm": 1.9261832237243652, + "learning_rate": 2.1628757108042246e-06, + "loss": 2.1025, + "step": 1687 + }, + { + "epoch": 1.8498630136986303, + "grad_norm": 2.0704665184020996, + "learning_rate": 2.160844841592202e-06, + "loss": 2.0943, + "step": 1688 + }, + { + "epoch": 1.850958904109589, + "grad_norm": 1.8882733583450317, + "learning_rate": 2.1588139723801787e-06, + "loss": 2.09, + "step": 1689 + }, + { + "epoch": 1.8520547945205479, + "grad_norm": 1.9136085510253906, + "learning_rate": 2.156783103168156e-06, + "loss": 2.1074, + "step": 1690 + }, + { + "epoch": 1.8520547945205479, + "eval_loss": 2.2687957286834717, + "eval_runtime": 44.1571, + "eval_samples_per_second": 22.646, + "eval_steps_per_second": 0.951, + "step": 1690 + }, + { + "epoch": 1.853150684931507, + "grad_norm": 1.9338082075119019, + "learning_rate": 2.1547522339561333e-06, + "loss": 2.0945, + "step": 1691 + }, + { + "epoch": 1.8542465753424657, + "grad_norm": 1.862281084060669, + "learning_rate": 2.1527213647441106e-06, + "loss": 2.1355, + "step": 1692 + }, + { + "epoch": 1.8553424657534245, + "grad_norm": 1.8184510469436646, + "learning_rate": 2.150690495532088e-06, + "loss": 2.1628, + "step": 1693 + }, + { + "epoch": 1.8564383561643836, + "grad_norm": 1.9667037725448608, + "learning_rate": 2.148659626320065e-06, + "loss": 2.1308, + "step": 1694 + }, + { + "epoch": 1.8575342465753426, + "grad_norm": 2.0214083194732666, + "learning_rate": 2.1466287571080424e-06, + "loss": 2.1363, + "step": 1695 + }, + { + "epoch": 1.8586301369863014, + "grad_norm": 1.8136624097824097, + "learning_rate": 2.1445978878960197e-06, + "loss": 2.1074, + "step": 1696 + }, + { + "epoch": 1.8597260273972602, + "grad_norm": 2.1120195388793945, + "learning_rate": 2.142567018683997e-06, + "loss": 2.1115, + "step": 1697 + }, + { + "epoch": 1.8608219178082193, + "grad_norm": 1.9148973226547241, + "learning_rate": 2.140536149471974e-06, + "loss": 2.1554, + "step": 1698 + }, + { + "epoch": 1.861917808219178, + "grad_norm": 1.9684176445007324, + "learning_rate": 2.1385052802599515e-06, + "loss": 2.0944, + "step": 1699 + }, + { + "epoch": 1.8630136986301369, + "grad_norm": 2.0816597938537598, + "learning_rate": 2.1364744110479288e-06, + "loss": 2.1063, + "step": 1700 + }, + { + "epoch": 1.8630136986301369, + "eval_loss": 2.2695536613464355, + "eval_runtime": 48.9529, + "eval_samples_per_second": 20.428, + "eval_steps_per_second": 0.858, + "step": 1700 + }, + { + "epoch": 1.864109589041096, + "grad_norm": 1.9451358318328857, + "learning_rate": 2.134443541835906e-06, + "loss": 2.1521, + "step": 1701 + }, + { + "epoch": 1.865205479452055, + "grad_norm": 1.9210634231567383, + "learning_rate": 2.132412672623883e-06, + "loss": 2.1216, + "step": 1702 + }, + { + "epoch": 1.8663013698630135, + "grad_norm": 2.0976240634918213, + "learning_rate": 2.13038180341186e-06, + "loss": 2.0947, + "step": 1703 + }, + { + "epoch": 1.8673972602739726, + "grad_norm": 1.9917279481887817, + "learning_rate": 2.1283509341998374e-06, + "loss": 2.1267, + "step": 1704 + }, + { + "epoch": 1.8684931506849316, + "grad_norm": 2.1310553550720215, + "learning_rate": 2.1263200649878147e-06, + "loss": 2.1566, + "step": 1705 + }, + { + "epoch": 1.8695890410958904, + "grad_norm": 2.0364415645599365, + "learning_rate": 2.124289195775792e-06, + "loss": 2.1077, + "step": 1706 + }, + { + "epoch": 1.8706849315068492, + "grad_norm": 1.8810032606124878, + "learning_rate": 2.1222583265637693e-06, + "loss": 2.1061, + "step": 1707 + }, + { + "epoch": 1.8717808219178083, + "grad_norm": 2.034438371658325, + "learning_rate": 2.1202274573517465e-06, + "loss": 2.0523, + "step": 1708 + }, + { + "epoch": 1.872876712328767, + "grad_norm": 2.0231120586395264, + "learning_rate": 2.118196588139724e-06, + "loss": 2.0829, + "step": 1709 + }, + { + "epoch": 1.8739726027397259, + "grad_norm": 2.0092573165893555, + "learning_rate": 2.116165718927701e-06, + "loss": 2.1101, + "step": 1710 + }, + { + "epoch": 1.8739726027397259, + "eval_loss": 2.2675492763519287, + "eval_runtime": 43.0128, + "eval_samples_per_second": 23.249, + "eval_steps_per_second": 0.976, + "step": 1710 + }, + { + "epoch": 1.875068493150685, + "grad_norm": 1.884216070175171, + "learning_rate": 2.1141348497156784e-06, + "loss": 2.1249, + "step": 1711 + }, + { + "epoch": 1.876164383561644, + "grad_norm": 1.954170823097229, + "learning_rate": 2.1121039805036556e-06, + "loss": 2.0915, + "step": 1712 + }, + { + "epoch": 1.8772602739726028, + "grad_norm": 2.0208632946014404, + "learning_rate": 2.110073111291633e-06, + "loss": 2.0824, + "step": 1713 + }, + { + "epoch": 1.8783561643835616, + "grad_norm": 1.9104353189468384, + "learning_rate": 2.10804224207961e-06, + "loss": 2.1265, + "step": 1714 + }, + { + "epoch": 1.8794520547945206, + "grad_norm": 1.945338249206543, + "learning_rate": 2.1060113728675875e-06, + "loss": 2.1403, + "step": 1715 + }, + { + "epoch": 1.8805479452054794, + "grad_norm": 2.006586790084839, + "learning_rate": 2.1039805036555647e-06, + "loss": 2.1439, + "step": 1716 + }, + { + "epoch": 1.8816438356164382, + "grad_norm": 1.941493034362793, + "learning_rate": 2.101949634443542e-06, + "loss": 2.0415, + "step": 1717 + }, + { + "epoch": 1.8827397260273973, + "grad_norm": 2.146116018295288, + "learning_rate": 2.0999187652315193e-06, + "loss": 2.1118, + "step": 1718 + }, + { + "epoch": 1.8838356164383563, + "grad_norm": 2.0794291496276855, + "learning_rate": 2.0978878960194966e-06, + "loss": 2.0731, + "step": 1719 + }, + { + "epoch": 1.884931506849315, + "grad_norm": 2.1100215911865234, + "learning_rate": 2.095857026807474e-06, + "loss": 2.1303, + "step": 1720 + }, + { + "epoch": 1.884931506849315, + "eval_loss": 2.267274856567383, + "eval_runtime": 48.0078, + "eval_samples_per_second": 20.83, + "eval_steps_per_second": 0.875, + "step": 1720 + }, + { + "epoch": 1.886027397260274, + "grad_norm": 1.9509919881820679, + "learning_rate": 2.093826157595451e-06, + "loss": 2.1191, + "step": 1721 + }, + { + "epoch": 1.887123287671233, + "grad_norm": 2.0161266326904297, + "learning_rate": 2.0917952883834284e-06, + "loss": 2.104, + "step": 1722 + }, + { + "epoch": 1.8882191780821918, + "grad_norm": 2.07547664642334, + "learning_rate": 2.0897644191714057e-06, + "loss": 2.1299, + "step": 1723 + }, + { + "epoch": 1.8893150684931506, + "grad_norm": 2.1406030654907227, + "learning_rate": 2.087733549959383e-06, + "loss": 2.1385, + "step": 1724 + }, + { + "epoch": 1.8904109589041096, + "grad_norm": 1.9793925285339355, + "learning_rate": 2.0857026807473602e-06, + "loss": 2.0814, + "step": 1725 + }, + { + "epoch": 1.8915068493150686, + "grad_norm": 1.9256374835968018, + "learning_rate": 2.0836718115353375e-06, + "loss": 2.0708, + "step": 1726 + }, + { + "epoch": 1.8926027397260274, + "grad_norm": 1.960571527481079, + "learning_rate": 2.0816409423233144e-06, + "loss": 2.156, + "step": 1727 + }, + { + "epoch": 1.8936986301369862, + "grad_norm": 2.1507158279418945, + "learning_rate": 2.0796100731112916e-06, + "loss": 2.0958, + "step": 1728 + }, + { + "epoch": 1.8947945205479453, + "grad_norm": 2.2358651161193848, + "learning_rate": 2.077579203899269e-06, + "loss": 2.11, + "step": 1729 + }, + { + "epoch": 1.895890410958904, + "grad_norm": 2.0311756134033203, + "learning_rate": 2.075548334687246e-06, + "loss": 2.072, + "step": 1730 + }, + { + "epoch": 1.895890410958904, + "eval_loss": 2.2663629055023193, + "eval_runtime": 50.5422, + "eval_samples_per_second": 19.785, + "eval_steps_per_second": 0.831, + "step": 1730 + }, + { + "epoch": 1.896986301369863, + "grad_norm": 1.98939049243927, + "learning_rate": 2.0735174654752235e-06, + "loss": 2.1327, + "step": 1731 + }, + { + "epoch": 1.898082191780822, + "grad_norm": 2.009493350982666, + "learning_rate": 2.0714865962632007e-06, + "loss": 2.0798, + "step": 1732 + }, + { + "epoch": 1.899178082191781, + "grad_norm": 2.1011271476745605, + "learning_rate": 2.069455727051178e-06, + "loss": 2.1194, + "step": 1733 + }, + { + "epoch": 1.9002739726027398, + "grad_norm": 1.9027225971221924, + "learning_rate": 2.0674248578391553e-06, + "loss": 2.0955, + "step": 1734 + }, + { + "epoch": 1.9013698630136986, + "grad_norm": 1.9784867763519287, + "learning_rate": 2.0653939886271326e-06, + "loss": 2.1012, + "step": 1735 + }, + { + "epoch": 1.9024657534246576, + "grad_norm": 2.0302555561065674, + "learning_rate": 2.06336311941511e-06, + "loss": 2.1472, + "step": 1736 + }, + { + "epoch": 1.9035616438356164, + "grad_norm": 1.81975257396698, + "learning_rate": 2.061332250203087e-06, + "loss": 2.1062, + "step": 1737 + }, + { + "epoch": 1.9046575342465752, + "grad_norm": 1.9868931770324707, + "learning_rate": 2.0593013809910644e-06, + "loss": 2.0829, + "step": 1738 + }, + { + "epoch": 1.9057534246575343, + "grad_norm": 2.001765012741089, + "learning_rate": 2.0572705117790417e-06, + "loss": 2.1113, + "step": 1739 + }, + { + "epoch": 1.9068493150684933, + "grad_norm": 1.891752004623413, + "learning_rate": 2.055239642567019e-06, + "loss": 2.1311, + "step": 1740 + }, + { + "epoch": 1.9068493150684933, + "eval_loss": 2.2680604457855225, + "eval_runtime": 50.0795, + "eval_samples_per_second": 19.968, + "eval_steps_per_second": 0.839, + "step": 1740 + }, + { + "epoch": 1.907945205479452, + "grad_norm": 1.978259563446045, + "learning_rate": 2.0532087733549962e-06, + "loss": 2.1187, + "step": 1741 + }, + { + "epoch": 1.909041095890411, + "grad_norm": 1.8946164846420288, + "learning_rate": 2.0511779041429735e-06, + "loss": 2.1189, + "step": 1742 + }, + { + "epoch": 1.91013698630137, + "grad_norm": 1.8812536001205444, + "learning_rate": 2.0491470349309508e-06, + "loss": 2.0895, + "step": 1743 + }, + { + "epoch": 1.9112328767123288, + "grad_norm": 1.990494728088379, + "learning_rate": 2.047116165718928e-06, + "loss": 2.1192, + "step": 1744 + }, + { + "epoch": 1.9123287671232876, + "grad_norm": 2.0288472175598145, + "learning_rate": 2.0450852965069053e-06, + "loss": 2.1182, + "step": 1745 + }, + { + "epoch": 1.9134246575342466, + "grad_norm": 2.0876214504241943, + "learning_rate": 2.0430544272948826e-06, + "loss": 2.1546, + "step": 1746 + }, + { + "epoch": 1.9145205479452054, + "grad_norm": 1.9842756986618042, + "learning_rate": 2.04102355808286e-06, + "loss": 2.1126, + "step": 1747 + }, + { + "epoch": 1.9156164383561642, + "grad_norm": 2.0303354263305664, + "learning_rate": 2.038992688870837e-06, + "loss": 2.0771, + "step": 1748 + }, + { + "epoch": 1.9167123287671233, + "grad_norm": 2.0054662227630615, + "learning_rate": 2.0369618196588144e-06, + "loss": 2.0895, + "step": 1749 + }, + { + "epoch": 1.9178082191780823, + "grad_norm": 1.9232923984527588, + "learning_rate": 2.0349309504467917e-06, + "loss": 2.0952, + "step": 1750 + }, + { + "epoch": 1.9178082191780823, + "eval_loss": 2.266294240951538, + "eval_runtime": 53.7831, + "eval_samples_per_second": 18.593, + "eval_steps_per_second": 0.781, + "step": 1750 + }, + { + "epoch": 1.9189041095890411, + "grad_norm": 1.9103394746780396, + "learning_rate": 2.032900081234769e-06, + "loss": 2.1324, + "step": 1751 + }, + { + "epoch": 1.92, + "grad_norm": 2.0642080307006836, + "learning_rate": 2.030869212022746e-06, + "loss": 2.1495, + "step": 1752 + }, + { + "epoch": 1.921095890410959, + "grad_norm": 1.9586466550827026, + "learning_rate": 2.028838342810723e-06, + "loss": 2.1036, + "step": 1753 + }, + { + "epoch": 1.9221917808219178, + "grad_norm": 2.024435043334961, + "learning_rate": 2.0268074735987004e-06, + "loss": 2.1049, + "step": 1754 + }, + { + "epoch": 1.9232876712328766, + "grad_norm": 1.9115983247756958, + "learning_rate": 2.0247766043866777e-06, + "loss": 2.0889, + "step": 1755 + }, + { + "epoch": 1.9243835616438356, + "grad_norm": 1.9067517518997192, + "learning_rate": 2.022745735174655e-06, + "loss": 2.1432, + "step": 1756 + }, + { + "epoch": 1.9254794520547946, + "grad_norm": 2.0182650089263916, + "learning_rate": 2.020714865962632e-06, + "loss": 2.1011, + "step": 1757 + }, + { + "epoch": 1.9265753424657535, + "grad_norm": 2.0648891925811768, + "learning_rate": 2.0186839967506095e-06, + "loss": 2.1266, + "step": 1758 + }, + { + "epoch": 1.9276712328767123, + "grad_norm": 2.0217037200927734, + "learning_rate": 2.0166531275385868e-06, + "loss": 2.1206, + "step": 1759 + }, + { + "epoch": 1.9287671232876713, + "grad_norm": 1.9545989036560059, + "learning_rate": 2.014622258326564e-06, + "loss": 2.0822, + "step": 1760 + }, + { + "epoch": 1.9287671232876713, + "eval_loss": 2.2664599418640137, + "eval_runtime": 53.4817, + "eval_samples_per_second": 18.698, + "eval_steps_per_second": 0.785, + "step": 1760 + }, + { + "epoch": 1.9298630136986301, + "grad_norm": 2.047677516937256, + "learning_rate": 2.0125913891145413e-06, + "loss": 2.1169, + "step": 1761 + }, + { + "epoch": 1.930958904109589, + "grad_norm": 2.170466184616089, + "learning_rate": 2.0105605199025186e-06, + "loss": 2.1211, + "step": 1762 + }, + { + "epoch": 1.932054794520548, + "grad_norm": 1.9144518375396729, + "learning_rate": 2.008529650690496e-06, + "loss": 2.0912, + "step": 1763 + }, + { + "epoch": 1.933150684931507, + "grad_norm": 2.004936933517456, + "learning_rate": 2.0064987814784727e-06, + "loss": 2.136, + "step": 1764 + }, + { + "epoch": 1.9342465753424658, + "grad_norm": 1.8312169313430786, + "learning_rate": 2.00446791226645e-06, + "loss": 2.1151, + "step": 1765 + }, + { + "epoch": 1.9353424657534246, + "grad_norm": 1.9144090414047241, + "learning_rate": 2.0024370430544273e-06, + "loss": 2.0835, + "step": 1766 + }, + { + "epoch": 1.9364383561643836, + "grad_norm": 2.020298719406128, + "learning_rate": 2.0004061738424045e-06, + "loss": 2.1203, + "step": 1767 + }, + { + "epoch": 1.9375342465753425, + "grad_norm": 1.9231114387512207, + "learning_rate": 1.998375304630382e-06, + "loss": 2.1425, + "step": 1768 + }, + { + "epoch": 1.9386301369863013, + "grad_norm": 1.8474186658859253, + "learning_rate": 1.996344435418359e-06, + "loss": 2.1071, + "step": 1769 + }, + { + "epoch": 1.9397260273972603, + "grad_norm": 2.171400785446167, + "learning_rate": 1.9943135662063364e-06, + "loss": 2.1344, + "step": 1770 + }, + { + "epoch": 1.9397260273972603, + "eval_loss": 2.266789436340332, + "eval_runtime": 50.4218, + "eval_samples_per_second": 19.833, + "eval_steps_per_second": 0.833, + "step": 1770 + }, + { + "epoch": 1.9408219178082193, + "grad_norm": 1.929115891456604, + "learning_rate": 1.9922826969943136e-06, + "loss": 2.1205, + "step": 1771 + }, + { + "epoch": 1.941917808219178, + "grad_norm": 1.961046814918518, + "learning_rate": 1.990251827782291e-06, + "loss": 2.1396, + "step": 1772 + }, + { + "epoch": 1.943013698630137, + "grad_norm": 2.056424617767334, + "learning_rate": 1.988220958570268e-06, + "loss": 2.125, + "step": 1773 + }, + { + "epoch": 1.944109589041096, + "grad_norm": 1.9814075231552124, + "learning_rate": 1.9861900893582455e-06, + "loss": 2.1297, + "step": 1774 + }, + { + "epoch": 1.9452054794520548, + "grad_norm": 1.937506079673767, + "learning_rate": 1.9841592201462227e-06, + "loss": 2.1288, + "step": 1775 + }, + { + "epoch": 1.9463013698630136, + "grad_norm": 1.902376413345337, + "learning_rate": 1.9821283509342e-06, + "loss": 2.0782, + "step": 1776 + }, + { + "epoch": 1.9473972602739726, + "grad_norm": 2.0806174278259277, + "learning_rate": 1.9800974817221773e-06, + "loss": 2.0778, + "step": 1777 + }, + { + "epoch": 1.9484931506849315, + "grad_norm": 2.0370235443115234, + "learning_rate": 1.9780666125101546e-06, + "loss": 2.1304, + "step": 1778 + }, + { + "epoch": 1.9495890410958903, + "grad_norm": 1.980531930923462, + "learning_rate": 1.976035743298132e-06, + "loss": 2.0989, + "step": 1779 + }, + { + "epoch": 1.9506849315068493, + "grad_norm": 1.96549654006958, + "learning_rate": 1.974004874086109e-06, + "loss": 2.1556, + "step": 1780 + }, + { + "epoch": 1.9506849315068493, + "eval_loss": 2.266958475112915, + "eval_runtime": 52.3331, + "eval_samples_per_second": 19.108, + "eval_steps_per_second": 0.803, + "step": 1780 + }, + { + "epoch": 1.9517808219178083, + "grad_norm": 2.1921818256378174, + "learning_rate": 1.9719740048740864e-06, + "loss": 2.0666, + "step": 1781 + }, + { + "epoch": 1.9528767123287671, + "grad_norm": 1.9451124668121338, + "learning_rate": 1.9699431356620637e-06, + "loss": 2.1078, + "step": 1782 + }, + { + "epoch": 1.953972602739726, + "grad_norm": 1.9163907766342163, + "learning_rate": 1.967912266450041e-06, + "loss": 2.0955, + "step": 1783 + }, + { + "epoch": 1.955068493150685, + "grad_norm": 2.1186892986297607, + "learning_rate": 1.9658813972380182e-06, + "loss": 2.0826, + "step": 1784 + }, + { + "epoch": 1.9561643835616438, + "grad_norm": 2.1072819232940674, + "learning_rate": 1.9638505280259955e-06, + "loss": 2.1869, + "step": 1785 + }, + { + "epoch": 1.9572602739726026, + "grad_norm": 1.9530322551727295, + "learning_rate": 1.9618196588139728e-06, + "loss": 2.0983, + "step": 1786 + }, + { + "epoch": 1.9583561643835616, + "grad_norm": 2.0034027099609375, + "learning_rate": 1.95978878960195e-06, + "loss": 2.1121, + "step": 1787 + }, + { + "epoch": 1.9594520547945207, + "grad_norm": 2.1277832984924316, + "learning_rate": 1.9577579203899273e-06, + "loss": 2.1769, + "step": 1788 + }, + { + "epoch": 1.9605479452054795, + "grad_norm": 1.9420982599258423, + "learning_rate": 1.955727051177904e-06, + "loss": 2.1424, + "step": 1789 + }, + { + "epoch": 1.9616438356164383, + "grad_norm": 1.9175841808319092, + "learning_rate": 1.9536961819658815e-06, + "loss": 2.0957, + "step": 1790 + }, + { + "epoch": 1.9616438356164383, + "eval_loss": 2.267153739929199, + "eval_runtime": 47.5802, + "eval_samples_per_second": 21.017, + "eval_steps_per_second": 0.883, + "step": 1790 + }, + { + "epoch": 1.9627397260273973, + "grad_norm": 1.9897676706314087, + "learning_rate": 1.9516653127538587e-06, + "loss": 2.0809, + "step": 1791 + }, + { + "epoch": 1.9638356164383561, + "grad_norm": 2.0855610370635986, + "learning_rate": 1.949634443541836e-06, + "loss": 2.0924, + "step": 1792 + }, + { + "epoch": 1.964931506849315, + "grad_norm": 2.0065109729766846, + "learning_rate": 1.9476035743298133e-06, + "loss": 2.0819, + "step": 1793 + }, + { + "epoch": 1.966027397260274, + "grad_norm": 1.9256092309951782, + "learning_rate": 1.9455727051177906e-06, + "loss": 2.1389, + "step": 1794 + }, + { + "epoch": 1.967123287671233, + "grad_norm": 1.9086920022964478, + "learning_rate": 1.943541835905768e-06, + "loss": 2.1106, + "step": 1795 + }, + { + "epoch": 1.9682191780821918, + "grad_norm": 2.0217244625091553, + "learning_rate": 1.941510966693745e-06, + "loss": 2.1624, + "step": 1796 + }, + { + "epoch": 1.9693150684931506, + "grad_norm": 2.005068063735962, + "learning_rate": 1.9394800974817224e-06, + "loss": 2.0751, + "step": 1797 + }, + { + "epoch": 1.9704109589041097, + "grad_norm": 2.0949530601501465, + "learning_rate": 1.9374492282696997e-06, + "loss": 2.1447, + "step": 1798 + }, + { + "epoch": 1.9715068493150685, + "grad_norm": 1.8868833780288696, + "learning_rate": 1.935418359057677e-06, + "loss": 2.1568, + "step": 1799 + }, + { + "epoch": 1.9726027397260273, + "grad_norm": 1.889626145362854, + "learning_rate": 1.9333874898456542e-06, + "loss": 2.1152, + "step": 1800 + }, + { + "epoch": 1.9726027397260273, + "eval_loss": 2.268470287322998, + "eval_runtime": 44.7042, + "eval_samples_per_second": 22.369, + "eval_steps_per_second": 0.94, + "step": 1800 + }, + { + "epoch": 1.9736986301369863, + "grad_norm": 1.9561820030212402, + "learning_rate": 1.931356620633631e-06, + "loss": 2.1584, + "step": 1801 + }, + { + "epoch": 1.9747945205479454, + "grad_norm": 1.9774333238601685, + "learning_rate": 1.9293257514216083e-06, + "loss": 2.0726, + "step": 1802 + }, + { + "epoch": 1.9758904109589042, + "grad_norm": 1.9008616209030151, + "learning_rate": 1.9272948822095856e-06, + "loss": 2.1222, + "step": 1803 + }, + { + "epoch": 1.976986301369863, + "grad_norm": 1.9981689453125, + "learning_rate": 1.925264012997563e-06, + "loss": 2.0761, + "step": 1804 + }, + { + "epoch": 1.978082191780822, + "grad_norm": 1.926156997680664, + "learning_rate": 1.92323314378554e-06, + "loss": 2.0962, + "step": 1805 + }, + { + "epoch": 1.9791780821917808, + "grad_norm": 1.9634530544281006, + "learning_rate": 1.9212022745735174e-06, + "loss": 2.1349, + "step": 1806 + }, + { + "epoch": 1.9802739726027396, + "grad_norm": 1.9713245630264282, + "learning_rate": 1.9191714053614947e-06, + "loss": 2.0881, + "step": 1807 + }, + { + "epoch": 1.9813698630136987, + "grad_norm": 1.9258323907852173, + "learning_rate": 1.917140536149472e-06, + "loss": 2.0744, + "step": 1808 + }, + { + "epoch": 1.9824657534246577, + "grad_norm": 1.9645394086837769, + "learning_rate": 1.9151096669374493e-06, + "loss": 2.0701, + "step": 1809 + }, + { + "epoch": 1.9835616438356163, + "grad_norm": 2.145430088043213, + "learning_rate": 1.9130787977254266e-06, + "loss": 2.1372, + "step": 1810 + }, + { + "epoch": 1.9835616438356163, + "eval_loss": 2.269702196121216, + "eval_runtime": 41.513, + "eval_samples_per_second": 24.089, + "eval_steps_per_second": 1.012, + "step": 1810 + }, + { + "epoch": 1.9846575342465753, + "grad_norm": 2.1812174320220947, + "learning_rate": 1.911047928513404e-06, + "loss": 2.1209, + "step": 1811 + }, + { + "epoch": 1.9857534246575343, + "grad_norm": 1.933090329170227, + "learning_rate": 1.909017059301381e-06, + "loss": 2.0905, + "step": 1812 + }, + { + "epoch": 1.9868493150684932, + "grad_norm": 2.0478832721710205, + "learning_rate": 1.9069861900893586e-06, + "loss": 2.0921, + "step": 1813 + }, + { + "epoch": 1.987945205479452, + "grad_norm": 1.9921154975891113, + "learning_rate": 1.9049553208773354e-06, + "loss": 2.1232, + "step": 1814 + }, + { + "epoch": 1.989041095890411, + "grad_norm": 1.8685261011123657, + "learning_rate": 1.9029244516653127e-06, + "loss": 2.0978, + "step": 1815 + }, + { + "epoch": 1.9901369863013698, + "grad_norm": 2.045020818710327, + "learning_rate": 1.90089358245329e-06, + "loss": 2.0614, + "step": 1816 + }, + { + "epoch": 1.9912328767123286, + "grad_norm": 2.059267520904541, + "learning_rate": 1.8988627132412673e-06, + "loss": 2.1098, + "step": 1817 + }, + { + "epoch": 1.9923287671232877, + "grad_norm": 2.0033164024353027, + "learning_rate": 1.8968318440292445e-06, + "loss": 2.135, + "step": 1818 + }, + { + "epoch": 1.9934246575342467, + "grad_norm": 1.8544939756393433, + "learning_rate": 1.8948009748172218e-06, + "loss": 2.0644, + "step": 1819 + }, + { + "epoch": 1.9945205479452055, + "grad_norm": 1.8415113687515259, + "learning_rate": 1.892770105605199e-06, + "loss": 2.1283, + "step": 1820 + }, + { + "epoch": 1.9945205479452055, + "eval_loss": 2.267810344696045, + "eval_runtime": 43.9738, + "eval_samples_per_second": 22.741, + "eval_steps_per_second": 0.955, + "step": 1820 + }, + { + "epoch": 1.9956164383561643, + "grad_norm": 1.90078866481781, + "learning_rate": 1.8907392363931764e-06, + "loss": 2.1274, + "step": 1821 + }, + { + "epoch": 1.9967123287671233, + "grad_norm": 1.8798222541809082, + "learning_rate": 1.8887083671811536e-06, + "loss": 2.1014, + "step": 1822 + }, + { + "epoch": 1.9978082191780822, + "grad_norm": 1.9333341121673584, + "learning_rate": 1.886677497969131e-06, + "loss": 2.1165, + "step": 1823 + }, + { + "epoch": 1.998904109589041, + "grad_norm": 1.9861575365066528, + "learning_rate": 1.8846466287571082e-06, + "loss": 2.036, + "step": 1824 + }, + { + "epoch": 2.0, + "grad_norm": 2.0310938358306885, + "learning_rate": 1.8826157595450855e-06, + "loss": 2.1324, + "step": 1825 + }, + { + "epoch": 2.001095890410959, + "grad_norm": 1.9574612379074097, + "learning_rate": 1.8805848903330625e-06, + "loss": 2.0485, + "step": 1826 + }, + { + "epoch": 2.0021917808219176, + "grad_norm": 2.0283970832824707, + "learning_rate": 1.8785540211210398e-06, + "loss": 2.1768, + "step": 1827 + }, + { + "epoch": 2.0032876712328767, + "grad_norm": 2.062384843826294, + "learning_rate": 1.876523151909017e-06, + "loss": 2.1199, + "step": 1828 + }, + { + "epoch": 2.0043835616438357, + "grad_norm": 2.0543339252471924, + "learning_rate": 1.8744922826969944e-06, + "loss": 2.0887, + "step": 1829 + }, + { + "epoch": 2.0054794520547947, + "grad_norm": 2.1769917011260986, + "learning_rate": 1.8724614134849716e-06, + "loss": 2.0821, + "step": 1830 + }, + { + "epoch": 2.0054794520547947, + "eval_loss": 2.2689950466156006, + "eval_runtime": 43.9448, + "eval_samples_per_second": 22.756, + "eval_steps_per_second": 0.956, + "step": 1830 + }, + { + "epoch": 2.0065753424657533, + "grad_norm": 2.0331335067749023, + "learning_rate": 1.870430544272949e-06, + "loss": 2.0916, + "step": 1831 + }, + { + "epoch": 2.0076712328767123, + "grad_norm": 2.1848011016845703, + "learning_rate": 1.8683996750609262e-06, + "loss": 2.1525, + "step": 1832 + }, + { + "epoch": 2.0087671232876714, + "grad_norm": 2.086441993713379, + "learning_rate": 1.8663688058489035e-06, + "loss": 2.0733, + "step": 1833 + }, + { + "epoch": 2.00986301369863, + "grad_norm": 2.1663239002227783, + "learning_rate": 1.8643379366368807e-06, + "loss": 2.1026, + "step": 1834 + }, + { + "epoch": 2.010958904109589, + "grad_norm": 2.1658689975738525, + "learning_rate": 1.862307067424858e-06, + "loss": 2.1178, + "step": 1835 + }, + { + "epoch": 2.012054794520548, + "grad_norm": 2.043203592300415, + "learning_rate": 1.8602761982128353e-06, + "loss": 2.1025, + "step": 1836 + }, + { + "epoch": 2.013150684931507, + "grad_norm": 2.1104307174682617, + "learning_rate": 1.8582453290008126e-06, + "loss": 2.1218, + "step": 1837 + }, + { + "epoch": 2.0142465753424657, + "grad_norm": 2.078554630279541, + "learning_rate": 1.8562144597887898e-06, + "loss": 2.082, + "step": 1838 + }, + { + "epoch": 2.0153424657534247, + "grad_norm": 2.1440582275390625, + "learning_rate": 1.854183590576767e-06, + "loss": 2.101, + "step": 1839 + }, + { + "epoch": 2.0164383561643837, + "grad_norm": 2.0824856758117676, + "learning_rate": 1.8521527213647442e-06, + "loss": 2.1377, + "step": 1840 + }, + { + "epoch": 2.0164383561643837, + "eval_loss": 2.268453359603882, + "eval_runtime": 43.8166, + "eval_samples_per_second": 22.822, + "eval_steps_per_second": 0.959, + "step": 1840 + }, + { + "epoch": 2.0175342465753423, + "grad_norm": 2.1607978343963623, + "learning_rate": 1.8501218521527215e-06, + "loss": 2.0734, + "step": 1841 + }, + { + "epoch": 2.0186301369863013, + "grad_norm": 2.044123649597168, + "learning_rate": 1.8480909829406987e-06, + "loss": 2.04, + "step": 1842 + }, + { + "epoch": 2.0197260273972604, + "grad_norm": 2.1130807399749756, + "learning_rate": 1.846060113728676e-06, + "loss": 2.069, + "step": 1843 + }, + { + "epoch": 2.020821917808219, + "grad_norm": 2.0879838466644287, + "learning_rate": 1.8440292445166533e-06, + "loss": 2.1006, + "step": 1844 + }, + { + "epoch": 2.021917808219178, + "grad_norm": 2.311849355697632, + "learning_rate": 1.8419983753046306e-06, + "loss": 2.0691, + "step": 1845 + }, + { + "epoch": 2.023013698630137, + "grad_norm": 2.122504711151123, + "learning_rate": 1.8399675060926078e-06, + "loss": 2.1297, + "step": 1846 + }, + { + "epoch": 2.024109589041096, + "grad_norm": 2.135181427001953, + "learning_rate": 1.8379366368805851e-06, + "loss": 2.0951, + "step": 1847 + }, + { + "epoch": 2.0252054794520546, + "grad_norm": 2.0710697174072266, + "learning_rate": 1.8359057676685624e-06, + "loss": 2.0754, + "step": 1848 + }, + { + "epoch": 2.0263013698630137, + "grad_norm": 2.0922486782073975, + "learning_rate": 1.8338748984565397e-06, + "loss": 2.068, + "step": 1849 + }, + { + "epoch": 2.0273972602739727, + "grad_norm": 2.0101380348205566, + "learning_rate": 1.831844029244517e-06, + "loss": 2.0831, + "step": 1850 + }, + { + "epoch": 2.0273972602739727, + "eval_loss": 2.267380952835083, + "eval_runtime": 44.4219, + "eval_samples_per_second": 22.511, + "eval_steps_per_second": 0.945, + "step": 1850 + }, + { + "epoch": 2.0284931506849313, + "grad_norm": 2.012316942214966, + "learning_rate": 1.829813160032494e-06, + "loss": 2.0571, + "step": 1851 + }, + { + "epoch": 2.0295890410958903, + "grad_norm": 2.039604663848877, + "learning_rate": 1.8277822908204713e-06, + "loss": 2.0691, + "step": 1852 + }, + { + "epoch": 2.0306849315068494, + "grad_norm": 2.242504119873047, + "learning_rate": 1.8257514216084486e-06, + "loss": 2.1025, + "step": 1853 + }, + { + "epoch": 2.0317808219178084, + "grad_norm": 2.0629303455352783, + "learning_rate": 1.8237205523964258e-06, + "loss": 2.0803, + "step": 1854 + }, + { + "epoch": 2.032876712328767, + "grad_norm": 2.0935115814208984, + "learning_rate": 1.8216896831844031e-06, + "loss": 2.101, + "step": 1855 + }, + { + "epoch": 2.033972602739726, + "grad_norm": 2.164381504058838, + "learning_rate": 1.8196588139723804e-06, + "loss": 2.1227, + "step": 1856 + }, + { + "epoch": 2.035068493150685, + "grad_norm": 2.240391492843628, + "learning_rate": 1.8176279447603577e-06, + "loss": 2.0549, + "step": 1857 + }, + { + "epoch": 2.0361643835616436, + "grad_norm": 2.1460423469543457, + "learning_rate": 1.815597075548335e-06, + "loss": 2.0858, + "step": 1858 + }, + { + "epoch": 2.0372602739726027, + "grad_norm": 2.0199930667877197, + "learning_rate": 1.8135662063363122e-06, + "loss": 2.1058, + "step": 1859 + }, + { + "epoch": 2.0383561643835617, + "grad_norm": 2.074615955352783, + "learning_rate": 1.8115353371242895e-06, + "loss": 2.113, + "step": 1860 + }, + { + "epoch": 2.0383561643835617, + "eval_loss": 2.2699100971221924, + "eval_runtime": 42.9781, + "eval_samples_per_second": 23.268, + "eval_steps_per_second": 0.977, + "step": 1860 + }, + { + "epoch": 2.0394520547945207, + "grad_norm": 2.091417074203491, + "learning_rate": 1.8095044679122668e-06, + "loss": 2.045, + "step": 1861 + }, + { + "epoch": 2.0405479452054793, + "grad_norm": 2.2032699584960938, + "learning_rate": 1.807473598700244e-06, + "loss": 2.081, + "step": 1862 + }, + { + "epoch": 2.0416438356164384, + "grad_norm": 2.0829155445098877, + "learning_rate": 1.805442729488221e-06, + "loss": 2.0798, + "step": 1863 + }, + { + "epoch": 2.0427397260273974, + "grad_norm": 1.97968590259552, + "learning_rate": 1.8034118602761982e-06, + "loss": 2.0851, + "step": 1864 + }, + { + "epoch": 2.043835616438356, + "grad_norm": 1.9853475093841553, + "learning_rate": 1.8013809910641754e-06, + "loss": 2.0826, + "step": 1865 + }, + { + "epoch": 2.044931506849315, + "grad_norm": 2.0638294219970703, + "learning_rate": 1.7993501218521527e-06, + "loss": 2.0806, + "step": 1866 + }, + { + "epoch": 2.046027397260274, + "grad_norm": 2.0647425651550293, + "learning_rate": 1.79731925264013e-06, + "loss": 2.0613, + "step": 1867 + }, + { + "epoch": 2.047123287671233, + "grad_norm": 2.1473264694213867, + "learning_rate": 1.7952883834281073e-06, + "loss": 2.1211, + "step": 1868 + }, + { + "epoch": 2.0482191780821917, + "grad_norm": 2.037551164627075, + "learning_rate": 1.7932575142160846e-06, + "loss": 2.0727, + "step": 1869 + }, + { + "epoch": 2.0493150684931507, + "grad_norm": 2.010626792907715, + "learning_rate": 1.7912266450040618e-06, + "loss": 2.1013, + "step": 1870 + }, + { + "epoch": 2.0493150684931507, + "eval_loss": 2.266191005706787, + "eval_runtime": 45.1806, + "eval_samples_per_second": 22.133, + "eval_steps_per_second": 0.93, + "step": 1870 + }, + { + "epoch": 2.0504109589041097, + "grad_norm": 1.955633282661438, + "learning_rate": 1.789195775792039e-06, + "loss": 2.114, + "step": 1871 + }, + { + "epoch": 2.0515068493150683, + "grad_norm": 2.1419057846069336, + "learning_rate": 1.7871649065800164e-06, + "loss": 2.0296, + "step": 1872 + }, + { + "epoch": 2.0526027397260274, + "grad_norm": 2.167240858078003, + "learning_rate": 1.7851340373679937e-06, + "loss": 2.1391, + "step": 1873 + }, + { + "epoch": 2.0536986301369864, + "grad_norm": 2.0724527835845947, + "learning_rate": 1.783103168155971e-06, + "loss": 2.0571, + "step": 1874 + }, + { + "epoch": 2.0547945205479454, + "grad_norm": 2.104583263397217, + "learning_rate": 1.7810722989439482e-06, + "loss": 2.0917, + "step": 1875 + }, + { + "epoch": 2.055890410958904, + "grad_norm": 2.083120584487915, + "learning_rate": 1.7790414297319253e-06, + "loss": 2.093, + "step": 1876 + }, + { + "epoch": 2.056986301369863, + "grad_norm": 2.136396646499634, + "learning_rate": 1.7770105605199025e-06, + "loss": 2.0654, + "step": 1877 + }, + { + "epoch": 2.058082191780822, + "grad_norm": 2.1515700817108154, + "learning_rate": 1.7749796913078798e-06, + "loss": 2.05, + "step": 1878 + }, + { + "epoch": 2.0591780821917807, + "grad_norm": 2.1882944107055664, + "learning_rate": 1.772948822095857e-06, + "loss": 2.0974, + "step": 1879 + }, + { + "epoch": 2.0602739726027397, + "grad_norm": 2.261822462081909, + "learning_rate": 1.7709179528838344e-06, + "loss": 2.1375, + "step": 1880 + }, + { + "epoch": 2.0602739726027397, + "eval_loss": 2.267698287963867, + "eval_runtime": 45.8285, + "eval_samples_per_second": 21.82, + "eval_steps_per_second": 0.916, + "step": 1880 + }, + { + "epoch": 2.0613698630136987, + "grad_norm": 2.121238946914673, + "learning_rate": 1.7688870836718116e-06, + "loss": 2.1102, + "step": 1881 + }, + { + "epoch": 2.0624657534246573, + "grad_norm": 2.183793544769287, + "learning_rate": 1.766856214459789e-06, + "loss": 2.0862, + "step": 1882 + }, + { + "epoch": 2.0635616438356164, + "grad_norm": 2.209182024002075, + "learning_rate": 1.7648253452477662e-06, + "loss": 2.088, + "step": 1883 + }, + { + "epoch": 2.0646575342465754, + "grad_norm": 1.9593517780303955, + "learning_rate": 1.7627944760357435e-06, + "loss": 2.0644, + "step": 1884 + }, + { + "epoch": 2.0657534246575344, + "grad_norm": 1.9904463291168213, + "learning_rate": 1.7607636068237208e-06, + "loss": 2.1313, + "step": 1885 + }, + { + "epoch": 2.066849315068493, + "grad_norm": 2.2765681743621826, + "learning_rate": 1.758732737611698e-06, + "loss": 2.0979, + "step": 1886 + }, + { + "epoch": 2.067945205479452, + "grad_norm": 2.066204309463501, + "learning_rate": 1.7567018683996753e-06, + "loss": 2.0934, + "step": 1887 + }, + { + "epoch": 2.069041095890411, + "grad_norm": 2.104768753051758, + "learning_rate": 1.7546709991876524e-06, + "loss": 2.0887, + "step": 1888 + }, + { + "epoch": 2.0701369863013697, + "grad_norm": 2.0848052501678467, + "learning_rate": 1.7526401299756296e-06, + "loss": 2.0635, + "step": 1889 + }, + { + "epoch": 2.0712328767123287, + "grad_norm": 2.052314519882202, + "learning_rate": 1.750609260763607e-06, + "loss": 2.0728, + "step": 1890 + }, + { + "epoch": 2.0712328767123287, + "eval_loss": 2.266915798187256, + "eval_runtime": 46.3568, + "eval_samples_per_second": 21.572, + "eval_steps_per_second": 0.906, + "step": 1890 + }, + { + "epoch": 2.0723287671232877, + "grad_norm": 2.209855318069458, + "learning_rate": 1.7485783915515842e-06, + "loss": 2.0951, + "step": 1891 + }, + { + "epoch": 2.0734246575342468, + "grad_norm": 2.1558358669281006, + "learning_rate": 1.7465475223395615e-06, + "loss": 2.0872, + "step": 1892 + }, + { + "epoch": 2.0745205479452054, + "grad_norm": 2.426797389984131, + "learning_rate": 1.7445166531275387e-06, + "loss": 2.0578, + "step": 1893 + }, + { + "epoch": 2.0756164383561644, + "grad_norm": 2.1906144618988037, + "learning_rate": 1.742485783915516e-06, + "loss": 2.1664, + "step": 1894 + }, + { + "epoch": 2.0767123287671234, + "grad_norm": 2.2041025161743164, + "learning_rate": 1.7404549147034933e-06, + "loss": 2.1532, + "step": 1895 + }, + { + "epoch": 2.077808219178082, + "grad_norm": 2.179513931274414, + "learning_rate": 1.7384240454914706e-06, + "loss": 2.0349, + "step": 1896 + }, + { + "epoch": 2.078904109589041, + "grad_norm": 2.239232301712036, + "learning_rate": 1.7363931762794478e-06, + "loss": 2.0926, + "step": 1897 + }, + { + "epoch": 2.08, + "grad_norm": 2.2452921867370605, + "learning_rate": 1.7343623070674251e-06, + "loss": 2.1082, + "step": 1898 + }, + { + "epoch": 2.081095890410959, + "grad_norm": 2.145571708679199, + "learning_rate": 1.7323314378554024e-06, + "loss": 2.0886, + "step": 1899 + }, + { + "epoch": 2.0821917808219177, + "grad_norm": 2.201815366744995, + "learning_rate": 1.7303005686433797e-06, + "loss": 2.0947, + "step": 1900 + }, + { + "epoch": 2.0821917808219177, + "eval_loss": 2.2664403915405273, + "eval_runtime": 46.4641, + "eval_samples_per_second": 21.522, + "eval_steps_per_second": 0.904, + "step": 1900 + }, + { + "epoch": 2.0832876712328767, + "grad_norm": 2.095641851425171, + "learning_rate": 1.7282696994313567e-06, + "loss": 2.0925, + "step": 1901 + }, + { + "epoch": 2.0843835616438358, + "grad_norm": 2.186495304107666, + "learning_rate": 1.726238830219334e-06, + "loss": 2.0971, + "step": 1902 + }, + { + "epoch": 2.0854794520547943, + "grad_norm": 2.135972023010254, + "learning_rate": 1.7242079610073113e-06, + "loss": 2.076, + "step": 1903 + }, + { + "epoch": 2.0865753424657534, + "grad_norm": 2.1793875694274902, + "learning_rate": 1.7221770917952886e-06, + "loss": 2.1336, + "step": 1904 + }, + { + "epoch": 2.0876712328767124, + "grad_norm": 2.178579330444336, + "learning_rate": 1.7201462225832658e-06, + "loss": 2.0959, + "step": 1905 + }, + { + "epoch": 2.0887671232876714, + "grad_norm": 2.1725568771362305, + "learning_rate": 1.7181153533712431e-06, + "loss": 2.0965, + "step": 1906 + }, + { + "epoch": 2.08986301369863, + "grad_norm": 2.126034736633301, + "learning_rate": 1.7160844841592204e-06, + "loss": 2.0225, + "step": 1907 + }, + { + "epoch": 2.090958904109589, + "grad_norm": 2.0938918590545654, + "learning_rate": 1.7140536149471977e-06, + "loss": 2.0447, + "step": 1908 + }, + { + "epoch": 2.092054794520548, + "grad_norm": 2.1260006427764893, + "learning_rate": 1.712022745735175e-06, + "loss": 2.0739, + "step": 1909 + }, + { + "epoch": 2.0931506849315067, + "grad_norm": 2.2151126861572266, + "learning_rate": 1.7099918765231522e-06, + "loss": 2.0786, + "step": 1910 + }, + { + "epoch": 2.0931506849315067, + "eval_loss": 2.2675349712371826, + "eval_runtime": 43.2795, + "eval_samples_per_second": 23.106, + "eval_steps_per_second": 0.97, + "step": 1910 + }, + { + "epoch": 2.0942465753424657, + "grad_norm": 2.146458625793457, + "learning_rate": 1.7079610073111295e-06, + "loss": 2.0928, + "step": 1911 + }, + { + "epoch": 2.0953424657534248, + "grad_norm": 2.1360816955566406, + "learning_rate": 1.7059301380991068e-06, + "loss": 2.0404, + "step": 1912 + }, + { + "epoch": 2.0964383561643833, + "grad_norm": 2.130659341812134, + "learning_rate": 1.7038992688870836e-06, + "loss": 2.0858, + "step": 1913 + }, + { + "epoch": 2.0975342465753424, + "grad_norm": 2.0875484943389893, + "learning_rate": 1.701868399675061e-06, + "loss": 2.0613, + "step": 1914 + }, + { + "epoch": 2.0986301369863014, + "grad_norm": 2.1744940280914307, + "learning_rate": 1.6998375304630382e-06, + "loss": 2.1012, + "step": 1915 + }, + { + "epoch": 2.0997260273972604, + "grad_norm": 2.195559024810791, + "learning_rate": 1.6978066612510155e-06, + "loss": 2.1595, + "step": 1916 + }, + { + "epoch": 2.100821917808219, + "grad_norm": 2.3818888664245605, + "learning_rate": 1.6957757920389927e-06, + "loss": 2.1246, + "step": 1917 + }, + { + "epoch": 2.101917808219178, + "grad_norm": 2.296870708465576, + "learning_rate": 1.69374492282697e-06, + "loss": 2.0893, + "step": 1918 + }, + { + "epoch": 2.103013698630137, + "grad_norm": 2.3288252353668213, + "learning_rate": 1.6917140536149473e-06, + "loss": 2.0764, + "step": 1919 + }, + { + "epoch": 2.1041095890410957, + "grad_norm": 2.1017708778381348, + "learning_rate": 1.6896831844029246e-06, + "loss": 2.1312, + "step": 1920 + }, + { + "epoch": 2.1041095890410957, + "eval_loss": 2.2681941986083984, + "eval_runtime": 51.4964, + "eval_samples_per_second": 19.419, + "eval_steps_per_second": 0.816, + "step": 1920 + }, + { + "epoch": 2.1052054794520547, + "grad_norm": 2.2251839637756348, + "learning_rate": 1.6876523151909018e-06, + "loss": 2.0967, + "step": 1921 + }, + { + "epoch": 2.1063013698630138, + "grad_norm": 2.1989998817443848, + "learning_rate": 1.6856214459788791e-06, + "loss": 2.1064, + "step": 1922 + }, + { + "epoch": 2.107397260273973, + "grad_norm": 2.2629430294036865, + "learning_rate": 1.6835905767668564e-06, + "loss": 2.0574, + "step": 1923 + }, + { + "epoch": 2.1084931506849314, + "grad_norm": 2.2153608798980713, + "learning_rate": 1.6815597075548337e-06, + "loss": 2.0736, + "step": 1924 + }, + { + "epoch": 2.1095890410958904, + "grad_norm": 2.1380763053894043, + "learning_rate": 1.679528838342811e-06, + "loss": 2.1437, + "step": 1925 + }, + { + "epoch": 2.1106849315068494, + "grad_norm": 2.198996067047119, + "learning_rate": 1.677497969130788e-06, + "loss": 2.0623, + "step": 1926 + }, + { + "epoch": 2.111780821917808, + "grad_norm": 2.1622812747955322, + "learning_rate": 1.6754670999187653e-06, + "loss": 2.072, + "step": 1927 + }, + { + "epoch": 2.112876712328767, + "grad_norm": 2.185896873474121, + "learning_rate": 1.6734362307067426e-06, + "loss": 2.0931, + "step": 1928 + }, + { + "epoch": 2.113972602739726, + "grad_norm": 2.118260383605957, + "learning_rate": 1.6714053614947198e-06, + "loss": 2.0555, + "step": 1929 + }, + { + "epoch": 2.115068493150685, + "grad_norm": 2.070236921310425, + "learning_rate": 1.669374492282697e-06, + "loss": 2.0503, + "step": 1930 + }, + { + "epoch": 2.115068493150685, + "eval_loss": 2.267728328704834, + "eval_runtime": 48.9617, + "eval_samples_per_second": 20.424, + "eval_steps_per_second": 0.858, + "step": 1930 + }, + { + "epoch": 2.1161643835616437, + "grad_norm": 2.2161452770233154, + "learning_rate": 1.6673436230706744e-06, + "loss": 2.1211, + "step": 1931 + }, + { + "epoch": 2.1172602739726027, + "grad_norm": 2.0978891849517822, + "learning_rate": 1.6653127538586517e-06, + "loss": 2.1055, + "step": 1932 + }, + { + "epoch": 2.118356164383562, + "grad_norm": 2.208021879196167, + "learning_rate": 1.663281884646629e-06, + "loss": 2.0754, + "step": 1933 + }, + { + "epoch": 2.1194520547945204, + "grad_norm": 2.0411078929901123, + "learning_rate": 1.6612510154346062e-06, + "loss": 2.0687, + "step": 1934 + }, + { + "epoch": 2.1205479452054794, + "grad_norm": 2.275355339050293, + "learning_rate": 1.6592201462225835e-06, + "loss": 2.1354, + "step": 1935 + }, + { + "epoch": 2.1216438356164384, + "grad_norm": 2.102735996246338, + "learning_rate": 1.6571892770105608e-06, + "loss": 2.105, + "step": 1936 + }, + { + "epoch": 2.1227397260273975, + "grad_norm": 2.1679234504699707, + "learning_rate": 1.655158407798538e-06, + "loss": 2.1314, + "step": 1937 + }, + { + "epoch": 2.123835616438356, + "grad_norm": 2.244574785232544, + "learning_rate": 1.653127538586515e-06, + "loss": 2.0923, + "step": 1938 + }, + { + "epoch": 2.124931506849315, + "grad_norm": 2.0957608222961426, + "learning_rate": 1.6510966693744924e-06, + "loss": 2.1115, + "step": 1939 + }, + { + "epoch": 2.126027397260274, + "grad_norm": 2.0705184936523438, + "learning_rate": 1.6490658001624696e-06, + "loss": 2.1125, + "step": 1940 + }, + { + "epoch": 2.126027397260274, + "eval_loss": 2.2668750286102295, + "eval_runtime": 48.4476, + "eval_samples_per_second": 20.641, + "eval_steps_per_second": 0.867, + "step": 1940 + }, + { + "epoch": 2.1271232876712327, + "grad_norm": 2.185532569885254, + "learning_rate": 1.647034930950447e-06, + "loss": 2.1148, + "step": 1941 + }, + { + "epoch": 2.1282191780821917, + "grad_norm": 2.1876089572906494, + "learning_rate": 1.6450040617384242e-06, + "loss": 2.0646, + "step": 1942 + }, + { + "epoch": 2.1293150684931508, + "grad_norm": 2.065310478210449, + "learning_rate": 1.6429731925264015e-06, + "loss": 2.0843, + "step": 1943 + }, + { + "epoch": 2.1304109589041094, + "grad_norm": 2.1671154499053955, + "learning_rate": 1.6409423233143788e-06, + "loss": 2.112, + "step": 1944 + }, + { + "epoch": 2.1315068493150684, + "grad_norm": 2.1573469638824463, + "learning_rate": 1.638911454102356e-06, + "loss": 2.0781, + "step": 1945 + }, + { + "epoch": 2.1326027397260274, + "grad_norm": 2.0912411212921143, + "learning_rate": 1.6368805848903333e-06, + "loss": 2.0621, + "step": 1946 + }, + { + "epoch": 2.1336986301369865, + "grad_norm": 2.1280713081359863, + "learning_rate": 1.6348497156783106e-06, + "loss": 2.1174, + "step": 1947 + }, + { + "epoch": 2.134794520547945, + "grad_norm": 2.1215052604675293, + "learning_rate": 1.6328188464662879e-06, + "loss": 2.0725, + "step": 1948 + }, + { + "epoch": 2.135890410958904, + "grad_norm": 2.0699117183685303, + "learning_rate": 1.6307879772542651e-06, + "loss": 2.0783, + "step": 1949 + }, + { + "epoch": 2.136986301369863, + "grad_norm": 2.0162057876586914, + "learning_rate": 1.628757108042242e-06, + "loss": 2.0868, + "step": 1950 + }, + { + "epoch": 2.136986301369863, + "eval_loss": 2.2651138305664062, + "eval_runtime": 53.3744, + "eval_samples_per_second": 18.736, + "eval_steps_per_second": 0.787, + "step": 1950 + }, + { + "epoch": 2.138082191780822, + "grad_norm": 2.1494126319885254, + "learning_rate": 1.6267262388302193e-06, + "loss": 2.1147, + "step": 1951 + }, + { + "epoch": 2.1391780821917807, + "grad_norm": 2.1003870964050293, + "learning_rate": 1.6246953696181965e-06, + "loss": 2.082, + "step": 1952 + }, + { + "epoch": 2.1402739726027398, + "grad_norm": 2.1687192916870117, + "learning_rate": 1.6226645004061738e-06, + "loss": 2.1084, + "step": 1953 + }, + { + "epoch": 2.141369863013699, + "grad_norm": 2.010176181793213, + "learning_rate": 1.620633631194151e-06, + "loss": 2.1568, + "step": 1954 + }, + { + "epoch": 2.1424657534246574, + "grad_norm": 2.13127064704895, + "learning_rate": 1.6186027619821284e-06, + "loss": 2.116, + "step": 1955 + }, + { + "epoch": 2.1435616438356164, + "grad_norm": 2.2229974269866943, + "learning_rate": 1.6165718927701056e-06, + "loss": 2.0596, + "step": 1956 + }, + { + "epoch": 2.1446575342465755, + "grad_norm": 2.155855894088745, + "learning_rate": 1.614541023558083e-06, + "loss": 2.1052, + "step": 1957 + }, + { + "epoch": 2.145753424657534, + "grad_norm": 2.1598658561706543, + "learning_rate": 1.6125101543460602e-06, + "loss": 2.097, + "step": 1958 + }, + { + "epoch": 2.146849315068493, + "grad_norm": 2.3428308963775635, + "learning_rate": 1.6104792851340375e-06, + "loss": 2.0991, + "step": 1959 + }, + { + "epoch": 2.147945205479452, + "grad_norm": 2.293128728866577, + "learning_rate": 1.6084484159220147e-06, + "loss": 2.1146, + "step": 1960 + }, + { + "epoch": 2.147945205479452, + "eval_loss": 2.265352964401245, + "eval_runtime": 49.6478, + "eval_samples_per_second": 20.142, + "eval_steps_per_second": 0.846, + "step": 1960 + }, + { + "epoch": 2.149041095890411, + "grad_norm": 2.1581852436065674, + "learning_rate": 1.606417546709992e-06, + "loss": 2.0901, + "step": 1961 + }, + { + "epoch": 2.1501369863013697, + "grad_norm": 2.177950382232666, + "learning_rate": 1.6043866774979693e-06, + "loss": 2.0765, + "step": 1962 + }, + { + "epoch": 2.1512328767123288, + "grad_norm": 2.176633596420288, + "learning_rate": 1.6023558082859464e-06, + "loss": 2.0917, + "step": 1963 + }, + { + "epoch": 2.152328767123288, + "grad_norm": 2.132002592086792, + "learning_rate": 1.6003249390739236e-06, + "loss": 2.0712, + "step": 1964 + }, + { + "epoch": 2.1534246575342464, + "grad_norm": 2.1170928478240967, + "learning_rate": 1.598294069861901e-06, + "loss": 2.0835, + "step": 1965 + }, + { + "epoch": 2.1545205479452054, + "grad_norm": 2.2532341480255127, + "learning_rate": 1.5962632006498782e-06, + "loss": 2.0638, + "step": 1966 + }, + { + "epoch": 2.1556164383561645, + "grad_norm": 2.1742208003997803, + "learning_rate": 1.5942323314378555e-06, + "loss": 2.1164, + "step": 1967 + }, + { + "epoch": 2.1567123287671235, + "grad_norm": 2.1353843212127686, + "learning_rate": 1.5922014622258327e-06, + "loss": 2.0447, + "step": 1968 + }, + { + "epoch": 2.157808219178082, + "grad_norm": 2.1103644371032715, + "learning_rate": 1.59017059301381e-06, + "loss": 2.0631, + "step": 1969 + }, + { + "epoch": 2.158904109589041, + "grad_norm": 2.129606246948242, + "learning_rate": 1.5881397238017873e-06, + "loss": 2.1184, + "step": 1970 + }, + { + "epoch": 2.158904109589041, + "eval_loss": 2.26619553565979, + "eval_runtime": 52.3189, + "eval_samples_per_second": 19.114, + "eval_steps_per_second": 0.803, + "step": 1970 + }, + { + "epoch": 2.16, + "grad_norm": 2.2155823707580566, + "learning_rate": 1.5861088545897646e-06, + "loss": 2.0885, + "step": 1971 + }, + { + "epoch": 2.1610958904109587, + "grad_norm": 2.1816463470458984, + "learning_rate": 1.5840779853777418e-06, + "loss": 2.0517, + "step": 1972 + }, + { + "epoch": 2.1621917808219178, + "grad_norm": 2.2179040908813477, + "learning_rate": 1.5820471161657191e-06, + "loss": 2.0884, + "step": 1973 + }, + { + "epoch": 2.163287671232877, + "grad_norm": 2.3558337688446045, + "learning_rate": 1.5800162469536964e-06, + "loss": 2.0682, + "step": 1974 + }, + { + "epoch": 2.1643835616438354, + "grad_norm": 2.406665086746216, + "learning_rate": 1.5779853777416735e-06, + "loss": 2.1172, + "step": 1975 + }, + { + "epoch": 2.1654794520547944, + "grad_norm": 2.3815410137176514, + "learning_rate": 1.5759545085296507e-06, + "loss": 2.115, + "step": 1976 + }, + { + "epoch": 2.1665753424657535, + "grad_norm": 2.2329678535461426, + "learning_rate": 1.573923639317628e-06, + "loss": 2.087, + "step": 1977 + }, + { + "epoch": 2.1676712328767125, + "grad_norm": 2.331451177597046, + "learning_rate": 1.5718927701056053e-06, + "loss": 2.0996, + "step": 1978 + }, + { + "epoch": 2.168767123287671, + "grad_norm": 2.158210039138794, + "learning_rate": 1.5698619008935826e-06, + "loss": 2.0905, + "step": 1979 + }, + { + "epoch": 2.16986301369863, + "grad_norm": 2.1256191730499268, + "learning_rate": 1.5678310316815598e-06, + "loss": 2.1099, + "step": 1980 + }, + { + "epoch": 2.16986301369863, + "eval_loss": 2.2658803462982178, + "eval_runtime": 51.5278, + "eval_samples_per_second": 19.407, + "eval_steps_per_second": 0.815, + "step": 1980 + }, + { + "epoch": 2.170958904109589, + "grad_norm": 2.2626984119415283, + "learning_rate": 1.5658001624695371e-06, + "loss": 2.0651, + "step": 1981 + }, + { + "epoch": 2.172054794520548, + "grad_norm": 2.132742404937744, + "learning_rate": 1.5637692932575144e-06, + "loss": 2.1246, + "step": 1982 + }, + { + "epoch": 2.1731506849315068, + "grad_norm": 2.0621824264526367, + "learning_rate": 1.5617384240454917e-06, + "loss": 2.0696, + "step": 1983 + }, + { + "epoch": 2.174246575342466, + "grad_norm": 2.0715548992156982, + "learning_rate": 1.559707554833469e-06, + "loss": 2.0894, + "step": 1984 + }, + { + "epoch": 2.175342465753425, + "grad_norm": 2.1211655139923096, + "learning_rate": 1.5576766856214462e-06, + "loss": 2.1122, + "step": 1985 + }, + { + "epoch": 2.1764383561643834, + "grad_norm": 2.14603328704834, + "learning_rate": 1.5556458164094235e-06, + "loss": 2.0239, + "step": 1986 + }, + { + "epoch": 2.1775342465753424, + "grad_norm": 2.092494249343872, + "learning_rate": 1.5536149471974008e-06, + "loss": 2.0795, + "step": 1987 + }, + { + "epoch": 2.1786301369863015, + "grad_norm": 2.14874267578125, + "learning_rate": 1.5515840779853778e-06, + "loss": 2.0476, + "step": 1988 + }, + { + "epoch": 2.17972602739726, + "grad_norm": 2.024716854095459, + "learning_rate": 1.549553208773355e-06, + "loss": 2.0904, + "step": 1989 + }, + { + "epoch": 2.180821917808219, + "grad_norm": 2.1700267791748047, + "learning_rate": 1.5475223395613324e-06, + "loss": 2.0806, + "step": 1990 + }, + { + "epoch": 2.180821917808219, + "eval_loss": 2.2657630443573, + "eval_runtime": 49.7954, + "eval_samples_per_second": 20.082, + "eval_steps_per_second": 0.843, + "step": 1990 + }, + { + "epoch": 2.181917808219178, + "grad_norm": 2.1046273708343506, + "learning_rate": 1.5454914703493097e-06, + "loss": 2.0738, + "step": 1991 + }, + { + "epoch": 2.183013698630137, + "grad_norm": 2.2489562034606934, + "learning_rate": 1.543460601137287e-06, + "loss": 2.1468, + "step": 1992 + }, + { + "epoch": 2.1841095890410958, + "grad_norm": 2.176658868789673, + "learning_rate": 1.5414297319252642e-06, + "loss": 2.0977, + "step": 1993 + }, + { + "epoch": 2.185205479452055, + "grad_norm": 2.2471182346343994, + "learning_rate": 1.5393988627132415e-06, + "loss": 2.1023, + "step": 1994 + }, + { + "epoch": 2.186301369863014, + "grad_norm": 2.306699514389038, + "learning_rate": 1.5373679935012188e-06, + "loss": 2.084, + "step": 1995 + }, + { + "epoch": 2.1873972602739724, + "grad_norm": 2.2053849697113037, + "learning_rate": 1.535337124289196e-06, + "loss": 2.0419, + "step": 1996 + }, + { + "epoch": 2.1884931506849314, + "grad_norm": 2.2459211349487305, + "learning_rate": 1.5333062550771733e-06, + "loss": 2.0882, + "step": 1997 + }, + { + "epoch": 2.1895890410958905, + "grad_norm": 2.574877977371216, + "learning_rate": 1.5312753858651506e-06, + "loss": 2.1627, + "step": 1998 + }, + { + "epoch": 2.1906849315068495, + "grad_norm": 2.111438274383545, + "learning_rate": 1.5292445166531279e-06, + "loss": 2.0762, + "step": 1999 + }, + { + "epoch": 2.191780821917808, + "grad_norm": 2.1831490993499756, + "learning_rate": 1.5272136474411047e-06, + "loss": 2.0954, + "step": 2000 + }, + { + "epoch": 2.191780821917808, + "eval_loss": 2.2667267322540283, + "eval_runtime": 46.6703, + "eval_samples_per_second": 21.427, + "eval_steps_per_second": 0.9, + "step": 2000 + }, + { + "epoch": 2.192876712328767, + "grad_norm": 2.161336660385132, + "learning_rate": 1.525182778229082e-06, + "loss": 2.1393, + "step": 2001 + }, + { + "epoch": 2.193972602739726, + "grad_norm": 2.210535764694214, + "learning_rate": 1.5231519090170593e-06, + "loss": 2.0925, + "step": 2002 + }, + { + "epoch": 2.1950684931506848, + "grad_norm": 2.123384714126587, + "learning_rate": 1.5211210398050365e-06, + "loss": 2.0623, + "step": 2003 + }, + { + "epoch": 2.196164383561644, + "grad_norm": 2.147090196609497, + "learning_rate": 1.5190901705930138e-06, + "loss": 2.0839, + "step": 2004 + }, + { + "epoch": 2.197260273972603, + "grad_norm": 2.261474609375, + "learning_rate": 1.517059301380991e-06, + "loss": 2.0987, + "step": 2005 + }, + { + "epoch": 2.198356164383562, + "grad_norm": 2.295656204223633, + "learning_rate": 1.5150284321689684e-06, + "loss": 2.0786, + "step": 2006 + }, + { + "epoch": 2.1994520547945204, + "grad_norm": 2.055387020111084, + "learning_rate": 1.5129975629569456e-06, + "loss": 2.0781, + "step": 2007 + }, + { + "epoch": 2.2005479452054795, + "grad_norm": 2.1994800567626953, + "learning_rate": 1.510966693744923e-06, + "loss": 2.088, + "step": 2008 + }, + { + "epoch": 2.2016438356164385, + "grad_norm": 2.3267500400543213, + "learning_rate": 1.5089358245329002e-06, + "loss": 2.0945, + "step": 2009 + }, + { + "epoch": 2.202739726027397, + "grad_norm": 2.1972532272338867, + "learning_rate": 1.5069049553208775e-06, + "loss": 2.0822, + "step": 2010 + }, + { + "epoch": 2.202739726027397, + "eval_loss": 2.26534104347229, + "eval_runtime": 41.6482, + "eval_samples_per_second": 24.011, + "eval_steps_per_second": 1.008, + "step": 2010 + }, + { + "epoch": 2.203835616438356, + "grad_norm": 2.161071538925171, + "learning_rate": 1.5048740861088547e-06, + "loss": 2.0908, + "step": 2011 + }, + { + "epoch": 2.204931506849315, + "grad_norm": 2.1437389850616455, + "learning_rate": 1.502843216896832e-06, + "loss": 2.1041, + "step": 2012 + }, + { + "epoch": 2.206027397260274, + "grad_norm": 2.16727614402771, + "learning_rate": 1.500812347684809e-06, + "loss": 2.0209, + "step": 2013 + }, + { + "epoch": 2.207123287671233, + "grad_norm": 2.1556594371795654, + "learning_rate": 1.4987814784727864e-06, + "loss": 2.09, + "step": 2014 + }, + { + "epoch": 2.208219178082192, + "grad_norm": 2.0891411304473877, + "learning_rate": 1.4967506092607636e-06, + "loss": 2.0864, + "step": 2015 + }, + { + "epoch": 2.209315068493151, + "grad_norm": 2.031249523162842, + "learning_rate": 1.494719740048741e-06, + "loss": 2.0363, + "step": 2016 + }, + { + "epoch": 2.2104109589041094, + "grad_norm": 2.2452943325042725, + "learning_rate": 1.4926888708367182e-06, + "loss": 2.114, + "step": 2017 + }, + { + "epoch": 2.2115068493150685, + "grad_norm": 2.1752209663391113, + "learning_rate": 1.4906580016246955e-06, + "loss": 2.1123, + "step": 2018 + }, + { + "epoch": 2.2126027397260275, + "grad_norm": 2.2148444652557373, + "learning_rate": 1.4886271324126727e-06, + "loss": 2.1132, + "step": 2019 + }, + { + "epoch": 2.213698630136986, + "grad_norm": 2.1530041694641113, + "learning_rate": 1.48659626320065e-06, + "loss": 2.1278, + "step": 2020 + }, + { + "epoch": 2.213698630136986, + "eval_loss": 2.264028310775757, + "eval_runtime": 43.6346, + "eval_samples_per_second": 22.918, + "eval_steps_per_second": 0.963, + "step": 2020 + }, + { + "epoch": 2.214794520547945, + "grad_norm": 2.223219871520996, + "learning_rate": 1.4845653939886273e-06, + "loss": 2.0293, + "step": 2021 + }, + { + "epoch": 2.215890410958904, + "grad_norm": 2.122913360595703, + "learning_rate": 1.4825345247766046e-06, + "loss": 2.0411, + "step": 2022 + }, + { + "epoch": 2.216986301369863, + "grad_norm": 2.1176650524139404, + "learning_rate": 1.4805036555645818e-06, + "loss": 2.0326, + "step": 2023 + }, + { + "epoch": 2.2180821917808218, + "grad_norm": 2.1717002391815186, + "learning_rate": 1.4784727863525591e-06, + "loss": 2.0835, + "step": 2024 + }, + { + "epoch": 2.219178082191781, + "grad_norm": 2.1869051456451416, + "learning_rate": 1.4764419171405362e-06, + "loss": 2.0571, + "step": 2025 + }, + { + "epoch": 2.22027397260274, + "grad_norm": 2.2699813842773438, + "learning_rate": 1.4744110479285135e-06, + "loss": 2.1156, + "step": 2026 + }, + { + "epoch": 2.2213698630136984, + "grad_norm": 2.243837594985962, + "learning_rate": 1.4723801787164907e-06, + "loss": 2.1231, + "step": 2027 + }, + { + "epoch": 2.2224657534246575, + "grad_norm": 2.1143088340759277, + "learning_rate": 1.470349309504468e-06, + "loss": 2.11, + "step": 2028 + }, + { + "epoch": 2.2235616438356165, + "grad_norm": 2.0839781761169434, + "learning_rate": 1.4683184402924453e-06, + "loss": 2.0971, + "step": 2029 + }, + { + "epoch": 2.2246575342465755, + "grad_norm": 2.182480812072754, + "learning_rate": 1.4662875710804226e-06, + "loss": 2.1094, + "step": 2030 + }, + { + "epoch": 2.2246575342465755, + "eval_loss": 2.2636313438415527, + "eval_runtime": 43.184, + "eval_samples_per_second": 23.157, + "eval_steps_per_second": 0.973, + "step": 2030 + }, + { + "epoch": 2.225753424657534, + "grad_norm": 2.2751855850219727, + "learning_rate": 1.4642567018683998e-06, + "loss": 2.1533, + "step": 2031 + }, + { + "epoch": 2.226849315068493, + "grad_norm": 2.049236536026001, + "learning_rate": 1.4622258326563771e-06, + "loss": 2.0847, + "step": 2032 + }, + { + "epoch": 2.227945205479452, + "grad_norm": 2.0883119106292725, + "learning_rate": 1.4601949634443544e-06, + "loss": 2.0687, + "step": 2033 + }, + { + "epoch": 2.2290410958904108, + "grad_norm": 2.152400016784668, + "learning_rate": 1.4581640942323317e-06, + "loss": 2.0979, + "step": 2034 + }, + { + "epoch": 2.23013698630137, + "grad_norm": 2.221442699432373, + "learning_rate": 1.456133225020309e-06, + "loss": 2.1601, + "step": 2035 + }, + { + "epoch": 2.231232876712329, + "grad_norm": 2.1799275875091553, + "learning_rate": 1.4541023558082862e-06, + "loss": 2.1133, + "step": 2036 + }, + { + "epoch": 2.232328767123288, + "grad_norm": 2.182882070541382, + "learning_rate": 1.4520714865962635e-06, + "loss": 2.1038, + "step": 2037 + }, + { + "epoch": 2.2334246575342465, + "grad_norm": 2.1438207626342773, + "learning_rate": 1.4500406173842406e-06, + "loss": 2.1176, + "step": 2038 + }, + { + "epoch": 2.2345205479452055, + "grad_norm": 2.1394078731536865, + "learning_rate": 1.4480097481722178e-06, + "loss": 2.0928, + "step": 2039 + }, + { + "epoch": 2.2356164383561645, + "grad_norm": 2.2296125888824463, + "learning_rate": 1.4459788789601951e-06, + "loss": 2.0938, + "step": 2040 + }, + { + "epoch": 2.2356164383561645, + "eval_loss": 2.2618958950042725, + "eval_runtime": 43.253, + "eval_samples_per_second": 23.12, + "eval_steps_per_second": 0.971, + "step": 2040 + }, + { + "epoch": 2.236712328767123, + "grad_norm": 2.248995542526245, + "learning_rate": 1.4439480097481724e-06, + "loss": 2.0666, + "step": 2041 + }, + { + "epoch": 2.237808219178082, + "grad_norm": 2.0892133712768555, + "learning_rate": 1.4419171405361497e-06, + "loss": 2.0437, + "step": 2042 + }, + { + "epoch": 2.238904109589041, + "grad_norm": 2.191180944442749, + "learning_rate": 1.439886271324127e-06, + "loss": 2.1038, + "step": 2043 + }, + { + "epoch": 2.24, + "grad_norm": 2.1049792766571045, + "learning_rate": 1.4378554021121042e-06, + "loss": 2.096, + "step": 2044 + }, + { + "epoch": 2.241095890410959, + "grad_norm": 2.166987419128418, + "learning_rate": 1.4358245329000815e-06, + "loss": 2.0704, + "step": 2045 + }, + { + "epoch": 2.242191780821918, + "grad_norm": 2.3071136474609375, + "learning_rate": 1.4337936636880588e-06, + "loss": 2.1118, + "step": 2046 + }, + { + "epoch": 2.243287671232877, + "grad_norm": 2.185621976852417, + "learning_rate": 1.431762794476036e-06, + "loss": 2.1032, + "step": 2047 + }, + { + "epoch": 2.2443835616438355, + "grad_norm": 2.2100017070770264, + "learning_rate": 1.4297319252640133e-06, + "loss": 2.0542, + "step": 2048 + }, + { + "epoch": 2.2454794520547945, + "grad_norm": 2.2313308715820312, + "learning_rate": 1.4277010560519906e-06, + "loss": 2.0812, + "step": 2049 + }, + { + "epoch": 2.2465753424657535, + "grad_norm": 2.1170578002929688, + "learning_rate": 1.4256701868399674e-06, + "loss": 2.0542, + "step": 2050 + }, + { + "epoch": 2.2465753424657535, + "eval_loss": 2.2627694606781006, + "eval_runtime": 44.8509, + "eval_samples_per_second": 22.296, + "eval_steps_per_second": 0.936, + "step": 2050 + }, + { + "epoch": 2.247671232876712, + "grad_norm": 2.207075357437134, + "learning_rate": 1.4236393176279447e-06, + "loss": 2.1467, + "step": 2051 + }, + { + "epoch": 2.248767123287671, + "grad_norm": 2.1787123680114746, + "learning_rate": 1.421608448415922e-06, + "loss": 2.0824, + "step": 2052 + }, + { + "epoch": 2.24986301369863, + "grad_norm": 2.179396867752075, + "learning_rate": 1.4195775792038993e-06, + "loss": 2.1394, + "step": 2053 + }, + { + "epoch": 2.250958904109589, + "grad_norm": 2.1963279247283936, + "learning_rate": 1.4175467099918765e-06, + "loss": 2.0594, + "step": 2054 + }, + { + "epoch": 2.252054794520548, + "grad_norm": 2.165109872817993, + "learning_rate": 1.4155158407798538e-06, + "loss": 2.0812, + "step": 2055 + }, + { + "epoch": 2.253150684931507, + "grad_norm": 2.033921957015991, + "learning_rate": 1.413484971567831e-06, + "loss": 2.1124, + "step": 2056 + }, + { + "epoch": 2.254246575342466, + "grad_norm": 2.12203311920166, + "learning_rate": 1.4114541023558084e-06, + "loss": 2.0361, + "step": 2057 + }, + { + "epoch": 2.255342465753425, + "grad_norm": 2.268758535385132, + "learning_rate": 1.4094232331437856e-06, + "loss": 2.0766, + "step": 2058 + }, + { + "epoch": 2.2564383561643835, + "grad_norm": 2.1977181434631348, + "learning_rate": 1.407392363931763e-06, + "loss": 2.124, + "step": 2059 + }, + { + "epoch": 2.2575342465753425, + "grad_norm": 2.100123882293701, + "learning_rate": 1.4053614947197402e-06, + "loss": 2.1554, + "step": 2060 + }, + { + "epoch": 2.2575342465753425, + "eval_loss": 2.263918399810791, + "eval_runtime": 44.5109, + "eval_samples_per_second": 22.466, + "eval_steps_per_second": 0.944, + "step": 2060 + }, + { + "epoch": 2.2586301369863016, + "grad_norm": 2.1824963092803955, + "learning_rate": 1.4033306255077175e-06, + "loss": 2.0998, + "step": 2061 + }, + { + "epoch": 2.25972602739726, + "grad_norm": 2.211451530456543, + "learning_rate": 1.4012997562956945e-06, + "loss": 2.0266, + "step": 2062 + }, + { + "epoch": 2.260821917808219, + "grad_norm": 2.2530057430267334, + "learning_rate": 1.3992688870836718e-06, + "loss": 2.1366, + "step": 2063 + }, + { + "epoch": 2.261917808219178, + "grad_norm": 2.121558904647827, + "learning_rate": 1.397238017871649e-06, + "loss": 2.0927, + "step": 2064 + }, + { + "epoch": 2.263013698630137, + "grad_norm": 2.1322062015533447, + "learning_rate": 1.3952071486596264e-06, + "loss": 2.0929, + "step": 2065 + }, + { + "epoch": 2.264109589041096, + "grad_norm": 2.1645631790161133, + "learning_rate": 1.3931762794476036e-06, + "loss": 2.057, + "step": 2066 + }, + { + "epoch": 2.265205479452055, + "grad_norm": 2.1870594024658203, + "learning_rate": 1.391145410235581e-06, + "loss": 2.1274, + "step": 2067 + }, + { + "epoch": 2.266301369863014, + "grad_norm": 2.1355020999908447, + "learning_rate": 1.3891145410235582e-06, + "loss": 2.0784, + "step": 2068 + }, + { + "epoch": 2.2673972602739725, + "grad_norm": 2.138497829437256, + "learning_rate": 1.3870836718115355e-06, + "loss": 2.0265, + "step": 2069 + }, + { + "epoch": 2.2684931506849315, + "grad_norm": 2.3041086196899414, + "learning_rate": 1.3850528025995127e-06, + "loss": 2.0716, + "step": 2070 + }, + { + "epoch": 2.2684931506849315, + "eval_loss": 2.263742685317993, + "eval_runtime": 45.8378, + "eval_samples_per_second": 21.816, + "eval_steps_per_second": 0.916, + "step": 2070 + }, + { + "epoch": 2.2695890410958905, + "grad_norm": 2.2722716331481934, + "learning_rate": 1.38302193338749e-06, + "loss": 2.0744, + "step": 2071 + }, + { + "epoch": 2.270684931506849, + "grad_norm": 2.267909288406372, + "learning_rate": 1.3809910641754673e-06, + "loss": 2.0269, + "step": 2072 + }, + { + "epoch": 2.271780821917808, + "grad_norm": 2.271430015563965, + "learning_rate": 1.3789601949634446e-06, + "loss": 2.1165, + "step": 2073 + }, + { + "epoch": 2.272876712328767, + "grad_norm": 2.3091483116149902, + "learning_rate": 1.3769293257514218e-06, + "loss": 2.0336, + "step": 2074 + }, + { + "epoch": 2.2739726027397262, + "grad_norm": 2.2311320304870605, + "learning_rate": 1.374898456539399e-06, + "loss": 2.0964, + "step": 2075 + }, + { + "epoch": 2.275068493150685, + "grad_norm": 2.2187438011169434, + "learning_rate": 1.3728675873273762e-06, + "loss": 2.0672, + "step": 2076 + }, + { + "epoch": 2.276164383561644, + "grad_norm": 2.0933640003204346, + "learning_rate": 1.3708367181153535e-06, + "loss": 2.1113, + "step": 2077 + }, + { + "epoch": 2.277260273972603, + "grad_norm": 2.2611358165740967, + "learning_rate": 1.3688058489033307e-06, + "loss": 2.101, + "step": 2078 + }, + { + "epoch": 2.2783561643835615, + "grad_norm": 2.269864797592163, + "learning_rate": 1.366774979691308e-06, + "loss": 2.0993, + "step": 2079 + }, + { + "epoch": 2.2794520547945205, + "grad_norm": 2.242854356765747, + "learning_rate": 1.3647441104792853e-06, + "loss": 2.107, + "step": 2080 + }, + { + "epoch": 2.2794520547945205, + "eval_loss": 2.2640912532806396, + "eval_runtime": 46.7194, + "eval_samples_per_second": 21.404, + "eval_steps_per_second": 0.899, + "step": 2080 + }, + { + "epoch": 2.2805479452054795, + "grad_norm": 2.2348811626434326, + "learning_rate": 1.3627132412672626e-06, + "loss": 2.0941, + "step": 2081 + }, + { + "epoch": 2.281643835616438, + "grad_norm": 2.214179277420044, + "learning_rate": 1.3606823720552398e-06, + "loss": 2.0548, + "step": 2082 + }, + { + "epoch": 2.282739726027397, + "grad_norm": 2.216902256011963, + "learning_rate": 1.3586515028432171e-06, + "loss": 2.0955, + "step": 2083 + }, + { + "epoch": 2.283835616438356, + "grad_norm": 2.1791675090789795, + "learning_rate": 1.3566206336311944e-06, + "loss": 2.0395, + "step": 2084 + }, + { + "epoch": 2.2849315068493152, + "grad_norm": 2.135376214981079, + "learning_rate": 1.3545897644191717e-06, + "loss": 2.0837, + "step": 2085 + }, + { + "epoch": 2.286027397260274, + "grad_norm": 2.244448184967041, + "learning_rate": 1.352558895207149e-06, + "loss": 2.0805, + "step": 2086 + }, + { + "epoch": 2.287123287671233, + "grad_norm": 2.28576922416687, + "learning_rate": 1.3505280259951258e-06, + "loss": 2.1014, + "step": 2087 + }, + { + "epoch": 2.288219178082192, + "grad_norm": 2.247220516204834, + "learning_rate": 1.348497156783103e-06, + "loss": 2.0773, + "step": 2088 + }, + { + "epoch": 2.289315068493151, + "grad_norm": 2.107109308242798, + "learning_rate": 1.3464662875710804e-06, + "loss": 2.1236, + "step": 2089 + }, + { + "epoch": 2.2904109589041095, + "grad_norm": 2.2805957794189453, + "learning_rate": 1.3444354183590576e-06, + "loss": 2.0516, + "step": 2090 + }, + { + "epoch": 2.2904109589041095, + "eval_loss": 2.262258291244507, + "eval_runtime": 46.3329, + "eval_samples_per_second": 21.583, + "eval_steps_per_second": 0.906, + "step": 2090 + }, + { + "epoch": 2.2915068493150685, + "grad_norm": 2.2838428020477295, + "learning_rate": 1.342404549147035e-06, + "loss": 2.1008, + "step": 2091 + }, + { + "epoch": 2.2926027397260276, + "grad_norm": 2.1949639320373535, + "learning_rate": 1.3403736799350122e-06, + "loss": 2.0634, + "step": 2092 + }, + { + "epoch": 2.293698630136986, + "grad_norm": 2.144423246383667, + "learning_rate": 1.3383428107229895e-06, + "loss": 2.0996, + "step": 2093 + }, + { + "epoch": 2.294794520547945, + "grad_norm": 2.293369770050049, + "learning_rate": 1.3363119415109667e-06, + "loss": 2.1206, + "step": 2094 + }, + { + "epoch": 2.2958904109589042, + "grad_norm": 2.2426724433898926, + "learning_rate": 1.334281072298944e-06, + "loss": 2.1026, + "step": 2095 + }, + { + "epoch": 2.296986301369863, + "grad_norm": 2.159067392349243, + "learning_rate": 1.3322502030869213e-06, + "loss": 2.0993, + "step": 2096 + }, + { + "epoch": 2.298082191780822, + "grad_norm": 2.2819197177886963, + "learning_rate": 1.3302193338748986e-06, + "loss": 2.0776, + "step": 2097 + }, + { + "epoch": 2.299178082191781, + "grad_norm": 2.253448963165283, + "learning_rate": 1.3281884646628758e-06, + "loss": 2.1211, + "step": 2098 + }, + { + "epoch": 2.30027397260274, + "grad_norm": 2.2652339935302734, + "learning_rate": 1.3261575954508531e-06, + "loss": 2.1175, + "step": 2099 + }, + { + "epoch": 2.3013698630136985, + "grad_norm": 2.1319172382354736, + "learning_rate": 1.3241267262388302e-06, + "loss": 2.1137, + "step": 2100 + }, + { + "epoch": 2.3013698630136985, + "eval_loss": 2.263416051864624, + "eval_runtime": 47.0463, + "eval_samples_per_second": 21.256, + "eval_steps_per_second": 0.893, + "step": 2100 + }, + { + "epoch": 2.3024657534246575, + "grad_norm": 2.1784651279449463, + "learning_rate": 1.3220958570268074e-06, + "loss": 2.0951, + "step": 2101 + }, + { + "epoch": 2.3035616438356166, + "grad_norm": 2.1439764499664307, + "learning_rate": 1.3200649878147847e-06, + "loss": 2.0483, + "step": 2102 + }, + { + "epoch": 2.304657534246575, + "grad_norm": 2.1723854541778564, + "learning_rate": 1.318034118602762e-06, + "loss": 2.0851, + "step": 2103 + }, + { + "epoch": 2.305753424657534, + "grad_norm": 2.098573923110962, + "learning_rate": 1.3160032493907393e-06, + "loss": 2.1204, + "step": 2104 + }, + { + "epoch": 2.3068493150684932, + "grad_norm": 2.1380534172058105, + "learning_rate": 1.3139723801787166e-06, + "loss": 2.1094, + "step": 2105 + }, + { + "epoch": 2.3079452054794523, + "grad_norm": 2.1696887016296387, + "learning_rate": 1.3119415109666938e-06, + "loss": 2.1411, + "step": 2106 + }, + { + "epoch": 2.309041095890411, + "grad_norm": 2.1761717796325684, + "learning_rate": 1.309910641754671e-06, + "loss": 2.0893, + "step": 2107 + }, + { + "epoch": 2.31013698630137, + "grad_norm": 2.2763733863830566, + "learning_rate": 1.3078797725426484e-06, + "loss": 2.0441, + "step": 2108 + }, + { + "epoch": 2.311232876712329, + "grad_norm": 2.1061692237854004, + "learning_rate": 1.3058489033306257e-06, + "loss": 2.0652, + "step": 2109 + }, + { + "epoch": 2.3123287671232875, + "grad_norm": 2.242868423461914, + "learning_rate": 1.303818034118603e-06, + "loss": 2.0636, + "step": 2110 + }, + { + "epoch": 2.3123287671232875, + "eval_loss": 2.2643768787384033, + "eval_runtime": 43.9201, + "eval_samples_per_second": 22.769, + "eval_steps_per_second": 0.956, + "step": 2110 + }, + { + "epoch": 2.3134246575342465, + "grad_norm": 2.4198052883148193, + "learning_rate": 1.3017871649065802e-06, + "loss": 2.0939, + "step": 2111 + }, + { + "epoch": 2.3145205479452056, + "grad_norm": 2.3702943325042725, + "learning_rate": 1.2997562956945573e-06, + "loss": 2.0663, + "step": 2112 + }, + { + "epoch": 2.315616438356164, + "grad_norm": 2.1991565227508545, + "learning_rate": 1.2977254264825345e-06, + "loss": 2.0594, + "step": 2113 + }, + { + "epoch": 2.316712328767123, + "grad_norm": 2.255929946899414, + "learning_rate": 1.2956945572705118e-06, + "loss": 2.0973, + "step": 2114 + }, + { + "epoch": 2.317808219178082, + "grad_norm": 2.307612419128418, + "learning_rate": 1.293663688058489e-06, + "loss": 2.1085, + "step": 2115 + }, + { + "epoch": 2.3189041095890413, + "grad_norm": 2.237325429916382, + "learning_rate": 1.2916328188464664e-06, + "loss": 2.0485, + "step": 2116 + }, + { + "epoch": 2.32, + "grad_norm": 2.3728206157684326, + "learning_rate": 1.2896019496344436e-06, + "loss": 2.0086, + "step": 2117 + }, + { + "epoch": 2.321095890410959, + "grad_norm": 2.138505697250366, + "learning_rate": 1.287571080422421e-06, + "loss": 2.0628, + "step": 2118 + }, + { + "epoch": 2.322191780821918, + "grad_norm": 2.1808021068573, + "learning_rate": 1.2855402112103982e-06, + "loss": 2.1146, + "step": 2119 + }, + { + "epoch": 2.323287671232877, + "grad_norm": 2.1783456802368164, + "learning_rate": 1.2835093419983755e-06, + "loss": 2.0589, + "step": 2120 + }, + { + "epoch": 2.323287671232877, + "eval_loss": 2.262885332107544, + "eval_runtime": 48.7798, + "eval_samples_per_second": 20.5, + "eval_steps_per_second": 0.861, + "step": 2120 + }, + { + "epoch": 2.3243835616438355, + "grad_norm": 2.3852620124816895, + "learning_rate": 1.2814784727863528e-06, + "loss": 2.0862, + "step": 2121 + }, + { + "epoch": 2.3254794520547946, + "grad_norm": 2.223008155822754, + "learning_rate": 1.27944760357433e-06, + "loss": 2.072, + "step": 2122 + }, + { + "epoch": 2.3265753424657536, + "grad_norm": 2.1848905086517334, + "learning_rate": 1.2774167343623073e-06, + "loss": 2.1234, + "step": 2123 + }, + { + "epoch": 2.327671232876712, + "grad_norm": 2.1176791191101074, + "learning_rate": 1.2753858651502846e-06, + "loss": 2.1188, + "step": 2124 + }, + { + "epoch": 2.328767123287671, + "grad_norm": 2.2725491523742676, + "learning_rate": 1.2733549959382616e-06, + "loss": 2.0892, + "step": 2125 + }, + { + "epoch": 2.3298630136986302, + "grad_norm": 2.218189001083374, + "learning_rate": 1.271324126726239e-06, + "loss": 2.0827, + "step": 2126 + }, + { + "epoch": 2.330958904109589, + "grad_norm": 2.2110207080841064, + "learning_rate": 1.2692932575142162e-06, + "loss": 2.0857, + "step": 2127 + }, + { + "epoch": 2.332054794520548, + "grad_norm": 2.175096273422241, + "learning_rate": 1.2672623883021935e-06, + "loss": 2.0461, + "step": 2128 + }, + { + "epoch": 2.333150684931507, + "grad_norm": 2.1493241786956787, + "learning_rate": 1.2652315190901707e-06, + "loss": 2.0361, + "step": 2129 + }, + { + "epoch": 2.334246575342466, + "grad_norm": 2.276071310043335, + "learning_rate": 1.263200649878148e-06, + "loss": 2.1336, + "step": 2130 + }, + { + "epoch": 2.334246575342466, + "eval_loss": 2.262963056564331, + "eval_runtime": 49.2242, + "eval_samples_per_second": 20.315, + "eval_steps_per_second": 0.853, + "step": 2130 + }, + { + "epoch": 2.3353424657534245, + "grad_norm": 2.1708502769470215, + "learning_rate": 1.2611697806661253e-06, + "loss": 2.081, + "step": 2131 + }, + { + "epoch": 2.3364383561643836, + "grad_norm": 2.219454526901245, + "learning_rate": 1.2591389114541026e-06, + "loss": 2.0941, + "step": 2132 + }, + { + "epoch": 2.3375342465753426, + "grad_norm": 2.110680103302002, + "learning_rate": 1.2571080422420798e-06, + "loss": 2.0692, + "step": 2133 + }, + { + "epoch": 2.338630136986301, + "grad_norm": 2.186999559402466, + "learning_rate": 1.2550771730300571e-06, + "loss": 2.0652, + "step": 2134 + }, + { + "epoch": 2.33972602739726, + "grad_norm": 2.182494878768921, + "learning_rate": 1.2530463038180344e-06, + "loss": 2.1238, + "step": 2135 + }, + { + "epoch": 2.3408219178082192, + "grad_norm": 2.163588047027588, + "learning_rate": 1.2510154346060117e-06, + "loss": 2.0788, + "step": 2136 + }, + { + "epoch": 2.3419178082191783, + "grad_norm": 2.1190285682678223, + "learning_rate": 1.2489845653939887e-06, + "loss": 2.0861, + "step": 2137 + }, + { + "epoch": 2.343013698630137, + "grad_norm": 2.187049388885498, + "learning_rate": 1.246953696181966e-06, + "loss": 2.1098, + "step": 2138 + }, + { + "epoch": 2.344109589041096, + "grad_norm": 2.3307504653930664, + "learning_rate": 1.2449228269699433e-06, + "loss": 2.1443, + "step": 2139 + }, + { + "epoch": 2.345205479452055, + "grad_norm": 2.149183511734009, + "learning_rate": 1.2428919577579204e-06, + "loss": 2.0855, + "step": 2140 + }, + { + "epoch": 2.345205479452055, + "eval_loss": 2.263383388519287, + "eval_runtime": 49.7579, + "eval_samples_per_second": 20.097, + "eval_steps_per_second": 0.844, + "step": 2140 + }, + { + "epoch": 2.3463013698630135, + "grad_norm": 2.190133571624756, + "learning_rate": 1.2408610885458976e-06, + "loss": 2.059, + "step": 2141 + }, + { + "epoch": 2.3473972602739726, + "grad_norm": 2.1775476932525635, + "learning_rate": 1.238830219333875e-06, + "loss": 2.0795, + "step": 2142 + }, + { + "epoch": 2.3484931506849316, + "grad_norm": 2.148383617401123, + "learning_rate": 1.2367993501218522e-06, + "loss": 2.0614, + "step": 2143 + }, + { + "epoch": 2.34958904109589, + "grad_norm": 2.184270143508911, + "learning_rate": 1.2347684809098295e-06, + "loss": 2.1111, + "step": 2144 + }, + { + "epoch": 2.350684931506849, + "grad_norm": 2.136432647705078, + "learning_rate": 1.2327376116978067e-06, + "loss": 2.1162, + "step": 2145 + }, + { + "epoch": 2.3517808219178082, + "grad_norm": 2.1389224529266357, + "learning_rate": 1.230706742485784e-06, + "loss": 2.0369, + "step": 2146 + }, + { + "epoch": 2.3528767123287673, + "grad_norm": 2.184678316116333, + "learning_rate": 1.2286758732737613e-06, + "loss": 2.0485, + "step": 2147 + }, + { + "epoch": 2.353972602739726, + "grad_norm": 2.1494438648223877, + "learning_rate": 1.2266450040617386e-06, + "loss": 2.1399, + "step": 2148 + }, + { + "epoch": 2.355068493150685, + "grad_norm": 2.197528839111328, + "learning_rate": 1.2246141348497158e-06, + "loss": 2.1553, + "step": 2149 + }, + { + "epoch": 2.356164383561644, + "grad_norm": 2.222090005874634, + "learning_rate": 1.2225832656376931e-06, + "loss": 2.0896, + "step": 2150 + }, + { + "epoch": 2.356164383561644, + "eval_loss": 2.261875867843628, + "eval_runtime": 50.0806, + "eval_samples_per_second": 19.968, + "eval_steps_per_second": 0.839, + "step": 2150 + }, + { + "epoch": 2.357260273972603, + "grad_norm": 2.157257318496704, + "learning_rate": 1.2205523964256704e-06, + "loss": 2.0824, + "step": 2151 + }, + { + "epoch": 2.3583561643835615, + "grad_norm": 2.1270229816436768, + "learning_rate": 1.2185215272136477e-06, + "loss": 2.08, + "step": 2152 + }, + { + "epoch": 2.3594520547945206, + "grad_norm": 2.1364707946777344, + "learning_rate": 1.2164906580016247e-06, + "loss": 2.091, + "step": 2153 + }, + { + "epoch": 2.3605479452054796, + "grad_norm": 2.1324431896209717, + "learning_rate": 1.214459788789602e-06, + "loss": 2.0935, + "step": 2154 + }, + { + "epoch": 2.361643835616438, + "grad_norm": 2.1576826572418213, + "learning_rate": 1.2124289195775793e-06, + "loss": 2.0128, + "step": 2155 + }, + { + "epoch": 2.3627397260273972, + "grad_norm": 2.145265579223633, + "learning_rate": 1.2103980503655566e-06, + "loss": 2.0418, + "step": 2156 + }, + { + "epoch": 2.3638356164383563, + "grad_norm": 2.1475625038146973, + "learning_rate": 1.2083671811535338e-06, + "loss": 2.0726, + "step": 2157 + }, + { + "epoch": 2.364931506849315, + "grad_norm": 2.2365241050720215, + "learning_rate": 1.2063363119415111e-06, + "loss": 2.0598, + "step": 2158 + }, + { + "epoch": 2.366027397260274, + "grad_norm": 2.208568572998047, + "learning_rate": 1.2043054427294882e-06, + "loss": 2.0614, + "step": 2159 + }, + { + "epoch": 2.367123287671233, + "grad_norm": 2.1937685012817383, + "learning_rate": 1.2022745735174654e-06, + "loss": 2.0799, + "step": 2160 + }, + { + "epoch": 2.367123287671233, + "eval_loss": 2.2633299827575684, + "eval_runtime": 49.4985, + "eval_samples_per_second": 20.203, + "eval_steps_per_second": 0.849, + "step": 2160 + }, + { + "epoch": 2.368219178082192, + "grad_norm": 2.1548843383789062, + "learning_rate": 1.2002437043054427e-06, + "loss": 2.092, + "step": 2161 + }, + { + "epoch": 2.3693150684931505, + "grad_norm": 2.139727830886841, + "learning_rate": 1.19821283509342e-06, + "loss": 2.1167, + "step": 2162 + }, + { + "epoch": 2.3704109589041096, + "grad_norm": 2.1580495834350586, + "learning_rate": 1.1961819658813973e-06, + "loss": 2.1222, + "step": 2163 + }, + { + "epoch": 2.3715068493150686, + "grad_norm": 2.1156060695648193, + "learning_rate": 1.1941510966693746e-06, + "loss": 2.0996, + "step": 2164 + }, + { + "epoch": 2.3726027397260276, + "grad_norm": 2.280191659927368, + "learning_rate": 1.1921202274573518e-06, + "loss": 2.0737, + "step": 2165 + }, + { + "epoch": 2.3736986301369862, + "grad_norm": 2.1379661560058594, + "learning_rate": 1.190089358245329e-06, + "loss": 2.0803, + "step": 2166 + }, + { + "epoch": 2.3747945205479453, + "grad_norm": 2.261583089828491, + "learning_rate": 1.1880584890333064e-06, + "loss": 2.1225, + "step": 2167 + }, + { + "epoch": 2.3758904109589043, + "grad_norm": 2.233469009399414, + "learning_rate": 1.1860276198212837e-06, + "loss": 2.0774, + "step": 2168 + }, + { + "epoch": 2.376986301369863, + "grad_norm": 2.2570717334747314, + "learning_rate": 1.183996750609261e-06, + "loss": 2.1057, + "step": 2169 + }, + { + "epoch": 2.378082191780822, + "grad_norm": 2.2743802070617676, + "learning_rate": 1.1819658813972382e-06, + "loss": 2.093, + "step": 2170 + }, + { + "epoch": 2.378082191780822, + "eval_loss": 2.2612462043762207, + "eval_runtime": 50.375, + "eval_samples_per_second": 19.851, + "eval_steps_per_second": 0.834, + "step": 2170 + }, + { + "epoch": 2.379178082191781, + "grad_norm": 2.2214903831481934, + "learning_rate": 1.1799350121852153e-06, + "loss": 2.084, + "step": 2171 + }, + { + "epoch": 2.3802739726027395, + "grad_norm": 2.143113136291504, + "learning_rate": 1.1779041429731925e-06, + "loss": 2.1153, + "step": 2172 + }, + { + "epoch": 2.3813698630136986, + "grad_norm": 2.243147134780884, + "learning_rate": 1.1758732737611698e-06, + "loss": 2.1186, + "step": 2173 + }, + { + "epoch": 2.3824657534246576, + "grad_norm": 2.2865614891052246, + "learning_rate": 1.173842404549147e-06, + "loss": 2.0712, + "step": 2174 + }, + { + "epoch": 2.383561643835616, + "grad_norm": 2.274543523788452, + "learning_rate": 1.1718115353371244e-06, + "loss": 2.1214, + "step": 2175 + }, + { + "epoch": 2.3846575342465752, + "grad_norm": 2.249037981033325, + "learning_rate": 1.1697806661251016e-06, + "loss": 2.0703, + "step": 2176 + }, + { + "epoch": 2.3857534246575343, + "grad_norm": 2.2552239894866943, + "learning_rate": 1.167749796913079e-06, + "loss": 2.1198, + "step": 2177 + }, + { + "epoch": 2.3868493150684933, + "grad_norm": 2.2714426517486572, + "learning_rate": 1.1657189277010562e-06, + "loss": 2.1037, + "step": 2178 + }, + { + "epoch": 2.387945205479452, + "grad_norm": 2.2843732833862305, + "learning_rate": 1.1636880584890335e-06, + "loss": 2.0833, + "step": 2179 + }, + { + "epoch": 2.389041095890411, + "grad_norm": 2.3192999362945557, + "learning_rate": 1.1616571892770108e-06, + "loss": 2.0807, + "step": 2180 + }, + { + "epoch": 2.389041095890411, + "eval_loss": 2.2609643936157227, + "eval_runtime": 45.7294, + "eval_samples_per_second": 21.868, + "eval_steps_per_second": 0.918, + "step": 2180 + }, + { + "epoch": 2.39013698630137, + "grad_norm": 2.2852554321289062, + "learning_rate": 1.159626320064988e-06, + "loss": 2.1198, + "step": 2181 + }, + { + "epoch": 2.391232876712329, + "grad_norm": 2.3606770038604736, + "learning_rate": 1.1575954508529653e-06, + "loss": 2.0938, + "step": 2182 + }, + { + "epoch": 2.3923287671232876, + "grad_norm": 2.2159781455993652, + "learning_rate": 1.1555645816409426e-06, + "loss": 2.1089, + "step": 2183 + }, + { + "epoch": 2.3934246575342466, + "grad_norm": 2.2246878147125244, + "learning_rate": 1.1535337124289196e-06, + "loss": 2.0974, + "step": 2184 + }, + { + "epoch": 2.3945205479452056, + "grad_norm": 2.180250406265259, + "learning_rate": 1.151502843216897e-06, + "loss": 2.0527, + "step": 2185 + }, + { + "epoch": 2.3956164383561642, + "grad_norm": 2.1486611366271973, + "learning_rate": 1.1494719740048742e-06, + "loss": 2.0251, + "step": 2186 + }, + { + "epoch": 2.3967123287671233, + "grad_norm": 2.242518424987793, + "learning_rate": 1.1474411047928515e-06, + "loss": 2.0885, + "step": 2187 + }, + { + "epoch": 2.3978082191780823, + "grad_norm": 2.1474642753601074, + "learning_rate": 1.1454102355808287e-06, + "loss": 2.0946, + "step": 2188 + }, + { + "epoch": 2.398904109589041, + "grad_norm": 2.230221748352051, + "learning_rate": 1.143379366368806e-06, + "loss": 2.0639, + "step": 2189 + }, + { + "epoch": 2.4, + "grad_norm": 2.1985819339752197, + "learning_rate": 1.141348497156783e-06, + "loss": 2.0578, + "step": 2190 + }, + { + "epoch": 2.4, + "eval_loss": 2.264763832092285, + "eval_runtime": 48.4178, + "eval_samples_per_second": 20.654, + "eval_steps_per_second": 0.867, + "step": 2190 + }, + { + "epoch": 2.401095890410959, + "grad_norm": 2.2248053550720215, + "learning_rate": 1.1393176279447604e-06, + "loss": 2.149, + "step": 2191 + }, + { + "epoch": 2.402191780821918, + "grad_norm": 2.197953939437866, + "learning_rate": 1.1372867587327376e-06, + "loss": 2.138, + "step": 2192 + }, + { + "epoch": 2.4032876712328766, + "grad_norm": 2.2678914070129395, + "learning_rate": 1.135255889520715e-06, + "loss": 2.0202, + "step": 2193 + }, + { + "epoch": 2.4043835616438356, + "grad_norm": 2.3155181407928467, + "learning_rate": 1.1332250203086922e-06, + "loss": 2.1037, + "step": 2194 + }, + { + "epoch": 2.4054794520547946, + "grad_norm": 2.191927909851074, + "learning_rate": 1.1311941510966695e-06, + "loss": 2.0933, + "step": 2195 + }, + { + "epoch": 2.4065753424657537, + "grad_norm": 2.28680419921875, + "learning_rate": 1.1291632818846467e-06, + "loss": 2.0694, + "step": 2196 + }, + { + "epoch": 2.4076712328767123, + "grad_norm": 2.1030619144439697, + "learning_rate": 1.127132412672624e-06, + "loss": 2.0599, + "step": 2197 + }, + { + "epoch": 2.4087671232876713, + "grad_norm": 2.170182228088379, + "learning_rate": 1.1251015434606013e-06, + "loss": 2.0769, + "step": 2198 + }, + { + "epoch": 2.4098630136986303, + "grad_norm": 2.372509241104126, + "learning_rate": 1.1230706742485786e-06, + "loss": 2.0848, + "step": 2199 + }, + { + "epoch": 2.410958904109589, + "grad_norm": 2.3988277912139893, + "learning_rate": 1.1210398050365558e-06, + "loss": 2.1207, + "step": 2200 + }, + { + "epoch": 2.410958904109589, + "eval_loss": 2.261070966720581, + "eval_runtime": 47.1736, + "eval_samples_per_second": 21.198, + "eval_steps_per_second": 0.89, + "step": 2200 + }, + { + "epoch": 2.412054794520548, + "grad_norm": 2.240713357925415, + "learning_rate": 1.1190089358245331e-06, + "loss": 2.0871, + "step": 2201 + }, + { + "epoch": 2.413150684931507, + "grad_norm": 2.161547899246216, + "learning_rate": 1.1169780666125102e-06, + "loss": 2.0417, + "step": 2202 + }, + { + "epoch": 2.4142465753424656, + "grad_norm": 2.179996967315674, + "learning_rate": 1.1149471974004875e-06, + "loss": 2.1289, + "step": 2203 + }, + { + "epoch": 2.4153424657534246, + "grad_norm": 2.2374696731567383, + "learning_rate": 1.1129163281884647e-06, + "loss": 2.073, + "step": 2204 + }, + { + "epoch": 2.4164383561643836, + "grad_norm": 2.237311363220215, + "learning_rate": 1.110885458976442e-06, + "loss": 2.1315, + "step": 2205 + }, + { + "epoch": 2.417534246575342, + "grad_norm": 2.150914192199707, + "learning_rate": 1.1088545897644193e-06, + "loss": 2.0558, + "step": 2206 + }, + { + "epoch": 2.4186301369863012, + "grad_norm": 2.122668504714966, + "learning_rate": 1.1068237205523966e-06, + "loss": 2.0954, + "step": 2207 + }, + { + "epoch": 2.4197260273972603, + "grad_norm": 2.133937358856201, + "learning_rate": 1.1047928513403736e-06, + "loss": 2.0999, + "step": 2208 + }, + { + "epoch": 2.4208219178082193, + "grad_norm": 2.394454002380371, + "learning_rate": 1.102761982128351e-06, + "loss": 2.1137, + "step": 2209 + }, + { + "epoch": 2.421917808219178, + "grad_norm": 2.22770094871521, + "learning_rate": 1.1007311129163282e-06, + "loss": 2.0568, + "step": 2210 + }, + { + "epoch": 2.421917808219178, + "eval_loss": 2.2609522342681885, + "eval_runtime": 42.8266, + "eval_samples_per_second": 23.35, + "eval_steps_per_second": 0.981, + "step": 2210 + }, + { + "epoch": 2.423013698630137, + "grad_norm": 2.177499294281006, + "learning_rate": 1.0987002437043055e-06, + "loss": 2.0884, + "step": 2211 + }, + { + "epoch": 2.424109589041096, + "grad_norm": 2.113409996032715, + "learning_rate": 1.0966693744922827e-06, + "loss": 2.0629, + "step": 2212 + }, + { + "epoch": 2.425205479452055, + "grad_norm": 2.185673952102661, + "learning_rate": 1.09463850528026e-06, + "loss": 2.0363, + "step": 2213 + }, + { + "epoch": 2.4263013698630136, + "grad_norm": 2.180579423904419, + "learning_rate": 1.0926076360682373e-06, + "loss": 2.0713, + "step": 2214 + }, + { + "epoch": 2.4273972602739726, + "grad_norm": 2.1550121307373047, + "learning_rate": 1.0905767668562146e-06, + "loss": 2.059, + "step": 2215 + }, + { + "epoch": 2.4284931506849317, + "grad_norm": 2.084537982940674, + "learning_rate": 1.0885458976441918e-06, + "loss": 2.0733, + "step": 2216 + }, + { + "epoch": 2.4295890410958902, + "grad_norm": 2.1601779460906982, + "learning_rate": 1.0865150284321691e-06, + "loss": 2.1492, + "step": 2217 + }, + { + "epoch": 2.4306849315068493, + "grad_norm": 2.2117674350738525, + "learning_rate": 1.0844841592201464e-06, + "loss": 2.0873, + "step": 2218 + }, + { + "epoch": 2.4317808219178083, + "grad_norm": 2.283747434616089, + "learning_rate": 1.0824532900081237e-06, + "loss": 2.104, + "step": 2219 + }, + { + "epoch": 2.432876712328767, + "grad_norm": 2.23386287689209, + "learning_rate": 1.080422420796101e-06, + "loss": 2.0922, + "step": 2220 + }, + { + "epoch": 2.432876712328767, + "eval_loss": 2.261997699737549, + "eval_runtime": 43.1121, + "eval_samples_per_second": 23.195, + "eval_steps_per_second": 0.974, + "step": 2220 + }, + { + "epoch": 2.433972602739726, + "grad_norm": 2.2691948413848877, + "learning_rate": 1.078391551584078e-06, + "loss": 2.0831, + "step": 2221 + }, + { + "epoch": 2.435068493150685, + "grad_norm": 2.2326037883758545, + "learning_rate": 1.0763606823720553e-06, + "loss": 2.1057, + "step": 2222 + }, + { + "epoch": 2.436164383561644, + "grad_norm": 2.159956455230713, + "learning_rate": 1.0743298131600326e-06, + "loss": 2.1423, + "step": 2223 + }, + { + "epoch": 2.4372602739726026, + "grad_norm": 2.1688168048858643, + "learning_rate": 1.0722989439480098e-06, + "loss": 2.1128, + "step": 2224 + }, + { + "epoch": 2.4383561643835616, + "grad_norm": 2.167076587677002, + "learning_rate": 1.070268074735987e-06, + "loss": 2.0763, + "step": 2225 + }, + { + "epoch": 2.4394520547945207, + "grad_norm": 2.1673460006713867, + "learning_rate": 1.0682372055239644e-06, + "loss": 2.0812, + "step": 2226 + }, + { + "epoch": 2.4405479452054797, + "grad_norm": 2.1561999320983887, + "learning_rate": 1.0662063363119414e-06, + "loss": 2.068, + "step": 2227 + }, + { + "epoch": 2.4416438356164383, + "grad_norm": 2.2114264965057373, + "learning_rate": 1.0641754670999187e-06, + "loss": 2.0561, + "step": 2228 + }, + { + "epoch": 2.4427397260273973, + "grad_norm": 2.3172457218170166, + "learning_rate": 1.062144597887896e-06, + "loss": 2.082, + "step": 2229 + }, + { + "epoch": 2.4438356164383563, + "grad_norm": 2.1823253631591797, + "learning_rate": 1.0601137286758733e-06, + "loss": 2.0863, + "step": 2230 + }, + { + "epoch": 2.4438356164383563, + "eval_loss": 2.261075019836426, + "eval_runtime": 42.9325, + "eval_samples_per_second": 23.292, + "eval_steps_per_second": 0.978, + "step": 2230 + }, + { + "epoch": 2.444931506849315, + "grad_norm": 2.210676908493042, + "learning_rate": 1.0580828594638505e-06, + "loss": 2.1205, + "step": 2231 + }, + { + "epoch": 2.446027397260274, + "grad_norm": 2.1634883880615234, + "learning_rate": 1.0560519902518278e-06, + "loss": 2.0283, + "step": 2232 + }, + { + "epoch": 2.447123287671233, + "grad_norm": 2.261366844177246, + "learning_rate": 1.054021121039805e-06, + "loss": 2.1171, + "step": 2233 + }, + { + "epoch": 2.4482191780821916, + "grad_norm": 2.2115914821624756, + "learning_rate": 1.0519902518277824e-06, + "loss": 2.0596, + "step": 2234 + }, + { + "epoch": 2.4493150684931506, + "grad_norm": 2.115755558013916, + "learning_rate": 1.0499593826157596e-06, + "loss": 2.0325, + "step": 2235 + }, + { + "epoch": 2.4504109589041096, + "grad_norm": 2.189357280731201, + "learning_rate": 1.047928513403737e-06, + "loss": 2.1108, + "step": 2236 + }, + { + "epoch": 2.4515068493150687, + "grad_norm": 2.2712552547454834, + "learning_rate": 1.0458976441917142e-06, + "loss": 2.0796, + "step": 2237 + }, + { + "epoch": 2.4526027397260273, + "grad_norm": 2.1391139030456543, + "learning_rate": 1.0438667749796915e-06, + "loss": 2.0221, + "step": 2238 + }, + { + "epoch": 2.4536986301369863, + "grad_norm": 2.1458547115325928, + "learning_rate": 1.0418359057676688e-06, + "loss": 2.0283, + "step": 2239 + }, + { + "epoch": 2.4547945205479453, + "grad_norm": 2.277369737625122, + "learning_rate": 1.0398050365556458e-06, + "loss": 2.0602, + "step": 2240 + }, + { + "epoch": 2.4547945205479453, + "eval_loss": 2.2614047527313232, + "eval_runtime": 43.3285, + "eval_samples_per_second": 23.079, + "eval_steps_per_second": 0.969, + "step": 2240 + }, + { + "epoch": 2.455890410958904, + "grad_norm": 2.2398059368133545, + "learning_rate": 1.037774167343623e-06, + "loss": 2.1035, + "step": 2241 + }, + { + "epoch": 2.456986301369863, + "grad_norm": 2.065532922744751, + "learning_rate": 1.0357432981316004e-06, + "loss": 2.0609, + "step": 2242 + }, + { + "epoch": 2.458082191780822, + "grad_norm": 2.128404140472412, + "learning_rate": 1.0337124289195776e-06, + "loss": 2.0687, + "step": 2243 + }, + { + "epoch": 2.459178082191781, + "grad_norm": 2.3261191844940186, + "learning_rate": 1.031681559707555e-06, + "loss": 2.1186, + "step": 2244 + }, + { + "epoch": 2.4602739726027396, + "grad_norm": 2.1382572650909424, + "learning_rate": 1.0296506904955322e-06, + "loss": 2.11, + "step": 2245 + }, + { + "epoch": 2.4613698630136986, + "grad_norm": 2.1222827434539795, + "learning_rate": 1.0276198212835095e-06, + "loss": 2.0832, + "step": 2246 + }, + { + "epoch": 2.4624657534246577, + "grad_norm": 2.1852123737335205, + "learning_rate": 1.0255889520714867e-06, + "loss": 2.1013, + "step": 2247 + }, + { + "epoch": 2.4635616438356163, + "grad_norm": 2.2199220657348633, + "learning_rate": 1.023558082859464e-06, + "loss": 2.0696, + "step": 2248 + }, + { + "epoch": 2.4646575342465753, + "grad_norm": 2.3017048835754395, + "learning_rate": 1.0215272136474413e-06, + "loss": 2.0473, + "step": 2249 + }, + { + "epoch": 2.4657534246575343, + "grad_norm": 2.2449405193328857, + "learning_rate": 1.0194963444354186e-06, + "loss": 2.1242, + "step": 2250 + }, + { + "epoch": 2.4657534246575343, + "eval_loss": 2.2596442699432373, + "eval_runtime": 43.6954, + "eval_samples_per_second": 22.886, + "eval_steps_per_second": 0.961, + "step": 2250 + }, + { + "epoch": 2.466849315068493, + "grad_norm": 2.2314534187316895, + "learning_rate": 1.0174654752233958e-06, + "loss": 2.0427, + "step": 2251 + }, + { + "epoch": 2.467945205479452, + "grad_norm": 2.1703758239746094, + "learning_rate": 1.015434606011373e-06, + "loss": 2.1062, + "step": 2252 + }, + { + "epoch": 2.469041095890411, + "grad_norm": 2.2300264835357666, + "learning_rate": 1.0134037367993502e-06, + "loss": 2.0712, + "step": 2253 + }, + { + "epoch": 2.47013698630137, + "grad_norm": 2.234168767929077, + "learning_rate": 1.0113728675873275e-06, + "loss": 2.0909, + "step": 2254 + }, + { + "epoch": 2.4712328767123286, + "grad_norm": 2.3168294429779053, + "learning_rate": 1.0093419983753047e-06, + "loss": 2.0677, + "step": 2255 + }, + { + "epoch": 2.4723287671232876, + "grad_norm": 2.3184220790863037, + "learning_rate": 1.007311129163282e-06, + "loss": 2.0826, + "step": 2256 + }, + { + "epoch": 2.4734246575342467, + "grad_norm": 2.1959195137023926, + "learning_rate": 1.0052802599512593e-06, + "loss": 2.0322, + "step": 2257 + }, + { + "epoch": 2.4745205479452057, + "grad_norm": 2.149315595626831, + "learning_rate": 1.0032493907392364e-06, + "loss": 2.0481, + "step": 2258 + }, + { + "epoch": 2.4756164383561643, + "grad_norm": 2.296046733856201, + "learning_rate": 1.0012185215272136e-06, + "loss": 2.0772, + "step": 2259 + }, + { + "epoch": 2.4767123287671233, + "grad_norm": 2.27614688873291, + "learning_rate": 9.99187652315191e-07, + "loss": 2.058, + "step": 2260 + }, + { + "epoch": 2.4767123287671233, + "eval_loss": 2.2602272033691406, + "eval_runtime": 44.8225, + "eval_samples_per_second": 22.31, + "eval_steps_per_second": 0.937, + "step": 2260 + }, + { + "epoch": 2.4778082191780824, + "grad_norm": 2.2012453079223633, + "learning_rate": 9.971567831031682e-07, + "loss": 2.1203, + "step": 2261 + }, + { + "epoch": 2.478904109589041, + "grad_norm": 2.163407802581787, + "learning_rate": 9.951259138911455e-07, + "loss": 2.1094, + "step": 2262 + }, + { + "epoch": 2.48, + "grad_norm": 2.3561294078826904, + "learning_rate": 9.930950446791227e-07, + "loss": 2.0397, + "step": 2263 + }, + { + "epoch": 2.481095890410959, + "grad_norm": 2.1073873043060303, + "learning_rate": 9.910641754671e-07, + "loss": 2.1113, + "step": 2264 + }, + { + "epoch": 2.4821917808219176, + "grad_norm": 2.164715528488159, + "learning_rate": 9.890333062550773e-07, + "loss": 2.1396, + "step": 2265 + }, + { + "epoch": 2.4832876712328766, + "grad_norm": 2.1889424324035645, + "learning_rate": 9.870024370430546e-07, + "loss": 2.0828, + "step": 2266 + }, + { + "epoch": 2.4843835616438357, + "grad_norm": 2.2024834156036377, + "learning_rate": 9.849715678310318e-07, + "loss": 2.0857, + "step": 2267 + }, + { + "epoch": 2.4854794520547947, + "grad_norm": 2.2106359004974365, + "learning_rate": 9.829406986190091e-07, + "loss": 2.088, + "step": 2268 + }, + { + "epoch": 2.4865753424657533, + "grad_norm": 2.1396069526672363, + "learning_rate": 9.809098294069864e-07, + "loss": 2.0961, + "step": 2269 + }, + { + "epoch": 2.4876712328767123, + "grad_norm": 2.240629196166992, + "learning_rate": 9.788789601949637e-07, + "loss": 2.0289, + "step": 2270 + }, + { + "epoch": 2.4876712328767123, + "eval_loss": 2.2604970932006836, + "eval_runtime": 44.21, + "eval_samples_per_second": 22.619, + "eval_steps_per_second": 0.95, + "step": 2270 + }, + { + "epoch": 2.4887671232876714, + "grad_norm": 2.154015064239502, + "learning_rate": 9.768480909829407e-07, + "loss": 2.0948, + "step": 2271 + }, + { + "epoch": 2.48986301369863, + "grad_norm": 2.1784112453460693, + "learning_rate": 9.74817221770918e-07, + "loss": 2.137, + "step": 2272 + }, + { + "epoch": 2.490958904109589, + "grad_norm": 2.2804994583129883, + "learning_rate": 9.727863525588953e-07, + "loss": 2.081, + "step": 2273 + }, + { + "epoch": 2.492054794520548, + "grad_norm": 2.2236075401306152, + "learning_rate": 9.707554833468726e-07, + "loss": 2.1024, + "step": 2274 + }, + { + "epoch": 2.493150684931507, + "grad_norm": 2.1746160984039307, + "learning_rate": 9.687246141348498e-07, + "loss": 2.0705, + "step": 2275 + }, + { + "epoch": 2.4942465753424656, + "grad_norm": 2.2176294326782227, + "learning_rate": 9.666937449228271e-07, + "loss": 2.0681, + "step": 2276 + }, + { + "epoch": 2.4953424657534247, + "grad_norm": 2.1879889965057373, + "learning_rate": 9.646628757108042e-07, + "loss": 2.0762, + "step": 2277 + }, + { + "epoch": 2.4964383561643837, + "grad_norm": 2.391286611557007, + "learning_rate": 9.626320064987814e-07, + "loss": 2.0389, + "step": 2278 + }, + { + "epoch": 2.4975342465753423, + "grad_norm": 2.1714260578155518, + "learning_rate": 9.606011372867587e-07, + "loss": 2.098, + "step": 2279 + }, + { + "epoch": 2.4986301369863013, + "grad_norm": 2.0793519020080566, + "learning_rate": 9.58570268074736e-07, + "loss": 2.0494, + "step": 2280 + }, + { + "epoch": 2.4986301369863013, + "eval_loss": 2.260951280593872, + "eval_runtime": 46.0038, + "eval_samples_per_second": 21.737, + "eval_steps_per_second": 0.913, + "step": 2280 + }, + { + "epoch": 2.4997260273972604, + "grad_norm": 2.296259880065918, + "learning_rate": 9.565393988627133e-07, + "loss": 2.0811, + "step": 2281 + }, + { + "epoch": 2.500821917808219, + "grad_norm": 2.1998987197875977, + "learning_rate": 9.545085296506906e-07, + "loss": 2.0685, + "step": 2282 + }, + { + "epoch": 2.501917808219178, + "grad_norm": 2.187220573425293, + "learning_rate": 9.524776604386677e-07, + "loss": 2.0941, + "step": 2283 + }, + { + "epoch": 2.503013698630137, + "grad_norm": 2.2390315532684326, + "learning_rate": 9.50446791226645e-07, + "loss": 2.0918, + "step": 2284 + }, + { + "epoch": 2.504109589041096, + "grad_norm": 2.225874423980713, + "learning_rate": 9.484159220146223e-07, + "loss": 2.0376, + "step": 2285 + }, + { + "epoch": 2.5052054794520546, + "grad_norm": 2.198423147201538, + "learning_rate": 9.463850528025995e-07, + "loss": 2.1163, + "step": 2286 + }, + { + "epoch": 2.5063013698630137, + "grad_norm": 2.277414560317993, + "learning_rate": 9.443541835905768e-07, + "loss": 2.0969, + "step": 2287 + }, + { + "epoch": 2.5073972602739727, + "grad_norm": 2.134843111038208, + "learning_rate": 9.423233143785541e-07, + "loss": 2.1279, + "step": 2288 + }, + { + "epoch": 2.5084931506849317, + "grad_norm": 2.2586472034454346, + "learning_rate": 9.402924451665313e-07, + "loss": 2.1381, + "step": 2289 + }, + { + "epoch": 2.5095890410958903, + "grad_norm": 2.1865246295928955, + "learning_rate": 9.382615759545085e-07, + "loss": 2.0601, + "step": 2290 + }, + { + "epoch": 2.5095890410958903, + "eval_loss": 2.2607827186584473, + "eval_runtime": 45.5132, + "eval_samples_per_second": 21.972, + "eval_steps_per_second": 0.923, + "step": 2290 + }, + { + "epoch": 2.5106849315068493, + "grad_norm": 2.169753074645996, + "learning_rate": 9.362307067424858e-07, + "loss": 2.092, + "step": 2291 + }, + { + "epoch": 2.5117808219178084, + "grad_norm": 2.221168041229248, + "learning_rate": 9.341998375304631e-07, + "loss": 2.0689, + "step": 2292 + }, + { + "epoch": 2.512876712328767, + "grad_norm": 2.2252960205078125, + "learning_rate": 9.321689683184404e-07, + "loss": 2.1182, + "step": 2293 + }, + { + "epoch": 2.513972602739726, + "grad_norm": 2.1615395545959473, + "learning_rate": 9.301380991064176e-07, + "loss": 2.0489, + "step": 2294 + }, + { + "epoch": 2.515068493150685, + "grad_norm": 2.184530258178711, + "learning_rate": 9.281072298943949e-07, + "loss": 2.0474, + "step": 2295 + }, + { + "epoch": 2.5161643835616436, + "grad_norm": 2.248020887374878, + "learning_rate": 9.260763606823721e-07, + "loss": 2.0676, + "step": 2296 + }, + { + "epoch": 2.5172602739726027, + "grad_norm": 2.3340086936950684, + "learning_rate": 9.240454914703494e-07, + "loss": 2.0891, + "step": 2297 + }, + { + "epoch": 2.5183561643835617, + "grad_norm": 2.283479690551758, + "learning_rate": 9.220146222583266e-07, + "loss": 2.0921, + "step": 2298 + }, + { + "epoch": 2.5194520547945203, + "grad_norm": 2.1698408126831055, + "learning_rate": 9.199837530463039e-07, + "loss": 2.0567, + "step": 2299 + }, + { + "epoch": 2.5205479452054793, + "grad_norm": 2.1792118549346924, + "learning_rate": 9.179528838342812e-07, + "loss": 2.0857, + "step": 2300 + }, + { + "epoch": 2.5205479452054793, + "eval_loss": 2.2594637870788574, + "eval_runtime": 46.1455, + "eval_samples_per_second": 21.671, + "eval_steps_per_second": 0.91, + "step": 2300 + }, + { + "epoch": 2.5216438356164383, + "grad_norm": 2.174304962158203, + "learning_rate": 9.159220146222585e-07, + "loss": 2.062, + "step": 2301 + }, + { + "epoch": 2.5227397260273974, + "grad_norm": 2.232606887817383, + "learning_rate": 9.138911454102356e-07, + "loss": 2.1047, + "step": 2302 + }, + { + "epoch": 2.5238356164383564, + "grad_norm": 2.328237295150757, + "learning_rate": 9.118602761982129e-07, + "loss": 2.0072, + "step": 2303 + }, + { + "epoch": 2.524931506849315, + "grad_norm": 2.2433090209960938, + "learning_rate": 9.098294069861902e-07, + "loss": 2.085, + "step": 2304 + }, + { + "epoch": 2.526027397260274, + "grad_norm": 2.2122716903686523, + "learning_rate": 9.077985377741675e-07, + "loss": 2.0907, + "step": 2305 + }, + { + "epoch": 2.527123287671233, + "grad_norm": 2.1627163887023926, + "learning_rate": 9.057676685621447e-07, + "loss": 2.0448, + "step": 2306 + }, + { + "epoch": 2.5282191780821917, + "grad_norm": 2.115935802459717, + "learning_rate": 9.03736799350122e-07, + "loss": 2.0359, + "step": 2307 + }, + { + "epoch": 2.5293150684931507, + "grad_norm": 2.152085781097412, + "learning_rate": 9.017059301380991e-07, + "loss": 2.1218, + "step": 2308 + }, + { + "epoch": 2.5304109589041097, + "grad_norm": 2.215233564376831, + "learning_rate": 8.996750609260764e-07, + "loss": 2.1038, + "step": 2309 + }, + { + "epoch": 2.5315068493150683, + "grad_norm": 2.1552557945251465, + "learning_rate": 8.976441917140536e-07, + "loss": 2.1263, + "step": 2310 + }, + { + "epoch": 2.5315068493150683, + "eval_loss": 2.2587778568267822, + "eval_runtime": 43.2321, + "eval_samples_per_second": 23.131, + "eval_steps_per_second": 0.972, + "step": 2310 + }, + { + "epoch": 2.5326027397260273, + "grad_norm": 2.1836440563201904, + "learning_rate": 8.956133225020309e-07, + "loss": 2.0988, + "step": 2311 + }, + { + "epoch": 2.5336986301369864, + "grad_norm": 2.1814589500427246, + "learning_rate": 8.935824532900082e-07, + "loss": 2.1343, + "step": 2312 + }, + { + "epoch": 2.534794520547945, + "grad_norm": 2.2270560264587402, + "learning_rate": 8.915515840779855e-07, + "loss": 2.0524, + "step": 2313 + }, + { + "epoch": 2.535890410958904, + "grad_norm": 2.145991563796997, + "learning_rate": 8.895207148659626e-07, + "loss": 2.0448, + "step": 2314 + }, + { + "epoch": 2.536986301369863, + "grad_norm": 2.150841236114502, + "learning_rate": 8.874898456539399e-07, + "loss": 2.1069, + "step": 2315 + }, + { + "epoch": 2.538082191780822, + "grad_norm": 2.1866137981414795, + "learning_rate": 8.854589764419172e-07, + "loss": 2.0986, + "step": 2316 + }, + { + "epoch": 2.539178082191781, + "grad_norm": 2.22078275680542, + "learning_rate": 8.834281072298945e-07, + "loss": 2.1077, + "step": 2317 + }, + { + "epoch": 2.5402739726027397, + "grad_norm": 2.119999885559082, + "learning_rate": 8.813972380178717e-07, + "loss": 2.0403, + "step": 2318 + }, + { + "epoch": 2.5413698630136987, + "grad_norm": 2.223712921142578, + "learning_rate": 8.79366368805849e-07, + "loss": 2.0819, + "step": 2319 + }, + { + "epoch": 2.5424657534246577, + "grad_norm": 2.168551206588745, + "learning_rate": 8.773354995938262e-07, + "loss": 2.0782, + "step": 2320 + }, + { + "epoch": 2.5424657534246577, + "eval_loss": 2.2597217559814453, + "eval_runtime": 45.3465, + "eval_samples_per_second": 22.052, + "eval_steps_per_second": 0.926, + "step": 2320 + }, + { + "epoch": 2.5435616438356163, + "grad_norm": 2.2165210247039795, + "learning_rate": 8.753046303818035e-07, + "loss": 2.0617, + "step": 2321 + }, + { + "epoch": 2.5446575342465754, + "grad_norm": 2.213984727859497, + "learning_rate": 8.732737611697807e-07, + "loss": 2.0452, + "step": 2322 + }, + { + "epoch": 2.5457534246575344, + "grad_norm": 2.1402530670166016, + "learning_rate": 8.71242891957758e-07, + "loss": 2.0576, + "step": 2323 + }, + { + "epoch": 2.546849315068493, + "grad_norm": 2.229064464569092, + "learning_rate": 8.692120227457353e-07, + "loss": 2.0925, + "step": 2324 + }, + { + "epoch": 2.547945205479452, + "grad_norm": 2.2119393348693848, + "learning_rate": 8.671811535337126e-07, + "loss": 2.114, + "step": 2325 + }, + { + "epoch": 2.549041095890411, + "grad_norm": 2.165804386138916, + "learning_rate": 8.651502843216898e-07, + "loss": 2.0913, + "step": 2326 + }, + { + "epoch": 2.5501369863013696, + "grad_norm": 2.2303950786590576, + "learning_rate": 8.63119415109667e-07, + "loss": 2.1396, + "step": 2327 + }, + { + "epoch": 2.5512328767123287, + "grad_norm": 2.2299821376800537, + "learning_rate": 8.610885458976443e-07, + "loss": 2.0932, + "step": 2328 + }, + { + "epoch": 2.5523287671232877, + "grad_norm": 2.199394464492798, + "learning_rate": 8.590576766856216e-07, + "loss": 2.1015, + "step": 2329 + }, + { + "epoch": 2.5534246575342463, + "grad_norm": 2.130966901779175, + "learning_rate": 8.570268074735988e-07, + "loss": 2.0562, + "step": 2330 + }, + { + "epoch": 2.5534246575342463, + "eval_loss": 2.2579281330108643, + "eval_runtime": 48.0173, + "eval_samples_per_second": 20.826, + "eval_steps_per_second": 0.875, + "step": 2330 + }, + { + "epoch": 2.5545205479452053, + "grad_norm": 2.16682767868042, + "learning_rate": 8.549959382615761e-07, + "loss": 2.0633, + "step": 2331 + }, + { + "epoch": 2.5556164383561644, + "grad_norm": 2.2167282104492188, + "learning_rate": 8.529650690495534e-07, + "loss": 2.0544, + "step": 2332 + }, + { + "epoch": 2.5567123287671234, + "grad_norm": 2.1946794986724854, + "learning_rate": 8.509341998375305e-07, + "loss": 2.0904, + "step": 2333 + }, + { + "epoch": 2.5578082191780824, + "grad_norm": 2.1869359016418457, + "learning_rate": 8.489033306255077e-07, + "loss": 2.0431, + "step": 2334 + }, + { + "epoch": 2.558904109589041, + "grad_norm": 2.154102325439453, + "learning_rate": 8.46872461413485e-07, + "loss": 2.0213, + "step": 2335 + }, + { + "epoch": 2.56, + "grad_norm": 2.218848705291748, + "learning_rate": 8.448415922014623e-07, + "loss": 2.0599, + "step": 2336 + }, + { + "epoch": 2.561095890410959, + "grad_norm": 2.2210631370544434, + "learning_rate": 8.428107229894396e-07, + "loss": 2.0589, + "step": 2337 + }, + { + "epoch": 2.5621917808219177, + "grad_norm": 2.153848648071289, + "learning_rate": 8.407798537774168e-07, + "loss": 2.0925, + "step": 2338 + }, + { + "epoch": 2.5632876712328767, + "grad_norm": 2.156040668487549, + "learning_rate": 8.38748984565394e-07, + "loss": 2.0773, + "step": 2339 + }, + { + "epoch": 2.5643835616438357, + "grad_norm": 2.2344610691070557, + "learning_rate": 8.367181153533713e-07, + "loss": 2.0588, + "step": 2340 + }, + { + "epoch": 2.5643835616438357, + "eval_loss": 2.257333993911743, + "eval_runtime": 48.3717, + "eval_samples_per_second": 20.673, + "eval_steps_per_second": 0.868, + "step": 2340 + }, + { + "epoch": 2.5654794520547943, + "grad_norm": 2.2611708641052246, + "learning_rate": 8.346872461413486e-07, + "loss": 2.069, + "step": 2341 + }, + { + "epoch": 2.5665753424657534, + "grad_norm": 2.2121822834014893, + "learning_rate": 8.326563769293258e-07, + "loss": 2.0699, + "step": 2342 + }, + { + "epoch": 2.5676712328767124, + "grad_norm": 2.2059433460235596, + "learning_rate": 8.306255077173031e-07, + "loss": 2.0773, + "step": 2343 + }, + { + "epoch": 2.568767123287671, + "grad_norm": 2.151614189147949, + "learning_rate": 8.285946385052804e-07, + "loss": 2.1359, + "step": 2344 + }, + { + "epoch": 2.56986301369863, + "grad_norm": 2.2343831062316895, + "learning_rate": 8.265637692932575e-07, + "loss": 2.0375, + "step": 2345 + }, + { + "epoch": 2.570958904109589, + "grad_norm": 2.113022565841675, + "learning_rate": 8.245329000812348e-07, + "loss": 2.0961, + "step": 2346 + }, + { + "epoch": 2.572054794520548, + "grad_norm": 2.270704746246338, + "learning_rate": 8.225020308692121e-07, + "loss": 2.0932, + "step": 2347 + }, + { + "epoch": 2.573150684931507, + "grad_norm": 2.1896655559539795, + "learning_rate": 8.204711616571894e-07, + "loss": 2.0501, + "step": 2348 + }, + { + "epoch": 2.5742465753424657, + "grad_norm": 2.197575569152832, + "learning_rate": 8.184402924451667e-07, + "loss": 2.0694, + "step": 2349 + }, + { + "epoch": 2.5753424657534247, + "grad_norm": 2.172053575515747, + "learning_rate": 8.164094232331439e-07, + "loss": 2.0344, + "step": 2350 + }, + { + "epoch": 2.5753424657534247, + "eval_loss": 2.25839900970459, + "eval_runtime": 54.584, + "eval_samples_per_second": 18.32, + "eval_steps_per_second": 0.769, + "step": 2350 + }, + { + "epoch": 2.5764383561643838, + "grad_norm": 2.2496328353881836, + "learning_rate": 8.14378554021121e-07, + "loss": 2.0822, + "step": 2351 + }, + { + "epoch": 2.5775342465753424, + "grad_norm": 2.2438595294952393, + "learning_rate": 8.123476848090983e-07, + "loss": 2.0683, + "step": 2352 + }, + { + "epoch": 2.5786301369863014, + "grad_norm": 2.158176898956299, + "learning_rate": 8.103168155970755e-07, + "loss": 2.0721, + "step": 2353 + }, + { + "epoch": 2.5797260273972604, + "grad_norm": 2.231637954711914, + "learning_rate": 8.082859463850528e-07, + "loss": 2.071, + "step": 2354 + }, + { + "epoch": 2.580821917808219, + "grad_norm": 2.1711671352386475, + "learning_rate": 8.062550771730301e-07, + "loss": 2.0881, + "step": 2355 + }, + { + "epoch": 2.581917808219178, + "grad_norm": 2.3087830543518066, + "learning_rate": 8.042242079610074e-07, + "loss": 2.0822, + "step": 2356 + }, + { + "epoch": 2.583013698630137, + "grad_norm": 2.2441811561584473, + "learning_rate": 8.021933387489846e-07, + "loss": 2.0681, + "step": 2357 + }, + { + "epoch": 2.5841095890410957, + "grad_norm": 2.1263461112976074, + "learning_rate": 8.001624695369618e-07, + "loss": 2.0591, + "step": 2358 + }, + { + "epoch": 2.5852054794520547, + "grad_norm": 2.2172417640686035, + "learning_rate": 7.981316003249391e-07, + "loss": 2.0718, + "step": 2359 + }, + { + "epoch": 2.5863013698630137, + "grad_norm": 2.2697091102600098, + "learning_rate": 7.961007311129164e-07, + "loss": 2.125, + "step": 2360 + }, + { + "epoch": 2.5863013698630137, + "eval_loss": 2.258852005004883, + "eval_runtime": 53.655, + "eval_samples_per_second": 18.638, + "eval_steps_per_second": 0.783, + "step": 2360 + }, + { + "epoch": 2.5873972602739728, + "grad_norm": 2.1597752571105957, + "learning_rate": 7.940698619008936e-07, + "loss": 2.0352, + "step": 2361 + }, + { + "epoch": 2.5884931506849314, + "grad_norm": 2.189544200897217, + "learning_rate": 7.920389926888709e-07, + "loss": 2.0962, + "step": 2362 + }, + { + "epoch": 2.5895890410958904, + "grad_norm": 2.2160205841064453, + "learning_rate": 7.900081234768482e-07, + "loss": 2.1394, + "step": 2363 + }, + { + "epoch": 2.5906849315068494, + "grad_norm": 2.272719621658325, + "learning_rate": 7.879772542648254e-07, + "loss": 2.0538, + "step": 2364 + }, + { + "epoch": 2.5917808219178085, + "grad_norm": 2.3767831325531006, + "learning_rate": 7.859463850528026e-07, + "loss": 2.0896, + "step": 2365 + }, + { + "epoch": 2.592876712328767, + "grad_norm": 2.3781375885009766, + "learning_rate": 7.839155158407799e-07, + "loss": 2.0949, + "step": 2366 + }, + { + "epoch": 2.593972602739726, + "grad_norm": 2.185418128967285, + "learning_rate": 7.818846466287572e-07, + "loss": 2.0896, + "step": 2367 + }, + { + "epoch": 2.595068493150685, + "grad_norm": 2.4299957752227783, + "learning_rate": 7.798537774167345e-07, + "loss": 2.0806, + "step": 2368 + }, + { + "epoch": 2.5961643835616437, + "grad_norm": 2.273036003112793, + "learning_rate": 7.778229082047117e-07, + "loss": 2.1109, + "step": 2369 + }, + { + "epoch": 2.5972602739726027, + "grad_norm": 2.2721755504608154, + "learning_rate": 7.757920389926889e-07, + "loss": 2.066, + "step": 2370 + }, + { + "epoch": 2.5972602739726027, + "eval_loss": 2.2589120864868164, + "eval_runtime": 55.0077, + "eval_samples_per_second": 18.179, + "eval_steps_per_second": 0.764, + "step": 2370 + }, + { + "epoch": 2.5983561643835618, + "grad_norm": 2.3353123664855957, + "learning_rate": 7.737611697806662e-07, + "loss": 2.0864, + "step": 2371 + }, + { + "epoch": 2.5994520547945204, + "grad_norm": 2.2153446674346924, + "learning_rate": 7.717303005686435e-07, + "loss": 2.0587, + "step": 2372 + }, + { + "epoch": 2.6005479452054794, + "grad_norm": 2.131464958190918, + "learning_rate": 7.696994313566207e-07, + "loss": 2.1023, + "step": 2373 + }, + { + "epoch": 2.6016438356164384, + "grad_norm": 2.3265395164489746, + "learning_rate": 7.67668562144598e-07, + "loss": 2.1326, + "step": 2374 + }, + { + "epoch": 2.602739726027397, + "grad_norm": 2.270179033279419, + "learning_rate": 7.656376929325753e-07, + "loss": 2.033, + "step": 2375 + }, + { + "epoch": 2.603835616438356, + "grad_norm": 2.1226251125335693, + "learning_rate": 7.636068237205524e-07, + "loss": 2.0529, + "step": 2376 + }, + { + "epoch": 2.604931506849315, + "grad_norm": 2.4038758277893066, + "learning_rate": 7.615759545085296e-07, + "loss": 2.1136, + "step": 2377 + }, + { + "epoch": 2.606027397260274, + "grad_norm": 2.2918882369995117, + "learning_rate": 7.595450852965069e-07, + "loss": 2.0471, + "step": 2378 + }, + { + "epoch": 2.607123287671233, + "grad_norm": 2.2518844604492188, + "learning_rate": 7.575142160844842e-07, + "loss": 2.0678, + "step": 2379 + }, + { + "epoch": 2.6082191780821917, + "grad_norm": 2.34464693069458, + "learning_rate": 7.554833468724615e-07, + "loss": 2.1027, + "step": 2380 + }, + { + "epoch": 2.6082191780821917, + "eval_loss": 2.2575674057006836, + "eval_runtime": 51.2418, + "eval_samples_per_second": 19.515, + "eval_steps_per_second": 0.82, + "step": 2380 + }, + { + "epoch": 2.6093150684931508, + "grad_norm": 2.1669251918792725, + "learning_rate": 7.534524776604387e-07, + "loss": 2.0488, + "step": 2381 + }, + { + "epoch": 2.61041095890411, + "grad_norm": 2.2391371726989746, + "learning_rate": 7.51421608448416e-07, + "loss": 2.0527, + "step": 2382 + }, + { + "epoch": 2.6115068493150684, + "grad_norm": 2.211881160736084, + "learning_rate": 7.493907392363932e-07, + "loss": 2.1028, + "step": 2383 + }, + { + "epoch": 2.6126027397260274, + "grad_norm": 2.1131086349487305, + "learning_rate": 7.473598700243705e-07, + "loss": 2.0648, + "step": 2384 + }, + { + "epoch": 2.6136986301369864, + "grad_norm": 2.174081325531006, + "learning_rate": 7.453290008123477e-07, + "loss": 2.0234, + "step": 2385 + }, + { + "epoch": 2.614794520547945, + "grad_norm": 2.182250738143921, + "learning_rate": 7.43298131600325e-07, + "loss": 2.0805, + "step": 2386 + }, + { + "epoch": 2.615890410958904, + "grad_norm": 2.202793836593628, + "learning_rate": 7.412672623883023e-07, + "loss": 2.0817, + "step": 2387 + }, + { + "epoch": 2.616986301369863, + "grad_norm": 2.203178644180298, + "learning_rate": 7.392363931762796e-07, + "loss": 2.0719, + "step": 2388 + }, + { + "epoch": 2.6180821917808217, + "grad_norm": 2.1379380226135254, + "learning_rate": 7.372055239642567e-07, + "loss": 2.0529, + "step": 2389 + }, + { + "epoch": 2.6191780821917807, + "grad_norm": 2.1674935817718506, + "learning_rate": 7.35174654752234e-07, + "loss": 2.079, + "step": 2390 + }, + { + "epoch": 2.6191780821917807, + "eval_loss": 2.2578635215759277, + "eval_runtime": 47.3602, + "eval_samples_per_second": 21.115, + "eval_steps_per_second": 0.887, + "step": 2390 + }, + { + "epoch": 2.6202739726027398, + "grad_norm": 2.191959857940674, + "learning_rate": 7.331437855402113e-07, + "loss": 2.0989, + "step": 2391 + }, + { + "epoch": 2.621369863013699, + "grad_norm": 2.1883914470672607, + "learning_rate": 7.311129163281886e-07, + "loss": 2.1036, + "step": 2392 + }, + { + "epoch": 2.6224657534246574, + "grad_norm": 2.20548152923584, + "learning_rate": 7.290820471161658e-07, + "loss": 2.0424, + "step": 2393 + }, + { + "epoch": 2.6235616438356164, + "grad_norm": 2.138023853302002, + "learning_rate": 7.270511779041431e-07, + "loss": 2.0993, + "step": 2394 + }, + { + "epoch": 2.6246575342465754, + "grad_norm": 2.145514965057373, + "learning_rate": 7.250203086921203e-07, + "loss": 2.1025, + "step": 2395 + }, + { + "epoch": 2.6257534246575345, + "grad_norm": 2.159749746322632, + "learning_rate": 7.229894394800976e-07, + "loss": 2.0261, + "step": 2396 + }, + { + "epoch": 2.626849315068493, + "grad_norm": 2.1717426776885986, + "learning_rate": 7.209585702680748e-07, + "loss": 2.0635, + "step": 2397 + }, + { + "epoch": 2.627945205479452, + "grad_norm": 2.123973846435547, + "learning_rate": 7.189277010560521e-07, + "loss": 2.0888, + "step": 2398 + }, + { + "epoch": 2.629041095890411, + "grad_norm": 2.154726266860962, + "learning_rate": 7.168968318440294e-07, + "loss": 2.043, + "step": 2399 + }, + { + "epoch": 2.6301369863013697, + "grad_norm": 2.1826655864715576, + "learning_rate": 7.148659626320067e-07, + "loss": 2.0598, + "step": 2400 + }, + { + "epoch": 2.6301369863013697, + "eval_loss": 2.257997751235962, + "eval_runtime": 45.2942, + "eval_samples_per_second": 22.078, + "eval_steps_per_second": 0.927, + "step": 2400 + }, + { + "epoch": 2.6312328767123287, + "grad_norm": 2.2959976196289062, + "learning_rate": 7.128350934199837e-07, + "loss": 2.0612, + "step": 2401 + }, + { + "epoch": 2.632328767123288, + "grad_norm": 2.2452211380004883, + "learning_rate": 7.10804224207961e-07, + "loss": 2.107, + "step": 2402 + }, + { + "epoch": 2.6334246575342464, + "grad_norm": 2.2367494106292725, + "learning_rate": 7.087733549959383e-07, + "loss": 2.1173, + "step": 2403 + }, + { + "epoch": 2.6345205479452054, + "grad_norm": 2.1162500381469727, + "learning_rate": 7.067424857839155e-07, + "loss": 2.072, + "step": 2404 + }, + { + "epoch": 2.6356164383561644, + "grad_norm": 2.157289981842041, + "learning_rate": 7.047116165718928e-07, + "loss": 2.0737, + "step": 2405 + }, + { + "epoch": 2.636712328767123, + "grad_norm": 2.2959725856781006, + "learning_rate": 7.026807473598701e-07, + "loss": 2.0737, + "step": 2406 + }, + { + "epoch": 2.637808219178082, + "grad_norm": 2.238600254058838, + "learning_rate": 7.006498781478473e-07, + "loss": 2.0778, + "step": 2407 + }, + { + "epoch": 2.638904109589041, + "grad_norm": 2.3807833194732666, + "learning_rate": 6.986190089358245e-07, + "loss": 2.1165, + "step": 2408 + }, + { + "epoch": 2.64, + "grad_norm": 2.2371649742126465, + "learning_rate": 6.965881397238018e-07, + "loss": 2.0547, + "step": 2409 + }, + { + "epoch": 2.641095890410959, + "grad_norm": 2.1728153228759766, + "learning_rate": 6.945572705117791e-07, + "loss": 2.0761, + "step": 2410 + }, + { + "epoch": 2.641095890410959, + "eval_loss": 2.25693941116333, + "eval_runtime": 41.2551, + "eval_samples_per_second": 24.239, + "eval_steps_per_second": 1.018, + "step": 2410 + }, + { + "epoch": 2.6421917808219177, + "grad_norm": 2.1825814247131348, + "learning_rate": 6.925264012997564e-07, + "loss": 2.0897, + "step": 2411 + }, + { + "epoch": 2.643287671232877, + "grad_norm": 2.3182337284088135, + "learning_rate": 6.904955320877336e-07, + "loss": 2.0859, + "step": 2412 + }, + { + "epoch": 2.644383561643836, + "grad_norm": 2.202728271484375, + "learning_rate": 6.884646628757109e-07, + "loss": 2.0826, + "step": 2413 + }, + { + "epoch": 2.6454794520547944, + "grad_norm": 2.1550540924072266, + "learning_rate": 6.864337936636881e-07, + "loss": 2.0936, + "step": 2414 + }, + { + "epoch": 2.6465753424657534, + "grad_norm": 2.310295581817627, + "learning_rate": 6.844029244516654e-07, + "loss": 2.1498, + "step": 2415 + }, + { + "epoch": 2.6476712328767125, + "grad_norm": 2.1691761016845703, + "learning_rate": 6.823720552396426e-07, + "loss": 2.1226, + "step": 2416 + }, + { + "epoch": 2.648767123287671, + "grad_norm": 2.200648307800293, + "learning_rate": 6.803411860276199e-07, + "loss": 2.0571, + "step": 2417 + }, + { + "epoch": 2.64986301369863, + "grad_norm": 2.3303613662719727, + "learning_rate": 6.783103168155972e-07, + "loss": 2.1067, + "step": 2418 + }, + { + "epoch": 2.650958904109589, + "grad_norm": 2.1697466373443604, + "learning_rate": 6.762794476035745e-07, + "loss": 2.0761, + "step": 2419 + }, + { + "epoch": 2.6520547945205477, + "grad_norm": 2.2162702083587646, + "learning_rate": 6.742485783915515e-07, + "loss": 2.0735, + "step": 2420 + }, + { + "epoch": 2.6520547945205477, + "eval_loss": 2.257615804672241, + "eval_runtime": 44.8342, + "eval_samples_per_second": 22.304, + "eval_steps_per_second": 0.937, + "step": 2420 + }, + { + "epoch": 2.6531506849315067, + "grad_norm": 2.234525442123413, + "learning_rate": 6.722177091795288e-07, + "loss": 2.0942, + "step": 2421 + }, + { + "epoch": 2.6542465753424658, + "grad_norm": 2.2796056270599365, + "learning_rate": 6.701868399675061e-07, + "loss": 2.1008, + "step": 2422 + }, + { + "epoch": 2.655342465753425, + "grad_norm": 2.259611129760742, + "learning_rate": 6.681559707554834e-07, + "loss": 2.095, + "step": 2423 + }, + { + "epoch": 2.6564383561643834, + "grad_norm": 2.4727160930633545, + "learning_rate": 6.661251015434606e-07, + "loss": 2.0695, + "step": 2424 + }, + { + "epoch": 2.6575342465753424, + "grad_norm": 2.2864344120025635, + "learning_rate": 6.640942323314379e-07, + "loss": 2.0479, + "step": 2425 + }, + { + "epoch": 2.6586301369863015, + "grad_norm": 2.16770601272583, + "learning_rate": 6.620633631194151e-07, + "loss": 2.0501, + "step": 2426 + }, + { + "epoch": 2.6597260273972605, + "grad_norm": 2.206383466720581, + "learning_rate": 6.600324939073924e-07, + "loss": 2.0645, + "step": 2427 + }, + { + "epoch": 2.660821917808219, + "grad_norm": 2.301408290863037, + "learning_rate": 6.580016246953696e-07, + "loss": 2.0962, + "step": 2428 + }, + { + "epoch": 2.661917808219178, + "grad_norm": 2.2411422729492188, + "learning_rate": 6.559707554833469e-07, + "loss": 2.0613, + "step": 2429 + }, + { + "epoch": 2.663013698630137, + "grad_norm": 2.176337480545044, + "learning_rate": 6.539398862713242e-07, + "loss": 2.1169, + "step": 2430 + }, + { + "epoch": 2.663013698630137, + "eval_loss": 2.2563741207122803, + "eval_runtime": 44.6519, + "eval_samples_per_second": 22.395, + "eval_steps_per_second": 0.941, + "step": 2430 + }, + { + "epoch": 2.6641095890410957, + "grad_norm": 2.250044107437134, + "learning_rate": 6.519090170593015e-07, + "loss": 2.0592, + "step": 2431 + }, + { + "epoch": 2.6652054794520548, + "grad_norm": 2.1671738624572754, + "learning_rate": 6.498781478472786e-07, + "loss": 2.1159, + "step": 2432 + }, + { + "epoch": 2.666301369863014, + "grad_norm": 2.152893543243408, + "learning_rate": 6.478472786352559e-07, + "loss": 2.0805, + "step": 2433 + }, + { + "epoch": 2.6673972602739724, + "grad_norm": 2.2559802532196045, + "learning_rate": 6.458164094232332e-07, + "loss": 2.108, + "step": 2434 + }, + { + "epoch": 2.6684931506849314, + "grad_norm": 2.273721694946289, + "learning_rate": 6.437855402112105e-07, + "loss": 2.0839, + "step": 2435 + }, + { + "epoch": 2.6695890410958905, + "grad_norm": 2.211909532546997, + "learning_rate": 6.417546709991877e-07, + "loss": 2.0292, + "step": 2436 + }, + { + "epoch": 2.670684931506849, + "grad_norm": 2.252609968185425, + "learning_rate": 6.39723801787165e-07, + "loss": 2.0378, + "step": 2437 + }, + { + "epoch": 2.671780821917808, + "grad_norm": 2.252904176712036, + "learning_rate": 6.376929325751423e-07, + "loss": 2.1113, + "step": 2438 + }, + { + "epoch": 2.672876712328767, + "grad_norm": 2.161658525466919, + "learning_rate": 6.356620633631195e-07, + "loss": 2.0765, + "step": 2439 + }, + { + "epoch": 2.673972602739726, + "grad_norm": 2.2181780338287354, + "learning_rate": 6.336311941510967e-07, + "loss": 2.0649, + "step": 2440 + }, + { + "epoch": 2.673972602739726, + "eval_loss": 2.2568325996398926, + "eval_runtime": 43.4163, + "eval_samples_per_second": 23.033, + "eval_steps_per_second": 0.967, + "step": 2440 + }, + { + "epoch": 2.675068493150685, + "grad_norm": 2.2615585327148438, + "learning_rate": 6.31600324939074e-07, + "loss": 2.0674, + "step": 2441 + }, + { + "epoch": 2.6761643835616438, + "grad_norm": 2.227696180343628, + "learning_rate": 6.295694557270513e-07, + "loss": 2.095, + "step": 2442 + }, + { + "epoch": 2.677260273972603, + "grad_norm": 2.1348302364349365, + "learning_rate": 6.275385865150286e-07, + "loss": 2.1081, + "step": 2443 + }, + { + "epoch": 2.678356164383562, + "grad_norm": 2.213954210281372, + "learning_rate": 6.255077173030058e-07, + "loss": 2.0819, + "step": 2444 + }, + { + "epoch": 2.6794520547945204, + "grad_norm": 2.2341015338897705, + "learning_rate": 6.23476848090983e-07, + "loss": 2.075, + "step": 2445 + }, + { + "epoch": 2.6805479452054795, + "grad_norm": 2.1238317489624023, + "learning_rate": 6.214459788789602e-07, + "loss": 2.0798, + "step": 2446 + }, + { + "epoch": 2.6816438356164385, + "grad_norm": 2.155360698699951, + "learning_rate": 6.194151096669375e-07, + "loss": 2.0748, + "step": 2447 + }, + { + "epoch": 2.682739726027397, + "grad_norm": 2.240307569503784, + "learning_rate": 6.173842404549147e-07, + "loss": 2.1168, + "step": 2448 + }, + { + "epoch": 2.683835616438356, + "grad_norm": 2.239638566970825, + "learning_rate": 6.15353371242892e-07, + "loss": 2.0743, + "step": 2449 + }, + { + "epoch": 2.684931506849315, + "grad_norm": 2.1331958770751953, + "learning_rate": 6.133225020308693e-07, + "loss": 2.0832, + "step": 2450 + }, + { + "epoch": 2.684931506849315, + "eval_loss": 2.2567331790924072, + "eval_runtime": 44.0159, + "eval_samples_per_second": 22.719, + "eval_steps_per_second": 0.954, + "step": 2450 + }, + { + "epoch": 2.6860273972602737, + "grad_norm": 2.113851547241211, + "learning_rate": 6.112916328188466e-07, + "loss": 2.0024, + "step": 2451 + }, + { + "epoch": 2.6871232876712328, + "grad_norm": 2.1780712604522705, + "learning_rate": 6.092607636068238e-07, + "loss": 2.0652, + "step": 2452 + }, + { + "epoch": 2.688219178082192, + "grad_norm": 2.1993205547332764, + "learning_rate": 6.07229894394801e-07, + "loss": 2.1325, + "step": 2453 + }, + { + "epoch": 2.689315068493151, + "grad_norm": 2.177851915359497, + "learning_rate": 6.051990251827783e-07, + "loss": 2.0678, + "step": 2454 + }, + { + "epoch": 2.69041095890411, + "grad_norm": 2.144575595855713, + "learning_rate": 6.031681559707556e-07, + "loss": 2.0625, + "step": 2455 + }, + { + "epoch": 2.6915068493150685, + "grad_norm": 2.1539716720581055, + "learning_rate": 6.011372867587327e-07, + "loss": 2.0884, + "step": 2456 + }, + { + "epoch": 2.6926027397260275, + "grad_norm": 2.1430931091308594, + "learning_rate": 5.9910641754671e-07, + "loss": 2.0809, + "step": 2457 + }, + { + "epoch": 2.6936986301369865, + "grad_norm": 2.1018853187561035, + "learning_rate": 5.970755483346873e-07, + "loss": 2.0608, + "step": 2458 + }, + { + "epoch": 2.694794520547945, + "grad_norm": 2.2417850494384766, + "learning_rate": 5.950446791226646e-07, + "loss": 2.0745, + "step": 2459 + }, + { + "epoch": 2.695890410958904, + "grad_norm": 2.232811689376831, + "learning_rate": 5.930138099106418e-07, + "loss": 2.056, + "step": 2460 + }, + { + "epoch": 2.695890410958904, + "eval_loss": 2.2569868564605713, + "eval_runtime": 44.1345, + "eval_samples_per_second": 22.658, + "eval_steps_per_second": 0.952, + "step": 2460 + }, + { + "epoch": 2.696986301369863, + "grad_norm": 2.2861359119415283, + "learning_rate": 5.909829406986191e-07, + "loss": 2.0758, + "step": 2461 + }, + { + "epoch": 2.6980821917808218, + "grad_norm": 2.235025644302368, + "learning_rate": 5.889520714865963e-07, + "loss": 2.0612, + "step": 2462 + }, + { + "epoch": 2.699178082191781, + "grad_norm": 2.2089755535125732, + "learning_rate": 5.869212022745735e-07, + "loss": 2.0818, + "step": 2463 + }, + { + "epoch": 2.70027397260274, + "grad_norm": 2.210655927658081, + "learning_rate": 5.848903330625508e-07, + "loss": 2.0369, + "step": 2464 + }, + { + "epoch": 2.7013698630136984, + "grad_norm": 2.2496445178985596, + "learning_rate": 5.828594638505281e-07, + "loss": 2.0831, + "step": 2465 + }, + { + "epoch": 2.7024657534246574, + "grad_norm": 2.198668956756592, + "learning_rate": 5.808285946385054e-07, + "loss": 2.0684, + "step": 2466 + }, + { + "epoch": 2.7035616438356165, + "grad_norm": 2.1553661823272705, + "learning_rate": 5.787977254264827e-07, + "loss": 2.0713, + "step": 2467 + }, + { + "epoch": 2.704657534246575, + "grad_norm": 2.2714855670928955, + "learning_rate": 5.767668562144598e-07, + "loss": 2.0923, + "step": 2468 + }, + { + "epoch": 2.705753424657534, + "grad_norm": 2.3484857082366943, + "learning_rate": 5.747359870024371e-07, + "loss": 2.1147, + "step": 2469 + }, + { + "epoch": 2.706849315068493, + "grad_norm": 2.2298500537872314, + "learning_rate": 5.727051177904144e-07, + "loss": 2.0691, + "step": 2470 + }, + { + "epoch": 2.706849315068493, + "eval_loss": 2.258481025695801, + "eval_runtime": 48.2537, + "eval_samples_per_second": 20.724, + "eval_steps_per_second": 0.87, + "step": 2470 + }, + { + "epoch": 2.707945205479452, + "grad_norm": 2.3507697582244873, + "learning_rate": 5.706742485783915e-07, + "loss": 2.0823, + "step": 2471 + }, + { + "epoch": 2.709041095890411, + "grad_norm": 2.276279926300049, + "learning_rate": 5.686433793663688e-07, + "loss": 2.1147, + "step": 2472 + }, + { + "epoch": 2.71013698630137, + "grad_norm": 2.277862787246704, + "learning_rate": 5.666125101543461e-07, + "loss": 2.046, + "step": 2473 + }, + { + "epoch": 2.711232876712329, + "grad_norm": 2.2062981128692627, + "learning_rate": 5.645816409423234e-07, + "loss": 2.037, + "step": 2474 + }, + { + "epoch": 2.712328767123288, + "grad_norm": 2.185692548751831, + "learning_rate": 5.625507717303006e-07, + "loss": 2.1187, + "step": 2475 + }, + { + "epoch": 2.7134246575342464, + "grad_norm": 2.1925277709960938, + "learning_rate": 5.605199025182779e-07, + "loss": 2.0874, + "step": 2476 + }, + { + "epoch": 2.7145205479452055, + "grad_norm": 2.142359733581543, + "learning_rate": 5.584890333062551e-07, + "loss": 2.0591, + "step": 2477 + }, + { + "epoch": 2.7156164383561645, + "grad_norm": 2.102414846420288, + "learning_rate": 5.564581640942324e-07, + "loss": 2.0437, + "step": 2478 + }, + { + "epoch": 2.716712328767123, + "grad_norm": 2.315833806991577, + "learning_rate": 5.544272948822096e-07, + "loss": 2.1203, + "step": 2479 + }, + { + "epoch": 2.717808219178082, + "grad_norm": 2.2440335750579834, + "learning_rate": 5.523964256701868e-07, + "loss": 2.0533, + "step": 2480 + }, + { + "epoch": 2.717808219178082, + "eval_loss": 2.2562665939331055, + "eval_runtime": 48.2587, + "eval_samples_per_second": 20.722, + "eval_steps_per_second": 0.87, + "step": 2480 + }, + { + "epoch": 2.718904109589041, + "grad_norm": 2.2396035194396973, + "learning_rate": 5.503655564581641e-07, + "loss": 2.081, + "step": 2481 + }, + { + "epoch": 2.7199999999999998, + "grad_norm": 2.1422877311706543, + "learning_rate": 5.483346872461414e-07, + "loss": 2.0968, + "step": 2482 + }, + { + "epoch": 2.721095890410959, + "grad_norm": 2.1234922409057617, + "learning_rate": 5.463038180341186e-07, + "loss": 2.0596, + "step": 2483 + }, + { + "epoch": 2.722191780821918, + "grad_norm": 2.2119977474212646, + "learning_rate": 5.442729488220959e-07, + "loss": 2.0518, + "step": 2484 + }, + { + "epoch": 2.723287671232877, + "grad_norm": 2.2316629886627197, + "learning_rate": 5.422420796100732e-07, + "loss": 2.0543, + "step": 2485 + }, + { + "epoch": 2.724383561643836, + "grad_norm": 2.2670505046844482, + "learning_rate": 5.402112103980505e-07, + "loss": 2.0821, + "step": 2486 + }, + { + "epoch": 2.7254794520547945, + "grad_norm": 2.233405590057373, + "learning_rate": 5.381803411860276e-07, + "loss": 2.0658, + "step": 2487 + }, + { + "epoch": 2.7265753424657535, + "grad_norm": 2.2305350303649902, + "learning_rate": 5.361494719740049e-07, + "loss": 2.0936, + "step": 2488 + }, + { + "epoch": 2.7276712328767125, + "grad_norm": 2.318533420562744, + "learning_rate": 5.341186027619822e-07, + "loss": 2.1449, + "step": 2489 + }, + { + "epoch": 2.728767123287671, + "grad_norm": 2.2384085655212402, + "learning_rate": 5.320877335499594e-07, + "loss": 2.0528, + "step": 2490 + }, + { + "epoch": 2.728767123287671, + "eval_loss": 2.2571513652801514, + "eval_runtime": 49.9845, + "eval_samples_per_second": 20.006, + "eval_steps_per_second": 0.84, + "step": 2490 + }, + { + "epoch": 2.72986301369863, + "grad_norm": 2.2665085792541504, + "learning_rate": 5.300568643379366e-07, + "loss": 2.0627, + "step": 2491 + }, + { + "epoch": 2.730958904109589, + "grad_norm": 2.1696274280548096, + "learning_rate": 5.280259951259139e-07, + "loss": 2.1144, + "step": 2492 + }, + { + "epoch": 2.732054794520548, + "grad_norm": 2.3332390785217285, + "learning_rate": 5.259951259138912e-07, + "loss": 2.084, + "step": 2493 + }, + { + "epoch": 2.733150684931507, + "grad_norm": 2.1803693771362305, + "learning_rate": 5.239642567018685e-07, + "loss": 2.0455, + "step": 2494 + }, + { + "epoch": 2.734246575342466, + "grad_norm": 2.279205560684204, + "learning_rate": 5.219333874898457e-07, + "loss": 2.0551, + "step": 2495 + }, + { + "epoch": 2.7353424657534244, + "grad_norm": 2.27457857131958, + "learning_rate": 5.199025182778229e-07, + "loss": 2.0663, + "step": 2496 + }, + { + "epoch": 2.7364383561643835, + "grad_norm": 2.1237573623657227, + "learning_rate": 5.178716490658002e-07, + "loss": 2.0792, + "step": 2497 + }, + { + "epoch": 2.7375342465753425, + "grad_norm": 2.2879316806793213, + "learning_rate": 5.158407798537775e-07, + "loss": 2.0621, + "step": 2498 + }, + { + "epoch": 2.7386301369863015, + "grad_norm": 2.236769199371338, + "learning_rate": 5.138099106417547e-07, + "loss": 2.0196, + "step": 2499 + }, + { + "epoch": 2.73972602739726, + "grad_norm": 2.2341625690460205, + "learning_rate": 5.11779041429732e-07, + "loss": 2.0114, + "step": 2500 + }, + { + "epoch": 2.73972602739726, + "eval_loss": 2.255844831466675, + "eval_runtime": 49.3954, + "eval_samples_per_second": 20.245, + "eval_steps_per_second": 0.85, + "step": 2500 + }, + { + "epoch": 2.740821917808219, + "grad_norm": 2.160499334335327, + "learning_rate": 5.097481722177093e-07, + "loss": 2.0632, + "step": 2501 + }, + { + "epoch": 2.741917808219178, + "grad_norm": 2.267280340194702, + "learning_rate": 5.077173030056865e-07, + "loss": 2.0638, + "step": 2502 + }, + { + "epoch": 2.743013698630137, + "grad_norm": 2.278404951095581, + "learning_rate": 5.056864337936637e-07, + "loss": 2.1048, + "step": 2503 + }, + { + "epoch": 2.744109589041096, + "grad_norm": 2.331587314605713, + "learning_rate": 5.03655564581641e-07, + "loss": 2.1125, + "step": 2504 + }, + { + "epoch": 2.745205479452055, + "grad_norm": 2.2172396183013916, + "learning_rate": 5.016246953696182e-07, + "loss": 2.0827, + "step": 2505 + }, + { + "epoch": 2.746301369863014, + "grad_norm": 2.1748032569885254, + "learning_rate": 4.995938261575955e-07, + "loss": 2.1537, + "step": 2506 + }, + { + "epoch": 2.7473972602739725, + "grad_norm": 2.3126235008239746, + "learning_rate": 4.975629569455727e-07, + "loss": 2.0634, + "step": 2507 + }, + { + "epoch": 2.7484931506849315, + "grad_norm": 2.1927216053009033, + "learning_rate": 4.9553208773355e-07, + "loss": 2.0858, + "step": 2508 + }, + { + "epoch": 2.7495890410958905, + "grad_norm": 2.1474738121032715, + "learning_rate": 4.935012185215273e-07, + "loss": 2.1293, + "step": 2509 + }, + { + "epoch": 2.750684931506849, + "grad_norm": 2.187898874282837, + "learning_rate": 4.914703493095046e-07, + "loss": 2.0519, + "step": 2510 + }, + { + "epoch": 2.750684931506849, + "eval_loss": 2.2556207180023193, + "eval_runtime": 42.5107, + "eval_samples_per_second": 23.523, + "eval_steps_per_second": 0.988, + "step": 2510 + }, + { + "epoch": 2.751780821917808, + "grad_norm": 2.2922189235687256, + "learning_rate": 4.894394800974818e-07, + "loss": 2.0232, + "step": 2511 + }, + { + "epoch": 2.752876712328767, + "grad_norm": 2.1553919315338135, + "learning_rate": 4.87408610885459e-07, + "loss": 2.063, + "step": 2512 + }, + { + "epoch": 2.7539726027397258, + "grad_norm": 2.1872823238372803, + "learning_rate": 4.853777416734363e-07, + "loss": 2.1445, + "step": 2513 + }, + { + "epoch": 2.755068493150685, + "grad_norm": 2.2216968536376953, + "learning_rate": 4.833468724614136e-07, + "loss": 2.097, + "step": 2514 + }, + { + "epoch": 2.756164383561644, + "grad_norm": 2.121187210083008, + "learning_rate": 4.813160032493907e-07, + "loss": 2.0653, + "step": 2515 + }, + { + "epoch": 2.757260273972603, + "grad_norm": 2.159050703048706, + "learning_rate": 4.79285134037368e-07, + "loss": 2.0655, + "step": 2516 + }, + { + "epoch": 2.758356164383562, + "grad_norm": 2.1701362133026123, + "learning_rate": 4.772542648253453e-07, + "loss": 2.0453, + "step": 2517 + }, + { + "epoch": 2.7594520547945205, + "grad_norm": 2.2150604724884033, + "learning_rate": 4.752233956133225e-07, + "loss": 2.0661, + "step": 2518 + }, + { + "epoch": 2.7605479452054795, + "grad_norm": 2.1968846321105957, + "learning_rate": 4.731925264012998e-07, + "loss": 2.0685, + "step": 2519 + }, + { + "epoch": 2.7616438356164386, + "grad_norm": 2.2282097339630127, + "learning_rate": 4.7116165718927705e-07, + "loss": 2.1185, + "step": 2520 + }, + { + "epoch": 2.7616438356164386, + "eval_loss": 2.2558414936065674, + "eval_runtime": 45.278, + "eval_samples_per_second": 22.086, + "eval_steps_per_second": 0.928, + "step": 2520 + }, + { + "epoch": 2.762739726027397, + "grad_norm": 2.1571805477142334, + "learning_rate": 4.6913078797725427e-07, + "loss": 2.0742, + "step": 2521 + }, + { + "epoch": 2.763835616438356, + "grad_norm": 2.166057825088501, + "learning_rate": 4.6709991876523155e-07, + "loss": 2.0852, + "step": 2522 + }, + { + "epoch": 2.764931506849315, + "grad_norm": 2.2078871726989746, + "learning_rate": 4.650690495532088e-07, + "loss": 2.0622, + "step": 2523 + }, + { + "epoch": 2.766027397260274, + "grad_norm": 2.2643191814422607, + "learning_rate": 4.6303818034118605e-07, + "loss": 2.1187, + "step": 2524 + }, + { + "epoch": 2.767123287671233, + "grad_norm": 2.2169249057769775, + "learning_rate": 4.610073111291633e-07, + "loss": 2.1023, + "step": 2525 + }, + { + "epoch": 2.768219178082192, + "grad_norm": 2.2126193046569824, + "learning_rate": 4.589764419171406e-07, + "loss": 2.0195, + "step": 2526 + }, + { + "epoch": 2.7693150684931505, + "grad_norm": 2.1085822582244873, + "learning_rate": 4.569455727051178e-07, + "loss": 2.0664, + "step": 2527 + }, + { + "epoch": 2.7704109589041095, + "grad_norm": 2.242666006088257, + "learning_rate": 4.549147034930951e-07, + "loss": 2.043, + "step": 2528 + }, + { + "epoch": 2.7715068493150685, + "grad_norm": 2.1822879314422607, + "learning_rate": 4.5288383428107237e-07, + "loss": 2.1123, + "step": 2529 + }, + { + "epoch": 2.7726027397260276, + "grad_norm": 2.312371253967285, + "learning_rate": 4.5085296506904954e-07, + "loss": 2.077, + "step": 2530 + }, + { + "epoch": 2.7726027397260276, + "eval_loss": 2.256312608718872, + "eval_runtime": 52.4906, + "eval_samples_per_second": 19.051, + "eval_steps_per_second": 0.8, + "step": 2530 + }, + { + "epoch": 2.773698630136986, + "grad_norm": 2.397677183151245, + "learning_rate": 4.488220958570268e-07, + "loss": 2.0727, + "step": 2531 + }, + { + "epoch": 2.774794520547945, + "grad_norm": 2.199512004852295, + "learning_rate": 4.467912266450041e-07, + "loss": 2.0606, + "step": 2532 + }, + { + "epoch": 2.775890410958904, + "grad_norm": 2.2825701236724854, + "learning_rate": 4.447603574329813e-07, + "loss": 2.0538, + "step": 2533 + }, + { + "epoch": 2.7769863013698632, + "grad_norm": 2.2335455417633057, + "learning_rate": 4.427294882209586e-07, + "loss": 2.0982, + "step": 2534 + }, + { + "epoch": 2.778082191780822, + "grad_norm": 2.1414973735809326, + "learning_rate": 4.4069861900893587e-07, + "loss": 2.0462, + "step": 2535 + }, + { + "epoch": 2.779178082191781, + "grad_norm": 2.314079761505127, + "learning_rate": 4.386677497969131e-07, + "loss": 2.0851, + "step": 2536 + }, + { + "epoch": 2.78027397260274, + "grad_norm": 2.190136671066284, + "learning_rate": 4.3663688058489037e-07, + "loss": 2.1396, + "step": 2537 + }, + { + "epoch": 2.7813698630136985, + "grad_norm": 2.215665340423584, + "learning_rate": 4.3460601137286764e-07, + "loss": 2.0799, + "step": 2538 + }, + { + "epoch": 2.7824657534246575, + "grad_norm": 2.301847219467163, + "learning_rate": 4.325751421608449e-07, + "loss": 2.1113, + "step": 2539 + }, + { + "epoch": 2.7835616438356166, + "grad_norm": 2.2280142307281494, + "learning_rate": 4.3054427294882214e-07, + "loss": 2.0816, + "step": 2540 + }, + { + "epoch": 2.7835616438356166, + "eval_loss": 2.2550714015960693, + "eval_runtime": 51.3991, + "eval_samples_per_second": 19.456, + "eval_steps_per_second": 0.817, + "step": 2540 + }, + { + "epoch": 2.784657534246575, + "grad_norm": 2.235395669937134, + "learning_rate": 4.285134037367994e-07, + "loss": 2.107, + "step": 2541 + }, + { + "epoch": 2.785753424657534, + "grad_norm": 2.3115875720977783, + "learning_rate": 4.264825345247767e-07, + "loss": 2.0729, + "step": 2542 + }, + { + "epoch": 2.786849315068493, + "grad_norm": 2.23592472076416, + "learning_rate": 4.2445166531275386e-07, + "loss": 2.026, + "step": 2543 + }, + { + "epoch": 2.787945205479452, + "grad_norm": 2.1707637310028076, + "learning_rate": 4.2242079610073114e-07, + "loss": 2.0536, + "step": 2544 + }, + { + "epoch": 2.789041095890411, + "grad_norm": 2.2537641525268555, + "learning_rate": 4.203899268887084e-07, + "loss": 2.0185, + "step": 2545 + }, + { + "epoch": 2.79013698630137, + "grad_norm": 2.2223923206329346, + "learning_rate": 4.1835905767668564e-07, + "loss": 2.1085, + "step": 2546 + }, + { + "epoch": 2.791232876712329, + "grad_norm": 2.2132439613342285, + "learning_rate": 4.163281884646629e-07, + "loss": 2.1329, + "step": 2547 + }, + { + "epoch": 2.792328767123288, + "grad_norm": 2.220302104949951, + "learning_rate": 4.142973192526402e-07, + "loss": 2.0627, + "step": 2548 + }, + { + "epoch": 2.7934246575342465, + "grad_norm": 2.367136240005493, + "learning_rate": 4.122664500406174e-07, + "loss": 2.1051, + "step": 2549 + }, + { + "epoch": 2.7945205479452055, + "grad_norm": 2.26678466796875, + "learning_rate": 4.102355808285947e-07, + "loss": 2.087, + "step": 2550 + }, + { + "epoch": 2.7945205479452055, + "eval_loss": 2.255298376083374, + "eval_runtime": 51.4984, + "eval_samples_per_second": 19.418, + "eval_steps_per_second": 0.816, + "step": 2550 + }, + { + "epoch": 2.7956164383561646, + "grad_norm": 3.4108686447143555, + "learning_rate": 4.0820471161657196e-07, + "loss": 2.0575, + "step": 2551 + }, + { + "epoch": 2.796712328767123, + "grad_norm": 2.231989860534668, + "learning_rate": 4.0617384240454913e-07, + "loss": 2.1014, + "step": 2552 + }, + { + "epoch": 2.797808219178082, + "grad_norm": 2.262299060821533, + "learning_rate": 4.041429731925264e-07, + "loss": 2.101, + "step": 2553 + }, + { + "epoch": 2.7989041095890412, + "grad_norm": 2.2188589572906494, + "learning_rate": 4.021121039805037e-07, + "loss": 2.0279, + "step": 2554 + }, + { + "epoch": 2.8, + "grad_norm": 2.200364112854004, + "learning_rate": 4.000812347684809e-07, + "loss": 2.0487, + "step": 2555 + }, + { + "epoch": 2.801095890410959, + "grad_norm": 2.2213754653930664, + "learning_rate": 3.980503655564582e-07, + "loss": 2.0588, + "step": 2556 + }, + { + "epoch": 2.802191780821918, + "grad_norm": 2.1714670658111572, + "learning_rate": 3.9601949634443546e-07, + "loss": 2.077, + "step": 2557 + }, + { + "epoch": 2.8032876712328765, + "grad_norm": 2.1523191928863525, + "learning_rate": 3.939886271324127e-07, + "loss": 2.0582, + "step": 2558 + }, + { + "epoch": 2.8043835616438355, + "grad_norm": 2.2840702533721924, + "learning_rate": 3.9195775792038996e-07, + "loss": 2.0734, + "step": 2559 + }, + { + "epoch": 2.8054794520547945, + "grad_norm": 2.2464489936828613, + "learning_rate": 3.8992688870836723e-07, + "loss": 2.0986, + "step": 2560 + }, + { + "epoch": 2.8054794520547945, + "eval_loss": 2.255234956741333, + "eval_runtime": 51.4682, + "eval_samples_per_second": 19.429, + "eval_steps_per_second": 0.816, + "step": 2560 + }, + { + "epoch": 2.8065753424657536, + "grad_norm": 2.2760167121887207, + "learning_rate": 3.8789601949634446e-07, + "loss": 2.0982, + "step": 2561 + }, + { + "epoch": 2.807671232876712, + "grad_norm": 2.171504020690918, + "learning_rate": 3.8586515028432173e-07, + "loss": 2.064, + "step": 2562 + }, + { + "epoch": 2.808767123287671, + "grad_norm": 2.2497222423553467, + "learning_rate": 3.83834281072299e-07, + "loss": 2.0083, + "step": 2563 + }, + { + "epoch": 2.8098630136986302, + "grad_norm": 2.2118873596191406, + "learning_rate": 3.818034118602762e-07, + "loss": 2.0908, + "step": 2564 + }, + { + "epoch": 2.8109589041095893, + "grad_norm": 2.1849124431610107, + "learning_rate": 3.7977254264825345e-07, + "loss": 2.0517, + "step": 2565 + }, + { + "epoch": 2.812054794520548, + "grad_norm": 2.1902389526367188, + "learning_rate": 3.7774167343623073e-07, + "loss": 2.0869, + "step": 2566 + }, + { + "epoch": 2.813150684931507, + "grad_norm": 2.165940523147583, + "learning_rate": 3.75710804224208e-07, + "loss": 2.0931, + "step": 2567 + }, + { + "epoch": 2.814246575342466, + "grad_norm": 2.3142664432525635, + "learning_rate": 3.7367993501218523e-07, + "loss": 2.1084, + "step": 2568 + }, + { + "epoch": 2.8153424657534245, + "grad_norm": 2.2530455589294434, + "learning_rate": 3.716490658001625e-07, + "loss": 2.0593, + "step": 2569 + }, + { + "epoch": 2.8164383561643835, + "grad_norm": 2.2404749393463135, + "learning_rate": 3.696181965881398e-07, + "loss": 2.0701, + "step": 2570 + }, + { + "epoch": 2.8164383561643835, + "eval_loss": 2.2546989917755127, + "eval_runtime": 53.9082, + "eval_samples_per_second": 18.55, + "eval_steps_per_second": 0.779, + "step": 2570 + }, + { + "epoch": 2.8175342465753426, + "grad_norm": 2.2436273097991943, + "learning_rate": 3.67587327376117e-07, + "loss": 2.0357, + "step": 2571 + }, + { + "epoch": 2.818630136986301, + "grad_norm": 2.1815578937530518, + "learning_rate": 3.655564581640943e-07, + "loss": 2.0893, + "step": 2572 + }, + { + "epoch": 2.81972602739726, + "grad_norm": 2.122631549835205, + "learning_rate": 3.6352558895207155e-07, + "loss": 2.07, + "step": 2573 + }, + { + "epoch": 2.8208219178082192, + "grad_norm": 2.2400054931640625, + "learning_rate": 3.614947197400488e-07, + "loss": 2.0464, + "step": 2574 + }, + { + "epoch": 2.821917808219178, + "grad_norm": 2.137629508972168, + "learning_rate": 3.5946385052802605e-07, + "loss": 2.0715, + "step": 2575 + }, + { + "epoch": 2.823013698630137, + "grad_norm": 2.1749675273895264, + "learning_rate": 3.5743298131600333e-07, + "loss": 2.0721, + "step": 2576 + }, + { + "epoch": 2.824109589041096, + "grad_norm": 2.2205114364624023, + "learning_rate": 3.554021121039805e-07, + "loss": 2.0646, + "step": 2577 + }, + { + "epoch": 2.825205479452055, + "grad_norm": 2.1990790367126465, + "learning_rate": 3.533712428919578e-07, + "loss": 2.1126, + "step": 2578 + }, + { + "epoch": 2.826301369863014, + "grad_norm": 2.2628259658813477, + "learning_rate": 3.5134037367993505e-07, + "loss": 2.0678, + "step": 2579 + }, + { + "epoch": 2.8273972602739725, + "grad_norm": 2.1200449466705322, + "learning_rate": 3.4930950446791227e-07, + "loss": 2.0742, + "step": 2580 + }, + { + "epoch": 2.8273972602739725, + "eval_loss": 2.255067825317383, + "eval_runtime": 52.3163, + "eval_samples_per_second": 19.114, + "eval_steps_per_second": 0.803, + "step": 2580 + }, + { + "epoch": 2.8284931506849316, + "grad_norm": 2.2298808097839355, + "learning_rate": 3.4727863525588955e-07, + "loss": 2.0633, + "step": 2581 + }, + { + "epoch": 2.8295890410958906, + "grad_norm": 2.308842420578003, + "learning_rate": 3.452477660438668e-07, + "loss": 2.0684, + "step": 2582 + }, + { + "epoch": 2.830684931506849, + "grad_norm": 2.1684608459472656, + "learning_rate": 3.4321689683184405e-07, + "loss": 2.1031, + "step": 2583 + }, + { + "epoch": 2.831780821917808, + "grad_norm": 2.248159408569336, + "learning_rate": 3.411860276198213e-07, + "loss": 2.0705, + "step": 2584 + }, + { + "epoch": 2.8328767123287673, + "grad_norm": 2.2653911113739014, + "learning_rate": 3.391551584077986e-07, + "loss": 2.0497, + "step": 2585 + }, + { + "epoch": 2.833972602739726, + "grad_norm": 2.2408978939056396, + "learning_rate": 3.3712428919577577e-07, + "loss": 2.1005, + "step": 2586 + }, + { + "epoch": 2.835068493150685, + "grad_norm": 2.260836601257324, + "learning_rate": 3.3509341998375304e-07, + "loss": 2.0846, + "step": 2587 + }, + { + "epoch": 2.836164383561644, + "grad_norm": 2.124600648880005, + "learning_rate": 3.330625507717303e-07, + "loss": 2.0343, + "step": 2588 + }, + { + "epoch": 2.8372602739726025, + "grad_norm": 2.1987783908843994, + "learning_rate": 3.3103168155970754e-07, + "loss": 2.0612, + "step": 2589 + }, + { + "epoch": 2.8383561643835615, + "grad_norm": 2.318005323410034, + "learning_rate": 3.290008123476848e-07, + "loss": 2.1269, + "step": 2590 + }, + { + "epoch": 2.8383561643835615, + "eval_loss": 2.2549877166748047, + "eval_runtime": 49.8147, + "eval_samples_per_second": 20.074, + "eval_steps_per_second": 0.843, + "step": 2590 + }, + { + "epoch": 2.8394520547945206, + "grad_norm": 2.209320306777954, + "learning_rate": 3.269699431356621e-07, + "loss": 2.1259, + "step": 2591 + }, + { + "epoch": 2.8405479452054796, + "grad_norm": 2.1967287063598633, + "learning_rate": 3.249390739236393e-07, + "loss": 2.055, + "step": 2592 + }, + { + "epoch": 2.8416438356164386, + "grad_norm": 2.1936452388763428, + "learning_rate": 3.229082047116166e-07, + "loss": 2.0248, + "step": 2593 + }, + { + "epoch": 2.842739726027397, + "grad_norm": 2.2553858757019043, + "learning_rate": 3.2087733549959387e-07, + "loss": 2.1085, + "step": 2594 + }, + { + "epoch": 2.8438356164383563, + "grad_norm": 2.166267156600952, + "learning_rate": 3.1884646628757114e-07, + "loss": 2.0994, + "step": 2595 + }, + { + "epoch": 2.8449315068493153, + "grad_norm": 2.1507043838500977, + "learning_rate": 3.1681559707554837e-07, + "loss": 2.0772, + "step": 2596 + }, + { + "epoch": 2.846027397260274, + "grad_norm": 2.2255585193634033, + "learning_rate": 3.1478472786352564e-07, + "loss": 2.0843, + "step": 2597 + }, + { + "epoch": 2.847123287671233, + "grad_norm": 2.1625001430511475, + "learning_rate": 3.127538586515029e-07, + "loss": 2.0401, + "step": 2598 + }, + { + "epoch": 2.848219178082192, + "grad_norm": 2.211390495300293, + "learning_rate": 3.107229894394801e-07, + "loss": 2.064, + "step": 2599 + }, + { + "epoch": 2.8493150684931505, + "grad_norm": 2.122021198272705, + "learning_rate": 3.0869212022745736e-07, + "loss": 2.0666, + "step": 2600 + }, + { + "epoch": 2.8493150684931505, + "eval_loss": 2.2553422451019287, + "eval_runtime": 50.6075, + "eval_samples_per_second": 19.76, + "eval_steps_per_second": 0.83, + "step": 2600 + }, + { + "epoch": 2.8504109589041096, + "grad_norm": 2.2089385986328125, + "learning_rate": 3.0666125101543464e-07, + "loss": 2.0213, + "step": 2601 + }, + { + "epoch": 2.8515068493150686, + "grad_norm": 2.2044339179992676, + "learning_rate": 3.046303818034119e-07, + "loss": 2.0725, + "step": 2602 + }, + { + "epoch": 2.852602739726027, + "grad_norm": 2.3303098678588867, + "learning_rate": 3.0259951259138914e-07, + "loss": 2.1329, + "step": 2603 + }, + { + "epoch": 2.853698630136986, + "grad_norm": 2.281168222427368, + "learning_rate": 3.0056864337936636e-07, + "loss": 2.0924, + "step": 2604 + }, + { + "epoch": 2.8547945205479452, + "grad_norm": 2.263659954071045, + "learning_rate": 2.9853777416734364e-07, + "loss": 1.9901, + "step": 2605 + }, + { + "epoch": 2.855890410958904, + "grad_norm": 2.2229549884796143, + "learning_rate": 2.965069049553209e-07, + "loss": 2.0711, + "step": 2606 + }, + { + "epoch": 2.856986301369863, + "grad_norm": 2.209564685821533, + "learning_rate": 2.9447603574329814e-07, + "loss": 2.0654, + "step": 2607 + }, + { + "epoch": 2.858082191780822, + "grad_norm": 2.2653021812438965, + "learning_rate": 2.924451665312754e-07, + "loss": 2.0381, + "step": 2608 + }, + { + "epoch": 2.859178082191781, + "grad_norm": 2.259511947631836, + "learning_rate": 2.904142973192527e-07, + "loss": 2.0569, + "step": 2609 + }, + { + "epoch": 2.86027397260274, + "grad_norm": 2.238955497741699, + "learning_rate": 2.883834281072299e-07, + "loss": 2.005, + "step": 2610 + }, + { + "epoch": 2.86027397260274, + "eval_loss": 2.2556395530700684, + "eval_runtime": 42.0282, + "eval_samples_per_second": 23.794, + "eval_steps_per_second": 0.999, + "step": 2610 + }, + { + "epoch": 2.8613698630136986, + "grad_norm": 2.332723379135132, + "learning_rate": 2.863525588952072e-07, + "loss": 2.0475, + "step": 2611 + }, + { + "epoch": 2.8624657534246576, + "grad_norm": 2.133765935897827, + "learning_rate": 2.843216896831844e-07, + "loss": 2.034, + "step": 2612 + }, + { + "epoch": 2.8635616438356166, + "grad_norm": 2.186910390853882, + "learning_rate": 2.822908204711617e-07, + "loss": 2.0233, + "step": 2613 + }, + { + "epoch": 2.864657534246575, + "grad_norm": 2.2730395793914795, + "learning_rate": 2.8025995125913896e-07, + "loss": 2.0948, + "step": 2614 + }, + { + "epoch": 2.8657534246575342, + "grad_norm": 2.1879565715789795, + "learning_rate": 2.782290820471162e-07, + "loss": 2.0557, + "step": 2615 + }, + { + "epoch": 2.8668493150684933, + "grad_norm": 2.2330384254455566, + "learning_rate": 2.761982128350934e-07, + "loss": 2.1018, + "step": 2616 + }, + { + "epoch": 2.867945205479452, + "grad_norm": 2.1939926147460938, + "learning_rate": 2.741673436230707e-07, + "loss": 2.07, + "step": 2617 + }, + { + "epoch": 2.869041095890411, + "grad_norm": 2.245387077331543, + "learning_rate": 2.7213647441104796e-07, + "loss": 2.0828, + "step": 2618 + }, + { + "epoch": 2.87013698630137, + "grad_norm": 2.2011120319366455, + "learning_rate": 2.7010560519902523e-07, + "loss": 2.0706, + "step": 2619 + }, + { + "epoch": 2.8712328767123285, + "grad_norm": 2.1083333492279053, + "learning_rate": 2.6807473598700246e-07, + "loss": 2.0073, + "step": 2620 + }, + { + "epoch": 2.8712328767123285, + "eval_loss": 2.2542226314544678, + "eval_runtime": 43.9483, + "eval_samples_per_second": 22.754, + "eval_steps_per_second": 0.956, + "step": 2620 + }, + { + "epoch": 2.8723287671232876, + "grad_norm": 2.251528739929199, + "learning_rate": 2.660438667749797e-07, + "loss": 2.0875, + "step": 2621 + }, + { + "epoch": 2.8734246575342466, + "grad_norm": 2.2272961139678955, + "learning_rate": 2.6401299756295696e-07, + "loss": 2.0383, + "step": 2622 + }, + { + "epoch": 2.8745205479452056, + "grad_norm": 2.1850523948669434, + "learning_rate": 2.6198212835093423e-07, + "loss": 2.0583, + "step": 2623 + }, + { + "epoch": 2.8756164383561647, + "grad_norm": 2.1318511962890625, + "learning_rate": 2.5995125913891145e-07, + "loss": 2.0367, + "step": 2624 + }, + { + "epoch": 2.8767123287671232, + "grad_norm": 2.3000693321228027, + "learning_rate": 2.5792038992688873e-07, + "loss": 2.0105, + "step": 2625 + }, + { + "epoch": 2.8778082191780823, + "grad_norm": 2.2896645069122314, + "learning_rate": 2.55889520714866e-07, + "loss": 2.0428, + "step": 2626 + }, + { + "epoch": 2.8789041095890413, + "grad_norm": 2.225282669067383, + "learning_rate": 2.5385865150284323e-07, + "loss": 2.071, + "step": 2627 + }, + { + "epoch": 2.88, + "grad_norm": 2.3280179500579834, + "learning_rate": 2.518277822908205e-07, + "loss": 2.1029, + "step": 2628 + }, + { + "epoch": 2.881095890410959, + "grad_norm": 2.247983932495117, + "learning_rate": 2.4979691307879773e-07, + "loss": 2.0822, + "step": 2629 + }, + { + "epoch": 2.882191780821918, + "grad_norm": 2.225578784942627, + "learning_rate": 2.47766043866775e-07, + "loss": 2.104, + "step": 2630 + }, + { + "epoch": 2.882191780821918, + "eval_loss": 2.2539989948272705, + "eval_runtime": 44.2896, + "eval_samples_per_second": 22.579, + "eval_steps_per_second": 0.948, + "step": 2630 + }, + { + "epoch": 2.8832876712328765, + "grad_norm": 2.2030534744262695, + "learning_rate": 2.457351746547523e-07, + "loss": 2.0346, + "step": 2631 + }, + { + "epoch": 2.8843835616438356, + "grad_norm": 2.2446820735931396, + "learning_rate": 2.437043054427295e-07, + "loss": 2.0318, + "step": 2632 + }, + { + "epoch": 2.8854794520547946, + "grad_norm": 2.211186170578003, + "learning_rate": 2.416734362307068e-07, + "loss": 2.0763, + "step": 2633 + }, + { + "epoch": 2.886575342465753, + "grad_norm": 2.174253225326538, + "learning_rate": 2.39642567018684e-07, + "loss": 2.0562, + "step": 2634 + }, + { + "epoch": 2.8876712328767122, + "grad_norm": 2.189344882965088, + "learning_rate": 2.3761169780666125e-07, + "loss": 2.0352, + "step": 2635 + }, + { + "epoch": 2.8887671232876713, + "grad_norm": 2.111955404281616, + "learning_rate": 2.3558082859463852e-07, + "loss": 2.0785, + "step": 2636 + }, + { + "epoch": 2.8898630136986303, + "grad_norm": 2.1549975872039795, + "learning_rate": 2.3354995938261577e-07, + "loss": 2.0585, + "step": 2637 + }, + { + "epoch": 2.890958904109589, + "grad_norm": 2.210564136505127, + "learning_rate": 2.3151909017059302e-07, + "loss": 2.007, + "step": 2638 + }, + { + "epoch": 2.892054794520548, + "grad_norm": 2.1878621578216553, + "learning_rate": 2.294882209585703e-07, + "loss": 2.0915, + "step": 2639 + }, + { + "epoch": 2.893150684931507, + "grad_norm": 2.306269645690918, + "learning_rate": 2.2745735174654755e-07, + "loss": 2.0897, + "step": 2640 + }, + { + "epoch": 2.893150684931507, + "eval_loss": 2.2534706592559814, + "eval_runtime": 43.673, + "eval_samples_per_second": 22.897, + "eval_steps_per_second": 0.962, + "step": 2640 + }, + { + "epoch": 2.894246575342466, + "grad_norm": 2.088963747024536, + "learning_rate": 2.2542648253452477e-07, + "loss": 2.0307, + "step": 2641 + }, + { + "epoch": 2.8953424657534246, + "grad_norm": 2.1883978843688965, + "learning_rate": 2.2339561332250205e-07, + "loss": 2.0799, + "step": 2642 + }, + { + "epoch": 2.8964383561643836, + "grad_norm": 2.2062344551086426, + "learning_rate": 2.213647441104793e-07, + "loss": 2.0982, + "step": 2643 + }, + { + "epoch": 2.8975342465753426, + "grad_norm": 2.2856674194335938, + "learning_rate": 2.1933387489845655e-07, + "loss": 2.08, + "step": 2644 + }, + { + "epoch": 2.8986301369863012, + "grad_norm": 2.2235584259033203, + "learning_rate": 2.1730300568643382e-07, + "loss": 2.052, + "step": 2645 + }, + { + "epoch": 2.8997260273972603, + "grad_norm": 2.197960376739502, + "learning_rate": 2.1527213647441107e-07, + "loss": 2.0717, + "step": 2646 + }, + { + "epoch": 2.9008219178082193, + "grad_norm": 2.1405436992645264, + "learning_rate": 2.1324126726238835e-07, + "loss": 2.0739, + "step": 2647 + }, + { + "epoch": 2.901917808219178, + "grad_norm": 2.2070910930633545, + "learning_rate": 2.1121039805036557e-07, + "loss": 2.0708, + "step": 2648 + }, + { + "epoch": 2.903013698630137, + "grad_norm": 2.1531498432159424, + "learning_rate": 2.0917952883834282e-07, + "loss": 2.0728, + "step": 2649 + }, + { + "epoch": 2.904109589041096, + "grad_norm": 2.181342840194702, + "learning_rate": 2.071486596263201e-07, + "loss": 2.0885, + "step": 2650 + }, + { + "epoch": 2.904109589041096, + "eval_loss": 2.253673791885376, + "eval_runtime": 44.1386, + "eval_samples_per_second": 22.656, + "eval_steps_per_second": 0.952, + "step": 2650 + }, + { + "epoch": 2.9052054794520545, + "grad_norm": 2.2566676139831543, + "learning_rate": 2.0511779041429734e-07, + "loss": 2.0426, + "step": 2651 + }, + { + "epoch": 2.9063013698630136, + "grad_norm": 2.2971630096435547, + "learning_rate": 2.0308692120227457e-07, + "loss": 2.0507, + "step": 2652 + }, + { + "epoch": 2.9073972602739726, + "grad_norm": 2.162501096725464, + "learning_rate": 2.0105605199025184e-07, + "loss": 2.098, + "step": 2653 + }, + { + "epoch": 2.9084931506849316, + "grad_norm": 2.2389981746673584, + "learning_rate": 1.990251827782291e-07, + "loss": 2.1044, + "step": 2654 + }, + { + "epoch": 2.9095890410958907, + "grad_norm": 2.221620559692383, + "learning_rate": 1.9699431356620634e-07, + "loss": 2.1208, + "step": 2655 + }, + { + "epoch": 2.9106849315068493, + "grad_norm": 2.192843198776245, + "learning_rate": 1.9496344435418362e-07, + "loss": 2.0252, + "step": 2656 + }, + { + "epoch": 2.9117808219178083, + "grad_norm": 2.2490060329437256, + "learning_rate": 1.9293257514216087e-07, + "loss": 2.0893, + "step": 2657 + }, + { + "epoch": 2.9128767123287673, + "grad_norm": 2.1450841426849365, + "learning_rate": 1.909017059301381e-07, + "loss": 2.0613, + "step": 2658 + }, + { + "epoch": 2.913972602739726, + "grad_norm": 2.1686902046203613, + "learning_rate": 1.8887083671811536e-07, + "loss": 2.0514, + "step": 2659 + }, + { + "epoch": 2.915068493150685, + "grad_norm": 2.153362989425659, + "learning_rate": 1.8683996750609261e-07, + "loss": 2.0405, + "step": 2660 + }, + { + "epoch": 2.915068493150685, + "eval_loss": 2.253669500350952, + "eval_runtime": 44.3924, + "eval_samples_per_second": 22.526, + "eval_steps_per_second": 0.946, + "step": 2660 + }, + { + "epoch": 2.916164383561644, + "grad_norm": 2.247509241104126, + "learning_rate": 1.848090982940699e-07, + "loss": 2.1075, + "step": 2661 + }, + { + "epoch": 2.9172602739726026, + "grad_norm": 2.143491268157959, + "learning_rate": 1.8277822908204714e-07, + "loss": 2.1169, + "step": 2662 + }, + { + "epoch": 2.9183561643835616, + "grad_norm": 2.2100136280059814, + "learning_rate": 1.807473598700244e-07, + "loss": 2.0462, + "step": 2663 + }, + { + "epoch": 2.9194520547945206, + "grad_norm": 2.2918028831481934, + "learning_rate": 1.7871649065800166e-07, + "loss": 2.0704, + "step": 2664 + }, + { + "epoch": 2.9205479452054792, + "grad_norm": 2.2782156467437744, + "learning_rate": 1.766856214459789e-07, + "loss": 2.1429, + "step": 2665 + }, + { + "epoch": 2.9216438356164383, + "grad_norm": 2.1859452724456787, + "learning_rate": 1.7465475223395614e-07, + "loss": 2.08, + "step": 2666 + }, + { + "epoch": 2.9227397260273973, + "grad_norm": 2.204322576522827, + "learning_rate": 1.726238830219334e-07, + "loss": 2.0767, + "step": 2667 + }, + { + "epoch": 2.9238356164383563, + "grad_norm": 2.126098394393921, + "learning_rate": 1.7059301380991066e-07, + "loss": 2.0525, + "step": 2668 + }, + { + "epoch": 2.924931506849315, + "grad_norm": 2.218139886856079, + "learning_rate": 1.6856214459788788e-07, + "loss": 2.0596, + "step": 2669 + }, + { + "epoch": 2.926027397260274, + "grad_norm": 2.176334857940674, + "learning_rate": 1.6653127538586516e-07, + "loss": 2.0142, + "step": 2670 + }, + { + "epoch": 2.926027397260274, + "eval_loss": 2.2536814212799072, + "eval_runtime": 43.7023, + "eval_samples_per_second": 22.882, + "eval_steps_per_second": 0.961, + "step": 2670 + }, + { + "epoch": 2.927123287671233, + "grad_norm": 2.199193239212036, + "learning_rate": 1.645004061738424e-07, + "loss": 2.1094, + "step": 2671 + }, + { + "epoch": 2.928219178082192, + "grad_norm": 2.207334280014038, + "learning_rate": 1.6246953696181966e-07, + "loss": 2.0413, + "step": 2672 + }, + { + "epoch": 2.9293150684931506, + "grad_norm": 2.1710855960845947, + "learning_rate": 1.6043866774979693e-07, + "loss": 2.075, + "step": 2673 + }, + { + "epoch": 2.9304109589041096, + "grad_norm": 2.2313950061798096, + "learning_rate": 1.5840779853777418e-07, + "loss": 2.0929, + "step": 2674 + }, + { + "epoch": 2.9315068493150687, + "grad_norm": 2.2185397148132324, + "learning_rate": 1.5637692932575146e-07, + "loss": 2.0633, + "step": 2675 + }, + { + "epoch": 2.9326027397260273, + "grad_norm": 2.2891173362731934, + "learning_rate": 1.5434606011372868e-07, + "loss": 2.0399, + "step": 2676 + }, + { + "epoch": 2.9336986301369863, + "grad_norm": 2.2687325477600098, + "learning_rate": 1.5231519090170596e-07, + "loss": 2.1174, + "step": 2677 + }, + { + "epoch": 2.9347945205479453, + "grad_norm": 2.1965270042419434, + "learning_rate": 1.5028432168968318e-07, + "loss": 2.0852, + "step": 2678 + }, + { + "epoch": 2.935890410958904, + "grad_norm": 2.1656129360198975, + "learning_rate": 1.4825345247766046e-07, + "loss": 2.0773, + "step": 2679 + }, + { + "epoch": 2.936986301369863, + "grad_norm": 2.122925281524658, + "learning_rate": 1.462225832656377e-07, + "loss": 2.1025, + "step": 2680 + }, + { + "epoch": 2.936986301369863, + "eval_loss": 2.253000259399414, + "eval_runtime": 45.5037, + "eval_samples_per_second": 21.976, + "eval_steps_per_second": 0.923, + "step": 2680 + }, + { + "epoch": 2.938082191780822, + "grad_norm": 2.2209091186523438, + "learning_rate": 1.4419171405361496e-07, + "loss": 2.0484, + "step": 2681 + }, + { + "epoch": 2.9391780821917806, + "grad_norm": 2.1906654834747314, + "learning_rate": 1.421608448415922e-07, + "loss": 2.0735, + "step": 2682 + }, + { + "epoch": 2.9402739726027396, + "grad_norm": 2.160768508911133, + "learning_rate": 1.4012997562956948e-07, + "loss": 2.0887, + "step": 2683 + }, + { + "epoch": 2.9413698630136986, + "grad_norm": 2.2567224502563477, + "learning_rate": 1.380991064175467e-07, + "loss": 2.1037, + "step": 2684 + }, + { + "epoch": 2.9424657534246577, + "grad_norm": 2.192812204360962, + "learning_rate": 1.3606823720552398e-07, + "loss": 2.0887, + "step": 2685 + }, + { + "epoch": 2.9435616438356167, + "grad_norm": 2.2602829933166504, + "learning_rate": 1.3403736799350123e-07, + "loss": 2.1062, + "step": 2686 + }, + { + "epoch": 2.9446575342465753, + "grad_norm": 2.238593339920044, + "learning_rate": 1.3200649878147848e-07, + "loss": 2.0787, + "step": 2687 + }, + { + "epoch": 2.9457534246575343, + "grad_norm": 2.3080153465270996, + "learning_rate": 1.2997562956945573e-07, + "loss": 2.096, + "step": 2688 + }, + { + "epoch": 2.9468493150684933, + "grad_norm": 2.229102611541748, + "learning_rate": 1.27944760357433e-07, + "loss": 2.1142, + "step": 2689 + }, + { + "epoch": 2.947945205479452, + "grad_norm": 2.1906065940856934, + "learning_rate": 1.2591389114541025e-07, + "loss": 2.0969, + "step": 2690 + }, + { + "epoch": 2.947945205479452, + "eval_loss": 2.2528958320617676, + "eval_runtime": 45.9927, + "eval_samples_per_second": 21.743, + "eval_steps_per_second": 0.913, + "step": 2690 + }, + { + "epoch": 2.949041095890411, + "grad_norm": 2.213803291320801, + "learning_rate": 1.238830219333875e-07, + "loss": 2.0503, + "step": 2691 + }, + { + "epoch": 2.95013698630137, + "grad_norm": 2.14281964302063, + "learning_rate": 1.2185215272136475e-07, + "loss": 2.0217, + "step": 2692 + }, + { + "epoch": 2.9512328767123286, + "grad_norm": 2.2080299854278564, + "learning_rate": 1.19821283509342e-07, + "loss": 2.0474, + "step": 2693 + }, + { + "epoch": 2.9523287671232876, + "grad_norm": 2.086970567703247, + "learning_rate": 1.1779041429731926e-07, + "loss": 2.0789, + "step": 2694 + }, + { + "epoch": 2.9534246575342467, + "grad_norm": 2.1364893913269043, + "learning_rate": 1.1575954508529651e-07, + "loss": 2.1075, + "step": 2695 + }, + { + "epoch": 2.9545205479452052, + "grad_norm": 2.1782732009887695, + "learning_rate": 1.1372867587327377e-07, + "loss": 2.0524, + "step": 2696 + }, + { + "epoch": 2.9556164383561643, + "grad_norm": 2.1796936988830566, + "learning_rate": 1.1169780666125102e-07, + "loss": 2.0833, + "step": 2697 + }, + { + "epoch": 2.9567123287671233, + "grad_norm": 2.145625591278076, + "learning_rate": 1.0966693744922827e-07, + "loss": 2.1162, + "step": 2698 + }, + { + "epoch": 2.9578082191780823, + "grad_norm": 2.2068347930908203, + "learning_rate": 1.0763606823720554e-07, + "loss": 2.0607, + "step": 2699 + }, + { + "epoch": 2.958904109589041, + "grad_norm": 2.2185962200164795, + "learning_rate": 1.0560519902518278e-07, + "loss": 2.0463, + "step": 2700 + }, + { + "epoch": 2.958904109589041, + "eval_loss": 2.2531802654266357, + "eval_runtime": 46.0392, + "eval_samples_per_second": 21.721, + "eval_steps_per_second": 0.912, + "step": 2700 + }, + { + "epoch": 2.96, + "grad_norm": 2.24737286567688, + "learning_rate": 1.0357432981316005e-07, + "loss": 2.0698, + "step": 2701 + }, + { + "epoch": 2.961095890410959, + "grad_norm": 2.153000593185425, + "learning_rate": 1.0154346060113728e-07, + "loss": 2.0896, + "step": 2702 + }, + { + "epoch": 2.962191780821918, + "grad_norm": 2.1966280937194824, + "learning_rate": 9.951259138911455e-08, + "loss": 2.1149, + "step": 2703 + }, + { + "epoch": 2.9632876712328766, + "grad_norm": 2.1699471473693848, + "learning_rate": 9.748172217709181e-08, + "loss": 2.0551, + "step": 2704 + }, + { + "epoch": 2.9643835616438357, + "grad_norm": 2.1531777381896973, + "learning_rate": 9.545085296506904e-08, + "loss": 2.0731, + "step": 2705 + }, + { + "epoch": 2.9654794520547947, + "grad_norm": 2.17754864692688, + "learning_rate": 9.341998375304631e-08, + "loss": 2.0601, + "step": 2706 + }, + { + "epoch": 2.9665753424657533, + "grad_norm": 2.2257227897644043, + "learning_rate": 9.138911454102357e-08, + "loss": 2.062, + "step": 2707 + }, + { + "epoch": 2.9676712328767123, + "grad_norm": 2.309396266937256, + "learning_rate": 8.935824532900083e-08, + "loss": 2.0665, + "step": 2708 + }, + { + "epoch": 2.9687671232876713, + "grad_norm": 2.2943942546844482, + "learning_rate": 8.732737611697807e-08, + "loss": 2.1113, + "step": 2709 + }, + { + "epoch": 2.96986301369863, + "grad_norm": 2.198591709136963, + "learning_rate": 8.529650690495533e-08, + "loss": 2.0774, + "step": 2710 + }, + { + "epoch": 2.96986301369863, + "eval_loss": 2.2531678676605225, + "eval_runtime": 44.0153, + "eval_samples_per_second": 22.719, + "eval_steps_per_second": 0.954, + "step": 2710 + }, + { + "epoch": 2.970958904109589, + "grad_norm": 2.1388938426971436, + "learning_rate": 8.326563769293258e-08, + "loss": 2.0122, + "step": 2711 + }, + { + "epoch": 2.972054794520548, + "grad_norm": 2.145585775375366, + "learning_rate": 8.123476848090983e-08, + "loss": 2.0754, + "step": 2712 + }, + { + "epoch": 2.9731506849315066, + "grad_norm": 2.2251172065734863, + "learning_rate": 7.920389926888709e-08, + "loss": 2.1059, + "step": 2713 + }, + { + "epoch": 2.9742465753424656, + "grad_norm": 2.233044147491455, + "learning_rate": 7.717303005686434e-08, + "loss": 2.0948, + "step": 2714 + }, + { + "epoch": 2.9753424657534246, + "grad_norm": 2.1500651836395264, + "learning_rate": 7.514216084484159e-08, + "loss": 2.084, + "step": 2715 + }, + { + "epoch": 2.9764383561643837, + "grad_norm": 2.1636741161346436, + "learning_rate": 7.311129163281885e-08, + "loss": 2.1244, + "step": 2716 + }, + { + "epoch": 2.9775342465753427, + "grad_norm": 2.275684356689453, + "learning_rate": 7.10804224207961e-08, + "loss": 2.0768, + "step": 2717 + }, + { + "epoch": 2.9786301369863013, + "grad_norm": 2.2161293029785156, + "learning_rate": 6.904955320877335e-08, + "loss": 2.1101, + "step": 2718 + }, + { + "epoch": 2.9797260273972603, + "grad_norm": 2.1446619033813477, + "learning_rate": 6.701868399675061e-08, + "loss": 2.0702, + "step": 2719 + }, + { + "epoch": 2.9808219178082194, + "grad_norm": 2.2171945571899414, + "learning_rate": 6.498781478472786e-08, + "loss": 2.0738, + "step": 2720 + }, + { + "epoch": 2.9808219178082194, + "eval_loss": 2.252962827682495, + "eval_runtime": 44.9074, + "eval_samples_per_second": 22.268, + "eval_steps_per_second": 0.935, + "step": 2720 + }, + { + "epoch": 2.981917808219178, + "grad_norm": 2.16827130317688, + "learning_rate": 6.295694557270513e-08, + "loss": 2.0624, + "step": 2721 + }, + { + "epoch": 2.983013698630137, + "grad_norm": 2.229531764984131, + "learning_rate": 6.092607636068238e-08, + "loss": 2.1472, + "step": 2722 + }, + { + "epoch": 2.984109589041096, + "grad_norm": 2.1417596340179443, + "learning_rate": 5.889520714865963e-08, + "loss": 2.0715, + "step": 2723 + }, + { + "epoch": 2.9852054794520546, + "grad_norm": 2.3001646995544434, + "learning_rate": 5.686433793663689e-08, + "loss": 2.0903, + "step": 2724 + }, + { + "epoch": 2.9863013698630136, + "grad_norm": 2.080298900604248, + "learning_rate": 5.4833468724614136e-08, + "loss": 2.0603, + "step": 2725 + }, + { + "epoch": 2.9873972602739727, + "grad_norm": 2.14170503616333, + "learning_rate": 5.280259951259139e-08, + "loss": 2.1339, + "step": 2726 + }, + { + "epoch": 2.9884931506849313, + "grad_norm": 2.1344692707061768, + "learning_rate": 5.077173030056864e-08, + "loss": 2.0696, + "step": 2727 + }, + { + "epoch": 2.9895890410958903, + "grad_norm": 2.270775556564331, + "learning_rate": 4.8740861088545904e-08, + "loss": 2.0305, + "step": 2728 + }, + { + "epoch": 2.9906849315068493, + "grad_norm": 2.2182064056396484, + "learning_rate": 4.6709991876523154e-08, + "loss": 2.0391, + "step": 2729 + }, + { + "epoch": 2.9917808219178084, + "grad_norm": 2.167832612991333, + "learning_rate": 4.4679122664500416e-08, + "loss": 2.1311, + "step": 2730 + }, + { + "epoch": 2.9917808219178084, + "eval_loss": 2.2530605792999268, + "eval_runtime": 47.8586, + "eval_samples_per_second": 20.895, + "eval_steps_per_second": 0.878, + "step": 2730 + }, + { + "epoch": 2.9928767123287674, + "grad_norm": 2.202209711074829, + "learning_rate": 4.2648253452477665e-08, + "loss": 2.1249, + "step": 2731 + }, + { + "epoch": 2.993972602739726, + "grad_norm": 2.139840602874756, + "learning_rate": 4.0617384240454915e-08, + "loss": 2.0717, + "step": 2732 + }, + { + "epoch": 2.995068493150685, + "grad_norm": 2.173750400543213, + "learning_rate": 3.858651502843217e-08, + "loss": 2.1009, + "step": 2733 + }, + { + "epoch": 2.996164383561644, + "grad_norm": 2.1986517906188965, + "learning_rate": 3.6555645816409427e-08, + "loss": 2.0892, + "step": 2734 + }, + { + "epoch": 2.9972602739726026, + "grad_norm": 2.1470165252685547, + "learning_rate": 3.4524776604386676e-08, + "loss": 2.0751, + "step": 2735 + }, + { + "epoch": 2.9983561643835617, + "grad_norm": 2.2216999530792236, + "learning_rate": 3.249390739236393e-08, + "loss": 2.1139, + "step": 2736 + } + ], + "logging_steps": 1, + "max_steps": 2736, + "num_input_tokens_seen": 0, + "num_train_epochs": 3, + "save_steps": 100, + "stateful_callbacks": { + "TrainerControl": { + "args": { + "should_epoch_stop": false, + "should_evaluate": false, + "should_log": false, + "should_save": true, + "should_training_stop": true + }, + "attributes": {} + } + }, + "total_flos": 8.528958544014213e+17, + "train_batch_size": 3, + "trial_name": null, + "trial_params": null +} diff --git a/babylm_shuffle_nondeterministic_10M_seed0/runs/checkpoint-400/latest b/babylm_shuffle_nondeterministic_10M_seed0/runs/checkpoint-400/latest new file mode 100644 index 0000000000000000000000000000000000000000..e5bdf58d4f29d34e909da25905fad376f73e7c29 --- /dev/null +++ b/babylm_shuffle_nondeterministic_10M_seed0/runs/checkpoint-400/latest @@ -0,0 +1 @@ +global_step400 \ No newline at end of file diff --git a/babylm_shuffle_nondeterministic_10M_seed0/runs/checkpoint-400/model.safetensors.index.json b/babylm_shuffle_nondeterministic_10M_seed0/runs/checkpoint-400/model.safetensors.index.json new file mode 100644 index 0000000000000000000000000000000000000000..ed64de846d720b9a7859dc20575fea8e8ca51940 --- /dev/null +++ b/babylm_shuffle_nondeterministic_10M_seed0/runs/checkpoint-400/model.safetensors.index.json @@ -0,0 +1,262 @@ +{ + "metadata": { + "total_size": 7213504512 + }, + "weight_map": { + "lm_head.weight": "model-00002-of-00002.safetensors", + "model.embed_tokens.weight": "model-00001-of-00002.safetensors", + "model.layers.0.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.0.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.0.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.0.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.0.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.0.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.0.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.0.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.0.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.1.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.1.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.1.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.1.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.1.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.1.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.1.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.1.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.1.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.10.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.10.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.10.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.10.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.10.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.10.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.10.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.10.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.10.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.11.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.11.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.11.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.11.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.11.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.11.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.11.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.11.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.11.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.12.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.12.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.12.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.12.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.12.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.12.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.12.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.12.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.12.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.13.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.13.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.13.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.13.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.13.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.13.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.13.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.13.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.13.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.14.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.14.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.14.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.14.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.14.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.14.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.14.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.14.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.14.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.15.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.15.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.15.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.15.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.15.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.15.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.15.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.15.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.15.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.16.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.16.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.16.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.16.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.16.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.16.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.16.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.16.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.16.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.17.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.17.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.17.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.17.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.17.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.17.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.17.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.17.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.17.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.18.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.18.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.18.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.18.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.18.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.18.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.18.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.18.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.18.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.19.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.19.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.19.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.19.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.19.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.19.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.19.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.19.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.19.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.2.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.2.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.2.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.2.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.2.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.2.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.2.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.2.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.2.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.20.input_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.20.mlp.down_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.20.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.20.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.20.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.20.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.20.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.20.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.20.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.21.input_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.21.mlp.down_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.21.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.21.mlp.up_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.21.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.21.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.21.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.21.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.21.self_attn.v_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.22.input_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.22.mlp.down_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.22.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.22.mlp.up_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.22.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.22.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.22.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.22.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.22.self_attn.v_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.23.input_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.23.mlp.down_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.23.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.23.mlp.up_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.23.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.23.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.23.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.23.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.23.self_attn.v_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.24.input_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.24.mlp.down_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.24.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.24.mlp.up_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.24.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.24.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.24.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.24.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.24.self_attn.v_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.25.input_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.25.mlp.down_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.25.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.25.mlp.up_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.25.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.25.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.25.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.25.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.25.self_attn.v_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.26.input_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.26.mlp.down_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.26.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.26.mlp.up_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.26.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.26.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.26.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.26.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.26.self_attn.v_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.27.input_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.27.mlp.down_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.27.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.27.mlp.up_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.27.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.27.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.27.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.27.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.27.self_attn.v_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.3.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.3.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.3.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.3.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.3.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.3.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.3.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.3.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.3.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.4.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.4.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.4.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.4.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.4.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.4.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.4.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.4.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.4.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.5.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.5.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.5.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.5.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.5.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.5.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.5.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.5.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.5.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.6.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.6.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.6.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.6.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.6.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.6.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.6.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.6.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.6.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.7.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.7.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.7.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.7.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.7.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.7.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.7.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.7.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.7.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.8.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.8.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.8.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.8.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.8.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.8.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.8.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.8.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.8.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.9.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.9.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.9.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.9.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.9.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.9.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.9.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.9.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.9.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.norm.weight": "model-00002-of-00002.safetensors" + } +} diff --git a/babylm_shuffle_nondeterministic_10M_seed0/runs/checkpoint-400/special_tokens_map.json b/babylm_shuffle_nondeterministic_10M_seed0/runs/checkpoint-400/special_tokens_map.json new file mode 100644 index 0000000000000000000000000000000000000000..6949c5975ee0e961ef61cf31010dce04df0a03f8 --- /dev/null +++ b/babylm_shuffle_nondeterministic_10M_seed0/runs/checkpoint-400/special_tokens_map.json @@ -0,0 +1,23 @@ +{ + "bos_token": { + "content": "<|begin_of_text|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false + }, + "eos_token": { + "content": "<|end_of_text|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false + }, + "pad_token": { + "content": "[PAD]", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false + } +} diff --git a/babylm_shuffle_nondeterministic_10M_seed0/runs/checkpoint-400/tokenizer_config.json b/babylm_shuffle_nondeterministic_10M_seed0/runs/checkpoint-400/tokenizer_config.json new file mode 100644 index 0000000000000000000000000000000000000000..5684b14e2756b56aa397c1324e239b442162b34a --- /dev/null +++ b/babylm_shuffle_nondeterministic_10M_seed0/runs/checkpoint-400/tokenizer_config.json @@ -0,0 +1,2070 @@ +{ + "added_tokens_decoder": { + "128000": { + "content": "<|begin_of_text|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128001": { + "content": "<|end_of_text|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128002": { + "content": "<|reserved_special_token_0|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128003": { + "content": "<|reserved_special_token_1|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128004": { + "content": "<|finetune_right_pad_id|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128005": { + "content": "<|reserved_special_token_2|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128006": { + "content": "<|start_header_id|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128007": { + "content": "<|end_header_id|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128008": { + "content": "<|eom_id|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128009": { + "content": "<|eot_id|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128010": { + "content": "<|python_tag|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128011": { + "content": "<|reserved_special_token_3|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128012": { + "content": "<|reserved_special_token_4|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128013": { + "content": "<|reserved_special_token_5|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128014": { + "content": "<|reserved_special_token_6|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128015": { + "content": "<|reserved_special_token_7|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128016": { + "content": "<|reserved_special_token_8|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128017": { + "content": "<|reserved_special_token_9|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128018": { + "content": "<|reserved_special_token_10|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128019": { + "content": "<|reserved_special_token_11|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128020": { + "content": "<|reserved_special_token_12|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128021": { + "content": "<|reserved_special_token_13|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128022": { + "content": "<|reserved_special_token_14|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128023": { + "content": "<|reserved_special_token_15|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128024": { + "content": "<|reserved_special_token_16|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128025": { + "content": "<|reserved_special_token_17|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128026": { + "content": "<|reserved_special_token_18|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128027": { + "content": "<|reserved_special_token_19|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128028": { + "content": "<|reserved_special_token_20|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128029": { + "content": "<|reserved_special_token_21|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128030": { + "content": "<|reserved_special_token_22|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128031": { + "content": "<|reserved_special_token_23|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128032": { + "content": "<|reserved_special_token_24|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128033": { + "content": "<|reserved_special_token_25|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128034": { + "content": "<|reserved_special_token_26|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128035": { + "content": "<|reserved_special_token_27|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128036": { + "content": "<|reserved_special_token_28|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128037": { + "content": "<|reserved_special_token_29|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128038": { + "content": "<|reserved_special_token_30|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128039": { + "content": "<|reserved_special_token_31|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128040": { + "content": "<|reserved_special_token_32|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128041": { + "content": "<|reserved_special_token_33|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128042": { + "content": "<|reserved_special_token_34|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128043": { + "content": "<|reserved_special_token_35|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128044": { + "content": "<|reserved_special_token_36|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128045": { + "content": "<|reserved_special_token_37|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128046": { + "content": "<|reserved_special_token_38|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128047": { + "content": "<|reserved_special_token_39|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128048": { + "content": "<|reserved_special_token_40|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128049": { + "content": "<|reserved_special_token_41|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128050": { + "content": "<|reserved_special_token_42|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128051": { + "content": "<|reserved_special_token_43|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128052": { + "content": "<|reserved_special_token_44|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128053": { + "content": "<|reserved_special_token_45|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128054": { + "content": "<|reserved_special_token_46|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128055": { + "content": "<|reserved_special_token_47|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128056": { + "content": "<|reserved_special_token_48|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128057": { + "content": "<|reserved_special_token_49|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128058": { + "content": "<|reserved_special_token_50|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128059": { + "content": "<|reserved_special_token_51|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128060": { + "content": "<|reserved_special_token_52|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128061": { + "content": "<|reserved_special_token_53|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128062": { + "content": "<|reserved_special_token_54|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128063": { + "content": "<|reserved_special_token_55|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128064": { + "content": "<|reserved_special_token_56|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128065": { + "content": "<|reserved_special_token_57|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128066": { + "content": "<|reserved_special_token_58|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128067": { + "content": "<|reserved_special_token_59|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128068": { + "content": "<|reserved_special_token_60|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128069": { + "content": "<|reserved_special_token_61|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128070": { + "content": "<|reserved_special_token_62|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128071": { + "content": "<|reserved_special_token_63|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128072": { + "content": "<|reserved_special_token_64|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128073": { + "content": "<|reserved_special_token_65|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128074": { + "content": "<|reserved_special_token_66|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128075": { + "content": "<|reserved_special_token_67|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128076": { + "content": "<|reserved_special_token_68|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128077": { + "content": "<|reserved_special_token_69|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128078": { + "content": "<|reserved_special_token_70|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128079": { + "content": "<|reserved_special_token_71|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128080": { + "content": "<|reserved_special_token_72|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128081": { + "content": "<|reserved_special_token_73|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128082": { + "content": "<|reserved_special_token_74|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128083": { + "content": "<|reserved_special_token_75|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128084": { + "content": "<|reserved_special_token_76|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128085": { + "content": "<|reserved_special_token_77|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128086": { + "content": "<|reserved_special_token_78|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128087": { + "content": "<|reserved_special_token_79|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128088": { + "content": "<|reserved_special_token_80|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128089": { + "content": "<|reserved_special_token_81|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128090": { + "content": "<|reserved_special_token_82|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128091": { + "content": "<|reserved_special_token_83|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128092": { + "content": "<|reserved_special_token_84|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128093": { + "content": "<|reserved_special_token_85|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128094": { + "content": "<|reserved_special_token_86|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128095": { + "content": "<|reserved_special_token_87|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128096": { + "content": "<|reserved_special_token_88|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128097": { + "content": "<|reserved_special_token_89|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128098": { + "content": "<|reserved_special_token_90|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128099": { + "content": "<|reserved_special_token_91|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128100": { + "content": "<|reserved_special_token_92|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128101": { + "content": "<|reserved_special_token_93|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128102": { + "content": "<|reserved_special_token_94|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128103": { + "content": "<|reserved_special_token_95|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128104": { + "content": "<|reserved_special_token_96|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128105": { + "content": "<|reserved_special_token_97|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128106": { + "content": "<|reserved_special_token_98|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128107": { + "content": "<|reserved_special_token_99|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128108": { + "content": "<|reserved_special_token_100|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128109": { + "content": "<|reserved_special_token_101|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128110": { + "content": "<|reserved_special_token_102|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128111": { + "content": "<|reserved_special_token_103|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128112": { + "content": "<|reserved_special_token_104|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128113": { + "content": "<|reserved_special_token_105|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128114": { + "content": "<|reserved_special_token_106|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128115": { + "content": "<|reserved_special_token_107|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128116": { + "content": "<|reserved_special_token_108|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128117": { + "content": "<|reserved_special_token_109|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128118": { + "content": "<|reserved_special_token_110|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128119": { + "content": "<|reserved_special_token_111|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128120": { + "content": "<|reserved_special_token_112|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128121": { + "content": "<|reserved_special_token_113|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128122": { + "content": "<|reserved_special_token_114|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128123": { + "content": "<|reserved_special_token_115|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128124": { + "content": "<|reserved_special_token_116|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128125": { + "content": "<|reserved_special_token_117|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128126": { + "content": "<|reserved_special_token_118|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128127": { + "content": "<|reserved_special_token_119|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128128": { + "content": "<|reserved_special_token_120|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128129": { + "content": "<|reserved_special_token_121|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128130": { + "content": "<|reserved_special_token_122|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128131": { + "content": "<|reserved_special_token_123|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128132": { + "content": "<|reserved_special_token_124|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128133": { + "content": "<|reserved_special_token_125|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128134": { + "content": "<|reserved_special_token_126|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128135": { + "content": "<|reserved_special_token_127|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128136": { + "content": "<|reserved_special_token_128|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128137": { + "content": "<|reserved_special_token_129|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128138": { + "content": "<|reserved_special_token_130|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128139": { + "content": "<|reserved_special_token_131|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128140": { + "content": "<|reserved_special_token_132|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128141": { + "content": "<|reserved_special_token_133|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128142": { + "content": "<|reserved_special_token_134|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128143": { + "content": "<|reserved_special_token_135|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128144": { + "content": "<|reserved_special_token_136|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128145": { + "content": "<|reserved_special_token_137|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128146": { + "content": "<|reserved_special_token_138|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128147": { + "content": "<|reserved_special_token_139|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128148": { + "content": "<|reserved_special_token_140|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128149": { + "content": "<|reserved_special_token_141|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128150": { + "content": "<|reserved_special_token_142|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128151": { + "content": "<|reserved_special_token_143|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128152": { + "content": "<|reserved_special_token_144|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128153": { + "content": "<|reserved_special_token_145|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128154": { + "content": "<|reserved_special_token_146|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128155": { + "content": "<|reserved_special_token_147|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128156": { + "content": "<|reserved_special_token_148|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128157": { + "content": "<|reserved_special_token_149|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128158": { + "content": "<|reserved_special_token_150|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128159": { + "content": "<|reserved_special_token_151|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128160": { + "content": "<|reserved_special_token_152|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128161": { + "content": "<|reserved_special_token_153|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128162": { + "content": "<|reserved_special_token_154|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128163": { + "content": "<|reserved_special_token_155|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128164": { + "content": "<|reserved_special_token_156|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128165": { + "content": "<|reserved_special_token_157|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128166": { + "content": "<|reserved_special_token_158|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128167": { + "content": "<|reserved_special_token_159|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128168": { + "content": "<|reserved_special_token_160|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128169": { + "content": "<|reserved_special_token_161|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128170": { + "content": "<|reserved_special_token_162|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128171": { + "content": "<|reserved_special_token_163|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128172": { + "content": "<|reserved_special_token_164|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128173": { + "content": "<|reserved_special_token_165|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128174": { + "content": "<|reserved_special_token_166|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128175": { + "content": "<|reserved_special_token_167|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128176": { + "content": "<|reserved_special_token_168|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128177": { + "content": "<|reserved_special_token_169|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128178": { + "content": "<|reserved_special_token_170|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128179": { + "content": "<|reserved_special_token_171|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128180": { + "content": "<|reserved_special_token_172|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128181": { + "content": "<|reserved_special_token_173|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128182": { + "content": "<|reserved_special_token_174|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128183": { + "content": "<|reserved_special_token_175|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128184": { + "content": "<|reserved_special_token_176|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128185": { + "content": "<|reserved_special_token_177|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128186": { + "content": "<|reserved_special_token_178|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128187": { + "content": "<|reserved_special_token_179|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128188": { + "content": "<|reserved_special_token_180|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128189": { + "content": "<|reserved_special_token_181|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128190": { + "content": "<|reserved_special_token_182|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128191": { + "content": "<|reserved_special_token_183|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128192": { + "content": "<|reserved_special_token_184|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128193": { + "content": "<|reserved_special_token_185|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128194": { + "content": "<|reserved_special_token_186|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128195": { + "content": "<|reserved_special_token_187|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128196": { + "content": "<|reserved_special_token_188|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128197": { + "content": "<|reserved_special_token_189|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128198": { + "content": "<|reserved_special_token_190|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128199": { + "content": "<|reserved_special_token_191|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128200": { + "content": "<|reserved_special_token_192|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128201": { + "content": "<|reserved_special_token_193|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128202": { + "content": "<|reserved_special_token_194|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128203": { + "content": "<|reserved_special_token_195|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128204": { + "content": "<|reserved_special_token_196|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128205": { + "content": "<|reserved_special_token_197|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128206": { + "content": "<|reserved_special_token_198|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128207": { + "content": "<|reserved_special_token_199|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128208": { + "content": "<|reserved_special_token_200|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128209": { + "content": "<|reserved_special_token_201|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128210": { + "content": "<|reserved_special_token_202|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128211": { + "content": "<|reserved_special_token_203|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128212": { + "content": "<|reserved_special_token_204|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128213": { + "content": "<|reserved_special_token_205|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128214": { + "content": "<|reserved_special_token_206|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128215": { + "content": "<|reserved_special_token_207|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128216": { + "content": "<|reserved_special_token_208|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128217": { + "content": "<|reserved_special_token_209|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128218": { + "content": "<|reserved_special_token_210|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128219": { + "content": "<|reserved_special_token_211|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128220": { + "content": "<|reserved_special_token_212|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128221": { + "content": "<|reserved_special_token_213|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128222": { + "content": "<|reserved_special_token_214|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128223": { + "content": "<|reserved_special_token_215|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128224": { + "content": "<|reserved_special_token_216|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128225": { + "content": "<|reserved_special_token_217|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128226": { + "content": "<|reserved_special_token_218|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128227": { + "content": "<|reserved_special_token_219|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128228": { + "content": "<|reserved_special_token_220|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128229": { + "content": "<|reserved_special_token_221|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128230": { + "content": "<|reserved_special_token_222|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128231": { + "content": "<|reserved_special_token_223|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128232": { + "content": "<|reserved_special_token_224|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128233": { + "content": "<|reserved_special_token_225|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128234": { + "content": "<|reserved_special_token_226|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128235": { + "content": "<|reserved_special_token_227|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128236": { + "content": "<|reserved_special_token_228|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128237": { + "content": "<|reserved_special_token_229|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128238": { + "content": "<|reserved_special_token_230|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128239": { + "content": "<|reserved_special_token_231|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128240": { + "content": "<|reserved_special_token_232|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128241": { + "content": "<|reserved_special_token_233|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128242": { + "content": "<|reserved_special_token_234|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128243": { + "content": "<|reserved_special_token_235|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128244": { + "content": "<|reserved_special_token_236|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128245": { + "content": "<|reserved_special_token_237|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128246": { + "content": "<|reserved_special_token_238|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128247": { + "content": "<|reserved_special_token_239|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128248": { + "content": "<|reserved_special_token_240|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128249": { + "content": "<|reserved_special_token_241|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128250": { + "content": "<|reserved_special_token_242|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128251": { + "content": "<|reserved_special_token_243|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128252": { + "content": "<|reserved_special_token_244|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128253": { + "content": "<|reserved_special_token_245|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128254": { + "content": "<|reserved_special_token_246|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128255": { + "content": "<|reserved_special_token_247|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128256": { + "content": "[PAD]", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + } + }, + "bos_token": "<|begin_of_text|>", + "clean_up_tokenization_spaces": true, + "eos_token": "<|end_of_text|>", + "model_input_names": [ + "input_ids", + "attention_mask" + ], + "model_max_length": 131072, + "pad_token": "[PAD]", + "tokenizer_class": "PreTrainedTokenizerFast" +} diff --git a/babylm_shuffle_nondeterministic_10M_seed0/runs/checkpoint-400/trainer_state.json b/babylm_shuffle_nondeterministic_10M_seed0/runs/checkpoint-400/trainer_state.json new file mode 100644 index 0000000000000000000000000000000000000000..8804fc9d59dfd48e0bc7912cb54f3bcc66e03f4b --- /dev/null +++ b/babylm_shuffle_nondeterministic_10M_seed0/runs/checkpoint-400/trainer_state.json @@ -0,0 +1,3153 @@ +{ + "best_metric": null, + "best_model_checkpoint": null, + "epoch": 0.4383561643835616, + "eval_steps": 10, + "global_step": 400, + "is_hyper_param_search": false, + "is_local_process_zero": true, + "is_world_process_zero": true, + "log_history": [ + { + "epoch": 0.001095890410958904, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.0998, + "step": 1 + }, + { + "epoch": 0.002191780821917808, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.0858, + "step": 2 + }, + { + "epoch": 0.0032876712328767125, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.0672, + "step": 3 + }, + { + "epoch": 0.004383561643835616, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.0802, + "step": 4 + }, + { + "epoch": 0.005479452054794521, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.0636, + "step": 5 + }, + { + "epoch": 0.006575342465753425, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.1672, + "step": 6 + }, + { + "epoch": 0.007671232876712329, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.0961, + "step": 7 + }, + { + "epoch": 0.008767123287671232, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.1258, + "step": 8 + }, + { + "epoch": 0.009863013698630137, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.0953, + "step": 9 + }, + { + "epoch": 0.010958904109589041, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.1404, + "step": 10 + }, + { + "epoch": 0.010958904109589041, + "eval_loss": 3.12593412399292, + "eval_runtime": 42.732, + "eval_samples_per_second": 23.402, + "eval_steps_per_second": 0.983, + "step": 10 + }, + { + "epoch": 0.012054794520547946, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.1187, + "step": 11 + }, + { + "epoch": 0.01315068493150685, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.0593, + "step": 12 + }, + { + "epoch": 0.014246575342465753, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.111, + "step": 13 + }, + { + "epoch": 0.015342465753424657, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.0974, + "step": 14 + }, + { + "epoch": 0.01643835616438356, + "grad_norm": 5.991453170776367, + "learning_rate": 1.8248175182481753e-08, + "loss": 3.135, + "step": 15 + }, + { + "epoch": 0.017534246575342465, + "grad_norm": 6.274210453033447, + "learning_rate": 3.6496350364963505e-08, + "loss": 3.083, + "step": 16 + }, + { + "epoch": 0.01863013698630137, + "grad_norm": 6.644786834716797, + "learning_rate": 5.4744525547445264e-08, + "loss": 3.0856, + "step": 17 + }, + { + "epoch": 0.019726027397260273, + "grad_norm": 6.769156455993652, + "learning_rate": 7.299270072992701e-08, + "loss": 3.1082, + "step": 18 + }, + { + "epoch": 0.020821917808219178, + "grad_norm": 6.771246910095215, + "learning_rate": 9.124087591240876e-08, + "loss": 3.0753, + "step": 19 + }, + { + "epoch": 0.021917808219178082, + "grad_norm": 6.468441486358643, + "learning_rate": 1.0948905109489053e-07, + "loss": 3.1478, + "step": 20 + }, + { + "epoch": 0.021917808219178082, + "eval_loss": 3.1245107650756836, + "eval_runtime": 43.6251, + "eval_samples_per_second": 22.923, + "eval_steps_per_second": 0.963, + "step": 20 + }, + { + "epoch": 0.023013698630136987, + "grad_norm": 6.83078670501709, + "learning_rate": 1.2773722627737227e-07, + "loss": 3.1213, + "step": 21 + }, + { + "epoch": 0.02410958904109589, + "grad_norm": 6.221287250518799, + "learning_rate": 1.4598540145985402e-07, + "loss": 3.0549, + "step": 22 + }, + { + "epoch": 0.025205479452054796, + "grad_norm": 5.564266204833984, + "learning_rate": 1.642335766423358e-07, + "loss": 3.0769, + "step": 23 + }, + { + "epoch": 0.0263013698630137, + "grad_norm": 5.735209941864014, + "learning_rate": 1.8248175182481753e-07, + "loss": 3.1124, + "step": 24 + }, + { + "epoch": 0.0273972602739726, + "grad_norm": 5.577925682067871, + "learning_rate": 2.0072992700729928e-07, + "loss": 3.1191, + "step": 25 + }, + { + "epoch": 0.028493150684931506, + "grad_norm": 4.814139366149902, + "learning_rate": 2.1897810218978106e-07, + "loss": 3.0947, + "step": 26 + }, + { + "epoch": 0.02958904109589041, + "grad_norm": 4.69775390625, + "learning_rate": 2.3722627737226278e-07, + "loss": 3.0997, + "step": 27 + }, + { + "epoch": 0.030684931506849315, + "grad_norm": 4.811586380004883, + "learning_rate": 2.5547445255474454e-07, + "loss": 3.0749, + "step": 28 + }, + { + "epoch": 0.03178082191780822, + "grad_norm": 4.623377323150635, + "learning_rate": 2.737226277372263e-07, + "loss": 3.0619, + "step": 29 + }, + { + "epoch": 0.03287671232876712, + "grad_norm": 4.721253395080566, + "learning_rate": 2.9197080291970804e-07, + "loss": 3.0371, + "step": 30 + }, + { + "epoch": 0.03287671232876712, + "eval_loss": 3.082857847213745, + "eval_runtime": 44.127, + "eval_samples_per_second": 22.662, + "eval_steps_per_second": 0.952, + "step": 30 + }, + { + "epoch": 0.03397260273972603, + "grad_norm": 3.968630790710449, + "learning_rate": 3.102189781021898e-07, + "loss": 3.0535, + "step": 31 + }, + { + "epoch": 0.03506849315068493, + "grad_norm": 4.4373779296875, + "learning_rate": 3.284671532846716e-07, + "loss": 3.0476, + "step": 32 + }, + { + "epoch": 0.03616438356164384, + "grad_norm": 4.479320049285889, + "learning_rate": 3.467153284671533e-07, + "loss": 3.0215, + "step": 33 + }, + { + "epoch": 0.03726027397260274, + "grad_norm": 4.122231960296631, + "learning_rate": 3.6496350364963505e-07, + "loss": 3.0106, + "step": 34 + }, + { + "epoch": 0.038356164383561646, + "grad_norm": 4.153995513916016, + "learning_rate": 3.8321167883211686e-07, + "loss": 3.0318, + "step": 35 + }, + { + "epoch": 0.03945205479452055, + "grad_norm": 3.6949076652526855, + "learning_rate": 4.0145985401459856e-07, + "loss": 3.046, + "step": 36 + }, + { + "epoch": 0.040547945205479455, + "grad_norm": 3.4610488414764404, + "learning_rate": 4.197080291970803e-07, + "loss": 3.002, + "step": 37 + }, + { + "epoch": 0.041643835616438356, + "grad_norm": 3.9523138999938965, + "learning_rate": 4.379562043795621e-07, + "loss": 2.9665, + "step": 38 + }, + { + "epoch": 0.042739726027397264, + "grad_norm": 3.68516206741333, + "learning_rate": 4.562043795620438e-07, + "loss": 3.0411, + "step": 39 + }, + { + "epoch": 0.043835616438356165, + "grad_norm": 3.692396402359009, + "learning_rate": 4.7445255474452557e-07, + "loss": 2.9666, + "step": 40 + }, + { + "epoch": 0.043835616438356165, + "eval_loss": 3.015512704849243, + "eval_runtime": 48.2438, + "eval_samples_per_second": 20.728, + "eval_steps_per_second": 0.871, + "step": 40 + }, + { + "epoch": 0.044931506849315066, + "grad_norm": 3.3765921592712402, + "learning_rate": 4.927007299270073e-07, + "loss": 2.9831, + "step": 41 + }, + { + "epoch": 0.046027397260273974, + "grad_norm": 2.9631125926971436, + "learning_rate": 5.109489051094891e-07, + "loss": 3.0126, + "step": 42 + }, + { + "epoch": 0.047123287671232875, + "grad_norm": 2.969730854034424, + "learning_rate": 5.291970802919708e-07, + "loss": 2.9803, + "step": 43 + }, + { + "epoch": 0.04821917808219178, + "grad_norm": 3.185237407684326, + "learning_rate": 5.474452554744526e-07, + "loss": 2.9888, + "step": 44 + }, + { + "epoch": 0.049315068493150684, + "grad_norm": 3.055579423904419, + "learning_rate": 5.656934306569343e-07, + "loss": 2.9715, + "step": 45 + }, + { + "epoch": 0.05041095890410959, + "grad_norm": 3.083284616470337, + "learning_rate": 5.839416058394161e-07, + "loss": 2.9293, + "step": 46 + }, + { + "epoch": 0.05150684931506849, + "grad_norm": 2.984781503677368, + "learning_rate": 6.021897810218979e-07, + "loss": 2.9228, + "step": 47 + }, + { + "epoch": 0.0526027397260274, + "grad_norm": 2.756173849105835, + "learning_rate": 6.204379562043796e-07, + "loss": 2.9406, + "step": 48 + }, + { + "epoch": 0.0536986301369863, + "grad_norm": 2.911132574081421, + "learning_rate": 6.386861313868614e-07, + "loss": 2.8829, + "step": 49 + }, + { + "epoch": 0.0547945205479452, + "grad_norm": 2.592890977859497, + "learning_rate": 6.569343065693432e-07, + "loss": 2.9839, + "step": 50 + }, + { + "epoch": 0.0547945205479452, + "eval_loss": 2.9409468173980713, + "eval_runtime": 47.1045, + "eval_samples_per_second": 21.229, + "eval_steps_per_second": 0.892, + "step": 50 + }, + { + "epoch": 0.05589041095890411, + "grad_norm": 3.0112955570220947, + "learning_rate": 6.751824817518248e-07, + "loss": 2.8847, + "step": 51 + }, + { + "epoch": 0.05698630136986301, + "grad_norm": 2.3701798915863037, + "learning_rate": 6.934306569343066e-07, + "loss": 2.9109, + "step": 52 + }, + { + "epoch": 0.05808219178082192, + "grad_norm": 2.268850088119507, + "learning_rate": 7.116788321167883e-07, + "loss": 2.8483, + "step": 53 + }, + { + "epoch": 0.05917808219178082, + "grad_norm": 2.419015645980835, + "learning_rate": 7.299270072992701e-07, + "loss": 2.8491, + "step": 54 + }, + { + "epoch": 0.06027397260273973, + "grad_norm": 2.4582653045654297, + "learning_rate": 7.48175182481752e-07, + "loss": 2.8622, + "step": 55 + }, + { + "epoch": 0.06136986301369863, + "grad_norm": 2.3471553325653076, + "learning_rate": 7.664233576642337e-07, + "loss": 2.8738, + "step": 56 + }, + { + "epoch": 0.06246575342465754, + "grad_norm": 2.479177236557007, + "learning_rate": 7.846715328467154e-07, + "loss": 2.8843, + "step": 57 + }, + { + "epoch": 0.06356164383561644, + "grad_norm": 2.1676292419433594, + "learning_rate": 8.029197080291971e-07, + "loss": 2.8674, + "step": 58 + }, + { + "epoch": 0.06465753424657535, + "grad_norm": 1.9862910509109497, + "learning_rate": 8.211678832116789e-07, + "loss": 2.8197, + "step": 59 + }, + { + "epoch": 0.06575342465753424, + "grad_norm": 2.211313486099243, + "learning_rate": 8.394160583941606e-07, + "loss": 2.7961, + "step": 60 + }, + { + "epoch": 0.06575342465753424, + "eval_loss": 2.880824327468872, + "eval_runtime": 51.3235, + "eval_samples_per_second": 19.484, + "eval_steps_per_second": 0.818, + "step": 60 + }, + { + "epoch": 0.06684931506849315, + "grad_norm": 2.3497281074523926, + "learning_rate": 8.576642335766425e-07, + "loss": 2.8233, + "step": 61 + }, + { + "epoch": 0.06794520547945206, + "grad_norm": 2.158515214920044, + "learning_rate": 8.759124087591242e-07, + "loss": 2.8136, + "step": 62 + }, + { + "epoch": 0.06904109589041096, + "grad_norm": 2.0219955444335938, + "learning_rate": 8.94160583941606e-07, + "loss": 2.8231, + "step": 63 + }, + { + "epoch": 0.07013698630136986, + "grad_norm": 1.8989133834838867, + "learning_rate": 9.124087591240876e-07, + "loss": 2.8268, + "step": 64 + }, + { + "epoch": 0.07123287671232877, + "grad_norm": 2.243851661682129, + "learning_rate": 9.306569343065694e-07, + "loss": 2.7886, + "step": 65 + }, + { + "epoch": 0.07232876712328767, + "grad_norm": 2.122464418411255, + "learning_rate": 9.489051094890511e-07, + "loss": 2.8166, + "step": 66 + }, + { + "epoch": 0.07342465753424658, + "grad_norm": 1.899857997894287, + "learning_rate": 9.67153284671533e-07, + "loss": 2.7997, + "step": 67 + }, + { + "epoch": 0.07452054794520548, + "grad_norm": 2.4424173831939697, + "learning_rate": 9.854014598540146e-07, + "loss": 2.8342, + "step": 68 + }, + { + "epoch": 0.07561643835616438, + "grad_norm": 2.5677411556243896, + "learning_rate": 1.0036496350364965e-06, + "loss": 2.7786, + "step": 69 + }, + { + "epoch": 0.07671232876712329, + "grad_norm": 1.9989523887634277, + "learning_rate": 1.0218978102189781e-06, + "loss": 2.7411, + "step": 70 + }, + { + "epoch": 0.07671232876712329, + "eval_loss": 2.8185322284698486, + "eval_runtime": 50.4998, + "eval_samples_per_second": 19.802, + "eval_steps_per_second": 0.832, + "step": 70 + }, + { + "epoch": 0.07780821917808219, + "grad_norm": 1.9376521110534668, + "learning_rate": 1.04014598540146e-06, + "loss": 2.7988, + "step": 71 + }, + { + "epoch": 0.0789041095890411, + "grad_norm": 2.5629725456237793, + "learning_rate": 1.0583941605839416e-06, + "loss": 2.7079, + "step": 72 + }, + { + "epoch": 0.08, + "grad_norm": 2.1634678840637207, + "learning_rate": 1.0766423357664233e-06, + "loss": 2.7049, + "step": 73 + }, + { + "epoch": 0.08109589041095891, + "grad_norm": 2.2143008708953857, + "learning_rate": 1.0948905109489052e-06, + "loss": 2.72, + "step": 74 + }, + { + "epoch": 0.0821917808219178, + "grad_norm": 2.2701101303100586, + "learning_rate": 1.113138686131387e-06, + "loss": 2.7418, + "step": 75 + }, + { + "epoch": 0.08328767123287671, + "grad_norm": 1.716569423675537, + "learning_rate": 1.1313868613138687e-06, + "loss": 2.7558, + "step": 76 + }, + { + "epoch": 0.08438356164383562, + "grad_norm": 2.79757022857666, + "learning_rate": 1.1496350364963505e-06, + "loss": 2.6865, + "step": 77 + }, + { + "epoch": 0.08547945205479453, + "grad_norm": 2.2189838886260986, + "learning_rate": 1.1678832116788322e-06, + "loss": 2.6684, + "step": 78 + }, + { + "epoch": 0.08657534246575342, + "grad_norm": 2.0695579051971436, + "learning_rate": 1.1861313868613138e-06, + "loss": 2.6779, + "step": 79 + }, + { + "epoch": 0.08767123287671233, + "grad_norm": 2.372326374053955, + "learning_rate": 1.2043795620437959e-06, + "loss": 2.6711, + "step": 80 + }, + { + "epoch": 0.08767123287671233, + "eval_loss": 2.7537453174591064, + "eval_runtime": 47.7274, + "eval_samples_per_second": 20.952, + "eval_steps_per_second": 0.88, + "step": 80 + }, + { + "epoch": 0.08876712328767124, + "grad_norm": 2.3398947715759277, + "learning_rate": 1.2226277372262775e-06, + "loss": 2.6776, + "step": 81 + }, + { + "epoch": 0.08986301369863013, + "grad_norm": 2.6619036197662354, + "learning_rate": 1.2408759124087592e-06, + "loss": 2.7077, + "step": 82 + }, + { + "epoch": 0.09095890410958904, + "grad_norm": 1.8245115280151367, + "learning_rate": 1.259124087591241e-06, + "loss": 2.662, + "step": 83 + }, + { + "epoch": 0.09205479452054795, + "grad_norm": 2.2627148628234863, + "learning_rate": 1.2773722627737229e-06, + "loss": 2.677, + "step": 84 + }, + { + "epoch": 0.09315068493150686, + "grad_norm": 2.4983835220336914, + "learning_rate": 1.2956204379562045e-06, + "loss": 2.6601, + "step": 85 + }, + { + "epoch": 0.09424657534246575, + "grad_norm": 2.147010087966919, + "learning_rate": 1.3138686131386864e-06, + "loss": 2.6213, + "step": 86 + }, + { + "epoch": 0.09534246575342466, + "grad_norm": 2.791682481765747, + "learning_rate": 1.332116788321168e-06, + "loss": 2.6281, + "step": 87 + }, + { + "epoch": 0.09643835616438357, + "grad_norm": 2.023970365524292, + "learning_rate": 1.3503649635036497e-06, + "loss": 2.5766, + "step": 88 + }, + { + "epoch": 0.09753424657534246, + "grad_norm": 2.8143956661224365, + "learning_rate": 1.3686131386861315e-06, + "loss": 2.6086, + "step": 89 + }, + { + "epoch": 0.09863013698630137, + "grad_norm": 2.226106882095337, + "learning_rate": 1.3868613138686132e-06, + "loss": 2.5477, + "step": 90 + }, + { + "epoch": 0.09863013698630137, + "eval_loss": 2.6880786418914795, + "eval_runtime": 47.8417, + "eval_samples_per_second": 20.902, + "eval_steps_per_second": 0.878, + "step": 90 + }, + { + "epoch": 0.09972602739726028, + "grad_norm": 2.5248148441314697, + "learning_rate": 1.405109489051095e-06, + "loss": 2.6232, + "step": 91 + }, + { + "epoch": 0.10082191780821918, + "grad_norm": 2.421330690383911, + "learning_rate": 1.4233576642335767e-06, + "loss": 2.599, + "step": 92 + }, + { + "epoch": 0.10191780821917808, + "grad_norm": 2.5562429428100586, + "learning_rate": 1.4416058394160583e-06, + "loss": 2.6197, + "step": 93 + }, + { + "epoch": 0.10301369863013699, + "grad_norm": 2.856201171875, + "learning_rate": 1.4598540145985402e-06, + "loss": 2.5825, + "step": 94 + }, + { + "epoch": 0.10410958904109589, + "grad_norm": 2.6576807498931885, + "learning_rate": 1.478102189781022e-06, + "loss": 2.5496, + "step": 95 + }, + { + "epoch": 0.1052054794520548, + "grad_norm": 3.1408190727233887, + "learning_rate": 1.496350364963504e-06, + "loss": 2.6142, + "step": 96 + }, + { + "epoch": 0.1063013698630137, + "grad_norm": 2.3270952701568604, + "learning_rate": 1.5145985401459856e-06, + "loss": 2.584, + "step": 97 + }, + { + "epoch": 0.1073972602739726, + "grad_norm": 2.3976614475250244, + "learning_rate": 1.5328467153284674e-06, + "loss": 2.5941, + "step": 98 + }, + { + "epoch": 0.10849315068493151, + "grad_norm": 2.783485174179077, + "learning_rate": 1.551094890510949e-06, + "loss": 2.5626, + "step": 99 + }, + { + "epoch": 0.1095890410958904, + "grad_norm": 2.2522788047790527, + "learning_rate": 1.5693430656934307e-06, + "loss": 2.517, + "step": 100 + }, + { + "epoch": 0.1095890410958904, + "eval_loss": 2.637315034866333, + "eval_runtime": 48.2107, + "eval_samples_per_second": 20.742, + "eval_steps_per_second": 0.871, + "step": 100 + }, + { + "epoch": 0.11068493150684931, + "grad_norm": 2.917799949645996, + "learning_rate": 1.5875912408759126e-06, + "loss": 2.5159, + "step": 101 + }, + { + "epoch": 0.11178082191780822, + "grad_norm": 2.955815076828003, + "learning_rate": 1.6058394160583942e-06, + "loss": 2.563, + "step": 102 + }, + { + "epoch": 0.11287671232876713, + "grad_norm": 2.8096001148223877, + "learning_rate": 1.624087591240876e-06, + "loss": 2.5995, + "step": 103 + }, + { + "epoch": 0.11397260273972602, + "grad_norm": 2.4967427253723145, + "learning_rate": 1.6423357664233577e-06, + "loss": 2.5336, + "step": 104 + }, + { + "epoch": 0.11506849315068493, + "grad_norm": 2.740717887878418, + "learning_rate": 1.6605839416058394e-06, + "loss": 2.4778, + "step": 105 + }, + { + "epoch": 0.11616438356164384, + "grad_norm": 2.5064632892608643, + "learning_rate": 1.6788321167883212e-06, + "loss": 2.5214, + "step": 106 + }, + { + "epoch": 0.11726027397260275, + "grad_norm": 2.2923007011413574, + "learning_rate": 1.6970802919708029e-06, + "loss": 2.5315, + "step": 107 + }, + { + "epoch": 0.11835616438356164, + "grad_norm": 3.339569568634033, + "learning_rate": 1.715328467153285e-06, + "loss": 2.4661, + "step": 108 + }, + { + "epoch": 0.11945205479452055, + "grad_norm": 2.1650891304016113, + "learning_rate": 1.7335766423357666e-06, + "loss": 2.5072, + "step": 109 + }, + { + "epoch": 0.12054794520547946, + "grad_norm": 3.1216914653778076, + "learning_rate": 1.7518248175182485e-06, + "loss": 2.5164, + "step": 110 + }, + { + "epoch": 0.12054794520547946, + "eval_loss": 2.594250202178955, + "eval_runtime": 41.8939, + "eval_samples_per_second": 23.87, + "eval_steps_per_second": 1.003, + "step": 110 + }, + { + "epoch": 0.12164383561643835, + "grad_norm": 2.191063404083252, + "learning_rate": 1.77007299270073e-06, + "loss": 2.5422, + "step": 111 + }, + { + "epoch": 0.12273972602739726, + "grad_norm": 2.922083854675293, + "learning_rate": 1.788321167883212e-06, + "loss": 2.5258, + "step": 112 + }, + { + "epoch": 0.12383561643835617, + "grad_norm": 2.518002986907959, + "learning_rate": 1.8065693430656936e-06, + "loss": 2.5588, + "step": 113 + }, + { + "epoch": 0.12493150684931507, + "grad_norm": 2.515194892883301, + "learning_rate": 1.8248175182481753e-06, + "loss": 2.5079, + "step": 114 + }, + { + "epoch": 0.12602739726027398, + "grad_norm": 2.5530874729156494, + "learning_rate": 1.8430656934306571e-06, + "loss": 2.4615, + "step": 115 + }, + { + "epoch": 0.12712328767123288, + "grad_norm": 2.3550198078155518, + "learning_rate": 1.8613138686131388e-06, + "loss": 2.4825, + "step": 116 + }, + { + "epoch": 0.12821917808219177, + "grad_norm": 2.295145034790039, + "learning_rate": 1.8795620437956206e-06, + "loss": 2.4749, + "step": 117 + }, + { + "epoch": 0.1293150684931507, + "grad_norm": 2.0158982276916504, + "learning_rate": 1.8978102189781023e-06, + "loss": 2.4756, + "step": 118 + }, + { + "epoch": 0.1304109589041096, + "grad_norm": 2.213057518005371, + "learning_rate": 1.916058394160584e-06, + "loss": 2.4744, + "step": 119 + }, + { + "epoch": 0.13150684931506848, + "grad_norm": 2.186222553253174, + "learning_rate": 1.934306569343066e-06, + "loss": 2.4957, + "step": 120 + }, + { + "epoch": 0.13150684931506848, + "eval_loss": 2.563931465148926, + "eval_runtime": 44.8714, + "eval_samples_per_second": 22.286, + "eval_steps_per_second": 0.936, + "step": 120 + }, + { + "epoch": 0.1326027397260274, + "grad_norm": 2.3143310546875, + "learning_rate": 1.952554744525548e-06, + "loss": 2.4, + "step": 121 + }, + { + "epoch": 0.1336986301369863, + "grad_norm": 2.0223965644836426, + "learning_rate": 1.9708029197080293e-06, + "loss": 2.4453, + "step": 122 + }, + { + "epoch": 0.13479452054794522, + "grad_norm": 2.2025411128997803, + "learning_rate": 1.989051094890511e-06, + "loss": 2.4463, + "step": 123 + }, + { + "epoch": 0.1358904109589041, + "grad_norm": 2.707930564880371, + "learning_rate": 2.007299270072993e-06, + "loss": 2.4589, + "step": 124 + }, + { + "epoch": 0.136986301369863, + "grad_norm": 2.146662473678589, + "learning_rate": 2.025547445255475e-06, + "loss": 2.5023, + "step": 125 + }, + { + "epoch": 0.13808219178082193, + "grad_norm": 2.520660161972046, + "learning_rate": 2.0437956204379563e-06, + "loss": 2.4464, + "step": 126 + }, + { + "epoch": 0.13917808219178082, + "grad_norm": 2.50632381439209, + "learning_rate": 2.062043795620438e-06, + "loss": 2.4556, + "step": 127 + }, + { + "epoch": 0.14027397260273972, + "grad_norm": 2.2919397354125977, + "learning_rate": 2.08029197080292e-06, + "loss": 2.4479, + "step": 128 + }, + { + "epoch": 0.14136986301369864, + "grad_norm": 2.521475315093994, + "learning_rate": 2.0985401459854014e-06, + "loss": 2.4982, + "step": 129 + }, + { + "epoch": 0.14246575342465753, + "grad_norm": 2.3812668323516846, + "learning_rate": 2.1167883211678833e-06, + "loss": 2.4163, + "step": 130 + }, + { + "epoch": 0.14246575342465753, + "eval_loss": 2.5390918254852295, + "eval_runtime": 43.8746, + "eval_samples_per_second": 22.792, + "eval_steps_per_second": 0.957, + "step": 130 + }, + { + "epoch": 0.14356164383561643, + "grad_norm": 2.302286148071289, + "learning_rate": 2.135036496350365e-06, + "loss": 2.4588, + "step": 131 + }, + { + "epoch": 0.14465753424657535, + "grad_norm": 2.4222044944763184, + "learning_rate": 2.1532846715328466e-06, + "loss": 2.492, + "step": 132 + }, + { + "epoch": 0.14575342465753424, + "grad_norm": 2.605835199356079, + "learning_rate": 2.171532846715329e-06, + "loss": 2.4055, + "step": 133 + }, + { + "epoch": 0.14684931506849316, + "grad_norm": 3.6261861324310303, + "learning_rate": 2.1897810218978103e-06, + "loss": 2.4221, + "step": 134 + }, + { + "epoch": 0.14794520547945206, + "grad_norm": 2.7647221088409424, + "learning_rate": 2.208029197080292e-06, + "loss": 2.4549, + "step": 135 + }, + { + "epoch": 0.14904109589041095, + "grad_norm": 3.2977371215820312, + "learning_rate": 2.226277372262774e-06, + "loss": 2.4331, + "step": 136 + }, + { + "epoch": 0.15013698630136987, + "grad_norm": 3.147094488143921, + "learning_rate": 2.244525547445256e-06, + "loss": 2.4268, + "step": 137 + }, + { + "epoch": 0.15123287671232877, + "grad_norm": 2.6363413333892822, + "learning_rate": 2.2627737226277373e-06, + "loss": 2.4524, + "step": 138 + }, + { + "epoch": 0.15232876712328766, + "grad_norm": 2.9288904666900635, + "learning_rate": 2.281021897810219e-06, + "loss": 2.4534, + "step": 139 + }, + { + "epoch": 0.15342465753424658, + "grad_norm": 2.5604870319366455, + "learning_rate": 2.299270072992701e-06, + "loss": 2.4138, + "step": 140 + }, + { + "epoch": 0.15342465753424658, + "eval_loss": 2.524790048599243, + "eval_runtime": 43.8851, + "eval_samples_per_second": 22.787, + "eval_steps_per_second": 0.957, + "step": 140 + }, + { + "epoch": 0.15452054794520548, + "grad_norm": 3.2746663093566895, + "learning_rate": 2.3175182481751825e-06, + "loss": 2.4094, + "step": 141 + }, + { + "epoch": 0.15561643835616437, + "grad_norm": 2.691483497619629, + "learning_rate": 2.3357664233576643e-06, + "loss": 2.4265, + "step": 142 + }, + { + "epoch": 0.1567123287671233, + "grad_norm": 2.5850117206573486, + "learning_rate": 2.354014598540146e-06, + "loss": 2.3922, + "step": 143 + }, + { + "epoch": 0.1578082191780822, + "grad_norm": 2.68204927444458, + "learning_rate": 2.3722627737226276e-06, + "loss": 2.4262, + "step": 144 + }, + { + "epoch": 0.1589041095890411, + "grad_norm": 2.952584981918335, + "learning_rate": 2.39051094890511e-06, + "loss": 2.4267, + "step": 145 + }, + { + "epoch": 0.16, + "grad_norm": 3.2354788780212402, + "learning_rate": 2.4087591240875918e-06, + "loss": 2.3836, + "step": 146 + }, + { + "epoch": 0.1610958904109589, + "grad_norm": 2.634521007537842, + "learning_rate": 2.427007299270073e-06, + "loss": 2.3953, + "step": 147 + }, + { + "epoch": 0.16219178082191782, + "grad_norm": 3.1115682125091553, + "learning_rate": 2.445255474452555e-06, + "loss": 2.409, + "step": 148 + }, + { + "epoch": 0.1632876712328767, + "grad_norm": 2.6824135780334473, + "learning_rate": 2.463503649635037e-06, + "loss": 2.4459, + "step": 149 + }, + { + "epoch": 0.1643835616438356, + "grad_norm": 2.617614507675171, + "learning_rate": 2.4817518248175183e-06, + "loss": 2.366, + "step": 150 + }, + { + "epoch": 0.1643835616438356, + "eval_loss": 2.5088160037994385, + "eval_runtime": 44.0888, + "eval_samples_per_second": 22.682, + "eval_steps_per_second": 0.953, + "step": 150 + }, + { + "epoch": 0.16547945205479453, + "grad_norm": 3.976992130279541, + "learning_rate": 2.5e-06, + "loss": 2.3517, + "step": 151 + }, + { + "epoch": 0.16657534246575342, + "grad_norm": 3.1433017253875732, + "learning_rate": 2.518248175182482e-06, + "loss": 2.4489, + "step": 152 + }, + { + "epoch": 0.16767123287671232, + "grad_norm": 4.100688934326172, + "learning_rate": 2.5364963503649635e-06, + "loss": 2.3683, + "step": 153 + }, + { + "epoch": 0.16876712328767124, + "grad_norm": 4.316102504730225, + "learning_rate": 2.5547445255474458e-06, + "loss": 2.3663, + "step": 154 + }, + { + "epoch": 0.16986301369863013, + "grad_norm": 2.3103132247924805, + "learning_rate": 2.5729927007299272e-06, + "loss": 2.4009, + "step": 155 + }, + { + "epoch": 0.17095890410958905, + "grad_norm": 3.576810836791992, + "learning_rate": 2.591240875912409e-06, + "loss": 2.3983, + "step": 156 + }, + { + "epoch": 0.17205479452054795, + "grad_norm": 2.337080955505371, + "learning_rate": 2.6094890510948905e-06, + "loss": 2.3849, + "step": 157 + }, + { + "epoch": 0.17315068493150684, + "grad_norm": 3.2523117065429688, + "learning_rate": 2.627737226277373e-06, + "loss": 2.4171, + "step": 158 + }, + { + "epoch": 0.17424657534246576, + "grad_norm": 2.529175281524658, + "learning_rate": 2.6459854014598542e-06, + "loss": 2.4303, + "step": 159 + }, + { + "epoch": 0.17534246575342466, + "grad_norm": 3.052212953567505, + "learning_rate": 2.664233576642336e-06, + "loss": 2.3895, + "step": 160 + }, + { + "epoch": 0.17534246575342466, + "eval_loss": 2.492485761642456, + "eval_runtime": 45.6738, + "eval_samples_per_second": 21.894, + "eval_steps_per_second": 0.92, + "step": 160 + }, + { + "epoch": 0.17643835616438355, + "grad_norm": 2.1544973850250244, + "learning_rate": 2.6824817518248175e-06, + "loss": 2.3649, + "step": 161 + }, + { + "epoch": 0.17753424657534247, + "grad_norm": 2.7351951599121094, + "learning_rate": 2.7007299270072994e-06, + "loss": 2.3908, + "step": 162 + }, + { + "epoch": 0.17863013698630137, + "grad_norm": 2.8431951999664307, + "learning_rate": 2.718978102189781e-06, + "loss": 2.374, + "step": 163 + }, + { + "epoch": 0.17972602739726026, + "grad_norm": 3.083190441131592, + "learning_rate": 2.737226277372263e-06, + "loss": 2.4193, + "step": 164 + }, + { + "epoch": 0.18082191780821918, + "grad_norm": 2.6523995399475098, + "learning_rate": 2.755474452554745e-06, + "loss": 2.3969, + "step": 165 + }, + { + "epoch": 0.18191780821917808, + "grad_norm": 2.626047372817993, + "learning_rate": 2.7737226277372264e-06, + "loss": 2.3742, + "step": 166 + }, + { + "epoch": 0.18301369863013697, + "grad_norm": 3.300917863845825, + "learning_rate": 2.7919708029197082e-06, + "loss": 2.4546, + "step": 167 + }, + { + "epoch": 0.1841095890410959, + "grad_norm": 2.797304391860962, + "learning_rate": 2.81021897810219e-06, + "loss": 2.4161, + "step": 168 + }, + { + "epoch": 0.1852054794520548, + "grad_norm": 2.777117967605591, + "learning_rate": 2.828467153284672e-06, + "loss": 2.3572, + "step": 169 + }, + { + "epoch": 0.1863013698630137, + "grad_norm": 2.887470006942749, + "learning_rate": 2.8467153284671534e-06, + "loss": 2.3887, + "step": 170 + }, + { + "epoch": 0.1863013698630137, + "eval_loss": 2.4854257106781006, + "eval_runtime": 44.5644, + "eval_samples_per_second": 22.439, + "eval_steps_per_second": 0.942, + "step": 170 + }, + { + "epoch": 0.1873972602739726, + "grad_norm": 3.0706076622009277, + "learning_rate": 2.8649635036496353e-06, + "loss": 2.3162, + "step": 171 + }, + { + "epoch": 0.1884931506849315, + "grad_norm": 3.6129744052886963, + "learning_rate": 2.8832116788321167e-06, + "loss": 2.3685, + "step": 172 + }, + { + "epoch": 0.18958904109589042, + "grad_norm": 2.4936370849609375, + "learning_rate": 2.901459854014599e-06, + "loss": 2.3297, + "step": 173 + }, + { + "epoch": 0.19068493150684931, + "grad_norm": 2.1041860580444336, + "learning_rate": 2.9197080291970804e-06, + "loss": 2.4085, + "step": 174 + }, + { + "epoch": 0.1917808219178082, + "grad_norm": 2.2066831588745117, + "learning_rate": 2.9379562043795623e-06, + "loss": 2.3905, + "step": 175 + }, + { + "epoch": 0.19287671232876713, + "grad_norm": 2.444652795791626, + "learning_rate": 2.956204379562044e-06, + "loss": 2.378, + "step": 176 + }, + { + "epoch": 0.19397260273972602, + "grad_norm": 2.501939058303833, + "learning_rate": 2.9744525547445256e-06, + "loss": 2.4136, + "step": 177 + }, + { + "epoch": 0.19506849315068492, + "grad_norm": 2.2167556285858154, + "learning_rate": 2.992700729927008e-06, + "loss": 2.385, + "step": 178 + }, + { + "epoch": 0.19616438356164384, + "grad_norm": 2.5298545360565186, + "learning_rate": 3.0109489051094893e-06, + "loss": 2.3508, + "step": 179 + }, + { + "epoch": 0.19726027397260273, + "grad_norm": 2.3774642944335938, + "learning_rate": 3.029197080291971e-06, + "loss": 2.3421, + "step": 180 + }, + { + "epoch": 0.19726027397260273, + "eval_loss": 2.476067304611206, + "eval_runtime": 44.3761, + "eval_samples_per_second": 22.535, + "eval_steps_per_second": 0.946, + "step": 180 + }, + { + "epoch": 0.19835616438356166, + "grad_norm": 2.9108712673187256, + "learning_rate": 3.0474452554744526e-06, + "loss": 2.3983, + "step": 181 + }, + { + "epoch": 0.19945205479452055, + "grad_norm": 2.5232033729553223, + "learning_rate": 3.065693430656935e-06, + "loss": 2.3946, + "step": 182 + }, + { + "epoch": 0.20054794520547944, + "grad_norm": 2.158168077468872, + "learning_rate": 3.0839416058394163e-06, + "loss": 2.3805, + "step": 183 + }, + { + "epoch": 0.20164383561643837, + "grad_norm": 2.3405354022979736, + "learning_rate": 3.102189781021898e-06, + "loss": 2.3974, + "step": 184 + }, + { + "epoch": 0.20273972602739726, + "grad_norm": 2.5308327674865723, + "learning_rate": 3.1204379562043796e-06, + "loss": 2.384, + "step": 185 + }, + { + "epoch": 0.20383561643835615, + "grad_norm": 2.6328237056732178, + "learning_rate": 3.1386861313868614e-06, + "loss": 2.3345, + "step": 186 + }, + { + "epoch": 0.20493150684931508, + "grad_norm": 3.0725512504577637, + "learning_rate": 3.156934306569343e-06, + "loss": 2.3191, + "step": 187 + }, + { + "epoch": 0.20602739726027397, + "grad_norm": 2.196218729019165, + "learning_rate": 3.175182481751825e-06, + "loss": 2.4073, + "step": 188 + }, + { + "epoch": 0.20712328767123286, + "grad_norm": 2.7722432613372803, + "learning_rate": 3.193430656934307e-06, + "loss": 2.3775, + "step": 189 + }, + { + "epoch": 0.20821917808219179, + "grad_norm": 2.4848241806030273, + "learning_rate": 3.2116788321167884e-06, + "loss": 2.3538, + "step": 190 + }, + { + "epoch": 0.20821917808219179, + "eval_loss": 2.462618112564087, + "eval_runtime": 43.9461, + "eval_samples_per_second": 22.755, + "eval_steps_per_second": 0.956, + "step": 190 + }, + { + "epoch": 0.20931506849315068, + "grad_norm": 2.4682741165161133, + "learning_rate": 3.2299270072992707e-06, + "loss": 2.3688, + "step": 191 + }, + { + "epoch": 0.2104109589041096, + "grad_norm": 2.518923282623291, + "learning_rate": 3.248175182481752e-06, + "loss": 2.3556, + "step": 192 + }, + { + "epoch": 0.2115068493150685, + "grad_norm": 2.4023561477661133, + "learning_rate": 3.266423357664234e-06, + "loss": 2.292, + "step": 193 + }, + { + "epoch": 0.2126027397260274, + "grad_norm": 2.5196943283081055, + "learning_rate": 3.2846715328467155e-06, + "loss": 2.3703, + "step": 194 + }, + { + "epoch": 0.2136986301369863, + "grad_norm": 2.762709617614746, + "learning_rate": 3.3029197080291973e-06, + "loss": 2.3578, + "step": 195 + }, + { + "epoch": 0.2147945205479452, + "grad_norm": 2.1287691593170166, + "learning_rate": 3.3211678832116788e-06, + "loss": 2.3687, + "step": 196 + }, + { + "epoch": 0.2158904109589041, + "grad_norm": 2.477341651916504, + "learning_rate": 3.339416058394161e-06, + "loss": 2.3545, + "step": 197 + }, + { + "epoch": 0.21698630136986302, + "grad_norm": 2.23933744430542, + "learning_rate": 3.3576642335766425e-06, + "loss": 2.3511, + "step": 198 + }, + { + "epoch": 0.21808219178082192, + "grad_norm": 2.2440133094787598, + "learning_rate": 3.3759124087591243e-06, + "loss": 2.3539, + "step": 199 + }, + { + "epoch": 0.2191780821917808, + "grad_norm": 2.5603177547454834, + "learning_rate": 3.3941605839416058e-06, + "loss": 2.36, + "step": 200 + }, + { + "epoch": 0.2191780821917808, + "eval_loss": 2.4507246017456055, + "eval_runtime": 43.7786, + "eval_samples_per_second": 22.842, + "eval_steps_per_second": 0.959, + "step": 200 + }, + { + "epoch": 0.22027397260273973, + "grad_norm": 2.4570510387420654, + "learning_rate": 3.412408759124088e-06, + "loss": 2.3702, + "step": 201 + }, + { + "epoch": 0.22136986301369863, + "grad_norm": 2.1975133419036865, + "learning_rate": 3.43065693430657e-06, + "loss": 2.3667, + "step": 202 + }, + { + "epoch": 0.22246575342465755, + "grad_norm": 2.710186004638672, + "learning_rate": 3.4489051094890513e-06, + "loss": 2.3159, + "step": 203 + }, + { + "epoch": 0.22356164383561644, + "grad_norm": 2.7799506187438965, + "learning_rate": 3.467153284671533e-06, + "loss": 2.3519, + "step": 204 + }, + { + "epoch": 0.22465753424657534, + "grad_norm": 2.2479400634765625, + "learning_rate": 3.4854014598540146e-06, + "loss": 2.3103, + "step": 205 + }, + { + "epoch": 0.22575342465753426, + "grad_norm": 3.1538405418395996, + "learning_rate": 3.503649635036497e-06, + "loss": 2.3506, + "step": 206 + }, + { + "epoch": 0.22684931506849315, + "grad_norm": 2.7607762813568115, + "learning_rate": 3.5218978102189783e-06, + "loss": 2.3386, + "step": 207 + }, + { + "epoch": 0.22794520547945205, + "grad_norm": 2.371105909347534, + "learning_rate": 3.54014598540146e-06, + "loss": 2.3583, + "step": 208 + }, + { + "epoch": 0.22904109589041097, + "grad_norm": 2.793933391571045, + "learning_rate": 3.5583941605839416e-06, + "loss": 2.3782, + "step": 209 + }, + { + "epoch": 0.23013698630136986, + "grad_norm": 2.372732400894165, + "learning_rate": 3.576642335766424e-06, + "loss": 2.417, + "step": 210 + }, + { + "epoch": 0.23013698630136986, + "eval_loss": 2.451474189758301, + "eval_runtime": 41.7197, + "eval_samples_per_second": 23.97, + "eval_steps_per_second": 1.007, + "step": 210 + }, + { + "epoch": 0.23123287671232876, + "grad_norm": 2.6989448070526123, + "learning_rate": 3.5948905109489054e-06, + "loss": 2.393, + "step": 211 + }, + { + "epoch": 0.23232876712328768, + "grad_norm": 2.4324285984039307, + "learning_rate": 3.6131386861313872e-06, + "loss": 2.2738, + "step": 212 + }, + { + "epoch": 0.23342465753424657, + "grad_norm": 3.5063016414642334, + "learning_rate": 3.6313868613138687e-06, + "loss": 2.2823, + "step": 213 + }, + { + "epoch": 0.2345205479452055, + "grad_norm": 3.953951358795166, + "learning_rate": 3.6496350364963505e-06, + "loss": 2.3132, + "step": 214 + }, + { + "epoch": 0.2356164383561644, + "grad_norm": 2.3483071327209473, + "learning_rate": 3.667883211678833e-06, + "loss": 2.3244, + "step": 215 + }, + { + "epoch": 0.23671232876712328, + "grad_norm": 2.884199380874634, + "learning_rate": 3.6861313868613142e-06, + "loss": 2.3067, + "step": 216 + }, + { + "epoch": 0.2378082191780822, + "grad_norm": 2.9514520168304443, + "learning_rate": 3.704379562043796e-06, + "loss": 2.304, + "step": 217 + }, + { + "epoch": 0.2389041095890411, + "grad_norm": 2.7530996799468994, + "learning_rate": 3.7226277372262775e-06, + "loss": 2.3691, + "step": 218 + }, + { + "epoch": 0.24, + "grad_norm": 2.8753647804260254, + "learning_rate": 3.7408759124087594e-06, + "loss": 2.3406, + "step": 219 + }, + { + "epoch": 0.2410958904109589, + "grad_norm": 2.670574903488159, + "learning_rate": 3.7591240875912412e-06, + "loss": 2.304, + "step": 220 + }, + { + "epoch": 0.2410958904109589, + "eval_loss": 2.4395809173583984, + "eval_runtime": 46.6723, + "eval_samples_per_second": 21.426, + "eval_steps_per_second": 0.9, + "step": 220 + }, + { + "epoch": 0.2421917808219178, + "grad_norm": 2.4152438640594482, + "learning_rate": 3.777372262773723e-06, + "loss": 2.3294, + "step": 221 + }, + { + "epoch": 0.2432876712328767, + "grad_norm": 2.8398540019989014, + "learning_rate": 3.7956204379562045e-06, + "loss": 2.3216, + "step": 222 + }, + { + "epoch": 0.24438356164383562, + "grad_norm": 2.6036765575408936, + "learning_rate": 3.8138686131386864e-06, + "loss": 2.3109, + "step": 223 + }, + { + "epoch": 0.24547945205479452, + "grad_norm": 2.7073473930358887, + "learning_rate": 3.832116788321168e-06, + "loss": 2.3457, + "step": 224 + }, + { + "epoch": 0.2465753424657534, + "grad_norm": 2.584404230117798, + "learning_rate": 3.85036496350365e-06, + "loss": 2.3159, + "step": 225 + }, + { + "epoch": 0.24767123287671233, + "grad_norm": 2.4220666885375977, + "learning_rate": 3.868613138686132e-06, + "loss": 2.2944, + "step": 226 + }, + { + "epoch": 0.24876712328767123, + "grad_norm": 2.5636494159698486, + "learning_rate": 3.886861313868613e-06, + "loss": 2.3219, + "step": 227 + }, + { + "epoch": 0.24986301369863015, + "grad_norm": 2.2990903854370117, + "learning_rate": 3.905109489051096e-06, + "loss": 2.3342, + "step": 228 + }, + { + "epoch": 0.25095890410958904, + "grad_norm": 2.50913143157959, + "learning_rate": 3.923357664233577e-06, + "loss": 2.3619, + "step": 229 + }, + { + "epoch": 0.25205479452054796, + "grad_norm": 2.1687769889831543, + "learning_rate": 3.9416058394160585e-06, + "loss": 2.3153, + "step": 230 + }, + { + "epoch": 0.25205479452054796, + "eval_loss": 2.4315240383148193, + "eval_runtime": 48.3542, + "eval_samples_per_second": 20.681, + "eval_steps_per_second": 0.869, + "step": 230 + }, + { + "epoch": 0.25315068493150683, + "grad_norm": 2.842125177383423, + "learning_rate": 3.95985401459854e-06, + "loss": 2.3644, + "step": 231 + }, + { + "epoch": 0.25424657534246575, + "grad_norm": 2.258049964904785, + "learning_rate": 3.978102189781022e-06, + "loss": 2.3727, + "step": 232 + }, + { + "epoch": 0.2553424657534247, + "grad_norm": 2.4630489349365234, + "learning_rate": 3.996350364963504e-06, + "loss": 2.267, + "step": 233 + }, + { + "epoch": 0.25643835616438354, + "grad_norm": 3.1469743251800537, + "learning_rate": 4.014598540145986e-06, + "loss": 2.3432, + "step": 234 + }, + { + "epoch": 0.25753424657534246, + "grad_norm": 2.6277477741241455, + "learning_rate": 4.032846715328467e-06, + "loss": 2.2545, + "step": 235 + }, + { + "epoch": 0.2586301369863014, + "grad_norm": 2.404141902923584, + "learning_rate": 4.05109489051095e-06, + "loss": 2.3109, + "step": 236 + }, + { + "epoch": 0.25972602739726025, + "grad_norm": 2.4545605182647705, + "learning_rate": 4.069343065693431e-06, + "loss": 2.3112, + "step": 237 + }, + { + "epoch": 0.2608219178082192, + "grad_norm": 2.237638235092163, + "learning_rate": 4.0875912408759126e-06, + "loss": 2.3, + "step": 238 + }, + { + "epoch": 0.2619178082191781, + "grad_norm": 2.192811965942383, + "learning_rate": 4.105839416058395e-06, + "loss": 2.3108, + "step": 239 + }, + { + "epoch": 0.26301369863013696, + "grad_norm": 2.5058720111846924, + "learning_rate": 4.124087591240876e-06, + "loss": 2.3635, + "step": 240 + }, + { + "epoch": 0.26301369863013696, + "eval_loss": 2.4219377040863037, + "eval_runtime": 45.3561, + "eval_samples_per_second": 22.048, + "eval_steps_per_second": 0.926, + "step": 240 + }, + { + "epoch": 0.2641095890410959, + "grad_norm": 2.0708935260772705, + "learning_rate": 4.1423357664233586e-06, + "loss": 2.3129, + "step": 241 + }, + { + "epoch": 0.2652054794520548, + "grad_norm": 2.6353116035461426, + "learning_rate": 4.16058394160584e-06, + "loss": 2.3011, + "step": 242 + }, + { + "epoch": 0.2663013698630137, + "grad_norm": 2.8392436504364014, + "learning_rate": 4.1788321167883214e-06, + "loss": 2.3643, + "step": 243 + }, + { + "epoch": 0.2673972602739726, + "grad_norm": 3.1773295402526855, + "learning_rate": 4.197080291970803e-06, + "loss": 2.2642, + "step": 244 + }, + { + "epoch": 0.2684931506849315, + "grad_norm": 2.4757354259490967, + "learning_rate": 4.215328467153285e-06, + "loss": 2.3211, + "step": 245 + }, + { + "epoch": 0.26958904109589044, + "grad_norm": 2.5799412727355957, + "learning_rate": 4.233576642335767e-06, + "loss": 2.2909, + "step": 246 + }, + { + "epoch": 0.2706849315068493, + "grad_norm": 2.368502378463745, + "learning_rate": 4.251824817518249e-06, + "loss": 2.2951, + "step": 247 + }, + { + "epoch": 0.2717808219178082, + "grad_norm": 2.114969253540039, + "learning_rate": 4.27007299270073e-06, + "loss": 2.3124, + "step": 248 + }, + { + "epoch": 0.27287671232876715, + "grad_norm": 3.056820869445801, + "learning_rate": 4.288321167883212e-06, + "loss": 2.2923, + "step": 249 + }, + { + "epoch": 0.273972602739726, + "grad_norm": 2.7752203941345215, + "learning_rate": 4.306569343065693e-06, + "loss": 2.2874, + "step": 250 + }, + { + "epoch": 0.273972602739726, + "eval_loss": 2.4229912757873535, + "eval_runtime": 52.5941, + "eval_samples_per_second": 19.014, + "eval_steps_per_second": 0.799, + "step": 250 + }, + { + "epoch": 0.27506849315068493, + "grad_norm": 3.045740842819214, + "learning_rate": 4.3248175182481755e-06, + "loss": 2.2686, + "step": 251 + }, + { + "epoch": 0.27616438356164386, + "grad_norm": 2.646402359008789, + "learning_rate": 4.343065693430658e-06, + "loss": 2.3247, + "step": 252 + }, + { + "epoch": 0.2772602739726027, + "grad_norm": 3.274994134902954, + "learning_rate": 4.361313868613139e-06, + "loss": 2.3093, + "step": 253 + }, + { + "epoch": 0.27835616438356164, + "grad_norm": 2.352246046066284, + "learning_rate": 4.379562043795621e-06, + "loss": 2.3124, + "step": 254 + }, + { + "epoch": 0.27945205479452057, + "grad_norm": 3.248717784881592, + "learning_rate": 4.397810218978102e-06, + "loss": 2.3417, + "step": 255 + }, + { + "epoch": 0.28054794520547943, + "grad_norm": 3.0919013023376465, + "learning_rate": 4.416058394160584e-06, + "loss": 2.3093, + "step": 256 + }, + { + "epoch": 0.28164383561643835, + "grad_norm": 2.1174702644348145, + "learning_rate": 4.434306569343066e-06, + "loss": 2.305, + "step": 257 + }, + { + "epoch": 0.2827397260273973, + "grad_norm": 2.913797378540039, + "learning_rate": 4.452554744525548e-06, + "loss": 2.2831, + "step": 258 + }, + { + "epoch": 0.28383561643835614, + "grad_norm": 2.693089723587036, + "learning_rate": 4.4708029197080295e-06, + "loss": 2.3206, + "step": 259 + }, + { + "epoch": 0.28493150684931506, + "grad_norm": 2.521411895751953, + "learning_rate": 4.489051094890512e-06, + "loss": 2.3196, + "step": 260 + }, + { + "epoch": 0.28493150684931506, + "eval_loss": 2.4178109169006348, + "eval_runtime": 50.6753, + "eval_samples_per_second": 19.733, + "eval_steps_per_second": 0.829, + "step": 260 + }, + { + "epoch": 0.286027397260274, + "grad_norm": 2.346346855163574, + "learning_rate": 4.507299270072993e-06, + "loss": 2.3433, + "step": 261 + }, + { + "epoch": 0.28712328767123285, + "grad_norm": 2.520368814468384, + "learning_rate": 4.525547445255475e-06, + "loss": 2.3174, + "step": 262 + }, + { + "epoch": 0.2882191780821918, + "grad_norm": 2.550347328186035, + "learning_rate": 4.543795620437956e-06, + "loss": 2.3072, + "step": 263 + }, + { + "epoch": 0.2893150684931507, + "grad_norm": 2.268662691116333, + "learning_rate": 4.562043795620438e-06, + "loss": 2.2763, + "step": 264 + }, + { + "epoch": 0.29041095890410956, + "grad_norm": 2.445798873901367, + "learning_rate": 4.580291970802921e-06, + "loss": 2.3395, + "step": 265 + }, + { + "epoch": 0.2915068493150685, + "grad_norm": 2.5391135215759277, + "learning_rate": 4.598540145985402e-06, + "loss": 2.3496, + "step": 266 + }, + { + "epoch": 0.2926027397260274, + "grad_norm": 2.3553097248077393, + "learning_rate": 4.6167883211678835e-06, + "loss": 2.3243, + "step": 267 + }, + { + "epoch": 0.2936986301369863, + "grad_norm": 2.448089599609375, + "learning_rate": 4.635036496350365e-06, + "loss": 2.2887, + "step": 268 + }, + { + "epoch": 0.2947945205479452, + "grad_norm": 2.1600170135498047, + "learning_rate": 4.653284671532847e-06, + "loss": 2.2816, + "step": 269 + }, + { + "epoch": 0.2958904109589041, + "grad_norm": 2.3783180713653564, + "learning_rate": 4.671532846715329e-06, + "loss": 2.3275, + "step": 270 + }, + { + "epoch": 0.2958904109589041, + "eval_loss": 2.413832902908325, + "eval_runtime": 48.7524, + "eval_samples_per_second": 20.512, + "eval_steps_per_second": 0.861, + "step": 270 + }, + { + "epoch": 0.29698630136986304, + "grad_norm": 2.468320369720459, + "learning_rate": 4.689781021897811e-06, + "loss": 2.3358, + "step": 271 + }, + { + "epoch": 0.2980821917808219, + "grad_norm": 2.401634454727173, + "learning_rate": 4.708029197080292e-06, + "loss": 2.2991, + "step": 272 + }, + { + "epoch": 0.2991780821917808, + "grad_norm": 2.4604203701019287, + "learning_rate": 4.726277372262774e-06, + "loss": 2.3496, + "step": 273 + }, + { + "epoch": 0.30027397260273975, + "grad_norm": 2.0338871479034424, + "learning_rate": 4.744525547445255e-06, + "loss": 2.3115, + "step": 274 + }, + { + "epoch": 0.3013698630136986, + "grad_norm": 2.5404436588287354, + "learning_rate": 4.7627737226277375e-06, + "loss": 2.2665, + "step": 275 + }, + { + "epoch": 0.30246575342465754, + "grad_norm": 2.5369458198547363, + "learning_rate": 4.78102189781022e-06, + "loss": 2.3072, + "step": 276 + }, + { + "epoch": 0.30356164383561646, + "grad_norm": 2.471522569656372, + "learning_rate": 4.799270072992701e-06, + "loss": 2.3113, + "step": 277 + }, + { + "epoch": 0.3046575342465753, + "grad_norm": 2.3890135288238525, + "learning_rate": 4.8175182481751835e-06, + "loss": 2.2838, + "step": 278 + }, + { + "epoch": 0.30575342465753425, + "grad_norm": 2.2485053539276123, + "learning_rate": 4.835766423357665e-06, + "loss": 2.3052, + "step": 279 + }, + { + "epoch": 0.30684931506849317, + "grad_norm": 2.1090753078460693, + "learning_rate": 4.854014598540146e-06, + "loss": 2.3054, + "step": 280 + }, + { + "epoch": 0.30684931506849317, + "eval_loss": 2.412548780441284, + "eval_runtime": 48.5034, + "eval_samples_per_second": 20.617, + "eval_steps_per_second": 0.866, + "step": 280 + }, + { + "epoch": 0.30794520547945203, + "grad_norm": 2.299133777618408, + "learning_rate": 4.872262773722628e-06, + "loss": 2.2599, + "step": 281 + }, + { + "epoch": 0.30904109589041096, + "grad_norm": 2.177905797958374, + "learning_rate": 4.89051094890511e-06, + "loss": 2.2745, + "step": 282 + }, + { + "epoch": 0.3101369863013699, + "grad_norm": 2.3178462982177734, + "learning_rate": 4.9087591240875915e-06, + "loss": 2.2562, + "step": 283 + }, + { + "epoch": 0.31123287671232874, + "grad_norm": 2.5088376998901367, + "learning_rate": 4.927007299270074e-06, + "loss": 2.2945, + "step": 284 + }, + { + "epoch": 0.31232876712328766, + "grad_norm": 2.095379590988159, + "learning_rate": 4.945255474452555e-06, + "loss": 2.3038, + "step": 285 + }, + { + "epoch": 0.3134246575342466, + "grad_norm": 2.1792995929718018, + "learning_rate": 4.963503649635037e-06, + "loss": 2.2702, + "step": 286 + }, + { + "epoch": 0.31452054794520545, + "grad_norm": 2.3001186847686768, + "learning_rate": 4.981751824817518e-06, + "loss": 2.3482, + "step": 287 + }, + { + "epoch": 0.3156164383561644, + "grad_norm": 2.1576106548309326, + "learning_rate": 5e-06, + "loss": 2.3391, + "step": 288 + }, + { + "epoch": 0.3167123287671233, + "grad_norm": 2.3715577125549316, + "learning_rate": 4.997969130787978e-06, + "loss": 2.3049, + "step": 289 + }, + { + "epoch": 0.3178082191780822, + "grad_norm": 2.0316715240478516, + "learning_rate": 4.995938261575955e-06, + "loss": 2.3111, + "step": 290 + }, + { + "epoch": 0.3178082191780822, + "eval_loss": 2.40299654006958, + "eval_runtime": 50.0873, + "eval_samples_per_second": 19.965, + "eval_steps_per_second": 0.839, + "step": 290 + }, + { + "epoch": 0.3189041095890411, + "grad_norm": 2.1135756969451904, + "learning_rate": 4.993907392363932e-06, + "loss": 2.2465, + "step": 291 + }, + { + "epoch": 0.32, + "grad_norm": 2.335655927658081, + "learning_rate": 4.9918765231519095e-06, + "loss": 2.31, + "step": 292 + }, + { + "epoch": 0.32109589041095893, + "grad_norm": 2.3255794048309326, + "learning_rate": 4.989845653939887e-06, + "loss": 2.2531, + "step": 293 + }, + { + "epoch": 0.3221917808219178, + "grad_norm": 2.111055612564087, + "learning_rate": 4.987814784727864e-06, + "loss": 2.2799, + "step": 294 + }, + { + "epoch": 0.3232876712328767, + "grad_norm": 2.1773171424865723, + "learning_rate": 4.985783915515841e-06, + "loss": 2.2884, + "step": 295 + }, + { + "epoch": 0.32438356164383564, + "grad_norm": 2.3559885025024414, + "learning_rate": 4.983753046303819e-06, + "loss": 2.2711, + "step": 296 + }, + { + "epoch": 0.3254794520547945, + "grad_norm": 2.120802879333496, + "learning_rate": 4.981722177091796e-06, + "loss": 2.2946, + "step": 297 + }, + { + "epoch": 0.3265753424657534, + "grad_norm": 2.2319388389587402, + "learning_rate": 4.979691307879773e-06, + "loss": 2.2771, + "step": 298 + }, + { + "epoch": 0.32767123287671235, + "grad_norm": 2.208740711212158, + "learning_rate": 4.9776604386677504e-06, + "loss": 2.3003, + "step": 299 + }, + { + "epoch": 0.3287671232876712, + "grad_norm": 2.543060541152954, + "learning_rate": 4.975629569455728e-06, + "loss": 2.2891, + "step": 300 + }, + { + "epoch": 0.3287671232876712, + "eval_loss": 2.398495674133301, + "eval_runtime": 51.1973, + "eval_samples_per_second": 19.532, + "eval_steps_per_second": 0.82, + "step": 300 + }, + { + "epoch": 0.32986301369863014, + "grad_norm": 2.086219072341919, + "learning_rate": 4.973598700243705e-06, + "loss": 2.3286, + "step": 301 + }, + { + "epoch": 0.33095890410958906, + "grad_norm": 2.3673319816589355, + "learning_rate": 4.971567831031681e-06, + "loss": 2.2672, + "step": 302 + }, + { + "epoch": 0.3320547945205479, + "grad_norm": 2.167593240737915, + "learning_rate": 4.9695369618196595e-06, + "loss": 2.264, + "step": 303 + }, + { + "epoch": 0.33315068493150685, + "grad_norm": 2.7218785285949707, + "learning_rate": 4.967506092607636e-06, + "loss": 2.1737, + "step": 304 + }, + { + "epoch": 0.33424657534246577, + "grad_norm": 2.673245429992676, + "learning_rate": 4.965475223395614e-06, + "loss": 2.2439, + "step": 305 + }, + { + "epoch": 0.33534246575342463, + "grad_norm": 2.0341989994049072, + "learning_rate": 4.9634443541835905e-06, + "loss": 2.2614, + "step": 306 + }, + { + "epoch": 0.33643835616438356, + "grad_norm": 1.9314247369766235, + "learning_rate": 4.961413484971569e-06, + "loss": 2.2592, + "step": 307 + }, + { + "epoch": 0.3375342465753425, + "grad_norm": 2.537794589996338, + "learning_rate": 4.959382615759545e-06, + "loss": 2.2859, + "step": 308 + }, + { + "epoch": 0.33863013698630134, + "grad_norm": 2.0811166763305664, + "learning_rate": 4.957351746547523e-06, + "loss": 2.2868, + "step": 309 + }, + { + "epoch": 0.33972602739726027, + "grad_norm": 2.783701181411743, + "learning_rate": 4.9553208773355e-06, + "loss": 2.2649, + "step": 310 + }, + { + "epoch": 0.33972602739726027, + "eval_loss": 2.400538921356201, + "eval_runtime": 43.185, + "eval_samples_per_second": 23.156, + "eval_steps_per_second": 0.973, + "step": 310 + }, + { + "epoch": 0.3408219178082192, + "grad_norm": 2.8541319370269775, + "learning_rate": 4.953290008123478e-06, + "loss": 2.3729, + "step": 311 + }, + { + "epoch": 0.3419178082191781, + "grad_norm": 2.1852617263793945, + "learning_rate": 4.951259138911454e-06, + "loss": 2.2885, + "step": 312 + }, + { + "epoch": 0.343013698630137, + "grad_norm": 2.872201919555664, + "learning_rate": 4.9492282696994315e-06, + "loss": 2.2353, + "step": 313 + }, + { + "epoch": 0.3441095890410959, + "grad_norm": 2.5140910148620605, + "learning_rate": 4.947197400487409e-06, + "loss": 2.3364, + "step": 314 + }, + { + "epoch": 0.3452054794520548, + "grad_norm": 2.619974374771118, + "learning_rate": 4.945166531275386e-06, + "loss": 2.2721, + "step": 315 + }, + { + "epoch": 0.3463013698630137, + "grad_norm": 2.4567618370056152, + "learning_rate": 4.943135662063363e-06, + "loss": 2.26, + "step": 316 + }, + { + "epoch": 0.3473972602739726, + "grad_norm": 2.9603192806243896, + "learning_rate": 4.9411047928513406e-06, + "loss": 2.2861, + "step": 317 + }, + { + "epoch": 0.34849315068493153, + "grad_norm": 2.115262985229492, + "learning_rate": 4.939073923639318e-06, + "loss": 2.3125, + "step": 318 + }, + { + "epoch": 0.3495890410958904, + "grad_norm": 2.693024158477783, + "learning_rate": 4.937043054427295e-06, + "loss": 2.3054, + "step": 319 + }, + { + "epoch": 0.3506849315068493, + "grad_norm": 2.3439996242523193, + "learning_rate": 4.935012185215272e-06, + "loss": 2.2576, + "step": 320 + }, + { + "epoch": 0.3506849315068493, + "eval_loss": 2.3920650482177734, + "eval_runtime": 47.7451, + "eval_samples_per_second": 20.945, + "eval_steps_per_second": 0.88, + "step": 320 + }, + { + "epoch": 0.35178082191780824, + "grad_norm": 2.3009634017944336, + "learning_rate": 4.93298131600325e-06, + "loss": 2.2454, + "step": 321 + }, + { + "epoch": 0.3528767123287671, + "grad_norm": 2.593203544616699, + "learning_rate": 4.930950446791227e-06, + "loss": 2.2965, + "step": 322 + }, + { + "epoch": 0.353972602739726, + "grad_norm": 2.9069650173187256, + "learning_rate": 4.928919577579204e-06, + "loss": 2.2785, + "step": 323 + }, + { + "epoch": 0.35506849315068495, + "grad_norm": 2.1021759510040283, + "learning_rate": 4.9268887083671815e-06, + "loss": 2.2857, + "step": 324 + }, + { + "epoch": 0.3561643835616438, + "grad_norm": 2.0330185890197754, + "learning_rate": 4.924857839155159e-06, + "loss": 2.2697, + "step": 325 + }, + { + "epoch": 0.35726027397260274, + "grad_norm": 2.1318063735961914, + "learning_rate": 4.922826969943136e-06, + "loss": 2.2698, + "step": 326 + }, + { + "epoch": 0.35835616438356166, + "grad_norm": 2.0157387256622314, + "learning_rate": 4.920796100731113e-06, + "loss": 2.292, + "step": 327 + }, + { + "epoch": 0.3594520547945205, + "grad_norm": 2.3266026973724365, + "learning_rate": 4.918765231519091e-06, + "loss": 2.2215, + "step": 328 + }, + { + "epoch": 0.36054794520547945, + "grad_norm": 2.066305160522461, + "learning_rate": 4.916734362307068e-06, + "loss": 2.2758, + "step": 329 + }, + { + "epoch": 0.36164383561643837, + "grad_norm": 2.0219993591308594, + "learning_rate": 4.914703493095045e-06, + "loss": 2.2381, + "step": 330 + }, + { + "epoch": 0.36164383561643837, + "eval_loss": 2.394064426422119, + "eval_runtime": 51.5482, + "eval_samples_per_second": 19.399, + "eval_steps_per_second": 0.815, + "step": 330 + }, + { + "epoch": 0.36273972602739724, + "grad_norm": 2.3964319229125977, + "learning_rate": 4.912672623883022e-06, + "loss": 2.2384, + "step": 331 + }, + { + "epoch": 0.36383561643835616, + "grad_norm": 2.5466530323028564, + "learning_rate": 4.910641754671e-06, + "loss": 2.2428, + "step": 332 + }, + { + "epoch": 0.3649315068493151, + "grad_norm": 2.236152410507202, + "learning_rate": 4.908610885458977e-06, + "loss": 2.2639, + "step": 333 + }, + { + "epoch": 0.36602739726027395, + "grad_norm": 2.7267303466796875, + "learning_rate": 4.906580016246954e-06, + "loss": 2.2608, + "step": 334 + }, + { + "epoch": 0.36712328767123287, + "grad_norm": 1.9403765201568604, + "learning_rate": 4.9045491470349315e-06, + "loss": 2.2962, + "step": 335 + }, + { + "epoch": 0.3682191780821918, + "grad_norm": 2.89070463180542, + "learning_rate": 4.902518277822909e-06, + "loss": 2.2847, + "step": 336 + }, + { + "epoch": 0.3693150684931507, + "grad_norm": 2.6427228450775146, + "learning_rate": 4.900487408610886e-06, + "loss": 2.2734, + "step": 337 + }, + { + "epoch": 0.3704109589041096, + "grad_norm": 2.1680362224578857, + "learning_rate": 4.898456539398863e-06, + "loss": 2.2869, + "step": 338 + }, + { + "epoch": 0.3715068493150685, + "grad_norm": 3.0160160064697266, + "learning_rate": 4.89642567018684e-06, + "loss": 2.3043, + "step": 339 + }, + { + "epoch": 0.3726027397260274, + "grad_norm": 1.9858418703079224, + "learning_rate": 4.894394800974818e-06, + "loss": 2.3422, + "step": 340 + }, + { + "epoch": 0.3726027397260274, + "eval_loss": 2.387559652328491, + "eval_runtime": 52.2455, + "eval_samples_per_second": 19.14, + "eval_steps_per_second": 0.804, + "step": 340 + }, + { + "epoch": 0.3736986301369863, + "grad_norm": 2.243699073791504, + "learning_rate": 4.892363931762794e-06, + "loss": 2.2394, + "step": 341 + }, + { + "epoch": 0.3747945205479452, + "grad_norm": 2.371299982070923, + "learning_rate": 4.8903330625507725e-06, + "loss": 2.2856, + "step": 342 + }, + { + "epoch": 0.37589041095890413, + "grad_norm": 2.3769736289978027, + "learning_rate": 4.888302193338749e-06, + "loss": 2.246, + "step": 343 + }, + { + "epoch": 0.376986301369863, + "grad_norm": 2.3209030628204346, + "learning_rate": 4.886271324126727e-06, + "loss": 2.2331, + "step": 344 + }, + { + "epoch": 0.3780821917808219, + "grad_norm": 2.747319221496582, + "learning_rate": 4.8842404549147034e-06, + "loss": 2.3115, + "step": 345 + }, + { + "epoch": 0.37917808219178084, + "grad_norm": 1.8599231243133545, + "learning_rate": 4.8822095857026816e-06, + "loss": 2.2994, + "step": 346 + }, + { + "epoch": 0.3802739726027397, + "grad_norm": 2.0358943939208984, + "learning_rate": 4.880178716490658e-06, + "loss": 2.271, + "step": 347 + }, + { + "epoch": 0.38136986301369863, + "grad_norm": 2.409106731414795, + "learning_rate": 4.878147847278636e-06, + "loss": 2.2281, + "step": 348 + }, + { + "epoch": 0.38246575342465755, + "grad_norm": 2.1641807556152344, + "learning_rate": 4.8761169780666125e-06, + "loss": 2.2437, + "step": 349 + }, + { + "epoch": 0.3835616438356164, + "grad_norm": 2.6917359828948975, + "learning_rate": 4.874086108854591e-06, + "loss": 2.226, + "step": 350 + }, + { + "epoch": 0.3835616438356164, + "eval_loss": 2.382991313934326, + "eval_runtime": 48.9534, + "eval_samples_per_second": 20.428, + "eval_steps_per_second": 0.858, + "step": 350 + }, + { + "epoch": 0.38465753424657534, + "grad_norm": 2.364215612411499, + "learning_rate": 4.872055239642567e-06, + "loss": 2.2319, + "step": 351 + }, + { + "epoch": 0.38575342465753426, + "grad_norm": 2.317643165588379, + "learning_rate": 4.870024370430544e-06, + "loss": 2.2096, + "step": 352 + }, + { + "epoch": 0.3868493150684931, + "grad_norm": 2.2283246517181396, + "learning_rate": 4.867993501218522e-06, + "loss": 2.2634, + "step": 353 + }, + { + "epoch": 0.38794520547945205, + "grad_norm": 2.498391628265381, + "learning_rate": 4.865962632006499e-06, + "loss": 2.2813, + "step": 354 + }, + { + "epoch": 0.38904109589041097, + "grad_norm": 2.3889403343200684, + "learning_rate": 4.863931762794476e-06, + "loss": 2.2361, + "step": 355 + }, + { + "epoch": 0.39013698630136984, + "grad_norm": 2.2049975395202637, + "learning_rate": 4.8619008935824535e-06, + "loss": 2.2583, + "step": 356 + }, + { + "epoch": 0.39123287671232876, + "grad_norm": 2.132840871810913, + "learning_rate": 4.859870024370431e-06, + "loss": 2.2823, + "step": 357 + }, + { + "epoch": 0.3923287671232877, + "grad_norm": 2.2692184448242188, + "learning_rate": 4.857839155158408e-06, + "loss": 2.2939, + "step": 358 + }, + { + "epoch": 0.3934246575342466, + "grad_norm": 2.0501933097839355, + "learning_rate": 4.855808285946385e-06, + "loss": 2.2976, + "step": 359 + }, + { + "epoch": 0.39452054794520547, + "grad_norm": 2.170236587524414, + "learning_rate": 4.8537774167343626e-06, + "loss": 2.2748, + "step": 360 + }, + { + "epoch": 0.39452054794520547, + "eval_loss": 2.3783886432647705, + "eval_runtime": 46.7324, + "eval_samples_per_second": 21.398, + "eval_steps_per_second": 0.899, + "step": 360 + }, + { + "epoch": 0.3956164383561644, + "grad_norm": 2.283137083053589, + "learning_rate": 4.85174654752234e-06, + "loss": 2.2705, + "step": 361 + }, + { + "epoch": 0.3967123287671233, + "grad_norm": 2.195984363555908, + "learning_rate": 4.849715678310317e-06, + "loss": 2.26, + "step": 362 + }, + { + "epoch": 0.3978082191780822, + "grad_norm": 2.399935007095337, + "learning_rate": 4.847684809098294e-06, + "loss": 2.2639, + "step": 363 + }, + { + "epoch": 0.3989041095890411, + "grad_norm": 2.411280393600464, + "learning_rate": 4.845653939886272e-06, + "loss": 2.2523, + "step": 364 + }, + { + "epoch": 0.4, + "grad_norm": 2.2033896446228027, + "learning_rate": 4.843623070674249e-06, + "loss": 2.2722, + "step": 365 + }, + { + "epoch": 0.4010958904109589, + "grad_norm": 2.2990238666534424, + "learning_rate": 4.841592201462226e-06, + "loss": 2.2411, + "step": 366 + }, + { + "epoch": 0.4021917808219178, + "grad_norm": 1.911453127861023, + "learning_rate": 4.8395613322502035e-06, + "loss": 2.2921, + "step": 367 + }, + { + "epoch": 0.40328767123287673, + "grad_norm": 2.453195095062256, + "learning_rate": 4.837530463038181e-06, + "loss": 2.218, + "step": 368 + }, + { + "epoch": 0.4043835616438356, + "grad_norm": 2.126992702484131, + "learning_rate": 4.835499593826158e-06, + "loss": 2.2628, + "step": 369 + }, + { + "epoch": 0.4054794520547945, + "grad_norm": 1.9305835962295532, + "learning_rate": 4.833468724614135e-06, + "loss": 2.2726, + "step": 370 + }, + { + "epoch": 0.4054794520547945, + "eval_loss": 2.3741114139556885, + "eval_runtime": 49.2956, + "eval_samples_per_second": 20.286, + "eval_steps_per_second": 0.852, + "step": 370 + }, + { + "epoch": 0.40657534246575344, + "grad_norm": 2.222024917602539, + "learning_rate": 4.831437855402113e-06, + "loss": 2.2634, + "step": 371 + }, + { + "epoch": 0.4076712328767123, + "grad_norm": 2.473958969116211, + "learning_rate": 4.82940698619009e-06, + "loss": 2.1817, + "step": 372 + }, + { + "epoch": 0.40876712328767123, + "grad_norm": 1.9208838939666748, + "learning_rate": 4.827376116978067e-06, + "loss": 2.2353, + "step": 373 + }, + { + "epoch": 0.40986301369863015, + "grad_norm": 2.7226450443267822, + "learning_rate": 4.8253452477660444e-06, + "loss": 2.2831, + "step": 374 + }, + { + "epoch": 0.410958904109589, + "grad_norm": 1.9078222513198853, + "learning_rate": 4.823314378554022e-06, + "loss": 2.264, + "step": 375 + }, + { + "epoch": 0.41205479452054794, + "grad_norm": 2.3946311473846436, + "learning_rate": 4.821283509341998e-06, + "loss": 2.251, + "step": 376 + }, + { + "epoch": 0.41315068493150686, + "grad_norm": 1.847343921661377, + "learning_rate": 4.819252640129976e-06, + "loss": 2.2441, + "step": 377 + }, + { + "epoch": 0.41424657534246573, + "grad_norm": 2.475385904312134, + "learning_rate": 4.817221770917953e-06, + "loss": 2.2639, + "step": 378 + }, + { + "epoch": 0.41534246575342465, + "grad_norm": 2.1396560668945312, + "learning_rate": 4.815190901705931e-06, + "loss": 2.2353, + "step": 379 + }, + { + "epoch": 0.41643835616438357, + "grad_norm": 2.3868885040283203, + "learning_rate": 4.813160032493907e-06, + "loss": 2.2551, + "step": 380 + }, + { + "epoch": 0.41643835616438357, + "eval_loss": 2.369168519973755, + "eval_runtime": 45.6131, + "eval_samples_per_second": 21.924, + "eval_steps_per_second": 0.921, + "step": 380 + }, + { + "epoch": 0.41753424657534244, + "grad_norm": 2.0193872451782227, + "learning_rate": 4.811129163281885e-06, + "loss": 2.2399, + "step": 381 + }, + { + "epoch": 0.41863013698630136, + "grad_norm": 2.4655935764312744, + "learning_rate": 4.809098294069862e-06, + "loss": 2.2729, + "step": 382 + }, + { + "epoch": 0.4197260273972603, + "grad_norm": 1.880974292755127, + "learning_rate": 4.80706742485784e-06, + "loss": 2.2681, + "step": 383 + }, + { + "epoch": 0.4208219178082192, + "grad_norm": 2.1912078857421875, + "learning_rate": 4.805036555645816e-06, + "loss": 2.2961, + "step": 384 + }, + { + "epoch": 0.42191780821917807, + "grad_norm": 1.985931634902954, + "learning_rate": 4.8030056864337945e-06, + "loss": 2.2407, + "step": 385 + }, + { + "epoch": 0.423013698630137, + "grad_norm": 2.4353220462799072, + "learning_rate": 4.800974817221771e-06, + "loss": 2.2867, + "step": 386 + }, + { + "epoch": 0.4241095890410959, + "grad_norm": 1.8810508251190186, + "learning_rate": 4.798943948009749e-06, + "loss": 2.2498, + "step": 387 + }, + { + "epoch": 0.4252054794520548, + "grad_norm": 2.195033073425293, + "learning_rate": 4.7969130787977254e-06, + "loss": 2.2532, + "step": 388 + }, + { + "epoch": 0.4263013698630137, + "grad_norm": 1.855372428894043, + "learning_rate": 4.794882209585703e-06, + "loss": 2.2538, + "step": 389 + }, + { + "epoch": 0.4273972602739726, + "grad_norm": 2.4465434551239014, + "learning_rate": 4.79285134037368e-06, + "loss": 2.2724, + "step": 390 + }, + { + "epoch": 0.4273972602739726, + "eval_loss": 2.364928960800171, + "eval_runtime": 48.0245, + "eval_samples_per_second": 20.823, + "eval_steps_per_second": 0.875, + "step": 390 + }, + { + "epoch": 0.4284931506849315, + "grad_norm": 1.8911314010620117, + "learning_rate": 4.790820471161657e-06, + "loss": 2.2291, + "step": 391 + }, + { + "epoch": 0.4295890410958904, + "grad_norm": 2.1975977420806885, + "learning_rate": 4.7887896019496345e-06, + "loss": 2.2373, + "step": 392 + }, + { + "epoch": 0.43068493150684933, + "grad_norm": 2.015331506729126, + "learning_rate": 4.786758732737612e-06, + "loss": 2.2483, + "step": 393 + }, + { + "epoch": 0.4317808219178082, + "grad_norm": 2.039339065551758, + "learning_rate": 4.784727863525589e-06, + "loss": 2.2746, + "step": 394 + }, + { + "epoch": 0.4328767123287671, + "grad_norm": 1.977513074874878, + "learning_rate": 4.782696994313566e-06, + "loss": 2.2275, + "step": 395 + }, + { + "epoch": 0.43397260273972604, + "grad_norm": 2.147217035293579, + "learning_rate": 4.780666125101544e-06, + "loss": 2.252, + "step": 396 + }, + { + "epoch": 0.4350684931506849, + "grad_norm": 2.1799919605255127, + "learning_rate": 4.778635255889521e-06, + "loss": 2.1975, + "step": 397 + }, + { + "epoch": 0.43616438356164383, + "grad_norm": 2.582240343093872, + "learning_rate": 4.776604386677498e-06, + "loss": 2.2294, + "step": 398 + }, + { + "epoch": 0.43726027397260275, + "grad_norm": 1.9127846956253052, + "learning_rate": 4.7745735174654755e-06, + "loss": 2.2319, + "step": 399 + }, + { + "epoch": 0.4383561643835616, + "grad_norm": 2.2444543838500977, + "learning_rate": 4.772542648253453e-06, + "loss": 2.2857, + "step": 400 + }, + { + "epoch": 0.4383561643835616, + "eval_loss": 2.3653724193573, + "eval_runtime": 43.8304, + "eval_samples_per_second": 22.815, + "eval_steps_per_second": 0.958, + "step": 400 + } + ], + "logging_steps": 1, + "max_steps": 2736, + "num_input_tokens_seen": 0, + "num_train_epochs": 3, + "save_steps": 100, + "stateful_callbacks": { + "TrainerControl": { + "args": { + "should_epoch_stop": false, + "should_evaluate": false, + "should_log": false, + "should_save": true, + "should_training_stop": false + }, + "attributes": {} + } + }, + "total_flos": 1.246923763744768e+17, + "train_batch_size": 3, + "trial_name": null, + "trial_params": null +} diff --git a/babylm_shuffle_nondeterministic_10M_seed0/runs/checkpoint-400/zero_to_fp32.py b/babylm_shuffle_nondeterministic_10M_seed0/runs/checkpoint-400/zero_to_fp32.py new file mode 100644 index 0000000000000000000000000000000000000000..24cc342e78d1a006c782b3a4cd68d9ce786d8fd8 --- /dev/null +++ b/babylm_shuffle_nondeterministic_10M_seed0/runs/checkpoint-400/zero_to_fp32.py @@ -0,0 +1,604 @@ +#!/usr/bin/env python + +# Copyright (c) Microsoft Corporation. +# SPDX-License-Identifier: Apache-2.0 + +# DeepSpeed Team + +# This script extracts fp32 consolidated weights from a zero 1, 2 and 3 DeepSpeed checkpoints. It gets +# copied into the top level checkpoint dir, so the user can easily do the conversion at any point in +# the future. Once extracted, the weights don't require DeepSpeed and can be used in any +# application. +# +# example: python zero_to_fp32.py . pytorch_model.bin + +import argparse +import torch +import glob +import math +import os +import re +from collections import OrderedDict +from dataclasses import dataclass + +# while this script doesn't use deepspeed to recover data, since the checkpoints are pickled with +# DeepSpeed data structures it has to be available in the current python environment. +from deepspeed.utils import logger +from deepspeed.checkpoint.constants import (DS_VERSION, OPTIMIZER_STATE_DICT, SINGLE_PARTITION_OF_FP32_GROUPS, + FP32_FLAT_GROUPS, ZERO_STAGE, PARTITION_COUNT, PARAM_SHAPES, BUFFER_NAMES, + FROZEN_PARAM_SHAPES, FROZEN_PARAM_FRAGMENTS) + + +@dataclass +class zero_model_state: + buffers: dict() + param_shapes: dict() + shared_params: list + ds_version: int + frozen_param_shapes: dict() + frozen_param_fragments: dict() + + +debug = 0 + +# load to cpu +device = torch.device('cpu') + + +def atoi(text): + return int(text) if text.isdigit() else text + + +def natural_keys(text): + ''' + alist.sort(key=natural_keys) sorts in human order + http://nedbatchelder.com/blog/200712/human_sorting.html + (See Toothy's implementation in the comments) + ''' + return [atoi(c) for c in re.split(r'(\d+)', text)] + + +def get_model_state_file(checkpoint_dir, zero_stage): + if not os.path.isdir(checkpoint_dir): + raise FileNotFoundError(f"Directory '{checkpoint_dir}' doesn't exist") + + # there should be only one file + if zero_stage <= 2: + file = os.path.join(checkpoint_dir, "mp_rank_00_model_states.pt") + elif zero_stage == 3: + file = os.path.join(checkpoint_dir, "zero_pp_rank_0_mp_rank_00_model_states.pt") + + if not os.path.exists(file): + raise FileNotFoundError(f"can't find model states file at '{file}'") + + return file + + +def get_checkpoint_files(checkpoint_dir, glob_pattern): + # XXX: need to test that this simple glob rule works for multi-node setup too + ckpt_files = sorted(glob.glob(os.path.join(checkpoint_dir, glob_pattern)), key=natural_keys) + + if len(ckpt_files) == 0: + raise FileNotFoundError(f"can't find {glob_pattern} files in directory '{checkpoint_dir}'") + + return ckpt_files + + +def get_optim_files(checkpoint_dir): + return get_checkpoint_files(checkpoint_dir, "*_optim_states.pt") + + +def get_model_state_files(checkpoint_dir): + return get_checkpoint_files(checkpoint_dir, "*_model_states.pt") + + +def parse_model_states(files): + zero_model_states = [] + for file in files: + state_dict = torch.load(file, map_location=device) + + if BUFFER_NAMES not in state_dict: + raise ValueError(f"{file} is not a model state checkpoint") + buffer_names = state_dict[BUFFER_NAMES] + if debug: + print("Found buffers:", buffer_names) + + # recover just the buffers while restoring them to fp32 if they were saved in fp16 + buffers = {k: v.float() for k, v in state_dict["module"].items() if k in buffer_names} + param_shapes = state_dict[PARAM_SHAPES] + + # collect parameters that are included in param_shapes + param_names = [] + for s in param_shapes: + for name in s.keys(): + param_names.append(name) + + # update with frozen parameters + frozen_param_shapes = state_dict.get(FROZEN_PARAM_SHAPES, None) + if frozen_param_shapes is not None: + if debug: + print(f"Found frozen_param_shapes: {frozen_param_shapes}") + param_names += list(frozen_param_shapes.keys()) + + # handle shared params + shared_params = [[k, v] for k, v in state_dict["shared_params"].items()] + + ds_version = state_dict.get(DS_VERSION, None) + + frozen_param_fragments = state_dict.get(FROZEN_PARAM_FRAGMENTS, None) + + z_model_state = zero_model_state(buffers=buffers, + param_shapes=param_shapes, + shared_params=shared_params, + ds_version=ds_version, + frozen_param_shapes=frozen_param_shapes, + frozen_param_fragments=frozen_param_fragments) + zero_model_states.append(z_model_state) + + return zero_model_states + + +def parse_optim_states(files, ds_checkpoint_dir): + + total_files = len(files) + state_dicts = [] + for f in files: + state_dict = torch.load(f, map_location=device) + # immediately discard the potentially huge 2 optimizer states as we only care for fp32 master weights + # and also handle the case where it was already removed by another helper script + state_dict["optimizer_state_dict"].pop("optimizer_state_dict", None) + state_dicts.append(state_dict) + + if not ZERO_STAGE in state_dicts[0][OPTIMIZER_STATE_DICT]: + raise ValueError(f"{files[0]} is not a zero checkpoint") + zero_stage = state_dicts[0][OPTIMIZER_STATE_DICT][ZERO_STAGE] + world_size = state_dicts[0][OPTIMIZER_STATE_DICT][PARTITION_COUNT] + + # For ZeRO-2 each param group can have different partition_count as data parallelism for expert + # parameters can be different from data parallelism for non-expert parameters. So we can just + # use the max of the partition_count to get the dp world_size. + + if type(world_size) is list: + world_size = max(world_size) + + if world_size != total_files: + raise ValueError( + f"Expected {world_size} of '*_optim_states.pt' under '{ds_checkpoint_dir}' but found {total_files} files. " + "Possibly due to an overwrite of an old checkpoint, or a checkpoint didn't get saved by one or more processes." + ) + + # the groups are named differently in each stage + if zero_stage <= 2: + fp32_groups_key = SINGLE_PARTITION_OF_FP32_GROUPS + elif zero_stage == 3: + fp32_groups_key = FP32_FLAT_GROUPS + else: + raise ValueError(f"unknown zero stage {zero_stage}") + + if zero_stage <= 2: + fp32_flat_groups = [state_dicts[i][OPTIMIZER_STATE_DICT][fp32_groups_key] for i in range(len(state_dicts))] + elif zero_stage == 3: + # if there is more than one param group, there will be multiple flattened tensors - one + # flattened tensor per group - for simplicity merge them into a single tensor + # + # XXX: could make the script more memory efficient for when there are multiple groups - it + # will require matching the sub-lists of param_shapes for each param group flattened tensor + + fp32_flat_groups = [ + torch.cat(state_dicts[i][OPTIMIZER_STATE_DICT][fp32_groups_key], 0) for i in range(len(state_dicts)) + ] + + return zero_stage, world_size, fp32_flat_groups + + +def _get_fp32_state_dict_from_zero_checkpoint(ds_checkpoint_dir, exclude_frozen_parameters): + """ + Returns fp32 state_dict reconstructed from ds checkpoint + + Args: + - ``ds_checkpoint_dir``: path to the deepspeed checkpoint folder (where the optimizer files are) + + """ + print(f"Processing zero checkpoint '{ds_checkpoint_dir}'") + + optim_files = get_optim_files(ds_checkpoint_dir) + zero_stage, world_size, fp32_flat_groups = parse_optim_states(optim_files, ds_checkpoint_dir) + print(f"Detected checkpoint of type zero stage {zero_stage}, world_size: {world_size}") + + model_files = get_model_state_files(ds_checkpoint_dir) + + zero_model_states = parse_model_states(model_files) + print(f'Parsing checkpoint created by deepspeed=={zero_model_states[0].ds_version}') + + if zero_stage <= 2: + return _get_fp32_state_dict_from_zero2_checkpoint(world_size, fp32_flat_groups, zero_model_states, + exclude_frozen_parameters) + elif zero_stage == 3: + return _get_fp32_state_dict_from_zero3_checkpoint(world_size, fp32_flat_groups, zero_model_states, + exclude_frozen_parameters) + + +def _zero2_merge_frozen_params(state_dict, zero_model_states): + if zero_model_states[0].frozen_param_shapes is None or len(zero_model_states[0].frozen_param_shapes) == 0: + return + + frozen_param_shapes = zero_model_states[0].frozen_param_shapes + frozen_param_fragments = zero_model_states[0].frozen_param_fragments + + if debug: + num_elem = sum(s.numel() for s in frozen_param_shapes.values()) + print(f'rank 0: {FROZEN_PARAM_SHAPES}.numel = {num_elem}') + + wanted_params = len(frozen_param_shapes) + wanted_numel = sum(s.numel() for s in frozen_param_shapes.values()) + avail_numel = sum([p.numel() for p in frozen_param_fragments.values()]) + print(f'Frozen params: Have {avail_numel} numels to process.') + print(f'Frozen params: Need {wanted_numel} numels in {wanted_params} params') + + total_params = 0 + total_numel = 0 + for name, shape in frozen_param_shapes.items(): + total_params += 1 + unpartitioned_numel = shape.numel() + total_numel += unpartitioned_numel + + state_dict[name] = frozen_param_fragments[name] + + if debug: + print(f"{name} full shape: {shape} unpartitioned numel {unpartitioned_numel} ") + + print(f"Reconstructed Frozen fp32 state dict with {total_params} params {total_numel} elements") + + +def _has_callable(obj, fn): + attr = getattr(obj, fn, None) + return callable(attr) + + +def _zero2_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states): + param_shapes = zero_model_states[0].param_shapes + + # Reconstruction protocol: + # + # XXX: document this + + if debug: + for i in range(world_size): + for j in range(len(fp32_flat_groups[0])): + print(f"{FP32_FLAT_GROUPS}[{i}][{j}].shape={fp32_flat_groups[i][j].shape}") + + # XXX: memory usage doubles here (zero2) + num_param_groups = len(fp32_flat_groups[0]) + merged_single_partition_of_fp32_groups = [] + for i in range(num_param_groups): + merged_partitions = [sd[i] for sd in fp32_flat_groups] + full_single_fp32_vector = torch.cat(merged_partitions, 0) + merged_single_partition_of_fp32_groups.append(full_single_fp32_vector) + avail_numel = sum( + [full_single_fp32_vector.numel() for full_single_fp32_vector in merged_single_partition_of_fp32_groups]) + + if debug: + wanted_params = sum([len(shapes) for shapes in param_shapes]) + wanted_numel = sum([sum(shape.numel() for shape in shapes.values()) for shapes in param_shapes]) + # not asserting if there is a mismatch due to possible padding + print(f"Have {avail_numel} numels to process.") + print(f"Need {wanted_numel} numels in {wanted_params} params.") + + # params + # XXX: for huge models that can't fit into the host's RAM we will have to recode this to support + # out-of-core computing solution + total_numel = 0 + total_params = 0 + for shapes, full_single_fp32_vector in zip(param_shapes, merged_single_partition_of_fp32_groups): + offset = 0 + avail_numel = full_single_fp32_vector.numel() + for name, shape in shapes.items(): + + unpartitioned_numel = shape.numel() if _has_callable(shape, 'numel') else math.prod(shape) + total_numel += unpartitioned_numel + total_params += 1 + + if debug: + print(f"{name} full shape: {shape} unpartitioned numel {unpartitioned_numel} ") + state_dict[name] = full_single_fp32_vector.narrow(0, offset, unpartitioned_numel).view(shape) + offset += unpartitioned_numel + + # Z2 started to align to 2*world_size to improve nccl performance. Therefore both offset and + # avail_numel can differ by anywhere between 0..2*world_size. Due to two unrelated complex + # paddings performed in the code it's almost impossible to predict the exact numbers w/o the + # live optimizer object, so we are checking that the numbers are within the right range + align_to = 2 * world_size + + def zero2_align(x): + return align_to * math.ceil(x / align_to) + + if debug: + print(f"original offset={offset}, avail_numel={avail_numel}") + + offset = zero2_align(offset) + avail_numel = zero2_align(avail_numel) + + if debug: + print(f"aligned offset={offset}, avail_numel={avail_numel}") + + # Sanity check + if offset != avail_numel: + raise ValueError(f"consumed {offset} numels out of {avail_numel} - something is wrong") + + print(f"Reconstructed fp32 state dict with {total_params} params {total_numel} elements") + + +def _get_fp32_state_dict_from_zero2_checkpoint(world_size, fp32_flat_groups, zero_model_states, + exclude_frozen_parameters): + state_dict = OrderedDict() + + # buffers + buffers = zero_model_states[0].buffers + state_dict.update(buffers) + if debug: + print(f"added {len(buffers)} buffers") + + if not exclude_frozen_parameters: + _zero2_merge_frozen_params(state_dict, zero_model_states) + + _zero2_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states) + + # recover shared parameters + for pair in zero_model_states[0].shared_params: + if pair[1] in state_dict: + state_dict[pair[0]] = state_dict[pair[1]] + + return state_dict + + +def zero3_partitioned_param_info(unpartitioned_numel, world_size): + remainder = unpartitioned_numel % world_size + padding_numel = (world_size - remainder) if remainder else 0 + partitioned_numel = math.ceil(unpartitioned_numel / world_size) + return partitioned_numel, padding_numel + + +def _zero3_merge_frozen_params(state_dict, world_size, zero_model_states): + if zero_model_states[0].frozen_param_shapes is None or len(zero_model_states[0].frozen_param_shapes) == 0: + return + + if debug: + for i in range(world_size): + num_elem = sum(s.numel() for s in zero_model_states[i].frozen_param_fragments.values()) + print(f'rank {i}: {FROZEN_PARAM_SHAPES}.numel = {num_elem}') + + frozen_param_shapes = zero_model_states[0].frozen_param_shapes + wanted_params = len(frozen_param_shapes) + wanted_numel = sum(s.numel() for s in frozen_param_shapes.values()) + avail_numel = sum([p.numel() for p in zero_model_states[0].frozen_param_fragments.values()]) * world_size + print(f'Frozen params: Have {avail_numel} numels to process.') + print(f'Frozen params: Need {wanted_numel} numels in {wanted_params} params') + + total_params = 0 + total_numel = 0 + for name, shape in zero_model_states[0].frozen_param_shapes.items(): + total_params += 1 + unpartitioned_numel = shape.numel() + total_numel += unpartitioned_numel + + param_frags = tuple(model_state.frozen_param_fragments[name] for model_state in zero_model_states) + state_dict[name] = torch.cat(param_frags, 0).narrow(0, 0, unpartitioned_numel).view(shape) + + partitioned_numel, partitioned_padding_numel = zero3_partitioned_param_info(unpartitioned_numel, world_size) + + if debug: + print( + f"Frozen params: {total_params} {name} full shape: {shape} partition0 numel={partitioned_numel} partitioned_padding_numel={partitioned_padding_numel}" + ) + + print(f"Reconstructed Frozen fp32 state dict with {total_params} params {total_numel} elements") + + +def _zero3_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states): + param_shapes = zero_model_states[0].param_shapes + avail_numel = fp32_flat_groups[0].numel() * world_size + # Reconstruction protocol: For zero3 we need to zip the partitions together at boundary of each + # param, re-consolidating each param, while dealing with padding if any + + # merge list of dicts, preserving order + param_shapes = {k: v for d in param_shapes for k, v in d.items()} + + if debug: + for i in range(world_size): + print(f"{FP32_FLAT_GROUPS}[{i}].shape={fp32_flat_groups[i].shape}") + + wanted_params = len(param_shapes) + wanted_numel = sum(shape.numel() for shape in param_shapes.values()) + # not asserting if there is a mismatch due to possible padding + avail_numel = fp32_flat_groups[0].numel() * world_size + print(f"Trainable params: Have {avail_numel} numels to process.") + print(f"Trainable params: Need {wanted_numel} numels in {wanted_params} params.") + + # params + # XXX: for huge models that can't fit into the host's RAM we will have to recode this to support + # out-of-core computing solution + offset = 0 + total_numel = 0 + total_params = 0 + for name, shape in param_shapes.items(): + + unpartitioned_numel = shape.numel() + total_numel += unpartitioned_numel + total_params += 1 + + partitioned_numel, partitioned_padding_numel = zero3_partitioned_param_info(unpartitioned_numel, world_size) + + if debug: + print( + f"Trainable params: {total_params} {name} full shape: {shape} partition0 numel={partitioned_numel} partitioned_padding_numel={partitioned_padding_numel}" + ) + + # XXX: memory usage doubles here + state_dict[name] = torch.cat( + tuple(fp32_flat_groups[i].narrow(0, offset, partitioned_numel) for i in range(world_size)), + 0).narrow(0, 0, unpartitioned_numel).view(shape) + offset += partitioned_numel + + offset *= world_size + + # Sanity check + if offset != avail_numel: + raise ValueError(f"consumed {offset} numels out of {avail_numel} - something is wrong") + + print(f"Reconstructed Trainable fp32 state dict with {total_params} params {total_numel} elements") + + +def _get_fp32_state_dict_from_zero3_checkpoint(world_size, fp32_flat_groups, zero_model_states, + exclude_frozen_parameters): + state_dict = OrderedDict() + + # buffers + buffers = zero_model_states[0].buffers + state_dict.update(buffers) + if debug: + print(f"added {len(buffers)} buffers") + + if not exclude_frozen_parameters: + _zero3_merge_frozen_params(state_dict, world_size, zero_model_states) + + _zero3_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states) + + # recover shared parameters + for pair in zero_model_states[0].shared_params: + if pair[1] in state_dict: + state_dict[pair[0]] = state_dict[pair[1]] + + return state_dict + + +def get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir, tag=None, exclude_frozen_parameters=False): + """ + Convert ZeRO 2 or 3 checkpoint into a single fp32 consolidated state_dict that can be loaded with + ``load_state_dict()`` and used for training without DeepSpeed or shared with others, for example + via a model hub. + + Args: + - ``checkpoint_dir``: path to the desired checkpoint folder + - ``tag``: checkpoint tag used as a unique identifier for checkpoint. If not provided will attempt to load tag in 'latest' file. e.g., ``global_step14`` + - ``exclude_frozen_parameters``: exclude frozen parameters + + Returns: + - pytorch ``state_dict`` + + Note: this approach may not work if your application doesn't have sufficient free CPU memory and + you may need to use the offline approach using the ``zero_to_fp32.py`` script that is saved with + the checkpoint. + + A typical usage might be :: + + from deepspeed.utils.zero_to_fp32 import get_fp32_state_dict_from_zero_checkpoint + # do the training and checkpoint saving + state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir) # already on cpu + model = model.cpu() # move to cpu + model.load_state_dict(state_dict) + # submit to model hub or save the model to share with others + + In this example the ``model`` will no longer be usable in the deepspeed context of the same + application. i.e. you will need to re-initialize the deepspeed engine, since + ``model.load_state_dict(state_dict)`` will remove all the deepspeed magic from it. + + If you want it all done for you, use ``load_state_dict_from_zero_checkpoint`` instead. + + """ + if tag is None: + latest_path = os.path.join(checkpoint_dir, 'latest') + if os.path.isfile(latest_path): + with open(latest_path, 'r') as fd: + tag = fd.read().strip() + else: + raise ValueError(f"Unable to find 'latest' file at {latest_path}") + + ds_checkpoint_dir = os.path.join(checkpoint_dir, tag) + + if not os.path.isdir(ds_checkpoint_dir): + raise FileNotFoundError(f"Directory '{ds_checkpoint_dir}' doesn't exist") + + return _get_fp32_state_dict_from_zero_checkpoint(ds_checkpoint_dir, exclude_frozen_parameters) + + +def convert_zero_checkpoint_to_fp32_state_dict(checkpoint_dir, output_file, tag=None, exclude_frozen_parameters=False): + """ + Convert ZeRO 2 or 3 checkpoint into a single fp32 consolidated ``state_dict`` file that can be + loaded with ``torch.load(file)`` + ``load_state_dict()`` and used for training without DeepSpeed. + + Args: + - ``checkpoint_dir``: path to the desired checkpoint folder. (one that contains the tag-folder, like ``global_step14``) + - ``output_file``: path to the pytorch fp32 state_dict output file (e.g. path/pytorch_model.bin) + - ``tag``: checkpoint tag used as a unique identifier for checkpoint. If not provided will attempt to load tag in the file named ``latest`` in the checkpoint folder, e.g., ``global_step14`` + - ``exclude_frozen_parameters``: exclude frozen parameters + """ + + state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir, tag, exclude_frozen_parameters) + print(f"Saving fp32 state dict to {output_file}") + torch.save(state_dict, output_file) + + +def load_state_dict_from_zero_checkpoint(model, checkpoint_dir, tag=None): + """ + 1. Put the provided model to cpu + 2. Convert ZeRO 2 or 3 checkpoint into a single fp32 consolidated ``state_dict`` + 3. Load it into the provided model + + Args: + - ``model``: the model object to update + - ``checkpoint_dir``: path to the desired checkpoint folder. (one that contains the tag-folder, like ``global_step14``) + - ``tag``: checkpoint tag used as a unique identifier for checkpoint. If not provided will attempt to load tag in the file named ``latest`` in the checkpoint folder, e.g., ``global_step14`` + + Returns: + - ``model`: modified model + + Make sure you have plenty of CPU memory available before you call this function. If you don't + have enough use the ``zero_to_fp32.py`` utility to do the conversion. You will find it + conveniently placed for you in the checkpoint folder. + + A typical usage might be :: + + from deepspeed.utils.zero_to_fp32 import load_state_dict_from_zero_checkpoint + model = load_state_dict_from_zero_checkpoint(trainer.model, checkpoint_dir) + # submit to model hub or save the model to share with others + + Note, that once this was run, the ``model`` will no longer be usable in the deepspeed context + of the same application. i.e. you will need to re-initialize the deepspeed engine, since + ``model.load_state_dict(state_dict)`` will remove all the deepspeed magic from it. + + """ + logger.info(f"Extracting fp32 weights") + state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir, tag) + + logger.info(f"Overwriting model with fp32 weights") + model = model.cpu() + model.load_state_dict(state_dict, strict=False) + + return model + + +if __name__ == "__main__": + + parser = argparse.ArgumentParser() + parser.add_argument("checkpoint_dir", + type=str, + help="path to the desired checkpoint folder, e.g., path/checkpoint-12") + parser.add_argument( + "output_file", + type=str, + help="path to the pytorch fp32 state_dict output file (e.g. path/checkpoint-12/pytorch_model.bin)") + parser.add_argument("-t", + "--tag", + type=str, + default=None, + help="checkpoint tag used as a unique identifier for checkpoint. e.g., global_step1") + parser.add_argument("--exclude_frozen_parameters", action='store_true', help="exclude frozen parameters") + parser.add_argument("-d", "--debug", action='store_true', help="enable debug") + args = parser.parse_args() + + debug = args.debug + + convert_zero_checkpoint_to_fp32_state_dict(args.checkpoint_dir, + args.output_file, + tag=args.tag, + exclude_frozen_parameters=args.exclude_frozen_parameters) diff --git a/babylm_shuffle_nondeterministic_10M_seed0/runs/checkpoint-600/config.json b/babylm_shuffle_nondeterministic_10M_seed0/runs/checkpoint-600/config.json new file mode 100644 index 0000000000000000000000000000000000000000..fb0f9a53a507ffcdb60410deb71d60ef801bf350 --- /dev/null +++ b/babylm_shuffle_nondeterministic_10M_seed0/runs/checkpoint-600/config.json @@ -0,0 +1,36 @@ +{ + "_name_or_path": "meta-llama/Llama-3.2-3B", + "architectures": [ + "LlamaForCausalLM" + ], + "attention_bias": false, + "attention_dropout": 0.0, + "bos_token_id": 128000, + "eos_token_id": 128001, + "head_dim": 128, + "hidden_act": "silu", + "hidden_size": 3072, + "initializer_range": 0.02, + "intermediate_size": 8192, + "max_position_embeddings": 131072, + "mlp_bias": false, + "model_type": "llama", + "num_attention_heads": 24, + "num_hidden_layers": 28, + "num_key_value_heads": 8, + "pretraining_tp": 1, + "rms_norm_eps": 1e-05, + "rope_scaling": { + "factor": 32.0, + "high_freq_factor": 4.0, + "low_freq_factor": 1.0, + "original_max_position_embeddings": 8192, + "rope_type": "llama3" + }, + "rope_theta": 500000.0, + "tie_word_embeddings": true, + "torch_dtype": "float16", + "transformers_version": "4.45.1", + "use_cache": true, + "vocab_size": 128256 +} diff --git a/babylm_shuffle_nondeterministic_10M_seed0/runs/checkpoint-600/generation_config.json b/babylm_shuffle_nondeterministic_10M_seed0/runs/checkpoint-600/generation_config.json new file mode 100644 index 0000000000000000000000000000000000000000..9c389b841a9a29ddd904e02d1eb0e08dcce82ad9 --- /dev/null +++ b/babylm_shuffle_nondeterministic_10M_seed0/runs/checkpoint-600/generation_config.json @@ -0,0 +1,9 @@ +{ + "_from_model_config": true, + "bos_token_id": 128000, + "do_sample": true, + "eos_token_id": 128001, + "temperature": 0.6, + "top_p": 0.9, + "transformers_version": "4.45.1" +} diff --git a/babylm_shuffle_nondeterministic_10M_seed0/runs/checkpoint-600/latest b/babylm_shuffle_nondeterministic_10M_seed0/runs/checkpoint-600/latest new file mode 100644 index 0000000000000000000000000000000000000000..12cae1adf3af8546b4141c6f62261c8e99839a54 --- /dev/null +++ b/babylm_shuffle_nondeterministic_10M_seed0/runs/checkpoint-600/latest @@ -0,0 +1 @@ +global_step600 \ No newline at end of file diff --git a/babylm_shuffle_nondeterministic_10M_seed0/runs/checkpoint-600/model.safetensors.index.json b/babylm_shuffle_nondeterministic_10M_seed0/runs/checkpoint-600/model.safetensors.index.json new file mode 100644 index 0000000000000000000000000000000000000000..ed64de846d720b9a7859dc20575fea8e8ca51940 --- /dev/null +++ b/babylm_shuffle_nondeterministic_10M_seed0/runs/checkpoint-600/model.safetensors.index.json @@ -0,0 +1,262 @@ +{ + "metadata": { + "total_size": 7213504512 + }, + "weight_map": { + "lm_head.weight": "model-00002-of-00002.safetensors", + "model.embed_tokens.weight": "model-00001-of-00002.safetensors", + "model.layers.0.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.0.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.0.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.0.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.0.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.0.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.0.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.0.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.0.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.1.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.1.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.1.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.1.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.1.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.1.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.1.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.1.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.1.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.10.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.10.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.10.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.10.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.10.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.10.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.10.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.10.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.10.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.11.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.11.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.11.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.11.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.11.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.11.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.11.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.11.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.11.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.12.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.12.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.12.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.12.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.12.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.12.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.12.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.12.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.12.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.13.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.13.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.13.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.13.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.13.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.13.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.13.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.13.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.13.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.14.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.14.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.14.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.14.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.14.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.14.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.14.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.14.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.14.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.15.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.15.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.15.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.15.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.15.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.15.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.15.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.15.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.15.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.16.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.16.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.16.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.16.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.16.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.16.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.16.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.16.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.16.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.17.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.17.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.17.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.17.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.17.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.17.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.17.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.17.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.17.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.18.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.18.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.18.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.18.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.18.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.18.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.18.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.18.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.18.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.19.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.19.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.19.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.19.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.19.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.19.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.19.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.19.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.19.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.2.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.2.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.2.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.2.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.2.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.2.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.2.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.2.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.2.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.20.input_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.20.mlp.down_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.20.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.20.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.20.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.20.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.20.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.20.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.20.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.21.input_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.21.mlp.down_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.21.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.21.mlp.up_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.21.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.21.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.21.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.21.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.21.self_attn.v_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.22.input_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.22.mlp.down_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.22.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.22.mlp.up_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.22.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.22.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.22.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.22.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.22.self_attn.v_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.23.input_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.23.mlp.down_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.23.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.23.mlp.up_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.23.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.23.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.23.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.23.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.23.self_attn.v_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.24.input_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.24.mlp.down_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.24.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.24.mlp.up_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.24.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.24.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.24.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.24.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.24.self_attn.v_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.25.input_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.25.mlp.down_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.25.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.25.mlp.up_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.25.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.25.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.25.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.25.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.25.self_attn.v_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.26.input_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.26.mlp.down_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.26.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.26.mlp.up_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.26.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.26.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.26.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.26.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.26.self_attn.v_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.27.input_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.27.mlp.down_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.27.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.27.mlp.up_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.27.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.27.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.27.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.27.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.27.self_attn.v_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.3.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.3.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.3.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.3.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.3.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.3.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.3.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.3.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.3.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.4.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.4.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.4.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.4.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.4.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.4.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.4.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.4.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.4.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.5.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.5.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.5.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.5.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.5.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.5.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.5.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.5.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.5.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.6.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.6.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.6.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.6.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.6.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.6.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.6.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.6.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.6.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.7.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.7.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.7.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.7.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.7.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.7.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.7.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.7.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.7.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.8.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.8.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.8.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.8.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.8.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.8.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.8.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.8.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.8.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.9.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.9.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.9.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.9.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.9.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.9.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.9.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.9.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.9.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.norm.weight": "model-00002-of-00002.safetensors" + } +} diff --git a/babylm_shuffle_nondeterministic_10M_seed0/runs/checkpoint-600/special_tokens_map.json b/babylm_shuffle_nondeterministic_10M_seed0/runs/checkpoint-600/special_tokens_map.json new file mode 100644 index 0000000000000000000000000000000000000000..6949c5975ee0e961ef61cf31010dce04df0a03f8 --- /dev/null +++ b/babylm_shuffle_nondeterministic_10M_seed0/runs/checkpoint-600/special_tokens_map.json @@ -0,0 +1,23 @@ +{ + "bos_token": { + "content": "<|begin_of_text|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false + }, + "eos_token": { + "content": "<|end_of_text|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false + }, + "pad_token": { + "content": "[PAD]", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false + } +} diff --git a/babylm_shuffle_nondeterministic_10M_seed0/runs/checkpoint-600/tokenizer_config.json b/babylm_shuffle_nondeterministic_10M_seed0/runs/checkpoint-600/tokenizer_config.json new file mode 100644 index 0000000000000000000000000000000000000000..5684b14e2756b56aa397c1324e239b442162b34a --- /dev/null +++ b/babylm_shuffle_nondeterministic_10M_seed0/runs/checkpoint-600/tokenizer_config.json @@ -0,0 +1,2070 @@ +{ + "added_tokens_decoder": { + "128000": { + "content": "<|begin_of_text|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128001": { + "content": "<|end_of_text|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128002": { + "content": "<|reserved_special_token_0|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128003": { + "content": "<|reserved_special_token_1|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128004": { + "content": "<|finetune_right_pad_id|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128005": { + "content": "<|reserved_special_token_2|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128006": { + "content": "<|start_header_id|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128007": { + "content": "<|end_header_id|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128008": { + "content": "<|eom_id|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128009": { + "content": "<|eot_id|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128010": { + "content": "<|python_tag|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128011": { + "content": "<|reserved_special_token_3|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128012": { + "content": "<|reserved_special_token_4|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128013": { + "content": "<|reserved_special_token_5|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128014": { + "content": "<|reserved_special_token_6|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128015": { + "content": "<|reserved_special_token_7|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128016": { + "content": "<|reserved_special_token_8|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128017": { + "content": "<|reserved_special_token_9|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128018": { + "content": "<|reserved_special_token_10|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128019": { + "content": "<|reserved_special_token_11|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128020": { + "content": "<|reserved_special_token_12|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128021": { + "content": "<|reserved_special_token_13|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128022": { + "content": "<|reserved_special_token_14|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128023": { + "content": "<|reserved_special_token_15|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128024": { + "content": "<|reserved_special_token_16|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128025": { + "content": "<|reserved_special_token_17|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128026": { + "content": "<|reserved_special_token_18|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128027": { + "content": "<|reserved_special_token_19|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128028": { + "content": "<|reserved_special_token_20|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128029": { + "content": "<|reserved_special_token_21|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128030": { + "content": "<|reserved_special_token_22|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128031": { + "content": "<|reserved_special_token_23|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128032": { + "content": "<|reserved_special_token_24|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128033": { + "content": "<|reserved_special_token_25|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128034": { + "content": "<|reserved_special_token_26|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128035": { + "content": "<|reserved_special_token_27|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128036": { + "content": "<|reserved_special_token_28|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128037": { + "content": "<|reserved_special_token_29|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128038": { + "content": "<|reserved_special_token_30|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128039": { + "content": "<|reserved_special_token_31|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128040": { + "content": "<|reserved_special_token_32|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128041": { + "content": "<|reserved_special_token_33|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128042": { + "content": "<|reserved_special_token_34|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128043": { + "content": "<|reserved_special_token_35|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128044": { + "content": "<|reserved_special_token_36|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128045": { + "content": "<|reserved_special_token_37|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128046": { + "content": "<|reserved_special_token_38|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128047": { + "content": "<|reserved_special_token_39|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128048": { + "content": "<|reserved_special_token_40|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128049": { + "content": "<|reserved_special_token_41|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128050": { + "content": "<|reserved_special_token_42|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128051": { + "content": "<|reserved_special_token_43|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128052": { + "content": "<|reserved_special_token_44|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128053": { + "content": "<|reserved_special_token_45|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128054": { + "content": "<|reserved_special_token_46|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128055": { + "content": "<|reserved_special_token_47|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128056": { + "content": "<|reserved_special_token_48|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128057": { + "content": "<|reserved_special_token_49|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128058": { + "content": "<|reserved_special_token_50|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128059": { + "content": "<|reserved_special_token_51|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128060": { + "content": "<|reserved_special_token_52|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128061": { + "content": "<|reserved_special_token_53|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128062": { + "content": "<|reserved_special_token_54|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128063": { + "content": "<|reserved_special_token_55|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128064": { + "content": "<|reserved_special_token_56|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128065": { + "content": "<|reserved_special_token_57|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128066": { + "content": "<|reserved_special_token_58|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128067": { + "content": "<|reserved_special_token_59|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128068": { + "content": "<|reserved_special_token_60|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128069": { + "content": "<|reserved_special_token_61|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128070": { + "content": "<|reserved_special_token_62|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128071": { + "content": "<|reserved_special_token_63|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128072": { + "content": "<|reserved_special_token_64|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128073": { + "content": "<|reserved_special_token_65|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128074": { + "content": "<|reserved_special_token_66|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128075": { + "content": "<|reserved_special_token_67|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128076": { + "content": "<|reserved_special_token_68|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128077": { + "content": "<|reserved_special_token_69|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128078": { + "content": "<|reserved_special_token_70|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128079": { + "content": "<|reserved_special_token_71|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128080": { + "content": "<|reserved_special_token_72|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128081": { + "content": "<|reserved_special_token_73|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128082": { + "content": "<|reserved_special_token_74|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128083": { + "content": "<|reserved_special_token_75|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128084": { + "content": "<|reserved_special_token_76|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128085": { + "content": "<|reserved_special_token_77|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128086": { + "content": "<|reserved_special_token_78|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128087": { + "content": "<|reserved_special_token_79|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128088": { + "content": "<|reserved_special_token_80|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128089": { + "content": "<|reserved_special_token_81|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128090": { + "content": "<|reserved_special_token_82|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128091": { + "content": "<|reserved_special_token_83|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128092": { + "content": "<|reserved_special_token_84|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128093": { + "content": "<|reserved_special_token_85|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128094": { + "content": "<|reserved_special_token_86|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128095": { + "content": "<|reserved_special_token_87|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128096": { + "content": "<|reserved_special_token_88|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128097": { + "content": "<|reserved_special_token_89|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128098": { + "content": "<|reserved_special_token_90|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128099": { + "content": "<|reserved_special_token_91|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128100": { + "content": "<|reserved_special_token_92|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128101": { + "content": "<|reserved_special_token_93|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128102": { + "content": "<|reserved_special_token_94|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128103": { + "content": "<|reserved_special_token_95|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128104": { + "content": "<|reserved_special_token_96|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128105": { + "content": "<|reserved_special_token_97|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128106": { + "content": "<|reserved_special_token_98|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128107": { + "content": "<|reserved_special_token_99|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128108": { + "content": "<|reserved_special_token_100|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128109": { + "content": "<|reserved_special_token_101|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128110": { + "content": "<|reserved_special_token_102|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128111": { + "content": "<|reserved_special_token_103|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128112": { + "content": "<|reserved_special_token_104|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128113": { + "content": "<|reserved_special_token_105|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128114": { + "content": "<|reserved_special_token_106|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128115": { + "content": "<|reserved_special_token_107|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128116": { + "content": "<|reserved_special_token_108|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128117": { + "content": "<|reserved_special_token_109|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128118": { + "content": "<|reserved_special_token_110|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128119": { + "content": "<|reserved_special_token_111|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128120": { + "content": "<|reserved_special_token_112|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128121": { + "content": "<|reserved_special_token_113|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128122": { + "content": "<|reserved_special_token_114|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128123": { + "content": "<|reserved_special_token_115|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128124": { + "content": "<|reserved_special_token_116|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128125": { + "content": "<|reserved_special_token_117|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128126": { + "content": "<|reserved_special_token_118|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128127": { + "content": "<|reserved_special_token_119|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128128": { + "content": "<|reserved_special_token_120|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128129": { + "content": "<|reserved_special_token_121|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128130": { + "content": "<|reserved_special_token_122|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128131": { + "content": "<|reserved_special_token_123|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128132": { + "content": "<|reserved_special_token_124|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128133": { + "content": "<|reserved_special_token_125|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128134": { + "content": "<|reserved_special_token_126|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128135": { + "content": "<|reserved_special_token_127|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128136": { + "content": "<|reserved_special_token_128|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128137": { + "content": "<|reserved_special_token_129|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128138": { + "content": "<|reserved_special_token_130|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128139": { + "content": "<|reserved_special_token_131|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128140": { + "content": "<|reserved_special_token_132|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128141": { + "content": "<|reserved_special_token_133|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128142": { + "content": "<|reserved_special_token_134|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128143": { + "content": "<|reserved_special_token_135|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128144": { + "content": "<|reserved_special_token_136|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128145": { + "content": "<|reserved_special_token_137|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128146": { + "content": "<|reserved_special_token_138|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128147": { + "content": "<|reserved_special_token_139|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128148": { + "content": "<|reserved_special_token_140|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128149": { + "content": "<|reserved_special_token_141|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128150": { + "content": "<|reserved_special_token_142|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128151": { + "content": "<|reserved_special_token_143|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128152": { + "content": "<|reserved_special_token_144|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128153": { + "content": "<|reserved_special_token_145|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128154": { + "content": "<|reserved_special_token_146|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128155": { + "content": "<|reserved_special_token_147|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128156": { + "content": "<|reserved_special_token_148|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128157": { + "content": "<|reserved_special_token_149|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128158": { + "content": "<|reserved_special_token_150|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128159": { + "content": "<|reserved_special_token_151|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128160": { + "content": "<|reserved_special_token_152|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128161": { + "content": "<|reserved_special_token_153|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128162": { + "content": "<|reserved_special_token_154|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128163": { + "content": "<|reserved_special_token_155|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128164": { + "content": "<|reserved_special_token_156|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128165": { + "content": "<|reserved_special_token_157|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128166": { + "content": "<|reserved_special_token_158|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128167": { + "content": "<|reserved_special_token_159|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128168": { + "content": "<|reserved_special_token_160|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128169": { + "content": "<|reserved_special_token_161|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128170": { + "content": "<|reserved_special_token_162|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128171": { + "content": "<|reserved_special_token_163|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128172": { + "content": "<|reserved_special_token_164|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128173": { + "content": "<|reserved_special_token_165|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128174": { + "content": "<|reserved_special_token_166|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128175": { + "content": "<|reserved_special_token_167|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128176": { + "content": "<|reserved_special_token_168|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128177": { + "content": "<|reserved_special_token_169|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128178": { + "content": "<|reserved_special_token_170|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128179": { + "content": "<|reserved_special_token_171|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128180": { + "content": "<|reserved_special_token_172|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128181": { + "content": "<|reserved_special_token_173|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128182": { + "content": "<|reserved_special_token_174|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128183": { + "content": "<|reserved_special_token_175|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128184": { + "content": "<|reserved_special_token_176|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128185": { + "content": "<|reserved_special_token_177|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128186": { + "content": "<|reserved_special_token_178|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128187": { + "content": "<|reserved_special_token_179|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128188": { + "content": "<|reserved_special_token_180|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128189": { + "content": "<|reserved_special_token_181|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128190": { + "content": "<|reserved_special_token_182|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128191": { + "content": "<|reserved_special_token_183|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128192": { + "content": "<|reserved_special_token_184|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128193": { + "content": "<|reserved_special_token_185|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128194": { + "content": "<|reserved_special_token_186|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128195": { + "content": "<|reserved_special_token_187|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128196": { + "content": "<|reserved_special_token_188|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128197": { + "content": "<|reserved_special_token_189|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128198": { + "content": "<|reserved_special_token_190|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128199": { + "content": "<|reserved_special_token_191|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128200": { + "content": "<|reserved_special_token_192|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128201": { + "content": "<|reserved_special_token_193|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128202": { + "content": "<|reserved_special_token_194|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128203": { + "content": "<|reserved_special_token_195|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128204": { + "content": "<|reserved_special_token_196|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128205": { + "content": "<|reserved_special_token_197|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128206": { + "content": "<|reserved_special_token_198|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128207": { + "content": "<|reserved_special_token_199|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128208": { + "content": "<|reserved_special_token_200|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128209": { + "content": "<|reserved_special_token_201|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128210": { + "content": "<|reserved_special_token_202|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128211": { + "content": "<|reserved_special_token_203|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128212": { + "content": "<|reserved_special_token_204|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128213": { + "content": "<|reserved_special_token_205|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128214": { + "content": "<|reserved_special_token_206|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128215": { + "content": "<|reserved_special_token_207|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128216": { + "content": "<|reserved_special_token_208|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128217": { + "content": "<|reserved_special_token_209|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128218": { + "content": "<|reserved_special_token_210|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128219": { + "content": "<|reserved_special_token_211|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128220": { + "content": "<|reserved_special_token_212|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128221": { + "content": "<|reserved_special_token_213|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128222": { + "content": "<|reserved_special_token_214|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128223": { + "content": "<|reserved_special_token_215|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128224": { + "content": "<|reserved_special_token_216|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128225": { + "content": "<|reserved_special_token_217|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128226": { + "content": "<|reserved_special_token_218|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128227": { + "content": "<|reserved_special_token_219|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128228": { + "content": "<|reserved_special_token_220|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128229": { + "content": "<|reserved_special_token_221|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128230": { + "content": "<|reserved_special_token_222|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128231": { + "content": "<|reserved_special_token_223|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128232": { + "content": "<|reserved_special_token_224|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128233": { + "content": "<|reserved_special_token_225|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128234": { + "content": "<|reserved_special_token_226|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128235": { + "content": "<|reserved_special_token_227|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128236": { + "content": "<|reserved_special_token_228|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128237": { + "content": "<|reserved_special_token_229|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128238": { + "content": "<|reserved_special_token_230|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128239": { + "content": "<|reserved_special_token_231|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128240": { + "content": "<|reserved_special_token_232|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128241": { + "content": "<|reserved_special_token_233|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128242": { + "content": "<|reserved_special_token_234|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128243": { + "content": "<|reserved_special_token_235|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128244": { + "content": "<|reserved_special_token_236|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128245": { + "content": "<|reserved_special_token_237|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128246": { + "content": "<|reserved_special_token_238|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128247": { + "content": "<|reserved_special_token_239|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128248": { + "content": "<|reserved_special_token_240|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128249": { + "content": "<|reserved_special_token_241|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128250": { + "content": "<|reserved_special_token_242|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128251": { + "content": "<|reserved_special_token_243|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128252": { + "content": "<|reserved_special_token_244|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128253": { + "content": "<|reserved_special_token_245|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128254": { + "content": "<|reserved_special_token_246|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128255": { + "content": "<|reserved_special_token_247|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128256": { + "content": "[PAD]", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + } + }, + "bos_token": "<|begin_of_text|>", + "clean_up_tokenization_spaces": true, + "eos_token": "<|end_of_text|>", + "model_input_names": [ + "input_ids", + "attention_mask" + ], + "model_max_length": 131072, + "pad_token": "[PAD]", + "tokenizer_class": "PreTrainedTokenizerFast" +} diff --git a/babylm_shuffle_nondeterministic_10M_seed0/runs/checkpoint-600/trainer_state.json b/babylm_shuffle_nondeterministic_10M_seed0/runs/checkpoint-600/trainer_state.json new file mode 100644 index 0000000000000000000000000000000000000000..514cecc771a8e4a14c3fd43dc094a3e65b37598f --- /dev/null +++ b/babylm_shuffle_nondeterministic_10M_seed0/runs/checkpoint-600/trainer_state.json @@ -0,0 +1,4713 @@ +{ + "best_metric": null, + "best_model_checkpoint": null, + "epoch": 0.6575342465753424, + "eval_steps": 10, + "global_step": 600, + "is_hyper_param_search": false, + "is_local_process_zero": true, + "is_world_process_zero": true, + "log_history": [ + { + "epoch": 0.001095890410958904, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.0998, + "step": 1 + }, + { + "epoch": 0.002191780821917808, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.0858, + "step": 2 + }, + { + "epoch": 0.0032876712328767125, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.0672, + "step": 3 + }, + { + "epoch": 0.004383561643835616, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.0802, + "step": 4 + }, + { + "epoch": 0.005479452054794521, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.0636, + "step": 5 + }, + { + "epoch": 0.006575342465753425, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.1672, + "step": 6 + }, + { + "epoch": 0.007671232876712329, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.0961, + "step": 7 + }, + { + "epoch": 0.008767123287671232, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.1258, + "step": 8 + }, + { + "epoch": 0.009863013698630137, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.0953, + "step": 9 + }, + { + "epoch": 0.010958904109589041, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.1404, + "step": 10 + }, + { + "epoch": 0.010958904109589041, + "eval_loss": 3.12593412399292, + "eval_runtime": 42.732, + "eval_samples_per_second": 23.402, + "eval_steps_per_second": 0.983, + "step": 10 + }, + { + "epoch": 0.012054794520547946, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.1187, + "step": 11 + }, + { + "epoch": 0.01315068493150685, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.0593, + "step": 12 + }, + { + "epoch": 0.014246575342465753, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.111, + "step": 13 + }, + { + "epoch": 0.015342465753424657, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.0974, + "step": 14 + }, + { + "epoch": 0.01643835616438356, + "grad_norm": 5.991453170776367, + "learning_rate": 1.8248175182481753e-08, + "loss": 3.135, + "step": 15 + }, + { + "epoch": 0.017534246575342465, + "grad_norm": 6.274210453033447, + "learning_rate": 3.6496350364963505e-08, + "loss": 3.083, + "step": 16 + }, + { + "epoch": 0.01863013698630137, + "grad_norm": 6.644786834716797, + "learning_rate": 5.4744525547445264e-08, + "loss": 3.0856, + "step": 17 + }, + { + "epoch": 0.019726027397260273, + "grad_norm": 6.769156455993652, + "learning_rate": 7.299270072992701e-08, + "loss": 3.1082, + "step": 18 + }, + { + "epoch": 0.020821917808219178, + "grad_norm": 6.771246910095215, + "learning_rate": 9.124087591240876e-08, + "loss": 3.0753, + "step": 19 + }, + { + "epoch": 0.021917808219178082, + "grad_norm": 6.468441486358643, + "learning_rate": 1.0948905109489053e-07, + "loss": 3.1478, + "step": 20 + }, + { + "epoch": 0.021917808219178082, + "eval_loss": 3.1245107650756836, + "eval_runtime": 43.6251, + "eval_samples_per_second": 22.923, + "eval_steps_per_second": 0.963, + "step": 20 + }, + { + "epoch": 0.023013698630136987, + "grad_norm": 6.83078670501709, + "learning_rate": 1.2773722627737227e-07, + "loss": 3.1213, + "step": 21 + }, + { + "epoch": 0.02410958904109589, + "grad_norm": 6.221287250518799, + "learning_rate": 1.4598540145985402e-07, + "loss": 3.0549, + "step": 22 + }, + { + "epoch": 0.025205479452054796, + "grad_norm": 5.564266204833984, + "learning_rate": 1.642335766423358e-07, + "loss": 3.0769, + "step": 23 + }, + { + "epoch": 0.0263013698630137, + "grad_norm": 5.735209941864014, + "learning_rate": 1.8248175182481753e-07, + "loss": 3.1124, + "step": 24 + }, + { + "epoch": 0.0273972602739726, + "grad_norm": 5.577925682067871, + "learning_rate": 2.0072992700729928e-07, + "loss": 3.1191, + "step": 25 + }, + { + "epoch": 0.028493150684931506, + "grad_norm": 4.814139366149902, + "learning_rate": 2.1897810218978106e-07, + "loss": 3.0947, + "step": 26 + }, + { + "epoch": 0.02958904109589041, + "grad_norm": 4.69775390625, + "learning_rate": 2.3722627737226278e-07, + "loss": 3.0997, + "step": 27 + }, + { + "epoch": 0.030684931506849315, + "grad_norm": 4.811586380004883, + "learning_rate": 2.5547445255474454e-07, + "loss": 3.0749, + "step": 28 + }, + { + "epoch": 0.03178082191780822, + "grad_norm": 4.623377323150635, + "learning_rate": 2.737226277372263e-07, + "loss": 3.0619, + "step": 29 + }, + { + "epoch": 0.03287671232876712, + "grad_norm": 4.721253395080566, + "learning_rate": 2.9197080291970804e-07, + "loss": 3.0371, + "step": 30 + }, + { + "epoch": 0.03287671232876712, + "eval_loss": 3.082857847213745, + "eval_runtime": 44.127, + "eval_samples_per_second": 22.662, + "eval_steps_per_second": 0.952, + "step": 30 + }, + { + "epoch": 0.03397260273972603, + "grad_norm": 3.968630790710449, + "learning_rate": 3.102189781021898e-07, + "loss": 3.0535, + "step": 31 + }, + { + "epoch": 0.03506849315068493, + "grad_norm": 4.4373779296875, + "learning_rate": 3.284671532846716e-07, + "loss": 3.0476, + "step": 32 + }, + { + "epoch": 0.03616438356164384, + "grad_norm": 4.479320049285889, + "learning_rate": 3.467153284671533e-07, + "loss": 3.0215, + "step": 33 + }, + { + "epoch": 0.03726027397260274, + "grad_norm": 4.122231960296631, + "learning_rate": 3.6496350364963505e-07, + "loss": 3.0106, + "step": 34 + }, + { + "epoch": 0.038356164383561646, + "grad_norm": 4.153995513916016, + "learning_rate": 3.8321167883211686e-07, + "loss": 3.0318, + "step": 35 + }, + { + "epoch": 0.03945205479452055, + "grad_norm": 3.6949076652526855, + "learning_rate": 4.0145985401459856e-07, + "loss": 3.046, + "step": 36 + }, + { + "epoch": 0.040547945205479455, + "grad_norm": 3.4610488414764404, + "learning_rate": 4.197080291970803e-07, + "loss": 3.002, + "step": 37 + }, + { + "epoch": 0.041643835616438356, + "grad_norm": 3.9523138999938965, + "learning_rate": 4.379562043795621e-07, + "loss": 2.9665, + "step": 38 + }, + { + "epoch": 0.042739726027397264, + "grad_norm": 3.68516206741333, + "learning_rate": 4.562043795620438e-07, + "loss": 3.0411, + "step": 39 + }, + { + "epoch": 0.043835616438356165, + "grad_norm": 3.692396402359009, + "learning_rate": 4.7445255474452557e-07, + "loss": 2.9666, + "step": 40 + }, + { + "epoch": 0.043835616438356165, + "eval_loss": 3.015512704849243, + "eval_runtime": 48.2438, + "eval_samples_per_second": 20.728, + "eval_steps_per_second": 0.871, + "step": 40 + }, + { + "epoch": 0.044931506849315066, + "grad_norm": 3.3765921592712402, + "learning_rate": 4.927007299270073e-07, + "loss": 2.9831, + "step": 41 + }, + { + "epoch": 0.046027397260273974, + "grad_norm": 2.9631125926971436, + "learning_rate": 5.109489051094891e-07, + "loss": 3.0126, + "step": 42 + }, + { + "epoch": 0.047123287671232875, + "grad_norm": 2.969730854034424, + "learning_rate": 5.291970802919708e-07, + "loss": 2.9803, + "step": 43 + }, + { + "epoch": 0.04821917808219178, + "grad_norm": 3.185237407684326, + "learning_rate": 5.474452554744526e-07, + "loss": 2.9888, + "step": 44 + }, + { + "epoch": 0.049315068493150684, + "grad_norm": 3.055579423904419, + "learning_rate": 5.656934306569343e-07, + "loss": 2.9715, + "step": 45 + }, + { + "epoch": 0.05041095890410959, + "grad_norm": 3.083284616470337, + "learning_rate": 5.839416058394161e-07, + "loss": 2.9293, + "step": 46 + }, + { + "epoch": 0.05150684931506849, + "grad_norm": 2.984781503677368, + "learning_rate": 6.021897810218979e-07, + "loss": 2.9228, + "step": 47 + }, + { + "epoch": 0.0526027397260274, + "grad_norm": 2.756173849105835, + "learning_rate": 6.204379562043796e-07, + "loss": 2.9406, + "step": 48 + }, + { + "epoch": 0.0536986301369863, + "grad_norm": 2.911132574081421, + "learning_rate": 6.386861313868614e-07, + "loss": 2.8829, + "step": 49 + }, + { + "epoch": 0.0547945205479452, + "grad_norm": 2.592890977859497, + "learning_rate": 6.569343065693432e-07, + "loss": 2.9839, + "step": 50 + }, + { + "epoch": 0.0547945205479452, + "eval_loss": 2.9409468173980713, + "eval_runtime": 47.1045, + "eval_samples_per_second": 21.229, + "eval_steps_per_second": 0.892, + "step": 50 + }, + { + "epoch": 0.05589041095890411, + "grad_norm": 3.0112955570220947, + "learning_rate": 6.751824817518248e-07, + "loss": 2.8847, + "step": 51 + }, + { + "epoch": 0.05698630136986301, + "grad_norm": 2.3701798915863037, + "learning_rate": 6.934306569343066e-07, + "loss": 2.9109, + "step": 52 + }, + { + "epoch": 0.05808219178082192, + "grad_norm": 2.268850088119507, + "learning_rate": 7.116788321167883e-07, + "loss": 2.8483, + "step": 53 + }, + { + "epoch": 0.05917808219178082, + "grad_norm": 2.419015645980835, + "learning_rate": 7.299270072992701e-07, + "loss": 2.8491, + "step": 54 + }, + { + "epoch": 0.06027397260273973, + "grad_norm": 2.4582653045654297, + "learning_rate": 7.48175182481752e-07, + "loss": 2.8622, + "step": 55 + }, + { + "epoch": 0.06136986301369863, + "grad_norm": 2.3471553325653076, + "learning_rate": 7.664233576642337e-07, + "loss": 2.8738, + "step": 56 + }, + { + "epoch": 0.06246575342465754, + "grad_norm": 2.479177236557007, + "learning_rate": 7.846715328467154e-07, + "loss": 2.8843, + "step": 57 + }, + { + "epoch": 0.06356164383561644, + "grad_norm": 2.1676292419433594, + "learning_rate": 8.029197080291971e-07, + "loss": 2.8674, + "step": 58 + }, + { + "epoch": 0.06465753424657535, + "grad_norm": 1.9862910509109497, + "learning_rate": 8.211678832116789e-07, + "loss": 2.8197, + "step": 59 + }, + { + "epoch": 0.06575342465753424, + "grad_norm": 2.211313486099243, + "learning_rate": 8.394160583941606e-07, + "loss": 2.7961, + "step": 60 + }, + { + "epoch": 0.06575342465753424, + "eval_loss": 2.880824327468872, + "eval_runtime": 51.3235, + "eval_samples_per_second": 19.484, + "eval_steps_per_second": 0.818, + "step": 60 + }, + { + "epoch": 0.06684931506849315, + "grad_norm": 2.3497281074523926, + "learning_rate": 8.576642335766425e-07, + "loss": 2.8233, + "step": 61 + }, + { + "epoch": 0.06794520547945206, + "grad_norm": 2.158515214920044, + "learning_rate": 8.759124087591242e-07, + "loss": 2.8136, + "step": 62 + }, + { + "epoch": 0.06904109589041096, + "grad_norm": 2.0219955444335938, + "learning_rate": 8.94160583941606e-07, + "loss": 2.8231, + "step": 63 + }, + { + "epoch": 0.07013698630136986, + "grad_norm": 1.8989133834838867, + "learning_rate": 9.124087591240876e-07, + "loss": 2.8268, + "step": 64 + }, + { + "epoch": 0.07123287671232877, + "grad_norm": 2.243851661682129, + "learning_rate": 9.306569343065694e-07, + "loss": 2.7886, + "step": 65 + }, + { + "epoch": 0.07232876712328767, + "grad_norm": 2.122464418411255, + "learning_rate": 9.489051094890511e-07, + "loss": 2.8166, + "step": 66 + }, + { + "epoch": 0.07342465753424658, + "grad_norm": 1.899857997894287, + "learning_rate": 9.67153284671533e-07, + "loss": 2.7997, + "step": 67 + }, + { + "epoch": 0.07452054794520548, + "grad_norm": 2.4424173831939697, + "learning_rate": 9.854014598540146e-07, + "loss": 2.8342, + "step": 68 + }, + { + "epoch": 0.07561643835616438, + "grad_norm": 2.5677411556243896, + "learning_rate": 1.0036496350364965e-06, + "loss": 2.7786, + "step": 69 + }, + { + "epoch": 0.07671232876712329, + "grad_norm": 1.9989523887634277, + "learning_rate": 1.0218978102189781e-06, + "loss": 2.7411, + "step": 70 + }, + { + "epoch": 0.07671232876712329, + "eval_loss": 2.8185322284698486, + "eval_runtime": 50.4998, + "eval_samples_per_second": 19.802, + "eval_steps_per_second": 0.832, + "step": 70 + }, + { + "epoch": 0.07780821917808219, + "grad_norm": 1.9376521110534668, + "learning_rate": 1.04014598540146e-06, + "loss": 2.7988, + "step": 71 + }, + { + "epoch": 0.0789041095890411, + "grad_norm": 2.5629725456237793, + "learning_rate": 1.0583941605839416e-06, + "loss": 2.7079, + "step": 72 + }, + { + "epoch": 0.08, + "grad_norm": 2.1634678840637207, + "learning_rate": 1.0766423357664233e-06, + "loss": 2.7049, + "step": 73 + }, + { + "epoch": 0.08109589041095891, + "grad_norm": 2.2143008708953857, + "learning_rate": 1.0948905109489052e-06, + "loss": 2.72, + "step": 74 + }, + { + "epoch": 0.0821917808219178, + "grad_norm": 2.2701101303100586, + "learning_rate": 1.113138686131387e-06, + "loss": 2.7418, + "step": 75 + }, + { + "epoch": 0.08328767123287671, + "grad_norm": 1.716569423675537, + "learning_rate": 1.1313868613138687e-06, + "loss": 2.7558, + "step": 76 + }, + { + "epoch": 0.08438356164383562, + "grad_norm": 2.79757022857666, + "learning_rate": 1.1496350364963505e-06, + "loss": 2.6865, + "step": 77 + }, + { + "epoch": 0.08547945205479453, + "grad_norm": 2.2189838886260986, + "learning_rate": 1.1678832116788322e-06, + "loss": 2.6684, + "step": 78 + }, + { + "epoch": 0.08657534246575342, + "grad_norm": 2.0695579051971436, + "learning_rate": 1.1861313868613138e-06, + "loss": 2.6779, + "step": 79 + }, + { + "epoch": 0.08767123287671233, + "grad_norm": 2.372326374053955, + "learning_rate": 1.2043795620437959e-06, + "loss": 2.6711, + "step": 80 + }, + { + "epoch": 0.08767123287671233, + "eval_loss": 2.7537453174591064, + "eval_runtime": 47.7274, + "eval_samples_per_second": 20.952, + "eval_steps_per_second": 0.88, + "step": 80 + }, + { + "epoch": 0.08876712328767124, + "grad_norm": 2.3398947715759277, + "learning_rate": 1.2226277372262775e-06, + "loss": 2.6776, + "step": 81 + }, + { + "epoch": 0.08986301369863013, + "grad_norm": 2.6619036197662354, + "learning_rate": 1.2408759124087592e-06, + "loss": 2.7077, + "step": 82 + }, + { + "epoch": 0.09095890410958904, + "grad_norm": 1.8245115280151367, + "learning_rate": 1.259124087591241e-06, + "loss": 2.662, + "step": 83 + }, + { + "epoch": 0.09205479452054795, + "grad_norm": 2.2627148628234863, + "learning_rate": 1.2773722627737229e-06, + "loss": 2.677, + "step": 84 + }, + { + "epoch": 0.09315068493150686, + "grad_norm": 2.4983835220336914, + "learning_rate": 1.2956204379562045e-06, + "loss": 2.6601, + "step": 85 + }, + { + "epoch": 0.09424657534246575, + "grad_norm": 2.147010087966919, + "learning_rate": 1.3138686131386864e-06, + "loss": 2.6213, + "step": 86 + }, + { + "epoch": 0.09534246575342466, + "grad_norm": 2.791682481765747, + "learning_rate": 1.332116788321168e-06, + "loss": 2.6281, + "step": 87 + }, + { + "epoch": 0.09643835616438357, + "grad_norm": 2.023970365524292, + "learning_rate": 1.3503649635036497e-06, + "loss": 2.5766, + "step": 88 + }, + { + "epoch": 0.09753424657534246, + "grad_norm": 2.8143956661224365, + "learning_rate": 1.3686131386861315e-06, + "loss": 2.6086, + "step": 89 + }, + { + "epoch": 0.09863013698630137, + "grad_norm": 2.226106882095337, + "learning_rate": 1.3868613138686132e-06, + "loss": 2.5477, + "step": 90 + }, + { + "epoch": 0.09863013698630137, + "eval_loss": 2.6880786418914795, + "eval_runtime": 47.8417, + "eval_samples_per_second": 20.902, + "eval_steps_per_second": 0.878, + "step": 90 + }, + { + "epoch": 0.09972602739726028, + "grad_norm": 2.5248148441314697, + "learning_rate": 1.405109489051095e-06, + "loss": 2.6232, + "step": 91 + }, + { + "epoch": 0.10082191780821918, + "grad_norm": 2.421330690383911, + "learning_rate": 1.4233576642335767e-06, + "loss": 2.599, + "step": 92 + }, + { + "epoch": 0.10191780821917808, + "grad_norm": 2.5562429428100586, + "learning_rate": 1.4416058394160583e-06, + "loss": 2.6197, + "step": 93 + }, + { + "epoch": 0.10301369863013699, + "grad_norm": 2.856201171875, + "learning_rate": 1.4598540145985402e-06, + "loss": 2.5825, + "step": 94 + }, + { + "epoch": 0.10410958904109589, + "grad_norm": 2.6576807498931885, + "learning_rate": 1.478102189781022e-06, + "loss": 2.5496, + "step": 95 + }, + { + "epoch": 0.1052054794520548, + "grad_norm": 3.1408190727233887, + "learning_rate": 1.496350364963504e-06, + "loss": 2.6142, + "step": 96 + }, + { + "epoch": 0.1063013698630137, + "grad_norm": 2.3270952701568604, + "learning_rate": 1.5145985401459856e-06, + "loss": 2.584, + "step": 97 + }, + { + "epoch": 0.1073972602739726, + "grad_norm": 2.3976614475250244, + "learning_rate": 1.5328467153284674e-06, + "loss": 2.5941, + "step": 98 + }, + { + "epoch": 0.10849315068493151, + "grad_norm": 2.783485174179077, + "learning_rate": 1.551094890510949e-06, + "loss": 2.5626, + "step": 99 + }, + { + "epoch": 0.1095890410958904, + "grad_norm": 2.2522788047790527, + "learning_rate": 1.5693430656934307e-06, + "loss": 2.517, + "step": 100 + }, + { + "epoch": 0.1095890410958904, + "eval_loss": 2.637315034866333, + "eval_runtime": 48.2107, + "eval_samples_per_second": 20.742, + "eval_steps_per_second": 0.871, + "step": 100 + }, + { + "epoch": 0.11068493150684931, + "grad_norm": 2.917799949645996, + "learning_rate": 1.5875912408759126e-06, + "loss": 2.5159, + "step": 101 + }, + { + "epoch": 0.11178082191780822, + "grad_norm": 2.955815076828003, + "learning_rate": 1.6058394160583942e-06, + "loss": 2.563, + "step": 102 + }, + { + "epoch": 0.11287671232876713, + "grad_norm": 2.8096001148223877, + "learning_rate": 1.624087591240876e-06, + "loss": 2.5995, + "step": 103 + }, + { + "epoch": 0.11397260273972602, + "grad_norm": 2.4967427253723145, + "learning_rate": 1.6423357664233577e-06, + "loss": 2.5336, + "step": 104 + }, + { + "epoch": 0.11506849315068493, + "grad_norm": 2.740717887878418, + "learning_rate": 1.6605839416058394e-06, + "loss": 2.4778, + "step": 105 + }, + { + "epoch": 0.11616438356164384, + "grad_norm": 2.5064632892608643, + "learning_rate": 1.6788321167883212e-06, + "loss": 2.5214, + "step": 106 + }, + { + "epoch": 0.11726027397260275, + "grad_norm": 2.2923007011413574, + "learning_rate": 1.6970802919708029e-06, + "loss": 2.5315, + "step": 107 + }, + { + "epoch": 0.11835616438356164, + "grad_norm": 3.339569568634033, + "learning_rate": 1.715328467153285e-06, + "loss": 2.4661, + "step": 108 + }, + { + "epoch": 0.11945205479452055, + "grad_norm": 2.1650891304016113, + "learning_rate": 1.7335766423357666e-06, + "loss": 2.5072, + "step": 109 + }, + { + "epoch": 0.12054794520547946, + "grad_norm": 3.1216914653778076, + "learning_rate": 1.7518248175182485e-06, + "loss": 2.5164, + "step": 110 + }, + { + "epoch": 0.12054794520547946, + "eval_loss": 2.594250202178955, + "eval_runtime": 41.8939, + "eval_samples_per_second": 23.87, + "eval_steps_per_second": 1.003, + "step": 110 + }, + { + "epoch": 0.12164383561643835, + "grad_norm": 2.191063404083252, + "learning_rate": 1.77007299270073e-06, + "loss": 2.5422, + "step": 111 + }, + { + "epoch": 0.12273972602739726, + "grad_norm": 2.922083854675293, + "learning_rate": 1.788321167883212e-06, + "loss": 2.5258, + "step": 112 + }, + { + "epoch": 0.12383561643835617, + "grad_norm": 2.518002986907959, + "learning_rate": 1.8065693430656936e-06, + "loss": 2.5588, + "step": 113 + }, + { + "epoch": 0.12493150684931507, + "grad_norm": 2.515194892883301, + "learning_rate": 1.8248175182481753e-06, + "loss": 2.5079, + "step": 114 + }, + { + "epoch": 0.12602739726027398, + "grad_norm": 2.5530874729156494, + "learning_rate": 1.8430656934306571e-06, + "loss": 2.4615, + "step": 115 + }, + { + "epoch": 0.12712328767123288, + "grad_norm": 2.3550198078155518, + "learning_rate": 1.8613138686131388e-06, + "loss": 2.4825, + "step": 116 + }, + { + "epoch": 0.12821917808219177, + "grad_norm": 2.295145034790039, + "learning_rate": 1.8795620437956206e-06, + "loss": 2.4749, + "step": 117 + }, + { + "epoch": 0.1293150684931507, + "grad_norm": 2.0158982276916504, + "learning_rate": 1.8978102189781023e-06, + "loss": 2.4756, + "step": 118 + }, + { + "epoch": 0.1304109589041096, + "grad_norm": 2.213057518005371, + "learning_rate": 1.916058394160584e-06, + "loss": 2.4744, + "step": 119 + }, + { + "epoch": 0.13150684931506848, + "grad_norm": 2.186222553253174, + "learning_rate": 1.934306569343066e-06, + "loss": 2.4957, + "step": 120 + }, + { + "epoch": 0.13150684931506848, + "eval_loss": 2.563931465148926, + "eval_runtime": 44.8714, + "eval_samples_per_second": 22.286, + "eval_steps_per_second": 0.936, + "step": 120 + }, + { + "epoch": 0.1326027397260274, + "grad_norm": 2.3143310546875, + "learning_rate": 1.952554744525548e-06, + "loss": 2.4, + "step": 121 + }, + { + "epoch": 0.1336986301369863, + "grad_norm": 2.0223965644836426, + "learning_rate": 1.9708029197080293e-06, + "loss": 2.4453, + "step": 122 + }, + { + "epoch": 0.13479452054794522, + "grad_norm": 2.2025411128997803, + "learning_rate": 1.989051094890511e-06, + "loss": 2.4463, + "step": 123 + }, + { + "epoch": 0.1358904109589041, + "grad_norm": 2.707930564880371, + "learning_rate": 2.007299270072993e-06, + "loss": 2.4589, + "step": 124 + }, + { + "epoch": 0.136986301369863, + "grad_norm": 2.146662473678589, + "learning_rate": 2.025547445255475e-06, + "loss": 2.5023, + "step": 125 + }, + { + "epoch": 0.13808219178082193, + "grad_norm": 2.520660161972046, + "learning_rate": 2.0437956204379563e-06, + "loss": 2.4464, + "step": 126 + }, + { + "epoch": 0.13917808219178082, + "grad_norm": 2.50632381439209, + "learning_rate": 2.062043795620438e-06, + "loss": 2.4556, + "step": 127 + }, + { + "epoch": 0.14027397260273972, + "grad_norm": 2.2919397354125977, + "learning_rate": 2.08029197080292e-06, + "loss": 2.4479, + "step": 128 + }, + { + "epoch": 0.14136986301369864, + "grad_norm": 2.521475315093994, + "learning_rate": 2.0985401459854014e-06, + "loss": 2.4982, + "step": 129 + }, + { + "epoch": 0.14246575342465753, + "grad_norm": 2.3812668323516846, + "learning_rate": 2.1167883211678833e-06, + "loss": 2.4163, + "step": 130 + }, + { + "epoch": 0.14246575342465753, + "eval_loss": 2.5390918254852295, + "eval_runtime": 43.8746, + "eval_samples_per_second": 22.792, + "eval_steps_per_second": 0.957, + "step": 130 + }, + { + "epoch": 0.14356164383561643, + "grad_norm": 2.302286148071289, + "learning_rate": 2.135036496350365e-06, + "loss": 2.4588, + "step": 131 + }, + { + "epoch": 0.14465753424657535, + "grad_norm": 2.4222044944763184, + "learning_rate": 2.1532846715328466e-06, + "loss": 2.492, + "step": 132 + }, + { + "epoch": 0.14575342465753424, + "grad_norm": 2.605835199356079, + "learning_rate": 2.171532846715329e-06, + "loss": 2.4055, + "step": 133 + }, + { + "epoch": 0.14684931506849316, + "grad_norm": 3.6261861324310303, + "learning_rate": 2.1897810218978103e-06, + "loss": 2.4221, + "step": 134 + }, + { + "epoch": 0.14794520547945206, + "grad_norm": 2.7647221088409424, + "learning_rate": 2.208029197080292e-06, + "loss": 2.4549, + "step": 135 + }, + { + "epoch": 0.14904109589041095, + "grad_norm": 3.2977371215820312, + "learning_rate": 2.226277372262774e-06, + "loss": 2.4331, + "step": 136 + }, + { + "epoch": 0.15013698630136987, + "grad_norm": 3.147094488143921, + "learning_rate": 2.244525547445256e-06, + "loss": 2.4268, + "step": 137 + }, + { + "epoch": 0.15123287671232877, + "grad_norm": 2.6363413333892822, + "learning_rate": 2.2627737226277373e-06, + "loss": 2.4524, + "step": 138 + }, + { + "epoch": 0.15232876712328766, + "grad_norm": 2.9288904666900635, + "learning_rate": 2.281021897810219e-06, + "loss": 2.4534, + "step": 139 + }, + { + "epoch": 0.15342465753424658, + "grad_norm": 2.5604870319366455, + "learning_rate": 2.299270072992701e-06, + "loss": 2.4138, + "step": 140 + }, + { + "epoch": 0.15342465753424658, + "eval_loss": 2.524790048599243, + "eval_runtime": 43.8851, + "eval_samples_per_second": 22.787, + "eval_steps_per_second": 0.957, + "step": 140 + }, + { + "epoch": 0.15452054794520548, + "grad_norm": 3.2746663093566895, + "learning_rate": 2.3175182481751825e-06, + "loss": 2.4094, + "step": 141 + }, + { + "epoch": 0.15561643835616437, + "grad_norm": 2.691483497619629, + "learning_rate": 2.3357664233576643e-06, + "loss": 2.4265, + "step": 142 + }, + { + "epoch": 0.1567123287671233, + "grad_norm": 2.5850117206573486, + "learning_rate": 2.354014598540146e-06, + "loss": 2.3922, + "step": 143 + }, + { + "epoch": 0.1578082191780822, + "grad_norm": 2.68204927444458, + "learning_rate": 2.3722627737226276e-06, + "loss": 2.4262, + "step": 144 + }, + { + "epoch": 0.1589041095890411, + "grad_norm": 2.952584981918335, + "learning_rate": 2.39051094890511e-06, + "loss": 2.4267, + "step": 145 + }, + { + "epoch": 0.16, + "grad_norm": 3.2354788780212402, + "learning_rate": 2.4087591240875918e-06, + "loss": 2.3836, + "step": 146 + }, + { + "epoch": 0.1610958904109589, + "grad_norm": 2.634521007537842, + "learning_rate": 2.427007299270073e-06, + "loss": 2.3953, + "step": 147 + }, + { + "epoch": 0.16219178082191782, + "grad_norm": 3.1115682125091553, + "learning_rate": 2.445255474452555e-06, + "loss": 2.409, + "step": 148 + }, + { + "epoch": 0.1632876712328767, + "grad_norm": 2.6824135780334473, + "learning_rate": 2.463503649635037e-06, + "loss": 2.4459, + "step": 149 + }, + { + "epoch": 0.1643835616438356, + "grad_norm": 2.617614507675171, + "learning_rate": 2.4817518248175183e-06, + "loss": 2.366, + "step": 150 + }, + { + "epoch": 0.1643835616438356, + "eval_loss": 2.5088160037994385, + "eval_runtime": 44.0888, + "eval_samples_per_second": 22.682, + "eval_steps_per_second": 0.953, + "step": 150 + }, + { + "epoch": 0.16547945205479453, + "grad_norm": 3.976992130279541, + "learning_rate": 2.5e-06, + "loss": 2.3517, + "step": 151 + }, + { + "epoch": 0.16657534246575342, + "grad_norm": 3.1433017253875732, + "learning_rate": 2.518248175182482e-06, + "loss": 2.4489, + "step": 152 + }, + { + "epoch": 0.16767123287671232, + "grad_norm": 4.100688934326172, + "learning_rate": 2.5364963503649635e-06, + "loss": 2.3683, + "step": 153 + }, + { + "epoch": 0.16876712328767124, + "grad_norm": 4.316102504730225, + "learning_rate": 2.5547445255474458e-06, + "loss": 2.3663, + "step": 154 + }, + { + "epoch": 0.16986301369863013, + "grad_norm": 2.3103132247924805, + "learning_rate": 2.5729927007299272e-06, + "loss": 2.4009, + "step": 155 + }, + { + "epoch": 0.17095890410958905, + "grad_norm": 3.576810836791992, + "learning_rate": 2.591240875912409e-06, + "loss": 2.3983, + "step": 156 + }, + { + "epoch": 0.17205479452054795, + "grad_norm": 2.337080955505371, + "learning_rate": 2.6094890510948905e-06, + "loss": 2.3849, + "step": 157 + }, + { + "epoch": 0.17315068493150684, + "grad_norm": 3.2523117065429688, + "learning_rate": 2.627737226277373e-06, + "loss": 2.4171, + "step": 158 + }, + { + "epoch": 0.17424657534246576, + "grad_norm": 2.529175281524658, + "learning_rate": 2.6459854014598542e-06, + "loss": 2.4303, + "step": 159 + }, + { + "epoch": 0.17534246575342466, + "grad_norm": 3.052212953567505, + "learning_rate": 2.664233576642336e-06, + "loss": 2.3895, + "step": 160 + }, + { + "epoch": 0.17534246575342466, + "eval_loss": 2.492485761642456, + "eval_runtime": 45.6738, + "eval_samples_per_second": 21.894, + "eval_steps_per_second": 0.92, + "step": 160 + }, + { + "epoch": 0.17643835616438355, + "grad_norm": 2.1544973850250244, + "learning_rate": 2.6824817518248175e-06, + "loss": 2.3649, + "step": 161 + }, + { + "epoch": 0.17753424657534247, + "grad_norm": 2.7351951599121094, + "learning_rate": 2.7007299270072994e-06, + "loss": 2.3908, + "step": 162 + }, + { + "epoch": 0.17863013698630137, + "grad_norm": 2.8431951999664307, + "learning_rate": 2.718978102189781e-06, + "loss": 2.374, + "step": 163 + }, + { + "epoch": 0.17972602739726026, + "grad_norm": 3.083190441131592, + "learning_rate": 2.737226277372263e-06, + "loss": 2.4193, + "step": 164 + }, + { + "epoch": 0.18082191780821918, + "grad_norm": 2.6523995399475098, + "learning_rate": 2.755474452554745e-06, + "loss": 2.3969, + "step": 165 + }, + { + "epoch": 0.18191780821917808, + "grad_norm": 2.626047372817993, + "learning_rate": 2.7737226277372264e-06, + "loss": 2.3742, + "step": 166 + }, + { + "epoch": 0.18301369863013697, + "grad_norm": 3.300917863845825, + "learning_rate": 2.7919708029197082e-06, + "loss": 2.4546, + "step": 167 + }, + { + "epoch": 0.1841095890410959, + "grad_norm": 2.797304391860962, + "learning_rate": 2.81021897810219e-06, + "loss": 2.4161, + "step": 168 + }, + { + "epoch": 0.1852054794520548, + "grad_norm": 2.777117967605591, + "learning_rate": 2.828467153284672e-06, + "loss": 2.3572, + "step": 169 + }, + { + "epoch": 0.1863013698630137, + "grad_norm": 2.887470006942749, + "learning_rate": 2.8467153284671534e-06, + "loss": 2.3887, + "step": 170 + }, + { + "epoch": 0.1863013698630137, + "eval_loss": 2.4854257106781006, + "eval_runtime": 44.5644, + "eval_samples_per_second": 22.439, + "eval_steps_per_second": 0.942, + "step": 170 + }, + { + "epoch": 0.1873972602739726, + "grad_norm": 3.0706076622009277, + "learning_rate": 2.8649635036496353e-06, + "loss": 2.3162, + "step": 171 + }, + { + "epoch": 0.1884931506849315, + "grad_norm": 3.6129744052886963, + "learning_rate": 2.8832116788321167e-06, + "loss": 2.3685, + "step": 172 + }, + { + "epoch": 0.18958904109589042, + "grad_norm": 2.4936370849609375, + "learning_rate": 2.901459854014599e-06, + "loss": 2.3297, + "step": 173 + }, + { + "epoch": 0.19068493150684931, + "grad_norm": 2.1041860580444336, + "learning_rate": 2.9197080291970804e-06, + "loss": 2.4085, + "step": 174 + }, + { + "epoch": 0.1917808219178082, + "grad_norm": 2.2066831588745117, + "learning_rate": 2.9379562043795623e-06, + "loss": 2.3905, + "step": 175 + }, + { + "epoch": 0.19287671232876713, + "grad_norm": 2.444652795791626, + "learning_rate": 2.956204379562044e-06, + "loss": 2.378, + "step": 176 + }, + { + "epoch": 0.19397260273972602, + "grad_norm": 2.501939058303833, + "learning_rate": 2.9744525547445256e-06, + "loss": 2.4136, + "step": 177 + }, + { + "epoch": 0.19506849315068492, + "grad_norm": 2.2167556285858154, + "learning_rate": 2.992700729927008e-06, + "loss": 2.385, + "step": 178 + }, + { + "epoch": 0.19616438356164384, + "grad_norm": 2.5298545360565186, + "learning_rate": 3.0109489051094893e-06, + "loss": 2.3508, + "step": 179 + }, + { + "epoch": 0.19726027397260273, + "grad_norm": 2.3774642944335938, + "learning_rate": 3.029197080291971e-06, + "loss": 2.3421, + "step": 180 + }, + { + "epoch": 0.19726027397260273, + "eval_loss": 2.476067304611206, + "eval_runtime": 44.3761, + "eval_samples_per_second": 22.535, + "eval_steps_per_second": 0.946, + "step": 180 + }, + { + "epoch": 0.19835616438356166, + "grad_norm": 2.9108712673187256, + "learning_rate": 3.0474452554744526e-06, + "loss": 2.3983, + "step": 181 + }, + { + "epoch": 0.19945205479452055, + "grad_norm": 2.5232033729553223, + "learning_rate": 3.065693430656935e-06, + "loss": 2.3946, + "step": 182 + }, + { + "epoch": 0.20054794520547944, + "grad_norm": 2.158168077468872, + "learning_rate": 3.0839416058394163e-06, + "loss": 2.3805, + "step": 183 + }, + { + "epoch": 0.20164383561643837, + "grad_norm": 2.3405354022979736, + "learning_rate": 3.102189781021898e-06, + "loss": 2.3974, + "step": 184 + }, + { + "epoch": 0.20273972602739726, + "grad_norm": 2.5308327674865723, + "learning_rate": 3.1204379562043796e-06, + "loss": 2.384, + "step": 185 + }, + { + "epoch": 0.20383561643835615, + "grad_norm": 2.6328237056732178, + "learning_rate": 3.1386861313868614e-06, + "loss": 2.3345, + "step": 186 + }, + { + "epoch": 0.20493150684931508, + "grad_norm": 3.0725512504577637, + "learning_rate": 3.156934306569343e-06, + "loss": 2.3191, + "step": 187 + }, + { + "epoch": 0.20602739726027397, + "grad_norm": 2.196218729019165, + "learning_rate": 3.175182481751825e-06, + "loss": 2.4073, + "step": 188 + }, + { + "epoch": 0.20712328767123286, + "grad_norm": 2.7722432613372803, + "learning_rate": 3.193430656934307e-06, + "loss": 2.3775, + "step": 189 + }, + { + "epoch": 0.20821917808219179, + "grad_norm": 2.4848241806030273, + "learning_rate": 3.2116788321167884e-06, + "loss": 2.3538, + "step": 190 + }, + { + "epoch": 0.20821917808219179, + "eval_loss": 2.462618112564087, + "eval_runtime": 43.9461, + "eval_samples_per_second": 22.755, + "eval_steps_per_second": 0.956, + "step": 190 + }, + { + "epoch": 0.20931506849315068, + "grad_norm": 2.4682741165161133, + "learning_rate": 3.2299270072992707e-06, + "loss": 2.3688, + "step": 191 + }, + { + "epoch": 0.2104109589041096, + "grad_norm": 2.518923282623291, + "learning_rate": 3.248175182481752e-06, + "loss": 2.3556, + "step": 192 + }, + { + "epoch": 0.2115068493150685, + "grad_norm": 2.4023561477661133, + "learning_rate": 3.266423357664234e-06, + "loss": 2.292, + "step": 193 + }, + { + "epoch": 0.2126027397260274, + "grad_norm": 2.5196943283081055, + "learning_rate": 3.2846715328467155e-06, + "loss": 2.3703, + "step": 194 + }, + { + "epoch": 0.2136986301369863, + "grad_norm": 2.762709617614746, + "learning_rate": 3.3029197080291973e-06, + "loss": 2.3578, + "step": 195 + }, + { + "epoch": 0.2147945205479452, + "grad_norm": 2.1287691593170166, + "learning_rate": 3.3211678832116788e-06, + "loss": 2.3687, + "step": 196 + }, + { + "epoch": 0.2158904109589041, + "grad_norm": 2.477341651916504, + "learning_rate": 3.339416058394161e-06, + "loss": 2.3545, + "step": 197 + }, + { + "epoch": 0.21698630136986302, + "grad_norm": 2.23933744430542, + "learning_rate": 3.3576642335766425e-06, + "loss": 2.3511, + "step": 198 + }, + { + "epoch": 0.21808219178082192, + "grad_norm": 2.2440133094787598, + "learning_rate": 3.3759124087591243e-06, + "loss": 2.3539, + "step": 199 + }, + { + "epoch": 0.2191780821917808, + "grad_norm": 2.5603177547454834, + "learning_rate": 3.3941605839416058e-06, + "loss": 2.36, + "step": 200 + }, + { + "epoch": 0.2191780821917808, + "eval_loss": 2.4507246017456055, + "eval_runtime": 43.7786, + "eval_samples_per_second": 22.842, + "eval_steps_per_second": 0.959, + "step": 200 + }, + { + "epoch": 0.22027397260273973, + "grad_norm": 2.4570510387420654, + "learning_rate": 3.412408759124088e-06, + "loss": 2.3702, + "step": 201 + }, + { + "epoch": 0.22136986301369863, + "grad_norm": 2.1975133419036865, + "learning_rate": 3.43065693430657e-06, + "loss": 2.3667, + "step": 202 + }, + { + "epoch": 0.22246575342465755, + "grad_norm": 2.710186004638672, + "learning_rate": 3.4489051094890513e-06, + "loss": 2.3159, + "step": 203 + }, + { + "epoch": 0.22356164383561644, + "grad_norm": 2.7799506187438965, + "learning_rate": 3.467153284671533e-06, + "loss": 2.3519, + "step": 204 + }, + { + "epoch": 0.22465753424657534, + "grad_norm": 2.2479400634765625, + "learning_rate": 3.4854014598540146e-06, + "loss": 2.3103, + "step": 205 + }, + { + "epoch": 0.22575342465753426, + "grad_norm": 3.1538405418395996, + "learning_rate": 3.503649635036497e-06, + "loss": 2.3506, + "step": 206 + }, + { + "epoch": 0.22684931506849315, + "grad_norm": 2.7607762813568115, + "learning_rate": 3.5218978102189783e-06, + "loss": 2.3386, + "step": 207 + }, + { + "epoch": 0.22794520547945205, + "grad_norm": 2.371105909347534, + "learning_rate": 3.54014598540146e-06, + "loss": 2.3583, + "step": 208 + }, + { + "epoch": 0.22904109589041097, + "grad_norm": 2.793933391571045, + "learning_rate": 3.5583941605839416e-06, + "loss": 2.3782, + "step": 209 + }, + { + "epoch": 0.23013698630136986, + "grad_norm": 2.372732400894165, + "learning_rate": 3.576642335766424e-06, + "loss": 2.417, + "step": 210 + }, + { + "epoch": 0.23013698630136986, + "eval_loss": 2.451474189758301, + "eval_runtime": 41.7197, + "eval_samples_per_second": 23.97, + "eval_steps_per_second": 1.007, + "step": 210 + }, + { + "epoch": 0.23123287671232876, + "grad_norm": 2.6989448070526123, + "learning_rate": 3.5948905109489054e-06, + "loss": 2.393, + "step": 211 + }, + { + "epoch": 0.23232876712328768, + "grad_norm": 2.4324285984039307, + "learning_rate": 3.6131386861313872e-06, + "loss": 2.2738, + "step": 212 + }, + { + "epoch": 0.23342465753424657, + "grad_norm": 3.5063016414642334, + "learning_rate": 3.6313868613138687e-06, + "loss": 2.2823, + "step": 213 + }, + { + "epoch": 0.2345205479452055, + "grad_norm": 3.953951358795166, + "learning_rate": 3.6496350364963505e-06, + "loss": 2.3132, + "step": 214 + }, + { + "epoch": 0.2356164383561644, + "grad_norm": 2.3483071327209473, + "learning_rate": 3.667883211678833e-06, + "loss": 2.3244, + "step": 215 + }, + { + "epoch": 0.23671232876712328, + "grad_norm": 2.884199380874634, + "learning_rate": 3.6861313868613142e-06, + "loss": 2.3067, + "step": 216 + }, + { + "epoch": 0.2378082191780822, + "grad_norm": 2.9514520168304443, + "learning_rate": 3.704379562043796e-06, + "loss": 2.304, + "step": 217 + }, + { + "epoch": 0.2389041095890411, + "grad_norm": 2.7530996799468994, + "learning_rate": 3.7226277372262775e-06, + "loss": 2.3691, + "step": 218 + }, + { + "epoch": 0.24, + "grad_norm": 2.8753647804260254, + "learning_rate": 3.7408759124087594e-06, + "loss": 2.3406, + "step": 219 + }, + { + "epoch": 0.2410958904109589, + "grad_norm": 2.670574903488159, + "learning_rate": 3.7591240875912412e-06, + "loss": 2.304, + "step": 220 + }, + { + "epoch": 0.2410958904109589, + "eval_loss": 2.4395809173583984, + "eval_runtime": 46.6723, + "eval_samples_per_second": 21.426, + "eval_steps_per_second": 0.9, + "step": 220 + }, + { + "epoch": 0.2421917808219178, + "grad_norm": 2.4152438640594482, + "learning_rate": 3.777372262773723e-06, + "loss": 2.3294, + "step": 221 + }, + { + "epoch": 0.2432876712328767, + "grad_norm": 2.8398540019989014, + "learning_rate": 3.7956204379562045e-06, + "loss": 2.3216, + "step": 222 + }, + { + "epoch": 0.24438356164383562, + "grad_norm": 2.6036765575408936, + "learning_rate": 3.8138686131386864e-06, + "loss": 2.3109, + "step": 223 + }, + { + "epoch": 0.24547945205479452, + "grad_norm": 2.7073473930358887, + "learning_rate": 3.832116788321168e-06, + "loss": 2.3457, + "step": 224 + }, + { + "epoch": 0.2465753424657534, + "grad_norm": 2.584404230117798, + "learning_rate": 3.85036496350365e-06, + "loss": 2.3159, + "step": 225 + }, + { + "epoch": 0.24767123287671233, + "grad_norm": 2.4220666885375977, + "learning_rate": 3.868613138686132e-06, + "loss": 2.2944, + "step": 226 + }, + { + "epoch": 0.24876712328767123, + "grad_norm": 2.5636494159698486, + "learning_rate": 3.886861313868613e-06, + "loss": 2.3219, + "step": 227 + }, + { + "epoch": 0.24986301369863015, + "grad_norm": 2.2990903854370117, + "learning_rate": 3.905109489051096e-06, + "loss": 2.3342, + "step": 228 + }, + { + "epoch": 0.25095890410958904, + "grad_norm": 2.50913143157959, + "learning_rate": 3.923357664233577e-06, + "loss": 2.3619, + "step": 229 + }, + { + "epoch": 0.25205479452054796, + "grad_norm": 2.1687769889831543, + "learning_rate": 3.9416058394160585e-06, + "loss": 2.3153, + "step": 230 + }, + { + "epoch": 0.25205479452054796, + "eval_loss": 2.4315240383148193, + "eval_runtime": 48.3542, + "eval_samples_per_second": 20.681, + "eval_steps_per_second": 0.869, + "step": 230 + }, + { + "epoch": 0.25315068493150683, + "grad_norm": 2.842125177383423, + "learning_rate": 3.95985401459854e-06, + "loss": 2.3644, + "step": 231 + }, + { + "epoch": 0.25424657534246575, + "grad_norm": 2.258049964904785, + "learning_rate": 3.978102189781022e-06, + "loss": 2.3727, + "step": 232 + }, + { + "epoch": 0.2553424657534247, + "grad_norm": 2.4630489349365234, + "learning_rate": 3.996350364963504e-06, + "loss": 2.267, + "step": 233 + }, + { + "epoch": 0.25643835616438354, + "grad_norm": 3.1469743251800537, + "learning_rate": 4.014598540145986e-06, + "loss": 2.3432, + "step": 234 + }, + { + "epoch": 0.25753424657534246, + "grad_norm": 2.6277477741241455, + "learning_rate": 4.032846715328467e-06, + "loss": 2.2545, + "step": 235 + }, + { + "epoch": 0.2586301369863014, + "grad_norm": 2.404141902923584, + "learning_rate": 4.05109489051095e-06, + "loss": 2.3109, + "step": 236 + }, + { + "epoch": 0.25972602739726025, + "grad_norm": 2.4545605182647705, + "learning_rate": 4.069343065693431e-06, + "loss": 2.3112, + "step": 237 + }, + { + "epoch": 0.2608219178082192, + "grad_norm": 2.237638235092163, + "learning_rate": 4.0875912408759126e-06, + "loss": 2.3, + "step": 238 + }, + { + "epoch": 0.2619178082191781, + "grad_norm": 2.192811965942383, + "learning_rate": 4.105839416058395e-06, + "loss": 2.3108, + "step": 239 + }, + { + "epoch": 0.26301369863013696, + "grad_norm": 2.5058720111846924, + "learning_rate": 4.124087591240876e-06, + "loss": 2.3635, + "step": 240 + }, + { + "epoch": 0.26301369863013696, + "eval_loss": 2.4219377040863037, + "eval_runtime": 45.3561, + "eval_samples_per_second": 22.048, + "eval_steps_per_second": 0.926, + "step": 240 + }, + { + "epoch": 0.2641095890410959, + "grad_norm": 2.0708935260772705, + "learning_rate": 4.1423357664233586e-06, + "loss": 2.3129, + "step": 241 + }, + { + "epoch": 0.2652054794520548, + "grad_norm": 2.6353116035461426, + "learning_rate": 4.16058394160584e-06, + "loss": 2.3011, + "step": 242 + }, + { + "epoch": 0.2663013698630137, + "grad_norm": 2.8392436504364014, + "learning_rate": 4.1788321167883214e-06, + "loss": 2.3643, + "step": 243 + }, + { + "epoch": 0.2673972602739726, + "grad_norm": 3.1773295402526855, + "learning_rate": 4.197080291970803e-06, + "loss": 2.2642, + "step": 244 + }, + { + "epoch": 0.2684931506849315, + "grad_norm": 2.4757354259490967, + "learning_rate": 4.215328467153285e-06, + "loss": 2.3211, + "step": 245 + }, + { + "epoch": 0.26958904109589044, + "grad_norm": 2.5799412727355957, + "learning_rate": 4.233576642335767e-06, + "loss": 2.2909, + "step": 246 + }, + { + "epoch": 0.2706849315068493, + "grad_norm": 2.368502378463745, + "learning_rate": 4.251824817518249e-06, + "loss": 2.2951, + "step": 247 + }, + { + "epoch": 0.2717808219178082, + "grad_norm": 2.114969253540039, + "learning_rate": 4.27007299270073e-06, + "loss": 2.3124, + "step": 248 + }, + { + "epoch": 0.27287671232876715, + "grad_norm": 3.056820869445801, + "learning_rate": 4.288321167883212e-06, + "loss": 2.2923, + "step": 249 + }, + { + "epoch": 0.273972602739726, + "grad_norm": 2.7752203941345215, + "learning_rate": 4.306569343065693e-06, + "loss": 2.2874, + "step": 250 + }, + { + "epoch": 0.273972602739726, + "eval_loss": 2.4229912757873535, + "eval_runtime": 52.5941, + "eval_samples_per_second": 19.014, + "eval_steps_per_second": 0.799, + "step": 250 + }, + { + "epoch": 0.27506849315068493, + "grad_norm": 3.045740842819214, + "learning_rate": 4.3248175182481755e-06, + "loss": 2.2686, + "step": 251 + }, + { + "epoch": 0.27616438356164386, + "grad_norm": 2.646402359008789, + "learning_rate": 4.343065693430658e-06, + "loss": 2.3247, + "step": 252 + }, + { + "epoch": 0.2772602739726027, + "grad_norm": 3.274994134902954, + "learning_rate": 4.361313868613139e-06, + "loss": 2.3093, + "step": 253 + }, + { + "epoch": 0.27835616438356164, + "grad_norm": 2.352246046066284, + "learning_rate": 4.379562043795621e-06, + "loss": 2.3124, + "step": 254 + }, + { + "epoch": 0.27945205479452057, + "grad_norm": 3.248717784881592, + "learning_rate": 4.397810218978102e-06, + "loss": 2.3417, + "step": 255 + }, + { + "epoch": 0.28054794520547943, + "grad_norm": 3.0919013023376465, + "learning_rate": 4.416058394160584e-06, + "loss": 2.3093, + "step": 256 + }, + { + "epoch": 0.28164383561643835, + "grad_norm": 2.1174702644348145, + "learning_rate": 4.434306569343066e-06, + "loss": 2.305, + "step": 257 + }, + { + "epoch": 0.2827397260273973, + "grad_norm": 2.913797378540039, + "learning_rate": 4.452554744525548e-06, + "loss": 2.2831, + "step": 258 + }, + { + "epoch": 0.28383561643835614, + "grad_norm": 2.693089723587036, + "learning_rate": 4.4708029197080295e-06, + "loss": 2.3206, + "step": 259 + }, + { + "epoch": 0.28493150684931506, + "grad_norm": 2.521411895751953, + "learning_rate": 4.489051094890512e-06, + "loss": 2.3196, + "step": 260 + }, + { + "epoch": 0.28493150684931506, + "eval_loss": 2.4178109169006348, + "eval_runtime": 50.6753, + "eval_samples_per_second": 19.733, + "eval_steps_per_second": 0.829, + "step": 260 + }, + { + "epoch": 0.286027397260274, + "grad_norm": 2.346346855163574, + "learning_rate": 4.507299270072993e-06, + "loss": 2.3433, + "step": 261 + }, + { + "epoch": 0.28712328767123285, + "grad_norm": 2.520368814468384, + "learning_rate": 4.525547445255475e-06, + "loss": 2.3174, + "step": 262 + }, + { + "epoch": 0.2882191780821918, + "grad_norm": 2.550347328186035, + "learning_rate": 4.543795620437956e-06, + "loss": 2.3072, + "step": 263 + }, + { + "epoch": 0.2893150684931507, + "grad_norm": 2.268662691116333, + "learning_rate": 4.562043795620438e-06, + "loss": 2.2763, + "step": 264 + }, + { + "epoch": 0.29041095890410956, + "grad_norm": 2.445798873901367, + "learning_rate": 4.580291970802921e-06, + "loss": 2.3395, + "step": 265 + }, + { + "epoch": 0.2915068493150685, + "grad_norm": 2.5391135215759277, + "learning_rate": 4.598540145985402e-06, + "loss": 2.3496, + "step": 266 + }, + { + "epoch": 0.2926027397260274, + "grad_norm": 2.3553097248077393, + "learning_rate": 4.6167883211678835e-06, + "loss": 2.3243, + "step": 267 + }, + { + "epoch": 0.2936986301369863, + "grad_norm": 2.448089599609375, + "learning_rate": 4.635036496350365e-06, + "loss": 2.2887, + "step": 268 + }, + { + "epoch": 0.2947945205479452, + "grad_norm": 2.1600170135498047, + "learning_rate": 4.653284671532847e-06, + "loss": 2.2816, + "step": 269 + }, + { + "epoch": 0.2958904109589041, + "grad_norm": 2.3783180713653564, + "learning_rate": 4.671532846715329e-06, + "loss": 2.3275, + "step": 270 + }, + { + "epoch": 0.2958904109589041, + "eval_loss": 2.413832902908325, + "eval_runtime": 48.7524, + "eval_samples_per_second": 20.512, + "eval_steps_per_second": 0.861, + "step": 270 + }, + { + "epoch": 0.29698630136986304, + "grad_norm": 2.468320369720459, + "learning_rate": 4.689781021897811e-06, + "loss": 2.3358, + "step": 271 + }, + { + "epoch": 0.2980821917808219, + "grad_norm": 2.401634454727173, + "learning_rate": 4.708029197080292e-06, + "loss": 2.2991, + "step": 272 + }, + { + "epoch": 0.2991780821917808, + "grad_norm": 2.4604203701019287, + "learning_rate": 4.726277372262774e-06, + "loss": 2.3496, + "step": 273 + }, + { + "epoch": 0.30027397260273975, + "grad_norm": 2.0338871479034424, + "learning_rate": 4.744525547445255e-06, + "loss": 2.3115, + "step": 274 + }, + { + "epoch": 0.3013698630136986, + "grad_norm": 2.5404436588287354, + "learning_rate": 4.7627737226277375e-06, + "loss": 2.2665, + "step": 275 + }, + { + "epoch": 0.30246575342465754, + "grad_norm": 2.5369458198547363, + "learning_rate": 4.78102189781022e-06, + "loss": 2.3072, + "step": 276 + }, + { + "epoch": 0.30356164383561646, + "grad_norm": 2.471522569656372, + "learning_rate": 4.799270072992701e-06, + "loss": 2.3113, + "step": 277 + }, + { + "epoch": 0.3046575342465753, + "grad_norm": 2.3890135288238525, + "learning_rate": 4.8175182481751835e-06, + "loss": 2.2838, + "step": 278 + }, + { + "epoch": 0.30575342465753425, + "grad_norm": 2.2485053539276123, + "learning_rate": 4.835766423357665e-06, + "loss": 2.3052, + "step": 279 + }, + { + "epoch": 0.30684931506849317, + "grad_norm": 2.1090753078460693, + "learning_rate": 4.854014598540146e-06, + "loss": 2.3054, + "step": 280 + }, + { + "epoch": 0.30684931506849317, + "eval_loss": 2.412548780441284, + "eval_runtime": 48.5034, + "eval_samples_per_second": 20.617, + "eval_steps_per_second": 0.866, + "step": 280 + }, + { + "epoch": 0.30794520547945203, + "grad_norm": 2.299133777618408, + "learning_rate": 4.872262773722628e-06, + "loss": 2.2599, + "step": 281 + }, + { + "epoch": 0.30904109589041096, + "grad_norm": 2.177905797958374, + "learning_rate": 4.89051094890511e-06, + "loss": 2.2745, + "step": 282 + }, + { + "epoch": 0.3101369863013699, + "grad_norm": 2.3178462982177734, + "learning_rate": 4.9087591240875915e-06, + "loss": 2.2562, + "step": 283 + }, + { + "epoch": 0.31123287671232874, + "grad_norm": 2.5088376998901367, + "learning_rate": 4.927007299270074e-06, + "loss": 2.2945, + "step": 284 + }, + { + "epoch": 0.31232876712328766, + "grad_norm": 2.095379590988159, + "learning_rate": 4.945255474452555e-06, + "loss": 2.3038, + "step": 285 + }, + { + "epoch": 0.3134246575342466, + "grad_norm": 2.1792995929718018, + "learning_rate": 4.963503649635037e-06, + "loss": 2.2702, + "step": 286 + }, + { + "epoch": 0.31452054794520545, + "grad_norm": 2.3001186847686768, + "learning_rate": 4.981751824817518e-06, + "loss": 2.3482, + "step": 287 + }, + { + "epoch": 0.3156164383561644, + "grad_norm": 2.1576106548309326, + "learning_rate": 5e-06, + "loss": 2.3391, + "step": 288 + }, + { + "epoch": 0.3167123287671233, + "grad_norm": 2.3715577125549316, + "learning_rate": 4.997969130787978e-06, + "loss": 2.3049, + "step": 289 + }, + { + "epoch": 0.3178082191780822, + "grad_norm": 2.0316715240478516, + "learning_rate": 4.995938261575955e-06, + "loss": 2.3111, + "step": 290 + }, + { + "epoch": 0.3178082191780822, + "eval_loss": 2.40299654006958, + "eval_runtime": 50.0873, + "eval_samples_per_second": 19.965, + "eval_steps_per_second": 0.839, + "step": 290 + }, + { + "epoch": 0.3189041095890411, + "grad_norm": 2.1135756969451904, + "learning_rate": 4.993907392363932e-06, + "loss": 2.2465, + "step": 291 + }, + { + "epoch": 0.32, + "grad_norm": 2.335655927658081, + "learning_rate": 4.9918765231519095e-06, + "loss": 2.31, + "step": 292 + }, + { + "epoch": 0.32109589041095893, + "grad_norm": 2.3255794048309326, + "learning_rate": 4.989845653939887e-06, + "loss": 2.2531, + "step": 293 + }, + { + "epoch": 0.3221917808219178, + "grad_norm": 2.111055612564087, + "learning_rate": 4.987814784727864e-06, + "loss": 2.2799, + "step": 294 + }, + { + "epoch": 0.3232876712328767, + "grad_norm": 2.1773171424865723, + "learning_rate": 4.985783915515841e-06, + "loss": 2.2884, + "step": 295 + }, + { + "epoch": 0.32438356164383564, + "grad_norm": 2.3559885025024414, + "learning_rate": 4.983753046303819e-06, + "loss": 2.2711, + "step": 296 + }, + { + "epoch": 0.3254794520547945, + "grad_norm": 2.120802879333496, + "learning_rate": 4.981722177091796e-06, + "loss": 2.2946, + "step": 297 + }, + { + "epoch": 0.3265753424657534, + "grad_norm": 2.2319388389587402, + "learning_rate": 4.979691307879773e-06, + "loss": 2.2771, + "step": 298 + }, + { + "epoch": 0.32767123287671235, + "grad_norm": 2.208740711212158, + "learning_rate": 4.9776604386677504e-06, + "loss": 2.3003, + "step": 299 + }, + { + "epoch": 0.3287671232876712, + "grad_norm": 2.543060541152954, + "learning_rate": 4.975629569455728e-06, + "loss": 2.2891, + "step": 300 + }, + { + "epoch": 0.3287671232876712, + "eval_loss": 2.398495674133301, + "eval_runtime": 51.1973, + "eval_samples_per_second": 19.532, + "eval_steps_per_second": 0.82, + "step": 300 + }, + { + "epoch": 0.32986301369863014, + "grad_norm": 2.086219072341919, + "learning_rate": 4.973598700243705e-06, + "loss": 2.3286, + "step": 301 + }, + { + "epoch": 0.33095890410958906, + "grad_norm": 2.3673319816589355, + "learning_rate": 4.971567831031681e-06, + "loss": 2.2672, + "step": 302 + }, + { + "epoch": 0.3320547945205479, + "grad_norm": 2.167593240737915, + "learning_rate": 4.9695369618196595e-06, + "loss": 2.264, + "step": 303 + }, + { + "epoch": 0.33315068493150685, + "grad_norm": 2.7218785285949707, + "learning_rate": 4.967506092607636e-06, + "loss": 2.1737, + "step": 304 + }, + { + "epoch": 0.33424657534246577, + "grad_norm": 2.673245429992676, + "learning_rate": 4.965475223395614e-06, + "loss": 2.2439, + "step": 305 + }, + { + "epoch": 0.33534246575342463, + "grad_norm": 2.0341989994049072, + "learning_rate": 4.9634443541835905e-06, + "loss": 2.2614, + "step": 306 + }, + { + "epoch": 0.33643835616438356, + "grad_norm": 1.9314247369766235, + "learning_rate": 4.961413484971569e-06, + "loss": 2.2592, + "step": 307 + }, + { + "epoch": 0.3375342465753425, + "grad_norm": 2.537794589996338, + "learning_rate": 4.959382615759545e-06, + "loss": 2.2859, + "step": 308 + }, + { + "epoch": 0.33863013698630134, + "grad_norm": 2.0811166763305664, + "learning_rate": 4.957351746547523e-06, + "loss": 2.2868, + "step": 309 + }, + { + "epoch": 0.33972602739726027, + "grad_norm": 2.783701181411743, + "learning_rate": 4.9553208773355e-06, + "loss": 2.2649, + "step": 310 + }, + { + "epoch": 0.33972602739726027, + "eval_loss": 2.400538921356201, + "eval_runtime": 43.185, + "eval_samples_per_second": 23.156, + "eval_steps_per_second": 0.973, + "step": 310 + }, + { + "epoch": 0.3408219178082192, + "grad_norm": 2.8541319370269775, + "learning_rate": 4.953290008123478e-06, + "loss": 2.3729, + "step": 311 + }, + { + "epoch": 0.3419178082191781, + "grad_norm": 2.1852617263793945, + "learning_rate": 4.951259138911454e-06, + "loss": 2.2885, + "step": 312 + }, + { + "epoch": 0.343013698630137, + "grad_norm": 2.872201919555664, + "learning_rate": 4.9492282696994315e-06, + "loss": 2.2353, + "step": 313 + }, + { + "epoch": 0.3441095890410959, + "grad_norm": 2.5140910148620605, + "learning_rate": 4.947197400487409e-06, + "loss": 2.3364, + "step": 314 + }, + { + "epoch": 0.3452054794520548, + "grad_norm": 2.619974374771118, + "learning_rate": 4.945166531275386e-06, + "loss": 2.2721, + "step": 315 + }, + { + "epoch": 0.3463013698630137, + "grad_norm": 2.4567618370056152, + "learning_rate": 4.943135662063363e-06, + "loss": 2.26, + "step": 316 + }, + { + "epoch": 0.3473972602739726, + "grad_norm": 2.9603192806243896, + "learning_rate": 4.9411047928513406e-06, + "loss": 2.2861, + "step": 317 + }, + { + "epoch": 0.34849315068493153, + "grad_norm": 2.115262985229492, + "learning_rate": 4.939073923639318e-06, + "loss": 2.3125, + "step": 318 + }, + { + "epoch": 0.3495890410958904, + "grad_norm": 2.693024158477783, + "learning_rate": 4.937043054427295e-06, + "loss": 2.3054, + "step": 319 + }, + { + "epoch": 0.3506849315068493, + "grad_norm": 2.3439996242523193, + "learning_rate": 4.935012185215272e-06, + "loss": 2.2576, + "step": 320 + }, + { + "epoch": 0.3506849315068493, + "eval_loss": 2.3920650482177734, + "eval_runtime": 47.7451, + "eval_samples_per_second": 20.945, + "eval_steps_per_second": 0.88, + "step": 320 + }, + { + "epoch": 0.35178082191780824, + "grad_norm": 2.3009634017944336, + "learning_rate": 4.93298131600325e-06, + "loss": 2.2454, + "step": 321 + }, + { + "epoch": 0.3528767123287671, + "grad_norm": 2.593203544616699, + "learning_rate": 4.930950446791227e-06, + "loss": 2.2965, + "step": 322 + }, + { + "epoch": 0.353972602739726, + "grad_norm": 2.9069650173187256, + "learning_rate": 4.928919577579204e-06, + "loss": 2.2785, + "step": 323 + }, + { + "epoch": 0.35506849315068495, + "grad_norm": 2.1021759510040283, + "learning_rate": 4.9268887083671815e-06, + "loss": 2.2857, + "step": 324 + }, + { + "epoch": 0.3561643835616438, + "grad_norm": 2.0330185890197754, + "learning_rate": 4.924857839155159e-06, + "loss": 2.2697, + "step": 325 + }, + { + "epoch": 0.35726027397260274, + "grad_norm": 2.1318063735961914, + "learning_rate": 4.922826969943136e-06, + "loss": 2.2698, + "step": 326 + }, + { + "epoch": 0.35835616438356166, + "grad_norm": 2.0157387256622314, + "learning_rate": 4.920796100731113e-06, + "loss": 2.292, + "step": 327 + }, + { + "epoch": 0.3594520547945205, + "grad_norm": 2.3266026973724365, + "learning_rate": 4.918765231519091e-06, + "loss": 2.2215, + "step": 328 + }, + { + "epoch": 0.36054794520547945, + "grad_norm": 2.066305160522461, + "learning_rate": 4.916734362307068e-06, + "loss": 2.2758, + "step": 329 + }, + { + "epoch": 0.36164383561643837, + "grad_norm": 2.0219993591308594, + "learning_rate": 4.914703493095045e-06, + "loss": 2.2381, + "step": 330 + }, + { + "epoch": 0.36164383561643837, + "eval_loss": 2.394064426422119, + "eval_runtime": 51.5482, + "eval_samples_per_second": 19.399, + "eval_steps_per_second": 0.815, + "step": 330 + }, + { + "epoch": 0.36273972602739724, + "grad_norm": 2.3964319229125977, + "learning_rate": 4.912672623883022e-06, + "loss": 2.2384, + "step": 331 + }, + { + "epoch": 0.36383561643835616, + "grad_norm": 2.5466530323028564, + "learning_rate": 4.910641754671e-06, + "loss": 2.2428, + "step": 332 + }, + { + "epoch": 0.3649315068493151, + "grad_norm": 2.236152410507202, + "learning_rate": 4.908610885458977e-06, + "loss": 2.2639, + "step": 333 + }, + { + "epoch": 0.36602739726027395, + "grad_norm": 2.7267303466796875, + "learning_rate": 4.906580016246954e-06, + "loss": 2.2608, + "step": 334 + }, + { + "epoch": 0.36712328767123287, + "grad_norm": 1.9403765201568604, + "learning_rate": 4.9045491470349315e-06, + "loss": 2.2962, + "step": 335 + }, + { + "epoch": 0.3682191780821918, + "grad_norm": 2.89070463180542, + "learning_rate": 4.902518277822909e-06, + "loss": 2.2847, + "step": 336 + }, + { + "epoch": 0.3693150684931507, + "grad_norm": 2.6427228450775146, + "learning_rate": 4.900487408610886e-06, + "loss": 2.2734, + "step": 337 + }, + { + "epoch": 0.3704109589041096, + "grad_norm": 2.1680362224578857, + "learning_rate": 4.898456539398863e-06, + "loss": 2.2869, + "step": 338 + }, + { + "epoch": 0.3715068493150685, + "grad_norm": 3.0160160064697266, + "learning_rate": 4.89642567018684e-06, + "loss": 2.3043, + "step": 339 + }, + { + "epoch": 0.3726027397260274, + "grad_norm": 1.9858418703079224, + "learning_rate": 4.894394800974818e-06, + "loss": 2.3422, + "step": 340 + }, + { + "epoch": 0.3726027397260274, + "eval_loss": 2.387559652328491, + "eval_runtime": 52.2455, + "eval_samples_per_second": 19.14, + "eval_steps_per_second": 0.804, + "step": 340 + }, + { + "epoch": 0.3736986301369863, + "grad_norm": 2.243699073791504, + "learning_rate": 4.892363931762794e-06, + "loss": 2.2394, + "step": 341 + }, + { + "epoch": 0.3747945205479452, + "grad_norm": 2.371299982070923, + "learning_rate": 4.8903330625507725e-06, + "loss": 2.2856, + "step": 342 + }, + { + "epoch": 0.37589041095890413, + "grad_norm": 2.3769736289978027, + "learning_rate": 4.888302193338749e-06, + "loss": 2.246, + "step": 343 + }, + { + "epoch": 0.376986301369863, + "grad_norm": 2.3209030628204346, + "learning_rate": 4.886271324126727e-06, + "loss": 2.2331, + "step": 344 + }, + { + "epoch": 0.3780821917808219, + "grad_norm": 2.747319221496582, + "learning_rate": 4.8842404549147034e-06, + "loss": 2.3115, + "step": 345 + }, + { + "epoch": 0.37917808219178084, + "grad_norm": 1.8599231243133545, + "learning_rate": 4.8822095857026816e-06, + "loss": 2.2994, + "step": 346 + }, + { + "epoch": 0.3802739726027397, + "grad_norm": 2.0358943939208984, + "learning_rate": 4.880178716490658e-06, + "loss": 2.271, + "step": 347 + }, + { + "epoch": 0.38136986301369863, + "grad_norm": 2.409106731414795, + "learning_rate": 4.878147847278636e-06, + "loss": 2.2281, + "step": 348 + }, + { + "epoch": 0.38246575342465755, + "grad_norm": 2.1641807556152344, + "learning_rate": 4.8761169780666125e-06, + "loss": 2.2437, + "step": 349 + }, + { + "epoch": 0.3835616438356164, + "grad_norm": 2.6917359828948975, + "learning_rate": 4.874086108854591e-06, + "loss": 2.226, + "step": 350 + }, + { + "epoch": 0.3835616438356164, + "eval_loss": 2.382991313934326, + "eval_runtime": 48.9534, + "eval_samples_per_second": 20.428, + "eval_steps_per_second": 0.858, + "step": 350 + }, + { + "epoch": 0.38465753424657534, + "grad_norm": 2.364215612411499, + "learning_rate": 4.872055239642567e-06, + "loss": 2.2319, + "step": 351 + }, + { + "epoch": 0.38575342465753426, + "grad_norm": 2.317643165588379, + "learning_rate": 4.870024370430544e-06, + "loss": 2.2096, + "step": 352 + }, + { + "epoch": 0.3868493150684931, + "grad_norm": 2.2283246517181396, + "learning_rate": 4.867993501218522e-06, + "loss": 2.2634, + "step": 353 + }, + { + "epoch": 0.38794520547945205, + "grad_norm": 2.498391628265381, + "learning_rate": 4.865962632006499e-06, + "loss": 2.2813, + "step": 354 + }, + { + "epoch": 0.38904109589041097, + "grad_norm": 2.3889403343200684, + "learning_rate": 4.863931762794476e-06, + "loss": 2.2361, + "step": 355 + }, + { + "epoch": 0.39013698630136984, + "grad_norm": 2.2049975395202637, + "learning_rate": 4.8619008935824535e-06, + "loss": 2.2583, + "step": 356 + }, + { + "epoch": 0.39123287671232876, + "grad_norm": 2.132840871810913, + "learning_rate": 4.859870024370431e-06, + "loss": 2.2823, + "step": 357 + }, + { + "epoch": 0.3923287671232877, + "grad_norm": 2.2692184448242188, + "learning_rate": 4.857839155158408e-06, + "loss": 2.2939, + "step": 358 + }, + { + "epoch": 0.3934246575342466, + "grad_norm": 2.0501933097839355, + "learning_rate": 4.855808285946385e-06, + "loss": 2.2976, + "step": 359 + }, + { + "epoch": 0.39452054794520547, + "grad_norm": 2.170236587524414, + "learning_rate": 4.8537774167343626e-06, + "loss": 2.2748, + "step": 360 + }, + { + "epoch": 0.39452054794520547, + "eval_loss": 2.3783886432647705, + "eval_runtime": 46.7324, + "eval_samples_per_second": 21.398, + "eval_steps_per_second": 0.899, + "step": 360 + }, + { + "epoch": 0.3956164383561644, + "grad_norm": 2.283137083053589, + "learning_rate": 4.85174654752234e-06, + "loss": 2.2705, + "step": 361 + }, + { + "epoch": 0.3967123287671233, + "grad_norm": 2.195984363555908, + "learning_rate": 4.849715678310317e-06, + "loss": 2.26, + "step": 362 + }, + { + "epoch": 0.3978082191780822, + "grad_norm": 2.399935007095337, + "learning_rate": 4.847684809098294e-06, + "loss": 2.2639, + "step": 363 + }, + { + "epoch": 0.3989041095890411, + "grad_norm": 2.411280393600464, + "learning_rate": 4.845653939886272e-06, + "loss": 2.2523, + "step": 364 + }, + { + "epoch": 0.4, + "grad_norm": 2.2033896446228027, + "learning_rate": 4.843623070674249e-06, + "loss": 2.2722, + "step": 365 + }, + { + "epoch": 0.4010958904109589, + "grad_norm": 2.2990238666534424, + "learning_rate": 4.841592201462226e-06, + "loss": 2.2411, + "step": 366 + }, + { + "epoch": 0.4021917808219178, + "grad_norm": 1.911453127861023, + "learning_rate": 4.8395613322502035e-06, + "loss": 2.2921, + "step": 367 + }, + { + "epoch": 0.40328767123287673, + "grad_norm": 2.453195095062256, + "learning_rate": 4.837530463038181e-06, + "loss": 2.218, + "step": 368 + }, + { + "epoch": 0.4043835616438356, + "grad_norm": 2.126992702484131, + "learning_rate": 4.835499593826158e-06, + "loss": 2.2628, + "step": 369 + }, + { + "epoch": 0.4054794520547945, + "grad_norm": 1.9305835962295532, + "learning_rate": 4.833468724614135e-06, + "loss": 2.2726, + "step": 370 + }, + { + "epoch": 0.4054794520547945, + "eval_loss": 2.3741114139556885, + "eval_runtime": 49.2956, + "eval_samples_per_second": 20.286, + "eval_steps_per_second": 0.852, + "step": 370 + }, + { + "epoch": 0.40657534246575344, + "grad_norm": 2.222024917602539, + "learning_rate": 4.831437855402113e-06, + "loss": 2.2634, + "step": 371 + }, + { + "epoch": 0.4076712328767123, + "grad_norm": 2.473958969116211, + "learning_rate": 4.82940698619009e-06, + "loss": 2.1817, + "step": 372 + }, + { + "epoch": 0.40876712328767123, + "grad_norm": 1.9208838939666748, + "learning_rate": 4.827376116978067e-06, + "loss": 2.2353, + "step": 373 + }, + { + "epoch": 0.40986301369863015, + "grad_norm": 2.7226450443267822, + "learning_rate": 4.8253452477660444e-06, + "loss": 2.2831, + "step": 374 + }, + { + "epoch": 0.410958904109589, + "grad_norm": 1.9078222513198853, + "learning_rate": 4.823314378554022e-06, + "loss": 2.264, + "step": 375 + }, + { + "epoch": 0.41205479452054794, + "grad_norm": 2.3946311473846436, + "learning_rate": 4.821283509341998e-06, + "loss": 2.251, + "step": 376 + }, + { + "epoch": 0.41315068493150686, + "grad_norm": 1.847343921661377, + "learning_rate": 4.819252640129976e-06, + "loss": 2.2441, + "step": 377 + }, + { + "epoch": 0.41424657534246573, + "grad_norm": 2.475385904312134, + "learning_rate": 4.817221770917953e-06, + "loss": 2.2639, + "step": 378 + }, + { + "epoch": 0.41534246575342465, + "grad_norm": 2.1396560668945312, + "learning_rate": 4.815190901705931e-06, + "loss": 2.2353, + "step": 379 + }, + { + "epoch": 0.41643835616438357, + "grad_norm": 2.3868885040283203, + "learning_rate": 4.813160032493907e-06, + "loss": 2.2551, + "step": 380 + }, + { + "epoch": 0.41643835616438357, + "eval_loss": 2.369168519973755, + "eval_runtime": 45.6131, + "eval_samples_per_second": 21.924, + "eval_steps_per_second": 0.921, + "step": 380 + }, + { + "epoch": 0.41753424657534244, + "grad_norm": 2.0193872451782227, + "learning_rate": 4.811129163281885e-06, + "loss": 2.2399, + "step": 381 + }, + { + "epoch": 0.41863013698630136, + "grad_norm": 2.4655935764312744, + "learning_rate": 4.809098294069862e-06, + "loss": 2.2729, + "step": 382 + }, + { + "epoch": 0.4197260273972603, + "grad_norm": 1.880974292755127, + "learning_rate": 4.80706742485784e-06, + "loss": 2.2681, + "step": 383 + }, + { + "epoch": 0.4208219178082192, + "grad_norm": 2.1912078857421875, + "learning_rate": 4.805036555645816e-06, + "loss": 2.2961, + "step": 384 + }, + { + "epoch": 0.42191780821917807, + "grad_norm": 1.985931634902954, + "learning_rate": 4.8030056864337945e-06, + "loss": 2.2407, + "step": 385 + }, + { + "epoch": 0.423013698630137, + "grad_norm": 2.4353220462799072, + "learning_rate": 4.800974817221771e-06, + "loss": 2.2867, + "step": 386 + }, + { + "epoch": 0.4241095890410959, + "grad_norm": 1.8810508251190186, + "learning_rate": 4.798943948009749e-06, + "loss": 2.2498, + "step": 387 + }, + { + "epoch": 0.4252054794520548, + "grad_norm": 2.195033073425293, + "learning_rate": 4.7969130787977254e-06, + "loss": 2.2532, + "step": 388 + }, + { + "epoch": 0.4263013698630137, + "grad_norm": 1.855372428894043, + "learning_rate": 4.794882209585703e-06, + "loss": 2.2538, + "step": 389 + }, + { + "epoch": 0.4273972602739726, + "grad_norm": 2.4465434551239014, + "learning_rate": 4.79285134037368e-06, + "loss": 2.2724, + "step": 390 + }, + { + "epoch": 0.4273972602739726, + "eval_loss": 2.364928960800171, + "eval_runtime": 48.0245, + "eval_samples_per_second": 20.823, + "eval_steps_per_second": 0.875, + "step": 390 + }, + { + "epoch": 0.4284931506849315, + "grad_norm": 1.8911314010620117, + "learning_rate": 4.790820471161657e-06, + "loss": 2.2291, + "step": 391 + }, + { + "epoch": 0.4295890410958904, + "grad_norm": 2.1975977420806885, + "learning_rate": 4.7887896019496345e-06, + "loss": 2.2373, + "step": 392 + }, + { + "epoch": 0.43068493150684933, + "grad_norm": 2.015331506729126, + "learning_rate": 4.786758732737612e-06, + "loss": 2.2483, + "step": 393 + }, + { + "epoch": 0.4317808219178082, + "grad_norm": 2.039339065551758, + "learning_rate": 4.784727863525589e-06, + "loss": 2.2746, + "step": 394 + }, + { + "epoch": 0.4328767123287671, + "grad_norm": 1.977513074874878, + "learning_rate": 4.782696994313566e-06, + "loss": 2.2275, + "step": 395 + }, + { + "epoch": 0.43397260273972604, + "grad_norm": 2.147217035293579, + "learning_rate": 4.780666125101544e-06, + "loss": 2.252, + "step": 396 + }, + { + "epoch": 0.4350684931506849, + "grad_norm": 2.1799919605255127, + "learning_rate": 4.778635255889521e-06, + "loss": 2.1975, + "step": 397 + }, + { + "epoch": 0.43616438356164383, + "grad_norm": 2.582240343093872, + "learning_rate": 4.776604386677498e-06, + "loss": 2.2294, + "step": 398 + }, + { + "epoch": 0.43726027397260275, + "grad_norm": 1.9127846956253052, + "learning_rate": 4.7745735174654755e-06, + "loss": 2.2319, + "step": 399 + }, + { + "epoch": 0.4383561643835616, + "grad_norm": 2.2444543838500977, + "learning_rate": 4.772542648253453e-06, + "loss": 2.2857, + "step": 400 + }, + { + "epoch": 0.4383561643835616, + "eval_loss": 2.3653724193573, + "eval_runtime": 43.8304, + "eval_samples_per_second": 22.815, + "eval_steps_per_second": 0.958, + "step": 400 + }, + { + "epoch": 0.43945205479452054, + "grad_norm": 1.9913676977157593, + "learning_rate": 4.77051177904143e-06, + "loss": 2.274, + "step": 401 + }, + { + "epoch": 0.44054794520547946, + "grad_norm": 2.2850067615509033, + "learning_rate": 4.768480909829407e-06, + "loss": 2.2209, + "step": 402 + }, + { + "epoch": 0.44164383561643833, + "grad_norm": 2.4984688758850098, + "learning_rate": 4.766450040617385e-06, + "loss": 2.263, + "step": 403 + }, + { + "epoch": 0.44273972602739725, + "grad_norm": 2.3244640827178955, + "learning_rate": 4.764419171405362e-06, + "loss": 2.2368, + "step": 404 + }, + { + "epoch": 0.4438356164383562, + "grad_norm": 2.7579681873321533, + "learning_rate": 4.762388302193339e-06, + "loss": 2.2762, + "step": 405 + }, + { + "epoch": 0.4449315068493151, + "grad_norm": 2.1996467113494873, + "learning_rate": 4.760357432981316e-06, + "loss": 2.2536, + "step": 406 + }, + { + "epoch": 0.44602739726027396, + "grad_norm": 2.3767571449279785, + "learning_rate": 4.758326563769294e-06, + "loss": 2.2353, + "step": 407 + }, + { + "epoch": 0.4471232876712329, + "grad_norm": 1.9148359298706055, + "learning_rate": 4.756295694557271e-06, + "loss": 2.2416, + "step": 408 + }, + { + "epoch": 0.4482191780821918, + "grad_norm": 2.6499507427215576, + "learning_rate": 4.754264825345248e-06, + "loss": 2.2641, + "step": 409 + }, + { + "epoch": 0.44931506849315067, + "grad_norm": 1.9890286922454834, + "learning_rate": 4.7522339561332255e-06, + "loss": 2.2148, + "step": 410 + }, + { + "epoch": 0.44931506849315067, + "eval_loss": 2.3670520782470703, + "eval_runtime": 41.1722, + "eval_samples_per_second": 24.288, + "eval_steps_per_second": 1.02, + "step": 410 + }, + { + "epoch": 0.4504109589041096, + "grad_norm": 2.575968027114868, + "learning_rate": 4.750203086921203e-06, + "loss": 2.2254, + "step": 411 + }, + { + "epoch": 0.4515068493150685, + "grad_norm": 2.015174388885498, + "learning_rate": 4.74817221770918e-06, + "loss": 2.1881, + "step": 412 + }, + { + "epoch": 0.4526027397260274, + "grad_norm": 2.44292950630188, + "learning_rate": 4.7461413484971565e-06, + "loss": 2.256, + "step": 413 + }, + { + "epoch": 0.4536986301369863, + "grad_norm": 1.9808441400527954, + "learning_rate": 4.744110479285135e-06, + "loss": 2.285, + "step": 414 + }, + { + "epoch": 0.4547945205479452, + "grad_norm": 2.009481906890869, + "learning_rate": 4.742079610073111e-06, + "loss": 2.2018, + "step": 415 + }, + { + "epoch": 0.4558904109589041, + "grad_norm": 2.1466031074523926, + "learning_rate": 4.740048740861089e-06, + "loss": 2.224, + "step": 416 + }, + { + "epoch": 0.456986301369863, + "grad_norm": 2.1337907314300537, + "learning_rate": 4.738017871649066e-06, + "loss": 2.2404, + "step": 417 + }, + { + "epoch": 0.45808219178082193, + "grad_norm": 1.94125235080719, + "learning_rate": 4.735987002437044e-06, + "loss": 2.2474, + "step": 418 + }, + { + "epoch": 0.4591780821917808, + "grad_norm": 2.379099130630493, + "learning_rate": 4.73395613322502e-06, + "loss": 2.2858, + "step": 419 + }, + { + "epoch": 0.4602739726027397, + "grad_norm": 1.844056248664856, + "learning_rate": 4.731925264012998e-06, + "loss": 2.2967, + "step": 420 + }, + { + "epoch": 0.4602739726027397, + "eval_loss": 2.3622241020202637, + "eval_runtime": 45.4674, + "eval_samples_per_second": 21.994, + "eval_steps_per_second": 0.924, + "step": 420 + }, + { + "epoch": 0.46136986301369864, + "grad_norm": 2.411895513534546, + "learning_rate": 4.729894394800975e-06, + "loss": 2.274, + "step": 421 + }, + { + "epoch": 0.4624657534246575, + "grad_norm": 2.014885425567627, + "learning_rate": 4.727863525588953e-06, + "loss": 2.2144, + "step": 422 + }, + { + "epoch": 0.46356164383561643, + "grad_norm": 2.112009048461914, + "learning_rate": 4.725832656376929e-06, + "loss": 2.1824, + "step": 423 + }, + { + "epoch": 0.46465753424657535, + "grad_norm": 2.111392021179199, + "learning_rate": 4.723801787164907e-06, + "loss": 2.2261, + "step": 424 + }, + { + "epoch": 0.4657534246575342, + "grad_norm": 2.0349185466766357, + "learning_rate": 4.721770917952885e-06, + "loss": 2.2622, + "step": 425 + }, + { + "epoch": 0.46684931506849314, + "grad_norm": 2.30222487449646, + "learning_rate": 4.719740048740861e-06, + "loss": 2.2206, + "step": 426 + }, + { + "epoch": 0.46794520547945206, + "grad_norm": 2.018789529800415, + "learning_rate": 4.717709179528839e-06, + "loss": 2.2196, + "step": 427 + }, + { + "epoch": 0.469041095890411, + "grad_norm": 2.046649932861328, + "learning_rate": 4.715678310316816e-06, + "loss": 2.2584, + "step": 428 + }, + { + "epoch": 0.47013698630136985, + "grad_norm": 2.0393869876861572, + "learning_rate": 4.713647441104794e-06, + "loss": 2.2537, + "step": 429 + }, + { + "epoch": 0.4712328767123288, + "grad_norm": 2.2788116931915283, + "learning_rate": 4.71161657189277e-06, + "loss": 2.2145, + "step": 430 + }, + { + "epoch": 0.4712328767123288, + "eval_loss": 2.3575785160064697, + "eval_runtime": 44.6687, + "eval_samples_per_second": 22.387, + "eval_steps_per_second": 0.94, + "step": 430 + }, + { + "epoch": 0.4723287671232877, + "grad_norm": 1.9950047731399536, + "learning_rate": 4.709585702680748e-06, + "loss": 2.2072, + "step": 431 + }, + { + "epoch": 0.47342465753424656, + "grad_norm": 1.9222451448440552, + "learning_rate": 4.707554833468725e-06, + "loss": 2.2654, + "step": 432 + }, + { + "epoch": 0.4745205479452055, + "grad_norm": 1.9735718965530396, + "learning_rate": 4.705523964256703e-06, + "loss": 2.3037, + "step": 433 + }, + { + "epoch": 0.4756164383561644, + "grad_norm": 2.3209776878356934, + "learning_rate": 4.703493095044679e-06, + "loss": 2.2083, + "step": 434 + }, + { + "epoch": 0.4767123287671233, + "grad_norm": 2.0993995666503906, + "learning_rate": 4.701462225832657e-06, + "loss": 2.2275, + "step": 435 + }, + { + "epoch": 0.4778082191780822, + "grad_norm": 2.334463119506836, + "learning_rate": 4.699431356620634e-06, + "loss": 2.1852, + "step": 436 + }, + { + "epoch": 0.4789041095890411, + "grad_norm": 1.9098213911056519, + "learning_rate": 4.697400487408612e-06, + "loss": 2.2829, + "step": 437 + }, + { + "epoch": 0.48, + "grad_norm": 2.3005056381225586, + "learning_rate": 4.695369618196588e-06, + "loss": 2.2532, + "step": 438 + }, + { + "epoch": 0.4810958904109589, + "grad_norm": 1.9069781303405762, + "learning_rate": 4.693338748984566e-06, + "loss": 2.2374, + "step": 439 + }, + { + "epoch": 0.4821917808219178, + "grad_norm": 2.050032138824463, + "learning_rate": 4.691307879772543e-06, + "loss": 2.2256, + "step": 440 + }, + { + "epoch": 0.4821917808219178, + "eval_loss": 2.3587422370910645, + "eval_runtime": 45.0941, + "eval_samples_per_second": 22.176, + "eval_steps_per_second": 0.931, + "step": 440 + }, + { + "epoch": 0.4832876712328767, + "grad_norm": 2.0563836097717285, + "learning_rate": 4.68927701056052e-06, + "loss": 2.1918, + "step": 441 + }, + { + "epoch": 0.4843835616438356, + "grad_norm": 2.1598386764526367, + "learning_rate": 4.6872461413484975e-06, + "loss": 2.2618, + "step": 442 + }, + { + "epoch": 0.48547945205479454, + "grad_norm": 1.9236539602279663, + "learning_rate": 4.685215272136475e-06, + "loss": 2.2477, + "step": 443 + }, + { + "epoch": 0.4865753424657534, + "grad_norm": 1.9923924207687378, + "learning_rate": 4.683184402924452e-06, + "loss": 2.2313, + "step": 444 + }, + { + "epoch": 0.4876712328767123, + "grad_norm": 2.162627696990967, + "learning_rate": 4.681153533712429e-06, + "loss": 2.2367, + "step": 445 + }, + { + "epoch": 0.48876712328767125, + "grad_norm": 2.083523988723755, + "learning_rate": 4.679122664500407e-06, + "loss": 2.2946, + "step": 446 + }, + { + "epoch": 0.4898630136986301, + "grad_norm": 2.504408359527588, + "learning_rate": 4.677091795288384e-06, + "loss": 2.2104, + "step": 447 + }, + { + "epoch": 0.49095890410958903, + "grad_norm": 1.8316532373428345, + "learning_rate": 4.675060926076361e-06, + "loss": 2.2027, + "step": 448 + }, + { + "epoch": 0.49205479452054796, + "grad_norm": 2.3337953090667725, + "learning_rate": 4.673030056864338e-06, + "loss": 2.278, + "step": 449 + }, + { + "epoch": 0.4931506849315068, + "grad_norm": 2.24102783203125, + "learning_rate": 4.670999187652316e-06, + "loss": 2.2018, + "step": 450 + }, + { + "epoch": 0.4931506849315068, + "eval_loss": 2.3545279502868652, + "eval_runtime": 45.1041, + "eval_samples_per_second": 22.171, + "eval_steps_per_second": 0.931, + "step": 450 + }, + { + "epoch": 0.49424657534246574, + "grad_norm": 2.07834792137146, + "learning_rate": 4.668968318440293e-06, + "loss": 2.1975, + "step": 451 + }, + { + "epoch": 0.49534246575342467, + "grad_norm": 1.7681723833084106, + "learning_rate": 4.66693744922827e-06, + "loss": 2.2033, + "step": 452 + }, + { + "epoch": 0.4964383561643836, + "grad_norm": 2.062134027481079, + "learning_rate": 4.6649065800162475e-06, + "loss": 2.2646, + "step": 453 + }, + { + "epoch": 0.49753424657534245, + "grad_norm": 1.94647216796875, + "learning_rate": 4.662875710804225e-06, + "loss": 2.1748, + "step": 454 + }, + { + "epoch": 0.4986301369863014, + "grad_norm": 1.9173872470855713, + "learning_rate": 4.660844841592202e-06, + "loss": 2.2435, + "step": 455 + }, + { + "epoch": 0.4997260273972603, + "grad_norm": 1.88880455493927, + "learning_rate": 4.658813972380179e-06, + "loss": 2.1711, + "step": 456 + }, + { + "epoch": 0.5008219178082192, + "grad_norm": 1.9628506898880005, + "learning_rate": 4.656783103168157e-06, + "loss": 2.2701, + "step": 457 + }, + { + "epoch": 0.5019178082191781, + "grad_norm": 2.0515286922454834, + "learning_rate": 4.654752233956134e-06, + "loss": 2.1656, + "step": 458 + }, + { + "epoch": 0.503013698630137, + "grad_norm": 1.978522539138794, + "learning_rate": 4.652721364744111e-06, + "loss": 2.246, + "step": 459 + }, + { + "epoch": 0.5041095890410959, + "grad_norm": 1.7993966341018677, + "learning_rate": 4.6506904955320885e-06, + "loss": 2.214, + "step": 460 + }, + { + "epoch": 0.5041095890410959, + "eval_loss": 2.3568837642669678, + "eval_runtime": 45.2086, + "eval_samples_per_second": 22.12, + "eval_steps_per_second": 0.929, + "step": 460 + }, + { + "epoch": 0.5052054794520547, + "grad_norm": 1.7809834480285645, + "learning_rate": 4.648659626320066e-06, + "loss": 2.2525, + "step": 461 + }, + { + "epoch": 0.5063013698630137, + "grad_norm": 1.9256166219711304, + "learning_rate": 4.646628757108043e-06, + "loss": 2.2611, + "step": 462 + }, + { + "epoch": 0.5073972602739726, + "grad_norm": 1.94192636013031, + "learning_rate": 4.6445978878960194e-06, + "loss": 2.2282, + "step": 463 + }, + { + "epoch": 0.5084931506849315, + "grad_norm": 1.8891818523406982, + "learning_rate": 4.6425670186839976e-06, + "loss": 2.2036, + "step": 464 + }, + { + "epoch": 0.5095890410958904, + "grad_norm": 2.352078676223755, + "learning_rate": 4.640536149471974e-06, + "loss": 2.2293, + "step": 465 + }, + { + "epoch": 0.5106849315068493, + "grad_norm": 1.8141659498214722, + "learning_rate": 4.638505280259952e-06, + "loss": 2.2551, + "step": 466 + }, + { + "epoch": 0.5117808219178083, + "grad_norm": 1.9492480754852295, + "learning_rate": 4.6364744110479285e-06, + "loss": 2.2641, + "step": 467 + }, + { + "epoch": 0.5128767123287671, + "grad_norm": 2.0210182666778564, + "learning_rate": 4.634443541835907e-06, + "loss": 2.2367, + "step": 468 + }, + { + "epoch": 0.513972602739726, + "grad_norm": 1.8936505317687988, + "learning_rate": 4.632412672623883e-06, + "loss": 2.265, + "step": 469 + }, + { + "epoch": 0.5150684931506849, + "grad_norm": 1.845888614654541, + "learning_rate": 4.630381803411861e-06, + "loss": 2.2169, + "step": 470 + }, + { + "epoch": 0.5150684931506849, + "eval_loss": 2.353489875793457, + "eval_runtime": 46.8739, + "eval_samples_per_second": 21.334, + "eval_steps_per_second": 0.896, + "step": 470 + }, + { + "epoch": 0.5161643835616438, + "grad_norm": 1.889675259590149, + "learning_rate": 4.628350934199838e-06, + "loss": 2.1383, + "step": 471 + }, + { + "epoch": 0.5172602739726028, + "grad_norm": 1.9799216985702515, + "learning_rate": 4.626320064987816e-06, + "loss": 2.2312, + "step": 472 + }, + { + "epoch": 0.5183561643835617, + "grad_norm": 2.0128607749938965, + "learning_rate": 4.624289195775792e-06, + "loss": 2.2529, + "step": 473 + }, + { + "epoch": 0.5194520547945205, + "grad_norm": 1.8791241645812988, + "learning_rate": 4.62225832656377e-06, + "loss": 2.226, + "step": 474 + }, + { + "epoch": 0.5205479452054794, + "grad_norm": 2.0497279167175293, + "learning_rate": 4.620227457351747e-06, + "loss": 2.2283, + "step": 475 + }, + { + "epoch": 0.5216438356164383, + "grad_norm": 1.8097424507141113, + "learning_rate": 4.618196588139724e-06, + "loss": 2.222, + "step": 476 + }, + { + "epoch": 0.5227397260273973, + "grad_norm": 1.8960500955581665, + "learning_rate": 4.616165718927701e-06, + "loss": 2.2342, + "step": 477 + }, + { + "epoch": 0.5238356164383562, + "grad_norm": 2.020730495452881, + "learning_rate": 4.6141348497156786e-06, + "loss": 2.2106, + "step": 478 + }, + { + "epoch": 0.5249315068493151, + "grad_norm": 1.8955297470092773, + "learning_rate": 4.612103980503656e-06, + "loss": 2.2755, + "step": 479 + }, + { + "epoch": 0.5260273972602739, + "grad_norm": 1.9767351150512695, + "learning_rate": 4.610073111291633e-06, + "loss": 2.2554, + "step": 480 + }, + { + "epoch": 0.5260273972602739, + "eval_loss": 2.347705841064453, + "eval_runtime": 46.9125, + "eval_samples_per_second": 21.316, + "eval_steps_per_second": 0.895, + "step": 480 + }, + { + "epoch": 0.5271232876712328, + "grad_norm": 1.936457633972168, + "learning_rate": 4.60804224207961e-06, + "loss": 2.2323, + "step": 481 + }, + { + "epoch": 0.5282191780821918, + "grad_norm": 1.9271862506866455, + "learning_rate": 4.606011372867588e-06, + "loss": 2.2151, + "step": 482 + }, + { + "epoch": 0.5293150684931507, + "grad_norm": 1.878153681755066, + "learning_rate": 4.603980503655565e-06, + "loss": 2.2322, + "step": 483 + }, + { + "epoch": 0.5304109589041096, + "grad_norm": 2.1009249687194824, + "learning_rate": 4.601949634443542e-06, + "loss": 2.2181, + "step": 484 + }, + { + "epoch": 0.5315068493150685, + "grad_norm": 1.9089181423187256, + "learning_rate": 4.5999187652315195e-06, + "loss": 2.1712, + "step": 485 + }, + { + "epoch": 0.5326027397260275, + "grad_norm": 2.207282781600952, + "learning_rate": 4.597887896019497e-06, + "loss": 2.1847, + "step": 486 + }, + { + "epoch": 0.5336986301369863, + "grad_norm": 2.001584053039551, + "learning_rate": 4.595857026807474e-06, + "loss": 2.214, + "step": 487 + }, + { + "epoch": 0.5347945205479452, + "grad_norm": 1.966700553894043, + "learning_rate": 4.593826157595451e-06, + "loss": 2.2237, + "step": 488 + }, + { + "epoch": 0.5358904109589041, + "grad_norm": 1.842469334602356, + "learning_rate": 4.591795288383429e-06, + "loss": 2.2089, + "step": 489 + }, + { + "epoch": 0.536986301369863, + "grad_norm": 2.0412986278533936, + "learning_rate": 4.589764419171406e-06, + "loss": 2.2734, + "step": 490 + }, + { + "epoch": 0.536986301369863, + "eval_loss": 2.3460772037506104, + "eval_runtime": 50.4048, + "eval_samples_per_second": 19.839, + "eval_steps_per_second": 0.833, + "step": 490 + }, + { + "epoch": 0.538082191780822, + "grad_norm": 1.9337031841278076, + "learning_rate": 4.587733549959383e-06, + "loss": 2.2314, + "step": 491 + }, + { + "epoch": 0.5391780821917809, + "grad_norm": 2.4448320865631104, + "learning_rate": 4.5857026807473604e-06, + "loss": 2.2098, + "step": 492 + }, + { + "epoch": 0.5402739726027397, + "grad_norm": 1.83162522315979, + "learning_rate": 4.583671811535338e-06, + "loss": 2.1902, + "step": 493 + }, + { + "epoch": 0.5413698630136986, + "grad_norm": 2.3614656925201416, + "learning_rate": 4.581640942323315e-06, + "loss": 2.1737, + "step": 494 + }, + { + "epoch": 0.5424657534246575, + "grad_norm": 2.001187801361084, + "learning_rate": 4.579610073111292e-06, + "loss": 2.2485, + "step": 495 + }, + { + "epoch": 0.5435616438356164, + "grad_norm": 2.3479933738708496, + "learning_rate": 4.5775792038992695e-06, + "loss": 2.2595, + "step": 496 + }, + { + "epoch": 0.5446575342465754, + "grad_norm": 2.135856866836548, + "learning_rate": 4.575548334687247e-06, + "loss": 2.2528, + "step": 497 + }, + { + "epoch": 0.5457534246575343, + "grad_norm": 1.983126163482666, + "learning_rate": 4.573517465475224e-06, + "loss": 2.2788, + "step": 498 + }, + { + "epoch": 0.5468493150684931, + "grad_norm": 2.18208909034729, + "learning_rate": 4.571486596263201e-06, + "loss": 2.1995, + "step": 499 + }, + { + "epoch": 0.547945205479452, + "grad_norm": 1.848747968673706, + "learning_rate": 4.569455727051178e-06, + "loss": 2.2403, + "step": 500 + }, + { + "epoch": 0.547945205479452, + "eval_loss": 2.344066619873047, + "eval_runtime": 51.1213, + "eval_samples_per_second": 19.561, + "eval_steps_per_second": 0.822, + "step": 500 + }, + { + "epoch": 0.549041095890411, + "grad_norm": 2.427860736846924, + "learning_rate": 4.567424857839156e-06, + "loss": 2.246, + "step": 501 + }, + { + "epoch": 0.5501369863013699, + "grad_norm": 1.913405418395996, + "learning_rate": 4.565393988627132e-06, + "loss": 2.2296, + "step": 502 + }, + { + "epoch": 0.5512328767123288, + "grad_norm": 2.0106029510498047, + "learning_rate": 4.5633631194151105e-06, + "loss": 2.2657, + "step": 503 + }, + { + "epoch": 0.5523287671232877, + "grad_norm": 2.2719595432281494, + "learning_rate": 4.561332250203087e-06, + "loss": 2.2341, + "step": 504 + }, + { + "epoch": 0.5534246575342465, + "grad_norm": 2.109786033630371, + "learning_rate": 4.559301380991065e-06, + "loss": 2.2181, + "step": 505 + }, + { + "epoch": 0.5545205479452054, + "grad_norm": 2.189779281616211, + "learning_rate": 4.5572705117790414e-06, + "loss": 2.1821, + "step": 506 + }, + { + "epoch": 0.5556164383561644, + "grad_norm": 2.0263590812683105, + "learning_rate": 4.5552396425670196e-06, + "loss": 2.2072, + "step": 507 + }, + { + "epoch": 0.5567123287671233, + "grad_norm": 2.215289354324341, + "learning_rate": 4.553208773354996e-06, + "loss": 2.2513, + "step": 508 + }, + { + "epoch": 0.5578082191780822, + "grad_norm": 2.3031351566314697, + "learning_rate": 4.551177904142974e-06, + "loss": 2.1988, + "step": 509 + }, + { + "epoch": 0.5589041095890411, + "grad_norm": 2.1830058097839355, + "learning_rate": 4.5491470349309505e-06, + "loss": 2.269, + "step": 510 + }, + { + "epoch": 0.5589041095890411, + "eval_loss": 2.342463731765747, + "eval_runtime": 43.1956, + "eval_samples_per_second": 23.151, + "eval_steps_per_second": 0.972, + "step": 510 + }, + { + "epoch": 0.56, + "grad_norm": 2.3210105895996094, + "learning_rate": 4.547116165718929e-06, + "loss": 2.2051, + "step": 511 + }, + { + "epoch": 0.5610958904109589, + "grad_norm": 1.9175361394882202, + "learning_rate": 4.545085296506905e-06, + "loss": 2.2304, + "step": 512 + }, + { + "epoch": 0.5621917808219178, + "grad_norm": 2.302947759628296, + "learning_rate": 4.543054427294882e-06, + "loss": 2.215, + "step": 513 + }, + { + "epoch": 0.5632876712328767, + "grad_norm": 1.8947396278381348, + "learning_rate": 4.54102355808286e-06, + "loss": 2.1926, + "step": 514 + }, + { + "epoch": 0.5643835616438356, + "grad_norm": 2.195268154144287, + "learning_rate": 4.538992688870837e-06, + "loss": 2.1937, + "step": 515 + }, + { + "epoch": 0.5654794520547946, + "grad_norm": 2.178874969482422, + "learning_rate": 4.536961819658814e-06, + "loss": 2.2248, + "step": 516 + }, + { + "epoch": 0.5665753424657535, + "grad_norm": 1.9856929779052734, + "learning_rate": 4.5349309504467915e-06, + "loss": 2.2077, + "step": 517 + }, + { + "epoch": 0.5676712328767123, + "grad_norm": 2.0686943531036377, + "learning_rate": 4.532900081234769e-06, + "loss": 2.2328, + "step": 518 + }, + { + "epoch": 0.5687671232876712, + "grad_norm": 1.9554919004440308, + "learning_rate": 4.530869212022746e-06, + "loss": 2.198, + "step": 519 + }, + { + "epoch": 0.5698630136986301, + "grad_norm": 1.9489670991897583, + "learning_rate": 4.528838342810723e-06, + "loss": 2.2287, + "step": 520 + }, + { + "epoch": 0.5698630136986301, + "eval_loss": 2.3415298461914062, + "eval_runtime": 48.7918, + "eval_samples_per_second": 20.495, + "eval_steps_per_second": 0.861, + "step": 520 + }, + { + "epoch": 0.570958904109589, + "grad_norm": 1.9868658781051636, + "learning_rate": 4.526807473598701e-06, + "loss": 2.2205, + "step": 521 + }, + { + "epoch": 0.572054794520548, + "grad_norm": 2.0965914726257324, + "learning_rate": 4.524776604386678e-06, + "loss": 2.2239, + "step": 522 + }, + { + "epoch": 0.5731506849315069, + "grad_norm": 2.041471481323242, + "learning_rate": 4.522745735174655e-06, + "loss": 2.2035, + "step": 523 + }, + { + "epoch": 0.5742465753424657, + "grad_norm": 1.885293960571289, + "learning_rate": 4.520714865962632e-06, + "loss": 2.2469, + "step": 524 + }, + { + "epoch": 0.5753424657534246, + "grad_norm": 1.9107296466827393, + "learning_rate": 4.51868399675061e-06, + "loss": 2.2386, + "step": 525 + }, + { + "epoch": 0.5764383561643835, + "grad_norm": 1.9945271015167236, + "learning_rate": 4.516653127538587e-06, + "loss": 2.235, + "step": 526 + }, + { + "epoch": 0.5775342465753425, + "grad_norm": 1.9787253141403198, + "learning_rate": 4.514622258326564e-06, + "loss": 2.2034, + "step": 527 + }, + { + "epoch": 0.5786301369863014, + "grad_norm": 2.0300636291503906, + "learning_rate": 4.5125913891145415e-06, + "loss": 2.1849, + "step": 528 + }, + { + "epoch": 0.5797260273972603, + "grad_norm": 2.0348970890045166, + "learning_rate": 4.510560519902519e-06, + "loss": 2.2174, + "step": 529 + }, + { + "epoch": 0.5808219178082191, + "grad_norm": 1.8589537143707275, + "learning_rate": 4.508529650690496e-06, + "loss": 2.2351, + "step": 530 + }, + { + "epoch": 0.5808219178082191, + "eval_loss": 2.3450186252593994, + "eval_runtime": 52.4548, + "eval_samples_per_second": 19.064, + "eval_steps_per_second": 0.801, + "step": 530 + }, + { + "epoch": 0.581917808219178, + "grad_norm": 2.0496020317077637, + "learning_rate": 4.506498781478473e-06, + "loss": 2.182, + "step": 531 + }, + { + "epoch": 0.583013698630137, + "grad_norm": 2.0649805068969727, + "learning_rate": 4.504467912266451e-06, + "loss": 2.1654, + "step": 532 + }, + { + "epoch": 0.5841095890410959, + "grad_norm": 1.9809486865997314, + "learning_rate": 4.502437043054428e-06, + "loss": 2.2182, + "step": 533 + }, + { + "epoch": 0.5852054794520548, + "grad_norm": 2.21915340423584, + "learning_rate": 4.500406173842405e-06, + "loss": 2.2697, + "step": 534 + }, + { + "epoch": 0.5863013698630137, + "grad_norm": 1.9822649955749512, + "learning_rate": 4.4983753046303824e-06, + "loss": 2.2297, + "step": 535 + }, + { + "epoch": 0.5873972602739727, + "grad_norm": 1.972575306892395, + "learning_rate": 4.49634443541836e-06, + "loss": 2.2164, + "step": 536 + }, + { + "epoch": 0.5884931506849315, + "grad_norm": 1.948667049407959, + "learning_rate": 4.494313566206336e-06, + "loss": 2.2087, + "step": 537 + }, + { + "epoch": 0.5895890410958904, + "grad_norm": 2.7486417293548584, + "learning_rate": 4.492282696994314e-06, + "loss": 2.2602, + "step": 538 + }, + { + "epoch": 0.5906849315068493, + "grad_norm": 2.3374552726745605, + "learning_rate": 4.490251827782291e-06, + "loss": 2.2054, + "step": 539 + }, + { + "epoch": 0.5917808219178082, + "grad_norm": 2.222116231918335, + "learning_rate": 4.488220958570269e-06, + "loss": 2.1868, + "step": 540 + }, + { + "epoch": 0.5917808219178082, + "eval_loss": 2.345618963241577, + "eval_runtime": 53.0302, + "eval_samples_per_second": 18.857, + "eval_steps_per_second": 0.792, + "step": 540 + }, + { + "epoch": 0.5928767123287672, + "grad_norm": 2.00752854347229, + "learning_rate": 4.486190089358245e-06, + "loss": 2.187, + "step": 541 + }, + { + "epoch": 0.5939726027397261, + "grad_norm": 2.2466132640838623, + "learning_rate": 4.484159220146223e-06, + "loss": 2.1611, + "step": 542 + }, + { + "epoch": 0.5950684931506849, + "grad_norm": 2.1279385089874268, + "learning_rate": 4.4821283509342e-06, + "loss": 2.248, + "step": 543 + }, + { + "epoch": 0.5961643835616438, + "grad_norm": 1.9140571355819702, + "learning_rate": 4.480097481722178e-06, + "loss": 2.2314, + "step": 544 + }, + { + "epoch": 0.5972602739726027, + "grad_norm": 2.120741605758667, + "learning_rate": 4.478066612510154e-06, + "loss": 2.212, + "step": 545 + }, + { + "epoch": 0.5983561643835617, + "grad_norm": 2.003713369369507, + "learning_rate": 4.4760357432981325e-06, + "loss": 2.1824, + "step": 546 + }, + { + "epoch": 0.5994520547945206, + "grad_norm": 1.9161221981048584, + "learning_rate": 4.474004874086109e-06, + "loss": 2.2015, + "step": 547 + }, + { + "epoch": 0.6005479452054795, + "grad_norm": 2.367110252380371, + "learning_rate": 4.471974004874087e-06, + "loss": 2.3122, + "step": 548 + }, + { + "epoch": 0.6016438356164383, + "grad_norm": 2.113380193710327, + "learning_rate": 4.4699431356620635e-06, + "loss": 2.1692, + "step": 549 + }, + { + "epoch": 0.6027397260273972, + "grad_norm": 1.974181890487671, + "learning_rate": 4.467912266450041e-06, + "loss": 2.2509, + "step": 550 + }, + { + "epoch": 0.6027397260273972, + "eval_loss": 2.342583656311035, + "eval_runtime": 50.7362, + "eval_samples_per_second": 19.71, + "eval_steps_per_second": 0.828, + "step": 550 + }, + { + "epoch": 0.6038356164383561, + "grad_norm": 1.889894962310791, + "learning_rate": 4.465881397238018e-06, + "loss": 2.2214, + "step": 551 + }, + { + "epoch": 0.6049315068493151, + "grad_norm": 1.8982343673706055, + "learning_rate": 4.463850528025995e-06, + "loss": 2.2135, + "step": 552 + }, + { + "epoch": 0.606027397260274, + "grad_norm": 1.9462087154388428, + "learning_rate": 4.4618196588139726e-06, + "loss": 2.2255, + "step": 553 + }, + { + "epoch": 0.6071232876712329, + "grad_norm": 1.8729931116104126, + "learning_rate": 4.45978878960195e-06, + "loss": 2.2358, + "step": 554 + }, + { + "epoch": 0.6082191780821918, + "grad_norm": 1.9230246543884277, + "learning_rate": 4.457757920389927e-06, + "loss": 2.222, + "step": 555 + }, + { + "epoch": 0.6093150684931506, + "grad_norm": 1.9495900869369507, + "learning_rate": 4.455727051177904e-06, + "loss": 2.1779, + "step": 556 + }, + { + "epoch": 0.6104109589041096, + "grad_norm": 1.983487606048584, + "learning_rate": 4.453696181965882e-06, + "loss": 2.2007, + "step": 557 + }, + { + "epoch": 0.6115068493150685, + "grad_norm": 2.1541049480438232, + "learning_rate": 4.451665312753859e-06, + "loss": 2.2513, + "step": 558 + }, + { + "epoch": 0.6126027397260274, + "grad_norm": 1.9133352041244507, + "learning_rate": 4.449634443541836e-06, + "loss": 2.2288, + "step": 559 + }, + { + "epoch": 0.6136986301369863, + "grad_norm": 2.3325741291046143, + "learning_rate": 4.4476035743298135e-06, + "loss": 2.2539, + "step": 560 + }, + { + "epoch": 0.6136986301369863, + "eval_loss": 2.3387348651885986, + "eval_runtime": 53.4301, + "eval_samples_per_second": 18.716, + "eval_steps_per_second": 0.786, + "step": 560 + }, + { + "epoch": 0.6147945205479453, + "grad_norm": 1.8099673986434937, + "learning_rate": 4.445572705117791e-06, + "loss": 2.1845, + "step": 561 + }, + { + "epoch": 0.6158904109589041, + "grad_norm": 2.201901912689209, + "learning_rate": 4.443541835905768e-06, + "loss": 2.195, + "step": 562 + }, + { + "epoch": 0.616986301369863, + "grad_norm": 1.8412714004516602, + "learning_rate": 4.441510966693745e-06, + "loss": 2.2551, + "step": 563 + }, + { + "epoch": 0.6180821917808219, + "grad_norm": 1.8761520385742188, + "learning_rate": 4.439480097481723e-06, + "loss": 2.1965, + "step": 564 + }, + { + "epoch": 0.6191780821917808, + "grad_norm": 1.9123467206954956, + "learning_rate": 4.4374492282697e-06, + "loss": 2.2502, + "step": 565 + }, + { + "epoch": 0.6202739726027398, + "grad_norm": 2.0631234645843506, + "learning_rate": 4.435418359057677e-06, + "loss": 2.1582, + "step": 566 + }, + { + "epoch": 0.6213698630136987, + "grad_norm": 2.3191616535186768, + "learning_rate": 4.433387489845654e-06, + "loss": 2.2271, + "step": 567 + }, + { + "epoch": 0.6224657534246575, + "grad_norm": 2.043057918548584, + "learning_rate": 4.431356620633632e-06, + "loss": 2.2292, + "step": 568 + }, + { + "epoch": 0.6235616438356164, + "grad_norm": 2.0716068744659424, + "learning_rate": 4.429325751421609e-06, + "loss": 2.196, + "step": 569 + }, + { + "epoch": 0.6246575342465753, + "grad_norm": 1.7669776678085327, + "learning_rate": 4.427294882209586e-06, + "loss": 2.1953, + "step": 570 + }, + { + "epoch": 0.6246575342465753, + "eval_loss": 2.336578607559204, + "eval_runtime": 50.0083, + "eval_samples_per_second": 19.997, + "eval_steps_per_second": 0.84, + "step": 570 + }, + { + "epoch": 0.6257534246575343, + "grad_norm": 1.8165771961212158, + "learning_rate": 4.4252640129975635e-06, + "loss": 2.2155, + "step": 571 + }, + { + "epoch": 0.6268493150684932, + "grad_norm": 1.8640249967575073, + "learning_rate": 4.423233143785541e-06, + "loss": 2.1648, + "step": 572 + }, + { + "epoch": 0.6279452054794521, + "grad_norm": 2.013869285583496, + "learning_rate": 4.421202274573518e-06, + "loss": 2.2131, + "step": 573 + }, + { + "epoch": 0.6290410958904109, + "grad_norm": 2.2650580406188965, + "learning_rate": 4.4191714053614945e-06, + "loss": 2.2287, + "step": 574 + }, + { + "epoch": 0.6301369863013698, + "grad_norm": 1.9232103824615479, + "learning_rate": 4.417140536149473e-06, + "loss": 2.1339, + "step": 575 + }, + { + "epoch": 0.6312328767123287, + "grad_norm": 1.910134196281433, + "learning_rate": 4.415109666937449e-06, + "loss": 2.1755, + "step": 576 + }, + { + "epoch": 0.6323287671232877, + "grad_norm": 2.2492005825042725, + "learning_rate": 4.413078797725427e-06, + "loss": 2.2362, + "step": 577 + }, + { + "epoch": 0.6334246575342466, + "grad_norm": 1.8584033250808716, + "learning_rate": 4.411047928513404e-06, + "loss": 2.1911, + "step": 578 + }, + { + "epoch": 0.6345205479452055, + "grad_norm": 2.156273126602173, + "learning_rate": 4.409017059301382e-06, + "loss": 2.2109, + "step": 579 + }, + { + "epoch": 0.6356164383561644, + "grad_norm": 1.7977535724639893, + "learning_rate": 4.406986190089358e-06, + "loss": 2.2195, + "step": 580 + }, + { + "epoch": 0.6356164383561644, + "eval_loss": 2.33555269241333, + "eval_runtime": 48.3919, + "eval_samples_per_second": 20.665, + "eval_steps_per_second": 0.868, + "step": 580 + }, + { + "epoch": 0.6367123287671232, + "grad_norm": 1.8262702226638794, + "learning_rate": 4.404955320877336e-06, + "loss": 2.2345, + "step": 581 + }, + { + "epoch": 0.6378082191780822, + "grad_norm": 1.9389383792877197, + "learning_rate": 4.402924451665313e-06, + "loss": 2.2347, + "step": 582 + }, + { + "epoch": 0.6389041095890411, + "grad_norm": 1.9548988342285156, + "learning_rate": 4.400893582453291e-06, + "loss": 2.1967, + "step": 583 + }, + { + "epoch": 0.64, + "grad_norm": 1.898092269897461, + "learning_rate": 4.398862713241267e-06, + "loss": 2.1795, + "step": 584 + }, + { + "epoch": 0.6410958904109589, + "grad_norm": 1.8183610439300537, + "learning_rate": 4.396831844029245e-06, + "loss": 2.2031, + "step": 585 + }, + { + "epoch": 0.6421917808219179, + "grad_norm": 2.1520016193389893, + "learning_rate": 4.394800974817222e-06, + "loss": 2.2051, + "step": 586 + }, + { + "epoch": 0.6432876712328767, + "grad_norm": 1.9930012226104736, + "learning_rate": 4.392770105605199e-06, + "loss": 2.2053, + "step": 587 + }, + { + "epoch": 0.6443835616438356, + "grad_norm": 2.311450958251953, + "learning_rate": 4.390739236393176e-06, + "loss": 2.2252, + "step": 588 + }, + { + "epoch": 0.6454794520547945, + "grad_norm": 2.0168697834014893, + "learning_rate": 4.388708367181154e-06, + "loss": 2.2286, + "step": 589 + }, + { + "epoch": 0.6465753424657534, + "grad_norm": 2.055964469909668, + "learning_rate": 4.386677497969131e-06, + "loss": 2.2278, + "step": 590 + }, + { + "epoch": 0.6465753424657534, + "eval_loss": 2.334367275238037, + "eval_runtime": 48.9414, + "eval_samples_per_second": 20.433, + "eval_steps_per_second": 0.858, + "step": 590 + }, + { + "epoch": 0.6476712328767124, + "grad_norm": 2.0161590576171875, + "learning_rate": 4.384646628757108e-06, + "loss": 2.2102, + "step": 591 + }, + { + "epoch": 0.6487671232876713, + "grad_norm": 2.0827667713165283, + "learning_rate": 4.3826157595450855e-06, + "loss": 2.2531, + "step": 592 + }, + { + "epoch": 0.6498630136986301, + "grad_norm": 2.077637195587158, + "learning_rate": 4.380584890333063e-06, + "loss": 2.2091, + "step": 593 + }, + { + "epoch": 0.650958904109589, + "grad_norm": 2.023892879486084, + "learning_rate": 4.37855402112104e-06, + "loss": 2.1684, + "step": 594 + }, + { + "epoch": 0.6520547945205479, + "grad_norm": 1.8509857654571533, + "learning_rate": 4.376523151909017e-06, + "loss": 2.1439, + "step": 595 + }, + { + "epoch": 0.6531506849315069, + "grad_norm": 2.1649608612060547, + "learning_rate": 4.3744922826969946e-06, + "loss": 2.2557, + "step": 596 + }, + { + "epoch": 0.6542465753424658, + "grad_norm": 2.1351773738861084, + "learning_rate": 4.372461413484972e-06, + "loss": 2.177, + "step": 597 + }, + { + "epoch": 0.6553424657534247, + "grad_norm": 2.1713974475860596, + "learning_rate": 4.370430544272949e-06, + "loss": 2.2223, + "step": 598 + }, + { + "epoch": 0.6564383561643835, + "grad_norm": 2.2905397415161133, + "learning_rate": 4.368399675060926e-06, + "loss": 2.2373, + "step": 599 + }, + { + "epoch": 0.6575342465753424, + "grad_norm": 2.007160186767578, + "learning_rate": 4.366368805848904e-06, + "loss": 2.2355, + "step": 600 + }, + { + "epoch": 0.6575342465753424, + "eval_loss": 2.3322324752807617, + "eval_runtime": 46.2769, + "eval_samples_per_second": 21.609, + "eval_steps_per_second": 0.908, + "step": 600 + } + ], + "logging_steps": 1, + "max_steps": 2736, + "num_input_tokens_seen": 0, + "num_train_epochs": 3, + "save_steps": 100, + "stateful_callbacks": { + "TrainerControl": { + "args": { + "should_epoch_stop": false, + "should_evaluate": false, + "should_log": false, + "should_save": true, + "should_training_stop": false + }, + "attributes": {} + } + }, + "total_flos": 1.870385645617152e+17, + "train_batch_size": 3, + "trial_name": null, + "trial_params": null +} diff --git a/babylm_shuffle_nondeterministic_10M_seed0/runs/checkpoint-600/zero_to_fp32.py b/babylm_shuffle_nondeterministic_10M_seed0/runs/checkpoint-600/zero_to_fp32.py new file mode 100644 index 0000000000000000000000000000000000000000..24cc342e78d1a006c782b3a4cd68d9ce786d8fd8 --- /dev/null +++ b/babylm_shuffle_nondeterministic_10M_seed0/runs/checkpoint-600/zero_to_fp32.py @@ -0,0 +1,604 @@ +#!/usr/bin/env python + +# Copyright (c) Microsoft Corporation. +# SPDX-License-Identifier: Apache-2.0 + +# DeepSpeed Team + +# This script extracts fp32 consolidated weights from a zero 1, 2 and 3 DeepSpeed checkpoints. It gets +# copied into the top level checkpoint dir, so the user can easily do the conversion at any point in +# the future. Once extracted, the weights don't require DeepSpeed and can be used in any +# application. +# +# example: python zero_to_fp32.py . pytorch_model.bin + +import argparse +import torch +import glob +import math +import os +import re +from collections import OrderedDict +from dataclasses import dataclass + +# while this script doesn't use deepspeed to recover data, since the checkpoints are pickled with +# DeepSpeed data structures it has to be available in the current python environment. +from deepspeed.utils import logger +from deepspeed.checkpoint.constants import (DS_VERSION, OPTIMIZER_STATE_DICT, SINGLE_PARTITION_OF_FP32_GROUPS, + FP32_FLAT_GROUPS, ZERO_STAGE, PARTITION_COUNT, PARAM_SHAPES, BUFFER_NAMES, + FROZEN_PARAM_SHAPES, FROZEN_PARAM_FRAGMENTS) + + +@dataclass +class zero_model_state: + buffers: dict() + param_shapes: dict() + shared_params: list + ds_version: int + frozen_param_shapes: dict() + frozen_param_fragments: dict() + + +debug = 0 + +# load to cpu +device = torch.device('cpu') + + +def atoi(text): + return int(text) if text.isdigit() else text + + +def natural_keys(text): + ''' + alist.sort(key=natural_keys) sorts in human order + http://nedbatchelder.com/blog/200712/human_sorting.html + (See Toothy's implementation in the comments) + ''' + return [atoi(c) for c in re.split(r'(\d+)', text)] + + +def get_model_state_file(checkpoint_dir, zero_stage): + if not os.path.isdir(checkpoint_dir): + raise FileNotFoundError(f"Directory '{checkpoint_dir}' doesn't exist") + + # there should be only one file + if zero_stage <= 2: + file = os.path.join(checkpoint_dir, "mp_rank_00_model_states.pt") + elif zero_stage == 3: + file = os.path.join(checkpoint_dir, "zero_pp_rank_0_mp_rank_00_model_states.pt") + + if not os.path.exists(file): + raise FileNotFoundError(f"can't find model states file at '{file}'") + + return file + + +def get_checkpoint_files(checkpoint_dir, glob_pattern): + # XXX: need to test that this simple glob rule works for multi-node setup too + ckpt_files = sorted(glob.glob(os.path.join(checkpoint_dir, glob_pattern)), key=natural_keys) + + if len(ckpt_files) == 0: + raise FileNotFoundError(f"can't find {glob_pattern} files in directory '{checkpoint_dir}'") + + return ckpt_files + + +def get_optim_files(checkpoint_dir): + return get_checkpoint_files(checkpoint_dir, "*_optim_states.pt") + + +def get_model_state_files(checkpoint_dir): + return get_checkpoint_files(checkpoint_dir, "*_model_states.pt") + + +def parse_model_states(files): + zero_model_states = [] + for file in files: + state_dict = torch.load(file, map_location=device) + + if BUFFER_NAMES not in state_dict: + raise ValueError(f"{file} is not a model state checkpoint") + buffer_names = state_dict[BUFFER_NAMES] + if debug: + print("Found buffers:", buffer_names) + + # recover just the buffers while restoring them to fp32 if they were saved in fp16 + buffers = {k: v.float() for k, v in state_dict["module"].items() if k in buffer_names} + param_shapes = state_dict[PARAM_SHAPES] + + # collect parameters that are included in param_shapes + param_names = [] + for s in param_shapes: + for name in s.keys(): + param_names.append(name) + + # update with frozen parameters + frozen_param_shapes = state_dict.get(FROZEN_PARAM_SHAPES, None) + if frozen_param_shapes is not None: + if debug: + print(f"Found frozen_param_shapes: {frozen_param_shapes}") + param_names += list(frozen_param_shapes.keys()) + + # handle shared params + shared_params = [[k, v] for k, v in state_dict["shared_params"].items()] + + ds_version = state_dict.get(DS_VERSION, None) + + frozen_param_fragments = state_dict.get(FROZEN_PARAM_FRAGMENTS, None) + + z_model_state = zero_model_state(buffers=buffers, + param_shapes=param_shapes, + shared_params=shared_params, + ds_version=ds_version, + frozen_param_shapes=frozen_param_shapes, + frozen_param_fragments=frozen_param_fragments) + zero_model_states.append(z_model_state) + + return zero_model_states + + +def parse_optim_states(files, ds_checkpoint_dir): + + total_files = len(files) + state_dicts = [] + for f in files: + state_dict = torch.load(f, map_location=device) + # immediately discard the potentially huge 2 optimizer states as we only care for fp32 master weights + # and also handle the case where it was already removed by another helper script + state_dict["optimizer_state_dict"].pop("optimizer_state_dict", None) + state_dicts.append(state_dict) + + if not ZERO_STAGE in state_dicts[0][OPTIMIZER_STATE_DICT]: + raise ValueError(f"{files[0]} is not a zero checkpoint") + zero_stage = state_dicts[0][OPTIMIZER_STATE_DICT][ZERO_STAGE] + world_size = state_dicts[0][OPTIMIZER_STATE_DICT][PARTITION_COUNT] + + # For ZeRO-2 each param group can have different partition_count as data parallelism for expert + # parameters can be different from data parallelism for non-expert parameters. So we can just + # use the max of the partition_count to get the dp world_size. + + if type(world_size) is list: + world_size = max(world_size) + + if world_size != total_files: + raise ValueError( + f"Expected {world_size} of '*_optim_states.pt' under '{ds_checkpoint_dir}' but found {total_files} files. " + "Possibly due to an overwrite of an old checkpoint, or a checkpoint didn't get saved by one or more processes." + ) + + # the groups are named differently in each stage + if zero_stage <= 2: + fp32_groups_key = SINGLE_PARTITION_OF_FP32_GROUPS + elif zero_stage == 3: + fp32_groups_key = FP32_FLAT_GROUPS + else: + raise ValueError(f"unknown zero stage {zero_stage}") + + if zero_stage <= 2: + fp32_flat_groups = [state_dicts[i][OPTIMIZER_STATE_DICT][fp32_groups_key] for i in range(len(state_dicts))] + elif zero_stage == 3: + # if there is more than one param group, there will be multiple flattened tensors - one + # flattened tensor per group - for simplicity merge them into a single tensor + # + # XXX: could make the script more memory efficient for when there are multiple groups - it + # will require matching the sub-lists of param_shapes for each param group flattened tensor + + fp32_flat_groups = [ + torch.cat(state_dicts[i][OPTIMIZER_STATE_DICT][fp32_groups_key], 0) for i in range(len(state_dicts)) + ] + + return zero_stage, world_size, fp32_flat_groups + + +def _get_fp32_state_dict_from_zero_checkpoint(ds_checkpoint_dir, exclude_frozen_parameters): + """ + Returns fp32 state_dict reconstructed from ds checkpoint + + Args: + - ``ds_checkpoint_dir``: path to the deepspeed checkpoint folder (where the optimizer files are) + + """ + print(f"Processing zero checkpoint '{ds_checkpoint_dir}'") + + optim_files = get_optim_files(ds_checkpoint_dir) + zero_stage, world_size, fp32_flat_groups = parse_optim_states(optim_files, ds_checkpoint_dir) + print(f"Detected checkpoint of type zero stage {zero_stage}, world_size: {world_size}") + + model_files = get_model_state_files(ds_checkpoint_dir) + + zero_model_states = parse_model_states(model_files) + print(f'Parsing checkpoint created by deepspeed=={zero_model_states[0].ds_version}') + + if zero_stage <= 2: + return _get_fp32_state_dict_from_zero2_checkpoint(world_size, fp32_flat_groups, zero_model_states, + exclude_frozen_parameters) + elif zero_stage == 3: + return _get_fp32_state_dict_from_zero3_checkpoint(world_size, fp32_flat_groups, zero_model_states, + exclude_frozen_parameters) + + +def _zero2_merge_frozen_params(state_dict, zero_model_states): + if zero_model_states[0].frozen_param_shapes is None or len(zero_model_states[0].frozen_param_shapes) == 0: + return + + frozen_param_shapes = zero_model_states[0].frozen_param_shapes + frozen_param_fragments = zero_model_states[0].frozen_param_fragments + + if debug: + num_elem = sum(s.numel() for s in frozen_param_shapes.values()) + print(f'rank 0: {FROZEN_PARAM_SHAPES}.numel = {num_elem}') + + wanted_params = len(frozen_param_shapes) + wanted_numel = sum(s.numel() for s in frozen_param_shapes.values()) + avail_numel = sum([p.numel() for p in frozen_param_fragments.values()]) + print(f'Frozen params: Have {avail_numel} numels to process.') + print(f'Frozen params: Need {wanted_numel} numels in {wanted_params} params') + + total_params = 0 + total_numel = 0 + for name, shape in frozen_param_shapes.items(): + total_params += 1 + unpartitioned_numel = shape.numel() + total_numel += unpartitioned_numel + + state_dict[name] = frozen_param_fragments[name] + + if debug: + print(f"{name} full shape: {shape} unpartitioned numel {unpartitioned_numel} ") + + print(f"Reconstructed Frozen fp32 state dict with {total_params} params {total_numel} elements") + + +def _has_callable(obj, fn): + attr = getattr(obj, fn, None) + return callable(attr) + + +def _zero2_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states): + param_shapes = zero_model_states[0].param_shapes + + # Reconstruction protocol: + # + # XXX: document this + + if debug: + for i in range(world_size): + for j in range(len(fp32_flat_groups[0])): + print(f"{FP32_FLAT_GROUPS}[{i}][{j}].shape={fp32_flat_groups[i][j].shape}") + + # XXX: memory usage doubles here (zero2) + num_param_groups = len(fp32_flat_groups[0]) + merged_single_partition_of_fp32_groups = [] + for i in range(num_param_groups): + merged_partitions = [sd[i] for sd in fp32_flat_groups] + full_single_fp32_vector = torch.cat(merged_partitions, 0) + merged_single_partition_of_fp32_groups.append(full_single_fp32_vector) + avail_numel = sum( + [full_single_fp32_vector.numel() for full_single_fp32_vector in merged_single_partition_of_fp32_groups]) + + if debug: + wanted_params = sum([len(shapes) for shapes in param_shapes]) + wanted_numel = sum([sum(shape.numel() for shape in shapes.values()) for shapes in param_shapes]) + # not asserting if there is a mismatch due to possible padding + print(f"Have {avail_numel} numels to process.") + print(f"Need {wanted_numel} numels in {wanted_params} params.") + + # params + # XXX: for huge models that can't fit into the host's RAM we will have to recode this to support + # out-of-core computing solution + total_numel = 0 + total_params = 0 + for shapes, full_single_fp32_vector in zip(param_shapes, merged_single_partition_of_fp32_groups): + offset = 0 + avail_numel = full_single_fp32_vector.numel() + for name, shape in shapes.items(): + + unpartitioned_numel = shape.numel() if _has_callable(shape, 'numel') else math.prod(shape) + total_numel += unpartitioned_numel + total_params += 1 + + if debug: + print(f"{name} full shape: {shape} unpartitioned numel {unpartitioned_numel} ") + state_dict[name] = full_single_fp32_vector.narrow(0, offset, unpartitioned_numel).view(shape) + offset += unpartitioned_numel + + # Z2 started to align to 2*world_size to improve nccl performance. Therefore both offset and + # avail_numel can differ by anywhere between 0..2*world_size. Due to two unrelated complex + # paddings performed in the code it's almost impossible to predict the exact numbers w/o the + # live optimizer object, so we are checking that the numbers are within the right range + align_to = 2 * world_size + + def zero2_align(x): + return align_to * math.ceil(x / align_to) + + if debug: + print(f"original offset={offset}, avail_numel={avail_numel}") + + offset = zero2_align(offset) + avail_numel = zero2_align(avail_numel) + + if debug: + print(f"aligned offset={offset}, avail_numel={avail_numel}") + + # Sanity check + if offset != avail_numel: + raise ValueError(f"consumed {offset} numels out of {avail_numel} - something is wrong") + + print(f"Reconstructed fp32 state dict with {total_params} params {total_numel} elements") + + +def _get_fp32_state_dict_from_zero2_checkpoint(world_size, fp32_flat_groups, zero_model_states, + exclude_frozen_parameters): + state_dict = OrderedDict() + + # buffers + buffers = zero_model_states[0].buffers + state_dict.update(buffers) + if debug: + print(f"added {len(buffers)} buffers") + + if not exclude_frozen_parameters: + _zero2_merge_frozen_params(state_dict, zero_model_states) + + _zero2_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states) + + # recover shared parameters + for pair in zero_model_states[0].shared_params: + if pair[1] in state_dict: + state_dict[pair[0]] = state_dict[pair[1]] + + return state_dict + + +def zero3_partitioned_param_info(unpartitioned_numel, world_size): + remainder = unpartitioned_numel % world_size + padding_numel = (world_size - remainder) if remainder else 0 + partitioned_numel = math.ceil(unpartitioned_numel / world_size) + return partitioned_numel, padding_numel + + +def _zero3_merge_frozen_params(state_dict, world_size, zero_model_states): + if zero_model_states[0].frozen_param_shapes is None or len(zero_model_states[0].frozen_param_shapes) == 0: + return + + if debug: + for i in range(world_size): + num_elem = sum(s.numel() for s in zero_model_states[i].frozen_param_fragments.values()) + print(f'rank {i}: {FROZEN_PARAM_SHAPES}.numel = {num_elem}') + + frozen_param_shapes = zero_model_states[0].frozen_param_shapes + wanted_params = len(frozen_param_shapes) + wanted_numel = sum(s.numel() for s in frozen_param_shapes.values()) + avail_numel = sum([p.numel() for p in zero_model_states[0].frozen_param_fragments.values()]) * world_size + print(f'Frozen params: Have {avail_numel} numels to process.') + print(f'Frozen params: Need {wanted_numel} numels in {wanted_params} params') + + total_params = 0 + total_numel = 0 + for name, shape in zero_model_states[0].frozen_param_shapes.items(): + total_params += 1 + unpartitioned_numel = shape.numel() + total_numel += unpartitioned_numel + + param_frags = tuple(model_state.frozen_param_fragments[name] for model_state in zero_model_states) + state_dict[name] = torch.cat(param_frags, 0).narrow(0, 0, unpartitioned_numel).view(shape) + + partitioned_numel, partitioned_padding_numel = zero3_partitioned_param_info(unpartitioned_numel, world_size) + + if debug: + print( + f"Frozen params: {total_params} {name} full shape: {shape} partition0 numel={partitioned_numel} partitioned_padding_numel={partitioned_padding_numel}" + ) + + print(f"Reconstructed Frozen fp32 state dict with {total_params} params {total_numel} elements") + + +def _zero3_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states): + param_shapes = zero_model_states[0].param_shapes + avail_numel = fp32_flat_groups[0].numel() * world_size + # Reconstruction protocol: For zero3 we need to zip the partitions together at boundary of each + # param, re-consolidating each param, while dealing with padding if any + + # merge list of dicts, preserving order + param_shapes = {k: v for d in param_shapes for k, v in d.items()} + + if debug: + for i in range(world_size): + print(f"{FP32_FLAT_GROUPS}[{i}].shape={fp32_flat_groups[i].shape}") + + wanted_params = len(param_shapes) + wanted_numel = sum(shape.numel() for shape in param_shapes.values()) + # not asserting if there is a mismatch due to possible padding + avail_numel = fp32_flat_groups[0].numel() * world_size + print(f"Trainable params: Have {avail_numel} numels to process.") + print(f"Trainable params: Need {wanted_numel} numels in {wanted_params} params.") + + # params + # XXX: for huge models that can't fit into the host's RAM we will have to recode this to support + # out-of-core computing solution + offset = 0 + total_numel = 0 + total_params = 0 + for name, shape in param_shapes.items(): + + unpartitioned_numel = shape.numel() + total_numel += unpartitioned_numel + total_params += 1 + + partitioned_numel, partitioned_padding_numel = zero3_partitioned_param_info(unpartitioned_numel, world_size) + + if debug: + print( + f"Trainable params: {total_params} {name} full shape: {shape} partition0 numel={partitioned_numel} partitioned_padding_numel={partitioned_padding_numel}" + ) + + # XXX: memory usage doubles here + state_dict[name] = torch.cat( + tuple(fp32_flat_groups[i].narrow(0, offset, partitioned_numel) for i in range(world_size)), + 0).narrow(0, 0, unpartitioned_numel).view(shape) + offset += partitioned_numel + + offset *= world_size + + # Sanity check + if offset != avail_numel: + raise ValueError(f"consumed {offset} numels out of {avail_numel} - something is wrong") + + print(f"Reconstructed Trainable fp32 state dict with {total_params} params {total_numel} elements") + + +def _get_fp32_state_dict_from_zero3_checkpoint(world_size, fp32_flat_groups, zero_model_states, + exclude_frozen_parameters): + state_dict = OrderedDict() + + # buffers + buffers = zero_model_states[0].buffers + state_dict.update(buffers) + if debug: + print(f"added {len(buffers)} buffers") + + if not exclude_frozen_parameters: + _zero3_merge_frozen_params(state_dict, world_size, zero_model_states) + + _zero3_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states) + + # recover shared parameters + for pair in zero_model_states[0].shared_params: + if pair[1] in state_dict: + state_dict[pair[0]] = state_dict[pair[1]] + + return state_dict + + +def get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir, tag=None, exclude_frozen_parameters=False): + """ + Convert ZeRO 2 or 3 checkpoint into a single fp32 consolidated state_dict that can be loaded with + ``load_state_dict()`` and used for training without DeepSpeed or shared with others, for example + via a model hub. + + Args: + - ``checkpoint_dir``: path to the desired checkpoint folder + - ``tag``: checkpoint tag used as a unique identifier for checkpoint. If not provided will attempt to load tag in 'latest' file. e.g., ``global_step14`` + - ``exclude_frozen_parameters``: exclude frozen parameters + + Returns: + - pytorch ``state_dict`` + + Note: this approach may not work if your application doesn't have sufficient free CPU memory and + you may need to use the offline approach using the ``zero_to_fp32.py`` script that is saved with + the checkpoint. + + A typical usage might be :: + + from deepspeed.utils.zero_to_fp32 import get_fp32_state_dict_from_zero_checkpoint + # do the training and checkpoint saving + state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir) # already on cpu + model = model.cpu() # move to cpu + model.load_state_dict(state_dict) + # submit to model hub or save the model to share with others + + In this example the ``model`` will no longer be usable in the deepspeed context of the same + application. i.e. you will need to re-initialize the deepspeed engine, since + ``model.load_state_dict(state_dict)`` will remove all the deepspeed magic from it. + + If you want it all done for you, use ``load_state_dict_from_zero_checkpoint`` instead. + + """ + if tag is None: + latest_path = os.path.join(checkpoint_dir, 'latest') + if os.path.isfile(latest_path): + with open(latest_path, 'r') as fd: + tag = fd.read().strip() + else: + raise ValueError(f"Unable to find 'latest' file at {latest_path}") + + ds_checkpoint_dir = os.path.join(checkpoint_dir, tag) + + if not os.path.isdir(ds_checkpoint_dir): + raise FileNotFoundError(f"Directory '{ds_checkpoint_dir}' doesn't exist") + + return _get_fp32_state_dict_from_zero_checkpoint(ds_checkpoint_dir, exclude_frozen_parameters) + + +def convert_zero_checkpoint_to_fp32_state_dict(checkpoint_dir, output_file, tag=None, exclude_frozen_parameters=False): + """ + Convert ZeRO 2 or 3 checkpoint into a single fp32 consolidated ``state_dict`` file that can be + loaded with ``torch.load(file)`` + ``load_state_dict()`` and used for training without DeepSpeed. + + Args: + - ``checkpoint_dir``: path to the desired checkpoint folder. (one that contains the tag-folder, like ``global_step14``) + - ``output_file``: path to the pytorch fp32 state_dict output file (e.g. path/pytorch_model.bin) + - ``tag``: checkpoint tag used as a unique identifier for checkpoint. If not provided will attempt to load tag in the file named ``latest`` in the checkpoint folder, e.g., ``global_step14`` + - ``exclude_frozen_parameters``: exclude frozen parameters + """ + + state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir, tag, exclude_frozen_parameters) + print(f"Saving fp32 state dict to {output_file}") + torch.save(state_dict, output_file) + + +def load_state_dict_from_zero_checkpoint(model, checkpoint_dir, tag=None): + """ + 1. Put the provided model to cpu + 2. Convert ZeRO 2 or 3 checkpoint into a single fp32 consolidated ``state_dict`` + 3. Load it into the provided model + + Args: + - ``model``: the model object to update + - ``checkpoint_dir``: path to the desired checkpoint folder. (one that contains the tag-folder, like ``global_step14``) + - ``tag``: checkpoint tag used as a unique identifier for checkpoint. If not provided will attempt to load tag in the file named ``latest`` in the checkpoint folder, e.g., ``global_step14`` + + Returns: + - ``model`: modified model + + Make sure you have plenty of CPU memory available before you call this function. If you don't + have enough use the ``zero_to_fp32.py`` utility to do the conversion. You will find it + conveniently placed for you in the checkpoint folder. + + A typical usage might be :: + + from deepspeed.utils.zero_to_fp32 import load_state_dict_from_zero_checkpoint + model = load_state_dict_from_zero_checkpoint(trainer.model, checkpoint_dir) + # submit to model hub or save the model to share with others + + Note, that once this was run, the ``model`` will no longer be usable in the deepspeed context + of the same application. i.e. you will need to re-initialize the deepspeed engine, since + ``model.load_state_dict(state_dict)`` will remove all the deepspeed magic from it. + + """ + logger.info(f"Extracting fp32 weights") + state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir, tag) + + logger.info(f"Overwriting model with fp32 weights") + model = model.cpu() + model.load_state_dict(state_dict, strict=False) + + return model + + +if __name__ == "__main__": + + parser = argparse.ArgumentParser() + parser.add_argument("checkpoint_dir", + type=str, + help="path to the desired checkpoint folder, e.g., path/checkpoint-12") + parser.add_argument( + "output_file", + type=str, + help="path to the pytorch fp32 state_dict output file (e.g. path/checkpoint-12/pytorch_model.bin)") + parser.add_argument("-t", + "--tag", + type=str, + default=None, + help="checkpoint tag used as a unique identifier for checkpoint. e.g., global_step1") + parser.add_argument("--exclude_frozen_parameters", action='store_true', help="exclude frozen parameters") + parser.add_argument("-d", "--debug", action='store_true', help="enable debug") + args = parser.parse_args() + + debug = args.debug + + convert_zero_checkpoint_to_fp32_state_dict(args.checkpoint_dir, + args.output_file, + tag=args.tag, + exclude_frozen_parameters=args.exclude_frozen_parameters) diff --git a/babylm_shuffle_nondeterministic_10M_seed0/runs/checkpoint-700/config.json b/babylm_shuffle_nondeterministic_10M_seed0/runs/checkpoint-700/config.json new file mode 100644 index 0000000000000000000000000000000000000000..fb0f9a53a507ffcdb60410deb71d60ef801bf350 --- /dev/null +++ b/babylm_shuffle_nondeterministic_10M_seed0/runs/checkpoint-700/config.json @@ -0,0 +1,36 @@ +{ + "_name_or_path": "meta-llama/Llama-3.2-3B", + "architectures": [ + "LlamaForCausalLM" + ], + "attention_bias": false, + "attention_dropout": 0.0, + "bos_token_id": 128000, + "eos_token_id": 128001, + "head_dim": 128, + "hidden_act": "silu", + "hidden_size": 3072, + "initializer_range": 0.02, + "intermediate_size": 8192, + "max_position_embeddings": 131072, + "mlp_bias": false, + "model_type": "llama", + "num_attention_heads": 24, + "num_hidden_layers": 28, + "num_key_value_heads": 8, + "pretraining_tp": 1, + "rms_norm_eps": 1e-05, + "rope_scaling": { + "factor": 32.0, + "high_freq_factor": 4.0, + "low_freq_factor": 1.0, + "original_max_position_embeddings": 8192, + "rope_type": "llama3" + }, + "rope_theta": 500000.0, + "tie_word_embeddings": true, + "torch_dtype": "float16", + "transformers_version": "4.45.1", + "use_cache": true, + "vocab_size": 128256 +} diff --git a/babylm_shuffle_nondeterministic_10M_seed0/runs/checkpoint-700/generation_config.json b/babylm_shuffle_nondeterministic_10M_seed0/runs/checkpoint-700/generation_config.json new file mode 100644 index 0000000000000000000000000000000000000000..9c389b841a9a29ddd904e02d1eb0e08dcce82ad9 --- /dev/null +++ b/babylm_shuffle_nondeterministic_10M_seed0/runs/checkpoint-700/generation_config.json @@ -0,0 +1,9 @@ +{ + "_from_model_config": true, + "bos_token_id": 128000, + "do_sample": true, + "eos_token_id": 128001, + "temperature": 0.6, + "top_p": 0.9, + "transformers_version": "4.45.1" +} diff --git a/babylm_shuffle_nondeterministic_10M_seed0/runs/checkpoint-700/latest b/babylm_shuffle_nondeterministic_10M_seed0/runs/checkpoint-700/latest new file mode 100644 index 0000000000000000000000000000000000000000..a0f3e526aa9af2ac647b278f006bb9616843c5d6 --- /dev/null +++ b/babylm_shuffle_nondeterministic_10M_seed0/runs/checkpoint-700/latest @@ -0,0 +1 @@ +global_step700 \ No newline at end of file diff --git a/babylm_shuffle_nondeterministic_10M_seed0/runs/checkpoint-700/model.safetensors.index.json b/babylm_shuffle_nondeterministic_10M_seed0/runs/checkpoint-700/model.safetensors.index.json new file mode 100644 index 0000000000000000000000000000000000000000..ed64de846d720b9a7859dc20575fea8e8ca51940 --- /dev/null +++ b/babylm_shuffle_nondeterministic_10M_seed0/runs/checkpoint-700/model.safetensors.index.json @@ -0,0 +1,262 @@ +{ + "metadata": { + "total_size": 7213504512 + }, + "weight_map": { + "lm_head.weight": "model-00002-of-00002.safetensors", + "model.embed_tokens.weight": "model-00001-of-00002.safetensors", + "model.layers.0.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.0.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.0.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.0.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.0.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.0.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.0.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.0.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.0.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.1.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.1.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.1.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.1.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.1.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.1.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.1.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.1.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.1.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.10.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.10.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.10.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.10.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.10.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.10.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.10.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.10.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.10.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.11.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.11.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.11.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.11.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.11.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.11.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.11.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.11.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.11.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.12.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.12.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.12.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.12.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.12.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.12.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.12.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.12.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.12.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.13.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.13.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.13.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.13.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.13.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.13.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.13.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.13.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.13.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.14.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.14.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.14.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.14.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.14.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.14.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.14.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.14.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.14.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.15.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.15.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.15.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.15.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.15.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.15.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.15.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.15.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.15.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.16.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.16.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.16.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.16.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.16.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.16.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.16.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.16.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.16.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.17.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.17.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.17.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.17.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.17.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.17.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.17.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.17.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.17.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.18.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.18.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.18.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.18.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.18.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.18.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.18.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.18.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.18.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.19.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.19.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.19.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.19.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.19.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.19.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.19.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.19.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.19.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.2.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.2.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.2.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.2.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.2.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.2.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.2.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.2.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.2.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.20.input_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.20.mlp.down_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.20.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.20.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.20.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.20.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.20.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.20.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.20.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.21.input_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.21.mlp.down_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.21.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.21.mlp.up_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.21.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.21.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.21.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.21.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.21.self_attn.v_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.22.input_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.22.mlp.down_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.22.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.22.mlp.up_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.22.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.22.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.22.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.22.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.22.self_attn.v_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.23.input_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.23.mlp.down_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.23.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.23.mlp.up_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.23.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.23.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.23.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.23.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.23.self_attn.v_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.24.input_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.24.mlp.down_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.24.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.24.mlp.up_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.24.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.24.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.24.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.24.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.24.self_attn.v_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.25.input_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.25.mlp.down_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.25.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.25.mlp.up_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.25.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.25.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.25.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.25.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.25.self_attn.v_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.26.input_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.26.mlp.down_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.26.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.26.mlp.up_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.26.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.26.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.26.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.26.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.26.self_attn.v_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.27.input_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.27.mlp.down_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.27.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.27.mlp.up_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.27.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.27.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.27.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.27.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.27.self_attn.v_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.3.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.3.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.3.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.3.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.3.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.3.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.3.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.3.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.3.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.4.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.4.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.4.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.4.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.4.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.4.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.4.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.4.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.4.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.5.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.5.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.5.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.5.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.5.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.5.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.5.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.5.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.5.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.6.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.6.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.6.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.6.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.6.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.6.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.6.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.6.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.6.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.7.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.7.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.7.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.7.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.7.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.7.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.7.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.7.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.7.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.8.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.8.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.8.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.8.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.8.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.8.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.8.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.8.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.8.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.9.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.9.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.9.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.9.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.9.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.9.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.9.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.9.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.9.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.norm.weight": "model-00002-of-00002.safetensors" + } +} diff --git a/babylm_shuffle_nondeterministic_10M_seed0/runs/checkpoint-700/special_tokens_map.json b/babylm_shuffle_nondeterministic_10M_seed0/runs/checkpoint-700/special_tokens_map.json new file mode 100644 index 0000000000000000000000000000000000000000..6949c5975ee0e961ef61cf31010dce04df0a03f8 --- /dev/null +++ b/babylm_shuffle_nondeterministic_10M_seed0/runs/checkpoint-700/special_tokens_map.json @@ -0,0 +1,23 @@ +{ + "bos_token": { + "content": "<|begin_of_text|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false + }, + "eos_token": { + "content": "<|end_of_text|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false + }, + "pad_token": { + "content": "[PAD]", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false + } +} diff --git a/babylm_shuffle_nondeterministic_10M_seed0/runs/checkpoint-700/tokenizer_config.json b/babylm_shuffle_nondeterministic_10M_seed0/runs/checkpoint-700/tokenizer_config.json new file mode 100644 index 0000000000000000000000000000000000000000..5684b14e2756b56aa397c1324e239b442162b34a --- /dev/null +++ b/babylm_shuffle_nondeterministic_10M_seed0/runs/checkpoint-700/tokenizer_config.json @@ -0,0 +1,2070 @@ +{ + "added_tokens_decoder": { + "128000": { + "content": "<|begin_of_text|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128001": { + "content": "<|end_of_text|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128002": { + "content": "<|reserved_special_token_0|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128003": { + "content": "<|reserved_special_token_1|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128004": { + "content": "<|finetune_right_pad_id|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128005": { + "content": "<|reserved_special_token_2|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128006": { + "content": "<|start_header_id|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128007": { + "content": "<|end_header_id|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128008": { + "content": "<|eom_id|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128009": { + "content": "<|eot_id|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128010": { + "content": "<|python_tag|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128011": { + "content": "<|reserved_special_token_3|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128012": { + "content": "<|reserved_special_token_4|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128013": { + "content": "<|reserved_special_token_5|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128014": { + "content": "<|reserved_special_token_6|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128015": { + "content": "<|reserved_special_token_7|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128016": { + "content": "<|reserved_special_token_8|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128017": { + "content": "<|reserved_special_token_9|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128018": { + "content": "<|reserved_special_token_10|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128019": { + "content": "<|reserved_special_token_11|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128020": { + "content": "<|reserved_special_token_12|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128021": { + "content": "<|reserved_special_token_13|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128022": { + "content": "<|reserved_special_token_14|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128023": { + "content": "<|reserved_special_token_15|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128024": { + "content": "<|reserved_special_token_16|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128025": { + "content": "<|reserved_special_token_17|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128026": { + "content": "<|reserved_special_token_18|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128027": { + "content": "<|reserved_special_token_19|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128028": { + "content": "<|reserved_special_token_20|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128029": { + "content": "<|reserved_special_token_21|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128030": { + "content": "<|reserved_special_token_22|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128031": { + "content": "<|reserved_special_token_23|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128032": { + "content": "<|reserved_special_token_24|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128033": { + "content": "<|reserved_special_token_25|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128034": { + "content": "<|reserved_special_token_26|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128035": { + "content": "<|reserved_special_token_27|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128036": { + "content": "<|reserved_special_token_28|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128037": { + "content": "<|reserved_special_token_29|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128038": { + "content": "<|reserved_special_token_30|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128039": { + "content": "<|reserved_special_token_31|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128040": { + "content": "<|reserved_special_token_32|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128041": { + "content": "<|reserved_special_token_33|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128042": { + "content": "<|reserved_special_token_34|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128043": { + "content": "<|reserved_special_token_35|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128044": { + "content": "<|reserved_special_token_36|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128045": { + "content": "<|reserved_special_token_37|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128046": { + "content": "<|reserved_special_token_38|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128047": { + "content": "<|reserved_special_token_39|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128048": { + "content": "<|reserved_special_token_40|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128049": { + "content": "<|reserved_special_token_41|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128050": { + "content": "<|reserved_special_token_42|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128051": { + "content": "<|reserved_special_token_43|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128052": { + "content": "<|reserved_special_token_44|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128053": { + "content": "<|reserved_special_token_45|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128054": { + "content": "<|reserved_special_token_46|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128055": { + "content": "<|reserved_special_token_47|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128056": { + "content": "<|reserved_special_token_48|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128057": { + "content": "<|reserved_special_token_49|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128058": { + "content": "<|reserved_special_token_50|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128059": { + "content": "<|reserved_special_token_51|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128060": { + "content": "<|reserved_special_token_52|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128061": { + "content": "<|reserved_special_token_53|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128062": { + "content": "<|reserved_special_token_54|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128063": { + "content": "<|reserved_special_token_55|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128064": { + "content": "<|reserved_special_token_56|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128065": { + "content": "<|reserved_special_token_57|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128066": { + "content": "<|reserved_special_token_58|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128067": { + "content": "<|reserved_special_token_59|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128068": { + "content": "<|reserved_special_token_60|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128069": { + "content": "<|reserved_special_token_61|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128070": { + "content": "<|reserved_special_token_62|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128071": { + "content": "<|reserved_special_token_63|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128072": { + "content": "<|reserved_special_token_64|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128073": { + "content": "<|reserved_special_token_65|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128074": { + "content": "<|reserved_special_token_66|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128075": { + "content": "<|reserved_special_token_67|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128076": { + "content": "<|reserved_special_token_68|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128077": { + "content": "<|reserved_special_token_69|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128078": { + "content": "<|reserved_special_token_70|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128079": { + "content": "<|reserved_special_token_71|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128080": { + "content": "<|reserved_special_token_72|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128081": { + "content": "<|reserved_special_token_73|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128082": { + "content": "<|reserved_special_token_74|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128083": { + "content": "<|reserved_special_token_75|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128084": { + "content": "<|reserved_special_token_76|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128085": { + "content": "<|reserved_special_token_77|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128086": { + "content": "<|reserved_special_token_78|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128087": { + "content": "<|reserved_special_token_79|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128088": { + "content": "<|reserved_special_token_80|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128089": { + "content": "<|reserved_special_token_81|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128090": { + "content": "<|reserved_special_token_82|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128091": { + "content": "<|reserved_special_token_83|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128092": { + "content": "<|reserved_special_token_84|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128093": { + "content": "<|reserved_special_token_85|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128094": { + "content": "<|reserved_special_token_86|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128095": { + "content": "<|reserved_special_token_87|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128096": { + "content": "<|reserved_special_token_88|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128097": { + "content": "<|reserved_special_token_89|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128098": { + "content": "<|reserved_special_token_90|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128099": { + "content": "<|reserved_special_token_91|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128100": { + "content": "<|reserved_special_token_92|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128101": { + "content": "<|reserved_special_token_93|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128102": { + "content": "<|reserved_special_token_94|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128103": { + "content": "<|reserved_special_token_95|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128104": { + "content": "<|reserved_special_token_96|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128105": { + "content": "<|reserved_special_token_97|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128106": { + "content": "<|reserved_special_token_98|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128107": { + "content": "<|reserved_special_token_99|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128108": { + "content": "<|reserved_special_token_100|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128109": { + "content": "<|reserved_special_token_101|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128110": { + "content": "<|reserved_special_token_102|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128111": { + "content": "<|reserved_special_token_103|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128112": { + "content": "<|reserved_special_token_104|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128113": { + "content": "<|reserved_special_token_105|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128114": { + "content": "<|reserved_special_token_106|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128115": { + "content": "<|reserved_special_token_107|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128116": { + "content": "<|reserved_special_token_108|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128117": { + "content": "<|reserved_special_token_109|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128118": { + "content": "<|reserved_special_token_110|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128119": { + "content": "<|reserved_special_token_111|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128120": { + "content": "<|reserved_special_token_112|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128121": { + "content": "<|reserved_special_token_113|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128122": { + "content": "<|reserved_special_token_114|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128123": { + "content": "<|reserved_special_token_115|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128124": { + "content": "<|reserved_special_token_116|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128125": { + "content": "<|reserved_special_token_117|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128126": { + "content": "<|reserved_special_token_118|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128127": { + "content": "<|reserved_special_token_119|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128128": { + "content": "<|reserved_special_token_120|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128129": { + "content": "<|reserved_special_token_121|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128130": { + "content": "<|reserved_special_token_122|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128131": { + "content": "<|reserved_special_token_123|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128132": { + "content": "<|reserved_special_token_124|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128133": { + "content": "<|reserved_special_token_125|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128134": { + "content": "<|reserved_special_token_126|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128135": { + "content": "<|reserved_special_token_127|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128136": { + "content": "<|reserved_special_token_128|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128137": { + "content": "<|reserved_special_token_129|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128138": { + "content": "<|reserved_special_token_130|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128139": { + "content": "<|reserved_special_token_131|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128140": { + "content": "<|reserved_special_token_132|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128141": { + "content": "<|reserved_special_token_133|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128142": { + "content": "<|reserved_special_token_134|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128143": { + "content": "<|reserved_special_token_135|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128144": { + "content": "<|reserved_special_token_136|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128145": { + "content": "<|reserved_special_token_137|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128146": { + "content": "<|reserved_special_token_138|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128147": { + "content": "<|reserved_special_token_139|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128148": { + "content": "<|reserved_special_token_140|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128149": { + "content": "<|reserved_special_token_141|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128150": { + "content": "<|reserved_special_token_142|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128151": { + "content": "<|reserved_special_token_143|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128152": { + "content": "<|reserved_special_token_144|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128153": { + "content": "<|reserved_special_token_145|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128154": { + "content": "<|reserved_special_token_146|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128155": { + "content": "<|reserved_special_token_147|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128156": { + "content": "<|reserved_special_token_148|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128157": { + "content": "<|reserved_special_token_149|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128158": { + "content": "<|reserved_special_token_150|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128159": { + "content": "<|reserved_special_token_151|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128160": { + "content": "<|reserved_special_token_152|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128161": { + "content": "<|reserved_special_token_153|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128162": { + "content": "<|reserved_special_token_154|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128163": { + "content": "<|reserved_special_token_155|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128164": { + "content": "<|reserved_special_token_156|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128165": { + "content": "<|reserved_special_token_157|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128166": { + "content": "<|reserved_special_token_158|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128167": { + "content": "<|reserved_special_token_159|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128168": { + "content": "<|reserved_special_token_160|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128169": { + "content": "<|reserved_special_token_161|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128170": { + "content": "<|reserved_special_token_162|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128171": { + "content": "<|reserved_special_token_163|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128172": { + "content": "<|reserved_special_token_164|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128173": { + "content": "<|reserved_special_token_165|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128174": { + "content": "<|reserved_special_token_166|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128175": { + "content": "<|reserved_special_token_167|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128176": { + "content": "<|reserved_special_token_168|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128177": { + "content": "<|reserved_special_token_169|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128178": { + "content": "<|reserved_special_token_170|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128179": { + "content": "<|reserved_special_token_171|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128180": { + "content": "<|reserved_special_token_172|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128181": { + "content": "<|reserved_special_token_173|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128182": { + "content": "<|reserved_special_token_174|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128183": { + "content": "<|reserved_special_token_175|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128184": { + "content": "<|reserved_special_token_176|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128185": { + "content": "<|reserved_special_token_177|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128186": { + "content": "<|reserved_special_token_178|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128187": { + "content": "<|reserved_special_token_179|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128188": { + "content": "<|reserved_special_token_180|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128189": { + "content": "<|reserved_special_token_181|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128190": { + "content": "<|reserved_special_token_182|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128191": { + "content": "<|reserved_special_token_183|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128192": { + "content": "<|reserved_special_token_184|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128193": { + "content": "<|reserved_special_token_185|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128194": { + "content": "<|reserved_special_token_186|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128195": { + "content": "<|reserved_special_token_187|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128196": { + "content": "<|reserved_special_token_188|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128197": { + "content": "<|reserved_special_token_189|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128198": { + "content": "<|reserved_special_token_190|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128199": { + "content": "<|reserved_special_token_191|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128200": { + "content": "<|reserved_special_token_192|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128201": { + "content": "<|reserved_special_token_193|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128202": { + "content": "<|reserved_special_token_194|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128203": { + "content": "<|reserved_special_token_195|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128204": { + "content": "<|reserved_special_token_196|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128205": { + "content": "<|reserved_special_token_197|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128206": { + "content": "<|reserved_special_token_198|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128207": { + "content": "<|reserved_special_token_199|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128208": { + "content": "<|reserved_special_token_200|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128209": { + "content": "<|reserved_special_token_201|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128210": { + "content": "<|reserved_special_token_202|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128211": { + "content": "<|reserved_special_token_203|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128212": { + "content": "<|reserved_special_token_204|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128213": { + "content": "<|reserved_special_token_205|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128214": { + "content": "<|reserved_special_token_206|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128215": { + "content": "<|reserved_special_token_207|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128216": { + "content": "<|reserved_special_token_208|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128217": { + "content": "<|reserved_special_token_209|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128218": { + "content": "<|reserved_special_token_210|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128219": { + "content": "<|reserved_special_token_211|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128220": { + "content": "<|reserved_special_token_212|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128221": { + "content": "<|reserved_special_token_213|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128222": { + "content": "<|reserved_special_token_214|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128223": { + "content": "<|reserved_special_token_215|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128224": { + "content": "<|reserved_special_token_216|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128225": { + "content": "<|reserved_special_token_217|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128226": { + "content": "<|reserved_special_token_218|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128227": { + "content": "<|reserved_special_token_219|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128228": { + "content": "<|reserved_special_token_220|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128229": { + "content": "<|reserved_special_token_221|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128230": { + "content": "<|reserved_special_token_222|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128231": { + "content": "<|reserved_special_token_223|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128232": { + "content": "<|reserved_special_token_224|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128233": { + "content": "<|reserved_special_token_225|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128234": { + "content": "<|reserved_special_token_226|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128235": { + "content": "<|reserved_special_token_227|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128236": { + "content": "<|reserved_special_token_228|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128237": { + "content": "<|reserved_special_token_229|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128238": { + "content": "<|reserved_special_token_230|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128239": { + "content": "<|reserved_special_token_231|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128240": { + "content": "<|reserved_special_token_232|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128241": { + "content": "<|reserved_special_token_233|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128242": { + "content": "<|reserved_special_token_234|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128243": { + "content": "<|reserved_special_token_235|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128244": { + "content": "<|reserved_special_token_236|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128245": { + "content": "<|reserved_special_token_237|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128246": { + "content": "<|reserved_special_token_238|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128247": { + "content": "<|reserved_special_token_239|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128248": { + "content": "<|reserved_special_token_240|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128249": { + "content": "<|reserved_special_token_241|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128250": { + "content": "<|reserved_special_token_242|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128251": { + "content": "<|reserved_special_token_243|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128252": { + "content": "<|reserved_special_token_244|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128253": { + "content": "<|reserved_special_token_245|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128254": { + "content": "<|reserved_special_token_246|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128255": { + "content": "<|reserved_special_token_247|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128256": { + "content": "[PAD]", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + } + }, + "bos_token": "<|begin_of_text|>", + "clean_up_tokenization_spaces": true, + "eos_token": "<|end_of_text|>", + "model_input_names": [ + "input_ids", + "attention_mask" + ], + "model_max_length": 131072, + "pad_token": "[PAD]", + "tokenizer_class": "PreTrainedTokenizerFast" +} diff --git a/babylm_shuffle_nondeterministic_10M_seed0/runs/checkpoint-700/trainer_state.json b/babylm_shuffle_nondeterministic_10M_seed0/runs/checkpoint-700/trainer_state.json new file mode 100644 index 0000000000000000000000000000000000000000..271cc5b40a0f861b4d11b6562bb974354c2308c4 --- /dev/null +++ b/babylm_shuffle_nondeterministic_10M_seed0/runs/checkpoint-700/trainer_state.json @@ -0,0 +1,5493 @@ +{ + "best_metric": null, + "best_model_checkpoint": null, + "epoch": 0.7671232876712328, + "eval_steps": 10, + "global_step": 700, + "is_hyper_param_search": false, + "is_local_process_zero": true, + "is_world_process_zero": true, + "log_history": [ + { + "epoch": 0.001095890410958904, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.0998, + "step": 1 + }, + { + "epoch": 0.002191780821917808, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.0858, + "step": 2 + }, + { + "epoch": 0.0032876712328767125, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.0672, + "step": 3 + }, + { + "epoch": 0.004383561643835616, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.0802, + "step": 4 + }, + { + "epoch": 0.005479452054794521, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.0636, + "step": 5 + }, + { + "epoch": 0.006575342465753425, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.1672, + "step": 6 + }, + { + "epoch": 0.007671232876712329, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.0961, + "step": 7 + }, + { + "epoch": 0.008767123287671232, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.1258, + "step": 8 + }, + { + "epoch": 0.009863013698630137, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.0953, + "step": 9 + }, + { + "epoch": 0.010958904109589041, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.1404, + "step": 10 + }, + { + "epoch": 0.010958904109589041, + "eval_loss": 3.12593412399292, + "eval_runtime": 42.732, + "eval_samples_per_second": 23.402, + "eval_steps_per_second": 0.983, + "step": 10 + }, + { + "epoch": 0.012054794520547946, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.1187, + "step": 11 + }, + { + "epoch": 0.01315068493150685, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.0593, + "step": 12 + }, + { + "epoch": 0.014246575342465753, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.111, + "step": 13 + }, + { + "epoch": 0.015342465753424657, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.0974, + "step": 14 + }, + { + "epoch": 0.01643835616438356, + "grad_norm": 5.991453170776367, + "learning_rate": 1.8248175182481753e-08, + "loss": 3.135, + "step": 15 + }, + { + "epoch": 0.017534246575342465, + "grad_norm": 6.274210453033447, + "learning_rate": 3.6496350364963505e-08, + "loss": 3.083, + "step": 16 + }, + { + "epoch": 0.01863013698630137, + "grad_norm": 6.644786834716797, + "learning_rate": 5.4744525547445264e-08, + "loss": 3.0856, + "step": 17 + }, + { + "epoch": 0.019726027397260273, + "grad_norm": 6.769156455993652, + "learning_rate": 7.299270072992701e-08, + "loss": 3.1082, + "step": 18 + }, + { + "epoch": 0.020821917808219178, + "grad_norm": 6.771246910095215, + "learning_rate": 9.124087591240876e-08, + "loss": 3.0753, + "step": 19 + }, + { + "epoch": 0.021917808219178082, + "grad_norm": 6.468441486358643, + "learning_rate": 1.0948905109489053e-07, + "loss": 3.1478, + "step": 20 + }, + { + "epoch": 0.021917808219178082, + "eval_loss": 3.1245107650756836, + "eval_runtime": 43.6251, + "eval_samples_per_second": 22.923, + "eval_steps_per_second": 0.963, + "step": 20 + }, + { + "epoch": 0.023013698630136987, + "grad_norm": 6.83078670501709, + "learning_rate": 1.2773722627737227e-07, + "loss": 3.1213, + "step": 21 + }, + { + "epoch": 0.02410958904109589, + "grad_norm": 6.221287250518799, + "learning_rate": 1.4598540145985402e-07, + "loss": 3.0549, + "step": 22 + }, + { + "epoch": 0.025205479452054796, + "grad_norm": 5.564266204833984, + "learning_rate": 1.642335766423358e-07, + "loss": 3.0769, + "step": 23 + }, + { + "epoch": 0.0263013698630137, + "grad_norm": 5.735209941864014, + "learning_rate": 1.8248175182481753e-07, + "loss": 3.1124, + "step": 24 + }, + { + "epoch": 0.0273972602739726, + "grad_norm": 5.577925682067871, + "learning_rate": 2.0072992700729928e-07, + "loss": 3.1191, + "step": 25 + }, + { + "epoch": 0.028493150684931506, + "grad_norm": 4.814139366149902, + "learning_rate": 2.1897810218978106e-07, + "loss": 3.0947, + "step": 26 + }, + { + "epoch": 0.02958904109589041, + "grad_norm": 4.69775390625, + "learning_rate": 2.3722627737226278e-07, + "loss": 3.0997, + "step": 27 + }, + { + "epoch": 0.030684931506849315, + "grad_norm": 4.811586380004883, + "learning_rate": 2.5547445255474454e-07, + "loss": 3.0749, + "step": 28 + }, + { + "epoch": 0.03178082191780822, + "grad_norm": 4.623377323150635, + "learning_rate": 2.737226277372263e-07, + "loss": 3.0619, + "step": 29 + }, + { + "epoch": 0.03287671232876712, + "grad_norm": 4.721253395080566, + "learning_rate": 2.9197080291970804e-07, + "loss": 3.0371, + "step": 30 + }, + { + "epoch": 0.03287671232876712, + "eval_loss": 3.082857847213745, + "eval_runtime": 44.127, + "eval_samples_per_second": 22.662, + "eval_steps_per_second": 0.952, + "step": 30 + }, + { + "epoch": 0.03397260273972603, + "grad_norm": 3.968630790710449, + "learning_rate": 3.102189781021898e-07, + "loss": 3.0535, + "step": 31 + }, + { + "epoch": 0.03506849315068493, + "grad_norm": 4.4373779296875, + "learning_rate": 3.284671532846716e-07, + "loss": 3.0476, + "step": 32 + }, + { + "epoch": 0.03616438356164384, + "grad_norm": 4.479320049285889, + "learning_rate": 3.467153284671533e-07, + "loss": 3.0215, + "step": 33 + }, + { + "epoch": 0.03726027397260274, + "grad_norm": 4.122231960296631, + "learning_rate": 3.6496350364963505e-07, + "loss": 3.0106, + "step": 34 + }, + { + "epoch": 0.038356164383561646, + "grad_norm": 4.153995513916016, + "learning_rate": 3.8321167883211686e-07, + "loss": 3.0318, + "step": 35 + }, + { + "epoch": 0.03945205479452055, + "grad_norm": 3.6949076652526855, + "learning_rate": 4.0145985401459856e-07, + "loss": 3.046, + "step": 36 + }, + { + "epoch": 0.040547945205479455, + "grad_norm": 3.4610488414764404, + "learning_rate": 4.197080291970803e-07, + "loss": 3.002, + "step": 37 + }, + { + "epoch": 0.041643835616438356, + "grad_norm": 3.9523138999938965, + "learning_rate": 4.379562043795621e-07, + "loss": 2.9665, + "step": 38 + }, + { + "epoch": 0.042739726027397264, + "grad_norm": 3.68516206741333, + "learning_rate": 4.562043795620438e-07, + "loss": 3.0411, + "step": 39 + }, + { + "epoch": 0.043835616438356165, + "grad_norm": 3.692396402359009, + "learning_rate": 4.7445255474452557e-07, + "loss": 2.9666, + "step": 40 + }, + { + "epoch": 0.043835616438356165, + "eval_loss": 3.015512704849243, + "eval_runtime": 48.2438, + "eval_samples_per_second": 20.728, + "eval_steps_per_second": 0.871, + "step": 40 + }, + { + "epoch": 0.044931506849315066, + "grad_norm": 3.3765921592712402, + "learning_rate": 4.927007299270073e-07, + "loss": 2.9831, + "step": 41 + }, + { + "epoch": 0.046027397260273974, + "grad_norm": 2.9631125926971436, + "learning_rate": 5.109489051094891e-07, + "loss": 3.0126, + "step": 42 + }, + { + "epoch": 0.047123287671232875, + "grad_norm": 2.969730854034424, + "learning_rate": 5.291970802919708e-07, + "loss": 2.9803, + "step": 43 + }, + { + "epoch": 0.04821917808219178, + "grad_norm": 3.185237407684326, + "learning_rate": 5.474452554744526e-07, + "loss": 2.9888, + "step": 44 + }, + { + "epoch": 0.049315068493150684, + "grad_norm": 3.055579423904419, + "learning_rate": 5.656934306569343e-07, + "loss": 2.9715, + "step": 45 + }, + { + "epoch": 0.05041095890410959, + "grad_norm": 3.083284616470337, + "learning_rate": 5.839416058394161e-07, + "loss": 2.9293, + "step": 46 + }, + { + "epoch": 0.05150684931506849, + "grad_norm": 2.984781503677368, + "learning_rate": 6.021897810218979e-07, + "loss": 2.9228, + "step": 47 + }, + { + "epoch": 0.0526027397260274, + "grad_norm": 2.756173849105835, + "learning_rate": 6.204379562043796e-07, + "loss": 2.9406, + "step": 48 + }, + { + "epoch": 0.0536986301369863, + "grad_norm": 2.911132574081421, + "learning_rate": 6.386861313868614e-07, + "loss": 2.8829, + "step": 49 + }, + { + "epoch": 0.0547945205479452, + "grad_norm": 2.592890977859497, + "learning_rate": 6.569343065693432e-07, + "loss": 2.9839, + "step": 50 + }, + { + "epoch": 0.0547945205479452, + "eval_loss": 2.9409468173980713, + "eval_runtime": 47.1045, + "eval_samples_per_second": 21.229, + "eval_steps_per_second": 0.892, + "step": 50 + }, + { + "epoch": 0.05589041095890411, + "grad_norm": 3.0112955570220947, + "learning_rate": 6.751824817518248e-07, + "loss": 2.8847, + "step": 51 + }, + { + "epoch": 0.05698630136986301, + "grad_norm": 2.3701798915863037, + "learning_rate": 6.934306569343066e-07, + "loss": 2.9109, + "step": 52 + }, + { + "epoch": 0.05808219178082192, + "grad_norm": 2.268850088119507, + "learning_rate": 7.116788321167883e-07, + "loss": 2.8483, + "step": 53 + }, + { + "epoch": 0.05917808219178082, + "grad_norm": 2.419015645980835, + "learning_rate": 7.299270072992701e-07, + "loss": 2.8491, + "step": 54 + }, + { + "epoch": 0.06027397260273973, + "grad_norm": 2.4582653045654297, + "learning_rate": 7.48175182481752e-07, + "loss": 2.8622, + "step": 55 + }, + { + "epoch": 0.06136986301369863, + "grad_norm": 2.3471553325653076, + "learning_rate": 7.664233576642337e-07, + "loss": 2.8738, + "step": 56 + }, + { + "epoch": 0.06246575342465754, + "grad_norm": 2.479177236557007, + "learning_rate": 7.846715328467154e-07, + "loss": 2.8843, + "step": 57 + }, + { + "epoch": 0.06356164383561644, + "grad_norm": 2.1676292419433594, + "learning_rate": 8.029197080291971e-07, + "loss": 2.8674, + "step": 58 + }, + { + "epoch": 0.06465753424657535, + "grad_norm": 1.9862910509109497, + "learning_rate": 8.211678832116789e-07, + "loss": 2.8197, + "step": 59 + }, + { + "epoch": 0.06575342465753424, + "grad_norm": 2.211313486099243, + "learning_rate": 8.394160583941606e-07, + "loss": 2.7961, + "step": 60 + }, + { + "epoch": 0.06575342465753424, + "eval_loss": 2.880824327468872, + "eval_runtime": 51.3235, + "eval_samples_per_second": 19.484, + "eval_steps_per_second": 0.818, + "step": 60 + }, + { + "epoch": 0.06684931506849315, + "grad_norm": 2.3497281074523926, + "learning_rate": 8.576642335766425e-07, + "loss": 2.8233, + "step": 61 + }, + { + "epoch": 0.06794520547945206, + "grad_norm": 2.158515214920044, + "learning_rate": 8.759124087591242e-07, + "loss": 2.8136, + "step": 62 + }, + { + "epoch": 0.06904109589041096, + "grad_norm": 2.0219955444335938, + "learning_rate": 8.94160583941606e-07, + "loss": 2.8231, + "step": 63 + }, + { + "epoch": 0.07013698630136986, + "grad_norm": 1.8989133834838867, + "learning_rate": 9.124087591240876e-07, + "loss": 2.8268, + "step": 64 + }, + { + "epoch": 0.07123287671232877, + "grad_norm": 2.243851661682129, + "learning_rate": 9.306569343065694e-07, + "loss": 2.7886, + "step": 65 + }, + { + "epoch": 0.07232876712328767, + "grad_norm": 2.122464418411255, + "learning_rate": 9.489051094890511e-07, + "loss": 2.8166, + "step": 66 + }, + { + "epoch": 0.07342465753424658, + "grad_norm": 1.899857997894287, + "learning_rate": 9.67153284671533e-07, + "loss": 2.7997, + "step": 67 + }, + { + "epoch": 0.07452054794520548, + "grad_norm": 2.4424173831939697, + "learning_rate": 9.854014598540146e-07, + "loss": 2.8342, + "step": 68 + }, + { + "epoch": 0.07561643835616438, + "grad_norm": 2.5677411556243896, + "learning_rate": 1.0036496350364965e-06, + "loss": 2.7786, + "step": 69 + }, + { + "epoch": 0.07671232876712329, + "grad_norm": 1.9989523887634277, + "learning_rate": 1.0218978102189781e-06, + "loss": 2.7411, + "step": 70 + }, + { + "epoch": 0.07671232876712329, + "eval_loss": 2.8185322284698486, + "eval_runtime": 50.4998, + "eval_samples_per_second": 19.802, + "eval_steps_per_second": 0.832, + "step": 70 + }, + { + "epoch": 0.07780821917808219, + "grad_norm": 1.9376521110534668, + "learning_rate": 1.04014598540146e-06, + "loss": 2.7988, + "step": 71 + }, + { + "epoch": 0.0789041095890411, + "grad_norm": 2.5629725456237793, + "learning_rate": 1.0583941605839416e-06, + "loss": 2.7079, + "step": 72 + }, + { + "epoch": 0.08, + "grad_norm": 2.1634678840637207, + "learning_rate": 1.0766423357664233e-06, + "loss": 2.7049, + "step": 73 + }, + { + "epoch": 0.08109589041095891, + "grad_norm": 2.2143008708953857, + "learning_rate": 1.0948905109489052e-06, + "loss": 2.72, + "step": 74 + }, + { + "epoch": 0.0821917808219178, + "grad_norm": 2.2701101303100586, + "learning_rate": 1.113138686131387e-06, + "loss": 2.7418, + "step": 75 + }, + { + "epoch": 0.08328767123287671, + "grad_norm": 1.716569423675537, + "learning_rate": 1.1313868613138687e-06, + "loss": 2.7558, + "step": 76 + }, + { + "epoch": 0.08438356164383562, + "grad_norm": 2.79757022857666, + "learning_rate": 1.1496350364963505e-06, + "loss": 2.6865, + "step": 77 + }, + { + "epoch": 0.08547945205479453, + "grad_norm": 2.2189838886260986, + "learning_rate": 1.1678832116788322e-06, + "loss": 2.6684, + "step": 78 + }, + { + "epoch": 0.08657534246575342, + "grad_norm": 2.0695579051971436, + "learning_rate": 1.1861313868613138e-06, + "loss": 2.6779, + "step": 79 + }, + { + "epoch": 0.08767123287671233, + "grad_norm": 2.372326374053955, + "learning_rate": 1.2043795620437959e-06, + "loss": 2.6711, + "step": 80 + }, + { + "epoch": 0.08767123287671233, + "eval_loss": 2.7537453174591064, + "eval_runtime": 47.7274, + "eval_samples_per_second": 20.952, + "eval_steps_per_second": 0.88, + "step": 80 + }, + { + "epoch": 0.08876712328767124, + "grad_norm": 2.3398947715759277, + "learning_rate": 1.2226277372262775e-06, + "loss": 2.6776, + "step": 81 + }, + { + "epoch": 0.08986301369863013, + "grad_norm": 2.6619036197662354, + "learning_rate": 1.2408759124087592e-06, + "loss": 2.7077, + "step": 82 + }, + { + "epoch": 0.09095890410958904, + "grad_norm": 1.8245115280151367, + "learning_rate": 1.259124087591241e-06, + "loss": 2.662, + "step": 83 + }, + { + "epoch": 0.09205479452054795, + "grad_norm": 2.2627148628234863, + "learning_rate": 1.2773722627737229e-06, + "loss": 2.677, + "step": 84 + }, + { + "epoch": 0.09315068493150686, + "grad_norm": 2.4983835220336914, + "learning_rate": 1.2956204379562045e-06, + "loss": 2.6601, + "step": 85 + }, + { + "epoch": 0.09424657534246575, + "grad_norm": 2.147010087966919, + "learning_rate": 1.3138686131386864e-06, + "loss": 2.6213, + "step": 86 + }, + { + "epoch": 0.09534246575342466, + "grad_norm": 2.791682481765747, + "learning_rate": 1.332116788321168e-06, + "loss": 2.6281, + "step": 87 + }, + { + "epoch": 0.09643835616438357, + "grad_norm": 2.023970365524292, + "learning_rate": 1.3503649635036497e-06, + "loss": 2.5766, + "step": 88 + }, + { + "epoch": 0.09753424657534246, + "grad_norm": 2.8143956661224365, + "learning_rate": 1.3686131386861315e-06, + "loss": 2.6086, + "step": 89 + }, + { + "epoch": 0.09863013698630137, + "grad_norm": 2.226106882095337, + "learning_rate": 1.3868613138686132e-06, + "loss": 2.5477, + "step": 90 + }, + { + "epoch": 0.09863013698630137, + "eval_loss": 2.6880786418914795, + "eval_runtime": 47.8417, + "eval_samples_per_second": 20.902, + "eval_steps_per_second": 0.878, + "step": 90 + }, + { + "epoch": 0.09972602739726028, + "grad_norm": 2.5248148441314697, + "learning_rate": 1.405109489051095e-06, + "loss": 2.6232, + "step": 91 + }, + { + "epoch": 0.10082191780821918, + "grad_norm": 2.421330690383911, + "learning_rate": 1.4233576642335767e-06, + "loss": 2.599, + "step": 92 + }, + { + "epoch": 0.10191780821917808, + "grad_norm": 2.5562429428100586, + "learning_rate": 1.4416058394160583e-06, + "loss": 2.6197, + "step": 93 + }, + { + "epoch": 0.10301369863013699, + "grad_norm": 2.856201171875, + "learning_rate": 1.4598540145985402e-06, + "loss": 2.5825, + "step": 94 + }, + { + "epoch": 0.10410958904109589, + "grad_norm": 2.6576807498931885, + "learning_rate": 1.478102189781022e-06, + "loss": 2.5496, + "step": 95 + }, + { + "epoch": 0.1052054794520548, + "grad_norm": 3.1408190727233887, + "learning_rate": 1.496350364963504e-06, + "loss": 2.6142, + "step": 96 + }, + { + "epoch": 0.1063013698630137, + "grad_norm": 2.3270952701568604, + "learning_rate": 1.5145985401459856e-06, + "loss": 2.584, + "step": 97 + }, + { + "epoch": 0.1073972602739726, + "grad_norm": 2.3976614475250244, + "learning_rate": 1.5328467153284674e-06, + "loss": 2.5941, + "step": 98 + }, + { + "epoch": 0.10849315068493151, + "grad_norm": 2.783485174179077, + "learning_rate": 1.551094890510949e-06, + "loss": 2.5626, + "step": 99 + }, + { + "epoch": 0.1095890410958904, + "grad_norm": 2.2522788047790527, + "learning_rate": 1.5693430656934307e-06, + "loss": 2.517, + "step": 100 + }, + { + "epoch": 0.1095890410958904, + "eval_loss": 2.637315034866333, + "eval_runtime": 48.2107, + "eval_samples_per_second": 20.742, + "eval_steps_per_second": 0.871, + "step": 100 + }, + { + "epoch": 0.11068493150684931, + "grad_norm": 2.917799949645996, + "learning_rate": 1.5875912408759126e-06, + "loss": 2.5159, + "step": 101 + }, + { + "epoch": 0.11178082191780822, + "grad_norm": 2.955815076828003, + "learning_rate": 1.6058394160583942e-06, + "loss": 2.563, + "step": 102 + }, + { + "epoch": 0.11287671232876713, + "grad_norm": 2.8096001148223877, + "learning_rate": 1.624087591240876e-06, + "loss": 2.5995, + "step": 103 + }, + { + "epoch": 0.11397260273972602, + "grad_norm": 2.4967427253723145, + "learning_rate": 1.6423357664233577e-06, + "loss": 2.5336, + "step": 104 + }, + { + "epoch": 0.11506849315068493, + "grad_norm": 2.740717887878418, + "learning_rate": 1.6605839416058394e-06, + "loss": 2.4778, + "step": 105 + }, + { + "epoch": 0.11616438356164384, + "grad_norm": 2.5064632892608643, + "learning_rate": 1.6788321167883212e-06, + "loss": 2.5214, + "step": 106 + }, + { + "epoch": 0.11726027397260275, + "grad_norm": 2.2923007011413574, + "learning_rate": 1.6970802919708029e-06, + "loss": 2.5315, + "step": 107 + }, + { + "epoch": 0.11835616438356164, + "grad_norm": 3.339569568634033, + "learning_rate": 1.715328467153285e-06, + "loss": 2.4661, + "step": 108 + }, + { + "epoch": 0.11945205479452055, + "grad_norm": 2.1650891304016113, + "learning_rate": 1.7335766423357666e-06, + "loss": 2.5072, + "step": 109 + }, + { + "epoch": 0.12054794520547946, + "grad_norm": 3.1216914653778076, + "learning_rate": 1.7518248175182485e-06, + "loss": 2.5164, + "step": 110 + }, + { + "epoch": 0.12054794520547946, + "eval_loss": 2.594250202178955, + "eval_runtime": 41.8939, + "eval_samples_per_second": 23.87, + "eval_steps_per_second": 1.003, + "step": 110 + }, + { + "epoch": 0.12164383561643835, + "grad_norm": 2.191063404083252, + "learning_rate": 1.77007299270073e-06, + "loss": 2.5422, + "step": 111 + }, + { + "epoch": 0.12273972602739726, + "grad_norm": 2.922083854675293, + "learning_rate": 1.788321167883212e-06, + "loss": 2.5258, + "step": 112 + }, + { + "epoch": 0.12383561643835617, + "grad_norm": 2.518002986907959, + "learning_rate": 1.8065693430656936e-06, + "loss": 2.5588, + "step": 113 + }, + { + "epoch": 0.12493150684931507, + "grad_norm": 2.515194892883301, + "learning_rate": 1.8248175182481753e-06, + "loss": 2.5079, + "step": 114 + }, + { + "epoch": 0.12602739726027398, + "grad_norm": 2.5530874729156494, + "learning_rate": 1.8430656934306571e-06, + "loss": 2.4615, + "step": 115 + }, + { + "epoch": 0.12712328767123288, + "grad_norm": 2.3550198078155518, + "learning_rate": 1.8613138686131388e-06, + "loss": 2.4825, + "step": 116 + }, + { + "epoch": 0.12821917808219177, + "grad_norm": 2.295145034790039, + "learning_rate": 1.8795620437956206e-06, + "loss": 2.4749, + "step": 117 + }, + { + "epoch": 0.1293150684931507, + "grad_norm": 2.0158982276916504, + "learning_rate": 1.8978102189781023e-06, + "loss": 2.4756, + "step": 118 + }, + { + "epoch": 0.1304109589041096, + "grad_norm": 2.213057518005371, + "learning_rate": 1.916058394160584e-06, + "loss": 2.4744, + "step": 119 + }, + { + "epoch": 0.13150684931506848, + "grad_norm": 2.186222553253174, + "learning_rate": 1.934306569343066e-06, + "loss": 2.4957, + "step": 120 + }, + { + "epoch": 0.13150684931506848, + "eval_loss": 2.563931465148926, + "eval_runtime": 44.8714, + "eval_samples_per_second": 22.286, + "eval_steps_per_second": 0.936, + "step": 120 + }, + { + "epoch": 0.1326027397260274, + "grad_norm": 2.3143310546875, + "learning_rate": 1.952554744525548e-06, + "loss": 2.4, + "step": 121 + }, + { + "epoch": 0.1336986301369863, + "grad_norm": 2.0223965644836426, + "learning_rate": 1.9708029197080293e-06, + "loss": 2.4453, + "step": 122 + }, + { + "epoch": 0.13479452054794522, + "grad_norm": 2.2025411128997803, + "learning_rate": 1.989051094890511e-06, + "loss": 2.4463, + "step": 123 + }, + { + "epoch": 0.1358904109589041, + "grad_norm": 2.707930564880371, + "learning_rate": 2.007299270072993e-06, + "loss": 2.4589, + "step": 124 + }, + { + "epoch": 0.136986301369863, + "grad_norm": 2.146662473678589, + "learning_rate": 2.025547445255475e-06, + "loss": 2.5023, + "step": 125 + }, + { + "epoch": 0.13808219178082193, + "grad_norm": 2.520660161972046, + "learning_rate": 2.0437956204379563e-06, + "loss": 2.4464, + "step": 126 + }, + { + "epoch": 0.13917808219178082, + "grad_norm": 2.50632381439209, + "learning_rate": 2.062043795620438e-06, + "loss": 2.4556, + "step": 127 + }, + { + "epoch": 0.14027397260273972, + "grad_norm": 2.2919397354125977, + "learning_rate": 2.08029197080292e-06, + "loss": 2.4479, + "step": 128 + }, + { + "epoch": 0.14136986301369864, + "grad_norm": 2.521475315093994, + "learning_rate": 2.0985401459854014e-06, + "loss": 2.4982, + "step": 129 + }, + { + "epoch": 0.14246575342465753, + "grad_norm": 2.3812668323516846, + "learning_rate": 2.1167883211678833e-06, + "loss": 2.4163, + "step": 130 + }, + { + "epoch": 0.14246575342465753, + "eval_loss": 2.5390918254852295, + "eval_runtime": 43.8746, + "eval_samples_per_second": 22.792, + "eval_steps_per_second": 0.957, + "step": 130 + }, + { + "epoch": 0.14356164383561643, + "grad_norm": 2.302286148071289, + "learning_rate": 2.135036496350365e-06, + "loss": 2.4588, + "step": 131 + }, + { + "epoch": 0.14465753424657535, + "grad_norm": 2.4222044944763184, + "learning_rate": 2.1532846715328466e-06, + "loss": 2.492, + "step": 132 + }, + { + "epoch": 0.14575342465753424, + "grad_norm": 2.605835199356079, + "learning_rate": 2.171532846715329e-06, + "loss": 2.4055, + "step": 133 + }, + { + "epoch": 0.14684931506849316, + "grad_norm": 3.6261861324310303, + "learning_rate": 2.1897810218978103e-06, + "loss": 2.4221, + "step": 134 + }, + { + "epoch": 0.14794520547945206, + "grad_norm": 2.7647221088409424, + "learning_rate": 2.208029197080292e-06, + "loss": 2.4549, + "step": 135 + }, + { + "epoch": 0.14904109589041095, + "grad_norm": 3.2977371215820312, + "learning_rate": 2.226277372262774e-06, + "loss": 2.4331, + "step": 136 + }, + { + "epoch": 0.15013698630136987, + "grad_norm": 3.147094488143921, + "learning_rate": 2.244525547445256e-06, + "loss": 2.4268, + "step": 137 + }, + { + "epoch": 0.15123287671232877, + "grad_norm": 2.6363413333892822, + "learning_rate": 2.2627737226277373e-06, + "loss": 2.4524, + "step": 138 + }, + { + "epoch": 0.15232876712328766, + "grad_norm": 2.9288904666900635, + "learning_rate": 2.281021897810219e-06, + "loss": 2.4534, + "step": 139 + }, + { + "epoch": 0.15342465753424658, + "grad_norm": 2.5604870319366455, + "learning_rate": 2.299270072992701e-06, + "loss": 2.4138, + "step": 140 + }, + { + "epoch": 0.15342465753424658, + "eval_loss": 2.524790048599243, + "eval_runtime": 43.8851, + "eval_samples_per_second": 22.787, + "eval_steps_per_second": 0.957, + "step": 140 + }, + { + "epoch": 0.15452054794520548, + "grad_norm": 3.2746663093566895, + "learning_rate": 2.3175182481751825e-06, + "loss": 2.4094, + "step": 141 + }, + { + "epoch": 0.15561643835616437, + "grad_norm": 2.691483497619629, + "learning_rate": 2.3357664233576643e-06, + "loss": 2.4265, + "step": 142 + }, + { + "epoch": 0.1567123287671233, + "grad_norm": 2.5850117206573486, + "learning_rate": 2.354014598540146e-06, + "loss": 2.3922, + "step": 143 + }, + { + "epoch": 0.1578082191780822, + "grad_norm": 2.68204927444458, + "learning_rate": 2.3722627737226276e-06, + "loss": 2.4262, + "step": 144 + }, + { + "epoch": 0.1589041095890411, + "grad_norm": 2.952584981918335, + "learning_rate": 2.39051094890511e-06, + "loss": 2.4267, + "step": 145 + }, + { + "epoch": 0.16, + "grad_norm": 3.2354788780212402, + "learning_rate": 2.4087591240875918e-06, + "loss": 2.3836, + "step": 146 + }, + { + "epoch": 0.1610958904109589, + "grad_norm": 2.634521007537842, + "learning_rate": 2.427007299270073e-06, + "loss": 2.3953, + "step": 147 + }, + { + "epoch": 0.16219178082191782, + "grad_norm": 3.1115682125091553, + "learning_rate": 2.445255474452555e-06, + "loss": 2.409, + "step": 148 + }, + { + "epoch": 0.1632876712328767, + "grad_norm": 2.6824135780334473, + "learning_rate": 2.463503649635037e-06, + "loss": 2.4459, + "step": 149 + }, + { + "epoch": 0.1643835616438356, + "grad_norm": 2.617614507675171, + "learning_rate": 2.4817518248175183e-06, + "loss": 2.366, + "step": 150 + }, + { + "epoch": 0.1643835616438356, + "eval_loss": 2.5088160037994385, + "eval_runtime": 44.0888, + "eval_samples_per_second": 22.682, + "eval_steps_per_second": 0.953, + "step": 150 + }, + { + "epoch": 0.16547945205479453, + "grad_norm": 3.976992130279541, + "learning_rate": 2.5e-06, + "loss": 2.3517, + "step": 151 + }, + { + "epoch": 0.16657534246575342, + "grad_norm": 3.1433017253875732, + "learning_rate": 2.518248175182482e-06, + "loss": 2.4489, + "step": 152 + }, + { + "epoch": 0.16767123287671232, + "grad_norm": 4.100688934326172, + "learning_rate": 2.5364963503649635e-06, + "loss": 2.3683, + "step": 153 + }, + { + "epoch": 0.16876712328767124, + "grad_norm": 4.316102504730225, + "learning_rate": 2.5547445255474458e-06, + "loss": 2.3663, + "step": 154 + }, + { + "epoch": 0.16986301369863013, + "grad_norm": 2.3103132247924805, + "learning_rate": 2.5729927007299272e-06, + "loss": 2.4009, + "step": 155 + }, + { + "epoch": 0.17095890410958905, + "grad_norm": 3.576810836791992, + "learning_rate": 2.591240875912409e-06, + "loss": 2.3983, + "step": 156 + }, + { + "epoch": 0.17205479452054795, + "grad_norm": 2.337080955505371, + "learning_rate": 2.6094890510948905e-06, + "loss": 2.3849, + "step": 157 + }, + { + "epoch": 0.17315068493150684, + "grad_norm": 3.2523117065429688, + "learning_rate": 2.627737226277373e-06, + "loss": 2.4171, + "step": 158 + }, + { + "epoch": 0.17424657534246576, + "grad_norm": 2.529175281524658, + "learning_rate": 2.6459854014598542e-06, + "loss": 2.4303, + "step": 159 + }, + { + "epoch": 0.17534246575342466, + "grad_norm": 3.052212953567505, + "learning_rate": 2.664233576642336e-06, + "loss": 2.3895, + "step": 160 + }, + { + "epoch": 0.17534246575342466, + "eval_loss": 2.492485761642456, + "eval_runtime": 45.6738, + "eval_samples_per_second": 21.894, + "eval_steps_per_second": 0.92, + "step": 160 + }, + { + "epoch": 0.17643835616438355, + "grad_norm": 2.1544973850250244, + "learning_rate": 2.6824817518248175e-06, + "loss": 2.3649, + "step": 161 + }, + { + "epoch": 0.17753424657534247, + "grad_norm": 2.7351951599121094, + "learning_rate": 2.7007299270072994e-06, + "loss": 2.3908, + "step": 162 + }, + { + "epoch": 0.17863013698630137, + "grad_norm": 2.8431951999664307, + "learning_rate": 2.718978102189781e-06, + "loss": 2.374, + "step": 163 + }, + { + "epoch": 0.17972602739726026, + "grad_norm": 3.083190441131592, + "learning_rate": 2.737226277372263e-06, + "loss": 2.4193, + "step": 164 + }, + { + "epoch": 0.18082191780821918, + "grad_norm": 2.6523995399475098, + "learning_rate": 2.755474452554745e-06, + "loss": 2.3969, + "step": 165 + }, + { + "epoch": 0.18191780821917808, + "grad_norm": 2.626047372817993, + "learning_rate": 2.7737226277372264e-06, + "loss": 2.3742, + "step": 166 + }, + { + "epoch": 0.18301369863013697, + "grad_norm": 3.300917863845825, + "learning_rate": 2.7919708029197082e-06, + "loss": 2.4546, + "step": 167 + }, + { + "epoch": 0.1841095890410959, + "grad_norm": 2.797304391860962, + "learning_rate": 2.81021897810219e-06, + "loss": 2.4161, + "step": 168 + }, + { + "epoch": 0.1852054794520548, + "grad_norm": 2.777117967605591, + "learning_rate": 2.828467153284672e-06, + "loss": 2.3572, + "step": 169 + }, + { + "epoch": 0.1863013698630137, + "grad_norm": 2.887470006942749, + "learning_rate": 2.8467153284671534e-06, + "loss": 2.3887, + "step": 170 + }, + { + "epoch": 0.1863013698630137, + "eval_loss": 2.4854257106781006, + "eval_runtime": 44.5644, + "eval_samples_per_second": 22.439, + "eval_steps_per_second": 0.942, + "step": 170 + }, + { + "epoch": 0.1873972602739726, + "grad_norm": 3.0706076622009277, + "learning_rate": 2.8649635036496353e-06, + "loss": 2.3162, + "step": 171 + }, + { + "epoch": 0.1884931506849315, + "grad_norm": 3.6129744052886963, + "learning_rate": 2.8832116788321167e-06, + "loss": 2.3685, + "step": 172 + }, + { + "epoch": 0.18958904109589042, + "grad_norm": 2.4936370849609375, + "learning_rate": 2.901459854014599e-06, + "loss": 2.3297, + "step": 173 + }, + { + "epoch": 0.19068493150684931, + "grad_norm": 2.1041860580444336, + "learning_rate": 2.9197080291970804e-06, + "loss": 2.4085, + "step": 174 + }, + { + "epoch": 0.1917808219178082, + "grad_norm": 2.2066831588745117, + "learning_rate": 2.9379562043795623e-06, + "loss": 2.3905, + "step": 175 + }, + { + "epoch": 0.19287671232876713, + "grad_norm": 2.444652795791626, + "learning_rate": 2.956204379562044e-06, + "loss": 2.378, + "step": 176 + }, + { + "epoch": 0.19397260273972602, + "grad_norm": 2.501939058303833, + "learning_rate": 2.9744525547445256e-06, + "loss": 2.4136, + "step": 177 + }, + { + "epoch": 0.19506849315068492, + "grad_norm": 2.2167556285858154, + "learning_rate": 2.992700729927008e-06, + "loss": 2.385, + "step": 178 + }, + { + "epoch": 0.19616438356164384, + "grad_norm": 2.5298545360565186, + "learning_rate": 3.0109489051094893e-06, + "loss": 2.3508, + "step": 179 + }, + { + "epoch": 0.19726027397260273, + "grad_norm": 2.3774642944335938, + "learning_rate": 3.029197080291971e-06, + "loss": 2.3421, + "step": 180 + }, + { + "epoch": 0.19726027397260273, + "eval_loss": 2.476067304611206, + "eval_runtime": 44.3761, + "eval_samples_per_second": 22.535, + "eval_steps_per_second": 0.946, + "step": 180 + }, + { + "epoch": 0.19835616438356166, + "grad_norm": 2.9108712673187256, + "learning_rate": 3.0474452554744526e-06, + "loss": 2.3983, + "step": 181 + }, + { + "epoch": 0.19945205479452055, + "grad_norm": 2.5232033729553223, + "learning_rate": 3.065693430656935e-06, + "loss": 2.3946, + "step": 182 + }, + { + "epoch": 0.20054794520547944, + "grad_norm": 2.158168077468872, + "learning_rate": 3.0839416058394163e-06, + "loss": 2.3805, + "step": 183 + }, + { + "epoch": 0.20164383561643837, + "grad_norm": 2.3405354022979736, + "learning_rate": 3.102189781021898e-06, + "loss": 2.3974, + "step": 184 + }, + { + "epoch": 0.20273972602739726, + "grad_norm": 2.5308327674865723, + "learning_rate": 3.1204379562043796e-06, + "loss": 2.384, + "step": 185 + }, + { + "epoch": 0.20383561643835615, + "grad_norm": 2.6328237056732178, + "learning_rate": 3.1386861313868614e-06, + "loss": 2.3345, + "step": 186 + }, + { + "epoch": 0.20493150684931508, + "grad_norm": 3.0725512504577637, + "learning_rate": 3.156934306569343e-06, + "loss": 2.3191, + "step": 187 + }, + { + "epoch": 0.20602739726027397, + "grad_norm": 2.196218729019165, + "learning_rate": 3.175182481751825e-06, + "loss": 2.4073, + "step": 188 + }, + { + "epoch": 0.20712328767123286, + "grad_norm": 2.7722432613372803, + "learning_rate": 3.193430656934307e-06, + "loss": 2.3775, + "step": 189 + }, + { + "epoch": 0.20821917808219179, + "grad_norm": 2.4848241806030273, + "learning_rate": 3.2116788321167884e-06, + "loss": 2.3538, + "step": 190 + }, + { + "epoch": 0.20821917808219179, + "eval_loss": 2.462618112564087, + "eval_runtime": 43.9461, + "eval_samples_per_second": 22.755, + "eval_steps_per_second": 0.956, + "step": 190 + }, + { + "epoch": 0.20931506849315068, + "grad_norm": 2.4682741165161133, + "learning_rate": 3.2299270072992707e-06, + "loss": 2.3688, + "step": 191 + }, + { + "epoch": 0.2104109589041096, + "grad_norm": 2.518923282623291, + "learning_rate": 3.248175182481752e-06, + "loss": 2.3556, + "step": 192 + }, + { + "epoch": 0.2115068493150685, + "grad_norm": 2.4023561477661133, + "learning_rate": 3.266423357664234e-06, + "loss": 2.292, + "step": 193 + }, + { + "epoch": 0.2126027397260274, + "grad_norm": 2.5196943283081055, + "learning_rate": 3.2846715328467155e-06, + "loss": 2.3703, + "step": 194 + }, + { + "epoch": 0.2136986301369863, + "grad_norm": 2.762709617614746, + "learning_rate": 3.3029197080291973e-06, + "loss": 2.3578, + "step": 195 + }, + { + "epoch": 0.2147945205479452, + "grad_norm": 2.1287691593170166, + "learning_rate": 3.3211678832116788e-06, + "loss": 2.3687, + "step": 196 + }, + { + "epoch": 0.2158904109589041, + "grad_norm": 2.477341651916504, + "learning_rate": 3.339416058394161e-06, + "loss": 2.3545, + "step": 197 + }, + { + "epoch": 0.21698630136986302, + "grad_norm": 2.23933744430542, + "learning_rate": 3.3576642335766425e-06, + "loss": 2.3511, + "step": 198 + }, + { + "epoch": 0.21808219178082192, + "grad_norm": 2.2440133094787598, + "learning_rate": 3.3759124087591243e-06, + "loss": 2.3539, + "step": 199 + }, + { + "epoch": 0.2191780821917808, + "grad_norm": 2.5603177547454834, + "learning_rate": 3.3941605839416058e-06, + "loss": 2.36, + "step": 200 + }, + { + "epoch": 0.2191780821917808, + "eval_loss": 2.4507246017456055, + "eval_runtime": 43.7786, + "eval_samples_per_second": 22.842, + "eval_steps_per_second": 0.959, + "step": 200 + }, + { + "epoch": 0.22027397260273973, + "grad_norm": 2.4570510387420654, + "learning_rate": 3.412408759124088e-06, + "loss": 2.3702, + "step": 201 + }, + { + "epoch": 0.22136986301369863, + "grad_norm": 2.1975133419036865, + "learning_rate": 3.43065693430657e-06, + "loss": 2.3667, + "step": 202 + }, + { + "epoch": 0.22246575342465755, + "grad_norm": 2.710186004638672, + "learning_rate": 3.4489051094890513e-06, + "loss": 2.3159, + "step": 203 + }, + { + "epoch": 0.22356164383561644, + "grad_norm": 2.7799506187438965, + "learning_rate": 3.467153284671533e-06, + "loss": 2.3519, + "step": 204 + }, + { + "epoch": 0.22465753424657534, + "grad_norm": 2.2479400634765625, + "learning_rate": 3.4854014598540146e-06, + "loss": 2.3103, + "step": 205 + }, + { + "epoch": 0.22575342465753426, + "grad_norm": 3.1538405418395996, + "learning_rate": 3.503649635036497e-06, + "loss": 2.3506, + "step": 206 + }, + { + "epoch": 0.22684931506849315, + "grad_norm": 2.7607762813568115, + "learning_rate": 3.5218978102189783e-06, + "loss": 2.3386, + "step": 207 + }, + { + "epoch": 0.22794520547945205, + "grad_norm": 2.371105909347534, + "learning_rate": 3.54014598540146e-06, + "loss": 2.3583, + "step": 208 + }, + { + "epoch": 0.22904109589041097, + "grad_norm": 2.793933391571045, + "learning_rate": 3.5583941605839416e-06, + "loss": 2.3782, + "step": 209 + }, + { + "epoch": 0.23013698630136986, + "grad_norm": 2.372732400894165, + "learning_rate": 3.576642335766424e-06, + "loss": 2.417, + "step": 210 + }, + { + "epoch": 0.23013698630136986, + "eval_loss": 2.451474189758301, + "eval_runtime": 41.7197, + "eval_samples_per_second": 23.97, + "eval_steps_per_second": 1.007, + "step": 210 + }, + { + "epoch": 0.23123287671232876, + "grad_norm": 2.6989448070526123, + "learning_rate": 3.5948905109489054e-06, + "loss": 2.393, + "step": 211 + }, + { + "epoch": 0.23232876712328768, + "grad_norm": 2.4324285984039307, + "learning_rate": 3.6131386861313872e-06, + "loss": 2.2738, + "step": 212 + }, + { + "epoch": 0.23342465753424657, + "grad_norm": 3.5063016414642334, + "learning_rate": 3.6313868613138687e-06, + "loss": 2.2823, + "step": 213 + }, + { + "epoch": 0.2345205479452055, + "grad_norm": 3.953951358795166, + "learning_rate": 3.6496350364963505e-06, + "loss": 2.3132, + "step": 214 + }, + { + "epoch": 0.2356164383561644, + "grad_norm": 2.3483071327209473, + "learning_rate": 3.667883211678833e-06, + "loss": 2.3244, + "step": 215 + }, + { + "epoch": 0.23671232876712328, + "grad_norm": 2.884199380874634, + "learning_rate": 3.6861313868613142e-06, + "loss": 2.3067, + "step": 216 + }, + { + "epoch": 0.2378082191780822, + "grad_norm": 2.9514520168304443, + "learning_rate": 3.704379562043796e-06, + "loss": 2.304, + "step": 217 + }, + { + "epoch": 0.2389041095890411, + "grad_norm": 2.7530996799468994, + "learning_rate": 3.7226277372262775e-06, + "loss": 2.3691, + "step": 218 + }, + { + "epoch": 0.24, + "grad_norm": 2.8753647804260254, + "learning_rate": 3.7408759124087594e-06, + "loss": 2.3406, + "step": 219 + }, + { + "epoch": 0.2410958904109589, + "grad_norm": 2.670574903488159, + "learning_rate": 3.7591240875912412e-06, + "loss": 2.304, + "step": 220 + }, + { + "epoch": 0.2410958904109589, + "eval_loss": 2.4395809173583984, + "eval_runtime": 46.6723, + "eval_samples_per_second": 21.426, + "eval_steps_per_second": 0.9, + "step": 220 + }, + { + "epoch": 0.2421917808219178, + "grad_norm": 2.4152438640594482, + "learning_rate": 3.777372262773723e-06, + "loss": 2.3294, + "step": 221 + }, + { + "epoch": 0.2432876712328767, + "grad_norm": 2.8398540019989014, + "learning_rate": 3.7956204379562045e-06, + "loss": 2.3216, + "step": 222 + }, + { + "epoch": 0.24438356164383562, + "grad_norm": 2.6036765575408936, + "learning_rate": 3.8138686131386864e-06, + "loss": 2.3109, + "step": 223 + }, + { + "epoch": 0.24547945205479452, + "grad_norm": 2.7073473930358887, + "learning_rate": 3.832116788321168e-06, + "loss": 2.3457, + "step": 224 + }, + { + "epoch": 0.2465753424657534, + "grad_norm": 2.584404230117798, + "learning_rate": 3.85036496350365e-06, + "loss": 2.3159, + "step": 225 + }, + { + "epoch": 0.24767123287671233, + "grad_norm": 2.4220666885375977, + "learning_rate": 3.868613138686132e-06, + "loss": 2.2944, + "step": 226 + }, + { + "epoch": 0.24876712328767123, + "grad_norm": 2.5636494159698486, + "learning_rate": 3.886861313868613e-06, + "loss": 2.3219, + "step": 227 + }, + { + "epoch": 0.24986301369863015, + "grad_norm": 2.2990903854370117, + "learning_rate": 3.905109489051096e-06, + "loss": 2.3342, + "step": 228 + }, + { + "epoch": 0.25095890410958904, + "grad_norm": 2.50913143157959, + "learning_rate": 3.923357664233577e-06, + "loss": 2.3619, + "step": 229 + }, + { + "epoch": 0.25205479452054796, + "grad_norm": 2.1687769889831543, + "learning_rate": 3.9416058394160585e-06, + "loss": 2.3153, + "step": 230 + }, + { + "epoch": 0.25205479452054796, + "eval_loss": 2.4315240383148193, + "eval_runtime": 48.3542, + "eval_samples_per_second": 20.681, + "eval_steps_per_second": 0.869, + "step": 230 + }, + { + "epoch": 0.25315068493150683, + "grad_norm": 2.842125177383423, + "learning_rate": 3.95985401459854e-06, + "loss": 2.3644, + "step": 231 + }, + { + "epoch": 0.25424657534246575, + "grad_norm": 2.258049964904785, + "learning_rate": 3.978102189781022e-06, + "loss": 2.3727, + "step": 232 + }, + { + "epoch": 0.2553424657534247, + "grad_norm": 2.4630489349365234, + "learning_rate": 3.996350364963504e-06, + "loss": 2.267, + "step": 233 + }, + { + "epoch": 0.25643835616438354, + "grad_norm": 3.1469743251800537, + "learning_rate": 4.014598540145986e-06, + "loss": 2.3432, + "step": 234 + }, + { + "epoch": 0.25753424657534246, + "grad_norm": 2.6277477741241455, + "learning_rate": 4.032846715328467e-06, + "loss": 2.2545, + "step": 235 + }, + { + "epoch": 0.2586301369863014, + "grad_norm": 2.404141902923584, + "learning_rate": 4.05109489051095e-06, + "loss": 2.3109, + "step": 236 + }, + { + "epoch": 0.25972602739726025, + "grad_norm": 2.4545605182647705, + "learning_rate": 4.069343065693431e-06, + "loss": 2.3112, + "step": 237 + }, + { + "epoch": 0.2608219178082192, + "grad_norm": 2.237638235092163, + "learning_rate": 4.0875912408759126e-06, + "loss": 2.3, + "step": 238 + }, + { + "epoch": 0.2619178082191781, + "grad_norm": 2.192811965942383, + "learning_rate": 4.105839416058395e-06, + "loss": 2.3108, + "step": 239 + }, + { + "epoch": 0.26301369863013696, + "grad_norm": 2.5058720111846924, + "learning_rate": 4.124087591240876e-06, + "loss": 2.3635, + "step": 240 + }, + { + "epoch": 0.26301369863013696, + "eval_loss": 2.4219377040863037, + "eval_runtime": 45.3561, + "eval_samples_per_second": 22.048, + "eval_steps_per_second": 0.926, + "step": 240 + }, + { + "epoch": 0.2641095890410959, + "grad_norm": 2.0708935260772705, + "learning_rate": 4.1423357664233586e-06, + "loss": 2.3129, + "step": 241 + }, + { + "epoch": 0.2652054794520548, + "grad_norm": 2.6353116035461426, + "learning_rate": 4.16058394160584e-06, + "loss": 2.3011, + "step": 242 + }, + { + "epoch": 0.2663013698630137, + "grad_norm": 2.8392436504364014, + "learning_rate": 4.1788321167883214e-06, + "loss": 2.3643, + "step": 243 + }, + { + "epoch": 0.2673972602739726, + "grad_norm": 3.1773295402526855, + "learning_rate": 4.197080291970803e-06, + "loss": 2.2642, + "step": 244 + }, + { + "epoch": 0.2684931506849315, + "grad_norm": 2.4757354259490967, + "learning_rate": 4.215328467153285e-06, + "loss": 2.3211, + "step": 245 + }, + { + "epoch": 0.26958904109589044, + "grad_norm": 2.5799412727355957, + "learning_rate": 4.233576642335767e-06, + "loss": 2.2909, + "step": 246 + }, + { + "epoch": 0.2706849315068493, + "grad_norm": 2.368502378463745, + "learning_rate": 4.251824817518249e-06, + "loss": 2.2951, + "step": 247 + }, + { + "epoch": 0.2717808219178082, + "grad_norm": 2.114969253540039, + "learning_rate": 4.27007299270073e-06, + "loss": 2.3124, + "step": 248 + }, + { + "epoch": 0.27287671232876715, + "grad_norm": 3.056820869445801, + "learning_rate": 4.288321167883212e-06, + "loss": 2.2923, + "step": 249 + }, + { + "epoch": 0.273972602739726, + "grad_norm": 2.7752203941345215, + "learning_rate": 4.306569343065693e-06, + "loss": 2.2874, + "step": 250 + }, + { + "epoch": 0.273972602739726, + "eval_loss": 2.4229912757873535, + "eval_runtime": 52.5941, + "eval_samples_per_second": 19.014, + "eval_steps_per_second": 0.799, + "step": 250 + }, + { + "epoch": 0.27506849315068493, + "grad_norm": 3.045740842819214, + "learning_rate": 4.3248175182481755e-06, + "loss": 2.2686, + "step": 251 + }, + { + "epoch": 0.27616438356164386, + "grad_norm": 2.646402359008789, + "learning_rate": 4.343065693430658e-06, + "loss": 2.3247, + "step": 252 + }, + { + "epoch": 0.2772602739726027, + "grad_norm": 3.274994134902954, + "learning_rate": 4.361313868613139e-06, + "loss": 2.3093, + "step": 253 + }, + { + "epoch": 0.27835616438356164, + "grad_norm": 2.352246046066284, + "learning_rate": 4.379562043795621e-06, + "loss": 2.3124, + "step": 254 + }, + { + "epoch": 0.27945205479452057, + "grad_norm": 3.248717784881592, + "learning_rate": 4.397810218978102e-06, + "loss": 2.3417, + "step": 255 + }, + { + "epoch": 0.28054794520547943, + "grad_norm": 3.0919013023376465, + "learning_rate": 4.416058394160584e-06, + "loss": 2.3093, + "step": 256 + }, + { + "epoch": 0.28164383561643835, + "grad_norm": 2.1174702644348145, + "learning_rate": 4.434306569343066e-06, + "loss": 2.305, + "step": 257 + }, + { + "epoch": 0.2827397260273973, + "grad_norm": 2.913797378540039, + "learning_rate": 4.452554744525548e-06, + "loss": 2.2831, + "step": 258 + }, + { + "epoch": 0.28383561643835614, + "grad_norm": 2.693089723587036, + "learning_rate": 4.4708029197080295e-06, + "loss": 2.3206, + "step": 259 + }, + { + "epoch": 0.28493150684931506, + "grad_norm": 2.521411895751953, + "learning_rate": 4.489051094890512e-06, + "loss": 2.3196, + "step": 260 + }, + { + "epoch": 0.28493150684931506, + "eval_loss": 2.4178109169006348, + "eval_runtime": 50.6753, + "eval_samples_per_second": 19.733, + "eval_steps_per_second": 0.829, + "step": 260 + }, + { + "epoch": 0.286027397260274, + "grad_norm": 2.346346855163574, + "learning_rate": 4.507299270072993e-06, + "loss": 2.3433, + "step": 261 + }, + { + "epoch": 0.28712328767123285, + "grad_norm": 2.520368814468384, + "learning_rate": 4.525547445255475e-06, + "loss": 2.3174, + "step": 262 + }, + { + "epoch": 0.2882191780821918, + "grad_norm": 2.550347328186035, + "learning_rate": 4.543795620437956e-06, + "loss": 2.3072, + "step": 263 + }, + { + "epoch": 0.2893150684931507, + "grad_norm": 2.268662691116333, + "learning_rate": 4.562043795620438e-06, + "loss": 2.2763, + "step": 264 + }, + { + "epoch": 0.29041095890410956, + "grad_norm": 2.445798873901367, + "learning_rate": 4.580291970802921e-06, + "loss": 2.3395, + "step": 265 + }, + { + "epoch": 0.2915068493150685, + "grad_norm": 2.5391135215759277, + "learning_rate": 4.598540145985402e-06, + "loss": 2.3496, + "step": 266 + }, + { + "epoch": 0.2926027397260274, + "grad_norm": 2.3553097248077393, + "learning_rate": 4.6167883211678835e-06, + "loss": 2.3243, + "step": 267 + }, + { + "epoch": 0.2936986301369863, + "grad_norm": 2.448089599609375, + "learning_rate": 4.635036496350365e-06, + "loss": 2.2887, + "step": 268 + }, + { + "epoch": 0.2947945205479452, + "grad_norm": 2.1600170135498047, + "learning_rate": 4.653284671532847e-06, + "loss": 2.2816, + "step": 269 + }, + { + "epoch": 0.2958904109589041, + "grad_norm": 2.3783180713653564, + "learning_rate": 4.671532846715329e-06, + "loss": 2.3275, + "step": 270 + }, + { + "epoch": 0.2958904109589041, + "eval_loss": 2.413832902908325, + "eval_runtime": 48.7524, + "eval_samples_per_second": 20.512, + "eval_steps_per_second": 0.861, + "step": 270 + }, + { + "epoch": 0.29698630136986304, + "grad_norm": 2.468320369720459, + "learning_rate": 4.689781021897811e-06, + "loss": 2.3358, + "step": 271 + }, + { + "epoch": 0.2980821917808219, + "grad_norm": 2.401634454727173, + "learning_rate": 4.708029197080292e-06, + "loss": 2.2991, + "step": 272 + }, + { + "epoch": 0.2991780821917808, + "grad_norm": 2.4604203701019287, + "learning_rate": 4.726277372262774e-06, + "loss": 2.3496, + "step": 273 + }, + { + "epoch": 0.30027397260273975, + "grad_norm": 2.0338871479034424, + "learning_rate": 4.744525547445255e-06, + "loss": 2.3115, + "step": 274 + }, + { + "epoch": 0.3013698630136986, + "grad_norm": 2.5404436588287354, + "learning_rate": 4.7627737226277375e-06, + "loss": 2.2665, + "step": 275 + }, + { + "epoch": 0.30246575342465754, + "grad_norm": 2.5369458198547363, + "learning_rate": 4.78102189781022e-06, + "loss": 2.3072, + "step": 276 + }, + { + "epoch": 0.30356164383561646, + "grad_norm": 2.471522569656372, + "learning_rate": 4.799270072992701e-06, + "loss": 2.3113, + "step": 277 + }, + { + "epoch": 0.3046575342465753, + "grad_norm": 2.3890135288238525, + "learning_rate": 4.8175182481751835e-06, + "loss": 2.2838, + "step": 278 + }, + { + "epoch": 0.30575342465753425, + "grad_norm": 2.2485053539276123, + "learning_rate": 4.835766423357665e-06, + "loss": 2.3052, + "step": 279 + }, + { + "epoch": 0.30684931506849317, + "grad_norm": 2.1090753078460693, + "learning_rate": 4.854014598540146e-06, + "loss": 2.3054, + "step": 280 + }, + { + "epoch": 0.30684931506849317, + "eval_loss": 2.412548780441284, + "eval_runtime": 48.5034, + "eval_samples_per_second": 20.617, + "eval_steps_per_second": 0.866, + "step": 280 + }, + { + "epoch": 0.30794520547945203, + "grad_norm": 2.299133777618408, + "learning_rate": 4.872262773722628e-06, + "loss": 2.2599, + "step": 281 + }, + { + "epoch": 0.30904109589041096, + "grad_norm": 2.177905797958374, + "learning_rate": 4.89051094890511e-06, + "loss": 2.2745, + "step": 282 + }, + { + "epoch": 0.3101369863013699, + "grad_norm": 2.3178462982177734, + "learning_rate": 4.9087591240875915e-06, + "loss": 2.2562, + "step": 283 + }, + { + "epoch": 0.31123287671232874, + "grad_norm": 2.5088376998901367, + "learning_rate": 4.927007299270074e-06, + "loss": 2.2945, + "step": 284 + }, + { + "epoch": 0.31232876712328766, + "grad_norm": 2.095379590988159, + "learning_rate": 4.945255474452555e-06, + "loss": 2.3038, + "step": 285 + }, + { + "epoch": 0.3134246575342466, + "grad_norm": 2.1792995929718018, + "learning_rate": 4.963503649635037e-06, + "loss": 2.2702, + "step": 286 + }, + { + "epoch": 0.31452054794520545, + "grad_norm": 2.3001186847686768, + "learning_rate": 4.981751824817518e-06, + "loss": 2.3482, + "step": 287 + }, + { + "epoch": 0.3156164383561644, + "grad_norm": 2.1576106548309326, + "learning_rate": 5e-06, + "loss": 2.3391, + "step": 288 + }, + { + "epoch": 0.3167123287671233, + "grad_norm": 2.3715577125549316, + "learning_rate": 4.997969130787978e-06, + "loss": 2.3049, + "step": 289 + }, + { + "epoch": 0.3178082191780822, + "grad_norm": 2.0316715240478516, + "learning_rate": 4.995938261575955e-06, + "loss": 2.3111, + "step": 290 + }, + { + "epoch": 0.3178082191780822, + "eval_loss": 2.40299654006958, + "eval_runtime": 50.0873, + "eval_samples_per_second": 19.965, + "eval_steps_per_second": 0.839, + "step": 290 + }, + { + "epoch": 0.3189041095890411, + "grad_norm": 2.1135756969451904, + "learning_rate": 4.993907392363932e-06, + "loss": 2.2465, + "step": 291 + }, + { + "epoch": 0.32, + "grad_norm": 2.335655927658081, + "learning_rate": 4.9918765231519095e-06, + "loss": 2.31, + "step": 292 + }, + { + "epoch": 0.32109589041095893, + "grad_norm": 2.3255794048309326, + "learning_rate": 4.989845653939887e-06, + "loss": 2.2531, + "step": 293 + }, + { + "epoch": 0.3221917808219178, + "grad_norm": 2.111055612564087, + "learning_rate": 4.987814784727864e-06, + "loss": 2.2799, + "step": 294 + }, + { + "epoch": 0.3232876712328767, + "grad_norm": 2.1773171424865723, + "learning_rate": 4.985783915515841e-06, + "loss": 2.2884, + "step": 295 + }, + { + "epoch": 0.32438356164383564, + "grad_norm": 2.3559885025024414, + "learning_rate": 4.983753046303819e-06, + "loss": 2.2711, + "step": 296 + }, + { + "epoch": 0.3254794520547945, + "grad_norm": 2.120802879333496, + "learning_rate": 4.981722177091796e-06, + "loss": 2.2946, + "step": 297 + }, + { + "epoch": 0.3265753424657534, + "grad_norm": 2.2319388389587402, + "learning_rate": 4.979691307879773e-06, + "loss": 2.2771, + "step": 298 + }, + { + "epoch": 0.32767123287671235, + "grad_norm": 2.208740711212158, + "learning_rate": 4.9776604386677504e-06, + "loss": 2.3003, + "step": 299 + }, + { + "epoch": 0.3287671232876712, + "grad_norm": 2.543060541152954, + "learning_rate": 4.975629569455728e-06, + "loss": 2.2891, + "step": 300 + }, + { + "epoch": 0.3287671232876712, + "eval_loss": 2.398495674133301, + "eval_runtime": 51.1973, + "eval_samples_per_second": 19.532, + "eval_steps_per_second": 0.82, + "step": 300 + }, + { + "epoch": 0.32986301369863014, + "grad_norm": 2.086219072341919, + "learning_rate": 4.973598700243705e-06, + "loss": 2.3286, + "step": 301 + }, + { + "epoch": 0.33095890410958906, + "grad_norm": 2.3673319816589355, + "learning_rate": 4.971567831031681e-06, + "loss": 2.2672, + "step": 302 + }, + { + "epoch": 0.3320547945205479, + "grad_norm": 2.167593240737915, + "learning_rate": 4.9695369618196595e-06, + "loss": 2.264, + "step": 303 + }, + { + "epoch": 0.33315068493150685, + "grad_norm": 2.7218785285949707, + "learning_rate": 4.967506092607636e-06, + "loss": 2.1737, + "step": 304 + }, + { + "epoch": 0.33424657534246577, + "grad_norm": 2.673245429992676, + "learning_rate": 4.965475223395614e-06, + "loss": 2.2439, + "step": 305 + }, + { + "epoch": 0.33534246575342463, + "grad_norm": 2.0341989994049072, + "learning_rate": 4.9634443541835905e-06, + "loss": 2.2614, + "step": 306 + }, + { + "epoch": 0.33643835616438356, + "grad_norm": 1.9314247369766235, + "learning_rate": 4.961413484971569e-06, + "loss": 2.2592, + "step": 307 + }, + { + "epoch": 0.3375342465753425, + "grad_norm": 2.537794589996338, + "learning_rate": 4.959382615759545e-06, + "loss": 2.2859, + "step": 308 + }, + { + "epoch": 0.33863013698630134, + "grad_norm": 2.0811166763305664, + "learning_rate": 4.957351746547523e-06, + "loss": 2.2868, + "step": 309 + }, + { + "epoch": 0.33972602739726027, + "grad_norm": 2.783701181411743, + "learning_rate": 4.9553208773355e-06, + "loss": 2.2649, + "step": 310 + }, + { + "epoch": 0.33972602739726027, + "eval_loss": 2.400538921356201, + "eval_runtime": 43.185, + "eval_samples_per_second": 23.156, + "eval_steps_per_second": 0.973, + "step": 310 + }, + { + "epoch": 0.3408219178082192, + "grad_norm": 2.8541319370269775, + "learning_rate": 4.953290008123478e-06, + "loss": 2.3729, + "step": 311 + }, + { + "epoch": 0.3419178082191781, + "grad_norm": 2.1852617263793945, + "learning_rate": 4.951259138911454e-06, + "loss": 2.2885, + "step": 312 + }, + { + "epoch": 0.343013698630137, + "grad_norm": 2.872201919555664, + "learning_rate": 4.9492282696994315e-06, + "loss": 2.2353, + "step": 313 + }, + { + "epoch": 0.3441095890410959, + "grad_norm": 2.5140910148620605, + "learning_rate": 4.947197400487409e-06, + "loss": 2.3364, + "step": 314 + }, + { + "epoch": 0.3452054794520548, + "grad_norm": 2.619974374771118, + "learning_rate": 4.945166531275386e-06, + "loss": 2.2721, + "step": 315 + }, + { + "epoch": 0.3463013698630137, + "grad_norm": 2.4567618370056152, + "learning_rate": 4.943135662063363e-06, + "loss": 2.26, + "step": 316 + }, + { + "epoch": 0.3473972602739726, + "grad_norm": 2.9603192806243896, + "learning_rate": 4.9411047928513406e-06, + "loss": 2.2861, + "step": 317 + }, + { + "epoch": 0.34849315068493153, + "grad_norm": 2.115262985229492, + "learning_rate": 4.939073923639318e-06, + "loss": 2.3125, + "step": 318 + }, + { + "epoch": 0.3495890410958904, + "grad_norm": 2.693024158477783, + "learning_rate": 4.937043054427295e-06, + "loss": 2.3054, + "step": 319 + }, + { + "epoch": 0.3506849315068493, + "grad_norm": 2.3439996242523193, + "learning_rate": 4.935012185215272e-06, + "loss": 2.2576, + "step": 320 + }, + { + "epoch": 0.3506849315068493, + "eval_loss": 2.3920650482177734, + "eval_runtime": 47.7451, + "eval_samples_per_second": 20.945, + "eval_steps_per_second": 0.88, + "step": 320 + }, + { + "epoch": 0.35178082191780824, + "grad_norm": 2.3009634017944336, + "learning_rate": 4.93298131600325e-06, + "loss": 2.2454, + "step": 321 + }, + { + "epoch": 0.3528767123287671, + "grad_norm": 2.593203544616699, + "learning_rate": 4.930950446791227e-06, + "loss": 2.2965, + "step": 322 + }, + { + "epoch": 0.353972602739726, + "grad_norm": 2.9069650173187256, + "learning_rate": 4.928919577579204e-06, + "loss": 2.2785, + "step": 323 + }, + { + "epoch": 0.35506849315068495, + "grad_norm": 2.1021759510040283, + "learning_rate": 4.9268887083671815e-06, + "loss": 2.2857, + "step": 324 + }, + { + "epoch": 0.3561643835616438, + "grad_norm": 2.0330185890197754, + "learning_rate": 4.924857839155159e-06, + "loss": 2.2697, + "step": 325 + }, + { + "epoch": 0.35726027397260274, + "grad_norm": 2.1318063735961914, + "learning_rate": 4.922826969943136e-06, + "loss": 2.2698, + "step": 326 + }, + { + "epoch": 0.35835616438356166, + "grad_norm": 2.0157387256622314, + "learning_rate": 4.920796100731113e-06, + "loss": 2.292, + "step": 327 + }, + { + "epoch": 0.3594520547945205, + "grad_norm": 2.3266026973724365, + "learning_rate": 4.918765231519091e-06, + "loss": 2.2215, + "step": 328 + }, + { + "epoch": 0.36054794520547945, + "grad_norm": 2.066305160522461, + "learning_rate": 4.916734362307068e-06, + "loss": 2.2758, + "step": 329 + }, + { + "epoch": 0.36164383561643837, + "grad_norm": 2.0219993591308594, + "learning_rate": 4.914703493095045e-06, + "loss": 2.2381, + "step": 330 + }, + { + "epoch": 0.36164383561643837, + "eval_loss": 2.394064426422119, + "eval_runtime": 51.5482, + "eval_samples_per_second": 19.399, + "eval_steps_per_second": 0.815, + "step": 330 + }, + { + "epoch": 0.36273972602739724, + "grad_norm": 2.3964319229125977, + "learning_rate": 4.912672623883022e-06, + "loss": 2.2384, + "step": 331 + }, + { + "epoch": 0.36383561643835616, + "grad_norm": 2.5466530323028564, + "learning_rate": 4.910641754671e-06, + "loss": 2.2428, + "step": 332 + }, + { + "epoch": 0.3649315068493151, + "grad_norm": 2.236152410507202, + "learning_rate": 4.908610885458977e-06, + "loss": 2.2639, + "step": 333 + }, + { + "epoch": 0.36602739726027395, + "grad_norm": 2.7267303466796875, + "learning_rate": 4.906580016246954e-06, + "loss": 2.2608, + "step": 334 + }, + { + "epoch": 0.36712328767123287, + "grad_norm": 1.9403765201568604, + "learning_rate": 4.9045491470349315e-06, + "loss": 2.2962, + "step": 335 + }, + { + "epoch": 0.3682191780821918, + "grad_norm": 2.89070463180542, + "learning_rate": 4.902518277822909e-06, + "loss": 2.2847, + "step": 336 + }, + { + "epoch": 0.3693150684931507, + "grad_norm": 2.6427228450775146, + "learning_rate": 4.900487408610886e-06, + "loss": 2.2734, + "step": 337 + }, + { + "epoch": 0.3704109589041096, + "grad_norm": 2.1680362224578857, + "learning_rate": 4.898456539398863e-06, + "loss": 2.2869, + "step": 338 + }, + { + "epoch": 0.3715068493150685, + "grad_norm": 3.0160160064697266, + "learning_rate": 4.89642567018684e-06, + "loss": 2.3043, + "step": 339 + }, + { + "epoch": 0.3726027397260274, + "grad_norm": 1.9858418703079224, + "learning_rate": 4.894394800974818e-06, + "loss": 2.3422, + "step": 340 + }, + { + "epoch": 0.3726027397260274, + "eval_loss": 2.387559652328491, + "eval_runtime": 52.2455, + "eval_samples_per_second": 19.14, + "eval_steps_per_second": 0.804, + "step": 340 + }, + { + "epoch": 0.3736986301369863, + "grad_norm": 2.243699073791504, + "learning_rate": 4.892363931762794e-06, + "loss": 2.2394, + "step": 341 + }, + { + "epoch": 0.3747945205479452, + "grad_norm": 2.371299982070923, + "learning_rate": 4.8903330625507725e-06, + "loss": 2.2856, + "step": 342 + }, + { + "epoch": 0.37589041095890413, + "grad_norm": 2.3769736289978027, + "learning_rate": 4.888302193338749e-06, + "loss": 2.246, + "step": 343 + }, + { + "epoch": 0.376986301369863, + "grad_norm": 2.3209030628204346, + "learning_rate": 4.886271324126727e-06, + "loss": 2.2331, + "step": 344 + }, + { + "epoch": 0.3780821917808219, + "grad_norm": 2.747319221496582, + "learning_rate": 4.8842404549147034e-06, + "loss": 2.3115, + "step": 345 + }, + { + "epoch": 0.37917808219178084, + "grad_norm": 1.8599231243133545, + "learning_rate": 4.8822095857026816e-06, + "loss": 2.2994, + "step": 346 + }, + { + "epoch": 0.3802739726027397, + "grad_norm": 2.0358943939208984, + "learning_rate": 4.880178716490658e-06, + "loss": 2.271, + "step": 347 + }, + { + "epoch": 0.38136986301369863, + "grad_norm": 2.409106731414795, + "learning_rate": 4.878147847278636e-06, + "loss": 2.2281, + "step": 348 + }, + { + "epoch": 0.38246575342465755, + "grad_norm": 2.1641807556152344, + "learning_rate": 4.8761169780666125e-06, + "loss": 2.2437, + "step": 349 + }, + { + "epoch": 0.3835616438356164, + "grad_norm": 2.6917359828948975, + "learning_rate": 4.874086108854591e-06, + "loss": 2.226, + "step": 350 + }, + { + "epoch": 0.3835616438356164, + "eval_loss": 2.382991313934326, + "eval_runtime": 48.9534, + "eval_samples_per_second": 20.428, + "eval_steps_per_second": 0.858, + "step": 350 + }, + { + "epoch": 0.38465753424657534, + "grad_norm": 2.364215612411499, + "learning_rate": 4.872055239642567e-06, + "loss": 2.2319, + "step": 351 + }, + { + "epoch": 0.38575342465753426, + "grad_norm": 2.317643165588379, + "learning_rate": 4.870024370430544e-06, + "loss": 2.2096, + "step": 352 + }, + { + "epoch": 0.3868493150684931, + "grad_norm": 2.2283246517181396, + "learning_rate": 4.867993501218522e-06, + "loss": 2.2634, + "step": 353 + }, + { + "epoch": 0.38794520547945205, + "grad_norm": 2.498391628265381, + "learning_rate": 4.865962632006499e-06, + "loss": 2.2813, + "step": 354 + }, + { + "epoch": 0.38904109589041097, + "grad_norm": 2.3889403343200684, + "learning_rate": 4.863931762794476e-06, + "loss": 2.2361, + "step": 355 + }, + { + "epoch": 0.39013698630136984, + "grad_norm": 2.2049975395202637, + "learning_rate": 4.8619008935824535e-06, + "loss": 2.2583, + "step": 356 + }, + { + "epoch": 0.39123287671232876, + "grad_norm": 2.132840871810913, + "learning_rate": 4.859870024370431e-06, + "loss": 2.2823, + "step": 357 + }, + { + "epoch": 0.3923287671232877, + "grad_norm": 2.2692184448242188, + "learning_rate": 4.857839155158408e-06, + "loss": 2.2939, + "step": 358 + }, + { + "epoch": 0.3934246575342466, + "grad_norm": 2.0501933097839355, + "learning_rate": 4.855808285946385e-06, + "loss": 2.2976, + "step": 359 + }, + { + "epoch": 0.39452054794520547, + "grad_norm": 2.170236587524414, + "learning_rate": 4.8537774167343626e-06, + "loss": 2.2748, + "step": 360 + }, + { + "epoch": 0.39452054794520547, + "eval_loss": 2.3783886432647705, + "eval_runtime": 46.7324, + "eval_samples_per_second": 21.398, + "eval_steps_per_second": 0.899, + "step": 360 + }, + { + "epoch": 0.3956164383561644, + "grad_norm": 2.283137083053589, + "learning_rate": 4.85174654752234e-06, + "loss": 2.2705, + "step": 361 + }, + { + "epoch": 0.3967123287671233, + "grad_norm": 2.195984363555908, + "learning_rate": 4.849715678310317e-06, + "loss": 2.26, + "step": 362 + }, + { + "epoch": 0.3978082191780822, + "grad_norm": 2.399935007095337, + "learning_rate": 4.847684809098294e-06, + "loss": 2.2639, + "step": 363 + }, + { + "epoch": 0.3989041095890411, + "grad_norm": 2.411280393600464, + "learning_rate": 4.845653939886272e-06, + "loss": 2.2523, + "step": 364 + }, + { + "epoch": 0.4, + "grad_norm": 2.2033896446228027, + "learning_rate": 4.843623070674249e-06, + "loss": 2.2722, + "step": 365 + }, + { + "epoch": 0.4010958904109589, + "grad_norm": 2.2990238666534424, + "learning_rate": 4.841592201462226e-06, + "loss": 2.2411, + "step": 366 + }, + { + "epoch": 0.4021917808219178, + "grad_norm": 1.911453127861023, + "learning_rate": 4.8395613322502035e-06, + "loss": 2.2921, + "step": 367 + }, + { + "epoch": 0.40328767123287673, + "grad_norm": 2.453195095062256, + "learning_rate": 4.837530463038181e-06, + "loss": 2.218, + "step": 368 + }, + { + "epoch": 0.4043835616438356, + "grad_norm": 2.126992702484131, + "learning_rate": 4.835499593826158e-06, + "loss": 2.2628, + "step": 369 + }, + { + "epoch": 0.4054794520547945, + "grad_norm": 1.9305835962295532, + "learning_rate": 4.833468724614135e-06, + "loss": 2.2726, + "step": 370 + }, + { + "epoch": 0.4054794520547945, + "eval_loss": 2.3741114139556885, + "eval_runtime": 49.2956, + "eval_samples_per_second": 20.286, + "eval_steps_per_second": 0.852, + "step": 370 + }, + { + "epoch": 0.40657534246575344, + "grad_norm": 2.222024917602539, + "learning_rate": 4.831437855402113e-06, + "loss": 2.2634, + "step": 371 + }, + { + "epoch": 0.4076712328767123, + "grad_norm": 2.473958969116211, + "learning_rate": 4.82940698619009e-06, + "loss": 2.1817, + "step": 372 + }, + { + "epoch": 0.40876712328767123, + "grad_norm": 1.9208838939666748, + "learning_rate": 4.827376116978067e-06, + "loss": 2.2353, + "step": 373 + }, + { + "epoch": 0.40986301369863015, + "grad_norm": 2.7226450443267822, + "learning_rate": 4.8253452477660444e-06, + "loss": 2.2831, + "step": 374 + }, + { + "epoch": 0.410958904109589, + "grad_norm": 1.9078222513198853, + "learning_rate": 4.823314378554022e-06, + "loss": 2.264, + "step": 375 + }, + { + "epoch": 0.41205479452054794, + "grad_norm": 2.3946311473846436, + "learning_rate": 4.821283509341998e-06, + "loss": 2.251, + "step": 376 + }, + { + "epoch": 0.41315068493150686, + "grad_norm": 1.847343921661377, + "learning_rate": 4.819252640129976e-06, + "loss": 2.2441, + "step": 377 + }, + { + "epoch": 0.41424657534246573, + "grad_norm": 2.475385904312134, + "learning_rate": 4.817221770917953e-06, + "loss": 2.2639, + "step": 378 + }, + { + "epoch": 0.41534246575342465, + "grad_norm": 2.1396560668945312, + "learning_rate": 4.815190901705931e-06, + "loss": 2.2353, + "step": 379 + }, + { + "epoch": 0.41643835616438357, + "grad_norm": 2.3868885040283203, + "learning_rate": 4.813160032493907e-06, + "loss": 2.2551, + "step": 380 + }, + { + "epoch": 0.41643835616438357, + "eval_loss": 2.369168519973755, + "eval_runtime": 45.6131, + "eval_samples_per_second": 21.924, + "eval_steps_per_second": 0.921, + "step": 380 + }, + { + "epoch": 0.41753424657534244, + "grad_norm": 2.0193872451782227, + "learning_rate": 4.811129163281885e-06, + "loss": 2.2399, + "step": 381 + }, + { + "epoch": 0.41863013698630136, + "grad_norm": 2.4655935764312744, + "learning_rate": 4.809098294069862e-06, + "loss": 2.2729, + "step": 382 + }, + { + "epoch": 0.4197260273972603, + "grad_norm": 1.880974292755127, + "learning_rate": 4.80706742485784e-06, + "loss": 2.2681, + "step": 383 + }, + { + "epoch": 0.4208219178082192, + "grad_norm": 2.1912078857421875, + "learning_rate": 4.805036555645816e-06, + "loss": 2.2961, + "step": 384 + }, + { + "epoch": 0.42191780821917807, + "grad_norm": 1.985931634902954, + "learning_rate": 4.8030056864337945e-06, + "loss": 2.2407, + "step": 385 + }, + { + "epoch": 0.423013698630137, + "grad_norm": 2.4353220462799072, + "learning_rate": 4.800974817221771e-06, + "loss": 2.2867, + "step": 386 + }, + { + "epoch": 0.4241095890410959, + "grad_norm": 1.8810508251190186, + "learning_rate": 4.798943948009749e-06, + "loss": 2.2498, + "step": 387 + }, + { + "epoch": 0.4252054794520548, + "grad_norm": 2.195033073425293, + "learning_rate": 4.7969130787977254e-06, + "loss": 2.2532, + "step": 388 + }, + { + "epoch": 0.4263013698630137, + "grad_norm": 1.855372428894043, + "learning_rate": 4.794882209585703e-06, + "loss": 2.2538, + "step": 389 + }, + { + "epoch": 0.4273972602739726, + "grad_norm": 2.4465434551239014, + "learning_rate": 4.79285134037368e-06, + "loss": 2.2724, + "step": 390 + }, + { + "epoch": 0.4273972602739726, + "eval_loss": 2.364928960800171, + "eval_runtime": 48.0245, + "eval_samples_per_second": 20.823, + "eval_steps_per_second": 0.875, + "step": 390 + }, + { + "epoch": 0.4284931506849315, + "grad_norm": 1.8911314010620117, + "learning_rate": 4.790820471161657e-06, + "loss": 2.2291, + "step": 391 + }, + { + "epoch": 0.4295890410958904, + "grad_norm": 2.1975977420806885, + "learning_rate": 4.7887896019496345e-06, + "loss": 2.2373, + "step": 392 + }, + { + "epoch": 0.43068493150684933, + "grad_norm": 2.015331506729126, + "learning_rate": 4.786758732737612e-06, + "loss": 2.2483, + "step": 393 + }, + { + "epoch": 0.4317808219178082, + "grad_norm": 2.039339065551758, + "learning_rate": 4.784727863525589e-06, + "loss": 2.2746, + "step": 394 + }, + { + "epoch": 0.4328767123287671, + "grad_norm": 1.977513074874878, + "learning_rate": 4.782696994313566e-06, + "loss": 2.2275, + "step": 395 + }, + { + "epoch": 0.43397260273972604, + "grad_norm": 2.147217035293579, + "learning_rate": 4.780666125101544e-06, + "loss": 2.252, + "step": 396 + }, + { + "epoch": 0.4350684931506849, + "grad_norm": 2.1799919605255127, + "learning_rate": 4.778635255889521e-06, + "loss": 2.1975, + "step": 397 + }, + { + "epoch": 0.43616438356164383, + "grad_norm": 2.582240343093872, + "learning_rate": 4.776604386677498e-06, + "loss": 2.2294, + "step": 398 + }, + { + "epoch": 0.43726027397260275, + "grad_norm": 1.9127846956253052, + "learning_rate": 4.7745735174654755e-06, + "loss": 2.2319, + "step": 399 + }, + { + "epoch": 0.4383561643835616, + "grad_norm": 2.2444543838500977, + "learning_rate": 4.772542648253453e-06, + "loss": 2.2857, + "step": 400 + }, + { + "epoch": 0.4383561643835616, + "eval_loss": 2.3653724193573, + "eval_runtime": 43.8304, + "eval_samples_per_second": 22.815, + "eval_steps_per_second": 0.958, + "step": 400 + }, + { + "epoch": 0.43945205479452054, + "grad_norm": 1.9913676977157593, + "learning_rate": 4.77051177904143e-06, + "loss": 2.274, + "step": 401 + }, + { + "epoch": 0.44054794520547946, + "grad_norm": 2.2850067615509033, + "learning_rate": 4.768480909829407e-06, + "loss": 2.2209, + "step": 402 + }, + { + "epoch": 0.44164383561643833, + "grad_norm": 2.4984688758850098, + "learning_rate": 4.766450040617385e-06, + "loss": 2.263, + "step": 403 + }, + { + "epoch": 0.44273972602739725, + "grad_norm": 2.3244640827178955, + "learning_rate": 4.764419171405362e-06, + "loss": 2.2368, + "step": 404 + }, + { + "epoch": 0.4438356164383562, + "grad_norm": 2.7579681873321533, + "learning_rate": 4.762388302193339e-06, + "loss": 2.2762, + "step": 405 + }, + { + "epoch": 0.4449315068493151, + "grad_norm": 2.1996467113494873, + "learning_rate": 4.760357432981316e-06, + "loss": 2.2536, + "step": 406 + }, + { + "epoch": 0.44602739726027396, + "grad_norm": 2.3767571449279785, + "learning_rate": 4.758326563769294e-06, + "loss": 2.2353, + "step": 407 + }, + { + "epoch": 0.4471232876712329, + "grad_norm": 1.9148359298706055, + "learning_rate": 4.756295694557271e-06, + "loss": 2.2416, + "step": 408 + }, + { + "epoch": 0.4482191780821918, + "grad_norm": 2.6499507427215576, + "learning_rate": 4.754264825345248e-06, + "loss": 2.2641, + "step": 409 + }, + { + "epoch": 0.44931506849315067, + "grad_norm": 1.9890286922454834, + "learning_rate": 4.7522339561332255e-06, + "loss": 2.2148, + "step": 410 + }, + { + "epoch": 0.44931506849315067, + "eval_loss": 2.3670520782470703, + "eval_runtime": 41.1722, + "eval_samples_per_second": 24.288, + "eval_steps_per_second": 1.02, + "step": 410 + }, + { + "epoch": 0.4504109589041096, + "grad_norm": 2.575968027114868, + "learning_rate": 4.750203086921203e-06, + "loss": 2.2254, + "step": 411 + }, + { + "epoch": 0.4515068493150685, + "grad_norm": 2.015174388885498, + "learning_rate": 4.74817221770918e-06, + "loss": 2.1881, + "step": 412 + }, + { + "epoch": 0.4526027397260274, + "grad_norm": 2.44292950630188, + "learning_rate": 4.7461413484971565e-06, + "loss": 2.256, + "step": 413 + }, + { + "epoch": 0.4536986301369863, + "grad_norm": 1.9808441400527954, + "learning_rate": 4.744110479285135e-06, + "loss": 2.285, + "step": 414 + }, + { + "epoch": 0.4547945205479452, + "grad_norm": 2.009481906890869, + "learning_rate": 4.742079610073111e-06, + "loss": 2.2018, + "step": 415 + }, + { + "epoch": 0.4558904109589041, + "grad_norm": 2.1466031074523926, + "learning_rate": 4.740048740861089e-06, + "loss": 2.224, + "step": 416 + }, + { + "epoch": 0.456986301369863, + "grad_norm": 2.1337907314300537, + "learning_rate": 4.738017871649066e-06, + "loss": 2.2404, + "step": 417 + }, + { + "epoch": 0.45808219178082193, + "grad_norm": 1.94125235080719, + "learning_rate": 4.735987002437044e-06, + "loss": 2.2474, + "step": 418 + }, + { + "epoch": 0.4591780821917808, + "grad_norm": 2.379099130630493, + "learning_rate": 4.73395613322502e-06, + "loss": 2.2858, + "step": 419 + }, + { + "epoch": 0.4602739726027397, + "grad_norm": 1.844056248664856, + "learning_rate": 4.731925264012998e-06, + "loss": 2.2967, + "step": 420 + }, + { + "epoch": 0.4602739726027397, + "eval_loss": 2.3622241020202637, + "eval_runtime": 45.4674, + "eval_samples_per_second": 21.994, + "eval_steps_per_second": 0.924, + "step": 420 + }, + { + "epoch": 0.46136986301369864, + "grad_norm": 2.411895513534546, + "learning_rate": 4.729894394800975e-06, + "loss": 2.274, + "step": 421 + }, + { + "epoch": 0.4624657534246575, + "grad_norm": 2.014885425567627, + "learning_rate": 4.727863525588953e-06, + "loss": 2.2144, + "step": 422 + }, + { + "epoch": 0.46356164383561643, + "grad_norm": 2.112009048461914, + "learning_rate": 4.725832656376929e-06, + "loss": 2.1824, + "step": 423 + }, + { + "epoch": 0.46465753424657535, + "grad_norm": 2.111392021179199, + "learning_rate": 4.723801787164907e-06, + "loss": 2.2261, + "step": 424 + }, + { + "epoch": 0.4657534246575342, + "grad_norm": 2.0349185466766357, + "learning_rate": 4.721770917952885e-06, + "loss": 2.2622, + "step": 425 + }, + { + "epoch": 0.46684931506849314, + "grad_norm": 2.30222487449646, + "learning_rate": 4.719740048740861e-06, + "loss": 2.2206, + "step": 426 + }, + { + "epoch": 0.46794520547945206, + "grad_norm": 2.018789529800415, + "learning_rate": 4.717709179528839e-06, + "loss": 2.2196, + "step": 427 + }, + { + "epoch": 0.469041095890411, + "grad_norm": 2.046649932861328, + "learning_rate": 4.715678310316816e-06, + "loss": 2.2584, + "step": 428 + }, + { + "epoch": 0.47013698630136985, + "grad_norm": 2.0393869876861572, + "learning_rate": 4.713647441104794e-06, + "loss": 2.2537, + "step": 429 + }, + { + "epoch": 0.4712328767123288, + "grad_norm": 2.2788116931915283, + "learning_rate": 4.71161657189277e-06, + "loss": 2.2145, + "step": 430 + }, + { + "epoch": 0.4712328767123288, + "eval_loss": 2.3575785160064697, + "eval_runtime": 44.6687, + "eval_samples_per_second": 22.387, + "eval_steps_per_second": 0.94, + "step": 430 + }, + { + "epoch": 0.4723287671232877, + "grad_norm": 1.9950047731399536, + "learning_rate": 4.709585702680748e-06, + "loss": 2.2072, + "step": 431 + }, + { + "epoch": 0.47342465753424656, + "grad_norm": 1.9222451448440552, + "learning_rate": 4.707554833468725e-06, + "loss": 2.2654, + "step": 432 + }, + { + "epoch": 0.4745205479452055, + "grad_norm": 1.9735718965530396, + "learning_rate": 4.705523964256703e-06, + "loss": 2.3037, + "step": 433 + }, + { + "epoch": 0.4756164383561644, + "grad_norm": 2.3209776878356934, + "learning_rate": 4.703493095044679e-06, + "loss": 2.2083, + "step": 434 + }, + { + "epoch": 0.4767123287671233, + "grad_norm": 2.0993995666503906, + "learning_rate": 4.701462225832657e-06, + "loss": 2.2275, + "step": 435 + }, + { + "epoch": 0.4778082191780822, + "grad_norm": 2.334463119506836, + "learning_rate": 4.699431356620634e-06, + "loss": 2.1852, + "step": 436 + }, + { + "epoch": 0.4789041095890411, + "grad_norm": 1.9098213911056519, + "learning_rate": 4.697400487408612e-06, + "loss": 2.2829, + "step": 437 + }, + { + "epoch": 0.48, + "grad_norm": 2.3005056381225586, + "learning_rate": 4.695369618196588e-06, + "loss": 2.2532, + "step": 438 + }, + { + "epoch": 0.4810958904109589, + "grad_norm": 1.9069781303405762, + "learning_rate": 4.693338748984566e-06, + "loss": 2.2374, + "step": 439 + }, + { + "epoch": 0.4821917808219178, + "grad_norm": 2.050032138824463, + "learning_rate": 4.691307879772543e-06, + "loss": 2.2256, + "step": 440 + }, + { + "epoch": 0.4821917808219178, + "eval_loss": 2.3587422370910645, + "eval_runtime": 45.0941, + "eval_samples_per_second": 22.176, + "eval_steps_per_second": 0.931, + "step": 440 + }, + { + "epoch": 0.4832876712328767, + "grad_norm": 2.0563836097717285, + "learning_rate": 4.68927701056052e-06, + "loss": 2.1918, + "step": 441 + }, + { + "epoch": 0.4843835616438356, + "grad_norm": 2.1598386764526367, + "learning_rate": 4.6872461413484975e-06, + "loss": 2.2618, + "step": 442 + }, + { + "epoch": 0.48547945205479454, + "grad_norm": 1.9236539602279663, + "learning_rate": 4.685215272136475e-06, + "loss": 2.2477, + "step": 443 + }, + { + "epoch": 0.4865753424657534, + "grad_norm": 1.9923924207687378, + "learning_rate": 4.683184402924452e-06, + "loss": 2.2313, + "step": 444 + }, + { + "epoch": 0.4876712328767123, + "grad_norm": 2.162627696990967, + "learning_rate": 4.681153533712429e-06, + "loss": 2.2367, + "step": 445 + }, + { + "epoch": 0.48876712328767125, + "grad_norm": 2.083523988723755, + "learning_rate": 4.679122664500407e-06, + "loss": 2.2946, + "step": 446 + }, + { + "epoch": 0.4898630136986301, + "grad_norm": 2.504408359527588, + "learning_rate": 4.677091795288384e-06, + "loss": 2.2104, + "step": 447 + }, + { + "epoch": 0.49095890410958903, + "grad_norm": 1.8316532373428345, + "learning_rate": 4.675060926076361e-06, + "loss": 2.2027, + "step": 448 + }, + { + "epoch": 0.49205479452054796, + "grad_norm": 2.3337953090667725, + "learning_rate": 4.673030056864338e-06, + "loss": 2.278, + "step": 449 + }, + { + "epoch": 0.4931506849315068, + "grad_norm": 2.24102783203125, + "learning_rate": 4.670999187652316e-06, + "loss": 2.2018, + "step": 450 + }, + { + "epoch": 0.4931506849315068, + "eval_loss": 2.3545279502868652, + "eval_runtime": 45.1041, + "eval_samples_per_second": 22.171, + "eval_steps_per_second": 0.931, + "step": 450 + }, + { + "epoch": 0.49424657534246574, + "grad_norm": 2.07834792137146, + "learning_rate": 4.668968318440293e-06, + "loss": 2.1975, + "step": 451 + }, + { + "epoch": 0.49534246575342467, + "grad_norm": 1.7681723833084106, + "learning_rate": 4.66693744922827e-06, + "loss": 2.2033, + "step": 452 + }, + { + "epoch": 0.4964383561643836, + "grad_norm": 2.062134027481079, + "learning_rate": 4.6649065800162475e-06, + "loss": 2.2646, + "step": 453 + }, + { + "epoch": 0.49753424657534245, + "grad_norm": 1.94647216796875, + "learning_rate": 4.662875710804225e-06, + "loss": 2.1748, + "step": 454 + }, + { + "epoch": 0.4986301369863014, + "grad_norm": 1.9173872470855713, + "learning_rate": 4.660844841592202e-06, + "loss": 2.2435, + "step": 455 + }, + { + "epoch": 0.4997260273972603, + "grad_norm": 1.88880455493927, + "learning_rate": 4.658813972380179e-06, + "loss": 2.1711, + "step": 456 + }, + { + "epoch": 0.5008219178082192, + "grad_norm": 1.9628506898880005, + "learning_rate": 4.656783103168157e-06, + "loss": 2.2701, + "step": 457 + }, + { + "epoch": 0.5019178082191781, + "grad_norm": 2.0515286922454834, + "learning_rate": 4.654752233956134e-06, + "loss": 2.1656, + "step": 458 + }, + { + "epoch": 0.503013698630137, + "grad_norm": 1.978522539138794, + "learning_rate": 4.652721364744111e-06, + "loss": 2.246, + "step": 459 + }, + { + "epoch": 0.5041095890410959, + "grad_norm": 1.7993966341018677, + "learning_rate": 4.6506904955320885e-06, + "loss": 2.214, + "step": 460 + }, + { + "epoch": 0.5041095890410959, + "eval_loss": 2.3568837642669678, + "eval_runtime": 45.2086, + "eval_samples_per_second": 22.12, + "eval_steps_per_second": 0.929, + "step": 460 + }, + { + "epoch": 0.5052054794520547, + "grad_norm": 1.7809834480285645, + "learning_rate": 4.648659626320066e-06, + "loss": 2.2525, + "step": 461 + }, + { + "epoch": 0.5063013698630137, + "grad_norm": 1.9256166219711304, + "learning_rate": 4.646628757108043e-06, + "loss": 2.2611, + "step": 462 + }, + { + "epoch": 0.5073972602739726, + "grad_norm": 1.94192636013031, + "learning_rate": 4.6445978878960194e-06, + "loss": 2.2282, + "step": 463 + }, + { + "epoch": 0.5084931506849315, + "grad_norm": 1.8891818523406982, + "learning_rate": 4.6425670186839976e-06, + "loss": 2.2036, + "step": 464 + }, + { + "epoch": 0.5095890410958904, + "grad_norm": 2.352078676223755, + "learning_rate": 4.640536149471974e-06, + "loss": 2.2293, + "step": 465 + }, + { + "epoch": 0.5106849315068493, + "grad_norm": 1.8141659498214722, + "learning_rate": 4.638505280259952e-06, + "loss": 2.2551, + "step": 466 + }, + { + "epoch": 0.5117808219178083, + "grad_norm": 1.9492480754852295, + "learning_rate": 4.6364744110479285e-06, + "loss": 2.2641, + "step": 467 + }, + { + "epoch": 0.5128767123287671, + "grad_norm": 2.0210182666778564, + "learning_rate": 4.634443541835907e-06, + "loss": 2.2367, + "step": 468 + }, + { + "epoch": 0.513972602739726, + "grad_norm": 1.8936505317687988, + "learning_rate": 4.632412672623883e-06, + "loss": 2.265, + "step": 469 + }, + { + "epoch": 0.5150684931506849, + "grad_norm": 1.845888614654541, + "learning_rate": 4.630381803411861e-06, + "loss": 2.2169, + "step": 470 + }, + { + "epoch": 0.5150684931506849, + "eval_loss": 2.353489875793457, + "eval_runtime": 46.8739, + "eval_samples_per_second": 21.334, + "eval_steps_per_second": 0.896, + "step": 470 + }, + { + "epoch": 0.5161643835616438, + "grad_norm": 1.889675259590149, + "learning_rate": 4.628350934199838e-06, + "loss": 2.1383, + "step": 471 + }, + { + "epoch": 0.5172602739726028, + "grad_norm": 1.9799216985702515, + "learning_rate": 4.626320064987816e-06, + "loss": 2.2312, + "step": 472 + }, + { + "epoch": 0.5183561643835617, + "grad_norm": 2.0128607749938965, + "learning_rate": 4.624289195775792e-06, + "loss": 2.2529, + "step": 473 + }, + { + "epoch": 0.5194520547945205, + "grad_norm": 1.8791241645812988, + "learning_rate": 4.62225832656377e-06, + "loss": 2.226, + "step": 474 + }, + { + "epoch": 0.5205479452054794, + "grad_norm": 2.0497279167175293, + "learning_rate": 4.620227457351747e-06, + "loss": 2.2283, + "step": 475 + }, + { + "epoch": 0.5216438356164383, + "grad_norm": 1.8097424507141113, + "learning_rate": 4.618196588139724e-06, + "loss": 2.222, + "step": 476 + }, + { + "epoch": 0.5227397260273973, + "grad_norm": 1.8960500955581665, + "learning_rate": 4.616165718927701e-06, + "loss": 2.2342, + "step": 477 + }, + { + "epoch": 0.5238356164383562, + "grad_norm": 2.020730495452881, + "learning_rate": 4.6141348497156786e-06, + "loss": 2.2106, + "step": 478 + }, + { + "epoch": 0.5249315068493151, + "grad_norm": 1.8955297470092773, + "learning_rate": 4.612103980503656e-06, + "loss": 2.2755, + "step": 479 + }, + { + "epoch": 0.5260273972602739, + "grad_norm": 1.9767351150512695, + "learning_rate": 4.610073111291633e-06, + "loss": 2.2554, + "step": 480 + }, + { + "epoch": 0.5260273972602739, + "eval_loss": 2.347705841064453, + "eval_runtime": 46.9125, + "eval_samples_per_second": 21.316, + "eval_steps_per_second": 0.895, + "step": 480 + }, + { + "epoch": 0.5271232876712328, + "grad_norm": 1.936457633972168, + "learning_rate": 4.60804224207961e-06, + "loss": 2.2323, + "step": 481 + }, + { + "epoch": 0.5282191780821918, + "grad_norm": 1.9271862506866455, + "learning_rate": 4.606011372867588e-06, + "loss": 2.2151, + "step": 482 + }, + { + "epoch": 0.5293150684931507, + "grad_norm": 1.878153681755066, + "learning_rate": 4.603980503655565e-06, + "loss": 2.2322, + "step": 483 + }, + { + "epoch": 0.5304109589041096, + "grad_norm": 2.1009249687194824, + "learning_rate": 4.601949634443542e-06, + "loss": 2.2181, + "step": 484 + }, + { + "epoch": 0.5315068493150685, + "grad_norm": 1.9089181423187256, + "learning_rate": 4.5999187652315195e-06, + "loss": 2.1712, + "step": 485 + }, + { + "epoch": 0.5326027397260275, + "grad_norm": 2.207282781600952, + "learning_rate": 4.597887896019497e-06, + "loss": 2.1847, + "step": 486 + }, + { + "epoch": 0.5336986301369863, + "grad_norm": 2.001584053039551, + "learning_rate": 4.595857026807474e-06, + "loss": 2.214, + "step": 487 + }, + { + "epoch": 0.5347945205479452, + "grad_norm": 1.966700553894043, + "learning_rate": 4.593826157595451e-06, + "loss": 2.2237, + "step": 488 + }, + { + "epoch": 0.5358904109589041, + "grad_norm": 1.842469334602356, + "learning_rate": 4.591795288383429e-06, + "loss": 2.2089, + "step": 489 + }, + { + "epoch": 0.536986301369863, + "grad_norm": 2.0412986278533936, + "learning_rate": 4.589764419171406e-06, + "loss": 2.2734, + "step": 490 + }, + { + "epoch": 0.536986301369863, + "eval_loss": 2.3460772037506104, + "eval_runtime": 50.4048, + "eval_samples_per_second": 19.839, + "eval_steps_per_second": 0.833, + "step": 490 + }, + { + "epoch": 0.538082191780822, + "grad_norm": 1.9337031841278076, + "learning_rate": 4.587733549959383e-06, + "loss": 2.2314, + "step": 491 + }, + { + "epoch": 0.5391780821917809, + "grad_norm": 2.4448320865631104, + "learning_rate": 4.5857026807473604e-06, + "loss": 2.2098, + "step": 492 + }, + { + "epoch": 0.5402739726027397, + "grad_norm": 1.83162522315979, + "learning_rate": 4.583671811535338e-06, + "loss": 2.1902, + "step": 493 + }, + { + "epoch": 0.5413698630136986, + "grad_norm": 2.3614656925201416, + "learning_rate": 4.581640942323315e-06, + "loss": 2.1737, + "step": 494 + }, + { + "epoch": 0.5424657534246575, + "grad_norm": 2.001187801361084, + "learning_rate": 4.579610073111292e-06, + "loss": 2.2485, + "step": 495 + }, + { + "epoch": 0.5435616438356164, + "grad_norm": 2.3479933738708496, + "learning_rate": 4.5775792038992695e-06, + "loss": 2.2595, + "step": 496 + }, + { + "epoch": 0.5446575342465754, + "grad_norm": 2.135856866836548, + "learning_rate": 4.575548334687247e-06, + "loss": 2.2528, + "step": 497 + }, + { + "epoch": 0.5457534246575343, + "grad_norm": 1.983126163482666, + "learning_rate": 4.573517465475224e-06, + "loss": 2.2788, + "step": 498 + }, + { + "epoch": 0.5468493150684931, + "grad_norm": 2.18208909034729, + "learning_rate": 4.571486596263201e-06, + "loss": 2.1995, + "step": 499 + }, + { + "epoch": 0.547945205479452, + "grad_norm": 1.848747968673706, + "learning_rate": 4.569455727051178e-06, + "loss": 2.2403, + "step": 500 + }, + { + "epoch": 0.547945205479452, + "eval_loss": 2.344066619873047, + "eval_runtime": 51.1213, + "eval_samples_per_second": 19.561, + "eval_steps_per_second": 0.822, + "step": 500 + }, + { + "epoch": 0.549041095890411, + "grad_norm": 2.427860736846924, + "learning_rate": 4.567424857839156e-06, + "loss": 2.246, + "step": 501 + }, + { + "epoch": 0.5501369863013699, + "grad_norm": 1.913405418395996, + "learning_rate": 4.565393988627132e-06, + "loss": 2.2296, + "step": 502 + }, + { + "epoch": 0.5512328767123288, + "grad_norm": 2.0106029510498047, + "learning_rate": 4.5633631194151105e-06, + "loss": 2.2657, + "step": 503 + }, + { + "epoch": 0.5523287671232877, + "grad_norm": 2.2719595432281494, + "learning_rate": 4.561332250203087e-06, + "loss": 2.2341, + "step": 504 + }, + { + "epoch": 0.5534246575342465, + "grad_norm": 2.109786033630371, + "learning_rate": 4.559301380991065e-06, + "loss": 2.2181, + "step": 505 + }, + { + "epoch": 0.5545205479452054, + "grad_norm": 2.189779281616211, + "learning_rate": 4.5572705117790414e-06, + "loss": 2.1821, + "step": 506 + }, + { + "epoch": 0.5556164383561644, + "grad_norm": 2.0263590812683105, + "learning_rate": 4.5552396425670196e-06, + "loss": 2.2072, + "step": 507 + }, + { + "epoch": 0.5567123287671233, + "grad_norm": 2.215289354324341, + "learning_rate": 4.553208773354996e-06, + "loss": 2.2513, + "step": 508 + }, + { + "epoch": 0.5578082191780822, + "grad_norm": 2.3031351566314697, + "learning_rate": 4.551177904142974e-06, + "loss": 2.1988, + "step": 509 + }, + { + "epoch": 0.5589041095890411, + "grad_norm": 2.1830058097839355, + "learning_rate": 4.5491470349309505e-06, + "loss": 2.269, + "step": 510 + }, + { + "epoch": 0.5589041095890411, + "eval_loss": 2.342463731765747, + "eval_runtime": 43.1956, + "eval_samples_per_second": 23.151, + "eval_steps_per_second": 0.972, + "step": 510 + }, + { + "epoch": 0.56, + "grad_norm": 2.3210105895996094, + "learning_rate": 4.547116165718929e-06, + "loss": 2.2051, + "step": 511 + }, + { + "epoch": 0.5610958904109589, + "grad_norm": 1.9175361394882202, + "learning_rate": 4.545085296506905e-06, + "loss": 2.2304, + "step": 512 + }, + { + "epoch": 0.5621917808219178, + "grad_norm": 2.302947759628296, + "learning_rate": 4.543054427294882e-06, + "loss": 2.215, + "step": 513 + }, + { + "epoch": 0.5632876712328767, + "grad_norm": 1.8947396278381348, + "learning_rate": 4.54102355808286e-06, + "loss": 2.1926, + "step": 514 + }, + { + "epoch": 0.5643835616438356, + "grad_norm": 2.195268154144287, + "learning_rate": 4.538992688870837e-06, + "loss": 2.1937, + "step": 515 + }, + { + "epoch": 0.5654794520547946, + "grad_norm": 2.178874969482422, + "learning_rate": 4.536961819658814e-06, + "loss": 2.2248, + "step": 516 + }, + { + "epoch": 0.5665753424657535, + "grad_norm": 1.9856929779052734, + "learning_rate": 4.5349309504467915e-06, + "loss": 2.2077, + "step": 517 + }, + { + "epoch": 0.5676712328767123, + "grad_norm": 2.0686943531036377, + "learning_rate": 4.532900081234769e-06, + "loss": 2.2328, + "step": 518 + }, + { + "epoch": 0.5687671232876712, + "grad_norm": 1.9554919004440308, + "learning_rate": 4.530869212022746e-06, + "loss": 2.198, + "step": 519 + }, + { + "epoch": 0.5698630136986301, + "grad_norm": 1.9489670991897583, + "learning_rate": 4.528838342810723e-06, + "loss": 2.2287, + "step": 520 + }, + { + "epoch": 0.5698630136986301, + "eval_loss": 2.3415298461914062, + "eval_runtime": 48.7918, + "eval_samples_per_second": 20.495, + "eval_steps_per_second": 0.861, + "step": 520 + }, + { + "epoch": 0.570958904109589, + "grad_norm": 1.9868658781051636, + "learning_rate": 4.526807473598701e-06, + "loss": 2.2205, + "step": 521 + }, + { + "epoch": 0.572054794520548, + "grad_norm": 2.0965914726257324, + "learning_rate": 4.524776604386678e-06, + "loss": 2.2239, + "step": 522 + }, + { + "epoch": 0.5731506849315069, + "grad_norm": 2.041471481323242, + "learning_rate": 4.522745735174655e-06, + "loss": 2.2035, + "step": 523 + }, + { + "epoch": 0.5742465753424657, + "grad_norm": 1.885293960571289, + "learning_rate": 4.520714865962632e-06, + "loss": 2.2469, + "step": 524 + }, + { + "epoch": 0.5753424657534246, + "grad_norm": 1.9107296466827393, + "learning_rate": 4.51868399675061e-06, + "loss": 2.2386, + "step": 525 + }, + { + "epoch": 0.5764383561643835, + "grad_norm": 1.9945271015167236, + "learning_rate": 4.516653127538587e-06, + "loss": 2.235, + "step": 526 + }, + { + "epoch": 0.5775342465753425, + "grad_norm": 1.9787253141403198, + "learning_rate": 4.514622258326564e-06, + "loss": 2.2034, + "step": 527 + }, + { + "epoch": 0.5786301369863014, + "grad_norm": 2.0300636291503906, + "learning_rate": 4.5125913891145415e-06, + "loss": 2.1849, + "step": 528 + }, + { + "epoch": 0.5797260273972603, + "grad_norm": 2.0348970890045166, + "learning_rate": 4.510560519902519e-06, + "loss": 2.2174, + "step": 529 + }, + { + "epoch": 0.5808219178082191, + "grad_norm": 1.8589537143707275, + "learning_rate": 4.508529650690496e-06, + "loss": 2.2351, + "step": 530 + }, + { + "epoch": 0.5808219178082191, + "eval_loss": 2.3450186252593994, + "eval_runtime": 52.4548, + "eval_samples_per_second": 19.064, + "eval_steps_per_second": 0.801, + "step": 530 + }, + { + "epoch": 0.581917808219178, + "grad_norm": 2.0496020317077637, + "learning_rate": 4.506498781478473e-06, + "loss": 2.182, + "step": 531 + }, + { + "epoch": 0.583013698630137, + "grad_norm": 2.0649805068969727, + "learning_rate": 4.504467912266451e-06, + "loss": 2.1654, + "step": 532 + }, + { + "epoch": 0.5841095890410959, + "grad_norm": 1.9809486865997314, + "learning_rate": 4.502437043054428e-06, + "loss": 2.2182, + "step": 533 + }, + { + "epoch": 0.5852054794520548, + "grad_norm": 2.21915340423584, + "learning_rate": 4.500406173842405e-06, + "loss": 2.2697, + "step": 534 + }, + { + "epoch": 0.5863013698630137, + "grad_norm": 1.9822649955749512, + "learning_rate": 4.4983753046303824e-06, + "loss": 2.2297, + "step": 535 + }, + { + "epoch": 0.5873972602739727, + "grad_norm": 1.972575306892395, + "learning_rate": 4.49634443541836e-06, + "loss": 2.2164, + "step": 536 + }, + { + "epoch": 0.5884931506849315, + "grad_norm": 1.948667049407959, + "learning_rate": 4.494313566206336e-06, + "loss": 2.2087, + "step": 537 + }, + { + "epoch": 0.5895890410958904, + "grad_norm": 2.7486417293548584, + "learning_rate": 4.492282696994314e-06, + "loss": 2.2602, + "step": 538 + }, + { + "epoch": 0.5906849315068493, + "grad_norm": 2.3374552726745605, + "learning_rate": 4.490251827782291e-06, + "loss": 2.2054, + "step": 539 + }, + { + "epoch": 0.5917808219178082, + "grad_norm": 2.222116231918335, + "learning_rate": 4.488220958570269e-06, + "loss": 2.1868, + "step": 540 + }, + { + "epoch": 0.5917808219178082, + "eval_loss": 2.345618963241577, + "eval_runtime": 53.0302, + "eval_samples_per_second": 18.857, + "eval_steps_per_second": 0.792, + "step": 540 + }, + { + "epoch": 0.5928767123287672, + "grad_norm": 2.00752854347229, + "learning_rate": 4.486190089358245e-06, + "loss": 2.187, + "step": 541 + }, + { + "epoch": 0.5939726027397261, + "grad_norm": 2.2466132640838623, + "learning_rate": 4.484159220146223e-06, + "loss": 2.1611, + "step": 542 + }, + { + "epoch": 0.5950684931506849, + "grad_norm": 2.1279385089874268, + "learning_rate": 4.4821283509342e-06, + "loss": 2.248, + "step": 543 + }, + { + "epoch": 0.5961643835616438, + "grad_norm": 1.9140571355819702, + "learning_rate": 4.480097481722178e-06, + "loss": 2.2314, + "step": 544 + }, + { + "epoch": 0.5972602739726027, + "grad_norm": 2.120741605758667, + "learning_rate": 4.478066612510154e-06, + "loss": 2.212, + "step": 545 + }, + { + "epoch": 0.5983561643835617, + "grad_norm": 2.003713369369507, + "learning_rate": 4.4760357432981325e-06, + "loss": 2.1824, + "step": 546 + }, + { + "epoch": 0.5994520547945206, + "grad_norm": 1.9161221981048584, + "learning_rate": 4.474004874086109e-06, + "loss": 2.2015, + "step": 547 + }, + { + "epoch": 0.6005479452054795, + "grad_norm": 2.367110252380371, + "learning_rate": 4.471974004874087e-06, + "loss": 2.3122, + "step": 548 + }, + { + "epoch": 0.6016438356164383, + "grad_norm": 2.113380193710327, + "learning_rate": 4.4699431356620635e-06, + "loss": 2.1692, + "step": 549 + }, + { + "epoch": 0.6027397260273972, + "grad_norm": 1.974181890487671, + "learning_rate": 4.467912266450041e-06, + "loss": 2.2509, + "step": 550 + }, + { + "epoch": 0.6027397260273972, + "eval_loss": 2.342583656311035, + "eval_runtime": 50.7362, + "eval_samples_per_second": 19.71, + "eval_steps_per_second": 0.828, + "step": 550 + }, + { + "epoch": 0.6038356164383561, + "grad_norm": 1.889894962310791, + "learning_rate": 4.465881397238018e-06, + "loss": 2.2214, + "step": 551 + }, + { + "epoch": 0.6049315068493151, + "grad_norm": 1.8982343673706055, + "learning_rate": 4.463850528025995e-06, + "loss": 2.2135, + "step": 552 + }, + { + "epoch": 0.606027397260274, + "grad_norm": 1.9462087154388428, + "learning_rate": 4.4618196588139726e-06, + "loss": 2.2255, + "step": 553 + }, + { + "epoch": 0.6071232876712329, + "grad_norm": 1.8729931116104126, + "learning_rate": 4.45978878960195e-06, + "loss": 2.2358, + "step": 554 + }, + { + "epoch": 0.6082191780821918, + "grad_norm": 1.9230246543884277, + "learning_rate": 4.457757920389927e-06, + "loss": 2.222, + "step": 555 + }, + { + "epoch": 0.6093150684931506, + "grad_norm": 1.9495900869369507, + "learning_rate": 4.455727051177904e-06, + "loss": 2.1779, + "step": 556 + }, + { + "epoch": 0.6104109589041096, + "grad_norm": 1.983487606048584, + "learning_rate": 4.453696181965882e-06, + "loss": 2.2007, + "step": 557 + }, + { + "epoch": 0.6115068493150685, + "grad_norm": 2.1541049480438232, + "learning_rate": 4.451665312753859e-06, + "loss": 2.2513, + "step": 558 + }, + { + "epoch": 0.6126027397260274, + "grad_norm": 1.9133352041244507, + "learning_rate": 4.449634443541836e-06, + "loss": 2.2288, + "step": 559 + }, + { + "epoch": 0.6136986301369863, + "grad_norm": 2.3325741291046143, + "learning_rate": 4.4476035743298135e-06, + "loss": 2.2539, + "step": 560 + }, + { + "epoch": 0.6136986301369863, + "eval_loss": 2.3387348651885986, + "eval_runtime": 53.4301, + "eval_samples_per_second": 18.716, + "eval_steps_per_second": 0.786, + "step": 560 + }, + { + "epoch": 0.6147945205479453, + "grad_norm": 1.8099673986434937, + "learning_rate": 4.445572705117791e-06, + "loss": 2.1845, + "step": 561 + }, + { + "epoch": 0.6158904109589041, + "grad_norm": 2.201901912689209, + "learning_rate": 4.443541835905768e-06, + "loss": 2.195, + "step": 562 + }, + { + "epoch": 0.616986301369863, + "grad_norm": 1.8412714004516602, + "learning_rate": 4.441510966693745e-06, + "loss": 2.2551, + "step": 563 + }, + { + "epoch": 0.6180821917808219, + "grad_norm": 1.8761520385742188, + "learning_rate": 4.439480097481723e-06, + "loss": 2.1965, + "step": 564 + }, + { + "epoch": 0.6191780821917808, + "grad_norm": 1.9123467206954956, + "learning_rate": 4.4374492282697e-06, + "loss": 2.2502, + "step": 565 + }, + { + "epoch": 0.6202739726027398, + "grad_norm": 2.0631234645843506, + "learning_rate": 4.435418359057677e-06, + "loss": 2.1582, + "step": 566 + }, + { + "epoch": 0.6213698630136987, + "grad_norm": 2.3191616535186768, + "learning_rate": 4.433387489845654e-06, + "loss": 2.2271, + "step": 567 + }, + { + "epoch": 0.6224657534246575, + "grad_norm": 2.043057918548584, + "learning_rate": 4.431356620633632e-06, + "loss": 2.2292, + "step": 568 + }, + { + "epoch": 0.6235616438356164, + "grad_norm": 2.0716068744659424, + "learning_rate": 4.429325751421609e-06, + "loss": 2.196, + "step": 569 + }, + { + "epoch": 0.6246575342465753, + "grad_norm": 1.7669776678085327, + "learning_rate": 4.427294882209586e-06, + "loss": 2.1953, + "step": 570 + }, + { + "epoch": 0.6246575342465753, + "eval_loss": 2.336578607559204, + "eval_runtime": 50.0083, + "eval_samples_per_second": 19.997, + "eval_steps_per_second": 0.84, + "step": 570 + }, + { + "epoch": 0.6257534246575343, + "grad_norm": 1.8165771961212158, + "learning_rate": 4.4252640129975635e-06, + "loss": 2.2155, + "step": 571 + }, + { + "epoch": 0.6268493150684932, + "grad_norm": 1.8640249967575073, + "learning_rate": 4.423233143785541e-06, + "loss": 2.1648, + "step": 572 + }, + { + "epoch": 0.6279452054794521, + "grad_norm": 2.013869285583496, + "learning_rate": 4.421202274573518e-06, + "loss": 2.2131, + "step": 573 + }, + { + "epoch": 0.6290410958904109, + "grad_norm": 2.2650580406188965, + "learning_rate": 4.4191714053614945e-06, + "loss": 2.2287, + "step": 574 + }, + { + "epoch": 0.6301369863013698, + "grad_norm": 1.9232103824615479, + "learning_rate": 4.417140536149473e-06, + "loss": 2.1339, + "step": 575 + }, + { + "epoch": 0.6312328767123287, + "grad_norm": 1.910134196281433, + "learning_rate": 4.415109666937449e-06, + "loss": 2.1755, + "step": 576 + }, + { + "epoch": 0.6323287671232877, + "grad_norm": 2.2492005825042725, + "learning_rate": 4.413078797725427e-06, + "loss": 2.2362, + "step": 577 + }, + { + "epoch": 0.6334246575342466, + "grad_norm": 1.8584033250808716, + "learning_rate": 4.411047928513404e-06, + "loss": 2.1911, + "step": 578 + }, + { + "epoch": 0.6345205479452055, + "grad_norm": 2.156273126602173, + "learning_rate": 4.409017059301382e-06, + "loss": 2.2109, + "step": 579 + }, + { + "epoch": 0.6356164383561644, + "grad_norm": 1.7977535724639893, + "learning_rate": 4.406986190089358e-06, + "loss": 2.2195, + "step": 580 + }, + { + "epoch": 0.6356164383561644, + "eval_loss": 2.33555269241333, + "eval_runtime": 48.3919, + "eval_samples_per_second": 20.665, + "eval_steps_per_second": 0.868, + "step": 580 + }, + { + "epoch": 0.6367123287671232, + "grad_norm": 1.8262702226638794, + "learning_rate": 4.404955320877336e-06, + "loss": 2.2345, + "step": 581 + }, + { + "epoch": 0.6378082191780822, + "grad_norm": 1.9389383792877197, + "learning_rate": 4.402924451665313e-06, + "loss": 2.2347, + "step": 582 + }, + { + "epoch": 0.6389041095890411, + "grad_norm": 1.9548988342285156, + "learning_rate": 4.400893582453291e-06, + "loss": 2.1967, + "step": 583 + }, + { + "epoch": 0.64, + "grad_norm": 1.898092269897461, + "learning_rate": 4.398862713241267e-06, + "loss": 2.1795, + "step": 584 + }, + { + "epoch": 0.6410958904109589, + "grad_norm": 1.8183610439300537, + "learning_rate": 4.396831844029245e-06, + "loss": 2.2031, + "step": 585 + }, + { + "epoch": 0.6421917808219179, + "grad_norm": 2.1520016193389893, + "learning_rate": 4.394800974817222e-06, + "loss": 2.2051, + "step": 586 + }, + { + "epoch": 0.6432876712328767, + "grad_norm": 1.9930012226104736, + "learning_rate": 4.392770105605199e-06, + "loss": 2.2053, + "step": 587 + }, + { + "epoch": 0.6443835616438356, + "grad_norm": 2.311450958251953, + "learning_rate": 4.390739236393176e-06, + "loss": 2.2252, + "step": 588 + }, + { + "epoch": 0.6454794520547945, + "grad_norm": 2.0168697834014893, + "learning_rate": 4.388708367181154e-06, + "loss": 2.2286, + "step": 589 + }, + { + "epoch": 0.6465753424657534, + "grad_norm": 2.055964469909668, + "learning_rate": 4.386677497969131e-06, + "loss": 2.2278, + "step": 590 + }, + { + "epoch": 0.6465753424657534, + "eval_loss": 2.334367275238037, + "eval_runtime": 48.9414, + "eval_samples_per_second": 20.433, + "eval_steps_per_second": 0.858, + "step": 590 + }, + { + "epoch": 0.6476712328767124, + "grad_norm": 2.0161590576171875, + "learning_rate": 4.384646628757108e-06, + "loss": 2.2102, + "step": 591 + }, + { + "epoch": 0.6487671232876713, + "grad_norm": 2.0827667713165283, + "learning_rate": 4.3826157595450855e-06, + "loss": 2.2531, + "step": 592 + }, + { + "epoch": 0.6498630136986301, + "grad_norm": 2.077637195587158, + "learning_rate": 4.380584890333063e-06, + "loss": 2.2091, + "step": 593 + }, + { + "epoch": 0.650958904109589, + "grad_norm": 2.023892879486084, + "learning_rate": 4.37855402112104e-06, + "loss": 2.1684, + "step": 594 + }, + { + "epoch": 0.6520547945205479, + "grad_norm": 1.8509857654571533, + "learning_rate": 4.376523151909017e-06, + "loss": 2.1439, + "step": 595 + }, + { + "epoch": 0.6531506849315069, + "grad_norm": 2.1649608612060547, + "learning_rate": 4.3744922826969946e-06, + "loss": 2.2557, + "step": 596 + }, + { + "epoch": 0.6542465753424658, + "grad_norm": 2.1351773738861084, + "learning_rate": 4.372461413484972e-06, + "loss": 2.177, + "step": 597 + }, + { + "epoch": 0.6553424657534247, + "grad_norm": 2.1713974475860596, + "learning_rate": 4.370430544272949e-06, + "loss": 2.2223, + "step": 598 + }, + { + "epoch": 0.6564383561643835, + "grad_norm": 2.2905397415161133, + "learning_rate": 4.368399675060926e-06, + "loss": 2.2373, + "step": 599 + }, + { + "epoch": 0.6575342465753424, + "grad_norm": 2.007160186767578, + "learning_rate": 4.366368805848904e-06, + "loss": 2.2355, + "step": 600 + }, + { + "epoch": 0.6575342465753424, + "eval_loss": 2.3322324752807617, + "eval_runtime": 46.2769, + "eval_samples_per_second": 21.609, + "eval_steps_per_second": 0.908, + "step": 600 + }, + { + "epoch": 0.6586301369863014, + "grad_norm": 2.309783697128296, + "learning_rate": 4.364337936636881e-06, + "loss": 2.2052, + "step": 601 + }, + { + "epoch": 0.6597260273972603, + "grad_norm": 2.498034954071045, + "learning_rate": 4.362307067424858e-06, + "loss": 2.1567, + "step": 602 + }, + { + "epoch": 0.6608219178082192, + "grad_norm": 1.8302545547485352, + "learning_rate": 4.3602761982128355e-06, + "loss": 2.2117, + "step": 603 + }, + { + "epoch": 0.6619178082191781, + "grad_norm": 2.226947546005249, + "learning_rate": 4.358245329000813e-06, + "loss": 2.1881, + "step": 604 + }, + { + "epoch": 0.663013698630137, + "grad_norm": 1.8484126329421997, + "learning_rate": 4.35621445978879e-06, + "loss": 2.1955, + "step": 605 + }, + { + "epoch": 0.6641095890410958, + "grad_norm": 1.9844266176223755, + "learning_rate": 4.354183590576767e-06, + "loss": 2.2063, + "step": 606 + }, + { + "epoch": 0.6652054794520548, + "grad_norm": 1.9292365312576294, + "learning_rate": 4.352152721364745e-06, + "loss": 2.205, + "step": 607 + }, + { + "epoch": 0.6663013698630137, + "grad_norm": 1.980895757675171, + "learning_rate": 4.350121852152722e-06, + "loss": 2.1727, + "step": 608 + }, + { + "epoch": 0.6673972602739726, + "grad_norm": 1.8331713676452637, + "learning_rate": 4.348090982940699e-06, + "loss": 2.226, + "step": 609 + }, + { + "epoch": 0.6684931506849315, + "grad_norm": 1.9666537046432495, + "learning_rate": 4.3460601137286764e-06, + "loss": 2.2198, + "step": 610 + }, + { + "epoch": 0.6684931506849315, + "eval_loss": 2.3297338485717773, + "eval_runtime": 42.6274, + "eval_samples_per_second": 23.459, + "eval_steps_per_second": 0.985, + "step": 610 + }, + { + "epoch": 0.6695890410958905, + "grad_norm": 1.9741110801696777, + "learning_rate": 4.344029244516654e-06, + "loss": 2.2132, + "step": 611 + }, + { + "epoch": 0.6706849315068493, + "grad_norm": 2.0747334957122803, + "learning_rate": 4.341998375304631e-06, + "loss": 2.2081, + "step": 612 + }, + { + "epoch": 0.6717808219178082, + "grad_norm": 2.031503915786743, + "learning_rate": 4.339967506092607e-06, + "loss": 2.2642, + "step": 613 + }, + { + "epoch": 0.6728767123287671, + "grad_norm": 1.8513256311416626, + "learning_rate": 4.3379366368805855e-06, + "loss": 2.2141, + "step": 614 + }, + { + "epoch": 0.673972602739726, + "grad_norm": 1.9098315238952637, + "learning_rate": 4.335905767668562e-06, + "loss": 2.2064, + "step": 615 + }, + { + "epoch": 0.675068493150685, + "grad_norm": 1.9704065322875977, + "learning_rate": 4.33387489845654e-06, + "loss": 2.1997, + "step": 616 + }, + { + "epoch": 0.6761643835616439, + "grad_norm": 2.1520018577575684, + "learning_rate": 4.3318440292445165e-06, + "loss": 2.2278, + "step": 617 + }, + { + "epoch": 0.6772602739726027, + "grad_norm": 1.898164987564087, + "learning_rate": 4.329813160032495e-06, + "loss": 2.2147, + "step": 618 + }, + { + "epoch": 0.6783561643835616, + "grad_norm": 1.8955962657928467, + "learning_rate": 4.327782290820471e-06, + "loss": 2.1825, + "step": 619 + }, + { + "epoch": 0.6794520547945205, + "grad_norm": 1.7579313516616821, + "learning_rate": 4.325751421608449e-06, + "loss": 2.1619, + "step": 620 + }, + { + "epoch": 0.6794520547945205, + "eval_loss": 2.329692840576172, + "eval_runtime": 46.3187, + "eval_samples_per_second": 21.59, + "eval_steps_per_second": 0.907, + "step": 620 + }, + { + "epoch": 0.6805479452054795, + "grad_norm": 1.8716063499450684, + "learning_rate": 4.323720552396426e-06, + "loss": 2.2289, + "step": 621 + }, + { + "epoch": 0.6816438356164384, + "grad_norm": 1.8798290491104126, + "learning_rate": 4.321689683184404e-06, + "loss": 2.1944, + "step": 622 + }, + { + "epoch": 0.6827397260273973, + "grad_norm": 1.8438303470611572, + "learning_rate": 4.31965881397238e-06, + "loss": 2.1879, + "step": 623 + }, + { + "epoch": 0.6838356164383562, + "grad_norm": 1.7844786643981934, + "learning_rate": 4.3176279447603574e-06, + "loss": 2.1936, + "step": 624 + }, + { + "epoch": 0.684931506849315, + "grad_norm": 1.955529808998108, + "learning_rate": 4.315597075548335e-06, + "loss": 2.2258, + "step": 625 + }, + { + "epoch": 0.686027397260274, + "grad_norm": 2.3826634883880615, + "learning_rate": 4.313566206336312e-06, + "loss": 2.2135, + "step": 626 + }, + { + "epoch": 0.6871232876712329, + "grad_norm": 1.8453123569488525, + "learning_rate": 4.311535337124289e-06, + "loss": 2.2362, + "step": 627 + }, + { + "epoch": 0.6882191780821918, + "grad_norm": 2.742682933807373, + "learning_rate": 4.3095044679122665e-06, + "loss": 2.194, + "step": 628 + }, + { + "epoch": 0.6893150684931507, + "grad_norm": 2.0299901962280273, + "learning_rate": 4.307473598700244e-06, + "loss": 2.2029, + "step": 629 + }, + { + "epoch": 0.6904109589041096, + "grad_norm": 2.29683780670166, + "learning_rate": 4.305442729488221e-06, + "loss": 2.2141, + "step": 630 + }, + { + "epoch": 0.6904109589041096, + "eval_loss": 2.329368829727173, + "eval_runtime": 45.1947, + "eval_samples_per_second": 22.126, + "eval_steps_per_second": 0.929, + "step": 630 + }, + { + "epoch": 0.6915068493150685, + "grad_norm": 2.208280086517334, + "learning_rate": 4.303411860276198e-06, + "loss": 2.1637, + "step": 631 + }, + { + "epoch": 0.6926027397260274, + "grad_norm": 1.943514108657837, + "learning_rate": 4.301380991064176e-06, + "loss": 2.234, + "step": 632 + }, + { + "epoch": 0.6936986301369863, + "grad_norm": 2.486440420150757, + "learning_rate": 4.299350121852153e-06, + "loss": 2.2239, + "step": 633 + }, + { + "epoch": 0.6947945205479452, + "grad_norm": 1.7243690490722656, + "learning_rate": 4.29731925264013e-06, + "loss": 2.2428, + "step": 634 + }, + { + "epoch": 0.6958904109589041, + "grad_norm": 2.31227445602417, + "learning_rate": 4.2952883834281075e-06, + "loss": 2.2403, + "step": 635 + }, + { + "epoch": 0.6969863013698631, + "grad_norm": 1.7361263036727905, + "learning_rate": 4.293257514216085e-06, + "loss": 2.1975, + "step": 636 + }, + { + "epoch": 0.6980821917808219, + "grad_norm": 2.010831356048584, + "learning_rate": 4.291226645004062e-06, + "loss": 2.1633, + "step": 637 + }, + { + "epoch": 0.6991780821917808, + "grad_norm": 2.2796833515167236, + "learning_rate": 4.289195775792039e-06, + "loss": 2.2307, + "step": 638 + }, + { + "epoch": 0.7002739726027397, + "grad_norm": 2.1069483757019043, + "learning_rate": 4.287164906580017e-06, + "loss": 2.2389, + "step": 639 + }, + { + "epoch": 0.7013698630136986, + "grad_norm": 2.19460129737854, + "learning_rate": 4.285134037367994e-06, + "loss": 2.1992, + "step": 640 + }, + { + "epoch": 0.7013698630136986, + "eval_loss": 2.3223369121551514, + "eval_runtime": 43.1093, + "eval_samples_per_second": 23.197, + "eval_steps_per_second": 0.974, + "step": 640 + }, + { + "epoch": 0.7024657534246576, + "grad_norm": 1.875261902809143, + "learning_rate": 4.283103168155971e-06, + "loss": 2.1767, + "step": 641 + }, + { + "epoch": 0.7035616438356165, + "grad_norm": 1.906762957572937, + "learning_rate": 4.281072298943948e-06, + "loss": 2.1734, + "step": 642 + }, + { + "epoch": 0.7046575342465753, + "grad_norm": 2.1114180088043213, + "learning_rate": 4.279041429731926e-06, + "loss": 2.1734, + "step": 643 + }, + { + "epoch": 0.7057534246575342, + "grad_norm": 1.844373106956482, + "learning_rate": 4.277010560519903e-06, + "loss": 2.1967, + "step": 644 + }, + { + "epoch": 0.7068493150684931, + "grad_norm": 2.3463754653930664, + "learning_rate": 4.27497969130788e-06, + "loss": 2.1735, + "step": 645 + }, + { + "epoch": 0.707945205479452, + "grad_norm": 2.3530960083007812, + "learning_rate": 4.2729488220958575e-06, + "loss": 2.189, + "step": 646 + }, + { + "epoch": 0.709041095890411, + "grad_norm": 2.110963821411133, + "learning_rate": 4.270917952883835e-06, + "loss": 2.218, + "step": 647 + }, + { + "epoch": 0.7101369863013699, + "grad_norm": 2.625005006790161, + "learning_rate": 4.268887083671812e-06, + "loss": 2.2235, + "step": 648 + }, + { + "epoch": 0.7112328767123288, + "grad_norm": 1.8411654233932495, + "learning_rate": 4.266856214459789e-06, + "loss": 2.1757, + "step": 649 + }, + { + "epoch": 0.7123287671232876, + "grad_norm": 2.0817489624023438, + "learning_rate": 4.264825345247766e-06, + "loss": 2.2698, + "step": 650 + }, + { + "epoch": 0.7123287671232876, + "eval_loss": 2.322544574737549, + "eval_runtime": 43.0771, + "eval_samples_per_second": 23.214, + "eval_steps_per_second": 0.975, + "step": 650 + }, + { + "epoch": 0.7134246575342466, + "grad_norm": 1.8953847885131836, + "learning_rate": 4.262794476035744e-06, + "loss": 2.1745, + "step": 651 + }, + { + "epoch": 0.7145205479452055, + "grad_norm": 1.8463294506072998, + "learning_rate": 4.26076360682372e-06, + "loss": 2.2042, + "step": 652 + }, + { + "epoch": 0.7156164383561644, + "grad_norm": 1.8776047229766846, + "learning_rate": 4.2587327376116984e-06, + "loss": 2.1512, + "step": 653 + }, + { + "epoch": 0.7167123287671233, + "grad_norm": 1.8949921131134033, + "learning_rate": 4.256701868399675e-06, + "loss": 2.1895, + "step": 654 + }, + { + "epoch": 0.7178082191780822, + "grad_norm": 1.8172632455825806, + "learning_rate": 4.254670999187653e-06, + "loss": 2.2, + "step": 655 + }, + { + "epoch": 0.718904109589041, + "grad_norm": 2.105499505996704, + "learning_rate": 4.252640129975629e-06, + "loss": 2.202, + "step": 656 + }, + { + "epoch": 0.72, + "grad_norm": 1.9515924453735352, + "learning_rate": 4.2506092607636075e-06, + "loss": 2.2116, + "step": 657 + }, + { + "epoch": 0.7210958904109589, + "grad_norm": 1.761770248413086, + "learning_rate": 4.248578391551584e-06, + "loss": 2.1921, + "step": 658 + }, + { + "epoch": 0.7221917808219178, + "grad_norm": 1.9848097562789917, + "learning_rate": 4.246547522339562e-06, + "loss": 2.1553, + "step": 659 + }, + { + "epoch": 0.7232876712328767, + "grad_norm": 2.041987180709839, + "learning_rate": 4.2445166531275385e-06, + "loss": 2.1975, + "step": 660 + }, + { + "epoch": 0.7232876712328767, + "eval_loss": 2.325580358505249, + "eval_runtime": 43.6207, + "eval_samples_per_second": 22.925, + "eval_steps_per_second": 0.963, + "step": 660 + }, + { + "epoch": 0.7243835616438357, + "grad_norm": 1.9755775928497314, + "learning_rate": 4.242485783915517e-06, + "loss": 2.1955, + "step": 661 + }, + { + "epoch": 0.7254794520547945, + "grad_norm": 1.9621502161026, + "learning_rate": 4.240454914703493e-06, + "loss": 2.1554, + "step": 662 + }, + { + "epoch": 0.7265753424657534, + "grad_norm": 1.848013162612915, + "learning_rate": 4.23842404549147e-06, + "loss": 2.16, + "step": 663 + }, + { + "epoch": 0.7276712328767123, + "grad_norm": 1.7370128631591797, + "learning_rate": 4.236393176279448e-06, + "loss": 2.1696, + "step": 664 + }, + { + "epoch": 0.7287671232876712, + "grad_norm": 2.0708158016204834, + "learning_rate": 4.234362307067425e-06, + "loss": 2.1592, + "step": 665 + }, + { + "epoch": 0.7298630136986302, + "grad_norm": 1.9864391088485718, + "learning_rate": 4.232331437855402e-06, + "loss": 2.2121, + "step": 666 + }, + { + "epoch": 0.7309589041095891, + "grad_norm": 1.9988250732421875, + "learning_rate": 4.2303005686433795e-06, + "loss": 2.2306, + "step": 667 + }, + { + "epoch": 0.7320547945205479, + "grad_norm": 1.89108407497406, + "learning_rate": 4.228269699431357e-06, + "loss": 2.222, + "step": 668 + }, + { + "epoch": 0.7331506849315068, + "grad_norm": 1.7941503524780273, + "learning_rate": 4.226238830219334e-06, + "loss": 2.1975, + "step": 669 + }, + { + "epoch": 0.7342465753424657, + "grad_norm": 1.8058613538742065, + "learning_rate": 4.224207961007311e-06, + "loss": 2.2037, + "step": 670 + }, + { + "epoch": 0.7342465753424657, + "eval_loss": 2.3266475200653076, + "eval_runtime": 47.6623, + "eval_samples_per_second": 20.981, + "eval_steps_per_second": 0.881, + "step": 670 + }, + { + "epoch": 0.7353424657534247, + "grad_norm": 1.764448642730713, + "learning_rate": 4.2221770917952886e-06, + "loss": 2.1904, + "step": 671 + }, + { + "epoch": 0.7364383561643836, + "grad_norm": 1.8252112865447998, + "learning_rate": 4.220146222583266e-06, + "loss": 2.1959, + "step": 672 + }, + { + "epoch": 0.7375342465753425, + "grad_norm": 2.02898907661438, + "learning_rate": 4.218115353371243e-06, + "loss": 2.1937, + "step": 673 + }, + { + "epoch": 0.7386301369863014, + "grad_norm": 1.8766248226165771, + "learning_rate": 4.21608448415922e-06, + "loss": 2.195, + "step": 674 + }, + { + "epoch": 0.7397260273972602, + "grad_norm": 2.2350332736968994, + "learning_rate": 4.214053614947198e-06, + "loss": 2.1798, + "step": 675 + }, + { + "epoch": 0.7408219178082192, + "grad_norm": 1.820399522781372, + "learning_rate": 4.212022745735175e-06, + "loss": 2.2371, + "step": 676 + }, + { + "epoch": 0.7419178082191781, + "grad_norm": 2.3411312103271484, + "learning_rate": 4.209991876523152e-06, + "loss": 2.1603, + "step": 677 + }, + { + "epoch": 0.743013698630137, + "grad_norm": 1.8562874794006348, + "learning_rate": 4.2079610073111295e-06, + "loss": 2.2082, + "step": 678 + }, + { + "epoch": 0.7441095890410959, + "grad_norm": 2.274214506149292, + "learning_rate": 4.205930138099107e-06, + "loss": 2.1716, + "step": 679 + }, + { + "epoch": 0.7452054794520548, + "grad_norm": 1.9173530340194702, + "learning_rate": 4.203899268887084e-06, + "loss": 2.2461, + "step": 680 + }, + { + "epoch": 0.7452054794520548, + "eval_loss": 2.3253841400146484, + "eval_runtime": 46.5732, + "eval_samples_per_second": 21.472, + "eval_steps_per_second": 0.902, + "step": 680 + }, + { + "epoch": 0.7463013698630137, + "grad_norm": 1.9732033014297485, + "learning_rate": 4.201868399675061e-06, + "loss": 2.2016, + "step": 681 + }, + { + "epoch": 0.7473972602739726, + "grad_norm": 1.8850247859954834, + "learning_rate": 4.199837530463039e-06, + "loss": 2.223, + "step": 682 + }, + { + "epoch": 0.7484931506849315, + "grad_norm": 1.8936313390731812, + "learning_rate": 4.197806661251016e-06, + "loss": 2.1381, + "step": 683 + }, + { + "epoch": 0.7495890410958904, + "grad_norm": 2.11995530128479, + "learning_rate": 4.195775792038993e-06, + "loss": 2.1962, + "step": 684 + }, + { + "epoch": 0.7506849315068493, + "grad_norm": 1.8207820653915405, + "learning_rate": 4.19374492282697e-06, + "loss": 2.1835, + "step": 685 + }, + { + "epoch": 0.7517808219178083, + "grad_norm": 2.06030535697937, + "learning_rate": 4.191714053614948e-06, + "loss": 2.1347, + "step": 686 + }, + { + "epoch": 0.7528767123287671, + "grad_norm": 2.048990488052368, + "learning_rate": 4.189683184402924e-06, + "loss": 2.2189, + "step": 687 + }, + { + "epoch": 0.753972602739726, + "grad_norm": 1.976369023323059, + "learning_rate": 4.187652315190902e-06, + "loss": 2.1805, + "step": 688 + }, + { + "epoch": 0.7550684931506849, + "grad_norm": 2.1160926818847656, + "learning_rate": 4.185621445978879e-06, + "loss": 2.2322, + "step": 689 + }, + { + "epoch": 0.7561643835616438, + "grad_norm": 1.9444516897201538, + "learning_rate": 4.183590576766857e-06, + "loss": 2.1696, + "step": 690 + }, + { + "epoch": 0.7561643835616438, + "eval_loss": 2.3242011070251465, + "eval_runtime": 48.7528, + "eval_samples_per_second": 20.512, + "eval_steps_per_second": 0.861, + "step": 690 + }, + { + "epoch": 0.7572602739726028, + "grad_norm": 1.9567183256149292, + "learning_rate": 4.181559707554833e-06, + "loss": 2.1891, + "step": 691 + }, + { + "epoch": 0.7583561643835617, + "grad_norm": 1.9185020923614502, + "learning_rate": 4.179528838342811e-06, + "loss": 2.2318, + "step": 692 + }, + { + "epoch": 0.7594520547945205, + "grad_norm": 1.833487629890442, + "learning_rate": 4.177497969130788e-06, + "loss": 2.1705, + "step": 693 + }, + { + "epoch": 0.7605479452054794, + "grad_norm": 2.2949867248535156, + "learning_rate": 4.175467099918766e-06, + "loss": 2.1712, + "step": 694 + }, + { + "epoch": 0.7616438356164383, + "grad_norm": 1.8663084506988525, + "learning_rate": 4.173436230706742e-06, + "loss": 2.2262, + "step": 695 + }, + { + "epoch": 0.7627397260273973, + "grad_norm": 2.4633052349090576, + "learning_rate": 4.1714053614947205e-06, + "loss": 2.185, + "step": 696 + }, + { + "epoch": 0.7638356164383562, + "grad_norm": 2.036524772644043, + "learning_rate": 4.169374492282697e-06, + "loss": 2.1697, + "step": 697 + }, + { + "epoch": 0.7649315068493151, + "grad_norm": 2.013810634613037, + "learning_rate": 4.167343623070675e-06, + "loss": 2.1826, + "step": 698 + }, + { + "epoch": 0.766027397260274, + "grad_norm": 2.214109182357788, + "learning_rate": 4.165312753858652e-06, + "loss": 2.2162, + "step": 699 + }, + { + "epoch": 0.7671232876712328, + "grad_norm": 1.8463964462280273, + "learning_rate": 4.163281884646629e-06, + "loss": 2.1865, + "step": 700 + }, + { + "epoch": 0.7671232876712328, + "eval_loss": 2.323822498321533, + "eval_runtime": 48.3307, + "eval_samples_per_second": 20.691, + "eval_steps_per_second": 0.869, + "step": 700 + } + ], + "logging_steps": 1, + "max_steps": 2736, + "num_input_tokens_seen": 0, + "num_train_epochs": 3, + "save_steps": 100, + "stateful_callbacks": { + "TrainerControl": { + "args": { + "should_epoch_stop": false, + "should_evaluate": false, + "should_log": false, + "should_save": true, + "should_training_stop": false + }, + "attributes": {} + } + }, + "total_flos": 2.182116586553344e+17, + "train_batch_size": 3, + "trial_name": null, + "trial_params": null +} diff --git a/babylm_shuffle_nondeterministic_10M_seed0/runs/checkpoint-700/zero_to_fp32.py b/babylm_shuffle_nondeterministic_10M_seed0/runs/checkpoint-700/zero_to_fp32.py new file mode 100644 index 0000000000000000000000000000000000000000..24cc342e78d1a006c782b3a4cd68d9ce786d8fd8 --- /dev/null +++ b/babylm_shuffle_nondeterministic_10M_seed0/runs/checkpoint-700/zero_to_fp32.py @@ -0,0 +1,604 @@ +#!/usr/bin/env python + +# Copyright (c) Microsoft Corporation. +# SPDX-License-Identifier: Apache-2.0 + +# DeepSpeed Team + +# This script extracts fp32 consolidated weights from a zero 1, 2 and 3 DeepSpeed checkpoints. It gets +# copied into the top level checkpoint dir, so the user can easily do the conversion at any point in +# the future. Once extracted, the weights don't require DeepSpeed and can be used in any +# application. +# +# example: python zero_to_fp32.py . pytorch_model.bin + +import argparse +import torch +import glob +import math +import os +import re +from collections import OrderedDict +from dataclasses import dataclass + +# while this script doesn't use deepspeed to recover data, since the checkpoints are pickled with +# DeepSpeed data structures it has to be available in the current python environment. +from deepspeed.utils import logger +from deepspeed.checkpoint.constants import (DS_VERSION, OPTIMIZER_STATE_DICT, SINGLE_PARTITION_OF_FP32_GROUPS, + FP32_FLAT_GROUPS, ZERO_STAGE, PARTITION_COUNT, PARAM_SHAPES, BUFFER_NAMES, + FROZEN_PARAM_SHAPES, FROZEN_PARAM_FRAGMENTS) + + +@dataclass +class zero_model_state: + buffers: dict() + param_shapes: dict() + shared_params: list + ds_version: int + frozen_param_shapes: dict() + frozen_param_fragments: dict() + + +debug = 0 + +# load to cpu +device = torch.device('cpu') + + +def atoi(text): + return int(text) if text.isdigit() else text + + +def natural_keys(text): + ''' + alist.sort(key=natural_keys) sorts in human order + http://nedbatchelder.com/blog/200712/human_sorting.html + (See Toothy's implementation in the comments) + ''' + return [atoi(c) for c in re.split(r'(\d+)', text)] + + +def get_model_state_file(checkpoint_dir, zero_stage): + if not os.path.isdir(checkpoint_dir): + raise FileNotFoundError(f"Directory '{checkpoint_dir}' doesn't exist") + + # there should be only one file + if zero_stage <= 2: + file = os.path.join(checkpoint_dir, "mp_rank_00_model_states.pt") + elif zero_stage == 3: + file = os.path.join(checkpoint_dir, "zero_pp_rank_0_mp_rank_00_model_states.pt") + + if not os.path.exists(file): + raise FileNotFoundError(f"can't find model states file at '{file}'") + + return file + + +def get_checkpoint_files(checkpoint_dir, glob_pattern): + # XXX: need to test that this simple glob rule works for multi-node setup too + ckpt_files = sorted(glob.glob(os.path.join(checkpoint_dir, glob_pattern)), key=natural_keys) + + if len(ckpt_files) == 0: + raise FileNotFoundError(f"can't find {glob_pattern} files in directory '{checkpoint_dir}'") + + return ckpt_files + + +def get_optim_files(checkpoint_dir): + return get_checkpoint_files(checkpoint_dir, "*_optim_states.pt") + + +def get_model_state_files(checkpoint_dir): + return get_checkpoint_files(checkpoint_dir, "*_model_states.pt") + + +def parse_model_states(files): + zero_model_states = [] + for file in files: + state_dict = torch.load(file, map_location=device) + + if BUFFER_NAMES not in state_dict: + raise ValueError(f"{file} is not a model state checkpoint") + buffer_names = state_dict[BUFFER_NAMES] + if debug: + print("Found buffers:", buffer_names) + + # recover just the buffers while restoring them to fp32 if they were saved in fp16 + buffers = {k: v.float() for k, v in state_dict["module"].items() if k in buffer_names} + param_shapes = state_dict[PARAM_SHAPES] + + # collect parameters that are included in param_shapes + param_names = [] + for s in param_shapes: + for name in s.keys(): + param_names.append(name) + + # update with frozen parameters + frozen_param_shapes = state_dict.get(FROZEN_PARAM_SHAPES, None) + if frozen_param_shapes is not None: + if debug: + print(f"Found frozen_param_shapes: {frozen_param_shapes}") + param_names += list(frozen_param_shapes.keys()) + + # handle shared params + shared_params = [[k, v] for k, v in state_dict["shared_params"].items()] + + ds_version = state_dict.get(DS_VERSION, None) + + frozen_param_fragments = state_dict.get(FROZEN_PARAM_FRAGMENTS, None) + + z_model_state = zero_model_state(buffers=buffers, + param_shapes=param_shapes, + shared_params=shared_params, + ds_version=ds_version, + frozen_param_shapes=frozen_param_shapes, + frozen_param_fragments=frozen_param_fragments) + zero_model_states.append(z_model_state) + + return zero_model_states + + +def parse_optim_states(files, ds_checkpoint_dir): + + total_files = len(files) + state_dicts = [] + for f in files: + state_dict = torch.load(f, map_location=device) + # immediately discard the potentially huge 2 optimizer states as we only care for fp32 master weights + # and also handle the case where it was already removed by another helper script + state_dict["optimizer_state_dict"].pop("optimizer_state_dict", None) + state_dicts.append(state_dict) + + if not ZERO_STAGE in state_dicts[0][OPTIMIZER_STATE_DICT]: + raise ValueError(f"{files[0]} is not a zero checkpoint") + zero_stage = state_dicts[0][OPTIMIZER_STATE_DICT][ZERO_STAGE] + world_size = state_dicts[0][OPTIMIZER_STATE_DICT][PARTITION_COUNT] + + # For ZeRO-2 each param group can have different partition_count as data parallelism for expert + # parameters can be different from data parallelism for non-expert parameters. So we can just + # use the max of the partition_count to get the dp world_size. + + if type(world_size) is list: + world_size = max(world_size) + + if world_size != total_files: + raise ValueError( + f"Expected {world_size} of '*_optim_states.pt' under '{ds_checkpoint_dir}' but found {total_files} files. " + "Possibly due to an overwrite of an old checkpoint, or a checkpoint didn't get saved by one or more processes." + ) + + # the groups are named differently in each stage + if zero_stage <= 2: + fp32_groups_key = SINGLE_PARTITION_OF_FP32_GROUPS + elif zero_stage == 3: + fp32_groups_key = FP32_FLAT_GROUPS + else: + raise ValueError(f"unknown zero stage {zero_stage}") + + if zero_stage <= 2: + fp32_flat_groups = [state_dicts[i][OPTIMIZER_STATE_DICT][fp32_groups_key] for i in range(len(state_dicts))] + elif zero_stage == 3: + # if there is more than one param group, there will be multiple flattened tensors - one + # flattened tensor per group - for simplicity merge them into a single tensor + # + # XXX: could make the script more memory efficient for when there are multiple groups - it + # will require matching the sub-lists of param_shapes for each param group flattened tensor + + fp32_flat_groups = [ + torch.cat(state_dicts[i][OPTIMIZER_STATE_DICT][fp32_groups_key], 0) for i in range(len(state_dicts)) + ] + + return zero_stage, world_size, fp32_flat_groups + + +def _get_fp32_state_dict_from_zero_checkpoint(ds_checkpoint_dir, exclude_frozen_parameters): + """ + Returns fp32 state_dict reconstructed from ds checkpoint + + Args: + - ``ds_checkpoint_dir``: path to the deepspeed checkpoint folder (where the optimizer files are) + + """ + print(f"Processing zero checkpoint '{ds_checkpoint_dir}'") + + optim_files = get_optim_files(ds_checkpoint_dir) + zero_stage, world_size, fp32_flat_groups = parse_optim_states(optim_files, ds_checkpoint_dir) + print(f"Detected checkpoint of type zero stage {zero_stage}, world_size: {world_size}") + + model_files = get_model_state_files(ds_checkpoint_dir) + + zero_model_states = parse_model_states(model_files) + print(f'Parsing checkpoint created by deepspeed=={zero_model_states[0].ds_version}') + + if zero_stage <= 2: + return _get_fp32_state_dict_from_zero2_checkpoint(world_size, fp32_flat_groups, zero_model_states, + exclude_frozen_parameters) + elif zero_stage == 3: + return _get_fp32_state_dict_from_zero3_checkpoint(world_size, fp32_flat_groups, zero_model_states, + exclude_frozen_parameters) + + +def _zero2_merge_frozen_params(state_dict, zero_model_states): + if zero_model_states[0].frozen_param_shapes is None or len(zero_model_states[0].frozen_param_shapes) == 0: + return + + frozen_param_shapes = zero_model_states[0].frozen_param_shapes + frozen_param_fragments = zero_model_states[0].frozen_param_fragments + + if debug: + num_elem = sum(s.numel() for s in frozen_param_shapes.values()) + print(f'rank 0: {FROZEN_PARAM_SHAPES}.numel = {num_elem}') + + wanted_params = len(frozen_param_shapes) + wanted_numel = sum(s.numel() for s in frozen_param_shapes.values()) + avail_numel = sum([p.numel() for p in frozen_param_fragments.values()]) + print(f'Frozen params: Have {avail_numel} numels to process.') + print(f'Frozen params: Need {wanted_numel} numels in {wanted_params} params') + + total_params = 0 + total_numel = 0 + for name, shape in frozen_param_shapes.items(): + total_params += 1 + unpartitioned_numel = shape.numel() + total_numel += unpartitioned_numel + + state_dict[name] = frozen_param_fragments[name] + + if debug: + print(f"{name} full shape: {shape} unpartitioned numel {unpartitioned_numel} ") + + print(f"Reconstructed Frozen fp32 state dict with {total_params} params {total_numel} elements") + + +def _has_callable(obj, fn): + attr = getattr(obj, fn, None) + return callable(attr) + + +def _zero2_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states): + param_shapes = zero_model_states[0].param_shapes + + # Reconstruction protocol: + # + # XXX: document this + + if debug: + for i in range(world_size): + for j in range(len(fp32_flat_groups[0])): + print(f"{FP32_FLAT_GROUPS}[{i}][{j}].shape={fp32_flat_groups[i][j].shape}") + + # XXX: memory usage doubles here (zero2) + num_param_groups = len(fp32_flat_groups[0]) + merged_single_partition_of_fp32_groups = [] + for i in range(num_param_groups): + merged_partitions = [sd[i] for sd in fp32_flat_groups] + full_single_fp32_vector = torch.cat(merged_partitions, 0) + merged_single_partition_of_fp32_groups.append(full_single_fp32_vector) + avail_numel = sum( + [full_single_fp32_vector.numel() for full_single_fp32_vector in merged_single_partition_of_fp32_groups]) + + if debug: + wanted_params = sum([len(shapes) for shapes in param_shapes]) + wanted_numel = sum([sum(shape.numel() for shape in shapes.values()) for shapes in param_shapes]) + # not asserting if there is a mismatch due to possible padding + print(f"Have {avail_numel} numels to process.") + print(f"Need {wanted_numel} numels in {wanted_params} params.") + + # params + # XXX: for huge models that can't fit into the host's RAM we will have to recode this to support + # out-of-core computing solution + total_numel = 0 + total_params = 0 + for shapes, full_single_fp32_vector in zip(param_shapes, merged_single_partition_of_fp32_groups): + offset = 0 + avail_numel = full_single_fp32_vector.numel() + for name, shape in shapes.items(): + + unpartitioned_numel = shape.numel() if _has_callable(shape, 'numel') else math.prod(shape) + total_numel += unpartitioned_numel + total_params += 1 + + if debug: + print(f"{name} full shape: {shape} unpartitioned numel {unpartitioned_numel} ") + state_dict[name] = full_single_fp32_vector.narrow(0, offset, unpartitioned_numel).view(shape) + offset += unpartitioned_numel + + # Z2 started to align to 2*world_size to improve nccl performance. Therefore both offset and + # avail_numel can differ by anywhere between 0..2*world_size. Due to two unrelated complex + # paddings performed in the code it's almost impossible to predict the exact numbers w/o the + # live optimizer object, so we are checking that the numbers are within the right range + align_to = 2 * world_size + + def zero2_align(x): + return align_to * math.ceil(x / align_to) + + if debug: + print(f"original offset={offset}, avail_numel={avail_numel}") + + offset = zero2_align(offset) + avail_numel = zero2_align(avail_numel) + + if debug: + print(f"aligned offset={offset}, avail_numel={avail_numel}") + + # Sanity check + if offset != avail_numel: + raise ValueError(f"consumed {offset} numels out of {avail_numel} - something is wrong") + + print(f"Reconstructed fp32 state dict with {total_params} params {total_numel} elements") + + +def _get_fp32_state_dict_from_zero2_checkpoint(world_size, fp32_flat_groups, zero_model_states, + exclude_frozen_parameters): + state_dict = OrderedDict() + + # buffers + buffers = zero_model_states[0].buffers + state_dict.update(buffers) + if debug: + print(f"added {len(buffers)} buffers") + + if not exclude_frozen_parameters: + _zero2_merge_frozen_params(state_dict, zero_model_states) + + _zero2_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states) + + # recover shared parameters + for pair in zero_model_states[0].shared_params: + if pair[1] in state_dict: + state_dict[pair[0]] = state_dict[pair[1]] + + return state_dict + + +def zero3_partitioned_param_info(unpartitioned_numel, world_size): + remainder = unpartitioned_numel % world_size + padding_numel = (world_size - remainder) if remainder else 0 + partitioned_numel = math.ceil(unpartitioned_numel / world_size) + return partitioned_numel, padding_numel + + +def _zero3_merge_frozen_params(state_dict, world_size, zero_model_states): + if zero_model_states[0].frozen_param_shapes is None or len(zero_model_states[0].frozen_param_shapes) == 0: + return + + if debug: + for i in range(world_size): + num_elem = sum(s.numel() for s in zero_model_states[i].frozen_param_fragments.values()) + print(f'rank {i}: {FROZEN_PARAM_SHAPES}.numel = {num_elem}') + + frozen_param_shapes = zero_model_states[0].frozen_param_shapes + wanted_params = len(frozen_param_shapes) + wanted_numel = sum(s.numel() for s in frozen_param_shapes.values()) + avail_numel = sum([p.numel() for p in zero_model_states[0].frozen_param_fragments.values()]) * world_size + print(f'Frozen params: Have {avail_numel} numels to process.') + print(f'Frozen params: Need {wanted_numel} numels in {wanted_params} params') + + total_params = 0 + total_numel = 0 + for name, shape in zero_model_states[0].frozen_param_shapes.items(): + total_params += 1 + unpartitioned_numel = shape.numel() + total_numel += unpartitioned_numel + + param_frags = tuple(model_state.frozen_param_fragments[name] for model_state in zero_model_states) + state_dict[name] = torch.cat(param_frags, 0).narrow(0, 0, unpartitioned_numel).view(shape) + + partitioned_numel, partitioned_padding_numel = zero3_partitioned_param_info(unpartitioned_numel, world_size) + + if debug: + print( + f"Frozen params: {total_params} {name} full shape: {shape} partition0 numel={partitioned_numel} partitioned_padding_numel={partitioned_padding_numel}" + ) + + print(f"Reconstructed Frozen fp32 state dict with {total_params} params {total_numel} elements") + + +def _zero3_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states): + param_shapes = zero_model_states[0].param_shapes + avail_numel = fp32_flat_groups[0].numel() * world_size + # Reconstruction protocol: For zero3 we need to zip the partitions together at boundary of each + # param, re-consolidating each param, while dealing with padding if any + + # merge list of dicts, preserving order + param_shapes = {k: v for d in param_shapes for k, v in d.items()} + + if debug: + for i in range(world_size): + print(f"{FP32_FLAT_GROUPS}[{i}].shape={fp32_flat_groups[i].shape}") + + wanted_params = len(param_shapes) + wanted_numel = sum(shape.numel() for shape in param_shapes.values()) + # not asserting if there is a mismatch due to possible padding + avail_numel = fp32_flat_groups[0].numel() * world_size + print(f"Trainable params: Have {avail_numel} numels to process.") + print(f"Trainable params: Need {wanted_numel} numels in {wanted_params} params.") + + # params + # XXX: for huge models that can't fit into the host's RAM we will have to recode this to support + # out-of-core computing solution + offset = 0 + total_numel = 0 + total_params = 0 + for name, shape in param_shapes.items(): + + unpartitioned_numel = shape.numel() + total_numel += unpartitioned_numel + total_params += 1 + + partitioned_numel, partitioned_padding_numel = zero3_partitioned_param_info(unpartitioned_numel, world_size) + + if debug: + print( + f"Trainable params: {total_params} {name} full shape: {shape} partition0 numel={partitioned_numel} partitioned_padding_numel={partitioned_padding_numel}" + ) + + # XXX: memory usage doubles here + state_dict[name] = torch.cat( + tuple(fp32_flat_groups[i].narrow(0, offset, partitioned_numel) for i in range(world_size)), + 0).narrow(0, 0, unpartitioned_numel).view(shape) + offset += partitioned_numel + + offset *= world_size + + # Sanity check + if offset != avail_numel: + raise ValueError(f"consumed {offset} numels out of {avail_numel} - something is wrong") + + print(f"Reconstructed Trainable fp32 state dict with {total_params} params {total_numel} elements") + + +def _get_fp32_state_dict_from_zero3_checkpoint(world_size, fp32_flat_groups, zero_model_states, + exclude_frozen_parameters): + state_dict = OrderedDict() + + # buffers + buffers = zero_model_states[0].buffers + state_dict.update(buffers) + if debug: + print(f"added {len(buffers)} buffers") + + if not exclude_frozen_parameters: + _zero3_merge_frozen_params(state_dict, world_size, zero_model_states) + + _zero3_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states) + + # recover shared parameters + for pair in zero_model_states[0].shared_params: + if pair[1] in state_dict: + state_dict[pair[0]] = state_dict[pair[1]] + + return state_dict + + +def get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir, tag=None, exclude_frozen_parameters=False): + """ + Convert ZeRO 2 or 3 checkpoint into a single fp32 consolidated state_dict that can be loaded with + ``load_state_dict()`` and used for training without DeepSpeed or shared with others, for example + via a model hub. + + Args: + - ``checkpoint_dir``: path to the desired checkpoint folder + - ``tag``: checkpoint tag used as a unique identifier for checkpoint. If not provided will attempt to load tag in 'latest' file. e.g., ``global_step14`` + - ``exclude_frozen_parameters``: exclude frozen parameters + + Returns: + - pytorch ``state_dict`` + + Note: this approach may not work if your application doesn't have sufficient free CPU memory and + you may need to use the offline approach using the ``zero_to_fp32.py`` script that is saved with + the checkpoint. + + A typical usage might be :: + + from deepspeed.utils.zero_to_fp32 import get_fp32_state_dict_from_zero_checkpoint + # do the training and checkpoint saving + state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir) # already on cpu + model = model.cpu() # move to cpu + model.load_state_dict(state_dict) + # submit to model hub or save the model to share with others + + In this example the ``model`` will no longer be usable in the deepspeed context of the same + application. i.e. you will need to re-initialize the deepspeed engine, since + ``model.load_state_dict(state_dict)`` will remove all the deepspeed magic from it. + + If you want it all done for you, use ``load_state_dict_from_zero_checkpoint`` instead. + + """ + if tag is None: + latest_path = os.path.join(checkpoint_dir, 'latest') + if os.path.isfile(latest_path): + with open(latest_path, 'r') as fd: + tag = fd.read().strip() + else: + raise ValueError(f"Unable to find 'latest' file at {latest_path}") + + ds_checkpoint_dir = os.path.join(checkpoint_dir, tag) + + if not os.path.isdir(ds_checkpoint_dir): + raise FileNotFoundError(f"Directory '{ds_checkpoint_dir}' doesn't exist") + + return _get_fp32_state_dict_from_zero_checkpoint(ds_checkpoint_dir, exclude_frozen_parameters) + + +def convert_zero_checkpoint_to_fp32_state_dict(checkpoint_dir, output_file, tag=None, exclude_frozen_parameters=False): + """ + Convert ZeRO 2 or 3 checkpoint into a single fp32 consolidated ``state_dict`` file that can be + loaded with ``torch.load(file)`` + ``load_state_dict()`` and used for training without DeepSpeed. + + Args: + - ``checkpoint_dir``: path to the desired checkpoint folder. (one that contains the tag-folder, like ``global_step14``) + - ``output_file``: path to the pytorch fp32 state_dict output file (e.g. path/pytorch_model.bin) + - ``tag``: checkpoint tag used as a unique identifier for checkpoint. If not provided will attempt to load tag in the file named ``latest`` in the checkpoint folder, e.g., ``global_step14`` + - ``exclude_frozen_parameters``: exclude frozen parameters + """ + + state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir, tag, exclude_frozen_parameters) + print(f"Saving fp32 state dict to {output_file}") + torch.save(state_dict, output_file) + + +def load_state_dict_from_zero_checkpoint(model, checkpoint_dir, tag=None): + """ + 1. Put the provided model to cpu + 2. Convert ZeRO 2 or 3 checkpoint into a single fp32 consolidated ``state_dict`` + 3. Load it into the provided model + + Args: + - ``model``: the model object to update + - ``checkpoint_dir``: path to the desired checkpoint folder. (one that contains the tag-folder, like ``global_step14``) + - ``tag``: checkpoint tag used as a unique identifier for checkpoint. If not provided will attempt to load tag in the file named ``latest`` in the checkpoint folder, e.g., ``global_step14`` + + Returns: + - ``model`: modified model + + Make sure you have plenty of CPU memory available before you call this function. If you don't + have enough use the ``zero_to_fp32.py`` utility to do the conversion. You will find it + conveniently placed for you in the checkpoint folder. + + A typical usage might be :: + + from deepspeed.utils.zero_to_fp32 import load_state_dict_from_zero_checkpoint + model = load_state_dict_from_zero_checkpoint(trainer.model, checkpoint_dir) + # submit to model hub or save the model to share with others + + Note, that once this was run, the ``model`` will no longer be usable in the deepspeed context + of the same application. i.e. you will need to re-initialize the deepspeed engine, since + ``model.load_state_dict(state_dict)`` will remove all the deepspeed magic from it. + + """ + logger.info(f"Extracting fp32 weights") + state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir, tag) + + logger.info(f"Overwriting model with fp32 weights") + model = model.cpu() + model.load_state_dict(state_dict, strict=False) + + return model + + +if __name__ == "__main__": + + parser = argparse.ArgumentParser() + parser.add_argument("checkpoint_dir", + type=str, + help="path to the desired checkpoint folder, e.g., path/checkpoint-12") + parser.add_argument( + "output_file", + type=str, + help="path to the pytorch fp32 state_dict output file (e.g. path/checkpoint-12/pytorch_model.bin)") + parser.add_argument("-t", + "--tag", + type=str, + default=None, + help="checkpoint tag used as a unique identifier for checkpoint. e.g., global_step1") + parser.add_argument("--exclude_frozen_parameters", action='store_true', help="exclude frozen parameters") + parser.add_argument("-d", "--debug", action='store_true', help="enable debug") + args = parser.parse_args() + + debug = args.debug + + convert_zero_checkpoint_to_fp32_state_dict(args.checkpoint_dir, + args.output_file, + tag=args.tag, + exclude_frozen_parameters=args.exclude_frozen_parameters)