diff --git a/.gitattributes b/.gitattributes index 3ae2c97e6e228f02cff703ca93ebe9b06d040fe6..7d4efa0e14051f609e8d50c9f1202efb2697dcaa 100644 --- a/.gitattributes +++ b/.gitattributes @@ -653,3 +653,16 @@ checkpoints/Llama-3.2-3B/babylm_reverse_control_10M_seed0/runs/checkpoint-2400/t checkpoints/Llama-3.2-3B/babylm_reverse_control_10M_seed0/runs/checkpoint-2300/tokenizer.json filter=lfs diff=lfs merge=lfs -text checkpoints/Llama-3.2-3B/babylm_reverse_control_10M_seed0/runs/checkpoint-2600/tokenizer.json filter=lfs diff=lfs merge=lfs -text checkpoints/Llama-3.2-3B/babylm_reverse_control_10M_seed0/runs/checkpoint-2200/tokenizer.json filter=lfs diff=lfs merge=lfs -text +checkpoints/Llama-3.2-3B/babylm_reverse_control_10M_seed0/runs/checkpoint-2500/tokenizer.json filter=lfs diff=lfs merge=lfs -text +checkpoints/Llama-3.2-3B/babylm_reverse_control_10M_seed0/artifacts/models--meta-llama--Llama-3.2-3B/blobs/584d8d3e3f82f7964955174dfe5e3b1cf117a9d859f022cfdf7fcb884856e002 filter=lfs diff=lfs merge=lfs -text +checkpoints/Llama-3.2-3B/babylm_reverse_control_10M_seed0/artifacts/models--meta-llama--Llama-3.2-3B/blobs/4719a04514ec2f060240711b7c33ab21187cac730ecaba3040b7a0fd95a9cefb filter=lfs diff=lfs merge=lfs -text +checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-900/tokenizer.json filter=lfs diff=lfs merge=lfs -text +checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-100/tokenizer.json filter=lfs diff=lfs merge=lfs -text +checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-1700/tokenizer.json filter=lfs diff=lfs merge=lfs -text +checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-1200/tokenizer.json filter=lfs diff=lfs merge=lfs -text +checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-1100/tokenizer.json filter=lfs diff=lfs merge=lfs -text +checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-1600/tokenizer.json filter=lfs diff=lfs merge=lfs -text +checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-600/tokenizer.json filter=lfs diff=lfs merge=lfs -text +checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-1500/tokenizer.json filter=lfs diff=lfs merge=lfs -text +checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-1300/tokenizer.json filter=lfs diff=lfs merge=lfs -text +checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-2700/tokenizer.json filter=lfs diff=lfs merge=lfs -text diff --git a/checkpoints/Llama-3.2-3B-FTP/babylm_shuffle_nondeterministic_10M_seed0/runs/checkpoint-200/global_step200/zero_pp_rank_0_mp_rank_00_optim_states.pt b/checkpoints/Llama-3.2-3B-FTP/babylm_shuffle_nondeterministic_10M_seed0/runs/checkpoint-200/global_step200/zero_pp_rank_0_mp_rank_00_optim_states.pt new file mode 100644 index 0000000000000000000000000000000000000000..225bb382b16593c12e4419bd54343808cc2d9b53 --- /dev/null +++ b/checkpoints/Llama-3.2-3B-FTP/babylm_shuffle_nondeterministic_10M_seed0/runs/checkpoint-200/global_step200/zero_pp_rank_0_mp_rank_00_optim_states.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:d431116b89bf76f2ca96d97404c904a83c21eebc2b1569ac25627fc67cbc0344 +size 25702104228 diff --git a/checkpoints/Llama-3.2-3B-FTP/babylm_shuffle_nondeterministic_10M_seed0/runs/checkpoint-200/global_step200/zero_pp_rank_2_mp_rank_00_optim_states.pt b/checkpoints/Llama-3.2-3B-FTP/babylm_shuffle_nondeterministic_10M_seed0/runs/checkpoint-200/global_step200/zero_pp_rank_2_mp_rank_00_optim_states.pt new file mode 100644 index 0000000000000000000000000000000000000000..e89630c68e205aba7841bd87475d62ff3b747c0e --- /dev/null +++ b/checkpoints/Llama-3.2-3B-FTP/babylm_shuffle_nondeterministic_10M_seed0/runs/checkpoint-200/global_step200/zero_pp_rank_2_mp_rank_00_optim_states.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:d46fe81e0c5bf68fb0045308959954d5b8906e202abe66b4ed3df086596ff8e7 +size 25702107428 diff --git a/checkpoints/Llama-3.2-3B/babylm_hop_words4_10M_seed0/runs/checkpoint-2200/global_step2200/zero_pp_rank_0_mp_rank_00_optim_states.pt b/checkpoints/Llama-3.2-3B/babylm_hop_words4_10M_seed0/runs/checkpoint-2200/global_step2200/zero_pp_rank_0_mp_rank_00_optim_states.pt new file mode 100644 index 0000000000000000000000000000000000000000..f69d152f4f7bcc631ab2d5e22bc2adba5095df48 --- /dev/null +++ b/checkpoints/Llama-3.2-3B/babylm_hop_words4_10M_seed0/runs/checkpoint-2200/global_step2200/zero_pp_rank_0_mp_rank_00_optim_states.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:574b6d6c0a6b18107c4de901853283067587f9ad499e423c96a6d71edd3d1ce1 +size 12851008339 diff --git a/checkpoints/Llama-3.2-3B/babylm_hop_words4_10M_seed0/runs/checkpoint-2200/global_step2200/zero_pp_rank_1_mp_rank_00_optim_states.pt b/checkpoints/Llama-3.2-3B/babylm_hop_words4_10M_seed0/runs/checkpoint-2200/global_step2200/zero_pp_rank_1_mp_rank_00_optim_states.pt new file mode 100644 index 0000000000000000000000000000000000000000..9b4e6204e16d8d76ef6925ac46adeb541fb10b7b --- /dev/null +++ b/checkpoints/Llama-3.2-3B/babylm_hop_words4_10M_seed0/runs/checkpoint-2200/global_step2200/zero_pp_rank_1_mp_rank_00_optim_states.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:57127b4ecb852003013e90027a32d91a3c2b9df0238363e543d36cba73d56df2 +size 12851010643 diff --git a/checkpoints/Llama-3.2-3B/babylm_reverse_control_10M_seed0/artifacts/models--meta-llama--Llama-3.2-3B/blobs/4719a04514ec2f060240711b7c33ab21187cac730ecaba3040b7a0fd95a9cefb b/checkpoints/Llama-3.2-3B/babylm_reverse_control_10M_seed0/artifacts/models--meta-llama--Llama-3.2-3B/blobs/4719a04514ec2f060240711b7c33ab21187cac730ecaba3040b7a0fd95a9cefb new file mode 100644 index 0000000000000000000000000000000000000000..82a877359be75c9c7ad1226b362efeeabf2edef4 --- /dev/null +++ b/checkpoints/Llama-3.2-3B/babylm_reverse_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/checkpoints/Llama-3.2-3B/babylm_reverse_control_10M_seed0/artifacts/models--meta-llama--Llama-3.2-3B/blobs/584d8d3e3f82f7964955174dfe5e3b1cf117a9d859f022cfdf7fcb884856e002 b/checkpoints/Llama-3.2-3B/babylm_reverse_control_10M_seed0/artifacts/models--meta-llama--Llama-3.2-3B/blobs/584d8d3e3f82f7964955174dfe5e3b1cf117a9d859f022cfdf7fcb884856e002 new file mode 100644 index 0000000000000000000000000000000000000000..e362b81bf93cef8c6d96b08171bab63416808c70 --- /dev/null +++ b/checkpoints/Llama-3.2-3B/babylm_reverse_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/checkpoints/Llama-3.2-3B/babylm_reverse_control_10M_seed0/artifacts/models--meta-llama--Llama-3.2-3B/snapshots/13afe5124825b4f3751f836b40dafda64c1ed062/model-00001-of-00002.safetensors b/checkpoints/Llama-3.2-3B/babylm_reverse_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/checkpoints/Llama-3.2-3B/babylm_reverse_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/checkpoints/Llama-3.2-3B/babylm_reverse_control_10M_seed0/artifacts/models--meta-llama--Llama-3.2-3B/snapshots/13afe5124825b4f3751f836b40dafda64c1ed062/model-00002-of-00002.safetensors b/checkpoints/Llama-3.2-3B/babylm_reverse_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/checkpoints/Llama-3.2-3B/babylm_reverse_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/checkpoints/Llama-3.2-3B/babylm_reverse_control_10M_seed0/runs/checkpoint-100/model-00001-of-00002.safetensors b/checkpoints/Llama-3.2-3B/babylm_reverse_control_10M_seed0/runs/checkpoint-100/model-00001-of-00002.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..5c22178a05d598bcf2727a05cbb9440a85f09b0b --- /dev/null +++ b/checkpoints/Llama-3.2-3B/babylm_reverse_control_10M_seed0/runs/checkpoint-100/model-00001-of-00002.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:76ed57adcc606cf2c634c7a0a342a8e51c9fa3bdee46273ab902173c0b5166cc +size 4965798912 diff --git a/checkpoints/Llama-3.2-3B/babylm_reverse_control_10M_seed0/runs/checkpoint-1400/model-00002-of-00002.safetensors b/checkpoints/Llama-3.2-3B/babylm_reverse_control_10M_seed0/runs/checkpoint-1400/model-00002-of-00002.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..ce8e519404533876ccd2f67fea43bd3016bbd2e4 --- /dev/null +++ b/checkpoints/Llama-3.2-3B/babylm_reverse_control_10M_seed0/runs/checkpoint-1400/model-00002-of-00002.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:a9d9c9544592c0876a3aff349499d66c6c146b0591bb6bdb777f915ef8ee3816 +size 2247734920 diff --git a/checkpoints/Llama-3.2-3B/babylm_reverse_control_10M_seed0/runs/checkpoint-1700/model-00001-of-00002.safetensors b/checkpoints/Llama-3.2-3B/babylm_reverse_control_10M_seed0/runs/checkpoint-1700/model-00001-of-00002.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..c133f3ae62fb85288cbfcd73e0f2d2c226895f23 --- /dev/null +++ b/checkpoints/Llama-3.2-3B/babylm_reverse_control_10M_seed0/runs/checkpoint-1700/model-00001-of-00002.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:e9bc0197901600a3f39b3e34bc6091a2d475e951221efedf81de2938fd1416cb +size 4965798912 diff --git a/checkpoints/Llama-3.2-3B/babylm_reverse_control_10M_seed0/runs/checkpoint-1800/model-00002-of-00002.safetensors b/checkpoints/Llama-3.2-3B/babylm_reverse_control_10M_seed0/runs/checkpoint-1800/model-00002-of-00002.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..e5651366d3d329d301115356a88ac275083bdd45 --- /dev/null +++ b/checkpoints/Llama-3.2-3B/babylm_reverse_control_10M_seed0/runs/checkpoint-1800/model-00002-of-00002.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:ac3e0b7e4515c73879d8990915f113dfd573cf665111e43c4b52d3e6483dc48f +size 2247734920 diff --git a/checkpoints/Llama-3.2-3B/babylm_reverse_control_10M_seed0/runs/checkpoint-1900/model-00002-of-00002.safetensors b/checkpoints/Llama-3.2-3B/babylm_reverse_control_10M_seed0/runs/checkpoint-1900/model-00002-of-00002.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..23190e9b5da1e69584c248909402bcfcf21c1e2e --- /dev/null +++ b/checkpoints/Llama-3.2-3B/babylm_reverse_control_10M_seed0/runs/checkpoint-1900/model-00002-of-00002.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:8928d125598d4b80649008968437776aa7adc363ae70575581a7ce30761349e8 +size 2247734920 diff --git a/checkpoints/Llama-3.2-3B/babylm_reverse_control_10M_seed0/runs/checkpoint-200/model-00002-of-00002.safetensors b/checkpoints/Llama-3.2-3B/babylm_reverse_control_10M_seed0/runs/checkpoint-200/model-00002-of-00002.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..5f06db76be6a4c3ed6a1033f42b21c6944a1486c --- /dev/null +++ b/checkpoints/Llama-3.2-3B/babylm_reverse_control_10M_seed0/runs/checkpoint-200/model-00002-of-00002.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:ccdb924c137a3ded423856e6309f9e3c0ad2fd944ef62fb4a5e696fd2f5db4c7 +size 2247734920 diff --git a/checkpoints/Llama-3.2-3B/babylm_reverse_control_10M_seed0/runs/checkpoint-2000/model-00002-of-00002.safetensors b/checkpoints/Llama-3.2-3B/babylm_reverse_control_10M_seed0/runs/checkpoint-2000/model-00002-of-00002.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..85c34de56fd43b10d22c476f9491e35074b55228 --- /dev/null +++ b/checkpoints/Llama-3.2-3B/babylm_reverse_control_10M_seed0/runs/checkpoint-2000/model-00002-of-00002.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:ab1d82fc2187083ca08f49e2ff04af120382f285853d3704d0d1f8db0535a154 +size 2247734920 diff --git a/checkpoints/Llama-3.2-3B/babylm_reverse_control_10M_seed0/runs/checkpoint-2200/rng_state_1.pth b/checkpoints/Llama-3.2-3B/babylm_reverse_control_10M_seed0/runs/checkpoint-2200/rng_state_1.pth new file mode 100644 index 0000000000000000000000000000000000000000..9a2039c6da7a85a1210a9c30d33ef8b01f90c45a --- /dev/null +++ b/checkpoints/Llama-3.2-3B/babylm_reverse_control_10M_seed0/runs/checkpoint-2200/rng_state_1.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:24be9519b492ae2ab5da4e7d258ad3f276da346422161ebc6874f26057c028cd +size 16567 diff --git a/checkpoints/Llama-3.2-3B/babylm_reverse_control_10M_seed0/runs/checkpoint-2200/training_args.bin b/checkpoints/Llama-3.2-3B/babylm_reverse_control_10M_seed0/runs/checkpoint-2200/training_args.bin new file mode 100644 index 0000000000000000000000000000000000000000..35c83e0aeb67b51f9041a0775518d818bd4514c1 --- /dev/null +++ b/checkpoints/Llama-3.2-3B/babylm_reverse_control_10M_seed0/runs/checkpoint-2200/training_args.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:892167fd00638a9a336920c633c1535594e6726eea0c8de0982206892bd302fd +size 6011 diff --git a/checkpoints/Llama-3.2-3B/babylm_reverse_control_10M_seed0/runs/checkpoint-2500/rng_state_0.pth b/checkpoints/Llama-3.2-3B/babylm_reverse_control_10M_seed0/runs/checkpoint-2500/rng_state_0.pth new file mode 100644 index 0000000000000000000000000000000000000000..f084998befba4c7b08cfeba516748af5fcfa6071 --- /dev/null +++ b/checkpoints/Llama-3.2-3B/babylm_reverse_control_10M_seed0/runs/checkpoint-2500/rng_state_0.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:8bfdf5ccc88de7823e7dc6081ffb5cb38ed71058da20638fbf7f61d94c58874c +size 16567 diff --git a/checkpoints/Llama-3.2-3B/babylm_reverse_control_10M_seed0/runs/checkpoint-2500/rng_state_1.pth b/checkpoints/Llama-3.2-3B/babylm_reverse_control_10M_seed0/runs/checkpoint-2500/rng_state_1.pth new file mode 100644 index 0000000000000000000000000000000000000000..a9f12d82a8fc9d14b8b35ce910d38bc891b2dafa --- /dev/null +++ b/checkpoints/Llama-3.2-3B/babylm_reverse_control_10M_seed0/runs/checkpoint-2500/rng_state_1.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:aa288bebadeaf65eb824b833596a55b793d13b5c957085cf3d7dc8c0e6de2ddf +size 16567 diff --git a/checkpoints/Llama-3.2-3B/babylm_reverse_control_10M_seed0/runs/checkpoint-2500/rng_state_2.pth b/checkpoints/Llama-3.2-3B/babylm_reverse_control_10M_seed0/runs/checkpoint-2500/rng_state_2.pth new file mode 100644 index 0000000000000000000000000000000000000000..cf71113371921ea2e7fb737a20484b1a62f38481 --- /dev/null +++ b/checkpoints/Llama-3.2-3B/babylm_reverse_control_10M_seed0/runs/checkpoint-2500/rng_state_2.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:a08b70b84bf331849cc389e31e88e5bf289f53f5f48e4ac516e4a788f9bfae7d +size 16567 diff --git a/checkpoints/Llama-3.2-3B/babylm_reverse_control_10M_seed0/runs/checkpoint-2500/scheduler.pt b/checkpoints/Llama-3.2-3B/babylm_reverse_control_10M_seed0/runs/checkpoint-2500/scheduler.pt new file mode 100644 index 0000000000000000000000000000000000000000..37144c7159b76a178058169c5e2b0aaee93a1afc --- /dev/null +++ b/checkpoints/Llama-3.2-3B/babylm_reverse_control_10M_seed0/runs/checkpoint-2500/scheduler.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:cf2ce44ba975caf32f04656b0621374f8b580738300b91f100cc656dd6882aac +size 627 diff --git a/checkpoints/Llama-3.2-3B/babylm_reverse_control_10M_seed0/runs/checkpoint-2500/tokenizer.json b/checkpoints/Llama-3.2-3B/babylm_reverse_control_10M_seed0/runs/checkpoint-2500/tokenizer.json new file mode 100644 index 0000000000000000000000000000000000000000..c76f0701115b50c463b66cc3645d7a3654ab5205 --- /dev/null +++ b/checkpoints/Llama-3.2-3B/babylm_reverse_control_10M_seed0/runs/checkpoint-2500/tokenizer.json @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:6fda2ecf7b5daf3cb77324ea6c5b9cb8bf47119249a6d756ef2dd4d62540ac6f +size 17210554 diff --git a/checkpoints/Llama-3.2-3B/babylm_reverse_control_10M_seed0/runs/checkpoint-2500/training_args.bin b/checkpoints/Llama-3.2-3B/babylm_reverse_control_10M_seed0/runs/checkpoint-2500/training_args.bin new file mode 100644 index 0000000000000000000000000000000000000000..35c83e0aeb67b51f9041a0775518d818bd4514c1 --- /dev/null +++ b/checkpoints/Llama-3.2-3B/babylm_reverse_control_10M_seed0/runs/checkpoint-2500/training_args.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:892167fd00638a9a336920c633c1535594e6726eea0c8de0982206892bd302fd +size 6011 diff --git a/checkpoints/Llama-3.2-3B/babylm_reverse_control_10M_seed0/runs/checkpoint-2919/model-00002-of-00002.safetensors b/checkpoints/Llama-3.2-3B/babylm_reverse_control_10M_seed0/runs/checkpoint-2919/model-00002-of-00002.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..e2d70a817b597bd765dd8864e71071bc0fdf8bf0 --- /dev/null +++ b/checkpoints/Llama-3.2-3B/babylm_reverse_control_10M_seed0/runs/checkpoint-2919/model-00002-of-00002.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:c891a6a03756ae9e52044ed2fad7a456ab017f313458554f965cb8b38e31000a +size 2247734920 diff --git a/checkpoints/Llama-3.2-3B/babylm_reverse_control_10M_seed0/runs/checkpoint-500/model-00002-of-00002.safetensors b/checkpoints/Llama-3.2-3B/babylm_reverse_control_10M_seed0/runs/checkpoint-500/model-00002-of-00002.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..5ee64b69f5ff43719904fed8a8ad4f3b641b7362 --- /dev/null +++ b/checkpoints/Llama-3.2-3B/babylm_reverse_control_10M_seed0/runs/checkpoint-500/model-00002-of-00002.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:a14aaeb71afd57a65d7bf5ad0ae05e187214bb500413de99abd7ab01b3ce3861 +size 2247734920 diff --git a/checkpoints/Llama-3.2-3B/babylm_reverse_control_10M_seed0/runs/checkpoint-700/model-00002-of-00002.safetensors b/checkpoints/Llama-3.2-3B/babylm_reverse_control_10M_seed0/runs/checkpoint-700/model-00002-of-00002.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..6dc2c04fcdc76e57a6ff10e959330513ab3806a0 --- /dev/null +++ b/checkpoints/Llama-3.2-3B/babylm_reverse_control_10M_seed0/runs/checkpoint-700/model-00002-of-00002.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:fbdf80c4466bae8018473078d3a1ded6544bc136b916ad09d733d96153283943 +size 2247734920 diff --git a/checkpoints/Llama-3.2-3B/babylm_reverse_control_10M_seed0/runs/checkpoint-800/model-00002-of-00002.safetensors b/checkpoints/Llama-3.2-3B/babylm_reverse_control_10M_seed0/runs/checkpoint-800/model-00002-of-00002.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..173f1b87f58c00d5cfead9e4ee8f4ba548e83724 --- /dev/null +++ b/checkpoints/Llama-3.2-3B/babylm_reverse_control_10M_seed0/runs/checkpoint-800/model-00002-of-00002.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:78e74dadf9a4362635c8c8c7b4f10e2cab5171713578ec1ab0d110b20cddc965 +size 2247734920 diff --git a/checkpoints/Llama-3.2-3B/babylm_reverse_control_10M_seed0/runs/checkpoint-900/model-00001-of-00002.safetensors b/checkpoints/Llama-3.2-3B/babylm_reverse_control_10M_seed0/runs/checkpoint-900/model-00001-of-00002.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..50fb8167338206028d3cb2fed5074d565a475826 --- /dev/null +++ b/checkpoints/Llama-3.2-3B/babylm_reverse_control_10M_seed0/runs/checkpoint-900/model-00001-of-00002.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:28531be3a7d16f3f82a7b89ed8904279ca3d7e2c0a6d44d00e2260eeeb36b97e +size 4965798912 diff --git a/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-100/rng_state_0.pth b/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-100/rng_state_0.pth new file mode 100644 index 0000000000000000000000000000000000000000..44995716d9133cdde1b0eafab9b2cbbb3010cad2 --- /dev/null +++ b/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-100/rng_state_0.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:01b934e6ec558bac6fe169663502cde3d438344e9336e1e7d8afe651e9135428 +size 16567 diff --git a/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-100/rng_state_1.pth b/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-100/rng_state_1.pth new file mode 100644 index 0000000000000000000000000000000000000000..ab8016bddf781c8ef313d7126b4b87d8b0c76b56 --- /dev/null +++ b/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-100/rng_state_1.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:1bc80f18f7e62b82a61f483f87081652089fe5a8da2188f74fe6f789b8c74274 +size 16567 diff --git a/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-100/rng_state_2.pth b/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-100/rng_state_2.pth new file mode 100644 index 0000000000000000000000000000000000000000..70cef6545138ddc2cc4b0c0ab18344cceb230f38 --- /dev/null +++ b/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-100/rng_state_2.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:74c577a7e7176d95ccd3d6ec57d9ef0875615f3974af670a48262ddf57ea767d +size 16567 diff --git a/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-100/scheduler.pt b/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-100/scheduler.pt new file mode 100644 index 0000000000000000000000000000000000000000..c03498f12c7817d4bbb55eae3a0bc6d0397b6056 --- /dev/null +++ b/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-100/scheduler.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:c2ea5e6938d34d22f9628dfc460e37fb652939f5bb51424123150e7dfaaab414 +size 627 diff --git a/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-100/tokenizer.json b/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-100/tokenizer.json new file mode 100644 index 0000000000000000000000000000000000000000..21e1edfa2183a57886f456b02efd11cce1a7df83 --- /dev/null +++ b/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-100/tokenizer.json @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:e8485b8900f2ef3a891091dbe5ef0edc56f6922c0a23c7dfbb4baf64854f7de4 +size 17210102 diff --git a/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-100/training_args.bin b/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-100/training_args.bin new file mode 100644 index 0000000000000000000000000000000000000000..0628b4aecba57cabec5230705ebfea1d004cfc77 --- /dev/null +++ b/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-100/training_args.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:f1b22d390bcc8f3a3c3fa0ce7ebfd361692a465dd5cb0c24a819e263b73e1c2e +size 6011 diff --git a/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-1000/config.json b/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-1000/config.json new file mode 100644 index 0000000000000000000000000000000000000000..fb0f9a53a507ffcdb60410deb71d60ef801bf350 --- /dev/null +++ b/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-1000/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/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-1000/generation_config.json b/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-1000/generation_config.json new file mode 100644 index 0000000000000000000000000000000000000000..9c389b841a9a29ddd904e02d1eb0e08dcce82ad9 --- /dev/null +++ b/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-1000/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/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-1000/model.safetensors.index.json b/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-1000/model.safetensors.index.json new file mode 100644 index 0000000000000000000000000000000000000000..ed64de846d720b9a7859dc20575fea8e8ca51940 --- /dev/null +++ b/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-1000/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/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-1000/special_tokens_map.json b/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-1000/special_tokens_map.json new file mode 100644 index 0000000000000000000000000000000000000000..6949c5975ee0e961ef61cf31010dce04df0a03f8 --- /dev/null +++ b/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-1000/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/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-1000/tokenizer_config.json b/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-1000/tokenizer_config.json new file mode 100644 index 0000000000000000000000000000000000000000..5684b14e2756b56aa397c1324e239b442162b34a --- /dev/null +++ b/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-1000/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/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-1000/trainer_state.json b/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-1000/trainer_state.json new file mode 100644 index 0000000000000000000000000000000000000000..8c5eddce305b7a88cf5edff4a5585a136362fb47 --- /dev/null +++ b/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-1000/trainer_state.json @@ -0,0 +1,7833 @@ +{ + "best_metric": null, + "best_model_checkpoint": null, + "epoch": 1.095890410958904, + "eval_steps": 10, + "global_step": 1000, + "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": 3.0476, + "step": 1 + }, + { + "epoch": 0.002191780821917808, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.0289, + "step": 2 + }, + { + "epoch": 0.0032876712328767125, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 2.9935, + "step": 3 + }, + { + "epoch": 0.004383561643835616, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.0139, + "step": 4 + }, + { + "epoch": 0.005479452054794521, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 2.9892, + "step": 5 + }, + { + "epoch": 0.006575342465753425, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.1252, + "step": 6 + }, + { + "epoch": 0.007671232876712329, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.0291, + "step": 7 + }, + { + "epoch": 0.008767123287671232, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.0607, + "step": 8 + }, + { + "epoch": 0.009863013698630137, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.0306, + "step": 9 + }, + { + "epoch": 0.010958904109589041, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.0811, + "step": 10 + }, + { + "epoch": 0.010958904109589041, + "eval_loss": 3.0768930912017822, + "eval_runtime": 38.0377, + "eval_samples_per_second": 26.29, + "eval_steps_per_second": 1.104, + "step": 10 + }, + { + "epoch": 0.012054794520547946, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.0693, + "step": 11 + }, + { + "epoch": 0.01315068493150685, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 2.9668, + "step": 12 + }, + { + "epoch": 0.014246575342465753, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.052, + "step": 13 + }, + { + "epoch": 0.015342465753424657, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.0363, + "step": 14 + }, + { + "epoch": 0.01643835616438356, + "grad_norm": 5.58175802230835, + "learning_rate": 1.8248175182481753e-08, + "loss": 3.0778, + "step": 15 + }, + { + "epoch": 0.017534246575342465, + "grad_norm": 6.360531806945801, + "learning_rate": 3.6496350364963505e-08, + "loss": 3.0257, + "step": 16 + }, + { + "epoch": 0.01863013698630137, + "grad_norm": 6.453113555908203, + "learning_rate": 5.4744525547445264e-08, + "loss": 3.0262, + "step": 17 + }, + { + "epoch": 0.019726027397260273, + "grad_norm": 6.789405345916748, + "learning_rate": 7.299270072992701e-08, + "loss": 3.0671, + "step": 18 + }, + { + "epoch": 0.020821917808219178, + "grad_norm": 6.185514450073242, + "learning_rate": 9.124087591240876e-08, + "loss": 3.0077, + "step": 19 + }, + { + "epoch": 0.021917808219178082, + "grad_norm": 5.719798564910889, + "learning_rate": 1.0948905109489053e-07, + "loss": 3.089, + "step": 20 + }, + { + "epoch": 0.021917808219178082, + "eval_loss": 3.07546329498291, + "eval_runtime": 37.5318, + "eval_samples_per_second": 26.644, + "eval_steps_per_second": 1.119, + "step": 20 + }, + { + "epoch": 0.023013698630136987, + "grad_norm": 6.307768821716309, + "learning_rate": 1.2773722627737227e-07, + "loss": 3.0641, + "step": 21 + }, + { + "epoch": 0.02410958904109589, + "grad_norm": 6.255112171173096, + "learning_rate": 1.4598540145985402e-07, + "loss": 2.9849, + "step": 22 + }, + { + "epoch": 0.025205479452054796, + "grad_norm": 5.812101364135742, + "learning_rate": 1.642335766423358e-07, + "loss": 3.0211, + "step": 23 + }, + { + "epoch": 0.0263013698630137, + "grad_norm": 6.262075901031494, + "learning_rate": 1.8248175182481753e-07, + "loss": 3.0654, + "step": 24 + }, + { + "epoch": 0.0273972602739726, + "grad_norm": 5.419229507446289, + "learning_rate": 2.0072992700729928e-07, + "loss": 3.0661, + "step": 25 + }, + { + "epoch": 0.028493150684931506, + "grad_norm": 4.750737190246582, + "learning_rate": 2.1897810218978106e-07, + "loss": 3.0494, + "step": 26 + }, + { + "epoch": 0.02958904109589041, + "grad_norm": 4.732361316680908, + "learning_rate": 2.3722627737226278e-07, + "loss": 3.0471, + "step": 27 + }, + { + "epoch": 0.030684931506849315, + "grad_norm": 4.749786853790283, + "learning_rate": 2.5547445255474454e-07, + "loss": 3.0107, + "step": 28 + }, + { + "epoch": 0.03178082191780822, + "grad_norm": 4.451244831085205, + "learning_rate": 2.737226277372263e-07, + "loss": 2.9987, + "step": 29 + }, + { + "epoch": 0.03287671232876712, + "grad_norm": 4.5158257484436035, + "learning_rate": 2.9197080291970804e-07, + "loss": 2.9686, + "step": 30 + }, + { + "epoch": 0.03287671232876712, + "eval_loss": 3.031872034072876, + "eval_runtime": 37.2608, + "eval_samples_per_second": 26.838, + "eval_steps_per_second": 1.127, + "step": 30 + }, + { + "epoch": 0.03397260273972603, + "grad_norm": 3.8239498138427734, + "learning_rate": 3.102189781021898e-07, + "loss": 3.0057, + "step": 31 + }, + { + "epoch": 0.03506849315068493, + "grad_norm": 4.037254810333252, + "learning_rate": 3.284671532846716e-07, + "loss": 2.9733, + "step": 32 + }, + { + "epoch": 0.03616438356164384, + "grad_norm": 4.283931732177734, + "learning_rate": 3.467153284671533e-07, + "loss": 2.9515, + "step": 33 + }, + { + "epoch": 0.03726027397260274, + "grad_norm": 4.1784987449646, + "learning_rate": 3.6496350364963505e-07, + "loss": 2.9483, + "step": 34 + }, + { + "epoch": 0.038356164383561646, + "grad_norm": 3.9210753440856934, + "learning_rate": 3.8321167883211686e-07, + "loss": 2.9679, + "step": 35 + }, + { + "epoch": 0.03945205479452055, + "grad_norm": 3.72884202003479, + "learning_rate": 4.0145985401459856e-07, + "loss": 2.9989, + "step": 36 + }, + { + "epoch": 0.040547945205479455, + "grad_norm": 3.2882330417633057, + "learning_rate": 4.197080291970803e-07, + "loss": 2.9384, + "step": 37 + }, + { + "epoch": 0.041643835616438356, + "grad_norm": 3.515293836593628, + "learning_rate": 4.379562043795621e-07, + "loss": 2.8986, + "step": 38 + }, + { + "epoch": 0.042739726027397264, + "grad_norm": 3.568453073501587, + "learning_rate": 4.562043795620438e-07, + "loss": 2.9854, + "step": 39 + }, + { + "epoch": 0.043835616438356165, + "grad_norm": 3.389789342880249, + "learning_rate": 4.7445255474452557e-07, + "loss": 2.8891, + "step": 40 + }, + { + "epoch": 0.043835616438356165, + "eval_loss": 2.964235544204712, + "eval_runtime": 37.3406, + "eval_samples_per_second": 26.781, + "eval_steps_per_second": 1.125, + "step": 40 + }, + { + "epoch": 0.044931506849315066, + "grad_norm": 3.1826283931732178, + "learning_rate": 4.927007299270073e-07, + "loss": 2.9271, + "step": 41 + }, + { + "epoch": 0.046027397260273974, + "grad_norm": 2.941539764404297, + "learning_rate": 5.109489051094891e-07, + "loss": 2.9486, + "step": 42 + }, + { + "epoch": 0.047123287671232875, + "grad_norm": 3.0076773166656494, + "learning_rate": 5.291970802919708e-07, + "loss": 2.9252, + "step": 43 + }, + { + "epoch": 0.04821917808219178, + "grad_norm": 2.79028582572937, + "learning_rate": 5.474452554744526e-07, + "loss": 2.9277, + "step": 44 + }, + { + "epoch": 0.049315068493150684, + "grad_norm": 2.603456735610962, + "learning_rate": 5.656934306569343e-07, + "loss": 2.9144, + "step": 45 + }, + { + "epoch": 0.05041095890410959, + "grad_norm": 2.674888849258423, + "learning_rate": 5.839416058394161e-07, + "loss": 2.8732, + "step": 46 + }, + { + "epoch": 0.05150684931506849, + "grad_norm": 2.972093343734741, + "learning_rate": 6.021897810218979e-07, + "loss": 2.8623, + "step": 47 + }, + { + "epoch": 0.0526027397260274, + "grad_norm": 2.757766008377075, + "learning_rate": 6.204379562043796e-07, + "loss": 2.8941, + "step": 48 + }, + { + "epoch": 0.0536986301369863, + "grad_norm": 2.910634994506836, + "learning_rate": 6.386861313868614e-07, + "loss": 2.8101, + "step": 49 + }, + { + "epoch": 0.0547945205479452, + "grad_norm": 2.5542638301849365, + "learning_rate": 6.569343065693432e-07, + "loss": 2.9262, + "step": 50 + }, + { + "epoch": 0.0547945205479452, + "eval_loss": 2.883579969406128, + "eval_runtime": 37.326, + "eval_samples_per_second": 26.791, + "eval_steps_per_second": 1.125, + "step": 50 + }, + { + "epoch": 0.05589041095890411, + "grad_norm": 2.333974599838257, + "learning_rate": 6.751824817518248e-07, + "loss": 2.8217, + "step": 51 + }, + { + "epoch": 0.05698630136986301, + "grad_norm": 2.181022882461548, + "learning_rate": 6.934306569343066e-07, + "loss": 2.8483, + "step": 52 + }, + { + "epoch": 0.05808219178082192, + "grad_norm": 2.12200927734375, + "learning_rate": 7.116788321167883e-07, + "loss": 2.7788, + "step": 53 + }, + { + "epoch": 0.05917808219178082, + "grad_norm": 2.052626371383667, + "learning_rate": 7.299270072992701e-07, + "loss": 2.7774, + "step": 54 + }, + { + "epoch": 0.06027397260273973, + "grad_norm": 2.11487078666687, + "learning_rate": 7.48175182481752e-07, + "loss": 2.8045, + "step": 55 + }, + { + "epoch": 0.06136986301369863, + "grad_norm": 2.083588123321533, + "learning_rate": 7.664233576642337e-07, + "loss": 2.8119, + "step": 56 + }, + { + "epoch": 0.06246575342465754, + "grad_norm": 2.2079808712005615, + "learning_rate": 7.846715328467154e-07, + "loss": 2.82, + "step": 57 + }, + { + "epoch": 0.06356164383561644, + "grad_norm": 2.100802421569824, + "learning_rate": 8.029197080291971e-07, + "loss": 2.794, + "step": 58 + }, + { + "epoch": 0.06465753424657535, + "grad_norm": 1.8322066068649292, + "learning_rate": 8.211678832116789e-07, + "loss": 2.7444, + "step": 59 + }, + { + "epoch": 0.06575342465753424, + "grad_norm": 2.077005386352539, + "learning_rate": 8.394160583941606e-07, + "loss": 2.7093, + "step": 60 + }, + { + "epoch": 0.06575342465753424, + "eval_loss": 2.8177874088287354, + "eval_runtime": 37.2903, + "eval_samples_per_second": 26.817, + "eval_steps_per_second": 1.126, + "step": 60 + }, + { + "epoch": 0.06684931506849315, + "grad_norm": 2.1688592433929443, + "learning_rate": 8.576642335766425e-07, + "loss": 2.7446, + "step": 61 + }, + { + "epoch": 0.06794520547945206, + "grad_norm": 2.1709237098693848, + "learning_rate": 8.759124087591242e-07, + "loss": 2.736, + "step": 62 + }, + { + "epoch": 0.06904109589041096, + "grad_norm": 2.3578364849090576, + "learning_rate": 8.94160583941606e-07, + "loss": 2.7583, + "step": 63 + }, + { + "epoch": 0.07013698630136986, + "grad_norm": 1.9025483131408691, + "learning_rate": 9.124087591240876e-07, + "loss": 2.7651, + "step": 64 + }, + { + "epoch": 0.07123287671232877, + "grad_norm": 1.9450963735580444, + "learning_rate": 9.306569343065694e-07, + "loss": 2.7186, + "step": 65 + }, + { + "epoch": 0.07232876712328767, + "grad_norm": 1.957723617553711, + "learning_rate": 9.489051094890511e-07, + "loss": 2.7421, + "step": 66 + }, + { + "epoch": 0.07342465753424658, + "grad_norm": 1.9823631048202515, + "learning_rate": 9.67153284671533e-07, + "loss": 2.7258, + "step": 67 + }, + { + "epoch": 0.07452054794520548, + "grad_norm": 1.945495843887329, + "learning_rate": 9.854014598540146e-07, + "loss": 2.7733, + "step": 68 + }, + { + "epoch": 0.07561643835616438, + "grad_norm": 1.9684759378433228, + "learning_rate": 1.0036496350364965e-06, + "loss": 2.7041, + "step": 69 + }, + { + "epoch": 0.07671232876712329, + "grad_norm": 2.087336540222168, + "learning_rate": 1.0218978102189781e-06, + "loss": 2.6662, + "step": 70 + }, + { + "epoch": 0.07671232876712329, + "eval_loss": 2.7529091835021973, + "eval_runtime": 37.3122, + "eval_samples_per_second": 26.801, + "eval_steps_per_second": 1.126, + "step": 70 + }, + { + "epoch": 0.07780821917808219, + "grad_norm": 1.7790117263793945, + "learning_rate": 1.04014598540146e-06, + "loss": 2.7379, + "step": 71 + }, + { + "epoch": 0.0789041095890411, + "grad_norm": 2.2780861854553223, + "learning_rate": 1.0583941605839416e-06, + "loss": 2.6234, + "step": 72 + }, + { + "epoch": 0.08, + "grad_norm": 2.1144700050354004, + "learning_rate": 1.0766423357664233e-06, + "loss": 2.6248, + "step": 73 + }, + { + "epoch": 0.08109589041095891, + "grad_norm": 2.0247490406036377, + "learning_rate": 1.0948905109489052e-06, + "loss": 2.6456, + "step": 74 + }, + { + "epoch": 0.0821917808219178, + "grad_norm": 2.702547311782837, + "learning_rate": 1.113138686131387e-06, + "loss": 2.6759, + "step": 75 + }, + { + "epoch": 0.08328767123287671, + "grad_norm": 1.8208750486373901, + "learning_rate": 1.1313868613138687e-06, + "loss": 2.6845, + "step": 76 + }, + { + "epoch": 0.08438356164383562, + "grad_norm": 2.3829221725463867, + "learning_rate": 1.1496350364963505e-06, + "loss": 2.61, + "step": 77 + }, + { + "epoch": 0.08547945205479453, + "grad_norm": 2.580885887145996, + "learning_rate": 1.1678832116788322e-06, + "loss": 2.5896, + "step": 78 + }, + { + "epoch": 0.08657534246575342, + "grad_norm": 2.2477641105651855, + "learning_rate": 1.1861313868613138e-06, + "loss": 2.596, + "step": 79 + }, + { + "epoch": 0.08767123287671233, + "grad_norm": 2.1895623207092285, + "learning_rate": 1.2043795620437959e-06, + "loss": 2.5906, + "step": 80 + }, + { + "epoch": 0.08767123287671233, + "eval_loss": 2.6815176010131836, + "eval_runtime": 37.3342, + "eval_samples_per_second": 26.785, + "eval_steps_per_second": 1.125, + "step": 80 + }, + { + "epoch": 0.08876712328767124, + "grad_norm": 2.217090606689453, + "learning_rate": 1.2226277372262775e-06, + "loss": 2.6005, + "step": 81 + }, + { + "epoch": 0.08986301369863013, + "grad_norm": 2.4679362773895264, + "learning_rate": 1.2408759124087592e-06, + "loss": 2.6333, + "step": 82 + }, + { + "epoch": 0.09095890410958904, + "grad_norm": 2.2590575218200684, + "learning_rate": 1.259124087591241e-06, + "loss": 2.5814, + "step": 83 + }, + { + "epoch": 0.09205479452054795, + "grad_norm": 1.877977728843689, + "learning_rate": 1.2773722627737229e-06, + "loss": 2.5948, + "step": 84 + }, + { + "epoch": 0.09315068493150686, + "grad_norm": 2.1535534858703613, + "learning_rate": 1.2956204379562045e-06, + "loss": 2.5781, + "step": 85 + }, + { + "epoch": 0.09424657534246575, + "grad_norm": 2.5001144409179688, + "learning_rate": 1.3138686131386864e-06, + "loss": 2.5339, + "step": 86 + }, + { + "epoch": 0.09534246575342466, + "grad_norm": 2.3279049396514893, + "learning_rate": 1.332116788321168e-06, + "loss": 2.545, + "step": 87 + }, + { + "epoch": 0.09643835616438357, + "grad_norm": 2.2230734825134277, + "learning_rate": 1.3503649635036497e-06, + "loss": 2.4745, + "step": 88 + }, + { + "epoch": 0.09753424657534246, + "grad_norm": 2.8762636184692383, + "learning_rate": 1.3686131386861315e-06, + "loss": 2.516, + "step": 89 + }, + { + "epoch": 0.09863013698630137, + "grad_norm": 2.153506278991699, + "learning_rate": 1.3868613138686132e-06, + "loss": 2.4547, + "step": 90 + }, + { + "epoch": 0.09863013698630137, + "eval_loss": 2.603238344192505, + "eval_runtime": 37.4162, + "eval_samples_per_second": 26.726, + "eval_steps_per_second": 1.123, + "step": 90 + }, + { + "epoch": 0.09972602739726028, + "grad_norm": 2.4143614768981934, + "learning_rate": 1.405109489051095e-06, + "loss": 2.547, + "step": 91 + }, + { + "epoch": 0.10082191780821918, + "grad_norm": 2.354632616043091, + "learning_rate": 1.4233576642335767e-06, + "loss": 2.5009, + "step": 92 + }, + { + "epoch": 0.10191780821917808, + "grad_norm": 2.6605563163757324, + "learning_rate": 1.4416058394160583e-06, + "loss": 2.5304, + "step": 93 + }, + { + "epoch": 0.10301369863013699, + "grad_norm": 2.3724122047424316, + "learning_rate": 1.4598540145985402e-06, + "loss": 2.4925, + "step": 94 + }, + { + "epoch": 0.10410958904109589, + "grad_norm": 3.6323022842407227, + "learning_rate": 1.478102189781022e-06, + "loss": 2.4592, + "step": 95 + }, + { + "epoch": 0.1052054794520548, + "grad_norm": 2.3407089710235596, + "learning_rate": 1.496350364963504e-06, + "loss": 2.5223, + "step": 96 + }, + { + "epoch": 0.1063013698630137, + "grad_norm": 2.8238413333892822, + "learning_rate": 1.5145985401459856e-06, + "loss": 2.489, + "step": 97 + }, + { + "epoch": 0.1073972602739726, + "grad_norm": 2.6433658599853516, + "learning_rate": 1.5328467153284674e-06, + "loss": 2.5101, + "step": 98 + }, + { + "epoch": 0.10849315068493151, + "grad_norm": 2.562761068344116, + "learning_rate": 1.551094890510949e-06, + "loss": 2.4642, + "step": 99 + }, + { + "epoch": 0.1095890410958904, + "grad_norm": 2.4501149654388428, + "learning_rate": 1.5693430656934307e-06, + "loss": 2.4102, + "step": 100 + }, + { + "epoch": 0.1095890410958904, + "eval_loss": 2.5452804565429688, + "eval_runtime": 37.4224, + "eval_samples_per_second": 26.722, + "eval_steps_per_second": 1.122, + "step": 100 + }, + { + "epoch": 0.11068493150684931, + "grad_norm": 2.475864887237549, + "learning_rate": 1.5875912408759126e-06, + "loss": 2.4147, + "step": 101 + }, + { + "epoch": 0.11178082191780822, + "grad_norm": 2.558875322341919, + "learning_rate": 1.6058394160583942e-06, + "loss": 2.4694, + "step": 102 + }, + { + "epoch": 0.11287671232876713, + "grad_norm": 2.1085734367370605, + "learning_rate": 1.624087591240876e-06, + "loss": 2.5099, + "step": 103 + }, + { + "epoch": 0.11397260273972602, + "grad_norm": 2.410731792449951, + "learning_rate": 1.6423357664233577e-06, + "loss": 2.432, + "step": 104 + }, + { + "epoch": 0.11506849315068493, + "grad_norm": 2.2800745964050293, + "learning_rate": 1.6605839416058394e-06, + "loss": 2.3669, + "step": 105 + }, + { + "epoch": 0.11616438356164384, + "grad_norm": 2.2234933376312256, + "learning_rate": 1.6788321167883212e-06, + "loss": 2.4145, + "step": 106 + }, + { + "epoch": 0.11726027397260275, + "grad_norm": 2.3367040157318115, + "learning_rate": 1.6970802919708029e-06, + "loss": 2.4272, + "step": 107 + }, + { + "epoch": 0.11835616438356164, + "grad_norm": 2.7002811431884766, + "learning_rate": 1.715328467153285e-06, + "loss": 2.3478, + "step": 108 + }, + { + "epoch": 0.11945205479452055, + "grad_norm": 2.6628332138061523, + "learning_rate": 1.7335766423357666e-06, + "loss": 2.4096, + "step": 109 + }, + { + "epoch": 0.12054794520547946, + "grad_norm": 2.4174230098724365, + "learning_rate": 1.7518248175182485e-06, + "loss": 2.407, + "step": 110 + }, + { + "epoch": 0.12054794520547946, + "eval_loss": 2.491746664047241, + "eval_runtime": 37.2555, + "eval_samples_per_second": 26.842, + "eval_steps_per_second": 1.127, + "step": 110 + }, + { + "epoch": 0.12164383561643835, + "grad_norm": 2.24796986579895, + "learning_rate": 1.77007299270073e-06, + "loss": 2.4369, + "step": 111 + }, + { + "epoch": 0.12273972602739726, + "grad_norm": 2.720939874649048, + "learning_rate": 1.788321167883212e-06, + "loss": 2.4116, + "step": 112 + }, + { + "epoch": 0.12383561643835617, + "grad_norm": 2.6907670497894287, + "learning_rate": 1.8065693430656936e-06, + "loss": 2.4577, + "step": 113 + }, + { + "epoch": 0.12493150684931507, + "grad_norm": 2.559636354446411, + "learning_rate": 1.8248175182481753e-06, + "loss": 2.3999, + "step": 114 + }, + { + "epoch": 0.12602739726027398, + "grad_norm": 2.6730551719665527, + "learning_rate": 1.8430656934306571e-06, + "loss": 2.3428, + "step": 115 + }, + { + "epoch": 0.12712328767123288, + "grad_norm": 3.172179937362671, + "learning_rate": 1.8613138686131388e-06, + "loss": 2.3722, + "step": 116 + }, + { + "epoch": 0.12821917808219177, + "grad_norm": 2.6757171154022217, + "learning_rate": 1.8795620437956206e-06, + "loss": 2.3572, + "step": 117 + }, + { + "epoch": 0.1293150684931507, + "grad_norm": 2.5840766429901123, + "learning_rate": 1.8978102189781023e-06, + "loss": 2.353, + "step": 118 + }, + { + "epoch": 0.1304109589041096, + "grad_norm": 2.6248581409454346, + "learning_rate": 1.916058394160584e-06, + "loss": 2.36, + "step": 119 + }, + { + "epoch": 0.13150684931506848, + "grad_norm": 2.428516387939453, + "learning_rate": 1.934306569343066e-06, + "loss": 2.3821, + "step": 120 + }, + { + "epoch": 0.13150684931506848, + "eval_loss": 2.451565980911255, + "eval_runtime": 37.3669, + "eval_samples_per_second": 26.762, + "eval_steps_per_second": 1.124, + "step": 120 + }, + { + "epoch": 0.1326027397260274, + "grad_norm": 2.9956233501434326, + "learning_rate": 1.952554744525548e-06, + "loss": 2.2778, + "step": 121 + }, + { + "epoch": 0.1336986301369863, + "grad_norm": 2.4120893478393555, + "learning_rate": 1.9708029197080293e-06, + "loss": 2.325, + "step": 122 + }, + { + "epoch": 0.13479452054794522, + "grad_norm": 2.1410024166107178, + "learning_rate": 1.989051094890511e-06, + "loss": 2.3248, + "step": 123 + }, + { + "epoch": 0.1358904109589041, + "grad_norm": 2.923799514770508, + "learning_rate": 2.007299270072993e-06, + "loss": 2.327, + "step": 124 + }, + { + "epoch": 0.136986301369863, + "grad_norm": 3.0770339965820312, + "learning_rate": 2.025547445255475e-06, + "loss": 2.3847, + "step": 125 + }, + { + "epoch": 0.13808219178082193, + "grad_norm": 2.6408181190490723, + "learning_rate": 2.0437956204379563e-06, + "loss": 2.3196, + "step": 126 + }, + { + "epoch": 0.13917808219178082, + "grad_norm": 2.9127118587493896, + "learning_rate": 2.062043795620438e-06, + "loss": 2.3287, + "step": 127 + }, + { + "epoch": 0.14027397260273972, + "grad_norm": 2.9805052280426025, + "learning_rate": 2.08029197080292e-06, + "loss": 2.3243, + "step": 128 + }, + { + "epoch": 0.14136986301369864, + "grad_norm": 3.1342010498046875, + "learning_rate": 2.0985401459854014e-06, + "loss": 2.38, + "step": 129 + }, + { + "epoch": 0.14246575342465753, + "grad_norm": 2.678422451019287, + "learning_rate": 2.1167883211678833e-06, + "loss": 2.2727, + "step": 130 + }, + { + "epoch": 0.14246575342465753, + "eval_loss": 2.4187021255493164, + "eval_runtime": 37.3117, + "eval_samples_per_second": 26.801, + "eval_steps_per_second": 1.126, + "step": 130 + }, + { + "epoch": 0.14356164383561643, + "grad_norm": 2.9281492233276367, + "learning_rate": 2.135036496350365e-06, + "loss": 2.3327, + "step": 131 + }, + { + "epoch": 0.14465753424657535, + "grad_norm": 2.9592933654785156, + "learning_rate": 2.1532846715328466e-06, + "loss": 2.3674, + "step": 132 + }, + { + "epoch": 0.14575342465753424, + "grad_norm": 3.404761552810669, + "learning_rate": 2.171532846715329e-06, + "loss": 2.2679, + "step": 133 + }, + { + "epoch": 0.14684931506849316, + "grad_norm": 3.635571241378784, + "learning_rate": 2.1897810218978103e-06, + "loss": 2.2799, + "step": 134 + }, + { + "epoch": 0.14794520547945206, + "grad_norm": 3.9130170345306396, + "learning_rate": 2.208029197080292e-06, + "loss": 2.3263, + "step": 135 + }, + { + "epoch": 0.14904109589041095, + "grad_norm": 4.300438404083252, + "learning_rate": 2.226277372262774e-06, + "loss": 2.2969, + "step": 136 + }, + { + "epoch": 0.15013698630136987, + "grad_norm": 2.863511085510254, + "learning_rate": 2.244525547445256e-06, + "loss": 2.297, + "step": 137 + }, + { + "epoch": 0.15123287671232877, + "grad_norm": 4.910321235656738, + "learning_rate": 2.2627737226277373e-06, + "loss": 2.3269, + "step": 138 + }, + { + "epoch": 0.15232876712328766, + "grad_norm": 3.3323862552642822, + "learning_rate": 2.281021897810219e-06, + "loss": 2.3195, + "step": 139 + }, + { + "epoch": 0.15342465753424658, + "grad_norm": 3.2672693729400635, + "learning_rate": 2.299270072992701e-06, + "loss": 2.2797, + "step": 140 + }, + { + "epoch": 0.15342465753424658, + "eval_loss": 2.3943886756896973, + "eval_runtime": 37.3137, + "eval_samples_per_second": 26.8, + "eval_steps_per_second": 1.126, + "step": 140 + }, + { + "epoch": 0.15452054794520548, + "grad_norm": 3.9229774475097656, + "learning_rate": 2.3175182481751825e-06, + "loss": 2.2718, + "step": 141 + }, + { + "epoch": 0.15561643835616437, + "grad_norm": 2.6409823894500732, + "learning_rate": 2.3357664233576643e-06, + "loss": 2.2828, + "step": 142 + }, + { + "epoch": 0.1567123287671233, + "grad_norm": 4.6122612953186035, + "learning_rate": 2.354014598540146e-06, + "loss": 2.247, + "step": 143 + }, + { + "epoch": 0.1578082191780822, + "grad_norm": 3.391357898712158, + "learning_rate": 2.3722627737226276e-06, + "loss": 2.2863, + "step": 144 + }, + { + "epoch": 0.1589041095890411, + "grad_norm": 3.3550853729248047, + "learning_rate": 2.39051094890511e-06, + "loss": 2.2789, + "step": 145 + }, + { + "epoch": 0.16, + "grad_norm": 3.7376883029937744, + "learning_rate": 2.4087591240875918e-06, + "loss": 2.2297, + "step": 146 + }, + { + "epoch": 0.1610958904109589, + "grad_norm": 3.005157709121704, + "learning_rate": 2.427007299270073e-06, + "loss": 2.2467, + "step": 147 + }, + { + "epoch": 0.16219178082191782, + "grad_norm": 3.4616103172302246, + "learning_rate": 2.445255474452555e-06, + "loss": 2.2612, + "step": 148 + }, + { + "epoch": 0.1632876712328767, + "grad_norm": 3.3595101833343506, + "learning_rate": 2.463503649635037e-06, + "loss": 2.3021, + "step": 149 + }, + { + "epoch": 0.1643835616438356, + "grad_norm": 4.831241607666016, + "learning_rate": 2.4817518248175183e-06, + "loss": 2.2174, + "step": 150 + }, + { + "epoch": 0.1643835616438356, + "eval_loss": 2.3677749633789062, + "eval_runtime": 37.3092, + "eval_samples_per_second": 26.803, + "eval_steps_per_second": 1.126, + "step": 150 + }, + { + "epoch": 0.16547945205479453, + "grad_norm": 2.778820037841797, + "learning_rate": 2.5e-06, + "loss": 2.1856, + "step": 151 + }, + { + "epoch": 0.16657534246575342, + "grad_norm": 4.566643714904785, + "learning_rate": 2.518248175182482e-06, + "loss": 2.3186, + "step": 152 + }, + { + "epoch": 0.16767123287671232, + "grad_norm": 2.9908089637756348, + "learning_rate": 2.5364963503649635e-06, + "loss": 2.2127, + "step": 153 + }, + { + "epoch": 0.16876712328767124, + "grad_norm": 2.9702792167663574, + "learning_rate": 2.5547445255474458e-06, + "loss": 2.2071, + "step": 154 + }, + { + "epoch": 0.16986301369863013, + "grad_norm": 2.850044012069702, + "learning_rate": 2.5729927007299272e-06, + "loss": 2.2585, + "step": 155 + }, + { + "epoch": 0.17095890410958905, + "grad_norm": 2.926091432571411, + "learning_rate": 2.591240875912409e-06, + "loss": 2.2397, + "step": 156 + }, + { + "epoch": 0.17205479452054795, + "grad_norm": 2.8720169067382812, + "learning_rate": 2.6094890510948905e-06, + "loss": 2.2306, + "step": 157 + }, + { + "epoch": 0.17315068493150684, + "grad_norm": 2.8837177753448486, + "learning_rate": 2.627737226277373e-06, + "loss": 2.2666, + "step": 158 + }, + { + "epoch": 0.17424657534246576, + "grad_norm": 3.119621753692627, + "learning_rate": 2.6459854014598542e-06, + "loss": 2.2722, + "step": 159 + }, + { + "epoch": 0.17534246575342466, + "grad_norm": 2.7085700035095215, + "learning_rate": 2.664233576642336e-06, + "loss": 2.2346, + "step": 160 + }, + { + "epoch": 0.17534246575342466, + "eval_loss": 2.352083921432495, + "eval_runtime": 37.245, + "eval_samples_per_second": 26.849, + "eval_steps_per_second": 1.128, + "step": 160 + }, + { + "epoch": 0.17643835616438355, + "grad_norm": 2.9964346885681152, + "learning_rate": 2.6824817518248175e-06, + "loss": 2.2155, + "step": 161 + }, + { + "epoch": 0.17753424657534247, + "grad_norm": 3.1535208225250244, + "learning_rate": 2.7007299270072994e-06, + "loss": 2.235, + "step": 162 + }, + { + "epoch": 0.17863013698630137, + "grad_norm": 2.7240943908691406, + "learning_rate": 2.718978102189781e-06, + "loss": 2.2095, + "step": 163 + }, + { + "epoch": 0.17972602739726026, + "grad_norm": 3.122654676437378, + "learning_rate": 2.737226277372263e-06, + "loss": 2.2667, + "step": 164 + }, + { + "epoch": 0.18082191780821918, + "grad_norm": 2.9718410968780518, + "learning_rate": 2.755474452554745e-06, + "loss": 2.2499, + "step": 165 + }, + { + "epoch": 0.18191780821917808, + "grad_norm": 3.0081896781921387, + "learning_rate": 2.7737226277372264e-06, + "loss": 2.2089, + "step": 166 + }, + { + "epoch": 0.18301369863013697, + "grad_norm": 2.829509973526001, + "learning_rate": 2.7919708029197082e-06, + "loss": 2.3139, + "step": 167 + }, + { + "epoch": 0.1841095890410959, + "grad_norm": 3.4480397701263428, + "learning_rate": 2.81021897810219e-06, + "loss": 2.2705, + "step": 168 + }, + { + "epoch": 0.1852054794520548, + "grad_norm": 3.1107025146484375, + "learning_rate": 2.828467153284672e-06, + "loss": 2.1895, + "step": 169 + }, + { + "epoch": 0.1863013698630137, + "grad_norm": 2.916790008544922, + "learning_rate": 2.8467153284671534e-06, + "loss": 2.2317, + "step": 170 + }, + { + "epoch": 0.1863013698630137, + "eval_loss": 2.336153030395508, + "eval_runtime": 37.2401, + "eval_samples_per_second": 26.853, + "eval_steps_per_second": 1.128, + "step": 170 + }, + { + "epoch": 0.1873972602739726, + "grad_norm": 3.2582287788391113, + "learning_rate": 2.8649635036496353e-06, + "loss": 2.143, + "step": 171 + }, + { + "epoch": 0.1884931506849315, + "grad_norm": 2.430973768234253, + "learning_rate": 2.8832116788321167e-06, + "loss": 2.1977, + "step": 172 + }, + { + "epoch": 0.18958904109589042, + "grad_norm": 2.8763599395751953, + "learning_rate": 2.901459854014599e-06, + "loss": 2.1551, + "step": 173 + }, + { + "epoch": 0.19068493150684931, + "grad_norm": 2.9242653846740723, + "learning_rate": 2.9197080291970804e-06, + "loss": 2.2569, + "step": 174 + }, + { + "epoch": 0.1917808219178082, + "grad_norm": 2.623551607131958, + "learning_rate": 2.9379562043795623e-06, + "loss": 2.2329, + "step": 175 + }, + { + "epoch": 0.19287671232876713, + "grad_norm": 2.9511337280273438, + "learning_rate": 2.956204379562044e-06, + "loss": 2.2154, + "step": 176 + }, + { + "epoch": 0.19397260273972602, + "grad_norm": 2.415118932723999, + "learning_rate": 2.9744525547445256e-06, + "loss": 2.2637, + "step": 177 + }, + { + "epoch": 0.19506849315068492, + "grad_norm": 2.9664924144744873, + "learning_rate": 2.992700729927008e-06, + "loss": 2.2263, + "step": 178 + }, + { + "epoch": 0.19616438356164384, + "grad_norm": 2.7214572429656982, + "learning_rate": 3.0109489051094893e-06, + "loss": 2.1667, + "step": 179 + }, + { + "epoch": 0.19726027397260273, + "grad_norm": 2.8712823390960693, + "learning_rate": 3.029197080291971e-06, + "loss": 2.1661, + "step": 180 + }, + { + "epoch": 0.19726027397260273, + "eval_loss": 2.3191800117492676, + "eval_runtime": 37.24, + "eval_samples_per_second": 26.853, + "eval_steps_per_second": 1.128, + "step": 180 + }, + { + "epoch": 0.19835616438356166, + "grad_norm": 2.959709405899048, + "learning_rate": 3.0474452554744526e-06, + "loss": 2.2483, + "step": 181 + }, + { + "epoch": 0.19945205479452055, + "grad_norm": 2.567089319229126, + "learning_rate": 3.065693430656935e-06, + "loss": 2.2382, + "step": 182 + }, + { + "epoch": 0.20054794520547944, + "grad_norm": 2.47188401222229, + "learning_rate": 3.0839416058394163e-06, + "loss": 2.2148, + "step": 183 + }, + { + "epoch": 0.20164383561643837, + "grad_norm": 2.3435261249542236, + "learning_rate": 3.102189781021898e-06, + "loss": 2.2334, + "step": 184 + }, + { + "epoch": 0.20273972602739726, + "grad_norm": 2.5280468463897705, + "learning_rate": 3.1204379562043796e-06, + "loss": 2.2214, + "step": 185 + }, + { + "epoch": 0.20383561643835615, + "grad_norm": 2.4808971881866455, + "learning_rate": 3.1386861313868614e-06, + "loss": 2.1554, + "step": 186 + }, + { + "epoch": 0.20493150684931508, + "grad_norm": 2.9937026500701904, + "learning_rate": 3.156934306569343e-06, + "loss": 2.1389, + "step": 187 + }, + { + "epoch": 0.20602739726027397, + "grad_norm": 2.9220476150512695, + "learning_rate": 3.175182481751825e-06, + "loss": 2.2515, + "step": 188 + }, + { + "epoch": 0.20712328767123286, + "grad_norm": 3.0449001789093018, + "learning_rate": 3.193430656934307e-06, + "loss": 2.2016, + "step": 189 + }, + { + "epoch": 0.20821917808219179, + "grad_norm": 2.8027074337005615, + "learning_rate": 3.2116788321167884e-06, + "loss": 2.1823, + "step": 190 + }, + { + "epoch": 0.20821917808219179, + "eval_loss": 2.309677839279175, + "eval_runtime": 37.2442, + "eval_samples_per_second": 26.85, + "eval_steps_per_second": 1.128, + "step": 190 + }, + { + "epoch": 0.20931506849315068, + "grad_norm": 2.9158318042755127, + "learning_rate": 3.2299270072992707e-06, + "loss": 2.1928, + "step": 191 + }, + { + "epoch": 0.2104109589041096, + "grad_norm": 2.5328080654144287, + "learning_rate": 3.248175182481752e-06, + "loss": 2.1828, + "step": 192 + }, + { + "epoch": 0.2115068493150685, + "grad_norm": 3.1709704399108887, + "learning_rate": 3.266423357664234e-06, + "loss": 2.1052, + "step": 193 + }, + { + "epoch": 0.2126027397260274, + "grad_norm": 2.9156367778778076, + "learning_rate": 3.2846715328467155e-06, + "loss": 2.1943, + "step": 194 + }, + { + "epoch": 0.2136986301369863, + "grad_norm": 2.879408836364746, + "learning_rate": 3.3029197080291973e-06, + "loss": 2.1782, + "step": 195 + }, + { + "epoch": 0.2147945205479452, + "grad_norm": 3.748483180999756, + "learning_rate": 3.3211678832116788e-06, + "loss": 2.2107, + "step": 196 + }, + { + "epoch": 0.2158904109589041, + "grad_norm": 3.197510004043579, + "learning_rate": 3.339416058394161e-06, + "loss": 2.1845, + "step": 197 + }, + { + "epoch": 0.21698630136986302, + "grad_norm": 3.2596049308776855, + "learning_rate": 3.3576642335766425e-06, + "loss": 2.1816, + "step": 198 + }, + { + "epoch": 0.21808219178082192, + "grad_norm": 3.3329780101776123, + "learning_rate": 3.3759124087591243e-06, + "loss": 2.181, + "step": 199 + }, + { + "epoch": 0.2191780821917808, + "grad_norm": 2.7302327156066895, + "learning_rate": 3.3941605839416058e-06, + "loss": 2.1761, + "step": 200 + }, + { + "epoch": 0.2191780821917808, + "eval_loss": 2.2909653186798096, + "eval_runtime": 37.284, + "eval_samples_per_second": 26.821, + "eval_steps_per_second": 1.126, + "step": 200 + }, + { + "epoch": 0.22027397260273973, + "grad_norm": 3.002255916595459, + "learning_rate": 3.412408759124088e-06, + "loss": 2.1868, + "step": 201 + }, + { + "epoch": 0.22136986301369863, + "grad_norm": 2.6694891452789307, + "learning_rate": 3.43065693430657e-06, + "loss": 2.192, + "step": 202 + }, + { + "epoch": 0.22246575342465755, + "grad_norm": 2.9225986003875732, + "learning_rate": 3.4489051094890513e-06, + "loss": 2.1303, + "step": 203 + }, + { + "epoch": 0.22356164383561644, + "grad_norm": 2.7844080924987793, + "learning_rate": 3.467153284671533e-06, + "loss": 2.1788, + "step": 204 + }, + { + "epoch": 0.22465753424657534, + "grad_norm": 2.726123332977295, + "learning_rate": 3.4854014598540146e-06, + "loss": 2.1193, + "step": 205 + }, + { + "epoch": 0.22575342465753426, + "grad_norm": 2.838040590286255, + "learning_rate": 3.503649635036497e-06, + "loss": 2.1674, + "step": 206 + }, + { + "epoch": 0.22684931506849315, + "grad_norm": 2.6863937377929688, + "learning_rate": 3.5218978102189783e-06, + "loss": 2.161, + "step": 207 + }, + { + "epoch": 0.22794520547945205, + "grad_norm": 2.5116159915924072, + "learning_rate": 3.54014598540146e-06, + "loss": 2.179, + "step": 208 + }, + { + "epoch": 0.22904109589041097, + "grad_norm": 3.381863594055176, + "learning_rate": 3.5583941605839416e-06, + "loss": 2.2035, + "step": 209 + }, + { + "epoch": 0.23013698630136986, + "grad_norm": 3.2741382122039795, + "learning_rate": 3.576642335766424e-06, + "loss": 2.2518, + "step": 210 + }, + { + "epoch": 0.23013698630136986, + "eval_loss": 2.2773635387420654, + "eval_runtime": 37.2658, + "eval_samples_per_second": 26.834, + "eval_steps_per_second": 1.127, + "step": 210 + }, + { + "epoch": 0.23123287671232876, + "grad_norm": 2.917370319366455, + "learning_rate": 3.5948905109489054e-06, + "loss": 2.2265, + "step": 211 + }, + { + "epoch": 0.23232876712328768, + "grad_norm": 2.8392834663391113, + "learning_rate": 3.6131386861313872e-06, + "loss": 2.0709, + "step": 212 + }, + { + "epoch": 0.23342465753424657, + "grad_norm": 2.753612518310547, + "learning_rate": 3.6313868613138687e-06, + "loss": 2.0752, + "step": 213 + }, + { + "epoch": 0.2345205479452055, + "grad_norm": 3.1477746963500977, + "learning_rate": 3.6496350364963505e-06, + "loss": 2.1169, + "step": 214 + }, + { + "epoch": 0.2356164383561644, + "grad_norm": 2.5448100566864014, + "learning_rate": 3.667883211678833e-06, + "loss": 2.1338, + "step": 215 + }, + { + "epoch": 0.23671232876712328, + "grad_norm": 3.1622560024261475, + "learning_rate": 3.6861313868613142e-06, + "loss": 2.1065, + "step": 216 + }, + { + "epoch": 0.2378082191780822, + "grad_norm": 2.7054336071014404, + "learning_rate": 3.704379562043796e-06, + "loss": 2.1104, + "step": 217 + }, + { + "epoch": 0.2389041095890411, + "grad_norm": 2.964141845703125, + "learning_rate": 3.7226277372262775e-06, + "loss": 2.1949, + "step": 218 + }, + { + "epoch": 0.24, + "grad_norm": 2.7586112022399902, + "learning_rate": 3.7408759124087594e-06, + "loss": 2.165, + "step": 219 + }, + { + "epoch": 0.2410958904109589, + "grad_norm": 2.9035964012145996, + "learning_rate": 3.7591240875912412e-06, + "loss": 2.1192, + "step": 220 + }, + { + "epoch": 0.2410958904109589, + "eval_loss": 2.2649471759796143, + "eval_runtime": 37.4311, + "eval_samples_per_second": 26.716, + "eval_steps_per_second": 1.122, + "step": 220 + }, + { + "epoch": 0.2421917808219178, + "grad_norm": 2.4212076663970947, + "learning_rate": 3.777372262773723e-06, + "loss": 2.1389, + "step": 221 + }, + { + "epoch": 0.2432876712328767, + "grad_norm": 2.973227024078369, + "learning_rate": 3.7956204379562045e-06, + "loss": 2.132, + "step": 222 + }, + { + "epoch": 0.24438356164383562, + "grad_norm": 2.9831430912017822, + "learning_rate": 3.8138686131386864e-06, + "loss": 2.1142, + "step": 223 + }, + { + "epoch": 0.24547945205479452, + "grad_norm": 3.2880353927612305, + "learning_rate": 3.832116788321168e-06, + "loss": 2.168, + "step": 224 + }, + { + "epoch": 0.2465753424657534, + "grad_norm": 3.3920998573303223, + "learning_rate": 3.85036496350365e-06, + "loss": 2.1277, + "step": 225 + }, + { + "epoch": 0.24767123287671233, + "grad_norm": 2.7170026302337646, + "learning_rate": 3.868613138686132e-06, + "loss": 2.0936, + "step": 226 + }, + { + "epoch": 0.24876712328767123, + "grad_norm": 3.6220011711120605, + "learning_rate": 3.886861313868613e-06, + "loss": 2.1259, + "step": 227 + }, + { + "epoch": 0.24986301369863015, + "grad_norm": 3.4229273796081543, + "learning_rate": 3.905109489051096e-06, + "loss": 2.1421, + "step": 228 + }, + { + "epoch": 0.25095890410958904, + "grad_norm": 3.0358338356018066, + "learning_rate": 3.923357664233577e-06, + "loss": 2.1747, + "step": 229 + }, + { + "epoch": 0.25205479452054796, + "grad_norm": 2.611771821975708, + "learning_rate": 3.9416058394160585e-06, + "loss": 2.1282, + "step": 230 + }, + { + "epoch": 0.25205479452054796, + "eval_loss": 2.254887580871582, + "eval_runtime": 37.5943, + "eval_samples_per_second": 26.6, + "eval_steps_per_second": 1.117, + "step": 230 + }, + { + "epoch": 0.25315068493150683, + "grad_norm": 3.846412181854248, + "learning_rate": 3.95985401459854e-06, + "loss": 2.1892, + "step": 231 + }, + { + "epoch": 0.25424657534246575, + "grad_norm": 2.488525390625, + "learning_rate": 3.978102189781022e-06, + "loss": 2.2064, + "step": 232 + }, + { + "epoch": 0.2553424657534247, + "grad_norm": 3.114964485168457, + "learning_rate": 3.996350364963504e-06, + "loss": 2.0653, + "step": 233 + }, + { + "epoch": 0.25643835616438354, + "grad_norm": 2.5559768676757812, + "learning_rate": 4.014598540145986e-06, + "loss": 2.1503, + "step": 234 + }, + { + "epoch": 0.25753424657534246, + "grad_norm": 3.2347540855407715, + "learning_rate": 4.032846715328467e-06, + "loss": 2.0523, + "step": 235 + }, + { + "epoch": 0.2586301369863014, + "grad_norm": 2.8086111545562744, + "learning_rate": 4.05109489051095e-06, + "loss": 2.1206, + "step": 236 + }, + { + "epoch": 0.25972602739726025, + "grad_norm": 2.909231185913086, + "learning_rate": 4.069343065693431e-06, + "loss": 2.139, + "step": 237 + }, + { + "epoch": 0.2608219178082192, + "grad_norm": 3.2171082496643066, + "learning_rate": 4.0875912408759126e-06, + "loss": 2.0908, + "step": 238 + }, + { + "epoch": 0.2619178082191781, + "grad_norm": 2.3851265907287598, + "learning_rate": 4.105839416058395e-06, + "loss": 2.1133, + "step": 239 + }, + { + "epoch": 0.26301369863013696, + "grad_norm": 3.3912737369537354, + "learning_rate": 4.124087591240876e-06, + "loss": 2.1911, + "step": 240 + }, + { + "epoch": 0.26301369863013696, + "eval_loss": 2.2436957359313965, + "eval_runtime": 37.2947, + "eval_samples_per_second": 26.813, + "eval_steps_per_second": 1.126, + "step": 240 + }, + { + "epoch": 0.2641095890410959, + "grad_norm": 2.82906174659729, + "learning_rate": 4.1423357664233586e-06, + "loss": 2.11, + "step": 241 + }, + { + "epoch": 0.2652054794520548, + "grad_norm": 3.0317952632904053, + "learning_rate": 4.16058394160584e-06, + "loss": 2.0953, + "step": 242 + }, + { + "epoch": 0.2663013698630137, + "grad_norm": 3.6438329219818115, + "learning_rate": 4.1788321167883214e-06, + "loss": 2.1746, + "step": 243 + }, + { + "epoch": 0.2673972602739726, + "grad_norm": 2.6690592765808105, + "learning_rate": 4.197080291970803e-06, + "loss": 2.0555, + "step": 244 + }, + { + "epoch": 0.2684931506849315, + "grad_norm": 3.135376453399658, + "learning_rate": 4.215328467153285e-06, + "loss": 2.1394, + "step": 245 + }, + { + "epoch": 0.26958904109589044, + "grad_norm": 2.5474987030029297, + "learning_rate": 4.233576642335767e-06, + "loss": 2.0822, + "step": 246 + }, + { + "epoch": 0.2706849315068493, + "grad_norm": 2.883108377456665, + "learning_rate": 4.251824817518249e-06, + "loss": 2.0869, + "step": 247 + }, + { + "epoch": 0.2717808219178082, + "grad_norm": 2.9801037311553955, + "learning_rate": 4.27007299270073e-06, + "loss": 2.1194, + "step": 248 + }, + { + "epoch": 0.27287671232876715, + "grad_norm": 2.7688217163085938, + "learning_rate": 4.288321167883212e-06, + "loss": 2.0908, + "step": 249 + }, + { + "epoch": 0.273972602739726, + "grad_norm": 3.205148458480835, + "learning_rate": 4.306569343065693e-06, + "loss": 2.0762, + "step": 250 + }, + { + "epoch": 0.273972602739726, + "eval_loss": 2.238699197769165, + "eval_runtime": 37.2428, + "eval_samples_per_second": 26.851, + "eval_steps_per_second": 1.128, + "step": 250 + }, + { + "epoch": 0.27506849315068493, + "grad_norm": 2.8262946605682373, + "learning_rate": 4.3248175182481755e-06, + "loss": 2.0509, + "step": 251 + }, + { + "epoch": 0.27616438356164386, + "grad_norm": 3.156261682510376, + "learning_rate": 4.343065693430658e-06, + "loss": 2.1213, + "step": 252 + }, + { + "epoch": 0.2772602739726027, + "grad_norm": 2.6367998123168945, + "learning_rate": 4.361313868613139e-06, + "loss": 2.1071, + "step": 253 + }, + { + "epoch": 0.27835616438356164, + "grad_norm": 2.871534824371338, + "learning_rate": 4.379562043795621e-06, + "loss": 2.1008, + "step": 254 + }, + { + "epoch": 0.27945205479452057, + "grad_norm": 2.839432716369629, + "learning_rate": 4.397810218978102e-06, + "loss": 2.1293, + "step": 255 + }, + { + "epoch": 0.28054794520547943, + "grad_norm": 2.627143383026123, + "learning_rate": 4.416058394160584e-06, + "loss": 2.1141, + "step": 256 + }, + { + "epoch": 0.28164383561643835, + "grad_norm": 2.8292996883392334, + "learning_rate": 4.434306569343066e-06, + "loss": 2.0981, + "step": 257 + }, + { + "epoch": 0.2827397260273973, + "grad_norm": 2.5711069107055664, + "learning_rate": 4.452554744525548e-06, + "loss": 2.0707, + "step": 258 + }, + { + "epoch": 0.28383561643835614, + "grad_norm": 2.488954544067383, + "learning_rate": 4.4708029197080295e-06, + "loss": 2.1172, + "step": 259 + }, + { + "epoch": 0.28493150684931506, + "grad_norm": 2.442209005355835, + "learning_rate": 4.489051094890512e-06, + "loss": 2.1119, + "step": 260 + }, + { + "epoch": 0.28493150684931506, + "eval_loss": 2.2306911945343018, + "eval_runtime": 37.2412, + "eval_samples_per_second": 26.852, + "eval_steps_per_second": 1.128, + "step": 260 + }, + { + "epoch": 0.286027397260274, + "grad_norm": 2.9147934913635254, + "learning_rate": 4.507299270072993e-06, + "loss": 2.1517, + "step": 261 + }, + { + "epoch": 0.28712328767123285, + "grad_norm": 2.549833297729492, + "learning_rate": 4.525547445255475e-06, + "loss": 2.1047, + "step": 262 + }, + { + "epoch": 0.2882191780821918, + "grad_norm": 3.3405206203460693, + "learning_rate": 4.543795620437956e-06, + "loss": 2.0975, + "step": 263 + }, + { + "epoch": 0.2893150684931507, + "grad_norm": 2.554431676864624, + "learning_rate": 4.562043795620438e-06, + "loss": 2.0622, + "step": 264 + }, + { + "epoch": 0.29041095890410956, + "grad_norm": 3.5529873371124268, + "learning_rate": 4.580291970802921e-06, + "loss": 2.1567, + "step": 265 + }, + { + "epoch": 0.2915068493150685, + "grad_norm": 2.421682119369507, + "learning_rate": 4.598540145985402e-06, + "loss": 2.1619, + "step": 266 + }, + { + "epoch": 0.2926027397260274, + "grad_norm": 2.918288230895996, + "learning_rate": 4.6167883211678835e-06, + "loss": 2.1238, + "step": 267 + }, + { + "epoch": 0.2936986301369863, + "grad_norm": 3.0737621784210205, + "learning_rate": 4.635036496350365e-06, + "loss": 2.0728, + "step": 268 + }, + { + "epoch": 0.2947945205479452, + "grad_norm": 2.5862410068511963, + "learning_rate": 4.653284671532847e-06, + "loss": 2.0675, + "step": 269 + }, + { + "epoch": 0.2958904109589041, + "grad_norm": 2.9004979133605957, + "learning_rate": 4.671532846715329e-06, + "loss": 2.1248, + "step": 270 + }, + { + "epoch": 0.2958904109589041, + "eval_loss": 2.2179365158081055, + "eval_runtime": 37.2352, + "eval_samples_per_second": 26.856, + "eval_steps_per_second": 1.128, + "step": 270 + }, + { + "epoch": 0.29698630136986304, + "grad_norm": 2.4784982204437256, + "learning_rate": 4.689781021897811e-06, + "loss": 2.1276, + "step": 271 + }, + { + "epoch": 0.2980821917808219, + "grad_norm": 2.7971739768981934, + "learning_rate": 4.708029197080292e-06, + "loss": 2.0931, + "step": 272 + }, + { + "epoch": 0.2991780821917808, + "grad_norm": 2.5839481353759766, + "learning_rate": 4.726277372262774e-06, + "loss": 2.1371, + "step": 273 + }, + { + "epoch": 0.30027397260273975, + "grad_norm": 2.7168564796447754, + "learning_rate": 4.744525547445255e-06, + "loss": 2.1086, + "step": 274 + }, + { + "epoch": 0.3013698630136986, + "grad_norm": 2.6566879749298096, + "learning_rate": 4.7627737226277375e-06, + "loss": 2.0368, + "step": 275 + }, + { + "epoch": 0.30246575342465754, + "grad_norm": 2.5068323612213135, + "learning_rate": 4.78102189781022e-06, + "loss": 2.1077, + "step": 276 + }, + { + "epoch": 0.30356164383561646, + "grad_norm": 2.631408214569092, + "learning_rate": 4.799270072992701e-06, + "loss": 2.1013, + "step": 277 + }, + { + "epoch": 0.3046575342465753, + "grad_norm": 2.5642812252044678, + "learning_rate": 4.8175182481751835e-06, + "loss": 2.0739, + "step": 278 + }, + { + "epoch": 0.30575342465753425, + "grad_norm": 2.3326497077941895, + "learning_rate": 4.835766423357665e-06, + "loss": 2.0971, + "step": 279 + }, + { + "epoch": 0.30684931506849317, + "grad_norm": 2.70701265335083, + "learning_rate": 4.854014598540146e-06, + "loss": 2.0904, + "step": 280 + }, + { + "epoch": 0.30684931506849317, + "eval_loss": 2.215972900390625, + "eval_runtime": 37.2196, + "eval_samples_per_second": 26.868, + "eval_steps_per_second": 1.128, + "step": 280 + }, + { + "epoch": 0.30794520547945203, + "grad_norm": 2.615201234817505, + "learning_rate": 4.872262773722628e-06, + "loss": 2.039, + "step": 281 + }, + { + "epoch": 0.30904109589041096, + "grad_norm": 2.2821998596191406, + "learning_rate": 4.89051094890511e-06, + "loss": 2.0529, + "step": 282 + }, + { + "epoch": 0.3101369863013699, + "grad_norm": 2.767153024673462, + "learning_rate": 4.9087591240875915e-06, + "loss": 2.0351, + "step": 283 + }, + { + "epoch": 0.31123287671232874, + "grad_norm": 2.537121295928955, + "learning_rate": 4.927007299270074e-06, + "loss": 2.0752, + "step": 284 + }, + { + "epoch": 0.31232876712328766, + "grad_norm": 2.835052013397217, + "learning_rate": 4.945255474452555e-06, + "loss": 2.0892, + "step": 285 + }, + { + "epoch": 0.3134246575342466, + "grad_norm": 2.5154924392700195, + "learning_rate": 4.963503649635037e-06, + "loss": 2.0569, + "step": 286 + }, + { + "epoch": 0.31452054794520545, + "grad_norm": 2.8212568759918213, + "learning_rate": 4.981751824817518e-06, + "loss": 2.1588, + "step": 287 + }, + { + "epoch": 0.3156164383561644, + "grad_norm": 2.3840994834899902, + "learning_rate": 5e-06, + "loss": 2.1409, + "step": 288 + }, + { + "epoch": 0.3167123287671233, + "grad_norm": 2.9641315937042236, + "learning_rate": 4.997969130787978e-06, + "loss": 2.0892, + "step": 289 + }, + { + "epoch": 0.3178082191780822, + "grad_norm": 2.4516749382019043, + "learning_rate": 4.995938261575955e-06, + "loss": 2.1105, + "step": 290 + }, + { + "epoch": 0.3178082191780822, + "eval_loss": 2.207075834274292, + "eval_runtime": 37.1304, + "eval_samples_per_second": 26.932, + "eval_steps_per_second": 1.131, + "step": 290 + }, + { + "epoch": 0.3189041095890411, + "grad_norm": 2.768869400024414, + "learning_rate": 4.993907392363932e-06, + "loss": 2.0231, + "step": 291 + }, + { + "epoch": 0.32, + "grad_norm": 2.771475315093994, + "learning_rate": 4.9918765231519095e-06, + "loss": 2.1018, + "step": 292 + }, + { + "epoch": 0.32109589041095893, + "grad_norm": 2.773983955383301, + "learning_rate": 4.989845653939887e-06, + "loss": 2.0317, + "step": 293 + }, + { + "epoch": 0.3221917808219178, + "grad_norm": 2.4447343349456787, + "learning_rate": 4.987814784727864e-06, + "loss": 2.0739, + "step": 294 + }, + { + "epoch": 0.3232876712328767, + "grad_norm": 2.444261312484741, + "learning_rate": 4.985783915515841e-06, + "loss": 2.073, + "step": 295 + }, + { + "epoch": 0.32438356164383564, + "grad_norm": 2.871792793273926, + "learning_rate": 4.983753046303819e-06, + "loss": 2.058, + "step": 296 + }, + { + "epoch": 0.3254794520547945, + "grad_norm": 2.3651537895202637, + "learning_rate": 4.981722177091796e-06, + "loss": 2.0749, + "step": 297 + }, + { + "epoch": 0.3265753424657534, + "grad_norm": 2.3395845890045166, + "learning_rate": 4.979691307879773e-06, + "loss": 2.065, + "step": 298 + }, + { + "epoch": 0.32767123287671235, + "grad_norm": 2.6275217533111572, + "learning_rate": 4.9776604386677504e-06, + "loss": 2.0945, + "step": 299 + }, + { + "epoch": 0.3287671232876712, + "grad_norm": 2.866285562515259, + "learning_rate": 4.975629569455728e-06, + "loss": 2.0702, + "step": 300 + }, + { + "epoch": 0.3287671232876712, + "eval_loss": 2.200012683868408, + "eval_runtime": 37.2082, + "eval_samples_per_second": 26.876, + "eval_steps_per_second": 1.129, + "step": 300 + }, + { + "epoch": 0.32986301369863014, + "grad_norm": 2.268401861190796, + "learning_rate": 4.973598700243705e-06, + "loss": 2.1257, + "step": 301 + }, + { + "epoch": 0.33095890410958906, + "grad_norm": 2.6197733879089355, + "learning_rate": 4.971567831031681e-06, + "loss": 2.0347, + "step": 302 + }, + { + "epoch": 0.3320547945205479, + "grad_norm": 2.4738848209381104, + "learning_rate": 4.9695369618196595e-06, + "loss": 2.0368, + "step": 303 + }, + { + "epoch": 0.33315068493150685, + "grad_norm": 2.705670118331909, + "learning_rate": 4.967506092607636e-06, + "loss": 1.9256, + "step": 304 + }, + { + "epoch": 0.33424657534246577, + "grad_norm": 2.813279628753662, + "learning_rate": 4.965475223395614e-06, + "loss": 2.0214, + "step": 305 + }, + { + "epoch": 0.33534246575342463, + "grad_norm": 2.320960283279419, + "learning_rate": 4.9634443541835905e-06, + "loss": 2.0342, + "step": 306 + }, + { + "epoch": 0.33643835616438356, + "grad_norm": 2.4495227336883545, + "learning_rate": 4.961413484971569e-06, + "loss": 2.0421, + "step": 307 + }, + { + "epoch": 0.3375342465753425, + "grad_norm": 2.5384018421173096, + "learning_rate": 4.959382615759545e-06, + "loss": 2.0572, + "step": 308 + }, + { + "epoch": 0.33863013698630134, + "grad_norm": 2.4018006324768066, + "learning_rate": 4.957351746547523e-06, + "loss": 2.0617, + "step": 309 + }, + { + "epoch": 0.33972602739726027, + "grad_norm": 2.552424907684326, + "learning_rate": 4.9553208773355e-06, + "loss": 2.0384, + "step": 310 + }, + { + "epoch": 0.33972602739726027, + "eval_loss": 2.1974849700927734, + "eval_runtime": 37.2308, + "eval_samples_per_second": 26.859, + "eval_steps_per_second": 1.128, + "step": 310 + }, + { + "epoch": 0.3408219178082192, + "grad_norm": 2.4587671756744385, + "learning_rate": 4.953290008123478e-06, + "loss": 2.1701, + "step": 311 + }, + { + "epoch": 0.3419178082191781, + "grad_norm": 2.4486496448516846, + "learning_rate": 4.951259138911454e-06, + "loss": 2.0654, + "step": 312 + }, + { + "epoch": 0.343013698630137, + "grad_norm": 2.123568058013916, + "learning_rate": 4.9492282696994315e-06, + "loss": 1.9961, + "step": 313 + }, + { + "epoch": 0.3441095890410959, + "grad_norm": 2.549788475036621, + "learning_rate": 4.947197400487409e-06, + "loss": 2.1265, + "step": 314 + }, + { + "epoch": 0.3452054794520548, + "grad_norm": 2.3453609943389893, + "learning_rate": 4.945166531275386e-06, + "loss": 2.0501, + "step": 315 + }, + { + "epoch": 0.3463013698630137, + "grad_norm": 2.3584256172180176, + "learning_rate": 4.943135662063363e-06, + "loss": 2.0402, + "step": 316 + }, + { + "epoch": 0.3473972602739726, + "grad_norm": 3.3100943565368652, + "learning_rate": 4.9411047928513406e-06, + "loss": 2.0545, + "step": 317 + }, + { + "epoch": 0.34849315068493153, + "grad_norm": 2.4266037940979004, + "learning_rate": 4.939073923639318e-06, + "loss": 2.0926, + "step": 318 + }, + { + "epoch": 0.3495890410958904, + "grad_norm": 3.101090908050537, + "learning_rate": 4.937043054427295e-06, + "loss": 2.0987, + "step": 319 + }, + { + "epoch": 0.3506849315068493, + "grad_norm": 2.743619203567505, + "learning_rate": 4.935012185215272e-06, + "loss": 2.0315, + "step": 320 + }, + { + "epoch": 0.3506849315068493, + "eval_loss": 2.188877820968628, + "eval_runtime": 37.4609, + "eval_samples_per_second": 26.694, + "eval_steps_per_second": 1.121, + "step": 320 + }, + { + "epoch": 0.35178082191780824, + "grad_norm": 2.6065986156463623, + "learning_rate": 4.93298131600325e-06, + "loss": 2.0166, + "step": 321 + }, + { + "epoch": 0.3528767123287671, + "grad_norm": 2.7554283142089844, + "learning_rate": 4.930950446791227e-06, + "loss": 2.0862, + "step": 322 + }, + { + "epoch": 0.353972602739726, + "grad_norm": 2.4773144721984863, + "learning_rate": 4.928919577579204e-06, + "loss": 2.0472, + "step": 323 + }, + { + "epoch": 0.35506849315068495, + "grad_norm": 2.695289134979248, + "learning_rate": 4.9268887083671815e-06, + "loss": 2.0567, + "step": 324 + }, + { + "epoch": 0.3561643835616438, + "grad_norm": 2.408220052719116, + "learning_rate": 4.924857839155159e-06, + "loss": 2.046, + "step": 325 + }, + { + "epoch": 0.35726027397260274, + "grad_norm": 2.3851559162139893, + "learning_rate": 4.922826969943136e-06, + "loss": 2.054, + "step": 326 + }, + { + "epoch": 0.35835616438356166, + "grad_norm": 2.8667280673980713, + "learning_rate": 4.920796100731113e-06, + "loss": 2.0717, + "step": 327 + }, + { + "epoch": 0.3594520547945205, + "grad_norm": 2.3814942836761475, + "learning_rate": 4.918765231519091e-06, + "loss": 1.9884, + "step": 328 + }, + { + "epoch": 0.36054794520547945, + "grad_norm": 2.341076612472534, + "learning_rate": 4.916734362307068e-06, + "loss": 2.0474, + "step": 329 + }, + { + "epoch": 0.36164383561643837, + "grad_norm": 2.3571853637695312, + "learning_rate": 4.914703493095045e-06, + "loss": 2.0009, + "step": 330 + }, + { + "epoch": 0.36164383561643837, + "eval_loss": 2.186048746109009, + "eval_runtime": 37.5746, + "eval_samples_per_second": 26.614, + "eval_steps_per_second": 1.118, + "step": 330 + }, + { + "epoch": 0.36273972602739724, + "grad_norm": 2.3770196437835693, + "learning_rate": 4.912672623883022e-06, + "loss": 2.0167, + "step": 331 + }, + { + "epoch": 0.36383561643835616, + "grad_norm": 2.3357701301574707, + "learning_rate": 4.910641754671e-06, + "loss": 1.9972, + "step": 332 + }, + { + "epoch": 0.3649315068493151, + "grad_norm": 2.4100499153137207, + "learning_rate": 4.908610885458977e-06, + "loss": 2.0319, + "step": 333 + }, + { + "epoch": 0.36602739726027395, + "grad_norm": 2.193018913269043, + "learning_rate": 4.906580016246954e-06, + "loss": 2.0285, + "step": 334 + }, + { + "epoch": 0.36712328767123287, + "grad_norm": 2.275637149810791, + "learning_rate": 4.9045491470349315e-06, + "loss": 2.0749, + "step": 335 + }, + { + "epoch": 0.3682191780821918, + "grad_norm": 2.347121238708496, + "learning_rate": 4.902518277822909e-06, + "loss": 2.0502, + "step": 336 + }, + { + "epoch": 0.3693150684931507, + "grad_norm": 2.210817575454712, + "learning_rate": 4.900487408610886e-06, + "loss": 2.0612, + "step": 337 + }, + { + "epoch": 0.3704109589041096, + "grad_norm": 2.15594482421875, + "learning_rate": 4.898456539398863e-06, + "loss": 2.0653, + "step": 338 + }, + { + "epoch": 0.3715068493150685, + "grad_norm": 2.3996052742004395, + "learning_rate": 4.89642567018684e-06, + "loss": 2.074, + "step": 339 + }, + { + "epoch": 0.3726027397260274, + "grad_norm": 2.384172201156616, + "learning_rate": 4.894394800974818e-06, + "loss": 2.1533, + "step": 340 + }, + { + "epoch": 0.3726027397260274, + "eval_loss": 2.175034284591675, + "eval_runtime": 37.2455, + "eval_samples_per_second": 26.849, + "eval_steps_per_second": 1.128, + "step": 340 + }, + { + "epoch": 0.3736986301369863, + "grad_norm": 2.3431336879730225, + "learning_rate": 4.892363931762794e-06, + "loss": 1.9919, + "step": 341 + }, + { + "epoch": 0.3747945205479452, + "grad_norm": 2.1828389167785645, + "learning_rate": 4.8903330625507725e-06, + "loss": 2.0703, + "step": 342 + }, + { + "epoch": 0.37589041095890413, + "grad_norm": 2.2117249965667725, + "learning_rate": 4.888302193338749e-06, + "loss": 2.0071, + "step": 343 + }, + { + "epoch": 0.376986301369863, + "grad_norm": 2.3595967292785645, + "learning_rate": 4.886271324126727e-06, + "loss": 1.9879, + "step": 344 + }, + { + "epoch": 0.3780821917808219, + "grad_norm": 2.5211799144744873, + "learning_rate": 4.8842404549147034e-06, + "loss": 2.0861, + "step": 345 + }, + { + "epoch": 0.37917808219178084, + "grad_norm": 2.1551895141601562, + "learning_rate": 4.8822095857026816e-06, + "loss": 2.0683, + "step": 346 + }, + { + "epoch": 0.3802739726027397, + "grad_norm": 2.4145829677581787, + "learning_rate": 4.880178716490658e-06, + "loss": 2.043, + "step": 347 + }, + { + "epoch": 0.38136986301369863, + "grad_norm": 2.4273521900177, + "learning_rate": 4.878147847278636e-06, + "loss": 1.9928, + "step": 348 + }, + { + "epoch": 0.38246575342465755, + "grad_norm": 2.261054754257202, + "learning_rate": 4.8761169780666125e-06, + "loss": 2.0065, + "step": 349 + }, + { + "epoch": 0.3835616438356164, + "grad_norm": 2.4703147411346436, + "learning_rate": 4.874086108854591e-06, + "loss": 1.9766, + "step": 350 + }, + { + "epoch": 0.3835616438356164, + "eval_loss": 2.176915168762207, + "eval_runtime": 37.3223, + "eval_samples_per_second": 26.794, + "eval_steps_per_second": 1.125, + "step": 350 + }, + { + "epoch": 0.38465753424657534, + "grad_norm": 2.3353374004364014, + "learning_rate": 4.872055239642567e-06, + "loss": 1.9874, + "step": 351 + }, + { + "epoch": 0.38575342465753426, + "grad_norm": 2.227834463119507, + "learning_rate": 4.870024370430544e-06, + "loss": 1.9644, + "step": 352 + }, + { + "epoch": 0.3868493150684931, + "grad_norm": 2.3155159950256348, + "learning_rate": 4.867993501218522e-06, + "loss": 2.0227, + "step": 353 + }, + { + "epoch": 0.38794520547945205, + "grad_norm": 2.2456047534942627, + "learning_rate": 4.865962632006499e-06, + "loss": 2.0566, + "step": 354 + }, + { + "epoch": 0.38904109589041097, + "grad_norm": 2.3364651203155518, + "learning_rate": 4.863931762794476e-06, + "loss": 1.9941, + "step": 355 + }, + { + "epoch": 0.39013698630136984, + "grad_norm": 2.2886059284210205, + "learning_rate": 4.8619008935824535e-06, + "loss": 2.0163, + "step": 356 + }, + { + "epoch": 0.39123287671232876, + "grad_norm": 2.3796911239624023, + "learning_rate": 4.859870024370431e-06, + "loss": 2.0525, + "step": 357 + }, + { + "epoch": 0.3923287671232877, + "grad_norm": 2.381986379623413, + "learning_rate": 4.857839155158408e-06, + "loss": 2.0682, + "step": 358 + }, + { + "epoch": 0.3934246575342466, + "grad_norm": 2.2048425674438477, + "learning_rate": 4.855808285946385e-06, + "loss": 2.0748, + "step": 359 + }, + { + "epoch": 0.39452054794520547, + "grad_norm": 2.162846088409424, + "learning_rate": 4.8537774167343626e-06, + "loss": 2.0512, + "step": 360 + }, + { + "epoch": 0.39452054794520547, + "eval_loss": 2.1645994186401367, + "eval_runtime": 37.3126, + "eval_samples_per_second": 26.801, + "eval_steps_per_second": 1.126, + "step": 360 + }, + { + "epoch": 0.3956164383561644, + "grad_norm": 2.4070873260498047, + "learning_rate": 4.85174654752234e-06, + "loss": 2.042, + "step": 361 + }, + { + "epoch": 0.3967123287671233, + "grad_norm": 2.11716628074646, + "learning_rate": 4.849715678310317e-06, + "loss": 2.0398, + "step": 362 + }, + { + "epoch": 0.3978082191780822, + "grad_norm": 2.265479803085327, + "learning_rate": 4.847684809098294e-06, + "loss": 2.0334, + "step": 363 + }, + { + "epoch": 0.3989041095890411, + "grad_norm": 2.2270994186401367, + "learning_rate": 4.845653939886272e-06, + "loss": 2.0292, + "step": 364 + }, + { + "epoch": 0.4, + "grad_norm": 2.406064033508301, + "learning_rate": 4.843623070674249e-06, + "loss": 2.0351, + "step": 365 + }, + { + "epoch": 0.4010958904109589, + "grad_norm": 2.216654062271118, + "learning_rate": 4.841592201462226e-06, + "loss": 2.005, + "step": 366 + }, + { + "epoch": 0.4021917808219178, + "grad_norm": 2.4241385459899902, + "learning_rate": 4.8395613322502035e-06, + "loss": 2.0647, + "step": 367 + }, + { + "epoch": 0.40328767123287673, + "grad_norm": 2.4315145015716553, + "learning_rate": 4.837530463038181e-06, + "loss": 1.9679, + "step": 368 + }, + { + "epoch": 0.4043835616438356, + "grad_norm": 2.3034439086914062, + "learning_rate": 4.835499593826158e-06, + "loss": 2.0259, + "step": 369 + }, + { + "epoch": 0.4054794520547945, + "grad_norm": 2.493436098098755, + "learning_rate": 4.833468724614135e-06, + "loss": 2.0468, + "step": 370 + }, + { + "epoch": 0.4054794520547945, + "eval_loss": 2.161508083343506, + "eval_runtime": 37.2692, + "eval_samples_per_second": 26.832, + "eval_steps_per_second": 1.127, + "step": 370 + }, + { + "epoch": 0.40657534246575344, + "grad_norm": 2.3887157440185547, + "learning_rate": 4.831437855402113e-06, + "loss": 2.0313, + "step": 371 + }, + { + "epoch": 0.4076712328767123, + "grad_norm": 2.563095808029175, + "learning_rate": 4.82940698619009e-06, + "loss": 1.9288, + "step": 372 + }, + { + "epoch": 0.40876712328767123, + "grad_norm": 2.2609968185424805, + "learning_rate": 4.827376116978067e-06, + "loss": 1.9975, + "step": 373 + }, + { + "epoch": 0.40986301369863015, + "grad_norm": 2.624257802963257, + "learning_rate": 4.8253452477660444e-06, + "loss": 2.0639, + "step": 374 + }, + { + "epoch": 0.410958904109589, + "grad_norm": 2.502197265625, + "learning_rate": 4.823314378554022e-06, + "loss": 2.0424, + "step": 375 + }, + { + "epoch": 0.41205479452054794, + "grad_norm": 2.36297345161438, + "learning_rate": 4.821283509341998e-06, + "loss": 2.027, + "step": 376 + }, + { + "epoch": 0.41315068493150686, + "grad_norm": 2.388918161392212, + "learning_rate": 4.819252640129976e-06, + "loss": 2.0002, + "step": 377 + }, + { + "epoch": 0.41424657534246573, + "grad_norm": 2.4262335300445557, + "learning_rate": 4.817221770917953e-06, + "loss": 2.0215, + "step": 378 + }, + { + "epoch": 0.41534246575342465, + "grad_norm": 2.8953094482421875, + "learning_rate": 4.815190901705931e-06, + "loss": 1.988, + "step": 379 + }, + { + "epoch": 0.41643835616438357, + "grad_norm": 2.3092124462127686, + "learning_rate": 4.813160032493907e-06, + "loss": 2.0199, + "step": 380 + }, + { + "epoch": 0.41643835616438357, + "eval_loss": 2.1589064598083496, + "eval_runtime": 37.2649, + "eval_samples_per_second": 26.835, + "eval_steps_per_second": 1.127, + "step": 380 + }, + { + "epoch": 0.41753424657534244, + "grad_norm": 3.0842413902282715, + "learning_rate": 4.811129163281885e-06, + "loss": 2.0134, + "step": 381 + }, + { + "epoch": 0.41863013698630136, + "grad_norm": 2.2015092372894287, + "learning_rate": 4.809098294069862e-06, + "loss": 2.0447, + "step": 382 + }, + { + "epoch": 0.4197260273972603, + "grad_norm": 2.93471097946167, + "learning_rate": 4.80706742485784e-06, + "loss": 2.0329, + "step": 383 + }, + { + "epoch": 0.4208219178082192, + "grad_norm": 2.3277227878570557, + "learning_rate": 4.805036555645816e-06, + "loss": 2.0805, + "step": 384 + }, + { + "epoch": 0.42191780821917807, + "grad_norm": 2.357966184616089, + "learning_rate": 4.8030056864337945e-06, + "loss": 2.0008, + "step": 385 + }, + { + "epoch": 0.423013698630137, + "grad_norm": 2.7505595684051514, + "learning_rate": 4.800974817221771e-06, + "loss": 2.059, + "step": 386 + }, + { + "epoch": 0.4241095890410959, + "grad_norm": 2.144338369369507, + "learning_rate": 4.798943948009749e-06, + "loss": 2.0125, + "step": 387 + }, + { + "epoch": 0.4252054794520548, + "grad_norm": 2.608407735824585, + "learning_rate": 4.7969130787977254e-06, + "loss": 2.0263, + "step": 388 + }, + { + "epoch": 0.4263013698630137, + "grad_norm": 2.4293761253356934, + "learning_rate": 4.794882209585703e-06, + "loss": 2.0107, + "step": 389 + }, + { + "epoch": 0.4273972602739726, + "grad_norm": 2.2136595249176025, + "learning_rate": 4.79285134037368e-06, + "loss": 2.0421, + "step": 390 + }, + { + "epoch": 0.4273972602739726, + "eval_loss": 2.1473641395568848, + "eval_runtime": 37.2895, + "eval_samples_per_second": 26.817, + "eval_steps_per_second": 1.126, + "step": 390 + }, + { + "epoch": 0.4284931506849315, + "grad_norm": 2.3706328868865967, + "learning_rate": 4.790820471161657e-06, + "loss": 1.9828, + "step": 391 + }, + { + "epoch": 0.4295890410958904, + "grad_norm": 2.1628777980804443, + "learning_rate": 4.7887896019496345e-06, + "loss": 2.0006, + "step": 392 + }, + { + "epoch": 0.43068493150684933, + "grad_norm": 2.5572149753570557, + "learning_rate": 4.786758732737612e-06, + "loss": 2.0046, + "step": 393 + }, + { + "epoch": 0.4317808219178082, + "grad_norm": 2.44661808013916, + "learning_rate": 4.784727863525589e-06, + "loss": 2.0513, + "step": 394 + }, + { + "epoch": 0.4328767123287671, + "grad_norm": 2.429487466812134, + "learning_rate": 4.782696994313566e-06, + "loss": 1.9619, + "step": 395 + }, + { + "epoch": 0.43397260273972604, + "grad_norm": 2.233353614807129, + "learning_rate": 4.780666125101544e-06, + "loss": 2.0141, + "step": 396 + }, + { + "epoch": 0.4350684931506849, + "grad_norm": 2.137549638748169, + "learning_rate": 4.778635255889521e-06, + "loss": 1.9512, + "step": 397 + }, + { + "epoch": 0.43616438356164383, + "grad_norm": 2.187638759613037, + "learning_rate": 4.776604386677498e-06, + "loss": 1.9802, + "step": 398 + }, + { + "epoch": 0.43726027397260275, + "grad_norm": 2.3084137439727783, + "learning_rate": 4.7745735174654755e-06, + "loss": 1.9887, + "step": 399 + }, + { + "epoch": 0.4383561643835616, + "grad_norm": 2.16654109954834, + "learning_rate": 4.772542648253453e-06, + "loss": 2.0535, + "step": 400 + }, + { + "epoch": 0.4383561643835616, + "eval_loss": 2.1444413661956787, + "eval_runtime": 37.2158, + "eval_samples_per_second": 26.87, + "eval_steps_per_second": 1.129, + "step": 400 + }, + { + "epoch": 0.43945205479452054, + "grad_norm": 2.0973072052001953, + "learning_rate": 4.77051177904143e-06, + "loss": 2.0529, + "step": 401 + }, + { + "epoch": 0.44054794520547946, + "grad_norm": 2.421881675720215, + "learning_rate": 4.768480909829407e-06, + "loss": 1.9893, + "step": 402 + }, + { + "epoch": 0.44164383561643833, + "grad_norm": 2.349296808242798, + "learning_rate": 4.766450040617385e-06, + "loss": 2.0237, + "step": 403 + }, + { + "epoch": 0.44273972602739725, + "grad_norm": 2.2154288291931152, + "learning_rate": 4.764419171405362e-06, + "loss": 1.996, + "step": 404 + }, + { + "epoch": 0.4438356164383562, + "grad_norm": 2.6355648040771484, + "learning_rate": 4.762388302193339e-06, + "loss": 2.0345, + "step": 405 + }, + { + "epoch": 0.4449315068493151, + "grad_norm": 2.352238655090332, + "learning_rate": 4.760357432981316e-06, + "loss": 2.0219, + "step": 406 + }, + { + "epoch": 0.44602739726027396, + "grad_norm": 2.1886510848999023, + "learning_rate": 4.758326563769294e-06, + "loss": 1.9923, + "step": 407 + }, + { + "epoch": 0.4471232876712329, + "grad_norm": 2.6483092308044434, + "learning_rate": 4.756295694557271e-06, + "loss": 2.0236, + "step": 408 + }, + { + "epoch": 0.4482191780821918, + "grad_norm": 2.382389545440674, + "learning_rate": 4.754264825345248e-06, + "loss": 2.0249, + "step": 409 + }, + { + "epoch": 0.44931506849315067, + "grad_norm": 2.4824411869049072, + "learning_rate": 4.7522339561332255e-06, + "loss": 1.9571, + "step": 410 + }, + { + "epoch": 0.44931506849315067, + "eval_loss": 2.1502621173858643, + "eval_runtime": 37.1789, + "eval_samples_per_second": 26.897, + "eval_steps_per_second": 1.13, + "step": 410 + }, + { + "epoch": 0.4504109589041096, + "grad_norm": 2.325589656829834, + "learning_rate": 4.750203086921203e-06, + "loss": 1.9753, + "step": 411 + }, + { + "epoch": 0.4515068493150685, + "grad_norm": 2.3618829250335693, + "learning_rate": 4.74817221770918e-06, + "loss": 1.9397, + "step": 412 + }, + { + "epoch": 0.4526027397260274, + "grad_norm": 2.3922839164733887, + "learning_rate": 4.7461413484971565e-06, + "loss": 2.0119, + "step": 413 + }, + { + "epoch": 0.4536986301369863, + "grad_norm": 2.403475046157837, + "learning_rate": 4.744110479285135e-06, + "loss": 2.0664, + "step": 414 + }, + { + "epoch": 0.4547945205479452, + "grad_norm": 2.225860834121704, + "learning_rate": 4.742079610073111e-06, + "loss": 1.9525, + "step": 415 + }, + { + "epoch": 0.4558904109589041, + "grad_norm": 2.344074010848999, + "learning_rate": 4.740048740861089e-06, + "loss": 1.9785, + "step": 416 + }, + { + "epoch": 0.456986301369863, + "grad_norm": 2.255382776260376, + "learning_rate": 4.738017871649066e-06, + "loss": 1.9902, + "step": 417 + }, + { + "epoch": 0.45808219178082193, + "grad_norm": 2.160628318786621, + "learning_rate": 4.735987002437044e-06, + "loss": 2.0075, + "step": 418 + }, + { + "epoch": 0.4591780821917808, + "grad_norm": 2.5667307376861572, + "learning_rate": 4.73395613322502e-06, + "loss": 2.0625, + "step": 419 + }, + { + "epoch": 0.4602739726027397, + "grad_norm": 2.3771040439605713, + "learning_rate": 4.731925264012998e-06, + "loss": 2.0724, + "step": 420 + }, + { + "epoch": 0.4602739726027397, + "eval_loss": 2.138829231262207, + "eval_runtime": 37.4361, + "eval_samples_per_second": 26.712, + "eval_steps_per_second": 1.122, + "step": 420 + }, + { + "epoch": 0.46136986301369864, + "grad_norm": 2.336482048034668, + "learning_rate": 4.729894394800975e-06, + "loss": 2.0234, + "step": 421 + }, + { + "epoch": 0.4624657534246575, + "grad_norm": 2.4206933975219727, + "learning_rate": 4.727863525588953e-06, + "loss": 1.9693, + "step": 422 + }, + { + "epoch": 0.46356164383561643, + "grad_norm": 2.4845645427703857, + "learning_rate": 4.725832656376929e-06, + "loss": 1.9239, + "step": 423 + }, + { + "epoch": 0.46465753424657535, + "grad_norm": 2.5199646949768066, + "learning_rate": 4.723801787164907e-06, + "loss": 1.9801, + "step": 424 + }, + { + "epoch": 0.4657534246575342, + "grad_norm": 2.2969748973846436, + "learning_rate": 4.721770917952885e-06, + "loss": 2.0192, + "step": 425 + }, + { + "epoch": 0.46684931506849314, + "grad_norm": 2.423503875732422, + "learning_rate": 4.719740048740861e-06, + "loss": 1.9684, + "step": 426 + }, + { + "epoch": 0.46794520547945206, + "grad_norm": 2.3681061267852783, + "learning_rate": 4.717709179528839e-06, + "loss": 1.9934, + "step": 427 + }, + { + "epoch": 0.469041095890411, + "grad_norm": 2.277965784072876, + "learning_rate": 4.715678310316816e-06, + "loss": 2.0166, + "step": 428 + }, + { + "epoch": 0.47013698630136985, + "grad_norm": 2.2730283737182617, + "learning_rate": 4.713647441104794e-06, + "loss": 2.011, + "step": 429 + }, + { + "epoch": 0.4712328767123288, + "grad_norm": 2.6694860458374023, + "learning_rate": 4.71161657189277e-06, + "loss": 1.9645, + "step": 430 + }, + { + "epoch": 0.4712328767123288, + "eval_loss": 2.1322011947631836, + "eval_runtime": 37.5723, + "eval_samples_per_second": 26.615, + "eval_steps_per_second": 1.118, + "step": 430 + }, + { + "epoch": 0.4723287671232877, + "grad_norm": 2.4100632667541504, + "learning_rate": 4.709585702680748e-06, + "loss": 1.9478, + "step": 431 + }, + { + "epoch": 0.47342465753424656, + "grad_norm": 2.2779541015625, + "learning_rate": 4.707554833468725e-06, + "loss": 2.0274, + "step": 432 + }, + { + "epoch": 0.4745205479452055, + "grad_norm": 2.55690336227417, + "learning_rate": 4.705523964256703e-06, + "loss": 2.0831, + "step": 433 + }, + { + "epoch": 0.4756164383561644, + "grad_norm": 2.455623149871826, + "learning_rate": 4.703493095044679e-06, + "loss": 1.9673, + "step": 434 + }, + { + "epoch": 0.4767123287671233, + "grad_norm": 2.2613699436187744, + "learning_rate": 4.701462225832657e-06, + "loss": 1.992, + "step": 435 + }, + { + "epoch": 0.4778082191780822, + "grad_norm": 2.4467933177948, + "learning_rate": 4.699431356620634e-06, + "loss": 1.9105, + "step": 436 + }, + { + "epoch": 0.4789041095890411, + "grad_norm": 2.4194023609161377, + "learning_rate": 4.697400487408612e-06, + "loss": 2.074, + "step": 437 + }, + { + "epoch": 0.48, + "grad_norm": 2.2165699005126953, + "learning_rate": 4.695369618196588e-06, + "loss": 2.0152, + "step": 438 + }, + { + "epoch": 0.4810958904109589, + "grad_norm": 2.1709225177764893, + "learning_rate": 4.693338748984566e-06, + "loss": 1.9888, + "step": 439 + }, + { + "epoch": 0.4821917808219178, + "grad_norm": 2.2325828075408936, + "learning_rate": 4.691307879772543e-06, + "loss": 1.9838, + "step": 440 + }, + { + "epoch": 0.4821917808219178, + "eval_loss": 2.128732919692993, + "eval_runtime": 37.205, + "eval_samples_per_second": 26.878, + "eval_steps_per_second": 1.129, + "step": 440 + }, + { + "epoch": 0.4832876712328767, + "grad_norm": 2.446331739425659, + "learning_rate": 4.68927701056052e-06, + "loss": 1.9295, + "step": 441 + }, + { + "epoch": 0.4843835616438356, + "grad_norm": 2.5459415912628174, + "learning_rate": 4.6872461413484975e-06, + "loss": 2.0351, + "step": 442 + }, + { + "epoch": 0.48547945205479454, + "grad_norm": 2.2877118587493896, + "learning_rate": 4.685215272136475e-06, + "loss": 2.0, + "step": 443 + }, + { + "epoch": 0.4865753424657534, + "grad_norm": 2.346625328063965, + "learning_rate": 4.683184402924452e-06, + "loss": 1.9928, + "step": 444 + }, + { + "epoch": 0.4876712328767123, + "grad_norm": 2.268467664718628, + "learning_rate": 4.681153533712429e-06, + "loss": 1.9969, + "step": 445 + }, + { + "epoch": 0.48876712328767125, + "grad_norm": 2.3699793815612793, + "learning_rate": 4.679122664500407e-06, + "loss": 2.0636, + "step": 446 + }, + { + "epoch": 0.4898630136986301, + "grad_norm": 2.2871041297912598, + "learning_rate": 4.677091795288384e-06, + "loss": 1.9535, + "step": 447 + }, + { + "epoch": 0.49095890410958903, + "grad_norm": 2.249393939971924, + "learning_rate": 4.675060926076361e-06, + "loss": 1.9358, + "step": 448 + }, + { + "epoch": 0.49205479452054796, + "grad_norm": 2.299863815307617, + "learning_rate": 4.673030056864338e-06, + "loss": 2.0299, + "step": 449 + }, + { + "epoch": 0.4931506849315068, + "grad_norm": 2.31239652633667, + "learning_rate": 4.670999187652316e-06, + "loss": 1.9372, + "step": 450 + }, + { + "epoch": 0.4931506849315068, + "eval_loss": 2.1267426013946533, + "eval_runtime": 37.1884, + "eval_samples_per_second": 26.89, + "eval_steps_per_second": 1.129, + "step": 450 + }, + { + "epoch": 0.49424657534246574, + "grad_norm": 2.2824623584747314, + "learning_rate": 4.668968318440293e-06, + "loss": 1.9435, + "step": 451 + }, + { + "epoch": 0.49534246575342467, + "grad_norm": 2.285339593887329, + "learning_rate": 4.66693744922827e-06, + "loss": 1.9455, + "step": 452 + }, + { + "epoch": 0.4964383561643836, + "grad_norm": 2.346592903137207, + "learning_rate": 4.6649065800162475e-06, + "loss": 2.0212, + "step": 453 + }, + { + "epoch": 0.49753424657534245, + "grad_norm": 2.280669689178467, + "learning_rate": 4.662875710804225e-06, + "loss": 1.9136, + "step": 454 + }, + { + "epoch": 0.4986301369863014, + "grad_norm": 2.178434133529663, + "learning_rate": 4.660844841592202e-06, + "loss": 1.9952, + "step": 455 + }, + { + "epoch": 0.4997260273972603, + "grad_norm": 2.108492851257324, + "learning_rate": 4.658813972380179e-06, + "loss": 1.9022, + "step": 456 + }, + { + "epoch": 0.5008219178082192, + "grad_norm": 2.3862147331237793, + "learning_rate": 4.656783103168157e-06, + "loss": 2.0332, + "step": 457 + }, + { + "epoch": 0.5019178082191781, + "grad_norm": 2.2017860412597656, + "learning_rate": 4.654752233956134e-06, + "loss": 1.8998, + "step": 458 + }, + { + "epoch": 0.503013698630137, + "grad_norm": 2.222134828567505, + "learning_rate": 4.652721364744111e-06, + "loss": 1.9836, + "step": 459 + }, + { + "epoch": 0.5041095890410959, + "grad_norm": 2.2974510192871094, + "learning_rate": 4.6506904955320885e-06, + "loss": 1.9626, + "step": 460 + }, + { + "epoch": 0.5041095890410959, + "eval_loss": 2.1345443725585938, + "eval_runtime": 37.2785, + "eval_samples_per_second": 26.825, + "eval_steps_per_second": 1.127, + "step": 460 + }, + { + "epoch": 0.5052054794520547, + "grad_norm": 2.107480764389038, + "learning_rate": 4.648659626320066e-06, + "loss": 2.0061, + "step": 461 + }, + { + "epoch": 0.5063013698630137, + "grad_norm": 2.3541181087493896, + "learning_rate": 4.646628757108043e-06, + "loss": 2.0133, + "step": 462 + }, + { + "epoch": 0.5073972602739726, + "grad_norm": 2.215571880340576, + "learning_rate": 4.6445978878960194e-06, + "loss": 1.9727, + "step": 463 + }, + { + "epoch": 0.5084931506849315, + "grad_norm": 2.2263615131378174, + "learning_rate": 4.6425670186839976e-06, + "loss": 1.9477, + "step": 464 + }, + { + "epoch": 0.5095890410958904, + "grad_norm": 2.178900957107544, + "learning_rate": 4.640536149471974e-06, + "loss": 1.9842, + "step": 465 + }, + { + "epoch": 0.5106849315068493, + "grad_norm": 2.3703105449676514, + "learning_rate": 4.638505280259952e-06, + "loss": 2.0082, + "step": 466 + }, + { + "epoch": 0.5117808219178083, + "grad_norm": 2.1587533950805664, + "learning_rate": 4.6364744110479285e-06, + "loss": 2.0271, + "step": 467 + }, + { + "epoch": 0.5128767123287671, + "grad_norm": 2.1626017093658447, + "learning_rate": 4.634443541835907e-06, + "loss": 1.9858, + "step": 468 + }, + { + "epoch": 0.513972602739726, + "grad_norm": 2.326869487762451, + "learning_rate": 4.632412672623883e-06, + "loss": 2.0413, + "step": 469 + }, + { + "epoch": 0.5150684931506849, + "grad_norm": 2.266502857208252, + "learning_rate": 4.630381803411861e-06, + "loss": 1.9584, + "step": 470 + }, + { + "epoch": 0.5150684931506849, + "eval_loss": 2.1273915767669678, + "eval_runtime": 37.2571, + "eval_samples_per_second": 26.841, + "eval_steps_per_second": 1.127, + "step": 470 + }, + { + "epoch": 0.5161643835616438, + "grad_norm": 2.131348133087158, + "learning_rate": 4.628350934199838e-06, + "loss": 1.863, + "step": 471 + }, + { + "epoch": 0.5172602739726028, + "grad_norm": 2.2635257244110107, + "learning_rate": 4.626320064987816e-06, + "loss": 1.9699, + "step": 472 + }, + { + "epoch": 0.5183561643835617, + "grad_norm": 2.1303744316101074, + "learning_rate": 4.624289195775792e-06, + "loss": 2.0001, + "step": 473 + }, + { + "epoch": 0.5194520547945205, + "grad_norm": 2.1736788749694824, + "learning_rate": 4.62225832656377e-06, + "loss": 1.9684, + "step": 474 + }, + { + "epoch": 0.5205479452054794, + "grad_norm": 2.1996278762817383, + "learning_rate": 4.620227457351747e-06, + "loss": 1.9739, + "step": 475 + }, + { + "epoch": 0.5216438356164383, + "grad_norm": 2.225456953048706, + "learning_rate": 4.618196588139724e-06, + "loss": 1.9649, + "step": 476 + }, + { + "epoch": 0.5227397260273973, + "grad_norm": 2.118537187576294, + "learning_rate": 4.616165718927701e-06, + "loss": 1.9911, + "step": 477 + }, + { + "epoch": 0.5238356164383562, + "grad_norm": 2.232940435409546, + "learning_rate": 4.6141348497156786e-06, + "loss": 1.944, + "step": 478 + }, + { + "epoch": 0.5249315068493151, + "grad_norm": 2.1776034832000732, + "learning_rate": 4.612103980503656e-06, + "loss": 2.0342, + "step": 479 + }, + { + "epoch": 0.5260273972602739, + "grad_norm": 2.042645215988159, + "learning_rate": 4.610073111291633e-06, + "loss": 2.0124, + "step": 480 + }, + { + "epoch": 0.5260273972602739, + "eval_loss": 2.118508815765381, + "eval_runtime": 37.2764, + "eval_samples_per_second": 26.827, + "eval_steps_per_second": 1.127, + "step": 480 + }, + { + "epoch": 0.5271232876712328, + "grad_norm": 2.302661895751953, + "learning_rate": 4.60804224207961e-06, + "loss": 1.9874, + "step": 481 + }, + { + "epoch": 0.5282191780821918, + "grad_norm": 2.1251118183135986, + "learning_rate": 4.606011372867588e-06, + "loss": 1.9594, + "step": 482 + }, + { + "epoch": 0.5293150684931507, + "grad_norm": 2.157043933868408, + "learning_rate": 4.603980503655565e-06, + "loss": 1.9955, + "step": 483 + }, + { + "epoch": 0.5304109589041096, + "grad_norm": 2.291428804397583, + "learning_rate": 4.601949634443542e-06, + "loss": 1.9695, + "step": 484 + }, + { + "epoch": 0.5315068493150685, + "grad_norm": 2.2341537475585938, + "learning_rate": 4.5999187652315195e-06, + "loss": 1.899, + "step": 485 + }, + { + "epoch": 0.5326027397260275, + "grad_norm": 2.1881206035614014, + "learning_rate": 4.597887896019497e-06, + "loss": 1.9066, + "step": 486 + }, + { + "epoch": 0.5336986301369863, + "grad_norm": 2.3057658672332764, + "learning_rate": 4.595857026807474e-06, + "loss": 1.9646, + "step": 487 + }, + { + "epoch": 0.5347945205479452, + "grad_norm": 2.2318356037139893, + "learning_rate": 4.593826157595451e-06, + "loss": 1.9662, + "step": 488 + }, + { + "epoch": 0.5358904109589041, + "grad_norm": 2.170509099960327, + "learning_rate": 4.591795288383429e-06, + "loss": 1.9522, + "step": 489 + }, + { + "epoch": 0.536986301369863, + "grad_norm": 2.101646661758423, + "learning_rate": 4.589764419171406e-06, + "loss": 2.0315, + "step": 490 + }, + { + "epoch": 0.536986301369863, + "eval_loss": 2.113968849182129, + "eval_runtime": 37.2751, + "eval_samples_per_second": 26.828, + "eval_steps_per_second": 1.127, + "step": 490 + }, + { + "epoch": 0.538082191780822, + "grad_norm": 2.1563706398010254, + "learning_rate": 4.587733549959383e-06, + "loss": 1.9843, + "step": 491 + }, + { + "epoch": 0.5391780821917809, + "grad_norm": 2.295722246170044, + "learning_rate": 4.5857026807473604e-06, + "loss": 1.9499, + "step": 492 + }, + { + "epoch": 0.5402739726027397, + "grad_norm": 1.974424123764038, + "learning_rate": 4.583671811535338e-06, + "loss": 1.9346, + "step": 493 + }, + { + "epoch": 0.5413698630136986, + "grad_norm": 2.4402668476104736, + "learning_rate": 4.581640942323315e-06, + "loss": 1.9053, + "step": 494 + }, + { + "epoch": 0.5424657534246575, + "grad_norm": 2.269000768661499, + "learning_rate": 4.579610073111292e-06, + "loss": 2.0108, + "step": 495 + }, + { + "epoch": 0.5435616438356164, + "grad_norm": 2.1182594299316406, + "learning_rate": 4.5775792038992695e-06, + "loss": 2.0155, + "step": 496 + }, + { + "epoch": 0.5446575342465754, + "grad_norm": 2.2181780338287354, + "learning_rate": 4.575548334687247e-06, + "loss": 2.0112, + "step": 497 + }, + { + "epoch": 0.5457534246575343, + "grad_norm": 2.2806756496429443, + "learning_rate": 4.573517465475224e-06, + "loss": 2.0418, + "step": 498 + }, + { + "epoch": 0.5468493150684931, + "grad_norm": 2.2690632343292236, + "learning_rate": 4.571486596263201e-06, + "loss": 1.9457, + "step": 499 + }, + { + "epoch": 0.547945205479452, + "grad_norm": 2.361335515975952, + "learning_rate": 4.569455727051178e-06, + "loss": 1.9949, + "step": 500 + }, + { + "epoch": 0.547945205479452, + "eval_loss": 2.1165614128112793, + "eval_runtime": 37.2664, + "eval_samples_per_second": 26.834, + "eval_steps_per_second": 1.127, + "step": 500 + }, + { + "epoch": 0.549041095890411, + "grad_norm": 2.3833138942718506, + "learning_rate": 4.567424857839156e-06, + "loss": 1.9932, + "step": 501 + }, + { + "epoch": 0.5501369863013699, + "grad_norm": 2.328500509262085, + "learning_rate": 4.565393988627132e-06, + "loss": 1.9773, + "step": 502 + }, + { + "epoch": 0.5512328767123288, + "grad_norm": 2.4191694259643555, + "learning_rate": 4.5633631194151105e-06, + "loss": 2.0389, + "step": 503 + }, + { + "epoch": 0.5523287671232877, + "grad_norm": 2.4106884002685547, + "learning_rate": 4.561332250203087e-06, + "loss": 1.965, + "step": 504 + }, + { + "epoch": 0.5534246575342465, + "grad_norm": 2.1452605724334717, + "learning_rate": 4.559301380991065e-06, + "loss": 1.9713, + "step": 505 + }, + { + "epoch": 0.5545205479452054, + "grad_norm": 2.2753357887268066, + "learning_rate": 4.5572705117790414e-06, + "loss": 1.9116, + "step": 506 + }, + { + "epoch": 0.5556164383561644, + "grad_norm": 2.6925032138824463, + "learning_rate": 4.5552396425670196e-06, + "loss": 1.9593, + "step": 507 + }, + { + "epoch": 0.5567123287671233, + "grad_norm": 2.3136658668518066, + "learning_rate": 4.553208773354996e-06, + "loss": 2.0072, + "step": 508 + }, + { + "epoch": 0.5578082191780822, + "grad_norm": 2.254849672317505, + "learning_rate": 4.551177904142974e-06, + "loss": 1.9193, + "step": 509 + }, + { + "epoch": 0.5589041095890411, + "grad_norm": 2.352755546569824, + "learning_rate": 4.5491470349309505e-06, + "loss": 2.0228, + "step": 510 + }, + { + "epoch": 0.5589041095890411, + "eval_loss": 2.114323377609253, + "eval_runtime": 37.3629, + "eval_samples_per_second": 26.764, + "eval_steps_per_second": 1.124, + "step": 510 + }, + { + "epoch": 0.56, + "grad_norm": 2.31339955329895, + "learning_rate": 4.547116165718929e-06, + "loss": 1.9416, + "step": 511 + }, + { + "epoch": 0.5610958904109589, + "grad_norm": 2.191164970397949, + "learning_rate": 4.545085296506905e-06, + "loss": 1.9709, + "step": 512 + }, + { + "epoch": 0.5621917808219178, + "grad_norm": 2.193551778793335, + "learning_rate": 4.543054427294882e-06, + "loss": 1.9475, + "step": 513 + }, + { + "epoch": 0.5632876712328767, + "grad_norm": 2.222710371017456, + "learning_rate": 4.54102355808286e-06, + "loss": 1.9283, + "step": 514 + }, + { + "epoch": 0.5643835616438356, + "grad_norm": 2.078907012939453, + "learning_rate": 4.538992688870837e-06, + "loss": 1.929, + "step": 515 + }, + { + "epoch": 0.5654794520547946, + "grad_norm": 2.2043654918670654, + "learning_rate": 4.536961819658814e-06, + "loss": 1.9591, + "step": 516 + }, + { + "epoch": 0.5665753424657535, + "grad_norm": 2.192481517791748, + "learning_rate": 4.5349309504467915e-06, + "loss": 1.9618, + "step": 517 + }, + { + "epoch": 0.5676712328767123, + "grad_norm": 2.3509292602539062, + "learning_rate": 4.532900081234769e-06, + "loss": 1.9782, + "step": 518 + }, + { + "epoch": 0.5687671232876712, + "grad_norm": 2.2415924072265625, + "learning_rate": 4.530869212022746e-06, + "loss": 1.9456, + "step": 519 + }, + { + "epoch": 0.5698630136986301, + "grad_norm": 2.2135009765625, + "learning_rate": 4.528838342810723e-06, + "loss": 1.9671, + "step": 520 + }, + { + "epoch": 0.5698630136986301, + "eval_loss": 2.109738826751709, + "eval_runtime": 37.5253, + "eval_samples_per_second": 26.649, + "eval_steps_per_second": 1.119, + "step": 520 + }, + { + "epoch": 0.570958904109589, + "grad_norm": 2.3513574600219727, + "learning_rate": 4.526807473598701e-06, + "loss": 1.9625, + "step": 521 + }, + { + "epoch": 0.572054794520548, + "grad_norm": 1.9949424266815186, + "learning_rate": 4.524776604386678e-06, + "loss": 1.9621, + "step": 522 + }, + { + "epoch": 0.5731506849315069, + "grad_norm": 2.199488401412964, + "learning_rate": 4.522745735174655e-06, + "loss": 1.9389, + "step": 523 + }, + { + "epoch": 0.5742465753424657, + "grad_norm": 2.2119452953338623, + "learning_rate": 4.520714865962632e-06, + "loss": 1.9936, + "step": 524 + }, + { + "epoch": 0.5753424657534246, + "grad_norm": 2.242812156677246, + "learning_rate": 4.51868399675061e-06, + "loss": 1.9728, + "step": 525 + }, + { + "epoch": 0.5764383561643835, + "grad_norm": 2.1713850498199463, + "learning_rate": 4.516653127538587e-06, + "loss": 1.98, + "step": 526 + }, + { + "epoch": 0.5775342465753425, + "grad_norm": 2.261742353439331, + "learning_rate": 4.514622258326564e-06, + "loss": 1.9412, + "step": 527 + }, + { + "epoch": 0.5786301369863014, + "grad_norm": 2.222663402557373, + "learning_rate": 4.5125913891145415e-06, + "loss": 1.9266, + "step": 528 + }, + { + "epoch": 0.5797260273972603, + "grad_norm": 2.1108906269073486, + "learning_rate": 4.510560519902519e-06, + "loss": 1.9539, + "step": 529 + }, + { + "epoch": 0.5808219178082191, + "grad_norm": 1.9594205617904663, + "learning_rate": 4.508529650690496e-06, + "loss": 1.9842, + "step": 530 + }, + { + "epoch": 0.5808219178082191, + "eval_loss": 2.1073970794677734, + "eval_runtime": 37.6167, + "eval_samples_per_second": 26.584, + "eval_steps_per_second": 1.117, + "step": 530 + }, + { + "epoch": 0.581917808219178, + "grad_norm": 2.1339495182037354, + "learning_rate": 4.506498781478473e-06, + "loss": 1.9095, + "step": 531 + }, + { + "epoch": 0.583013698630137, + "grad_norm": 2.116248607635498, + "learning_rate": 4.504467912266451e-06, + "loss": 1.9091, + "step": 532 + }, + { + "epoch": 0.5841095890410959, + "grad_norm": 2.063962459564209, + "learning_rate": 4.502437043054428e-06, + "loss": 1.9537, + "step": 533 + }, + { + "epoch": 0.5852054794520548, + "grad_norm": 2.162569522857666, + "learning_rate": 4.500406173842405e-06, + "loss": 2.0253, + "step": 534 + }, + { + "epoch": 0.5863013698630137, + "grad_norm": 2.1783549785614014, + "learning_rate": 4.4983753046303824e-06, + "loss": 1.9714, + "step": 535 + }, + { + "epoch": 0.5873972602739727, + "grad_norm": 2.0791046619415283, + "learning_rate": 4.49634443541836e-06, + "loss": 1.9629, + "step": 536 + }, + { + "epoch": 0.5884931506849315, + "grad_norm": 2.3923251628875732, + "learning_rate": 4.494313566206336e-06, + "loss": 1.9445, + "step": 537 + }, + { + "epoch": 0.5895890410958904, + "grad_norm": 2.1848347187042236, + "learning_rate": 4.492282696994314e-06, + "loss": 2.007, + "step": 538 + }, + { + "epoch": 0.5906849315068493, + "grad_norm": 2.0824925899505615, + "learning_rate": 4.490251827782291e-06, + "loss": 1.9409, + "step": 539 + }, + { + "epoch": 0.5917808219178082, + "grad_norm": 2.057790756225586, + "learning_rate": 4.488220958570269e-06, + "loss": 1.9061, + "step": 540 + }, + { + "epoch": 0.5917808219178082, + "eval_loss": 2.1101415157318115, + "eval_runtime": 37.2235, + "eval_samples_per_second": 26.865, + "eval_steps_per_second": 1.128, + "step": 540 + }, + { + "epoch": 0.5928767123287672, + "grad_norm": 2.3652148246765137, + "learning_rate": 4.486190089358245e-06, + "loss": 1.9379, + "step": 541 + }, + { + "epoch": 0.5939726027397261, + "grad_norm": 2.195535659790039, + "learning_rate": 4.484159220146223e-06, + "loss": 1.8881, + "step": 542 + }, + { + "epoch": 0.5950684931506849, + "grad_norm": 2.258000135421753, + "learning_rate": 4.4821283509342e-06, + "loss": 1.9923, + "step": 543 + }, + { + "epoch": 0.5961643835616438, + "grad_norm": 2.097041606903076, + "learning_rate": 4.480097481722178e-06, + "loss": 1.9715, + "step": 544 + }, + { + "epoch": 0.5972602739726027, + "grad_norm": 2.1741156578063965, + "learning_rate": 4.478066612510154e-06, + "loss": 1.9487, + "step": 545 + }, + { + "epoch": 0.5983561643835617, + "grad_norm": 2.135338306427002, + "learning_rate": 4.4760357432981325e-06, + "loss": 1.9172, + "step": 546 + }, + { + "epoch": 0.5994520547945206, + "grad_norm": 2.3435418605804443, + "learning_rate": 4.474004874086109e-06, + "loss": 1.945, + "step": 547 + }, + { + "epoch": 0.6005479452054795, + "grad_norm": 2.2242488861083984, + "learning_rate": 4.471974004874087e-06, + "loss": 2.0717, + "step": 548 + }, + { + "epoch": 0.6016438356164383, + "grad_norm": 2.3371622562408447, + "learning_rate": 4.4699431356620635e-06, + "loss": 1.9082, + "step": 549 + }, + { + "epoch": 0.6027397260273972, + "grad_norm": 2.176064968109131, + "learning_rate": 4.467912266450041e-06, + "loss": 2.0154, + "step": 550 + }, + { + "epoch": 0.6027397260273972, + "eval_loss": 2.105236291885376, + "eval_runtime": 37.1885, + "eval_samples_per_second": 26.89, + "eval_steps_per_second": 1.129, + "step": 550 + }, + { + "epoch": 0.6038356164383561, + "grad_norm": 2.131166934967041, + "learning_rate": 4.465881397238018e-06, + "loss": 1.9575, + "step": 551 + }, + { + "epoch": 0.6049315068493151, + "grad_norm": 2.094514846801758, + "learning_rate": 4.463850528025995e-06, + "loss": 1.9544, + "step": 552 + }, + { + "epoch": 0.606027397260274, + "grad_norm": 2.0827765464782715, + "learning_rate": 4.4618196588139726e-06, + "loss": 1.9539, + "step": 553 + }, + { + "epoch": 0.6071232876712329, + "grad_norm": 2.225749969482422, + "learning_rate": 4.45978878960195e-06, + "loss": 1.983, + "step": 554 + }, + { + "epoch": 0.6082191780821918, + "grad_norm": 2.1363680362701416, + "learning_rate": 4.457757920389927e-06, + "loss": 1.9769, + "step": 555 + }, + { + "epoch": 0.6093150684931506, + "grad_norm": 2.238170623779297, + "learning_rate": 4.455727051177904e-06, + "loss": 1.9094, + "step": 556 + }, + { + "epoch": 0.6104109589041096, + "grad_norm": 2.228384494781494, + "learning_rate": 4.453696181965882e-06, + "loss": 1.9464, + "step": 557 + }, + { + "epoch": 0.6115068493150685, + "grad_norm": 2.479428768157959, + "learning_rate": 4.451665312753859e-06, + "loss": 2.0017, + "step": 558 + }, + { + "epoch": 0.6126027397260274, + "grad_norm": 2.2331902980804443, + "learning_rate": 4.449634443541836e-06, + "loss": 1.9728, + "step": 559 + }, + { + "epoch": 0.6136986301369863, + "grad_norm": 2.1728556156158447, + "learning_rate": 4.4476035743298135e-06, + "loss": 2.0153, + "step": 560 + }, + { + "epoch": 0.6136986301369863, + "eval_loss": 2.0978519916534424, + "eval_runtime": 37.233, + "eval_samples_per_second": 26.858, + "eval_steps_per_second": 1.128, + "step": 560 + }, + { + "epoch": 0.6147945205479453, + "grad_norm": 2.377814292907715, + "learning_rate": 4.445572705117791e-06, + "loss": 1.9242, + "step": 561 + }, + { + "epoch": 0.6158904109589041, + "grad_norm": 2.190436601638794, + "learning_rate": 4.443541835905768e-06, + "loss": 1.9454, + "step": 562 + }, + { + "epoch": 0.616986301369863, + "grad_norm": 2.052201986312866, + "learning_rate": 4.441510966693745e-06, + "loss": 2.0048, + "step": 563 + }, + { + "epoch": 0.6180821917808219, + "grad_norm": 2.2706589698791504, + "learning_rate": 4.439480097481723e-06, + "loss": 1.9426, + "step": 564 + }, + { + "epoch": 0.6191780821917808, + "grad_norm": 2.3478105068206787, + "learning_rate": 4.4374492282697e-06, + "loss": 2.0047, + "step": 565 + }, + { + "epoch": 0.6202739726027398, + "grad_norm": 2.465078115463257, + "learning_rate": 4.435418359057677e-06, + "loss": 1.8739, + "step": 566 + }, + { + "epoch": 0.6213698630136987, + "grad_norm": 2.0979490280151367, + "learning_rate": 4.433387489845654e-06, + "loss": 1.9588, + "step": 567 + }, + { + "epoch": 0.6224657534246575, + "grad_norm": 2.1239588260650635, + "learning_rate": 4.431356620633632e-06, + "loss": 1.9695, + "step": 568 + }, + { + "epoch": 0.6235616438356164, + "grad_norm": 2.014007806777954, + "learning_rate": 4.429325751421609e-06, + "loss": 1.9232, + "step": 569 + }, + { + "epoch": 0.6246575342465753, + "grad_norm": 2.212414026260376, + "learning_rate": 4.427294882209586e-06, + "loss": 1.928, + "step": 570 + }, + { + "epoch": 0.6246575342465753, + "eval_loss": 2.103149890899658, + "eval_runtime": 37.2641, + "eval_samples_per_second": 26.835, + "eval_steps_per_second": 1.127, + "step": 570 + }, + { + "epoch": 0.6257534246575343, + "grad_norm": 2.0944411754608154, + "learning_rate": 4.4252640129975635e-06, + "loss": 1.9568, + "step": 571 + }, + { + "epoch": 0.6268493150684932, + "grad_norm": 2.1068055629730225, + "learning_rate": 4.423233143785541e-06, + "loss": 1.8917, + "step": 572 + }, + { + "epoch": 0.6279452054794521, + "grad_norm": 2.1992151737213135, + "learning_rate": 4.421202274573518e-06, + "loss": 1.9536, + "step": 573 + }, + { + "epoch": 0.6290410958904109, + "grad_norm": 2.2079834938049316, + "learning_rate": 4.4191714053614945e-06, + "loss": 1.9603, + "step": 574 + }, + { + "epoch": 0.6301369863013698, + "grad_norm": 2.1230316162109375, + "learning_rate": 4.417140536149473e-06, + "loss": 1.8529, + "step": 575 + }, + { + "epoch": 0.6312328767123287, + "grad_norm": 2.1069579124450684, + "learning_rate": 4.415109666937449e-06, + "loss": 1.8997, + "step": 576 + }, + { + "epoch": 0.6323287671232877, + "grad_norm": 2.1683409214019775, + "learning_rate": 4.413078797725427e-06, + "loss": 1.971, + "step": 577 + }, + { + "epoch": 0.6334246575342466, + "grad_norm": 2.0599024295806885, + "learning_rate": 4.411047928513404e-06, + "loss": 1.9154, + "step": 578 + }, + { + "epoch": 0.6345205479452055, + "grad_norm": 2.1605849266052246, + "learning_rate": 4.409017059301382e-06, + "loss": 1.9562, + "step": 579 + }, + { + "epoch": 0.6356164383561644, + "grad_norm": 2.151249408721924, + "learning_rate": 4.406986190089358e-06, + "loss": 1.9571, + "step": 580 + }, + { + "epoch": 0.6356164383561644, + "eval_loss": 2.098177671432495, + "eval_runtime": 37.2578, + "eval_samples_per_second": 26.84, + "eval_steps_per_second": 1.127, + "step": 580 + }, + { + "epoch": 0.6367123287671232, + "grad_norm": 2.1819236278533936, + "learning_rate": 4.404955320877336e-06, + "loss": 1.9816, + "step": 581 + }, + { + "epoch": 0.6378082191780822, + "grad_norm": 2.1131205558776855, + "learning_rate": 4.402924451665313e-06, + "loss": 1.9738, + "step": 582 + }, + { + "epoch": 0.6389041095890411, + "grad_norm": 2.0852456092834473, + "learning_rate": 4.400893582453291e-06, + "loss": 1.924, + "step": 583 + }, + { + "epoch": 0.64, + "grad_norm": 2.0997941493988037, + "learning_rate": 4.398862713241267e-06, + "loss": 1.9067, + "step": 584 + }, + { + "epoch": 0.6410958904109589, + "grad_norm": 2.0848538875579834, + "learning_rate": 4.396831844029245e-06, + "loss": 1.9393, + "step": 585 + }, + { + "epoch": 0.6421917808219179, + "grad_norm": 2.2121758460998535, + "learning_rate": 4.394800974817222e-06, + "loss": 1.9458, + "step": 586 + }, + { + "epoch": 0.6432876712328767, + "grad_norm": 2.194462299346924, + "learning_rate": 4.392770105605199e-06, + "loss": 1.9528, + "step": 587 + }, + { + "epoch": 0.6443835616438356, + "grad_norm": 2.1936254501342773, + "learning_rate": 4.390739236393176e-06, + "loss": 1.9584, + "step": 588 + }, + { + "epoch": 0.6454794520547945, + "grad_norm": 2.1685783863067627, + "learning_rate": 4.388708367181154e-06, + "loss": 1.9666, + "step": 589 + }, + { + "epoch": 0.6465753424657534, + "grad_norm": 2.334571123123169, + "learning_rate": 4.386677497969131e-06, + "loss": 1.9748, + "step": 590 + }, + { + "epoch": 0.6465753424657534, + "eval_loss": 2.0937106609344482, + "eval_runtime": 37.2877, + "eval_samples_per_second": 26.818, + "eval_steps_per_second": 1.126, + "step": 590 + }, + { + "epoch": 0.6476712328767124, + "grad_norm": 2.2901110649108887, + "learning_rate": 4.384646628757108e-06, + "loss": 1.9597, + "step": 591 + }, + { + "epoch": 0.6487671232876713, + "grad_norm": 2.2364838123321533, + "learning_rate": 4.3826157595450855e-06, + "loss": 2.004, + "step": 592 + }, + { + "epoch": 0.6498630136986301, + "grad_norm": 2.1088364124298096, + "learning_rate": 4.380584890333063e-06, + "loss": 1.9319, + "step": 593 + }, + { + "epoch": 0.650958904109589, + "grad_norm": 2.164210319519043, + "learning_rate": 4.37855402112104e-06, + "loss": 1.8906, + "step": 594 + }, + { + "epoch": 0.6520547945205479, + "grad_norm": 2.205200672149658, + "learning_rate": 4.376523151909017e-06, + "loss": 1.8474, + "step": 595 + }, + { + "epoch": 0.6531506849315069, + "grad_norm": 2.239530086517334, + "learning_rate": 4.3744922826969946e-06, + "loss": 2.0033, + "step": 596 + }, + { + "epoch": 0.6542465753424658, + "grad_norm": 2.214747190475464, + "learning_rate": 4.372461413484972e-06, + "loss": 1.9258, + "step": 597 + }, + { + "epoch": 0.6553424657534247, + "grad_norm": 2.1746792793273926, + "learning_rate": 4.370430544272949e-06, + "loss": 1.9479, + "step": 598 + }, + { + "epoch": 0.6564383561643835, + "grad_norm": 2.194131851196289, + "learning_rate": 4.368399675060926e-06, + "loss": 1.9825, + "step": 599 + }, + { + "epoch": 0.6575342465753424, + "grad_norm": 2.3334717750549316, + "learning_rate": 4.366368805848904e-06, + "loss": 1.9831, + "step": 600 + }, + { + "epoch": 0.6575342465753424, + "eval_loss": 2.0881216526031494, + "eval_runtime": 37.3226, + "eval_samples_per_second": 26.793, + "eval_steps_per_second": 1.125, + "step": 600 + }, + { + "epoch": 0.6586301369863014, + "grad_norm": 2.269484758377075, + "learning_rate": 4.364337936636881e-06, + "loss": 1.9407, + "step": 601 + }, + { + "epoch": 0.6597260273972603, + "grad_norm": 2.470773935317993, + "learning_rate": 4.362307067424858e-06, + "loss": 1.8905, + "step": 602 + }, + { + "epoch": 0.6608219178082192, + "grad_norm": 2.485574960708618, + "learning_rate": 4.3602761982128355e-06, + "loss": 1.953, + "step": 603 + }, + { + "epoch": 0.6619178082191781, + "grad_norm": 2.2443618774414062, + "learning_rate": 4.358245329000813e-06, + "loss": 1.9244, + "step": 604 + }, + { + "epoch": 0.663013698630137, + "grad_norm": 2.039600372314453, + "learning_rate": 4.35621445978879e-06, + "loss": 1.9364, + "step": 605 + }, + { + "epoch": 0.6641095890410958, + "grad_norm": 2.093001127243042, + "learning_rate": 4.354183590576767e-06, + "loss": 1.9439, + "step": 606 + }, + { + "epoch": 0.6652054794520548, + "grad_norm": 2.3433260917663574, + "learning_rate": 4.352152721364745e-06, + "loss": 1.9322, + "step": 607 + }, + { + "epoch": 0.6663013698630137, + "grad_norm": 2.2403812408447266, + "learning_rate": 4.350121852152722e-06, + "loss": 1.8829, + "step": 608 + }, + { + "epoch": 0.6673972602739726, + "grad_norm": 2.1542811393737793, + "learning_rate": 4.348090982940699e-06, + "loss": 1.9549, + "step": 609 + }, + { + "epoch": 0.6684931506849315, + "grad_norm": 2.1526548862457275, + "learning_rate": 4.3460601137286764e-06, + "loss": 1.9433, + "step": 610 + }, + { + "epoch": 0.6684931506849315, + "eval_loss": 2.083638906478882, + "eval_runtime": 37.2677, + "eval_samples_per_second": 26.833, + "eval_steps_per_second": 1.127, + "step": 610 + }, + { + "epoch": 0.6695890410958905, + "grad_norm": 2.093421459197998, + "learning_rate": 4.344029244516654e-06, + "loss": 1.9643, + "step": 611 + }, + { + "epoch": 0.6706849315068493, + "grad_norm": 2.2205088138580322, + "learning_rate": 4.341998375304631e-06, + "loss": 1.9309, + "step": 612 + }, + { + "epoch": 0.6717808219178082, + "grad_norm": 2.059763193130493, + "learning_rate": 4.339967506092607e-06, + "loss": 2.0064, + "step": 613 + }, + { + "epoch": 0.6728767123287671, + "grad_norm": 2.0340707302093506, + "learning_rate": 4.3379366368805855e-06, + "loss": 1.9394, + "step": 614 + }, + { + "epoch": 0.673972602739726, + "grad_norm": 2.0716137886047363, + "learning_rate": 4.335905767668562e-06, + "loss": 1.9421, + "step": 615 + }, + { + "epoch": 0.675068493150685, + "grad_norm": 2.176809549331665, + "learning_rate": 4.33387489845654e-06, + "loss": 1.9261, + "step": 616 + }, + { + "epoch": 0.6761643835616439, + "grad_norm": 2.144385814666748, + "learning_rate": 4.3318440292445165e-06, + "loss": 1.9635, + "step": 617 + }, + { + "epoch": 0.6772602739726027, + "grad_norm": 2.0949580669403076, + "learning_rate": 4.329813160032495e-06, + "loss": 1.9417, + "step": 618 + }, + { + "epoch": 0.6783561643835616, + "grad_norm": 2.1727259159088135, + "learning_rate": 4.327782290820471e-06, + "loss": 1.9041, + "step": 619 + }, + { + "epoch": 0.6794520547945205, + "grad_norm": 2.012868881225586, + "learning_rate": 4.325751421608449e-06, + "loss": 1.8973, + "step": 620 + }, + { + "epoch": 0.6794520547945205, + "eval_loss": 2.0839755535125732, + "eval_runtime": 37.5302, + "eval_samples_per_second": 26.645, + "eval_steps_per_second": 1.119, + "step": 620 + }, + { + "epoch": 0.6805479452054795, + "grad_norm": 2.204153299331665, + "learning_rate": 4.323720552396426e-06, + "loss": 1.9719, + "step": 621 + }, + { + "epoch": 0.6816438356164384, + "grad_norm": 2.2421770095825195, + "learning_rate": 4.321689683184404e-06, + "loss": 1.9207, + "step": 622 + }, + { + "epoch": 0.6827397260273973, + "grad_norm": 2.2263410091400146, + "learning_rate": 4.31965881397238e-06, + "loss": 1.9146, + "step": 623 + }, + { + "epoch": 0.6838356164383562, + "grad_norm": 2.2015221118927, + "learning_rate": 4.3176279447603574e-06, + "loss": 1.9311, + "step": 624 + }, + { + "epoch": 0.684931506849315, + "grad_norm": 2.1459643840789795, + "learning_rate": 4.315597075548335e-06, + "loss": 1.9673, + "step": 625 + }, + { + "epoch": 0.686027397260274, + "grad_norm": 2.278928756713867, + "learning_rate": 4.313566206336312e-06, + "loss": 1.954, + "step": 626 + }, + { + "epoch": 0.6871232876712329, + "grad_norm": 2.05305552482605, + "learning_rate": 4.311535337124289e-06, + "loss": 1.9789, + "step": 627 + }, + { + "epoch": 0.6882191780821918, + "grad_norm": 2.3979289531707764, + "learning_rate": 4.3095044679122665e-06, + "loss": 1.926, + "step": 628 + }, + { + "epoch": 0.6893150684931507, + "grad_norm": 2.1635828018188477, + "learning_rate": 4.307473598700244e-06, + "loss": 1.9316, + "step": 629 + }, + { + "epoch": 0.6904109589041096, + "grad_norm": 2.237713098526001, + "learning_rate": 4.305442729488221e-06, + "loss": 1.9409, + "step": 630 + }, + { + "epoch": 0.6904109589041096, + "eval_loss": 2.084226131439209, + "eval_runtime": 37.6039, + "eval_samples_per_second": 26.593, + "eval_steps_per_second": 1.117, + "step": 630 + }, + { + "epoch": 0.6915068493150685, + "grad_norm": 2.157435894012451, + "learning_rate": 4.303411860276198e-06, + "loss": 1.8777, + "step": 631 + }, + { + "epoch": 0.6926027397260274, + "grad_norm": 2.20766019821167, + "learning_rate": 4.301380991064176e-06, + "loss": 1.9803, + "step": 632 + }, + { + "epoch": 0.6936986301369863, + "grad_norm": 2.173079490661621, + "learning_rate": 4.299350121852153e-06, + "loss": 1.9541, + "step": 633 + }, + { + "epoch": 0.6947945205479452, + "grad_norm": 1.9727106094360352, + "learning_rate": 4.29731925264013e-06, + "loss": 1.9766, + "step": 634 + }, + { + "epoch": 0.6958904109589041, + "grad_norm": 2.268538475036621, + "learning_rate": 4.2952883834281075e-06, + "loss": 1.9737, + "step": 635 + }, + { + "epoch": 0.6969863013698631, + "grad_norm": 2.2118406295776367, + "learning_rate": 4.293257514216085e-06, + "loss": 1.9291, + "step": 636 + }, + { + "epoch": 0.6980821917808219, + "grad_norm": 2.2939369678497314, + "learning_rate": 4.291226645004062e-06, + "loss": 1.8868, + "step": 637 + }, + { + "epoch": 0.6991780821917808, + "grad_norm": 2.4940083026885986, + "learning_rate": 4.289195775792039e-06, + "loss": 1.9694, + "step": 638 + }, + { + "epoch": 0.7002739726027397, + "grad_norm": 2.1988351345062256, + "learning_rate": 4.287164906580017e-06, + "loss": 1.9781, + "step": 639 + }, + { + "epoch": 0.7013698630136986, + "grad_norm": 2.099578619003296, + "learning_rate": 4.285134037367994e-06, + "loss": 1.9145, + "step": 640 + }, + { + "epoch": 0.7013698630136986, + "eval_loss": 2.07913875579834, + "eval_runtime": 37.3458, + "eval_samples_per_second": 26.777, + "eval_steps_per_second": 1.125, + "step": 640 + }, + { + "epoch": 0.7024657534246576, + "grad_norm": 2.2437307834625244, + "learning_rate": 4.283103168155971e-06, + "loss": 1.9075, + "step": 641 + }, + { + "epoch": 0.7035616438356165, + "grad_norm": 2.320880174636841, + "learning_rate": 4.281072298943948e-06, + "loss": 1.8929, + "step": 642 + }, + { + "epoch": 0.7046575342465753, + "grad_norm": 2.1174306869506836, + "learning_rate": 4.279041429731926e-06, + "loss": 1.8872, + "step": 643 + }, + { + "epoch": 0.7057534246575342, + "grad_norm": 2.2826361656188965, + "learning_rate": 4.277010560519903e-06, + "loss": 1.9287, + "step": 644 + }, + { + "epoch": 0.7068493150684931, + "grad_norm": 2.7543559074401855, + "learning_rate": 4.27497969130788e-06, + "loss": 1.9119, + "step": 645 + }, + { + "epoch": 0.707945205479452, + "grad_norm": 2.2582201957702637, + "learning_rate": 4.2729488220958575e-06, + "loss": 1.9118, + "step": 646 + }, + { + "epoch": 0.709041095890411, + "grad_norm": 2.3121371269226074, + "learning_rate": 4.270917952883835e-06, + "loss": 1.959, + "step": 647 + }, + { + "epoch": 0.7101369863013699, + "grad_norm": 2.419102430343628, + "learning_rate": 4.268887083671812e-06, + "loss": 1.9679, + "step": 648 + }, + { + "epoch": 0.7112328767123288, + "grad_norm": 2.285771369934082, + "learning_rate": 4.266856214459789e-06, + "loss": 1.8939, + "step": 649 + }, + { + "epoch": 0.7123287671232876, + "grad_norm": 2.193603754043579, + "learning_rate": 4.264825345247766e-06, + "loss": 2.0097, + "step": 650 + }, + { + "epoch": 0.7123287671232876, + "eval_loss": 2.0785164833068848, + "eval_runtime": 37.2595, + "eval_samples_per_second": 26.839, + "eval_steps_per_second": 1.127, + "step": 650 + }, + { + "epoch": 0.7134246575342466, + "grad_norm": 2.2645511627197266, + "learning_rate": 4.262794476035744e-06, + "loss": 1.8865, + "step": 651 + }, + { + "epoch": 0.7145205479452055, + "grad_norm": 2.5553910732269287, + "learning_rate": 4.26076360682372e-06, + "loss": 1.9331, + "step": 652 + }, + { + "epoch": 0.7156164383561644, + "grad_norm": 2.2770845890045166, + "learning_rate": 4.2587327376116984e-06, + "loss": 1.847, + "step": 653 + }, + { + "epoch": 0.7167123287671233, + "grad_norm": 2.1486358642578125, + "learning_rate": 4.256701868399675e-06, + "loss": 1.9058, + "step": 654 + }, + { + "epoch": 0.7178082191780822, + "grad_norm": 2.168975353240967, + "learning_rate": 4.254670999187653e-06, + "loss": 1.9408, + "step": 655 + }, + { + "epoch": 0.718904109589041, + "grad_norm": 2.2156167030334473, + "learning_rate": 4.252640129975629e-06, + "loss": 1.932, + "step": 656 + }, + { + "epoch": 0.72, + "grad_norm": 2.148214101791382, + "learning_rate": 4.2506092607636075e-06, + "loss": 1.9327, + "step": 657 + }, + { + "epoch": 0.7210958904109589, + "grad_norm": 2.3860998153686523, + "learning_rate": 4.248578391551584e-06, + "loss": 1.9191, + "step": 658 + }, + { + "epoch": 0.7221917808219178, + "grad_norm": 2.3917558193206787, + "learning_rate": 4.246547522339562e-06, + "loss": 1.8607, + "step": 659 + }, + { + "epoch": 0.7232876712328767, + "grad_norm": 2.3024795055389404, + "learning_rate": 4.2445166531275385e-06, + "loss": 1.9083, + "step": 660 + }, + { + "epoch": 0.7232876712328767, + "eval_loss": 2.07985782623291, + "eval_runtime": 37.1583, + "eval_samples_per_second": 26.912, + "eval_steps_per_second": 1.13, + "step": 660 + }, + { + "epoch": 0.7243835616438357, + "grad_norm": 2.261615037918091, + "learning_rate": 4.242485783915517e-06, + "loss": 1.9155, + "step": 661 + }, + { + "epoch": 0.7254794520547945, + "grad_norm": 2.190124034881592, + "learning_rate": 4.240454914703493e-06, + "loss": 1.858, + "step": 662 + }, + { + "epoch": 0.7265753424657534, + "grad_norm": 2.3396036624908447, + "learning_rate": 4.23842404549147e-06, + "loss": 1.878, + "step": 663 + }, + { + "epoch": 0.7276712328767123, + "grad_norm": 2.198410987854004, + "learning_rate": 4.236393176279448e-06, + "loss": 1.8755, + "step": 664 + }, + { + "epoch": 0.7287671232876712, + "grad_norm": 2.108875274658203, + "learning_rate": 4.234362307067425e-06, + "loss": 1.8709, + "step": 665 + }, + { + "epoch": 0.7298630136986302, + "grad_norm": 2.0908193588256836, + "learning_rate": 4.232331437855402e-06, + "loss": 1.9395, + "step": 666 + }, + { + "epoch": 0.7309589041095891, + "grad_norm": 2.208134651184082, + "learning_rate": 4.2303005686433795e-06, + "loss": 1.9675, + "step": 667 + }, + { + "epoch": 0.7320547945205479, + "grad_norm": 2.2779526710510254, + "learning_rate": 4.228269699431357e-06, + "loss": 1.9661, + "step": 668 + }, + { + "epoch": 0.7331506849315068, + "grad_norm": 2.053279161453247, + "learning_rate": 4.226238830219334e-06, + "loss": 1.9244, + "step": 669 + }, + { + "epoch": 0.7342465753424657, + "grad_norm": 2.2060821056365967, + "learning_rate": 4.224207961007311e-06, + "loss": 1.9252, + "step": 670 + }, + { + "epoch": 0.7342465753424657, + "eval_loss": 2.0804648399353027, + "eval_runtime": 37.1852, + "eval_samples_per_second": 26.892, + "eval_steps_per_second": 1.129, + "step": 670 + }, + { + "epoch": 0.7353424657534247, + "grad_norm": 2.1996090412139893, + "learning_rate": 4.2221770917952886e-06, + "loss": 1.926, + "step": 671 + }, + { + "epoch": 0.7364383561643836, + "grad_norm": 2.0817034244537354, + "learning_rate": 4.220146222583266e-06, + "loss": 1.9123, + "step": 672 + }, + { + "epoch": 0.7375342465753425, + "grad_norm": 2.2505035400390625, + "learning_rate": 4.218115353371243e-06, + "loss": 1.9369, + "step": 673 + }, + { + "epoch": 0.7386301369863014, + "grad_norm": 2.2053000926971436, + "learning_rate": 4.21608448415922e-06, + "loss": 1.9225, + "step": 674 + }, + { + "epoch": 0.7397260273972602, + "grad_norm": 2.039905071258545, + "learning_rate": 4.214053614947198e-06, + "loss": 1.8876, + "step": 675 + }, + { + "epoch": 0.7408219178082192, + "grad_norm": 2.047445297241211, + "learning_rate": 4.212022745735175e-06, + "loss": 1.9724, + "step": 676 + }, + { + "epoch": 0.7419178082191781, + "grad_norm": 2.277803897857666, + "learning_rate": 4.209991876523152e-06, + "loss": 1.8687, + "step": 677 + }, + { + "epoch": 0.743013698630137, + "grad_norm": 2.0048105716705322, + "learning_rate": 4.2079610073111295e-06, + "loss": 1.9349, + "step": 678 + }, + { + "epoch": 0.7441095890410959, + "grad_norm": 2.121614456176758, + "learning_rate": 4.205930138099107e-06, + "loss": 1.8723, + "step": 679 + }, + { + "epoch": 0.7452054794520548, + "grad_norm": 2.169165849685669, + "learning_rate": 4.203899268887084e-06, + "loss": 1.9713, + "step": 680 + }, + { + "epoch": 0.7452054794520548, + "eval_loss": 2.0704824924468994, + "eval_runtime": 37.2015, + "eval_samples_per_second": 26.881, + "eval_steps_per_second": 1.129, + "step": 680 + }, + { + "epoch": 0.7463013698630137, + "grad_norm": 2.004974603652954, + "learning_rate": 4.201868399675061e-06, + "loss": 1.9237, + "step": 681 + }, + { + "epoch": 0.7473972602739726, + "grad_norm": 2.042402505874634, + "learning_rate": 4.199837530463039e-06, + "loss": 1.9686, + "step": 682 + }, + { + "epoch": 0.7484931506849315, + "grad_norm": 2.152040719985962, + "learning_rate": 4.197806661251016e-06, + "loss": 1.8498, + "step": 683 + }, + { + "epoch": 0.7495890410958904, + "grad_norm": 2.116483211517334, + "learning_rate": 4.195775792038993e-06, + "loss": 1.9241, + "step": 684 + }, + { + "epoch": 0.7506849315068493, + "grad_norm": 2.1628384590148926, + "learning_rate": 4.19374492282697e-06, + "loss": 1.9019, + "step": 685 + }, + { + "epoch": 0.7517808219178083, + "grad_norm": 2.1851415634155273, + "learning_rate": 4.191714053614948e-06, + "loss": 1.836, + "step": 686 + }, + { + "epoch": 0.7528767123287671, + "grad_norm": 2.173215866088867, + "learning_rate": 4.189683184402924e-06, + "loss": 1.9412, + "step": 687 + }, + { + "epoch": 0.753972602739726, + "grad_norm": 2.1002466678619385, + "learning_rate": 4.187652315190902e-06, + "loss": 1.8871, + "step": 688 + }, + { + "epoch": 0.7550684931506849, + "grad_norm": 2.1932122707366943, + "learning_rate": 4.185621445978879e-06, + "loss": 1.9671, + "step": 689 + }, + { + "epoch": 0.7561643835616438, + "grad_norm": 2.2915258407592773, + "learning_rate": 4.183590576766857e-06, + "loss": 1.8957, + "step": 690 + }, + { + "epoch": 0.7561643835616438, + "eval_loss": 2.067124128341675, + "eval_runtime": 37.2362, + "eval_samples_per_second": 26.856, + "eval_steps_per_second": 1.128, + "step": 690 + }, + { + "epoch": 0.7572602739726028, + "grad_norm": 2.060879707336426, + "learning_rate": 4.181559707554833e-06, + "loss": 1.9091, + "step": 691 + }, + { + "epoch": 0.7583561643835617, + "grad_norm": 2.031156539916992, + "learning_rate": 4.179528838342811e-06, + "loss": 1.9606, + "step": 692 + }, + { + "epoch": 0.7594520547945205, + "grad_norm": 2.121628999710083, + "learning_rate": 4.177497969130788e-06, + "loss": 1.88, + "step": 693 + }, + { + "epoch": 0.7605479452054794, + "grad_norm": 2.2009618282318115, + "learning_rate": 4.175467099918766e-06, + "loss": 1.8916, + "step": 694 + }, + { + "epoch": 0.7616438356164383, + "grad_norm": 2.124082326889038, + "learning_rate": 4.173436230706742e-06, + "loss": 1.9805, + "step": 695 + }, + { + "epoch": 0.7627397260273973, + "grad_norm": 2.3914906978607178, + "learning_rate": 4.1714053614947205e-06, + "loss": 1.9052, + "step": 696 + }, + { + "epoch": 0.7638356164383562, + "grad_norm": 2.2398853302001953, + "learning_rate": 4.169374492282697e-06, + "loss": 1.8965, + "step": 697 + }, + { + "epoch": 0.7649315068493151, + "grad_norm": 2.224860191345215, + "learning_rate": 4.167343623070675e-06, + "loss": 1.8962, + "step": 698 + }, + { + "epoch": 0.766027397260274, + "grad_norm": 2.3356244564056396, + "learning_rate": 4.165312753858652e-06, + "loss": 1.9543, + "step": 699 + }, + { + "epoch": 0.7671232876712328, + "grad_norm": 2.1308751106262207, + "learning_rate": 4.163281884646629e-06, + "loss": 1.9095, + "step": 700 + }, + { + "epoch": 0.7671232876712328, + "eval_loss": 2.066145896911621, + "eval_runtime": 37.2629, + "eval_samples_per_second": 26.836, + "eval_steps_per_second": 1.127, + "step": 700 + }, + { + "epoch": 0.7682191780821918, + "grad_norm": 2.1655938625335693, + "learning_rate": 4.161251015434607e-06, + "loss": 1.8923, + "step": 701 + }, + { + "epoch": 0.7693150684931507, + "grad_norm": 2.1985316276550293, + "learning_rate": 4.159220146222583e-06, + "loss": 1.9216, + "step": 702 + }, + { + "epoch": 0.7704109589041096, + "grad_norm": 2.1537251472473145, + "learning_rate": 4.157189277010561e-06, + "loss": 1.879, + "step": 703 + }, + { + "epoch": 0.7715068493150685, + "grad_norm": 2.1867849826812744, + "learning_rate": 4.155158407798538e-06, + "loss": 1.905, + "step": 704 + }, + { + "epoch": 0.7726027397260274, + "grad_norm": 2.342320203781128, + "learning_rate": 4.153127538586516e-06, + "loss": 1.9403, + "step": 705 + }, + { + "epoch": 0.7736986301369863, + "grad_norm": 2.2250113487243652, + "learning_rate": 4.151096669374492e-06, + "loss": 1.9109, + "step": 706 + }, + { + "epoch": 0.7747945205479452, + "grad_norm": 2.179103374481201, + "learning_rate": 4.1490658001624705e-06, + "loss": 1.9288, + "step": 707 + }, + { + "epoch": 0.7758904109589041, + "grad_norm": 2.3618645668029785, + "learning_rate": 4.147034930950447e-06, + "loss": 1.8982, + "step": 708 + }, + { + "epoch": 0.776986301369863, + "grad_norm": 2.150402307510376, + "learning_rate": 4.145004061738425e-06, + "loss": 1.9, + "step": 709 + }, + { + "epoch": 0.7780821917808219, + "grad_norm": 2.295571804046631, + "learning_rate": 4.1429731925264015e-06, + "loss": 1.9003, + "step": 710 + }, + { + "epoch": 0.7780821917808219, + "eval_loss": 2.067948341369629, + "eval_runtime": 37.2379, + "eval_samples_per_second": 26.854, + "eval_steps_per_second": 1.128, + "step": 710 + }, + { + "epoch": 0.7791780821917809, + "grad_norm": 2.3279545307159424, + "learning_rate": 4.140942323314379e-06, + "loss": 1.8307, + "step": 711 + }, + { + "epoch": 0.7802739726027397, + "grad_norm": 2.3993380069732666, + "learning_rate": 4.138911454102356e-06, + "loss": 1.8755, + "step": 712 + }, + { + "epoch": 0.7813698630136986, + "grad_norm": 2.139653444290161, + "learning_rate": 4.136880584890333e-06, + "loss": 1.8914, + "step": 713 + }, + { + "epoch": 0.7824657534246575, + "grad_norm": 2.239544153213501, + "learning_rate": 4.1348497156783106e-06, + "loss": 1.9289, + "step": 714 + }, + { + "epoch": 0.7835616438356164, + "grad_norm": 2.2039642333984375, + "learning_rate": 4.132818846466288e-06, + "loss": 1.9237, + "step": 715 + }, + { + "epoch": 0.7846575342465754, + "grad_norm": 2.1453804969787598, + "learning_rate": 4.130787977254265e-06, + "loss": 1.9291, + "step": 716 + }, + { + "epoch": 0.7857534246575343, + "grad_norm": 2.2350032329559326, + "learning_rate": 4.128757108042242e-06, + "loss": 1.9413, + "step": 717 + }, + { + "epoch": 0.7868493150684932, + "grad_norm": 2.1718318462371826, + "learning_rate": 4.12672623883022e-06, + "loss": 1.9318, + "step": 718 + }, + { + "epoch": 0.787945205479452, + "grad_norm": 2.1860601902008057, + "learning_rate": 4.124695369618197e-06, + "loss": 1.9934, + "step": 719 + }, + { + "epoch": 0.7890410958904109, + "grad_norm": 2.205166816711426, + "learning_rate": 4.122664500406174e-06, + "loss": 1.9038, + "step": 720 + }, + { + "epoch": 0.7890410958904109, + "eval_loss": 2.06229829788208, + "eval_runtime": 37.6121, + "eval_samples_per_second": 26.587, + "eval_steps_per_second": 1.117, + "step": 720 + }, + { + "epoch": 0.7901369863013699, + "grad_norm": 2.2094533443450928, + "learning_rate": 4.1206336311941515e-06, + "loss": 1.906, + "step": 721 + }, + { + "epoch": 0.7912328767123288, + "grad_norm": 2.178891658782959, + "learning_rate": 4.118602761982129e-06, + "loss": 1.936, + "step": 722 + }, + { + "epoch": 0.7923287671232877, + "grad_norm": 2.051077365875244, + "learning_rate": 4.116571892770106e-06, + "loss": 1.8736, + "step": 723 + }, + { + "epoch": 0.7934246575342466, + "grad_norm": 2.2430427074432373, + "learning_rate": 4.114541023558083e-06, + "loss": 1.8396, + "step": 724 + }, + { + "epoch": 0.7945205479452054, + "grad_norm": 2.2255303859710693, + "learning_rate": 4.112510154346061e-06, + "loss": 1.9668, + "step": 725 + }, + { + "epoch": 0.7956164383561644, + "grad_norm": 2.280313491821289, + "learning_rate": 4.110479285134038e-06, + "loss": 1.8945, + "step": 726 + }, + { + "epoch": 0.7967123287671233, + "grad_norm": 2.2516801357269287, + "learning_rate": 4.108448415922015e-06, + "loss": 1.9157, + "step": 727 + }, + { + "epoch": 0.7978082191780822, + "grad_norm": 2.1379809379577637, + "learning_rate": 4.1064175467099924e-06, + "loss": 1.935, + "step": 728 + }, + { + "epoch": 0.7989041095890411, + "grad_norm": 2.149425983428955, + "learning_rate": 4.10438667749797e-06, + "loss": 1.935, + "step": 729 + }, + { + "epoch": 0.8, + "grad_norm": 2.275270700454712, + "learning_rate": 4.102355808285947e-06, + "loss": 1.7889, + "step": 730 + }, + { + "epoch": 0.8, + "eval_loss": 2.0633974075317383, + "eval_runtime": 37.6895, + "eval_samples_per_second": 26.533, + "eval_steps_per_second": 1.114, + "step": 730 + }, + { + "epoch": 0.8010958904109589, + "grad_norm": 2.1550278663635254, + "learning_rate": 4.100324939073924e-06, + "loss": 1.9833, + "step": 731 + }, + { + "epoch": 0.8021917808219178, + "grad_norm": 2.3050060272216797, + "learning_rate": 4.0982940698619015e-06, + "loss": 1.8969, + "step": 732 + }, + { + "epoch": 0.8032876712328767, + "grad_norm": 2.4548778533935547, + "learning_rate": 4.096263200649879e-06, + "loss": 1.8945, + "step": 733 + }, + { + "epoch": 0.8043835616438356, + "grad_norm": 2.2611639499664307, + "learning_rate": 4.094232331437856e-06, + "loss": 1.9071, + "step": 734 + }, + { + "epoch": 0.8054794520547945, + "grad_norm": 2.1806774139404297, + "learning_rate": 4.092201462225833e-06, + "loss": 1.8787, + "step": 735 + }, + { + "epoch": 0.8065753424657535, + "grad_norm": 2.4374301433563232, + "learning_rate": 4.090170593013811e-06, + "loss": 1.88, + "step": 736 + }, + { + "epoch": 0.8076712328767123, + "grad_norm": 2.328333854675293, + "learning_rate": 4.088139723801787e-06, + "loss": 1.9178, + "step": 737 + }, + { + "epoch": 0.8087671232876712, + "grad_norm": 2.33552885055542, + "learning_rate": 4.086108854589765e-06, + "loss": 1.8587, + "step": 738 + }, + { + "epoch": 0.8098630136986301, + "grad_norm": 2.3472790718078613, + "learning_rate": 4.084077985377742e-06, + "loss": 1.877, + "step": 739 + }, + { + "epoch": 0.810958904109589, + "grad_norm": 2.307123899459839, + "learning_rate": 4.08204711616572e-06, + "loss": 1.8888, + "step": 740 + }, + { + "epoch": 0.810958904109589, + "eval_loss": 2.065826892852783, + "eval_runtime": 37.2747, + "eval_samples_per_second": 26.828, + "eval_steps_per_second": 1.127, + "step": 740 + }, + { + "epoch": 0.812054794520548, + "grad_norm": 2.3688747882843018, + "learning_rate": 4.080016246953696e-06, + "loss": 1.9571, + "step": 741 + }, + { + "epoch": 0.8131506849315069, + "grad_norm": 2.385007381439209, + "learning_rate": 4.077985377741674e-06, + "loss": 1.8657, + "step": 742 + }, + { + "epoch": 0.8142465753424658, + "grad_norm": 2.156766414642334, + "learning_rate": 4.075954508529651e-06, + "loss": 1.9421, + "step": 743 + }, + { + "epoch": 0.8153424657534246, + "grad_norm": 2.2074286937713623, + "learning_rate": 4.073923639317629e-06, + "loss": 1.9396, + "step": 744 + }, + { + "epoch": 0.8164383561643835, + "grad_norm": 2.185458183288574, + "learning_rate": 4.071892770105605e-06, + "loss": 1.9612, + "step": 745 + }, + { + "epoch": 0.8175342465753425, + "grad_norm": 2.324209451675415, + "learning_rate": 4.069861900893583e-06, + "loss": 1.892, + "step": 746 + }, + { + "epoch": 0.8186301369863014, + "grad_norm": 2.1312127113342285, + "learning_rate": 4.06783103168156e-06, + "loss": 1.8522, + "step": 747 + }, + { + "epoch": 0.8197260273972603, + "grad_norm": 2.0959460735321045, + "learning_rate": 4.065800162469538e-06, + "loss": 1.8732, + "step": 748 + }, + { + "epoch": 0.8208219178082192, + "grad_norm": 2.085519552230835, + "learning_rate": 4.063769293257514e-06, + "loss": 1.8832, + "step": 749 + }, + { + "epoch": 0.821917808219178, + "grad_norm": 2.310145854949951, + "learning_rate": 4.061738424045492e-06, + "loss": 1.919, + "step": 750 + }, + { + "epoch": 0.821917808219178, + "eval_loss": 2.0637967586517334, + "eval_runtime": 37.2246, + "eval_samples_per_second": 26.864, + "eval_steps_per_second": 1.128, + "step": 750 + }, + { + "epoch": 0.823013698630137, + "grad_norm": 2.3171160221099854, + "learning_rate": 4.059707554833469e-06, + "loss": 1.8874, + "step": 751 + }, + { + "epoch": 0.8241095890410959, + "grad_norm": 2.1065480709075928, + "learning_rate": 4.057676685621446e-06, + "loss": 1.9064, + "step": 752 + }, + { + "epoch": 0.8252054794520548, + "grad_norm": 2.0539677143096924, + "learning_rate": 4.0556458164094235e-06, + "loss": 1.9376, + "step": 753 + }, + { + "epoch": 0.8263013698630137, + "grad_norm": 2.4440360069274902, + "learning_rate": 4.053614947197401e-06, + "loss": 1.9129, + "step": 754 + }, + { + "epoch": 0.8273972602739726, + "grad_norm": 2.2324604988098145, + "learning_rate": 4.051584077985378e-06, + "loss": 1.8281, + "step": 755 + }, + { + "epoch": 0.8284931506849315, + "grad_norm": 2.262627363204956, + "learning_rate": 4.049553208773355e-06, + "loss": 1.9036, + "step": 756 + }, + { + "epoch": 0.8295890410958904, + "grad_norm": 2.2473418712615967, + "learning_rate": 4.047522339561333e-06, + "loss": 1.9455, + "step": 757 + }, + { + "epoch": 0.8306849315068493, + "grad_norm": 2.473444938659668, + "learning_rate": 4.04549147034931e-06, + "loss": 1.9003, + "step": 758 + }, + { + "epoch": 0.8317808219178082, + "grad_norm": 2.3454110622406006, + "learning_rate": 4.043460601137287e-06, + "loss": 1.8941, + "step": 759 + }, + { + "epoch": 0.8328767123287671, + "grad_norm": 2.4391231536865234, + "learning_rate": 4.041429731925264e-06, + "loss": 2.0146, + "step": 760 + }, + { + "epoch": 0.8328767123287671, + "eval_loss": 2.0618460178375244, + "eval_runtime": 37.1704, + "eval_samples_per_second": 26.903, + "eval_steps_per_second": 1.13, + "step": 760 + }, + { + "epoch": 0.8339726027397261, + "grad_norm": 2.3129498958587646, + "learning_rate": 4.039398862713242e-06, + "loss": 1.9, + "step": 761 + }, + { + "epoch": 0.8350684931506849, + "grad_norm": 2.3766934871673584, + "learning_rate": 4.037367993501219e-06, + "loss": 1.9017, + "step": 762 + }, + { + "epoch": 0.8361643835616438, + "grad_norm": 2.1698460578918457, + "learning_rate": 4.035337124289196e-06, + "loss": 1.9657, + "step": 763 + }, + { + "epoch": 0.8372602739726027, + "grad_norm": 2.384904146194458, + "learning_rate": 4.0333062550771735e-06, + "loss": 1.8867, + "step": 764 + }, + { + "epoch": 0.8383561643835616, + "grad_norm": 2.572391986846924, + "learning_rate": 4.031275385865151e-06, + "loss": 1.8561, + "step": 765 + }, + { + "epoch": 0.8394520547945206, + "grad_norm": 2.216606616973877, + "learning_rate": 4.029244516653128e-06, + "loss": 1.8926, + "step": 766 + }, + { + "epoch": 0.8405479452054795, + "grad_norm": 2.215756893157959, + "learning_rate": 4.027213647441105e-06, + "loss": 1.928, + "step": 767 + }, + { + "epoch": 0.8416438356164384, + "grad_norm": 2.2713284492492676, + "learning_rate": 4.025182778229083e-06, + "loss": 1.8286, + "step": 768 + }, + { + "epoch": 0.8427397260273972, + "grad_norm": 2.341747283935547, + "learning_rate": 4.02315190901706e-06, + "loss": 1.8789, + "step": 769 + }, + { + "epoch": 0.8438356164383561, + "grad_norm": 2.1485848426818848, + "learning_rate": 4.021121039805037e-06, + "loss": 1.8998, + "step": 770 + }, + { + "epoch": 0.8438356164383561, + "eval_loss": 2.0596234798431396, + "eval_runtime": 37.1854, + "eval_samples_per_second": 26.892, + "eval_steps_per_second": 1.129, + "step": 770 + }, + { + "epoch": 0.8449315068493151, + "grad_norm": 2.234192132949829, + "learning_rate": 4.0190901705930144e-06, + "loss": 1.8995, + "step": 771 + }, + { + "epoch": 0.846027397260274, + "grad_norm": 2.289328098297119, + "learning_rate": 4.017059301380992e-06, + "loss": 1.9882, + "step": 772 + }, + { + "epoch": 0.8471232876712329, + "grad_norm": 2.1001455783843994, + "learning_rate": 4.015028432168969e-06, + "loss": 1.8706, + "step": 773 + }, + { + "epoch": 0.8482191780821918, + "grad_norm": 2.2200522422790527, + "learning_rate": 4.012997562956945e-06, + "loss": 1.9443, + "step": 774 + }, + { + "epoch": 0.8493150684931506, + "grad_norm": 2.2540805339813232, + "learning_rate": 4.0109666937449235e-06, + "loss": 1.8552, + "step": 775 + }, + { + "epoch": 0.8504109589041096, + "grad_norm": 2.1214053630828857, + "learning_rate": 4.0089358245329e-06, + "loss": 1.8741, + "step": 776 + }, + { + "epoch": 0.8515068493150685, + "grad_norm": 2.0221869945526123, + "learning_rate": 4.006904955320878e-06, + "loss": 1.8798, + "step": 777 + }, + { + "epoch": 0.8526027397260274, + "grad_norm": 2.217261791229248, + "learning_rate": 4.0048740861088545e-06, + "loss": 1.8542, + "step": 778 + }, + { + "epoch": 0.8536986301369863, + "grad_norm": 2.099202871322632, + "learning_rate": 4.002843216896833e-06, + "loss": 1.9342, + "step": 779 + }, + { + "epoch": 0.8547945205479452, + "grad_norm": 2.139526844024658, + "learning_rate": 4.000812347684809e-06, + "loss": 1.9252, + "step": 780 + }, + { + "epoch": 0.8547945205479452, + "eval_loss": 2.054051399230957, + "eval_runtime": 37.1125, + "eval_samples_per_second": 26.945, + "eval_steps_per_second": 1.132, + "step": 780 + }, + { + "epoch": 0.8558904109589041, + "grad_norm": 2.2231476306915283, + "learning_rate": 3.998781478472787e-06, + "loss": 1.8671, + "step": 781 + }, + { + "epoch": 0.856986301369863, + "grad_norm": 2.143979072570801, + "learning_rate": 3.996750609260764e-06, + "loss": 1.8759, + "step": 782 + }, + { + "epoch": 0.8580821917808219, + "grad_norm": 2.1353135108947754, + "learning_rate": 3.994719740048742e-06, + "loss": 1.8559, + "step": 783 + }, + { + "epoch": 0.8591780821917808, + "grad_norm": 2.222564697265625, + "learning_rate": 3.992688870836718e-06, + "loss": 1.999, + "step": 784 + }, + { + "epoch": 0.8602739726027397, + "grad_norm": 1.958031177520752, + "learning_rate": 3.990658001624696e-06, + "loss": 1.9019, + "step": 785 + }, + { + "epoch": 0.8613698630136987, + "grad_norm": 2.2048962116241455, + "learning_rate": 3.988627132412673e-06, + "loss": 1.8561, + "step": 786 + }, + { + "epoch": 0.8624657534246576, + "grad_norm": 2.150332450866699, + "learning_rate": 3.98659626320065e-06, + "loss": 1.9399, + "step": 787 + }, + { + "epoch": 0.8635616438356164, + "grad_norm": 2.182742118835449, + "learning_rate": 3.984565393988627e-06, + "loss": 1.8933, + "step": 788 + }, + { + "epoch": 0.8646575342465753, + "grad_norm": 2.19435715675354, + "learning_rate": 3.9825345247766046e-06, + "loss": 1.8918, + "step": 789 + }, + { + "epoch": 0.8657534246575342, + "grad_norm": 2.3382859230041504, + "learning_rate": 3.980503655564582e-06, + "loss": 1.8937, + "step": 790 + }, + { + "epoch": 0.8657534246575342, + "eval_loss": 2.051997423171997, + "eval_runtime": 37.1866, + "eval_samples_per_second": 26.891, + "eval_steps_per_second": 1.129, + "step": 790 + }, + { + "epoch": 0.8668493150684932, + "grad_norm": 2.2700726985931396, + "learning_rate": 3.978472786352559e-06, + "loss": 1.9148, + "step": 791 + }, + { + "epoch": 0.8679452054794521, + "grad_norm": 2.2998759746551514, + "learning_rate": 3.976441917140536e-06, + "loss": 1.9268, + "step": 792 + }, + { + "epoch": 0.869041095890411, + "grad_norm": 2.2367377281188965, + "learning_rate": 3.974411047928514e-06, + "loss": 1.901, + "step": 793 + }, + { + "epoch": 0.8701369863013698, + "grad_norm": 2.0868773460388184, + "learning_rate": 3.972380178716491e-06, + "loss": 1.9677, + "step": 794 + }, + { + "epoch": 0.8712328767123287, + "grad_norm": 2.183732509613037, + "learning_rate": 3.970349309504468e-06, + "loss": 1.9271, + "step": 795 + }, + { + "epoch": 0.8723287671232877, + "grad_norm": 2.187282085418701, + "learning_rate": 3.9683184402924455e-06, + "loss": 1.9437, + "step": 796 + }, + { + "epoch": 0.8734246575342466, + "grad_norm": 2.1906070709228516, + "learning_rate": 3.966287571080423e-06, + "loss": 1.8437, + "step": 797 + }, + { + "epoch": 0.8745205479452055, + "grad_norm": 2.16979718208313, + "learning_rate": 3.9642567018684e-06, + "loss": 1.829, + "step": 798 + }, + { + "epoch": 0.8756164383561644, + "grad_norm": 2.2387869358062744, + "learning_rate": 3.962225832656377e-06, + "loss": 1.9143, + "step": 799 + }, + { + "epoch": 0.8767123287671232, + "grad_norm": 2.208221197128296, + "learning_rate": 3.960194963444355e-06, + "loss": 1.8894, + "step": 800 + }, + { + "epoch": 0.8767123287671232, + "eval_loss": 2.047945976257324, + "eval_runtime": 37.1778, + "eval_samples_per_second": 26.898, + "eval_steps_per_second": 1.13, + "step": 800 + }, + { + "epoch": 0.8778082191780822, + "grad_norm": 2.335108518600464, + "learning_rate": 3.958164094232332e-06, + "loss": 1.8155, + "step": 801 + }, + { + "epoch": 0.8789041095890411, + "grad_norm": 2.168006181716919, + "learning_rate": 3.956133225020309e-06, + "loss": 1.8919, + "step": 802 + }, + { + "epoch": 0.88, + "grad_norm": 2.3029897212982178, + "learning_rate": 3.954102355808286e-06, + "loss": 1.8996, + "step": 803 + }, + { + "epoch": 0.8810958904109589, + "grad_norm": 2.156141519546509, + "learning_rate": 3.952071486596264e-06, + "loss": 1.8965, + "step": 804 + }, + { + "epoch": 0.8821917808219178, + "grad_norm": 2.1522068977355957, + "learning_rate": 3.950040617384241e-06, + "loss": 1.8922, + "step": 805 + }, + { + "epoch": 0.8832876712328767, + "grad_norm": 2.153430700302124, + "learning_rate": 3.948009748172218e-06, + "loss": 1.8839, + "step": 806 + }, + { + "epoch": 0.8843835616438356, + "grad_norm": 2.1891674995422363, + "learning_rate": 3.9459788789601955e-06, + "loss": 1.8756, + "step": 807 + }, + { + "epoch": 0.8854794520547945, + "grad_norm": 2.1092092990875244, + "learning_rate": 3.943948009748173e-06, + "loss": 1.9452, + "step": 808 + }, + { + "epoch": 0.8865753424657534, + "grad_norm": 2.0465025901794434, + "learning_rate": 3.94191714053615e-06, + "loss": 1.9067, + "step": 809 + }, + { + "epoch": 0.8876712328767123, + "grad_norm": 2.1150474548339844, + "learning_rate": 3.939886271324127e-06, + "loss": 1.8874, + "step": 810 + }, + { + "epoch": 0.8876712328767123, + "eval_loss": 2.046292304992676, + "eval_runtime": 37.2421, + "eval_samples_per_second": 26.851, + "eval_steps_per_second": 1.128, + "step": 810 + }, + { + "epoch": 0.8887671232876713, + "grad_norm": 2.039989948272705, + "learning_rate": 3.937855402112104e-06, + "loss": 1.9212, + "step": 811 + }, + { + "epoch": 0.8898630136986302, + "grad_norm": 2.17354679107666, + "learning_rate": 3.935824532900082e-06, + "loss": 1.8473, + "step": 812 + }, + { + "epoch": 0.890958904109589, + "grad_norm": 2.3101940155029297, + "learning_rate": 3.933793663688058e-06, + "loss": 1.8413, + "step": 813 + }, + { + "epoch": 0.8920547945205479, + "grad_norm": 2.1300928592681885, + "learning_rate": 3.9317627944760365e-06, + "loss": 1.9126, + "step": 814 + }, + { + "epoch": 0.8931506849315068, + "grad_norm": 2.2317357063293457, + "learning_rate": 3.929731925264013e-06, + "loss": 1.938, + "step": 815 + }, + { + "epoch": 0.8942465753424658, + "grad_norm": 2.1097419261932373, + "learning_rate": 3.927701056051991e-06, + "loss": 1.8426, + "step": 816 + }, + { + "epoch": 0.8953424657534247, + "grad_norm": 2.1709141731262207, + "learning_rate": 3.9256701868399674e-06, + "loss": 1.8779, + "step": 817 + }, + { + "epoch": 0.8964383561643836, + "grad_norm": 2.104193925857544, + "learning_rate": 3.9236393176279456e-06, + "loss": 1.8736, + "step": 818 + }, + { + "epoch": 0.8975342465753424, + "grad_norm": 2.094637632369995, + "learning_rate": 3.921608448415922e-06, + "loss": 1.8556, + "step": 819 + }, + { + "epoch": 0.8986301369863013, + "grad_norm": 2.217918872833252, + "learning_rate": 3.9195775792039e-06, + "loss": 1.8429, + "step": 820 + }, + { + "epoch": 0.8986301369863013, + "eval_loss": 2.046095848083496, + "eval_runtime": 37.5659, + "eval_samples_per_second": 26.62, + "eval_steps_per_second": 1.118, + "step": 820 + }, + { + "epoch": 0.8997260273972603, + "grad_norm": 2.220355272293091, + "learning_rate": 3.9175467099918765e-06, + "loss": 1.8973, + "step": 821 + }, + { + "epoch": 0.9008219178082192, + "grad_norm": 2.0766689777374268, + "learning_rate": 3.915515840779855e-06, + "loss": 1.8728, + "step": 822 + }, + { + "epoch": 0.9019178082191781, + "grad_norm": 2.2046821117401123, + "learning_rate": 3.913484971567831e-06, + "loss": 1.9211, + "step": 823 + }, + { + "epoch": 0.903013698630137, + "grad_norm": 2.286024332046509, + "learning_rate": 3.911454102355808e-06, + "loss": 1.913, + "step": 824 + }, + { + "epoch": 0.9041095890410958, + "grad_norm": 2.2495479583740234, + "learning_rate": 3.909423233143786e-06, + "loss": 1.8214, + "step": 825 + }, + { + "epoch": 0.9052054794520548, + "grad_norm": 2.1300203800201416, + "learning_rate": 3.907392363931763e-06, + "loss": 1.9033, + "step": 826 + }, + { + "epoch": 0.9063013698630137, + "grad_norm": 2.092094898223877, + "learning_rate": 3.90536149471974e-06, + "loss": 1.9007, + "step": 827 + }, + { + "epoch": 0.9073972602739726, + "grad_norm": 2.1571428775787354, + "learning_rate": 3.9033306255077175e-06, + "loss": 1.9606, + "step": 828 + }, + { + "epoch": 0.9084931506849315, + "grad_norm": 2.1052281856536865, + "learning_rate": 3.901299756295695e-06, + "loss": 1.9916, + "step": 829 + }, + { + "epoch": 0.9095890410958904, + "grad_norm": 2.113666534423828, + "learning_rate": 3.899268887083672e-06, + "loss": 2.0136, + "step": 830 + }, + { + "epoch": 0.9095890410958904, + "eval_loss": 2.0450072288513184, + "eval_runtime": 37.6746, + "eval_samples_per_second": 26.543, + "eval_steps_per_second": 1.115, + "step": 830 + }, + { + "epoch": 0.9106849315068493, + "grad_norm": 2.112591028213501, + "learning_rate": 3.897238017871649e-06, + "loss": 1.8784, + "step": 831 + }, + { + "epoch": 0.9117808219178082, + "grad_norm": 2.1584672927856445, + "learning_rate": 3.8952071486596266e-06, + "loss": 1.8876, + "step": 832 + }, + { + "epoch": 0.9128767123287671, + "grad_norm": 2.2744317054748535, + "learning_rate": 3.893176279447604e-06, + "loss": 1.8981, + "step": 833 + }, + { + "epoch": 0.913972602739726, + "grad_norm": 2.21824312210083, + "learning_rate": 3.891145410235581e-06, + "loss": 1.85, + "step": 834 + }, + { + "epoch": 0.915068493150685, + "grad_norm": 2.2203729152679443, + "learning_rate": 3.889114541023558e-06, + "loss": 1.9649, + "step": 835 + }, + { + "epoch": 0.9161643835616439, + "grad_norm": 2.1767687797546387, + "learning_rate": 3.887083671811536e-06, + "loss": 1.9044, + "step": 836 + }, + { + "epoch": 0.9172602739726028, + "grad_norm": 2.0724353790283203, + "learning_rate": 3.885052802599513e-06, + "loss": 1.8787, + "step": 837 + }, + { + "epoch": 0.9183561643835616, + "grad_norm": 2.1283740997314453, + "learning_rate": 3.88302193338749e-06, + "loss": 1.8284, + "step": 838 + }, + { + "epoch": 0.9194520547945205, + "grad_norm": 2.5087451934814453, + "learning_rate": 3.8809910641754675e-06, + "loss": 1.9546, + "step": 839 + }, + { + "epoch": 0.9205479452054794, + "grad_norm": 2.0633344650268555, + "learning_rate": 3.878960194963445e-06, + "loss": 1.8969, + "step": 840 + }, + { + "epoch": 0.9205479452054794, + "eval_loss": 2.047938346862793, + "eval_runtime": 37.2293, + "eval_samples_per_second": 26.861, + "eval_steps_per_second": 1.128, + "step": 840 + }, + { + "epoch": 0.9216438356164384, + "grad_norm": 2.178361654281616, + "learning_rate": 3.876929325751422e-06, + "loss": 1.8855, + "step": 841 + }, + { + "epoch": 0.9227397260273973, + "grad_norm": 2.3223721981048584, + "learning_rate": 3.874898456539399e-06, + "loss": 1.9363, + "step": 842 + }, + { + "epoch": 0.9238356164383562, + "grad_norm": 2.3330960273742676, + "learning_rate": 3.872867587327377e-06, + "loss": 1.8374, + "step": 843 + }, + { + "epoch": 0.924931506849315, + "grad_norm": 2.0584356784820557, + "learning_rate": 3.870836718115354e-06, + "loss": 1.911, + "step": 844 + }, + { + "epoch": 0.9260273972602739, + "grad_norm": 2.1834609508514404, + "learning_rate": 3.868805848903331e-06, + "loss": 1.8645, + "step": 845 + }, + { + "epoch": 0.9271232876712329, + "grad_norm": 2.2576122283935547, + "learning_rate": 3.8667749796913084e-06, + "loss": 1.8662, + "step": 846 + }, + { + "epoch": 0.9282191780821918, + "grad_norm": 2.1271183490753174, + "learning_rate": 3.864744110479286e-06, + "loss": 1.8935, + "step": 847 + }, + { + "epoch": 0.9293150684931507, + "grad_norm": 2.226727247238159, + "learning_rate": 3.862713241267262e-06, + "loss": 1.8969, + "step": 848 + }, + { + "epoch": 0.9304109589041096, + "grad_norm": 2.3301939964294434, + "learning_rate": 3.86068237205524e-06, + "loss": 1.9393, + "step": 849 + }, + { + "epoch": 0.9315068493150684, + "grad_norm": 2.180250883102417, + "learning_rate": 3.858651502843217e-06, + "loss": 1.9331, + "step": 850 + }, + { + "epoch": 0.9315068493150684, + "eval_loss": 2.0509681701660156, + "eval_runtime": 37.2111, + "eval_samples_per_second": 26.874, + "eval_steps_per_second": 1.129, + "step": 850 + }, + { + "epoch": 0.9326027397260274, + "grad_norm": 2.223548173904419, + "learning_rate": 3.856620633631195e-06, + "loss": 1.894, + "step": 851 + }, + { + "epoch": 0.9336986301369863, + "grad_norm": 2.1904265880584717, + "learning_rate": 3.854589764419171e-06, + "loss": 1.8265, + "step": 852 + }, + { + "epoch": 0.9347945205479452, + "grad_norm": 2.1374316215515137, + "learning_rate": 3.852558895207149e-06, + "loss": 1.8755, + "step": 853 + }, + { + "epoch": 0.9358904109589041, + "grad_norm": 2.2175488471984863, + "learning_rate": 3.850528025995126e-06, + "loss": 1.9649, + "step": 854 + }, + { + "epoch": 0.936986301369863, + "grad_norm": 2.192126989364624, + "learning_rate": 3.848497156783104e-06, + "loss": 1.919, + "step": 855 + }, + { + "epoch": 0.938082191780822, + "grad_norm": 2.219954013824463, + "learning_rate": 3.84646628757108e-06, + "loss": 1.8651, + "step": 856 + }, + { + "epoch": 0.9391780821917808, + "grad_norm": 2.2897071838378906, + "learning_rate": 3.8444354183590585e-06, + "loss": 1.8744, + "step": 857 + }, + { + "epoch": 0.9402739726027397, + "grad_norm": 2.1656205654144287, + "learning_rate": 3.842404549147035e-06, + "loss": 1.8622, + "step": 858 + }, + { + "epoch": 0.9413698630136986, + "grad_norm": 2.20072603225708, + "learning_rate": 3.840373679935013e-06, + "loss": 1.8678, + "step": 859 + }, + { + "epoch": 0.9424657534246575, + "grad_norm": 2.1474406719207764, + "learning_rate": 3.8383428107229894e-06, + "loss": 1.8462, + "step": 860 + }, + { + "epoch": 0.9424657534246575, + "eval_loss": 2.0451602935791016, + "eval_runtime": 37.1383, + "eval_samples_per_second": 26.926, + "eval_steps_per_second": 1.131, + "step": 860 + }, + { + "epoch": 0.9435616438356165, + "grad_norm": 1.9542036056518555, + "learning_rate": 3.836311941510967e-06, + "loss": 1.8739, + "step": 861 + }, + { + "epoch": 0.9446575342465754, + "grad_norm": 2.3951354026794434, + "learning_rate": 3.834281072298944e-06, + "loss": 1.8565, + "step": 862 + }, + { + "epoch": 0.9457534246575342, + "grad_norm": 2.4916961193084717, + "learning_rate": 3.832250203086921e-06, + "loss": 1.9138, + "step": 863 + }, + { + "epoch": 0.9468493150684931, + "grad_norm": 2.1650452613830566, + "learning_rate": 3.8302193338748985e-06, + "loss": 1.8499, + "step": 864 + }, + { + "epoch": 0.947945205479452, + "grad_norm": 2.2215073108673096, + "learning_rate": 3.828188464662876e-06, + "loss": 1.9376, + "step": 865 + }, + { + "epoch": 0.949041095890411, + "grad_norm": 2.3827078342437744, + "learning_rate": 3.826157595450853e-06, + "loss": 1.9726, + "step": 866 + }, + { + "epoch": 0.9501369863013699, + "grad_norm": 2.37194561958313, + "learning_rate": 3.82412672623883e-06, + "loss": 1.9009, + "step": 867 + }, + { + "epoch": 0.9512328767123288, + "grad_norm": 2.116562604904175, + "learning_rate": 3.822095857026808e-06, + "loss": 1.8342, + "step": 868 + }, + { + "epoch": 0.9523287671232876, + "grad_norm": 2.3410849571228027, + "learning_rate": 3.820064987814785e-06, + "loss": 1.9035, + "step": 869 + }, + { + "epoch": 0.9534246575342465, + "grad_norm": 2.207545280456543, + "learning_rate": 3.818034118602762e-06, + "loss": 1.8539, + "step": 870 + }, + { + "epoch": 0.9534246575342465, + "eval_loss": 2.0433826446533203, + "eval_runtime": 37.114, + "eval_samples_per_second": 26.944, + "eval_steps_per_second": 1.132, + "step": 870 + }, + { + "epoch": 0.9545205479452055, + "grad_norm": 2.1716196537017822, + "learning_rate": 3.8160032493907395e-06, + "loss": 1.8557, + "step": 871 + }, + { + "epoch": 0.9556164383561644, + "grad_norm": 2.213124990463257, + "learning_rate": 3.813972380178717e-06, + "loss": 1.8638, + "step": 872 + }, + { + "epoch": 0.9567123287671233, + "grad_norm": 2.3185672760009766, + "learning_rate": 3.811941510966694e-06, + "loss": 1.8645, + "step": 873 + }, + { + "epoch": 0.9578082191780822, + "grad_norm": 2.453956127166748, + "learning_rate": 3.809910641754671e-06, + "loss": 1.9183, + "step": 874 + }, + { + "epoch": 0.958904109589041, + "grad_norm": 2.1971592903137207, + "learning_rate": 3.8078797725426486e-06, + "loss": 1.8445, + "step": 875 + }, + { + "epoch": 0.96, + "grad_norm": 2.273247241973877, + "learning_rate": 3.8058489033306254e-06, + "loss": 1.8024, + "step": 876 + }, + { + "epoch": 0.9610958904109589, + "grad_norm": 2.1687841415405273, + "learning_rate": 3.803818034118603e-06, + "loss": 1.8931, + "step": 877 + }, + { + "epoch": 0.9621917808219178, + "grad_norm": 2.326498508453369, + "learning_rate": 3.80178716490658e-06, + "loss": 1.8587, + "step": 878 + }, + { + "epoch": 0.9632876712328767, + "grad_norm": 2.0317249298095703, + "learning_rate": 3.7997562956945577e-06, + "loss": 1.9394, + "step": 879 + }, + { + "epoch": 0.9643835616438357, + "grad_norm": 2.3171892166137695, + "learning_rate": 3.7977254264825345e-06, + "loss": 1.92, + "step": 880 + }, + { + "epoch": 0.9643835616438357, + "eval_loss": 2.0431556701660156, + "eval_runtime": 37.1174, + "eval_samples_per_second": 26.942, + "eval_steps_per_second": 1.132, + "step": 880 + }, + { + "epoch": 0.9654794520547946, + "grad_norm": 2.2215981483459473, + "learning_rate": 3.7956945572705122e-06, + "loss": 1.8658, + "step": 881 + }, + { + "epoch": 0.9665753424657534, + "grad_norm": 2.2940685749053955, + "learning_rate": 3.793663688058489e-06, + "loss": 1.8528, + "step": 882 + }, + { + "epoch": 0.9676712328767123, + "grad_norm": 2.2218081951141357, + "learning_rate": 3.791632818846467e-06, + "loss": 1.9446, + "step": 883 + }, + { + "epoch": 0.9687671232876712, + "grad_norm": 2.2608420848846436, + "learning_rate": 3.7896019496344436e-06, + "loss": 1.9433, + "step": 884 + }, + { + "epoch": 0.9698630136986301, + "grad_norm": 2.0247788429260254, + "learning_rate": 3.787571080422421e-06, + "loss": 1.9286, + "step": 885 + }, + { + "epoch": 0.9709589041095891, + "grad_norm": 2.164071798324585, + "learning_rate": 3.785540211210398e-06, + "loss": 1.9349, + "step": 886 + }, + { + "epoch": 0.972054794520548, + "grad_norm": 2.089215040206909, + "learning_rate": 3.7835093419983755e-06, + "loss": 1.9077, + "step": 887 + }, + { + "epoch": 0.9731506849315068, + "grad_norm": 2.245256185531616, + "learning_rate": 3.7814784727863527e-06, + "loss": 1.8911, + "step": 888 + }, + { + "epoch": 0.9742465753424657, + "grad_norm": 2.1227362155914307, + "learning_rate": 3.77944760357433e-06, + "loss": 1.8761, + "step": 889 + }, + { + "epoch": 0.9753424657534246, + "grad_norm": 2.1797730922698975, + "learning_rate": 3.7774167343623073e-06, + "loss": 1.9954, + "step": 890 + }, + { + "epoch": 0.9753424657534246, + "eval_loss": 2.034895896911621, + "eval_runtime": 37.1682, + "eval_samples_per_second": 26.905, + "eval_steps_per_second": 1.13, + "step": 890 + }, + { + "epoch": 0.9764383561643836, + "grad_norm": 2.171062707901001, + "learning_rate": 3.7753858651502846e-06, + "loss": 1.9287, + "step": 891 + }, + { + "epoch": 0.9775342465753425, + "grad_norm": 2.1950814723968506, + "learning_rate": 3.773354995938262e-06, + "loss": 1.8587, + "step": 892 + }, + { + "epoch": 0.9786301369863014, + "grad_norm": 2.2582995891571045, + "learning_rate": 3.771324126726239e-06, + "loss": 1.9269, + "step": 893 + }, + { + "epoch": 0.9797260273972602, + "grad_norm": 2.218113660812378, + "learning_rate": 3.7692932575142164e-06, + "loss": 1.9007, + "step": 894 + }, + { + "epoch": 0.9808219178082191, + "grad_norm": 2.187178134918213, + "learning_rate": 3.7672623883021937e-06, + "loss": 1.9326, + "step": 895 + }, + { + "epoch": 0.9819178082191781, + "grad_norm": 2.20451283454895, + "learning_rate": 3.765231519090171e-06, + "loss": 1.8644, + "step": 896 + }, + { + "epoch": 0.983013698630137, + "grad_norm": 2.1399290561676025, + "learning_rate": 3.7632006498781482e-06, + "loss": 1.7875, + "step": 897 + }, + { + "epoch": 0.9841095890410959, + "grad_norm": 2.049473524093628, + "learning_rate": 3.761169780666125e-06, + "loss": 1.8987, + "step": 898 + }, + { + "epoch": 0.9852054794520548, + "grad_norm": 2.154444456100464, + "learning_rate": 3.7591389114541028e-06, + "loss": 1.8475, + "step": 899 + }, + { + "epoch": 0.9863013698630136, + "grad_norm": 2.1840224266052246, + "learning_rate": 3.7571080422420796e-06, + "loss": 1.8966, + "step": 900 + }, + { + "epoch": 0.9863013698630136, + "eval_loss": 2.0326430797576904, + "eval_runtime": 37.1551, + "eval_samples_per_second": 26.914, + "eval_steps_per_second": 1.13, + "step": 900 + }, + { + "epoch": 0.9873972602739726, + "grad_norm": 2.0891942977905273, + "learning_rate": 3.7550771730300573e-06, + "loss": 1.9113, + "step": 901 + }, + { + "epoch": 0.9884931506849315, + "grad_norm": 2.2646753787994385, + "learning_rate": 3.753046303818034e-06, + "loss": 1.8624, + "step": 902 + }, + { + "epoch": 0.9895890410958904, + "grad_norm": 2.15173602104187, + "learning_rate": 3.751015434606012e-06, + "loss": 1.933, + "step": 903 + }, + { + "epoch": 0.9906849315068493, + "grad_norm": 2.1751456260681152, + "learning_rate": 3.7489845653939887e-06, + "loss": 1.9114, + "step": 904 + }, + { + "epoch": 0.9917808219178083, + "grad_norm": 2.161860942840576, + "learning_rate": 3.7469536961819664e-06, + "loss": 1.9508, + "step": 905 + }, + { + "epoch": 0.9928767123287672, + "grad_norm": 2.0921733379364014, + "learning_rate": 3.7449228269699433e-06, + "loss": 1.8482, + "step": 906 + }, + { + "epoch": 0.993972602739726, + "grad_norm": 2.0823893547058105, + "learning_rate": 3.742891957757921e-06, + "loss": 1.8744, + "step": 907 + }, + { + "epoch": 0.9950684931506849, + "grad_norm": 2.136939525604248, + "learning_rate": 3.740861088545898e-06, + "loss": 1.9037, + "step": 908 + }, + { + "epoch": 0.9961643835616438, + "grad_norm": 2.0600836277008057, + "learning_rate": 3.7388302193338755e-06, + "loss": 1.8922, + "step": 909 + }, + { + "epoch": 0.9972602739726028, + "grad_norm": 2.235780715942383, + "learning_rate": 3.7367993501218524e-06, + "loss": 1.9222, + "step": 910 + }, + { + "epoch": 0.9972602739726028, + "eval_loss": 2.034031629562378, + "eval_runtime": 37.2911, + "eval_samples_per_second": 26.816, + "eval_steps_per_second": 1.126, + "step": 910 + }, + { + "epoch": 0.9983561643835617, + "grad_norm": 2.124662160873413, + "learning_rate": 3.7347684809098297e-06, + "loss": 1.8584, + "step": 911 + }, + { + "epoch": 0.9994520547945206, + "grad_norm": 2.1237940788269043, + "learning_rate": 3.732737611697807e-06, + "loss": 1.8579, + "step": 912 + }, + { + "epoch": 1.0005479452054795, + "grad_norm": 2.1175124645233154, + "learning_rate": 3.7307067424857842e-06, + "loss": 1.8348, + "step": 913 + }, + { + "epoch": 1.0016438356164383, + "grad_norm": 2.349686861038208, + "learning_rate": 3.7286758732737615e-06, + "loss": 1.7822, + "step": 914 + }, + { + "epoch": 1.0027397260273974, + "grad_norm": 2.3278446197509766, + "learning_rate": 3.7266450040617388e-06, + "loss": 1.8346, + "step": 915 + }, + { + "epoch": 1.0038356164383562, + "grad_norm": 2.1633124351501465, + "learning_rate": 3.724614134849716e-06, + "loss": 1.8137, + "step": 916 + }, + { + "epoch": 1.004931506849315, + "grad_norm": 2.5075809955596924, + "learning_rate": 3.7225832656376933e-06, + "loss": 1.955, + "step": 917 + }, + { + "epoch": 1.006027397260274, + "grad_norm": 2.3370068073272705, + "learning_rate": 3.7205523964256706e-06, + "loss": 1.7985, + "step": 918 + }, + { + "epoch": 1.0071232876712328, + "grad_norm": 2.288229465484619, + "learning_rate": 3.718521527213648e-06, + "loss": 1.923, + "step": 919 + }, + { + "epoch": 1.0082191780821919, + "grad_norm": 2.1989684104919434, + "learning_rate": 3.716490658001625e-06, + "loss": 1.8876, + "step": 920 + }, + { + "epoch": 1.0082191780821919, + "eval_loss": 2.03442120552063, + "eval_runtime": 37.5659, + "eval_samples_per_second": 26.62, + "eval_steps_per_second": 1.118, + "step": 920 + }, + { + "epoch": 1.0093150684931507, + "grad_norm": 2.202122211456299, + "learning_rate": 3.7144597887896024e-06, + "loss": 1.879, + "step": 921 + }, + { + "epoch": 1.0104109589041095, + "grad_norm": 2.325223445892334, + "learning_rate": 3.7124289195775797e-06, + "loss": 1.7632, + "step": 922 + }, + { + "epoch": 1.0115068493150685, + "grad_norm": 2.2964577674865723, + "learning_rate": 3.710398050365557e-06, + "loss": 1.8983, + "step": 923 + }, + { + "epoch": 1.0126027397260273, + "grad_norm": 2.548537015914917, + "learning_rate": 3.708367181153534e-06, + "loss": 1.8666, + "step": 924 + }, + { + "epoch": 1.0136986301369864, + "grad_norm": 2.229745626449585, + "learning_rate": 3.7063363119415115e-06, + "loss": 1.8395, + "step": 925 + }, + { + "epoch": 1.0147945205479452, + "grad_norm": 2.3639235496520996, + "learning_rate": 3.7043054427294884e-06, + "loss": 1.8864, + "step": 926 + }, + { + "epoch": 1.0158904109589042, + "grad_norm": 2.3403055667877197, + "learning_rate": 3.702274573517466e-06, + "loss": 1.8385, + "step": 927 + }, + { + "epoch": 1.016986301369863, + "grad_norm": 2.2838001251220703, + "learning_rate": 3.700243704305443e-06, + "loss": 1.8135, + "step": 928 + }, + { + "epoch": 1.0180821917808218, + "grad_norm": 2.4730923175811768, + "learning_rate": 3.6982128350934206e-06, + "loss": 1.837, + "step": 929 + }, + { + "epoch": 1.0191780821917809, + "grad_norm": 2.5711073875427246, + "learning_rate": 3.6961819658813975e-06, + "loss": 1.8276, + "step": 930 + }, + { + "epoch": 1.0191780821917809, + "eval_loss": 2.0346033573150635, + "eval_runtime": 37.6741, + "eval_samples_per_second": 26.543, + "eval_steps_per_second": 1.115, + "step": 930 + }, + { + "epoch": 1.0202739726027397, + "grad_norm": 2.296823740005493, + "learning_rate": 3.694151096669375e-06, + "loss": 1.8054, + "step": 931 + }, + { + "epoch": 1.0213698630136987, + "grad_norm": 2.321107864379883, + "learning_rate": 3.692120227457352e-06, + "loss": 1.9296, + "step": 932 + }, + { + "epoch": 1.0224657534246575, + "grad_norm": 2.3478338718414307, + "learning_rate": 3.6900893582453297e-06, + "loss": 1.8503, + "step": 933 + }, + { + "epoch": 1.0235616438356165, + "grad_norm": 2.3698856830596924, + "learning_rate": 3.6880584890333066e-06, + "loss": 1.8766, + "step": 934 + }, + { + "epoch": 1.0246575342465754, + "grad_norm": 2.2751212120056152, + "learning_rate": 3.6860276198212834e-06, + "loss": 1.7575, + "step": 935 + }, + { + "epoch": 1.0257534246575342, + "grad_norm": 2.368741035461426, + "learning_rate": 3.683996750609261e-06, + "loss": 1.8555, + "step": 936 + }, + { + "epoch": 1.0268493150684932, + "grad_norm": 2.4507620334625244, + "learning_rate": 3.681965881397238e-06, + "loss": 1.8765, + "step": 937 + }, + { + "epoch": 1.027945205479452, + "grad_norm": 2.1907308101654053, + "learning_rate": 3.6799350121852157e-06, + "loss": 1.8652, + "step": 938 + }, + { + "epoch": 1.029041095890411, + "grad_norm": 2.2696914672851562, + "learning_rate": 3.6779041429731925e-06, + "loss": 1.8854, + "step": 939 + }, + { + "epoch": 1.0301369863013699, + "grad_norm": 2.3696796894073486, + "learning_rate": 3.6758732737611702e-06, + "loss": 1.8303, + "step": 940 + }, + { + "epoch": 1.0301369863013699, + "eval_loss": 2.030118703842163, + "eval_runtime": 37.1755, + "eval_samples_per_second": 26.899, + "eval_steps_per_second": 1.13, + "step": 940 + }, + { + "epoch": 1.0312328767123287, + "grad_norm": 2.440284013748169, + "learning_rate": 3.673842404549147e-06, + "loss": 1.8498, + "step": 941 + }, + { + "epoch": 1.0323287671232877, + "grad_norm": 2.250570297241211, + "learning_rate": 3.671811535337125e-06, + "loss": 1.8361, + "step": 942 + }, + { + "epoch": 1.0334246575342465, + "grad_norm": 2.5550990104675293, + "learning_rate": 3.6697806661251016e-06, + "loss": 1.8464, + "step": 943 + }, + { + "epoch": 1.0345205479452055, + "grad_norm": 2.502474784851074, + "learning_rate": 3.6677497969130793e-06, + "loss": 1.8471, + "step": 944 + }, + { + "epoch": 1.0356164383561643, + "grad_norm": 2.3815231323242188, + "learning_rate": 3.665718927701056e-06, + "loss": 1.9786, + "step": 945 + }, + { + "epoch": 1.0367123287671234, + "grad_norm": 2.3053736686706543, + "learning_rate": 3.663688058489034e-06, + "loss": 1.8625, + "step": 946 + }, + { + "epoch": 1.0378082191780822, + "grad_norm": 2.2623229026794434, + "learning_rate": 3.6616571892770107e-06, + "loss": 1.9206, + "step": 947 + }, + { + "epoch": 1.038904109589041, + "grad_norm": 2.634559392929077, + "learning_rate": 3.659626320064988e-06, + "loss": 1.8475, + "step": 948 + }, + { + "epoch": 1.04, + "grad_norm": 2.264615297317505, + "learning_rate": 3.6575954508529653e-06, + "loss": 1.7857, + "step": 949 + }, + { + "epoch": 1.0410958904109588, + "grad_norm": 2.4125282764434814, + "learning_rate": 3.6555645816409426e-06, + "loss": 1.841, + "step": 950 + }, + { + "epoch": 1.0410958904109588, + "eval_loss": 2.0278165340423584, + "eval_runtime": 37.1503, + "eval_samples_per_second": 26.918, + "eval_steps_per_second": 1.131, + "step": 950 + }, + { + "epoch": 1.0421917808219179, + "grad_norm": 2.365743637084961, + "learning_rate": 3.65353371242892e-06, + "loss": 1.8352, + "step": 951 + }, + { + "epoch": 1.0432876712328767, + "grad_norm": 2.6469883918762207, + "learning_rate": 3.651502843216897e-06, + "loss": 1.8814, + "step": 952 + }, + { + "epoch": 1.0443835616438357, + "grad_norm": 2.3646206855773926, + "learning_rate": 3.6494719740048744e-06, + "loss": 1.7282, + "step": 953 + }, + { + "epoch": 1.0454794520547945, + "grad_norm": 2.250856876373291, + "learning_rate": 3.6474411047928517e-06, + "loss": 1.8647, + "step": 954 + }, + { + "epoch": 1.0465753424657533, + "grad_norm": 2.5404837131500244, + "learning_rate": 3.645410235580829e-06, + "loss": 1.8175, + "step": 955 + }, + { + "epoch": 1.0476712328767124, + "grad_norm": 2.4388837814331055, + "learning_rate": 3.6433793663688062e-06, + "loss": 1.8643, + "step": 956 + }, + { + "epoch": 1.0487671232876712, + "grad_norm": 2.285959005355835, + "learning_rate": 3.6413484971567835e-06, + "loss": 1.8367, + "step": 957 + }, + { + "epoch": 1.0498630136986302, + "grad_norm": 2.1190242767333984, + "learning_rate": 3.6393176279447608e-06, + "loss": 1.8242, + "step": 958 + }, + { + "epoch": 1.050958904109589, + "grad_norm": 2.397578716278076, + "learning_rate": 3.637286758732738e-06, + "loss": 1.8681, + "step": 959 + }, + { + "epoch": 1.0520547945205478, + "grad_norm": 2.510599374771118, + "learning_rate": 3.6352558895207153e-06, + "loss": 1.8804, + "step": 960 + }, + { + "epoch": 1.0520547945205478, + "eval_loss": 2.0267090797424316, + "eval_runtime": 37.1226, + "eval_samples_per_second": 26.938, + "eval_steps_per_second": 1.131, + "step": 960 + }, + { + "epoch": 1.0531506849315069, + "grad_norm": 2.4743072986602783, + "learning_rate": 3.633225020308692e-06, + "loss": 1.8895, + "step": 961 + }, + { + "epoch": 1.0542465753424657, + "grad_norm": 2.3666775226593018, + "learning_rate": 3.63119415109667e-06, + "loss": 1.8836, + "step": 962 + }, + { + "epoch": 1.0553424657534247, + "grad_norm": 2.301119327545166, + "learning_rate": 3.6291632818846467e-06, + "loss": 1.8541, + "step": 963 + }, + { + "epoch": 1.0564383561643835, + "grad_norm": 2.482811212539673, + "learning_rate": 3.6271324126726244e-06, + "loss": 1.88, + "step": 964 + }, + { + "epoch": 1.0575342465753426, + "grad_norm": 2.2962276935577393, + "learning_rate": 3.6251015434606013e-06, + "loss": 1.7834, + "step": 965 + }, + { + "epoch": 1.0586301369863014, + "grad_norm": 2.365130662918091, + "learning_rate": 3.623070674248579e-06, + "loss": 1.7946, + "step": 966 + }, + { + "epoch": 1.0597260273972602, + "grad_norm": 2.4644675254821777, + "learning_rate": 3.621039805036556e-06, + "loss": 1.8379, + "step": 967 + }, + { + "epoch": 1.0608219178082192, + "grad_norm": 2.31742787361145, + "learning_rate": 3.6190089358245335e-06, + "loss": 1.8192, + "step": 968 + }, + { + "epoch": 1.061917808219178, + "grad_norm": 2.1614131927490234, + "learning_rate": 3.6169780666125104e-06, + "loss": 1.8794, + "step": 969 + }, + { + "epoch": 1.063013698630137, + "grad_norm": 2.4367032051086426, + "learning_rate": 3.614947197400488e-06, + "loss": 1.8673, + "step": 970 + }, + { + "epoch": 1.063013698630137, + "eval_loss": 2.0261292457580566, + "eval_runtime": 37.1116, + "eval_samples_per_second": 26.946, + "eval_steps_per_second": 1.132, + "step": 970 + }, + { + "epoch": 1.0641095890410959, + "grad_norm": 2.262789726257324, + "learning_rate": 3.612916328188465e-06, + "loss": 1.8808, + "step": 971 + }, + { + "epoch": 1.0652054794520547, + "grad_norm": 2.316316843032837, + "learning_rate": 3.610885458976442e-06, + "loss": 1.7803, + "step": 972 + }, + { + "epoch": 1.0663013698630137, + "grad_norm": 2.2936418056488037, + "learning_rate": 3.6088545897644195e-06, + "loss": 1.8783, + "step": 973 + }, + { + "epoch": 1.0673972602739725, + "grad_norm": 2.667069673538208, + "learning_rate": 3.6068237205523963e-06, + "loss": 1.8748, + "step": 974 + }, + { + "epoch": 1.0684931506849316, + "grad_norm": 2.31111216545105, + "learning_rate": 3.604792851340374e-06, + "loss": 1.8119, + "step": 975 + }, + { + "epoch": 1.0695890410958904, + "grad_norm": 2.268258571624756, + "learning_rate": 3.602761982128351e-06, + "loss": 1.8055, + "step": 976 + }, + { + "epoch": 1.0706849315068494, + "grad_norm": 2.307137966156006, + "learning_rate": 3.6007311129163286e-06, + "loss": 1.8976, + "step": 977 + }, + { + "epoch": 1.0717808219178082, + "grad_norm": 2.406036853790283, + "learning_rate": 3.5987002437043054e-06, + "loss": 1.8608, + "step": 978 + }, + { + "epoch": 1.072876712328767, + "grad_norm": 2.3303062915802, + "learning_rate": 3.596669374492283e-06, + "loss": 1.8038, + "step": 979 + }, + { + "epoch": 1.073972602739726, + "grad_norm": 2.2280168533325195, + "learning_rate": 3.59463850528026e-06, + "loss": 1.8342, + "step": 980 + }, + { + "epoch": 1.073972602739726, + "eval_loss": 2.0273854732513428, + "eval_runtime": 37.1546, + "eval_samples_per_second": 26.915, + "eval_steps_per_second": 1.13, + "step": 980 + }, + { + "epoch": 1.0750684931506849, + "grad_norm": 2.4218759536743164, + "learning_rate": 3.5926076360682377e-06, + "loss": 1.9104, + "step": 981 + }, + { + "epoch": 1.076164383561644, + "grad_norm": 2.2140634059906006, + "learning_rate": 3.5905767668562145e-06, + "loss": 1.7927, + "step": 982 + }, + { + "epoch": 1.0772602739726027, + "grad_norm": 2.179877996444702, + "learning_rate": 3.5885458976441922e-06, + "loss": 1.8868, + "step": 983 + }, + { + "epoch": 1.0783561643835617, + "grad_norm": 2.1988537311553955, + "learning_rate": 3.586515028432169e-06, + "loss": 1.8516, + "step": 984 + }, + { + "epoch": 1.0794520547945206, + "grad_norm": 2.1724448204040527, + "learning_rate": 3.5844841592201464e-06, + "loss": 1.8869, + "step": 985 + }, + { + "epoch": 1.0805479452054794, + "grad_norm": 2.2685718536376953, + "learning_rate": 3.5824532900081237e-06, + "loss": 1.8955, + "step": 986 + }, + { + "epoch": 1.0816438356164384, + "grad_norm": 2.39284348487854, + "learning_rate": 3.580422420796101e-06, + "loss": 1.7769, + "step": 987 + }, + { + "epoch": 1.0827397260273972, + "grad_norm": 2.3448750972747803, + "learning_rate": 3.578391551584078e-06, + "loss": 1.8417, + "step": 988 + }, + { + "epoch": 1.0838356164383562, + "grad_norm": 2.3518283367156982, + "learning_rate": 3.5763606823720555e-06, + "loss": 1.8524, + "step": 989 + }, + { + "epoch": 1.084931506849315, + "grad_norm": 2.1696958541870117, + "learning_rate": 3.5743298131600328e-06, + "loss": 1.7668, + "step": 990 + }, + { + "epoch": 1.084931506849315, + "eval_loss": 2.0244593620300293, + "eval_runtime": 37.1929, + "eval_samples_per_second": 26.887, + "eval_steps_per_second": 1.129, + "step": 990 + }, + { + "epoch": 1.086027397260274, + "grad_norm": 2.18300724029541, + "learning_rate": 3.57229894394801e-06, + "loss": 1.852, + "step": 991 + }, + { + "epoch": 1.087123287671233, + "grad_norm": 2.2075037956237793, + "learning_rate": 3.5702680747359873e-06, + "loss": 1.8799, + "step": 992 + }, + { + "epoch": 1.0882191780821917, + "grad_norm": 2.2172231674194336, + "learning_rate": 3.5682372055239646e-06, + "loss": 1.8573, + "step": 993 + }, + { + "epoch": 1.0893150684931507, + "grad_norm": 2.25169038772583, + "learning_rate": 3.566206336311942e-06, + "loss": 1.8125, + "step": 994 + }, + { + "epoch": 1.0904109589041096, + "grad_norm": 2.168386697769165, + "learning_rate": 3.564175467099919e-06, + "loss": 1.857, + "step": 995 + }, + { + "epoch": 1.0915068493150686, + "grad_norm": 2.2610065937042236, + "learning_rate": 3.5621445978878964e-06, + "loss": 1.817, + "step": 996 + }, + { + "epoch": 1.0926027397260274, + "grad_norm": 2.2561864852905273, + "learning_rate": 3.5601137286758737e-06, + "loss": 1.7754, + "step": 997 + }, + { + "epoch": 1.0936986301369862, + "grad_norm": 2.143625497817993, + "learning_rate": 3.5580828594638505e-06, + "loss": 1.8734, + "step": 998 + }, + { + "epoch": 1.0947945205479452, + "grad_norm": 2.433831214904785, + "learning_rate": 3.5560519902518282e-06, + "loss": 1.8248, + "step": 999 + }, + { + "epoch": 1.095890410958904, + "grad_norm": 2.237497091293335, + "learning_rate": 3.554021121039805e-06, + "loss": 1.8543, + "step": 1000 + }, + { + "epoch": 1.095890410958904, + "eval_loss": 2.020202159881592, + "eval_runtime": 37.2413, + "eval_samples_per_second": 26.852, + "eval_steps_per_second": 1.128, + "step": 1000 + } + ], + "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": 3.11730940936192e+17, + "train_batch_size": 3, + "trial_name": null, + "trial_params": null +} diff --git a/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-1100/rng_state_0.pth b/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-1100/rng_state_0.pth new file mode 100644 index 0000000000000000000000000000000000000000..fad444f8143e1badd2a8d15a75e619abd485ceab --- /dev/null +++ b/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-1100/rng_state_0.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:186b472c4073bfc086d5b4584aeaefc23f9c40f1a8199963bda08775617e1e12 +size 16567 diff --git a/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-1100/rng_state_1.pth b/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-1100/rng_state_1.pth new file mode 100644 index 0000000000000000000000000000000000000000..39b6620f3f8d750109fe44053f80f2e221c6e290 --- /dev/null +++ b/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-1100/rng_state_1.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:d8957d748b592f49a56723f1936535313c244cee56956dbc893c99d40f6ef84b +size 16567 diff --git a/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-1100/rng_state_2.pth b/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-1100/rng_state_2.pth new file mode 100644 index 0000000000000000000000000000000000000000..04d30c4b88bb88b6860f56816d3db2c6c7742594 --- /dev/null +++ b/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-1100/rng_state_2.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:be2cd13083af0074c65198c01c90c0a3c599fbe6e0ea705a8835ca51f393d0c0 +size 16567 diff --git a/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-1100/scheduler.pt b/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-1100/scheduler.pt new file mode 100644 index 0000000000000000000000000000000000000000..6b4783ab76151fc39cdf28c9968f1003541739ce --- /dev/null +++ b/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-1100/scheduler.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:119455f8a72d175f2d5ebb240c025ad3aad0c3ba4d75fea2ee75899fff26e8c4 +size 627 diff --git a/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-1100/tokenizer.json b/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-1100/tokenizer.json new file mode 100644 index 0000000000000000000000000000000000000000..21e1edfa2183a57886f456b02efd11cce1a7df83 --- /dev/null +++ b/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-1100/tokenizer.json @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:e8485b8900f2ef3a891091dbe5ef0edc56f6922c0a23c7dfbb4baf64854f7de4 +size 17210102 diff --git a/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-1100/training_args.bin b/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-1100/training_args.bin new file mode 100644 index 0000000000000000000000000000000000000000..0628b4aecba57cabec5230705ebfea1d004cfc77 --- /dev/null +++ b/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-1100/training_args.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:f1b22d390bcc8f3a3c3fa0ce7ebfd361692a465dd5cb0c24a819e263b73e1c2e +size 6011 diff --git a/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-1200/rng_state_0.pth b/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-1200/rng_state_0.pth new file mode 100644 index 0000000000000000000000000000000000000000..3259c6605b4fec3d2abdf15c7c3ccbdef85f5d96 --- /dev/null +++ b/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-1200/rng_state_0.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:6b6aefb2eb9cf046f058c1e4bd20c70066eedd4535bf2e9f2b3f178ddbfbfd9e +size 16567 diff --git a/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-1200/rng_state_1.pth b/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-1200/rng_state_1.pth new file mode 100644 index 0000000000000000000000000000000000000000..9f4723485ff30930b59224e8afe0cb0b2b3c51bc --- /dev/null +++ b/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-1200/rng_state_1.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:f34c8972483cae48cde438f54c14179a5393e31298d634393f2e166c6a9bfb3e +size 16567 diff --git a/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-1200/rng_state_2.pth b/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-1200/rng_state_2.pth new file mode 100644 index 0000000000000000000000000000000000000000..6f245457df6ae96d9dae4e4269752c2f036e6ee3 --- /dev/null +++ b/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-1200/rng_state_2.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:0d0068d515416f5010564895d620c337c347a04bf8d5382015b5833708d52850 +size 16567 diff --git a/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-1200/scheduler.pt b/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-1200/scheduler.pt new file mode 100644 index 0000000000000000000000000000000000000000..c919d14fa610e4574f4fd9331755cee562da4ff0 --- /dev/null +++ b/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-1200/scheduler.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:3f6fbc606ea3eb86afb6064c77008ae1eee71ea66df54dd503817e7f8de93f0f +size 627 diff --git a/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-1200/tokenizer.json b/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-1200/tokenizer.json new file mode 100644 index 0000000000000000000000000000000000000000..21e1edfa2183a57886f456b02efd11cce1a7df83 --- /dev/null +++ b/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-1200/tokenizer.json @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:e8485b8900f2ef3a891091dbe5ef0edc56f6922c0a23c7dfbb4baf64854f7de4 +size 17210102 diff --git a/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-1200/training_args.bin b/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-1200/training_args.bin new file mode 100644 index 0000000000000000000000000000000000000000..0628b4aecba57cabec5230705ebfea1d004cfc77 --- /dev/null +++ b/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-1200/training_args.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:f1b22d390bcc8f3a3c3fa0ce7ebfd361692a465dd5cb0c24a819e263b73e1c2e +size 6011 diff --git a/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-1300/rng_state_0.pth b/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-1300/rng_state_0.pth new file mode 100644 index 0000000000000000000000000000000000000000..3429b5bbf5e65631c22532fed743338bede54d6b --- /dev/null +++ b/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-1300/rng_state_0.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:f7020c2ba12cf7959bf7033d7ad83dcf637e59ddbd3ff5b17fc4b9a1ebed3202 +size 16567 diff --git a/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-1300/rng_state_1.pth b/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-1300/rng_state_1.pth new file mode 100644 index 0000000000000000000000000000000000000000..cbba424cb2109da26bffbb01960eb52b2dc47b0d --- /dev/null +++ b/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-1300/rng_state_1.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:f29ab0036e394c39267ad86bf247bf1558e2213add95383fce34a980f109b42e +size 16567 diff --git a/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-1300/rng_state_2.pth b/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-1300/rng_state_2.pth new file mode 100644 index 0000000000000000000000000000000000000000..b47aeb757c57446497f8f5d1181a50c3faf2dd79 --- /dev/null +++ b/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-1300/rng_state_2.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:14413139ea5b777791c35297ed8a0a717bb3d6a3553268cc92b15fa61be1ae39 +size 16567 diff --git a/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-1300/scheduler.pt b/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-1300/scheduler.pt new file mode 100644 index 0000000000000000000000000000000000000000..a75a8c9f5dc9325555ce389e4aa850f175eb93ca --- /dev/null +++ b/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-1300/scheduler.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:7a575040a93498a27fba2ffaede0287594be19bdd971f34f5de353afde0fd1f3 +size 627 diff --git a/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-1300/tokenizer.json b/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-1300/tokenizer.json new file mode 100644 index 0000000000000000000000000000000000000000..21e1edfa2183a57886f456b02efd11cce1a7df83 --- /dev/null +++ b/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-1300/tokenizer.json @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:e8485b8900f2ef3a891091dbe5ef0edc56f6922c0a23c7dfbb4baf64854f7de4 +size 17210102 diff --git a/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-1300/training_args.bin b/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-1300/training_args.bin new file mode 100644 index 0000000000000000000000000000000000000000..0628b4aecba57cabec5230705ebfea1d004cfc77 --- /dev/null +++ b/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-1300/training_args.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:f1b22d390bcc8f3a3c3fa0ce7ebfd361692a465dd5cb0c24a819e263b73e1c2e +size 6011 diff --git a/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-1500/rng_state_0.pth b/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-1500/rng_state_0.pth new file mode 100644 index 0000000000000000000000000000000000000000..491d7aa6873ac345ad7d2d8238b3751df06748aa --- /dev/null +++ b/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-1500/rng_state_0.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:dc77a0f3d732b76deec0b9f9ee77f2ad22c4c8a301af0167ff41a26d22d5beb3 +size 16567 diff --git a/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-1500/rng_state_1.pth b/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-1500/rng_state_1.pth new file mode 100644 index 0000000000000000000000000000000000000000..0c1653c04de59e53b4d05673f04ed4b2c7039c73 --- /dev/null +++ b/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-1500/rng_state_1.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:7a2973146fd35e5e1a515f6332b8a2b33728c4360853ce2c03e1a7a90b79c2cc +size 16567 diff --git a/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-1500/rng_state_2.pth b/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-1500/rng_state_2.pth new file mode 100644 index 0000000000000000000000000000000000000000..4cf5bfa682e62dbe5c2e496536a84208dc8433d0 --- /dev/null +++ b/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-1500/rng_state_2.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:8d1658c107e9585fdf02f4eb4b846349794600f9e1c2f5151d959824c6021559 +size 16567 diff --git a/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-1500/scheduler.pt b/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-1500/scheduler.pt new file mode 100644 index 0000000000000000000000000000000000000000..62f5a834bf01763776e29085543fb94d551f7728 --- /dev/null +++ b/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-1500/scheduler.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:f4abadbee26e217bf2e89e4f87751042b0e000146aaac1d8b7c1cbce30a3dfa1 +size 627 diff --git a/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-1500/tokenizer.json b/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-1500/tokenizer.json new file mode 100644 index 0000000000000000000000000000000000000000..21e1edfa2183a57886f456b02efd11cce1a7df83 --- /dev/null +++ b/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-1500/tokenizer.json @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:e8485b8900f2ef3a891091dbe5ef0edc56f6922c0a23c7dfbb4baf64854f7de4 +size 17210102 diff --git a/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-1500/training_args.bin b/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-1500/training_args.bin new file mode 100644 index 0000000000000000000000000000000000000000..0628b4aecba57cabec5230705ebfea1d004cfc77 --- /dev/null +++ b/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-1500/training_args.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:f1b22d390bcc8f3a3c3fa0ce7ebfd361692a465dd5cb0c24a819e263b73e1c2e +size 6011 diff --git a/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-1600/rng_state_0.pth b/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-1600/rng_state_0.pth new file mode 100644 index 0000000000000000000000000000000000000000..5509f0e1ece25901639e38b63337f0f28e33b98d --- /dev/null +++ b/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_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/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-1600/rng_state_1.pth b/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-1600/rng_state_1.pth new file mode 100644 index 0000000000000000000000000000000000000000..bdaddc4981c690246dfa665b10da40b38a679bc3 --- /dev/null +++ b/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_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/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-1600/rng_state_2.pth b/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-1600/rng_state_2.pth new file mode 100644 index 0000000000000000000000000000000000000000..fd1ca9202bb0004e7be63108d043e05db918daf3 --- /dev/null +++ b/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_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/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-1600/scheduler.pt b/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-1600/scheduler.pt new file mode 100644 index 0000000000000000000000000000000000000000..d1562189f8cda31a70d795f024169d4f59752074 --- /dev/null +++ b/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-1600/scheduler.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:e18b0e7e9e00ba355ada3ada0702d9b891c2fd6c15a02fbfa0e9c5860e6c540d +size 627 diff --git a/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-1600/tokenizer.json b/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-1600/tokenizer.json new file mode 100644 index 0000000000000000000000000000000000000000..21e1edfa2183a57886f456b02efd11cce1a7df83 --- /dev/null +++ b/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-1600/tokenizer.json @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:e8485b8900f2ef3a891091dbe5ef0edc56f6922c0a23c7dfbb4baf64854f7de4 +size 17210102 diff --git a/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-1600/training_args.bin b/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-1600/training_args.bin new file mode 100644 index 0000000000000000000000000000000000000000..0628b4aecba57cabec5230705ebfea1d004cfc77 --- /dev/null +++ b/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-1600/training_args.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:f1b22d390bcc8f3a3c3fa0ce7ebfd361692a465dd5cb0c24a819e263b73e1c2e +size 6011 diff --git a/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-1700/rng_state_0.pth b/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-1700/rng_state_0.pth new file mode 100644 index 0000000000000000000000000000000000000000..173f12bde4ca76ddf4b03b0964b7524eec1d0cf8 --- /dev/null +++ b/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-1700/rng_state_0.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:103551ff1c3e292b45450c5a23cdaf1958c78f242d8cf8b564df9603b5e96d5c +size 16567 diff --git a/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-1700/rng_state_1.pth b/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-1700/rng_state_1.pth new file mode 100644 index 0000000000000000000000000000000000000000..b0aed44f2fa8fd35795e9b8bfde272b7a98d0ec9 --- /dev/null +++ b/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-1700/rng_state_1.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:9d69b787213a6aea61f516b084db235c8c80de8bef9f33a5157ae51d85fa9135 +size 16567 diff --git a/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-1700/rng_state_2.pth b/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-1700/rng_state_2.pth new file mode 100644 index 0000000000000000000000000000000000000000..487463491926bb630fa016cfe776df854285e817 --- /dev/null +++ b/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-1700/rng_state_2.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:ffe85d84d1459bc77a8988c2ddaa541ec71b9112c567063ec91af255e984793b +size 16567 diff --git a/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-1700/scheduler.pt b/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-1700/scheduler.pt new file mode 100644 index 0000000000000000000000000000000000000000..4e2599b24bc77b7cbf5b941e14863dc395eb0aee --- /dev/null +++ b/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-1700/scheduler.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:0d06926a97f807ed50063a0fbaaef06f637672c0bf846d322b5f2fd5d678dc96 +size 627 diff --git a/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-1700/tokenizer.json b/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-1700/tokenizer.json new file mode 100644 index 0000000000000000000000000000000000000000..21e1edfa2183a57886f456b02efd11cce1a7df83 --- /dev/null +++ b/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-1700/tokenizer.json @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:e8485b8900f2ef3a891091dbe5ef0edc56f6922c0a23c7dfbb4baf64854f7de4 +size 17210102 diff --git a/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-1700/training_args.bin b/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-1700/training_args.bin new file mode 100644 index 0000000000000000000000000000000000000000..0628b4aecba57cabec5230705ebfea1d004cfc77 --- /dev/null +++ b/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-1700/training_args.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:f1b22d390bcc8f3a3c3fa0ce7ebfd361692a465dd5cb0c24a819e263b73e1c2e +size 6011 diff --git a/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-2100/config.json b/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-2100/config.json new file mode 100644 index 0000000000000000000000000000000000000000..fb0f9a53a507ffcdb60410deb71d60ef801bf350 --- /dev/null +++ b/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-2100/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/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-2100/generation_config.json b/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-2100/generation_config.json new file mode 100644 index 0000000000000000000000000000000000000000..9c389b841a9a29ddd904e02d1eb0e08dcce82ad9 --- /dev/null +++ b/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-2100/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/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-2100/latest b/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-2100/latest new file mode 100644 index 0000000000000000000000000000000000000000..b05c33b7aff64e27ba9c8b2e7dc2454bea58ee7f --- /dev/null +++ b/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-2100/latest @@ -0,0 +1 @@ +global_step2100 \ No newline at end of file diff --git a/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-2100/model.safetensors.index.json b/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-2100/model.safetensors.index.json new file mode 100644 index 0000000000000000000000000000000000000000..ed64de846d720b9a7859dc20575fea8e8ca51940 --- /dev/null +++ b/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-2100/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/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-2100/special_tokens_map.json b/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-2100/special_tokens_map.json new file mode 100644 index 0000000000000000000000000000000000000000..6949c5975ee0e961ef61cf31010dce04df0a03f8 --- /dev/null +++ b/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-2100/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/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-2100/tokenizer_config.json b/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-2100/tokenizer_config.json new file mode 100644 index 0000000000000000000000000000000000000000..5684b14e2756b56aa397c1324e239b442162b34a --- /dev/null +++ b/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-2100/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/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-2100/trainer_state.json b/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-2100/trainer_state.json new file mode 100644 index 0000000000000000000000000000000000000000..21273159afba1c9dde79304e4e81bfabcb3f0d4c --- /dev/null +++ b/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-2100/trainer_state.json @@ -0,0 +1,16413 @@ +{ + "best_metric": null, + "best_model_checkpoint": null, + "epoch": 2.3013698630136985, + "eval_steps": 10, + "global_step": 2100, + "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": 3.0476, + "step": 1 + }, + { + "epoch": 0.002191780821917808, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.0289, + "step": 2 + }, + { + "epoch": 0.0032876712328767125, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 2.9935, + "step": 3 + }, + { + "epoch": 0.004383561643835616, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.0139, + "step": 4 + }, + { + "epoch": 0.005479452054794521, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 2.9892, + "step": 5 + }, + { + "epoch": 0.006575342465753425, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.1252, + "step": 6 + }, + { + "epoch": 0.007671232876712329, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.0291, + "step": 7 + }, + { + "epoch": 0.008767123287671232, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.0607, + "step": 8 + }, + { + "epoch": 0.009863013698630137, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.0306, + "step": 9 + }, + { + "epoch": 0.010958904109589041, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.0811, + "step": 10 + }, + { + "epoch": 0.010958904109589041, + "eval_loss": 3.0768930912017822, + "eval_runtime": 38.0377, + "eval_samples_per_second": 26.29, + "eval_steps_per_second": 1.104, + "step": 10 + }, + { + "epoch": 0.012054794520547946, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.0693, + "step": 11 + }, + { + "epoch": 0.01315068493150685, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 2.9668, + "step": 12 + }, + { + "epoch": 0.014246575342465753, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.052, + "step": 13 + }, + { + "epoch": 0.015342465753424657, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.0363, + "step": 14 + }, + { + "epoch": 0.01643835616438356, + "grad_norm": 5.58175802230835, + "learning_rate": 1.8248175182481753e-08, + "loss": 3.0778, + "step": 15 + }, + { + "epoch": 0.017534246575342465, + "grad_norm": 6.360531806945801, + "learning_rate": 3.6496350364963505e-08, + "loss": 3.0257, + "step": 16 + }, + { + "epoch": 0.01863013698630137, + "grad_norm": 6.453113555908203, + "learning_rate": 5.4744525547445264e-08, + "loss": 3.0262, + "step": 17 + }, + { + "epoch": 0.019726027397260273, + "grad_norm": 6.789405345916748, + "learning_rate": 7.299270072992701e-08, + "loss": 3.0671, + "step": 18 + }, + { + "epoch": 0.020821917808219178, + "grad_norm": 6.185514450073242, + "learning_rate": 9.124087591240876e-08, + "loss": 3.0077, + "step": 19 + }, + { + "epoch": 0.021917808219178082, + "grad_norm": 5.719798564910889, + "learning_rate": 1.0948905109489053e-07, + "loss": 3.089, + "step": 20 + }, + { + "epoch": 0.021917808219178082, + "eval_loss": 3.07546329498291, + "eval_runtime": 37.5318, + "eval_samples_per_second": 26.644, + "eval_steps_per_second": 1.119, + "step": 20 + }, + { + "epoch": 0.023013698630136987, + "grad_norm": 6.307768821716309, + "learning_rate": 1.2773722627737227e-07, + "loss": 3.0641, + "step": 21 + }, + { + "epoch": 0.02410958904109589, + "grad_norm": 6.255112171173096, + "learning_rate": 1.4598540145985402e-07, + "loss": 2.9849, + "step": 22 + }, + { + "epoch": 0.025205479452054796, + "grad_norm": 5.812101364135742, + "learning_rate": 1.642335766423358e-07, + "loss": 3.0211, + "step": 23 + }, + { + "epoch": 0.0263013698630137, + "grad_norm": 6.262075901031494, + "learning_rate": 1.8248175182481753e-07, + "loss": 3.0654, + "step": 24 + }, + { + "epoch": 0.0273972602739726, + "grad_norm": 5.419229507446289, + "learning_rate": 2.0072992700729928e-07, + "loss": 3.0661, + "step": 25 + }, + { + "epoch": 0.028493150684931506, + "grad_norm": 4.750737190246582, + "learning_rate": 2.1897810218978106e-07, + "loss": 3.0494, + "step": 26 + }, + { + "epoch": 0.02958904109589041, + "grad_norm": 4.732361316680908, + "learning_rate": 2.3722627737226278e-07, + "loss": 3.0471, + "step": 27 + }, + { + "epoch": 0.030684931506849315, + "grad_norm": 4.749786853790283, + "learning_rate": 2.5547445255474454e-07, + "loss": 3.0107, + "step": 28 + }, + { + "epoch": 0.03178082191780822, + "grad_norm": 4.451244831085205, + "learning_rate": 2.737226277372263e-07, + "loss": 2.9987, + "step": 29 + }, + { + "epoch": 0.03287671232876712, + "grad_norm": 4.5158257484436035, + "learning_rate": 2.9197080291970804e-07, + "loss": 2.9686, + "step": 30 + }, + { + "epoch": 0.03287671232876712, + "eval_loss": 3.031872034072876, + "eval_runtime": 37.2608, + "eval_samples_per_second": 26.838, + "eval_steps_per_second": 1.127, + "step": 30 + }, + { + "epoch": 0.03397260273972603, + "grad_norm": 3.8239498138427734, + "learning_rate": 3.102189781021898e-07, + "loss": 3.0057, + "step": 31 + }, + { + "epoch": 0.03506849315068493, + "grad_norm": 4.037254810333252, + "learning_rate": 3.284671532846716e-07, + "loss": 2.9733, + "step": 32 + }, + { + "epoch": 0.03616438356164384, + "grad_norm": 4.283931732177734, + "learning_rate": 3.467153284671533e-07, + "loss": 2.9515, + "step": 33 + }, + { + "epoch": 0.03726027397260274, + "grad_norm": 4.1784987449646, + "learning_rate": 3.6496350364963505e-07, + "loss": 2.9483, + "step": 34 + }, + { + "epoch": 0.038356164383561646, + "grad_norm": 3.9210753440856934, + "learning_rate": 3.8321167883211686e-07, + "loss": 2.9679, + "step": 35 + }, + { + "epoch": 0.03945205479452055, + "grad_norm": 3.72884202003479, + "learning_rate": 4.0145985401459856e-07, + "loss": 2.9989, + "step": 36 + }, + { + "epoch": 0.040547945205479455, + "grad_norm": 3.2882330417633057, + "learning_rate": 4.197080291970803e-07, + "loss": 2.9384, + "step": 37 + }, + { + "epoch": 0.041643835616438356, + "grad_norm": 3.515293836593628, + "learning_rate": 4.379562043795621e-07, + "loss": 2.8986, + "step": 38 + }, + { + "epoch": 0.042739726027397264, + "grad_norm": 3.568453073501587, + "learning_rate": 4.562043795620438e-07, + "loss": 2.9854, + "step": 39 + }, + { + "epoch": 0.043835616438356165, + "grad_norm": 3.389789342880249, + "learning_rate": 4.7445255474452557e-07, + "loss": 2.8891, + "step": 40 + }, + { + "epoch": 0.043835616438356165, + "eval_loss": 2.964235544204712, + "eval_runtime": 37.3406, + "eval_samples_per_second": 26.781, + "eval_steps_per_second": 1.125, + "step": 40 + }, + { + "epoch": 0.044931506849315066, + "grad_norm": 3.1826283931732178, + "learning_rate": 4.927007299270073e-07, + "loss": 2.9271, + "step": 41 + }, + { + "epoch": 0.046027397260273974, + "grad_norm": 2.941539764404297, + "learning_rate": 5.109489051094891e-07, + "loss": 2.9486, + "step": 42 + }, + { + "epoch": 0.047123287671232875, + "grad_norm": 3.0076773166656494, + "learning_rate": 5.291970802919708e-07, + "loss": 2.9252, + "step": 43 + }, + { + "epoch": 0.04821917808219178, + "grad_norm": 2.79028582572937, + "learning_rate": 5.474452554744526e-07, + "loss": 2.9277, + "step": 44 + }, + { + "epoch": 0.049315068493150684, + "grad_norm": 2.603456735610962, + "learning_rate": 5.656934306569343e-07, + "loss": 2.9144, + "step": 45 + }, + { + "epoch": 0.05041095890410959, + "grad_norm": 2.674888849258423, + "learning_rate": 5.839416058394161e-07, + "loss": 2.8732, + "step": 46 + }, + { + "epoch": 0.05150684931506849, + "grad_norm": 2.972093343734741, + "learning_rate": 6.021897810218979e-07, + "loss": 2.8623, + "step": 47 + }, + { + "epoch": 0.0526027397260274, + "grad_norm": 2.757766008377075, + "learning_rate": 6.204379562043796e-07, + "loss": 2.8941, + "step": 48 + }, + { + "epoch": 0.0536986301369863, + "grad_norm": 2.910634994506836, + "learning_rate": 6.386861313868614e-07, + "loss": 2.8101, + "step": 49 + }, + { + "epoch": 0.0547945205479452, + "grad_norm": 2.5542638301849365, + "learning_rate": 6.569343065693432e-07, + "loss": 2.9262, + "step": 50 + }, + { + "epoch": 0.0547945205479452, + "eval_loss": 2.883579969406128, + "eval_runtime": 37.326, + "eval_samples_per_second": 26.791, + "eval_steps_per_second": 1.125, + "step": 50 + }, + { + "epoch": 0.05589041095890411, + "grad_norm": 2.333974599838257, + "learning_rate": 6.751824817518248e-07, + "loss": 2.8217, + "step": 51 + }, + { + "epoch": 0.05698630136986301, + "grad_norm": 2.181022882461548, + "learning_rate": 6.934306569343066e-07, + "loss": 2.8483, + "step": 52 + }, + { + "epoch": 0.05808219178082192, + "grad_norm": 2.12200927734375, + "learning_rate": 7.116788321167883e-07, + "loss": 2.7788, + "step": 53 + }, + { + "epoch": 0.05917808219178082, + "grad_norm": 2.052626371383667, + "learning_rate": 7.299270072992701e-07, + "loss": 2.7774, + "step": 54 + }, + { + "epoch": 0.06027397260273973, + "grad_norm": 2.11487078666687, + "learning_rate": 7.48175182481752e-07, + "loss": 2.8045, + "step": 55 + }, + { + "epoch": 0.06136986301369863, + "grad_norm": 2.083588123321533, + "learning_rate": 7.664233576642337e-07, + "loss": 2.8119, + "step": 56 + }, + { + "epoch": 0.06246575342465754, + "grad_norm": 2.2079808712005615, + "learning_rate": 7.846715328467154e-07, + "loss": 2.82, + "step": 57 + }, + { + "epoch": 0.06356164383561644, + "grad_norm": 2.100802421569824, + "learning_rate": 8.029197080291971e-07, + "loss": 2.794, + "step": 58 + }, + { + "epoch": 0.06465753424657535, + "grad_norm": 1.8322066068649292, + "learning_rate": 8.211678832116789e-07, + "loss": 2.7444, + "step": 59 + }, + { + "epoch": 0.06575342465753424, + "grad_norm": 2.077005386352539, + "learning_rate": 8.394160583941606e-07, + "loss": 2.7093, + "step": 60 + }, + { + "epoch": 0.06575342465753424, + "eval_loss": 2.8177874088287354, + "eval_runtime": 37.2903, + "eval_samples_per_second": 26.817, + "eval_steps_per_second": 1.126, + "step": 60 + }, + { + "epoch": 0.06684931506849315, + "grad_norm": 2.1688592433929443, + "learning_rate": 8.576642335766425e-07, + "loss": 2.7446, + "step": 61 + }, + { + "epoch": 0.06794520547945206, + "grad_norm": 2.1709237098693848, + "learning_rate": 8.759124087591242e-07, + "loss": 2.736, + "step": 62 + }, + { + "epoch": 0.06904109589041096, + "grad_norm": 2.3578364849090576, + "learning_rate": 8.94160583941606e-07, + "loss": 2.7583, + "step": 63 + }, + { + "epoch": 0.07013698630136986, + "grad_norm": 1.9025483131408691, + "learning_rate": 9.124087591240876e-07, + "loss": 2.7651, + "step": 64 + }, + { + "epoch": 0.07123287671232877, + "grad_norm": 1.9450963735580444, + "learning_rate": 9.306569343065694e-07, + "loss": 2.7186, + "step": 65 + }, + { + "epoch": 0.07232876712328767, + "grad_norm": 1.957723617553711, + "learning_rate": 9.489051094890511e-07, + "loss": 2.7421, + "step": 66 + }, + { + "epoch": 0.07342465753424658, + "grad_norm": 1.9823631048202515, + "learning_rate": 9.67153284671533e-07, + "loss": 2.7258, + "step": 67 + }, + { + "epoch": 0.07452054794520548, + "grad_norm": 1.945495843887329, + "learning_rate": 9.854014598540146e-07, + "loss": 2.7733, + "step": 68 + }, + { + "epoch": 0.07561643835616438, + "grad_norm": 1.9684759378433228, + "learning_rate": 1.0036496350364965e-06, + "loss": 2.7041, + "step": 69 + }, + { + "epoch": 0.07671232876712329, + "grad_norm": 2.087336540222168, + "learning_rate": 1.0218978102189781e-06, + "loss": 2.6662, + "step": 70 + }, + { + "epoch": 0.07671232876712329, + "eval_loss": 2.7529091835021973, + "eval_runtime": 37.3122, + "eval_samples_per_second": 26.801, + "eval_steps_per_second": 1.126, + "step": 70 + }, + { + "epoch": 0.07780821917808219, + "grad_norm": 1.7790117263793945, + "learning_rate": 1.04014598540146e-06, + "loss": 2.7379, + "step": 71 + }, + { + "epoch": 0.0789041095890411, + "grad_norm": 2.2780861854553223, + "learning_rate": 1.0583941605839416e-06, + "loss": 2.6234, + "step": 72 + }, + { + "epoch": 0.08, + "grad_norm": 2.1144700050354004, + "learning_rate": 1.0766423357664233e-06, + "loss": 2.6248, + "step": 73 + }, + { + "epoch": 0.08109589041095891, + "grad_norm": 2.0247490406036377, + "learning_rate": 1.0948905109489052e-06, + "loss": 2.6456, + "step": 74 + }, + { + "epoch": 0.0821917808219178, + "grad_norm": 2.702547311782837, + "learning_rate": 1.113138686131387e-06, + "loss": 2.6759, + "step": 75 + }, + { + "epoch": 0.08328767123287671, + "grad_norm": 1.8208750486373901, + "learning_rate": 1.1313868613138687e-06, + "loss": 2.6845, + "step": 76 + }, + { + "epoch": 0.08438356164383562, + "grad_norm": 2.3829221725463867, + "learning_rate": 1.1496350364963505e-06, + "loss": 2.61, + "step": 77 + }, + { + "epoch": 0.08547945205479453, + "grad_norm": 2.580885887145996, + "learning_rate": 1.1678832116788322e-06, + "loss": 2.5896, + "step": 78 + }, + { + "epoch": 0.08657534246575342, + "grad_norm": 2.2477641105651855, + "learning_rate": 1.1861313868613138e-06, + "loss": 2.596, + "step": 79 + }, + { + "epoch": 0.08767123287671233, + "grad_norm": 2.1895623207092285, + "learning_rate": 1.2043795620437959e-06, + "loss": 2.5906, + "step": 80 + }, + { + "epoch": 0.08767123287671233, + "eval_loss": 2.6815176010131836, + "eval_runtime": 37.3342, + "eval_samples_per_second": 26.785, + "eval_steps_per_second": 1.125, + "step": 80 + }, + { + "epoch": 0.08876712328767124, + "grad_norm": 2.217090606689453, + "learning_rate": 1.2226277372262775e-06, + "loss": 2.6005, + "step": 81 + }, + { + "epoch": 0.08986301369863013, + "grad_norm": 2.4679362773895264, + "learning_rate": 1.2408759124087592e-06, + "loss": 2.6333, + "step": 82 + }, + { + "epoch": 0.09095890410958904, + "grad_norm": 2.2590575218200684, + "learning_rate": 1.259124087591241e-06, + "loss": 2.5814, + "step": 83 + }, + { + "epoch": 0.09205479452054795, + "grad_norm": 1.877977728843689, + "learning_rate": 1.2773722627737229e-06, + "loss": 2.5948, + "step": 84 + }, + { + "epoch": 0.09315068493150686, + "grad_norm": 2.1535534858703613, + "learning_rate": 1.2956204379562045e-06, + "loss": 2.5781, + "step": 85 + }, + { + "epoch": 0.09424657534246575, + "grad_norm": 2.5001144409179688, + "learning_rate": 1.3138686131386864e-06, + "loss": 2.5339, + "step": 86 + }, + { + "epoch": 0.09534246575342466, + "grad_norm": 2.3279049396514893, + "learning_rate": 1.332116788321168e-06, + "loss": 2.545, + "step": 87 + }, + { + "epoch": 0.09643835616438357, + "grad_norm": 2.2230734825134277, + "learning_rate": 1.3503649635036497e-06, + "loss": 2.4745, + "step": 88 + }, + { + "epoch": 0.09753424657534246, + "grad_norm": 2.8762636184692383, + "learning_rate": 1.3686131386861315e-06, + "loss": 2.516, + "step": 89 + }, + { + "epoch": 0.09863013698630137, + "grad_norm": 2.153506278991699, + "learning_rate": 1.3868613138686132e-06, + "loss": 2.4547, + "step": 90 + }, + { + "epoch": 0.09863013698630137, + "eval_loss": 2.603238344192505, + "eval_runtime": 37.4162, + "eval_samples_per_second": 26.726, + "eval_steps_per_second": 1.123, + "step": 90 + }, + { + "epoch": 0.09972602739726028, + "grad_norm": 2.4143614768981934, + "learning_rate": 1.405109489051095e-06, + "loss": 2.547, + "step": 91 + }, + { + "epoch": 0.10082191780821918, + "grad_norm": 2.354632616043091, + "learning_rate": 1.4233576642335767e-06, + "loss": 2.5009, + "step": 92 + }, + { + "epoch": 0.10191780821917808, + "grad_norm": 2.6605563163757324, + "learning_rate": 1.4416058394160583e-06, + "loss": 2.5304, + "step": 93 + }, + { + "epoch": 0.10301369863013699, + "grad_norm": 2.3724122047424316, + "learning_rate": 1.4598540145985402e-06, + "loss": 2.4925, + "step": 94 + }, + { + "epoch": 0.10410958904109589, + "grad_norm": 3.6323022842407227, + "learning_rate": 1.478102189781022e-06, + "loss": 2.4592, + "step": 95 + }, + { + "epoch": 0.1052054794520548, + "grad_norm": 2.3407089710235596, + "learning_rate": 1.496350364963504e-06, + "loss": 2.5223, + "step": 96 + }, + { + "epoch": 0.1063013698630137, + "grad_norm": 2.8238413333892822, + "learning_rate": 1.5145985401459856e-06, + "loss": 2.489, + "step": 97 + }, + { + "epoch": 0.1073972602739726, + "grad_norm": 2.6433658599853516, + "learning_rate": 1.5328467153284674e-06, + "loss": 2.5101, + "step": 98 + }, + { + "epoch": 0.10849315068493151, + "grad_norm": 2.562761068344116, + "learning_rate": 1.551094890510949e-06, + "loss": 2.4642, + "step": 99 + }, + { + "epoch": 0.1095890410958904, + "grad_norm": 2.4501149654388428, + "learning_rate": 1.5693430656934307e-06, + "loss": 2.4102, + "step": 100 + }, + { + "epoch": 0.1095890410958904, + "eval_loss": 2.5452804565429688, + "eval_runtime": 37.4224, + "eval_samples_per_second": 26.722, + "eval_steps_per_second": 1.122, + "step": 100 + }, + { + "epoch": 0.11068493150684931, + "grad_norm": 2.475864887237549, + "learning_rate": 1.5875912408759126e-06, + "loss": 2.4147, + "step": 101 + }, + { + "epoch": 0.11178082191780822, + "grad_norm": 2.558875322341919, + "learning_rate": 1.6058394160583942e-06, + "loss": 2.4694, + "step": 102 + }, + { + "epoch": 0.11287671232876713, + "grad_norm": 2.1085734367370605, + "learning_rate": 1.624087591240876e-06, + "loss": 2.5099, + "step": 103 + }, + { + "epoch": 0.11397260273972602, + "grad_norm": 2.410731792449951, + "learning_rate": 1.6423357664233577e-06, + "loss": 2.432, + "step": 104 + }, + { + "epoch": 0.11506849315068493, + "grad_norm": 2.2800745964050293, + "learning_rate": 1.6605839416058394e-06, + "loss": 2.3669, + "step": 105 + }, + { + "epoch": 0.11616438356164384, + "grad_norm": 2.2234933376312256, + "learning_rate": 1.6788321167883212e-06, + "loss": 2.4145, + "step": 106 + }, + { + "epoch": 0.11726027397260275, + "grad_norm": 2.3367040157318115, + "learning_rate": 1.6970802919708029e-06, + "loss": 2.4272, + "step": 107 + }, + { + "epoch": 0.11835616438356164, + "grad_norm": 2.7002811431884766, + "learning_rate": 1.715328467153285e-06, + "loss": 2.3478, + "step": 108 + }, + { + "epoch": 0.11945205479452055, + "grad_norm": 2.6628332138061523, + "learning_rate": 1.7335766423357666e-06, + "loss": 2.4096, + "step": 109 + }, + { + "epoch": 0.12054794520547946, + "grad_norm": 2.4174230098724365, + "learning_rate": 1.7518248175182485e-06, + "loss": 2.407, + "step": 110 + }, + { + "epoch": 0.12054794520547946, + "eval_loss": 2.491746664047241, + "eval_runtime": 37.2555, + "eval_samples_per_second": 26.842, + "eval_steps_per_second": 1.127, + "step": 110 + }, + { + "epoch": 0.12164383561643835, + "grad_norm": 2.24796986579895, + "learning_rate": 1.77007299270073e-06, + "loss": 2.4369, + "step": 111 + }, + { + "epoch": 0.12273972602739726, + "grad_norm": 2.720939874649048, + "learning_rate": 1.788321167883212e-06, + "loss": 2.4116, + "step": 112 + }, + { + "epoch": 0.12383561643835617, + "grad_norm": 2.6907670497894287, + "learning_rate": 1.8065693430656936e-06, + "loss": 2.4577, + "step": 113 + }, + { + "epoch": 0.12493150684931507, + "grad_norm": 2.559636354446411, + "learning_rate": 1.8248175182481753e-06, + "loss": 2.3999, + "step": 114 + }, + { + "epoch": 0.12602739726027398, + "grad_norm": 2.6730551719665527, + "learning_rate": 1.8430656934306571e-06, + "loss": 2.3428, + "step": 115 + }, + { + "epoch": 0.12712328767123288, + "grad_norm": 3.172179937362671, + "learning_rate": 1.8613138686131388e-06, + "loss": 2.3722, + "step": 116 + }, + { + "epoch": 0.12821917808219177, + "grad_norm": 2.6757171154022217, + "learning_rate": 1.8795620437956206e-06, + "loss": 2.3572, + "step": 117 + }, + { + "epoch": 0.1293150684931507, + "grad_norm": 2.5840766429901123, + "learning_rate": 1.8978102189781023e-06, + "loss": 2.353, + "step": 118 + }, + { + "epoch": 0.1304109589041096, + "grad_norm": 2.6248581409454346, + "learning_rate": 1.916058394160584e-06, + "loss": 2.36, + "step": 119 + }, + { + "epoch": 0.13150684931506848, + "grad_norm": 2.428516387939453, + "learning_rate": 1.934306569343066e-06, + "loss": 2.3821, + "step": 120 + }, + { + "epoch": 0.13150684931506848, + "eval_loss": 2.451565980911255, + "eval_runtime": 37.3669, + "eval_samples_per_second": 26.762, + "eval_steps_per_second": 1.124, + "step": 120 + }, + { + "epoch": 0.1326027397260274, + "grad_norm": 2.9956233501434326, + "learning_rate": 1.952554744525548e-06, + "loss": 2.2778, + "step": 121 + }, + { + "epoch": 0.1336986301369863, + "grad_norm": 2.4120893478393555, + "learning_rate": 1.9708029197080293e-06, + "loss": 2.325, + "step": 122 + }, + { + "epoch": 0.13479452054794522, + "grad_norm": 2.1410024166107178, + "learning_rate": 1.989051094890511e-06, + "loss": 2.3248, + "step": 123 + }, + { + "epoch": 0.1358904109589041, + "grad_norm": 2.923799514770508, + "learning_rate": 2.007299270072993e-06, + "loss": 2.327, + "step": 124 + }, + { + "epoch": 0.136986301369863, + "grad_norm": 3.0770339965820312, + "learning_rate": 2.025547445255475e-06, + "loss": 2.3847, + "step": 125 + }, + { + "epoch": 0.13808219178082193, + "grad_norm": 2.6408181190490723, + "learning_rate": 2.0437956204379563e-06, + "loss": 2.3196, + "step": 126 + }, + { + "epoch": 0.13917808219178082, + "grad_norm": 2.9127118587493896, + "learning_rate": 2.062043795620438e-06, + "loss": 2.3287, + "step": 127 + }, + { + "epoch": 0.14027397260273972, + "grad_norm": 2.9805052280426025, + "learning_rate": 2.08029197080292e-06, + "loss": 2.3243, + "step": 128 + }, + { + "epoch": 0.14136986301369864, + "grad_norm": 3.1342010498046875, + "learning_rate": 2.0985401459854014e-06, + "loss": 2.38, + "step": 129 + }, + { + "epoch": 0.14246575342465753, + "grad_norm": 2.678422451019287, + "learning_rate": 2.1167883211678833e-06, + "loss": 2.2727, + "step": 130 + }, + { + "epoch": 0.14246575342465753, + "eval_loss": 2.4187021255493164, + "eval_runtime": 37.3117, + "eval_samples_per_second": 26.801, + "eval_steps_per_second": 1.126, + "step": 130 + }, + { + "epoch": 0.14356164383561643, + "grad_norm": 2.9281492233276367, + "learning_rate": 2.135036496350365e-06, + "loss": 2.3327, + "step": 131 + }, + { + "epoch": 0.14465753424657535, + "grad_norm": 2.9592933654785156, + "learning_rate": 2.1532846715328466e-06, + "loss": 2.3674, + "step": 132 + }, + { + "epoch": 0.14575342465753424, + "grad_norm": 3.404761552810669, + "learning_rate": 2.171532846715329e-06, + "loss": 2.2679, + "step": 133 + }, + { + "epoch": 0.14684931506849316, + "grad_norm": 3.635571241378784, + "learning_rate": 2.1897810218978103e-06, + "loss": 2.2799, + "step": 134 + }, + { + "epoch": 0.14794520547945206, + "grad_norm": 3.9130170345306396, + "learning_rate": 2.208029197080292e-06, + "loss": 2.3263, + "step": 135 + }, + { + "epoch": 0.14904109589041095, + "grad_norm": 4.300438404083252, + "learning_rate": 2.226277372262774e-06, + "loss": 2.2969, + "step": 136 + }, + { + "epoch": 0.15013698630136987, + "grad_norm": 2.863511085510254, + "learning_rate": 2.244525547445256e-06, + "loss": 2.297, + "step": 137 + }, + { + "epoch": 0.15123287671232877, + "grad_norm": 4.910321235656738, + "learning_rate": 2.2627737226277373e-06, + "loss": 2.3269, + "step": 138 + }, + { + "epoch": 0.15232876712328766, + "grad_norm": 3.3323862552642822, + "learning_rate": 2.281021897810219e-06, + "loss": 2.3195, + "step": 139 + }, + { + "epoch": 0.15342465753424658, + "grad_norm": 3.2672693729400635, + "learning_rate": 2.299270072992701e-06, + "loss": 2.2797, + "step": 140 + }, + { + "epoch": 0.15342465753424658, + "eval_loss": 2.3943886756896973, + "eval_runtime": 37.3137, + "eval_samples_per_second": 26.8, + "eval_steps_per_second": 1.126, + "step": 140 + }, + { + "epoch": 0.15452054794520548, + "grad_norm": 3.9229774475097656, + "learning_rate": 2.3175182481751825e-06, + "loss": 2.2718, + "step": 141 + }, + { + "epoch": 0.15561643835616437, + "grad_norm": 2.6409823894500732, + "learning_rate": 2.3357664233576643e-06, + "loss": 2.2828, + "step": 142 + }, + { + "epoch": 0.1567123287671233, + "grad_norm": 4.6122612953186035, + "learning_rate": 2.354014598540146e-06, + "loss": 2.247, + "step": 143 + }, + { + "epoch": 0.1578082191780822, + "grad_norm": 3.391357898712158, + "learning_rate": 2.3722627737226276e-06, + "loss": 2.2863, + "step": 144 + }, + { + "epoch": 0.1589041095890411, + "grad_norm": 3.3550853729248047, + "learning_rate": 2.39051094890511e-06, + "loss": 2.2789, + "step": 145 + }, + { + "epoch": 0.16, + "grad_norm": 3.7376883029937744, + "learning_rate": 2.4087591240875918e-06, + "loss": 2.2297, + "step": 146 + }, + { + "epoch": 0.1610958904109589, + "grad_norm": 3.005157709121704, + "learning_rate": 2.427007299270073e-06, + "loss": 2.2467, + "step": 147 + }, + { + "epoch": 0.16219178082191782, + "grad_norm": 3.4616103172302246, + "learning_rate": 2.445255474452555e-06, + "loss": 2.2612, + "step": 148 + }, + { + "epoch": 0.1632876712328767, + "grad_norm": 3.3595101833343506, + "learning_rate": 2.463503649635037e-06, + "loss": 2.3021, + "step": 149 + }, + { + "epoch": 0.1643835616438356, + "grad_norm": 4.831241607666016, + "learning_rate": 2.4817518248175183e-06, + "loss": 2.2174, + "step": 150 + }, + { + "epoch": 0.1643835616438356, + "eval_loss": 2.3677749633789062, + "eval_runtime": 37.3092, + "eval_samples_per_second": 26.803, + "eval_steps_per_second": 1.126, + "step": 150 + }, + { + "epoch": 0.16547945205479453, + "grad_norm": 2.778820037841797, + "learning_rate": 2.5e-06, + "loss": 2.1856, + "step": 151 + }, + { + "epoch": 0.16657534246575342, + "grad_norm": 4.566643714904785, + "learning_rate": 2.518248175182482e-06, + "loss": 2.3186, + "step": 152 + }, + { + "epoch": 0.16767123287671232, + "grad_norm": 2.9908089637756348, + "learning_rate": 2.5364963503649635e-06, + "loss": 2.2127, + "step": 153 + }, + { + "epoch": 0.16876712328767124, + "grad_norm": 2.9702792167663574, + "learning_rate": 2.5547445255474458e-06, + "loss": 2.2071, + "step": 154 + }, + { + "epoch": 0.16986301369863013, + "grad_norm": 2.850044012069702, + "learning_rate": 2.5729927007299272e-06, + "loss": 2.2585, + "step": 155 + }, + { + "epoch": 0.17095890410958905, + "grad_norm": 2.926091432571411, + "learning_rate": 2.591240875912409e-06, + "loss": 2.2397, + "step": 156 + }, + { + "epoch": 0.17205479452054795, + "grad_norm": 2.8720169067382812, + "learning_rate": 2.6094890510948905e-06, + "loss": 2.2306, + "step": 157 + }, + { + "epoch": 0.17315068493150684, + "grad_norm": 2.8837177753448486, + "learning_rate": 2.627737226277373e-06, + "loss": 2.2666, + "step": 158 + }, + { + "epoch": 0.17424657534246576, + "grad_norm": 3.119621753692627, + "learning_rate": 2.6459854014598542e-06, + "loss": 2.2722, + "step": 159 + }, + { + "epoch": 0.17534246575342466, + "grad_norm": 2.7085700035095215, + "learning_rate": 2.664233576642336e-06, + "loss": 2.2346, + "step": 160 + }, + { + "epoch": 0.17534246575342466, + "eval_loss": 2.352083921432495, + "eval_runtime": 37.245, + "eval_samples_per_second": 26.849, + "eval_steps_per_second": 1.128, + "step": 160 + }, + { + "epoch": 0.17643835616438355, + "grad_norm": 2.9964346885681152, + "learning_rate": 2.6824817518248175e-06, + "loss": 2.2155, + "step": 161 + }, + { + "epoch": 0.17753424657534247, + "grad_norm": 3.1535208225250244, + "learning_rate": 2.7007299270072994e-06, + "loss": 2.235, + "step": 162 + }, + { + "epoch": 0.17863013698630137, + "grad_norm": 2.7240943908691406, + "learning_rate": 2.718978102189781e-06, + "loss": 2.2095, + "step": 163 + }, + { + "epoch": 0.17972602739726026, + "grad_norm": 3.122654676437378, + "learning_rate": 2.737226277372263e-06, + "loss": 2.2667, + "step": 164 + }, + { + "epoch": 0.18082191780821918, + "grad_norm": 2.9718410968780518, + "learning_rate": 2.755474452554745e-06, + "loss": 2.2499, + "step": 165 + }, + { + "epoch": 0.18191780821917808, + "grad_norm": 3.0081896781921387, + "learning_rate": 2.7737226277372264e-06, + "loss": 2.2089, + "step": 166 + }, + { + "epoch": 0.18301369863013697, + "grad_norm": 2.829509973526001, + "learning_rate": 2.7919708029197082e-06, + "loss": 2.3139, + "step": 167 + }, + { + "epoch": 0.1841095890410959, + "grad_norm": 3.4480397701263428, + "learning_rate": 2.81021897810219e-06, + "loss": 2.2705, + "step": 168 + }, + { + "epoch": 0.1852054794520548, + "grad_norm": 3.1107025146484375, + "learning_rate": 2.828467153284672e-06, + "loss": 2.1895, + "step": 169 + }, + { + "epoch": 0.1863013698630137, + "grad_norm": 2.916790008544922, + "learning_rate": 2.8467153284671534e-06, + "loss": 2.2317, + "step": 170 + }, + { + "epoch": 0.1863013698630137, + "eval_loss": 2.336153030395508, + "eval_runtime": 37.2401, + "eval_samples_per_second": 26.853, + "eval_steps_per_second": 1.128, + "step": 170 + }, + { + "epoch": 0.1873972602739726, + "grad_norm": 3.2582287788391113, + "learning_rate": 2.8649635036496353e-06, + "loss": 2.143, + "step": 171 + }, + { + "epoch": 0.1884931506849315, + "grad_norm": 2.430973768234253, + "learning_rate": 2.8832116788321167e-06, + "loss": 2.1977, + "step": 172 + }, + { + "epoch": 0.18958904109589042, + "grad_norm": 2.8763599395751953, + "learning_rate": 2.901459854014599e-06, + "loss": 2.1551, + "step": 173 + }, + { + "epoch": 0.19068493150684931, + "grad_norm": 2.9242653846740723, + "learning_rate": 2.9197080291970804e-06, + "loss": 2.2569, + "step": 174 + }, + { + "epoch": 0.1917808219178082, + "grad_norm": 2.623551607131958, + "learning_rate": 2.9379562043795623e-06, + "loss": 2.2329, + "step": 175 + }, + { + "epoch": 0.19287671232876713, + "grad_norm": 2.9511337280273438, + "learning_rate": 2.956204379562044e-06, + "loss": 2.2154, + "step": 176 + }, + { + "epoch": 0.19397260273972602, + "grad_norm": 2.415118932723999, + "learning_rate": 2.9744525547445256e-06, + "loss": 2.2637, + "step": 177 + }, + { + "epoch": 0.19506849315068492, + "grad_norm": 2.9664924144744873, + "learning_rate": 2.992700729927008e-06, + "loss": 2.2263, + "step": 178 + }, + { + "epoch": 0.19616438356164384, + "grad_norm": 2.7214572429656982, + "learning_rate": 3.0109489051094893e-06, + "loss": 2.1667, + "step": 179 + }, + { + "epoch": 0.19726027397260273, + "grad_norm": 2.8712823390960693, + "learning_rate": 3.029197080291971e-06, + "loss": 2.1661, + "step": 180 + }, + { + "epoch": 0.19726027397260273, + "eval_loss": 2.3191800117492676, + "eval_runtime": 37.24, + "eval_samples_per_second": 26.853, + "eval_steps_per_second": 1.128, + "step": 180 + }, + { + "epoch": 0.19835616438356166, + "grad_norm": 2.959709405899048, + "learning_rate": 3.0474452554744526e-06, + "loss": 2.2483, + "step": 181 + }, + { + "epoch": 0.19945205479452055, + "grad_norm": 2.567089319229126, + "learning_rate": 3.065693430656935e-06, + "loss": 2.2382, + "step": 182 + }, + { + "epoch": 0.20054794520547944, + "grad_norm": 2.47188401222229, + "learning_rate": 3.0839416058394163e-06, + "loss": 2.2148, + "step": 183 + }, + { + "epoch": 0.20164383561643837, + "grad_norm": 2.3435261249542236, + "learning_rate": 3.102189781021898e-06, + "loss": 2.2334, + "step": 184 + }, + { + "epoch": 0.20273972602739726, + "grad_norm": 2.5280468463897705, + "learning_rate": 3.1204379562043796e-06, + "loss": 2.2214, + "step": 185 + }, + { + "epoch": 0.20383561643835615, + "grad_norm": 2.4808971881866455, + "learning_rate": 3.1386861313868614e-06, + "loss": 2.1554, + "step": 186 + }, + { + "epoch": 0.20493150684931508, + "grad_norm": 2.9937026500701904, + "learning_rate": 3.156934306569343e-06, + "loss": 2.1389, + "step": 187 + }, + { + "epoch": 0.20602739726027397, + "grad_norm": 2.9220476150512695, + "learning_rate": 3.175182481751825e-06, + "loss": 2.2515, + "step": 188 + }, + { + "epoch": 0.20712328767123286, + "grad_norm": 3.0449001789093018, + "learning_rate": 3.193430656934307e-06, + "loss": 2.2016, + "step": 189 + }, + { + "epoch": 0.20821917808219179, + "grad_norm": 2.8027074337005615, + "learning_rate": 3.2116788321167884e-06, + "loss": 2.1823, + "step": 190 + }, + { + "epoch": 0.20821917808219179, + "eval_loss": 2.309677839279175, + "eval_runtime": 37.2442, + "eval_samples_per_second": 26.85, + "eval_steps_per_second": 1.128, + "step": 190 + }, + { + "epoch": 0.20931506849315068, + "grad_norm": 2.9158318042755127, + "learning_rate": 3.2299270072992707e-06, + "loss": 2.1928, + "step": 191 + }, + { + "epoch": 0.2104109589041096, + "grad_norm": 2.5328080654144287, + "learning_rate": 3.248175182481752e-06, + "loss": 2.1828, + "step": 192 + }, + { + "epoch": 0.2115068493150685, + "grad_norm": 3.1709704399108887, + "learning_rate": 3.266423357664234e-06, + "loss": 2.1052, + "step": 193 + }, + { + "epoch": 0.2126027397260274, + "grad_norm": 2.9156367778778076, + "learning_rate": 3.2846715328467155e-06, + "loss": 2.1943, + "step": 194 + }, + { + "epoch": 0.2136986301369863, + "grad_norm": 2.879408836364746, + "learning_rate": 3.3029197080291973e-06, + "loss": 2.1782, + "step": 195 + }, + { + "epoch": 0.2147945205479452, + "grad_norm": 3.748483180999756, + "learning_rate": 3.3211678832116788e-06, + "loss": 2.2107, + "step": 196 + }, + { + "epoch": 0.2158904109589041, + "grad_norm": 3.197510004043579, + "learning_rate": 3.339416058394161e-06, + "loss": 2.1845, + "step": 197 + }, + { + "epoch": 0.21698630136986302, + "grad_norm": 3.2596049308776855, + "learning_rate": 3.3576642335766425e-06, + "loss": 2.1816, + "step": 198 + }, + { + "epoch": 0.21808219178082192, + "grad_norm": 3.3329780101776123, + "learning_rate": 3.3759124087591243e-06, + "loss": 2.181, + "step": 199 + }, + { + "epoch": 0.2191780821917808, + "grad_norm": 2.7302327156066895, + "learning_rate": 3.3941605839416058e-06, + "loss": 2.1761, + "step": 200 + }, + { + "epoch": 0.2191780821917808, + "eval_loss": 2.2909653186798096, + "eval_runtime": 37.284, + "eval_samples_per_second": 26.821, + "eval_steps_per_second": 1.126, + "step": 200 + }, + { + "epoch": 0.22027397260273973, + "grad_norm": 3.002255916595459, + "learning_rate": 3.412408759124088e-06, + "loss": 2.1868, + "step": 201 + }, + { + "epoch": 0.22136986301369863, + "grad_norm": 2.6694891452789307, + "learning_rate": 3.43065693430657e-06, + "loss": 2.192, + "step": 202 + }, + { + "epoch": 0.22246575342465755, + "grad_norm": 2.9225986003875732, + "learning_rate": 3.4489051094890513e-06, + "loss": 2.1303, + "step": 203 + }, + { + "epoch": 0.22356164383561644, + "grad_norm": 2.7844080924987793, + "learning_rate": 3.467153284671533e-06, + "loss": 2.1788, + "step": 204 + }, + { + "epoch": 0.22465753424657534, + "grad_norm": 2.726123332977295, + "learning_rate": 3.4854014598540146e-06, + "loss": 2.1193, + "step": 205 + }, + { + "epoch": 0.22575342465753426, + "grad_norm": 2.838040590286255, + "learning_rate": 3.503649635036497e-06, + "loss": 2.1674, + "step": 206 + }, + { + "epoch": 0.22684931506849315, + "grad_norm": 2.6863937377929688, + "learning_rate": 3.5218978102189783e-06, + "loss": 2.161, + "step": 207 + }, + { + "epoch": 0.22794520547945205, + "grad_norm": 2.5116159915924072, + "learning_rate": 3.54014598540146e-06, + "loss": 2.179, + "step": 208 + }, + { + "epoch": 0.22904109589041097, + "grad_norm": 3.381863594055176, + "learning_rate": 3.5583941605839416e-06, + "loss": 2.2035, + "step": 209 + }, + { + "epoch": 0.23013698630136986, + "grad_norm": 3.2741382122039795, + "learning_rate": 3.576642335766424e-06, + "loss": 2.2518, + "step": 210 + }, + { + "epoch": 0.23013698630136986, + "eval_loss": 2.2773635387420654, + "eval_runtime": 37.2658, + "eval_samples_per_second": 26.834, + "eval_steps_per_second": 1.127, + "step": 210 + }, + { + "epoch": 0.23123287671232876, + "grad_norm": 2.917370319366455, + "learning_rate": 3.5948905109489054e-06, + "loss": 2.2265, + "step": 211 + }, + { + "epoch": 0.23232876712328768, + "grad_norm": 2.8392834663391113, + "learning_rate": 3.6131386861313872e-06, + "loss": 2.0709, + "step": 212 + }, + { + "epoch": 0.23342465753424657, + "grad_norm": 2.753612518310547, + "learning_rate": 3.6313868613138687e-06, + "loss": 2.0752, + "step": 213 + }, + { + "epoch": 0.2345205479452055, + "grad_norm": 3.1477746963500977, + "learning_rate": 3.6496350364963505e-06, + "loss": 2.1169, + "step": 214 + }, + { + "epoch": 0.2356164383561644, + "grad_norm": 2.5448100566864014, + "learning_rate": 3.667883211678833e-06, + "loss": 2.1338, + "step": 215 + }, + { + "epoch": 0.23671232876712328, + "grad_norm": 3.1622560024261475, + "learning_rate": 3.6861313868613142e-06, + "loss": 2.1065, + "step": 216 + }, + { + "epoch": 0.2378082191780822, + "grad_norm": 2.7054336071014404, + "learning_rate": 3.704379562043796e-06, + "loss": 2.1104, + "step": 217 + }, + { + "epoch": 0.2389041095890411, + "grad_norm": 2.964141845703125, + "learning_rate": 3.7226277372262775e-06, + "loss": 2.1949, + "step": 218 + }, + { + "epoch": 0.24, + "grad_norm": 2.7586112022399902, + "learning_rate": 3.7408759124087594e-06, + "loss": 2.165, + "step": 219 + }, + { + "epoch": 0.2410958904109589, + "grad_norm": 2.9035964012145996, + "learning_rate": 3.7591240875912412e-06, + "loss": 2.1192, + "step": 220 + }, + { + "epoch": 0.2410958904109589, + "eval_loss": 2.2649471759796143, + "eval_runtime": 37.4311, + "eval_samples_per_second": 26.716, + "eval_steps_per_second": 1.122, + "step": 220 + }, + { + "epoch": 0.2421917808219178, + "grad_norm": 2.4212076663970947, + "learning_rate": 3.777372262773723e-06, + "loss": 2.1389, + "step": 221 + }, + { + "epoch": 0.2432876712328767, + "grad_norm": 2.973227024078369, + "learning_rate": 3.7956204379562045e-06, + "loss": 2.132, + "step": 222 + }, + { + "epoch": 0.24438356164383562, + "grad_norm": 2.9831430912017822, + "learning_rate": 3.8138686131386864e-06, + "loss": 2.1142, + "step": 223 + }, + { + "epoch": 0.24547945205479452, + "grad_norm": 3.2880353927612305, + "learning_rate": 3.832116788321168e-06, + "loss": 2.168, + "step": 224 + }, + { + "epoch": 0.2465753424657534, + "grad_norm": 3.3920998573303223, + "learning_rate": 3.85036496350365e-06, + "loss": 2.1277, + "step": 225 + }, + { + "epoch": 0.24767123287671233, + "grad_norm": 2.7170026302337646, + "learning_rate": 3.868613138686132e-06, + "loss": 2.0936, + "step": 226 + }, + { + "epoch": 0.24876712328767123, + "grad_norm": 3.6220011711120605, + "learning_rate": 3.886861313868613e-06, + "loss": 2.1259, + "step": 227 + }, + { + "epoch": 0.24986301369863015, + "grad_norm": 3.4229273796081543, + "learning_rate": 3.905109489051096e-06, + "loss": 2.1421, + "step": 228 + }, + { + "epoch": 0.25095890410958904, + "grad_norm": 3.0358338356018066, + "learning_rate": 3.923357664233577e-06, + "loss": 2.1747, + "step": 229 + }, + { + "epoch": 0.25205479452054796, + "grad_norm": 2.611771821975708, + "learning_rate": 3.9416058394160585e-06, + "loss": 2.1282, + "step": 230 + }, + { + "epoch": 0.25205479452054796, + "eval_loss": 2.254887580871582, + "eval_runtime": 37.5943, + "eval_samples_per_second": 26.6, + "eval_steps_per_second": 1.117, + "step": 230 + }, + { + "epoch": 0.25315068493150683, + "grad_norm": 3.846412181854248, + "learning_rate": 3.95985401459854e-06, + "loss": 2.1892, + "step": 231 + }, + { + "epoch": 0.25424657534246575, + "grad_norm": 2.488525390625, + "learning_rate": 3.978102189781022e-06, + "loss": 2.2064, + "step": 232 + }, + { + "epoch": 0.2553424657534247, + "grad_norm": 3.114964485168457, + "learning_rate": 3.996350364963504e-06, + "loss": 2.0653, + "step": 233 + }, + { + "epoch": 0.25643835616438354, + "grad_norm": 2.5559768676757812, + "learning_rate": 4.014598540145986e-06, + "loss": 2.1503, + "step": 234 + }, + { + "epoch": 0.25753424657534246, + "grad_norm": 3.2347540855407715, + "learning_rate": 4.032846715328467e-06, + "loss": 2.0523, + "step": 235 + }, + { + "epoch": 0.2586301369863014, + "grad_norm": 2.8086111545562744, + "learning_rate": 4.05109489051095e-06, + "loss": 2.1206, + "step": 236 + }, + { + "epoch": 0.25972602739726025, + "grad_norm": 2.909231185913086, + "learning_rate": 4.069343065693431e-06, + "loss": 2.139, + "step": 237 + }, + { + "epoch": 0.2608219178082192, + "grad_norm": 3.2171082496643066, + "learning_rate": 4.0875912408759126e-06, + "loss": 2.0908, + "step": 238 + }, + { + "epoch": 0.2619178082191781, + "grad_norm": 2.3851265907287598, + "learning_rate": 4.105839416058395e-06, + "loss": 2.1133, + "step": 239 + }, + { + "epoch": 0.26301369863013696, + "grad_norm": 3.3912737369537354, + "learning_rate": 4.124087591240876e-06, + "loss": 2.1911, + "step": 240 + }, + { + "epoch": 0.26301369863013696, + "eval_loss": 2.2436957359313965, + "eval_runtime": 37.2947, + "eval_samples_per_second": 26.813, + "eval_steps_per_second": 1.126, + "step": 240 + }, + { + "epoch": 0.2641095890410959, + "grad_norm": 2.82906174659729, + "learning_rate": 4.1423357664233586e-06, + "loss": 2.11, + "step": 241 + }, + { + "epoch": 0.2652054794520548, + "grad_norm": 3.0317952632904053, + "learning_rate": 4.16058394160584e-06, + "loss": 2.0953, + "step": 242 + }, + { + "epoch": 0.2663013698630137, + "grad_norm": 3.6438329219818115, + "learning_rate": 4.1788321167883214e-06, + "loss": 2.1746, + "step": 243 + }, + { + "epoch": 0.2673972602739726, + "grad_norm": 2.6690592765808105, + "learning_rate": 4.197080291970803e-06, + "loss": 2.0555, + "step": 244 + }, + { + "epoch": 0.2684931506849315, + "grad_norm": 3.135376453399658, + "learning_rate": 4.215328467153285e-06, + "loss": 2.1394, + "step": 245 + }, + { + "epoch": 0.26958904109589044, + "grad_norm": 2.5474987030029297, + "learning_rate": 4.233576642335767e-06, + "loss": 2.0822, + "step": 246 + }, + { + "epoch": 0.2706849315068493, + "grad_norm": 2.883108377456665, + "learning_rate": 4.251824817518249e-06, + "loss": 2.0869, + "step": 247 + }, + { + "epoch": 0.2717808219178082, + "grad_norm": 2.9801037311553955, + "learning_rate": 4.27007299270073e-06, + "loss": 2.1194, + "step": 248 + }, + { + "epoch": 0.27287671232876715, + "grad_norm": 2.7688217163085938, + "learning_rate": 4.288321167883212e-06, + "loss": 2.0908, + "step": 249 + }, + { + "epoch": 0.273972602739726, + "grad_norm": 3.205148458480835, + "learning_rate": 4.306569343065693e-06, + "loss": 2.0762, + "step": 250 + }, + { + "epoch": 0.273972602739726, + "eval_loss": 2.238699197769165, + "eval_runtime": 37.2428, + "eval_samples_per_second": 26.851, + "eval_steps_per_second": 1.128, + "step": 250 + }, + { + "epoch": 0.27506849315068493, + "grad_norm": 2.8262946605682373, + "learning_rate": 4.3248175182481755e-06, + "loss": 2.0509, + "step": 251 + }, + { + "epoch": 0.27616438356164386, + "grad_norm": 3.156261682510376, + "learning_rate": 4.343065693430658e-06, + "loss": 2.1213, + "step": 252 + }, + { + "epoch": 0.2772602739726027, + "grad_norm": 2.6367998123168945, + "learning_rate": 4.361313868613139e-06, + "loss": 2.1071, + "step": 253 + }, + { + "epoch": 0.27835616438356164, + "grad_norm": 2.871534824371338, + "learning_rate": 4.379562043795621e-06, + "loss": 2.1008, + "step": 254 + }, + { + "epoch": 0.27945205479452057, + "grad_norm": 2.839432716369629, + "learning_rate": 4.397810218978102e-06, + "loss": 2.1293, + "step": 255 + }, + { + "epoch": 0.28054794520547943, + "grad_norm": 2.627143383026123, + "learning_rate": 4.416058394160584e-06, + "loss": 2.1141, + "step": 256 + }, + { + "epoch": 0.28164383561643835, + "grad_norm": 2.8292996883392334, + "learning_rate": 4.434306569343066e-06, + "loss": 2.0981, + "step": 257 + }, + { + "epoch": 0.2827397260273973, + "grad_norm": 2.5711069107055664, + "learning_rate": 4.452554744525548e-06, + "loss": 2.0707, + "step": 258 + }, + { + "epoch": 0.28383561643835614, + "grad_norm": 2.488954544067383, + "learning_rate": 4.4708029197080295e-06, + "loss": 2.1172, + "step": 259 + }, + { + "epoch": 0.28493150684931506, + "grad_norm": 2.442209005355835, + "learning_rate": 4.489051094890512e-06, + "loss": 2.1119, + "step": 260 + }, + { + "epoch": 0.28493150684931506, + "eval_loss": 2.2306911945343018, + "eval_runtime": 37.2412, + "eval_samples_per_second": 26.852, + "eval_steps_per_second": 1.128, + "step": 260 + }, + { + "epoch": 0.286027397260274, + "grad_norm": 2.9147934913635254, + "learning_rate": 4.507299270072993e-06, + "loss": 2.1517, + "step": 261 + }, + { + "epoch": 0.28712328767123285, + "grad_norm": 2.549833297729492, + "learning_rate": 4.525547445255475e-06, + "loss": 2.1047, + "step": 262 + }, + { + "epoch": 0.2882191780821918, + "grad_norm": 3.3405206203460693, + "learning_rate": 4.543795620437956e-06, + "loss": 2.0975, + "step": 263 + }, + { + "epoch": 0.2893150684931507, + "grad_norm": 2.554431676864624, + "learning_rate": 4.562043795620438e-06, + "loss": 2.0622, + "step": 264 + }, + { + "epoch": 0.29041095890410956, + "grad_norm": 3.5529873371124268, + "learning_rate": 4.580291970802921e-06, + "loss": 2.1567, + "step": 265 + }, + { + "epoch": 0.2915068493150685, + "grad_norm": 2.421682119369507, + "learning_rate": 4.598540145985402e-06, + "loss": 2.1619, + "step": 266 + }, + { + "epoch": 0.2926027397260274, + "grad_norm": 2.918288230895996, + "learning_rate": 4.6167883211678835e-06, + "loss": 2.1238, + "step": 267 + }, + { + "epoch": 0.2936986301369863, + "grad_norm": 3.0737621784210205, + "learning_rate": 4.635036496350365e-06, + "loss": 2.0728, + "step": 268 + }, + { + "epoch": 0.2947945205479452, + "grad_norm": 2.5862410068511963, + "learning_rate": 4.653284671532847e-06, + "loss": 2.0675, + "step": 269 + }, + { + "epoch": 0.2958904109589041, + "grad_norm": 2.9004979133605957, + "learning_rate": 4.671532846715329e-06, + "loss": 2.1248, + "step": 270 + }, + { + "epoch": 0.2958904109589041, + "eval_loss": 2.2179365158081055, + "eval_runtime": 37.2352, + "eval_samples_per_second": 26.856, + "eval_steps_per_second": 1.128, + "step": 270 + }, + { + "epoch": 0.29698630136986304, + "grad_norm": 2.4784982204437256, + "learning_rate": 4.689781021897811e-06, + "loss": 2.1276, + "step": 271 + }, + { + "epoch": 0.2980821917808219, + "grad_norm": 2.7971739768981934, + "learning_rate": 4.708029197080292e-06, + "loss": 2.0931, + "step": 272 + }, + { + "epoch": 0.2991780821917808, + "grad_norm": 2.5839481353759766, + "learning_rate": 4.726277372262774e-06, + "loss": 2.1371, + "step": 273 + }, + { + "epoch": 0.30027397260273975, + "grad_norm": 2.7168564796447754, + "learning_rate": 4.744525547445255e-06, + "loss": 2.1086, + "step": 274 + }, + { + "epoch": 0.3013698630136986, + "grad_norm": 2.6566879749298096, + "learning_rate": 4.7627737226277375e-06, + "loss": 2.0368, + "step": 275 + }, + { + "epoch": 0.30246575342465754, + "grad_norm": 2.5068323612213135, + "learning_rate": 4.78102189781022e-06, + "loss": 2.1077, + "step": 276 + }, + { + "epoch": 0.30356164383561646, + "grad_norm": 2.631408214569092, + "learning_rate": 4.799270072992701e-06, + "loss": 2.1013, + "step": 277 + }, + { + "epoch": 0.3046575342465753, + "grad_norm": 2.5642812252044678, + "learning_rate": 4.8175182481751835e-06, + "loss": 2.0739, + "step": 278 + }, + { + "epoch": 0.30575342465753425, + "grad_norm": 2.3326497077941895, + "learning_rate": 4.835766423357665e-06, + "loss": 2.0971, + "step": 279 + }, + { + "epoch": 0.30684931506849317, + "grad_norm": 2.70701265335083, + "learning_rate": 4.854014598540146e-06, + "loss": 2.0904, + "step": 280 + }, + { + "epoch": 0.30684931506849317, + "eval_loss": 2.215972900390625, + "eval_runtime": 37.2196, + "eval_samples_per_second": 26.868, + "eval_steps_per_second": 1.128, + "step": 280 + }, + { + "epoch": 0.30794520547945203, + "grad_norm": 2.615201234817505, + "learning_rate": 4.872262773722628e-06, + "loss": 2.039, + "step": 281 + }, + { + "epoch": 0.30904109589041096, + "grad_norm": 2.2821998596191406, + "learning_rate": 4.89051094890511e-06, + "loss": 2.0529, + "step": 282 + }, + { + "epoch": 0.3101369863013699, + "grad_norm": 2.767153024673462, + "learning_rate": 4.9087591240875915e-06, + "loss": 2.0351, + "step": 283 + }, + { + "epoch": 0.31123287671232874, + "grad_norm": 2.537121295928955, + "learning_rate": 4.927007299270074e-06, + "loss": 2.0752, + "step": 284 + }, + { + "epoch": 0.31232876712328766, + "grad_norm": 2.835052013397217, + "learning_rate": 4.945255474452555e-06, + "loss": 2.0892, + "step": 285 + }, + { + "epoch": 0.3134246575342466, + "grad_norm": 2.5154924392700195, + "learning_rate": 4.963503649635037e-06, + "loss": 2.0569, + "step": 286 + }, + { + "epoch": 0.31452054794520545, + "grad_norm": 2.8212568759918213, + "learning_rate": 4.981751824817518e-06, + "loss": 2.1588, + "step": 287 + }, + { + "epoch": 0.3156164383561644, + "grad_norm": 2.3840994834899902, + "learning_rate": 5e-06, + "loss": 2.1409, + "step": 288 + }, + { + "epoch": 0.3167123287671233, + "grad_norm": 2.9641315937042236, + "learning_rate": 4.997969130787978e-06, + "loss": 2.0892, + "step": 289 + }, + { + "epoch": 0.3178082191780822, + "grad_norm": 2.4516749382019043, + "learning_rate": 4.995938261575955e-06, + "loss": 2.1105, + "step": 290 + }, + { + "epoch": 0.3178082191780822, + "eval_loss": 2.207075834274292, + "eval_runtime": 37.1304, + "eval_samples_per_second": 26.932, + "eval_steps_per_second": 1.131, + "step": 290 + }, + { + "epoch": 0.3189041095890411, + "grad_norm": 2.768869400024414, + "learning_rate": 4.993907392363932e-06, + "loss": 2.0231, + "step": 291 + }, + { + "epoch": 0.32, + "grad_norm": 2.771475315093994, + "learning_rate": 4.9918765231519095e-06, + "loss": 2.1018, + "step": 292 + }, + { + "epoch": 0.32109589041095893, + "grad_norm": 2.773983955383301, + "learning_rate": 4.989845653939887e-06, + "loss": 2.0317, + "step": 293 + }, + { + "epoch": 0.3221917808219178, + "grad_norm": 2.4447343349456787, + "learning_rate": 4.987814784727864e-06, + "loss": 2.0739, + "step": 294 + }, + { + "epoch": 0.3232876712328767, + "grad_norm": 2.444261312484741, + "learning_rate": 4.985783915515841e-06, + "loss": 2.073, + "step": 295 + }, + { + "epoch": 0.32438356164383564, + "grad_norm": 2.871792793273926, + "learning_rate": 4.983753046303819e-06, + "loss": 2.058, + "step": 296 + }, + { + "epoch": 0.3254794520547945, + "grad_norm": 2.3651537895202637, + "learning_rate": 4.981722177091796e-06, + "loss": 2.0749, + "step": 297 + }, + { + "epoch": 0.3265753424657534, + "grad_norm": 2.3395845890045166, + "learning_rate": 4.979691307879773e-06, + "loss": 2.065, + "step": 298 + }, + { + "epoch": 0.32767123287671235, + "grad_norm": 2.6275217533111572, + "learning_rate": 4.9776604386677504e-06, + "loss": 2.0945, + "step": 299 + }, + { + "epoch": 0.3287671232876712, + "grad_norm": 2.866285562515259, + "learning_rate": 4.975629569455728e-06, + "loss": 2.0702, + "step": 300 + }, + { + "epoch": 0.3287671232876712, + "eval_loss": 2.200012683868408, + "eval_runtime": 37.2082, + "eval_samples_per_second": 26.876, + "eval_steps_per_second": 1.129, + "step": 300 + }, + { + "epoch": 0.32986301369863014, + "grad_norm": 2.268401861190796, + "learning_rate": 4.973598700243705e-06, + "loss": 2.1257, + "step": 301 + }, + { + "epoch": 0.33095890410958906, + "grad_norm": 2.6197733879089355, + "learning_rate": 4.971567831031681e-06, + "loss": 2.0347, + "step": 302 + }, + { + "epoch": 0.3320547945205479, + "grad_norm": 2.4738848209381104, + "learning_rate": 4.9695369618196595e-06, + "loss": 2.0368, + "step": 303 + }, + { + "epoch": 0.33315068493150685, + "grad_norm": 2.705670118331909, + "learning_rate": 4.967506092607636e-06, + "loss": 1.9256, + "step": 304 + }, + { + "epoch": 0.33424657534246577, + "grad_norm": 2.813279628753662, + "learning_rate": 4.965475223395614e-06, + "loss": 2.0214, + "step": 305 + }, + { + "epoch": 0.33534246575342463, + "grad_norm": 2.320960283279419, + "learning_rate": 4.9634443541835905e-06, + "loss": 2.0342, + "step": 306 + }, + { + "epoch": 0.33643835616438356, + "grad_norm": 2.4495227336883545, + "learning_rate": 4.961413484971569e-06, + "loss": 2.0421, + "step": 307 + }, + { + "epoch": 0.3375342465753425, + "grad_norm": 2.5384018421173096, + "learning_rate": 4.959382615759545e-06, + "loss": 2.0572, + "step": 308 + }, + { + "epoch": 0.33863013698630134, + "grad_norm": 2.4018006324768066, + "learning_rate": 4.957351746547523e-06, + "loss": 2.0617, + "step": 309 + }, + { + "epoch": 0.33972602739726027, + "grad_norm": 2.552424907684326, + "learning_rate": 4.9553208773355e-06, + "loss": 2.0384, + "step": 310 + }, + { + "epoch": 0.33972602739726027, + "eval_loss": 2.1974849700927734, + "eval_runtime": 37.2308, + "eval_samples_per_second": 26.859, + "eval_steps_per_second": 1.128, + "step": 310 + }, + { + "epoch": 0.3408219178082192, + "grad_norm": 2.4587671756744385, + "learning_rate": 4.953290008123478e-06, + "loss": 2.1701, + "step": 311 + }, + { + "epoch": 0.3419178082191781, + "grad_norm": 2.4486496448516846, + "learning_rate": 4.951259138911454e-06, + "loss": 2.0654, + "step": 312 + }, + { + "epoch": 0.343013698630137, + "grad_norm": 2.123568058013916, + "learning_rate": 4.9492282696994315e-06, + "loss": 1.9961, + "step": 313 + }, + { + "epoch": 0.3441095890410959, + "grad_norm": 2.549788475036621, + "learning_rate": 4.947197400487409e-06, + "loss": 2.1265, + "step": 314 + }, + { + "epoch": 0.3452054794520548, + "grad_norm": 2.3453609943389893, + "learning_rate": 4.945166531275386e-06, + "loss": 2.0501, + "step": 315 + }, + { + "epoch": 0.3463013698630137, + "grad_norm": 2.3584256172180176, + "learning_rate": 4.943135662063363e-06, + "loss": 2.0402, + "step": 316 + }, + { + "epoch": 0.3473972602739726, + "grad_norm": 3.3100943565368652, + "learning_rate": 4.9411047928513406e-06, + "loss": 2.0545, + "step": 317 + }, + { + "epoch": 0.34849315068493153, + "grad_norm": 2.4266037940979004, + "learning_rate": 4.939073923639318e-06, + "loss": 2.0926, + "step": 318 + }, + { + "epoch": 0.3495890410958904, + "grad_norm": 3.101090908050537, + "learning_rate": 4.937043054427295e-06, + "loss": 2.0987, + "step": 319 + }, + { + "epoch": 0.3506849315068493, + "grad_norm": 2.743619203567505, + "learning_rate": 4.935012185215272e-06, + "loss": 2.0315, + "step": 320 + }, + { + "epoch": 0.3506849315068493, + "eval_loss": 2.188877820968628, + "eval_runtime": 37.4609, + "eval_samples_per_second": 26.694, + "eval_steps_per_second": 1.121, + "step": 320 + }, + { + "epoch": 0.35178082191780824, + "grad_norm": 2.6065986156463623, + "learning_rate": 4.93298131600325e-06, + "loss": 2.0166, + "step": 321 + }, + { + "epoch": 0.3528767123287671, + "grad_norm": 2.7554283142089844, + "learning_rate": 4.930950446791227e-06, + "loss": 2.0862, + "step": 322 + }, + { + "epoch": 0.353972602739726, + "grad_norm": 2.4773144721984863, + "learning_rate": 4.928919577579204e-06, + "loss": 2.0472, + "step": 323 + }, + { + "epoch": 0.35506849315068495, + "grad_norm": 2.695289134979248, + "learning_rate": 4.9268887083671815e-06, + "loss": 2.0567, + "step": 324 + }, + { + "epoch": 0.3561643835616438, + "grad_norm": 2.408220052719116, + "learning_rate": 4.924857839155159e-06, + "loss": 2.046, + "step": 325 + }, + { + "epoch": 0.35726027397260274, + "grad_norm": 2.3851559162139893, + "learning_rate": 4.922826969943136e-06, + "loss": 2.054, + "step": 326 + }, + { + "epoch": 0.35835616438356166, + "grad_norm": 2.8667280673980713, + "learning_rate": 4.920796100731113e-06, + "loss": 2.0717, + "step": 327 + }, + { + "epoch": 0.3594520547945205, + "grad_norm": 2.3814942836761475, + "learning_rate": 4.918765231519091e-06, + "loss": 1.9884, + "step": 328 + }, + { + "epoch": 0.36054794520547945, + "grad_norm": 2.341076612472534, + "learning_rate": 4.916734362307068e-06, + "loss": 2.0474, + "step": 329 + }, + { + "epoch": 0.36164383561643837, + "grad_norm": 2.3571853637695312, + "learning_rate": 4.914703493095045e-06, + "loss": 2.0009, + "step": 330 + }, + { + "epoch": 0.36164383561643837, + "eval_loss": 2.186048746109009, + "eval_runtime": 37.5746, + "eval_samples_per_second": 26.614, + "eval_steps_per_second": 1.118, + "step": 330 + }, + { + "epoch": 0.36273972602739724, + "grad_norm": 2.3770196437835693, + "learning_rate": 4.912672623883022e-06, + "loss": 2.0167, + "step": 331 + }, + { + "epoch": 0.36383561643835616, + "grad_norm": 2.3357701301574707, + "learning_rate": 4.910641754671e-06, + "loss": 1.9972, + "step": 332 + }, + { + "epoch": 0.3649315068493151, + "grad_norm": 2.4100499153137207, + "learning_rate": 4.908610885458977e-06, + "loss": 2.0319, + "step": 333 + }, + { + "epoch": 0.36602739726027395, + "grad_norm": 2.193018913269043, + "learning_rate": 4.906580016246954e-06, + "loss": 2.0285, + "step": 334 + }, + { + "epoch": 0.36712328767123287, + "grad_norm": 2.275637149810791, + "learning_rate": 4.9045491470349315e-06, + "loss": 2.0749, + "step": 335 + }, + { + "epoch": 0.3682191780821918, + "grad_norm": 2.347121238708496, + "learning_rate": 4.902518277822909e-06, + "loss": 2.0502, + "step": 336 + }, + { + "epoch": 0.3693150684931507, + "grad_norm": 2.210817575454712, + "learning_rate": 4.900487408610886e-06, + "loss": 2.0612, + "step": 337 + }, + { + "epoch": 0.3704109589041096, + "grad_norm": 2.15594482421875, + "learning_rate": 4.898456539398863e-06, + "loss": 2.0653, + "step": 338 + }, + { + "epoch": 0.3715068493150685, + "grad_norm": 2.3996052742004395, + "learning_rate": 4.89642567018684e-06, + "loss": 2.074, + "step": 339 + }, + { + "epoch": 0.3726027397260274, + "grad_norm": 2.384172201156616, + "learning_rate": 4.894394800974818e-06, + "loss": 2.1533, + "step": 340 + }, + { + "epoch": 0.3726027397260274, + "eval_loss": 2.175034284591675, + "eval_runtime": 37.2455, + "eval_samples_per_second": 26.849, + "eval_steps_per_second": 1.128, + "step": 340 + }, + { + "epoch": 0.3736986301369863, + "grad_norm": 2.3431336879730225, + "learning_rate": 4.892363931762794e-06, + "loss": 1.9919, + "step": 341 + }, + { + "epoch": 0.3747945205479452, + "grad_norm": 2.1828389167785645, + "learning_rate": 4.8903330625507725e-06, + "loss": 2.0703, + "step": 342 + }, + { + "epoch": 0.37589041095890413, + "grad_norm": 2.2117249965667725, + "learning_rate": 4.888302193338749e-06, + "loss": 2.0071, + "step": 343 + }, + { + "epoch": 0.376986301369863, + "grad_norm": 2.3595967292785645, + "learning_rate": 4.886271324126727e-06, + "loss": 1.9879, + "step": 344 + }, + { + "epoch": 0.3780821917808219, + "grad_norm": 2.5211799144744873, + "learning_rate": 4.8842404549147034e-06, + "loss": 2.0861, + "step": 345 + }, + { + "epoch": 0.37917808219178084, + "grad_norm": 2.1551895141601562, + "learning_rate": 4.8822095857026816e-06, + "loss": 2.0683, + "step": 346 + }, + { + "epoch": 0.3802739726027397, + "grad_norm": 2.4145829677581787, + "learning_rate": 4.880178716490658e-06, + "loss": 2.043, + "step": 347 + }, + { + "epoch": 0.38136986301369863, + "grad_norm": 2.4273521900177, + "learning_rate": 4.878147847278636e-06, + "loss": 1.9928, + "step": 348 + }, + { + "epoch": 0.38246575342465755, + "grad_norm": 2.261054754257202, + "learning_rate": 4.8761169780666125e-06, + "loss": 2.0065, + "step": 349 + }, + { + "epoch": 0.3835616438356164, + "grad_norm": 2.4703147411346436, + "learning_rate": 4.874086108854591e-06, + "loss": 1.9766, + "step": 350 + }, + { + "epoch": 0.3835616438356164, + "eval_loss": 2.176915168762207, + "eval_runtime": 37.3223, + "eval_samples_per_second": 26.794, + "eval_steps_per_second": 1.125, + "step": 350 + }, + { + "epoch": 0.38465753424657534, + "grad_norm": 2.3353374004364014, + "learning_rate": 4.872055239642567e-06, + "loss": 1.9874, + "step": 351 + }, + { + "epoch": 0.38575342465753426, + "grad_norm": 2.227834463119507, + "learning_rate": 4.870024370430544e-06, + "loss": 1.9644, + "step": 352 + }, + { + "epoch": 0.3868493150684931, + "grad_norm": 2.3155159950256348, + "learning_rate": 4.867993501218522e-06, + "loss": 2.0227, + "step": 353 + }, + { + "epoch": 0.38794520547945205, + "grad_norm": 2.2456047534942627, + "learning_rate": 4.865962632006499e-06, + "loss": 2.0566, + "step": 354 + }, + { + "epoch": 0.38904109589041097, + "grad_norm": 2.3364651203155518, + "learning_rate": 4.863931762794476e-06, + "loss": 1.9941, + "step": 355 + }, + { + "epoch": 0.39013698630136984, + "grad_norm": 2.2886059284210205, + "learning_rate": 4.8619008935824535e-06, + "loss": 2.0163, + "step": 356 + }, + { + "epoch": 0.39123287671232876, + "grad_norm": 2.3796911239624023, + "learning_rate": 4.859870024370431e-06, + "loss": 2.0525, + "step": 357 + }, + { + "epoch": 0.3923287671232877, + "grad_norm": 2.381986379623413, + "learning_rate": 4.857839155158408e-06, + "loss": 2.0682, + "step": 358 + }, + { + "epoch": 0.3934246575342466, + "grad_norm": 2.2048425674438477, + "learning_rate": 4.855808285946385e-06, + "loss": 2.0748, + "step": 359 + }, + { + "epoch": 0.39452054794520547, + "grad_norm": 2.162846088409424, + "learning_rate": 4.8537774167343626e-06, + "loss": 2.0512, + "step": 360 + }, + { + "epoch": 0.39452054794520547, + "eval_loss": 2.1645994186401367, + "eval_runtime": 37.3126, + "eval_samples_per_second": 26.801, + "eval_steps_per_second": 1.126, + "step": 360 + }, + { + "epoch": 0.3956164383561644, + "grad_norm": 2.4070873260498047, + "learning_rate": 4.85174654752234e-06, + "loss": 2.042, + "step": 361 + }, + { + "epoch": 0.3967123287671233, + "grad_norm": 2.11716628074646, + "learning_rate": 4.849715678310317e-06, + "loss": 2.0398, + "step": 362 + }, + { + "epoch": 0.3978082191780822, + "grad_norm": 2.265479803085327, + "learning_rate": 4.847684809098294e-06, + "loss": 2.0334, + "step": 363 + }, + { + "epoch": 0.3989041095890411, + "grad_norm": 2.2270994186401367, + "learning_rate": 4.845653939886272e-06, + "loss": 2.0292, + "step": 364 + }, + { + "epoch": 0.4, + "grad_norm": 2.406064033508301, + "learning_rate": 4.843623070674249e-06, + "loss": 2.0351, + "step": 365 + }, + { + "epoch": 0.4010958904109589, + "grad_norm": 2.216654062271118, + "learning_rate": 4.841592201462226e-06, + "loss": 2.005, + "step": 366 + }, + { + "epoch": 0.4021917808219178, + "grad_norm": 2.4241385459899902, + "learning_rate": 4.8395613322502035e-06, + "loss": 2.0647, + "step": 367 + }, + { + "epoch": 0.40328767123287673, + "grad_norm": 2.4315145015716553, + "learning_rate": 4.837530463038181e-06, + "loss": 1.9679, + "step": 368 + }, + { + "epoch": 0.4043835616438356, + "grad_norm": 2.3034439086914062, + "learning_rate": 4.835499593826158e-06, + "loss": 2.0259, + "step": 369 + }, + { + "epoch": 0.4054794520547945, + "grad_norm": 2.493436098098755, + "learning_rate": 4.833468724614135e-06, + "loss": 2.0468, + "step": 370 + }, + { + "epoch": 0.4054794520547945, + "eval_loss": 2.161508083343506, + "eval_runtime": 37.2692, + "eval_samples_per_second": 26.832, + "eval_steps_per_second": 1.127, + "step": 370 + }, + { + "epoch": 0.40657534246575344, + "grad_norm": 2.3887157440185547, + "learning_rate": 4.831437855402113e-06, + "loss": 2.0313, + "step": 371 + }, + { + "epoch": 0.4076712328767123, + "grad_norm": 2.563095808029175, + "learning_rate": 4.82940698619009e-06, + "loss": 1.9288, + "step": 372 + }, + { + "epoch": 0.40876712328767123, + "grad_norm": 2.2609968185424805, + "learning_rate": 4.827376116978067e-06, + "loss": 1.9975, + "step": 373 + }, + { + "epoch": 0.40986301369863015, + "grad_norm": 2.624257802963257, + "learning_rate": 4.8253452477660444e-06, + "loss": 2.0639, + "step": 374 + }, + { + "epoch": 0.410958904109589, + "grad_norm": 2.502197265625, + "learning_rate": 4.823314378554022e-06, + "loss": 2.0424, + "step": 375 + }, + { + "epoch": 0.41205479452054794, + "grad_norm": 2.36297345161438, + "learning_rate": 4.821283509341998e-06, + "loss": 2.027, + "step": 376 + }, + { + "epoch": 0.41315068493150686, + "grad_norm": 2.388918161392212, + "learning_rate": 4.819252640129976e-06, + "loss": 2.0002, + "step": 377 + }, + { + "epoch": 0.41424657534246573, + "grad_norm": 2.4262335300445557, + "learning_rate": 4.817221770917953e-06, + "loss": 2.0215, + "step": 378 + }, + { + "epoch": 0.41534246575342465, + "grad_norm": 2.8953094482421875, + "learning_rate": 4.815190901705931e-06, + "loss": 1.988, + "step": 379 + }, + { + "epoch": 0.41643835616438357, + "grad_norm": 2.3092124462127686, + "learning_rate": 4.813160032493907e-06, + "loss": 2.0199, + "step": 380 + }, + { + "epoch": 0.41643835616438357, + "eval_loss": 2.1589064598083496, + "eval_runtime": 37.2649, + "eval_samples_per_second": 26.835, + "eval_steps_per_second": 1.127, + "step": 380 + }, + { + "epoch": 0.41753424657534244, + "grad_norm": 3.0842413902282715, + "learning_rate": 4.811129163281885e-06, + "loss": 2.0134, + "step": 381 + }, + { + "epoch": 0.41863013698630136, + "grad_norm": 2.2015092372894287, + "learning_rate": 4.809098294069862e-06, + "loss": 2.0447, + "step": 382 + }, + { + "epoch": 0.4197260273972603, + "grad_norm": 2.93471097946167, + "learning_rate": 4.80706742485784e-06, + "loss": 2.0329, + "step": 383 + }, + { + "epoch": 0.4208219178082192, + "grad_norm": 2.3277227878570557, + "learning_rate": 4.805036555645816e-06, + "loss": 2.0805, + "step": 384 + }, + { + "epoch": 0.42191780821917807, + "grad_norm": 2.357966184616089, + "learning_rate": 4.8030056864337945e-06, + "loss": 2.0008, + "step": 385 + }, + { + "epoch": 0.423013698630137, + "grad_norm": 2.7505595684051514, + "learning_rate": 4.800974817221771e-06, + "loss": 2.059, + "step": 386 + }, + { + "epoch": 0.4241095890410959, + "grad_norm": 2.144338369369507, + "learning_rate": 4.798943948009749e-06, + "loss": 2.0125, + "step": 387 + }, + { + "epoch": 0.4252054794520548, + "grad_norm": 2.608407735824585, + "learning_rate": 4.7969130787977254e-06, + "loss": 2.0263, + "step": 388 + }, + { + "epoch": 0.4263013698630137, + "grad_norm": 2.4293761253356934, + "learning_rate": 4.794882209585703e-06, + "loss": 2.0107, + "step": 389 + }, + { + "epoch": 0.4273972602739726, + "grad_norm": 2.2136595249176025, + "learning_rate": 4.79285134037368e-06, + "loss": 2.0421, + "step": 390 + }, + { + "epoch": 0.4273972602739726, + "eval_loss": 2.1473641395568848, + "eval_runtime": 37.2895, + "eval_samples_per_second": 26.817, + "eval_steps_per_second": 1.126, + "step": 390 + }, + { + "epoch": 0.4284931506849315, + "grad_norm": 2.3706328868865967, + "learning_rate": 4.790820471161657e-06, + "loss": 1.9828, + "step": 391 + }, + { + "epoch": 0.4295890410958904, + "grad_norm": 2.1628777980804443, + "learning_rate": 4.7887896019496345e-06, + "loss": 2.0006, + "step": 392 + }, + { + "epoch": 0.43068493150684933, + "grad_norm": 2.5572149753570557, + "learning_rate": 4.786758732737612e-06, + "loss": 2.0046, + "step": 393 + }, + { + "epoch": 0.4317808219178082, + "grad_norm": 2.44661808013916, + "learning_rate": 4.784727863525589e-06, + "loss": 2.0513, + "step": 394 + }, + { + "epoch": 0.4328767123287671, + "grad_norm": 2.429487466812134, + "learning_rate": 4.782696994313566e-06, + "loss": 1.9619, + "step": 395 + }, + { + "epoch": 0.43397260273972604, + "grad_norm": 2.233353614807129, + "learning_rate": 4.780666125101544e-06, + "loss": 2.0141, + "step": 396 + }, + { + "epoch": 0.4350684931506849, + "grad_norm": 2.137549638748169, + "learning_rate": 4.778635255889521e-06, + "loss": 1.9512, + "step": 397 + }, + { + "epoch": 0.43616438356164383, + "grad_norm": 2.187638759613037, + "learning_rate": 4.776604386677498e-06, + "loss": 1.9802, + "step": 398 + }, + { + "epoch": 0.43726027397260275, + "grad_norm": 2.3084137439727783, + "learning_rate": 4.7745735174654755e-06, + "loss": 1.9887, + "step": 399 + }, + { + "epoch": 0.4383561643835616, + "grad_norm": 2.16654109954834, + "learning_rate": 4.772542648253453e-06, + "loss": 2.0535, + "step": 400 + }, + { + "epoch": 0.4383561643835616, + "eval_loss": 2.1444413661956787, + "eval_runtime": 37.2158, + "eval_samples_per_second": 26.87, + "eval_steps_per_second": 1.129, + "step": 400 + }, + { + "epoch": 0.43945205479452054, + "grad_norm": 2.0973072052001953, + "learning_rate": 4.77051177904143e-06, + "loss": 2.0529, + "step": 401 + }, + { + "epoch": 0.44054794520547946, + "grad_norm": 2.421881675720215, + "learning_rate": 4.768480909829407e-06, + "loss": 1.9893, + "step": 402 + }, + { + "epoch": 0.44164383561643833, + "grad_norm": 2.349296808242798, + "learning_rate": 4.766450040617385e-06, + "loss": 2.0237, + "step": 403 + }, + { + "epoch": 0.44273972602739725, + "grad_norm": 2.2154288291931152, + "learning_rate": 4.764419171405362e-06, + "loss": 1.996, + "step": 404 + }, + { + "epoch": 0.4438356164383562, + "grad_norm": 2.6355648040771484, + "learning_rate": 4.762388302193339e-06, + "loss": 2.0345, + "step": 405 + }, + { + "epoch": 0.4449315068493151, + "grad_norm": 2.352238655090332, + "learning_rate": 4.760357432981316e-06, + "loss": 2.0219, + "step": 406 + }, + { + "epoch": 0.44602739726027396, + "grad_norm": 2.1886510848999023, + "learning_rate": 4.758326563769294e-06, + "loss": 1.9923, + "step": 407 + }, + { + "epoch": 0.4471232876712329, + "grad_norm": 2.6483092308044434, + "learning_rate": 4.756295694557271e-06, + "loss": 2.0236, + "step": 408 + }, + { + "epoch": 0.4482191780821918, + "grad_norm": 2.382389545440674, + "learning_rate": 4.754264825345248e-06, + "loss": 2.0249, + "step": 409 + }, + { + "epoch": 0.44931506849315067, + "grad_norm": 2.4824411869049072, + "learning_rate": 4.7522339561332255e-06, + "loss": 1.9571, + "step": 410 + }, + { + "epoch": 0.44931506849315067, + "eval_loss": 2.1502621173858643, + "eval_runtime": 37.1789, + "eval_samples_per_second": 26.897, + "eval_steps_per_second": 1.13, + "step": 410 + }, + { + "epoch": 0.4504109589041096, + "grad_norm": 2.325589656829834, + "learning_rate": 4.750203086921203e-06, + "loss": 1.9753, + "step": 411 + }, + { + "epoch": 0.4515068493150685, + "grad_norm": 2.3618829250335693, + "learning_rate": 4.74817221770918e-06, + "loss": 1.9397, + "step": 412 + }, + { + "epoch": 0.4526027397260274, + "grad_norm": 2.3922839164733887, + "learning_rate": 4.7461413484971565e-06, + "loss": 2.0119, + "step": 413 + }, + { + "epoch": 0.4536986301369863, + "grad_norm": 2.403475046157837, + "learning_rate": 4.744110479285135e-06, + "loss": 2.0664, + "step": 414 + }, + { + "epoch": 0.4547945205479452, + "grad_norm": 2.225860834121704, + "learning_rate": 4.742079610073111e-06, + "loss": 1.9525, + "step": 415 + }, + { + "epoch": 0.4558904109589041, + "grad_norm": 2.344074010848999, + "learning_rate": 4.740048740861089e-06, + "loss": 1.9785, + "step": 416 + }, + { + "epoch": 0.456986301369863, + "grad_norm": 2.255382776260376, + "learning_rate": 4.738017871649066e-06, + "loss": 1.9902, + "step": 417 + }, + { + "epoch": 0.45808219178082193, + "grad_norm": 2.160628318786621, + "learning_rate": 4.735987002437044e-06, + "loss": 2.0075, + "step": 418 + }, + { + "epoch": 0.4591780821917808, + "grad_norm": 2.5667307376861572, + "learning_rate": 4.73395613322502e-06, + "loss": 2.0625, + "step": 419 + }, + { + "epoch": 0.4602739726027397, + "grad_norm": 2.3771040439605713, + "learning_rate": 4.731925264012998e-06, + "loss": 2.0724, + "step": 420 + }, + { + "epoch": 0.4602739726027397, + "eval_loss": 2.138829231262207, + "eval_runtime": 37.4361, + "eval_samples_per_second": 26.712, + "eval_steps_per_second": 1.122, + "step": 420 + }, + { + "epoch": 0.46136986301369864, + "grad_norm": 2.336482048034668, + "learning_rate": 4.729894394800975e-06, + "loss": 2.0234, + "step": 421 + }, + { + "epoch": 0.4624657534246575, + "grad_norm": 2.4206933975219727, + "learning_rate": 4.727863525588953e-06, + "loss": 1.9693, + "step": 422 + }, + { + "epoch": 0.46356164383561643, + "grad_norm": 2.4845645427703857, + "learning_rate": 4.725832656376929e-06, + "loss": 1.9239, + "step": 423 + }, + { + "epoch": 0.46465753424657535, + "grad_norm": 2.5199646949768066, + "learning_rate": 4.723801787164907e-06, + "loss": 1.9801, + "step": 424 + }, + { + "epoch": 0.4657534246575342, + "grad_norm": 2.2969748973846436, + "learning_rate": 4.721770917952885e-06, + "loss": 2.0192, + "step": 425 + }, + { + "epoch": 0.46684931506849314, + "grad_norm": 2.423503875732422, + "learning_rate": 4.719740048740861e-06, + "loss": 1.9684, + "step": 426 + }, + { + "epoch": 0.46794520547945206, + "grad_norm": 2.3681061267852783, + "learning_rate": 4.717709179528839e-06, + "loss": 1.9934, + "step": 427 + }, + { + "epoch": 0.469041095890411, + "grad_norm": 2.277965784072876, + "learning_rate": 4.715678310316816e-06, + "loss": 2.0166, + "step": 428 + }, + { + "epoch": 0.47013698630136985, + "grad_norm": 2.2730283737182617, + "learning_rate": 4.713647441104794e-06, + "loss": 2.011, + "step": 429 + }, + { + "epoch": 0.4712328767123288, + "grad_norm": 2.6694860458374023, + "learning_rate": 4.71161657189277e-06, + "loss": 1.9645, + "step": 430 + }, + { + "epoch": 0.4712328767123288, + "eval_loss": 2.1322011947631836, + "eval_runtime": 37.5723, + "eval_samples_per_second": 26.615, + "eval_steps_per_second": 1.118, + "step": 430 + }, + { + "epoch": 0.4723287671232877, + "grad_norm": 2.4100632667541504, + "learning_rate": 4.709585702680748e-06, + "loss": 1.9478, + "step": 431 + }, + { + "epoch": 0.47342465753424656, + "grad_norm": 2.2779541015625, + "learning_rate": 4.707554833468725e-06, + "loss": 2.0274, + "step": 432 + }, + { + "epoch": 0.4745205479452055, + "grad_norm": 2.55690336227417, + "learning_rate": 4.705523964256703e-06, + "loss": 2.0831, + "step": 433 + }, + { + "epoch": 0.4756164383561644, + "grad_norm": 2.455623149871826, + "learning_rate": 4.703493095044679e-06, + "loss": 1.9673, + "step": 434 + }, + { + "epoch": 0.4767123287671233, + "grad_norm": 2.2613699436187744, + "learning_rate": 4.701462225832657e-06, + "loss": 1.992, + "step": 435 + }, + { + "epoch": 0.4778082191780822, + "grad_norm": 2.4467933177948, + "learning_rate": 4.699431356620634e-06, + "loss": 1.9105, + "step": 436 + }, + { + "epoch": 0.4789041095890411, + "grad_norm": 2.4194023609161377, + "learning_rate": 4.697400487408612e-06, + "loss": 2.074, + "step": 437 + }, + { + "epoch": 0.48, + "grad_norm": 2.2165699005126953, + "learning_rate": 4.695369618196588e-06, + "loss": 2.0152, + "step": 438 + }, + { + "epoch": 0.4810958904109589, + "grad_norm": 2.1709225177764893, + "learning_rate": 4.693338748984566e-06, + "loss": 1.9888, + "step": 439 + }, + { + "epoch": 0.4821917808219178, + "grad_norm": 2.2325828075408936, + "learning_rate": 4.691307879772543e-06, + "loss": 1.9838, + "step": 440 + }, + { + "epoch": 0.4821917808219178, + "eval_loss": 2.128732919692993, + "eval_runtime": 37.205, + "eval_samples_per_second": 26.878, + "eval_steps_per_second": 1.129, + "step": 440 + }, + { + "epoch": 0.4832876712328767, + "grad_norm": 2.446331739425659, + "learning_rate": 4.68927701056052e-06, + "loss": 1.9295, + "step": 441 + }, + { + "epoch": 0.4843835616438356, + "grad_norm": 2.5459415912628174, + "learning_rate": 4.6872461413484975e-06, + "loss": 2.0351, + "step": 442 + }, + { + "epoch": 0.48547945205479454, + "grad_norm": 2.2877118587493896, + "learning_rate": 4.685215272136475e-06, + "loss": 2.0, + "step": 443 + }, + { + "epoch": 0.4865753424657534, + "grad_norm": 2.346625328063965, + "learning_rate": 4.683184402924452e-06, + "loss": 1.9928, + "step": 444 + }, + { + "epoch": 0.4876712328767123, + "grad_norm": 2.268467664718628, + "learning_rate": 4.681153533712429e-06, + "loss": 1.9969, + "step": 445 + }, + { + "epoch": 0.48876712328767125, + "grad_norm": 2.3699793815612793, + "learning_rate": 4.679122664500407e-06, + "loss": 2.0636, + "step": 446 + }, + { + "epoch": 0.4898630136986301, + "grad_norm": 2.2871041297912598, + "learning_rate": 4.677091795288384e-06, + "loss": 1.9535, + "step": 447 + }, + { + "epoch": 0.49095890410958903, + "grad_norm": 2.249393939971924, + "learning_rate": 4.675060926076361e-06, + "loss": 1.9358, + "step": 448 + }, + { + "epoch": 0.49205479452054796, + "grad_norm": 2.299863815307617, + "learning_rate": 4.673030056864338e-06, + "loss": 2.0299, + "step": 449 + }, + { + "epoch": 0.4931506849315068, + "grad_norm": 2.31239652633667, + "learning_rate": 4.670999187652316e-06, + "loss": 1.9372, + "step": 450 + }, + { + "epoch": 0.4931506849315068, + "eval_loss": 2.1267426013946533, + "eval_runtime": 37.1884, + "eval_samples_per_second": 26.89, + "eval_steps_per_second": 1.129, + "step": 450 + }, + { + "epoch": 0.49424657534246574, + "grad_norm": 2.2824623584747314, + "learning_rate": 4.668968318440293e-06, + "loss": 1.9435, + "step": 451 + }, + { + "epoch": 0.49534246575342467, + "grad_norm": 2.285339593887329, + "learning_rate": 4.66693744922827e-06, + "loss": 1.9455, + "step": 452 + }, + { + "epoch": 0.4964383561643836, + "grad_norm": 2.346592903137207, + "learning_rate": 4.6649065800162475e-06, + "loss": 2.0212, + "step": 453 + }, + { + "epoch": 0.49753424657534245, + "grad_norm": 2.280669689178467, + "learning_rate": 4.662875710804225e-06, + "loss": 1.9136, + "step": 454 + }, + { + "epoch": 0.4986301369863014, + "grad_norm": 2.178434133529663, + "learning_rate": 4.660844841592202e-06, + "loss": 1.9952, + "step": 455 + }, + { + "epoch": 0.4997260273972603, + "grad_norm": 2.108492851257324, + "learning_rate": 4.658813972380179e-06, + "loss": 1.9022, + "step": 456 + }, + { + "epoch": 0.5008219178082192, + "grad_norm": 2.3862147331237793, + "learning_rate": 4.656783103168157e-06, + "loss": 2.0332, + "step": 457 + }, + { + "epoch": 0.5019178082191781, + "grad_norm": 2.2017860412597656, + "learning_rate": 4.654752233956134e-06, + "loss": 1.8998, + "step": 458 + }, + { + "epoch": 0.503013698630137, + "grad_norm": 2.222134828567505, + "learning_rate": 4.652721364744111e-06, + "loss": 1.9836, + "step": 459 + }, + { + "epoch": 0.5041095890410959, + "grad_norm": 2.2974510192871094, + "learning_rate": 4.6506904955320885e-06, + "loss": 1.9626, + "step": 460 + }, + { + "epoch": 0.5041095890410959, + "eval_loss": 2.1345443725585938, + "eval_runtime": 37.2785, + "eval_samples_per_second": 26.825, + "eval_steps_per_second": 1.127, + "step": 460 + }, + { + "epoch": 0.5052054794520547, + "grad_norm": 2.107480764389038, + "learning_rate": 4.648659626320066e-06, + "loss": 2.0061, + "step": 461 + }, + { + "epoch": 0.5063013698630137, + "grad_norm": 2.3541181087493896, + "learning_rate": 4.646628757108043e-06, + "loss": 2.0133, + "step": 462 + }, + { + "epoch": 0.5073972602739726, + "grad_norm": 2.215571880340576, + "learning_rate": 4.6445978878960194e-06, + "loss": 1.9727, + "step": 463 + }, + { + "epoch": 0.5084931506849315, + "grad_norm": 2.2263615131378174, + "learning_rate": 4.6425670186839976e-06, + "loss": 1.9477, + "step": 464 + }, + { + "epoch": 0.5095890410958904, + "grad_norm": 2.178900957107544, + "learning_rate": 4.640536149471974e-06, + "loss": 1.9842, + "step": 465 + }, + { + "epoch": 0.5106849315068493, + "grad_norm": 2.3703105449676514, + "learning_rate": 4.638505280259952e-06, + "loss": 2.0082, + "step": 466 + }, + { + "epoch": 0.5117808219178083, + "grad_norm": 2.1587533950805664, + "learning_rate": 4.6364744110479285e-06, + "loss": 2.0271, + "step": 467 + }, + { + "epoch": 0.5128767123287671, + "grad_norm": 2.1626017093658447, + "learning_rate": 4.634443541835907e-06, + "loss": 1.9858, + "step": 468 + }, + { + "epoch": 0.513972602739726, + "grad_norm": 2.326869487762451, + "learning_rate": 4.632412672623883e-06, + "loss": 2.0413, + "step": 469 + }, + { + "epoch": 0.5150684931506849, + "grad_norm": 2.266502857208252, + "learning_rate": 4.630381803411861e-06, + "loss": 1.9584, + "step": 470 + }, + { + "epoch": 0.5150684931506849, + "eval_loss": 2.1273915767669678, + "eval_runtime": 37.2571, + "eval_samples_per_second": 26.841, + "eval_steps_per_second": 1.127, + "step": 470 + }, + { + "epoch": 0.5161643835616438, + "grad_norm": 2.131348133087158, + "learning_rate": 4.628350934199838e-06, + "loss": 1.863, + "step": 471 + }, + { + "epoch": 0.5172602739726028, + "grad_norm": 2.2635257244110107, + "learning_rate": 4.626320064987816e-06, + "loss": 1.9699, + "step": 472 + }, + { + "epoch": 0.5183561643835617, + "grad_norm": 2.1303744316101074, + "learning_rate": 4.624289195775792e-06, + "loss": 2.0001, + "step": 473 + }, + { + "epoch": 0.5194520547945205, + "grad_norm": 2.1736788749694824, + "learning_rate": 4.62225832656377e-06, + "loss": 1.9684, + "step": 474 + }, + { + "epoch": 0.5205479452054794, + "grad_norm": 2.1996278762817383, + "learning_rate": 4.620227457351747e-06, + "loss": 1.9739, + "step": 475 + }, + { + "epoch": 0.5216438356164383, + "grad_norm": 2.225456953048706, + "learning_rate": 4.618196588139724e-06, + "loss": 1.9649, + "step": 476 + }, + { + "epoch": 0.5227397260273973, + "grad_norm": 2.118537187576294, + "learning_rate": 4.616165718927701e-06, + "loss": 1.9911, + "step": 477 + }, + { + "epoch": 0.5238356164383562, + "grad_norm": 2.232940435409546, + "learning_rate": 4.6141348497156786e-06, + "loss": 1.944, + "step": 478 + }, + { + "epoch": 0.5249315068493151, + "grad_norm": 2.1776034832000732, + "learning_rate": 4.612103980503656e-06, + "loss": 2.0342, + "step": 479 + }, + { + "epoch": 0.5260273972602739, + "grad_norm": 2.042645215988159, + "learning_rate": 4.610073111291633e-06, + "loss": 2.0124, + "step": 480 + }, + { + "epoch": 0.5260273972602739, + "eval_loss": 2.118508815765381, + "eval_runtime": 37.2764, + "eval_samples_per_second": 26.827, + "eval_steps_per_second": 1.127, + "step": 480 + }, + { + "epoch": 0.5271232876712328, + "grad_norm": 2.302661895751953, + "learning_rate": 4.60804224207961e-06, + "loss": 1.9874, + "step": 481 + }, + { + "epoch": 0.5282191780821918, + "grad_norm": 2.1251118183135986, + "learning_rate": 4.606011372867588e-06, + "loss": 1.9594, + "step": 482 + }, + { + "epoch": 0.5293150684931507, + "grad_norm": 2.157043933868408, + "learning_rate": 4.603980503655565e-06, + "loss": 1.9955, + "step": 483 + }, + { + "epoch": 0.5304109589041096, + "grad_norm": 2.291428804397583, + "learning_rate": 4.601949634443542e-06, + "loss": 1.9695, + "step": 484 + }, + { + "epoch": 0.5315068493150685, + "grad_norm": 2.2341537475585938, + "learning_rate": 4.5999187652315195e-06, + "loss": 1.899, + "step": 485 + }, + { + "epoch": 0.5326027397260275, + "grad_norm": 2.1881206035614014, + "learning_rate": 4.597887896019497e-06, + "loss": 1.9066, + "step": 486 + }, + { + "epoch": 0.5336986301369863, + "grad_norm": 2.3057658672332764, + "learning_rate": 4.595857026807474e-06, + "loss": 1.9646, + "step": 487 + }, + { + "epoch": 0.5347945205479452, + "grad_norm": 2.2318356037139893, + "learning_rate": 4.593826157595451e-06, + "loss": 1.9662, + "step": 488 + }, + { + "epoch": 0.5358904109589041, + "grad_norm": 2.170509099960327, + "learning_rate": 4.591795288383429e-06, + "loss": 1.9522, + "step": 489 + }, + { + "epoch": 0.536986301369863, + "grad_norm": 2.101646661758423, + "learning_rate": 4.589764419171406e-06, + "loss": 2.0315, + "step": 490 + }, + { + "epoch": 0.536986301369863, + "eval_loss": 2.113968849182129, + "eval_runtime": 37.2751, + "eval_samples_per_second": 26.828, + "eval_steps_per_second": 1.127, + "step": 490 + }, + { + "epoch": 0.538082191780822, + "grad_norm": 2.1563706398010254, + "learning_rate": 4.587733549959383e-06, + "loss": 1.9843, + "step": 491 + }, + { + "epoch": 0.5391780821917809, + "grad_norm": 2.295722246170044, + "learning_rate": 4.5857026807473604e-06, + "loss": 1.9499, + "step": 492 + }, + { + "epoch": 0.5402739726027397, + "grad_norm": 1.974424123764038, + "learning_rate": 4.583671811535338e-06, + "loss": 1.9346, + "step": 493 + }, + { + "epoch": 0.5413698630136986, + "grad_norm": 2.4402668476104736, + "learning_rate": 4.581640942323315e-06, + "loss": 1.9053, + "step": 494 + }, + { + "epoch": 0.5424657534246575, + "grad_norm": 2.269000768661499, + "learning_rate": 4.579610073111292e-06, + "loss": 2.0108, + "step": 495 + }, + { + "epoch": 0.5435616438356164, + "grad_norm": 2.1182594299316406, + "learning_rate": 4.5775792038992695e-06, + "loss": 2.0155, + "step": 496 + }, + { + "epoch": 0.5446575342465754, + "grad_norm": 2.2181780338287354, + "learning_rate": 4.575548334687247e-06, + "loss": 2.0112, + "step": 497 + }, + { + "epoch": 0.5457534246575343, + "grad_norm": 2.2806756496429443, + "learning_rate": 4.573517465475224e-06, + "loss": 2.0418, + "step": 498 + }, + { + "epoch": 0.5468493150684931, + "grad_norm": 2.2690632343292236, + "learning_rate": 4.571486596263201e-06, + "loss": 1.9457, + "step": 499 + }, + { + "epoch": 0.547945205479452, + "grad_norm": 2.361335515975952, + "learning_rate": 4.569455727051178e-06, + "loss": 1.9949, + "step": 500 + }, + { + "epoch": 0.547945205479452, + "eval_loss": 2.1165614128112793, + "eval_runtime": 37.2664, + "eval_samples_per_second": 26.834, + "eval_steps_per_second": 1.127, + "step": 500 + }, + { + "epoch": 0.549041095890411, + "grad_norm": 2.3833138942718506, + "learning_rate": 4.567424857839156e-06, + "loss": 1.9932, + "step": 501 + }, + { + "epoch": 0.5501369863013699, + "grad_norm": 2.328500509262085, + "learning_rate": 4.565393988627132e-06, + "loss": 1.9773, + "step": 502 + }, + { + "epoch": 0.5512328767123288, + "grad_norm": 2.4191694259643555, + "learning_rate": 4.5633631194151105e-06, + "loss": 2.0389, + "step": 503 + }, + { + "epoch": 0.5523287671232877, + "grad_norm": 2.4106884002685547, + "learning_rate": 4.561332250203087e-06, + "loss": 1.965, + "step": 504 + }, + { + "epoch": 0.5534246575342465, + "grad_norm": 2.1452605724334717, + "learning_rate": 4.559301380991065e-06, + "loss": 1.9713, + "step": 505 + }, + { + "epoch": 0.5545205479452054, + "grad_norm": 2.2753357887268066, + "learning_rate": 4.5572705117790414e-06, + "loss": 1.9116, + "step": 506 + }, + { + "epoch": 0.5556164383561644, + "grad_norm": 2.6925032138824463, + "learning_rate": 4.5552396425670196e-06, + "loss": 1.9593, + "step": 507 + }, + { + "epoch": 0.5567123287671233, + "grad_norm": 2.3136658668518066, + "learning_rate": 4.553208773354996e-06, + "loss": 2.0072, + "step": 508 + }, + { + "epoch": 0.5578082191780822, + "grad_norm": 2.254849672317505, + "learning_rate": 4.551177904142974e-06, + "loss": 1.9193, + "step": 509 + }, + { + "epoch": 0.5589041095890411, + "grad_norm": 2.352755546569824, + "learning_rate": 4.5491470349309505e-06, + "loss": 2.0228, + "step": 510 + }, + { + "epoch": 0.5589041095890411, + "eval_loss": 2.114323377609253, + "eval_runtime": 37.3629, + "eval_samples_per_second": 26.764, + "eval_steps_per_second": 1.124, + "step": 510 + }, + { + "epoch": 0.56, + "grad_norm": 2.31339955329895, + "learning_rate": 4.547116165718929e-06, + "loss": 1.9416, + "step": 511 + }, + { + "epoch": 0.5610958904109589, + "grad_norm": 2.191164970397949, + "learning_rate": 4.545085296506905e-06, + "loss": 1.9709, + "step": 512 + }, + { + "epoch": 0.5621917808219178, + "grad_norm": 2.193551778793335, + "learning_rate": 4.543054427294882e-06, + "loss": 1.9475, + "step": 513 + }, + { + "epoch": 0.5632876712328767, + "grad_norm": 2.222710371017456, + "learning_rate": 4.54102355808286e-06, + "loss": 1.9283, + "step": 514 + }, + { + "epoch": 0.5643835616438356, + "grad_norm": 2.078907012939453, + "learning_rate": 4.538992688870837e-06, + "loss": 1.929, + "step": 515 + }, + { + "epoch": 0.5654794520547946, + "grad_norm": 2.2043654918670654, + "learning_rate": 4.536961819658814e-06, + "loss": 1.9591, + "step": 516 + }, + { + "epoch": 0.5665753424657535, + "grad_norm": 2.192481517791748, + "learning_rate": 4.5349309504467915e-06, + "loss": 1.9618, + "step": 517 + }, + { + "epoch": 0.5676712328767123, + "grad_norm": 2.3509292602539062, + "learning_rate": 4.532900081234769e-06, + "loss": 1.9782, + "step": 518 + }, + { + "epoch": 0.5687671232876712, + "grad_norm": 2.2415924072265625, + "learning_rate": 4.530869212022746e-06, + "loss": 1.9456, + "step": 519 + }, + { + "epoch": 0.5698630136986301, + "grad_norm": 2.2135009765625, + "learning_rate": 4.528838342810723e-06, + "loss": 1.9671, + "step": 520 + }, + { + "epoch": 0.5698630136986301, + "eval_loss": 2.109738826751709, + "eval_runtime": 37.5253, + "eval_samples_per_second": 26.649, + "eval_steps_per_second": 1.119, + "step": 520 + }, + { + "epoch": 0.570958904109589, + "grad_norm": 2.3513574600219727, + "learning_rate": 4.526807473598701e-06, + "loss": 1.9625, + "step": 521 + }, + { + "epoch": 0.572054794520548, + "grad_norm": 1.9949424266815186, + "learning_rate": 4.524776604386678e-06, + "loss": 1.9621, + "step": 522 + }, + { + "epoch": 0.5731506849315069, + "grad_norm": 2.199488401412964, + "learning_rate": 4.522745735174655e-06, + "loss": 1.9389, + "step": 523 + }, + { + "epoch": 0.5742465753424657, + "grad_norm": 2.2119452953338623, + "learning_rate": 4.520714865962632e-06, + "loss": 1.9936, + "step": 524 + }, + { + "epoch": 0.5753424657534246, + "grad_norm": 2.242812156677246, + "learning_rate": 4.51868399675061e-06, + "loss": 1.9728, + "step": 525 + }, + { + "epoch": 0.5764383561643835, + "grad_norm": 2.1713850498199463, + "learning_rate": 4.516653127538587e-06, + "loss": 1.98, + "step": 526 + }, + { + "epoch": 0.5775342465753425, + "grad_norm": 2.261742353439331, + "learning_rate": 4.514622258326564e-06, + "loss": 1.9412, + "step": 527 + }, + { + "epoch": 0.5786301369863014, + "grad_norm": 2.222663402557373, + "learning_rate": 4.5125913891145415e-06, + "loss": 1.9266, + "step": 528 + }, + { + "epoch": 0.5797260273972603, + "grad_norm": 2.1108906269073486, + "learning_rate": 4.510560519902519e-06, + "loss": 1.9539, + "step": 529 + }, + { + "epoch": 0.5808219178082191, + "grad_norm": 1.9594205617904663, + "learning_rate": 4.508529650690496e-06, + "loss": 1.9842, + "step": 530 + }, + { + "epoch": 0.5808219178082191, + "eval_loss": 2.1073970794677734, + "eval_runtime": 37.6167, + "eval_samples_per_second": 26.584, + "eval_steps_per_second": 1.117, + "step": 530 + }, + { + "epoch": 0.581917808219178, + "grad_norm": 2.1339495182037354, + "learning_rate": 4.506498781478473e-06, + "loss": 1.9095, + "step": 531 + }, + { + "epoch": 0.583013698630137, + "grad_norm": 2.116248607635498, + "learning_rate": 4.504467912266451e-06, + "loss": 1.9091, + "step": 532 + }, + { + "epoch": 0.5841095890410959, + "grad_norm": 2.063962459564209, + "learning_rate": 4.502437043054428e-06, + "loss": 1.9537, + "step": 533 + }, + { + "epoch": 0.5852054794520548, + "grad_norm": 2.162569522857666, + "learning_rate": 4.500406173842405e-06, + "loss": 2.0253, + "step": 534 + }, + { + "epoch": 0.5863013698630137, + "grad_norm": 2.1783549785614014, + "learning_rate": 4.4983753046303824e-06, + "loss": 1.9714, + "step": 535 + }, + { + "epoch": 0.5873972602739727, + "grad_norm": 2.0791046619415283, + "learning_rate": 4.49634443541836e-06, + "loss": 1.9629, + "step": 536 + }, + { + "epoch": 0.5884931506849315, + "grad_norm": 2.3923251628875732, + "learning_rate": 4.494313566206336e-06, + "loss": 1.9445, + "step": 537 + }, + { + "epoch": 0.5895890410958904, + "grad_norm": 2.1848347187042236, + "learning_rate": 4.492282696994314e-06, + "loss": 2.007, + "step": 538 + }, + { + "epoch": 0.5906849315068493, + "grad_norm": 2.0824925899505615, + "learning_rate": 4.490251827782291e-06, + "loss": 1.9409, + "step": 539 + }, + { + "epoch": 0.5917808219178082, + "grad_norm": 2.057790756225586, + "learning_rate": 4.488220958570269e-06, + "loss": 1.9061, + "step": 540 + }, + { + "epoch": 0.5917808219178082, + "eval_loss": 2.1101415157318115, + "eval_runtime": 37.2235, + "eval_samples_per_second": 26.865, + "eval_steps_per_second": 1.128, + "step": 540 + }, + { + "epoch": 0.5928767123287672, + "grad_norm": 2.3652148246765137, + "learning_rate": 4.486190089358245e-06, + "loss": 1.9379, + "step": 541 + }, + { + "epoch": 0.5939726027397261, + "grad_norm": 2.195535659790039, + "learning_rate": 4.484159220146223e-06, + "loss": 1.8881, + "step": 542 + }, + { + "epoch": 0.5950684931506849, + "grad_norm": 2.258000135421753, + "learning_rate": 4.4821283509342e-06, + "loss": 1.9923, + "step": 543 + }, + { + "epoch": 0.5961643835616438, + "grad_norm": 2.097041606903076, + "learning_rate": 4.480097481722178e-06, + "loss": 1.9715, + "step": 544 + }, + { + "epoch": 0.5972602739726027, + "grad_norm": 2.1741156578063965, + "learning_rate": 4.478066612510154e-06, + "loss": 1.9487, + "step": 545 + }, + { + "epoch": 0.5983561643835617, + "grad_norm": 2.135338306427002, + "learning_rate": 4.4760357432981325e-06, + "loss": 1.9172, + "step": 546 + }, + { + "epoch": 0.5994520547945206, + "grad_norm": 2.3435418605804443, + "learning_rate": 4.474004874086109e-06, + "loss": 1.945, + "step": 547 + }, + { + "epoch": 0.6005479452054795, + "grad_norm": 2.2242488861083984, + "learning_rate": 4.471974004874087e-06, + "loss": 2.0717, + "step": 548 + }, + { + "epoch": 0.6016438356164383, + "grad_norm": 2.3371622562408447, + "learning_rate": 4.4699431356620635e-06, + "loss": 1.9082, + "step": 549 + }, + { + "epoch": 0.6027397260273972, + "grad_norm": 2.176064968109131, + "learning_rate": 4.467912266450041e-06, + "loss": 2.0154, + "step": 550 + }, + { + "epoch": 0.6027397260273972, + "eval_loss": 2.105236291885376, + "eval_runtime": 37.1885, + "eval_samples_per_second": 26.89, + "eval_steps_per_second": 1.129, + "step": 550 + }, + { + "epoch": 0.6038356164383561, + "grad_norm": 2.131166934967041, + "learning_rate": 4.465881397238018e-06, + "loss": 1.9575, + "step": 551 + }, + { + "epoch": 0.6049315068493151, + "grad_norm": 2.094514846801758, + "learning_rate": 4.463850528025995e-06, + "loss": 1.9544, + "step": 552 + }, + { + "epoch": 0.606027397260274, + "grad_norm": 2.0827765464782715, + "learning_rate": 4.4618196588139726e-06, + "loss": 1.9539, + "step": 553 + }, + { + "epoch": 0.6071232876712329, + "grad_norm": 2.225749969482422, + "learning_rate": 4.45978878960195e-06, + "loss": 1.983, + "step": 554 + }, + { + "epoch": 0.6082191780821918, + "grad_norm": 2.1363680362701416, + "learning_rate": 4.457757920389927e-06, + "loss": 1.9769, + "step": 555 + }, + { + "epoch": 0.6093150684931506, + "grad_norm": 2.238170623779297, + "learning_rate": 4.455727051177904e-06, + "loss": 1.9094, + "step": 556 + }, + { + "epoch": 0.6104109589041096, + "grad_norm": 2.228384494781494, + "learning_rate": 4.453696181965882e-06, + "loss": 1.9464, + "step": 557 + }, + { + "epoch": 0.6115068493150685, + "grad_norm": 2.479428768157959, + "learning_rate": 4.451665312753859e-06, + "loss": 2.0017, + "step": 558 + }, + { + "epoch": 0.6126027397260274, + "grad_norm": 2.2331902980804443, + "learning_rate": 4.449634443541836e-06, + "loss": 1.9728, + "step": 559 + }, + { + "epoch": 0.6136986301369863, + "grad_norm": 2.1728556156158447, + "learning_rate": 4.4476035743298135e-06, + "loss": 2.0153, + "step": 560 + }, + { + "epoch": 0.6136986301369863, + "eval_loss": 2.0978519916534424, + "eval_runtime": 37.233, + "eval_samples_per_second": 26.858, + "eval_steps_per_second": 1.128, + "step": 560 + }, + { + "epoch": 0.6147945205479453, + "grad_norm": 2.377814292907715, + "learning_rate": 4.445572705117791e-06, + "loss": 1.9242, + "step": 561 + }, + { + "epoch": 0.6158904109589041, + "grad_norm": 2.190436601638794, + "learning_rate": 4.443541835905768e-06, + "loss": 1.9454, + "step": 562 + }, + { + "epoch": 0.616986301369863, + "grad_norm": 2.052201986312866, + "learning_rate": 4.441510966693745e-06, + "loss": 2.0048, + "step": 563 + }, + { + "epoch": 0.6180821917808219, + "grad_norm": 2.2706589698791504, + "learning_rate": 4.439480097481723e-06, + "loss": 1.9426, + "step": 564 + }, + { + "epoch": 0.6191780821917808, + "grad_norm": 2.3478105068206787, + "learning_rate": 4.4374492282697e-06, + "loss": 2.0047, + "step": 565 + }, + { + "epoch": 0.6202739726027398, + "grad_norm": 2.465078115463257, + "learning_rate": 4.435418359057677e-06, + "loss": 1.8739, + "step": 566 + }, + { + "epoch": 0.6213698630136987, + "grad_norm": 2.0979490280151367, + "learning_rate": 4.433387489845654e-06, + "loss": 1.9588, + "step": 567 + }, + { + "epoch": 0.6224657534246575, + "grad_norm": 2.1239588260650635, + "learning_rate": 4.431356620633632e-06, + "loss": 1.9695, + "step": 568 + }, + { + "epoch": 0.6235616438356164, + "grad_norm": 2.014007806777954, + "learning_rate": 4.429325751421609e-06, + "loss": 1.9232, + "step": 569 + }, + { + "epoch": 0.6246575342465753, + "grad_norm": 2.212414026260376, + "learning_rate": 4.427294882209586e-06, + "loss": 1.928, + "step": 570 + }, + { + "epoch": 0.6246575342465753, + "eval_loss": 2.103149890899658, + "eval_runtime": 37.2641, + "eval_samples_per_second": 26.835, + "eval_steps_per_second": 1.127, + "step": 570 + }, + { + "epoch": 0.6257534246575343, + "grad_norm": 2.0944411754608154, + "learning_rate": 4.4252640129975635e-06, + "loss": 1.9568, + "step": 571 + }, + { + "epoch": 0.6268493150684932, + "grad_norm": 2.1068055629730225, + "learning_rate": 4.423233143785541e-06, + "loss": 1.8917, + "step": 572 + }, + { + "epoch": 0.6279452054794521, + "grad_norm": 2.1992151737213135, + "learning_rate": 4.421202274573518e-06, + "loss": 1.9536, + "step": 573 + }, + { + "epoch": 0.6290410958904109, + "grad_norm": 2.2079834938049316, + "learning_rate": 4.4191714053614945e-06, + "loss": 1.9603, + "step": 574 + }, + { + "epoch": 0.6301369863013698, + "grad_norm": 2.1230316162109375, + "learning_rate": 4.417140536149473e-06, + "loss": 1.8529, + "step": 575 + }, + { + "epoch": 0.6312328767123287, + "grad_norm": 2.1069579124450684, + "learning_rate": 4.415109666937449e-06, + "loss": 1.8997, + "step": 576 + }, + { + "epoch": 0.6323287671232877, + "grad_norm": 2.1683409214019775, + "learning_rate": 4.413078797725427e-06, + "loss": 1.971, + "step": 577 + }, + { + "epoch": 0.6334246575342466, + "grad_norm": 2.0599024295806885, + "learning_rate": 4.411047928513404e-06, + "loss": 1.9154, + "step": 578 + }, + { + "epoch": 0.6345205479452055, + "grad_norm": 2.1605849266052246, + "learning_rate": 4.409017059301382e-06, + "loss": 1.9562, + "step": 579 + }, + { + "epoch": 0.6356164383561644, + "grad_norm": 2.151249408721924, + "learning_rate": 4.406986190089358e-06, + "loss": 1.9571, + "step": 580 + }, + { + "epoch": 0.6356164383561644, + "eval_loss": 2.098177671432495, + "eval_runtime": 37.2578, + "eval_samples_per_second": 26.84, + "eval_steps_per_second": 1.127, + "step": 580 + }, + { + "epoch": 0.6367123287671232, + "grad_norm": 2.1819236278533936, + "learning_rate": 4.404955320877336e-06, + "loss": 1.9816, + "step": 581 + }, + { + "epoch": 0.6378082191780822, + "grad_norm": 2.1131205558776855, + "learning_rate": 4.402924451665313e-06, + "loss": 1.9738, + "step": 582 + }, + { + "epoch": 0.6389041095890411, + "grad_norm": 2.0852456092834473, + "learning_rate": 4.400893582453291e-06, + "loss": 1.924, + "step": 583 + }, + { + "epoch": 0.64, + "grad_norm": 2.0997941493988037, + "learning_rate": 4.398862713241267e-06, + "loss": 1.9067, + "step": 584 + }, + { + "epoch": 0.6410958904109589, + "grad_norm": 2.0848538875579834, + "learning_rate": 4.396831844029245e-06, + "loss": 1.9393, + "step": 585 + }, + { + "epoch": 0.6421917808219179, + "grad_norm": 2.2121758460998535, + "learning_rate": 4.394800974817222e-06, + "loss": 1.9458, + "step": 586 + }, + { + "epoch": 0.6432876712328767, + "grad_norm": 2.194462299346924, + "learning_rate": 4.392770105605199e-06, + "loss": 1.9528, + "step": 587 + }, + { + "epoch": 0.6443835616438356, + "grad_norm": 2.1936254501342773, + "learning_rate": 4.390739236393176e-06, + "loss": 1.9584, + "step": 588 + }, + { + "epoch": 0.6454794520547945, + "grad_norm": 2.1685783863067627, + "learning_rate": 4.388708367181154e-06, + "loss": 1.9666, + "step": 589 + }, + { + "epoch": 0.6465753424657534, + "grad_norm": 2.334571123123169, + "learning_rate": 4.386677497969131e-06, + "loss": 1.9748, + "step": 590 + }, + { + "epoch": 0.6465753424657534, + "eval_loss": 2.0937106609344482, + "eval_runtime": 37.2877, + "eval_samples_per_second": 26.818, + "eval_steps_per_second": 1.126, + "step": 590 + }, + { + "epoch": 0.6476712328767124, + "grad_norm": 2.2901110649108887, + "learning_rate": 4.384646628757108e-06, + "loss": 1.9597, + "step": 591 + }, + { + "epoch": 0.6487671232876713, + "grad_norm": 2.2364838123321533, + "learning_rate": 4.3826157595450855e-06, + "loss": 2.004, + "step": 592 + }, + { + "epoch": 0.6498630136986301, + "grad_norm": 2.1088364124298096, + "learning_rate": 4.380584890333063e-06, + "loss": 1.9319, + "step": 593 + }, + { + "epoch": 0.650958904109589, + "grad_norm": 2.164210319519043, + "learning_rate": 4.37855402112104e-06, + "loss": 1.8906, + "step": 594 + }, + { + "epoch": 0.6520547945205479, + "grad_norm": 2.205200672149658, + "learning_rate": 4.376523151909017e-06, + "loss": 1.8474, + "step": 595 + }, + { + "epoch": 0.6531506849315069, + "grad_norm": 2.239530086517334, + "learning_rate": 4.3744922826969946e-06, + "loss": 2.0033, + "step": 596 + }, + { + "epoch": 0.6542465753424658, + "grad_norm": 2.214747190475464, + "learning_rate": 4.372461413484972e-06, + "loss": 1.9258, + "step": 597 + }, + { + "epoch": 0.6553424657534247, + "grad_norm": 2.1746792793273926, + "learning_rate": 4.370430544272949e-06, + "loss": 1.9479, + "step": 598 + }, + { + "epoch": 0.6564383561643835, + "grad_norm": 2.194131851196289, + "learning_rate": 4.368399675060926e-06, + "loss": 1.9825, + "step": 599 + }, + { + "epoch": 0.6575342465753424, + "grad_norm": 2.3334717750549316, + "learning_rate": 4.366368805848904e-06, + "loss": 1.9831, + "step": 600 + }, + { + "epoch": 0.6575342465753424, + "eval_loss": 2.0881216526031494, + "eval_runtime": 37.3226, + "eval_samples_per_second": 26.793, + "eval_steps_per_second": 1.125, + "step": 600 + }, + { + "epoch": 0.6586301369863014, + "grad_norm": 2.269484758377075, + "learning_rate": 4.364337936636881e-06, + "loss": 1.9407, + "step": 601 + }, + { + "epoch": 0.6597260273972603, + "grad_norm": 2.470773935317993, + "learning_rate": 4.362307067424858e-06, + "loss": 1.8905, + "step": 602 + }, + { + "epoch": 0.6608219178082192, + "grad_norm": 2.485574960708618, + "learning_rate": 4.3602761982128355e-06, + "loss": 1.953, + "step": 603 + }, + { + "epoch": 0.6619178082191781, + "grad_norm": 2.2443618774414062, + "learning_rate": 4.358245329000813e-06, + "loss": 1.9244, + "step": 604 + }, + { + "epoch": 0.663013698630137, + "grad_norm": 2.039600372314453, + "learning_rate": 4.35621445978879e-06, + "loss": 1.9364, + "step": 605 + }, + { + "epoch": 0.6641095890410958, + "grad_norm": 2.093001127243042, + "learning_rate": 4.354183590576767e-06, + "loss": 1.9439, + "step": 606 + }, + { + "epoch": 0.6652054794520548, + "grad_norm": 2.3433260917663574, + "learning_rate": 4.352152721364745e-06, + "loss": 1.9322, + "step": 607 + }, + { + "epoch": 0.6663013698630137, + "grad_norm": 2.2403812408447266, + "learning_rate": 4.350121852152722e-06, + "loss": 1.8829, + "step": 608 + }, + { + "epoch": 0.6673972602739726, + "grad_norm": 2.1542811393737793, + "learning_rate": 4.348090982940699e-06, + "loss": 1.9549, + "step": 609 + }, + { + "epoch": 0.6684931506849315, + "grad_norm": 2.1526548862457275, + "learning_rate": 4.3460601137286764e-06, + "loss": 1.9433, + "step": 610 + }, + { + "epoch": 0.6684931506849315, + "eval_loss": 2.083638906478882, + "eval_runtime": 37.2677, + "eval_samples_per_second": 26.833, + "eval_steps_per_second": 1.127, + "step": 610 + }, + { + "epoch": 0.6695890410958905, + "grad_norm": 2.093421459197998, + "learning_rate": 4.344029244516654e-06, + "loss": 1.9643, + "step": 611 + }, + { + "epoch": 0.6706849315068493, + "grad_norm": 2.2205088138580322, + "learning_rate": 4.341998375304631e-06, + "loss": 1.9309, + "step": 612 + }, + { + "epoch": 0.6717808219178082, + "grad_norm": 2.059763193130493, + "learning_rate": 4.339967506092607e-06, + "loss": 2.0064, + "step": 613 + }, + { + "epoch": 0.6728767123287671, + "grad_norm": 2.0340707302093506, + "learning_rate": 4.3379366368805855e-06, + "loss": 1.9394, + "step": 614 + }, + { + "epoch": 0.673972602739726, + "grad_norm": 2.0716137886047363, + "learning_rate": 4.335905767668562e-06, + "loss": 1.9421, + "step": 615 + }, + { + "epoch": 0.675068493150685, + "grad_norm": 2.176809549331665, + "learning_rate": 4.33387489845654e-06, + "loss": 1.9261, + "step": 616 + }, + { + "epoch": 0.6761643835616439, + "grad_norm": 2.144385814666748, + "learning_rate": 4.3318440292445165e-06, + "loss": 1.9635, + "step": 617 + }, + { + "epoch": 0.6772602739726027, + "grad_norm": 2.0949580669403076, + "learning_rate": 4.329813160032495e-06, + "loss": 1.9417, + "step": 618 + }, + { + "epoch": 0.6783561643835616, + "grad_norm": 2.1727259159088135, + "learning_rate": 4.327782290820471e-06, + "loss": 1.9041, + "step": 619 + }, + { + "epoch": 0.6794520547945205, + "grad_norm": 2.012868881225586, + "learning_rate": 4.325751421608449e-06, + "loss": 1.8973, + "step": 620 + }, + { + "epoch": 0.6794520547945205, + "eval_loss": 2.0839755535125732, + "eval_runtime": 37.5302, + "eval_samples_per_second": 26.645, + "eval_steps_per_second": 1.119, + "step": 620 + }, + { + "epoch": 0.6805479452054795, + "grad_norm": 2.204153299331665, + "learning_rate": 4.323720552396426e-06, + "loss": 1.9719, + "step": 621 + }, + { + "epoch": 0.6816438356164384, + "grad_norm": 2.2421770095825195, + "learning_rate": 4.321689683184404e-06, + "loss": 1.9207, + "step": 622 + }, + { + "epoch": 0.6827397260273973, + "grad_norm": 2.2263410091400146, + "learning_rate": 4.31965881397238e-06, + "loss": 1.9146, + "step": 623 + }, + { + "epoch": 0.6838356164383562, + "grad_norm": 2.2015221118927, + "learning_rate": 4.3176279447603574e-06, + "loss": 1.9311, + "step": 624 + }, + { + "epoch": 0.684931506849315, + "grad_norm": 2.1459643840789795, + "learning_rate": 4.315597075548335e-06, + "loss": 1.9673, + "step": 625 + }, + { + "epoch": 0.686027397260274, + "grad_norm": 2.278928756713867, + "learning_rate": 4.313566206336312e-06, + "loss": 1.954, + "step": 626 + }, + { + "epoch": 0.6871232876712329, + "grad_norm": 2.05305552482605, + "learning_rate": 4.311535337124289e-06, + "loss": 1.9789, + "step": 627 + }, + { + "epoch": 0.6882191780821918, + "grad_norm": 2.3979289531707764, + "learning_rate": 4.3095044679122665e-06, + "loss": 1.926, + "step": 628 + }, + { + "epoch": 0.6893150684931507, + "grad_norm": 2.1635828018188477, + "learning_rate": 4.307473598700244e-06, + "loss": 1.9316, + "step": 629 + }, + { + "epoch": 0.6904109589041096, + "grad_norm": 2.237713098526001, + "learning_rate": 4.305442729488221e-06, + "loss": 1.9409, + "step": 630 + }, + { + "epoch": 0.6904109589041096, + "eval_loss": 2.084226131439209, + "eval_runtime": 37.6039, + "eval_samples_per_second": 26.593, + "eval_steps_per_second": 1.117, + "step": 630 + }, + { + "epoch": 0.6915068493150685, + "grad_norm": 2.157435894012451, + "learning_rate": 4.303411860276198e-06, + "loss": 1.8777, + "step": 631 + }, + { + "epoch": 0.6926027397260274, + "grad_norm": 2.20766019821167, + "learning_rate": 4.301380991064176e-06, + "loss": 1.9803, + "step": 632 + }, + { + "epoch": 0.6936986301369863, + "grad_norm": 2.173079490661621, + "learning_rate": 4.299350121852153e-06, + "loss": 1.9541, + "step": 633 + }, + { + "epoch": 0.6947945205479452, + "grad_norm": 1.9727106094360352, + "learning_rate": 4.29731925264013e-06, + "loss": 1.9766, + "step": 634 + }, + { + "epoch": 0.6958904109589041, + "grad_norm": 2.268538475036621, + "learning_rate": 4.2952883834281075e-06, + "loss": 1.9737, + "step": 635 + }, + { + "epoch": 0.6969863013698631, + "grad_norm": 2.2118406295776367, + "learning_rate": 4.293257514216085e-06, + "loss": 1.9291, + "step": 636 + }, + { + "epoch": 0.6980821917808219, + "grad_norm": 2.2939369678497314, + "learning_rate": 4.291226645004062e-06, + "loss": 1.8868, + "step": 637 + }, + { + "epoch": 0.6991780821917808, + "grad_norm": 2.4940083026885986, + "learning_rate": 4.289195775792039e-06, + "loss": 1.9694, + "step": 638 + }, + { + "epoch": 0.7002739726027397, + "grad_norm": 2.1988351345062256, + "learning_rate": 4.287164906580017e-06, + "loss": 1.9781, + "step": 639 + }, + { + "epoch": 0.7013698630136986, + "grad_norm": 2.099578619003296, + "learning_rate": 4.285134037367994e-06, + "loss": 1.9145, + "step": 640 + }, + { + "epoch": 0.7013698630136986, + "eval_loss": 2.07913875579834, + "eval_runtime": 37.3458, + "eval_samples_per_second": 26.777, + "eval_steps_per_second": 1.125, + "step": 640 + }, + { + "epoch": 0.7024657534246576, + "grad_norm": 2.2437307834625244, + "learning_rate": 4.283103168155971e-06, + "loss": 1.9075, + "step": 641 + }, + { + "epoch": 0.7035616438356165, + "grad_norm": 2.320880174636841, + "learning_rate": 4.281072298943948e-06, + "loss": 1.8929, + "step": 642 + }, + { + "epoch": 0.7046575342465753, + "grad_norm": 2.1174306869506836, + "learning_rate": 4.279041429731926e-06, + "loss": 1.8872, + "step": 643 + }, + { + "epoch": 0.7057534246575342, + "grad_norm": 2.2826361656188965, + "learning_rate": 4.277010560519903e-06, + "loss": 1.9287, + "step": 644 + }, + { + "epoch": 0.7068493150684931, + "grad_norm": 2.7543559074401855, + "learning_rate": 4.27497969130788e-06, + "loss": 1.9119, + "step": 645 + }, + { + "epoch": 0.707945205479452, + "grad_norm": 2.2582201957702637, + "learning_rate": 4.2729488220958575e-06, + "loss": 1.9118, + "step": 646 + }, + { + "epoch": 0.709041095890411, + "grad_norm": 2.3121371269226074, + "learning_rate": 4.270917952883835e-06, + "loss": 1.959, + "step": 647 + }, + { + "epoch": 0.7101369863013699, + "grad_norm": 2.419102430343628, + "learning_rate": 4.268887083671812e-06, + "loss": 1.9679, + "step": 648 + }, + { + "epoch": 0.7112328767123288, + "grad_norm": 2.285771369934082, + "learning_rate": 4.266856214459789e-06, + "loss": 1.8939, + "step": 649 + }, + { + "epoch": 0.7123287671232876, + "grad_norm": 2.193603754043579, + "learning_rate": 4.264825345247766e-06, + "loss": 2.0097, + "step": 650 + }, + { + "epoch": 0.7123287671232876, + "eval_loss": 2.0785164833068848, + "eval_runtime": 37.2595, + "eval_samples_per_second": 26.839, + "eval_steps_per_second": 1.127, + "step": 650 + }, + { + "epoch": 0.7134246575342466, + "grad_norm": 2.2645511627197266, + "learning_rate": 4.262794476035744e-06, + "loss": 1.8865, + "step": 651 + }, + { + "epoch": 0.7145205479452055, + "grad_norm": 2.5553910732269287, + "learning_rate": 4.26076360682372e-06, + "loss": 1.9331, + "step": 652 + }, + { + "epoch": 0.7156164383561644, + "grad_norm": 2.2770845890045166, + "learning_rate": 4.2587327376116984e-06, + "loss": 1.847, + "step": 653 + }, + { + "epoch": 0.7167123287671233, + "grad_norm": 2.1486358642578125, + "learning_rate": 4.256701868399675e-06, + "loss": 1.9058, + "step": 654 + }, + { + "epoch": 0.7178082191780822, + "grad_norm": 2.168975353240967, + "learning_rate": 4.254670999187653e-06, + "loss": 1.9408, + "step": 655 + }, + { + "epoch": 0.718904109589041, + "grad_norm": 2.2156167030334473, + "learning_rate": 4.252640129975629e-06, + "loss": 1.932, + "step": 656 + }, + { + "epoch": 0.72, + "grad_norm": 2.148214101791382, + "learning_rate": 4.2506092607636075e-06, + "loss": 1.9327, + "step": 657 + }, + { + "epoch": 0.7210958904109589, + "grad_norm": 2.3860998153686523, + "learning_rate": 4.248578391551584e-06, + "loss": 1.9191, + "step": 658 + }, + { + "epoch": 0.7221917808219178, + "grad_norm": 2.3917558193206787, + "learning_rate": 4.246547522339562e-06, + "loss": 1.8607, + "step": 659 + }, + { + "epoch": 0.7232876712328767, + "grad_norm": 2.3024795055389404, + "learning_rate": 4.2445166531275385e-06, + "loss": 1.9083, + "step": 660 + }, + { + "epoch": 0.7232876712328767, + "eval_loss": 2.07985782623291, + "eval_runtime": 37.1583, + "eval_samples_per_second": 26.912, + "eval_steps_per_second": 1.13, + "step": 660 + }, + { + "epoch": 0.7243835616438357, + "grad_norm": 2.261615037918091, + "learning_rate": 4.242485783915517e-06, + "loss": 1.9155, + "step": 661 + }, + { + "epoch": 0.7254794520547945, + "grad_norm": 2.190124034881592, + "learning_rate": 4.240454914703493e-06, + "loss": 1.858, + "step": 662 + }, + { + "epoch": 0.7265753424657534, + "grad_norm": 2.3396036624908447, + "learning_rate": 4.23842404549147e-06, + "loss": 1.878, + "step": 663 + }, + { + "epoch": 0.7276712328767123, + "grad_norm": 2.198410987854004, + "learning_rate": 4.236393176279448e-06, + "loss": 1.8755, + "step": 664 + }, + { + "epoch": 0.7287671232876712, + "grad_norm": 2.108875274658203, + "learning_rate": 4.234362307067425e-06, + "loss": 1.8709, + "step": 665 + }, + { + "epoch": 0.7298630136986302, + "grad_norm": 2.0908193588256836, + "learning_rate": 4.232331437855402e-06, + "loss": 1.9395, + "step": 666 + }, + { + "epoch": 0.7309589041095891, + "grad_norm": 2.208134651184082, + "learning_rate": 4.2303005686433795e-06, + "loss": 1.9675, + "step": 667 + }, + { + "epoch": 0.7320547945205479, + "grad_norm": 2.2779526710510254, + "learning_rate": 4.228269699431357e-06, + "loss": 1.9661, + "step": 668 + }, + { + "epoch": 0.7331506849315068, + "grad_norm": 2.053279161453247, + "learning_rate": 4.226238830219334e-06, + "loss": 1.9244, + "step": 669 + }, + { + "epoch": 0.7342465753424657, + "grad_norm": 2.2060821056365967, + "learning_rate": 4.224207961007311e-06, + "loss": 1.9252, + "step": 670 + }, + { + "epoch": 0.7342465753424657, + "eval_loss": 2.0804648399353027, + "eval_runtime": 37.1852, + "eval_samples_per_second": 26.892, + "eval_steps_per_second": 1.129, + "step": 670 + }, + { + "epoch": 0.7353424657534247, + "grad_norm": 2.1996090412139893, + "learning_rate": 4.2221770917952886e-06, + "loss": 1.926, + "step": 671 + }, + { + "epoch": 0.7364383561643836, + "grad_norm": 2.0817034244537354, + "learning_rate": 4.220146222583266e-06, + "loss": 1.9123, + "step": 672 + }, + { + "epoch": 0.7375342465753425, + "grad_norm": 2.2505035400390625, + "learning_rate": 4.218115353371243e-06, + "loss": 1.9369, + "step": 673 + }, + { + "epoch": 0.7386301369863014, + "grad_norm": 2.2053000926971436, + "learning_rate": 4.21608448415922e-06, + "loss": 1.9225, + "step": 674 + }, + { + "epoch": 0.7397260273972602, + "grad_norm": 2.039905071258545, + "learning_rate": 4.214053614947198e-06, + "loss": 1.8876, + "step": 675 + }, + { + "epoch": 0.7408219178082192, + "grad_norm": 2.047445297241211, + "learning_rate": 4.212022745735175e-06, + "loss": 1.9724, + "step": 676 + }, + { + "epoch": 0.7419178082191781, + "grad_norm": 2.277803897857666, + "learning_rate": 4.209991876523152e-06, + "loss": 1.8687, + "step": 677 + }, + { + "epoch": 0.743013698630137, + "grad_norm": 2.0048105716705322, + "learning_rate": 4.2079610073111295e-06, + "loss": 1.9349, + "step": 678 + }, + { + "epoch": 0.7441095890410959, + "grad_norm": 2.121614456176758, + "learning_rate": 4.205930138099107e-06, + "loss": 1.8723, + "step": 679 + }, + { + "epoch": 0.7452054794520548, + "grad_norm": 2.169165849685669, + "learning_rate": 4.203899268887084e-06, + "loss": 1.9713, + "step": 680 + }, + { + "epoch": 0.7452054794520548, + "eval_loss": 2.0704824924468994, + "eval_runtime": 37.2015, + "eval_samples_per_second": 26.881, + "eval_steps_per_second": 1.129, + "step": 680 + }, + { + "epoch": 0.7463013698630137, + "grad_norm": 2.004974603652954, + "learning_rate": 4.201868399675061e-06, + "loss": 1.9237, + "step": 681 + }, + { + "epoch": 0.7473972602739726, + "grad_norm": 2.042402505874634, + "learning_rate": 4.199837530463039e-06, + "loss": 1.9686, + "step": 682 + }, + { + "epoch": 0.7484931506849315, + "grad_norm": 2.152040719985962, + "learning_rate": 4.197806661251016e-06, + "loss": 1.8498, + "step": 683 + }, + { + "epoch": 0.7495890410958904, + "grad_norm": 2.116483211517334, + "learning_rate": 4.195775792038993e-06, + "loss": 1.9241, + "step": 684 + }, + { + "epoch": 0.7506849315068493, + "grad_norm": 2.1628384590148926, + "learning_rate": 4.19374492282697e-06, + "loss": 1.9019, + "step": 685 + }, + { + "epoch": 0.7517808219178083, + "grad_norm": 2.1851415634155273, + "learning_rate": 4.191714053614948e-06, + "loss": 1.836, + "step": 686 + }, + { + "epoch": 0.7528767123287671, + "grad_norm": 2.173215866088867, + "learning_rate": 4.189683184402924e-06, + "loss": 1.9412, + "step": 687 + }, + { + "epoch": 0.753972602739726, + "grad_norm": 2.1002466678619385, + "learning_rate": 4.187652315190902e-06, + "loss": 1.8871, + "step": 688 + }, + { + "epoch": 0.7550684931506849, + "grad_norm": 2.1932122707366943, + "learning_rate": 4.185621445978879e-06, + "loss": 1.9671, + "step": 689 + }, + { + "epoch": 0.7561643835616438, + "grad_norm": 2.2915258407592773, + "learning_rate": 4.183590576766857e-06, + "loss": 1.8957, + "step": 690 + }, + { + "epoch": 0.7561643835616438, + "eval_loss": 2.067124128341675, + "eval_runtime": 37.2362, + "eval_samples_per_second": 26.856, + "eval_steps_per_second": 1.128, + "step": 690 + }, + { + "epoch": 0.7572602739726028, + "grad_norm": 2.060879707336426, + "learning_rate": 4.181559707554833e-06, + "loss": 1.9091, + "step": 691 + }, + { + "epoch": 0.7583561643835617, + "grad_norm": 2.031156539916992, + "learning_rate": 4.179528838342811e-06, + "loss": 1.9606, + "step": 692 + }, + { + "epoch": 0.7594520547945205, + "grad_norm": 2.121628999710083, + "learning_rate": 4.177497969130788e-06, + "loss": 1.88, + "step": 693 + }, + { + "epoch": 0.7605479452054794, + "grad_norm": 2.2009618282318115, + "learning_rate": 4.175467099918766e-06, + "loss": 1.8916, + "step": 694 + }, + { + "epoch": 0.7616438356164383, + "grad_norm": 2.124082326889038, + "learning_rate": 4.173436230706742e-06, + "loss": 1.9805, + "step": 695 + }, + { + "epoch": 0.7627397260273973, + "grad_norm": 2.3914906978607178, + "learning_rate": 4.1714053614947205e-06, + "loss": 1.9052, + "step": 696 + }, + { + "epoch": 0.7638356164383562, + "grad_norm": 2.2398853302001953, + "learning_rate": 4.169374492282697e-06, + "loss": 1.8965, + "step": 697 + }, + { + "epoch": 0.7649315068493151, + "grad_norm": 2.224860191345215, + "learning_rate": 4.167343623070675e-06, + "loss": 1.8962, + "step": 698 + }, + { + "epoch": 0.766027397260274, + "grad_norm": 2.3356244564056396, + "learning_rate": 4.165312753858652e-06, + "loss": 1.9543, + "step": 699 + }, + { + "epoch": 0.7671232876712328, + "grad_norm": 2.1308751106262207, + "learning_rate": 4.163281884646629e-06, + "loss": 1.9095, + "step": 700 + }, + { + "epoch": 0.7671232876712328, + "eval_loss": 2.066145896911621, + "eval_runtime": 37.2629, + "eval_samples_per_second": 26.836, + "eval_steps_per_second": 1.127, + "step": 700 + }, + { + "epoch": 0.7682191780821918, + "grad_norm": 2.1655938625335693, + "learning_rate": 4.161251015434607e-06, + "loss": 1.8923, + "step": 701 + }, + { + "epoch": 0.7693150684931507, + "grad_norm": 2.1985316276550293, + "learning_rate": 4.159220146222583e-06, + "loss": 1.9216, + "step": 702 + }, + { + "epoch": 0.7704109589041096, + "grad_norm": 2.1537251472473145, + "learning_rate": 4.157189277010561e-06, + "loss": 1.879, + "step": 703 + }, + { + "epoch": 0.7715068493150685, + "grad_norm": 2.1867849826812744, + "learning_rate": 4.155158407798538e-06, + "loss": 1.905, + "step": 704 + }, + { + "epoch": 0.7726027397260274, + "grad_norm": 2.342320203781128, + "learning_rate": 4.153127538586516e-06, + "loss": 1.9403, + "step": 705 + }, + { + "epoch": 0.7736986301369863, + "grad_norm": 2.2250113487243652, + "learning_rate": 4.151096669374492e-06, + "loss": 1.9109, + "step": 706 + }, + { + "epoch": 0.7747945205479452, + "grad_norm": 2.179103374481201, + "learning_rate": 4.1490658001624705e-06, + "loss": 1.9288, + "step": 707 + }, + { + "epoch": 0.7758904109589041, + "grad_norm": 2.3618645668029785, + "learning_rate": 4.147034930950447e-06, + "loss": 1.8982, + "step": 708 + }, + { + "epoch": 0.776986301369863, + "grad_norm": 2.150402307510376, + "learning_rate": 4.145004061738425e-06, + "loss": 1.9, + "step": 709 + }, + { + "epoch": 0.7780821917808219, + "grad_norm": 2.295571804046631, + "learning_rate": 4.1429731925264015e-06, + "loss": 1.9003, + "step": 710 + }, + { + "epoch": 0.7780821917808219, + "eval_loss": 2.067948341369629, + "eval_runtime": 37.2379, + "eval_samples_per_second": 26.854, + "eval_steps_per_second": 1.128, + "step": 710 + }, + { + "epoch": 0.7791780821917809, + "grad_norm": 2.3279545307159424, + "learning_rate": 4.140942323314379e-06, + "loss": 1.8307, + "step": 711 + }, + { + "epoch": 0.7802739726027397, + "grad_norm": 2.3993380069732666, + "learning_rate": 4.138911454102356e-06, + "loss": 1.8755, + "step": 712 + }, + { + "epoch": 0.7813698630136986, + "grad_norm": 2.139653444290161, + "learning_rate": 4.136880584890333e-06, + "loss": 1.8914, + "step": 713 + }, + { + "epoch": 0.7824657534246575, + "grad_norm": 2.239544153213501, + "learning_rate": 4.1348497156783106e-06, + "loss": 1.9289, + "step": 714 + }, + { + "epoch": 0.7835616438356164, + "grad_norm": 2.2039642333984375, + "learning_rate": 4.132818846466288e-06, + "loss": 1.9237, + "step": 715 + }, + { + "epoch": 0.7846575342465754, + "grad_norm": 2.1453804969787598, + "learning_rate": 4.130787977254265e-06, + "loss": 1.9291, + "step": 716 + }, + { + "epoch": 0.7857534246575343, + "grad_norm": 2.2350032329559326, + "learning_rate": 4.128757108042242e-06, + "loss": 1.9413, + "step": 717 + }, + { + "epoch": 0.7868493150684932, + "grad_norm": 2.1718318462371826, + "learning_rate": 4.12672623883022e-06, + "loss": 1.9318, + "step": 718 + }, + { + "epoch": 0.787945205479452, + "grad_norm": 2.1860601902008057, + "learning_rate": 4.124695369618197e-06, + "loss": 1.9934, + "step": 719 + }, + { + "epoch": 0.7890410958904109, + "grad_norm": 2.205166816711426, + "learning_rate": 4.122664500406174e-06, + "loss": 1.9038, + "step": 720 + }, + { + "epoch": 0.7890410958904109, + "eval_loss": 2.06229829788208, + "eval_runtime": 37.6121, + "eval_samples_per_second": 26.587, + "eval_steps_per_second": 1.117, + "step": 720 + }, + { + "epoch": 0.7901369863013699, + "grad_norm": 2.2094533443450928, + "learning_rate": 4.1206336311941515e-06, + "loss": 1.906, + "step": 721 + }, + { + "epoch": 0.7912328767123288, + "grad_norm": 2.178891658782959, + "learning_rate": 4.118602761982129e-06, + "loss": 1.936, + "step": 722 + }, + { + "epoch": 0.7923287671232877, + "grad_norm": 2.051077365875244, + "learning_rate": 4.116571892770106e-06, + "loss": 1.8736, + "step": 723 + }, + { + "epoch": 0.7934246575342466, + "grad_norm": 2.2430427074432373, + "learning_rate": 4.114541023558083e-06, + "loss": 1.8396, + "step": 724 + }, + { + "epoch": 0.7945205479452054, + "grad_norm": 2.2255303859710693, + "learning_rate": 4.112510154346061e-06, + "loss": 1.9668, + "step": 725 + }, + { + "epoch": 0.7956164383561644, + "grad_norm": 2.280313491821289, + "learning_rate": 4.110479285134038e-06, + "loss": 1.8945, + "step": 726 + }, + { + "epoch": 0.7967123287671233, + "grad_norm": 2.2516801357269287, + "learning_rate": 4.108448415922015e-06, + "loss": 1.9157, + "step": 727 + }, + { + "epoch": 0.7978082191780822, + "grad_norm": 2.1379809379577637, + "learning_rate": 4.1064175467099924e-06, + "loss": 1.935, + "step": 728 + }, + { + "epoch": 0.7989041095890411, + "grad_norm": 2.149425983428955, + "learning_rate": 4.10438667749797e-06, + "loss": 1.935, + "step": 729 + }, + { + "epoch": 0.8, + "grad_norm": 2.275270700454712, + "learning_rate": 4.102355808285947e-06, + "loss": 1.7889, + "step": 730 + }, + { + "epoch": 0.8, + "eval_loss": 2.0633974075317383, + "eval_runtime": 37.6895, + "eval_samples_per_second": 26.533, + "eval_steps_per_second": 1.114, + "step": 730 + }, + { + "epoch": 0.8010958904109589, + "grad_norm": 2.1550278663635254, + "learning_rate": 4.100324939073924e-06, + "loss": 1.9833, + "step": 731 + }, + { + "epoch": 0.8021917808219178, + "grad_norm": 2.3050060272216797, + "learning_rate": 4.0982940698619015e-06, + "loss": 1.8969, + "step": 732 + }, + { + "epoch": 0.8032876712328767, + "grad_norm": 2.4548778533935547, + "learning_rate": 4.096263200649879e-06, + "loss": 1.8945, + "step": 733 + }, + { + "epoch": 0.8043835616438356, + "grad_norm": 2.2611639499664307, + "learning_rate": 4.094232331437856e-06, + "loss": 1.9071, + "step": 734 + }, + { + "epoch": 0.8054794520547945, + "grad_norm": 2.1806774139404297, + "learning_rate": 4.092201462225833e-06, + "loss": 1.8787, + "step": 735 + }, + { + "epoch": 0.8065753424657535, + "grad_norm": 2.4374301433563232, + "learning_rate": 4.090170593013811e-06, + "loss": 1.88, + "step": 736 + }, + { + "epoch": 0.8076712328767123, + "grad_norm": 2.328333854675293, + "learning_rate": 4.088139723801787e-06, + "loss": 1.9178, + "step": 737 + }, + { + "epoch": 0.8087671232876712, + "grad_norm": 2.33552885055542, + "learning_rate": 4.086108854589765e-06, + "loss": 1.8587, + "step": 738 + }, + { + "epoch": 0.8098630136986301, + "grad_norm": 2.3472790718078613, + "learning_rate": 4.084077985377742e-06, + "loss": 1.877, + "step": 739 + }, + { + "epoch": 0.810958904109589, + "grad_norm": 2.307123899459839, + "learning_rate": 4.08204711616572e-06, + "loss": 1.8888, + "step": 740 + }, + { + "epoch": 0.810958904109589, + "eval_loss": 2.065826892852783, + "eval_runtime": 37.2747, + "eval_samples_per_second": 26.828, + "eval_steps_per_second": 1.127, + "step": 740 + }, + { + "epoch": 0.812054794520548, + "grad_norm": 2.3688747882843018, + "learning_rate": 4.080016246953696e-06, + "loss": 1.9571, + "step": 741 + }, + { + "epoch": 0.8131506849315069, + "grad_norm": 2.385007381439209, + "learning_rate": 4.077985377741674e-06, + "loss": 1.8657, + "step": 742 + }, + { + "epoch": 0.8142465753424658, + "grad_norm": 2.156766414642334, + "learning_rate": 4.075954508529651e-06, + "loss": 1.9421, + "step": 743 + }, + { + "epoch": 0.8153424657534246, + "grad_norm": 2.2074286937713623, + "learning_rate": 4.073923639317629e-06, + "loss": 1.9396, + "step": 744 + }, + { + "epoch": 0.8164383561643835, + "grad_norm": 2.185458183288574, + "learning_rate": 4.071892770105605e-06, + "loss": 1.9612, + "step": 745 + }, + { + "epoch": 0.8175342465753425, + "grad_norm": 2.324209451675415, + "learning_rate": 4.069861900893583e-06, + "loss": 1.892, + "step": 746 + }, + { + "epoch": 0.8186301369863014, + "grad_norm": 2.1312127113342285, + "learning_rate": 4.06783103168156e-06, + "loss": 1.8522, + "step": 747 + }, + { + "epoch": 0.8197260273972603, + "grad_norm": 2.0959460735321045, + "learning_rate": 4.065800162469538e-06, + "loss": 1.8732, + "step": 748 + }, + { + "epoch": 0.8208219178082192, + "grad_norm": 2.085519552230835, + "learning_rate": 4.063769293257514e-06, + "loss": 1.8832, + "step": 749 + }, + { + "epoch": 0.821917808219178, + "grad_norm": 2.310145854949951, + "learning_rate": 4.061738424045492e-06, + "loss": 1.919, + "step": 750 + }, + { + "epoch": 0.821917808219178, + "eval_loss": 2.0637967586517334, + "eval_runtime": 37.2246, + "eval_samples_per_second": 26.864, + "eval_steps_per_second": 1.128, + "step": 750 + }, + { + "epoch": 0.823013698630137, + "grad_norm": 2.3171160221099854, + "learning_rate": 4.059707554833469e-06, + "loss": 1.8874, + "step": 751 + }, + { + "epoch": 0.8241095890410959, + "grad_norm": 2.1065480709075928, + "learning_rate": 4.057676685621446e-06, + "loss": 1.9064, + "step": 752 + }, + { + "epoch": 0.8252054794520548, + "grad_norm": 2.0539677143096924, + "learning_rate": 4.0556458164094235e-06, + "loss": 1.9376, + "step": 753 + }, + { + "epoch": 0.8263013698630137, + "grad_norm": 2.4440360069274902, + "learning_rate": 4.053614947197401e-06, + "loss": 1.9129, + "step": 754 + }, + { + "epoch": 0.8273972602739726, + "grad_norm": 2.2324604988098145, + "learning_rate": 4.051584077985378e-06, + "loss": 1.8281, + "step": 755 + }, + { + "epoch": 0.8284931506849315, + "grad_norm": 2.262627363204956, + "learning_rate": 4.049553208773355e-06, + "loss": 1.9036, + "step": 756 + }, + { + "epoch": 0.8295890410958904, + "grad_norm": 2.2473418712615967, + "learning_rate": 4.047522339561333e-06, + "loss": 1.9455, + "step": 757 + }, + { + "epoch": 0.8306849315068493, + "grad_norm": 2.473444938659668, + "learning_rate": 4.04549147034931e-06, + "loss": 1.9003, + "step": 758 + }, + { + "epoch": 0.8317808219178082, + "grad_norm": 2.3454110622406006, + "learning_rate": 4.043460601137287e-06, + "loss": 1.8941, + "step": 759 + }, + { + "epoch": 0.8328767123287671, + "grad_norm": 2.4391231536865234, + "learning_rate": 4.041429731925264e-06, + "loss": 2.0146, + "step": 760 + }, + { + "epoch": 0.8328767123287671, + "eval_loss": 2.0618460178375244, + "eval_runtime": 37.1704, + "eval_samples_per_second": 26.903, + "eval_steps_per_second": 1.13, + "step": 760 + }, + { + "epoch": 0.8339726027397261, + "grad_norm": 2.3129498958587646, + "learning_rate": 4.039398862713242e-06, + "loss": 1.9, + "step": 761 + }, + { + "epoch": 0.8350684931506849, + "grad_norm": 2.3766934871673584, + "learning_rate": 4.037367993501219e-06, + "loss": 1.9017, + "step": 762 + }, + { + "epoch": 0.8361643835616438, + "grad_norm": 2.1698460578918457, + "learning_rate": 4.035337124289196e-06, + "loss": 1.9657, + "step": 763 + }, + { + "epoch": 0.8372602739726027, + "grad_norm": 2.384904146194458, + "learning_rate": 4.0333062550771735e-06, + "loss": 1.8867, + "step": 764 + }, + { + "epoch": 0.8383561643835616, + "grad_norm": 2.572391986846924, + "learning_rate": 4.031275385865151e-06, + "loss": 1.8561, + "step": 765 + }, + { + "epoch": 0.8394520547945206, + "grad_norm": 2.216606616973877, + "learning_rate": 4.029244516653128e-06, + "loss": 1.8926, + "step": 766 + }, + { + "epoch": 0.8405479452054795, + "grad_norm": 2.215756893157959, + "learning_rate": 4.027213647441105e-06, + "loss": 1.928, + "step": 767 + }, + { + "epoch": 0.8416438356164384, + "grad_norm": 2.2713284492492676, + "learning_rate": 4.025182778229083e-06, + "loss": 1.8286, + "step": 768 + }, + { + "epoch": 0.8427397260273972, + "grad_norm": 2.341747283935547, + "learning_rate": 4.02315190901706e-06, + "loss": 1.8789, + "step": 769 + }, + { + "epoch": 0.8438356164383561, + "grad_norm": 2.1485848426818848, + "learning_rate": 4.021121039805037e-06, + "loss": 1.8998, + "step": 770 + }, + { + "epoch": 0.8438356164383561, + "eval_loss": 2.0596234798431396, + "eval_runtime": 37.1854, + "eval_samples_per_second": 26.892, + "eval_steps_per_second": 1.129, + "step": 770 + }, + { + "epoch": 0.8449315068493151, + "grad_norm": 2.234192132949829, + "learning_rate": 4.0190901705930144e-06, + "loss": 1.8995, + "step": 771 + }, + { + "epoch": 0.846027397260274, + "grad_norm": 2.289328098297119, + "learning_rate": 4.017059301380992e-06, + "loss": 1.9882, + "step": 772 + }, + { + "epoch": 0.8471232876712329, + "grad_norm": 2.1001455783843994, + "learning_rate": 4.015028432168969e-06, + "loss": 1.8706, + "step": 773 + }, + { + "epoch": 0.8482191780821918, + "grad_norm": 2.2200522422790527, + "learning_rate": 4.012997562956945e-06, + "loss": 1.9443, + "step": 774 + }, + { + "epoch": 0.8493150684931506, + "grad_norm": 2.2540805339813232, + "learning_rate": 4.0109666937449235e-06, + "loss": 1.8552, + "step": 775 + }, + { + "epoch": 0.8504109589041096, + "grad_norm": 2.1214053630828857, + "learning_rate": 4.0089358245329e-06, + "loss": 1.8741, + "step": 776 + }, + { + "epoch": 0.8515068493150685, + "grad_norm": 2.0221869945526123, + "learning_rate": 4.006904955320878e-06, + "loss": 1.8798, + "step": 777 + }, + { + "epoch": 0.8526027397260274, + "grad_norm": 2.217261791229248, + "learning_rate": 4.0048740861088545e-06, + "loss": 1.8542, + "step": 778 + }, + { + "epoch": 0.8536986301369863, + "grad_norm": 2.099202871322632, + "learning_rate": 4.002843216896833e-06, + "loss": 1.9342, + "step": 779 + }, + { + "epoch": 0.8547945205479452, + "grad_norm": 2.139526844024658, + "learning_rate": 4.000812347684809e-06, + "loss": 1.9252, + "step": 780 + }, + { + "epoch": 0.8547945205479452, + "eval_loss": 2.054051399230957, + "eval_runtime": 37.1125, + "eval_samples_per_second": 26.945, + "eval_steps_per_second": 1.132, + "step": 780 + }, + { + "epoch": 0.8558904109589041, + "grad_norm": 2.2231476306915283, + "learning_rate": 3.998781478472787e-06, + "loss": 1.8671, + "step": 781 + }, + { + "epoch": 0.856986301369863, + "grad_norm": 2.143979072570801, + "learning_rate": 3.996750609260764e-06, + "loss": 1.8759, + "step": 782 + }, + { + "epoch": 0.8580821917808219, + "grad_norm": 2.1353135108947754, + "learning_rate": 3.994719740048742e-06, + "loss": 1.8559, + "step": 783 + }, + { + "epoch": 0.8591780821917808, + "grad_norm": 2.222564697265625, + "learning_rate": 3.992688870836718e-06, + "loss": 1.999, + "step": 784 + }, + { + "epoch": 0.8602739726027397, + "grad_norm": 1.958031177520752, + "learning_rate": 3.990658001624696e-06, + "loss": 1.9019, + "step": 785 + }, + { + "epoch": 0.8613698630136987, + "grad_norm": 2.2048962116241455, + "learning_rate": 3.988627132412673e-06, + "loss": 1.8561, + "step": 786 + }, + { + "epoch": 0.8624657534246576, + "grad_norm": 2.150332450866699, + "learning_rate": 3.98659626320065e-06, + "loss": 1.9399, + "step": 787 + }, + { + "epoch": 0.8635616438356164, + "grad_norm": 2.182742118835449, + "learning_rate": 3.984565393988627e-06, + "loss": 1.8933, + "step": 788 + }, + { + "epoch": 0.8646575342465753, + "grad_norm": 2.19435715675354, + "learning_rate": 3.9825345247766046e-06, + "loss": 1.8918, + "step": 789 + }, + { + "epoch": 0.8657534246575342, + "grad_norm": 2.3382859230041504, + "learning_rate": 3.980503655564582e-06, + "loss": 1.8937, + "step": 790 + }, + { + "epoch": 0.8657534246575342, + "eval_loss": 2.051997423171997, + "eval_runtime": 37.1866, + "eval_samples_per_second": 26.891, + "eval_steps_per_second": 1.129, + "step": 790 + }, + { + "epoch": 0.8668493150684932, + "grad_norm": 2.2700726985931396, + "learning_rate": 3.978472786352559e-06, + "loss": 1.9148, + "step": 791 + }, + { + "epoch": 0.8679452054794521, + "grad_norm": 2.2998759746551514, + "learning_rate": 3.976441917140536e-06, + "loss": 1.9268, + "step": 792 + }, + { + "epoch": 0.869041095890411, + "grad_norm": 2.2367377281188965, + "learning_rate": 3.974411047928514e-06, + "loss": 1.901, + "step": 793 + }, + { + "epoch": 0.8701369863013698, + "grad_norm": 2.0868773460388184, + "learning_rate": 3.972380178716491e-06, + "loss": 1.9677, + "step": 794 + }, + { + "epoch": 0.8712328767123287, + "grad_norm": 2.183732509613037, + "learning_rate": 3.970349309504468e-06, + "loss": 1.9271, + "step": 795 + }, + { + "epoch": 0.8723287671232877, + "grad_norm": 2.187282085418701, + "learning_rate": 3.9683184402924455e-06, + "loss": 1.9437, + "step": 796 + }, + { + "epoch": 0.8734246575342466, + "grad_norm": 2.1906070709228516, + "learning_rate": 3.966287571080423e-06, + "loss": 1.8437, + "step": 797 + }, + { + "epoch": 0.8745205479452055, + "grad_norm": 2.16979718208313, + "learning_rate": 3.9642567018684e-06, + "loss": 1.829, + "step": 798 + }, + { + "epoch": 0.8756164383561644, + "grad_norm": 2.2387869358062744, + "learning_rate": 3.962225832656377e-06, + "loss": 1.9143, + "step": 799 + }, + { + "epoch": 0.8767123287671232, + "grad_norm": 2.208221197128296, + "learning_rate": 3.960194963444355e-06, + "loss": 1.8894, + "step": 800 + }, + { + "epoch": 0.8767123287671232, + "eval_loss": 2.047945976257324, + "eval_runtime": 37.1778, + "eval_samples_per_second": 26.898, + "eval_steps_per_second": 1.13, + "step": 800 + }, + { + "epoch": 0.8778082191780822, + "grad_norm": 2.335108518600464, + "learning_rate": 3.958164094232332e-06, + "loss": 1.8155, + "step": 801 + }, + { + "epoch": 0.8789041095890411, + "grad_norm": 2.168006181716919, + "learning_rate": 3.956133225020309e-06, + "loss": 1.8919, + "step": 802 + }, + { + "epoch": 0.88, + "grad_norm": 2.3029897212982178, + "learning_rate": 3.954102355808286e-06, + "loss": 1.8996, + "step": 803 + }, + { + "epoch": 0.8810958904109589, + "grad_norm": 2.156141519546509, + "learning_rate": 3.952071486596264e-06, + "loss": 1.8965, + "step": 804 + }, + { + "epoch": 0.8821917808219178, + "grad_norm": 2.1522068977355957, + "learning_rate": 3.950040617384241e-06, + "loss": 1.8922, + "step": 805 + }, + { + "epoch": 0.8832876712328767, + "grad_norm": 2.153430700302124, + "learning_rate": 3.948009748172218e-06, + "loss": 1.8839, + "step": 806 + }, + { + "epoch": 0.8843835616438356, + "grad_norm": 2.1891674995422363, + "learning_rate": 3.9459788789601955e-06, + "loss": 1.8756, + "step": 807 + }, + { + "epoch": 0.8854794520547945, + "grad_norm": 2.1092092990875244, + "learning_rate": 3.943948009748173e-06, + "loss": 1.9452, + "step": 808 + }, + { + "epoch": 0.8865753424657534, + "grad_norm": 2.0465025901794434, + "learning_rate": 3.94191714053615e-06, + "loss": 1.9067, + "step": 809 + }, + { + "epoch": 0.8876712328767123, + "grad_norm": 2.1150474548339844, + "learning_rate": 3.939886271324127e-06, + "loss": 1.8874, + "step": 810 + }, + { + "epoch": 0.8876712328767123, + "eval_loss": 2.046292304992676, + "eval_runtime": 37.2421, + "eval_samples_per_second": 26.851, + "eval_steps_per_second": 1.128, + "step": 810 + }, + { + "epoch": 0.8887671232876713, + "grad_norm": 2.039989948272705, + "learning_rate": 3.937855402112104e-06, + "loss": 1.9212, + "step": 811 + }, + { + "epoch": 0.8898630136986302, + "grad_norm": 2.17354679107666, + "learning_rate": 3.935824532900082e-06, + "loss": 1.8473, + "step": 812 + }, + { + "epoch": 0.890958904109589, + "grad_norm": 2.3101940155029297, + "learning_rate": 3.933793663688058e-06, + "loss": 1.8413, + "step": 813 + }, + { + "epoch": 0.8920547945205479, + "grad_norm": 2.1300928592681885, + "learning_rate": 3.9317627944760365e-06, + "loss": 1.9126, + "step": 814 + }, + { + "epoch": 0.8931506849315068, + "grad_norm": 2.2317357063293457, + "learning_rate": 3.929731925264013e-06, + "loss": 1.938, + "step": 815 + }, + { + "epoch": 0.8942465753424658, + "grad_norm": 2.1097419261932373, + "learning_rate": 3.927701056051991e-06, + "loss": 1.8426, + "step": 816 + }, + { + "epoch": 0.8953424657534247, + "grad_norm": 2.1709141731262207, + "learning_rate": 3.9256701868399674e-06, + "loss": 1.8779, + "step": 817 + }, + { + "epoch": 0.8964383561643836, + "grad_norm": 2.104193925857544, + "learning_rate": 3.9236393176279456e-06, + "loss": 1.8736, + "step": 818 + }, + { + "epoch": 0.8975342465753424, + "grad_norm": 2.094637632369995, + "learning_rate": 3.921608448415922e-06, + "loss": 1.8556, + "step": 819 + }, + { + "epoch": 0.8986301369863013, + "grad_norm": 2.217918872833252, + "learning_rate": 3.9195775792039e-06, + "loss": 1.8429, + "step": 820 + }, + { + "epoch": 0.8986301369863013, + "eval_loss": 2.046095848083496, + "eval_runtime": 37.5659, + "eval_samples_per_second": 26.62, + "eval_steps_per_second": 1.118, + "step": 820 + }, + { + "epoch": 0.8997260273972603, + "grad_norm": 2.220355272293091, + "learning_rate": 3.9175467099918765e-06, + "loss": 1.8973, + "step": 821 + }, + { + "epoch": 0.9008219178082192, + "grad_norm": 2.0766689777374268, + "learning_rate": 3.915515840779855e-06, + "loss": 1.8728, + "step": 822 + }, + { + "epoch": 0.9019178082191781, + "grad_norm": 2.2046821117401123, + "learning_rate": 3.913484971567831e-06, + "loss": 1.9211, + "step": 823 + }, + { + "epoch": 0.903013698630137, + "grad_norm": 2.286024332046509, + "learning_rate": 3.911454102355808e-06, + "loss": 1.913, + "step": 824 + }, + { + "epoch": 0.9041095890410958, + "grad_norm": 2.2495479583740234, + "learning_rate": 3.909423233143786e-06, + "loss": 1.8214, + "step": 825 + }, + { + "epoch": 0.9052054794520548, + "grad_norm": 2.1300203800201416, + "learning_rate": 3.907392363931763e-06, + "loss": 1.9033, + "step": 826 + }, + { + "epoch": 0.9063013698630137, + "grad_norm": 2.092094898223877, + "learning_rate": 3.90536149471974e-06, + "loss": 1.9007, + "step": 827 + }, + { + "epoch": 0.9073972602739726, + "grad_norm": 2.1571428775787354, + "learning_rate": 3.9033306255077175e-06, + "loss": 1.9606, + "step": 828 + }, + { + "epoch": 0.9084931506849315, + "grad_norm": 2.1052281856536865, + "learning_rate": 3.901299756295695e-06, + "loss": 1.9916, + "step": 829 + }, + { + "epoch": 0.9095890410958904, + "grad_norm": 2.113666534423828, + "learning_rate": 3.899268887083672e-06, + "loss": 2.0136, + "step": 830 + }, + { + "epoch": 0.9095890410958904, + "eval_loss": 2.0450072288513184, + "eval_runtime": 37.6746, + "eval_samples_per_second": 26.543, + "eval_steps_per_second": 1.115, + "step": 830 + }, + { + "epoch": 0.9106849315068493, + "grad_norm": 2.112591028213501, + "learning_rate": 3.897238017871649e-06, + "loss": 1.8784, + "step": 831 + }, + { + "epoch": 0.9117808219178082, + "grad_norm": 2.1584672927856445, + "learning_rate": 3.8952071486596266e-06, + "loss": 1.8876, + "step": 832 + }, + { + "epoch": 0.9128767123287671, + "grad_norm": 2.2744317054748535, + "learning_rate": 3.893176279447604e-06, + "loss": 1.8981, + "step": 833 + }, + { + "epoch": 0.913972602739726, + "grad_norm": 2.21824312210083, + "learning_rate": 3.891145410235581e-06, + "loss": 1.85, + "step": 834 + }, + { + "epoch": 0.915068493150685, + "grad_norm": 2.2203729152679443, + "learning_rate": 3.889114541023558e-06, + "loss": 1.9649, + "step": 835 + }, + { + "epoch": 0.9161643835616439, + "grad_norm": 2.1767687797546387, + "learning_rate": 3.887083671811536e-06, + "loss": 1.9044, + "step": 836 + }, + { + "epoch": 0.9172602739726028, + "grad_norm": 2.0724353790283203, + "learning_rate": 3.885052802599513e-06, + "loss": 1.8787, + "step": 837 + }, + { + "epoch": 0.9183561643835616, + "grad_norm": 2.1283740997314453, + "learning_rate": 3.88302193338749e-06, + "loss": 1.8284, + "step": 838 + }, + { + "epoch": 0.9194520547945205, + "grad_norm": 2.5087451934814453, + "learning_rate": 3.8809910641754675e-06, + "loss": 1.9546, + "step": 839 + }, + { + "epoch": 0.9205479452054794, + "grad_norm": 2.0633344650268555, + "learning_rate": 3.878960194963445e-06, + "loss": 1.8969, + "step": 840 + }, + { + "epoch": 0.9205479452054794, + "eval_loss": 2.047938346862793, + "eval_runtime": 37.2293, + "eval_samples_per_second": 26.861, + "eval_steps_per_second": 1.128, + "step": 840 + }, + { + "epoch": 0.9216438356164384, + "grad_norm": 2.178361654281616, + "learning_rate": 3.876929325751422e-06, + "loss": 1.8855, + "step": 841 + }, + { + "epoch": 0.9227397260273973, + "grad_norm": 2.3223721981048584, + "learning_rate": 3.874898456539399e-06, + "loss": 1.9363, + "step": 842 + }, + { + "epoch": 0.9238356164383562, + "grad_norm": 2.3330960273742676, + "learning_rate": 3.872867587327377e-06, + "loss": 1.8374, + "step": 843 + }, + { + "epoch": 0.924931506849315, + "grad_norm": 2.0584356784820557, + "learning_rate": 3.870836718115354e-06, + "loss": 1.911, + "step": 844 + }, + { + "epoch": 0.9260273972602739, + "grad_norm": 2.1834609508514404, + "learning_rate": 3.868805848903331e-06, + "loss": 1.8645, + "step": 845 + }, + { + "epoch": 0.9271232876712329, + "grad_norm": 2.2576122283935547, + "learning_rate": 3.8667749796913084e-06, + "loss": 1.8662, + "step": 846 + }, + { + "epoch": 0.9282191780821918, + "grad_norm": 2.1271183490753174, + "learning_rate": 3.864744110479286e-06, + "loss": 1.8935, + "step": 847 + }, + { + "epoch": 0.9293150684931507, + "grad_norm": 2.226727247238159, + "learning_rate": 3.862713241267262e-06, + "loss": 1.8969, + "step": 848 + }, + { + "epoch": 0.9304109589041096, + "grad_norm": 2.3301939964294434, + "learning_rate": 3.86068237205524e-06, + "loss": 1.9393, + "step": 849 + }, + { + "epoch": 0.9315068493150684, + "grad_norm": 2.180250883102417, + "learning_rate": 3.858651502843217e-06, + "loss": 1.9331, + "step": 850 + }, + { + "epoch": 0.9315068493150684, + "eval_loss": 2.0509681701660156, + "eval_runtime": 37.2111, + "eval_samples_per_second": 26.874, + "eval_steps_per_second": 1.129, + "step": 850 + }, + { + "epoch": 0.9326027397260274, + "grad_norm": 2.223548173904419, + "learning_rate": 3.856620633631195e-06, + "loss": 1.894, + "step": 851 + }, + { + "epoch": 0.9336986301369863, + "grad_norm": 2.1904265880584717, + "learning_rate": 3.854589764419171e-06, + "loss": 1.8265, + "step": 852 + }, + { + "epoch": 0.9347945205479452, + "grad_norm": 2.1374316215515137, + "learning_rate": 3.852558895207149e-06, + "loss": 1.8755, + "step": 853 + }, + { + "epoch": 0.9358904109589041, + "grad_norm": 2.2175488471984863, + "learning_rate": 3.850528025995126e-06, + "loss": 1.9649, + "step": 854 + }, + { + "epoch": 0.936986301369863, + "grad_norm": 2.192126989364624, + "learning_rate": 3.848497156783104e-06, + "loss": 1.919, + "step": 855 + }, + { + "epoch": 0.938082191780822, + "grad_norm": 2.219954013824463, + "learning_rate": 3.84646628757108e-06, + "loss": 1.8651, + "step": 856 + }, + { + "epoch": 0.9391780821917808, + "grad_norm": 2.2897071838378906, + "learning_rate": 3.8444354183590585e-06, + "loss": 1.8744, + "step": 857 + }, + { + "epoch": 0.9402739726027397, + "grad_norm": 2.1656205654144287, + "learning_rate": 3.842404549147035e-06, + "loss": 1.8622, + "step": 858 + }, + { + "epoch": 0.9413698630136986, + "grad_norm": 2.20072603225708, + "learning_rate": 3.840373679935013e-06, + "loss": 1.8678, + "step": 859 + }, + { + "epoch": 0.9424657534246575, + "grad_norm": 2.1474406719207764, + "learning_rate": 3.8383428107229894e-06, + "loss": 1.8462, + "step": 860 + }, + { + "epoch": 0.9424657534246575, + "eval_loss": 2.0451602935791016, + "eval_runtime": 37.1383, + "eval_samples_per_second": 26.926, + "eval_steps_per_second": 1.131, + "step": 860 + }, + { + "epoch": 0.9435616438356165, + "grad_norm": 1.9542036056518555, + "learning_rate": 3.836311941510967e-06, + "loss": 1.8739, + "step": 861 + }, + { + "epoch": 0.9446575342465754, + "grad_norm": 2.3951354026794434, + "learning_rate": 3.834281072298944e-06, + "loss": 1.8565, + "step": 862 + }, + { + "epoch": 0.9457534246575342, + "grad_norm": 2.4916961193084717, + "learning_rate": 3.832250203086921e-06, + "loss": 1.9138, + "step": 863 + }, + { + "epoch": 0.9468493150684931, + "grad_norm": 2.1650452613830566, + "learning_rate": 3.8302193338748985e-06, + "loss": 1.8499, + "step": 864 + }, + { + "epoch": 0.947945205479452, + "grad_norm": 2.2215073108673096, + "learning_rate": 3.828188464662876e-06, + "loss": 1.9376, + "step": 865 + }, + { + "epoch": 0.949041095890411, + "grad_norm": 2.3827078342437744, + "learning_rate": 3.826157595450853e-06, + "loss": 1.9726, + "step": 866 + }, + { + "epoch": 0.9501369863013699, + "grad_norm": 2.37194561958313, + "learning_rate": 3.82412672623883e-06, + "loss": 1.9009, + "step": 867 + }, + { + "epoch": 0.9512328767123288, + "grad_norm": 2.116562604904175, + "learning_rate": 3.822095857026808e-06, + "loss": 1.8342, + "step": 868 + }, + { + "epoch": 0.9523287671232876, + "grad_norm": 2.3410849571228027, + "learning_rate": 3.820064987814785e-06, + "loss": 1.9035, + "step": 869 + }, + { + "epoch": 0.9534246575342465, + "grad_norm": 2.207545280456543, + "learning_rate": 3.818034118602762e-06, + "loss": 1.8539, + "step": 870 + }, + { + "epoch": 0.9534246575342465, + "eval_loss": 2.0433826446533203, + "eval_runtime": 37.114, + "eval_samples_per_second": 26.944, + "eval_steps_per_second": 1.132, + "step": 870 + }, + { + "epoch": 0.9545205479452055, + "grad_norm": 2.1716196537017822, + "learning_rate": 3.8160032493907395e-06, + "loss": 1.8557, + "step": 871 + }, + { + "epoch": 0.9556164383561644, + "grad_norm": 2.213124990463257, + "learning_rate": 3.813972380178717e-06, + "loss": 1.8638, + "step": 872 + }, + { + "epoch": 0.9567123287671233, + "grad_norm": 2.3185672760009766, + "learning_rate": 3.811941510966694e-06, + "loss": 1.8645, + "step": 873 + }, + { + "epoch": 0.9578082191780822, + "grad_norm": 2.453956127166748, + "learning_rate": 3.809910641754671e-06, + "loss": 1.9183, + "step": 874 + }, + { + "epoch": 0.958904109589041, + "grad_norm": 2.1971592903137207, + "learning_rate": 3.8078797725426486e-06, + "loss": 1.8445, + "step": 875 + }, + { + "epoch": 0.96, + "grad_norm": 2.273247241973877, + "learning_rate": 3.8058489033306254e-06, + "loss": 1.8024, + "step": 876 + }, + { + "epoch": 0.9610958904109589, + "grad_norm": 2.1687841415405273, + "learning_rate": 3.803818034118603e-06, + "loss": 1.8931, + "step": 877 + }, + { + "epoch": 0.9621917808219178, + "grad_norm": 2.326498508453369, + "learning_rate": 3.80178716490658e-06, + "loss": 1.8587, + "step": 878 + }, + { + "epoch": 0.9632876712328767, + "grad_norm": 2.0317249298095703, + "learning_rate": 3.7997562956945577e-06, + "loss": 1.9394, + "step": 879 + }, + { + "epoch": 0.9643835616438357, + "grad_norm": 2.3171892166137695, + "learning_rate": 3.7977254264825345e-06, + "loss": 1.92, + "step": 880 + }, + { + "epoch": 0.9643835616438357, + "eval_loss": 2.0431556701660156, + "eval_runtime": 37.1174, + "eval_samples_per_second": 26.942, + "eval_steps_per_second": 1.132, + "step": 880 + }, + { + "epoch": 0.9654794520547946, + "grad_norm": 2.2215981483459473, + "learning_rate": 3.7956945572705122e-06, + "loss": 1.8658, + "step": 881 + }, + { + "epoch": 0.9665753424657534, + "grad_norm": 2.2940685749053955, + "learning_rate": 3.793663688058489e-06, + "loss": 1.8528, + "step": 882 + }, + { + "epoch": 0.9676712328767123, + "grad_norm": 2.2218081951141357, + "learning_rate": 3.791632818846467e-06, + "loss": 1.9446, + "step": 883 + }, + { + "epoch": 0.9687671232876712, + "grad_norm": 2.2608420848846436, + "learning_rate": 3.7896019496344436e-06, + "loss": 1.9433, + "step": 884 + }, + { + "epoch": 0.9698630136986301, + "grad_norm": 2.0247788429260254, + "learning_rate": 3.787571080422421e-06, + "loss": 1.9286, + "step": 885 + }, + { + "epoch": 0.9709589041095891, + "grad_norm": 2.164071798324585, + "learning_rate": 3.785540211210398e-06, + "loss": 1.9349, + "step": 886 + }, + { + "epoch": 0.972054794520548, + "grad_norm": 2.089215040206909, + "learning_rate": 3.7835093419983755e-06, + "loss": 1.9077, + "step": 887 + }, + { + "epoch": 0.9731506849315068, + "grad_norm": 2.245256185531616, + "learning_rate": 3.7814784727863527e-06, + "loss": 1.8911, + "step": 888 + }, + { + "epoch": 0.9742465753424657, + "grad_norm": 2.1227362155914307, + "learning_rate": 3.77944760357433e-06, + "loss": 1.8761, + "step": 889 + }, + { + "epoch": 0.9753424657534246, + "grad_norm": 2.1797730922698975, + "learning_rate": 3.7774167343623073e-06, + "loss": 1.9954, + "step": 890 + }, + { + "epoch": 0.9753424657534246, + "eval_loss": 2.034895896911621, + "eval_runtime": 37.1682, + "eval_samples_per_second": 26.905, + "eval_steps_per_second": 1.13, + "step": 890 + }, + { + "epoch": 0.9764383561643836, + "grad_norm": 2.171062707901001, + "learning_rate": 3.7753858651502846e-06, + "loss": 1.9287, + "step": 891 + }, + { + "epoch": 0.9775342465753425, + "grad_norm": 2.1950814723968506, + "learning_rate": 3.773354995938262e-06, + "loss": 1.8587, + "step": 892 + }, + { + "epoch": 0.9786301369863014, + "grad_norm": 2.2582995891571045, + "learning_rate": 3.771324126726239e-06, + "loss": 1.9269, + "step": 893 + }, + { + "epoch": 0.9797260273972602, + "grad_norm": 2.218113660812378, + "learning_rate": 3.7692932575142164e-06, + "loss": 1.9007, + "step": 894 + }, + { + "epoch": 0.9808219178082191, + "grad_norm": 2.187178134918213, + "learning_rate": 3.7672623883021937e-06, + "loss": 1.9326, + "step": 895 + }, + { + "epoch": 0.9819178082191781, + "grad_norm": 2.20451283454895, + "learning_rate": 3.765231519090171e-06, + "loss": 1.8644, + "step": 896 + }, + { + "epoch": 0.983013698630137, + "grad_norm": 2.1399290561676025, + "learning_rate": 3.7632006498781482e-06, + "loss": 1.7875, + "step": 897 + }, + { + "epoch": 0.9841095890410959, + "grad_norm": 2.049473524093628, + "learning_rate": 3.761169780666125e-06, + "loss": 1.8987, + "step": 898 + }, + { + "epoch": 0.9852054794520548, + "grad_norm": 2.154444456100464, + "learning_rate": 3.7591389114541028e-06, + "loss": 1.8475, + "step": 899 + }, + { + "epoch": 0.9863013698630136, + "grad_norm": 2.1840224266052246, + "learning_rate": 3.7571080422420796e-06, + "loss": 1.8966, + "step": 900 + }, + { + "epoch": 0.9863013698630136, + "eval_loss": 2.0326430797576904, + "eval_runtime": 37.1551, + "eval_samples_per_second": 26.914, + "eval_steps_per_second": 1.13, + "step": 900 + }, + { + "epoch": 0.9873972602739726, + "grad_norm": 2.0891942977905273, + "learning_rate": 3.7550771730300573e-06, + "loss": 1.9113, + "step": 901 + }, + { + "epoch": 0.9884931506849315, + "grad_norm": 2.2646753787994385, + "learning_rate": 3.753046303818034e-06, + "loss": 1.8624, + "step": 902 + }, + { + "epoch": 0.9895890410958904, + "grad_norm": 2.15173602104187, + "learning_rate": 3.751015434606012e-06, + "loss": 1.933, + "step": 903 + }, + { + "epoch": 0.9906849315068493, + "grad_norm": 2.1751456260681152, + "learning_rate": 3.7489845653939887e-06, + "loss": 1.9114, + "step": 904 + }, + { + "epoch": 0.9917808219178083, + "grad_norm": 2.161860942840576, + "learning_rate": 3.7469536961819664e-06, + "loss": 1.9508, + "step": 905 + }, + { + "epoch": 0.9928767123287672, + "grad_norm": 2.0921733379364014, + "learning_rate": 3.7449228269699433e-06, + "loss": 1.8482, + "step": 906 + }, + { + "epoch": 0.993972602739726, + "grad_norm": 2.0823893547058105, + "learning_rate": 3.742891957757921e-06, + "loss": 1.8744, + "step": 907 + }, + { + "epoch": 0.9950684931506849, + "grad_norm": 2.136939525604248, + "learning_rate": 3.740861088545898e-06, + "loss": 1.9037, + "step": 908 + }, + { + "epoch": 0.9961643835616438, + "grad_norm": 2.0600836277008057, + "learning_rate": 3.7388302193338755e-06, + "loss": 1.8922, + "step": 909 + }, + { + "epoch": 0.9972602739726028, + "grad_norm": 2.235780715942383, + "learning_rate": 3.7367993501218524e-06, + "loss": 1.9222, + "step": 910 + }, + { + "epoch": 0.9972602739726028, + "eval_loss": 2.034031629562378, + "eval_runtime": 37.2911, + "eval_samples_per_second": 26.816, + "eval_steps_per_second": 1.126, + "step": 910 + }, + { + "epoch": 0.9983561643835617, + "grad_norm": 2.124662160873413, + "learning_rate": 3.7347684809098297e-06, + "loss": 1.8584, + "step": 911 + }, + { + "epoch": 0.9994520547945206, + "grad_norm": 2.1237940788269043, + "learning_rate": 3.732737611697807e-06, + "loss": 1.8579, + "step": 912 + }, + { + "epoch": 1.0005479452054795, + "grad_norm": 2.1175124645233154, + "learning_rate": 3.7307067424857842e-06, + "loss": 1.8348, + "step": 913 + }, + { + "epoch": 1.0016438356164383, + "grad_norm": 2.349686861038208, + "learning_rate": 3.7286758732737615e-06, + "loss": 1.7822, + "step": 914 + }, + { + "epoch": 1.0027397260273974, + "grad_norm": 2.3278446197509766, + "learning_rate": 3.7266450040617388e-06, + "loss": 1.8346, + "step": 915 + }, + { + "epoch": 1.0038356164383562, + "grad_norm": 2.1633124351501465, + "learning_rate": 3.724614134849716e-06, + "loss": 1.8137, + "step": 916 + }, + { + "epoch": 1.004931506849315, + "grad_norm": 2.5075809955596924, + "learning_rate": 3.7225832656376933e-06, + "loss": 1.955, + "step": 917 + }, + { + "epoch": 1.006027397260274, + "grad_norm": 2.3370068073272705, + "learning_rate": 3.7205523964256706e-06, + "loss": 1.7985, + "step": 918 + }, + { + "epoch": 1.0071232876712328, + "grad_norm": 2.288229465484619, + "learning_rate": 3.718521527213648e-06, + "loss": 1.923, + "step": 919 + }, + { + "epoch": 1.0082191780821919, + "grad_norm": 2.1989684104919434, + "learning_rate": 3.716490658001625e-06, + "loss": 1.8876, + "step": 920 + }, + { + "epoch": 1.0082191780821919, + "eval_loss": 2.03442120552063, + "eval_runtime": 37.5659, + "eval_samples_per_second": 26.62, + "eval_steps_per_second": 1.118, + "step": 920 + }, + { + "epoch": 1.0093150684931507, + "grad_norm": 2.202122211456299, + "learning_rate": 3.7144597887896024e-06, + "loss": 1.879, + "step": 921 + }, + { + "epoch": 1.0104109589041095, + "grad_norm": 2.325223445892334, + "learning_rate": 3.7124289195775797e-06, + "loss": 1.7632, + "step": 922 + }, + { + "epoch": 1.0115068493150685, + "grad_norm": 2.2964577674865723, + "learning_rate": 3.710398050365557e-06, + "loss": 1.8983, + "step": 923 + }, + { + "epoch": 1.0126027397260273, + "grad_norm": 2.548537015914917, + "learning_rate": 3.708367181153534e-06, + "loss": 1.8666, + "step": 924 + }, + { + "epoch": 1.0136986301369864, + "grad_norm": 2.229745626449585, + "learning_rate": 3.7063363119415115e-06, + "loss": 1.8395, + "step": 925 + }, + { + "epoch": 1.0147945205479452, + "grad_norm": 2.3639235496520996, + "learning_rate": 3.7043054427294884e-06, + "loss": 1.8864, + "step": 926 + }, + { + "epoch": 1.0158904109589042, + "grad_norm": 2.3403055667877197, + "learning_rate": 3.702274573517466e-06, + "loss": 1.8385, + "step": 927 + }, + { + "epoch": 1.016986301369863, + "grad_norm": 2.2838001251220703, + "learning_rate": 3.700243704305443e-06, + "loss": 1.8135, + "step": 928 + }, + { + "epoch": 1.0180821917808218, + "grad_norm": 2.4730923175811768, + "learning_rate": 3.6982128350934206e-06, + "loss": 1.837, + "step": 929 + }, + { + "epoch": 1.0191780821917809, + "grad_norm": 2.5711073875427246, + "learning_rate": 3.6961819658813975e-06, + "loss": 1.8276, + "step": 930 + }, + { + "epoch": 1.0191780821917809, + "eval_loss": 2.0346033573150635, + "eval_runtime": 37.6741, + "eval_samples_per_second": 26.543, + "eval_steps_per_second": 1.115, + "step": 930 + }, + { + "epoch": 1.0202739726027397, + "grad_norm": 2.296823740005493, + "learning_rate": 3.694151096669375e-06, + "loss": 1.8054, + "step": 931 + }, + { + "epoch": 1.0213698630136987, + "grad_norm": 2.321107864379883, + "learning_rate": 3.692120227457352e-06, + "loss": 1.9296, + "step": 932 + }, + { + "epoch": 1.0224657534246575, + "grad_norm": 2.3478338718414307, + "learning_rate": 3.6900893582453297e-06, + "loss": 1.8503, + "step": 933 + }, + { + "epoch": 1.0235616438356165, + "grad_norm": 2.3698856830596924, + "learning_rate": 3.6880584890333066e-06, + "loss": 1.8766, + "step": 934 + }, + { + "epoch": 1.0246575342465754, + "grad_norm": 2.2751212120056152, + "learning_rate": 3.6860276198212834e-06, + "loss": 1.7575, + "step": 935 + }, + { + "epoch": 1.0257534246575342, + "grad_norm": 2.368741035461426, + "learning_rate": 3.683996750609261e-06, + "loss": 1.8555, + "step": 936 + }, + { + "epoch": 1.0268493150684932, + "grad_norm": 2.4507620334625244, + "learning_rate": 3.681965881397238e-06, + "loss": 1.8765, + "step": 937 + }, + { + "epoch": 1.027945205479452, + "grad_norm": 2.1907308101654053, + "learning_rate": 3.6799350121852157e-06, + "loss": 1.8652, + "step": 938 + }, + { + "epoch": 1.029041095890411, + "grad_norm": 2.2696914672851562, + "learning_rate": 3.6779041429731925e-06, + "loss": 1.8854, + "step": 939 + }, + { + "epoch": 1.0301369863013699, + "grad_norm": 2.3696796894073486, + "learning_rate": 3.6758732737611702e-06, + "loss": 1.8303, + "step": 940 + }, + { + "epoch": 1.0301369863013699, + "eval_loss": 2.030118703842163, + "eval_runtime": 37.1755, + "eval_samples_per_second": 26.899, + "eval_steps_per_second": 1.13, + "step": 940 + }, + { + "epoch": 1.0312328767123287, + "grad_norm": 2.440284013748169, + "learning_rate": 3.673842404549147e-06, + "loss": 1.8498, + "step": 941 + }, + { + "epoch": 1.0323287671232877, + "grad_norm": 2.250570297241211, + "learning_rate": 3.671811535337125e-06, + "loss": 1.8361, + "step": 942 + }, + { + "epoch": 1.0334246575342465, + "grad_norm": 2.5550990104675293, + "learning_rate": 3.6697806661251016e-06, + "loss": 1.8464, + "step": 943 + }, + { + "epoch": 1.0345205479452055, + "grad_norm": 2.502474784851074, + "learning_rate": 3.6677497969130793e-06, + "loss": 1.8471, + "step": 944 + }, + { + "epoch": 1.0356164383561643, + "grad_norm": 2.3815231323242188, + "learning_rate": 3.665718927701056e-06, + "loss": 1.9786, + "step": 945 + }, + { + "epoch": 1.0367123287671234, + "grad_norm": 2.3053736686706543, + "learning_rate": 3.663688058489034e-06, + "loss": 1.8625, + "step": 946 + }, + { + "epoch": 1.0378082191780822, + "grad_norm": 2.2623229026794434, + "learning_rate": 3.6616571892770107e-06, + "loss": 1.9206, + "step": 947 + }, + { + "epoch": 1.038904109589041, + "grad_norm": 2.634559392929077, + "learning_rate": 3.659626320064988e-06, + "loss": 1.8475, + "step": 948 + }, + { + "epoch": 1.04, + "grad_norm": 2.264615297317505, + "learning_rate": 3.6575954508529653e-06, + "loss": 1.7857, + "step": 949 + }, + { + "epoch": 1.0410958904109588, + "grad_norm": 2.4125282764434814, + "learning_rate": 3.6555645816409426e-06, + "loss": 1.841, + "step": 950 + }, + { + "epoch": 1.0410958904109588, + "eval_loss": 2.0278165340423584, + "eval_runtime": 37.1503, + "eval_samples_per_second": 26.918, + "eval_steps_per_second": 1.131, + "step": 950 + }, + { + "epoch": 1.0421917808219179, + "grad_norm": 2.365743637084961, + "learning_rate": 3.65353371242892e-06, + "loss": 1.8352, + "step": 951 + }, + { + "epoch": 1.0432876712328767, + "grad_norm": 2.6469883918762207, + "learning_rate": 3.651502843216897e-06, + "loss": 1.8814, + "step": 952 + }, + { + "epoch": 1.0443835616438357, + "grad_norm": 2.3646206855773926, + "learning_rate": 3.6494719740048744e-06, + "loss": 1.7282, + "step": 953 + }, + { + "epoch": 1.0454794520547945, + "grad_norm": 2.250856876373291, + "learning_rate": 3.6474411047928517e-06, + "loss": 1.8647, + "step": 954 + }, + { + "epoch": 1.0465753424657533, + "grad_norm": 2.5404837131500244, + "learning_rate": 3.645410235580829e-06, + "loss": 1.8175, + "step": 955 + }, + { + "epoch": 1.0476712328767124, + "grad_norm": 2.4388837814331055, + "learning_rate": 3.6433793663688062e-06, + "loss": 1.8643, + "step": 956 + }, + { + "epoch": 1.0487671232876712, + "grad_norm": 2.285959005355835, + "learning_rate": 3.6413484971567835e-06, + "loss": 1.8367, + "step": 957 + }, + { + "epoch": 1.0498630136986302, + "grad_norm": 2.1190242767333984, + "learning_rate": 3.6393176279447608e-06, + "loss": 1.8242, + "step": 958 + }, + { + "epoch": 1.050958904109589, + "grad_norm": 2.397578716278076, + "learning_rate": 3.637286758732738e-06, + "loss": 1.8681, + "step": 959 + }, + { + "epoch": 1.0520547945205478, + "grad_norm": 2.510599374771118, + "learning_rate": 3.6352558895207153e-06, + "loss": 1.8804, + "step": 960 + }, + { + "epoch": 1.0520547945205478, + "eval_loss": 2.0267090797424316, + "eval_runtime": 37.1226, + "eval_samples_per_second": 26.938, + "eval_steps_per_second": 1.131, + "step": 960 + }, + { + "epoch": 1.0531506849315069, + "grad_norm": 2.4743072986602783, + "learning_rate": 3.633225020308692e-06, + "loss": 1.8895, + "step": 961 + }, + { + "epoch": 1.0542465753424657, + "grad_norm": 2.3666775226593018, + "learning_rate": 3.63119415109667e-06, + "loss": 1.8836, + "step": 962 + }, + { + "epoch": 1.0553424657534247, + "grad_norm": 2.301119327545166, + "learning_rate": 3.6291632818846467e-06, + "loss": 1.8541, + "step": 963 + }, + { + "epoch": 1.0564383561643835, + "grad_norm": 2.482811212539673, + "learning_rate": 3.6271324126726244e-06, + "loss": 1.88, + "step": 964 + }, + { + "epoch": 1.0575342465753426, + "grad_norm": 2.2962276935577393, + "learning_rate": 3.6251015434606013e-06, + "loss": 1.7834, + "step": 965 + }, + { + "epoch": 1.0586301369863014, + "grad_norm": 2.365130662918091, + "learning_rate": 3.623070674248579e-06, + "loss": 1.7946, + "step": 966 + }, + { + "epoch": 1.0597260273972602, + "grad_norm": 2.4644675254821777, + "learning_rate": 3.621039805036556e-06, + "loss": 1.8379, + "step": 967 + }, + { + "epoch": 1.0608219178082192, + "grad_norm": 2.31742787361145, + "learning_rate": 3.6190089358245335e-06, + "loss": 1.8192, + "step": 968 + }, + { + "epoch": 1.061917808219178, + "grad_norm": 2.1614131927490234, + "learning_rate": 3.6169780666125104e-06, + "loss": 1.8794, + "step": 969 + }, + { + "epoch": 1.063013698630137, + "grad_norm": 2.4367032051086426, + "learning_rate": 3.614947197400488e-06, + "loss": 1.8673, + "step": 970 + }, + { + "epoch": 1.063013698630137, + "eval_loss": 2.0261292457580566, + "eval_runtime": 37.1116, + "eval_samples_per_second": 26.946, + "eval_steps_per_second": 1.132, + "step": 970 + }, + { + "epoch": 1.0641095890410959, + "grad_norm": 2.262789726257324, + "learning_rate": 3.612916328188465e-06, + "loss": 1.8808, + "step": 971 + }, + { + "epoch": 1.0652054794520547, + "grad_norm": 2.316316843032837, + "learning_rate": 3.610885458976442e-06, + "loss": 1.7803, + "step": 972 + }, + { + "epoch": 1.0663013698630137, + "grad_norm": 2.2936418056488037, + "learning_rate": 3.6088545897644195e-06, + "loss": 1.8783, + "step": 973 + }, + { + "epoch": 1.0673972602739725, + "grad_norm": 2.667069673538208, + "learning_rate": 3.6068237205523963e-06, + "loss": 1.8748, + "step": 974 + }, + { + "epoch": 1.0684931506849316, + "grad_norm": 2.31111216545105, + "learning_rate": 3.604792851340374e-06, + "loss": 1.8119, + "step": 975 + }, + { + "epoch": 1.0695890410958904, + "grad_norm": 2.268258571624756, + "learning_rate": 3.602761982128351e-06, + "loss": 1.8055, + "step": 976 + }, + { + "epoch": 1.0706849315068494, + "grad_norm": 2.307137966156006, + "learning_rate": 3.6007311129163286e-06, + "loss": 1.8976, + "step": 977 + }, + { + "epoch": 1.0717808219178082, + "grad_norm": 2.406036853790283, + "learning_rate": 3.5987002437043054e-06, + "loss": 1.8608, + "step": 978 + }, + { + "epoch": 1.072876712328767, + "grad_norm": 2.3303062915802, + "learning_rate": 3.596669374492283e-06, + "loss": 1.8038, + "step": 979 + }, + { + "epoch": 1.073972602739726, + "grad_norm": 2.2280168533325195, + "learning_rate": 3.59463850528026e-06, + "loss": 1.8342, + "step": 980 + }, + { + "epoch": 1.073972602739726, + "eval_loss": 2.0273854732513428, + "eval_runtime": 37.1546, + "eval_samples_per_second": 26.915, + "eval_steps_per_second": 1.13, + "step": 980 + }, + { + "epoch": 1.0750684931506849, + "grad_norm": 2.4218759536743164, + "learning_rate": 3.5926076360682377e-06, + "loss": 1.9104, + "step": 981 + }, + { + "epoch": 1.076164383561644, + "grad_norm": 2.2140634059906006, + "learning_rate": 3.5905767668562145e-06, + "loss": 1.7927, + "step": 982 + }, + { + "epoch": 1.0772602739726027, + "grad_norm": 2.179877996444702, + "learning_rate": 3.5885458976441922e-06, + "loss": 1.8868, + "step": 983 + }, + { + "epoch": 1.0783561643835617, + "grad_norm": 2.1988537311553955, + "learning_rate": 3.586515028432169e-06, + "loss": 1.8516, + "step": 984 + }, + { + "epoch": 1.0794520547945206, + "grad_norm": 2.1724448204040527, + "learning_rate": 3.5844841592201464e-06, + "loss": 1.8869, + "step": 985 + }, + { + "epoch": 1.0805479452054794, + "grad_norm": 2.2685718536376953, + "learning_rate": 3.5824532900081237e-06, + "loss": 1.8955, + "step": 986 + }, + { + "epoch": 1.0816438356164384, + "grad_norm": 2.39284348487854, + "learning_rate": 3.580422420796101e-06, + "loss": 1.7769, + "step": 987 + }, + { + "epoch": 1.0827397260273972, + "grad_norm": 2.3448750972747803, + "learning_rate": 3.578391551584078e-06, + "loss": 1.8417, + "step": 988 + }, + { + "epoch": 1.0838356164383562, + "grad_norm": 2.3518283367156982, + "learning_rate": 3.5763606823720555e-06, + "loss": 1.8524, + "step": 989 + }, + { + "epoch": 1.084931506849315, + "grad_norm": 2.1696958541870117, + "learning_rate": 3.5743298131600328e-06, + "loss": 1.7668, + "step": 990 + }, + { + "epoch": 1.084931506849315, + "eval_loss": 2.0244593620300293, + "eval_runtime": 37.1929, + "eval_samples_per_second": 26.887, + "eval_steps_per_second": 1.129, + "step": 990 + }, + { + "epoch": 1.086027397260274, + "grad_norm": 2.18300724029541, + "learning_rate": 3.57229894394801e-06, + "loss": 1.852, + "step": 991 + }, + { + "epoch": 1.087123287671233, + "grad_norm": 2.2075037956237793, + "learning_rate": 3.5702680747359873e-06, + "loss": 1.8799, + "step": 992 + }, + { + "epoch": 1.0882191780821917, + "grad_norm": 2.2172231674194336, + "learning_rate": 3.5682372055239646e-06, + "loss": 1.8573, + "step": 993 + }, + { + "epoch": 1.0893150684931507, + "grad_norm": 2.25169038772583, + "learning_rate": 3.566206336311942e-06, + "loss": 1.8125, + "step": 994 + }, + { + "epoch": 1.0904109589041096, + "grad_norm": 2.168386697769165, + "learning_rate": 3.564175467099919e-06, + "loss": 1.857, + "step": 995 + }, + { + "epoch": 1.0915068493150686, + "grad_norm": 2.2610065937042236, + "learning_rate": 3.5621445978878964e-06, + "loss": 1.817, + "step": 996 + }, + { + "epoch": 1.0926027397260274, + "grad_norm": 2.2561864852905273, + "learning_rate": 3.5601137286758737e-06, + "loss": 1.7754, + "step": 997 + }, + { + "epoch": 1.0936986301369862, + "grad_norm": 2.143625497817993, + "learning_rate": 3.5580828594638505e-06, + "loss": 1.8734, + "step": 998 + }, + { + "epoch": 1.0947945205479452, + "grad_norm": 2.433831214904785, + "learning_rate": 3.5560519902518282e-06, + "loss": 1.8248, + "step": 999 + }, + { + "epoch": 1.095890410958904, + "grad_norm": 2.237497091293335, + "learning_rate": 3.554021121039805e-06, + "loss": 1.8543, + "step": 1000 + }, + { + "epoch": 1.095890410958904, + "eval_loss": 2.020202159881592, + "eval_runtime": 37.2413, + "eval_samples_per_second": 26.852, + "eval_steps_per_second": 1.128, + "step": 1000 + }, + { + "epoch": 1.096986301369863, + "grad_norm": 2.253833532333374, + "learning_rate": 3.551990251827783e-06, + "loss": 1.8799, + "step": 1001 + }, + { + "epoch": 1.098082191780822, + "grad_norm": 2.205322504043579, + "learning_rate": 3.5499593826157596e-06, + "loss": 1.834, + "step": 1002 + }, + { + "epoch": 1.099178082191781, + "grad_norm": 2.3098037242889404, + "learning_rate": 3.5479285134037373e-06, + "loss": 1.8184, + "step": 1003 + }, + { + "epoch": 1.1002739726027397, + "grad_norm": 2.361888885498047, + "learning_rate": 3.545897644191714e-06, + "loss": 1.827, + "step": 1004 + }, + { + "epoch": 1.1013698630136985, + "grad_norm": 2.341947555541992, + "learning_rate": 3.543866774979692e-06, + "loss": 1.8184, + "step": 1005 + }, + { + "epoch": 1.1024657534246576, + "grad_norm": 2.2927780151367188, + "learning_rate": 3.5418359057676687e-06, + "loss": 1.805, + "step": 1006 + }, + { + "epoch": 1.1035616438356164, + "grad_norm": 2.432988166809082, + "learning_rate": 3.5398050365556464e-06, + "loss": 1.8465, + "step": 1007 + }, + { + "epoch": 1.1046575342465754, + "grad_norm": 2.428126335144043, + "learning_rate": 3.5377741673436233e-06, + "loss": 1.8115, + "step": 1008 + }, + { + "epoch": 1.1057534246575342, + "grad_norm": 2.32551646232605, + "learning_rate": 3.535743298131601e-06, + "loss": 1.8522, + "step": 1009 + }, + { + "epoch": 1.106849315068493, + "grad_norm": 2.297980308532715, + "learning_rate": 3.533712428919578e-06, + "loss": 1.8137, + "step": 1010 + }, + { + "epoch": 1.106849315068493, + "eval_loss": 2.0190494060516357, + "eval_runtime": 37.2149, + "eval_samples_per_second": 26.871, + "eval_steps_per_second": 1.129, + "step": 1010 + }, + { + "epoch": 1.107945205479452, + "grad_norm": 2.364342451095581, + "learning_rate": 3.5316815597075547e-06, + "loss": 1.8677, + "step": 1011 + }, + { + "epoch": 1.1090410958904109, + "grad_norm": 2.355008125305176, + "learning_rate": 3.5296506904955324e-06, + "loss": 1.8937, + "step": 1012 + }, + { + "epoch": 1.11013698630137, + "grad_norm": 2.2288317680358887, + "learning_rate": 3.5276198212835093e-06, + "loss": 1.8188, + "step": 1013 + }, + { + "epoch": 1.1112328767123287, + "grad_norm": 2.2801361083984375, + "learning_rate": 3.525588952071487e-06, + "loss": 1.8386, + "step": 1014 + }, + { + "epoch": 1.1123287671232878, + "grad_norm": 2.2801361083984375, + "learning_rate": 3.525588952071487e-06, + "loss": 1.8845, + "step": 1015 + }, + { + "epoch": 1.1134246575342466, + "grad_norm": 2.4301352500915527, + "learning_rate": 3.523558082859464e-06, + "loss": 1.8898, + "step": 1016 + }, + { + "epoch": 1.1145205479452054, + "grad_norm": 2.3492274284362793, + "learning_rate": 3.5215272136474415e-06, + "loss": 1.8711, + "step": 1017 + }, + { + "epoch": 1.1156164383561644, + "grad_norm": 2.1788229942321777, + "learning_rate": 3.5194963444354184e-06, + "loss": 1.8177, + "step": 1018 + }, + { + "epoch": 1.1167123287671232, + "grad_norm": 2.269767999649048, + "learning_rate": 3.517465475223396e-06, + "loss": 1.8259, + "step": 1019 + }, + { + "epoch": 1.1178082191780823, + "grad_norm": 2.2890219688415527, + "learning_rate": 3.515434606011373e-06, + "loss": 1.7873, + "step": 1020 + }, + { + "epoch": 1.1178082191780823, + "eval_loss": 2.023195266723633, + "eval_runtime": 37.5701, + "eval_samples_per_second": 26.617, + "eval_steps_per_second": 1.118, + "step": 1020 + }, + { + "epoch": 1.118904109589041, + "grad_norm": 2.5528533458709717, + "learning_rate": 3.5134037367993506e-06, + "loss": 1.8624, + "step": 1021 + }, + { + "epoch": 1.12, + "grad_norm": 2.5109457969665527, + "learning_rate": 3.5113728675873275e-06, + "loss": 1.9178, + "step": 1022 + }, + { + "epoch": 1.121095890410959, + "grad_norm": 2.288877010345459, + "learning_rate": 3.5093419983753047e-06, + "loss": 1.8604, + "step": 1023 + }, + { + "epoch": 1.1221917808219177, + "grad_norm": 2.460153818130493, + "learning_rate": 3.507311129163282e-06, + "loss": 1.8551, + "step": 1024 + }, + { + "epoch": 1.1232876712328768, + "grad_norm": 2.3520290851593018, + "learning_rate": 3.5052802599512593e-06, + "loss": 1.7818, + "step": 1025 + }, + { + "epoch": 1.1243835616438356, + "grad_norm": 2.32812237739563, + "learning_rate": 3.5032493907392366e-06, + "loss": 1.8255, + "step": 1026 + }, + { + "epoch": 1.1254794520547946, + "grad_norm": 2.3118042945861816, + "learning_rate": 3.501218521527214e-06, + "loss": 1.8493, + "step": 1027 + }, + { + "epoch": 1.1265753424657534, + "grad_norm": 2.2750277519226074, + "learning_rate": 3.499187652315191e-06, + "loss": 1.7759, + "step": 1028 + }, + { + "epoch": 1.1276712328767124, + "grad_norm": 2.272120475769043, + "learning_rate": 3.4971567831031684e-06, + "loss": 1.8399, + "step": 1029 + }, + { + "epoch": 1.1287671232876713, + "grad_norm": 2.2075254917144775, + "learning_rate": 3.4951259138911457e-06, + "loss": 1.8774, + "step": 1030 + }, + { + "epoch": 1.1287671232876713, + "eval_loss": 2.0199804306030273, + "eval_runtime": 37.6753, + "eval_samples_per_second": 26.543, + "eval_steps_per_second": 1.115, + "step": 1030 + }, + { + "epoch": 1.12986301369863, + "grad_norm": 2.359877586364746, + "learning_rate": 3.493095044679123e-06, + "loss": 1.7942, + "step": 1031 + }, + { + "epoch": 1.130958904109589, + "grad_norm": 2.3057703971862793, + "learning_rate": 3.4910641754671002e-06, + "loss": 1.8996, + "step": 1032 + }, + { + "epoch": 1.132054794520548, + "grad_norm": 2.3109569549560547, + "learning_rate": 3.4890333062550775e-06, + "loss": 1.7336, + "step": 1033 + }, + { + "epoch": 1.133150684931507, + "grad_norm": 2.3021819591522217, + "learning_rate": 3.4870024370430548e-06, + "loss": 1.869, + "step": 1034 + }, + { + "epoch": 1.1342465753424658, + "grad_norm": 2.4741852283477783, + "learning_rate": 3.484971567831032e-06, + "loss": 1.7652, + "step": 1035 + }, + { + "epoch": 1.1353424657534246, + "grad_norm": 2.326500415802002, + "learning_rate": 3.482940698619009e-06, + "loss": 1.809, + "step": 1036 + }, + { + "epoch": 1.1364383561643836, + "grad_norm": 2.2562193870544434, + "learning_rate": 3.4809098294069866e-06, + "loss": 1.8806, + "step": 1037 + }, + { + "epoch": 1.1375342465753424, + "grad_norm": 2.2849345207214355, + "learning_rate": 3.4788789601949634e-06, + "loss": 1.8329, + "step": 1038 + }, + { + "epoch": 1.1386301369863014, + "grad_norm": 2.333707332611084, + "learning_rate": 3.476848090982941e-06, + "loss": 1.8541, + "step": 1039 + }, + { + "epoch": 1.1397260273972603, + "grad_norm": 2.234175682067871, + "learning_rate": 3.474817221770918e-06, + "loss": 1.8126, + "step": 1040 + }, + { + "epoch": 1.1397260273972603, + "eval_loss": 2.0176141262054443, + "eval_runtime": 37.232, + "eval_samples_per_second": 26.859, + "eval_steps_per_second": 1.128, + "step": 1040 + }, + { + "epoch": 1.140821917808219, + "grad_norm": 2.4212377071380615, + "learning_rate": 3.4727863525588957e-06, + "loss": 1.9325, + "step": 1041 + }, + { + "epoch": 1.141917808219178, + "grad_norm": 2.1526010036468506, + "learning_rate": 3.4707554833468725e-06, + "loss": 1.8674, + "step": 1042 + }, + { + "epoch": 1.143013698630137, + "grad_norm": 2.336186408996582, + "learning_rate": 3.4687246141348502e-06, + "loss": 1.8275, + "step": 1043 + }, + { + "epoch": 1.144109589041096, + "grad_norm": 2.2991745471954346, + "learning_rate": 3.466693744922827e-06, + "loss": 1.8057, + "step": 1044 + }, + { + "epoch": 1.1452054794520548, + "grad_norm": 2.3816370964050293, + "learning_rate": 3.464662875710805e-06, + "loss": 1.8212, + "step": 1045 + }, + { + "epoch": 1.1463013698630138, + "grad_norm": 2.2383193969726562, + "learning_rate": 3.4626320064987817e-06, + "loss": 1.8444, + "step": 1046 + }, + { + "epoch": 1.1473972602739726, + "grad_norm": 2.377574920654297, + "learning_rate": 3.4606011372867594e-06, + "loss": 1.8319, + "step": 1047 + }, + { + "epoch": 1.1484931506849314, + "grad_norm": 2.1803700923919678, + "learning_rate": 3.458570268074736e-06, + "loss": 1.8484, + "step": 1048 + }, + { + "epoch": 1.1495890410958904, + "grad_norm": 2.1810641288757324, + "learning_rate": 3.4565393988627135e-06, + "loss": 1.8187, + "step": 1049 + }, + { + "epoch": 1.1506849315068493, + "grad_norm": 2.1565394401550293, + "learning_rate": 3.4545085296506908e-06, + "loss": 1.8117, + "step": 1050 + }, + { + "epoch": 1.1506849315068493, + "eval_loss": 2.0156562328338623, + "eval_runtime": 37.2072, + "eval_samples_per_second": 26.877, + "eval_steps_per_second": 1.129, + "step": 1050 + }, + { + "epoch": 1.1517808219178083, + "grad_norm": 2.1753759384155273, + "learning_rate": 3.452477660438668e-06, + "loss": 1.7822, + "step": 1051 + }, + { + "epoch": 1.152876712328767, + "grad_norm": 2.250089645385742, + "learning_rate": 3.4504467912266453e-06, + "loss": 1.8634, + "step": 1052 + }, + { + "epoch": 1.1539726027397261, + "grad_norm": 2.3043293952941895, + "learning_rate": 3.4484159220146226e-06, + "loss": 1.7743, + "step": 1053 + }, + { + "epoch": 1.155068493150685, + "grad_norm": 2.2193732261657715, + "learning_rate": 3.4463850528026e-06, + "loss": 1.8328, + "step": 1054 + }, + { + "epoch": 1.1561643835616437, + "grad_norm": 2.4204866886138916, + "learning_rate": 3.444354183590577e-06, + "loss": 1.8133, + "step": 1055 + }, + { + "epoch": 1.1572602739726028, + "grad_norm": 2.3088018894195557, + "learning_rate": 3.4423233143785544e-06, + "loss": 1.8523, + "step": 1056 + }, + { + "epoch": 1.1583561643835616, + "grad_norm": 2.38910174369812, + "learning_rate": 3.4402924451665317e-06, + "loss": 1.7846, + "step": 1057 + }, + { + "epoch": 1.1594520547945206, + "grad_norm": 2.2473702430725098, + "learning_rate": 3.438261575954509e-06, + "loss": 1.7632, + "step": 1058 + }, + { + "epoch": 1.1605479452054794, + "grad_norm": 2.3356645107269287, + "learning_rate": 3.4362307067424862e-06, + "loss": 1.8461, + "step": 1059 + }, + { + "epoch": 1.1616438356164385, + "grad_norm": 2.2460737228393555, + "learning_rate": 3.434199837530463e-06, + "loss": 1.8853, + "step": 1060 + }, + { + "epoch": 1.1616438356164385, + "eval_loss": 2.016077995300293, + "eval_runtime": 37.1943, + "eval_samples_per_second": 26.886, + "eval_steps_per_second": 1.129, + "step": 1060 + }, + { + "epoch": 1.1627397260273973, + "grad_norm": 2.28434157371521, + "learning_rate": 3.432168968318441e-06, + "loss": 1.8451, + "step": 1061 + }, + { + "epoch": 1.163835616438356, + "grad_norm": 2.299049139022827, + "learning_rate": 3.4301380991064176e-06, + "loss": 1.7777, + "step": 1062 + }, + { + "epoch": 1.1649315068493151, + "grad_norm": 2.364074468612671, + "learning_rate": 3.4281072298943953e-06, + "loss": 1.8517, + "step": 1063 + }, + { + "epoch": 1.166027397260274, + "grad_norm": 2.2381322383880615, + "learning_rate": 3.426076360682372e-06, + "loss": 1.889, + "step": 1064 + }, + { + "epoch": 1.167123287671233, + "grad_norm": 2.2984025478363037, + "learning_rate": 3.42404549147035e-06, + "loss": 1.8907, + "step": 1065 + }, + { + "epoch": 1.1682191780821918, + "grad_norm": 2.2215473651885986, + "learning_rate": 3.4220146222583267e-06, + "loss": 1.8313, + "step": 1066 + }, + { + "epoch": 1.1693150684931506, + "grad_norm": 2.3414270877838135, + "learning_rate": 3.4199837530463044e-06, + "loss": 1.783, + "step": 1067 + }, + { + "epoch": 1.1704109589041096, + "grad_norm": 2.0636470317840576, + "learning_rate": 3.4179528838342813e-06, + "loss": 1.8185, + "step": 1068 + }, + { + "epoch": 1.1715068493150684, + "grad_norm": 2.2158873081207275, + "learning_rate": 3.415922014622259e-06, + "loss": 1.9319, + "step": 1069 + }, + { + "epoch": 1.1726027397260275, + "grad_norm": 2.244051694869995, + "learning_rate": 3.413891145410236e-06, + "loss": 1.8732, + "step": 1070 + }, + { + "epoch": 1.1726027397260275, + "eval_loss": 2.0180234909057617, + "eval_runtime": 37.1822, + "eval_samples_per_second": 26.895, + "eval_steps_per_second": 1.13, + "step": 1070 + }, + { + "epoch": 1.1736986301369863, + "grad_norm": 2.1608433723449707, + "learning_rate": 3.4118602761982135e-06, + "loss": 1.8709, + "step": 1071 + }, + { + "epoch": 1.174794520547945, + "grad_norm": 2.2612338066101074, + "learning_rate": 3.4098294069861904e-06, + "loss": 1.8882, + "step": 1072 + }, + { + "epoch": 1.1758904109589041, + "grad_norm": 2.3257741928100586, + "learning_rate": 3.4077985377741673e-06, + "loss": 1.8377, + "step": 1073 + }, + { + "epoch": 1.176986301369863, + "grad_norm": 2.2486586570739746, + "learning_rate": 3.405767668562145e-06, + "loss": 1.8105, + "step": 1074 + }, + { + "epoch": 1.178082191780822, + "grad_norm": 2.2840020656585693, + "learning_rate": 3.403736799350122e-06, + "loss": 1.8868, + "step": 1075 + }, + { + "epoch": 1.1791780821917808, + "grad_norm": 2.3264284133911133, + "learning_rate": 3.4017059301380995e-06, + "loss": 1.8701, + "step": 1076 + }, + { + "epoch": 1.1802739726027398, + "grad_norm": 2.311424493789673, + "learning_rate": 3.3996750609260764e-06, + "loss": 1.801, + "step": 1077 + }, + { + "epoch": 1.1813698630136986, + "grad_norm": 2.2177181243896484, + "learning_rate": 3.397644191714054e-06, + "loss": 1.9188, + "step": 1078 + }, + { + "epoch": 1.1824657534246574, + "grad_norm": 2.413118362426758, + "learning_rate": 3.395613322502031e-06, + "loss": 1.8198, + "step": 1079 + }, + { + "epoch": 1.1835616438356165, + "grad_norm": 2.334688186645508, + "learning_rate": 3.3935824532900086e-06, + "loss": 1.822, + "step": 1080 + }, + { + "epoch": 1.1835616438356165, + "eval_loss": 2.012676954269409, + "eval_runtime": 37.2415, + "eval_samples_per_second": 26.852, + "eval_steps_per_second": 1.128, + "step": 1080 + }, + { + "epoch": 1.1846575342465753, + "grad_norm": 2.3901584148406982, + "learning_rate": 3.3915515840779855e-06, + "loss": 1.791, + "step": 1081 + }, + { + "epoch": 1.1857534246575343, + "grad_norm": 2.2731425762176514, + "learning_rate": 3.389520714865963e-06, + "loss": 1.8367, + "step": 1082 + }, + { + "epoch": 1.1868493150684931, + "grad_norm": 2.311553478240967, + "learning_rate": 3.38748984565394e-06, + "loss": 1.8302, + "step": 1083 + }, + { + "epoch": 1.1879452054794521, + "grad_norm": 2.3030776977539062, + "learning_rate": 3.3854589764419177e-06, + "loss": 1.8652, + "step": 1084 + }, + { + "epoch": 1.189041095890411, + "grad_norm": 2.341188430786133, + "learning_rate": 3.3834281072298946e-06, + "loss": 1.8094, + "step": 1085 + }, + { + "epoch": 1.1901369863013698, + "grad_norm": 2.331962823867798, + "learning_rate": 3.381397238017872e-06, + "loss": 1.827, + "step": 1086 + }, + { + "epoch": 1.1912328767123288, + "grad_norm": 2.275923728942871, + "learning_rate": 3.379366368805849e-06, + "loss": 1.7935, + "step": 1087 + }, + { + "epoch": 1.1923287671232876, + "grad_norm": 2.4608163833618164, + "learning_rate": 3.3773354995938264e-06, + "loss": 1.8965, + "step": 1088 + }, + { + "epoch": 1.1934246575342466, + "grad_norm": 2.2208592891693115, + "learning_rate": 3.3753046303818037e-06, + "loss": 1.8447, + "step": 1089 + }, + { + "epoch": 1.1945205479452055, + "grad_norm": 2.4284770488739014, + "learning_rate": 3.373273761169781e-06, + "loss": 1.8306, + "step": 1090 + }, + { + "epoch": 1.1945205479452055, + "eval_loss": 2.010664939880371, + "eval_runtime": 37.237, + "eval_samples_per_second": 26.855, + "eval_steps_per_second": 1.128, + "step": 1090 + }, + { + "epoch": 1.1956164383561645, + "grad_norm": 2.329601764678955, + "learning_rate": 3.3712428919577582e-06, + "loss": 1.8275, + "step": 1091 + }, + { + "epoch": 1.1967123287671233, + "grad_norm": 2.277939796447754, + "learning_rate": 3.3692120227457355e-06, + "loss": 1.8161, + "step": 1092 + }, + { + "epoch": 1.1978082191780821, + "grad_norm": 2.446780204772949, + "learning_rate": 3.3671811535337128e-06, + "loss": 1.8511, + "step": 1093 + }, + { + "epoch": 1.1989041095890411, + "grad_norm": 2.382477283477783, + "learning_rate": 3.36515028432169e-06, + "loss": 1.7834, + "step": 1094 + }, + { + "epoch": 1.2, + "grad_norm": 2.366361379623413, + "learning_rate": 3.3631194151096673e-06, + "loss": 1.8105, + "step": 1095 + }, + { + "epoch": 1.201095890410959, + "grad_norm": 2.3439717292785645, + "learning_rate": 3.3610885458976446e-06, + "loss": 1.8418, + "step": 1096 + }, + { + "epoch": 1.2021917808219178, + "grad_norm": 2.2739830017089844, + "learning_rate": 3.359057676685622e-06, + "loss": 1.8818, + "step": 1097 + }, + { + "epoch": 1.2032876712328768, + "grad_norm": 2.1636738777160645, + "learning_rate": 3.357026807473599e-06, + "loss": 1.8757, + "step": 1098 + }, + { + "epoch": 1.2043835616438356, + "grad_norm": 2.386441946029663, + "learning_rate": 3.354995938261576e-06, + "loss": 1.756, + "step": 1099 + }, + { + "epoch": 1.2054794520547945, + "grad_norm": 2.2241456508636475, + "learning_rate": 3.3529650690495537e-06, + "loss": 1.8034, + "step": 1100 + }, + { + "epoch": 1.2054794520547945, + "eval_loss": 2.008504867553711, + "eval_runtime": 37.2855, + "eval_samples_per_second": 26.82, + "eval_steps_per_second": 1.126, + "step": 1100 + }, + { + "epoch": 1.2065753424657535, + "grad_norm": 2.196992874145508, + "learning_rate": 3.3509341998375305e-06, + "loss": 1.7722, + "step": 1101 + }, + { + "epoch": 1.2076712328767123, + "grad_norm": 2.3572094440460205, + "learning_rate": 3.3489033306255082e-06, + "loss": 1.8109, + "step": 1102 + }, + { + "epoch": 1.208767123287671, + "grad_norm": 2.443281650543213, + "learning_rate": 3.346872461413485e-06, + "loss": 1.843, + "step": 1103 + }, + { + "epoch": 1.2098630136986301, + "grad_norm": 2.2613234519958496, + "learning_rate": 3.344841592201463e-06, + "loss": 1.8023, + "step": 1104 + }, + { + "epoch": 1.210958904109589, + "grad_norm": 2.2466726303100586, + "learning_rate": 3.3428107229894397e-06, + "loss": 1.7838, + "step": 1105 + }, + { + "epoch": 1.212054794520548, + "grad_norm": 2.2314462661743164, + "learning_rate": 3.3407798537774174e-06, + "loss": 1.8077, + "step": 1106 + }, + { + "epoch": 1.2131506849315068, + "grad_norm": 2.3288400173187256, + "learning_rate": 3.338748984565394e-06, + "loss": 1.8319, + "step": 1107 + }, + { + "epoch": 1.2142465753424658, + "grad_norm": 2.3477249145507812, + "learning_rate": 3.336718115353372e-06, + "loss": 1.8772, + "step": 1108 + }, + { + "epoch": 1.2153424657534246, + "grad_norm": 2.3588409423828125, + "learning_rate": 3.3346872461413488e-06, + "loss": 1.8832, + "step": 1109 + }, + { + "epoch": 1.2164383561643834, + "grad_norm": 2.327796220779419, + "learning_rate": 3.3326563769293256e-06, + "loss": 1.8033, + "step": 1110 + }, + { + "epoch": 1.2164383561643834, + "eval_loss": 2.0094382762908936, + "eval_runtime": 37.2666, + "eval_samples_per_second": 26.834, + "eval_steps_per_second": 1.127, + "step": 1110 + }, + { + "epoch": 1.2175342465753425, + "grad_norm": 2.257244825363159, + "learning_rate": 3.3306255077173033e-06, + "loss": 1.7903, + "step": 1111 + }, + { + "epoch": 1.2186301369863013, + "grad_norm": 2.3321614265441895, + "learning_rate": 3.32859463850528e-06, + "loss": 1.8295, + "step": 1112 + }, + { + "epoch": 1.2197260273972603, + "grad_norm": 2.4073803424835205, + "learning_rate": 3.326563769293258e-06, + "loss": 1.8395, + "step": 1113 + }, + { + "epoch": 1.2208219178082191, + "grad_norm": 2.3631434440612793, + "learning_rate": 3.3245329000812347e-06, + "loss": 1.7548, + "step": 1114 + }, + { + "epoch": 1.2219178082191782, + "grad_norm": 2.275766134262085, + "learning_rate": 3.3225020308692124e-06, + "loss": 1.808, + "step": 1115 + }, + { + "epoch": 1.223013698630137, + "grad_norm": 2.438011407852173, + "learning_rate": 3.3204711616571893e-06, + "loss": 1.8274, + "step": 1116 + }, + { + "epoch": 1.2241095890410958, + "grad_norm": 2.371826171875, + "learning_rate": 3.318440292445167e-06, + "loss": 1.7742, + "step": 1117 + }, + { + "epoch": 1.2252054794520548, + "grad_norm": 2.422004461288452, + "learning_rate": 3.316409423233144e-06, + "loss": 1.8268, + "step": 1118 + }, + { + "epoch": 1.2263013698630136, + "grad_norm": 2.377979278564453, + "learning_rate": 3.3143785540211215e-06, + "loss": 1.7857, + "step": 1119 + }, + { + "epoch": 1.2273972602739727, + "grad_norm": 2.288482904434204, + "learning_rate": 3.3123476848090984e-06, + "loss": 1.876, + "step": 1120 + }, + { + "epoch": 1.2273972602739727, + "eval_loss": 2.0104730129241943, + "eval_runtime": 37.5365, + "eval_samples_per_second": 26.641, + "eval_steps_per_second": 1.119, + "step": 1120 + }, + { + "epoch": 1.2284931506849315, + "grad_norm": 2.451533555984497, + "learning_rate": 3.310316815597076e-06, + "loss": 1.8324, + "step": 1121 + }, + { + "epoch": 1.2295890410958905, + "grad_norm": 2.3785181045532227, + "learning_rate": 3.308285946385053e-06, + "loss": 1.8239, + "step": 1122 + }, + { + "epoch": 1.2306849315068493, + "grad_norm": 2.2276721000671387, + "learning_rate": 3.30625507717303e-06, + "loss": 1.9132, + "step": 1123 + }, + { + "epoch": 1.2317808219178081, + "grad_norm": 2.2575314044952393, + "learning_rate": 3.3042242079610075e-06, + "loss": 1.769, + "step": 1124 + }, + { + "epoch": 1.2328767123287672, + "grad_norm": 2.5311717987060547, + "learning_rate": 3.3021933387489847e-06, + "loss": 1.8605, + "step": 1125 + }, + { + "epoch": 1.233972602739726, + "grad_norm": 2.2395896911621094, + "learning_rate": 3.300162469536962e-06, + "loss": 1.8725, + "step": 1126 + }, + { + "epoch": 1.235068493150685, + "grad_norm": 2.2402031421661377, + "learning_rate": 3.2981316003249393e-06, + "loss": 1.7955, + "step": 1127 + }, + { + "epoch": 1.2361643835616438, + "grad_norm": 2.308950424194336, + "learning_rate": 3.2961007311129166e-06, + "loss": 1.8812, + "step": 1128 + }, + { + "epoch": 1.2372602739726029, + "grad_norm": 2.437767267227173, + "learning_rate": 3.294069861900894e-06, + "loss": 1.8489, + "step": 1129 + }, + { + "epoch": 1.2383561643835617, + "grad_norm": 2.3188693523406982, + "learning_rate": 3.292038992688871e-06, + "loss": 1.9082, + "step": 1130 + }, + { + "epoch": 1.2383561643835617, + "eval_loss": 2.0110676288604736, + "eval_runtime": 37.6437, + "eval_samples_per_second": 26.565, + "eval_steps_per_second": 1.116, + "step": 1130 + }, + { + "epoch": 1.2394520547945205, + "grad_norm": 2.312760591506958, + "learning_rate": 3.2900081234768484e-06, + "loss": 1.8344, + "step": 1131 + }, + { + "epoch": 1.2405479452054795, + "grad_norm": 2.5837206840515137, + "learning_rate": 3.2879772542648257e-06, + "loss": 1.8503, + "step": 1132 + }, + { + "epoch": 1.2416438356164383, + "grad_norm": 2.1864733695983887, + "learning_rate": 3.285946385052803e-06, + "loss": 1.8822, + "step": 1133 + }, + { + "epoch": 1.2427397260273974, + "grad_norm": 2.2644550800323486, + "learning_rate": 3.2839155158407802e-06, + "loss": 1.7541, + "step": 1134 + }, + { + "epoch": 1.2438356164383562, + "grad_norm": 2.292646646499634, + "learning_rate": 3.2818846466287575e-06, + "loss": 1.8575, + "step": 1135 + }, + { + "epoch": 1.244931506849315, + "grad_norm": 2.2162070274353027, + "learning_rate": 3.2798537774167344e-06, + "loss": 1.7525, + "step": 1136 + }, + { + "epoch": 1.246027397260274, + "grad_norm": 2.32688307762146, + "learning_rate": 3.277822908204712e-06, + "loss": 1.8955, + "step": 1137 + }, + { + "epoch": 1.2471232876712328, + "grad_norm": 2.3226473331451416, + "learning_rate": 3.275792038992689e-06, + "loss": 1.8255, + "step": 1138 + }, + { + "epoch": 1.2482191780821918, + "grad_norm": 2.3533706665039062, + "learning_rate": 3.2737611697806666e-06, + "loss": 1.783, + "step": 1139 + }, + { + "epoch": 1.2493150684931507, + "grad_norm": 2.237661123275757, + "learning_rate": 3.2717303005686435e-06, + "loss": 1.8367, + "step": 1140 + }, + { + "epoch": 1.2493150684931507, + "eval_loss": 2.0123534202575684, + "eval_runtime": 37.3573, + "eval_samples_per_second": 26.769, + "eval_steps_per_second": 1.124, + "step": 1140 + }, + { + "epoch": 1.2504109589041095, + "grad_norm": 2.283047676086426, + "learning_rate": 3.269699431356621e-06, + "loss": 1.7987, + "step": 1141 + }, + { + "epoch": 1.2515068493150685, + "grad_norm": 2.2264456748962402, + "learning_rate": 3.267668562144598e-06, + "loss": 1.8307, + "step": 1142 + }, + { + "epoch": 1.2526027397260273, + "grad_norm": 2.2724905014038086, + "learning_rate": 3.2656376929325757e-06, + "loss": 1.8237, + "step": 1143 + }, + { + "epoch": 1.2536986301369863, + "grad_norm": 2.270782947540283, + "learning_rate": 3.2636068237205526e-06, + "loss": 1.7846, + "step": 1144 + }, + { + "epoch": 1.2547945205479452, + "grad_norm": 2.2032551765441895, + "learning_rate": 3.2615759545085303e-06, + "loss": 1.8438, + "step": 1145 + }, + { + "epoch": 1.2558904109589042, + "grad_norm": 2.199150800704956, + "learning_rate": 3.259545085296507e-06, + "loss": 1.8199, + "step": 1146 + }, + { + "epoch": 1.256986301369863, + "grad_norm": 2.4404170513153076, + "learning_rate": 3.257514216084484e-06, + "loss": 1.8817, + "step": 1147 + }, + { + "epoch": 1.2580821917808218, + "grad_norm": 2.281348943710327, + "learning_rate": 3.2554833468724617e-06, + "loss": 1.771, + "step": 1148 + }, + { + "epoch": 1.2591780821917808, + "grad_norm": 2.19482159614563, + "learning_rate": 3.2534524776604385e-06, + "loss": 1.8684, + "step": 1149 + }, + { + "epoch": 1.2602739726027397, + "grad_norm": 2.221618413925171, + "learning_rate": 3.2514216084484162e-06, + "loss": 1.8398, + "step": 1150 + }, + { + "epoch": 1.2602739726027397, + "eval_loss": 2.007969617843628, + "eval_runtime": 37.3679, + "eval_samples_per_second": 26.761, + "eval_steps_per_second": 1.124, + "step": 1150 + }, + { + "epoch": 1.2613698630136987, + "grad_norm": 2.193432569503784, + "learning_rate": 3.249390739236393e-06, + "loss": 1.8122, + "step": 1151 + }, + { + "epoch": 1.2624657534246575, + "grad_norm": 2.246462106704712, + "learning_rate": 3.2473598700243708e-06, + "loss": 1.8447, + "step": 1152 + }, + { + "epoch": 1.2635616438356165, + "grad_norm": 2.3411879539489746, + "learning_rate": 3.2453290008123476e-06, + "loss": 1.7988, + "step": 1153 + }, + { + "epoch": 1.2646575342465753, + "grad_norm": 2.245286464691162, + "learning_rate": 3.2432981316003253e-06, + "loss": 1.8209, + "step": 1154 + }, + { + "epoch": 1.2657534246575342, + "grad_norm": 2.292067527770996, + "learning_rate": 3.241267262388302e-06, + "loss": 1.8285, + "step": 1155 + }, + { + "epoch": 1.2668493150684932, + "grad_norm": 2.3058111667633057, + "learning_rate": 3.23923639317628e-06, + "loss": 1.7842, + "step": 1156 + }, + { + "epoch": 1.267945205479452, + "grad_norm": 2.306243896484375, + "learning_rate": 3.2372055239642567e-06, + "loss": 1.8221, + "step": 1157 + }, + { + "epoch": 1.269041095890411, + "grad_norm": 2.399759531021118, + "learning_rate": 3.2351746547522344e-06, + "loss": 1.7836, + "step": 1158 + }, + { + "epoch": 1.2701369863013698, + "grad_norm": 2.3405187129974365, + "learning_rate": 3.2331437855402113e-06, + "loss": 1.8365, + "step": 1159 + }, + { + "epoch": 1.2712328767123289, + "grad_norm": 2.3855013847351074, + "learning_rate": 3.2311129163281885e-06, + "loss": 1.8973, + "step": 1160 + }, + { + "epoch": 1.2712328767123289, + "eval_loss": 2.005983591079712, + "eval_runtime": 37.3061, + "eval_samples_per_second": 26.805, + "eval_steps_per_second": 1.126, + "step": 1160 + }, + { + "epoch": 1.2723287671232877, + "grad_norm": 2.2767140865325928, + "learning_rate": 3.229082047116166e-06, + "loss": 1.7791, + "step": 1161 + }, + { + "epoch": 1.2734246575342465, + "grad_norm": 2.289494276046753, + "learning_rate": 3.227051177904143e-06, + "loss": 1.8508, + "step": 1162 + }, + { + "epoch": 1.2745205479452055, + "grad_norm": 2.1836023330688477, + "learning_rate": 3.2250203086921204e-06, + "loss": 1.8275, + "step": 1163 + }, + { + "epoch": 1.2756164383561643, + "grad_norm": 2.1917505264282227, + "learning_rate": 3.2229894394800977e-06, + "loss": 1.8289, + "step": 1164 + }, + { + "epoch": 1.2767123287671232, + "grad_norm": 2.315983295440674, + "learning_rate": 3.220958570268075e-06, + "loss": 1.8655, + "step": 1165 + }, + { + "epoch": 1.2778082191780822, + "grad_norm": 2.3628008365631104, + "learning_rate": 3.218927701056052e-06, + "loss": 1.7396, + "step": 1166 + }, + { + "epoch": 1.2789041095890412, + "grad_norm": 2.3200488090515137, + "learning_rate": 3.2168968318440295e-06, + "loss": 1.8196, + "step": 1167 + }, + { + "epoch": 1.28, + "grad_norm": 2.2591991424560547, + "learning_rate": 3.2148659626320068e-06, + "loss": 1.7323, + "step": 1168 + }, + { + "epoch": 1.2810958904109588, + "grad_norm": 2.2458770275115967, + "learning_rate": 3.212835093419984e-06, + "loss": 1.8881, + "step": 1169 + }, + { + "epoch": 1.2821917808219179, + "grad_norm": 2.202010154724121, + "learning_rate": 3.2108042242079613e-06, + "loss": 1.7776, + "step": 1170 + }, + { + "epoch": 1.2821917808219179, + "eval_loss": 2.005174398422241, + "eval_runtime": 37.3597, + "eval_samples_per_second": 26.767, + "eval_steps_per_second": 1.124, + "step": 1170 + }, + { + "epoch": 1.2832876712328767, + "grad_norm": 2.2375359535217285, + "learning_rate": 3.2087733549959386e-06, + "loss": 1.7994, + "step": 1171 + }, + { + "epoch": 1.2843835616438355, + "grad_norm": 2.3152217864990234, + "learning_rate": 3.206742485783916e-06, + "loss": 1.8767, + "step": 1172 + }, + { + "epoch": 1.2854794520547945, + "grad_norm": 2.3317618370056152, + "learning_rate": 3.2047116165718927e-06, + "loss": 1.772, + "step": 1173 + }, + { + "epoch": 1.2865753424657536, + "grad_norm": 2.249943971633911, + "learning_rate": 3.2026807473598704e-06, + "loss": 1.7793, + "step": 1174 + }, + { + "epoch": 1.2876712328767124, + "grad_norm": 2.320481538772583, + "learning_rate": 3.2006498781478473e-06, + "loss": 1.8275, + "step": 1175 + }, + { + "epoch": 1.2887671232876712, + "grad_norm": 2.21785569190979, + "learning_rate": 3.198619008935825e-06, + "loss": 1.8449, + "step": 1176 + }, + { + "epoch": 1.2898630136986302, + "grad_norm": 2.469965934753418, + "learning_rate": 3.196588139723802e-06, + "loss": 1.779, + "step": 1177 + }, + { + "epoch": 1.290958904109589, + "grad_norm": 2.1989049911499023, + "learning_rate": 3.1945572705117795e-06, + "loss": 1.8188, + "step": 1178 + }, + { + "epoch": 1.2920547945205478, + "grad_norm": 2.293513298034668, + "learning_rate": 3.1925264012997564e-06, + "loss": 1.8377, + "step": 1179 + }, + { + "epoch": 1.2931506849315069, + "grad_norm": 2.3462443351745605, + "learning_rate": 3.190495532087734e-06, + "loss": 1.8418, + "step": 1180 + }, + { + "epoch": 1.2931506849315069, + "eval_loss": 2.0061001777648926, + "eval_runtime": 37.2951, + "eval_samples_per_second": 26.813, + "eval_steps_per_second": 1.126, + "step": 1180 + }, + { + "epoch": 1.2942465753424657, + "grad_norm": 2.2071478366851807, + "learning_rate": 3.188464662875711e-06, + "loss": 1.8419, + "step": 1181 + }, + { + "epoch": 1.2953424657534247, + "grad_norm": 2.233333110809326, + "learning_rate": 3.1864337936636886e-06, + "loss": 1.8931, + "step": 1182 + }, + { + "epoch": 1.2964383561643835, + "grad_norm": 2.383085250854492, + "learning_rate": 3.1844029244516655e-06, + "loss": 1.8047, + "step": 1183 + }, + { + "epoch": 1.2975342465753426, + "grad_norm": 2.473808765411377, + "learning_rate": 3.182372055239643e-06, + "loss": 1.7794, + "step": 1184 + }, + { + "epoch": 1.2986301369863014, + "grad_norm": 2.3357670307159424, + "learning_rate": 3.18034118602762e-06, + "loss": 1.7717, + "step": 1185 + }, + { + "epoch": 1.2997260273972602, + "grad_norm": 2.376835823059082, + "learning_rate": 3.1783103168155973e-06, + "loss": 1.8534, + "step": 1186 + }, + { + "epoch": 1.3008219178082192, + "grad_norm": 2.447415590286255, + "learning_rate": 3.1762794476035746e-06, + "loss": 1.8251, + "step": 1187 + }, + { + "epoch": 1.301917808219178, + "grad_norm": 2.5145719051361084, + "learning_rate": 3.174248578391552e-06, + "loss": 1.8117, + "step": 1188 + }, + { + "epoch": 1.303013698630137, + "grad_norm": 2.3134653568267822, + "learning_rate": 3.172217709179529e-06, + "loss": 1.8111, + "step": 1189 + }, + { + "epoch": 1.3041095890410959, + "grad_norm": 2.293311834335327, + "learning_rate": 3.1701868399675064e-06, + "loss": 1.791, + "step": 1190 + }, + { + "epoch": 1.3041095890410959, + "eval_loss": 2.0038933753967285, + "eval_runtime": 37.2282, + "eval_samples_per_second": 26.861, + "eval_steps_per_second": 1.128, + "step": 1190 + }, + { + "epoch": 1.305205479452055, + "grad_norm": 2.298020124435425, + "learning_rate": 3.1681559707554837e-06, + "loss": 1.8301, + "step": 1191 + }, + { + "epoch": 1.3063013698630137, + "grad_norm": 2.5261130332946777, + "learning_rate": 3.166125101543461e-06, + "loss": 1.7714, + "step": 1192 + }, + { + "epoch": 1.3073972602739725, + "grad_norm": 2.270820379257202, + "learning_rate": 3.1640942323314382e-06, + "loss": 1.8246, + "step": 1193 + }, + { + "epoch": 1.3084931506849315, + "grad_norm": 2.373260021209717, + "learning_rate": 3.1620633631194155e-06, + "loss": 1.8316, + "step": 1194 + }, + { + "epoch": 1.3095890410958904, + "grad_norm": 2.3265364170074463, + "learning_rate": 3.1600324939073928e-06, + "loss": 1.7679, + "step": 1195 + }, + { + "epoch": 1.3106849315068494, + "grad_norm": 2.2378532886505127, + "learning_rate": 3.15800162469537e-06, + "loss": 1.8149, + "step": 1196 + }, + { + "epoch": 1.3117808219178082, + "grad_norm": 2.2459309101104736, + "learning_rate": 3.155970755483347e-06, + "loss": 1.8086, + "step": 1197 + }, + { + "epoch": 1.3128767123287672, + "grad_norm": 2.2769811153411865, + "learning_rate": 3.1539398862713246e-06, + "loss": 1.8109, + "step": 1198 + }, + { + "epoch": 1.313972602739726, + "grad_norm": 2.310483694076538, + "learning_rate": 3.1519090170593015e-06, + "loss": 1.8297, + "step": 1199 + }, + { + "epoch": 1.3150684931506849, + "grad_norm": 2.302478313446045, + "learning_rate": 3.149878147847279e-06, + "loss": 1.8237, + "step": 1200 + }, + { + "epoch": 1.3150684931506849, + "eval_loss": 2.0011062622070312, + "eval_runtime": 37.2077, + "eval_samples_per_second": 26.876, + "eval_steps_per_second": 1.129, + "step": 1200 + }, + { + "epoch": 1.316164383561644, + "grad_norm": 2.2240796089172363, + "learning_rate": 3.147847278635256e-06, + "loss": 1.8745, + "step": 1201 + }, + { + "epoch": 1.3172602739726027, + "grad_norm": 2.303537130355835, + "learning_rate": 3.1458164094232337e-06, + "loss": 1.9051, + "step": 1202 + }, + { + "epoch": 1.3183561643835615, + "grad_norm": 2.4038636684417725, + "learning_rate": 3.1437855402112106e-06, + "loss": 1.7974, + "step": 1203 + }, + { + "epoch": 1.3194520547945205, + "grad_norm": 2.3458874225616455, + "learning_rate": 3.1417546709991883e-06, + "loss": 1.7601, + "step": 1204 + }, + { + "epoch": 1.3205479452054796, + "grad_norm": 2.3399100303649902, + "learning_rate": 3.139723801787165e-06, + "loss": 1.8314, + "step": 1205 + }, + { + "epoch": 1.3216438356164384, + "grad_norm": 2.293229818344116, + "learning_rate": 3.137692932575143e-06, + "loss": 1.8117, + "step": 1206 + }, + { + "epoch": 1.3227397260273972, + "grad_norm": 2.5378005504608154, + "learning_rate": 3.1356620633631197e-06, + "loss": 1.8614, + "step": 1207 + }, + { + "epoch": 1.3238356164383562, + "grad_norm": 2.3644745349884033, + "learning_rate": 3.1336311941510974e-06, + "loss": 1.7843, + "step": 1208 + }, + { + "epoch": 1.324931506849315, + "grad_norm": 2.1592626571655273, + "learning_rate": 3.1316003249390742e-06, + "loss": 1.8128, + "step": 1209 + }, + { + "epoch": 1.3260273972602739, + "grad_norm": 2.155649185180664, + "learning_rate": 3.129569455727051e-06, + "loss": 1.8636, + "step": 1210 + }, + { + "epoch": 1.3260273972602739, + "eval_loss": 1.9952574968338013, + "eval_runtime": 37.3004, + "eval_samples_per_second": 26.809, + "eval_steps_per_second": 1.126, + "step": 1210 + }, + { + "epoch": 1.3271232876712329, + "grad_norm": 2.1906397342681885, + "learning_rate": 3.1275385865150288e-06, + "loss": 1.8583, + "step": 1211 + }, + { + "epoch": 1.3282191780821917, + "grad_norm": 2.2708194255828857, + "learning_rate": 3.1255077173030056e-06, + "loss": 1.8363, + "step": 1212 + }, + { + "epoch": 1.3293150684931507, + "grad_norm": 2.1753969192504883, + "learning_rate": 3.1234768480909833e-06, + "loss": 1.8372, + "step": 1213 + }, + { + "epoch": 1.3304109589041095, + "grad_norm": 2.229363441467285, + "learning_rate": 3.12144597887896e-06, + "loss": 1.8179, + "step": 1214 + }, + { + "epoch": 1.3315068493150686, + "grad_norm": 2.362152338027954, + "learning_rate": 3.119415109666938e-06, + "loss": 1.8596, + "step": 1215 + }, + { + "epoch": 1.3326027397260274, + "grad_norm": 2.3278255462646484, + "learning_rate": 3.1173842404549147e-06, + "loss": 1.8778, + "step": 1216 + }, + { + "epoch": 1.3336986301369862, + "grad_norm": 2.477710247039795, + "learning_rate": 3.1153533712428924e-06, + "loss": 1.8297, + "step": 1217 + }, + { + "epoch": 1.3347945205479452, + "grad_norm": 2.2611043453216553, + "learning_rate": 3.1133225020308693e-06, + "loss": 1.8374, + "step": 1218 + }, + { + "epoch": 1.335890410958904, + "grad_norm": 2.198275566101074, + "learning_rate": 3.111291632818847e-06, + "loss": 1.8686, + "step": 1219 + }, + { + "epoch": 1.336986301369863, + "grad_norm": 2.227508068084717, + "learning_rate": 3.109260763606824e-06, + "loss": 1.8642, + "step": 1220 + }, + { + "epoch": 1.336986301369863, + "eval_loss": 1.9998422861099243, + "eval_runtime": 37.5152, + "eval_samples_per_second": 26.656, + "eval_steps_per_second": 1.12, + "step": 1220 + }, + { + "epoch": 1.3380821917808219, + "grad_norm": 2.386310338973999, + "learning_rate": 3.1072298943948015e-06, + "loss": 1.8347, + "step": 1221 + }, + { + "epoch": 1.339178082191781, + "grad_norm": 2.290313482284546, + "learning_rate": 3.1051990251827784e-06, + "loss": 1.8455, + "step": 1222 + }, + { + "epoch": 1.3402739726027397, + "grad_norm": 2.3063457012176514, + "learning_rate": 3.1031681559707557e-06, + "loss": 1.811, + "step": 1223 + }, + { + "epoch": 1.3413698630136985, + "grad_norm": 2.297840118408203, + "learning_rate": 3.101137286758733e-06, + "loss": 1.8558, + "step": 1224 + }, + { + "epoch": 1.3424657534246576, + "grad_norm": 2.40864896774292, + "learning_rate": 3.09910641754671e-06, + "loss": 1.8604, + "step": 1225 + }, + { + "epoch": 1.3435616438356164, + "grad_norm": 2.397599458694458, + "learning_rate": 3.0970755483346875e-06, + "loss": 1.8556, + "step": 1226 + }, + { + "epoch": 1.3446575342465754, + "grad_norm": 2.409029245376587, + "learning_rate": 3.0950446791226648e-06, + "loss": 1.8424, + "step": 1227 + }, + { + "epoch": 1.3457534246575342, + "grad_norm": 2.2789814472198486, + "learning_rate": 3.093013809910642e-06, + "loss": 1.8945, + "step": 1228 + }, + { + "epoch": 1.3468493150684933, + "grad_norm": 2.3749048709869385, + "learning_rate": 3.0909829406986193e-06, + "loss": 1.8387, + "step": 1229 + }, + { + "epoch": 1.347945205479452, + "grad_norm": 2.44486141204834, + "learning_rate": 3.0889520714865966e-06, + "loss": 1.8838, + "step": 1230 + }, + { + "epoch": 1.347945205479452, + "eval_loss": 1.9961321353912354, + "eval_runtime": 37.6273, + "eval_samples_per_second": 26.576, + "eval_steps_per_second": 1.116, + "step": 1230 + }, + { + "epoch": 1.3490410958904109, + "grad_norm": 2.2852954864501953, + "learning_rate": 3.086921202274574e-06, + "loss": 1.8551, + "step": 1231 + }, + { + "epoch": 1.35013698630137, + "grad_norm": 2.3131206035614014, + "learning_rate": 3.084890333062551e-06, + "loss": 1.8183, + "step": 1232 + }, + { + "epoch": 1.3512328767123287, + "grad_norm": 2.4095048904418945, + "learning_rate": 3.0828594638505284e-06, + "loss": 1.7946, + "step": 1233 + }, + { + "epoch": 1.3523287671232875, + "grad_norm": 2.429971933364868, + "learning_rate": 3.0808285946385057e-06, + "loss": 1.8349, + "step": 1234 + }, + { + "epoch": 1.3534246575342466, + "grad_norm": 2.2360565662384033, + "learning_rate": 3.078797725426483e-06, + "loss": 1.794, + "step": 1235 + }, + { + "epoch": 1.3545205479452056, + "grad_norm": 2.3630571365356445, + "learning_rate": 3.07676685621446e-06, + "loss": 1.7377, + "step": 1236 + }, + { + "epoch": 1.3556164383561644, + "grad_norm": 2.2604660987854004, + "learning_rate": 3.0747359870024375e-06, + "loss": 1.8239, + "step": 1237 + }, + { + "epoch": 1.3567123287671232, + "grad_norm": 2.209717035293579, + "learning_rate": 3.0727051177904144e-06, + "loss": 1.7835, + "step": 1238 + }, + { + "epoch": 1.3578082191780823, + "grad_norm": 2.3221793174743652, + "learning_rate": 3.070674248578392e-06, + "loss": 1.8192, + "step": 1239 + }, + { + "epoch": 1.358904109589041, + "grad_norm": 2.357313394546509, + "learning_rate": 3.068643379366369e-06, + "loss": 1.8246, + "step": 1240 + }, + { + "epoch": 1.358904109589041, + "eval_loss": 1.9935922622680664, + "eval_runtime": 37.2832, + "eval_samples_per_second": 26.822, + "eval_steps_per_second": 1.127, + "step": 1240 + }, + { + "epoch": 1.3599999999999999, + "grad_norm": 2.2563390731811523, + "learning_rate": 3.0666125101543466e-06, + "loss": 1.8175, + "step": 1241 + }, + { + "epoch": 1.361095890410959, + "grad_norm": 2.2531445026397705, + "learning_rate": 3.0645816409423235e-06, + "loss": 1.7713, + "step": 1242 + }, + { + "epoch": 1.362191780821918, + "grad_norm": 2.2369470596313477, + "learning_rate": 3.062550771730301e-06, + "loss": 1.8377, + "step": 1243 + }, + { + "epoch": 1.3632876712328768, + "grad_norm": 2.4063796997070312, + "learning_rate": 3.060519902518278e-06, + "loss": 1.8116, + "step": 1244 + }, + { + "epoch": 1.3643835616438356, + "grad_norm": 2.4628145694732666, + "learning_rate": 3.0584890333062557e-06, + "loss": 1.7998, + "step": 1245 + }, + { + "epoch": 1.3654794520547946, + "grad_norm": 2.3018743991851807, + "learning_rate": 3.0564581640942326e-06, + "loss": 1.7413, + "step": 1246 + }, + { + "epoch": 1.3665753424657534, + "grad_norm": 2.2432262897491455, + "learning_rate": 3.0544272948822094e-06, + "loss": 1.82, + "step": 1247 + }, + { + "epoch": 1.3676712328767122, + "grad_norm": 2.368435859680176, + "learning_rate": 3.052396425670187e-06, + "loss": 1.7976, + "step": 1248 + }, + { + "epoch": 1.3687671232876713, + "grad_norm": 2.2668025493621826, + "learning_rate": 3.050365556458164e-06, + "loss": 1.8379, + "step": 1249 + }, + { + "epoch": 1.36986301369863, + "grad_norm": 2.2622179985046387, + "learning_rate": 3.0483346872461417e-06, + "loss": 1.794, + "step": 1250 + }, + { + "epoch": 1.36986301369863, + "eval_loss": 1.9959099292755127, + "eval_runtime": 37.3358, + "eval_samples_per_second": 26.784, + "eval_steps_per_second": 1.125, + "step": 1250 + }, + { + "epoch": 1.370958904109589, + "grad_norm": 2.3414738178253174, + "learning_rate": 3.0463038180341185e-06, + "loss": 1.7993, + "step": 1251 + }, + { + "epoch": 1.372054794520548, + "grad_norm": 2.2678418159484863, + "learning_rate": 3.0442729488220962e-06, + "loss": 1.8581, + "step": 1252 + }, + { + "epoch": 1.373150684931507, + "grad_norm": 2.268195867538452, + "learning_rate": 3.042242079610073e-06, + "loss": 1.8486, + "step": 1253 + }, + { + "epoch": 1.3742465753424657, + "grad_norm": 2.230173110961914, + "learning_rate": 3.0402112103980508e-06, + "loss": 1.8468, + "step": 1254 + }, + { + "epoch": 1.3753424657534246, + "grad_norm": 2.3020238876342773, + "learning_rate": 3.0381803411860276e-06, + "loss": 1.7623, + "step": 1255 + }, + { + "epoch": 1.3764383561643836, + "grad_norm": 2.385343313217163, + "learning_rate": 3.0361494719740053e-06, + "loss": 1.7976, + "step": 1256 + }, + { + "epoch": 1.3775342465753424, + "grad_norm": 2.344768524169922, + "learning_rate": 3.034118602761982e-06, + "loss": 1.807, + "step": 1257 + }, + { + "epoch": 1.3786301369863014, + "grad_norm": 2.411760091781616, + "learning_rate": 3.03208773354996e-06, + "loss": 1.8079, + "step": 1258 + }, + { + "epoch": 1.3797260273972602, + "grad_norm": 2.3160128593444824, + "learning_rate": 3.0300568643379367e-06, + "loss": 1.7868, + "step": 1259 + }, + { + "epoch": 1.3808219178082193, + "grad_norm": 2.334632635116577, + "learning_rate": 3.028025995125914e-06, + "loss": 1.8466, + "step": 1260 + }, + { + "epoch": 1.3808219178082193, + "eval_loss": 1.9903327226638794, + "eval_runtime": 37.2256, + "eval_samples_per_second": 26.863, + "eval_steps_per_second": 1.128, + "step": 1260 + }, + { + "epoch": 1.381917808219178, + "grad_norm": 2.3443334102630615, + "learning_rate": 3.0259951259138913e-06, + "loss": 1.8302, + "step": 1261 + }, + { + "epoch": 1.383013698630137, + "grad_norm": 2.311563014984131, + "learning_rate": 3.0239642567018686e-06, + "loss": 1.7727, + "step": 1262 + }, + { + "epoch": 1.384109589041096, + "grad_norm": 2.2501776218414307, + "learning_rate": 3.021933387489846e-06, + "loss": 1.8058, + "step": 1263 + }, + { + "epoch": 1.3852054794520547, + "grad_norm": 2.259157657623291, + "learning_rate": 3.019902518277823e-06, + "loss": 1.7627, + "step": 1264 + }, + { + "epoch": 1.3863013698630138, + "grad_norm": 2.4380314350128174, + "learning_rate": 3.0178716490658004e-06, + "loss": 1.8237, + "step": 1265 + }, + { + "epoch": 1.3873972602739726, + "grad_norm": 2.3203837871551514, + "learning_rate": 3.0158407798537777e-06, + "loss": 1.8557, + "step": 1266 + }, + { + "epoch": 1.3884931506849316, + "grad_norm": 2.220996141433716, + "learning_rate": 3.013809910641755e-06, + "loss": 1.7641, + "step": 1267 + }, + { + "epoch": 1.3895890410958904, + "grad_norm": 2.190006971359253, + "learning_rate": 3.0117790414297322e-06, + "loss": 1.7705, + "step": 1268 + }, + { + "epoch": 1.3906849315068492, + "grad_norm": 2.229748249053955, + "learning_rate": 3.0097481722177095e-06, + "loss": 1.7709, + "step": 1269 + }, + { + "epoch": 1.3917808219178083, + "grad_norm": 2.3755619525909424, + "learning_rate": 3.0077173030056868e-06, + "loss": 1.8172, + "step": 1270 + }, + { + "epoch": 1.3917808219178083, + "eval_loss": 1.992952585220337, + "eval_runtime": 37.2403, + "eval_samples_per_second": 26.853, + "eval_steps_per_second": 1.128, + "step": 1270 + }, + { + "epoch": 1.392876712328767, + "grad_norm": 2.2514004707336426, + "learning_rate": 3.005686433793664e-06, + "loss": 1.7719, + "step": 1271 + }, + { + "epoch": 1.393972602739726, + "grad_norm": 2.3010191917419434, + "learning_rate": 3.0036555645816413e-06, + "loss": 1.8369, + "step": 1272 + }, + { + "epoch": 1.395068493150685, + "grad_norm": 2.68790602684021, + "learning_rate": 3.001624695369618e-06, + "loss": 1.7974, + "step": 1273 + }, + { + "epoch": 1.396164383561644, + "grad_norm": 2.454904794692993, + "learning_rate": 2.999593826157596e-06, + "loss": 1.8509, + "step": 1274 + }, + { + "epoch": 1.3972602739726028, + "grad_norm": 2.251556396484375, + "learning_rate": 2.9975629569455727e-06, + "loss": 1.8011, + "step": 1275 + }, + { + "epoch": 1.3983561643835616, + "grad_norm": 2.3789098262786865, + "learning_rate": 2.9955320877335504e-06, + "loss": 1.8488, + "step": 1276 + }, + { + "epoch": 1.3994520547945206, + "grad_norm": 2.2659196853637695, + "learning_rate": 2.9935012185215273e-06, + "loss": 1.7946, + "step": 1277 + }, + { + "epoch": 1.4005479452054794, + "grad_norm": 2.2148971557617188, + "learning_rate": 2.991470349309505e-06, + "loss": 1.8388, + "step": 1278 + }, + { + "epoch": 1.4016438356164382, + "grad_norm": 2.1883089542388916, + "learning_rate": 2.989439480097482e-06, + "loss": 1.7925, + "step": 1279 + }, + { + "epoch": 1.4027397260273973, + "grad_norm": 2.290208339691162, + "learning_rate": 2.9874086108854595e-06, + "loss": 1.8099, + "step": 1280 + }, + { + "epoch": 1.4027397260273973, + "eval_loss": 1.9924782514572144, + "eval_runtime": 37.1891, + "eval_samples_per_second": 26.89, + "eval_steps_per_second": 1.129, + "step": 1280 + }, + { + "epoch": 1.403835616438356, + "grad_norm": 2.121649980545044, + "learning_rate": 2.9853777416734364e-06, + "loss": 1.7991, + "step": 1281 + }, + { + "epoch": 1.4049315068493151, + "grad_norm": 2.3528690338134766, + "learning_rate": 2.983346872461414e-06, + "loss": 1.8487, + "step": 1282 + }, + { + "epoch": 1.406027397260274, + "grad_norm": 2.391491174697876, + "learning_rate": 2.981316003249391e-06, + "loss": 1.8564, + "step": 1283 + }, + { + "epoch": 1.407123287671233, + "grad_norm": 2.336615800857544, + "learning_rate": 2.9792851340373678e-06, + "loss": 1.8216, + "step": 1284 + }, + { + "epoch": 1.4082191780821918, + "grad_norm": 2.2958548069000244, + "learning_rate": 2.9772542648253455e-06, + "loss": 1.8003, + "step": 1285 + }, + { + "epoch": 1.4093150684931506, + "grad_norm": 2.2062184810638428, + "learning_rate": 2.9752233956133223e-06, + "loss": 1.8488, + "step": 1286 + }, + { + "epoch": 1.4104109589041096, + "grad_norm": 2.486133575439453, + "learning_rate": 2.9731925264013e-06, + "loss": 1.7914, + "step": 1287 + }, + { + "epoch": 1.4115068493150684, + "grad_norm": 2.2933201789855957, + "learning_rate": 2.971161657189277e-06, + "loss": 1.8164, + "step": 1288 + }, + { + "epoch": 1.4126027397260275, + "grad_norm": 2.3301920890808105, + "learning_rate": 2.9691307879772546e-06, + "loss": 1.7883, + "step": 1289 + }, + { + "epoch": 1.4136986301369863, + "grad_norm": 2.248832941055298, + "learning_rate": 2.9670999187652314e-06, + "loss": 1.7916, + "step": 1290 + }, + { + "epoch": 1.4136986301369863, + "eval_loss": 1.9919097423553467, + "eval_runtime": 37.1283, + "eval_samples_per_second": 26.934, + "eval_steps_per_second": 1.131, + "step": 1290 + }, + { + "epoch": 1.4147945205479453, + "grad_norm": 2.37760853767395, + "learning_rate": 2.965069049553209e-06, + "loss": 1.7854, + "step": 1291 + }, + { + "epoch": 1.415890410958904, + "grad_norm": 2.327718734741211, + "learning_rate": 2.963038180341186e-06, + "loss": 1.874, + "step": 1292 + }, + { + "epoch": 1.416986301369863, + "grad_norm": 2.2413878440856934, + "learning_rate": 2.9610073111291637e-06, + "loss": 1.7591, + "step": 1293 + }, + { + "epoch": 1.418082191780822, + "grad_norm": 2.2203164100646973, + "learning_rate": 2.9589764419171405e-06, + "loss": 1.8466, + "step": 1294 + }, + { + "epoch": 1.4191780821917808, + "grad_norm": 2.2070505619049072, + "learning_rate": 2.9569455727051182e-06, + "loss": 1.8689, + "step": 1295 + }, + { + "epoch": 1.4202739726027398, + "grad_norm": 2.2122223377227783, + "learning_rate": 2.954914703493095e-06, + "loss": 1.7689, + "step": 1296 + }, + { + "epoch": 1.4213698630136986, + "grad_norm": 2.331303358078003, + "learning_rate": 2.9528838342810724e-06, + "loss": 1.8089, + "step": 1297 + }, + { + "epoch": 1.4224657534246576, + "grad_norm": 2.408184289932251, + "learning_rate": 2.9508529650690496e-06, + "loss": 1.7847, + "step": 1298 + }, + { + "epoch": 1.4235616438356165, + "grad_norm": 2.212766170501709, + "learning_rate": 2.948822095857027e-06, + "loss": 1.7899, + "step": 1299 + }, + { + "epoch": 1.4246575342465753, + "grad_norm": 2.259953737258911, + "learning_rate": 2.946791226645004e-06, + "loss": 1.8435, + "step": 1300 + }, + { + "epoch": 1.4246575342465753, + "eval_loss": 1.9893293380737305, + "eval_runtime": 37.1675, + "eval_samples_per_second": 26.905, + "eval_steps_per_second": 1.13, + "step": 1300 + }, + { + "epoch": 1.4257534246575343, + "grad_norm": 2.342400312423706, + "learning_rate": 2.9447603574329815e-06, + "loss": 1.7779, + "step": 1301 + }, + { + "epoch": 1.426849315068493, + "grad_norm": 2.237618923187256, + "learning_rate": 2.9427294882209587e-06, + "loss": 1.7509, + "step": 1302 + }, + { + "epoch": 1.427945205479452, + "grad_norm": 2.3418633937835693, + "learning_rate": 2.940698619008936e-06, + "loss": 1.7963, + "step": 1303 + }, + { + "epoch": 1.429041095890411, + "grad_norm": 2.470827579498291, + "learning_rate": 2.9386677497969133e-06, + "loss": 1.8045, + "step": 1304 + }, + { + "epoch": 1.43013698630137, + "grad_norm": 2.508345365524292, + "learning_rate": 2.9366368805848906e-06, + "loss": 1.7926, + "step": 1305 + }, + { + "epoch": 1.4312328767123288, + "grad_norm": 2.261319160461426, + "learning_rate": 2.934606011372868e-06, + "loss": 1.831, + "step": 1306 + }, + { + "epoch": 1.4323287671232876, + "grad_norm": 2.2703914642333984, + "learning_rate": 2.932575142160845e-06, + "loss": 1.8352, + "step": 1307 + }, + { + "epoch": 1.4334246575342466, + "grad_norm": 2.243626117706299, + "learning_rate": 2.9305442729488224e-06, + "loss": 1.8162, + "step": 1308 + }, + { + "epoch": 1.4345205479452054, + "grad_norm": 2.3859238624572754, + "learning_rate": 2.9285134037367997e-06, + "loss": 1.8171, + "step": 1309 + }, + { + "epoch": 1.4356164383561643, + "grad_norm": 2.3350934982299805, + "learning_rate": 2.9264825345247765e-06, + "loss": 1.8066, + "step": 1310 + }, + { + "epoch": 1.4356164383561643, + "eval_loss": 1.9905259609222412, + "eval_runtime": 37.2009, + "eval_samples_per_second": 26.881, + "eval_steps_per_second": 1.129, + "step": 1310 + }, + { + "epoch": 1.4367123287671233, + "grad_norm": 2.1943347454071045, + "learning_rate": 2.9244516653127542e-06, + "loss": 1.8435, + "step": 1311 + }, + { + "epoch": 1.4378082191780823, + "grad_norm": 2.429128885269165, + "learning_rate": 2.922420796100731e-06, + "loss": 1.8179, + "step": 1312 + }, + { + "epoch": 1.4389041095890411, + "grad_norm": 2.3677234649658203, + "learning_rate": 2.9203899268887088e-06, + "loss": 1.7245, + "step": 1313 + }, + { + "epoch": 1.44, + "grad_norm": 2.255570888519287, + "learning_rate": 2.9183590576766856e-06, + "loss": 1.8545, + "step": 1314 + }, + { + "epoch": 1.441095890410959, + "grad_norm": 2.5176682472229004, + "learning_rate": 2.9163281884646633e-06, + "loss": 1.7509, + "step": 1315 + }, + { + "epoch": 1.4421917808219178, + "grad_norm": 2.29972767829895, + "learning_rate": 2.91429731925264e-06, + "loss": 1.8116, + "step": 1316 + }, + { + "epoch": 1.4432876712328766, + "grad_norm": 2.2853286266326904, + "learning_rate": 2.912266450040618e-06, + "loss": 1.8552, + "step": 1317 + }, + { + "epoch": 1.4443835616438356, + "grad_norm": 2.505075216293335, + "learning_rate": 2.9102355808285947e-06, + "loss": 1.811, + "step": 1318 + }, + { + "epoch": 1.4454794520547944, + "grad_norm": 2.229064464569092, + "learning_rate": 2.9082047116165724e-06, + "loss": 1.8068, + "step": 1319 + }, + { + "epoch": 1.4465753424657535, + "grad_norm": 2.199187994003296, + "learning_rate": 2.9061738424045493e-06, + "loss": 1.7825, + "step": 1320 + }, + { + "epoch": 1.4465753424657535, + "eval_loss": 1.9851222038269043, + "eval_runtime": 37.4614, + "eval_samples_per_second": 26.694, + "eval_steps_per_second": 1.121, + "step": 1320 + }, + { + "epoch": 1.4476712328767123, + "grad_norm": 2.4318959712982178, + "learning_rate": 2.904142973192527e-06, + "loss": 1.77, + "step": 1321 + }, + { + "epoch": 1.4487671232876713, + "grad_norm": 2.480703592300415, + "learning_rate": 2.902112103980504e-06, + "loss": 1.8774, + "step": 1322 + }, + { + "epoch": 1.4498630136986301, + "grad_norm": 2.367624521255493, + "learning_rate": 2.900081234768481e-06, + "loss": 1.818, + "step": 1323 + }, + { + "epoch": 1.450958904109589, + "grad_norm": 2.245974540710449, + "learning_rate": 2.8980503655564584e-06, + "loss": 1.8054, + "step": 1324 + }, + { + "epoch": 1.452054794520548, + "grad_norm": 2.352687120437622, + "learning_rate": 2.8960194963444357e-06, + "loss": 1.7922, + "step": 1325 + }, + { + "epoch": 1.4531506849315068, + "grad_norm": 2.308253526687622, + "learning_rate": 2.893988627132413e-06, + "loss": 1.7637, + "step": 1326 + }, + { + "epoch": 1.4542465753424658, + "grad_norm": 2.309495687484741, + "learning_rate": 2.8919577579203902e-06, + "loss": 1.7905, + "step": 1327 + }, + { + "epoch": 1.4553424657534246, + "grad_norm": 2.354365348815918, + "learning_rate": 2.8899268887083675e-06, + "loss": 1.7321, + "step": 1328 + }, + { + "epoch": 1.4564383561643837, + "grad_norm": 2.3886210918426514, + "learning_rate": 2.8878960194963448e-06, + "loss": 1.7932, + "step": 1329 + }, + { + "epoch": 1.4575342465753425, + "grad_norm": 2.395714044570923, + "learning_rate": 2.885865150284322e-06, + "loss": 1.813, + "step": 1330 + }, + { + "epoch": 1.4575342465753425, + "eval_loss": 1.9851990938186646, + "eval_runtime": 37.2966, + "eval_samples_per_second": 26.812, + "eval_steps_per_second": 1.126, + "step": 1330 + }, + { + "epoch": 1.4586301369863013, + "grad_norm": 2.433069944381714, + "learning_rate": 2.8838342810722993e-06, + "loss": 1.8346, + "step": 1331 + }, + { + "epoch": 1.4597260273972603, + "grad_norm": 2.295894145965576, + "learning_rate": 2.8818034118602766e-06, + "loss": 1.7855, + "step": 1332 + }, + { + "epoch": 1.4608219178082191, + "grad_norm": 2.495742082595825, + "learning_rate": 2.879772542648254e-06, + "loss": 1.7601, + "step": 1333 + }, + { + "epoch": 1.4619178082191782, + "grad_norm": 2.35984206199646, + "learning_rate": 2.8777416734362307e-06, + "loss": 1.8329, + "step": 1334 + }, + { + "epoch": 1.463013698630137, + "grad_norm": 2.39632248878479, + "learning_rate": 2.8757108042242084e-06, + "loss": 1.7426, + "step": 1335 + }, + { + "epoch": 1.464109589041096, + "grad_norm": 2.4151153564453125, + "learning_rate": 2.8736799350121853e-06, + "loss": 1.8899, + "step": 1336 + }, + { + "epoch": 1.4652054794520548, + "grad_norm": 2.43914532661438, + "learning_rate": 2.871649065800163e-06, + "loss": 1.8464, + "step": 1337 + }, + { + "epoch": 1.4663013698630136, + "grad_norm": 2.44490122795105, + "learning_rate": 2.86961819658814e-06, + "loss": 1.8643, + "step": 1338 + }, + { + "epoch": 1.4673972602739727, + "grad_norm": 2.2639644145965576, + "learning_rate": 2.8675873273761175e-06, + "loss": 1.792, + "step": 1339 + }, + { + "epoch": 1.4684931506849315, + "grad_norm": 2.2803847789764404, + "learning_rate": 2.8655564581640944e-06, + "loss": 1.8109, + "step": 1340 + }, + { + "epoch": 1.4684931506849315, + "eval_loss": 1.9856781959533691, + "eval_runtime": 37.2519, + "eval_samples_per_second": 26.844, + "eval_steps_per_second": 1.127, + "step": 1340 + }, + { + "epoch": 1.4695890410958903, + "grad_norm": 2.494727849960327, + "learning_rate": 2.863525588952072e-06, + "loss": 1.8032, + "step": 1341 + }, + { + "epoch": 1.4706849315068493, + "grad_norm": 2.239595413208008, + "learning_rate": 2.861494719740049e-06, + "loss": 1.8659, + "step": 1342 + }, + { + "epoch": 1.4717808219178083, + "grad_norm": 2.2922589778900146, + "learning_rate": 2.8594638505280266e-06, + "loss": 1.7587, + "step": 1343 + }, + { + "epoch": 1.4728767123287672, + "grad_norm": 2.345224142074585, + "learning_rate": 2.8574329813160035e-06, + "loss": 1.7602, + "step": 1344 + }, + { + "epoch": 1.473972602739726, + "grad_norm": 2.3415307998657227, + "learning_rate": 2.855402112103981e-06, + "loss": 1.8044, + "step": 1345 + }, + { + "epoch": 1.475068493150685, + "grad_norm": 2.331557035446167, + "learning_rate": 2.853371242891958e-06, + "loss": 1.7808, + "step": 1346 + }, + { + "epoch": 1.4761643835616438, + "grad_norm": 2.293572425842285, + "learning_rate": 2.851340373679935e-06, + "loss": 1.8017, + "step": 1347 + }, + { + "epoch": 1.4772602739726026, + "grad_norm": 2.3477773666381836, + "learning_rate": 2.8493095044679126e-06, + "loss": 1.8113, + "step": 1348 + }, + { + "epoch": 1.4783561643835617, + "grad_norm": 2.3740994930267334, + "learning_rate": 2.8472786352558894e-06, + "loss": 1.8801, + "step": 1349 + }, + { + "epoch": 1.4794520547945205, + "grad_norm": 2.4202206134796143, + "learning_rate": 2.845247766043867e-06, + "loss": 1.821, + "step": 1350 + }, + { + "epoch": 1.4794520547945205, + "eval_loss": 1.9867810010910034, + "eval_runtime": 37.2597, + "eval_samples_per_second": 26.839, + "eval_steps_per_second": 1.127, + "step": 1350 + }, + { + "epoch": 1.4805479452054795, + "grad_norm": 2.2323198318481445, + "learning_rate": 2.843216896831844e-06, + "loss": 1.7574, + "step": 1351 + }, + { + "epoch": 1.4816438356164383, + "grad_norm": 2.2192625999450684, + "learning_rate": 2.8411860276198217e-06, + "loss": 1.7662, + "step": 1352 + }, + { + "epoch": 1.4827397260273973, + "grad_norm": 2.276507616043091, + "learning_rate": 2.8391551584077985e-06, + "loss": 1.7748, + "step": 1353 + }, + { + "epoch": 1.4838356164383562, + "grad_norm": 2.292773723602295, + "learning_rate": 2.8371242891957762e-06, + "loss": 1.8388, + "step": 1354 + }, + { + "epoch": 1.484931506849315, + "grad_norm": 2.3127009868621826, + "learning_rate": 2.835093419983753e-06, + "loss": 1.861, + "step": 1355 + }, + { + "epoch": 1.486027397260274, + "grad_norm": 2.2752184867858887, + "learning_rate": 2.833062550771731e-06, + "loss": 1.7883, + "step": 1356 + }, + { + "epoch": 1.4871232876712328, + "grad_norm": 2.288158416748047, + "learning_rate": 2.8310316815597076e-06, + "loss": 1.7963, + "step": 1357 + }, + { + "epoch": 1.4882191780821918, + "grad_norm": 2.3287482261657715, + "learning_rate": 2.8290008123476853e-06, + "loss": 1.7534, + "step": 1358 + }, + { + "epoch": 1.4893150684931507, + "grad_norm": 2.441279888153076, + "learning_rate": 2.826969943135662e-06, + "loss": 1.8073, + "step": 1359 + }, + { + "epoch": 1.4904109589041097, + "grad_norm": 2.3399839401245117, + "learning_rate": 2.8249390739236395e-06, + "loss": 1.7876, + "step": 1360 + }, + { + "epoch": 1.4904109589041097, + "eval_loss": 1.9868030548095703, + "eval_runtime": 37.2891, + "eval_samples_per_second": 26.817, + "eval_steps_per_second": 1.126, + "step": 1360 + }, + { + "epoch": 1.4915068493150685, + "grad_norm": 2.234246253967285, + "learning_rate": 2.8229082047116167e-06, + "loss": 1.8031, + "step": 1361 + }, + { + "epoch": 1.4926027397260273, + "grad_norm": 2.3136820793151855, + "learning_rate": 2.820877335499594e-06, + "loss": 1.7669, + "step": 1362 + }, + { + "epoch": 1.4936986301369863, + "grad_norm": 2.389420747756958, + "learning_rate": 2.8188464662875713e-06, + "loss": 1.7879, + "step": 1363 + }, + { + "epoch": 1.4947945205479451, + "grad_norm": 2.222306489944458, + "learning_rate": 2.8168155970755486e-06, + "loss": 1.7412, + "step": 1364 + }, + { + "epoch": 1.4958904109589042, + "grad_norm": 2.1689319610595703, + "learning_rate": 2.814784727863526e-06, + "loss": 1.7651, + "step": 1365 + }, + { + "epoch": 1.496986301369863, + "grad_norm": 2.262047529220581, + "learning_rate": 2.812753858651503e-06, + "loss": 1.7798, + "step": 1366 + }, + { + "epoch": 1.498082191780822, + "grad_norm": 2.266894817352295, + "learning_rate": 2.8107229894394804e-06, + "loss": 1.8172, + "step": 1367 + }, + { + "epoch": 1.4991780821917808, + "grad_norm": 2.6400444507598877, + "learning_rate": 2.8086921202274577e-06, + "loss": 1.7871, + "step": 1368 + }, + { + "epoch": 1.5002739726027396, + "grad_norm": 2.3868865966796875, + "learning_rate": 2.806661251015435e-06, + "loss": 1.7427, + "step": 1369 + }, + { + "epoch": 1.5013698630136987, + "grad_norm": 2.5206542015075684, + "learning_rate": 2.8046303818034122e-06, + "loss": 1.817, + "step": 1370 + }, + { + "epoch": 1.5013698630136987, + "eval_loss": 1.9870396852493286, + "eval_runtime": 37.2372, + "eval_samples_per_second": 26.855, + "eval_steps_per_second": 1.128, + "step": 1370 + }, + { + "epoch": 1.5024657534246575, + "grad_norm": 2.5236382484436035, + "learning_rate": 2.802599512591389e-06, + "loss": 1.8662, + "step": 1371 + }, + { + "epoch": 1.5035616438356163, + "grad_norm": 2.5095479488372803, + "learning_rate": 2.8005686433793668e-06, + "loss": 1.9168, + "step": 1372 + }, + { + "epoch": 1.5046575342465753, + "grad_norm": 2.307508707046509, + "learning_rate": 2.7985377741673436e-06, + "loss": 1.799, + "step": 1373 + }, + { + "epoch": 1.5057534246575344, + "grad_norm": 2.410879373550415, + "learning_rate": 2.7965069049553213e-06, + "loss": 1.7447, + "step": 1374 + }, + { + "epoch": 1.5068493150684932, + "grad_norm": 2.738558530807495, + "learning_rate": 2.794476035743298e-06, + "loss": 1.8073, + "step": 1375 + }, + { + "epoch": 1.507945205479452, + "grad_norm": 2.3826518058776855, + "learning_rate": 2.792445166531276e-06, + "loss": 1.8506, + "step": 1376 + }, + { + "epoch": 1.509041095890411, + "grad_norm": 2.334123373031616, + "learning_rate": 2.7904142973192527e-06, + "loss": 1.8209, + "step": 1377 + }, + { + "epoch": 1.5101369863013698, + "grad_norm": 2.668513298034668, + "learning_rate": 2.7883834281072304e-06, + "loss": 1.8285, + "step": 1378 + }, + { + "epoch": 1.5112328767123286, + "grad_norm": 2.471938133239746, + "learning_rate": 2.7863525588952073e-06, + "loss": 1.7668, + "step": 1379 + }, + { + "epoch": 1.5123287671232877, + "grad_norm": 2.3712313175201416, + "learning_rate": 2.784321689683185e-06, + "loss": 1.7565, + "step": 1380 + }, + { + "epoch": 1.5123287671232877, + "eval_loss": 1.9826414585113525, + "eval_runtime": 37.2426, + "eval_samples_per_second": 26.851, + "eval_steps_per_second": 1.128, + "step": 1380 + }, + { + "epoch": 1.5134246575342467, + "grad_norm": 2.3364150524139404, + "learning_rate": 2.782290820471162e-06, + "loss": 1.7596, + "step": 1381 + }, + { + "epoch": 1.5145205479452055, + "grad_norm": 2.3088393211364746, + "learning_rate": 2.7802599512591395e-06, + "loss": 1.7568, + "step": 1382 + }, + { + "epoch": 1.5156164383561643, + "grad_norm": 2.2584595680236816, + "learning_rate": 2.7782290820471164e-06, + "loss": 1.833, + "step": 1383 + }, + { + "epoch": 1.5167123287671234, + "grad_norm": 2.2440273761749268, + "learning_rate": 2.7761982128350932e-06, + "loss": 1.8237, + "step": 1384 + }, + { + "epoch": 1.5178082191780822, + "grad_norm": 2.3552212715148926, + "learning_rate": 2.774167343623071e-06, + "loss": 1.8288, + "step": 1385 + }, + { + "epoch": 1.518904109589041, + "grad_norm": 2.3001790046691895, + "learning_rate": 2.772136474411048e-06, + "loss": 1.8519, + "step": 1386 + }, + { + "epoch": 1.52, + "grad_norm": 2.3691201210021973, + "learning_rate": 2.7701056051990255e-06, + "loss": 1.7759, + "step": 1387 + }, + { + "epoch": 1.521095890410959, + "grad_norm": 2.327988624572754, + "learning_rate": 2.7680747359870023e-06, + "loss": 1.7132, + "step": 1388 + }, + { + "epoch": 1.5221917808219179, + "grad_norm": 2.30245304107666, + "learning_rate": 2.76604386677498e-06, + "loss": 1.8167, + "step": 1389 + }, + { + "epoch": 1.5232876712328767, + "grad_norm": 2.290121078491211, + "learning_rate": 2.764012997562957e-06, + "loss": 1.7913, + "step": 1390 + }, + { + "epoch": 1.5232876712328767, + "eval_loss": 1.98329758644104, + "eval_runtime": 37.2462, + "eval_samples_per_second": 26.848, + "eval_steps_per_second": 1.128, + "step": 1390 + }, + { + "epoch": 1.5243835616438357, + "grad_norm": 2.294076681137085, + "learning_rate": 2.7619821283509346e-06, + "loss": 1.8598, + "step": 1391 + }, + { + "epoch": 1.5254794520547945, + "grad_norm": 2.3325250148773193, + "learning_rate": 2.7599512591389114e-06, + "loss": 1.7392, + "step": 1392 + }, + { + "epoch": 1.5265753424657533, + "grad_norm": 2.2802610397338867, + "learning_rate": 2.757920389926889e-06, + "loss": 1.8332, + "step": 1393 + }, + { + "epoch": 1.5276712328767124, + "grad_norm": 2.284207344055176, + "learning_rate": 2.755889520714866e-06, + "loss": 1.7875, + "step": 1394 + }, + { + "epoch": 1.5287671232876714, + "grad_norm": 2.325998306274414, + "learning_rate": 2.7538586515028437e-06, + "loss": 1.7657, + "step": 1395 + }, + { + "epoch": 1.52986301369863, + "grad_norm": 2.285393476486206, + "learning_rate": 2.7518277822908205e-06, + "loss": 1.7416, + "step": 1396 + }, + { + "epoch": 1.530958904109589, + "grad_norm": 2.3110482692718506, + "learning_rate": 2.749796913078798e-06, + "loss": 1.7931, + "step": 1397 + }, + { + "epoch": 1.532054794520548, + "grad_norm": 2.4234025478363037, + "learning_rate": 2.747766043866775e-06, + "loss": 1.8472, + "step": 1398 + }, + { + "epoch": 1.5331506849315069, + "grad_norm": 2.384767532348633, + "learning_rate": 2.7457351746547524e-06, + "loss": 1.7961, + "step": 1399 + }, + { + "epoch": 1.5342465753424657, + "grad_norm": 2.1857433319091797, + "learning_rate": 2.7437043054427297e-06, + "loss": 1.8654, + "step": 1400 + }, + { + "epoch": 1.5342465753424657, + "eval_loss": 1.9802658557891846, + "eval_runtime": 37.1443, + "eval_samples_per_second": 26.922, + "eval_steps_per_second": 1.131, + "step": 1400 + }, + { + "epoch": 1.5353424657534247, + "grad_norm": 2.1302895545959473, + "learning_rate": 2.741673436230707e-06, + "loss": 1.9051, + "step": 1401 + }, + { + "epoch": 1.5364383561643835, + "grad_norm": 2.377307415008545, + "learning_rate": 2.739642567018684e-06, + "loss": 1.7942, + "step": 1402 + }, + { + "epoch": 1.5375342465753423, + "grad_norm": 2.316549062728882, + "learning_rate": 2.7376116978066615e-06, + "loss": 1.7734, + "step": 1403 + }, + { + "epoch": 1.5386301369863014, + "grad_norm": 2.3452532291412354, + "learning_rate": 2.7355808285946388e-06, + "loss": 1.8243, + "step": 1404 + }, + { + "epoch": 1.5397260273972604, + "grad_norm": 2.3139419555664062, + "learning_rate": 2.733549959382616e-06, + "loss": 1.812, + "step": 1405 + }, + { + "epoch": 1.5408219178082192, + "grad_norm": 2.3135571479797363, + "learning_rate": 2.7315190901705933e-06, + "loss": 1.7751, + "step": 1406 + }, + { + "epoch": 1.541917808219178, + "grad_norm": 2.403231620788574, + "learning_rate": 2.7294882209585706e-06, + "loss": 1.8561, + "step": 1407 + }, + { + "epoch": 1.543013698630137, + "grad_norm": 2.4028689861297607, + "learning_rate": 2.727457351746548e-06, + "loss": 1.8249, + "step": 1408 + }, + { + "epoch": 1.5441095890410959, + "grad_norm": 2.4028851985931396, + "learning_rate": 2.725426482534525e-06, + "loss": 1.779, + "step": 1409 + }, + { + "epoch": 1.5452054794520547, + "grad_norm": 2.270709753036499, + "learning_rate": 2.723395613322502e-06, + "loss": 1.7754, + "step": 1410 + }, + { + "epoch": 1.5452054794520547, + "eval_loss": 1.983272910118103, + "eval_runtime": 37.2405, + "eval_samples_per_second": 26.853, + "eval_steps_per_second": 1.128, + "step": 1410 + }, + { + "epoch": 1.5463013698630137, + "grad_norm": 2.2672812938690186, + "learning_rate": 2.7213647441104797e-06, + "loss": 1.814, + "step": 1411 + }, + { + "epoch": 1.5473972602739727, + "grad_norm": 2.3539397716522217, + "learning_rate": 2.7193338748984565e-06, + "loss": 1.7907, + "step": 1412 + }, + { + "epoch": 1.5484931506849315, + "grad_norm": 2.3229010105133057, + "learning_rate": 2.7173030056864342e-06, + "loss": 1.8061, + "step": 1413 + }, + { + "epoch": 1.5495890410958904, + "grad_norm": 2.482874870300293, + "learning_rate": 2.715272136474411e-06, + "loss": 1.9577, + "step": 1414 + }, + { + "epoch": 1.5506849315068494, + "grad_norm": 2.211798906326294, + "learning_rate": 2.713241267262389e-06, + "loss": 1.7309, + "step": 1415 + }, + { + "epoch": 1.5517808219178082, + "grad_norm": 2.3159289360046387, + "learning_rate": 2.7112103980503656e-06, + "loss": 1.8132, + "step": 1416 + }, + { + "epoch": 1.552876712328767, + "grad_norm": 2.486035108566284, + "learning_rate": 2.7091795288383433e-06, + "loss": 1.7734, + "step": 1417 + }, + { + "epoch": 1.553972602739726, + "grad_norm": 2.3025195598602295, + "learning_rate": 2.70714865962632e-06, + "loss": 1.8232, + "step": 1418 + }, + { + "epoch": 1.555068493150685, + "grad_norm": 2.234581470489502, + "learning_rate": 2.705117790414298e-06, + "loss": 1.7814, + "step": 1419 + }, + { + "epoch": 1.5561643835616439, + "grad_norm": 2.2855048179626465, + "learning_rate": 2.7030869212022747e-06, + "loss": 1.777, + "step": 1420 + }, + { + "epoch": 1.5561643835616439, + "eval_loss": 1.9851160049438477, + "eval_runtime": 37.4606, + "eval_samples_per_second": 26.695, + "eval_steps_per_second": 1.121, + "step": 1420 + }, + { + "epoch": 1.5572602739726027, + "grad_norm": 2.4190328121185303, + "learning_rate": 2.7010560519902516e-06, + "loss": 1.8456, + "step": 1421 + }, + { + "epoch": 1.5583561643835617, + "grad_norm": 2.225623369216919, + "learning_rate": 2.6990251827782293e-06, + "loss": 1.7638, + "step": 1422 + }, + { + "epoch": 1.5594520547945205, + "grad_norm": 2.1516661643981934, + "learning_rate": 2.696994313566206e-06, + "loss": 1.7861, + "step": 1423 + }, + { + "epoch": 1.5605479452054793, + "grad_norm": 2.2461135387420654, + "learning_rate": 2.694963444354184e-06, + "loss": 1.7763, + "step": 1424 + }, + { + "epoch": 1.5616438356164384, + "grad_norm": 2.280869245529175, + "learning_rate": 2.6929325751421607e-06, + "loss": 1.8589, + "step": 1425 + }, + { + "epoch": 1.5627397260273974, + "grad_norm": 2.2034990787506104, + "learning_rate": 2.6909017059301384e-06, + "loss": 1.7705, + "step": 1426 + }, + { + "epoch": 1.563835616438356, + "grad_norm": 2.1346969604492188, + "learning_rate": 2.6888708367181153e-06, + "loss": 1.8318, + "step": 1427 + }, + { + "epoch": 1.564931506849315, + "grad_norm": 2.37400221824646, + "learning_rate": 2.686839967506093e-06, + "loss": 1.746, + "step": 1428 + }, + { + "epoch": 1.566027397260274, + "grad_norm": 2.392561435699463, + "learning_rate": 2.68480909829407e-06, + "loss": 1.8105, + "step": 1429 + }, + { + "epoch": 1.5671232876712329, + "grad_norm": 2.299830198287964, + "learning_rate": 2.6827782290820475e-06, + "loss": 1.7103, + "step": 1430 + }, + { + "epoch": 1.5671232876712329, + "eval_loss": 1.9877182245254517, + "eval_runtime": 37.4813, + "eval_samples_per_second": 26.68, + "eval_steps_per_second": 1.121, + "step": 1430 + }, + { + "epoch": 1.5682191780821917, + "grad_norm": 2.3431026935577393, + "learning_rate": 2.6807473598700244e-06, + "loss": 1.7217, + "step": 1431 + }, + { + "epoch": 1.5693150684931507, + "grad_norm": 2.3474628925323486, + "learning_rate": 2.678716490658002e-06, + "loss": 1.8419, + "step": 1432 + }, + { + "epoch": 1.5704109589041098, + "grad_norm": 2.2877044677734375, + "learning_rate": 2.676685621445979e-06, + "loss": 1.8031, + "step": 1433 + }, + { + "epoch": 1.5715068493150683, + "grad_norm": 2.278613805770874, + "learning_rate": 2.674654752233956e-06, + "loss": 1.8066, + "step": 1434 + }, + { + "epoch": 1.5726027397260274, + "grad_norm": 2.3357837200164795, + "learning_rate": 2.6726238830219335e-06, + "loss": 1.8079, + "step": 1435 + }, + { + "epoch": 1.5736986301369864, + "grad_norm": 2.1932482719421387, + "learning_rate": 2.6705930138099107e-06, + "loss": 1.8032, + "step": 1436 + }, + { + "epoch": 1.5747945205479452, + "grad_norm": 2.298553705215454, + "learning_rate": 2.668562144597888e-06, + "loss": 1.8439, + "step": 1437 + }, + { + "epoch": 1.575890410958904, + "grad_norm": 2.2906084060668945, + "learning_rate": 2.6665312753858653e-06, + "loss": 1.7144, + "step": 1438 + }, + { + "epoch": 1.576986301369863, + "grad_norm": 2.2512989044189453, + "learning_rate": 2.6645004061738426e-06, + "loss": 1.8031, + "step": 1439 + }, + { + "epoch": 1.5780821917808219, + "grad_norm": 2.177103281021118, + "learning_rate": 2.66246953696182e-06, + "loss": 1.8073, + "step": 1440 + }, + { + "epoch": 1.5780821917808219, + "eval_loss": 1.9914618730545044, + "eval_runtime": 37.2706, + "eval_samples_per_second": 26.831, + "eval_steps_per_second": 1.127, + "step": 1440 + }, + { + "epoch": 1.5791780821917807, + "grad_norm": 2.4129281044006348, + "learning_rate": 2.660438667749797e-06, + "loss": 1.7552, + "step": 1441 + }, + { + "epoch": 1.5802739726027397, + "grad_norm": 2.2975244522094727, + "learning_rate": 2.6584077985377744e-06, + "loss": 1.7318, + "step": 1442 + }, + { + "epoch": 1.5813698630136988, + "grad_norm": 2.4061625003814697, + "learning_rate": 2.6563769293257517e-06, + "loss": 1.7839, + "step": 1443 + }, + { + "epoch": 1.5824657534246576, + "grad_norm": 2.236353635787964, + "learning_rate": 2.654346060113729e-06, + "loss": 1.7995, + "step": 1444 + }, + { + "epoch": 1.5835616438356164, + "grad_norm": 2.2464141845703125, + "learning_rate": 2.6523151909017062e-06, + "loss": 1.7496, + "step": 1445 + }, + { + "epoch": 1.5846575342465754, + "grad_norm": 2.2744123935699463, + "learning_rate": 2.6502843216896835e-06, + "loss": 1.7703, + "step": 1446 + }, + { + "epoch": 1.5857534246575342, + "grad_norm": 2.1578054428100586, + "learning_rate": 2.6482534524776603e-06, + "loss": 1.7561, + "step": 1447 + }, + { + "epoch": 1.586849315068493, + "grad_norm": 2.1924850940704346, + "learning_rate": 2.646222583265638e-06, + "loss": 1.867, + "step": 1448 + }, + { + "epoch": 1.587945205479452, + "grad_norm": 2.303420305252075, + "learning_rate": 2.644191714053615e-06, + "loss": 1.8345, + "step": 1449 + }, + { + "epoch": 1.589041095890411, + "grad_norm": 2.308837413787842, + "learning_rate": 2.6421608448415926e-06, + "loss": 1.7979, + "step": 1450 + }, + { + "epoch": 1.589041095890411, + "eval_loss": 1.9836338758468628, + "eval_runtime": 37.253, + "eval_samples_per_second": 26.843, + "eval_steps_per_second": 1.127, + "step": 1450 + }, + { + "epoch": 1.59013698630137, + "grad_norm": 2.2896456718444824, + "learning_rate": 2.6401299756295694e-06, + "loss": 1.7679, + "step": 1451 + }, + { + "epoch": 1.5912328767123287, + "grad_norm": 2.2482378482818604, + "learning_rate": 2.638099106417547e-06, + "loss": 1.7481, + "step": 1452 + }, + { + "epoch": 1.5923287671232877, + "grad_norm": 2.2620620727539062, + "learning_rate": 2.636068237205524e-06, + "loss": 1.801, + "step": 1453 + }, + { + "epoch": 1.5934246575342466, + "grad_norm": 2.216510772705078, + "learning_rate": 2.6340373679935017e-06, + "loss": 1.7485, + "step": 1454 + }, + { + "epoch": 1.5945205479452054, + "grad_norm": 2.354858160018921, + "learning_rate": 2.6320064987814785e-06, + "loss": 1.8636, + "step": 1455 + }, + { + "epoch": 1.5956164383561644, + "grad_norm": 2.251880168914795, + "learning_rate": 2.6299756295694562e-06, + "loss": 1.7087, + "step": 1456 + }, + { + "epoch": 1.5967123287671234, + "grad_norm": 2.2605879306793213, + "learning_rate": 2.627944760357433e-06, + "loss": 1.8141, + "step": 1457 + }, + { + "epoch": 1.5978082191780822, + "grad_norm": 2.3315021991729736, + "learning_rate": 2.6259138911454104e-06, + "loss": 1.8565, + "step": 1458 + }, + { + "epoch": 1.598904109589041, + "grad_norm": 2.1999425888061523, + "learning_rate": 2.6238830219333877e-06, + "loss": 1.8683, + "step": 1459 + }, + { + "epoch": 1.6, + "grad_norm": 2.2652416229248047, + "learning_rate": 2.621852152721365e-06, + "loss": 1.8489, + "step": 1460 + }, + { + "epoch": 1.6, + "eval_loss": 1.9816856384277344, + "eval_runtime": 37.2727, + "eval_samples_per_second": 26.829, + "eval_steps_per_second": 1.127, + "step": 1460 + }, + { + "epoch": 1.601095890410959, + "grad_norm": 2.469440221786499, + "learning_rate": 2.619821283509342e-06, + "loss": 1.7335, + "step": 1461 + }, + { + "epoch": 1.6021917808219177, + "grad_norm": 2.434267044067383, + "learning_rate": 2.6177904142973195e-06, + "loss": 1.8291, + "step": 1462 + }, + { + "epoch": 1.6032876712328767, + "grad_norm": 2.300492286682129, + "learning_rate": 2.6157595450852968e-06, + "loss": 1.7504, + "step": 1463 + }, + { + "epoch": 1.6043835616438358, + "grad_norm": 2.3906843662261963, + "learning_rate": 2.613728675873274e-06, + "loss": 1.8114, + "step": 1464 + }, + { + "epoch": 1.6054794520547944, + "grad_norm": 2.5111773014068604, + "learning_rate": 2.6116978066612513e-06, + "loss": 1.836, + "step": 1465 + }, + { + "epoch": 1.6065753424657534, + "grad_norm": 2.3834846019744873, + "learning_rate": 2.6096669374492286e-06, + "loss": 1.8589, + "step": 1466 + }, + { + "epoch": 1.6076712328767124, + "grad_norm": 2.2709615230560303, + "learning_rate": 2.607636068237206e-06, + "loss": 1.8052, + "step": 1467 + }, + { + "epoch": 1.6087671232876712, + "grad_norm": 2.3351972103118896, + "learning_rate": 2.605605199025183e-06, + "loss": 1.7559, + "step": 1468 + }, + { + "epoch": 1.60986301369863, + "grad_norm": 2.3994698524475098, + "learning_rate": 2.6035743298131604e-06, + "loss": 1.8099, + "step": 1469 + }, + { + "epoch": 1.610958904109589, + "grad_norm": 2.352590322494507, + "learning_rate": 2.6015434606011377e-06, + "loss": 1.8043, + "step": 1470 + }, + { + "epoch": 1.610958904109589, + "eval_loss": 1.9790339469909668, + "eval_runtime": 37.2619, + "eval_samples_per_second": 26.837, + "eval_steps_per_second": 1.127, + "step": 1470 + }, + { + "epoch": 1.612054794520548, + "grad_norm": 2.35754656791687, + "learning_rate": 2.5995125913891145e-06, + "loss": 1.8252, + "step": 1471 + }, + { + "epoch": 1.6131506849315067, + "grad_norm": 2.4011359214782715, + "learning_rate": 2.5974817221770922e-06, + "loss": 1.8005, + "step": 1472 + }, + { + "epoch": 1.6142465753424657, + "grad_norm": 3.143979787826538, + "learning_rate": 2.595450852965069e-06, + "loss": 1.7952, + "step": 1473 + }, + { + "epoch": 1.6153424657534248, + "grad_norm": 2.3928465843200684, + "learning_rate": 2.593419983753047e-06, + "loss": 1.8344, + "step": 1474 + }, + { + "epoch": 1.6164383561643836, + "grad_norm": 2.370368480682373, + "learning_rate": 2.5913891145410236e-06, + "loss": 1.8358, + "step": 1475 + }, + { + "epoch": 1.6175342465753424, + "grad_norm": 2.6163346767425537, + "learning_rate": 2.5893582453290013e-06, + "loss": 1.7734, + "step": 1476 + }, + { + "epoch": 1.6186301369863014, + "grad_norm": 2.370539903640747, + "learning_rate": 2.587327376116978e-06, + "loss": 1.8383, + "step": 1477 + }, + { + "epoch": 1.6197260273972602, + "grad_norm": 2.397007465362549, + "learning_rate": 2.585296506904956e-06, + "loss": 1.7702, + "step": 1478 + }, + { + "epoch": 1.620821917808219, + "grad_norm": 2.4692702293395996, + "learning_rate": 2.5832656376929327e-06, + "loss": 1.7887, + "step": 1479 + }, + { + "epoch": 1.621917808219178, + "grad_norm": 2.5779969692230225, + "learning_rate": 2.5812347684809104e-06, + "loss": 1.7751, + "step": 1480 + }, + { + "epoch": 1.621917808219178, + "eval_loss": 1.9826369285583496, + "eval_runtime": 37.2779, + "eval_samples_per_second": 26.826, + "eval_steps_per_second": 1.127, + "step": 1480 + }, + { + "epoch": 1.6230136986301371, + "grad_norm": 2.3182761669158936, + "learning_rate": 2.5792038992688873e-06, + "loss": 1.7136, + "step": 1481 + }, + { + "epoch": 1.624109589041096, + "grad_norm": 2.4109292030334473, + "learning_rate": 2.577173030056865e-06, + "loss": 1.8804, + "step": 1482 + }, + { + "epoch": 1.6252054794520547, + "grad_norm": 2.426166534423828, + "learning_rate": 2.575142160844842e-06, + "loss": 1.6996, + "step": 1483 + }, + { + "epoch": 1.6263013698630138, + "grad_norm": 2.5459423065185547, + "learning_rate": 2.5731112916328187e-06, + "loss": 1.8186, + "step": 1484 + }, + { + "epoch": 1.6273972602739726, + "grad_norm": 2.583578586578369, + "learning_rate": 2.5710804224207964e-06, + "loss": 1.8275, + "step": 1485 + }, + { + "epoch": 1.6284931506849314, + "grad_norm": 2.442936658859253, + "learning_rate": 2.5690495532087733e-06, + "loss": 1.7553, + "step": 1486 + }, + { + "epoch": 1.6295890410958904, + "grad_norm": 2.587864637374878, + "learning_rate": 2.567018683996751e-06, + "loss": 1.7171, + "step": 1487 + }, + { + "epoch": 1.6306849315068495, + "grad_norm": 2.4119534492492676, + "learning_rate": 2.564987814784728e-06, + "loss": 1.8296, + "step": 1488 + }, + { + "epoch": 1.6317808219178083, + "grad_norm": 2.3582253456115723, + "learning_rate": 2.5629569455727055e-06, + "loss": 1.7595, + "step": 1489 + }, + { + "epoch": 1.632876712328767, + "grad_norm": 2.522674083709717, + "learning_rate": 2.5609260763606824e-06, + "loss": 1.7908, + "step": 1490 + }, + { + "epoch": 1.632876712328767, + "eval_loss": 1.9791436195373535, + "eval_runtime": 37.2967, + "eval_samples_per_second": 26.812, + "eval_steps_per_second": 1.126, + "step": 1490 + }, + { + "epoch": 1.633972602739726, + "grad_norm": 2.2927470207214355, + "learning_rate": 2.55889520714866e-06, + "loss": 1.7789, + "step": 1491 + }, + { + "epoch": 1.635068493150685, + "grad_norm": 2.4469518661499023, + "learning_rate": 2.556864337936637e-06, + "loss": 1.8937, + "step": 1492 + }, + { + "epoch": 1.6361643835616437, + "grad_norm": 2.36860728263855, + "learning_rate": 2.5548334687246146e-06, + "loss": 1.778, + "step": 1493 + }, + { + "epoch": 1.6372602739726028, + "grad_norm": 2.4953925609588623, + "learning_rate": 2.5528025995125915e-06, + "loss": 1.7894, + "step": 1494 + }, + { + "epoch": 1.6383561643835618, + "grad_norm": 2.3478338718414307, + "learning_rate": 2.550771730300569e-06, + "loss": 1.8327, + "step": 1495 + }, + { + "epoch": 1.6394520547945204, + "grad_norm": 2.4399492740631104, + "learning_rate": 2.548740861088546e-06, + "loss": 1.8062, + "step": 1496 + }, + { + "epoch": 1.6405479452054794, + "grad_norm": 2.4457216262817383, + "learning_rate": 2.5467099918765233e-06, + "loss": 1.8268, + "step": 1497 + }, + { + "epoch": 1.6416438356164385, + "grad_norm": 2.365879535675049, + "learning_rate": 2.5446791226645006e-06, + "loss": 1.7288, + "step": 1498 + }, + { + "epoch": 1.6427397260273973, + "grad_norm": 2.4942879676818848, + "learning_rate": 2.542648253452478e-06, + "loss": 1.7111, + "step": 1499 + }, + { + "epoch": 1.643835616438356, + "grad_norm": 2.275982618331909, + "learning_rate": 2.540617384240455e-06, + "loss": 1.8292, + "step": 1500 + }, + { + "epoch": 1.643835616438356, + "eval_loss": 1.9746516942977905, + "eval_runtime": 37.2247, + "eval_samples_per_second": 26.864, + "eval_steps_per_second": 1.128, + "step": 1500 + }, + { + "epoch": 1.644931506849315, + "grad_norm": 2.4430429935455322, + "learning_rate": 2.5385865150284324e-06, + "loss": 1.8034, + "step": 1501 + }, + { + "epoch": 1.6460273972602741, + "grad_norm": 2.2778332233428955, + "learning_rate": 2.5365556458164097e-06, + "loss": 1.7589, + "step": 1502 + }, + { + "epoch": 1.6471232876712327, + "grad_norm": 2.2818830013275146, + "learning_rate": 2.534524776604387e-06, + "loss": 1.7743, + "step": 1503 + }, + { + "epoch": 1.6482191780821918, + "grad_norm": 2.321328639984131, + "learning_rate": 2.5324939073923642e-06, + "loss": 1.7922, + "step": 1504 + }, + { + "epoch": 1.6493150684931508, + "grad_norm": 2.4312188625335693, + "learning_rate": 2.5304630381803415e-06, + "loss": 1.6784, + "step": 1505 + }, + { + "epoch": 1.6504109589041096, + "grad_norm": 2.298741102218628, + "learning_rate": 2.5284321689683188e-06, + "loss": 1.7745, + "step": 1506 + }, + { + "epoch": 1.6515068493150684, + "grad_norm": 2.2882816791534424, + "learning_rate": 2.526401299756296e-06, + "loss": 1.7896, + "step": 1507 + }, + { + "epoch": 1.6526027397260274, + "grad_norm": 2.2336251735687256, + "learning_rate": 2.524370430544273e-06, + "loss": 1.7548, + "step": 1508 + }, + { + "epoch": 1.6536986301369863, + "grad_norm": 2.4369094371795654, + "learning_rate": 2.5223395613322506e-06, + "loss": 1.8049, + "step": 1509 + }, + { + "epoch": 1.654794520547945, + "grad_norm": 2.2978999614715576, + "learning_rate": 2.5203086921202274e-06, + "loss": 1.7489, + "step": 1510 + }, + { + "epoch": 1.654794520547945, + "eval_loss": 1.9760501384735107, + "eval_runtime": 37.3461, + "eval_samples_per_second": 26.777, + "eval_steps_per_second": 1.125, + "step": 1510 + }, + { + "epoch": 1.655890410958904, + "grad_norm": 2.331296920776367, + "learning_rate": 2.518277822908205e-06, + "loss": 1.8293, + "step": 1511 + }, + { + "epoch": 1.6569863013698631, + "grad_norm": 2.4544663429260254, + "learning_rate": 2.516246953696182e-06, + "loss": 1.7617, + "step": 1512 + }, + { + "epoch": 1.658082191780822, + "grad_norm": 2.4682042598724365, + "learning_rate": 2.5142160844841597e-06, + "loss": 1.8054, + "step": 1513 + }, + { + "epoch": 1.6591780821917808, + "grad_norm": 2.2827956676483154, + "learning_rate": 2.5121852152721365e-06, + "loss": 1.8137, + "step": 1514 + }, + { + "epoch": 1.6602739726027398, + "grad_norm": 2.2961761951446533, + "learning_rate": 2.5101543460601142e-06, + "loss": 1.7938, + "step": 1515 + }, + { + "epoch": 1.6613698630136986, + "grad_norm": 2.3464298248291016, + "learning_rate": 2.508123476848091e-06, + "loss": 1.8026, + "step": 1516 + }, + { + "epoch": 1.6624657534246574, + "grad_norm": 2.4975106716156006, + "learning_rate": 2.506092607636069e-06, + "loss": 1.7457, + "step": 1517 + }, + { + "epoch": 1.6635616438356164, + "grad_norm": 2.4823005199432373, + "learning_rate": 2.5040617384240457e-06, + "loss": 1.8282, + "step": 1518 + }, + { + "epoch": 1.6646575342465755, + "grad_norm": 2.2875800132751465, + "learning_rate": 2.5020308692120234e-06, + "loss": 1.7569, + "step": 1519 + }, + { + "epoch": 1.6657534246575343, + "grad_norm": 2.451022148132324, + "learning_rate": 2.5e-06, + "loss": 1.7483, + "step": 1520 + }, + { + "epoch": 1.6657534246575343, + "eval_loss": 1.9748642444610596, + "eval_runtime": 37.4512, + "eval_samples_per_second": 26.701, + "eval_steps_per_second": 1.121, + "step": 1520 + }, + { + "epoch": 1.666849315068493, + "grad_norm": 2.3427352905273438, + "learning_rate": 2.4979691307879775e-06, + "loss": 1.8493, + "step": 1521 + }, + { + "epoch": 1.6679452054794521, + "grad_norm": 2.4496946334838867, + "learning_rate": 2.4959382615759548e-06, + "loss": 1.7569, + "step": 1522 + }, + { + "epoch": 1.669041095890411, + "grad_norm": 2.141963481903076, + "learning_rate": 2.493907392363932e-06, + "loss": 1.7553, + "step": 1523 + }, + { + "epoch": 1.6701369863013698, + "grad_norm": 2.4115428924560547, + "learning_rate": 2.4918765231519093e-06, + "loss": 1.8606, + "step": 1524 + }, + { + "epoch": 1.6712328767123288, + "grad_norm": 2.614616632461548, + "learning_rate": 2.4898456539398866e-06, + "loss": 1.8217, + "step": 1525 + }, + { + "epoch": 1.6723287671232878, + "grad_norm": 2.338557243347168, + "learning_rate": 2.487814784727864e-06, + "loss": 1.7672, + "step": 1526 + }, + { + "epoch": 1.6734246575342466, + "grad_norm": 2.4309446811676025, + "learning_rate": 2.4857839155158407e-06, + "loss": 1.8476, + "step": 1527 + }, + { + "epoch": 1.6745205479452054, + "grad_norm": 2.25883412361145, + "learning_rate": 2.483753046303818e-06, + "loss": 1.7827, + "step": 1528 + }, + { + "epoch": 1.6756164383561645, + "grad_norm": 2.264845371246338, + "learning_rate": 2.4817221770917953e-06, + "loss": 1.7877, + "step": 1529 + }, + { + "epoch": 1.6767123287671233, + "grad_norm": 2.3584723472595215, + "learning_rate": 2.4796913078797725e-06, + "loss": 1.7891, + "step": 1530 + }, + { + "epoch": 1.6767123287671233, + "eval_loss": 1.9751919507980347, + "eval_runtime": 37.558, + "eval_samples_per_second": 26.625, + "eval_steps_per_second": 1.118, + "step": 1530 + }, + { + "epoch": 1.677808219178082, + "grad_norm": 2.3555185794830322, + "learning_rate": 2.47766043866775e-06, + "loss": 1.8235, + "step": 1531 + }, + { + "epoch": 1.6789041095890411, + "grad_norm": 2.3086600303649902, + "learning_rate": 2.475629569455727e-06, + "loss": 1.8036, + "step": 1532 + }, + { + "epoch": 1.6800000000000002, + "grad_norm": 2.286921262741089, + "learning_rate": 2.4735987002437044e-06, + "loss": 1.7885, + "step": 1533 + }, + { + "epoch": 1.6810958904109587, + "grad_norm": 2.2671141624450684, + "learning_rate": 2.4715678310316816e-06, + "loss": 1.758, + "step": 1534 + }, + { + "epoch": 1.6821917808219178, + "grad_norm": 2.4795749187469482, + "learning_rate": 2.469536961819659e-06, + "loss": 1.7791, + "step": 1535 + }, + { + "epoch": 1.6832876712328768, + "grad_norm": 2.201650857925415, + "learning_rate": 2.467506092607636e-06, + "loss": 1.7802, + "step": 1536 + }, + { + "epoch": 1.6843835616438356, + "grad_norm": 2.3804843425750732, + "learning_rate": 2.4654752233956135e-06, + "loss": 1.7739, + "step": 1537 + }, + { + "epoch": 1.6854794520547944, + "grad_norm": 2.322216510772705, + "learning_rate": 2.4634443541835907e-06, + "loss": 1.758, + "step": 1538 + }, + { + "epoch": 1.6865753424657535, + "grad_norm": 2.3582146167755127, + "learning_rate": 2.461413484971568e-06, + "loss": 1.8439, + "step": 1539 + }, + { + "epoch": 1.6876712328767123, + "grad_norm": 2.3597848415374756, + "learning_rate": 2.4593826157595453e-06, + "loss": 1.7772, + "step": 1540 + }, + { + "epoch": 1.6876712328767123, + "eval_loss": 1.9731802940368652, + "eval_runtime": 37.2439, + "eval_samples_per_second": 26.85, + "eval_steps_per_second": 1.128, + "step": 1540 + }, + { + "epoch": 1.688767123287671, + "grad_norm": 2.404569149017334, + "learning_rate": 2.4573517465475226e-06, + "loss": 1.8683, + "step": 1541 + }, + { + "epoch": 1.6898630136986301, + "grad_norm": 2.318737268447876, + "learning_rate": 2.4553208773355e-06, + "loss": 1.7223, + "step": 1542 + }, + { + "epoch": 1.6909589041095892, + "grad_norm": 2.2535340785980225, + "learning_rate": 2.453290008123477e-06, + "loss": 1.7693, + "step": 1543 + }, + { + "epoch": 1.692054794520548, + "grad_norm": 2.2533493041992188, + "learning_rate": 2.4512591389114544e-06, + "loss": 1.7876, + "step": 1544 + }, + { + "epoch": 1.6931506849315068, + "grad_norm": 2.253278970718384, + "learning_rate": 2.4492282696994317e-06, + "loss": 1.788, + "step": 1545 + }, + { + "epoch": 1.6942465753424658, + "grad_norm": 2.296807289123535, + "learning_rate": 2.447197400487409e-06, + "loss": 1.7877, + "step": 1546 + }, + { + "epoch": 1.6953424657534246, + "grad_norm": 2.4022111892700195, + "learning_rate": 2.4451665312753862e-06, + "loss": 1.7986, + "step": 1547 + }, + { + "epoch": 1.6964383561643834, + "grad_norm": 2.2671430110931396, + "learning_rate": 2.4431356620633635e-06, + "loss": 1.8269, + "step": 1548 + }, + { + "epoch": 1.6975342465753425, + "grad_norm": 2.338252067565918, + "learning_rate": 2.4411047928513408e-06, + "loss": 1.8279, + "step": 1549 + }, + { + "epoch": 1.6986301369863015, + "grad_norm": 2.365365505218506, + "learning_rate": 2.439073923639318e-06, + "loss": 1.7895, + "step": 1550 + }, + { + "epoch": 1.6986301369863015, + "eval_loss": 1.9711860418319702, + "eval_runtime": 37.2757, + "eval_samples_per_second": 26.827, + "eval_steps_per_second": 1.127, + "step": 1550 + }, + { + "epoch": 1.6997260273972603, + "grad_norm": 2.407580614089966, + "learning_rate": 2.4370430544272953e-06, + "loss": 1.7985, + "step": 1551 + }, + { + "epoch": 1.7008219178082191, + "grad_norm": 2.3145346641540527, + "learning_rate": 2.435012185215272e-06, + "loss": 1.7589, + "step": 1552 + }, + { + "epoch": 1.7019178082191782, + "grad_norm": 2.3809280395507812, + "learning_rate": 2.4329813160032495e-06, + "loss": 1.7334, + "step": 1553 + }, + { + "epoch": 1.703013698630137, + "grad_norm": 2.4224488735198975, + "learning_rate": 2.4309504467912267e-06, + "loss": 1.7549, + "step": 1554 + }, + { + "epoch": 1.7041095890410958, + "grad_norm": 2.374133348464966, + "learning_rate": 2.428919577579204e-06, + "loss": 1.7923, + "step": 1555 + }, + { + "epoch": 1.7052054794520548, + "grad_norm": 2.258222818374634, + "learning_rate": 2.4268887083671813e-06, + "loss": 1.7683, + "step": 1556 + }, + { + "epoch": 1.7063013698630138, + "grad_norm": 2.4481396675109863, + "learning_rate": 2.4248578391551586e-06, + "loss": 1.8609, + "step": 1557 + }, + { + "epoch": 1.7073972602739727, + "grad_norm": 2.3932790756225586, + "learning_rate": 2.422826969943136e-06, + "loss": 1.7475, + "step": 1558 + }, + { + "epoch": 1.7084931506849315, + "grad_norm": 2.3034355640411377, + "learning_rate": 2.420796100731113e-06, + "loss": 1.7664, + "step": 1559 + }, + { + "epoch": 1.7095890410958905, + "grad_norm": 2.2018799781799316, + "learning_rate": 2.4187652315190904e-06, + "loss": 1.7328, + "step": 1560 + }, + { + "epoch": 1.7095890410958905, + "eval_loss": 1.9715769290924072, + "eval_runtime": 37.2507, + "eval_samples_per_second": 26.845, + "eval_steps_per_second": 1.127, + "step": 1560 + }, + { + "epoch": 1.7106849315068493, + "grad_norm": 2.396462917327881, + "learning_rate": 2.4167343623070677e-06, + "loss": 1.7773, + "step": 1561 + }, + { + "epoch": 1.7117808219178081, + "grad_norm": 2.3212695121765137, + "learning_rate": 2.414703493095045e-06, + "loss": 1.7966, + "step": 1562 + }, + { + "epoch": 1.7128767123287671, + "grad_norm": 2.4157631397247314, + "learning_rate": 2.4126726238830222e-06, + "loss": 1.7797, + "step": 1563 + }, + { + "epoch": 1.7139726027397262, + "grad_norm": 2.404907464981079, + "learning_rate": 2.410641754670999e-06, + "loss": 1.8093, + "step": 1564 + }, + { + "epoch": 1.7150684931506848, + "grad_norm": 2.3798625469207764, + "learning_rate": 2.4086108854589763e-06, + "loss": 1.7635, + "step": 1565 + }, + { + "epoch": 1.7161643835616438, + "grad_norm": 2.325932025909424, + "learning_rate": 2.4065800162469536e-06, + "loss": 1.8281, + "step": 1566 + }, + { + "epoch": 1.7172602739726028, + "grad_norm": 2.413975238800049, + "learning_rate": 2.404549147034931e-06, + "loss": 1.7609, + "step": 1567 + }, + { + "epoch": 1.7183561643835616, + "grad_norm": 2.512775421142578, + "learning_rate": 2.402518277822908e-06, + "loss": 1.7793, + "step": 1568 + }, + { + "epoch": 1.7194520547945205, + "grad_norm": 2.581679582595825, + "learning_rate": 2.4004874086108854e-06, + "loss": 1.8302, + "step": 1569 + }, + { + "epoch": 1.7205479452054795, + "grad_norm": 2.389822244644165, + "learning_rate": 2.3984565393988627e-06, + "loss": 1.7573, + "step": 1570 + }, + { + "epoch": 1.7205479452054795, + "eval_loss": 1.9670331478118896, + "eval_runtime": 37.2901, + "eval_samples_per_second": 26.817, + "eval_steps_per_second": 1.126, + "step": 1570 + }, + { + "epoch": 1.7216438356164385, + "grad_norm": 2.6191556453704834, + "learning_rate": 2.39642567018684e-06, + "loss": 1.8252, + "step": 1571 + }, + { + "epoch": 1.7227397260273971, + "grad_norm": 2.394347906112671, + "learning_rate": 2.3943948009748173e-06, + "loss": 1.7561, + "step": 1572 + }, + { + "epoch": 1.7238356164383561, + "grad_norm": 2.3016326427459717, + "learning_rate": 2.3923639317627945e-06, + "loss": 1.7619, + "step": 1573 + }, + { + "epoch": 1.7249315068493152, + "grad_norm": 2.3663971424102783, + "learning_rate": 2.390333062550772e-06, + "loss": 1.7922, + "step": 1574 + }, + { + "epoch": 1.726027397260274, + "grad_norm": 2.3328938484191895, + "learning_rate": 2.388302193338749e-06, + "loss": 1.7374, + "step": 1575 + }, + { + "epoch": 1.7271232876712328, + "grad_norm": 2.313525676727295, + "learning_rate": 2.3862713241267264e-06, + "loss": 1.6585, + "step": 1576 + }, + { + "epoch": 1.7282191780821918, + "grad_norm": 2.4609642028808594, + "learning_rate": 2.3842404549147037e-06, + "loss": 1.8757, + "step": 1577 + }, + { + "epoch": 1.7293150684931506, + "grad_norm": 2.3502120971679688, + "learning_rate": 2.382209585702681e-06, + "loss": 1.7786, + "step": 1578 + }, + { + "epoch": 1.7304109589041095, + "grad_norm": 2.383025884628296, + "learning_rate": 2.380178716490658e-06, + "loss": 1.7502, + "step": 1579 + }, + { + "epoch": 1.7315068493150685, + "grad_norm": 2.280890464782715, + "learning_rate": 2.3781478472786355e-06, + "loss": 1.8192, + "step": 1580 + }, + { + "epoch": 1.7315068493150685, + "eval_loss": 1.9658632278442383, + "eval_runtime": 37.3714, + "eval_samples_per_second": 26.758, + "eval_steps_per_second": 1.124, + "step": 1580 + }, + { + "epoch": 1.7326027397260275, + "grad_norm": 2.342240333557129, + "learning_rate": 2.3761169780666128e-06, + "loss": 1.7786, + "step": 1581 + }, + { + "epoch": 1.7336986301369863, + "grad_norm": 2.2916715145111084, + "learning_rate": 2.37408610885459e-06, + "loss": 1.8161, + "step": 1582 + }, + { + "epoch": 1.7347945205479451, + "grad_norm": 2.24537992477417, + "learning_rate": 2.3720552396425673e-06, + "loss": 1.8442, + "step": 1583 + }, + { + "epoch": 1.7358904109589042, + "grad_norm": 2.3003411293029785, + "learning_rate": 2.3700243704305446e-06, + "loss": 1.7775, + "step": 1584 + }, + { + "epoch": 1.736986301369863, + "grad_norm": 2.3062400817871094, + "learning_rate": 2.367993501218522e-06, + "loss": 1.8374, + "step": 1585 + }, + { + "epoch": 1.7380821917808218, + "grad_norm": 2.3534083366394043, + "learning_rate": 2.365962632006499e-06, + "loss": 1.7829, + "step": 1586 + }, + { + "epoch": 1.7391780821917808, + "grad_norm": 2.2612221240997314, + "learning_rate": 2.3639317627944764e-06, + "loss": 1.7934, + "step": 1587 + }, + { + "epoch": 1.7402739726027399, + "grad_norm": 2.4197304248809814, + "learning_rate": 2.3619008935824537e-06, + "loss": 1.8226, + "step": 1588 + }, + { + "epoch": 1.7413698630136987, + "grad_norm": 2.435908079147339, + "learning_rate": 2.3598700243704305e-06, + "loss": 1.8148, + "step": 1589 + }, + { + "epoch": 1.7424657534246575, + "grad_norm": 2.3617560863494873, + "learning_rate": 2.357839155158408e-06, + "loss": 1.8085, + "step": 1590 + }, + { + "epoch": 1.7424657534246575, + "eval_loss": 1.9636695384979248, + "eval_runtime": 37.2968, + "eval_samples_per_second": 26.812, + "eval_steps_per_second": 1.126, + "step": 1590 + }, + { + "epoch": 1.7435616438356165, + "grad_norm": 2.295182228088379, + "learning_rate": 2.355808285946385e-06, + "loss": 1.8091, + "step": 1591 + }, + { + "epoch": 1.7446575342465753, + "grad_norm": 2.3347294330596924, + "learning_rate": 2.3537774167343624e-06, + "loss": 1.8086, + "step": 1592 + }, + { + "epoch": 1.7457534246575341, + "grad_norm": 2.4760799407958984, + "learning_rate": 2.3517465475223396e-06, + "loss": 1.8227, + "step": 1593 + }, + { + "epoch": 1.7468493150684932, + "grad_norm": 2.4243907928466797, + "learning_rate": 2.349715678310317e-06, + "loss": 1.7921, + "step": 1594 + }, + { + "epoch": 1.7479452054794522, + "grad_norm": 2.271735668182373, + "learning_rate": 2.347684809098294e-06, + "loss": 1.7987, + "step": 1595 + }, + { + "epoch": 1.749041095890411, + "grad_norm": 2.5412354469299316, + "learning_rate": 2.3456539398862715e-06, + "loss": 1.7558, + "step": 1596 + }, + { + "epoch": 1.7501369863013698, + "grad_norm": 2.464944362640381, + "learning_rate": 2.3436230706742487e-06, + "loss": 1.8087, + "step": 1597 + }, + { + "epoch": 1.7512328767123289, + "grad_norm": 2.4210381507873535, + "learning_rate": 2.341592201462226e-06, + "loss": 1.756, + "step": 1598 + }, + { + "epoch": 1.7523287671232877, + "grad_norm": 2.284945011138916, + "learning_rate": 2.3395613322502033e-06, + "loss": 1.7909, + "step": 1599 + }, + { + "epoch": 1.7534246575342465, + "grad_norm": 2.3740732669830322, + "learning_rate": 2.3375304630381806e-06, + "loss": 1.7683, + "step": 1600 + }, + { + "epoch": 1.7534246575342465, + "eval_loss": 1.9646408557891846, + "eval_runtime": 37.3266, + "eval_samples_per_second": 26.791, + "eval_steps_per_second": 1.125, + "step": 1600 + }, + { + "epoch": 1.7545205479452055, + "grad_norm": 2.399583578109741, + "learning_rate": 2.335499593826158e-06, + "loss": 1.7342, + "step": 1601 + }, + { + "epoch": 1.7556164383561645, + "grad_norm": 2.4731597900390625, + "learning_rate": 2.333468724614135e-06, + "loss": 1.7507, + "step": 1602 + }, + { + "epoch": 1.7567123287671231, + "grad_norm": 2.365433692932129, + "learning_rate": 2.3314378554021124e-06, + "loss": 1.7944, + "step": 1603 + }, + { + "epoch": 1.7578082191780822, + "grad_norm": 2.4500601291656494, + "learning_rate": 2.3294069861900897e-06, + "loss": 1.822, + "step": 1604 + }, + { + "epoch": 1.7589041095890412, + "grad_norm": 2.3339014053344727, + "learning_rate": 2.327376116978067e-06, + "loss": 1.7464, + "step": 1605 + }, + { + "epoch": 1.76, + "grad_norm": 2.343973159790039, + "learning_rate": 2.3253452477660442e-06, + "loss": 1.7408, + "step": 1606 + }, + { + "epoch": 1.7610958904109588, + "grad_norm": 2.453462839126587, + "learning_rate": 2.3233143785540215e-06, + "loss": 1.8261, + "step": 1607 + }, + { + "epoch": 1.7621917808219179, + "grad_norm": 2.357698678970337, + "learning_rate": 2.3212835093419988e-06, + "loss": 1.762, + "step": 1608 + }, + { + "epoch": 1.7632876712328767, + "grad_norm": 2.353257417678833, + "learning_rate": 2.319252640129976e-06, + "loss": 1.8076, + "step": 1609 + }, + { + "epoch": 1.7643835616438355, + "grad_norm": 2.3269948959350586, + "learning_rate": 2.3172217709179533e-06, + "loss": 1.7516, + "step": 1610 + }, + { + "epoch": 1.7643835616438355, + "eval_loss": 1.9678465127944946, + "eval_runtime": 37.2052, + "eval_samples_per_second": 26.878, + "eval_steps_per_second": 1.129, + "step": 1610 + }, + { + "epoch": 1.7654794520547945, + "grad_norm": 2.4023501873016357, + "learning_rate": 2.3151909017059306e-06, + "loss": 1.8477, + "step": 1611 + }, + { + "epoch": 1.7665753424657535, + "grad_norm": 2.3356990814208984, + "learning_rate": 2.313160032493908e-06, + "loss": 1.8236, + "step": 1612 + }, + { + "epoch": 1.7676712328767124, + "grad_norm": 2.3571391105651855, + "learning_rate": 2.311129163281885e-06, + "loss": 1.7335, + "step": 1613 + }, + { + "epoch": 1.7687671232876712, + "grad_norm": 2.3510208129882812, + "learning_rate": 2.309098294069862e-06, + "loss": 1.8088, + "step": 1614 + }, + { + "epoch": 1.7698630136986302, + "grad_norm": 2.3852577209472656, + "learning_rate": 2.3070674248578393e-06, + "loss": 1.7312, + "step": 1615 + }, + { + "epoch": 1.770958904109589, + "grad_norm": 2.3085618019104004, + "learning_rate": 2.3050365556458166e-06, + "loss": 1.7379, + "step": 1616 + }, + { + "epoch": 1.7720547945205478, + "grad_norm": 2.3454065322875977, + "learning_rate": 2.303005686433794e-06, + "loss": 1.8054, + "step": 1617 + }, + { + "epoch": 1.7731506849315068, + "grad_norm": 2.4034247398376465, + "learning_rate": 2.300974817221771e-06, + "loss": 1.7955, + "step": 1618 + }, + { + "epoch": 1.7742465753424659, + "grad_norm": 2.4181671142578125, + "learning_rate": 2.2989439480097484e-06, + "loss": 1.7732, + "step": 1619 + }, + { + "epoch": 1.7753424657534247, + "grad_norm": 2.3408010005950928, + "learning_rate": 2.2969130787977257e-06, + "loss": 1.7231, + "step": 1620 + }, + { + "epoch": 1.7753424657534247, + "eval_loss": 1.9655163288116455, + "eval_runtime": 37.4878, + "eval_samples_per_second": 26.675, + "eval_steps_per_second": 1.12, + "step": 1620 + }, + { + "epoch": 1.7764383561643835, + "grad_norm": 2.3113718032836914, + "learning_rate": 2.294882209585703e-06, + "loss": 1.7904, + "step": 1621 + }, + { + "epoch": 1.7775342465753425, + "grad_norm": 2.3037896156311035, + "learning_rate": 2.2928513403736802e-06, + "loss": 1.8069, + "step": 1622 + }, + { + "epoch": 1.7786301369863013, + "grad_norm": 2.3962693214416504, + "learning_rate": 2.2908204711616575e-06, + "loss": 1.7354, + "step": 1623 + }, + { + "epoch": 1.7797260273972602, + "grad_norm": 2.3499252796173096, + "learning_rate": 2.2887896019496348e-06, + "loss": 1.779, + "step": 1624 + }, + { + "epoch": 1.7808219178082192, + "grad_norm": 2.390958070755005, + "learning_rate": 2.286758732737612e-06, + "loss": 1.8257, + "step": 1625 + }, + { + "epoch": 1.7819178082191782, + "grad_norm": 2.365626573562622, + "learning_rate": 2.284727863525589e-06, + "loss": 1.7618, + "step": 1626 + }, + { + "epoch": 1.783013698630137, + "grad_norm": 2.3558530807495117, + "learning_rate": 2.282696994313566e-06, + "loss": 1.7729, + "step": 1627 + }, + { + "epoch": 1.7841095890410958, + "grad_norm": 2.3276045322418213, + "learning_rate": 2.2806661251015434e-06, + "loss": 1.8231, + "step": 1628 + }, + { + "epoch": 1.7852054794520549, + "grad_norm": 2.3369953632354736, + "learning_rate": 2.2786352558895207e-06, + "loss": 1.8011, + "step": 1629 + }, + { + "epoch": 1.7863013698630137, + "grad_norm": 2.3588311672210693, + "learning_rate": 2.276604386677498e-06, + "loss": 1.7682, + "step": 1630 + }, + { + "epoch": 1.7863013698630137, + "eval_loss": 1.963529348373413, + "eval_runtime": 37.5589, + "eval_samples_per_second": 26.625, + "eval_steps_per_second": 1.118, + "step": 1630 + }, + { + "epoch": 1.7873972602739725, + "grad_norm": 2.2918803691864014, + "learning_rate": 2.2745735174654753e-06, + "loss": 1.7256, + "step": 1631 + }, + { + "epoch": 1.7884931506849315, + "grad_norm": 2.423738479614258, + "learning_rate": 2.2725426482534526e-06, + "loss": 1.7582, + "step": 1632 + }, + { + "epoch": 1.7895890410958906, + "grad_norm": 2.3500850200653076, + "learning_rate": 2.27051177904143e-06, + "loss": 1.8255, + "step": 1633 + }, + { + "epoch": 1.7906849315068492, + "grad_norm": 2.3280837535858154, + "learning_rate": 2.268480909829407e-06, + "loss": 1.7845, + "step": 1634 + }, + { + "epoch": 1.7917808219178082, + "grad_norm": 2.5055465698242188, + "learning_rate": 2.2664500406173844e-06, + "loss": 1.78, + "step": 1635 + }, + { + "epoch": 1.7928767123287672, + "grad_norm": 2.3132143020629883, + "learning_rate": 2.2644191714053617e-06, + "loss": 1.7814, + "step": 1636 + }, + { + "epoch": 1.793972602739726, + "grad_norm": 2.316361427307129, + "learning_rate": 2.262388302193339e-06, + "loss": 1.8161, + "step": 1637 + }, + { + "epoch": 1.7950684931506848, + "grad_norm": 2.337318181991577, + "learning_rate": 2.260357432981316e-06, + "loss": 1.7942, + "step": 1638 + }, + { + "epoch": 1.7961643835616439, + "grad_norm": 2.362031936645508, + "learning_rate": 2.2583265637692935e-06, + "loss": 1.7771, + "step": 1639 + }, + { + "epoch": 1.7972602739726027, + "grad_norm": 2.3260183334350586, + "learning_rate": 2.2562956945572708e-06, + "loss": 1.7975, + "step": 1640 + }, + { + "epoch": 1.7972602739726027, + "eval_loss": 1.9619382619857788, + "eval_runtime": 37.2606, + "eval_samples_per_second": 26.838, + "eval_steps_per_second": 1.127, + "step": 1640 + }, + { + "epoch": 1.7983561643835615, + "grad_norm": 2.4260714054107666, + "learning_rate": 2.254264825345248e-06, + "loss": 1.7654, + "step": 1641 + }, + { + "epoch": 1.7994520547945205, + "grad_norm": 2.297428846359253, + "learning_rate": 2.2522339561332253e-06, + "loss": 1.775, + "step": 1642 + }, + { + "epoch": 1.8005479452054796, + "grad_norm": 2.5153353214263916, + "learning_rate": 2.2502030869212026e-06, + "loss": 1.7796, + "step": 1643 + }, + { + "epoch": 1.8016438356164384, + "grad_norm": 2.5102577209472656, + "learning_rate": 2.24817221770918e-06, + "loss": 1.7917, + "step": 1644 + }, + { + "epoch": 1.8027397260273972, + "grad_norm": 2.3693106174468994, + "learning_rate": 2.246141348497157e-06, + "loss": 1.8459, + "step": 1645 + }, + { + "epoch": 1.8038356164383562, + "grad_norm": 2.2353153228759766, + "learning_rate": 2.2441104792851344e-06, + "loss": 1.8095, + "step": 1646 + }, + { + "epoch": 1.804931506849315, + "grad_norm": 2.265465497970581, + "learning_rate": 2.2420796100731117e-06, + "loss": 1.7623, + "step": 1647 + }, + { + "epoch": 1.8060273972602738, + "grad_norm": 2.2820940017700195, + "learning_rate": 2.240048740861089e-06, + "loss": 1.7946, + "step": 1648 + }, + { + "epoch": 1.8071232876712329, + "grad_norm": 2.378035068511963, + "learning_rate": 2.2380178716490662e-06, + "loss": 1.8059, + "step": 1649 + }, + { + "epoch": 1.808219178082192, + "grad_norm": 2.328341245651245, + "learning_rate": 2.2359870024370435e-06, + "loss": 1.6611, + "step": 1650 + }, + { + "epoch": 1.808219178082192, + "eval_loss": 1.9600144624710083, + "eval_runtime": 37.2601, + "eval_samples_per_second": 26.838, + "eval_steps_per_second": 1.127, + "step": 1650 + }, + { + "epoch": 1.8093150684931507, + "grad_norm": 2.2819113731384277, + "learning_rate": 2.2339561332250204e-06, + "loss": 1.8058, + "step": 1651 + }, + { + "epoch": 1.8104109589041095, + "grad_norm": 2.4178121089935303, + "learning_rate": 2.2319252640129976e-06, + "loss": 1.8204, + "step": 1652 + }, + { + "epoch": 1.8115068493150686, + "grad_norm": 2.497008800506592, + "learning_rate": 2.229894394800975e-06, + "loss": 1.7968, + "step": 1653 + }, + { + "epoch": 1.8126027397260274, + "grad_norm": 2.4681501388549805, + "learning_rate": 2.227863525588952e-06, + "loss": 1.7292, + "step": 1654 + }, + { + "epoch": 1.8136986301369862, + "grad_norm": 2.2983527183532715, + "learning_rate": 2.2258326563769295e-06, + "loss": 1.7097, + "step": 1655 + }, + { + "epoch": 1.8147945205479452, + "grad_norm": 2.349454164505005, + "learning_rate": 2.2238017871649067e-06, + "loss": 1.7256, + "step": 1656 + }, + { + "epoch": 1.8158904109589042, + "grad_norm": 2.4247677326202393, + "learning_rate": 2.221770917952884e-06, + "loss": 1.7165, + "step": 1657 + }, + { + "epoch": 1.816986301369863, + "grad_norm": 2.3035342693328857, + "learning_rate": 2.2197400487408613e-06, + "loss": 1.7811, + "step": 1658 + }, + { + "epoch": 1.8180821917808219, + "grad_norm": 2.318772554397583, + "learning_rate": 2.2177091795288386e-06, + "loss": 1.8082, + "step": 1659 + }, + { + "epoch": 1.819178082191781, + "grad_norm": 2.2477188110351562, + "learning_rate": 2.215678310316816e-06, + "loss": 1.7206, + "step": 1660 + }, + { + "epoch": 1.819178082191781, + "eval_loss": 1.962982416152954, + "eval_runtime": 37.2709, + "eval_samples_per_second": 26.831, + "eval_steps_per_second": 1.127, + "step": 1660 + }, + { + "epoch": 1.8202739726027397, + "grad_norm": 2.355264186859131, + "learning_rate": 2.213647441104793e-06, + "loss": 1.7634, + "step": 1661 + }, + { + "epoch": 1.8213698630136985, + "grad_norm": 2.267458200454712, + "learning_rate": 2.2116165718927704e-06, + "loss": 1.7819, + "step": 1662 + }, + { + "epoch": 1.8224657534246576, + "grad_norm": 2.2487895488739014, + "learning_rate": 2.2095857026807473e-06, + "loss": 1.759, + "step": 1663 + }, + { + "epoch": 1.8235616438356166, + "grad_norm": 2.4253578186035156, + "learning_rate": 2.2075548334687245e-06, + "loss": 1.7937, + "step": 1664 + }, + { + "epoch": 1.8246575342465754, + "grad_norm": 2.478050470352173, + "learning_rate": 2.205523964256702e-06, + "loss": 1.7241, + "step": 1665 + }, + { + "epoch": 1.8257534246575342, + "grad_norm": 2.289905309677124, + "learning_rate": 2.203493095044679e-06, + "loss": 1.7531, + "step": 1666 + }, + { + "epoch": 1.8268493150684932, + "grad_norm": 2.303421974182129, + "learning_rate": 2.2014622258326564e-06, + "loss": 1.7711, + "step": 1667 + }, + { + "epoch": 1.827945205479452, + "grad_norm": 2.3670248985290527, + "learning_rate": 2.1994313566206336e-06, + "loss": 1.8093, + "step": 1668 + }, + { + "epoch": 1.8290410958904109, + "grad_norm": 2.3721048831939697, + "learning_rate": 2.197400487408611e-06, + "loss": 1.8273, + "step": 1669 + }, + { + "epoch": 1.83013698630137, + "grad_norm": 2.3000190258026123, + "learning_rate": 2.195369618196588e-06, + "loss": 1.75, + "step": 1670 + }, + { + "epoch": 1.83013698630137, + "eval_loss": 1.9577263593673706, + "eval_runtime": 37.2953, + "eval_samples_per_second": 26.813, + "eval_steps_per_second": 1.126, + "step": 1670 + }, + { + "epoch": 1.831232876712329, + "grad_norm": 2.303584575653076, + "learning_rate": 2.1933387489845655e-06, + "loss": 1.7278, + "step": 1671 + }, + { + "epoch": 1.8323287671232875, + "grad_norm": 2.260579824447632, + "learning_rate": 2.1913078797725427e-06, + "loss": 1.7524, + "step": 1672 + }, + { + "epoch": 1.8334246575342465, + "grad_norm": 2.311967372894287, + "learning_rate": 2.18927701056052e-06, + "loss": 1.7799, + "step": 1673 + }, + { + "epoch": 1.8345205479452056, + "grad_norm": 2.338989496231079, + "learning_rate": 2.1872461413484973e-06, + "loss": 1.8468, + "step": 1674 + }, + { + "epoch": 1.8356164383561644, + "grad_norm": 2.338989496231079, + "learning_rate": 2.1872461413484973e-06, + "loss": 1.7853, + "step": 1675 + }, + { + "epoch": 1.8367123287671232, + "grad_norm": 2.526724100112915, + "learning_rate": 2.1852152721364746e-06, + "loss": 1.78, + "step": 1676 + }, + { + "epoch": 1.8378082191780822, + "grad_norm": 2.4637107849121094, + "learning_rate": 2.183184402924452e-06, + "loss": 1.7328, + "step": 1677 + }, + { + "epoch": 1.838904109589041, + "grad_norm": 2.34454345703125, + "learning_rate": 2.181153533712429e-06, + "loss": 1.8058, + "step": 1678 + }, + { + "epoch": 1.8399999999999999, + "grad_norm": 2.5518760681152344, + "learning_rate": 2.1791226645004064e-06, + "loss": 1.7746, + "step": 1679 + }, + { + "epoch": 1.841095890410959, + "grad_norm": 2.437443971633911, + "learning_rate": 2.1770917952883837e-06, + "loss": 1.7517, + "step": 1680 + }, + { + "epoch": 1.841095890410959, + "eval_loss": 1.9572285413742065, + "eval_runtime": 37.3402, + "eval_samples_per_second": 26.781, + "eval_steps_per_second": 1.125, + "step": 1680 + }, + { + "epoch": 1.842191780821918, + "grad_norm": 2.344287157058716, + "learning_rate": 2.175060926076361e-06, + "loss": 1.8229, + "step": 1681 + }, + { + "epoch": 1.8432876712328767, + "grad_norm": 2.333840847015381, + "learning_rate": 2.1730300568643382e-06, + "loss": 1.8534, + "step": 1682 + }, + { + "epoch": 1.8443835616438355, + "grad_norm": 2.372926950454712, + "learning_rate": 2.1709991876523155e-06, + "loss": 1.7609, + "step": 1683 + }, + { + "epoch": 1.8454794520547946, + "grad_norm": 2.2858645915985107, + "learning_rate": 2.1689683184402928e-06, + "loss": 1.7508, + "step": 1684 + }, + { + "epoch": 1.8465753424657534, + "grad_norm": 2.326586961746216, + "learning_rate": 2.16693744922827e-06, + "loss": 1.8122, + "step": 1685 + }, + { + "epoch": 1.8476712328767122, + "grad_norm": 2.3605616092681885, + "learning_rate": 2.1649065800162473e-06, + "loss": 1.7794, + "step": 1686 + }, + { + "epoch": 1.8487671232876712, + "grad_norm": 2.4188039302825928, + "learning_rate": 2.1628757108042246e-06, + "loss": 1.7552, + "step": 1687 + }, + { + "epoch": 1.8498630136986303, + "grad_norm": 2.335617780685425, + "learning_rate": 2.160844841592202e-06, + "loss": 1.7259, + "step": 1688 + }, + { + "epoch": 1.850958904109589, + "grad_norm": 2.245882272720337, + "learning_rate": 2.1588139723801787e-06, + "loss": 1.7259, + "step": 1689 + }, + { + "epoch": 1.8520547945205479, + "grad_norm": 2.4447500705718994, + "learning_rate": 2.156783103168156e-06, + "loss": 1.7605, + "step": 1690 + }, + { + "epoch": 1.8520547945205479, + "eval_loss": 1.9608982801437378, + "eval_runtime": 37.3756, + "eval_samples_per_second": 26.755, + "eval_steps_per_second": 1.124, + "step": 1690 + }, + { + "epoch": 1.853150684931507, + "grad_norm": 2.363443613052368, + "learning_rate": 2.1547522339561333e-06, + "loss": 1.7336, + "step": 1691 + }, + { + "epoch": 1.8542465753424657, + "grad_norm": 2.3573989868164062, + "learning_rate": 2.1527213647441106e-06, + "loss": 1.7953, + "step": 1692 + }, + { + "epoch": 1.8553424657534245, + "grad_norm": 2.4292593002319336, + "learning_rate": 2.150690495532088e-06, + "loss": 1.8276, + "step": 1693 + }, + { + "epoch": 1.8564383561643836, + "grad_norm": 2.374563694000244, + "learning_rate": 2.148659626320065e-06, + "loss": 1.7975, + "step": 1694 + }, + { + "epoch": 1.8575342465753426, + "grad_norm": 2.3935306072235107, + "learning_rate": 2.1466287571080424e-06, + "loss": 1.7824, + "step": 1695 + }, + { + "epoch": 1.8586301369863014, + "grad_norm": 2.343773365020752, + "learning_rate": 2.1445978878960197e-06, + "loss": 1.7574, + "step": 1696 + }, + { + "epoch": 1.8597260273972602, + "grad_norm": 2.2765426635742188, + "learning_rate": 2.142567018683997e-06, + "loss": 1.7682, + "step": 1697 + }, + { + "epoch": 1.8608219178082193, + "grad_norm": 2.339470863342285, + "learning_rate": 2.140536149471974e-06, + "loss": 1.8229, + "step": 1698 + }, + { + "epoch": 1.861917808219178, + "grad_norm": 2.341775894165039, + "learning_rate": 2.1385052802599515e-06, + "loss": 1.7421, + "step": 1699 + }, + { + "epoch": 1.8630136986301369, + "grad_norm": 2.343876838684082, + "learning_rate": 2.1364744110479288e-06, + "loss": 1.7519, + "step": 1700 + }, + { + "epoch": 1.8630136986301369, + "eval_loss": 1.9587067365646362, + "eval_runtime": 37.4192, + "eval_samples_per_second": 26.724, + "eval_steps_per_second": 1.122, + "step": 1700 + }, + { + "epoch": 1.864109589041096, + "grad_norm": 2.388373613357544, + "learning_rate": 2.134443541835906e-06, + "loss": 1.8039, + "step": 1701 + }, + { + "epoch": 1.865205479452055, + "grad_norm": 2.379742383956909, + "learning_rate": 2.132412672623883e-06, + "loss": 1.7549, + "step": 1702 + }, + { + "epoch": 1.8663013698630135, + "grad_norm": 2.25833797454834, + "learning_rate": 2.13038180341186e-06, + "loss": 1.7349, + "step": 1703 + }, + { + "epoch": 1.8673972602739726, + "grad_norm": 2.4541401863098145, + "learning_rate": 2.1283509341998374e-06, + "loss": 1.7785, + "step": 1704 + }, + { + "epoch": 1.8684931506849316, + "grad_norm": 2.338456392288208, + "learning_rate": 2.1263200649878147e-06, + "loss": 1.8272, + "step": 1705 + }, + { + "epoch": 1.8695890410958904, + "grad_norm": 2.474292755126953, + "learning_rate": 2.124289195775792e-06, + "loss": 1.7497, + "step": 1706 + }, + { + "epoch": 1.8706849315068492, + "grad_norm": 2.206749439239502, + "learning_rate": 2.1222583265637693e-06, + "loss": 1.7697, + "step": 1707 + }, + { + "epoch": 1.8717808219178083, + "grad_norm": 2.378093957901001, + "learning_rate": 2.1202274573517465e-06, + "loss": 1.6851, + "step": 1708 + }, + { + "epoch": 1.872876712328767, + "grad_norm": 2.2442147731781006, + "learning_rate": 2.118196588139724e-06, + "loss": 1.7268, + "step": 1709 + }, + { + "epoch": 1.8739726027397259, + "grad_norm": 2.324852228164673, + "learning_rate": 2.116165718927701e-06, + "loss": 1.7369, + "step": 1710 + }, + { + "epoch": 1.8739726027397259, + "eval_loss": 1.9555498361587524, + "eval_runtime": 37.2599, + "eval_samples_per_second": 26.838, + "eval_steps_per_second": 1.127, + "step": 1710 + }, + { + "epoch": 1.875068493150685, + "grad_norm": 2.3466620445251465, + "learning_rate": 2.1141348497156784e-06, + "loss": 1.7642, + "step": 1711 + }, + { + "epoch": 1.876164383561644, + "grad_norm": 2.303654432296753, + "learning_rate": 2.1121039805036556e-06, + "loss": 1.7406, + "step": 1712 + }, + { + "epoch": 1.8772602739726028, + "grad_norm": 2.242719888687134, + "learning_rate": 2.110073111291633e-06, + "loss": 1.7269, + "step": 1713 + }, + { + "epoch": 1.8783561643835616, + "grad_norm": 2.4076626300811768, + "learning_rate": 2.10804224207961e-06, + "loss": 1.7749, + "step": 1714 + }, + { + "epoch": 1.8794520547945206, + "grad_norm": 2.510887861251831, + "learning_rate": 2.1060113728675875e-06, + "loss": 1.8056, + "step": 1715 + }, + { + "epoch": 1.8805479452054794, + "grad_norm": 2.4038093090057373, + "learning_rate": 2.1039805036555647e-06, + "loss": 1.7915, + "step": 1716 + }, + { + "epoch": 1.8816438356164382, + "grad_norm": 2.4485573768615723, + "learning_rate": 2.101949634443542e-06, + "loss": 1.6845, + "step": 1717 + }, + { + "epoch": 1.8827397260273973, + "grad_norm": 2.4619247913360596, + "learning_rate": 2.0999187652315193e-06, + "loss": 1.7597, + "step": 1718 + }, + { + "epoch": 1.8838356164383563, + "grad_norm": 2.5079665184020996, + "learning_rate": 2.0978878960194966e-06, + "loss": 1.7207, + "step": 1719 + }, + { + "epoch": 1.884931506849315, + "grad_norm": 2.38696551322937, + "learning_rate": 2.095857026807474e-06, + "loss": 1.7855, + "step": 1720 + }, + { + "epoch": 1.884931506849315, + "eval_loss": 1.9538354873657227, + "eval_runtime": 37.5269, + "eval_samples_per_second": 26.648, + "eval_steps_per_second": 1.119, + "step": 1720 + }, + { + "epoch": 1.886027397260274, + "grad_norm": 2.2652602195739746, + "learning_rate": 2.093826157595451e-06, + "loss": 1.7806, + "step": 1721 + }, + { + "epoch": 1.887123287671233, + "grad_norm": 2.4074227809906006, + "learning_rate": 2.0917952883834284e-06, + "loss": 1.7397, + "step": 1722 + }, + { + "epoch": 1.8882191780821918, + "grad_norm": 2.4302561283111572, + "learning_rate": 2.0897644191714057e-06, + "loss": 1.7883, + "step": 1723 + }, + { + "epoch": 1.8893150684931506, + "grad_norm": 2.311933755874634, + "learning_rate": 2.087733549959383e-06, + "loss": 1.8186, + "step": 1724 + }, + { + "epoch": 1.8904109589041096, + "grad_norm": 2.4028313159942627, + "learning_rate": 2.0857026807473602e-06, + "loss": 1.7208, + "step": 1725 + }, + { + "epoch": 1.8915068493150686, + "grad_norm": 2.2129130363464355, + "learning_rate": 2.0836718115353375e-06, + "loss": 1.7007, + "step": 1726 + }, + { + "epoch": 1.8926027397260274, + "grad_norm": 2.3893744945526123, + "learning_rate": 2.0816409423233144e-06, + "loss": 1.8281, + "step": 1727 + }, + { + "epoch": 1.8936986301369862, + "grad_norm": 2.2396240234375, + "learning_rate": 2.0796100731112916e-06, + "loss": 1.7252, + "step": 1728 + }, + { + "epoch": 1.8947945205479453, + "grad_norm": 2.4073901176452637, + "learning_rate": 2.077579203899269e-06, + "loss": 1.7533, + "step": 1729 + }, + { + "epoch": 1.895890410958904, + "grad_norm": 2.3669373989105225, + "learning_rate": 2.075548334687246e-06, + "loss": 1.7013, + "step": 1730 + }, + { + "epoch": 1.895890410958904, + "eval_loss": 1.95528244972229, + "eval_runtime": 37.6174, + "eval_samples_per_second": 26.583, + "eval_steps_per_second": 1.117, + "step": 1730 + }, + { + "epoch": 1.896986301369863, + "grad_norm": 2.5254933834075928, + "learning_rate": 2.0735174654752235e-06, + "loss": 1.7848, + "step": 1731 + }, + { + "epoch": 1.898082191780822, + "grad_norm": 2.4138967990875244, + "learning_rate": 2.0714865962632007e-06, + "loss": 1.7204, + "step": 1732 + }, + { + "epoch": 1.899178082191781, + "grad_norm": 2.417165994644165, + "learning_rate": 2.069455727051178e-06, + "loss": 1.7741, + "step": 1733 + }, + { + "epoch": 1.9002739726027398, + "grad_norm": 2.4026505947113037, + "learning_rate": 2.0674248578391553e-06, + "loss": 1.7456, + "step": 1734 + }, + { + "epoch": 1.9013698630136986, + "grad_norm": 2.4637675285339355, + "learning_rate": 2.0653939886271326e-06, + "loss": 1.7679, + "step": 1735 + }, + { + "epoch": 1.9024657534246576, + "grad_norm": 2.269869089126587, + "learning_rate": 2.06336311941511e-06, + "loss": 1.8059, + "step": 1736 + }, + { + "epoch": 1.9035616438356164, + "grad_norm": 2.321094512939453, + "learning_rate": 2.061332250203087e-06, + "loss": 1.743, + "step": 1737 + }, + { + "epoch": 1.9046575342465752, + "grad_norm": 2.477288246154785, + "learning_rate": 2.0593013809910644e-06, + "loss": 1.7143, + "step": 1738 + }, + { + "epoch": 1.9057534246575343, + "grad_norm": 2.3924875259399414, + "learning_rate": 2.0572705117790417e-06, + "loss": 1.75, + "step": 1739 + }, + { + "epoch": 1.9068493150684933, + "grad_norm": 2.4452881813049316, + "learning_rate": 2.055239642567019e-06, + "loss": 1.7831, + "step": 1740 + }, + { + "epoch": 1.9068493150684933, + "eval_loss": 1.9599454402923584, + "eval_runtime": 37.3026, + "eval_samples_per_second": 26.808, + "eval_steps_per_second": 1.126, + "step": 1740 + }, + { + "epoch": 1.907945205479452, + "grad_norm": 2.2778351306915283, + "learning_rate": 2.0532087733549962e-06, + "loss": 1.7633, + "step": 1741 + }, + { + "epoch": 1.909041095890411, + "grad_norm": 2.3171589374542236, + "learning_rate": 2.0511779041429735e-06, + "loss": 1.7726, + "step": 1742 + }, + { + "epoch": 1.91013698630137, + "grad_norm": 2.286094903945923, + "learning_rate": 2.0491470349309508e-06, + "loss": 1.7458, + "step": 1743 + }, + { + "epoch": 1.9112328767123288, + "grad_norm": 2.5381743907928467, + "learning_rate": 2.047116165718928e-06, + "loss": 1.777, + "step": 1744 + }, + { + "epoch": 1.9123287671232876, + "grad_norm": 2.3261141777038574, + "learning_rate": 2.0450852965069053e-06, + "loss": 1.7715, + "step": 1745 + }, + { + "epoch": 1.9134246575342466, + "grad_norm": 2.351914167404175, + "learning_rate": 2.0430544272948826e-06, + "loss": 1.8214, + "step": 1746 + }, + { + "epoch": 1.9145205479452054, + "grad_norm": 2.274226427078247, + "learning_rate": 2.04102355808286e-06, + "loss": 1.763, + "step": 1747 + }, + { + "epoch": 1.9156164383561642, + "grad_norm": 2.4025075435638428, + "learning_rate": 2.038992688870837e-06, + "loss": 1.7072, + "step": 1748 + }, + { + "epoch": 1.9167123287671233, + "grad_norm": 2.334024429321289, + "learning_rate": 2.0369618196588144e-06, + "loss": 1.7381, + "step": 1749 + }, + { + "epoch": 1.9178082191780823, + "grad_norm": 2.3011622428894043, + "learning_rate": 2.0349309504467917e-06, + "loss": 1.7432, + "step": 1750 + }, + { + "epoch": 1.9178082191780823, + "eval_loss": 1.958349347114563, + "eval_runtime": 37.2591, + "eval_samples_per_second": 26.839, + "eval_steps_per_second": 1.127, + "step": 1750 + }, + { + "epoch": 1.9189041095890411, + "grad_norm": 2.643063545227051, + "learning_rate": 2.032900081234769e-06, + "loss": 1.7931, + "step": 1751 + }, + { + "epoch": 1.92, + "grad_norm": 2.3357834815979004, + "learning_rate": 2.030869212022746e-06, + "loss": 1.7933, + "step": 1752 + }, + { + "epoch": 1.921095890410959, + "grad_norm": 2.2614479064941406, + "learning_rate": 2.028838342810723e-06, + "loss": 1.7611, + "step": 1753 + }, + { + "epoch": 1.9221917808219178, + "grad_norm": 2.3057546615600586, + "learning_rate": 2.0268074735987004e-06, + "loss": 1.7406, + "step": 1754 + }, + { + "epoch": 1.9232876712328766, + "grad_norm": 2.6138038635253906, + "learning_rate": 2.0247766043866777e-06, + "loss": 1.7303, + "step": 1755 + }, + { + "epoch": 1.9243835616438356, + "grad_norm": 2.3803296089172363, + "learning_rate": 2.022745735174655e-06, + "loss": 1.8023, + "step": 1756 + }, + { + "epoch": 1.9254794520547946, + "grad_norm": 2.3436431884765625, + "learning_rate": 2.020714865962632e-06, + "loss": 1.7654, + "step": 1757 + }, + { + "epoch": 1.9265753424657535, + "grad_norm": 2.4103317260742188, + "learning_rate": 2.0186839967506095e-06, + "loss": 1.7932, + "step": 1758 + }, + { + "epoch": 1.9276712328767123, + "grad_norm": 2.4537675380706787, + "learning_rate": 2.0166531275385868e-06, + "loss": 1.761, + "step": 1759 + }, + { + "epoch": 1.9287671232876713, + "grad_norm": 2.3816988468170166, + "learning_rate": 2.014622258326564e-06, + "loss": 1.7276, + "step": 1760 + }, + { + "epoch": 1.9287671232876713, + "eval_loss": 1.9558053016662598, + "eval_runtime": 37.2417, + "eval_samples_per_second": 26.852, + "eval_steps_per_second": 1.128, + "step": 1760 + }, + { + "epoch": 1.9298630136986301, + "grad_norm": 2.37233304977417, + "learning_rate": 2.0125913891145413e-06, + "loss": 1.7674, + "step": 1761 + }, + { + "epoch": 1.930958904109589, + "grad_norm": 2.4043285846710205, + "learning_rate": 2.0105605199025186e-06, + "loss": 1.7638, + "step": 1762 + }, + { + "epoch": 1.932054794520548, + "grad_norm": 2.3402836322784424, + "learning_rate": 2.008529650690496e-06, + "loss": 1.7269, + "step": 1763 + }, + { + "epoch": 1.933150684931507, + "grad_norm": 2.42535400390625, + "learning_rate": 2.0064987814784727e-06, + "loss": 1.7965, + "step": 1764 + }, + { + "epoch": 1.9342465753424658, + "grad_norm": 2.361659288406372, + "learning_rate": 2.00446791226645e-06, + "loss": 1.7467, + "step": 1765 + }, + { + "epoch": 1.9353424657534246, + "grad_norm": 2.455253839492798, + "learning_rate": 2.0024370430544273e-06, + "loss": 1.7359, + "step": 1766 + }, + { + "epoch": 1.9364383561643836, + "grad_norm": 2.3062705993652344, + "learning_rate": 2.0004061738424045e-06, + "loss": 1.751, + "step": 1767 + }, + { + "epoch": 1.9375342465753425, + "grad_norm": 2.3683881759643555, + "learning_rate": 1.998375304630382e-06, + "loss": 1.8032, + "step": 1768 + }, + { + "epoch": 1.9386301369863013, + "grad_norm": 2.2175045013427734, + "learning_rate": 1.996344435418359e-06, + "loss": 1.7629, + "step": 1769 + }, + { + "epoch": 1.9397260273972603, + "grad_norm": 2.4331724643707275, + "learning_rate": 1.9943135662063364e-06, + "loss": 1.7899, + "step": 1770 + }, + { + "epoch": 1.9397260273972603, + "eval_loss": 1.9554719924926758, + "eval_runtime": 37.2636, + "eval_samples_per_second": 26.836, + "eval_steps_per_second": 1.127, + "step": 1770 + }, + { + "epoch": 1.9408219178082193, + "grad_norm": 2.370785713195801, + "learning_rate": 1.9922826969943136e-06, + "loss": 1.7687, + "step": 1771 + }, + { + "epoch": 1.941917808219178, + "grad_norm": 2.4047751426696777, + "learning_rate": 1.990251827782291e-06, + "loss": 1.804, + "step": 1772 + }, + { + "epoch": 1.943013698630137, + "grad_norm": 2.4597785472869873, + "learning_rate": 1.988220958570268e-06, + "loss": 1.7731, + "step": 1773 + }, + { + "epoch": 1.944109589041096, + "grad_norm": 2.388749837875366, + "learning_rate": 1.9861900893582455e-06, + "loss": 1.7945, + "step": 1774 + }, + { + "epoch": 1.9452054794520548, + "grad_norm": 2.3419551849365234, + "learning_rate": 1.9841592201462227e-06, + "loss": 1.7987, + "step": 1775 + }, + { + "epoch": 1.9463013698630136, + "grad_norm": 2.43770432472229, + "learning_rate": 1.9821283509342e-06, + "loss": 1.7183, + "step": 1776 + }, + { + "epoch": 1.9473972602739726, + "grad_norm": 2.333385467529297, + "learning_rate": 1.9800974817221773e-06, + "loss": 1.7283, + "step": 1777 + }, + { + "epoch": 1.9484931506849315, + "grad_norm": 2.391052722930908, + "learning_rate": 1.9780666125101546e-06, + "loss": 1.776, + "step": 1778 + }, + { + "epoch": 1.9495890410958903, + "grad_norm": 2.375338077545166, + "learning_rate": 1.976035743298132e-06, + "loss": 1.7495, + "step": 1779 + }, + { + "epoch": 1.9506849315068493, + "grad_norm": 2.4482579231262207, + "learning_rate": 1.974004874086109e-06, + "loss": 1.8256, + "step": 1780 + }, + { + "epoch": 1.9506849315068493, + "eval_loss": 1.9600462913513184, + "eval_runtime": 37.2638, + "eval_samples_per_second": 26.836, + "eval_steps_per_second": 1.127, + "step": 1780 + }, + { + "epoch": 1.9517808219178083, + "grad_norm": 2.296039342880249, + "learning_rate": 1.9719740048740864e-06, + "loss": 1.6996, + "step": 1781 + }, + { + "epoch": 1.9528767123287671, + "grad_norm": 2.2494969367980957, + "learning_rate": 1.9699431356620637e-06, + "loss": 1.7401, + "step": 1782 + }, + { + "epoch": 1.953972602739726, + "grad_norm": 2.4078574180603027, + "learning_rate": 1.967912266450041e-06, + "loss": 1.7374, + "step": 1783 + }, + { + "epoch": 1.955068493150685, + "grad_norm": 2.4018476009368896, + "learning_rate": 1.9658813972380182e-06, + "loss": 1.731, + "step": 1784 + }, + { + "epoch": 1.9561643835616438, + "grad_norm": 2.3429362773895264, + "learning_rate": 1.9638505280259955e-06, + "loss": 1.8531, + "step": 1785 + }, + { + "epoch": 1.9572602739726026, + "grad_norm": 2.4287197589874268, + "learning_rate": 1.9618196588139728e-06, + "loss": 1.7271, + "step": 1786 + }, + { + "epoch": 1.9583561643835616, + "grad_norm": 2.338541030883789, + "learning_rate": 1.95978878960195e-06, + "loss": 1.7594, + "step": 1787 + }, + { + "epoch": 1.9594520547945207, + "grad_norm": 2.531425952911377, + "learning_rate": 1.9577579203899273e-06, + "loss": 1.8374, + "step": 1788 + }, + { + "epoch": 1.9605479452054795, + "grad_norm": 2.413638114929199, + "learning_rate": 1.955727051177904e-06, + "loss": 1.8178, + "step": 1789 + }, + { + "epoch": 1.9616438356164383, + "grad_norm": 2.303318977355957, + "learning_rate": 1.9536961819658815e-06, + "loss": 1.7429, + "step": 1790 + }, + { + "epoch": 1.9616438356164383, + "eval_loss": 1.9622509479522705, + "eval_runtime": 37.2699, + "eval_samples_per_second": 26.831, + "eval_steps_per_second": 1.127, + "step": 1790 + }, + { + "epoch": 1.9627397260273973, + "grad_norm": 2.400160074234009, + "learning_rate": 1.9516653127538587e-06, + "loss": 1.7136, + "step": 1791 + }, + { + "epoch": 1.9638356164383561, + "grad_norm": 2.3861589431762695, + "learning_rate": 1.949634443541836e-06, + "loss": 1.7146, + "step": 1792 + }, + { + "epoch": 1.964931506849315, + "grad_norm": 2.272711753845215, + "learning_rate": 1.9476035743298133e-06, + "loss": 1.7177, + "step": 1793 + }, + { + "epoch": 1.966027397260274, + "grad_norm": 2.4745068550109863, + "learning_rate": 1.9455727051177906e-06, + "loss": 1.7938, + "step": 1794 + }, + { + "epoch": 1.967123287671233, + "grad_norm": 2.4643466472625732, + "learning_rate": 1.943541835905768e-06, + "loss": 1.7601, + "step": 1795 + }, + { + "epoch": 1.9682191780821918, + "grad_norm": 2.5346457958221436, + "learning_rate": 1.941510966693745e-06, + "loss": 1.8132, + "step": 1796 + }, + { + "epoch": 1.9693150684931506, + "grad_norm": 2.280719041824341, + "learning_rate": 1.9394800974817224e-06, + "loss": 1.7101, + "step": 1797 + }, + { + "epoch": 1.9704109589041097, + "grad_norm": 2.3304555416107178, + "learning_rate": 1.9374492282696997e-06, + "loss": 1.7809, + "step": 1798 + }, + { + "epoch": 1.9715068493150685, + "grad_norm": 2.409796714782715, + "learning_rate": 1.935418359057677e-06, + "loss": 1.8127, + "step": 1799 + }, + { + "epoch": 1.9726027397260273, + "grad_norm": 2.208312749862671, + "learning_rate": 1.9333874898456542e-06, + "loss": 1.7752, + "step": 1800 + }, + { + "epoch": 1.9726027397260273, + "eval_loss": 1.96150541305542, + "eval_runtime": 37.3221, + "eval_samples_per_second": 26.794, + "eval_steps_per_second": 1.125, + "step": 1800 + }, + { + "epoch": 1.9736986301369863, + "grad_norm": 2.374393939971924, + "learning_rate": 1.931356620633631e-06, + "loss": 1.8188, + "step": 1801 + }, + { + "epoch": 1.9747945205479454, + "grad_norm": 2.3343780040740967, + "learning_rate": 1.9293257514216083e-06, + "loss": 1.7035, + "step": 1802 + }, + { + "epoch": 1.9758904109589042, + "grad_norm": 2.307894229888916, + "learning_rate": 1.9272948822095856e-06, + "loss": 1.7651, + "step": 1803 + }, + { + "epoch": 1.976986301369863, + "grad_norm": 2.387434244155884, + "learning_rate": 1.925264012997563e-06, + "loss": 1.7021, + "step": 1804 + }, + { + "epoch": 1.978082191780822, + "grad_norm": 2.3444128036499023, + "learning_rate": 1.92323314378554e-06, + "loss": 1.7585, + "step": 1805 + }, + { + "epoch": 1.9791780821917808, + "grad_norm": 2.371931791305542, + "learning_rate": 1.9212022745735174e-06, + "loss": 1.7932, + "step": 1806 + }, + { + "epoch": 1.9802739726027396, + "grad_norm": 2.3159632682800293, + "learning_rate": 1.9191714053614947e-06, + "loss": 1.722, + "step": 1807 + }, + { + "epoch": 1.9813698630136987, + "grad_norm": 2.4669857025146484, + "learning_rate": 1.917140536149472e-06, + "loss": 1.7326, + "step": 1808 + }, + { + "epoch": 1.9824657534246577, + "grad_norm": 2.365138292312622, + "learning_rate": 1.9151096669374493e-06, + "loss": 1.7013, + "step": 1809 + }, + { + "epoch": 1.9835616438356163, + "grad_norm": 2.539698362350464, + "learning_rate": 1.9130787977254266e-06, + "loss": 1.8082, + "step": 1810 + }, + { + "epoch": 1.9835616438356163, + "eval_loss": 1.9535884857177734, + "eval_runtime": 37.3497, + "eval_samples_per_second": 26.774, + "eval_steps_per_second": 1.125, + "step": 1810 + }, + { + "epoch": 1.9846575342465753, + "grad_norm": 2.386599540710449, + "learning_rate": 1.911047928513404e-06, + "loss": 1.7835, + "step": 1811 + }, + { + "epoch": 1.9857534246575343, + "grad_norm": 2.376086711883545, + "learning_rate": 1.909017059301381e-06, + "loss": 1.7161, + "step": 1812 + }, + { + "epoch": 1.9868493150684932, + "grad_norm": 2.5276663303375244, + "learning_rate": 1.9069861900893586e-06, + "loss": 1.7222, + "step": 1813 + }, + { + "epoch": 1.987945205479452, + "grad_norm": 2.466137170791626, + "learning_rate": 1.9049553208773354e-06, + "loss": 1.7841, + "step": 1814 + }, + { + "epoch": 1.989041095890411, + "grad_norm": 2.361128091812134, + "learning_rate": 1.9029244516653127e-06, + "loss": 1.7354, + "step": 1815 + }, + { + "epoch": 1.9901369863013698, + "grad_norm": 2.4348530769348145, + "learning_rate": 1.90089358245329e-06, + "loss": 1.6832, + "step": 1816 + }, + { + "epoch": 1.9912328767123286, + "grad_norm": 2.342285633087158, + "learning_rate": 1.8988627132412673e-06, + "loss": 1.7351, + "step": 1817 + }, + { + "epoch": 1.9923287671232877, + "grad_norm": 2.521137237548828, + "learning_rate": 1.8968318440292445e-06, + "loss": 1.794, + "step": 1818 + }, + { + "epoch": 1.9934246575342467, + "grad_norm": 2.5968360900878906, + "learning_rate": 1.8948009748172218e-06, + "loss": 1.6976, + "step": 1819 + }, + { + "epoch": 1.9945205479452055, + "grad_norm": 2.734095335006714, + "learning_rate": 1.892770105605199e-06, + "loss": 1.7919, + "step": 1820 + }, + { + "epoch": 1.9945205479452055, + "eval_loss": 1.9536023139953613, + "eval_runtime": 37.5805, + "eval_samples_per_second": 26.61, + "eval_steps_per_second": 1.118, + "step": 1820 + }, + { + "epoch": 1.9956164383561643, + "grad_norm": 2.327333927154541, + "learning_rate": 1.8907392363931764e-06, + "loss": 1.7891, + "step": 1821 + }, + { + "epoch": 1.9967123287671233, + "grad_norm": 2.2902543544769287, + "learning_rate": 1.8887083671811536e-06, + "loss": 1.7552, + "step": 1822 + }, + { + "epoch": 1.9978082191780822, + "grad_norm": 2.3587660789489746, + "learning_rate": 1.886677497969131e-06, + "loss": 1.7641, + "step": 1823 + }, + { + "epoch": 1.998904109589041, + "grad_norm": 2.4068408012390137, + "learning_rate": 1.8846466287571082e-06, + "loss": 1.6622, + "step": 1824 + }, + { + "epoch": 2.0, + "grad_norm": 2.453514575958252, + "learning_rate": 1.8826157595450855e-06, + "loss": 1.7962, + "step": 1825 + }, + { + "epoch": 2.001095890410959, + "grad_norm": 2.4694175720214844, + "learning_rate": 1.8805848903330625e-06, + "loss": 1.6823, + "step": 1826 + }, + { + "epoch": 2.0021917808219176, + "grad_norm": 2.594013214111328, + "learning_rate": 1.8785540211210398e-06, + "loss": 1.8453, + "step": 1827 + }, + { + "epoch": 2.0032876712328767, + "grad_norm": 2.455796718597412, + "learning_rate": 1.876523151909017e-06, + "loss": 1.7604, + "step": 1828 + }, + { + "epoch": 2.0043835616438357, + "grad_norm": 2.819857120513916, + "learning_rate": 1.8744922826969944e-06, + "loss": 1.7177, + "step": 1829 + }, + { + "epoch": 2.0054794520547947, + "grad_norm": 2.845775842666626, + "learning_rate": 1.8724614134849716e-06, + "loss": 1.7007, + "step": 1830 + }, + { + "epoch": 2.0054794520547947, + "eval_loss": 1.9594684839248657, + "eval_runtime": 37.6856, + "eval_samples_per_second": 26.535, + "eval_steps_per_second": 1.114, + "step": 1830 + }, + { + "epoch": 2.0065753424657533, + "grad_norm": 2.6129424571990967, + "learning_rate": 1.870430544272949e-06, + "loss": 1.7301, + "step": 1831 + }, + { + "epoch": 2.0076712328767123, + "grad_norm": 2.762805223464966, + "learning_rate": 1.8683996750609262e-06, + "loss": 1.8294, + "step": 1832 + }, + { + "epoch": 2.0087671232876714, + "grad_norm": 2.6397714614868164, + "learning_rate": 1.8663688058489035e-06, + "loss": 1.7176, + "step": 1833 + }, + { + "epoch": 2.00986301369863, + "grad_norm": 2.6637110710144043, + "learning_rate": 1.8643379366368807e-06, + "loss": 1.7363, + "step": 1834 + }, + { + "epoch": 2.010958904109589, + "grad_norm": 2.732534408569336, + "learning_rate": 1.862307067424858e-06, + "loss": 1.7496, + "step": 1835 + }, + { + "epoch": 2.012054794520548, + "grad_norm": 2.7163467407226562, + "learning_rate": 1.8602761982128353e-06, + "loss": 1.7551, + "step": 1836 + }, + { + "epoch": 2.013150684931507, + "grad_norm": 2.442484140396118, + "learning_rate": 1.8582453290008126e-06, + "loss": 1.7761, + "step": 1837 + }, + { + "epoch": 2.0142465753424657, + "grad_norm": 2.544834852218628, + "learning_rate": 1.8562144597887898e-06, + "loss": 1.7251, + "step": 1838 + }, + { + "epoch": 2.0153424657534247, + "grad_norm": 2.6588494777679443, + "learning_rate": 1.854183590576767e-06, + "loss": 1.7381, + "step": 1839 + }, + { + "epoch": 2.0164383561643837, + "grad_norm": 2.5374786853790283, + "learning_rate": 1.8521527213647442e-06, + "loss": 1.7831, + "step": 1840 + }, + { + "epoch": 2.0164383561643837, + "eval_loss": 1.9550480842590332, + "eval_runtime": 37.3312, + "eval_samples_per_second": 26.787, + "eval_steps_per_second": 1.125, + "step": 1840 + }, + { + "epoch": 2.0175342465753423, + "grad_norm": 2.4791345596313477, + "learning_rate": 1.8501218521527215e-06, + "loss": 1.6937, + "step": 1841 + }, + { + "epoch": 2.0186301369863013, + "grad_norm": 2.7243971824645996, + "learning_rate": 1.8480909829406987e-06, + "loss": 1.6674, + "step": 1842 + }, + { + "epoch": 2.0197260273972604, + "grad_norm": 2.618680477142334, + "learning_rate": 1.846060113728676e-06, + "loss": 1.6994, + "step": 1843 + }, + { + "epoch": 2.020821917808219, + "grad_norm": 2.597929000854492, + "learning_rate": 1.8440292445166533e-06, + "loss": 1.7375, + "step": 1844 + }, + { + "epoch": 2.021917808219178, + "grad_norm": 2.860177993774414, + "learning_rate": 1.8419983753046306e-06, + "loss": 1.7128, + "step": 1845 + }, + { + "epoch": 2.023013698630137, + "grad_norm": 2.606898307800293, + "learning_rate": 1.8399675060926078e-06, + "loss": 1.7735, + "step": 1846 + }, + { + "epoch": 2.024109589041096, + "grad_norm": 2.5152604579925537, + "learning_rate": 1.8379366368805851e-06, + "loss": 1.75, + "step": 1847 + }, + { + "epoch": 2.0252054794520546, + "grad_norm": 2.678071975708008, + "learning_rate": 1.8359057676685624e-06, + "loss": 1.7114, + "step": 1848 + }, + { + "epoch": 2.0263013698630137, + "grad_norm": 2.673515558242798, + "learning_rate": 1.8338748984565397e-06, + "loss": 1.6992, + "step": 1849 + }, + { + "epoch": 2.0273972602739727, + "grad_norm": 2.5902516841888428, + "learning_rate": 1.831844029244517e-06, + "loss": 1.7195, + "step": 1850 + }, + { + "epoch": 2.0273972602739727, + "eval_loss": 1.953579068183899, + "eval_runtime": 37.304, + "eval_samples_per_second": 26.807, + "eval_steps_per_second": 1.126, + "step": 1850 + }, + { + "epoch": 2.0284931506849313, + "grad_norm": 2.4912915229797363, + "learning_rate": 1.829813160032494e-06, + "loss": 1.6955, + "step": 1851 + }, + { + "epoch": 2.0295890410958903, + "grad_norm": 2.683821678161621, + "learning_rate": 1.8277822908204713e-06, + "loss": 1.7057, + "step": 1852 + }, + { + "epoch": 2.0306849315068494, + "grad_norm": 2.9541101455688477, + "learning_rate": 1.8257514216084486e-06, + "loss": 1.7377, + "step": 1853 + }, + { + "epoch": 2.0317808219178084, + "grad_norm": 2.783851146697998, + "learning_rate": 1.8237205523964258e-06, + "loss": 1.7197, + "step": 1854 + }, + { + "epoch": 2.032876712328767, + "grad_norm": 2.5224950313568115, + "learning_rate": 1.8216896831844031e-06, + "loss": 1.7396, + "step": 1855 + }, + { + "epoch": 2.033972602739726, + "grad_norm": 2.627767562866211, + "learning_rate": 1.8196588139723804e-06, + "loss": 1.7773, + "step": 1856 + }, + { + "epoch": 2.035068493150685, + "grad_norm": 2.443873405456543, + "learning_rate": 1.8176279447603577e-06, + "loss": 1.6656, + "step": 1857 + }, + { + "epoch": 2.0361643835616436, + "grad_norm": 2.648056745529175, + "learning_rate": 1.815597075548335e-06, + "loss": 1.6954, + "step": 1858 + }, + { + "epoch": 2.0372602739726027, + "grad_norm": 2.536935329437256, + "learning_rate": 1.8135662063363122e-06, + "loss": 1.7427, + "step": 1859 + }, + { + "epoch": 2.0383561643835617, + "grad_norm": 2.42183256149292, + "learning_rate": 1.8115353371242895e-06, + "loss": 1.7634, + "step": 1860 + }, + { + "epoch": 2.0383561643835617, + "eval_loss": 1.953397512435913, + "eval_runtime": 37.2886, + "eval_samples_per_second": 26.818, + "eval_steps_per_second": 1.126, + "step": 1860 + }, + { + "epoch": 2.0394520547945207, + "grad_norm": 2.538804769515991, + "learning_rate": 1.8095044679122668e-06, + "loss": 1.6712, + "step": 1861 + }, + { + "epoch": 2.0405479452054793, + "grad_norm": 2.5820095539093018, + "learning_rate": 1.807473598700244e-06, + "loss": 1.7215, + "step": 1862 + }, + { + "epoch": 2.0416438356164384, + "grad_norm": 2.6997382640838623, + "learning_rate": 1.805442729488221e-06, + "loss": 1.7117, + "step": 1863 + }, + { + "epoch": 2.0427397260273974, + "grad_norm": 2.775996685028076, + "learning_rate": 1.8034118602761982e-06, + "loss": 1.7355, + "step": 1864 + }, + { + "epoch": 2.043835616438356, + "grad_norm": 2.6333167552948, + "learning_rate": 1.8013809910641754e-06, + "loss": 1.7343, + "step": 1865 + }, + { + "epoch": 2.044931506849315, + "grad_norm": 2.5830278396606445, + "learning_rate": 1.7993501218521527e-06, + "loss": 1.7339, + "step": 1866 + }, + { + "epoch": 2.046027397260274, + "grad_norm": 2.6145055294036865, + "learning_rate": 1.79731925264013e-06, + "loss": 1.6835, + "step": 1867 + }, + { + "epoch": 2.047123287671233, + "grad_norm": 2.568159580230713, + "learning_rate": 1.7952883834281073e-06, + "loss": 1.7574, + "step": 1868 + }, + { + "epoch": 2.0482191780821917, + "grad_norm": 2.5235254764556885, + "learning_rate": 1.7932575142160846e-06, + "loss": 1.7067, + "step": 1869 + }, + { + "epoch": 2.0493150684931507, + "grad_norm": 2.464035987854004, + "learning_rate": 1.7912266450040618e-06, + "loss": 1.7451, + "step": 1870 + }, + { + "epoch": 2.0493150684931507, + "eval_loss": 1.9485031366348267, + "eval_runtime": 37.2648, + "eval_samples_per_second": 26.835, + "eval_steps_per_second": 1.127, + "step": 1870 + }, + { + "epoch": 2.0504109589041097, + "grad_norm": 2.4328689575195312, + "learning_rate": 1.789195775792039e-06, + "loss": 1.7508, + "step": 1871 + }, + { + "epoch": 2.0515068493150683, + "grad_norm": 2.5483932495117188, + "learning_rate": 1.7871649065800164e-06, + "loss": 1.6488, + "step": 1872 + }, + { + "epoch": 2.0526027397260274, + "grad_norm": 2.5290846824645996, + "learning_rate": 1.7851340373679937e-06, + "loss": 1.7939, + "step": 1873 + }, + { + "epoch": 2.0536986301369864, + "grad_norm": 2.506943702697754, + "learning_rate": 1.783103168155971e-06, + "loss": 1.6813, + "step": 1874 + }, + { + "epoch": 2.0547945205479454, + "grad_norm": 2.5863115787506104, + "learning_rate": 1.7810722989439482e-06, + "loss": 1.7104, + "step": 1875 + }, + { + "epoch": 2.055890410958904, + "grad_norm": 2.6402666568756104, + "learning_rate": 1.7790414297319253e-06, + "loss": 1.7311, + "step": 1876 + }, + { + "epoch": 2.056986301369863, + "grad_norm": 2.6700708866119385, + "learning_rate": 1.7770105605199025e-06, + "loss": 1.695, + "step": 1877 + }, + { + "epoch": 2.058082191780822, + "grad_norm": 2.580693244934082, + "learning_rate": 1.7749796913078798e-06, + "loss": 1.6763, + "step": 1878 + }, + { + "epoch": 2.0591780821917807, + "grad_norm": 2.6035709381103516, + "learning_rate": 1.772948822095857e-06, + "loss": 1.7389, + "step": 1879 + }, + { + "epoch": 2.0602739726027397, + "grad_norm": 2.508784532546997, + "learning_rate": 1.7709179528838344e-06, + "loss": 1.7762, + "step": 1880 + }, + { + "epoch": 2.0602739726027397, + "eval_loss": 1.9516879320144653, + "eval_runtime": 37.2284, + "eval_samples_per_second": 26.861, + "eval_steps_per_second": 1.128, + "step": 1880 + }, + { + "epoch": 2.0613698630136987, + "grad_norm": 2.47988224029541, + "learning_rate": 1.7688870836718116e-06, + "loss": 1.7659, + "step": 1881 + }, + { + "epoch": 2.0624657534246573, + "grad_norm": 2.5782723426818848, + "learning_rate": 1.766856214459789e-06, + "loss": 1.7179, + "step": 1882 + }, + { + "epoch": 2.0635616438356164, + "grad_norm": 2.5616347789764404, + "learning_rate": 1.7648253452477662e-06, + "loss": 1.7163, + "step": 1883 + }, + { + "epoch": 2.0646575342465754, + "grad_norm": 2.4464340209960938, + "learning_rate": 1.7627944760357435e-06, + "loss": 1.7033, + "step": 1884 + }, + { + "epoch": 2.0657534246575344, + "grad_norm": 2.6022887229919434, + "learning_rate": 1.7607636068237208e-06, + "loss": 1.7823, + "step": 1885 + }, + { + "epoch": 2.066849315068493, + "grad_norm": 2.5734951496124268, + "learning_rate": 1.758732737611698e-06, + "loss": 1.7486, + "step": 1886 + }, + { + "epoch": 2.067945205479452, + "grad_norm": 2.6162502765655518, + "learning_rate": 1.7567018683996753e-06, + "loss": 1.7285, + "step": 1887 + }, + { + "epoch": 2.069041095890411, + "grad_norm": 2.654008388519287, + "learning_rate": 1.7546709991876524e-06, + "loss": 1.7195, + "step": 1888 + }, + { + "epoch": 2.0701369863013697, + "grad_norm": 2.64469575881958, + "learning_rate": 1.7526401299756296e-06, + "loss": 1.7247, + "step": 1889 + }, + { + "epoch": 2.0712328767123287, + "grad_norm": 2.5627946853637695, + "learning_rate": 1.750609260763607e-06, + "loss": 1.7254, + "step": 1890 + }, + { + "epoch": 2.0712328767123287, + "eval_loss": 1.951324224472046, + "eval_runtime": 37.2566, + "eval_samples_per_second": 26.841, + "eval_steps_per_second": 1.127, + "step": 1890 + }, + { + "epoch": 2.0723287671232877, + "grad_norm": 2.656520128250122, + "learning_rate": 1.7485783915515842e-06, + "loss": 1.7236, + "step": 1891 + }, + { + "epoch": 2.0734246575342468, + "grad_norm": 2.6355814933776855, + "learning_rate": 1.7465475223395615e-06, + "loss": 1.7118, + "step": 1892 + }, + { + "epoch": 2.0745205479452054, + "grad_norm": 2.5709362030029297, + "learning_rate": 1.7445166531275387e-06, + "loss": 1.685, + "step": 1893 + }, + { + "epoch": 2.0756164383561644, + "grad_norm": 2.555422782897949, + "learning_rate": 1.742485783915516e-06, + "loss": 1.8356, + "step": 1894 + }, + { + "epoch": 2.0767123287671234, + "grad_norm": 2.6279077529907227, + "learning_rate": 1.7404549147034933e-06, + "loss": 1.8079, + "step": 1895 + }, + { + "epoch": 2.077808219178082, + "grad_norm": 2.5886247158050537, + "learning_rate": 1.7384240454914706e-06, + "loss": 1.639, + "step": 1896 + }, + { + "epoch": 2.078904109589041, + "grad_norm": 2.562755584716797, + "learning_rate": 1.7363931762794478e-06, + "loss": 1.7345, + "step": 1897 + }, + { + "epoch": 2.08, + "grad_norm": 2.670344591140747, + "learning_rate": 1.7343623070674251e-06, + "loss": 1.7594, + "step": 1898 + }, + { + "epoch": 2.081095890410959, + "grad_norm": 2.567506790161133, + "learning_rate": 1.7323314378554024e-06, + "loss": 1.7161, + "step": 1899 + }, + { + "epoch": 2.0821917808219177, + "grad_norm": 2.562826156616211, + "learning_rate": 1.7303005686433797e-06, + "loss": 1.7275, + "step": 1900 + }, + { + "epoch": 2.0821917808219177, + "eval_loss": 1.9498006105422974, + "eval_runtime": 37.3069, + "eval_samples_per_second": 26.805, + "eval_steps_per_second": 1.126, + "step": 1900 + }, + { + "epoch": 2.0832876712328767, + "grad_norm": 2.538393974304199, + "learning_rate": 1.7282696994313567e-06, + "loss": 1.7187, + "step": 1901 + }, + { + "epoch": 2.0843835616438358, + "grad_norm": 2.687603235244751, + "learning_rate": 1.726238830219334e-06, + "loss": 1.7349, + "step": 1902 + }, + { + "epoch": 2.0854794520547943, + "grad_norm": 2.5806703567504883, + "learning_rate": 1.7242079610073113e-06, + "loss": 1.7135, + "step": 1903 + }, + { + "epoch": 2.0865753424657534, + "grad_norm": 2.468123435974121, + "learning_rate": 1.7221770917952886e-06, + "loss": 1.7946, + "step": 1904 + }, + { + "epoch": 2.0876712328767124, + "grad_norm": 2.514617443084717, + "learning_rate": 1.7201462225832658e-06, + "loss": 1.7403, + "step": 1905 + }, + { + "epoch": 2.0887671232876714, + "grad_norm": 2.7283825874328613, + "learning_rate": 1.7181153533712431e-06, + "loss": 1.7369, + "step": 1906 + }, + { + "epoch": 2.08986301369863, + "grad_norm": 2.5634899139404297, + "learning_rate": 1.7160844841592204e-06, + "loss": 1.6337, + "step": 1907 + }, + { + "epoch": 2.090958904109589, + "grad_norm": 2.635281801223755, + "learning_rate": 1.7140536149471977e-06, + "loss": 1.6836, + "step": 1908 + }, + { + "epoch": 2.092054794520548, + "grad_norm": 2.550048351287842, + "learning_rate": 1.712022745735175e-06, + "loss": 1.7171, + "step": 1909 + }, + { + "epoch": 2.0931506849315067, + "grad_norm": 2.62712025642395, + "learning_rate": 1.7099918765231522e-06, + "loss": 1.7027, + "step": 1910 + }, + { + "epoch": 2.0931506849315067, + "eval_loss": 1.9503322839736938, + "eval_runtime": 37.3435, + "eval_samples_per_second": 26.778, + "eval_steps_per_second": 1.125, + "step": 1910 + }, + { + "epoch": 2.0942465753424657, + "grad_norm": 2.540708541870117, + "learning_rate": 1.7079610073111295e-06, + "loss": 1.7358, + "step": 1911 + }, + { + "epoch": 2.0953424657534248, + "grad_norm": 2.6046950817108154, + "learning_rate": 1.7059301380991068e-06, + "loss": 1.6538, + "step": 1912 + }, + { + "epoch": 2.0964383561643833, + "grad_norm": 2.4760568141937256, + "learning_rate": 1.7038992688870836e-06, + "loss": 1.7207, + "step": 1913 + }, + { + "epoch": 2.0975342465753424, + "grad_norm": 2.428621292114258, + "learning_rate": 1.701868399675061e-06, + "loss": 1.7092, + "step": 1914 + }, + { + "epoch": 2.0986301369863014, + "grad_norm": 2.6177377700805664, + "learning_rate": 1.6998375304630382e-06, + "loss": 1.7429, + "step": 1915 + }, + { + "epoch": 2.0997260273972604, + "grad_norm": 2.4981932640075684, + "learning_rate": 1.6978066612510155e-06, + "loss": 1.7996, + "step": 1916 + }, + { + "epoch": 2.100821917808219, + "grad_norm": 2.617943286895752, + "learning_rate": 1.6957757920389927e-06, + "loss": 1.7716, + "step": 1917 + }, + { + "epoch": 2.101917808219178, + "grad_norm": 2.7581987380981445, + "learning_rate": 1.69374492282697e-06, + "loss": 1.7194, + "step": 1918 + }, + { + "epoch": 2.103013698630137, + "grad_norm": 2.622145652770996, + "learning_rate": 1.6917140536149473e-06, + "loss": 1.6887, + "step": 1919 + }, + { + "epoch": 2.1041095890410957, + "grad_norm": 2.742100477218628, + "learning_rate": 1.6896831844029246e-06, + "loss": 1.7787, + "step": 1920 + }, + { + "epoch": 2.1041095890410957, + "eval_loss": 1.9521653652191162, + "eval_runtime": 37.5529, + "eval_samples_per_second": 26.629, + "eval_steps_per_second": 1.118, + "step": 1920 + }, + { + "epoch": 2.1052054794520547, + "grad_norm": 2.5809876918792725, + "learning_rate": 1.6876523151909018e-06, + "loss": 1.7392, + "step": 1921 + }, + { + "epoch": 2.1063013698630138, + "grad_norm": 2.7006125450134277, + "learning_rate": 1.6856214459788791e-06, + "loss": 1.7442, + "step": 1922 + }, + { + "epoch": 2.107397260273973, + "grad_norm": 2.566088914871216, + "learning_rate": 1.6835905767668564e-06, + "loss": 1.6862, + "step": 1923 + }, + { + "epoch": 2.1084931506849314, + "grad_norm": 2.508967638015747, + "learning_rate": 1.6815597075548337e-06, + "loss": 1.695, + "step": 1924 + }, + { + "epoch": 2.1095890410958904, + "grad_norm": 2.6175615787506104, + "learning_rate": 1.679528838342811e-06, + "loss": 1.785, + "step": 1925 + }, + { + "epoch": 2.1106849315068494, + "grad_norm": 2.6141958236694336, + "learning_rate": 1.677497969130788e-06, + "loss": 1.6952, + "step": 1926 + }, + { + "epoch": 2.111780821917808, + "grad_norm": 2.4894959926605225, + "learning_rate": 1.6754670999187653e-06, + "loss": 1.6947, + "step": 1927 + }, + { + "epoch": 2.112876712328767, + "grad_norm": 2.5664544105529785, + "learning_rate": 1.6734362307067426e-06, + "loss": 1.731, + "step": 1928 + }, + { + "epoch": 2.113972602739726, + "grad_norm": 2.5422523021698, + "learning_rate": 1.6714053614947198e-06, + "loss": 1.6765, + "step": 1929 + }, + { + "epoch": 2.115068493150685, + "grad_norm": 2.6428608894348145, + "learning_rate": 1.669374492282697e-06, + "loss": 1.6834, + "step": 1930 + }, + { + "epoch": 2.115068493150685, + "eval_loss": 1.952999472618103, + "eval_runtime": 37.6765, + "eval_samples_per_second": 26.542, + "eval_steps_per_second": 1.115, + "step": 1930 + }, + { + "epoch": 2.1161643835616437, + "grad_norm": 2.6088430881500244, + "learning_rate": 1.6673436230706744e-06, + "loss": 1.7713, + "step": 1931 + }, + { + "epoch": 2.1172602739726027, + "grad_norm": 2.49733304977417, + "learning_rate": 1.6653127538586517e-06, + "loss": 1.7461, + "step": 1932 + }, + { + "epoch": 2.118356164383562, + "grad_norm": 2.5760915279388428, + "learning_rate": 1.663281884646629e-06, + "loss": 1.6837, + "step": 1933 + }, + { + "epoch": 2.1194520547945204, + "grad_norm": 2.4771194458007812, + "learning_rate": 1.6612510154346062e-06, + "loss": 1.7191, + "step": 1934 + }, + { + "epoch": 2.1205479452054794, + "grad_norm": 2.6267881393432617, + "learning_rate": 1.6592201462225835e-06, + "loss": 1.7819, + "step": 1935 + }, + { + "epoch": 2.1216438356164384, + "grad_norm": 2.6205623149871826, + "learning_rate": 1.6571892770105608e-06, + "loss": 1.7443, + "step": 1936 + }, + { + "epoch": 2.1227397260273975, + "grad_norm": 2.660942792892456, + "learning_rate": 1.655158407798538e-06, + "loss": 1.783, + "step": 1937 + }, + { + "epoch": 2.123835616438356, + "grad_norm": 2.5452628135681152, + "learning_rate": 1.653127538586515e-06, + "loss": 1.7316, + "step": 1938 + }, + { + "epoch": 2.124931506849315, + "grad_norm": 2.542837142944336, + "learning_rate": 1.6510966693744924e-06, + "loss": 1.7624, + "step": 1939 + }, + { + "epoch": 2.126027397260274, + "grad_norm": 2.535205841064453, + "learning_rate": 1.6490658001624696e-06, + "loss": 1.7664, + "step": 1940 + }, + { + "epoch": 2.126027397260274, + "eval_loss": 1.9492406845092773, + "eval_runtime": 37.3497, + "eval_samples_per_second": 26.774, + "eval_steps_per_second": 1.125, + "step": 1940 + }, + { + "epoch": 2.1271232876712327, + "grad_norm": 2.7622485160827637, + "learning_rate": 1.647034930950447e-06, + "loss": 1.7603, + "step": 1941 + }, + { + "epoch": 2.1282191780821917, + "grad_norm": 2.629058599472046, + "learning_rate": 1.6450040617384242e-06, + "loss": 1.7036, + "step": 1942 + }, + { + "epoch": 2.1293150684931508, + "grad_norm": 2.543135643005371, + "learning_rate": 1.6429731925264015e-06, + "loss": 1.7172, + "step": 1943 + }, + { + "epoch": 2.1304109589041094, + "grad_norm": 2.792086601257324, + "learning_rate": 1.6409423233143788e-06, + "loss": 1.7486, + "step": 1944 + }, + { + "epoch": 2.1315068493150684, + "grad_norm": 2.7788681983947754, + "learning_rate": 1.638911454102356e-06, + "loss": 1.7136, + "step": 1945 + }, + { + "epoch": 2.1326027397260274, + "grad_norm": 2.5364747047424316, + "learning_rate": 1.6368805848903333e-06, + "loss": 1.6766, + "step": 1946 + }, + { + "epoch": 2.1336986301369865, + "grad_norm": 2.5749800205230713, + "learning_rate": 1.6348497156783106e-06, + "loss": 1.7627, + "step": 1947 + }, + { + "epoch": 2.134794520547945, + "grad_norm": 2.4931676387786865, + "learning_rate": 1.6328188464662879e-06, + "loss": 1.7049, + "step": 1948 + }, + { + "epoch": 2.135890410958904, + "grad_norm": 2.547396183013916, + "learning_rate": 1.6307879772542651e-06, + "loss": 1.7165, + "step": 1949 + }, + { + "epoch": 2.136986301369863, + "grad_norm": 2.5881359577178955, + "learning_rate": 1.628757108042242e-06, + "loss": 1.7508, + "step": 1950 + }, + { + "epoch": 2.136986301369863, + "eval_loss": 1.9464337825775146, + "eval_runtime": 37.284, + "eval_samples_per_second": 26.821, + "eval_steps_per_second": 1.126, + "step": 1950 + }, + { + "epoch": 2.138082191780822, + "grad_norm": 2.6131961345672607, + "learning_rate": 1.6267262388302193e-06, + "loss": 1.7717, + "step": 1951 + }, + { + "epoch": 2.1391780821917807, + "grad_norm": 2.5644335746765137, + "learning_rate": 1.6246953696181965e-06, + "loss": 1.7139, + "step": 1952 + }, + { + "epoch": 2.1402739726027398, + "grad_norm": 2.5851385593414307, + "learning_rate": 1.6226645004061738e-06, + "loss": 1.7473, + "step": 1953 + }, + { + "epoch": 2.141369863013699, + "grad_norm": 2.493952989578247, + "learning_rate": 1.620633631194151e-06, + "loss": 1.8209, + "step": 1954 + }, + { + "epoch": 2.1424657534246574, + "grad_norm": 2.5686917304992676, + "learning_rate": 1.6186027619821284e-06, + "loss": 1.7527, + "step": 1955 + }, + { + "epoch": 2.1435616438356164, + "grad_norm": 2.6349050998687744, + "learning_rate": 1.6165718927701056e-06, + "loss": 1.7033, + "step": 1956 + }, + { + "epoch": 2.1446575342465755, + "grad_norm": 2.7004778385162354, + "learning_rate": 1.614541023558083e-06, + "loss": 1.7498, + "step": 1957 + }, + { + "epoch": 2.145753424657534, + "grad_norm": 2.5406992435455322, + "learning_rate": 1.6125101543460602e-06, + "loss": 1.7605, + "step": 1958 + }, + { + "epoch": 2.146849315068493, + "grad_norm": 2.796125650405884, + "learning_rate": 1.6104792851340375e-06, + "loss": 1.7087, + "step": 1959 + }, + { + "epoch": 2.147945205479452, + "grad_norm": 2.703362226486206, + "learning_rate": 1.6084484159220147e-06, + "loss": 1.751, + "step": 1960 + }, + { + "epoch": 2.147945205479452, + "eval_loss": 1.9470759630203247, + "eval_runtime": 37.2721, + "eval_samples_per_second": 26.83, + "eval_steps_per_second": 1.127, + "step": 1960 + }, + { + "epoch": 2.149041095890411, + "grad_norm": 2.6464223861694336, + "learning_rate": 1.606417546709992e-06, + "loss": 1.7147, + "step": 1961 + }, + { + "epoch": 2.1501369863013697, + "grad_norm": 2.6499273777008057, + "learning_rate": 1.6043866774979693e-06, + "loss": 1.7015, + "step": 1962 + }, + { + "epoch": 2.1512328767123288, + "grad_norm": 2.583768844604492, + "learning_rate": 1.6023558082859464e-06, + "loss": 1.7158, + "step": 1963 + }, + { + "epoch": 2.152328767123288, + "grad_norm": 2.6245381832122803, + "learning_rate": 1.6003249390739236e-06, + "loss": 1.6824, + "step": 1964 + }, + { + "epoch": 2.1534246575342464, + "grad_norm": 2.497676134109497, + "learning_rate": 1.598294069861901e-06, + "loss": 1.7345, + "step": 1965 + }, + { + "epoch": 2.1545205479452054, + "grad_norm": 2.7218856811523438, + "learning_rate": 1.5962632006498782e-06, + "loss": 1.6988, + "step": 1966 + }, + { + "epoch": 2.1556164383561645, + "grad_norm": 2.7654666900634766, + "learning_rate": 1.5942323314378555e-06, + "loss": 1.7635, + "step": 1967 + }, + { + "epoch": 2.1567123287671235, + "grad_norm": 2.598526954650879, + "learning_rate": 1.5922014622258327e-06, + "loss": 1.6769, + "step": 1968 + }, + { + "epoch": 2.157808219178082, + "grad_norm": 2.5995986461639404, + "learning_rate": 1.59017059301381e-06, + "loss": 1.6928, + "step": 1969 + }, + { + "epoch": 2.158904109589041, + "grad_norm": 2.5692837238311768, + "learning_rate": 1.5881397238017873e-06, + "loss": 1.7415, + "step": 1970 + }, + { + "epoch": 2.158904109589041, + "eval_loss": 1.9472142457962036, + "eval_runtime": 37.258, + "eval_samples_per_second": 26.84, + "eval_steps_per_second": 1.127, + "step": 1970 + }, + { + "epoch": 2.16, + "grad_norm": 2.622732162475586, + "learning_rate": 1.5861088545897646e-06, + "loss": 1.7383, + "step": 1971 + }, + { + "epoch": 2.1610958904109587, + "grad_norm": 2.573005437850952, + "learning_rate": 1.5840779853777418e-06, + "loss": 1.6696, + "step": 1972 + }, + { + "epoch": 2.1621917808219178, + "grad_norm": 2.7022411823272705, + "learning_rate": 1.5820471161657191e-06, + "loss": 1.7231, + "step": 1973 + }, + { + "epoch": 2.163287671232877, + "grad_norm": 2.739145278930664, + "learning_rate": 1.5800162469536964e-06, + "loss": 1.7012, + "step": 1974 + }, + { + "epoch": 2.1643835616438354, + "grad_norm": 2.7034459114074707, + "learning_rate": 1.5779853777416735e-06, + "loss": 1.769, + "step": 1975 + }, + { + "epoch": 2.1654794520547944, + "grad_norm": 2.745359182357788, + "learning_rate": 1.5759545085296507e-06, + "loss": 1.7424, + "step": 1976 + }, + { + "epoch": 2.1665753424657535, + "grad_norm": 2.722721576690674, + "learning_rate": 1.573923639317628e-06, + "loss": 1.733, + "step": 1977 + }, + { + "epoch": 2.1676712328767125, + "grad_norm": 2.689074754714966, + "learning_rate": 1.5718927701056053e-06, + "loss": 1.7468, + "step": 1978 + }, + { + "epoch": 2.168767123287671, + "grad_norm": 2.6735317707061768, + "learning_rate": 1.5698619008935826e-06, + "loss": 1.7312, + "step": 1979 + }, + { + "epoch": 2.16986301369863, + "grad_norm": 2.6908175945281982, + "learning_rate": 1.5678310316815598e-06, + "loss": 1.7502, + "step": 1980 + }, + { + "epoch": 2.16986301369863, + "eval_loss": 1.947867512702942, + "eval_runtime": 37.1858, + "eval_samples_per_second": 26.892, + "eval_steps_per_second": 1.129, + "step": 1980 + }, + { + "epoch": 2.170958904109589, + "grad_norm": 2.524951934814453, + "learning_rate": 1.5658001624695371e-06, + "loss": 1.6863, + "step": 1981 + }, + { + "epoch": 2.172054794520548, + "grad_norm": 2.746180295944214, + "learning_rate": 1.5637692932575144e-06, + "loss": 1.7621, + "step": 1982 + }, + { + "epoch": 2.1731506849315068, + "grad_norm": 2.4799301624298096, + "learning_rate": 1.5617384240454917e-06, + "loss": 1.7121, + "step": 1983 + }, + { + "epoch": 2.174246575342466, + "grad_norm": 2.6821515560150146, + "learning_rate": 1.559707554833469e-06, + "loss": 1.7271, + "step": 1984 + }, + { + "epoch": 2.175342465753425, + "grad_norm": 2.572122812271118, + "learning_rate": 1.5576766856214462e-06, + "loss": 1.7671, + "step": 1985 + }, + { + "epoch": 2.1764383561643834, + "grad_norm": 2.614806652069092, + "learning_rate": 1.5556458164094235e-06, + "loss": 1.6411, + "step": 1986 + }, + { + "epoch": 2.1775342465753424, + "grad_norm": 2.602883815765381, + "learning_rate": 1.5536149471974008e-06, + "loss": 1.7143, + "step": 1987 + }, + { + "epoch": 2.1786301369863015, + "grad_norm": 2.5893354415893555, + "learning_rate": 1.5515840779853778e-06, + "loss": 1.6733, + "step": 1988 + }, + { + "epoch": 2.17972602739726, + "grad_norm": 2.652308702468872, + "learning_rate": 1.549553208773355e-06, + "loss": 1.7311, + "step": 1989 + }, + { + "epoch": 2.180821917808219, + "grad_norm": 2.7129695415496826, + "learning_rate": 1.5475223395613324e-06, + "loss": 1.7001, + "step": 1990 + }, + { + "epoch": 2.180821917808219, + "eval_loss": 1.9512702226638794, + "eval_runtime": 37.225, + "eval_samples_per_second": 26.864, + "eval_steps_per_second": 1.128, + "step": 1990 + }, + { + "epoch": 2.181917808219178, + "grad_norm": 2.563638210296631, + "learning_rate": 1.5454914703493097e-06, + "loss": 1.7118, + "step": 1991 + }, + { + "epoch": 2.183013698630137, + "grad_norm": 2.629056453704834, + "learning_rate": 1.543460601137287e-06, + "loss": 1.8041, + "step": 1992 + }, + { + "epoch": 2.1841095890410958, + "grad_norm": 2.6072700023651123, + "learning_rate": 1.5414297319252642e-06, + "loss": 1.7568, + "step": 1993 + }, + { + "epoch": 2.185205479452055, + "grad_norm": 2.525784730911255, + "learning_rate": 1.5393988627132415e-06, + "loss": 1.748, + "step": 1994 + }, + { + "epoch": 2.186301369863014, + "grad_norm": 2.564699649810791, + "learning_rate": 1.5373679935012188e-06, + "loss": 1.7148, + "step": 1995 + }, + { + "epoch": 2.1873972602739724, + "grad_norm": 2.7365031242370605, + "learning_rate": 1.535337124289196e-06, + "loss": 1.6714, + "step": 1996 + }, + { + "epoch": 2.1884931506849314, + "grad_norm": 2.719954252243042, + "learning_rate": 1.5333062550771733e-06, + "loss": 1.7301, + "step": 1997 + }, + { + "epoch": 2.1895890410958905, + "grad_norm": 2.7979655265808105, + "learning_rate": 1.5312753858651506e-06, + "loss": 1.8163, + "step": 1998 + }, + { + "epoch": 2.1906849315068495, + "grad_norm": 2.6182446479797363, + "learning_rate": 1.5292445166531279e-06, + "loss": 1.713, + "step": 1999 + }, + { + "epoch": 2.191780821917808, + "grad_norm": 2.7218992710113525, + "learning_rate": 1.5272136474411047e-06, + "loss": 1.7355, + "step": 2000 + }, + { + "epoch": 2.191780821917808, + "eval_loss": 1.950900673866272, + "eval_runtime": 37.2927, + "eval_samples_per_second": 26.815, + "eval_steps_per_second": 1.126, + "step": 2000 + }, + { + "epoch": 2.192876712328767, + "grad_norm": 2.7384393215179443, + "learning_rate": 1.525182778229082e-06, + "loss": 1.7777, + "step": 2001 + }, + { + "epoch": 2.193972602739726, + "grad_norm": 2.7223575115203857, + "learning_rate": 1.5231519090170593e-06, + "loss": 1.7136, + "step": 2002 + }, + { + "epoch": 2.1950684931506848, + "grad_norm": 2.5952577590942383, + "learning_rate": 1.5211210398050365e-06, + "loss": 1.6964, + "step": 2003 + }, + { + "epoch": 2.196164383561644, + "grad_norm": 2.626406669616699, + "learning_rate": 1.5190901705930138e-06, + "loss": 1.6996, + "step": 2004 + }, + { + "epoch": 2.197260273972603, + "grad_norm": 2.6563360691070557, + "learning_rate": 1.517059301380991e-06, + "loss": 1.7428, + "step": 2005 + }, + { + "epoch": 2.198356164383562, + "grad_norm": 2.647664785385132, + "learning_rate": 1.5150284321689684e-06, + "loss": 1.6953, + "step": 2006 + }, + { + "epoch": 2.1994520547945204, + "grad_norm": 2.5670430660247803, + "learning_rate": 1.5129975629569456e-06, + "loss": 1.7215, + "step": 2007 + }, + { + "epoch": 2.2005479452054795, + "grad_norm": 2.6148288249969482, + "learning_rate": 1.510966693744923e-06, + "loss": 1.724, + "step": 2008 + }, + { + "epoch": 2.2016438356164385, + "grad_norm": 2.8430588245391846, + "learning_rate": 1.5089358245329002e-06, + "loss": 1.7255, + "step": 2009 + }, + { + "epoch": 2.202739726027397, + "grad_norm": 2.5937509536743164, + "learning_rate": 1.5069049553208775e-06, + "loss": 1.7055, + "step": 2010 + }, + { + "epoch": 2.202739726027397, + "eval_loss": 1.9469048976898193, + "eval_runtime": 37.2173, + "eval_samples_per_second": 26.869, + "eval_steps_per_second": 1.129, + "step": 2010 + }, + { + "epoch": 2.203835616438356, + "grad_norm": 2.7167937755584717, + "learning_rate": 1.5048740861088547e-06, + "loss": 1.7474, + "step": 2011 + }, + { + "epoch": 2.204931506849315, + "grad_norm": 2.53837251663208, + "learning_rate": 1.502843216896832e-06, + "loss": 1.74, + "step": 2012 + }, + { + "epoch": 2.206027397260274, + "grad_norm": 2.604931592941284, + "learning_rate": 1.500812347684809e-06, + "loss": 1.6573, + "step": 2013 + }, + { + "epoch": 2.207123287671233, + "grad_norm": 2.6670846939086914, + "learning_rate": 1.4987814784727864e-06, + "loss": 1.7208, + "step": 2014 + }, + { + "epoch": 2.208219178082192, + "grad_norm": 2.677746534347534, + "learning_rate": 1.4967506092607636e-06, + "loss": 1.728, + "step": 2015 + }, + { + "epoch": 2.209315068493151, + "grad_norm": 2.6490612030029297, + "learning_rate": 1.494719740048741e-06, + "loss": 1.6728, + "step": 2016 + }, + { + "epoch": 2.2104109589041094, + "grad_norm": 2.6571807861328125, + "learning_rate": 1.4926888708367182e-06, + "loss": 1.7528, + "step": 2017 + }, + { + "epoch": 2.2115068493150685, + "grad_norm": 2.5933258533477783, + "learning_rate": 1.4906580016246955e-06, + "loss": 1.757, + "step": 2018 + }, + { + "epoch": 2.2126027397260275, + "grad_norm": 2.6451070308685303, + "learning_rate": 1.4886271324126727e-06, + "loss": 1.763, + "step": 2019 + }, + { + "epoch": 2.213698630136986, + "grad_norm": 2.5259275436401367, + "learning_rate": 1.48659626320065e-06, + "loss": 1.7979, + "step": 2020 + }, + { + "epoch": 2.213698630136986, + "eval_loss": 1.9480665922164917, + "eval_runtime": 37.5325, + "eval_samples_per_second": 26.644, + "eval_steps_per_second": 1.119, + "step": 2020 + }, + { + "epoch": 2.214794520547945, + "grad_norm": 2.6959381103515625, + "learning_rate": 1.4845653939886273e-06, + "loss": 1.6376, + "step": 2021 + }, + { + "epoch": 2.215890410958904, + "grad_norm": 2.6179497241973877, + "learning_rate": 1.4825345247766046e-06, + "loss": 1.6575, + "step": 2022 + }, + { + "epoch": 2.216986301369863, + "grad_norm": 2.5618393421173096, + "learning_rate": 1.4805036555645818e-06, + "loss": 1.6486, + "step": 2023 + }, + { + "epoch": 2.2180821917808218, + "grad_norm": 2.545051336288452, + "learning_rate": 1.4784727863525591e-06, + "loss": 1.7218, + "step": 2024 + }, + { + "epoch": 2.219178082191781, + "grad_norm": 2.6248512268066406, + "learning_rate": 1.4764419171405362e-06, + "loss": 1.691, + "step": 2025 + }, + { + "epoch": 2.22027397260274, + "grad_norm": 2.65645170211792, + "learning_rate": 1.4744110479285135e-06, + "loss": 1.7602, + "step": 2026 + }, + { + "epoch": 2.2213698630136984, + "grad_norm": 2.6771113872528076, + "learning_rate": 1.4723801787164907e-06, + "loss": 1.7573, + "step": 2027 + }, + { + "epoch": 2.2224657534246575, + "grad_norm": 2.5466039180755615, + "learning_rate": 1.470349309504468e-06, + "loss": 1.7604, + "step": 2028 + }, + { + "epoch": 2.2235616438356165, + "grad_norm": 2.685079574584961, + "learning_rate": 1.4683184402924453e-06, + "loss": 1.7341, + "step": 2029 + }, + { + "epoch": 2.2246575342465755, + "grad_norm": 2.6854352951049805, + "learning_rate": 1.4662875710804226e-06, + "loss": 1.7427, + "step": 2030 + }, + { + "epoch": 2.2246575342465755, + "eval_loss": 1.9459377527236938, + "eval_runtime": 37.5255, + "eval_samples_per_second": 26.649, + "eval_steps_per_second": 1.119, + "step": 2030 + }, + { + "epoch": 2.225753424657534, + "grad_norm": 2.667675733566284, + "learning_rate": 1.4642567018683998e-06, + "loss": 1.8328, + "step": 2031 + }, + { + "epoch": 2.226849315068493, + "grad_norm": 2.603386640548706, + "learning_rate": 1.4622258326563771e-06, + "loss": 1.7193, + "step": 2032 + }, + { + "epoch": 2.227945205479452, + "grad_norm": 2.4960451126098633, + "learning_rate": 1.4601949634443544e-06, + "loss": 1.695, + "step": 2033 + }, + { + "epoch": 2.2290410958904108, + "grad_norm": 2.755707025527954, + "learning_rate": 1.4581640942323317e-06, + "loss": 1.7475, + "step": 2034 + }, + { + "epoch": 2.23013698630137, + "grad_norm": 2.5849006175994873, + "learning_rate": 1.456133225020309e-06, + "loss": 1.8106, + "step": 2035 + }, + { + "epoch": 2.231232876712329, + "grad_norm": 2.690680980682373, + "learning_rate": 1.4541023558082862e-06, + "loss": 1.7368, + "step": 2036 + }, + { + "epoch": 2.232328767123288, + "grad_norm": 2.6760425567626953, + "learning_rate": 1.4520714865962635e-06, + "loss": 1.722, + "step": 2037 + }, + { + "epoch": 2.2334246575342465, + "grad_norm": 2.656379461288452, + "learning_rate": 1.4500406173842406e-06, + "loss": 1.7623, + "step": 2038 + }, + { + "epoch": 2.2345205479452055, + "grad_norm": 2.8213517665863037, + "learning_rate": 1.4480097481722178e-06, + "loss": 1.7344, + "step": 2039 + }, + { + "epoch": 2.2356164383561645, + "grad_norm": 2.877163887023926, + "learning_rate": 1.4459788789601951e-06, + "loss": 1.7184, + "step": 2040 + }, + { + "epoch": 2.2356164383561645, + "eval_loss": 1.9430633783340454, + "eval_runtime": 37.3102, + "eval_samples_per_second": 26.802, + "eval_steps_per_second": 1.126, + "step": 2040 + }, + { + "epoch": 2.236712328767123, + "grad_norm": 2.7285237312316895, + "learning_rate": 1.4439480097481724e-06, + "loss": 1.6826, + "step": 2041 + }, + { + "epoch": 2.237808219178082, + "grad_norm": 2.5208237171173096, + "learning_rate": 1.4419171405361497e-06, + "loss": 1.6788, + "step": 2042 + }, + { + "epoch": 2.238904109589041, + "grad_norm": 2.7415900230407715, + "learning_rate": 1.439886271324127e-06, + "loss": 1.7423, + "step": 2043 + }, + { + "epoch": 2.24, + "grad_norm": 2.760897397994995, + "learning_rate": 1.4378554021121042e-06, + "loss": 1.7341, + "step": 2044 + }, + { + "epoch": 2.241095890410959, + "grad_norm": 2.5650556087493896, + "learning_rate": 1.4358245329000815e-06, + "loss": 1.7036, + "step": 2045 + }, + { + "epoch": 2.242191780821918, + "grad_norm": 2.60780930519104, + "learning_rate": 1.4337936636880588e-06, + "loss": 1.7719, + "step": 2046 + }, + { + "epoch": 2.243287671232877, + "grad_norm": 2.592473268508911, + "learning_rate": 1.431762794476036e-06, + "loss": 1.7421, + "step": 2047 + }, + { + "epoch": 2.2443835616438355, + "grad_norm": 2.6672041416168213, + "learning_rate": 1.4297319252640133e-06, + "loss": 1.6865, + "step": 2048 + }, + { + "epoch": 2.2454794520547945, + "grad_norm": 2.6182847023010254, + "learning_rate": 1.4277010560519906e-06, + "loss": 1.731, + "step": 2049 + }, + { + "epoch": 2.2465753424657535, + "grad_norm": 2.8213913440704346, + "learning_rate": 1.4256701868399674e-06, + "loss": 1.6845, + "step": 2050 + }, + { + "epoch": 2.2465753424657535, + "eval_loss": 1.9434115886688232, + "eval_runtime": 37.3048, + "eval_samples_per_second": 26.806, + "eval_steps_per_second": 1.126, + "step": 2050 + }, + { + "epoch": 2.247671232876712, + "grad_norm": 2.7278084754943848, + "learning_rate": 1.4236393176279447e-06, + "loss": 1.8077, + "step": 2051 + }, + { + "epoch": 2.248767123287671, + "grad_norm": 2.5665392875671387, + "learning_rate": 1.421608448415922e-06, + "loss": 1.7235, + "step": 2052 + }, + { + "epoch": 2.24986301369863, + "grad_norm": 2.680154323577881, + "learning_rate": 1.4195775792038993e-06, + "loss": 1.8164, + "step": 2053 + }, + { + "epoch": 2.250958904109589, + "grad_norm": 2.770051956176758, + "learning_rate": 1.4175467099918765e-06, + "loss": 1.6948, + "step": 2054 + }, + { + "epoch": 2.252054794520548, + "grad_norm": 2.6189041137695312, + "learning_rate": 1.4155158407798538e-06, + "loss": 1.7369, + "step": 2055 + }, + { + "epoch": 2.253150684931507, + "grad_norm": 2.6655054092407227, + "learning_rate": 1.413484971567831e-06, + "loss": 1.7456, + "step": 2056 + }, + { + "epoch": 2.254246575342466, + "grad_norm": 2.6520164012908936, + "learning_rate": 1.4114541023558084e-06, + "loss": 1.6719, + "step": 2057 + }, + { + "epoch": 2.255342465753425, + "grad_norm": 2.594533681869507, + "learning_rate": 1.4094232331437856e-06, + "loss": 1.7078, + "step": 2058 + }, + { + "epoch": 2.2564383561643835, + "grad_norm": 2.592308759689331, + "learning_rate": 1.407392363931763e-06, + "loss": 1.7517, + "step": 2059 + }, + { + "epoch": 2.2575342465753425, + "grad_norm": 2.5815367698669434, + "learning_rate": 1.4053614947197402e-06, + "loss": 1.8262, + "step": 2060 + }, + { + "epoch": 2.2575342465753425, + "eval_loss": 1.945846438407898, + "eval_runtime": 37.186, + "eval_samples_per_second": 26.892, + "eval_steps_per_second": 1.129, + "step": 2060 + }, + { + "epoch": 2.2586301369863016, + "grad_norm": 2.643691301345825, + "learning_rate": 1.4033306255077175e-06, + "loss": 1.7559, + "step": 2061 + }, + { + "epoch": 2.25972602739726, + "grad_norm": 2.6091558933258057, + "learning_rate": 1.4012997562956945e-06, + "loss": 1.6433, + "step": 2062 + }, + { + "epoch": 2.260821917808219, + "grad_norm": 2.7172136306762695, + "learning_rate": 1.3992688870836718e-06, + "loss": 1.7735, + "step": 2063 + }, + { + "epoch": 2.261917808219178, + "grad_norm": 2.5673139095306396, + "learning_rate": 1.397238017871649e-06, + "loss": 1.7402, + "step": 2064 + }, + { + "epoch": 2.263013698630137, + "grad_norm": 2.711242198944092, + "learning_rate": 1.3952071486596264e-06, + "loss": 1.7183, + "step": 2065 + }, + { + "epoch": 2.264109589041096, + "grad_norm": 2.6999948024749756, + "learning_rate": 1.3931762794476036e-06, + "loss": 1.6856, + "step": 2066 + }, + { + "epoch": 2.265205479452055, + "grad_norm": 2.670964002609253, + "learning_rate": 1.391145410235581e-06, + "loss": 1.78, + "step": 2067 + }, + { + "epoch": 2.266301369863014, + "grad_norm": 2.4599525928497314, + "learning_rate": 1.3891145410235582e-06, + "loss": 1.7143, + "step": 2068 + }, + { + "epoch": 2.2673972602739725, + "grad_norm": 2.552266836166382, + "learning_rate": 1.3870836718115355e-06, + "loss": 1.6503, + "step": 2069 + }, + { + "epoch": 2.2684931506849315, + "grad_norm": 2.5764405727386475, + "learning_rate": 1.3850528025995127e-06, + "loss": 1.7046, + "step": 2070 + }, + { + "epoch": 2.2684931506849315, + "eval_loss": 1.9457523822784424, + "eval_runtime": 37.217, + "eval_samples_per_second": 26.869, + "eval_steps_per_second": 1.129, + "step": 2070 + }, + { + "epoch": 2.2695890410958905, + "grad_norm": 2.524228572845459, + "learning_rate": 1.38302193338749e-06, + "loss": 1.7078, + "step": 2071 + }, + { + "epoch": 2.270684931506849, + "grad_norm": 2.5531418323516846, + "learning_rate": 1.3809910641754673e-06, + "loss": 1.646, + "step": 2072 + }, + { + "epoch": 2.271780821917808, + "grad_norm": 2.705824613571167, + "learning_rate": 1.3789601949634446e-06, + "loss": 1.7407, + "step": 2073 + }, + { + "epoch": 2.272876712328767, + "grad_norm": 2.5266356468200684, + "learning_rate": 1.3769293257514218e-06, + "loss": 1.6472, + "step": 2074 + }, + { + "epoch": 2.2739726027397262, + "grad_norm": 2.5785422325134277, + "learning_rate": 1.374898456539399e-06, + "loss": 1.7561, + "step": 2075 + }, + { + "epoch": 2.275068493150685, + "grad_norm": 2.663120985031128, + "learning_rate": 1.3728675873273762e-06, + "loss": 1.6901, + "step": 2076 + }, + { + "epoch": 2.276164383561644, + "grad_norm": 2.6085517406463623, + "learning_rate": 1.3708367181153535e-06, + "loss": 1.7829, + "step": 2077 + }, + { + "epoch": 2.277260273972603, + "grad_norm": 2.6439929008483887, + "learning_rate": 1.3688058489033307e-06, + "loss": 1.7684, + "step": 2078 + }, + { + "epoch": 2.2783561643835615, + "grad_norm": 2.6614694595336914, + "learning_rate": 1.366774979691308e-06, + "loss": 1.726, + "step": 2079 + }, + { + "epoch": 2.2794520547945205, + "grad_norm": 2.7339539527893066, + "learning_rate": 1.3647441104792853e-06, + "loss": 1.7577, + "step": 2080 + }, + { + "epoch": 2.2794520547945205, + "eval_loss": 1.9456545114517212, + "eval_runtime": 37.174, + "eval_samples_per_second": 26.901, + "eval_steps_per_second": 1.13, + "step": 2080 + }, + { + "epoch": 2.2805479452054795, + "grad_norm": 2.603545665740967, + "learning_rate": 1.3627132412672626e-06, + "loss": 1.7297, + "step": 2081 + }, + { + "epoch": 2.281643835616438, + "grad_norm": 2.548466682434082, + "learning_rate": 1.3606823720552398e-06, + "loss": 1.6701, + "step": 2082 + }, + { + "epoch": 2.282739726027397, + "grad_norm": 2.6043875217437744, + "learning_rate": 1.3586515028432171e-06, + "loss": 1.7369, + "step": 2083 + }, + { + "epoch": 2.283835616438356, + "grad_norm": 2.7412586212158203, + "learning_rate": 1.3566206336311944e-06, + "loss": 1.654, + "step": 2084 + }, + { + "epoch": 2.2849315068493152, + "grad_norm": 2.669410228729248, + "learning_rate": 1.3545897644191717e-06, + "loss": 1.7186, + "step": 2085 + }, + { + "epoch": 2.286027397260274, + "grad_norm": 2.7052674293518066, + "learning_rate": 1.352558895207149e-06, + "loss": 1.7005, + "step": 2086 + }, + { + "epoch": 2.287123287671233, + "grad_norm": 2.5840418338775635, + "learning_rate": 1.3505280259951258e-06, + "loss": 1.7459, + "step": 2087 + }, + { + "epoch": 2.288219178082192, + "grad_norm": 2.6355416774749756, + "learning_rate": 1.348497156783103e-06, + "loss": 1.7243, + "step": 2088 + }, + { + "epoch": 2.289315068493151, + "grad_norm": 2.6697628498077393, + "learning_rate": 1.3464662875710804e-06, + "loss": 1.7597, + "step": 2089 + }, + { + "epoch": 2.2904109589041095, + "grad_norm": 2.679842233657837, + "learning_rate": 1.3444354183590576e-06, + "loss": 1.6557, + "step": 2090 + }, + { + "epoch": 2.2904109589041095, + "eval_loss": 1.9469218254089355, + "eval_runtime": 37.1615, + "eval_samples_per_second": 26.91, + "eval_steps_per_second": 1.13, + "step": 2090 + }, + { + "epoch": 2.2915068493150685, + "grad_norm": 2.8354430198669434, + "learning_rate": 1.342404549147035e-06, + "loss": 1.7496, + "step": 2091 + }, + { + "epoch": 2.2926027397260276, + "grad_norm": 2.6957387924194336, + "learning_rate": 1.3403736799350122e-06, + "loss": 1.6872, + "step": 2092 + }, + { + "epoch": 2.293698630136986, + "grad_norm": 2.5815868377685547, + "learning_rate": 1.3383428107229895e-06, + "loss": 1.7497, + "step": 2093 + }, + { + "epoch": 2.294794520547945, + "grad_norm": 2.727085828781128, + "learning_rate": 1.3363119415109667e-06, + "loss": 1.7723, + "step": 2094 + }, + { + "epoch": 2.2958904109589042, + "grad_norm": 2.6119000911712646, + "learning_rate": 1.334281072298944e-06, + "loss": 1.7414, + "step": 2095 + }, + { + "epoch": 2.296986301369863, + "grad_norm": 2.6403088569641113, + "learning_rate": 1.3322502030869213e-06, + "loss": 1.7398, + "step": 2096 + }, + { + "epoch": 2.298082191780822, + "grad_norm": 2.5066816806793213, + "learning_rate": 1.3302193338748986e-06, + "loss": 1.6974, + "step": 2097 + }, + { + "epoch": 2.299178082191781, + "grad_norm": 2.6601762771606445, + "learning_rate": 1.3281884646628758e-06, + "loss": 1.7578, + "step": 2098 + }, + { + "epoch": 2.30027397260274, + "grad_norm": 2.571499824523926, + "learning_rate": 1.3261575954508531e-06, + "loss": 1.7573, + "step": 2099 + }, + { + "epoch": 2.3013698630136985, + "grad_norm": 2.533543825149536, + "learning_rate": 1.3241267262388302e-06, + "loss": 1.7566, + "step": 2100 + }, + { + "epoch": 2.3013698630136985, + "eval_loss": 1.9441688060760498, + "eval_runtime": 37.2158, + "eval_samples_per_second": 26.87, + "eval_steps_per_second": 1.129, + "step": 2100 + } + ], + "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": 6.546349759660032e+17, + "train_batch_size": 3, + "trial_name": null, + "trial_params": null +} diff --git a/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-2100/zero_to_fp32.py b/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-2100/zero_to_fp32.py new file mode 100644 index 0000000000000000000000000000000000000000..24cc342e78d1a006c782b3a4cd68d9ce786d8fd8 --- /dev/null +++ b/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-2100/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/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-2200/config.json b/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-2200/config.json new file mode 100644 index 0000000000000000000000000000000000000000..fb0f9a53a507ffcdb60410deb71d60ef801bf350 --- /dev/null +++ b/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_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/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-2200/generation_config.json b/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-2200/generation_config.json new file mode 100644 index 0000000000000000000000000000000000000000..9c389b841a9a29ddd904e02d1eb0e08dcce82ad9 --- /dev/null +++ b/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_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/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-2200/latest b/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-2200/latest new file mode 100644 index 0000000000000000000000000000000000000000..adbdf3f70e1e0db88a47bff3edcb0592ed0a2e61 --- /dev/null +++ b/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-2200/latest @@ -0,0 +1 @@ +global_step2200 \ No newline at end of file diff --git a/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-2200/model.safetensors.index.json b/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-2200/model.safetensors.index.json new file mode 100644 index 0000000000000000000000000000000000000000..ed64de846d720b9a7859dc20575fea8e8ca51940 --- /dev/null +++ b/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_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/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-2200/special_tokens_map.json b/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-2200/special_tokens_map.json new file mode 100644 index 0000000000000000000000000000000000000000..6949c5975ee0e961ef61cf31010dce04df0a03f8 --- /dev/null +++ b/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_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/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-2200/tokenizer_config.json b/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-2200/tokenizer_config.json new file mode 100644 index 0000000000000000000000000000000000000000..5684b14e2756b56aa397c1324e239b442162b34a --- /dev/null +++ b/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_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/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-2200/trainer_state.json b/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-2200/trainer_state.json new file mode 100644 index 0000000000000000000000000000000000000000..2057e473243af805a62175348303e9408d58a015 --- /dev/null +++ b/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_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": 3.0476, + "step": 1 + }, + { + "epoch": 0.002191780821917808, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.0289, + "step": 2 + }, + { + "epoch": 0.0032876712328767125, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 2.9935, + "step": 3 + }, + { + "epoch": 0.004383561643835616, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.0139, + "step": 4 + }, + { + "epoch": 0.005479452054794521, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 2.9892, + "step": 5 + }, + { + "epoch": 0.006575342465753425, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.1252, + "step": 6 + }, + { + "epoch": 0.007671232876712329, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.0291, + "step": 7 + }, + { + "epoch": 0.008767123287671232, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.0607, + "step": 8 + }, + { + "epoch": 0.009863013698630137, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.0306, + "step": 9 + }, + { + "epoch": 0.010958904109589041, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.0811, + "step": 10 + }, + { + "epoch": 0.010958904109589041, + "eval_loss": 3.0768930912017822, + "eval_runtime": 38.0377, + "eval_samples_per_second": 26.29, + "eval_steps_per_second": 1.104, + "step": 10 + }, + { + "epoch": 0.012054794520547946, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.0693, + "step": 11 + }, + { + "epoch": 0.01315068493150685, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 2.9668, + "step": 12 + }, + { + "epoch": 0.014246575342465753, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.052, + "step": 13 + }, + { + "epoch": 0.015342465753424657, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.0363, + "step": 14 + }, + { + "epoch": 0.01643835616438356, + "grad_norm": 5.58175802230835, + "learning_rate": 1.8248175182481753e-08, + "loss": 3.0778, + "step": 15 + }, + { + "epoch": 0.017534246575342465, + "grad_norm": 6.360531806945801, + "learning_rate": 3.6496350364963505e-08, + "loss": 3.0257, + "step": 16 + }, + { + "epoch": 0.01863013698630137, + "grad_norm": 6.453113555908203, + "learning_rate": 5.4744525547445264e-08, + "loss": 3.0262, + "step": 17 + }, + { + "epoch": 0.019726027397260273, + "grad_norm": 6.789405345916748, + "learning_rate": 7.299270072992701e-08, + "loss": 3.0671, + "step": 18 + }, + { + "epoch": 0.020821917808219178, + "grad_norm": 6.185514450073242, + "learning_rate": 9.124087591240876e-08, + "loss": 3.0077, + "step": 19 + }, + { + "epoch": 0.021917808219178082, + "grad_norm": 5.719798564910889, + "learning_rate": 1.0948905109489053e-07, + "loss": 3.089, + "step": 20 + }, + { + "epoch": 0.021917808219178082, + "eval_loss": 3.07546329498291, + "eval_runtime": 37.5318, + "eval_samples_per_second": 26.644, + "eval_steps_per_second": 1.119, + "step": 20 + }, + { + "epoch": 0.023013698630136987, + "grad_norm": 6.307768821716309, + "learning_rate": 1.2773722627737227e-07, + "loss": 3.0641, + "step": 21 + }, + { + "epoch": 0.02410958904109589, + "grad_norm": 6.255112171173096, + "learning_rate": 1.4598540145985402e-07, + "loss": 2.9849, + "step": 22 + }, + { + "epoch": 0.025205479452054796, + "grad_norm": 5.812101364135742, + "learning_rate": 1.642335766423358e-07, + "loss": 3.0211, + "step": 23 + }, + { + "epoch": 0.0263013698630137, + "grad_norm": 6.262075901031494, + "learning_rate": 1.8248175182481753e-07, + "loss": 3.0654, + "step": 24 + }, + { + "epoch": 0.0273972602739726, + "grad_norm": 5.419229507446289, + "learning_rate": 2.0072992700729928e-07, + "loss": 3.0661, + "step": 25 + }, + { + "epoch": 0.028493150684931506, + "grad_norm": 4.750737190246582, + "learning_rate": 2.1897810218978106e-07, + "loss": 3.0494, + "step": 26 + }, + { + "epoch": 0.02958904109589041, + "grad_norm": 4.732361316680908, + "learning_rate": 2.3722627737226278e-07, + "loss": 3.0471, + "step": 27 + }, + { + "epoch": 0.030684931506849315, + "grad_norm": 4.749786853790283, + "learning_rate": 2.5547445255474454e-07, + "loss": 3.0107, + "step": 28 + }, + { + "epoch": 0.03178082191780822, + "grad_norm": 4.451244831085205, + "learning_rate": 2.737226277372263e-07, + "loss": 2.9987, + "step": 29 + }, + { + "epoch": 0.03287671232876712, + "grad_norm": 4.5158257484436035, + "learning_rate": 2.9197080291970804e-07, + "loss": 2.9686, + "step": 30 + }, + { + "epoch": 0.03287671232876712, + "eval_loss": 3.031872034072876, + "eval_runtime": 37.2608, + "eval_samples_per_second": 26.838, + "eval_steps_per_second": 1.127, + "step": 30 + }, + { + "epoch": 0.03397260273972603, + "grad_norm": 3.8239498138427734, + "learning_rate": 3.102189781021898e-07, + "loss": 3.0057, + "step": 31 + }, + { + "epoch": 0.03506849315068493, + "grad_norm": 4.037254810333252, + "learning_rate": 3.284671532846716e-07, + "loss": 2.9733, + "step": 32 + }, + { + "epoch": 0.03616438356164384, + "grad_norm": 4.283931732177734, + "learning_rate": 3.467153284671533e-07, + "loss": 2.9515, + "step": 33 + }, + { + "epoch": 0.03726027397260274, + "grad_norm": 4.1784987449646, + "learning_rate": 3.6496350364963505e-07, + "loss": 2.9483, + "step": 34 + }, + { + "epoch": 0.038356164383561646, + "grad_norm": 3.9210753440856934, + "learning_rate": 3.8321167883211686e-07, + "loss": 2.9679, + "step": 35 + }, + { + "epoch": 0.03945205479452055, + "grad_norm": 3.72884202003479, + "learning_rate": 4.0145985401459856e-07, + "loss": 2.9989, + "step": 36 + }, + { + "epoch": 0.040547945205479455, + "grad_norm": 3.2882330417633057, + "learning_rate": 4.197080291970803e-07, + "loss": 2.9384, + "step": 37 + }, + { + "epoch": 0.041643835616438356, + "grad_norm": 3.515293836593628, + "learning_rate": 4.379562043795621e-07, + "loss": 2.8986, + "step": 38 + }, + { + "epoch": 0.042739726027397264, + "grad_norm": 3.568453073501587, + "learning_rate": 4.562043795620438e-07, + "loss": 2.9854, + "step": 39 + }, + { + "epoch": 0.043835616438356165, + "grad_norm": 3.389789342880249, + "learning_rate": 4.7445255474452557e-07, + "loss": 2.8891, + "step": 40 + }, + { + "epoch": 0.043835616438356165, + "eval_loss": 2.964235544204712, + "eval_runtime": 37.3406, + "eval_samples_per_second": 26.781, + "eval_steps_per_second": 1.125, + "step": 40 + }, + { + "epoch": 0.044931506849315066, + "grad_norm": 3.1826283931732178, + "learning_rate": 4.927007299270073e-07, + "loss": 2.9271, + "step": 41 + }, + { + "epoch": 0.046027397260273974, + "grad_norm": 2.941539764404297, + "learning_rate": 5.109489051094891e-07, + "loss": 2.9486, + "step": 42 + }, + { + "epoch": 0.047123287671232875, + "grad_norm": 3.0076773166656494, + "learning_rate": 5.291970802919708e-07, + "loss": 2.9252, + "step": 43 + }, + { + "epoch": 0.04821917808219178, + "grad_norm": 2.79028582572937, + "learning_rate": 5.474452554744526e-07, + "loss": 2.9277, + "step": 44 + }, + { + "epoch": 0.049315068493150684, + "grad_norm": 2.603456735610962, + "learning_rate": 5.656934306569343e-07, + "loss": 2.9144, + "step": 45 + }, + { + "epoch": 0.05041095890410959, + "grad_norm": 2.674888849258423, + "learning_rate": 5.839416058394161e-07, + "loss": 2.8732, + "step": 46 + }, + { + "epoch": 0.05150684931506849, + "grad_norm": 2.972093343734741, + "learning_rate": 6.021897810218979e-07, + "loss": 2.8623, + "step": 47 + }, + { + "epoch": 0.0526027397260274, + "grad_norm": 2.757766008377075, + "learning_rate": 6.204379562043796e-07, + "loss": 2.8941, + "step": 48 + }, + { + "epoch": 0.0536986301369863, + "grad_norm": 2.910634994506836, + "learning_rate": 6.386861313868614e-07, + "loss": 2.8101, + "step": 49 + }, + { + "epoch": 0.0547945205479452, + "grad_norm": 2.5542638301849365, + "learning_rate": 6.569343065693432e-07, + "loss": 2.9262, + "step": 50 + }, + { + "epoch": 0.0547945205479452, + "eval_loss": 2.883579969406128, + "eval_runtime": 37.326, + "eval_samples_per_second": 26.791, + "eval_steps_per_second": 1.125, + "step": 50 + }, + { + "epoch": 0.05589041095890411, + "grad_norm": 2.333974599838257, + "learning_rate": 6.751824817518248e-07, + "loss": 2.8217, + "step": 51 + }, + { + "epoch": 0.05698630136986301, + "grad_norm": 2.181022882461548, + "learning_rate": 6.934306569343066e-07, + "loss": 2.8483, + "step": 52 + }, + { + "epoch": 0.05808219178082192, + "grad_norm": 2.12200927734375, + "learning_rate": 7.116788321167883e-07, + "loss": 2.7788, + "step": 53 + }, + { + "epoch": 0.05917808219178082, + "grad_norm": 2.052626371383667, + "learning_rate": 7.299270072992701e-07, + "loss": 2.7774, + "step": 54 + }, + { + "epoch": 0.06027397260273973, + "grad_norm": 2.11487078666687, + "learning_rate": 7.48175182481752e-07, + "loss": 2.8045, + "step": 55 + }, + { + "epoch": 0.06136986301369863, + "grad_norm": 2.083588123321533, + "learning_rate": 7.664233576642337e-07, + "loss": 2.8119, + "step": 56 + }, + { + "epoch": 0.06246575342465754, + "grad_norm": 2.2079808712005615, + "learning_rate": 7.846715328467154e-07, + "loss": 2.82, + "step": 57 + }, + { + "epoch": 0.06356164383561644, + "grad_norm": 2.100802421569824, + "learning_rate": 8.029197080291971e-07, + "loss": 2.794, + "step": 58 + }, + { + "epoch": 0.06465753424657535, + "grad_norm": 1.8322066068649292, + "learning_rate": 8.211678832116789e-07, + "loss": 2.7444, + "step": 59 + }, + { + "epoch": 0.06575342465753424, + "grad_norm": 2.077005386352539, + "learning_rate": 8.394160583941606e-07, + "loss": 2.7093, + "step": 60 + }, + { + "epoch": 0.06575342465753424, + "eval_loss": 2.8177874088287354, + "eval_runtime": 37.2903, + "eval_samples_per_second": 26.817, + "eval_steps_per_second": 1.126, + "step": 60 + }, + { + "epoch": 0.06684931506849315, + "grad_norm": 2.1688592433929443, + "learning_rate": 8.576642335766425e-07, + "loss": 2.7446, + "step": 61 + }, + { + "epoch": 0.06794520547945206, + "grad_norm": 2.1709237098693848, + "learning_rate": 8.759124087591242e-07, + "loss": 2.736, + "step": 62 + }, + { + "epoch": 0.06904109589041096, + "grad_norm": 2.3578364849090576, + "learning_rate": 8.94160583941606e-07, + "loss": 2.7583, + "step": 63 + }, + { + "epoch": 0.07013698630136986, + "grad_norm": 1.9025483131408691, + "learning_rate": 9.124087591240876e-07, + "loss": 2.7651, + "step": 64 + }, + { + "epoch": 0.07123287671232877, + "grad_norm": 1.9450963735580444, + "learning_rate": 9.306569343065694e-07, + "loss": 2.7186, + "step": 65 + }, + { + "epoch": 0.07232876712328767, + "grad_norm": 1.957723617553711, + "learning_rate": 9.489051094890511e-07, + "loss": 2.7421, + "step": 66 + }, + { + "epoch": 0.07342465753424658, + "grad_norm": 1.9823631048202515, + "learning_rate": 9.67153284671533e-07, + "loss": 2.7258, + "step": 67 + }, + { + "epoch": 0.07452054794520548, + "grad_norm": 1.945495843887329, + "learning_rate": 9.854014598540146e-07, + "loss": 2.7733, + "step": 68 + }, + { + "epoch": 0.07561643835616438, + "grad_norm": 1.9684759378433228, + "learning_rate": 1.0036496350364965e-06, + "loss": 2.7041, + "step": 69 + }, + { + "epoch": 0.07671232876712329, + "grad_norm": 2.087336540222168, + "learning_rate": 1.0218978102189781e-06, + "loss": 2.6662, + "step": 70 + }, + { + "epoch": 0.07671232876712329, + "eval_loss": 2.7529091835021973, + "eval_runtime": 37.3122, + "eval_samples_per_second": 26.801, + "eval_steps_per_second": 1.126, + "step": 70 + }, + { + "epoch": 0.07780821917808219, + "grad_norm": 1.7790117263793945, + "learning_rate": 1.04014598540146e-06, + "loss": 2.7379, + "step": 71 + }, + { + "epoch": 0.0789041095890411, + "grad_norm": 2.2780861854553223, + "learning_rate": 1.0583941605839416e-06, + "loss": 2.6234, + "step": 72 + }, + { + "epoch": 0.08, + "grad_norm": 2.1144700050354004, + "learning_rate": 1.0766423357664233e-06, + "loss": 2.6248, + "step": 73 + }, + { + "epoch": 0.08109589041095891, + "grad_norm": 2.0247490406036377, + "learning_rate": 1.0948905109489052e-06, + "loss": 2.6456, + "step": 74 + }, + { + "epoch": 0.0821917808219178, + "grad_norm": 2.702547311782837, + "learning_rate": 1.113138686131387e-06, + "loss": 2.6759, + "step": 75 + }, + { + "epoch": 0.08328767123287671, + "grad_norm": 1.8208750486373901, + "learning_rate": 1.1313868613138687e-06, + "loss": 2.6845, + "step": 76 + }, + { + "epoch": 0.08438356164383562, + "grad_norm": 2.3829221725463867, + "learning_rate": 1.1496350364963505e-06, + "loss": 2.61, + "step": 77 + }, + { + "epoch": 0.08547945205479453, + "grad_norm": 2.580885887145996, + "learning_rate": 1.1678832116788322e-06, + "loss": 2.5896, + "step": 78 + }, + { + "epoch": 0.08657534246575342, + "grad_norm": 2.2477641105651855, + "learning_rate": 1.1861313868613138e-06, + "loss": 2.596, + "step": 79 + }, + { + "epoch": 0.08767123287671233, + "grad_norm": 2.1895623207092285, + "learning_rate": 1.2043795620437959e-06, + "loss": 2.5906, + "step": 80 + }, + { + "epoch": 0.08767123287671233, + "eval_loss": 2.6815176010131836, + "eval_runtime": 37.3342, + "eval_samples_per_second": 26.785, + "eval_steps_per_second": 1.125, + "step": 80 + }, + { + "epoch": 0.08876712328767124, + "grad_norm": 2.217090606689453, + "learning_rate": 1.2226277372262775e-06, + "loss": 2.6005, + "step": 81 + }, + { + "epoch": 0.08986301369863013, + "grad_norm": 2.4679362773895264, + "learning_rate": 1.2408759124087592e-06, + "loss": 2.6333, + "step": 82 + }, + { + "epoch": 0.09095890410958904, + "grad_norm": 2.2590575218200684, + "learning_rate": 1.259124087591241e-06, + "loss": 2.5814, + "step": 83 + }, + { + "epoch": 0.09205479452054795, + "grad_norm": 1.877977728843689, + "learning_rate": 1.2773722627737229e-06, + "loss": 2.5948, + "step": 84 + }, + { + "epoch": 0.09315068493150686, + "grad_norm": 2.1535534858703613, + "learning_rate": 1.2956204379562045e-06, + "loss": 2.5781, + "step": 85 + }, + { + "epoch": 0.09424657534246575, + "grad_norm": 2.5001144409179688, + "learning_rate": 1.3138686131386864e-06, + "loss": 2.5339, + "step": 86 + }, + { + "epoch": 0.09534246575342466, + "grad_norm": 2.3279049396514893, + "learning_rate": 1.332116788321168e-06, + "loss": 2.545, + "step": 87 + }, + { + "epoch": 0.09643835616438357, + "grad_norm": 2.2230734825134277, + "learning_rate": 1.3503649635036497e-06, + "loss": 2.4745, + "step": 88 + }, + { + "epoch": 0.09753424657534246, + "grad_norm": 2.8762636184692383, + "learning_rate": 1.3686131386861315e-06, + "loss": 2.516, + "step": 89 + }, + { + "epoch": 0.09863013698630137, + "grad_norm": 2.153506278991699, + "learning_rate": 1.3868613138686132e-06, + "loss": 2.4547, + "step": 90 + }, + { + "epoch": 0.09863013698630137, + "eval_loss": 2.603238344192505, + "eval_runtime": 37.4162, + "eval_samples_per_second": 26.726, + "eval_steps_per_second": 1.123, + "step": 90 + }, + { + "epoch": 0.09972602739726028, + "grad_norm": 2.4143614768981934, + "learning_rate": 1.405109489051095e-06, + "loss": 2.547, + "step": 91 + }, + { + "epoch": 0.10082191780821918, + "grad_norm": 2.354632616043091, + "learning_rate": 1.4233576642335767e-06, + "loss": 2.5009, + "step": 92 + }, + { + "epoch": 0.10191780821917808, + "grad_norm": 2.6605563163757324, + "learning_rate": 1.4416058394160583e-06, + "loss": 2.5304, + "step": 93 + }, + { + "epoch": 0.10301369863013699, + "grad_norm": 2.3724122047424316, + "learning_rate": 1.4598540145985402e-06, + "loss": 2.4925, + "step": 94 + }, + { + "epoch": 0.10410958904109589, + "grad_norm": 3.6323022842407227, + "learning_rate": 1.478102189781022e-06, + "loss": 2.4592, + "step": 95 + }, + { + "epoch": 0.1052054794520548, + "grad_norm": 2.3407089710235596, + "learning_rate": 1.496350364963504e-06, + "loss": 2.5223, + "step": 96 + }, + { + "epoch": 0.1063013698630137, + "grad_norm": 2.8238413333892822, + "learning_rate": 1.5145985401459856e-06, + "loss": 2.489, + "step": 97 + }, + { + "epoch": 0.1073972602739726, + "grad_norm": 2.6433658599853516, + "learning_rate": 1.5328467153284674e-06, + "loss": 2.5101, + "step": 98 + }, + { + "epoch": 0.10849315068493151, + "grad_norm": 2.562761068344116, + "learning_rate": 1.551094890510949e-06, + "loss": 2.4642, + "step": 99 + }, + { + "epoch": 0.1095890410958904, + "grad_norm": 2.4501149654388428, + "learning_rate": 1.5693430656934307e-06, + "loss": 2.4102, + "step": 100 + }, + { + "epoch": 0.1095890410958904, + "eval_loss": 2.5452804565429688, + "eval_runtime": 37.4224, + "eval_samples_per_second": 26.722, + "eval_steps_per_second": 1.122, + "step": 100 + }, + { + "epoch": 0.11068493150684931, + "grad_norm": 2.475864887237549, + "learning_rate": 1.5875912408759126e-06, + "loss": 2.4147, + "step": 101 + }, + { + "epoch": 0.11178082191780822, + "grad_norm": 2.558875322341919, + "learning_rate": 1.6058394160583942e-06, + "loss": 2.4694, + "step": 102 + }, + { + "epoch": 0.11287671232876713, + "grad_norm": 2.1085734367370605, + "learning_rate": 1.624087591240876e-06, + "loss": 2.5099, + "step": 103 + }, + { + "epoch": 0.11397260273972602, + "grad_norm": 2.410731792449951, + "learning_rate": 1.6423357664233577e-06, + "loss": 2.432, + "step": 104 + }, + { + "epoch": 0.11506849315068493, + "grad_norm": 2.2800745964050293, + "learning_rate": 1.6605839416058394e-06, + "loss": 2.3669, + "step": 105 + }, + { + "epoch": 0.11616438356164384, + "grad_norm": 2.2234933376312256, + "learning_rate": 1.6788321167883212e-06, + "loss": 2.4145, + "step": 106 + }, + { + "epoch": 0.11726027397260275, + "grad_norm": 2.3367040157318115, + "learning_rate": 1.6970802919708029e-06, + "loss": 2.4272, + "step": 107 + }, + { + "epoch": 0.11835616438356164, + "grad_norm": 2.7002811431884766, + "learning_rate": 1.715328467153285e-06, + "loss": 2.3478, + "step": 108 + }, + { + "epoch": 0.11945205479452055, + "grad_norm": 2.6628332138061523, + "learning_rate": 1.7335766423357666e-06, + "loss": 2.4096, + "step": 109 + }, + { + "epoch": 0.12054794520547946, + "grad_norm": 2.4174230098724365, + "learning_rate": 1.7518248175182485e-06, + "loss": 2.407, + "step": 110 + }, + { + "epoch": 0.12054794520547946, + "eval_loss": 2.491746664047241, + "eval_runtime": 37.2555, + "eval_samples_per_second": 26.842, + "eval_steps_per_second": 1.127, + "step": 110 + }, + { + "epoch": 0.12164383561643835, + "grad_norm": 2.24796986579895, + "learning_rate": 1.77007299270073e-06, + "loss": 2.4369, + "step": 111 + }, + { + "epoch": 0.12273972602739726, + "grad_norm": 2.720939874649048, + "learning_rate": 1.788321167883212e-06, + "loss": 2.4116, + "step": 112 + }, + { + "epoch": 0.12383561643835617, + "grad_norm": 2.6907670497894287, + "learning_rate": 1.8065693430656936e-06, + "loss": 2.4577, + "step": 113 + }, + { + "epoch": 0.12493150684931507, + "grad_norm": 2.559636354446411, + "learning_rate": 1.8248175182481753e-06, + "loss": 2.3999, + "step": 114 + }, + { + "epoch": 0.12602739726027398, + "grad_norm": 2.6730551719665527, + "learning_rate": 1.8430656934306571e-06, + "loss": 2.3428, + "step": 115 + }, + { + "epoch": 0.12712328767123288, + "grad_norm": 3.172179937362671, + "learning_rate": 1.8613138686131388e-06, + "loss": 2.3722, + "step": 116 + }, + { + "epoch": 0.12821917808219177, + "grad_norm": 2.6757171154022217, + "learning_rate": 1.8795620437956206e-06, + "loss": 2.3572, + "step": 117 + }, + { + "epoch": 0.1293150684931507, + "grad_norm": 2.5840766429901123, + "learning_rate": 1.8978102189781023e-06, + "loss": 2.353, + "step": 118 + }, + { + "epoch": 0.1304109589041096, + "grad_norm": 2.6248581409454346, + "learning_rate": 1.916058394160584e-06, + "loss": 2.36, + "step": 119 + }, + { + "epoch": 0.13150684931506848, + "grad_norm": 2.428516387939453, + "learning_rate": 1.934306569343066e-06, + "loss": 2.3821, + "step": 120 + }, + { + "epoch": 0.13150684931506848, + "eval_loss": 2.451565980911255, + "eval_runtime": 37.3669, + "eval_samples_per_second": 26.762, + "eval_steps_per_second": 1.124, + "step": 120 + }, + { + "epoch": 0.1326027397260274, + "grad_norm": 2.9956233501434326, + "learning_rate": 1.952554744525548e-06, + "loss": 2.2778, + "step": 121 + }, + { + "epoch": 0.1336986301369863, + "grad_norm": 2.4120893478393555, + "learning_rate": 1.9708029197080293e-06, + "loss": 2.325, + "step": 122 + }, + { + "epoch": 0.13479452054794522, + "grad_norm": 2.1410024166107178, + "learning_rate": 1.989051094890511e-06, + "loss": 2.3248, + "step": 123 + }, + { + "epoch": 0.1358904109589041, + "grad_norm": 2.923799514770508, + "learning_rate": 2.007299270072993e-06, + "loss": 2.327, + "step": 124 + }, + { + "epoch": 0.136986301369863, + "grad_norm": 3.0770339965820312, + "learning_rate": 2.025547445255475e-06, + "loss": 2.3847, + "step": 125 + }, + { + "epoch": 0.13808219178082193, + "grad_norm": 2.6408181190490723, + "learning_rate": 2.0437956204379563e-06, + "loss": 2.3196, + "step": 126 + }, + { + "epoch": 0.13917808219178082, + "grad_norm": 2.9127118587493896, + "learning_rate": 2.062043795620438e-06, + "loss": 2.3287, + "step": 127 + }, + { + "epoch": 0.14027397260273972, + "grad_norm": 2.9805052280426025, + "learning_rate": 2.08029197080292e-06, + "loss": 2.3243, + "step": 128 + }, + { + "epoch": 0.14136986301369864, + "grad_norm": 3.1342010498046875, + "learning_rate": 2.0985401459854014e-06, + "loss": 2.38, + "step": 129 + }, + { + "epoch": 0.14246575342465753, + "grad_norm": 2.678422451019287, + "learning_rate": 2.1167883211678833e-06, + "loss": 2.2727, + "step": 130 + }, + { + "epoch": 0.14246575342465753, + "eval_loss": 2.4187021255493164, + "eval_runtime": 37.3117, + "eval_samples_per_second": 26.801, + "eval_steps_per_second": 1.126, + "step": 130 + }, + { + "epoch": 0.14356164383561643, + "grad_norm": 2.9281492233276367, + "learning_rate": 2.135036496350365e-06, + "loss": 2.3327, + "step": 131 + }, + { + "epoch": 0.14465753424657535, + "grad_norm": 2.9592933654785156, + "learning_rate": 2.1532846715328466e-06, + "loss": 2.3674, + "step": 132 + }, + { + "epoch": 0.14575342465753424, + "grad_norm": 3.404761552810669, + "learning_rate": 2.171532846715329e-06, + "loss": 2.2679, + "step": 133 + }, + { + "epoch": 0.14684931506849316, + "grad_norm": 3.635571241378784, + "learning_rate": 2.1897810218978103e-06, + "loss": 2.2799, + "step": 134 + }, + { + "epoch": 0.14794520547945206, + "grad_norm": 3.9130170345306396, + "learning_rate": 2.208029197080292e-06, + "loss": 2.3263, + "step": 135 + }, + { + "epoch": 0.14904109589041095, + "grad_norm": 4.300438404083252, + "learning_rate": 2.226277372262774e-06, + "loss": 2.2969, + "step": 136 + }, + { + "epoch": 0.15013698630136987, + "grad_norm": 2.863511085510254, + "learning_rate": 2.244525547445256e-06, + "loss": 2.297, + "step": 137 + }, + { + "epoch": 0.15123287671232877, + "grad_norm": 4.910321235656738, + "learning_rate": 2.2627737226277373e-06, + "loss": 2.3269, + "step": 138 + }, + { + "epoch": 0.15232876712328766, + "grad_norm": 3.3323862552642822, + "learning_rate": 2.281021897810219e-06, + "loss": 2.3195, + "step": 139 + }, + { + "epoch": 0.15342465753424658, + "grad_norm": 3.2672693729400635, + "learning_rate": 2.299270072992701e-06, + "loss": 2.2797, + "step": 140 + }, + { + "epoch": 0.15342465753424658, + "eval_loss": 2.3943886756896973, + "eval_runtime": 37.3137, + "eval_samples_per_second": 26.8, + "eval_steps_per_second": 1.126, + "step": 140 + }, + { + "epoch": 0.15452054794520548, + "grad_norm": 3.9229774475097656, + "learning_rate": 2.3175182481751825e-06, + "loss": 2.2718, + "step": 141 + }, + { + "epoch": 0.15561643835616437, + "grad_norm": 2.6409823894500732, + "learning_rate": 2.3357664233576643e-06, + "loss": 2.2828, + "step": 142 + }, + { + "epoch": 0.1567123287671233, + "grad_norm": 4.6122612953186035, + "learning_rate": 2.354014598540146e-06, + "loss": 2.247, + "step": 143 + }, + { + "epoch": 0.1578082191780822, + "grad_norm": 3.391357898712158, + "learning_rate": 2.3722627737226276e-06, + "loss": 2.2863, + "step": 144 + }, + { + "epoch": 0.1589041095890411, + "grad_norm": 3.3550853729248047, + "learning_rate": 2.39051094890511e-06, + "loss": 2.2789, + "step": 145 + }, + { + "epoch": 0.16, + "grad_norm": 3.7376883029937744, + "learning_rate": 2.4087591240875918e-06, + "loss": 2.2297, + "step": 146 + }, + { + "epoch": 0.1610958904109589, + "grad_norm": 3.005157709121704, + "learning_rate": 2.427007299270073e-06, + "loss": 2.2467, + "step": 147 + }, + { + "epoch": 0.16219178082191782, + "grad_norm": 3.4616103172302246, + "learning_rate": 2.445255474452555e-06, + "loss": 2.2612, + "step": 148 + }, + { + "epoch": 0.1632876712328767, + "grad_norm": 3.3595101833343506, + "learning_rate": 2.463503649635037e-06, + "loss": 2.3021, + "step": 149 + }, + { + "epoch": 0.1643835616438356, + "grad_norm": 4.831241607666016, + "learning_rate": 2.4817518248175183e-06, + "loss": 2.2174, + "step": 150 + }, + { + "epoch": 0.1643835616438356, + "eval_loss": 2.3677749633789062, + "eval_runtime": 37.3092, + "eval_samples_per_second": 26.803, + "eval_steps_per_second": 1.126, + "step": 150 + }, + { + "epoch": 0.16547945205479453, + "grad_norm": 2.778820037841797, + "learning_rate": 2.5e-06, + "loss": 2.1856, + "step": 151 + }, + { + "epoch": 0.16657534246575342, + "grad_norm": 4.566643714904785, + "learning_rate": 2.518248175182482e-06, + "loss": 2.3186, + "step": 152 + }, + { + "epoch": 0.16767123287671232, + "grad_norm": 2.9908089637756348, + "learning_rate": 2.5364963503649635e-06, + "loss": 2.2127, + "step": 153 + }, + { + "epoch": 0.16876712328767124, + "grad_norm": 2.9702792167663574, + "learning_rate": 2.5547445255474458e-06, + "loss": 2.2071, + "step": 154 + }, + { + "epoch": 0.16986301369863013, + "grad_norm": 2.850044012069702, + "learning_rate": 2.5729927007299272e-06, + "loss": 2.2585, + "step": 155 + }, + { + "epoch": 0.17095890410958905, + "grad_norm": 2.926091432571411, + "learning_rate": 2.591240875912409e-06, + "loss": 2.2397, + "step": 156 + }, + { + "epoch": 0.17205479452054795, + "grad_norm": 2.8720169067382812, + "learning_rate": 2.6094890510948905e-06, + "loss": 2.2306, + "step": 157 + }, + { + "epoch": 0.17315068493150684, + "grad_norm": 2.8837177753448486, + "learning_rate": 2.627737226277373e-06, + "loss": 2.2666, + "step": 158 + }, + { + "epoch": 0.17424657534246576, + "grad_norm": 3.119621753692627, + "learning_rate": 2.6459854014598542e-06, + "loss": 2.2722, + "step": 159 + }, + { + "epoch": 0.17534246575342466, + "grad_norm": 2.7085700035095215, + "learning_rate": 2.664233576642336e-06, + "loss": 2.2346, + "step": 160 + }, + { + "epoch": 0.17534246575342466, + "eval_loss": 2.352083921432495, + "eval_runtime": 37.245, + "eval_samples_per_second": 26.849, + "eval_steps_per_second": 1.128, + "step": 160 + }, + { + "epoch": 0.17643835616438355, + "grad_norm": 2.9964346885681152, + "learning_rate": 2.6824817518248175e-06, + "loss": 2.2155, + "step": 161 + }, + { + "epoch": 0.17753424657534247, + "grad_norm": 3.1535208225250244, + "learning_rate": 2.7007299270072994e-06, + "loss": 2.235, + "step": 162 + }, + { + "epoch": 0.17863013698630137, + "grad_norm": 2.7240943908691406, + "learning_rate": 2.718978102189781e-06, + "loss": 2.2095, + "step": 163 + }, + { + "epoch": 0.17972602739726026, + "grad_norm": 3.122654676437378, + "learning_rate": 2.737226277372263e-06, + "loss": 2.2667, + "step": 164 + }, + { + "epoch": 0.18082191780821918, + "grad_norm": 2.9718410968780518, + "learning_rate": 2.755474452554745e-06, + "loss": 2.2499, + "step": 165 + }, + { + "epoch": 0.18191780821917808, + "grad_norm": 3.0081896781921387, + "learning_rate": 2.7737226277372264e-06, + "loss": 2.2089, + "step": 166 + }, + { + "epoch": 0.18301369863013697, + "grad_norm": 2.829509973526001, + "learning_rate": 2.7919708029197082e-06, + "loss": 2.3139, + "step": 167 + }, + { + "epoch": 0.1841095890410959, + "grad_norm": 3.4480397701263428, + "learning_rate": 2.81021897810219e-06, + "loss": 2.2705, + "step": 168 + }, + { + "epoch": 0.1852054794520548, + "grad_norm": 3.1107025146484375, + "learning_rate": 2.828467153284672e-06, + "loss": 2.1895, + "step": 169 + }, + { + "epoch": 0.1863013698630137, + "grad_norm": 2.916790008544922, + "learning_rate": 2.8467153284671534e-06, + "loss": 2.2317, + "step": 170 + }, + { + "epoch": 0.1863013698630137, + "eval_loss": 2.336153030395508, + "eval_runtime": 37.2401, + "eval_samples_per_second": 26.853, + "eval_steps_per_second": 1.128, + "step": 170 + }, + { + "epoch": 0.1873972602739726, + "grad_norm": 3.2582287788391113, + "learning_rate": 2.8649635036496353e-06, + "loss": 2.143, + "step": 171 + }, + { + "epoch": 0.1884931506849315, + "grad_norm": 2.430973768234253, + "learning_rate": 2.8832116788321167e-06, + "loss": 2.1977, + "step": 172 + }, + { + "epoch": 0.18958904109589042, + "grad_norm": 2.8763599395751953, + "learning_rate": 2.901459854014599e-06, + "loss": 2.1551, + "step": 173 + }, + { + "epoch": 0.19068493150684931, + "grad_norm": 2.9242653846740723, + "learning_rate": 2.9197080291970804e-06, + "loss": 2.2569, + "step": 174 + }, + { + "epoch": 0.1917808219178082, + "grad_norm": 2.623551607131958, + "learning_rate": 2.9379562043795623e-06, + "loss": 2.2329, + "step": 175 + }, + { + "epoch": 0.19287671232876713, + "grad_norm": 2.9511337280273438, + "learning_rate": 2.956204379562044e-06, + "loss": 2.2154, + "step": 176 + }, + { + "epoch": 0.19397260273972602, + "grad_norm": 2.415118932723999, + "learning_rate": 2.9744525547445256e-06, + "loss": 2.2637, + "step": 177 + }, + { + "epoch": 0.19506849315068492, + "grad_norm": 2.9664924144744873, + "learning_rate": 2.992700729927008e-06, + "loss": 2.2263, + "step": 178 + }, + { + "epoch": 0.19616438356164384, + "grad_norm": 2.7214572429656982, + "learning_rate": 3.0109489051094893e-06, + "loss": 2.1667, + "step": 179 + }, + { + "epoch": 0.19726027397260273, + "grad_norm": 2.8712823390960693, + "learning_rate": 3.029197080291971e-06, + "loss": 2.1661, + "step": 180 + }, + { + "epoch": 0.19726027397260273, + "eval_loss": 2.3191800117492676, + "eval_runtime": 37.24, + "eval_samples_per_second": 26.853, + "eval_steps_per_second": 1.128, + "step": 180 + }, + { + "epoch": 0.19835616438356166, + "grad_norm": 2.959709405899048, + "learning_rate": 3.0474452554744526e-06, + "loss": 2.2483, + "step": 181 + }, + { + "epoch": 0.19945205479452055, + "grad_norm": 2.567089319229126, + "learning_rate": 3.065693430656935e-06, + "loss": 2.2382, + "step": 182 + }, + { + "epoch": 0.20054794520547944, + "grad_norm": 2.47188401222229, + "learning_rate": 3.0839416058394163e-06, + "loss": 2.2148, + "step": 183 + }, + { + "epoch": 0.20164383561643837, + "grad_norm": 2.3435261249542236, + "learning_rate": 3.102189781021898e-06, + "loss": 2.2334, + "step": 184 + }, + { + "epoch": 0.20273972602739726, + "grad_norm": 2.5280468463897705, + "learning_rate": 3.1204379562043796e-06, + "loss": 2.2214, + "step": 185 + }, + { + "epoch": 0.20383561643835615, + "grad_norm": 2.4808971881866455, + "learning_rate": 3.1386861313868614e-06, + "loss": 2.1554, + "step": 186 + }, + { + "epoch": 0.20493150684931508, + "grad_norm": 2.9937026500701904, + "learning_rate": 3.156934306569343e-06, + "loss": 2.1389, + "step": 187 + }, + { + "epoch": 0.20602739726027397, + "grad_norm": 2.9220476150512695, + "learning_rate": 3.175182481751825e-06, + "loss": 2.2515, + "step": 188 + }, + { + "epoch": 0.20712328767123286, + "grad_norm": 3.0449001789093018, + "learning_rate": 3.193430656934307e-06, + "loss": 2.2016, + "step": 189 + }, + { + "epoch": 0.20821917808219179, + "grad_norm": 2.8027074337005615, + "learning_rate": 3.2116788321167884e-06, + "loss": 2.1823, + "step": 190 + }, + { + "epoch": 0.20821917808219179, + "eval_loss": 2.309677839279175, + "eval_runtime": 37.2442, + "eval_samples_per_second": 26.85, + "eval_steps_per_second": 1.128, + "step": 190 + }, + { + "epoch": 0.20931506849315068, + "grad_norm": 2.9158318042755127, + "learning_rate": 3.2299270072992707e-06, + "loss": 2.1928, + "step": 191 + }, + { + "epoch": 0.2104109589041096, + "grad_norm": 2.5328080654144287, + "learning_rate": 3.248175182481752e-06, + "loss": 2.1828, + "step": 192 + }, + { + "epoch": 0.2115068493150685, + "grad_norm": 3.1709704399108887, + "learning_rate": 3.266423357664234e-06, + "loss": 2.1052, + "step": 193 + }, + { + "epoch": 0.2126027397260274, + "grad_norm": 2.9156367778778076, + "learning_rate": 3.2846715328467155e-06, + "loss": 2.1943, + "step": 194 + }, + { + "epoch": 0.2136986301369863, + "grad_norm": 2.879408836364746, + "learning_rate": 3.3029197080291973e-06, + "loss": 2.1782, + "step": 195 + }, + { + "epoch": 0.2147945205479452, + "grad_norm": 3.748483180999756, + "learning_rate": 3.3211678832116788e-06, + "loss": 2.2107, + "step": 196 + }, + { + "epoch": 0.2158904109589041, + "grad_norm": 3.197510004043579, + "learning_rate": 3.339416058394161e-06, + "loss": 2.1845, + "step": 197 + }, + { + "epoch": 0.21698630136986302, + "grad_norm": 3.2596049308776855, + "learning_rate": 3.3576642335766425e-06, + "loss": 2.1816, + "step": 198 + }, + { + "epoch": 0.21808219178082192, + "grad_norm": 3.3329780101776123, + "learning_rate": 3.3759124087591243e-06, + "loss": 2.181, + "step": 199 + }, + { + "epoch": 0.2191780821917808, + "grad_norm": 2.7302327156066895, + "learning_rate": 3.3941605839416058e-06, + "loss": 2.1761, + "step": 200 + }, + { + "epoch": 0.2191780821917808, + "eval_loss": 2.2909653186798096, + "eval_runtime": 37.284, + "eval_samples_per_second": 26.821, + "eval_steps_per_second": 1.126, + "step": 200 + }, + { + "epoch": 0.22027397260273973, + "grad_norm": 3.002255916595459, + "learning_rate": 3.412408759124088e-06, + "loss": 2.1868, + "step": 201 + }, + { + "epoch": 0.22136986301369863, + "grad_norm": 2.6694891452789307, + "learning_rate": 3.43065693430657e-06, + "loss": 2.192, + "step": 202 + }, + { + "epoch": 0.22246575342465755, + "grad_norm": 2.9225986003875732, + "learning_rate": 3.4489051094890513e-06, + "loss": 2.1303, + "step": 203 + }, + { + "epoch": 0.22356164383561644, + "grad_norm": 2.7844080924987793, + "learning_rate": 3.467153284671533e-06, + "loss": 2.1788, + "step": 204 + }, + { + "epoch": 0.22465753424657534, + "grad_norm": 2.726123332977295, + "learning_rate": 3.4854014598540146e-06, + "loss": 2.1193, + "step": 205 + }, + { + "epoch": 0.22575342465753426, + "grad_norm": 2.838040590286255, + "learning_rate": 3.503649635036497e-06, + "loss": 2.1674, + "step": 206 + }, + { + "epoch": 0.22684931506849315, + "grad_norm": 2.6863937377929688, + "learning_rate": 3.5218978102189783e-06, + "loss": 2.161, + "step": 207 + }, + { + "epoch": 0.22794520547945205, + "grad_norm": 2.5116159915924072, + "learning_rate": 3.54014598540146e-06, + "loss": 2.179, + "step": 208 + }, + { + "epoch": 0.22904109589041097, + "grad_norm": 3.381863594055176, + "learning_rate": 3.5583941605839416e-06, + "loss": 2.2035, + "step": 209 + }, + { + "epoch": 0.23013698630136986, + "grad_norm": 3.2741382122039795, + "learning_rate": 3.576642335766424e-06, + "loss": 2.2518, + "step": 210 + }, + { + "epoch": 0.23013698630136986, + "eval_loss": 2.2773635387420654, + "eval_runtime": 37.2658, + "eval_samples_per_second": 26.834, + "eval_steps_per_second": 1.127, + "step": 210 + }, + { + "epoch": 0.23123287671232876, + "grad_norm": 2.917370319366455, + "learning_rate": 3.5948905109489054e-06, + "loss": 2.2265, + "step": 211 + }, + { + "epoch": 0.23232876712328768, + "grad_norm": 2.8392834663391113, + "learning_rate": 3.6131386861313872e-06, + "loss": 2.0709, + "step": 212 + }, + { + "epoch": 0.23342465753424657, + "grad_norm": 2.753612518310547, + "learning_rate": 3.6313868613138687e-06, + "loss": 2.0752, + "step": 213 + }, + { + "epoch": 0.2345205479452055, + "grad_norm": 3.1477746963500977, + "learning_rate": 3.6496350364963505e-06, + "loss": 2.1169, + "step": 214 + }, + { + "epoch": 0.2356164383561644, + "grad_norm": 2.5448100566864014, + "learning_rate": 3.667883211678833e-06, + "loss": 2.1338, + "step": 215 + }, + { + "epoch": 0.23671232876712328, + "grad_norm": 3.1622560024261475, + "learning_rate": 3.6861313868613142e-06, + "loss": 2.1065, + "step": 216 + }, + { + "epoch": 0.2378082191780822, + "grad_norm": 2.7054336071014404, + "learning_rate": 3.704379562043796e-06, + "loss": 2.1104, + "step": 217 + }, + { + "epoch": 0.2389041095890411, + "grad_norm": 2.964141845703125, + "learning_rate": 3.7226277372262775e-06, + "loss": 2.1949, + "step": 218 + }, + { + "epoch": 0.24, + "grad_norm": 2.7586112022399902, + "learning_rate": 3.7408759124087594e-06, + "loss": 2.165, + "step": 219 + }, + { + "epoch": 0.2410958904109589, + "grad_norm": 2.9035964012145996, + "learning_rate": 3.7591240875912412e-06, + "loss": 2.1192, + "step": 220 + }, + { + "epoch": 0.2410958904109589, + "eval_loss": 2.2649471759796143, + "eval_runtime": 37.4311, + "eval_samples_per_second": 26.716, + "eval_steps_per_second": 1.122, + "step": 220 + }, + { + "epoch": 0.2421917808219178, + "grad_norm": 2.4212076663970947, + "learning_rate": 3.777372262773723e-06, + "loss": 2.1389, + "step": 221 + }, + { + "epoch": 0.2432876712328767, + "grad_norm": 2.973227024078369, + "learning_rate": 3.7956204379562045e-06, + "loss": 2.132, + "step": 222 + }, + { + "epoch": 0.24438356164383562, + "grad_norm": 2.9831430912017822, + "learning_rate": 3.8138686131386864e-06, + "loss": 2.1142, + "step": 223 + }, + { + "epoch": 0.24547945205479452, + "grad_norm": 3.2880353927612305, + "learning_rate": 3.832116788321168e-06, + "loss": 2.168, + "step": 224 + }, + { + "epoch": 0.2465753424657534, + "grad_norm": 3.3920998573303223, + "learning_rate": 3.85036496350365e-06, + "loss": 2.1277, + "step": 225 + }, + { + "epoch": 0.24767123287671233, + "grad_norm": 2.7170026302337646, + "learning_rate": 3.868613138686132e-06, + "loss": 2.0936, + "step": 226 + }, + { + "epoch": 0.24876712328767123, + "grad_norm": 3.6220011711120605, + "learning_rate": 3.886861313868613e-06, + "loss": 2.1259, + "step": 227 + }, + { + "epoch": 0.24986301369863015, + "grad_norm": 3.4229273796081543, + "learning_rate": 3.905109489051096e-06, + "loss": 2.1421, + "step": 228 + }, + { + "epoch": 0.25095890410958904, + "grad_norm": 3.0358338356018066, + "learning_rate": 3.923357664233577e-06, + "loss": 2.1747, + "step": 229 + }, + { + "epoch": 0.25205479452054796, + "grad_norm": 2.611771821975708, + "learning_rate": 3.9416058394160585e-06, + "loss": 2.1282, + "step": 230 + }, + { + "epoch": 0.25205479452054796, + "eval_loss": 2.254887580871582, + "eval_runtime": 37.5943, + "eval_samples_per_second": 26.6, + "eval_steps_per_second": 1.117, + "step": 230 + }, + { + "epoch": 0.25315068493150683, + "grad_norm": 3.846412181854248, + "learning_rate": 3.95985401459854e-06, + "loss": 2.1892, + "step": 231 + }, + { + "epoch": 0.25424657534246575, + "grad_norm": 2.488525390625, + "learning_rate": 3.978102189781022e-06, + "loss": 2.2064, + "step": 232 + }, + { + "epoch": 0.2553424657534247, + "grad_norm": 3.114964485168457, + "learning_rate": 3.996350364963504e-06, + "loss": 2.0653, + "step": 233 + }, + { + "epoch": 0.25643835616438354, + "grad_norm": 2.5559768676757812, + "learning_rate": 4.014598540145986e-06, + "loss": 2.1503, + "step": 234 + }, + { + "epoch": 0.25753424657534246, + "grad_norm": 3.2347540855407715, + "learning_rate": 4.032846715328467e-06, + "loss": 2.0523, + "step": 235 + }, + { + "epoch": 0.2586301369863014, + "grad_norm": 2.8086111545562744, + "learning_rate": 4.05109489051095e-06, + "loss": 2.1206, + "step": 236 + }, + { + "epoch": 0.25972602739726025, + "grad_norm": 2.909231185913086, + "learning_rate": 4.069343065693431e-06, + "loss": 2.139, + "step": 237 + }, + { + "epoch": 0.2608219178082192, + "grad_norm": 3.2171082496643066, + "learning_rate": 4.0875912408759126e-06, + "loss": 2.0908, + "step": 238 + }, + { + "epoch": 0.2619178082191781, + "grad_norm": 2.3851265907287598, + "learning_rate": 4.105839416058395e-06, + "loss": 2.1133, + "step": 239 + }, + { + "epoch": 0.26301369863013696, + "grad_norm": 3.3912737369537354, + "learning_rate": 4.124087591240876e-06, + "loss": 2.1911, + "step": 240 + }, + { + "epoch": 0.26301369863013696, + "eval_loss": 2.2436957359313965, + "eval_runtime": 37.2947, + "eval_samples_per_second": 26.813, + "eval_steps_per_second": 1.126, + "step": 240 + }, + { + "epoch": 0.2641095890410959, + "grad_norm": 2.82906174659729, + "learning_rate": 4.1423357664233586e-06, + "loss": 2.11, + "step": 241 + }, + { + "epoch": 0.2652054794520548, + "grad_norm": 3.0317952632904053, + "learning_rate": 4.16058394160584e-06, + "loss": 2.0953, + "step": 242 + }, + { + "epoch": 0.2663013698630137, + "grad_norm": 3.6438329219818115, + "learning_rate": 4.1788321167883214e-06, + "loss": 2.1746, + "step": 243 + }, + { + "epoch": 0.2673972602739726, + "grad_norm": 2.6690592765808105, + "learning_rate": 4.197080291970803e-06, + "loss": 2.0555, + "step": 244 + }, + { + "epoch": 0.2684931506849315, + "grad_norm": 3.135376453399658, + "learning_rate": 4.215328467153285e-06, + "loss": 2.1394, + "step": 245 + }, + { + "epoch": 0.26958904109589044, + "grad_norm": 2.5474987030029297, + "learning_rate": 4.233576642335767e-06, + "loss": 2.0822, + "step": 246 + }, + { + "epoch": 0.2706849315068493, + "grad_norm": 2.883108377456665, + "learning_rate": 4.251824817518249e-06, + "loss": 2.0869, + "step": 247 + }, + { + "epoch": 0.2717808219178082, + "grad_norm": 2.9801037311553955, + "learning_rate": 4.27007299270073e-06, + "loss": 2.1194, + "step": 248 + }, + { + "epoch": 0.27287671232876715, + "grad_norm": 2.7688217163085938, + "learning_rate": 4.288321167883212e-06, + "loss": 2.0908, + "step": 249 + }, + { + "epoch": 0.273972602739726, + "grad_norm": 3.205148458480835, + "learning_rate": 4.306569343065693e-06, + "loss": 2.0762, + "step": 250 + }, + { + "epoch": 0.273972602739726, + "eval_loss": 2.238699197769165, + "eval_runtime": 37.2428, + "eval_samples_per_second": 26.851, + "eval_steps_per_second": 1.128, + "step": 250 + }, + { + "epoch": 0.27506849315068493, + "grad_norm": 2.8262946605682373, + "learning_rate": 4.3248175182481755e-06, + "loss": 2.0509, + "step": 251 + }, + { + "epoch": 0.27616438356164386, + "grad_norm": 3.156261682510376, + "learning_rate": 4.343065693430658e-06, + "loss": 2.1213, + "step": 252 + }, + { + "epoch": 0.2772602739726027, + "grad_norm": 2.6367998123168945, + "learning_rate": 4.361313868613139e-06, + "loss": 2.1071, + "step": 253 + }, + { + "epoch": 0.27835616438356164, + "grad_norm": 2.871534824371338, + "learning_rate": 4.379562043795621e-06, + "loss": 2.1008, + "step": 254 + }, + { + "epoch": 0.27945205479452057, + "grad_norm": 2.839432716369629, + "learning_rate": 4.397810218978102e-06, + "loss": 2.1293, + "step": 255 + }, + { + "epoch": 0.28054794520547943, + "grad_norm": 2.627143383026123, + "learning_rate": 4.416058394160584e-06, + "loss": 2.1141, + "step": 256 + }, + { + "epoch": 0.28164383561643835, + "grad_norm": 2.8292996883392334, + "learning_rate": 4.434306569343066e-06, + "loss": 2.0981, + "step": 257 + }, + { + "epoch": 0.2827397260273973, + "grad_norm": 2.5711069107055664, + "learning_rate": 4.452554744525548e-06, + "loss": 2.0707, + "step": 258 + }, + { + "epoch": 0.28383561643835614, + "grad_norm": 2.488954544067383, + "learning_rate": 4.4708029197080295e-06, + "loss": 2.1172, + "step": 259 + }, + { + "epoch": 0.28493150684931506, + "grad_norm": 2.442209005355835, + "learning_rate": 4.489051094890512e-06, + "loss": 2.1119, + "step": 260 + }, + { + "epoch": 0.28493150684931506, + "eval_loss": 2.2306911945343018, + "eval_runtime": 37.2412, + "eval_samples_per_second": 26.852, + "eval_steps_per_second": 1.128, + "step": 260 + }, + { + "epoch": 0.286027397260274, + "grad_norm": 2.9147934913635254, + "learning_rate": 4.507299270072993e-06, + "loss": 2.1517, + "step": 261 + }, + { + "epoch": 0.28712328767123285, + "grad_norm": 2.549833297729492, + "learning_rate": 4.525547445255475e-06, + "loss": 2.1047, + "step": 262 + }, + { + "epoch": 0.2882191780821918, + "grad_norm": 3.3405206203460693, + "learning_rate": 4.543795620437956e-06, + "loss": 2.0975, + "step": 263 + }, + { + "epoch": 0.2893150684931507, + "grad_norm": 2.554431676864624, + "learning_rate": 4.562043795620438e-06, + "loss": 2.0622, + "step": 264 + }, + { + "epoch": 0.29041095890410956, + "grad_norm": 3.5529873371124268, + "learning_rate": 4.580291970802921e-06, + "loss": 2.1567, + "step": 265 + }, + { + "epoch": 0.2915068493150685, + "grad_norm": 2.421682119369507, + "learning_rate": 4.598540145985402e-06, + "loss": 2.1619, + "step": 266 + }, + { + "epoch": 0.2926027397260274, + "grad_norm": 2.918288230895996, + "learning_rate": 4.6167883211678835e-06, + "loss": 2.1238, + "step": 267 + }, + { + "epoch": 0.2936986301369863, + "grad_norm": 3.0737621784210205, + "learning_rate": 4.635036496350365e-06, + "loss": 2.0728, + "step": 268 + }, + { + "epoch": 0.2947945205479452, + "grad_norm": 2.5862410068511963, + "learning_rate": 4.653284671532847e-06, + "loss": 2.0675, + "step": 269 + }, + { + "epoch": 0.2958904109589041, + "grad_norm": 2.9004979133605957, + "learning_rate": 4.671532846715329e-06, + "loss": 2.1248, + "step": 270 + }, + { + "epoch": 0.2958904109589041, + "eval_loss": 2.2179365158081055, + "eval_runtime": 37.2352, + "eval_samples_per_second": 26.856, + "eval_steps_per_second": 1.128, + "step": 270 + }, + { + "epoch": 0.29698630136986304, + "grad_norm": 2.4784982204437256, + "learning_rate": 4.689781021897811e-06, + "loss": 2.1276, + "step": 271 + }, + { + "epoch": 0.2980821917808219, + "grad_norm": 2.7971739768981934, + "learning_rate": 4.708029197080292e-06, + "loss": 2.0931, + "step": 272 + }, + { + "epoch": 0.2991780821917808, + "grad_norm": 2.5839481353759766, + "learning_rate": 4.726277372262774e-06, + "loss": 2.1371, + "step": 273 + }, + { + "epoch": 0.30027397260273975, + "grad_norm": 2.7168564796447754, + "learning_rate": 4.744525547445255e-06, + "loss": 2.1086, + "step": 274 + }, + { + "epoch": 0.3013698630136986, + "grad_norm": 2.6566879749298096, + "learning_rate": 4.7627737226277375e-06, + "loss": 2.0368, + "step": 275 + }, + { + "epoch": 0.30246575342465754, + "grad_norm": 2.5068323612213135, + "learning_rate": 4.78102189781022e-06, + "loss": 2.1077, + "step": 276 + }, + { + "epoch": 0.30356164383561646, + "grad_norm": 2.631408214569092, + "learning_rate": 4.799270072992701e-06, + "loss": 2.1013, + "step": 277 + }, + { + "epoch": 0.3046575342465753, + "grad_norm": 2.5642812252044678, + "learning_rate": 4.8175182481751835e-06, + "loss": 2.0739, + "step": 278 + }, + { + "epoch": 0.30575342465753425, + "grad_norm": 2.3326497077941895, + "learning_rate": 4.835766423357665e-06, + "loss": 2.0971, + "step": 279 + }, + { + "epoch": 0.30684931506849317, + "grad_norm": 2.70701265335083, + "learning_rate": 4.854014598540146e-06, + "loss": 2.0904, + "step": 280 + }, + { + "epoch": 0.30684931506849317, + "eval_loss": 2.215972900390625, + "eval_runtime": 37.2196, + "eval_samples_per_second": 26.868, + "eval_steps_per_second": 1.128, + "step": 280 + }, + { + "epoch": 0.30794520547945203, + "grad_norm": 2.615201234817505, + "learning_rate": 4.872262773722628e-06, + "loss": 2.039, + "step": 281 + }, + { + "epoch": 0.30904109589041096, + "grad_norm": 2.2821998596191406, + "learning_rate": 4.89051094890511e-06, + "loss": 2.0529, + "step": 282 + }, + { + "epoch": 0.3101369863013699, + "grad_norm": 2.767153024673462, + "learning_rate": 4.9087591240875915e-06, + "loss": 2.0351, + "step": 283 + }, + { + "epoch": 0.31123287671232874, + "grad_norm": 2.537121295928955, + "learning_rate": 4.927007299270074e-06, + "loss": 2.0752, + "step": 284 + }, + { + "epoch": 0.31232876712328766, + "grad_norm": 2.835052013397217, + "learning_rate": 4.945255474452555e-06, + "loss": 2.0892, + "step": 285 + }, + { + "epoch": 0.3134246575342466, + "grad_norm": 2.5154924392700195, + "learning_rate": 4.963503649635037e-06, + "loss": 2.0569, + "step": 286 + }, + { + "epoch": 0.31452054794520545, + "grad_norm": 2.8212568759918213, + "learning_rate": 4.981751824817518e-06, + "loss": 2.1588, + "step": 287 + }, + { + "epoch": 0.3156164383561644, + "grad_norm": 2.3840994834899902, + "learning_rate": 5e-06, + "loss": 2.1409, + "step": 288 + }, + { + "epoch": 0.3167123287671233, + "grad_norm": 2.9641315937042236, + "learning_rate": 4.997969130787978e-06, + "loss": 2.0892, + "step": 289 + }, + { + "epoch": 0.3178082191780822, + "grad_norm": 2.4516749382019043, + "learning_rate": 4.995938261575955e-06, + "loss": 2.1105, + "step": 290 + }, + { + "epoch": 0.3178082191780822, + "eval_loss": 2.207075834274292, + "eval_runtime": 37.1304, + "eval_samples_per_second": 26.932, + "eval_steps_per_second": 1.131, + "step": 290 + }, + { + "epoch": 0.3189041095890411, + "grad_norm": 2.768869400024414, + "learning_rate": 4.993907392363932e-06, + "loss": 2.0231, + "step": 291 + }, + { + "epoch": 0.32, + "grad_norm": 2.771475315093994, + "learning_rate": 4.9918765231519095e-06, + "loss": 2.1018, + "step": 292 + }, + { + "epoch": 0.32109589041095893, + "grad_norm": 2.773983955383301, + "learning_rate": 4.989845653939887e-06, + "loss": 2.0317, + "step": 293 + }, + { + "epoch": 0.3221917808219178, + "grad_norm": 2.4447343349456787, + "learning_rate": 4.987814784727864e-06, + "loss": 2.0739, + "step": 294 + }, + { + "epoch": 0.3232876712328767, + "grad_norm": 2.444261312484741, + "learning_rate": 4.985783915515841e-06, + "loss": 2.073, + "step": 295 + }, + { + "epoch": 0.32438356164383564, + "grad_norm": 2.871792793273926, + "learning_rate": 4.983753046303819e-06, + "loss": 2.058, + "step": 296 + }, + { + "epoch": 0.3254794520547945, + "grad_norm": 2.3651537895202637, + "learning_rate": 4.981722177091796e-06, + "loss": 2.0749, + "step": 297 + }, + { + "epoch": 0.3265753424657534, + "grad_norm": 2.3395845890045166, + "learning_rate": 4.979691307879773e-06, + "loss": 2.065, + "step": 298 + }, + { + "epoch": 0.32767123287671235, + "grad_norm": 2.6275217533111572, + "learning_rate": 4.9776604386677504e-06, + "loss": 2.0945, + "step": 299 + }, + { + "epoch": 0.3287671232876712, + "grad_norm": 2.866285562515259, + "learning_rate": 4.975629569455728e-06, + "loss": 2.0702, + "step": 300 + }, + { + "epoch": 0.3287671232876712, + "eval_loss": 2.200012683868408, + "eval_runtime": 37.2082, + "eval_samples_per_second": 26.876, + "eval_steps_per_second": 1.129, + "step": 300 + }, + { + "epoch": 0.32986301369863014, + "grad_norm": 2.268401861190796, + "learning_rate": 4.973598700243705e-06, + "loss": 2.1257, + "step": 301 + }, + { + "epoch": 0.33095890410958906, + "grad_norm": 2.6197733879089355, + "learning_rate": 4.971567831031681e-06, + "loss": 2.0347, + "step": 302 + }, + { + "epoch": 0.3320547945205479, + "grad_norm": 2.4738848209381104, + "learning_rate": 4.9695369618196595e-06, + "loss": 2.0368, + "step": 303 + }, + { + "epoch": 0.33315068493150685, + "grad_norm": 2.705670118331909, + "learning_rate": 4.967506092607636e-06, + "loss": 1.9256, + "step": 304 + }, + { + "epoch": 0.33424657534246577, + "grad_norm": 2.813279628753662, + "learning_rate": 4.965475223395614e-06, + "loss": 2.0214, + "step": 305 + }, + { + "epoch": 0.33534246575342463, + "grad_norm": 2.320960283279419, + "learning_rate": 4.9634443541835905e-06, + "loss": 2.0342, + "step": 306 + }, + { + "epoch": 0.33643835616438356, + "grad_norm": 2.4495227336883545, + "learning_rate": 4.961413484971569e-06, + "loss": 2.0421, + "step": 307 + }, + { + "epoch": 0.3375342465753425, + "grad_norm": 2.5384018421173096, + "learning_rate": 4.959382615759545e-06, + "loss": 2.0572, + "step": 308 + }, + { + "epoch": 0.33863013698630134, + "grad_norm": 2.4018006324768066, + "learning_rate": 4.957351746547523e-06, + "loss": 2.0617, + "step": 309 + }, + { + "epoch": 0.33972602739726027, + "grad_norm": 2.552424907684326, + "learning_rate": 4.9553208773355e-06, + "loss": 2.0384, + "step": 310 + }, + { + "epoch": 0.33972602739726027, + "eval_loss": 2.1974849700927734, + "eval_runtime": 37.2308, + "eval_samples_per_second": 26.859, + "eval_steps_per_second": 1.128, + "step": 310 + }, + { + "epoch": 0.3408219178082192, + "grad_norm": 2.4587671756744385, + "learning_rate": 4.953290008123478e-06, + "loss": 2.1701, + "step": 311 + }, + { + "epoch": 0.3419178082191781, + "grad_norm": 2.4486496448516846, + "learning_rate": 4.951259138911454e-06, + "loss": 2.0654, + "step": 312 + }, + { + "epoch": 0.343013698630137, + "grad_norm": 2.123568058013916, + "learning_rate": 4.9492282696994315e-06, + "loss": 1.9961, + "step": 313 + }, + { + "epoch": 0.3441095890410959, + "grad_norm": 2.549788475036621, + "learning_rate": 4.947197400487409e-06, + "loss": 2.1265, + "step": 314 + }, + { + "epoch": 0.3452054794520548, + "grad_norm": 2.3453609943389893, + "learning_rate": 4.945166531275386e-06, + "loss": 2.0501, + "step": 315 + }, + { + "epoch": 0.3463013698630137, + "grad_norm": 2.3584256172180176, + "learning_rate": 4.943135662063363e-06, + "loss": 2.0402, + "step": 316 + }, + { + "epoch": 0.3473972602739726, + "grad_norm": 3.3100943565368652, + "learning_rate": 4.9411047928513406e-06, + "loss": 2.0545, + "step": 317 + }, + { + "epoch": 0.34849315068493153, + "grad_norm": 2.4266037940979004, + "learning_rate": 4.939073923639318e-06, + "loss": 2.0926, + "step": 318 + }, + { + "epoch": 0.3495890410958904, + "grad_norm": 3.101090908050537, + "learning_rate": 4.937043054427295e-06, + "loss": 2.0987, + "step": 319 + }, + { + "epoch": 0.3506849315068493, + "grad_norm": 2.743619203567505, + "learning_rate": 4.935012185215272e-06, + "loss": 2.0315, + "step": 320 + }, + { + "epoch": 0.3506849315068493, + "eval_loss": 2.188877820968628, + "eval_runtime": 37.4609, + "eval_samples_per_second": 26.694, + "eval_steps_per_second": 1.121, + "step": 320 + }, + { + "epoch": 0.35178082191780824, + "grad_norm": 2.6065986156463623, + "learning_rate": 4.93298131600325e-06, + "loss": 2.0166, + "step": 321 + }, + { + "epoch": 0.3528767123287671, + "grad_norm": 2.7554283142089844, + "learning_rate": 4.930950446791227e-06, + "loss": 2.0862, + "step": 322 + }, + { + "epoch": 0.353972602739726, + "grad_norm": 2.4773144721984863, + "learning_rate": 4.928919577579204e-06, + "loss": 2.0472, + "step": 323 + }, + { + "epoch": 0.35506849315068495, + "grad_norm": 2.695289134979248, + "learning_rate": 4.9268887083671815e-06, + "loss": 2.0567, + "step": 324 + }, + { + "epoch": 0.3561643835616438, + "grad_norm": 2.408220052719116, + "learning_rate": 4.924857839155159e-06, + "loss": 2.046, + "step": 325 + }, + { + "epoch": 0.35726027397260274, + "grad_norm": 2.3851559162139893, + "learning_rate": 4.922826969943136e-06, + "loss": 2.054, + "step": 326 + }, + { + "epoch": 0.35835616438356166, + "grad_norm": 2.8667280673980713, + "learning_rate": 4.920796100731113e-06, + "loss": 2.0717, + "step": 327 + }, + { + "epoch": 0.3594520547945205, + "grad_norm": 2.3814942836761475, + "learning_rate": 4.918765231519091e-06, + "loss": 1.9884, + "step": 328 + }, + { + "epoch": 0.36054794520547945, + "grad_norm": 2.341076612472534, + "learning_rate": 4.916734362307068e-06, + "loss": 2.0474, + "step": 329 + }, + { + "epoch": 0.36164383561643837, + "grad_norm": 2.3571853637695312, + "learning_rate": 4.914703493095045e-06, + "loss": 2.0009, + "step": 330 + }, + { + "epoch": 0.36164383561643837, + "eval_loss": 2.186048746109009, + "eval_runtime": 37.5746, + "eval_samples_per_second": 26.614, + "eval_steps_per_second": 1.118, + "step": 330 + }, + { + "epoch": 0.36273972602739724, + "grad_norm": 2.3770196437835693, + "learning_rate": 4.912672623883022e-06, + "loss": 2.0167, + "step": 331 + }, + { + "epoch": 0.36383561643835616, + "grad_norm": 2.3357701301574707, + "learning_rate": 4.910641754671e-06, + "loss": 1.9972, + "step": 332 + }, + { + "epoch": 0.3649315068493151, + "grad_norm": 2.4100499153137207, + "learning_rate": 4.908610885458977e-06, + "loss": 2.0319, + "step": 333 + }, + { + "epoch": 0.36602739726027395, + "grad_norm": 2.193018913269043, + "learning_rate": 4.906580016246954e-06, + "loss": 2.0285, + "step": 334 + }, + { + "epoch": 0.36712328767123287, + "grad_norm": 2.275637149810791, + "learning_rate": 4.9045491470349315e-06, + "loss": 2.0749, + "step": 335 + }, + { + "epoch": 0.3682191780821918, + "grad_norm": 2.347121238708496, + "learning_rate": 4.902518277822909e-06, + "loss": 2.0502, + "step": 336 + }, + { + "epoch": 0.3693150684931507, + "grad_norm": 2.210817575454712, + "learning_rate": 4.900487408610886e-06, + "loss": 2.0612, + "step": 337 + }, + { + "epoch": 0.3704109589041096, + "grad_norm": 2.15594482421875, + "learning_rate": 4.898456539398863e-06, + "loss": 2.0653, + "step": 338 + }, + { + "epoch": 0.3715068493150685, + "grad_norm": 2.3996052742004395, + "learning_rate": 4.89642567018684e-06, + "loss": 2.074, + "step": 339 + }, + { + "epoch": 0.3726027397260274, + "grad_norm": 2.384172201156616, + "learning_rate": 4.894394800974818e-06, + "loss": 2.1533, + "step": 340 + }, + { + "epoch": 0.3726027397260274, + "eval_loss": 2.175034284591675, + "eval_runtime": 37.2455, + "eval_samples_per_second": 26.849, + "eval_steps_per_second": 1.128, + "step": 340 + }, + { + "epoch": 0.3736986301369863, + "grad_norm": 2.3431336879730225, + "learning_rate": 4.892363931762794e-06, + "loss": 1.9919, + "step": 341 + }, + { + "epoch": 0.3747945205479452, + "grad_norm": 2.1828389167785645, + "learning_rate": 4.8903330625507725e-06, + "loss": 2.0703, + "step": 342 + }, + { + "epoch": 0.37589041095890413, + "grad_norm": 2.2117249965667725, + "learning_rate": 4.888302193338749e-06, + "loss": 2.0071, + "step": 343 + }, + { + "epoch": 0.376986301369863, + "grad_norm": 2.3595967292785645, + "learning_rate": 4.886271324126727e-06, + "loss": 1.9879, + "step": 344 + }, + { + "epoch": 0.3780821917808219, + "grad_norm": 2.5211799144744873, + "learning_rate": 4.8842404549147034e-06, + "loss": 2.0861, + "step": 345 + }, + { + "epoch": 0.37917808219178084, + "grad_norm": 2.1551895141601562, + "learning_rate": 4.8822095857026816e-06, + "loss": 2.0683, + "step": 346 + }, + { + "epoch": 0.3802739726027397, + "grad_norm": 2.4145829677581787, + "learning_rate": 4.880178716490658e-06, + "loss": 2.043, + "step": 347 + }, + { + "epoch": 0.38136986301369863, + "grad_norm": 2.4273521900177, + "learning_rate": 4.878147847278636e-06, + "loss": 1.9928, + "step": 348 + }, + { + "epoch": 0.38246575342465755, + "grad_norm": 2.261054754257202, + "learning_rate": 4.8761169780666125e-06, + "loss": 2.0065, + "step": 349 + }, + { + "epoch": 0.3835616438356164, + "grad_norm": 2.4703147411346436, + "learning_rate": 4.874086108854591e-06, + "loss": 1.9766, + "step": 350 + }, + { + "epoch": 0.3835616438356164, + "eval_loss": 2.176915168762207, + "eval_runtime": 37.3223, + "eval_samples_per_second": 26.794, + "eval_steps_per_second": 1.125, + "step": 350 + }, + { + "epoch": 0.38465753424657534, + "grad_norm": 2.3353374004364014, + "learning_rate": 4.872055239642567e-06, + "loss": 1.9874, + "step": 351 + }, + { + "epoch": 0.38575342465753426, + "grad_norm": 2.227834463119507, + "learning_rate": 4.870024370430544e-06, + "loss": 1.9644, + "step": 352 + }, + { + "epoch": 0.3868493150684931, + "grad_norm": 2.3155159950256348, + "learning_rate": 4.867993501218522e-06, + "loss": 2.0227, + "step": 353 + }, + { + "epoch": 0.38794520547945205, + "grad_norm": 2.2456047534942627, + "learning_rate": 4.865962632006499e-06, + "loss": 2.0566, + "step": 354 + }, + { + "epoch": 0.38904109589041097, + "grad_norm": 2.3364651203155518, + "learning_rate": 4.863931762794476e-06, + "loss": 1.9941, + "step": 355 + }, + { + "epoch": 0.39013698630136984, + "grad_norm": 2.2886059284210205, + "learning_rate": 4.8619008935824535e-06, + "loss": 2.0163, + "step": 356 + }, + { + "epoch": 0.39123287671232876, + "grad_norm": 2.3796911239624023, + "learning_rate": 4.859870024370431e-06, + "loss": 2.0525, + "step": 357 + }, + { + "epoch": 0.3923287671232877, + "grad_norm": 2.381986379623413, + "learning_rate": 4.857839155158408e-06, + "loss": 2.0682, + "step": 358 + }, + { + "epoch": 0.3934246575342466, + "grad_norm": 2.2048425674438477, + "learning_rate": 4.855808285946385e-06, + "loss": 2.0748, + "step": 359 + }, + { + "epoch": 0.39452054794520547, + "grad_norm": 2.162846088409424, + "learning_rate": 4.8537774167343626e-06, + "loss": 2.0512, + "step": 360 + }, + { + "epoch": 0.39452054794520547, + "eval_loss": 2.1645994186401367, + "eval_runtime": 37.3126, + "eval_samples_per_second": 26.801, + "eval_steps_per_second": 1.126, + "step": 360 + }, + { + "epoch": 0.3956164383561644, + "grad_norm": 2.4070873260498047, + "learning_rate": 4.85174654752234e-06, + "loss": 2.042, + "step": 361 + }, + { + "epoch": 0.3967123287671233, + "grad_norm": 2.11716628074646, + "learning_rate": 4.849715678310317e-06, + "loss": 2.0398, + "step": 362 + }, + { + "epoch": 0.3978082191780822, + "grad_norm": 2.265479803085327, + "learning_rate": 4.847684809098294e-06, + "loss": 2.0334, + "step": 363 + }, + { + "epoch": 0.3989041095890411, + "grad_norm": 2.2270994186401367, + "learning_rate": 4.845653939886272e-06, + "loss": 2.0292, + "step": 364 + }, + { + "epoch": 0.4, + "grad_norm": 2.406064033508301, + "learning_rate": 4.843623070674249e-06, + "loss": 2.0351, + "step": 365 + }, + { + "epoch": 0.4010958904109589, + "grad_norm": 2.216654062271118, + "learning_rate": 4.841592201462226e-06, + "loss": 2.005, + "step": 366 + }, + { + "epoch": 0.4021917808219178, + "grad_norm": 2.4241385459899902, + "learning_rate": 4.8395613322502035e-06, + "loss": 2.0647, + "step": 367 + }, + { + "epoch": 0.40328767123287673, + "grad_norm": 2.4315145015716553, + "learning_rate": 4.837530463038181e-06, + "loss": 1.9679, + "step": 368 + }, + { + "epoch": 0.4043835616438356, + "grad_norm": 2.3034439086914062, + "learning_rate": 4.835499593826158e-06, + "loss": 2.0259, + "step": 369 + }, + { + "epoch": 0.4054794520547945, + "grad_norm": 2.493436098098755, + "learning_rate": 4.833468724614135e-06, + "loss": 2.0468, + "step": 370 + }, + { + "epoch": 0.4054794520547945, + "eval_loss": 2.161508083343506, + "eval_runtime": 37.2692, + "eval_samples_per_second": 26.832, + "eval_steps_per_second": 1.127, + "step": 370 + }, + { + "epoch": 0.40657534246575344, + "grad_norm": 2.3887157440185547, + "learning_rate": 4.831437855402113e-06, + "loss": 2.0313, + "step": 371 + }, + { + "epoch": 0.4076712328767123, + "grad_norm": 2.563095808029175, + "learning_rate": 4.82940698619009e-06, + "loss": 1.9288, + "step": 372 + }, + { + "epoch": 0.40876712328767123, + "grad_norm": 2.2609968185424805, + "learning_rate": 4.827376116978067e-06, + "loss": 1.9975, + "step": 373 + }, + { + "epoch": 0.40986301369863015, + "grad_norm": 2.624257802963257, + "learning_rate": 4.8253452477660444e-06, + "loss": 2.0639, + "step": 374 + }, + { + "epoch": 0.410958904109589, + "grad_norm": 2.502197265625, + "learning_rate": 4.823314378554022e-06, + "loss": 2.0424, + "step": 375 + }, + { + "epoch": 0.41205479452054794, + "grad_norm": 2.36297345161438, + "learning_rate": 4.821283509341998e-06, + "loss": 2.027, + "step": 376 + }, + { + "epoch": 0.41315068493150686, + "grad_norm": 2.388918161392212, + "learning_rate": 4.819252640129976e-06, + "loss": 2.0002, + "step": 377 + }, + { + "epoch": 0.41424657534246573, + "grad_norm": 2.4262335300445557, + "learning_rate": 4.817221770917953e-06, + "loss": 2.0215, + "step": 378 + }, + { + "epoch": 0.41534246575342465, + "grad_norm": 2.8953094482421875, + "learning_rate": 4.815190901705931e-06, + "loss": 1.988, + "step": 379 + }, + { + "epoch": 0.41643835616438357, + "grad_norm": 2.3092124462127686, + "learning_rate": 4.813160032493907e-06, + "loss": 2.0199, + "step": 380 + }, + { + "epoch": 0.41643835616438357, + "eval_loss": 2.1589064598083496, + "eval_runtime": 37.2649, + "eval_samples_per_second": 26.835, + "eval_steps_per_second": 1.127, + "step": 380 + }, + { + "epoch": 0.41753424657534244, + "grad_norm": 3.0842413902282715, + "learning_rate": 4.811129163281885e-06, + "loss": 2.0134, + "step": 381 + }, + { + "epoch": 0.41863013698630136, + "grad_norm": 2.2015092372894287, + "learning_rate": 4.809098294069862e-06, + "loss": 2.0447, + "step": 382 + }, + { + "epoch": 0.4197260273972603, + "grad_norm": 2.93471097946167, + "learning_rate": 4.80706742485784e-06, + "loss": 2.0329, + "step": 383 + }, + { + "epoch": 0.4208219178082192, + "grad_norm": 2.3277227878570557, + "learning_rate": 4.805036555645816e-06, + "loss": 2.0805, + "step": 384 + }, + { + "epoch": 0.42191780821917807, + "grad_norm": 2.357966184616089, + "learning_rate": 4.8030056864337945e-06, + "loss": 2.0008, + "step": 385 + }, + { + "epoch": 0.423013698630137, + "grad_norm": 2.7505595684051514, + "learning_rate": 4.800974817221771e-06, + "loss": 2.059, + "step": 386 + }, + { + "epoch": 0.4241095890410959, + "grad_norm": 2.144338369369507, + "learning_rate": 4.798943948009749e-06, + "loss": 2.0125, + "step": 387 + }, + { + "epoch": 0.4252054794520548, + "grad_norm": 2.608407735824585, + "learning_rate": 4.7969130787977254e-06, + "loss": 2.0263, + "step": 388 + }, + { + "epoch": 0.4263013698630137, + "grad_norm": 2.4293761253356934, + "learning_rate": 4.794882209585703e-06, + "loss": 2.0107, + "step": 389 + }, + { + "epoch": 0.4273972602739726, + "grad_norm": 2.2136595249176025, + "learning_rate": 4.79285134037368e-06, + "loss": 2.0421, + "step": 390 + }, + { + "epoch": 0.4273972602739726, + "eval_loss": 2.1473641395568848, + "eval_runtime": 37.2895, + "eval_samples_per_second": 26.817, + "eval_steps_per_second": 1.126, + "step": 390 + }, + { + "epoch": 0.4284931506849315, + "grad_norm": 2.3706328868865967, + "learning_rate": 4.790820471161657e-06, + "loss": 1.9828, + "step": 391 + }, + { + "epoch": 0.4295890410958904, + "grad_norm": 2.1628777980804443, + "learning_rate": 4.7887896019496345e-06, + "loss": 2.0006, + "step": 392 + }, + { + "epoch": 0.43068493150684933, + "grad_norm": 2.5572149753570557, + "learning_rate": 4.786758732737612e-06, + "loss": 2.0046, + "step": 393 + }, + { + "epoch": 0.4317808219178082, + "grad_norm": 2.44661808013916, + "learning_rate": 4.784727863525589e-06, + "loss": 2.0513, + "step": 394 + }, + { + "epoch": 0.4328767123287671, + "grad_norm": 2.429487466812134, + "learning_rate": 4.782696994313566e-06, + "loss": 1.9619, + "step": 395 + }, + { + "epoch": 0.43397260273972604, + "grad_norm": 2.233353614807129, + "learning_rate": 4.780666125101544e-06, + "loss": 2.0141, + "step": 396 + }, + { + "epoch": 0.4350684931506849, + "grad_norm": 2.137549638748169, + "learning_rate": 4.778635255889521e-06, + "loss": 1.9512, + "step": 397 + }, + { + "epoch": 0.43616438356164383, + "grad_norm": 2.187638759613037, + "learning_rate": 4.776604386677498e-06, + "loss": 1.9802, + "step": 398 + }, + { + "epoch": 0.43726027397260275, + "grad_norm": 2.3084137439727783, + "learning_rate": 4.7745735174654755e-06, + "loss": 1.9887, + "step": 399 + }, + { + "epoch": 0.4383561643835616, + "grad_norm": 2.16654109954834, + "learning_rate": 4.772542648253453e-06, + "loss": 2.0535, + "step": 400 + }, + { + "epoch": 0.4383561643835616, + "eval_loss": 2.1444413661956787, + "eval_runtime": 37.2158, + "eval_samples_per_second": 26.87, + "eval_steps_per_second": 1.129, + "step": 400 + }, + { + "epoch": 0.43945205479452054, + "grad_norm": 2.0973072052001953, + "learning_rate": 4.77051177904143e-06, + "loss": 2.0529, + "step": 401 + }, + { + "epoch": 0.44054794520547946, + "grad_norm": 2.421881675720215, + "learning_rate": 4.768480909829407e-06, + "loss": 1.9893, + "step": 402 + }, + { + "epoch": 0.44164383561643833, + "grad_norm": 2.349296808242798, + "learning_rate": 4.766450040617385e-06, + "loss": 2.0237, + "step": 403 + }, + { + "epoch": 0.44273972602739725, + "grad_norm": 2.2154288291931152, + "learning_rate": 4.764419171405362e-06, + "loss": 1.996, + "step": 404 + }, + { + "epoch": 0.4438356164383562, + "grad_norm": 2.6355648040771484, + "learning_rate": 4.762388302193339e-06, + "loss": 2.0345, + "step": 405 + }, + { + "epoch": 0.4449315068493151, + "grad_norm": 2.352238655090332, + "learning_rate": 4.760357432981316e-06, + "loss": 2.0219, + "step": 406 + }, + { + "epoch": 0.44602739726027396, + "grad_norm": 2.1886510848999023, + "learning_rate": 4.758326563769294e-06, + "loss": 1.9923, + "step": 407 + }, + { + "epoch": 0.4471232876712329, + "grad_norm": 2.6483092308044434, + "learning_rate": 4.756295694557271e-06, + "loss": 2.0236, + "step": 408 + }, + { + "epoch": 0.4482191780821918, + "grad_norm": 2.382389545440674, + "learning_rate": 4.754264825345248e-06, + "loss": 2.0249, + "step": 409 + }, + { + "epoch": 0.44931506849315067, + "grad_norm": 2.4824411869049072, + "learning_rate": 4.7522339561332255e-06, + "loss": 1.9571, + "step": 410 + }, + { + "epoch": 0.44931506849315067, + "eval_loss": 2.1502621173858643, + "eval_runtime": 37.1789, + "eval_samples_per_second": 26.897, + "eval_steps_per_second": 1.13, + "step": 410 + }, + { + "epoch": 0.4504109589041096, + "grad_norm": 2.325589656829834, + "learning_rate": 4.750203086921203e-06, + "loss": 1.9753, + "step": 411 + }, + { + "epoch": 0.4515068493150685, + "grad_norm": 2.3618829250335693, + "learning_rate": 4.74817221770918e-06, + "loss": 1.9397, + "step": 412 + }, + { + "epoch": 0.4526027397260274, + "grad_norm": 2.3922839164733887, + "learning_rate": 4.7461413484971565e-06, + "loss": 2.0119, + "step": 413 + }, + { + "epoch": 0.4536986301369863, + "grad_norm": 2.403475046157837, + "learning_rate": 4.744110479285135e-06, + "loss": 2.0664, + "step": 414 + }, + { + "epoch": 0.4547945205479452, + "grad_norm": 2.225860834121704, + "learning_rate": 4.742079610073111e-06, + "loss": 1.9525, + "step": 415 + }, + { + "epoch": 0.4558904109589041, + "grad_norm": 2.344074010848999, + "learning_rate": 4.740048740861089e-06, + "loss": 1.9785, + "step": 416 + }, + { + "epoch": 0.456986301369863, + "grad_norm": 2.255382776260376, + "learning_rate": 4.738017871649066e-06, + "loss": 1.9902, + "step": 417 + }, + { + "epoch": 0.45808219178082193, + "grad_norm": 2.160628318786621, + "learning_rate": 4.735987002437044e-06, + "loss": 2.0075, + "step": 418 + }, + { + "epoch": 0.4591780821917808, + "grad_norm": 2.5667307376861572, + "learning_rate": 4.73395613322502e-06, + "loss": 2.0625, + "step": 419 + }, + { + "epoch": 0.4602739726027397, + "grad_norm": 2.3771040439605713, + "learning_rate": 4.731925264012998e-06, + "loss": 2.0724, + "step": 420 + }, + { + "epoch": 0.4602739726027397, + "eval_loss": 2.138829231262207, + "eval_runtime": 37.4361, + "eval_samples_per_second": 26.712, + "eval_steps_per_second": 1.122, + "step": 420 + }, + { + "epoch": 0.46136986301369864, + "grad_norm": 2.336482048034668, + "learning_rate": 4.729894394800975e-06, + "loss": 2.0234, + "step": 421 + }, + { + "epoch": 0.4624657534246575, + "grad_norm": 2.4206933975219727, + "learning_rate": 4.727863525588953e-06, + "loss": 1.9693, + "step": 422 + }, + { + "epoch": 0.46356164383561643, + "grad_norm": 2.4845645427703857, + "learning_rate": 4.725832656376929e-06, + "loss": 1.9239, + "step": 423 + }, + { + "epoch": 0.46465753424657535, + "grad_norm": 2.5199646949768066, + "learning_rate": 4.723801787164907e-06, + "loss": 1.9801, + "step": 424 + }, + { + "epoch": 0.4657534246575342, + "grad_norm": 2.2969748973846436, + "learning_rate": 4.721770917952885e-06, + "loss": 2.0192, + "step": 425 + }, + { + "epoch": 0.46684931506849314, + "grad_norm": 2.423503875732422, + "learning_rate": 4.719740048740861e-06, + "loss": 1.9684, + "step": 426 + }, + { + "epoch": 0.46794520547945206, + "grad_norm": 2.3681061267852783, + "learning_rate": 4.717709179528839e-06, + "loss": 1.9934, + "step": 427 + }, + { + "epoch": 0.469041095890411, + "grad_norm": 2.277965784072876, + "learning_rate": 4.715678310316816e-06, + "loss": 2.0166, + "step": 428 + }, + { + "epoch": 0.47013698630136985, + "grad_norm": 2.2730283737182617, + "learning_rate": 4.713647441104794e-06, + "loss": 2.011, + "step": 429 + }, + { + "epoch": 0.4712328767123288, + "grad_norm": 2.6694860458374023, + "learning_rate": 4.71161657189277e-06, + "loss": 1.9645, + "step": 430 + }, + { + "epoch": 0.4712328767123288, + "eval_loss": 2.1322011947631836, + "eval_runtime": 37.5723, + "eval_samples_per_second": 26.615, + "eval_steps_per_second": 1.118, + "step": 430 + }, + { + "epoch": 0.4723287671232877, + "grad_norm": 2.4100632667541504, + "learning_rate": 4.709585702680748e-06, + "loss": 1.9478, + "step": 431 + }, + { + "epoch": 0.47342465753424656, + "grad_norm": 2.2779541015625, + "learning_rate": 4.707554833468725e-06, + "loss": 2.0274, + "step": 432 + }, + { + "epoch": 0.4745205479452055, + "grad_norm": 2.55690336227417, + "learning_rate": 4.705523964256703e-06, + "loss": 2.0831, + "step": 433 + }, + { + "epoch": 0.4756164383561644, + "grad_norm": 2.455623149871826, + "learning_rate": 4.703493095044679e-06, + "loss": 1.9673, + "step": 434 + }, + { + "epoch": 0.4767123287671233, + "grad_norm": 2.2613699436187744, + "learning_rate": 4.701462225832657e-06, + "loss": 1.992, + "step": 435 + }, + { + "epoch": 0.4778082191780822, + "grad_norm": 2.4467933177948, + "learning_rate": 4.699431356620634e-06, + "loss": 1.9105, + "step": 436 + }, + { + "epoch": 0.4789041095890411, + "grad_norm": 2.4194023609161377, + "learning_rate": 4.697400487408612e-06, + "loss": 2.074, + "step": 437 + }, + { + "epoch": 0.48, + "grad_norm": 2.2165699005126953, + "learning_rate": 4.695369618196588e-06, + "loss": 2.0152, + "step": 438 + }, + { + "epoch": 0.4810958904109589, + "grad_norm": 2.1709225177764893, + "learning_rate": 4.693338748984566e-06, + "loss": 1.9888, + "step": 439 + }, + { + "epoch": 0.4821917808219178, + "grad_norm": 2.2325828075408936, + "learning_rate": 4.691307879772543e-06, + "loss": 1.9838, + "step": 440 + }, + { + "epoch": 0.4821917808219178, + "eval_loss": 2.128732919692993, + "eval_runtime": 37.205, + "eval_samples_per_second": 26.878, + "eval_steps_per_second": 1.129, + "step": 440 + }, + { + "epoch": 0.4832876712328767, + "grad_norm": 2.446331739425659, + "learning_rate": 4.68927701056052e-06, + "loss": 1.9295, + "step": 441 + }, + { + "epoch": 0.4843835616438356, + "grad_norm": 2.5459415912628174, + "learning_rate": 4.6872461413484975e-06, + "loss": 2.0351, + "step": 442 + }, + { + "epoch": 0.48547945205479454, + "grad_norm": 2.2877118587493896, + "learning_rate": 4.685215272136475e-06, + "loss": 2.0, + "step": 443 + }, + { + "epoch": 0.4865753424657534, + "grad_norm": 2.346625328063965, + "learning_rate": 4.683184402924452e-06, + "loss": 1.9928, + "step": 444 + }, + { + "epoch": 0.4876712328767123, + "grad_norm": 2.268467664718628, + "learning_rate": 4.681153533712429e-06, + "loss": 1.9969, + "step": 445 + }, + { + "epoch": 0.48876712328767125, + "grad_norm": 2.3699793815612793, + "learning_rate": 4.679122664500407e-06, + "loss": 2.0636, + "step": 446 + }, + { + "epoch": 0.4898630136986301, + "grad_norm": 2.2871041297912598, + "learning_rate": 4.677091795288384e-06, + "loss": 1.9535, + "step": 447 + }, + { + "epoch": 0.49095890410958903, + "grad_norm": 2.249393939971924, + "learning_rate": 4.675060926076361e-06, + "loss": 1.9358, + "step": 448 + }, + { + "epoch": 0.49205479452054796, + "grad_norm": 2.299863815307617, + "learning_rate": 4.673030056864338e-06, + "loss": 2.0299, + "step": 449 + }, + { + "epoch": 0.4931506849315068, + "grad_norm": 2.31239652633667, + "learning_rate": 4.670999187652316e-06, + "loss": 1.9372, + "step": 450 + }, + { + "epoch": 0.4931506849315068, + "eval_loss": 2.1267426013946533, + "eval_runtime": 37.1884, + "eval_samples_per_second": 26.89, + "eval_steps_per_second": 1.129, + "step": 450 + }, + { + "epoch": 0.49424657534246574, + "grad_norm": 2.2824623584747314, + "learning_rate": 4.668968318440293e-06, + "loss": 1.9435, + "step": 451 + }, + { + "epoch": 0.49534246575342467, + "grad_norm": 2.285339593887329, + "learning_rate": 4.66693744922827e-06, + "loss": 1.9455, + "step": 452 + }, + { + "epoch": 0.4964383561643836, + "grad_norm": 2.346592903137207, + "learning_rate": 4.6649065800162475e-06, + "loss": 2.0212, + "step": 453 + }, + { + "epoch": 0.49753424657534245, + "grad_norm": 2.280669689178467, + "learning_rate": 4.662875710804225e-06, + "loss": 1.9136, + "step": 454 + }, + { + "epoch": 0.4986301369863014, + "grad_norm": 2.178434133529663, + "learning_rate": 4.660844841592202e-06, + "loss": 1.9952, + "step": 455 + }, + { + "epoch": 0.4997260273972603, + "grad_norm": 2.108492851257324, + "learning_rate": 4.658813972380179e-06, + "loss": 1.9022, + "step": 456 + }, + { + "epoch": 0.5008219178082192, + "grad_norm": 2.3862147331237793, + "learning_rate": 4.656783103168157e-06, + "loss": 2.0332, + "step": 457 + }, + { + "epoch": 0.5019178082191781, + "grad_norm": 2.2017860412597656, + "learning_rate": 4.654752233956134e-06, + "loss": 1.8998, + "step": 458 + }, + { + "epoch": 0.503013698630137, + "grad_norm": 2.222134828567505, + "learning_rate": 4.652721364744111e-06, + "loss": 1.9836, + "step": 459 + }, + { + "epoch": 0.5041095890410959, + "grad_norm": 2.2974510192871094, + "learning_rate": 4.6506904955320885e-06, + "loss": 1.9626, + "step": 460 + }, + { + "epoch": 0.5041095890410959, + "eval_loss": 2.1345443725585938, + "eval_runtime": 37.2785, + "eval_samples_per_second": 26.825, + "eval_steps_per_second": 1.127, + "step": 460 + }, + { + "epoch": 0.5052054794520547, + "grad_norm": 2.107480764389038, + "learning_rate": 4.648659626320066e-06, + "loss": 2.0061, + "step": 461 + }, + { + "epoch": 0.5063013698630137, + "grad_norm": 2.3541181087493896, + "learning_rate": 4.646628757108043e-06, + "loss": 2.0133, + "step": 462 + }, + { + "epoch": 0.5073972602739726, + "grad_norm": 2.215571880340576, + "learning_rate": 4.6445978878960194e-06, + "loss": 1.9727, + "step": 463 + }, + { + "epoch": 0.5084931506849315, + "grad_norm": 2.2263615131378174, + "learning_rate": 4.6425670186839976e-06, + "loss": 1.9477, + "step": 464 + }, + { + "epoch": 0.5095890410958904, + "grad_norm": 2.178900957107544, + "learning_rate": 4.640536149471974e-06, + "loss": 1.9842, + "step": 465 + }, + { + "epoch": 0.5106849315068493, + "grad_norm": 2.3703105449676514, + "learning_rate": 4.638505280259952e-06, + "loss": 2.0082, + "step": 466 + }, + { + "epoch": 0.5117808219178083, + "grad_norm": 2.1587533950805664, + "learning_rate": 4.6364744110479285e-06, + "loss": 2.0271, + "step": 467 + }, + { + "epoch": 0.5128767123287671, + "grad_norm": 2.1626017093658447, + "learning_rate": 4.634443541835907e-06, + "loss": 1.9858, + "step": 468 + }, + { + "epoch": 0.513972602739726, + "grad_norm": 2.326869487762451, + "learning_rate": 4.632412672623883e-06, + "loss": 2.0413, + "step": 469 + }, + { + "epoch": 0.5150684931506849, + "grad_norm": 2.266502857208252, + "learning_rate": 4.630381803411861e-06, + "loss": 1.9584, + "step": 470 + }, + { + "epoch": 0.5150684931506849, + "eval_loss": 2.1273915767669678, + "eval_runtime": 37.2571, + "eval_samples_per_second": 26.841, + "eval_steps_per_second": 1.127, + "step": 470 + }, + { + "epoch": 0.5161643835616438, + "grad_norm": 2.131348133087158, + "learning_rate": 4.628350934199838e-06, + "loss": 1.863, + "step": 471 + }, + { + "epoch": 0.5172602739726028, + "grad_norm": 2.2635257244110107, + "learning_rate": 4.626320064987816e-06, + "loss": 1.9699, + "step": 472 + }, + { + "epoch": 0.5183561643835617, + "grad_norm": 2.1303744316101074, + "learning_rate": 4.624289195775792e-06, + "loss": 2.0001, + "step": 473 + }, + { + "epoch": 0.5194520547945205, + "grad_norm": 2.1736788749694824, + "learning_rate": 4.62225832656377e-06, + "loss": 1.9684, + "step": 474 + }, + { + "epoch": 0.5205479452054794, + "grad_norm": 2.1996278762817383, + "learning_rate": 4.620227457351747e-06, + "loss": 1.9739, + "step": 475 + }, + { + "epoch": 0.5216438356164383, + "grad_norm": 2.225456953048706, + "learning_rate": 4.618196588139724e-06, + "loss": 1.9649, + "step": 476 + }, + { + "epoch": 0.5227397260273973, + "grad_norm": 2.118537187576294, + "learning_rate": 4.616165718927701e-06, + "loss": 1.9911, + "step": 477 + }, + { + "epoch": 0.5238356164383562, + "grad_norm": 2.232940435409546, + "learning_rate": 4.6141348497156786e-06, + "loss": 1.944, + "step": 478 + }, + { + "epoch": 0.5249315068493151, + "grad_norm": 2.1776034832000732, + "learning_rate": 4.612103980503656e-06, + "loss": 2.0342, + "step": 479 + }, + { + "epoch": 0.5260273972602739, + "grad_norm": 2.042645215988159, + "learning_rate": 4.610073111291633e-06, + "loss": 2.0124, + "step": 480 + }, + { + "epoch": 0.5260273972602739, + "eval_loss": 2.118508815765381, + "eval_runtime": 37.2764, + "eval_samples_per_second": 26.827, + "eval_steps_per_second": 1.127, + "step": 480 + }, + { + "epoch": 0.5271232876712328, + "grad_norm": 2.302661895751953, + "learning_rate": 4.60804224207961e-06, + "loss": 1.9874, + "step": 481 + }, + { + "epoch": 0.5282191780821918, + "grad_norm": 2.1251118183135986, + "learning_rate": 4.606011372867588e-06, + "loss": 1.9594, + "step": 482 + }, + { + "epoch": 0.5293150684931507, + "grad_norm": 2.157043933868408, + "learning_rate": 4.603980503655565e-06, + "loss": 1.9955, + "step": 483 + }, + { + "epoch": 0.5304109589041096, + "grad_norm": 2.291428804397583, + "learning_rate": 4.601949634443542e-06, + "loss": 1.9695, + "step": 484 + }, + { + "epoch": 0.5315068493150685, + "grad_norm": 2.2341537475585938, + "learning_rate": 4.5999187652315195e-06, + "loss": 1.899, + "step": 485 + }, + { + "epoch": 0.5326027397260275, + "grad_norm": 2.1881206035614014, + "learning_rate": 4.597887896019497e-06, + "loss": 1.9066, + "step": 486 + }, + { + "epoch": 0.5336986301369863, + "grad_norm": 2.3057658672332764, + "learning_rate": 4.595857026807474e-06, + "loss": 1.9646, + "step": 487 + }, + { + "epoch": 0.5347945205479452, + "grad_norm": 2.2318356037139893, + "learning_rate": 4.593826157595451e-06, + "loss": 1.9662, + "step": 488 + }, + { + "epoch": 0.5358904109589041, + "grad_norm": 2.170509099960327, + "learning_rate": 4.591795288383429e-06, + "loss": 1.9522, + "step": 489 + }, + { + "epoch": 0.536986301369863, + "grad_norm": 2.101646661758423, + "learning_rate": 4.589764419171406e-06, + "loss": 2.0315, + "step": 490 + }, + { + "epoch": 0.536986301369863, + "eval_loss": 2.113968849182129, + "eval_runtime": 37.2751, + "eval_samples_per_second": 26.828, + "eval_steps_per_second": 1.127, + "step": 490 + }, + { + "epoch": 0.538082191780822, + "grad_norm": 2.1563706398010254, + "learning_rate": 4.587733549959383e-06, + "loss": 1.9843, + "step": 491 + }, + { + "epoch": 0.5391780821917809, + "grad_norm": 2.295722246170044, + "learning_rate": 4.5857026807473604e-06, + "loss": 1.9499, + "step": 492 + }, + { + "epoch": 0.5402739726027397, + "grad_norm": 1.974424123764038, + "learning_rate": 4.583671811535338e-06, + "loss": 1.9346, + "step": 493 + }, + { + "epoch": 0.5413698630136986, + "grad_norm": 2.4402668476104736, + "learning_rate": 4.581640942323315e-06, + "loss": 1.9053, + "step": 494 + }, + { + "epoch": 0.5424657534246575, + "grad_norm": 2.269000768661499, + "learning_rate": 4.579610073111292e-06, + "loss": 2.0108, + "step": 495 + }, + { + "epoch": 0.5435616438356164, + "grad_norm": 2.1182594299316406, + "learning_rate": 4.5775792038992695e-06, + "loss": 2.0155, + "step": 496 + }, + { + "epoch": 0.5446575342465754, + "grad_norm": 2.2181780338287354, + "learning_rate": 4.575548334687247e-06, + "loss": 2.0112, + "step": 497 + }, + { + "epoch": 0.5457534246575343, + "grad_norm": 2.2806756496429443, + "learning_rate": 4.573517465475224e-06, + "loss": 2.0418, + "step": 498 + }, + { + "epoch": 0.5468493150684931, + "grad_norm": 2.2690632343292236, + "learning_rate": 4.571486596263201e-06, + "loss": 1.9457, + "step": 499 + }, + { + "epoch": 0.547945205479452, + "grad_norm": 2.361335515975952, + "learning_rate": 4.569455727051178e-06, + "loss": 1.9949, + "step": 500 + }, + { + "epoch": 0.547945205479452, + "eval_loss": 2.1165614128112793, + "eval_runtime": 37.2664, + "eval_samples_per_second": 26.834, + "eval_steps_per_second": 1.127, + "step": 500 + }, + { + "epoch": 0.549041095890411, + "grad_norm": 2.3833138942718506, + "learning_rate": 4.567424857839156e-06, + "loss": 1.9932, + "step": 501 + }, + { + "epoch": 0.5501369863013699, + "grad_norm": 2.328500509262085, + "learning_rate": 4.565393988627132e-06, + "loss": 1.9773, + "step": 502 + }, + { + "epoch": 0.5512328767123288, + "grad_norm": 2.4191694259643555, + "learning_rate": 4.5633631194151105e-06, + "loss": 2.0389, + "step": 503 + }, + { + "epoch": 0.5523287671232877, + "grad_norm": 2.4106884002685547, + "learning_rate": 4.561332250203087e-06, + "loss": 1.965, + "step": 504 + }, + { + "epoch": 0.5534246575342465, + "grad_norm": 2.1452605724334717, + "learning_rate": 4.559301380991065e-06, + "loss": 1.9713, + "step": 505 + }, + { + "epoch": 0.5545205479452054, + "grad_norm": 2.2753357887268066, + "learning_rate": 4.5572705117790414e-06, + "loss": 1.9116, + "step": 506 + }, + { + "epoch": 0.5556164383561644, + "grad_norm": 2.6925032138824463, + "learning_rate": 4.5552396425670196e-06, + "loss": 1.9593, + "step": 507 + }, + { + "epoch": 0.5567123287671233, + "grad_norm": 2.3136658668518066, + "learning_rate": 4.553208773354996e-06, + "loss": 2.0072, + "step": 508 + }, + { + "epoch": 0.5578082191780822, + "grad_norm": 2.254849672317505, + "learning_rate": 4.551177904142974e-06, + "loss": 1.9193, + "step": 509 + }, + { + "epoch": 0.5589041095890411, + "grad_norm": 2.352755546569824, + "learning_rate": 4.5491470349309505e-06, + "loss": 2.0228, + "step": 510 + }, + { + "epoch": 0.5589041095890411, + "eval_loss": 2.114323377609253, + "eval_runtime": 37.3629, + "eval_samples_per_second": 26.764, + "eval_steps_per_second": 1.124, + "step": 510 + }, + { + "epoch": 0.56, + "grad_norm": 2.31339955329895, + "learning_rate": 4.547116165718929e-06, + "loss": 1.9416, + "step": 511 + }, + { + "epoch": 0.5610958904109589, + "grad_norm": 2.191164970397949, + "learning_rate": 4.545085296506905e-06, + "loss": 1.9709, + "step": 512 + }, + { + "epoch": 0.5621917808219178, + "grad_norm": 2.193551778793335, + "learning_rate": 4.543054427294882e-06, + "loss": 1.9475, + "step": 513 + }, + { + "epoch": 0.5632876712328767, + "grad_norm": 2.222710371017456, + "learning_rate": 4.54102355808286e-06, + "loss": 1.9283, + "step": 514 + }, + { + "epoch": 0.5643835616438356, + "grad_norm": 2.078907012939453, + "learning_rate": 4.538992688870837e-06, + "loss": 1.929, + "step": 515 + }, + { + "epoch": 0.5654794520547946, + "grad_norm": 2.2043654918670654, + "learning_rate": 4.536961819658814e-06, + "loss": 1.9591, + "step": 516 + }, + { + "epoch": 0.5665753424657535, + "grad_norm": 2.192481517791748, + "learning_rate": 4.5349309504467915e-06, + "loss": 1.9618, + "step": 517 + }, + { + "epoch": 0.5676712328767123, + "grad_norm": 2.3509292602539062, + "learning_rate": 4.532900081234769e-06, + "loss": 1.9782, + "step": 518 + }, + { + "epoch": 0.5687671232876712, + "grad_norm": 2.2415924072265625, + "learning_rate": 4.530869212022746e-06, + "loss": 1.9456, + "step": 519 + }, + { + "epoch": 0.5698630136986301, + "grad_norm": 2.2135009765625, + "learning_rate": 4.528838342810723e-06, + "loss": 1.9671, + "step": 520 + }, + { + "epoch": 0.5698630136986301, + "eval_loss": 2.109738826751709, + "eval_runtime": 37.5253, + "eval_samples_per_second": 26.649, + "eval_steps_per_second": 1.119, + "step": 520 + }, + { + "epoch": 0.570958904109589, + "grad_norm": 2.3513574600219727, + "learning_rate": 4.526807473598701e-06, + "loss": 1.9625, + "step": 521 + }, + { + "epoch": 0.572054794520548, + "grad_norm": 1.9949424266815186, + "learning_rate": 4.524776604386678e-06, + "loss": 1.9621, + "step": 522 + }, + { + "epoch": 0.5731506849315069, + "grad_norm": 2.199488401412964, + "learning_rate": 4.522745735174655e-06, + "loss": 1.9389, + "step": 523 + }, + { + "epoch": 0.5742465753424657, + "grad_norm": 2.2119452953338623, + "learning_rate": 4.520714865962632e-06, + "loss": 1.9936, + "step": 524 + }, + { + "epoch": 0.5753424657534246, + "grad_norm": 2.242812156677246, + "learning_rate": 4.51868399675061e-06, + "loss": 1.9728, + "step": 525 + }, + { + "epoch": 0.5764383561643835, + "grad_norm": 2.1713850498199463, + "learning_rate": 4.516653127538587e-06, + "loss": 1.98, + "step": 526 + }, + { + "epoch": 0.5775342465753425, + "grad_norm": 2.261742353439331, + "learning_rate": 4.514622258326564e-06, + "loss": 1.9412, + "step": 527 + }, + { + "epoch": 0.5786301369863014, + "grad_norm": 2.222663402557373, + "learning_rate": 4.5125913891145415e-06, + "loss": 1.9266, + "step": 528 + }, + { + "epoch": 0.5797260273972603, + "grad_norm": 2.1108906269073486, + "learning_rate": 4.510560519902519e-06, + "loss": 1.9539, + "step": 529 + }, + { + "epoch": 0.5808219178082191, + "grad_norm": 1.9594205617904663, + "learning_rate": 4.508529650690496e-06, + "loss": 1.9842, + "step": 530 + }, + { + "epoch": 0.5808219178082191, + "eval_loss": 2.1073970794677734, + "eval_runtime": 37.6167, + "eval_samples_per_second": 26.584, + "eval_steps_per_second": 1.117, + "step": 530 + }, + { + "epoch": 0.581917808219178, + "grad_norm": 2.1339495182037354, + "learning_rate": 4.506498781478473e-06, + "loss": 1.9095, + "step": 531 + }, + { + "epoch": 0.583013698630137, + "grad_norm": 2.116248607635498, + "learning_rate": 4.504467912266451e-06, + "loss": 1.9091, + "step": 532 + }, + { + "epoch": 0.5841095890410959, + "grad_norm": 2.063962459564209, + "learning_rate": 4.502437043054428e-06, + "loss": 1.9537, + "step": 533 + }, + { + "epoch": 0.5852054794520548, + "grad_norm": 2.162569522857666, + "learning_rate": 4.500406173842405e-06, + "loss": 2.0253, + "step": 534 + }, + { + "epoch": 0.5863013698630137, + "grad_norm": 2.1783549785614014, + "learning_rate": 4.4983753046303824e-06, + "loss": 1.9714, + "step": 535 + }, + { + "epoch": 0.5873972602739727, + "grad_norm": 2.0791046619415283, + "learning_rate": 4.49634443541836e-06, + "loss": 1.9629, + "step": 536 + }, + { + "epoch": 0.5884931506849315, + "grad_norm": 2.3923251628875732, + "learning_rate": 4.494313566206336e-06, + "loss": 1.9445, + "step": 537 + }, + { + "epoch": 0.5895890410958904, + "grad_norm": 2.1848347187042236, + "learning_rate": 4.492282696994314e-06, + "loss": 2.007, + "step": 538 + }, + { + "epoch": 0.5906849315068493, + "grad_norm": 2.0824925899505615, + "learning_rate": 4.490251827782291e-06, + "loss": 1.9409, + "step": 539 + }, + { + "epoch": 0.5917808219178082, + "grad_norm": 2.057790756225586, + "learning_rate": 4.488220958570269e-06, + "loss": 1.9061, + "step": 540 + }, + { + "epoch": 0.5917808219178082, + "eval_loss": 2.1101415157318115, + "eval_runtime": 37.2235, + "eval_samples_per_second": 26.865, + "eval_steps_per_second": 1.128, + "step": 540 + }, + { + "epoch": 0.5928767123287672, + "grad_norm": 2.3652148246765137, + "learning_rate": 4.486190089358245e-06, + "loss": 1.9379, + "step": 541 + }, + { + "epoch": 0.5939726027397261, + "grad_norm": 2.195535659790039, + "learning_rate": 4.484159220146223e-06, + "loss": 1.8881, + "step": 542 + }, + { + "epoch": 0.5950684931506849, + "grad_norm": 2.258000135421753, + "learning_rate": 4.4821283509342e-06, + "loss": 1.9923, + "step": 543 + }, + { + "epoch": 0.5961643835616438, + "grad_norm": 2.097041606903076, + "learning_rate": 4.480097481722178e-06, + "loss": 1.9715, + "step": 544 + }, + { + "epoch": 0.5972602739726027, + "grad_norm": 2.1741156578063965, + "learning_rate": 4.478066612510154e-06, + "loss": 1.9487, + "step": 545 + }, + { + "epoch": 0.5983561643835617, + "grad_norm": 2.135338306427002, + "learning_rate": 4.4760357432981325e-06, + "loss": 1.9172, + "step": 546 + }, + { + "epoch": 0.5994520547945206, + "grad_norm": 2.3435418605804443, + "learning_rate": 4.474004874086109e-06, + "loss": 1.945, + "step": 547 + }, + { + "epoch": 0.6005479452054795, + "grad_norm": 2.2242488861083984, + "learning_rate": 4.471974004874087e-06, + "loss": 2.0717, + "step": 548 + }, + { + "epoch": 0.6016438356164383, + "grad_norm": 2.3371622562408447, + "learning_rate": 4.4699431356620635e-06, + "loss": 1.9082, + "step": 549 + }, + { + "epoch": 0.6027397260273972, + "grad_norm": 2.176064968109131, + "learning_rate": 4.467912266450041e-06, + "loss": 2.0154, + "step": 550 + }, + { + "epoch": 0.6027397260273972, + "eval_loss": 2.105236291885376, + "eval_runtime": 37.1885, + "eval_samples_per_second": 26.89, + "eval_steps_per_second": 1.129, + "step": 550 + }, + { + "epoch": 0.6038356164383561, + "grad_norm": 2.131166934967041, + "learning_rate": 4.465881397238018e-06, + "loss": 1.9575, + "step": 551 + }, + { + "epoch": 0.6049315068493151, + "grad_norm": 2.094514846801758, + "learning_rate": 4.463850528025995e-06, + "loss": 1.9544, + "step": 552 + }, + { + "epoch": 0.606027397260274, + "grad_norm": 2.0827765464782715, + "learning_rate": 4.4618196588139726e-06, + "loss": 1.9539, + "step": 553 + }, + { + "epoch": 0.6071232876712329, + "grad_norm": 2.225749969482422, + "learning_rate": 4.45978878960195e-06, + "loss": 1.983, + "step": 554 + }, + { + "epoch": 0.6082191780821918, + "grad_norm": 2.1363680362701416, + "learning_rate": 4.457757920389927e-06, + "loss": 1.9769, + "step": 555 + }, + { + "epoch": 0.6093150684931506, + "grad_norm": 2.238170623779297, + "learning_rate": 4.455727051177904e-06, + "loss": 1.9094, + "step": 556 + }, + { + "epoch": 0.6104109589041096, + "grad_norm": 2.228384494781494, + "learning_rate": 4.453696181965882e-06, + "loss": 1.9464, + "step": 557 + }, + { + "epoch": 0.6115068493150685, + "grad_norm": 2.479428768157959, + "learning_rate": 4.451665312753859e-06, + "loss": 2.0017, + "step": 558 + }, + { + "epoch": 0.6126027397260274, + "grad_norm": 2.2331902980804443, + "learning_rate": 4.449634443541836e-06, + "loss": 1.9728, + "step": 559 + }, + { + "epoch": 0.6136986301369863, + "grad_norm": 2.1728556156158447, + "learning_rate": 4.4476035743298135e-06, + "loss": 2.0153, + "step": 560 + }, + { + "epoch": 0.6136986301369863, + "eval_loss": 2.0978519916534424, + "eval_runtime": 37.233, + "eval_samples_per_second": 26.858, + "eval_steps_per_second": 1.128, + "step": 560 + }, + { + "epoch": 0.6147945205479453, + "grad_norm": 2.377814292907715, + "learning_rate": 4.445572705117791e-06, + "loss": 1.9242, + "step": 561 + }, + { + "epoch": 0.6158904109589041, + "grad_norm": 2.190436601638794, + "learning_rate": 4.443541835905768e-06, + "loss": 1.9454, + "step": 562 + }, + { + "epoch": 0.616986301369863, + "grad_norm": 2.052201986312866, + "learning_rate": 4.441510966693745e-06, + "loss": 2.0048, + "step": 563 + }, + { + "epoch": 0.6180821917808219, + "grad_norm": 2.2706589698791504, + "learning_rate": 4.439480097481723e-06, + "loss": 1.9426, + "step": 564 + }, + { + "epoch": 0.6191780821917808, + "grad_norm": 2.3478105068206787, + "learning_rate": 4.4374492282697e-06, + "loss": 2.0047, + "step": 565 + }, + { + "epoch": 0.6202739726027398, + "grad_norm": 2.465078115463257, + "learning_rate": 4.435418359057677e-06, + "loss": 1.8739, + "step": 566 + }, + { + "epoch": 0.6213698630136987, + "grad_norm": 2.0979490280151367, + "learning_rate": 4.433387489845654e-06, + "loss": 1.9588, + "step": 567 + }, + { + "epoch": 0.6224657534246575, + "grad_norm": 2.1239588260650635, + "learning_rate": 4.431356620633632e-06, + "loss": 1.9695, + "step": 568 + }, + { + "epoch": 0.6235616438356164, + "grad_norm": 2.014007806777954, + "learning_rate": 4.429325751421609e-06, + "loss": 1.9232, + "step": 569 + }, + { + "epoch": 0.6246575342465753, + "grad_norm": 2.212414026260376, + "learning_rate": 4.427294882209586e-06, + "loss": 1.928, + "step": 570 + }, + { + "epoch": 0.6246575342465753, + "eval_loss": 2.103149890899658, + "eval_runtime": 37.2641, + "eval_samples_per_second": 26.835, + "eval_steps_per_second": 1.127, + "step": 570 + }, + { + "epoch": 0.6257534246575343, + "grad_norm": 2.0944411754608154, + "learning_rate": 4.4252640129975635e-06, + "loss": 1.9568, + "step": 571 + }, + { + "epoch": 0.6268493150684932, + "grad_norm": 2.1068055629730225, + "learning_rate": 4.423233143785541e-06, + "loss": 1.8917, + "step": 572 + }, + { + "epoch": 0.6279452054794521, + "grad_norm": 2.1992151737213135, + "learning_rate": 4.421202274573518e-06, + "loss": 1.9536, + "step": 573 + }, + { + "epoch": 0.6290410958904109, + "grad_norm": 2.2079834938049316, + "learning_rate": 4.4191714053614945e-06, + "loss": 1.9603, + "step": 574 + }, + { + "epoch": 0.6301369863013698, + "grad_norm": 2.1230316162109375, + "learning_rate": 4.417140536149473e-06, + "loss": 1.8529, + "step": 575 + }, + { + "epoch": 0.6312328767123287, + "grad_norm": 2.1069579124450684, + "learning_rate": 4.415109666937449e-06, + "loss": 1.8997, + "step": 576 + }, + { + "epoch": 0.6323287671232877, + "grad_norm": 2.1683409214019775, + "learning_rate": 4.413078797725427e-06, + "loss": 1.971, + "step": 577 + }, + { + "epoch": 0.6334246575342466, + "grad_norm": 2.0599024295806885, + "learning_rate": 4.411047928513404e-06, + "loss": 1.9154, + "step": 578 + }, + { + "epoch": 0.6345205479452055, + "grad_norm": 2.1605849266052246, + "learning_rate": 4.409017059301382e-06, + "loss": 1.9562, + "step": 579 + }, + { + "epoch": 0.6356164383561644, + "grad_norm": 2.151249408721924, + "learning_rate": 4.406986190089358e-06, + "loss": 1.9571, + "step": 580 + }, + { + "epoch": 0.6356164383561644, + "eval_loss": 2.098177671432495, + "eval_runtime": 37.2578, + "eval_samples_per_second": 26.84, + "eval_steps_per_second": 1.127, + "step": 580 + }, + { + "epoch": 0.6367123287671232, + "grad_norm": 2.1819236278533936, + "learning_rate": 4.404955320877336e-06, + "loss": 1.9816, + "step": 581 + }, + { + "epoch": 0.6378082191780822, + "grad_norm": 2.1131205558776855, + "learning_rate": 4.402924451665313e-06, + "loss": 1.9738, + "step": 582 + }, + { + "epoch": 0.6389041095890411, + "grad_norm": 2.0852456092834473, + "learning_rate": 4.400893582453291e-06, + "loss": 1.924, + "step": 583 + }, + { + "epoch": 0.64, + "grad_norm": 2.0997941493988037, + "learning_rate": 4.398862713241267e-06, + "loss": 1.9067, + "step": 584 + }, + { + "epoch": 0.6410958904109589, + "grad_norm": 2.0848538875579834, + "learning_rate": 4.396831844029245e-06, + "loss": 1.9393, + "step": 585 + }, + { + "epoch": 0.6421917808219179, + "grad_norm": 2.2121758460998535, + "learning_rate": 4.394800974817222e-06, + "loss": 1.9458, + "step": 586 + }, + { + "epoch": 0.6432876712328767, + "grad_norm": 2.194462299346924, + "learning_rate": 4.392770105605199e-06, + "loss": 1.9528, + "step": 587 + }, + { + "epoch": 0.6443835616438356, + "grad_norm": 2.1936254501342773, + "learning_rate": 4.390739236393176e-06, + "loss": 1.9584, + "step": 588 + }, + { + "epoch": 0.6454794520547945, + "grad_norm": 2.1685783863067627, + "learning_rate": 4.388708367181154e-06, + "loss": 1.9666, + "step": 589 + }, + { + "epoch": 0.6465753424657534, + "grad_norm": 2.334571123123169, + "learning_rate": 4.386677497969131e-06, + "loss": 1.9748, + "step": 590 + }, + { + "epoch": 0.6465753424657534, + "eval_loss": 2.0937106609344482, + "eval_runtime": 37.2877, + "eval_samples_per_second": 26.818, + "eval_steps_per_second": 1.126, + "step": 590 + }, + { + "epoch": 0.6476712328767124, + "grad_norm": 2.2901110649108887, + "learning_rate": 4.384646628757108e-06, + "loss": 1.9597, + "step": 591 + }, + { + "epoch": 0.6487671232876713, + "grad_norm": 2.2364838123321533, + "learning_rate": 4.3826157595450855e-06, + "loss": 2.004, + "step": 592 + }, + { + "epoch": 0.6498630136986301, + "grad_norm": 2.1088364124298096, + "learning_rate": 4.380584890333063e-06, + "loss": 1.9319, + "step": 593 + }, + { + "epoch": 0.650958904109589, + "grad_norm": 2.164210319519043, + "learning_rate": 4.37855402112104e-06, + "loss": 1.8906, + "step": 594 + }, + { + "epoch": 0.6520547945205479, + "grad_norm": 2.205200672149658, + "learning_rate": 4.376523151909017e-06, + "loss": 1.8474, + "step": 595 + }, + { + "epoch": 0.6531506849315069, + "grad_norm": 2.239530086517334, + "learning_rate": 4.3744922826969946e-06, + "loss": 2.0033, + "step": 596 + }, + { + "epoch": 0.6542465753424658, + "grad_norm": 2.214747190475464, + "learning_rate": 4.372461413484972e-06, + "loss": 1.9258, + "step": 597 + }, + { + "epoch": 0.6553424657534247, + "grad_norm": 2.1746792793273926, + "learning_rate": 4.370430544272949e-06, + "loss": 1.9479, + "step": 598 + }, + { + "epoch": 0.6564383561643835, + "grad_norm": 2.194131851196289, + "learning_rate": 4.368399675060926e-06, + "loss": 1.9825, + "step": 599 + }, + { + "epoch": 0.6575342465753424, + "grad_norm": 2.3334717750549316, + "learning_rate": 4.366368805848904e-06, + "loss": 1.9831, + "step": 600 + }, + { + "epoch": 0.6575342465753424, + "eval_loss": 2.0881216526031494, + "eval_runtime": 37.3226, + "eval_samples_per_second": 26.793, + "eval_steps_per_second": 1.125, + "step": 600 + }, + { + "epoch": 0.6586301369863014, + "grad_norm": 2.269484758377075, + "learning_rate": 4.364337936636881e-06, + "loss": 1.9407, + "step": 601 + }, + { + "epoch": 0.6597260273972603, + "grad_norm": 2.470773935317993, + "learning_rate": 4.362307067424858e-06, + "loss": 1.8905, + "step": 602 + }, + { + "epoch": 0.6608219178082192, + "grad_norm": 2.485574960708618, + "learning_rate": 4.3602761982128355e-06, + "loss": 1.953, + "step": 603 + }, + { + "epoch": 0.6619178082191781, + "grad_norm": 2.2443618774414062, + "learning_rate": 4.358245329000813e-06, + "loss": 1.9244, + "step": 604 + }, + { + "epoch": 0.663013698630137, + "grad_norm": 2.039600372314453, + "learning_rate": 4.35621445978879e-06, + "loss": 1.9364, + "step": 605 + }, + { + "epoch": 0.6641095890410958, + "grad_norm": 2.093001127243042, + "learning_rate": 4.354183590576767e-06, + "loss": 1.9439, + "step": 606 + }, + { + "epoch": 0.6652054794520548, + "grad_norm": 2.3433260917663574, + "learning_rate": 4.352152721364745e-06, + "loss": 1.9322, + "step": 607 + }, + { + "epoch": 0.6663013698630137, + "grad_norm": 2.2403812408447266, + "learning_rate": 4.350121852152722e-06, + "loss": 1.8829, + "step": 608 + }, + { + "epoch": 0.6673972602739726, + "grad_norm": 2.1542811393737793, + "learning_rate": 4.348090982940699e-06, + "loss": 1.9549, + "step": 609 + }, + { + "epoch": 0.6684931506849315, + "grad_norm": 2.1526548862457275, + "learning_rate": 4.3460601137286764e-06, + "loss": 1.9433, + "step": 610 + }, + { + "epoch": 0.6684931506849315, + "eval_loss": 2.083638906478882, + "eval_runtime": 37.2677, + "eval_samples_per_second": 26.833, + "eval_steps_per_second": 1.127, + "step": 610 + }, + { + "epoch": 0.6695890410958905, + "grad_norm": 2.093421459197998, + "learning_rate": 4.344029244516654e-06, + "loss": 1.9643, + "step": 611 + }, + { + "epoch": 0.6706849315068493, + "grad_norm": 2.2205088138580322, + "learning_rate": 4.341998375304631e-06, + "loss": 1.9309, + "step": 612 + }, + { + "epoch": 0.6717808219178082, + "grad_norm": 2.059763193130493, + "learning_rate": 4.339967506092607e-06, + "loss": 2.0064, + "step": 613 + }, + { + "epoch": 0.6728767123287671, + "grad_norm": 2.0340707302093506, + "learning_rate": 4.3379366368805855e-06, + "loss": 1.9394, + "step": 614 + }, + { + "epoch": 0.673972602739726, + "grad_norm": 2.0716137886047363, + "learning_rate": 4.335905767668562e-06, + "loss": 1.9421, + "step": 615 + }, + { + "epoch": 0.675068493150685, + "grad_norm": 2.176809549331665, + "learning_rate": 4.33387489845654e-06, + "loss": 1.9261, + "step": 616 + }, + { + "epoch": 0.6761643835616439, + "grad_norm": 2.144385814666748, + "learning_rate": 4.3318440292445165e-06, + "loss": 1.9635, + "step": 617 + }, + { + "epoch": 0.6772602739726027, + "grad_norm": 2.0949580669403076, + "learning_rate": 4.329813160032495e-06, + "loss": 1.9417, + "step": 618 + }, + { + "epoch": 0.6783561643835616, + "grad_norm": 2.1727259159088135, + "learning_rate": 4.327782290820471e-06, + "loss": 1.9041, + "step": 619 + }, + { + "epoch": 0.6794520547945205, + "grad_norm": 2.012868881225586, + "learning_rate": 4.325751421608449e-06, + "loss": 1.8973, + "step": 620 + }, + { + "epoch": 0.6794520547945205, + "eval_loss": 2.0839755535125732, + "eval_runtime": 37.5302, + "eval_samples_per_second": 26.645, + "eval_steps_per_second": 1.119, + "step": 620 + }, + { + "epoch": 0.6805479452054795, + "grad_norm": 2.204153299331665, + "learning_rate": 4.323720552396426e-06, + "loss": 1.9719, + "step": 621 + }, + { + "epoch": 0.6816438356164384, + "grad_norm": 2.2421770095825195, + "learning_rate": 4.321689683184404e-06, + "loss": 1.9207, + "step": 622 + }, + { + "epoch": 0.6827397260273973, + "grad_norm": 2.2263410091400146, + "learning_rate": 4.31965881397238e-06, + "loss": 1.9146, + "step": 623 + }, + { + "epoch": 0.6838356164383562, + "grad_norm": 2.2015221118927, + "learning_rate": 4.3176279447603574e-06, + "loss": 1.9311, + "step": 624 + }, + { + "epoch": 0.684931506849315, + "grad_norm": 2.1459643840789795, + "learning_rate": 4.315597075548335e-06, + "loss": 1.9673, + "step": 625 + }, + { + "epoch": 0.686027397260274, + "grad_norm": 2.278928756713867, + "learning_rate": 4.313566206336312e-06, + "loss": 1.954, + "step": 626 + }, + { + "epoch": 0.6871232876712329, + "grad_norm": 2.05305552482605, + "learning_rate": 4.311535337124289e-06, + "loss": 1.9789, + "step": 627 + }, + { + "epoch": 0.6882191780821918, + "grad_norm": 2.3979289531707764, + "learning_rate": 4.3095044679122665e-06, + "loss": 1.926, + "step": 628 + }, + { + "epoch": 0.6893150684931507, + "grad_norm": 2.1635828018188477, + "learning_rate": 4.307473598700244e-06, + "loss": 1.9316, + "step": 629 + }, + { + "epoch": 0.6904109589041096, + "grad_norm": 2.237713098526001, + "learning_rate": 4.305442729488221e-06, + "loss": 1.9409, + "step": 630 + }, + { + "epoch": 0.6904109589041096, + "eval_loss": 2.084226131439209, + "eval_runtime": 37.6039, + "eval_samples_per_second": 26.593, + "eval_steps_per_second": 1.117, + "step": 630 + }, + { + "epoch": 0.6915068493150685, + "grad_norm": 2.157435894012451, + "learning_rate": 4.303411860276198e-06, + "loss": 1.8777, + "step": 631 + }, + { + "epoch": 0.6926027397260274, + "grad_norm": 2.20766019821167, + "learning_rate": 4.301380991064176e-06, + "loss": 1.9803, + "step": 632 + }, + { + "epoch": 0.6936986301369863, + "grad_norm": 2.173079490661621, + "learning_rate": 4.299350121852153e-06, + "loss": 1.9541, + "step": 633 + }, + { + "epoch": 0.6947945205479452, + "grad_norm": 1.9727106094360352, + "learning_rate": 4.29731925264013e-06, + "loss": 1.9766, + "step": 634 + }, + { + "epoch": 0.6958904109589041, + "grad_norm": 2.268538475036621, + "learning_rate": 4.2952883834281075e-06, + "loss": 1.9737, + "step": 635 + }, + { + "epoch": 0.6969863013698631, + "grad_norm": 2.2118406295776367, + "learning_rate": 4.293257514216085e-06, + "loss": 1.9291, + "step": 636 + }, + { + "epoch": 0.6980821917808219, + "grad_norm": 2.2939369678497314, + "learning_rate": 4.291226645004062e-06, + "loss": 1.8868, + "step": 637 + }, + { + "epoch": 0.6991780821917808, + "grad_norm": 2.4940083026885986, + "learning_rate": 4.289195775792039e-06, + "loss": 1.9694, + "step": 638 + }, + { + "epoch": 0.7002739726027397, + "grad_norm": 2.1988351345062256, + "learning_rate": 4.287164906580017e-06, + "loss": 1.9781, + "step": 639 + }, + { + "epoch": 0.7013698630136986, + "grad_norm": 2.099578619003296, + "learning_rate": 4.285134037367994e-06, + "loss": 1.9145, + "step": 640 + }, + { + "epoch": 0.7013698630136986, + "eval_loss": 2.07913875579834, + "eval_runtime": 37.3458, + "eval_samples_per_second": 26.777, + "eval_steps_per_second": 1.125, + "step": 640 + }, + { + "epoch": 0.7024657534246576, + "grad_norm": 2.2437307834625244, + "learning_rate": 4.283103168155971e-06, + "loss": 1.9075, + "step": 641 + }, + { + "epoch": 0.7035616438356165, + "grad_norm": 2.320880174636841, + "learning_rate": 4.281072298943948e-06, + "loss": 1.8929, + "step": 642 + }, + { + "epoch": 0.7046575342465753, + "grad_norm": 2.1174306869506836, + "learning_rate": 4.279041429731926e-06, + "loss": 1.8872, + "step": 643 + }, + { + "epoch": 0.7057534246575342, + "grad_norm": 2.2826361656188965, + "learning_rate": 4.277010560519903e-06, + "loss": 1.9287, + "step": 644 + }, + { + "epoch": 0.7068493150684931, + "grad_norm": 2.7543559074401855, + "learning_rate": 4.27497969130788e-06, + "loss": 1.9119, + "step": 645 + }, + { + "epoch": 0.707945205479452, + "grad_norm": 2.2582201957702637, + "learning_rate": 4.2729488220958575e-06, + "loss": 1.9118, + "step": 646 + }, + { + "epoch": 0.709041095890411, + "grad_norm": 2.3121371269226074, + "learning_rate": 4.270917952883835e-06, + "loss": 1.959, + "step": 647 + }, + { + "epoch": 0.7101369863013699, + "grad_norm": 2.419102430343628, + "learning_rate": 4.268887083671812e-06, + "loss": 1.9679, + "step": 648 + }, + { + "epoch": 0.7112328767123288, + "grad_norm": 2.285771369934082, + "learning_rate": 4.266856214459789e-06, + "loss": 1.8939, + "step": 649 + }, + { + "epoch": 0.7123287671232876, + "grad_norm": 2.193603754043579, + "learning_rate": 4.264825345247766e-06, + "loss": 2.0097, + "step": 650 + }, + { + "epoch": 0.7123287671232876, + "eval_loss": 2.0785164833068848, + "eval_runtime": 37.2595, + "eval_samples_per_second": 26.839, + "eval_steps_per_second": 1.127, + "step": 650 + }, + { + "epoch": 0.7134246575342466, + "grad_norm": 2.2645511627197266, + "learning_rate": 4.262794476035744e-06, + "loss": 1.8865, + "step": 651 + }, + { + "epoch": 0.7145205479452055, + "grad_norm": 2.5553910732269287, + "learning_rate": 4.26076360682372e-06, + "loss": 1.9331, + "step": 652 + }, + { + "epoch": 0.7156164383561644, + "grad_norm": 2.2770845890045166, + "learning_rate": 4.2587327376116984e-06, + "loss": 1.847, + "step": 653 + }, + { + "epoch": 0.7167123287671233, + "grad_norm": 2.1486358642578125, + "learning_rate": 4.256701868399675e-06, + "loss": 1.9058, + "step": 654 + }, + { + "epoch": 0.7178082191780822, + "grad_norm": 2.168975353240967, + "learning_rate": 4.254670999187653e-06, + "loss": 1.9408, + "step": 655 + }, + { + "epoch": 0.718904109589041, + "grad_norm": 2.2156167030334473, + "learning_rate": 4.252640129975629e-06, + "loss": 1.932, + "step": 656 + }, + { + "epoch": 0.72, + "grad_norm": 2.148214101791382, + "learning_rate": 4.2506092607636075e-06, + "loss": 1.9327, + "step": 657 + }, + { + "epoch": 0.7210958904109589, + "grad_norm": 2.3860998153686523, + "learning_rate": 4.248578391551584e-06, + "loss": 1.9191, + "step": 658 + }, + { + "epoch": 0.7221917808219178, + "grad_norm": 2.3917558193206787, + "learning_rate": 4.246547522339562e-06, + "loss": 1.8607, + "step": 659 + }, + { + "epoch": 0.7232876712328767, + "grad_norm": 2.3024795055389404, + "learning_rate": 4.2445166531275385e-06, + "loss": 1.9083, + "step": 660 + }, + { + "epoch": 0.7232876712328767, + "eval_loss": 2.07985782623291, + "eval_runtime": 37.1583, + "eval_samples_per_second": 26.912, + "eval_steps_per_second": 1.13, + "step": 660 + }, + { + "epoch": 0.7243835616438357, + "grad_norm": 2.261615037918091, + "learning_rate": 4.242485783915517e-06, + "loss": 1.9155, + "step": 661 + }, + { + "epoch": 0.7254794520547945, + "grad_norm": 2.190124034881592, + "learning_rate": 4.240454914703493e-06, + "loss": 1.858, + "step": 662 + }, + { + "epoch": 0.7265753424657534, + "grad_norm": 2.3396036624908447, + "learning_rate": 4.23842404549147e-06, + "loss": 1.878, + "step": 663 + }, + { + "epoch": 0.7276712328767123, + "grad_norm": 2.198410987854004, + "learning_rate": 4.236393176279448e-06, + "loss": 1.8755, + "step": 664 + }, + { + "epoch": 0.7287671232876712, + "grad_norm": 2.108875274658203, + "learning_rate": 4.234362307067425e-06, + "loss": 1.8709, + "step": 665 + }, + { + "epoch": 0.7298630136986302, + "grad_norm": 2.0908193588256836, + "learning_rate": 4.232331437855402e-06, + "loss": 1.9395, + "step": 666 + }, + { + "epoch": 0.7309589041095891, + "grad_norm": 2.208134651184082, + "learning_rate": 4.2303005686433795e-06, + "loss": 1.9675, + "step": 667 + }, + { + "epoch": 0.7320547945205479, + "grad_norm": 2.2779526710510254, + "learning_rate": 4.228269699431357e-06, + "loss": 1.9661, + "step": 668 + }, + { + "epoch": 0.7331506849315068, + "grad_norm": 2.053279161453247, + "learning_rate": 4.226238830219334e-06, + "loss": 1.9244, + "step": 669 + }, + { + "epoch": 0.7342465753424657, + "grad_norm": 2.2060821056365967, + "learning_rate": 4.224207961007311e-06, + "loss": 1.9252, + "step": 670 + }, + { + "epoch": 0.7342465753424657, + "eval_loss": 2.0804648399353027, + "eval_runtime": 37.1852, + "eval_samples_per_second": 26.892, + "eval_steps_per_second": 1.129, + "step": 670 + }, + { + "epoch": 0.7353424657534247, + "grad_norm": 2.1996090412139893, + "learning_rate": 4.2221770917952886e-06, + "loss": 1.926, + "step": 671 + }, + { + "epoch": 0.7364383561643836, + "grad_norm": 2.0817034244537354, + "learning_rate": 4.220146222583266e-06, + "loss": 1.9123, + "step": 672 + }, + { + "epoch": 0.7375342465753425, + "grad_norm": 2.2505035400390625, + "learning_rate": 4.218115353371243e-06, + "loss": 1.9369, + "step": 673 + }, + { + "epoch": 0.7386301369863014, + "grad_norm": 2.2053000926971436, + "learning_rate": 4.21608448415922e-06, + "loss": 1.9225, + "step": 674 + }, + { + "epoch": 0.7397260273972602, + "grad_norm": 2.039905071258545, + "learning_rate": 4.214053614947198e-06, + "loss": 1.8876, + "step": 675 + }, + { + "epoch": 0.7408219178082192, + "grad_norm": 2.047445297241211, + "learning_rate": 4.212022745735175e-06, + "loss": 1.9724, + "step": 676 + }, + { + "epoch": 0.7419178082191781, + "grad_norm": 2.277803897857666, + "learning_rate": 4.209991876523152e-06, + "loss": 1.8687, + "step": 677 + }, + { + "epoch": 0.743013698630137, + "grad_norm": 2.0048105716705322, + "learning_rate": 4.2079610073111295e-06, + "loss": 1.9349, + "step": 678 + }, + { + "epoch": 0.7441095890410959, + "grad_norm": 2.121614456176758, + "learning_rate": 4.205930138099107e-06, + "loss": 1.8723, + "step": 679 + }, + { + "epoch": 0.7452054794520548, + "grad_norm": 2.169165849685669, + "learning_rate": 4.203899268887084e-06, + "loss": 1.9713, + "step": 680 + }, + { + "epoch": 0.7452054794520548, + "eval_loss": 2.0704824924468994, + "eval_runtime": 37.2015, + "eval_samples_per_second": 26.881, + "eval_steps_per_second": 1.129, + "step": 680 + }, + { + "epoch": 0.7463013698630137, + "grad_norm": 2.004974603652954, + "learning_rate": 4.201868399675061e-06, + "loss": 1.9237, + "step": 681 + }, + { + "epoch": 0.7473972602739726, + "grad_norm": 2.042402505874634, + "learning_rate": 4.199837530463039e-06, + "loss": 1.9686, + "step": 682 + }, + { + "epoch": 0.7484931506849315, + "grad_norm": 2.152040719985962, + "learning_rate": 4.197806661251016e-06, + "loss": 1.8498, + "step": 683 + }, + { + "epoch": 0.7495890410958904, + "grad_norm": 2.116483211517334, + "learning_rate": 4.195775792038993e-06, + "loss": 1.9241, + "step": 684 + }, + { + "epoch": 0.7506849315068493, + "grad_norm": 2.1628384590148926, + "learning_rate": 4.19374492282697e-06, + "loss": 1.9019, + "step": 685 + }, + { + "epoch": 0.7517808219178083, + "grad_norm": 2.1851415634155273, + "learning_rate": 4.191714053614948e-06, + "loss": 1.836, + "step": 686 + }, + { + "epoch": 0.7528767123287671, + "grad_norm": 2.173215866088867, + "learning_rate": 4.189683184402924e-06, + "loss": 1.9412, + "step": 687 + }, + { + "epoch": 0.753972602739726, + "grad_norm": 2.1002466678619385, + "learning_rate": 4.187652315190902e-06, + "loss": 1.8871, + "step": 688 + }, + { + "epoch": 0.7550684931506849, + "grad_norm": 2.1932122707366943, + "learning_rate": 4.185621445978879e-06, + "loss": 1.9671, + "step": 689 + }, + { + "epoch": 0.7561643835616438, + "grad_norm": 2.2915258407592773, + "learning_rate": 4.183590576766857e-06, + "loss": 1.8957, + "step": 690 + }, + { + "epoch": 0.7561643835616438, + "eval_loss": 2.067124128341675, + "eval_runtime": 37.2362, + "eval_samples_per_second": 26.856, + "eval_steps_per_second": 1.128, + "step": 690 + }, + { + "epoch": 0.7572602739726028, + "grad_norm": 2.060879707336426, + "learning_rate": 4.181559707554833e-06, + "loss": 1.9091, + "step": 691 + }, + { + "epoch": 0.7583561643835617, + "grad_norm": 2.031156539916992, + "learning_rate": 4.179528838342811e-06, + "loss": 1.9606, + "step": 692 + }, + { + "epoch": 0.7594520547945205, + "grad_norm": 2.121628999710083, + "learning_rate": 4.177497969130788e-06, + "loss": 1.88, + "step": 693 + }, + { + "epoch": 0.7605479452054794, + "grad_norm": 2.2009618282318115, + "learning_rate": 4.175467099918766e-06, + "loss": 1.8916, + "step": 694 + }, + { + "epoch": 0.7616438356164383, + "grad_norm": 2.124082326889038, + "learning_rate": 4.173436230706742e-06, + "loss": 1.9805, + "step": 695 + }, + { + "epoch": 0.7627397260273973, + "grad_norm": 2.3914906978607178, + "learning_rate": 4.1714053614947205e-06, + "loss": 1.9052, + "step": 696 + }, + { + "epoch": 0.7638356164383562, + "grad_norm": 2.2398853302001953, + "learning_rate": 4.169374492282697e-06, + "loss": 1.8965, + "step": 697 + }, + { + "epoch": 0.7649315068493151, + "grad_norm": 2.224860191345215, + "learning_rate": 4.167343623070675e-06, + "loss": 1.8962, + "step": 698 + }, + { + "epoch": 0.766027397260274, + "grad_norm": 2.3356244564056396, + "learning_rate": 4.165312753858652e-06, + "loss": 1.9543, + "step": 699 + }, + { + "epoch": 0.7671232876712328, + "grad_norm": 2.1308751106262207, + "learning_rate": 4.163281884646629e-06, + "loss": 1.9095, + "step": 700 + }, + { + "epoch": 0.7671232876712328, + "eval_loss": 2.066145896911621, + "eval_runtime": 37.2629, + "eval_samples_per_second": 26.836, + "eval_steps_per_second": 1.127, + "step": 700 + }, + { + "epoch": 0.7682191780821918, + "grad_norm": 2.1655938625335693, + "learning_rate": 4.161251015434607e-06, + "loss": 1.8923, + "step": 701 + }, + { + "epoch": 0.7693150684931507, + "grad_norm": 2.1985316276550293, + "learning_rate": 4.159220146222583e-06, + "loss": 1.9216, + "step": 702 + }, + { + "epoch": 0.7704109589041096, + "grad_norm": 2.1537251472473145, + "learning_rate": 4.157189277010561e-06, + "loss": 1.879, + "step": 703 + }, + { + "epoch": 0.7715068493150685, + "grad_norm": 2.1867849826812744, + "learning_rate": 4.155158407798538e-06, + "loss": 1.905, + "step": 704 + }, + { + "epoch": 0.7726027397260274, + "grad_norm": 2.342320203781128, + "learning_rate": 4.153127538586516e-06, + "loss": 1.9403, + "step": 705 + }, + { + "epoch": 0.7736986301369863, + "grad_norm": 2.2250113487243652, + "learning_rate": 4.151096669374492e-06, + "loss": 1.9109, + "step": 706 + }, + { + "epoch": 0.7747945205479452, + "grad_norm": 2.179103374481201, + "learning_rate": 4.1490658001624705e-06, + "loss": 1.9288, + "step": 707 + }, + { + "epoch": 0.7758904109589041, + "grad_norm": 2.3618645668029785, + "learning_rate": 4.147034930950447e-06, + "loss": 1.8982, + "step": 708 + }, + { + "epoch": 0.776986301369863, + "grad_norm": 2.150402307510376, + "learning_rate": 4.145004061738425e-06, + "loss": 1.9, + "step": 709 + }, + { + "epoch": 0.7780821917808219, + "grad_norm": 2.295571804046631, + "learning_rate": 4.1429731925264015e-06, + "loss": 1.9003, + "step": 710 + }, + { + "epoch": 0.7780821917808219, + "eval_loss": 2.067948341369629, + "eval_runtime": 37.2379, + "eval_samples_per_second": 26.854, + "eval_steps_per_second": 1.128, + "step": 710 + }, + { + "epoch": 0.7791780821917809, + "grad_norm": 2.3279545307159424, + "learning_rate": 4.140942323314379e-06, + "loss": 1.8307, + "step": 711 + }, + { + "epoch": 0.7802739726027397, + "grad_norm": 2.3993380069732666, + "learning_rate": 4.138911454102356e-06, + "loss": 1.8755, + "step": 712 + }, + { + "epoch": 0.7813698630136986, + "grad_norm": 2.139653444290161, + "learning_rate": 4.136880584890333e-06, + "loss": 1.8914, + "step": 713 + }, + { + "epoch": 0.7824657534246575, + "grad_norm": 2.239544153213501, + "learning_rate": 4.1348497156783106e-06, + "loss": 1.9289, + "step": 714 + }, + { + "epoch": 0.7835616438356164, + "grad_norm": 2.2039642333984375, + "learning_rate": 4.132818846466288e-06, + "loss": 1.9237, + "step": 715 + }, + { + "epoch": 0.7846575342465754, + "grad_norm": 2.1453804969787598, + "learning_rate": 4.130787977254265e-06, + "loss": 1.9291, + "step": 716 + }, + { + "epoch": 0.7857534246575343, + "grad_norm": 2.2350032329559326, + "learning_rate": 4.128757108042242e-06, + "loss": 1.9413, + "step": 717 + }, + { + "epoch": 0.7868493150684932, + "grad_norm": 2.1718318462371826, + "learning_rate": 4.12672623883022e-06, + "loss": 1.9318, + "step": 718 + }, + { + "epoch": 0.787945205479452, + "grad_norm": 2.1860601902008057, + "learning_rate": 4.124695369618197e-06, + "loss": 1.9934, + "step": 719 + }, + { + "epoch": 0.7890410958904109, + "grad_norm": 2.205166816711426, + "learning_rate": 4.122664500406174e-06, + "loss": 1.9038, + "step": 720 + }, + { + "epoch": 0.7890410958904109, + "eval_loss": 2.06229829788208, + "eval_runtime": 37.6121, + "eval_samples_per_second": 26.587, + "eval_steps_per_second": 1.117, + "step": 720 + }, + { + "epoch": 0.7901369863013699, + "grad_norm": 2.2094533443450928, + "learning_rate": 4.1206336311941515e-06, + "loss": 1.906, + "step": 721 + }, + { + "epoch": 0.7912328767123288, + "grad_norm": 2.178891658782959, + "learning_rate": 4.118602761982129e-06, + "loss": 1.936, + "step": 722 + }, + { + "epoch": 0.7923287671232877, + "grad_norm": 2.051077365875244, + "learning_rate": 4.116571892770106e-06, + "loss": 1.8736, + "step": 723 + }, + { + "epoch": 0.7934246575342466, + "grad_norm": 2.2430427074432373, + "learning_rate": 4.114541023558083e-06, + "loss": 1.8396, + "step": 724 + }, + { + "epoch": 0.7945205479452054, + "grad_norm": 2.2255303859710693, + "learning_rate": 4.112510154346061e-06, + "loss": 1.9668, + "step": 725 + }, + { + "epoch": 0.7956164383561644, + "grad_norm": 2.280313491821289, + "learning_rate": 4.110479285134038e-06, + "loss": 1.8945, + "step": 726 + }, + { + "epoch": 0.7967123287671233, + "grad_norm": 2.2516801357269287, + "learning_rate": 4.108448415922015e-06, + "loss": 1.9157, + "step": 727 + }, + { + "epoch": 0.7978082191780822, + "grad_norm": 2.1379809379577637, + "learning_rate": 4.1064175467099924e-06, + "loss": 1.935, + "step": 728 + }, + { + "epoch": 0.7989041095890411, + "grad_norm": 2.149425983428955, + "learning_rate": 4.10438667749797e-06, + "loss": 1.935, + "step": 729 + }, + { + "epoch": 0.8, + "grad_norm": 2.275270700454712, + "learning_rate": 4.102355808285947e-06, + "loss": 1.7889, + "step": 730 + }, + { + "epoch": 0.8, + "eval_loss": 2.0633974075317383, + "eval_runtime": 37.6895, + "eval_samples_per_second": 26.533, + "eval_steps_per_second": 1.114, + "step": 730 + }, + { + "epoch": 0.8010958904109589, + "grad_norm": 2.1550278663635254, + "learning_rate": 4.100324939073924e-06, + "loss": 1.9833, + "step": 731 + }, + { + "epoch": 0.8021917808219178, + "grad_norm": 2.3050060272216797, + "learning_rate": 4.0982940698619015e-06, + "loss": 1.8969, + "step": 732 + }, + { + "epoch": 0.8032876712328767, + "grad_norm": 2.4548778533935547, + "learning_rate": 4.096263200649879e-06, + "loss": 1.8945, + "step": 733 + }, + { + "epoch": 0.8043835616438356, + "grad_norm": 2.2611639499664307, + "learning_rate": 4.094232331437856e-06, + "loss": 1.9071, + "step": 734 + }, + { + "epoch": 0.8054794520547945, + "grad_norm": 2.1806774139404297, + "learning_rate": 4.092201462225833e-06, + "loss": 1.8787, + "step": 735 + }, + { + "epoch": 0.8065753424657535, + "grad_norm": 2.4374301433563232, + "learning_rate": 4.090170593013811e-06, + "loss": 1.88, + "step": 736 + }, + { + "epoch": 0.8076712328767123, + "grad_norm": 2.328333854675293, + "learning_rate": 4.088139723801787e-06, + "loss": 1.9178, + "step": 737 + }, + { + "epoch": 0.8087671232876712, + "grad_norm": 2.33552885055542, + "learning_rate": 4.086108854589765e-06, + "loss": 1.8587, + "step": 738 + }, + { + "epoch": 0.8098630136986301, + "grad_norm": 2.3472790718078613, + "learning_rate": 4.084077985377742e-06, + "loss": 1.877, + "step": 739 + }, + { + "epoch": 0.810958904109589, + "grad_norm": 2.307123899459839, + "learning_rate": 4.08204711616572e-06, + "loss": 1.8888, + "step": 740 + }, + { + "epoch": 0.810958904109589, + "eval_loss": 2.065826892852783, + "eval_runtime": 37.2747, + "eval_samples_per_second": 26.828, + "eval_steps_per_second": 1.127, + "step": 740 + }, + { + "epoch": 0.812054794520548, + "grad_norm": 2.3688747882843018, + "learning_rate": 4.080016246953696e-06, + "loss": 1.9571, + "step": 741 + }, + { + "epoch": 0.8131506849315069, + "grad_norm": 2.385007381439209, + "learning_rate": 4.077985377741674e-06, + "loss": 1.8657, + "step": 742 + }, + { + "epoch": 0.8142465753424658, + "grad_norm": 2.156766414642334, + "learning_rate": 4.075954508529651e-06, + "loss": 1.9421, + "step": 743 + }, + { + "epoch": 0.8153424657534246, + "grad_norm": 2.2074286937713623, + "learning_rate": 4.073923639317629e-06, + "loss": 1.9396, + "step": 744 + }, + { + "epoch": 0.8164383561643835, + "grad_norm": 2.185458183288574, + "learning_rate": 4.071892770105605e-06, + "loss": 1.9612, + "step": 745 + }, + { + "epoch": 0.8175342465753425, + "grad_norm": 2.324209451675415, + "learning_rate": 4.069861900893583e-06, + "loss": 1.892, + "step": 746 + }, + { + "epoch": 0.8186301369863014, + "grad_norm": 2.1312127113342285, + "learning_rate": 4.06783103168156e-06, + "loss": 1.8522, + "step": 747 + }, + { + "epoch": 0.8197260273972603, + "grad_norm": 2.0959460735321045, + "learning_rate": 4.065800162469538e-06, + "loss": 1.8732, + "step": 748 + }, + { + "epoch": 0.8208219178082192, + "grad_norm": 2.085519552230835, + "learning_rate": 4.063769293257514e-06, + "loss": 1.8832, + "step": 749 + }, + { + "epoch": 0.821917808219178, + "grad_norm": 2.310145854949951, + "learning_rate": 4.061738424045492e-06, + "loss": 1.919, + "step": 750 + }, + { + "epoch": 0.821917808219178, + "eval_loss": 2.0637967586517334, + "eval_runtime": 37.2246, + "eval_samples_per_second": 26.864, + "eval_steps_per_second": 1.128, + "step": 750 + }, + { + "epoch": 0.823013698630137, + "grad_norm": 2.3171160221099854, + "learning_rate": 4.059707554833469e-06, + "loss": 1.8874, + "step": 751 + }, + { + "epoch": 0.8241095890410959, + "grad_norm": 2.1065480709075928, + "learning_rate": 4.057676685621446e-06, + "loss": 1.9064, + "step": 752 + }, + { + "epoch": 0.8252054794520548, + "grad_norm": 2.0539677143096924, + "learning_rate": 4.0556458164094235e-06, + "loss": 1.9376, + "step": 753 + }, + { + "epoch": 0.8263013698630137, + "grad_norm": 2.4440360069274902, + "learning_rate": 4.053614947197401e-06, + "loss": 1.9129, + "step": 754 + }, + { + "epoch": 0.8273972602739726, + "grad_norm": 2.2324604988098145, + "learning_rate": 4.051584077985378e-06, + "loss": 1.8281, + "step": 755 + }, + { + "epoch": 0.8284931506849315, + "grad_norm": 2.262627363204956, + "learning_rate": 4.049553208773355e-06, + "loss": 1.9036, + "step": 756 + }, + { + "epoch": 0.8295890410958904, + "grad_norm": 2.2473418712615967, + "learning_rate": 4.047522339561333e-06, + "loss": 1.9455, + "step": 757 + }, + { + "epoch": 0.8306849315068493, + "grad_norm": 2.473444938659668, + "learning_rate": 4.04549147034931e-06, + "loss": 1.9003, + "step": 758 + }, + { + "epoch": 0.8317808219178082, + "grad_norm": 2.3454110622406006, + "learning_rate": 4.043460601137287e-06, + "loss": 1.8941, + "step": 759 + }, + { + "epoch": 0.8328767123287671, + "grad_norm": 2.4391231536865234, + "learning_rate": 4.041429731925264e-06, + "loss": 2.0146, + "step": 760 + }, + { + "epoch": 0.8328767123287671, + "eval_loss": 2.0618460178375244, + "eval_runtime": 37.1704, + "eval_samples_per_second": 26.903, + "eval_steps_per_second": 1.13, + "step": 760 + }, + { + "epoch": 0.8339726027397261, + "grad_norm": 2.3129498958587646, + "learning_rate": 4.039398862713242e-06, + "loss": 1.9, + "step": 761 + }, + { + "epoch": 0.8350684931506849, + "grad_norm": 2.3766934871673584, + "learning_rate": 4.037367993501219e-06, + "loss": 1.9017, + "step": 762 + }, + { + "epoch": 0.8361643835616438, + "grad_norm": 2.1698460578918457, + "learning_rate": 4.035337124289196e-06, + "loss": 1.9657, + "step": 763 + }, + { + "epoch": 0.8372602739726027, + "grad_norm": 2.384904146194458, + "learning_rate": 4.0333062550771735e-06, + "loss": 1.8867, + "step": 764 + }, + { + "epoch": 0.8383561643835616, + "grad_norm": 2.572391986846924, + "learning_rate": 4.031275385865151e-06, + "loss": 1.8561, + "step": 765 + }, + { + "epoch": 0.8394520547945206, + "grad_norm": 2.216606616973877, + "learning_rate": 4.029244516653128e-06, + "loss": 1.8926, + "step": 766 + }, + { + "epoch": 0.8405479452054795, + "grad_norm": 2.215756893157959, + "learning_rate": 4.027213647441105e-06, + "loss": 1.928, + "step": 767 + }, + { + "epoch": 0.8416438356164384, + "grad_norm": 2.2713284492492676, + "learning_rate": 4.025182778229083e-06, + "loss": 1.8286, + "step": 768 + }, + { + "epoch": 0.8427397260273972, + "grad_norm": 2.341747283935547, + "learning_rate": 4.02315190901706e-06, + "loss": 1.8789, + "step": 769 + }, + { + "epoch": 0.8438356164383561, + "grad_norm": 2.1485848426818848, + "learning_rate": 4.021121039805037e-06, + "loss": 1.8998, + "step": 770 + }, + { + "epoch": 0.8438356164383561, + "eval_loss": 2.0596234798431396, + "eval_runtime": 37.1854, + "eval_samples_per_second": 26.892, + "eval_steps_per_second": 1.129, + "step": 770 + }, + { + "epoch": 0.8449315068493151, + "grad_norm": 2.234192132949829, + "learning_rate": 4.0190901705930144e-06, + "loss": 1.8995, + "step": 771 + }, + { + "epoch": 0.846027397260274, + "grad_norm": 2.289328098297119, + "learning_rate": 4.017059301380992e-06, + "loss": 1.9882, + "step": 772 + }, + { + "epoch": 0.8471232876712329, + "grad_norm": 2.1001455783843994, + "learning_rate": 4.015028432168969e-06, + "loss": 1.8706, + "step": 773 + }, + { + "epoch": 0.8482191780821918, + "grad_norm": 2.2200522422790527, + "learning_rate": 4.012997562956945e-06, + "loss": 1.9443, + "step": 774 + }, + { + "epoch": 0.8493150684931506, + "grad_norm": 2.2540805339813232, + "learning_rate": 4.0109666937449235e-06, + "loss": 1.8552, + "step": 775 + }, + { + "epoch": 0.8504109589041096, + "grad_norm": 2.1214053630828857, + "learning_rate": 4.0089358245329e-06, + "loss": 1.8741, + "step": 776 + }, + { + "epoch": 0.8515068493150685, + "grad_norm": 2.0221869945526123, + "learning_rate": 4.006904955320878e-06, + "loss": 1.8798, + "step": 777 + }, + { + "epoch": 0.8526027397260274, + "grad_norm": 2.217261791229248, + "learning_rate": 4.0048740861088545e-06, + "loss": 1.8542, + "step": 778 + }, + { + "epoch": 0.8536986301369863, + "grad_norm": 2.099202871322632, + "learning_rate": 4.002843216896833e-06, + "loss": 1.9342, + "step": 779 + }, + { + "epoch": 0.8547945205479452, + "grad_norm": 2.139526844024658, + "learning_rate": 4.000812347684809e-06, + "loss": 1.9252, + "step": 780 + }, + { + "epoch": 0.8547945205479452, + "eval_loss": 2.054051399230957, + "eval_runtime": 37.1125, + "eval_samples_per_second": 26.945, + "eval_steps_per_second": 1.132, + "step": 780 + }, + { + "epoch": 0.8558904109589041, + "grad_norm": 2.2231476306915283, + "learning_rate": 3.998781478472787e-06, + "loss": 1.8671, + "step": 781 + }, + { + "epoch": 0.856986301369863, + "grad_norm": 2.143979072570801, + "learning_rate": 3.996750609260764e-06, + "loss": 1.8759, + "step": 782 + }, + { + "epoch": 0.8580821917808219, + "grad_norm": 2.1353135108947754, + "learning_rate": 3.994719740048742e-06, + "loss": 1.8559, + "step": 783 + }, + { + "epoch": 0.8591780821917808, + "grad_norm": 2.222564697265625, + "learning_rate": 3.992688870836718e-06, + "loss": 1.999, + "step": 784 + }, + { + "epoch": 0.8602739726027397, + "grad_norm": 1.958031177520752, + "learning_rate": 3.990658001624696e-06, + "loss": 1.9019, + "step": 785 + }, + { + "epoch": 0.8613698630136987, + "grad_norm": 2.2048962116241455, + "learning_rate": 3.988627132412673e-06, + "loss": 1.8561, + "step": 786 + }, + { + "epoch": 0.8624657534246576, + "grad_norm": 2.150332450866699, + "learning_rate": 3.98659626320065e-06, + "loss": 1.9399, + "step": 787 + }, + { + "epoch": 0.8635616438356164, + "grad_norm": 2.182742118835449, + "learning_rate": 3.984565393988627e-06, + "loss": 1.8933, + "step": 788 + }, + { + "epoch": 0.8646575342465753, + "grad_norm": 2.19435715675354, + "learning_rate": 3.9825345247766046e-06, + "loss": 1.8918, + "step": 789 + }, + { + "epoch": 0.8657534246575342, + "grad_norm": 2.3382859230041504, + "learning_rate": 3.980503655564582e-06, + "loss": 1.8937, + "step": 790 + }, + { + "epoch": 0.8657534246575342, + "eval_loss": 2.051997423171997, + "eval_runtime": 37.1866, + "eval_samples_per_second": 26.891, + "eval_steps_per_second": 1.129, + "step": 790 + }, + { + "epoch": 0.8668493150684932, + "grad_norm": 2.2700726985931396, + "learning_rate": 3.978472786352559e-06, + "loss": 1.9148, + "step": 791 + }, + { + "epoch": 0.8679452054794521, + "grad_norm": 2.2998759746551514, + "learning_rate": 3.976441917140536e-06, + "loss": 1.9268, + "step": 792 + }, + { + "epoch": 0.869041095890411, + "grad_norm": 2.2367377281188965, + "learning_rate": 3.974411047928514e-06, + "loss": 1.901, + "step": 793 + }, + { + "epoch": 0.8701369863013698, + "grad_norm": 2.0868773460388184, + "learning_rate": 3.972380178716491e-06, + "loss": 1.9677, + "step": 794 + }, + { + "epoch": 0.8712328767123287, + "grad_norm": 2.183732509613037, + "learning_rate": 3.970349309504468e-06, + "loss": 1.9271, + "step": 795 + }, + { + "epoch": 0.8723287671232877, + "grad_norm": 2.187282085418701, + "learning_rate": 3.9683184402924455e-06, + "loss": 1.9437, + "step": 796 + }, + { + "epoch": 0.8734246575342466, + "grad_norm": 2.1906070709228516, + "learning_rate": 3.966287571080423e-06, + "loss": 1.8437, + "step": 797 + }, + { + "epoch": 0.8745205479452055, + "grad_norm": 2.16979718208313, + "learning_rate": 3.9642567018684e-06, + "loss": 1.829, + "step": 798 + }, + { + "epoch": 0.8756164383561644, + "grad_norm": 2.2387869358062744, + "learning_rate": 3.962225832656377e-06, + "loss": 1.9143, + "step": 799 + }, + { + "epoch": 0.8767123287671232, + "grad_norm": 2.208221197128296, + "learning_rate": 3.960194963444355e-06, + "loss": 1.8894, + "step": 800 + }, + { + "epoch": 0.8767123287671232, + "eval_loss": 2.047945976257324, + "eval_runtime": 37.1778, + "eval_samples_per_second": 26.898, + "eval_steps_per_second": 1.13, + "step": 800 + }, + { + "epoch": 0.8778082191780822, + "grad_norm": 2.335108518600464, + "learning_rate": 3.958164094232332e-06, + "loss": 1.8155, + "step": 801 + }, + { + "epoch": 0.8789041095890411, + "grad_norm": 2.168006181716919, + "learning_rate": 3.956133225020309e-06, + "loss": 1.8919, + "step": 802 + }, + { + "epoch": 0.88, + "grad_norm": 2.3029897212982178, + "learning_rate": 3.954102355808286e-06, + "loss": 1.8996, + "step": 803 + }, + { + "epoch": 0.8810958904109589, + "grad_norm": 2.156141519546509, + "learning_rate": 3.952071486596264e-06, + "loss": 1.8965, + "step": 804 + }, + { + "epoch": 0.8821917808219178, + "grad_norm": 2.1522068977355957, + "learning_rate": 3.950040617384241e-06, + "loss": 1.8922, + "step": 805 + }, + { + "epoch": 0.8832876712328767, + "grad_norm": 2.153430700302124, + "learning_rate": 3.948009748172218e-06, + "loss": 1.8839, + "step": 806 + }, + { + "epoch": 0.8843835616438356, + "grad_norm": 2.1891674995422363, + "learning_rate": 3.9459788789601955e-06, + "loss": 1.8756, + "step": 807 + }, + { + "epoch": 0.8854794520547945, + "grad_norm": 2.1092092990875244, + "learning_rate": 3.943948009748173e-06, + "loss": 1.9452, + "step": 808 + }, + { + "epoch": 0.8865753424657534, + "grad_norm": 2.0465025901794434, + "learning_rate": 3.94191714053615e-06, + "loss": 1.9067, + "step": 809 + }, + { + "epoch": 0.8876712328767123, + "grad_norm": 2.1150474548339844, + "learning_rate": 3.939886271324127e-06, + "loss": 1.8874, + "step": 810 + }, + { + "epoch": 0.8876712328767123, + "eval_loss": 2.046292304992676, + "eval_runtime": 37.2421, + "eval_samples_per_second": 26.851, + "eval_steps_per_second": 1.128, + "step": 810 + }, + { + "epoch": 0.8887671232876713, + "grad_norm": 2.039989948272705, + "learning_rate": 3.937855402112104e-06, + "loss": 1.9212, + "step": 811 + }, + { + "epoch": 0.8898630136986302, + "grad_norm": 2.17354679107666, + "learning_rate": 3.935824532900082e-06, + "loss": 1.8473, + "step": 812 + }, + { + "epoch": 0.890958904109589, + "grad_norm": 2.3101940155029297, + "learning_rate": 3.933793663688058e-06, + "loss": 1.8413, + "step": 813 + }, + { + "epoch": 0.8920547945205479, + "grad_norm": 2.1300928592681885, + "learning_rate": 3.9317627944760365e-06, + "loss": 1.9126, + "step": 814 + }, + { + "epoch": 0.8931506849315068, + "grad_norm": 2.2317357063293457, + "learning_rate": 3.929731925264013e-06, + "loss": 1.938, + "step": 815 + }, + { + "epoch": 0.8942465753424658, + "grad_norm": 2.1097419261932373, + "learning_rate": 3.927701056051991e-06, + "loss": 1.8426, + "step": 816 + }, + { + "epoch": 0.8953424657534247, + "grad_norm": 2.1709141731262207, + "learning_rate": 3.9256701868399674e-06, + "loss": 1.8779, + "step": 817 + }, + { + "epoch": 0.8964383561643836, + "grad_norm": 2.104193925857544, + "learning_rate": 3.9236393176279456e-06, + "loss": 1.8736, + "step": 818 + }, + { + "epoch": 0.8975342465753424, + "grad_norm": 2.094637632369995, + "learning_rate": 3.921608448415922e-06, + "loss": 1.8556, + "step": 819 + }, + { + "epoch": 0.8986301369863013, + "grad_norm": 2.217918872833252, + "learning_rate": 3.9195775792039e-06, + "loss": 1.8429, + "step": 820 + }, + { + "epoch": 0.8986301369863013, + "eval_loss": 2.046095848083496, + "eval_runtime": 37.5659, + "eval_samples_per_second": 26.62, + "eval_steps_per_second": 1.118, + "step": 820 + }, + { + "epoch": 0.8997260273972603, + "grad_norm": 2.220355272293091, + "learning_rate": 3.9175467099918765e-06, + "loss": 1.8973, + "step": 821 + }, + { + "epoch": 0.9008219178082192, + "grad_norm": 2.0766689777374268, + "learning_rate": 3.915515840779855e-06, + "loss": 1.8728, + "step": 822 + }, + { + "epoch": 0.9019178082191781, + "grad_norm": 2.2046821117401123, + "learning_rate": 3.913484971567831e-06, + "loss": 1.9211, + "step": 823 + }, + { + "epoch": 0.903013698630137, + "grad_norm": 2.286024332046509, + "learning_rate": 3.911454102355808e-06, + "loss": 1.913, + "step": 824 + }, + { + "epoch": 0.9041095890410958, + "grad_norm": 2.2495479583740234, + "learning_rate": 3.909423233143786e-06, + "loss": 1.8214, + "step": 825 + }, + { + "epoch": 0.9052054794520548, + "grad_norm": 2.1300203800201416, + "learning_rate": 3.907392363931763e-06, + "loss": 1.9033, + "step": 826 + }, + { + "epoch": 0.9063013698630137, + "grad_norm": 2.092094898223877, + "learning_rate": 3.90536149471974e-06, + "loss": 1.9007, + "step": 827 + }, + { + "epoch": 0.9073972602739726, + "grad_norm": 2.1571428775787354, + "learning_rate": 3.9033306255077175e-06, + "loss": 1.9606, + "step": 828 + }, + { + "epoch": 0.9084931506849315, + "grad_norm": 2.1052281856536865, + "learning_rate": 3.901299756295695e-06, + "loss": 1.9916, + "step": 829 + }, + { + "epoch": 0.9095890410958904, + "grad_norm": 2.113666534423828, + "learning_rate": 3.899268887083672e-06, + "loss": 2.0136, + "step": 830 + }, + { + "epoch": 0.9095890410958904, + "eval_loss": 2.0450072288513184, + "eval_runtime": 37.6746, + "eval_samples_per_second": 26.543, + "eval_steps_per_second": 1.115, + "step": 830 + }, + { + "epoch": 0.9106849315068493, + "grad_norm": 2.112591028213501, + "learning_rate": 3.897238017871649e-06, + "loss": 1.8784, + "step": 831 + }, + { + "epoch": 0.9117808219178082, + "grad_norm": 2.1584672927856445, + "learning_rate": 3.8952071486596266e-06, + "loss": 1.8876, + "step": 832 + }, + { + "epoch": 0.9128767123287671, + "grad_norm": 2.2744317054748535, + "learning_rate": 3.893176279447604e-06, + "loss": 1.8981, + "step": 833 + }, + { + "epoch": 0.913972602739726, + "grad_norm": 2.21824312210083, + "learning_rate": 3.891145410235581e-06, + "loss": 1.85, + "step": 834 + }, + { + "epoch": 0.915068493150685, + "grad_norm": 2.2203729152679443, + "learning_rate": 3.889114541023558e-06, + "loss": 1.9649, + "step": 835 + }, + { + "epoch": 0.9161643835616439, + "grad_norm": 2.1767687797546387, + "learning_rate": 3.887083671811536e-06, + "loss": 1.9044, + "step": 836 + }, + { + "epoch": 0.9172602739726028, + "grad_norm": 2.0724353790283203, + "learning_rate": 3.885052802599513e-06, + "loss": 1.8787, + "step": 837 + }, + { + "epoch": 0.9183561643835616, + "grad_norm": 2.1283740997314453, + "learning_rate": 3.88302193338749e-06, + "loss": 1.8284, + "step": 838 + }, + { + "epoch": 0.9194520547945205, + "grad_norm": 2.5087451934814453, + "learning_rate": 3.8809910641754675e-06, + "loss": 1.9546, + "step": 839 + }, + { + "epoch": 0.9205479452054794, + "grad_norm": 2.0633344650268555, + "learning_rate": 3.878960194963445e-06, + "loss": 1.8969, + "step": 840 + }, + { + "epoch": 0.9205479452054794, + "eval_loss": 2.047938346862793, + "eval_runtime": 37.2293, + "eval_samples_per_second": 26.861, + "eval_steps_per_second": 1.128, + "step": 840 + }, + { + "epoch": 0.9216438356164384, + "grad_norm": 2.178361654281616, + "learning_rate": 3.876929325751422e-06, + "loss": 1.8855, + "step": 841 + }, + { + "epoch": 0.9227397260273973, + "grad_norm": 2.3223721981048584, + "learning_rate": 3.874898456539399e-06, + "loss": 1.9363, + "step": 842 + }, + { + "epoch": 0.9238356164383562, + "grad_norm": 2.3330960273742676, + "learning_rate": 3.872867587327377e-06, + "loss": 1.8374, + "step": 843 + }, + { + "epoch": 0.924931506849315, + "grad_norm": 2.0584356784820557, + "learning_rate": 3.870836718115354e-06, + "loss": 1.911, + "step": 844 + }, + { + "epoch": 0.9260273972602739, + "grad_norm": 2.1834609508514404, + "learning_rate": 3.868805848903331e-06, + "loss": 1.8645, + "step": 845 + }, + { + "epoch": 0.9271232876712329, + "grad_norm": 2.2576122283935547, + "learning_rate": 3.8667749796913084e-06, + "loss": 1.8662, + "step": 846 + }, + { + "epoch": 0.9282191780821918, + "grad_norm": 2.1271183490753174, + "learning_rate": 3.864744110479286e-06, + "loss": 1.8935, + "step": 847 + }, + { + "epoch": 0.9293150684931507, + "grad_norm": 2.226727247238159, + "learning_rate": 3.862713241267262e-06, + "loss": 1.8969, + "step": 848 + }, + { + "epoch": 0.9304109589041096, + "grad_norm": 2.3301939964294434, + "learning_rate": 3.86068237205524e-06, + "loss": 1.9393, + "step": 849 + }, + { + "epoch": 0.9315068493150684, + "grad_norm": 2.180250883102417, + "learning_rate": 3.858651502843217e-06, + "loss": 1.9331, + "step": 850 + }, + { + "epoch": 0.9315068493150684, + "eval_loss": 2.0509681701660156, + "eval_runtime": 37.2111, + "eval_samples_per_second": 26.874, + "eval_steps_per_second": 1.129, + "step": 850 + }, + { + "epoch": 0.9326027397260274, + "grad_norm": 2.223548173904419, + "learning_rate": 3.856620633631195e-06, + "loss": 1.894, + "step": 851 + }, + { + "epoch": 0.9336986301369863, + "grad_norm": 2.1904265880584717, + "learning_rate": 3.854589764419171e-06, + "loss": 1.8265, + "step": 852 + }, + { + "epoch": 0.9347945205479452, + "grad_norm": 2.1374316215515137, + "learning_rate": 3.852558895207149e-06, + "loss": 1.8755, + "step": 853 + }, + { + "epoch": 0.9358904109589041, + "grad_norm": 2.2175488471984863, + "learning_rate": 3.850528025995126e-06, + "loss": 1.9649, + "step": 854 + }, + { + "epoch": 0.936986301369863, + "grad_norm": 2.192126989364624, + "learning_rate": 3.848497156783104e-06, + "loss": 1.919, + "step": 855 + }, + { + "epoch": 0.938082191780822, + "grad_norm": 2.219954013824463, + "learning_rate": 3.84646628757108e-06, + "loss": 1.8651, + "step": 856 + }, + { + "epoch": 0.9391780821917808, + "grad_norm": 2.2897071838378906, + "learning_rate": 3.8444354183590585e-06, + "loss": 1.8744, + "step": 857 + }, + { + "epoch": 0.9402739726027397, + "grad_norm": 2.1656205654144287, + "learning_rate": 3.842404549147035e-06, + "loss": 1.8622, + "step": 858 + }, + { + "epoch": 0.9413698630136986, + "grad_norm": 2.20072603225708, + "learning_rate": 3.840373679935013e-06, + "loss": 1.8678, + "step": 859 + }, + { + "epoch": 0.9424657534246575, + "grad_norm": 2.1474406719207764, + "learning_rate": 3.8383428107229894e-06, + "loss": 1.8462, + "step": 860 + }, + { + "epoch": 0.9424657534246575, + "eval_loss": 2.0451602935791016, + "eval_runtime": 37.1383, + "eval_samples_per_second": 26.926, + "eval_steps_per_second": 1.131, + "step": 860 + }, + { + "epoch": 0.9435616438356165, + "grad_norm": 1.9542036056518555, + "learning_rate": 3.836311941510967e-06, + "loss": 1.8739, + "step": 861 + }, + { + "epoch": 0.9446575342465754, + "grad_norm": 2.3951354026794434, + "learning_rate": 3.834281072298944e-06, + "loss": 1.8565, + "step": 862 + }, + { + "epoch": 0.9457534246575342, + "grad_norm": 2.4916961193084717, + "learning_rate": 3.832250203086921e-06, + "loss": 1.9138, + "step": 863 + }, + { + "epoch": 0.9468493150684931, + "grad_norm": 2.1650452613830566, + "learning_rate": 3.8302193338748985e-06, + "loss": 1.8499, + "step": 864 + }, + { + "epoch": 0.947945205479452, + "grad_norm": 2.2215073108673096, + "learning_rate": 3.828188464662876e-06, + "loss": 1.9376, + "step": 865 + }, + { + "epoch": 0.949041095890411, + "grad_norm": 2.3827078342437744, + "learning_rate": 3.826157595450853e-06, + "loss": 1.9726, + "step": 866 + }, + { + "epoch": 0.9501369863013699, + "grad_norm": 2.37194561958313, + "learning_rate": 3.82412672623883e-06, + "loss": 1.9009, + "step": 867 + }, + { + "epoch": 0.9512328767123288, + "grad_norm": 2.116562604904175, + "learning_rate": 3.822095857026808e-06, + "loss": 1.8342, + "step": 868 + }, + { + "epoch": 0.9523287671232876, + "grad_norm": 2.3410849571228027, + "learning_rate": 3.820064987814785e-06, + "loss": 1.9035, + "step": 869 + }, + { + "epoch": 0.9534246575342465, + "grad_norm": 2.207545280456543, + "learning_rate": 3.818034118602762e-06, + "loss": 1.8539, + "step": 870 + }, + { + "epoch": 0.9534246575342465, + "eval_loss": 2.0433826446533203, + "eval_runtime": 37.114, + "eval_samples_per_second": 26.944, + "eval_steps_per_second": 1.132, + "step": 870 + }, + { + "epoch": 0.9545205479452055, + "grad_norm": 2.1716196537017822, + "learning_rate": 3.8160032493907395e-06, + "loss": 1.8557, + "step": 871 + }, + { + "epoch": 0.9556164383561644, + "grad_norm": 2.213124990463257, + "learning_rate": 3.813972380178717e-06, + "loss": 1.8638, + "step": 872 + }, + { + "epoch": 0.9567123287671233, + "grad_norm": 2.3185672760009766, + "learning_rate": 3.811941510966694e-06, + "loss": 1.8645, + "step": 873 + }, + { + "epoch": 0.9578082191780822, + "grad_norm": 2.453956127166748, + "learning_rate": 3.809910641754671e-06, + "loss": 1.9183, + "step": 874 + }, + { + "epoch": 0.958904109589041, + "grad_norm": 2.1971592903137207, + "learning_rate": 3.8078797725426486e-06, + "loss": 1.8445, + "step": 875 + }, + { + "epoch": 0.96, + "grad_norm": 2.273247241973877, + "learning_rate": 3.8058489033306254e-06, + "loss": 1.8024, + "step": 876 + }, + { + "epoch": 0.9610958904109589, + "grad_norm": 2.1687841415405273, + "learning_rate": 3.803818034118603e-06, + "loss": 1.8931, + "step": 877 + }, + { + "epoch": 0.9621917808219178, + "grad_norm": 2.326498508453369, + "learning_rate": 3.80178716490658e-06, + "loss": 1.8587, + "step": 878 + }, + { + "epoch": 0.9632876712328767, + "grad_norm": 2.0317249298095703, + "learning_rate": 3.7997562956945577e-06, + "loss": 1.9394, + "step": 879 + }, + { + "epoch": 0.9643835616438357, + "grad_norm": 2.3171892166137695, + "learning_rate": 3.7977254264825345e-06, + "loss": 1.92, + "step": 880 + }, + { + "epoch": 0.9643835616438357, + "eval_loss": 2.0431556701660156, + "eval_runtime": 37.1174, + "eval_samples_per_second": 26.942, + "eval_steps_per_second": 1.132, + "step": 880 + }, + { + "epoch": 0.9654794520547946, + "grad_norm": 2.2215981483459473, + "learning_rate": 3.7956945572705122e-06, + "loss": 1.8658, + "step": 881 + }, + { + "epoch": 0.9665753424657534, + "grad_norm": 2.2940685749053955, + "learning_rate": 3.793663688058489e-06, + "loss": 1.8528, + "step": 882 + }, + { + "epoch": 0.9676712328767123, + "grad_norm": 2.2218081951141357, + "learning_rate": 3.791632818846467e-06, + "loss": 1.9446, + "step": 883 + }, + { + "epoch": 0.9687671232876712, + "grad_norm": 2.2608420848846436, + "learning_rate": 3.7896019496344436e-06, + "loss": 1.9433, + "step": 884 + }, + { + "epoch": 0.9698630136986301, + "grad_norm": 2.0247788429260254, + "learning_rate": 3.787571080422421e-06, + "loss": 1.9286, + "step": 885 + }, + { + "epoch": 0.9709589041095891, + "grad_norm": 2.164071798324585, + "learning_rate": 3.785540211210398e-06, + "loss": 1.9349, + "step": 886 + }, + { + "epoch": 0.972054794520548, + "grad_norm": 2.089215040206909, + "learning_rate": 3.7835093419983755e-06, + "loss": 1.9077, + "step": 887 + }, + { + "epoch": 0.9731506849315068, + "grad_norm": 2.245256185531616, + "learning_rate": 3.7814784727863527e-06, + "loss": 1.8911, + "step": 888 + }, + { + "epoch": 0.9742465753424657, + "grad_norm": 2.1227362155914307, + "learning_rate": 3.77944760357433e-06, + "loss": 1.8761, + "step": 889 + }, + { + "epoch": 0.9753424657534246, + "grad_norm": 2.1797730922698975, + "learning_rate": 3.7774167343623073e-06, + "loss": 1.9954, + "step": 890 + }, + { + "epoch": 0.9753424657534246, + "eval_loss": 2.034895896911621, + "eval_runtime": 37.1682, + "eval_samples_per_second": 26.905, + "eval_steps_per_second": 1.13, + "step": 890 + }, + { + "epoch": 0.9764383561643836, + "grad_norm": 2.171062707901001, + "learning_rate": 3.7753858651502846e-06, + "loss": 1.9287, + "step": 891 + }, + { + "epoch": 0.9775342465753425, + "grad_norm": 2.1950814723968506, + "learning_rate": 3.773354995938262e-06, + "loss": 1.8587, + "step": 892 + }, + { + "epoch": 0.9786301369863014, + "grad_norm": 2.2582995891571045, + "learning_rate": 3.771324126726239e-06, + "loss": 1.9269, + "step": 893 + }, + { + "epoch": 0.9797260273972602, + "grad_norm": 2.218113660812378, + "learning_rate": 3.7692932575142164e-06, + "loss": 1.9007, + "step": 894 + }, + { + "epoch": 0.9808219178082191, + "grad_norm": 2.187178134918213, + "learning_rate": 3.7672623883021937e-06, + "loss": 1.9326, + "step": 895 + }, + { + "epoch": 0.9819178082191781, + "grad_norm": 2.20451283454895, + "learning_rate": 3.765231519090171e-06, + "loss": 1.8644, + "step": 896 + }, + { + "epoch": 0.983013698630137, + "grad_norm": 2.1399290561676025, + "learning_rate": 3.7632006498781482e-06, + "loss": 1.7875, + "step": 897 + }, + { + "epoch": 0.9841095890410959, + "grad_norm": 2.049473524093628, + "learning_rate": 3.761169780666125e-06, + "loss": 1.8987, + "step": 898 + }, + { + "epoch": 0.9852054794520548, + "grad_norm": 2.154444456100464, + "learning_rate": 3.7591389114541028e-06, + "loss": 1.8475, + "step": 899 + }, + { + "epoch": 0.9863013698630136, + "grad_norm": 2.1840224266052246, + "learning_rate": 3.7571080422420796e-06, + "loss": 1.8966, + "step": 900 + }, + { + "epoch": 0.9863013698630136, + "eval_loss": 2.0326430797576904, + "eval_runtime": 37.1551, + "eval_samples_per_second": 26.914, + "eval_steps_per_second": 1.13, + "step": 900 + }, + { + "epoch": 0.9873972602739726, + "grad_norm": 2.0891942977905273, + "learning_rate": 3.7550771730300573e-06, + "loss": 1.9113, + "step": 901 + }, + { + "epoch": 0.9884931506849315, + "grad_norm": 2.2646753787994385, + "learning_rate": 3.753046303818034e-06, + "loss": 1.8624, + "step": 902 + }, + { + "epoch": 0.9895890410958904, + "grad_norm": 2.15173602104187, + "learning_rate": 3.751015434606012e-06, + "loss": 1.933, + "step": 903 + }, + { + "epoch": 0.9906849315068493, + "grad_norm": 2.1751456260681152, + "learning_rate": 3.7489845653939887e-06, + "loss": 1.9114, + "step": 904 + }, + { + "epoch": 0.9917808219178083, + "grad_norm": 2.161860942840576, + "learning_rate": 3.7469536961819664e-06, + "loss": 1.9508, + "step": 905 + }, + { + "epoch": 0.9928767123287672, + "grad_norm": 2.0921733379364014, + "learning_rate": 3.7449228269699433e-06, + "loss": 1.8482, + "step": 906 + }, + { + "epoch": 0.993972602739726, + "grad_norm": 2.0823893547058105, + "learning_rate": 3.742891957757921e-06, + "loss": 1.8744, + "step": 907 + }, + { + "epoch": 0.9950684931506849, + "grad_norm": 2.136939525604248, + "learning_rate": 3.740861088545898e-06, + "loss": 1.9037, + "step": 908 + }, + { + "epoch": 0.9961643835616438, + "grad_norm": 2.0600836277008057, + "learning_rate": 3.7388302193338755e-06, + "loss": 1.8922, + "step": 909 + }, + { + "epoch": 0.9972602739726028, + "grad_norm": 2.235780715942383, + "learning_rate": 3.7367993501218524e-06, + "loss": 1.9222, + "step": 910 + }, + { + "epoch": 0.9972602739726028, + "eval_loss": 2.034031629562378, + "eval_runtime": 37.2911, + "eval_samples_per_second": 26.816, + "eval_steps_per_second": 1.126, + "step": 910 + }, + { + "epoch": 0.9983561643835617, + "grad_norm": 2.124662160873413, + "learning_rate": 3.7347684809098297e-06, + "loss": 1.8584, + "step": 911 + }, + { + "epoch": 0.9994520547945206, + "grad_norm": 2.1237940788269043, + "learning_rate": 3.732737611697807e-06, + "loss": 1.8579, + "step": 912 + }, + { + "epoch": 1.0005479452054795, + "grad_norm": 2.1175124645233154, + "learning_rate": 3.7307067424857842e-06, + "loss": 1.8348, + "step": 913 + }, + { + "epoch": 1.0016438356164383, + "grad_norm": 2.349686861038208, + "learning_rate": 3.7286758732737615e-06, + "loss": 1.7822, + "step": 914 + }, + { + "epoch": 1.0027397260273974, + "grad_norm": 2.3278446197509766, + "learning_rate": 3.7266450040617388e-06, + "loss": 1.8346, + "step": 915 + }, + { + "epoch": 1.0038356164383562, + "grad_norm": 2.1633124351501465, + "learning_rate": 3.724614134849716e-06, + "loss": 1.8137, + "step": 916 + }, + { + "epoch": 1.004931506849315, + "grad_norm": 2.5075809955596924, + "learning_rate": 3.7225832656376933e-06, + "loss": 1.955, + "step": 917 + }, + { + "epoch": 1.006027397260274, + "grad_norm": 2.3370068073272705, + "learning_rate": 3.7205523964256706e-06, + "loss": 1.7985, + "step": 918 + }, + { + "epoch": 1.0071232876712328, + "grad_norm": 2.288229465484619, + "learning_rate": 3.718521527213648e-06, + "loss": 1.923, + "step": 919 + }, + { + "epoch": 1.0082191780821919, + "grad_norm": 2.1989684104919434, + "learning_rate": 3.716490658001625e-06, + "loss": 1.8876, + "step": 920 + }, + { + "epoch": 1.0082191780821919, + "eval_loss": 2.03442120552063, + "eval_runtime": 37.5659, + "eval_samples_per_second": 26.62, + "eval_steps_per_second": 1.118, + "step": 920 + }, + { + "epoch": 1.0093150684931507, + "grad_norm": 2.202122211456299, + "learning_rate": 3.7144597887896024e-06, + "loss": 1.879, + "step": 921 + }, + { + "epoch": 1.0104109589041095, + "grad_norm": 2.325223445892334, + "learning_rate": 3.7124289195775797e-06, + "loss": 1.7632, + "step": 922 + }, + { + "epoch": 1.0115068493150685, + "grad_norm": 2.2964577674865723, + "learning_rate": 3.710398050365557e-06, + "loss": 1.8983, + "step": 923 + }, + { + "epoch": 1.0126027397260273, + "grad_norm": 2.548537015914917, + "learning_rate": 3.708367181153534e-06, + "loss": 1.8666, + "step": 924 + }, + { + "epoch": 1.0136986301369864, + "grad_norm": 2.229745626449585, + "learning_rate": 3.7063363119415115e-06, + "loss": 1.8395, + "step": 925 + }, + { + "epoch": 1.0147945205479452, + "grad_norm": 2.3639235496520996, + "learning_rate": 3.7043054427294884e-06, + "loss": 1.8864, + "step": 926 + }, + { + "epoch": 1.0158904109589042, + "grad_norm": 2.3403055667877197, + "learning_rate": 3.702274573517466e-06, + "loss": 1.8385, + "step": 927 + }, + { + "epoch": 1.016986301369863, + "grad_norm": 2.2838001251220703, + "learning_rate": 3.700243704305443e-06, + "loss": 1.8135, + "step": 928 + }, + { + "epoch": 1.0180821917808218, + "grad_norm": 2.4730923175811768, + "learning_rate": 3.6982128350934206e-06, + "loss": 1.837, + "step": 929 + }, + { + "epoch": 1.0191780821917809, + "grad_norm": 2.5711073875427246, + "learning_rate": 3.6961819658813975e-06, + "loss": 1.8276, + "step": 930 + }, + { + "epoch": 1.0191780821917809, + "eval_loss": 2.0346033573150635, + "eval_runtime": 37.6741, + "eval_samples_per_second": 26.543, + "eval_steps_per_second": 1.115, + "step": 930 + }, + { + "epoch": 1.0202739726027397, + "grad_norm": 2.296823740005493, + "learning_rate": 3.694151096669375e-06, + "loss": 1.8054, + "step": 931 + }, + { + "epoch": 1.0213698630136987, + "grad_norm": 2.321107864379883, + "learning_rate": 3.692120227457352e-06, + "loss": 1.9296, + "step": 932 + }, + { + "epoch": 1.0224657534246575, + "grad_norm": 2.3478338718414307, + "learning_rate": 3.6900893582453297e-06, + "loss": 1.8503, + "step": 933 + }, + { + "epoch": 1.0235616438356165, + "grad_norm": 2.3698856830596924, + "learning_rate": 3.6880584890333066e-06, + "loss": 1.8766, + "step": 934 + }, + { + "epoch": 1.0246575342465754, + "grad_norm": 2.2751212120056152, + "learning_rate": 3.6860276198212834e-06, + "loss": 1.7575, + "step": 935 + }, + { + "epoch": 1.0257534246575342, + "grad_norm": 2.368741035461426, + "learning_rate": 3.683996750609261e-06, + "loss": 1.8555, + "step": 936 + }, + { + "epoch": 1.0268493150684932, + "grad_norm": 2.4507620334625244, + "learning_rate": 3.681965881397238e-06, + "loss": 1.8765, + "step": 937 + }, + { + "epoch": 1.027945205479452, + "grad_norm": 2.1907308101654053, + "learning_rate": 3.6799350121852157e-06, + "loss": 1.8652, + "step": 938 + }, + { + "epoch": 1.029041095890411, + "grad_norm": 2.2696914672851562, + "learning_rate": 3.6779041429731925e-06, + "loss": 1.8854, + "step": 939 + }, + { + "epoch": 1.0301369863013699, + "grad_norm": 2.3696796894073486, + "learning_rate": 3.6758732737611702e-06, + "loss": 1.8303, + "step": 940 + }, + { + "epoch": 1.0301369863013699, + "eval_loss": 2.030118703842163, + "eval_runtime": 37.1755, + "eval_samples_per_second": 26.899, + "eval_steps_per_second": 1.13, + "step": 940 + }, + { + "epoch": 1.0312328767123287, + "grad_norm": 2.440284013748169, + "learning_rate": 3.673842404549147e-06, + "loss": 1.8498, + "step": 941 + }, + { + "epoch": 1.0323287671232877, + "grad_norm": 2.250570297241211, + "learning_rate": 3.671811535337125e-06, + "loss": 1.8361, + "step": 942 + }, + { + "epoch": 1.0334246575342465, + "grad_norm": 2.5550990104675293, + "learning_rate": 3.6697806661251016e-06, + "loss": 1.8464, + "step": 943 + }, + { + "epoch": 1.0345205479452055, + "grad_norm": 2.502474784851074, + "learning_rate": 3.6677497969130793e-06, + "loss": 1.8471, + "step": 944 + }, + { + "epoch": 1.0356164383561643, + "grad_norm": 2.3815231323242188, + "learning_rate": 3.665718927701056e-06, + "loss": 1.9786, + "step": 945 + }, + { + "epoch": 1.0367123287671234, + "grad_norm": 2.3053736686706543, + "learning_rate": 3.663688058489034e-06, + "loss": 1.8625, + "step": 946 + }, + { + "epoch": 1.0378082191780822, + "grad_norm": 2.2623229026794434, + "learning_rate": 3.6616571892770107e-06, + "loss": 1.9206, + "step": 947 + }, + { + "epoch": 1.038904109589041, + "grad_norm": 2.634559392929077, + "learning_rate": 3.659626320064988e-06, + "loss": 1.8475, + "step": 948 + }, + { + "epoch": 1.04, + "grad_norm": 2.264615297317505, + "learning_rate": 3.6575954508529653e-06, + "loss": 1.7857, + "step": 949 + }, + { + "epoch": 1.0410958904109588, + "grad_norm": 2.4125282764434814, + "learning_rate": 3.6555645816409426e-06, + "loss": 1.841, + "step": 950 + }, + { + "epoch": 1.0410958904109588, + "eval_loss": 2.0278165340423584, + "eval_runtime": 37.1503, + "eval_samples_per_second": 26.918, + "eval_steps_per_second": 1.131, + "step": 950 + }, + { + "epoch": 1.0421917808219179, + "grad_norm": 2.365743637084961, + "learning_rate": 3.65353371242892e-06, + "loss": 1.8352, + "step": 951 + }, + { + "epoch": 1.0432876712328767, + "grad_norm": 2.6469883918762207, + "learning_rate": 3.651502843216897e-06, + "loss": 1.8814, + "step": 952 + }, + { + "epoch": 1.0443835616438357, + "grad_norm": 2.3646206855773926, + "learning_rate": 3.6494719740048744e-06, + "loss": 1.7282, + "step": 953 + }, + { + "epoch": 1.0454794520547945, + "grad_norm": 2.250856876373291, + "learning_rate": 3.6474411047928517e-06, + "loss": 1.8647, + "step": 954 + }, + { + "epoch": 1.0465753424657533, + "grad_norm": 2.5404837131500244, + "learning_rate": 3.645410235580829e-06, + "loss": 1.8175, + "step": 955 + }, + { + "epoch": 1.0476712328767124, + "grad_norm": 2.4388837814331055, + "learning_rate": 3.6433793663688062e-06, + "loss": 1.8643, + "step": 956 + }, + { + "epoch": 1.0487671232876712, + "grad_norm": 2.285959005355835, + "learning_rate": 3.6413484971567835e-06, + "loss": 1.8367, + "step": 957 + }, + { + "epoch": 1.0498630136986302, + "grad_norm": 2.1190242767333984, + "learning_rate": 3.6393176279447608e-06, + "loss": 1.8242, + "step": 958 + }, + { + "epoch": 1.050958904109589, + "grad_norm": 2.397578716278076, + "learning_rate": 3.637286758732738e-06, + "loss": 1.8681, + "step": 959 + }, + { + "epoch": 1.0520547945205478, + "grad_norm": 2.510599374771118, + "learning_rate": 3.6352558895207153e-06, + "loss": 1.8804, + "step": 960 + }, + { + "epoch": 1.0520547945205478, + "eval_loss": 2.0267090797424316, + "eval_runtime": 37.1226, + "eval_samples_per_second": 26.938, + "eval_steps_per_second": 1.131, + "step": 960 + }, + { + "epoch": 1.0531506849315069, + "grad_norm": 2.4743072986602783, + "learning_rate": 3.633225020308692e-06, + "loss": 1.8895, + "step": 961 + }, + { + "epoch": 1.0542465753424657, + "grad_norm": 2.3666775226593018, + "learning_rate": 3.63119415109667e-06, + "loss": 1.8836, + "step": 962 + }, + { + "epoch": 1.0553424657534247, + "grad_norm": 2.301119327545166, + "learning_rate": 3.6291632818846467e-06, + "loss": 1.8541, + "step": 963 + }, + { + "epoch": 1.0564383561643835, + "grad_norm": 2.482811212539673, + "learning_rate": 3.6271324126726244e-06, + "loss": 1.88, + "step": 964 + }, + { + "epoch": 1.0575342465753426, + "grad_norm": 2.2962276935577393, + "learning_rate": 3.6251015434606013e-06, + "loss": 1.7834, + "step": 965 + }, + { + "epoch": 1.0586301369863014, + "grad_norm": 2.365130662918091, + "learning_rate": 3.623070674248579e-06, + "loss": 1.7946, + "step": 966 + }, + { + "epoch": 1.0597260273972602, + "grad_norm": 2.4644675254821777, + "learning_rate": 3.621039805036556e-06, + "loss": 1.8379, + "step": 967 + }, + { + "epoch": 1.0608219178082192, + "grad_norm": 2.31742787361145, + "learning_rate": 3.6190089358245335e-06, + "loss": 1.8192, + "step": 968 + }, + { + "epoch": 1.061917808219178, + "grad_norm": 2.1614131927490234, + "learning_rate": 3.6169780666125104e-06, + "loss": 1.8794, + "step": 969 + }, + { + "epoch": 1.063013698630137, + "grad_norm": 2.4367032051086426, + "learning_rate": 3.614947197400488e-06, + "loss": 1.8673, + "step": 970 + }, + { + "epoch": 1.063013698630137, + "eval_loss": 2.0261292457580566, + "eval_runtime": 37.1116, + "eval_samples_per_second": 26.946, + "eval_steps_per_second": 1.132, + "step": 970 + }, + { + "epoch": 1.0641095890410959, + "grad_norm": 2.262789726257324, + "learning_rate": 3.612916328188465e-06, + "loss": 1.8808, + "step": 971 + }, + { + "epoch": 1.0652054794520547, + "grad_norm": 2.316316843032837, + "learning_rate": 3.610885458976442e-06, + "loss": 1.7803, + "step": 972 + }, + { + "epoch": 1.0663013698630137, + "grad_norm": 2.2936418056488037, + "learning_rate": 3.6088545897644195e-06, + "loss": 1.8783, + "step": 973 + }, + { + "epoch": 1.0673972602739725, + "grad_norm": 2.667069673538208, + "learning_rate": 3.6068237205523963e-06, + "loss": 1.8748, + "step": 974 + }, + { + "epoch": 1.0684931506849316, + "grad_norm": 2.31111216545105, + "learning_rate": 3.604792851340374e-06, + "loss": 1.8119, + "step": 975 + }, + { + "epoch": 1.0695890410958904, + "grad_norm": 2.268258571624756, + "learning_rate": 3.602761982128351e-06, + "loss": 1.8055, + "step": 976 + }, + { + "epoch": 1.0706849315068494, + "grad_norm": 2.307137966156006, + "learning_rate": 3.6007311129163286e-06, + "loss": 1.8976, + "step": 977 + }, + { + "epoch": 1.0717808219178082, + "grad_norm": 2.406036853790283, + "learning_rate": 3.5987002437043054e-06, + "loss": 1.8608, + "step": 978 + }, + { + "epoch": 1.072876712328767, + "grad_norm": 2.3303062915802, + "learning_rate": 3.596669374492283e-06, + "loss": 1.8038, + "step": 979 + }, + { + "epoch": 1.073972602739726, + "grad_norm": 2.2280168533325195, + "learning_rate": 3.59463850528026e-06, + "loss": 1.8342, + "step": 980 + }, + { + "epoch": 1.073972602739726, + "eval_loss": 2.0273854732513428, + "eval_runtime": 37.1546, + "eval_samples_per_second": 26.915, + "eval_steps_per_second": 1.13, + "step": 980 + }, + { + "epoch": 1.0750684931506849, + "grad_norm": 2.4218759536743164, + "learning_rate": 3.5926076360682377e-06, + "loss": 1.9104, + "step": 981 + }, + { + "epoch": 1.076164383561644, + "grad_norm": 2.2140634059906006, + "learning_rate": 3.5905767668562145e-06, + "loss": 1.7927, + "step": 982 + }, + { + "epoch": 1.0772602739726027, + "grad_norm": 2.179877996444702, + "learning_rate": 3.5885458976441922e-06, + "loss": 1.8868, + "step": 983 + }, + { + "epoch": 1.0783561643835617, + "grad_norm": 2.1988537311553955, + "learning_rate": 3.586515028432169e-06, + "loss": 1.8516, + "step": 984 + }, + { + "epoch": 1.0794520547945206, + "grad_norm": 2.1724448204040527, + "learning_rate": 3.5844841592201464e-06, + "loss": 1.8869, + "step": 985 + }, + { + "epoch": 1.0805479452054794, + "grad_norm": 2.2685718536376953, + "learning_rate": 3.5824532900081237e-06, + "loss": 1.8955, + "step": 986 + }, + { + "epoch": 1.0816438356164384, + "grad_norm": 2.39284348487854, + "learning_rate": 3.580422420796101e-06, + "loss": 1.7769, + "step": 987 + }, + { + "epoch": 1.0827397260273972, + "grad_norm": 2.3448750972747803, + "learning_rate": 3.578391551584078e-06, + "loss": 1.8417, + "step": 988 + }, + { + "epoch": 1.0838356164383562, + "grad_norm": 2.3518283367156982, + "learning_rate": 3.5763606823720555e-06, + "loss": 1.8524, + "step": 989 + }, + { + "epoch": 1.084931506849315, + "grad_norm": 2.1696958541870117, + "learning_rate": 3.5743298131600328e-06, + "loss": 1.7668, + "step": 990 + }, + { + "epoch": 1.084931506849315, + "eval_loss": 2.0244593620300293, + "eval_runtime": 37.1929, + "eval_samples_per_second": 26.887, + "eval_steps_per_second": 1.129, + "step": 990 + }, + { + "epoch": 1.086027397260274, + "grad_norm": 2.18300724029541, + "learning_rate": 3.57229894394801e-06, + "loss": 1.852, + "step": 991 + }, + { + "epoch": 1.087123287671233, + "grad_norm": 2.2075037956237793, + "learning_rate": 3.5702680747359873e-06, + "loss": 1.8799, + "step": 992 + }, + { + "epoch": 1.0882191780821917, + "grad_norm": 2.2172231674194336, + "learning_rate": 3.5682372055239646e-06, + "loss": 1.8573, + "step": 993 + }, + { + "epoch": 1.0893150684931507, + "grad_norm": 2.25169038772583, + "learning_rate": 3.566206336311942e-06, + "loss": 1.8125, + "step": 994 + }, + { + "epoch": 1.0904109589041096, + "grad_norm": 2.168386697769165, + "learning_rate": 3.564175467099919e-06, + "loss": 1.857, + "step": 995 + }, + { + "epoch": 1.0915068493150686, + "grad_norm": 2.2610065937042236, + "learning_rate": 3.5621445978878964e-06, + "loss": 1.817, + "step": 996 + }, + { + "epoch": 1.0926027397260274, + "grad_norm": 2.2561864852905273, + "learning_rate": 3.5601137286758737e-06, + "loss": 1.7754, + "step": 997 + }, + { + "epoch": 1.0936986301369862, + "grad_norm": 2.143625497817993, + "learning_rate": 3.5580828594638505e-06, + "loss": 1.8734, + "step": 998 + }, + { + "epoch": 1.0947945205479452, + "grad_norm": 2.433831214904785, + "learning_rate": 3.5560519902518282e-06, + "loss": 1.8248, + "step": 999 + }, + { + "epoch": 1.095890410958904, + "grad_norm": 2.237497091293335, + "learning_rate": 3.554021121039805e-06, + "loss": 1.8543, + "step": 1000 + }, + { + "epoch": 1.095890410958904, + "eval_loss": 2.020202159881592, + "eval_runtime": 37.2413, + "eval_samples_per_second": 26.852, + "eval_steps_per_second": 1.128, + "step": 1000 + }, + { + "epoch": 1.096986301369863, + "grad_norm": 2.253833532333374, + "learning_rate": 3.551990251827783e-06, + "loss": 1.8799, + "step": 1001 + }, + { + "epoch": 1.098082191780822, + "grad_norm": 2.205322504043579, + "learning_rate": 3.5499593826157596e-06, + "loss": 1.834, + "step": 1002 + }, + { + "epoch": 1.099178082191781, + "grad_norm": 2.3098037242889404, + "learning_rate": 3.5479285134037373e-06, + "loss": 1.8184, + "step": 1003 + }, + { + "epoch": 1.1002739726027397, + "grad_norm": 2.361888885498047, + "learning_rate": 3.545897644191714e-06, + "loss": 1.827, + "step": 1004 + }, + { + "epoch": 1.1013698630136985, + "grad_norm": 2.341947555541992, + "learning_rate": 3.543866774979692e-06, + "loss": 1.8184, + "step": 1005 + }, + { + "epoch": 1.1024657534246576, + "grad_norm": 2.2927780151367188, + "learning_rate": 3.5418359057676687e-06, + "loss": 1.805, + "step": 1006 + }, + { + "epoch": 1.1035616438356164, + "grad_norm": 2.432988166809082, + "learning_rate": 3.5398050365556464e-06, + "loss": 1.8465, + "step": 1007 + }, + { + "epoch": 1.1046575342465754, + "grad_norm": 2.428126335144043, + "learning_rate": 3.5377741673436233e-06, + "loss": 1.8115, + "step": 1008 + }, + { + "epoch": 1.1057534246575342, + "grad_norm": 2.32551646232605, + "learning_rate": 3.535743298131601e-06, + "loss": 1.8522, + "step": 1009 + }, + { + "epoch": 1.106849315068493, + "grad_norm": 2.297980308532715, + "learning_rate": 3.533712428919578e-06, + "loss": 1.8137, + "step": 1010 + }, + { + "epoch": 1.106849315068493, + "eval_loss": 2.0190494060516357, + "eval_runtime": 37.2149, + "eval_samples_per_second": 26.871, + "eval_steps_per_second": 1.129, + "step": 1010 + }, + { + "epoch": 1.107945205479452, + "grad_norm": 2.364342451095581, + "learning_rate": 3.5316815597075547e-06, + "loss": 1.8677, + "step": 1011 + }, + { + "epoch": 1.1090410958904109, + "grad_norm": 2.355008125305176, + "learning_rate": 3.5296506904955324e-06, + "loss": 1.8937, + "step": 1012 + }, + { + "epoch": 1.11013698630137, + "grad_norm": 2.2288317680358887, + "learning_rate": 3.5276198212835093e-06, + "loss": 1.8188, + "step": 1013 + }, + { + "epoch": 1.1112328767123287, + "grad_norm": 2.2801361083984375, + "learning_rate": 3.525588952071487e-06, + "loss": 1.8386, + "step": 1014 + }, + { + "epoch": 1.1123287671232878, + "grad_norm": 2.2801361083984375, + "learning_rate": 3.525588952071487e-06, + "loss": 1.8845, + "step": 1015 + }, + { + "epoch": 1.1134246575342466, + "grad_norm": 2.4301352500915527, + "learning_rate": 3.523558082859464e-06, + "loss": 1.8898, + "step": 1016 + }, + { + "epoch": 1.1145205479452054, + "grad_norm": 2.3492274284362793, + "learning_rate": 3.5215272136474415e-06, + "loss": 1.8711, + "step": 1017 + }, + { + "epoch": 1.1156164383561644, + "grad_norm": 2.1788229942321777, + "learning_rate": 3.5194963444354184e-06, + "loss": 1.8177, + "step": 1018 + }, + { + "epoch": 1.1167123287671232, + "grad_norm": 2.269767999649048, + "learning_rate": 3.517465475223396e-06, + "loss": 1.8259, + "step": 1019 + }, + { + "epoch": 1.1178082191780823, + "grad_norm": 2.2890219688415527, + "learning_rate": 3.515434606011373e-06, + "loss": 1.7873, + "step": 1020 + }, + { + "epoch": 1.1178082191780823, + "eval_loss": 2.023195266723633, + "eval_runtime": 37.5701, + "eval_samples_per_second": 26.617, + "eval_steps_per_second": 1.118, + "step": 1020 + }, + { + "epoch": 1.118904109589041, + "grad_norm": 2.5528533458709717, + "learning_rate": 3.5134037367993506e-06, + "loss": 1.8624, + "step": 1021 + }, + { + "epoch": 1.12, + "grad_norm": 2.5109457969665527, + "learning_rate": 3.5113728675873275e-06, + "loss": 1.9178, + "step": 1022 + }, + { + "epoch": 1.121095890410959, + "grad_norm": 2.288877010345459, + "learning_rate": 3.5093419983753047e-06, + "loss": 1.8604, + "step": 1023 + }, + { + "epoch": 1.1221917808219177, + "grad_norm": 2.460153818130493, + "learning_rate": 3.507311129163282e-06, + "loss": 1.8551, + "step": 1024 + }, + { + "epoch": 1.1232876712328768, + "grad_norm": 2.3520290851593018, + "learning_rate": 3.5052802599512593e-06, + "loss": 1.7818, + "step": 1025 + }, + { + "epoch": 1.1243835616438356, + "grad_norm": 2.32812237739563, + "learning_rate": 3.5032493907392366e-06, + "loss": 1.8255, + "step": 1026 + }, + { + "epoch": 1.1254794520547946, + "grad_norm": 2.3118042945861816, + "learning_rate": 3.501218521527214e-06, + "loss": 1.8493, + "step": 1027 + }, + { + "epoch": 1.1265753424657534, + "grad_norm": 2.2750277519226074, + "learning_rate": 3.499187652315191e-06, + "loss": 1.7759, + "step": 1028 + }, + { + "epoch": 1.1276712328767124, + "grad_norm": 2.272120475769043, + "learning_rate": 3.4971567831031684e-06, + "loss": 1.8399, + "step": 1029 + }, + { + "epoch": 1.1287671232876713, + "grad_norm": 2.2075254917144775, + "learning_rate": 3.4951259138911457e-06, + "loss": 1.8774, + "step": 1030 + }, + { + "epoch": 1.1287671232876713, + "eval_loss": 2.0199804306030273, + "eval_runtime": 37.6753, + "eval_samples_per_second": 26.543, + "eval_steps_per_second": 1.115, + "step": 1030 + }, + { + "epoch": 1.12986301369863, + "grad_norm": 2.359877586364746, + "learning_rate": 3.493095044679123e-06, + "loss": 1.7942, + "step": 1031 + }, + { + "epoch": 1.130958904109589, + "grad_norm": 2.3057703971862793, + "learning_rate": 3.4910641754671002e-06, + "loss": 1.8996, + "step": 1032 + }, + { + "epoch": 1.132054794520548, + "grad_norm": 2.3109569549560547, + "learning_rate": 3.4890333062550775e-06, + "loss": 1.7336, + "step": 1033 + }, + { + "epoch": 1.133150684931507, + "grad_norm": 2.3021819591522217, + "learning_rate": 3.4870024370430548e-06, + "loss": 1.869, + "step": 1034 + }, + { + "epoch": 1.1342465753424658, + "grad_norm": 2.4741852283477783, + "learning_rate": 3.484971567831032e-06, + "loss": 1.7652, + "step": 1035 + }, + { + "epoch": 1.1353424657534246, + "grad_norm": 2.326500415802002, + "learning_rate": 3.482940698619009e-06, + "loss": 1.809, + "step": 1036 + }, + { + "epoch": 1.1364383561643836, + "grad_norm": 2.2562193870544434, + "learning_rate": 3.4809098294069866e-06, + "loss": 1.8806, + "step": 1037 + }, + { + "epoch": 1.1375342465753424, + "grad_norm": 2.2849345207214355, + "learning_rate": 3.4788789601949634e-06, + "loss": 1.8329, + "step": 1038 + }, + { + "epoch": 1.1386301369863014, + "grad_norm": 2.333707332611084, + "learning_rate": 3.476848090982941e-06, + "loss": 1.8541, + "step": 1039 + }, + { + "epoch": 1.1397260273972603, + "grad_norm": 2.234175682067871, + "learning_rate": 3.474817221770918e-06, + "loss": 1.8126, + "step": 1040 + }, + { + "epoch": 1.1397260273972603, + "eval_loss": 2.0176141262054443, + "eval_runtime": 37.232, + "eval_samples_per_second": 26.859, + "eval_steps_per_second": 1.128, + "step": 1040 + }, + { + "epoch": 1.140821917808219, + "grad_norm": 2.4212377071380615, + "learning_rate": 3.4727863525588957e-06, + "loss": 1.9325, + "step": 1041 + }, + { + "epoch": 1.141917808219178, + "grad_norm": 2.1526010036468506, + "learning_rate": 3.4707554833468725e-06, + "loss": 1.8674, + "step": 1042 + }, + { + "epoch": 1.143013698630137, + "grad_norm": 2.336186408996582, + "learning_rate": 3.4687246141348502e-06, + "loss": 1.8275, + "step": 1043 + }, + { + "epoch": 1.144109589041096, + "grad_norm": 2.2991745471954346, + "learning_rate": 3.466693744922827e-06, + "loss": 1.8057, + "step": 1044 + }, + { + "epoch": 1.1452054794520548, + "grad_norm": 2.3816370964050293, + "learning_rate": 3.464662875710805e-06, + "loss": 1.8212, + "step": 1045 + }, + { + "epoch": 1.1463013698630138, + "grad_norm": 2.2383193969726562, + "learning_rate": 3.4626320064987817e-06, + "loss": 1.8444, + "step": 1046 + }, + { + "epoch": 1.1473972602739726, + "grad_norm": 2.377574920654297, + "learning_rate": 3.4606011372867594e-06, + "loss": 1.8319, + "step": 1047 + }, + { + "epoch": 1.1484931506849314, + "grad_norm": 2.1803700923919678, + "learning_rate": 3.458570268074736e-06, + "loss": 1.8484, + "step": 1048 + }, + { + "epoch": 1.1495890410958904, + "grad_norm": 2.1810641288757324, + "learning_rate": 3.4565393988627135e-06, + "loss": 1.8187, + "step": 1049 + }, + { + "epoch": 1.1506849315068493, + "grad_norm": 2.1565394401550293, + "learning_rate": 3.4545085296506908e-06, + "loss": 1.8117, + "step": 1050 + }, + { + "epoch": 1.1506849315068493, + "eval_loss": 2.0156562328338623, + "eval_runtime": 37.2072, + "eval_samples_per_second": 26.877, + "eval_steps_per_second": 1.129, + "step": 1050 + }, + { + "epoch": 1.1517808219178083, + "grad_norm": 2.1753759384155273, + "learning_rate": 3.452477660438668e-06, + "loss": 1.7822, + "step": 1051 + }, + { + "epoch": 1.152876712328767, + "grad_norm": 2.250089645385742, + "learning_rate": 3.4504467912266453e-06, + "loss": 1.8634, + "step": 1052 + }, + { + "epoch": 1.1539726027397261, + "grad_norm": 2.3043293952941895, + "learning_rate": 3.4484159220146226e-06, + "loss": 1.7743, + "step": 1053 + }, + { + "epoch": 1.155068493150685, + "grad_norm": 2.2193732261657715, + "learning_rate": 3.4463850528026e-06, + "loss": 1.8328, + "step": 1054 + }, + { + "epoch": 1.1561643835616437, + "grad_norm": 2.4204866886138916, + "learning_rate": 3.444354183590577e-06, + "loss": 1.8133, + "step": 1055 + }, + { + "epoch": 1.1572602739726028, + "grad_norm": 2.3088018894195557, + "learning_rate": 3.4423233143785544e-06, + "loss": 1.8523, + "step": 1056 + }, + { + "epoch": 1.1583561643835616, + "grad_norm": 2.38910174369812, + "learning_rate": 3.4402924451665317e-06, + "loss": 1.7846, + "step": 1057 + }, + { + "epoch": 1.1594520547945206, + "grad_norm": 2.2473702430725098, + "learning_rate": 3.438261575954509e-06, + "loss": 1.7632, + "step": 1058 + }, + { + "epoch": 1.1605479452054794, + "grad_norm": 2.3356645107269287, + "learning_rate": 3.4362307067424862e-06, + "loss": 1.8461, + "step": 1059 + }, + { + "epoch": 1.1616438356164385, + "grad_norm": 2.2460737228393555, + "learning_rate": 3.434199837530463e-06, + "loss": 1.8853, + "step": 1060 + }, + { + "epoch": 1.1616438356164385, + "eval_loss": 2.016077995300293, + "eval_runtime": 37.1943, + "eval_samples_per_second": 26.886, + "eval_steps_per_second": 1.129, + "step": 1060 + }, + { + "epoch": 1.1627397260273973, + "grad_norm": 2.28434157371521, + "learning_rate": 3.432168968318441e-06, + "loss": 1.8451, + "step": 1061 + }, + { + "epoch": 1.163835616438356, + "grad_norm": 2.299049139022827, + "learning_rate": 3.4301380991064176e-06, + "loss": 1.7777, + "step": 1062 + }, + { + "epoch": 1.1649315068493151, + "grad_norm": 2.364074468612671, + "learning_rate": 3.4281072298943953e-06, + "loss": 1.8517, + "step": 1063 + }, + { + "epoch": 1.166027397260274, + "grad_norm": 2.2381322383880615, + "learning_rate": 3.426076360682372e-06, + "loss": 1.889, + "step": 1064 + }, + { + "epoch": 1.167123287671233, + "grad_norm": 2.2984025478363037, + "learning_rate": 3.42404549147035e-06, + "loss": 1.8907, + "step": 1065 + }, + { + "epoch": 1.1682191780821918, + "grad_norm": 2.2215473651885986, + "learning_rate": 3.4220146222583267e-06, + "loss": 1.8313, + "step": 1066 + }, + { + "epoch": 1.1693150684931506, + "grad_norm": 2.3414270877838135, + "learning_rate": 3.4199837530463044e-06, + "loss": 1.783, + "step": 1067 + }, + { + "epoch": 1.1704109589041096, + "grad_norm": 2.0636470317840576, + "learning_rate": 3.4179528838342813e-06, + "loss": 1.8185, + "step": 1068 + }, + { + "epoch": 1.1715068493150684, + "grad_norm": 2.2158873081207275, + "learning_rate": 3.415922014622259e-06, + "loss": 1.9319, + "step": 1069 + }, + { + "epoch": 1.1726027397260275, + "grad_norm": 2.244051694869995, + "learning_rate": 3.413891145410236e-06, + "loss": 1.8732, + "step": 1070 + }, + { + "epoch": 1.1726027397260275, + "eval_loss": 2.0180234909057617, + "eval_runtime": 37.1822, + "eval_samples_per_second": 26.895, + "eval_steps_per_second": 1.13, + "step": 1070 + }, + { + "epoch": 1.1736986301369863, + "grad_norm": 2.1608433723449707, + "learning_rate": 3.4118602761982135e-06, + "loss": 1.8709, + "step": 1071 + }, + { + "epoch": 1.174794520547945, + "grad_norm": 2.2612338066101074, + "learning_rate": 3.4098294069861904e-06, + "loss": 1.8882, + "step": 1072 + }, + { + "epoch": 1.1758904109589041, + "grad_norm": 2.3257741928100586, + "learning_rate": 3.4077985377741673e-06, + "loss": 1.8377, + "step": 1073 + }, + { + "epoch": 1.176986301369863, + "grad_norm": 2.2486586570739746, + "learning_rate": 3.405767668562145e-06, + "loss": 1.8105, + "step": 1074 + }, + { + "epoch": 1.178082191780822, + "grad_norm": 2.2840020656585693, + "learning_rate": 3.403736799350122e-06, + "loss": 1.8868, + "step": 1075 + }, + { + "epoch": 1.1791780821917808, + "grad_norm": 2.3264284133911133, + "learning_rate": 3.4017059301380995e-06, + "loss": 1.8701, + "step": 1076 + }, + { + "epoch": 1.1802739726027398, + "grad_norm": 2.311424493789673, + "learning_rate": 3.3996750609260764e-06, + "loss": 1.801, + "step": 1077 + }, + { + "epoch": 1.1813698630136986, + "grad_norm": 2.2177181243896484, + "learning_rate": 3.397644191714054e-06, + "loss": 1.9188, + "step": 1078 + }, + { + "epoch": 1.1824657534246574, + "grad_norm": 2.413118362426758, + "learning_rate": 3.395613322502031e-06, + "loss": 1.8198, + "step": 1079 + }, + { + "epoch": 1.1835616438356165, + "grad_norm": 2.334688186645508, + "learning_rate": 3.3935824532900086e-06, + "loss": 1.822, + "step": 1080 + }, + { + "epoch": 1.1835616438356165, + "eval_loss": 2.012676954269409, + "eval_runtime": 37.2415, + "eval_samples_per_second": 26.852, + "eval_steps_per_second": 1.128, + "step": 1080 + }, + { + "epoch": 1.1846575342465753, + "grad_norm": 2.3901584148406982, + "learning_rate": 3.3915515840779855e-06, + "loss": 1.791, + "step": 1081 + }, + { + "epoch": 1.1857534246575343, + "grad_norm": 2.2731425762176514, + "learning_rate": 3.389520714865963e-06, + "loss": 1.8367, + "step": 1082 + }, + { + "epoch": 1.1868493150684931, + "grad_norm": 2.311553478240967, + "learning_rate": 3.38748984565394e-06, + "loss": 1.8302, + "step": 1083 + }, + { + "epoch": 1.1879452054794521, + "grad_norm": 2.3030776977539062, + "learning_rate": 3.3854589764419177e-06, + "loss": 1.8652, + "step": 1084 + }, + { + "epoch": 1.189041095890411, + "grad_norm": 2.341188430786133, + "learning_rate": 3.3834281072298946e-06, + "loss": 1.8094, + "step": 1085 + }, + { + "epoch": 1.1901369863013698, + "grad_norm": 2.331962823867798, + "learning_rate": 3.381397238017872e-06, + "loss": 1.827, + "step": 1086 + }, + { + "epoch": 1.1912328767123288, + "grad_norm": 2.275923728942871, + "learning_rate": 3.379366368805849e-06, + "loss": 1.7935, + "step": 1087 + }, + { + "epoch": 1.1923287671232876, + "grad_norm": 2.4608163833618164, + "learning_rate": 3.3773354995938264e-06, + "loss": 1.8965, + "step": 1088 + }, + { + "epoch": 1.1934246575342466, + "grad_norm": 2.2208592891693115, + "learning_rate": 3.3753046303818037e-06, + "loss": 1.8447, + "step": 1089 + }, + { + "epoch": 1.1945205479452055, + "grad_norm": 2.4284770488739014, + "learning_rate": 3.373273761169781e-06, + "loss": 1.8306, + "step": 1090 + }, + { + "epoch": 1.1945205479452055, + "eval_loss": 2.010664939880371, + "eval_runtime": 37.237, + "eval_samples_per_second": 26.855, + "eval_steps_per_second": 1.128, + "step": 1090 + }, + { + "epoch": 1.1956164383561645, + "grad_norm": 2.329601764678955, + "learning_rate": 3.3712428919577582e-06, + "loss": 1.8275, + "step": 1091 + }, + { + "epoch": 1.1967123287671233, + "grad_norm": 2.277939796447754, + "learning_rate": 3.3692120227457355e-06, + "loss": 1.8161, + "step": 1092 + }, + { + "epoch": 1.1978082191780821, + "grad_norm": 2.446780204772949, + "learning_rate": 3.3671811535337128e-06, + "loss": 1.8511, + "step": 1093 + }, + { + "epoch": 1.1989041095890411, + "grad_norm": 2.382477283477783, + "learning_rate": 3.36515028432169e-06, + "loss": 1.7834, + "step": 1094 + }, + { + "epoch": 1.2, + "grad_norm": 2.366361379623413, + "learning_rate": 3.3631194151096673e-06, + "loss": 1.8105, + "step": 1095 + }, + { + "epoch": 1.201095890410959, + "grad_norm": 2.3439717292785645, + "learning_rate": 3.3610885458976446e-06, + "loss": 1.8418, + "step": 1096 + }, + { + "epoch": 1.2021917808219178, + "grad_norm": 2.2739830017089844, + "learning_rate": 3.359057676685622e-06, + "loss": 1.8818, + "step": 1097 + }, + { + "epoch": 1.2032876712328768, + "grad_norm": 2.1636738777160645, + "learning_rate": 3.357026807473599e-06, + "loss": 1.8757, + "step": 1098 + }, + { + "epoch": 1.2043835616438356, + "grad_norm": 2.386441946029663, + "learning_rate": 3.354995938261576e-06, + "loss": 1.756, + "step": 1099 + }, + { + "epoch": 1.2054794520547945, + "grad_norm": 2.2241456508636475, + "learning_rate": 3.3529650690495537e-06, + "loss": 1.8034, + "step": 1100 + }, + { + "epoch": 1.2054794520547945, + "eval_loss": 2.008504867553711, + "eval_runtime": 37.2855, + "eval_samples_per_second": 26.82, + "eval_steps_per_second": 1.126, + "step": 1100 + }, + { + "epoch": 1.2065753424657535, + "grad_norm": 2.196992874145508, + "learning_rate": 3.3509341998375305e-06, + "loss": 1.7722, + "step": 1101 + }, + { + "epoch": 1.2076712328767123, + "grad_norm": 2.3572094440460205, + "learning_rate": 3.3489033306255082e-06, + "loss": 1.8109, + "step": 1102 + }, + { + "epoch": 1.208767123287671, + "grad_norm": 2.443281650543213, + "learning_rate": 3.346872461413485e-06, + "loss": 1.843, + "step": 1103 + }, + { + "epoch": 1.2098630136986301, + "grad_norm": 2.2613234519958496, + "learning_rate": 3.344841592201463e-06, + "loss": 1.8023, + "step": 1104 + }, + { + "epoch": 1.210958904109589, + "grad_norm": 2.2466726303100586, + "learning_rate": 3.3428107229894397e-06, + "loss": 1.7838, + "step": 1105 + }, + { + "epoch": 1.212054794520548, + "grad_norm": 2.2314462661743164, + "learning_rate": 3.3407798537774174e-06, + "loss": 1.8077, + "step": 1106 + }, + { + "epoch": 1.2131506849315068, + "grad_norm": 2.3288400173187256, + "learning_rate": 3.338748984565394e-06, + "loss": 1.8319, + "step": 1107 + }, + { + "epoch": 1.2142465753424658, + "grad_norm": 2.3477249145507812, + "learning_rate": 3.336718115353372e-06, + "loss": 1.8772, + "step": 1108 + }, + { + "epoch": 1.2153424657534246, + "grad_norm": 2.3588409423828125, + "learning_rate": 3.3346872461413488e-06, + "loss": 1.8832, + "step": 1109 + }, + { + "epoch": 1.2164383561643834, + "grad_norm": 2.327796220779419, + "learning_rate": 3.3326563769293256e-06, + "loss": 1.8033, + "step": 1110 + }, + { + "epoch": 1.2164383561643834, + "eval_loss": 2.0094382762908936, + "eval_runtime": 37.2666, + "eval_samples_per_second": 26.834, + "eval_steps_per_second": 1.127, + "step": 1110 + }, + { + "epoch": 1.2175342465753425, + "grad_norm": 2.257244825363159, + "learning_rate": 3.3306255077173033e-06, + "loss": 1.7903, + "step": 1111 + }, + { + "epoch": 1.2186301369863013, + "grad_norm": 2.3321614265441895, + "learning_rate": 3.32859463850528e-06, + "loss": 1.8295, + "step": 1112 + }, + { + "epoch": 1.2197260273972603, + "grad_norm": 2.4073803424835205, + "learning_rate": 3.326563769293258e-06, + "loss": 1.8395, + "step": 1113 + }, + { + "epoch": 1.2208219178082191, + "grad_norm": 2.3631434440612793, + "learning_rate": 3.3245329000812347e-06, + "loss": 1.7548, + "step": 1114 + }, + { + "epoch": 1.2219178082191782, + "grad_norm": 2.275766134262085, + "learning_rate": 3.3225020308692124e-06, + "loss": 1.808, + "step": 1115 + }, + { + "epoch": 1.223013698630137, + "grad_norm": 2.438011407852173, + "learning_rate": 3.3204711616571893e-06, + "loss": 1.8274, + "step": 1116 + }, + { + "epoch": 1.2241095890410958, + "grad_norm": 2.371826171875, + "learning_rate": 3.318440292445167e-06, + "loss": 1.7742, + "step": 1117 + }, + { + "epoch": 1.2252054794520548, + "grad_norm": 2.422004461288452, + "learning_rate": 3.316409423233144e-06, + "loss": 1.8268, + "step": 1118 + }, + { + "epoch": 1.2263013698630136, + "grad_norm": 2.377979278564453, + "learning_rate": 3.3143785540211215e-06, + "loss": 1.7857, + "step": 1119 + }, + { + "epoch": 1.2273972602739727, + "grad_norm": 2.288482904434204, + "learning_rate": 3.3123476848090984e-06, + "loss": 1.876, + "step": 1120 + }, + { + "epoch": 1.2273972602739727, + "eval_loss": 2.0104730129241943, + "eval_runtime": 37.5365, + "eval_samples_per_second": 26.641, + "eval_steps_per_second": 1.119, + "step": 1120 + }, + { + "epoch": 1.2284931506849315, + "grad_norm": 2.451533555984497, + "learning_rate": 3.310316815597076e-06, + "loss": 1.8324, + "step": 1121 + }, + { + "epoch": 1.2295890410958905, + "grad_norm": 2.3785181045532227, + "learning_rate": 3.308285946385053e-06, + "loss": 1.8239, + "step": 1122 + }, + { + "epoch": 1.2306849315068493, + "grad_norm": 2.2276721000671387, + "learning_rate": 3.30625507717303e-06, + "loss": 1.9132, + "step": 1123 + }, + { + "epoch": 1.2317808219178081, + "grad_norm": 2.2575314044952393, + "learning_rate": 3.3042242079610075e-06, + "loss": 1.769, + "step": 1124 + }, + { + "epoch": 1.2328767123287672, + "grad_norm": 2.5311717987060547, + "learning_rate": 3.3021933387489847e-06, + "loss": 1.8605, + "step": 1125 + }, + { + "epoch": 1.233972602739726, + "grad_norm": 2.2395896911621094, + "learning_rate": 3.300162469536962e-06, + "loss": 1.8725, + "step": 1126 + }, + { + "epoch": 1.235068493150685, + "grad_norm": 2.2402031421661377, + "learning_rate": 3.2981316003249393e-06, + "loss": 1.7955, + "step": 1127 + }, + { + "epoch": 1.2361643835616438, + "grad_norm": 2.308950424194336, + "learning_rate": 3.2961007311129166e-06, + "loss": 1.8812, + "step": 1128 + }, + { + "epoch": 1.2372602739726029, + "grad_norm": 2.437767267227173, + "learning_rate": 3.294069861900894e-06, + "loss": 1.8489, + "step": 1129 + }, + { + "epoch": 1.2383561643835617, + "grad_norm": 2.3188693523406982, + "learning_rate": 3.292038992688871e-06, + "loss": 1.9082, + "step": 1130 + }, + { + "epoch": 1.2383561643835617, + "eval_loss": 2.0110676288604736, + "eval_runtime": 37.6437, + "eval_samples_per_second": 26.565, + "eval_steps_per_second": 1.116, + "step": 1130 + }, + { + "epoch": 1.2394520547945205, + "grad_norm": 2.312760591506958, + "learning_rate": 3.2900081234768484e-06, + "loss": 1.8344, + "step": 1131 + }, + { + "epoch": 1.2405479452054795, + "grad_norm": 2.5837206840515137, + "learning_rate": 3.2879772542648257e-06, + "loss": 1.8503, + "step": 1132 + }, + { + "epoch": 1.2416438356164383, + "grad_norm": 2.1864733695983887, + "learning_rate": 3.285946385052803e-06, + "loss": 1.8822, + "step": 1133 + }, + { + "epoch": 1.2427397260273974, + "grad_norm": 2.2644550800323486, + "learning_rate": 3.2839155158407802e-06, + "loss": 1.7541, + "step": 1134 + }, + { + "epoch": 1.2438356164383562, + "grad_norm": 2.292646646499634, + "learning_rate": 3.2818846466287575e-06, + "loss": 1.8575, + "step": 1135 + }, + { + "epoch": 1.244931506849315, + "grad_norm": 2.2162070274353027, + "learning_rate": 3.2798537774167344e-06, + "loss": 1.7525, + "step": 1136 + }, + { + "epoch": 1.246027397260274, + "grad_norm": 2.32688307762146, + "learning_rate": 3.277822908204712e-06, + "loss": 1.8955, + "step": 1137 + }, + { + "epoch": 1.2471232876712328, + "grad_norm": 2.3226473331451416, + "learning_rate": 3.275792038992689e-06, + "loss": 1.8255, + "step": 1138 + }, + { + "epoch": 1.2482191780821918, + "grad_norm": 2.3533706665039062, + "learning_rate": 3.2737611697806666e-06, + "loss": 1.783, + "step": 1139 + }, + { + "epoch": 1.2493150684931507, + "grad_norm": 2.237661123275757, + "learning_rate": 3.2717303005686435e-06, + "loss": 1.8367, + "step": 1140 + }, + { + "epoch": 1.2493150684931507, + "eval_loss": 2.0123534202575684, + "eval_runtime": 37.3573, + "eval_samples_per_second": 26.769, + "eval_steps_per_second": 1.124, + "step": 1140 + }, + { + "epoch": 1.2504109589041095, + "grad_norm": 2.283047676086426, + "learning_rate": 3.269699431356621e-06, + "loss": 1.7987, + "step": 1141 + }, + { + "epoch": 1.2515068493150685, + "grad_norm": 2.2264456748962402, + "learning_rate": 3.267668562144598e-06, + "loss": 1.8307, + "step": 1142 + }, + { + "epoch": 1.2526027397260273, + "grad_norm": 2.2724905014038086, + "learning_rate": 3.2656376929325757e-06, + "loss": 1.8237, + "step": 1143 + }, + { + "epoch": 1.2536986301369863, + "grad_norm": 2.270782947540283, + "learning_rate": 3.2636068237205526e-06, + "loss": 1.7846, + "step": 1144 + }, + { + "epoch": 1.2547945205479452, + "grad_norm": 2.2032551765441895, + "learning_rate": 3.2615759545085303e-06, + "loss": 1.8438, + "step": 1145 + }, + { + "epoch": 1.2558904109589042, + "grad_norm": 2.199150800704956, + "learning_rate": 3.259545085296507e-06, + "loss": 1.8199, + "step": 1146 + }, + { + "epoch": 1.256986301369863, + "grad_norm": 2.4404170513153076, + "learning_rate": 3.257514216084484e-06, + "loss": 1.8817, + "step": 1147 + }, + { + "epoch": 1.2580821917808218, + "grad_norm": 2.281348943710327, + "learning_rate": 3.2554833468724617e-06, + "loss": 1.771, + "step": 1148 + }, + { + "epoch": 1.2591780821917808, + "grad_norm": 2.19482159614563, + "learning_rate": 3.2534524776604385e-06, + "loss": 1.8684, + "step": 1149 + }, + { + "epoch": 1.2602739726027397, + "grad_norm": 2.221618413925171, + "learning_rate": 3.2514216084484162e-06, + "loss": 1.8398, + "step": 1150 + }, + { + "epoch": 1.2602739726027397, + "eval_loss": 2.007969617843628, + "eval_runtime": 37.3679, + "eval_samples_per_second": 26.761, + "eval_steps_per_second": 1.124, + "step": 1150 + }, + { + "epoch": 1.2613698630136987, + "grad_norm": 2.193432569503784, + "learning_rate": 3.249390739236393e-06, + "loss": 1.8122, + "step": 1151 + }, + { + "epoch": 1.2624657534246575, + "grad_norm": 2.246462106704712, + "learning_rate": 3.2473598700243708e-06, + "loss": 1.8447, + "step": 1152 + }, + { + "epoch": 1.2635616438356165, + "grad_norm": 2.3411879539489746, + "learning_rate": 3.2453290008123476e-06, + "loss": 1.7988, + "step": 1153 + }, + { + "epoch": 1.2646575342465753, + "grad_norm": 2.245286464691162, + "learning_rate": 3.2432981316003253e-06, + "loss": 1.8209, + "step": 1154 + }, + { + "epoch": 1.2657534246575342, + "grad_norm": 2.292067527770996, + "learning_rate": 3.241267262388302e-06, + "loss": 1.8285, + "step": 1155 + }, + { + "epoch": 1.2668493150684932, + "grad_norm": 2.3058111667633057, + "learning_rate": 3.23923639317628e-06, + "loss": 1.7842, + "step": 1156 + }, + { + "epoch": 1.267945205479452, + "grad_norm": 2.306243896484375, + "learning_rate": 3.2372055239642567e-06, + "loss": 1.8221, + "step": 1157 + }, + { + "epoch": 1.269041095890411, + "grad_norm": 2.399759531021118, + "learning_rate": 3.2351746547522344e-06, + "loss": 1.7836, + "step": 1158 + }, + { + "epoch": 1.2701369863013698, + "grad_norm": 2.3405187129974365, + "learning_rate": 3.2331437855402113e-06, + "loss": 1.8365, + "step": 1159 + }, + { + "epoch": 1.2712328767123289, + "grad_norm": 2.3855013847351074, + "learning_rate": 3.2311129163281885e-06, + "loss": 1.8973, + "step": 1160 + }, + { + "epoch": 1.2712328767123289, + "eval_loss": 2.005983591079712, + "eval_runtime": 37.3061, + "eval_samples_per_second": 26.805, + "eval_steps_per_second": 1.126, + "step": 1160 + }, + { + "epoch": 1.2723287671232877, + "grad_norm": 2.2767140865325928, + "learning_rate": 3.229082047116166e-06, + "loss": 1.7791, + "step": 1161 + }, + { + "epoch": 1.2734246575342465, + "grad_norm": 2.289494276046753, + "learning_rate": 3.227051177904143e-06, + "loss": 1.8508, + "step": 1162 + }, + { + "epoch": 1.2745205479452055, + "grad_norm": 2.1836023330688477, + "learning_rate": 3.2250203086921204e-06, + "loss": 1.8275, + "step": 1163 + }, + { + "epoch": 1.2756164383561643, + "grad_norm": 2.1917505264282227, + "learning_rate": 3.2229894394800977e-06, + "loss": 1.8289, + "step": 1164 + }, + { + "epoch": 1.2767123287671232, + "grad_norm": 2.315983295440674, + "learning_rate": 3.220958570268075e-06, + "loss": 1.8655, + "step": 1165 + }, + { + "epoch": 1.2778082191780822, + "grad_norm": 2.3628008365631104, + "learning_rate": 3.218927701056052e-06, + "loss": 1.7396, + "step": 1166 + }, + { + "epoch": 1.2789041095890412, + "grad_norm": 2.3200488090515137, + "learning_rate": 3.2168968318440295e-06, + "loss": 1.8196, + "step": 1167 + }, + { + "epoch": 1.28, + "grad_norm": 2.2591991424560547, + "learning_rate": 3.2148659626320068e-06, + "loss": 1.7323, + "step": 1168 + }, + { + "epoch": 1.2810958904109588, + "grad_norm": 2.2458770275115967, + "learning_rate": 3.212835093419984e-06, + "loss": 1.8881, + "step": 1169 + }, + { + "epoch": 1.2821917808219179, + "grad_norm": 2.202010154724121, + "learning_rate": 3.2108042242079613e-06, + "loss": 1.7776, + "step": 1170 + }, + { + "epoch": 1.2821917808219179, + "eval_loss": 2.005174398422241, + "eval_runtime": 37.3597, + "eval_samples_per_second": 26.767, + "eval_steps_per_second": 1.124, + "step": 1170 + }, + { + "epoch": 1.2832876712328767, + "grad_norm": 2.2375359535217285, + "learning_rate": 3.2087733549959386e-06, + "loss": 1.7994, + "step": 1171 + }, + { + "epoch": 1.2843835616438355, + "grad_norm": 2.3152217864990234, + "learning_rate": 3.206742485783916e-06, + "loss": 1.8767, + "step": 1172 + }, + { + "epoch": 1.2854794520547945, + "grad_norm": 2.3317618370056152, + "learning_rate": 3.2047116165718927e-06, + "loss": 1.772, + "step": 1173 + }, + { + "epoch": 1.2865753424657536, + "grad_norm": 2.249943971633911, + "learning_rate": 3.2026807473598704e-06, + "loss": 1.7793, + "step": 1174 + }, + { + "epoch": 1.2876712328767124, + "grad_norm": 2.320481538772583, + "learning_rate": 3.2006498781478473e-06, + "loss": 1.8275, + "step": 1175 + }, + { + "epoch": 1.2887671232876712, + "grad_norm": 2.21785569190979, + "learning_rate": 3.198619008935825e-06, + "loss": 1.8449, + "step": 1176 + }, + { + "epoch": 1.2898630136986302, + "grad_norm": 2.469965934753418, + "learning_rate": 3.196588139723802e-06, + "loss": 1.779, + "step": 1177 + }, + { + "epoch": 1.290958904109589, + "grad_norm": 2.1989049911499023, + "learning_rate": 3.1945572705117795e-06, + "loss": 1.8188, + "step": 1178 + }, + { + "epoch": 1.2920547945205478, + "grad_norm": 2.293513298034668, + "learning_rate": 3.1925264012997564e-06, + "loss": 1.8377, + "step": 1179 + }, + { + "epoch": 1.2931506849315069, + "grad_norm": 2.3462443351745605, + "learning_rate": 3.190495532087734e-06, + "loss": 1.8418, + "step": 1180 + }, + { + "epoch": 1.2931506849315069, + "eval_loss": 2.0061001777648926, + "eval_runtime": 37.2951, + "eval_samples_per_second": 26.813, + "eval_steps_per_second": 1.126, + "step": 1180 + }, + { + "epoch": 1.2942465753424657, + "grad_norm": 2.2071478366851807, + "learning_rate": 3.188464662875711e-06, + "loss": 1.8419, + "step": 1181 + }, + { + "epoch": 1.2953424657534247, + "grad_norm": 2.233333110809326, + "learning_rate": 3.1864337936636886e-06, + "loss": 1.8931, + "step": 1182 + }, + { + "epoch": 1.2964383561643835, + "grad_norm": 2.383085250854492, + "learning_rate": 3.1844029244516655e-06, + "loss": 1.8047, + "step": 1183 + }, + { + "epoch": 1.2975342465753426, + "grad_norm": 2.473808765411377, + "learning_rate": 3.182372055239643e-06, + "loss": 1.7794, + "step": 1184 + }, + { + "epoch": 1.2986301369863014, + "grad_norm": 2.3357670307159424, + "learning_rate": 3.18034118602762e-06, + "loss": 1.7717, + "step": 1185 + }, + { + "epoch": 1.2997260273972602, + "grad_norm": 2.376835823059082, + "learning_rate": 3.1783103168155973e-06, + "loss": 1.8534, + "step": 1186 + }, + { + "epoch": 1.3008219178082192, + "grad_norm": 2.447415590286255, + "learning_rate": 3.1762794476035746e-06, + "loss": 1.8251, + "step": 1187 + }, + { + "epoch": 1.301917808219178, + "grad_norm": 2.5145719051361084, + "learning_rate": 3.174248578391552e-06, + "loss": 1.8117, + "step": 1188 + }, + { + "epoch": 1.303013698630137, + "grad_norm": 2.3134653568267822, + "learning_rate": 3.172217709179529e-06, + "loss": 1.8111, + "step": 1189 + }, + { + "epoch": 1.3041095890410959, + "grad_norm": 2.293311834335327, + "learning_rate": 3.1701868399675064e-06, + "loss": 1.791, + "step": 1190 + }, + { + "epoch": 1.3041095890410959, + "eval_loss": 2.0038933753967285, + "eval_runtime": 37.2282, + "eval_samples_per_second": 26.861, + "eval_steps_per_second": 1.128, + "step": 1190 + }, + { + "epoch": 1.305205479452055, + "grad_norm": 2.298020124435425, + "learning_rate": 3.1681559707554837e-06, + "loss": 1.8301, + "step": 1191 + }, + { + "epoch": 1.3063013698630137, + "grad_norm": 2.5261130332946777, + "learning_rate": 3.166125101543461e-06, + "loss": 1.7714, + "step": 1192 + }, + { + "epoch": 1.3073972602739725, + "grad_norm": 2.270820379257202, + "learning_rate": 3.1640942323314382e-06, + "loss": 1.8246, + "step": 1193 + }, + { + "epoch": 1.3084931506849315, + "grad_norm": 2.373260021209717, + "learning_rate": 3.1620633631194155e-06, + "loss": 1.8316, + "step": 1194 + }, + { + "epoch": 1.3095890410958904, + "grad_norm": 2.3265364170074463, + "learning_rate": 3.1600324939073928e-06, + "loss": 1.7679, + "step": 1195 + }, + { + "epoch": 1.3106849315068494, + "grad_norm": 2.2378532886505127, + "learning_rate": 3.15800162469537e-06, + "loss": 1.8149, + "step": 1196 + }, + { + "epoch": 1.3117808219178082, + "grad_norm": 2.2459309101104736, + "learning_rate": 3.155970755483347e-06, + "loss": 1.8086, + "step": 1197 + }, + { + "epoch": 1.3128767123287672, + "grad_norm": 2.2769811153411865, + "learning_rate": 3.1539398862713246e-06, + "loss": 1.8109, + "step": 1198 + }, + { + "epoch": 1.313972602739726, + "grad_norm": 2.310483694076538, + "learning_rate": 3.1519090170593015e-06, + "loss": 1.8297, + "step": 1199 + }, + { + "epoch": 1.3150684931506849, + "grad_norm": 2.302478313446045, + "learning_rate": 3.149878147847279e-06, + "loss": 1.8237, + "step": 1200 + }, + { + "epoch": 1.3150684931506849, + "eval_loss": 2.0011062622070312, + "eval_runtime": 37.2077, + "eval_samples_per_second": 26.876, + "eval_steps_per_second": 1.129, + "step": 1200 + }, + { + "epoch": 1.316164383561644, + "grad_norm": 2.2240796089172363, + "learning_rate": 3.147847278635256e-06, + "loss": 1.8745, + "step": 1201 + }, + { + "epoch": 1.3172602739726027, + "grad_norm": 2.303537130355835, + "learning_rate": 3.1458164094232337e-06, + "loss": 1.9051, + "step": 1202 + }, + { + "epoch": 1.3183561643835615, + "grad_norm": 2.4038636684417725, + "learning_rate": 3.1437855402112106e-06, + "loss": 1.7974, + "step": 1203 + }, + { + "epoch": 1.3194520547945205, + "grad_norm": 2.3458874225616455, + "learning_rate": 3.1417546709991883e-06, + "loss": 1.7601, + "step": 1204 + }, + { + "epoch": 1.3205479452054796, + "grad_norm": 2.3399100303649902, + "learning_rate": 3.139723801787165e-06, + "loss": 1.8314, + "step": 1205 + }, + { + "epoch": 1.3216438356164384, + "grad_norm": 2.293229818344116, + "learning_rate": 3.137692932575143e-06, + "loss": 1.8117, + "step": 1206 + }, + { + "epoch": 1.3227397260273972, + "grad_norm": 2.5378005504608154, + "learning_rate": 3.1356620633631197e-06, + "loss": 1.8614, + "step": 1207 + }, + { + "epoch": 1.3238356164383562, + "grad_norm": 2.3644745349884033, + "learning_rate": 3.1336311941510974e-06, + "loss": 1.7843, + "step": 1208 + }, + { + "epoch": 1.324931506849315, + "grad_norm": 2.1592626571655273, + "learning_rate": 3.1316003249390742e-06, + "loss": 1.8128, + "step": 1209 + }, + { + "epoch": 1.3260273972602739, + "grad_norm": 2.155649185180664, + "learning_rate": 3.129569455727051e-06, + "loss": 1.8636, + "step": 1210 + }, + { + "epoch": 1.3260273972602739, + "eval_loss": 1.9952574968338013, + "eval_runtime": 37.3004, + "eval_samples_per_second": 26.809, + "eval_steps_per_second": 1.126, + "step": 1210 + }, + { + "epoch": 1.3271232876712329, + "grad_norm": 2.1906397342681885, + "learning_rate": 3.1275385865150288e-06, + "loss": 1.8583, + "step": 1211 + }, + { + "epoch": 1.3282191780821917, + "grad_norm": 2.2708194255828857, + "learning_rate": 3.1255077173030056e-06, + "loss": 1.8363, + "step": 1212 + }, + { + "epoch": 1.3293150684931507, + "grad_norm": 2.1753969192504883, + "learning_rate": 3.1234768480909833e-06, + "loss": 1.8372, + "step": 1213 + }, + { + "epoch": 1.3304109589041095, + "grad_norm": 2.229363441467285, + "learning_rate": 3.12144597887896e-06, + "loss": 1.8179, + "step": 1214 + }, + { + "epoch": 1.3315068493150686, + "grad_norm": 2.362152338027954, + "learning_rate": 3.119415109666938e-06, + "loss": 1.8596, + "step": 1215 + }, + { + "epoch": 1.3326027397260274, + "grad_norm": 2.3278255462646484, + "learning_rate": 3.1173842404549147e-06, + "loss": 1.8778, + "step": 1216 + }, + { + "epoch": 1.3336986301369862, + "grad_norm": 2.477710247039795, + "learning_rate": 3.1153533712428924e-06, + "loss": 1.8297, + "step": 1217 + }, + { + "epoch": 1.3347945205479452, + "grad_norm": 2.2611043453216553, + "learning_rate": 3.1133225020308693e-06, + "loss": 1.8374, + "step": 1218 + }, + { + "epoch": 1.335890410958904, + "grad_norm": 2.198275566101074, + "learning_rate": 3.111291632818847e-06, + "loss": 1.8686, + "step": 1219 + }, + { + "epoch": 1.336986301369863, + "grad_norm": 2.227508068084717, + "learning_rate": 3.109260763606824e-06, + "loss": 1.8642, + "step": 1220 + }, + { + "epoch": 1.336986301369863, + "eval_loss": 1.9998422861099243, + "eval_runtime": 37.5152, + "eval_samples_per_second": 26.656, + "eval_steps_per_second": 1.12, + "step": 1220 + }, + { + "epoch": 1.3380821917808219, + "grad_norm": 2.386310338973999, + "learning_rate": 3.1072298943948015e-06, + "loss": 1.8347, + "step": 1221 + }, + { + "epoch": 1.339178082191781, + "grad_norm": 2.290313482284546, + "learning_rate": 3.1051990251827784e-06, + "loss": 1.8455, + "step": 1222 + }, + { + "epoch": 1.3402739726027397, + "grad_norm": 2.3063457012176514, + "learning_rate": 3.1031681559707557e-06, + "loss": 1.811, + "step": 1223 + }, + { + "epoch": 1.3413698630136985, + "grad_norm": 2.297840118408203, + "learning_rate": 3.101137286758733e-06, + "loss": 1.8558, + "step": 1224 + }, + { + "epoch": 1.3424657534246576, + "grad_norm": 2.40864896774292, + "learning_rate": 3.09910641754671e-06, + "loss": 1.8604, + "step": 1225 + }, + { + "epoch": 1.3435616438356164, + "grad_norm": 2.397599458694458, + "learning_rate": 3.0970755483346875e-06, + "loss": 1.8556, + "step": 1226 + }, + { + "epoch": 1.3446575342465754, + "grad_norm": 2.409029245376587, + "learning_rate": 3.0950446791226648e-06, + "loss": 1.8424, + "step": 1227 + }, + { + "epoch": 1.3457534246575342, + "grad_norm": 2.2789814472198486, + "learning_rate": 3.093013809910642e-06, + "loss": 1.8945, + "step": 1228 + }, + { + "epoch": 1.3468493150684933, + "grad_norm": 2.3749048709869385, + "learning_rate": 3.0909829406986193e-06, + "loss": 1.8387, + "step": 1229 + }, + { + "epoch": 1.347945205479452, + "grad_norm": 2.44486141204834, + "learning_rate": 3.0889520714865966e-06, + "loss": 1.8838, + "step": 1230 + }, + { + "epoch": 1.347945205479452, + "eval_loss": 1.9961321353912354, + "eval_runtime": 37.6273, + "eval_samples_per_second": 26.576, + "eval_steps_per_second": 1.116, + "step": 1230 + }, + { + "epoch": 1.3490410958904109, + "grad_norm": 2.2852954864501953, + "learning_rate": 3.086921202274574e-06, + "loss": 1.8551, + "step": 1231 + }, + { + "epoch": 1.35013698630137, + "grad_norm": 2.3131206035614014, + "learning_rate": 3.084890333062551e-06, + "loss": 1.8183, + "step": 1232 + }, + { + "epoch": 1.3512328767123287, + "grad_norm": 2.4095048904418945, + "learning_rate": 3.0828594638505284e-06, + "loss": 1.7946, + "step": 1233 + }, + { + "epoch": 1.3523287671232875, + "grad_norm": 2.429971933364868, + "learning_rate": 3.0808285946385057e-06, + "loss": 1.8349, + "step": 1234 + }, + { + "epoch": 1.3534246575342466, + "grad_norm": 2.2360565662384033, + "learning_rate": 3.078797725426483e-06, + "loss": 1.794, + "step": 1235 + }, + { + "epoch": 1.3545205479452056, + "grad_norm": 2.3630571365356445, + "learning_rate": 3.07676685621446e-06, + "loss": 1.7377, + "step": 1236 + }, + { + "epoch": 1.3556164383561644, + "grad_norm": 2.2604660987854004, + "learning_rate": 3.0747359870024375e-06, + "loss": 1.8239, + "step": 1237 + }, + { + "epoch": 1.3567123287671232, + "grad_norm": 2.209717035293579, + "learning_rate": 3.0727051177904144e-06, + "loss": 1.7835, + "step": 1238 + }, + { + "epoch": 1.3578082191780823, + "grad_norm": 2.3221793174743652, + "learning_rate": 3.070674248578392e-06, + "loss": 1.8192, + "step": 1239 + }, + { + "epoch": 1.358904109589041, + "grad_norm": 2.357313394546509, + "learning_rate": 3.068643379366369e-06, + "loss": 1.8246, + "step": 1240 + }, + { + "epoch": 1.358904109589041, + "eval_loss": 1.9935922622680664, + "eval_runtime": 37.2832, + "eval_samples_per_second": 26.822, + "eval_steps_per_second": 1.127, + "step": 1240 + }, + { + "epoch": 1.3599999999999999, + "grad_norm": 2.2563390731811523, + "learning_rate": 3.0666125101543466e-06, + "loss": 1.8175, + "step": 1241 + }, + { + "epoch": 1.361095890410959, + "grad_norm": 2.2531445026397705, + "learning_rate": 3.0645816409423235e-06, + "loss": 1.7713, + "step": 1242 + }, + { + "epoch": 1.362191780821918, + "grad_norm": 2.2369470596313477, + "learning_rate": 3.062550771730301e-06, + "loss": 1.8377, + "step": 1243 + }, + { + "epoch": 1.3632876712328768, + "grad_norm": 2.4063796997070312, + "learning_rate": 3.060519902518278e-06, + "loss": 1.8116, + "step": 1244 + }, + { + "epoch": 1.3643835616438356, + "grad_norm": 2.4628145694732666, + "learning_rate": 3.0584890333062557e-06, + "loss": 1.7998, + "step": 1245 + }, + { + "epoch": 1.3654794520547946, + "grad_norm": 2.3018743991851807, + "learning_rate": 3.0564581640942326e-06, + "loss": 1.7413, + "step": 1246 + }, + { + "epoch": 1.3665753424657534, + "grad_norm": 2.2432262897491455, + "learning_rate": 3.0544272948822094e-06, + "loss": 1.82, + "step": 1247 + }, + { + "epoch": 1.3676712328767122, + "grad_norm": 2.368435859680176, + "learning_rate": 3.052396425670187e-06, + "loss": 1.7976, + "step": 1248 + }, + { + "epoch": 1.3687671232876713, + "grad_norm": 2.2668025493621826, + "learning_rate": 3.050365556458164e-06, + "loss": 1.8379, + "step": 1249 + }, + { + "epoch": 1.36986301369863, + "grad_norm": 2.2622179985046387, + "learning_rate": 3.0483346872461417e-06, + "loss": 1.794, + "step": 1250 + }, + { + "epoch": 1.36986301369863, + "eval_loss": 1.9959099292755127, + "eval_runtime": 37.3358, + "eval_samples_per_second": 26.784, + "eval_steps_per_second": 1.125, + "step": 1250 + }, + { + "epoch": 1.370958904109589, + "grad_norm": 2.3414738178253174, + "learning_rate": 3.0463038180341185e-06, + "loss": 1.7993, + "step": 1251 + }, + { + "epoch": 1.372054794520548, + "grad_norm": 2.2678418159484863, + "learning_rate": 3.0442729488220962e-06, + "loss": 1.8581, + "step": 1252 + }, + { + "epoch": 1.373150684931507, + "grad_norm": 2.268195867538452, + "learning_rate": 3.042242079610073e-06, + "loss": 1.8486, + "step": 1253 + }, + { + "epoch": 1.3742465753424657, + "grad_norm": 2.230173110961914, + "learning_rate": 3.0402112103980508e-06, + "loss": 1.8468, + "step": 1254 + }, + { + "epoch": 1.3753424657534246, + "grad_norm": 2.3020238876342773, + "learning_rate": 3.0381803411860276e-06, + "loss": 1.7623, + "step": 1255 + }, + { + "epoch": 1.3764383561643836, + "grad_norm": 2.385343313217163, + "learning_rate": 3.0361494719740053e-06, + "loss": 1.7976, + "step": 1256 + }, + { + "epoch": 1.3775342465753424, + "grad_norm": 2.344768524169922, + "learning_rate": 3.034118602761982e-06, + "loss": 1.807, + "step": 1257 + }, + { + "epoch": 1.3786301369863014, + "grad_norm": 2.411760091781616, + "learning_rate": 3.03208773354996e-06, + "loss": 1.8079, + "step": 1258 + }, + { + "epoch": 1.3797260273972602, + "grad_norm": 2.3160128593444824, + "learning_rate": 3.0300568643379367e-06, + "loss": 1.7868, + "step": 1259 + }, + { + "epoch": 1.3808219178082193, + "grad_norm": 2.334632635116577, + "learning_rate": 3.028025995125914e-06, + "loss": 1.8466, + "step": 1260 + }, + { + "epoch": 1.3808219178082193, + "eval_loss": 1.9903327226638794, + "eval_runtime": 37.2256, + "eval_samples_per_second": 26.863, + "eval_steps_per_second": 1.128, + "step": 1260 + }, + { + "epoch": 1.381917808219178, + "grad_norm": 2.3443334102630615, + "learning_rate": 3.0259951259138913e-06, + "loss": 1.8302, + "step": 1261 + }, + { + "epoch": 1.383013698630137, + "grad_norm": 2.311563014984131, + "learning_rate": 3.0239642567018686e-06, + "loss": 1.7727, + "step": 1262 + }, + { + "epoch": 1.384109589041096, + "grad_norm": 2.2501776218414307, + "learning_rate": 3.021933387489846e-06, + "loss": 1.8058, + "step": 1263 + }, + { + "epoch": 1.3852054794520547, + "grad_norm": 2.259157657623291, + "learning_rate": 3.019902518277823e-06, + "loss": 1.7627, + "step": 1264 + }, + { + "epoch": 1.3863013698630138, + "grad_norm": 2.4380314350128174, + "learning_rate": 3.0178716490658004e-06, + "loss": 1.8237, + "step": 1265 + }, + { + "epoch": 1.3873972602739726, + "grad_norm": 2.3203837871551514, + "learning_rate": 3.0158407798537777e-06, + "loss": 1.8557, + "step": 1266 + }, + { + "epoch": 1.3884931506849316, + "grad_norm": 2.220996141433716, + "learning_rate": 3.013809910641755e-06, + "loss": 1.7641, + "step": 1267 + }, + { + "epoch": 1.3895890410958904, + "grad_norm": 2.190006971359253, + "learning_rate": 3.0117790414297322e-06, + "loss": 1.7705, + "step": 1268 + }, + { + "epoch": 1.3906849315068492, + "grad_norm": 2.229748249053955, + "learning_rate": 3.0097481722177095e-06, + "loss": 1.7709, + "step": 1269 + }, + { + "epoch": 1.3917808219178083, + "grad_norm": 2.3755619525909424, + "learning_rate": 3.0077173030056868e-06, + "loss": 1.8172, + "step": 1270 + }, + { + "epoch": 1.3917808219178083, + "eval_loss": 1.992952585220337, + "eval_runtime": 37.2403, + "eval_samples_per_second": 26.853, + "eval_steps_per_second": 1.128, + "step": 1270 + }, + { + "epoch": 1.392876712328767, + "grad_norm": 2.2514004707336426, + "learning_rate": 3.005686433793664e-06, + "loss": 1.7719, + "step": 1271 + }, + { + "epoch": 1.393972602739726, + "grad_norm": 2.3010191917419434, + "learning_rate": 3.0036555645816413e-06, + "loss": 1.8369, + "step": 1272 + }, + { + "epoch": 1.395068493150685, + "grad_norm": 2.68790602684021, + "learning_rate": 3.001624695369618e-06, + "loss": 1.7974, + "step": 1273 + }, + { + "epoch": 1.396164383561644, + "grad_norm": 2.454904794692993, + "learning_rate": 2.999593826157596e-06, + "loss": 1.8509, + "step": 1274 + }, + { + "epoch": 1.3972602739726028, + "grad_norm": 2.251556396484375, + "learning_rate": 2.9975629569455727e-06, + "loss": 1.8011, + "step": 1275 + }, + { + "epoch": 1.3983561643835616, + "grad_norm": 2.3789098262786865, + "learning_rate": 2.9955320877335504e-06, + "loss": 1.8488, + "step": 1276 + }, + { + "epoch": 1.3994520547945206, + "grad_norm": 2.2659196853637695, + "learning_rate": 2.9935012185215273e-06, + "loss": 1.7946, + "step": 1277 + }, + { + "epoch": 1.4005479452054794, + "grad_norm": 2.2148971557617188, + "learning_rate": 2.991470349309505e-06, + "loss": 1.8388, + "step": 1278 + }, + { + "epoch": 1.4016438356164382, + "grad_norm": 2.1883089542388916, + "learning_rate": 2.989439480097482e-06, + "loss": 1.7925, + "step": 1279 + }, + { + "epoch": 1.4027397260273973, + "grad_norm": 2.290208339691162, + "learning_rate": 2.9874086108854595e-06, + "loss": 1.8099, + "step": 1280 + }, + { + "epoch": 1.4027397260273973, + "eval_loss": 1.9924782514572144, + "eval_runtime": 37.1891, + "eval_samples_per_second": 26.89, + "eval_steps_per_second": 1.129, + "step": 1280 + }, + { + "epoch": 1.403835616438356, + "grad_norm": 2.121649980545044, + "learning_rate": 2.9853777416734364e-06, + "loss": 1.7991, + "step": 1281 + }, + { + "epoch": 1.4049315068493151, + "grad_norm": 2.3528690338134766, + "learning_rate": 2.983346872461414e-06, + "loss": 1.8487, + "step": 1282 + }, + { + "epoch": 1.406027397260274, + "grad_norm": 2.391491174697876, + "learning_rate": 2.981316003249391e-06, + "loss": 1.8564, + "step": 1283 + }, + { + "epoch": 1.407123287671233, + "grad_norm": 2.336615800857544, + "learning_rate": 2.9792851340373678e-06, + "loss": 1.8216, + "step": 1284 + }, + { + "epoch": 1.4082191780821918, + "grad_norm": 2.2958548069000244, + "learning_rate": 2.9772542648253455e-06, + "loss": 1.8003, + "step": 1285 + }, + { + "epoch": 1.4093150684931506, + "grad_norm": 2.2062184810638428, + "learning_rate": 2.9752233956133223e-06, + "loss": 1.8488, + "step": 1286 + }, + { + "epoch": 1.4104109589041096, + "grad_norm": 2.486133575439453, + "learning_rate": 2.9731925264013e-06, + "loss": 1.7914, + "step": 1287 + }, + { + "epoch": 1.4115068493150684, + "grad_norm": 2.2933201789855957, + "learning_rate": 2.971161657189277e-06, + "loss": 1.8164, + "step": 1288 + }, + { + "epoch": 1.4126027397260275, + "grad_norm": 2.3301920890808105, + "learning_rate": 2.9691307879772546e-06, + "loss": 1.7883, + "step": 1289 + }, + { + "epoch": 1.4136986301369863, + "grad_norm": 2.248832941055298, + "learning_rate": 2.9670999187652314e-06, + "loss": 1.7916, + "step": 1290 + }, + { + "epoch": 1.4136986301369863, + "eval_loss": 1.9919097423553467, + "eval_runtime": 37.1283, + "eval_samples_per_second": 26.934, + "eval_steps_per_second": 1.131, + "step": 1290 + }, + { + "epoch": 1.4147945205479453, + "grad_norm": 2.37760853767395, + "learning_rate": 2.965069049553209e-06, + "loss": 1.7854, + "step": 1291 + }, + { + "epoch": 1.415890410958904, + "grad_norm": 2.327718734741211, + "learning_rate": 2.963038180341186e-06, + "loss": 1.874, + "step": 1292 + }, + { + "epoch": 1.416986301369863, + "grad_norm": 2.2413878440856934, + "learning_rate": 2.9610073111291637e-06, + "loss": 1.7591, + "step": 1293 + }, + { + "epoch": 1.418082191780822, + "grad_norm": 2.2203164100646973, + "learning_rate": 2.9589764419171405e-06, + "loss": 1.8466, + "step": 1294 + }, + { + "epoch": 1.4191780821917808, + "grad_norm": 2.2070505619049072, + "learning_rate": 2.9569455727051182e-06, + "loss": 1.8689, + "step": 1295 + }, + { + "epoch": 1.4202739726027398, + "grad_norm": 2.2122223377227783, + "learning_rate": 2.954914703493095e-06, + "loss": 1.7689, + "step": 1296 + }, + { + "epoch": 1.4213698630136986, + "grad_norm": 2.331303358078003, + "learning_rate": 2.9528838342810724e-06, + "loss": 1.8089, + "step": 1297 + }, + { + "epoch": 1.4224657534246576, + "grad_norm": 2.408184289932251, + "learning_rate": 2.9508529650690496e-06, + "loss": 1.7847, + "step": 1298 + }, + { + "epoch": 1.4235616438356165, + "grad_norm": 2.212766170501709, + "learning_rate": 2.948822095857027e-06, + "loss": 1.7899, + "step": 1299 + }, + { + "epoch": 1.4246575342465753, + "grad_norm": 2.259953737258911, + "learning_rate": 2.946791226645004e-06, + "loss": 1.8435, + "step": 1300 + }, + { + "epoch": 1.4246575342465753, + "eval_loss": 1.9893293380737305, + "eval_runtime": 37.1675, + "eval_samples_per_second": 26.905, + "eval_steps_per_second": 1.13, + "step": 1300 + }, + { + "epoch": 1.4257534246575343, + "grad_norm": 2.342400312423706, + "learning_rate": 2.9447603574329815e-06, + "loss": 1.7779, + "step": 1301 + }, + { + "epoch": 1.426849315068493, + "grad_norm": 2.237618923187256, + "learning_rate": 2.9427294882209587e-06, + "loss": 1.7509, + "step": 1302 + }, + { + "epoch": 1.427945205479452, + "grad_norm": 2.3418633937835693, + "learning_rate": 2.940698619008936e-06, + "loss": 1.7963, + "step": 1303 + }, + { + "epoch": 1.429041095890411, + "grad_norm": 2.470827579498291, + "learning_rate": 2.9386677497969133e-06, + "loss": 1.8045, + "step": 1304 + }, + { + "epoch": 1.43013698630137, + "grad_norm": 2.508345365524292, + "learning_rate": 2.9366368805848906e-06, + "loss": 1.7926, + "step": 1305 + }, + { + "epoch": 1.4312328767123288, + "grad_norm": 2.261319160461426, + "learning_rate": 2.934606011372868e-06, + "loss": 1.831, + "step": 1306 + }, + { + "epoch": 1.4323287671232876, + "grad_norm": 2.2703914642333984, + "learning_rate": 2.932575142160845e-06, + "loss": 1.8352, + "step": 1307 + }, + { + "epoch": 1.4334246575342466, + "grad_norm": 2.243626117706299, + "learning_rate": 2.9305442729488224e-06, + "loss": 1.8162, + "step": 1308 + }, + { + "epoch": 1.4345205479452054, + "grad_norm": 2.3859238624572754, + "learning_rate": 2.9285134037367997e-06, + "loss": 1.8171, + "step": 1309 + }, + { + "epoch": 1.4356164383561643, + "grad_norm": 2.3350934982299805, + "learning_rate": 2.9264825345247765e-06, + "loss": 1.8066, + "step": 1310 + }, + { + "epoch": 1.4356164383561643, + "eval_loss": 1.9905259609222412, + "eval_runtime": 37.2009, + "eval_samples_per_second": 26.881, + "eval_steps_per_second": 1.129, + "step": 1310 + }, + { + "epoch": 1.4367123287671233, + "grad_norm": 2.1943347454071045, + "learning_rate": 2.9244516653127542e-06, + "loss": 1.8435, + "step": 1311 + }, + { + "epoch": 1.4378082191780823, + "grad_norm": 2.429128885269165, + "learning_rate": 2.922420796100731e-06, + "loss": 1.8179, + "step": 1312 + }, + { + "epoch": 1.4389041095890411, + "grad_norm": 2.3677234649658203, + "learning_rate": 2.9203899268887088e-06, + "loss": 1.7245, + "step": 1313 + }, + { + "epoch": 1.44, + "grad_norm": 2.255570888519287, + "learning_rate": 2.9183590576766856e-06, + "loss": 1.8545, + "step": 1314 + }, + { + "epoch": 1.441095890410959, + "grad_norm": 2.5176682472229004, + "learning_rate": 2.9163281884646633e-06, + "loss": 1.7509, + "step": 1315 + }, + { + "epoch": 1.4421917808219178, + "grad_norm": 2.29972767829895, + "learning_rate": 2.91429731925264e-06, + "loss": 1.8116, + "step": 1316 + }, + { + "epoch": 1.4432876712328766, + "grad_norm": 2.2853286266326904, + "learning_rate": 2.912266450040618e-06, + "loss": 1.8552, + "step": 1317 + }, + { + "epoch": 1.4443835616438356, + "grad_norm": 2.505075216293335, + "learning_rate": 2.9102355808285947e-06, + "loss": 1.811, + "step": 1318 + }, + { + "epoch": 1.4454794520547944, + "grad_norm": 2.229064464569092, + "learning_rate": 2.9082047116165724e-06, + "loss": 1.8068, + "step": 1319 + }, + { + "epoch": 1.4465753424657535, + "grad_norm": 2.199187994003296, + "learning_rate": 2.9061738424045493e-06, + "loss": 1.7825, + "step": 1320 + }, + { + "epoch": 1.4465753424657535, + "eval_loss": 1.9851222038269043, + "eval_runtime": 37.4614, + "eval_samples_per_second": 26.694, + "eval_steps_per_second": 1.121, + "step": 1320 + }, + { + "epoch": 1.4476712328767123, + "grad_norm": 2.4318959712982178, + "learning_rate": 2.904142973192527e-06, + "loss": 1.77, + "step": 1321 + }, + { + "epoch": 1.4487671232876713, + "grad_norm": 2.480703592300415, + "learning_rate": 2.902112103980504e-06, + "loss": 1.8774, + "step": 1322 + }, + { + "epoch": 1.4498630136986301, + "grad_norm": 2.367624521255493, + "learning_rate": 2.900081234768481e-06, + "loss": 1.818, + "step": 1323 + }, + { + "epoch": 1.450958904109589, + "grad_norm": 2.245974540710449, + "learning_rate": 2.8980503655564584e-06, + "loss": 1.8054, + "step": 1324 + }, + { + "epoch": 1.452054794520548, + "grad_norm": 2.352687120437622, + "learning_rate": 2.8960194963444357e-06, + "loss": 1.7922, + "step": 1325 + }, + { + "epoch": 1.4531506849315068, + "grad_norm": 2.308253526687622, + "learning_rate": 2.893988627132413e-06, + "loss": 1.7637, + "step": 1326 + }, + { + "epoch": 1.4542465753424658, + "grad_norm": 2.309495687484741, + "learning_rate": 2.8919577579203902e-06, + "loss": 1.7905, + "step": 1327 + }, + { + "epoch": 1.4553424657534246, + "grad_norm": 2.354365348815918, + "learning_rate": 2.8899268887083675e-06, + "loss": 1.7321, + "step": 1328 + }, + { + "epoch": 1.4564383561643837, + "grad_norm": 2.3886210918426514, + "learning_rate": 2.8878960194963448e-06, + "loss": 1.7932, + "step": 1329 + }, + { + "epoch": 1.4575342465753425, + "grad_norm": 2.395714044570923, + "learning_rate": 2.885865150284322e-06, + "loss": 1.813, + "step": 1330 + }, + { + "epoch": 1.4575342465753425, + "eval_loss": 1.9851990938186646, + "eval_runtime": 37.2966, + "eval_samples_per_second": 26.812, + "eval_steps_per_second": 1.126, + "step": 1330 + }, + { + "epoch": 1.4586301369863013, + "grad_norm": 2.433069944381714, + "learning_rate": 2.8838342810722993e-06, + "loss": 1.8346, + "step": 1331 + }, + { + "epoch": 1.4597260273972603, + "grad_norm": 2.295894145965576, + "learning_rate": 2.8818034118602766e-06, + "loss": 1.7855, + "step": 1332 + }, + { + "epoch": 1.4608219178082191, + "grad_norm": 2.495742082595825, + "learning_rate": 2.879772542648254e-06, + "loss": 1.7601, + "step": 1333 + }, + { + "epoch": 1.4619178082191782, + "grad_norm": 2.35984206199646, + "learning_rate": 2.8777416734362307e-06, + "loss": 1.8329, + "step": 1334 + }, + { + "epoch": 1.463013698630137, + "grad_norm": 2.39632248878479, + "learning_rate": 2.8757108042242084e-06, + "loss": 1.7426, + "step": 1335 + }, + { + "epoch": 1.464109589041096, + "grad_norm": 2.4151153564453125, + "learning_rate": 2.8736799350121853e-06, + "loss": 1.8899, + "step": 1336 + }, + { + "epoch": 1.4652054794520548, + "grad_norm": 2.43914532661438, + "learning_rate": 2.871649065800163e-06, + "loss": 1.8464, + "step": 1337 + }, + { + "epoch": 1.4663013698630136, + "grad_norm": 2.44490122795105, + "learning_rate": 2.86961819658814e-06, + "loss": 1.8643, + "step": 1338 + }, + { + "epoch": 1.4673972602739727, + "grad_norm": 2.2639644145965576, + "learning_rate": 2.8675873273761175e-06, + "loss": 1.792, + "step": 1339 + }, + { + "epoch": 1.4684931506849315, + "grad_norm": 2.2803847789764404, + "learning_rate": 2.8655564581640944e-06, + "loss": 1.8109, + "step": 1340 + }, + { + "epoch": 1.4684931506849315, + "eval_loss": 1.9856781959533691, + "eval_runtime": 37.2519, + "eval_samples_per_second": 26.844, + "eval_steps_per_second": 1.127, + "step": 1340 + }, + { + "epoch": 1.4695890410958903, + "grad_norm": 2.494727849960327, + "learning_rate": 2.863525588952072e-06, + "loss": 1.8032, + "step": 1341 + }, + { + "epoch": 1.4706849315068493, + "grad_norm": 2.239595413208008, + "learning_rate": 2.861494719740049e-06, + "loss": 1.8659, + "step": 1342 + }, + { + "epoch": 1.4717808219178083, + "grad_norm": 2.2922589778900146, + "learning_rate": 2.8594638505280266e-06, + "loss": 1.7587, + "step": 1343 + }, + { + "epoch": 1.4728767123287672, + "grad_norm": 2.345224142074585, + "learning_rate": 2.8574329813160035e-06, + "loss": 1.7602, + "step": 1344 + }, + { + "epoch": 1.473972602739726, + "grad_norm": 2.3415307998657227, + "learning_rate": 2.855402112103981e-06, + "loss": 1.8044, + "step": 1345 + }, + { + "epoch": 1.475068493150685, + "grad_norm": 2.331557035446167, + "learning_rate": 2.853371242891958e-06, + "loss": 1.7808, + "step": 1346 + }, + { + "epoch": 1.4761643835616438, + "grad_norm": 2.293572425842285, + "learning_rate": 2.851340373679935e-06, + "loss": 1.8017, + "step": 1347 + }, + { + "epoch": 1.4772602739726026, + "grad_norm": 2.3477773666381836, + "learning_rate": 2.8493095044679126e-06, + "loss": 1.8113, + "step": 1348 + }, + { + "epoch": 1.4783561643835617, + "grad_norm": 2.3740994930267334, + "learning_rate": 2.8472786352558894e-06, + "loss": 1.8801, + "step": 1349 + }, + { + "epoch": 1.4794520547945205, + "grad_norm": 2.4202206134796143, + "learning_rate": 2.845247766043867e-06, + "loss": 1.821, + "step": 1350 + }, + { + "epoch": 1.4794520547945205, + "eval_loss": 1.9867810010910034, + "eval_runtime": 37.2597, + "eval_samples_per_second": 26.839, + "eval_steps_per_second": 1.127, + "step": 1350 + }, + { + "epoch": 1.4805479452054795, + "grad_norm": 2.2323198318481445, + "learning_rate": 2.843216896831844e-06, + "loss": 1.7574, + "step": 1351 + }, + { + "epoch": 1.4816438356164383, + "grad_norm": 2.2192625999450684, + "learning_rate": 2.8411860276198217e-06, + "loss": 1.7662, + "step": 1352 + }, + { + "epoch": 1.4827397260273973, + "grad_norm": 2.276507616043091, + "learning_rate": 2.8391551584077985e-06, + "loss": 1.7748, + "step": 1353 + }, + { + "epoch": 1.4838356164383562, + "grad_norm": 2.292773723602295, + "learning_rate": 2.8371242891957762e-06, + "loss": 1.8388, + "step": 1354 + }, + { + "epoch": 1.484931506849315, + "grad_norm": 2.3127009868621826, + "learning_rate": 2.835093419983753e-06, + "loss": 1.861, + "step": 1355 + }, + { + "epoch": 1.486027397260274, + "grad_norm": 2.2752184867858887, + "learning_rate": 2.833062550771731e-06, + "loss": 1.7883, + "step": 1356 + }, + { + "epoch": 1.4871232876712328, + "grad_norm": 2.288158416748047, + "learning_rate": 2.8310316815597076e-06, + "loss": 1.7963, + "step": 1357 + }, + { + "epoch": 1.4882191780821918, + "grad_norm": 2.3287482261657715, + "learning_rate": 2.8290008123476853e-06, + "loss": 1.7534, + "step": 1358 + }, + { + "epoch": 1.4893150684931507, + "grad_norm": 2.441279888153076, + "learning_rate": 2.826969943135662e-06, + "loss": 1.8073, + "step": 1359 + }, + { + "epoch": 1.4904109589041097, + "grad_norm": 2.3399839401245117, + "learning_rate": 2.8249390739236395e-06, + "loss": 1.7876, + "step": 1360 + }, + { + "epoch": 1.4904109589041097, + "eval_loss": 1.9868030548095703, + "eval_runtime": 37.2891, + "eval_samples_per_second": 26.817, + "eval_steps_per_second": 1.126, + "step": 1360 + }, + { + "epoch": 1.4915068493150685, + "grad_norm": 2.234246253967285, + "learning_rate": 2.8229082047116167e-06, + "loss": 1.8031, + "step": 1361 + }, + { + "epoch": 1.4926027397260273, + "grad_norm": 2.3136820793151855, + "learning_rate": 2.820877335499594e-06, + "loss": 1.7669, + "step": 1362 + }, + { + "epoch": 1.4936986301369863, + "grad_norm": 2.389420747756958, + "learning_rate": 2.8188464662875713e-06, + "loss": 1.7879, + "step": 1363 + }, + { + "epoch": 1.4947945205479451, + "grad_norm": 2.222306489944458, + "learning_rate": 2.8168155970755486e-06, + "loss": 1.7412, + "step": 1364 + }, + { + "epoch": 1.4958904109589042, + "grad_norm": 2.1689319610595703, + "learning_rate": 2.814784727863526e-06, + "loss": 1.7651, + "step": 1365 + }, + { + "epoch": 1.496986301369863, + "grad_norm": 2.262047529220581, + "learning_rate": 2.812753858651503e-06, + "loss": 1.7798, + "step": 1366 + }, + { + "epoch": 1.498082191780822, + "grad_norm": 2.266894817352295, + "learning_rate": 2.8107229894394804e-06, + "loss": 1.8172, + "step": 1367 + }, + { + "epoch": 1.4991780821917808, + "grad_norm": 2.6400444507598877, + "learning_rate": 2.8086921202274577e-06, + "loss": 1.7871, + "step": 1368 + }, + { + "epoch": 1.5002739726027396, + "grad_norm": 2.3868865966796875, + "learning_rate": 2.806661251015435e-06, + "loss": 1.7427, + "step": 1369 + }, + { + "epoch": 1.5013698630136987, + "grad_norm": 2.5206542015075684, + "learning_rate": 2.8046303818034122e-06, + "loss": 1.817, + "step": 1370 + }, + { + "epoch": 1.5013698630136987, + "eval_loss": 1.9870396852493286, + "eval_runtime": 37.2372, + "eval_samples_per_second": 26.855, + "eval_steps_per_second": 1.128, + "step": 1370 + }, + { + "epoch": 1.5024657534246575, + "grad_norm": 2.5236382484436035, + "learning_rate": 2.802599512591389e-06, + "loss": 1.8662, + "step": 1371 + }, + { + "epoch": 1.5035616438356163, + "grad_norm": 2.5095479488372803, + "learning_rate": 2.8005686433793668e-06, + "loss": 1.9168, + "step": 1372 + }, + { + "epoch": 1.5046575342465753, + "grad_norm": 2.307508707046509, + "learning_rate": 2.7985377741673436e-06, + "loss": 1.799, + "step": 1373 + }, + { + "epoch": 1.5057534246575344, + "grad_norm": 2.410879373550415, + "learning_rate": 2.7965069049553213e-06, + "loss": 1.7447, + "step": 1374 + }, + { + "epoch": 1.5068493150684932, + "grad_norm": 2.738558530807495, + "learning_rate": 2.794476035743298e-06, + "loss": 1.8073, + "step": 1375 + }, + { + "epoch": 1.507945205479452, + "grad_norm": 2.3826518058776855, + "learning_rate": 2.792445166531276e-06, + "loss": 1.8506, + "step": 1376 + }, + { + "epoch": 1.509041095890411, + "grad_norm": 2.334123373031616, + "learning_rate": 2.7904142973192527e-06, + "loss": 1.8209, + "step": 1377 + }, + { + "epoch": 1.5101369863013698, + "grad_norm": 2.668513298034668, + "learning_rate": 2.7883834281072304e-06, + "loss": 1.8285, + "step": 1378 + }, + { + "epoch": 1.5112328767123286, + "grad_norm": 2.471938133239746, + "learning_rate": 2.7863525588952073e-06, + "loss": 1.7668, + "step": 1379 + }, + { + "epoch": 1.5123287671232877, + "grad_norm": 2.3712313175201416, + "learning_rate": 2.784321689683185e-06, + "loss": 1.7565, + "step": 1380 + }, + { + "epoch": 1.5123287671232877, + "eval_loss": 1.9826414585113525, + "eval_runtime": 37.2426, + "eval_samples_per_second": 26.851, + "eval_steps_per_second": 1.128, + "step": 1380 + }, + { + "epoch": 1.5134246575342467, + "grad_norm": 2.3364150524139404, + "learning_rate": 2.782290820471162e-06, + "loss": 1.7596, + "step": 1381 + }, + { + "epoch": 1.5145205479452055, + "grad_norm": 2.3088393211364746, + "learning_rate": 2.7802599512591395e-06, + "loss": 1.7568, + "step": 1382 + }, + { + "epoch": 1.5156164383561643, + "grad_norm": 2.2584595680236816, + "learning_rate": 2.7782290820471164e-06, + "loss": 1.833, + "step": 1383 + }, + { + "epoch": 1.5167123287671234, + "grad_norm": 2.2440273761749268, + "learning_rate": 2.7761982128350932e-06, + "loss": 1.8237, + "step": 1384 + }, + { + "epoch": 1.5178082191780822, + "grad_norm": 2.3552212715148926, + "learning_rate": 2.774167343623071e-06, + "loss": 1.8288, + "step": 1385 + }, + { + "epoch": 1.518904109589041, + "grad_norm": 2.3001790046691895, + "learning_rate": 2.772136474411048e-06, + "loss": 1.8519, + "step": 1386 + }, + { + "epoch": 1.52, + "grad_norm": 2.3691201210021973, + "learning_rate": 2.7701056051990255e-06, + "loss": 1.7759, + "step": 1387 + }, + { + "epoch": 1.521095890410959, + "grad_norm": 2.327988624572754, + "learning_rate": 2.7680747359870023e-06, + "loss": 1.7132, + "step": 1388 + }, + { + "epoch": 1.5221917808219179, + "grad_norm": 2.30245304107666, + "learning_rate": 2.76604386677498e-06, + "loss": 1.8167, + "step": 1389 + }, + { + "epoch": 1.5232876712328767, + "grad_norm": 2.290121078491211, + "learning_rate": 2.764012997562957e-06, + "loss": 1.7913, + "step": 1390 + }, + { + "epoch": 1.5232876712328767, + "eval_loss": 1.98329758644104, + "eval_runtime": 37.2462, + "eval_samples_per_second": 26.848, + "eval_steps_per_second": 1.128, + "step": 1390 + }, + { + "epoch": 1.5243835616438357, + "grad_norm": 2.294076681137085, + "learning_rate": 2.7619821283509346e-06, + "loss": 1.8598, + "step": 1391 + }, + { + "epoch": 1.5254794520547945, + "grad_norm": 2.3325250148773193, + "learning_rate": 2.7599512591389114e-06, + "loss": 1.7392, + "step": 1392 + }, + { + "epoch": 1.5265753424657533, + "grad_norm": 2.2802610397338867, + "learning_rate": 2.757920389926889e-06, + "loss": 1.8332, + "step": 1393 + }, + { + "epoch": 1.5276712328767124, + "grad_norm": 2.284207344055176, + "learning_rate": 2.755889520714866e-06, + "loss": 1.7875, + "step": 1394 + }, + { + "epoch": 1.5287671232876714, + "grad_norm": 2.325998306274414, + "learning_rate": 2.7538586515028437e-06, + "loss": 1.7657, + "step": 1395 + }, + { + "epoch": 1.52986301369863, + "grad_norm": 2.285393476486206, + "learning_rate": 2.7518277822908205e-06, + "loss": 1.7416, + "step": 1396 + }, + { + "epoch": 1.530958904109589, + "grad_norm": 2.3110482692718506, + "learning_rate": 2.749796913078798e-06, + "loss": 1.7931, + "step": 1397 + }, + { + "epoch": 1.532054794520548, + "grad_norm": 2.4234025478363037, + "learning_rate": 2.747766043866775e-06, + "loss": 1.8472, + "step": 1398 + }, + { + "epoch": 1.5331506849315069, + "grad_norm": 2.384767532348633, + "learning_rate": 2.7457351746547524e-06, + "loss": 1.7961, + "step": 1399 + }, + { + "epoch": 1.5342465753424657, + "grad_norm": 2.1857433319091797, + "learning_rate": 2.7437043054427297e-06, + "loss": 1.8654, + "step": 1400 + }, + { + "epoch": 1.5342465753424657, + "eval_loss": 1.9802658557891846, + "eval_runtime": 37.1443, + "eval_samples_per_second": 26.922, + "eval_steps_per_second": 1.131, + "step": 1400 + }, + { + "epoch": 1.5353424657534247, + "grad_norm": 2.1302895545959473, + "learning_rate": 2.741673436230707e-06, + "loss": 1.9051, + "step": 1401 + }, + { + "epoch": 1.5364383561643835, + "grad_norm": 2.377307415008545, + "learning_rate": 2.739642567018684e-06, + "loss": 1.7942, + "step": 1402 + }, + { + "epoch": 1.5375342465753423, + "grad_norm": 2.316549062728882, + "learning_rate": 2.7376116978066615e-06, + "loss": 1.7734, + "step": 1403 + }, + { + "epoch": 1.5386301369863014, + "grad_norm": 2.3452532291412354, + "learning_rate": 2.7355808285946388e-06, + "loss": 1.8243, + "step": 1404 + }, + { + "epoch": 1.5397260273972604, + "grad_norm": 2.3139419555664062, + "learning_rate": 2.733549959382616e-06, + "loss": 1.812, + "step": 1405 + }, + { + "epoch": 1.5408219178082192, + "grad_norm": 2.3135571479797363, + "learning_rate": 2.7315190901705933e-06, + "loss": 1.7751, + "step": 1406 + }, + { + "epoch": 1.541917808219178, + "grad_norm": 2.403231620788574, + "learning_rate": 2.7294882209585706e-06, + "loss": 1.8561, + "step": 1407 + }, + { + "epoch": 1.543013698630137, + "grad_norm": 2.4028689861297607, + "learning_rate": 2.727457351746548e-06, + "loss": 1.8249, + "step": 1408 + }, + { + "epoch": 1.5441095890410959, + "grad_norm": 2.4028851985931396, + "learning_rate": 2.725426482534525e-06, + "loss": 1.779, + "step": 1409 + }, + { + "epoch": 1.5452054794520547, + "grad_norm": 2.270709753036499, + "learning_rate": 2.723395613322502e-06, + "loss": 1.7754, + "step": 1410 + }, + { + "epoch": 1.5452054794520547, + "eval_loss": 1.983272910118103, + "eval_runtime": 37.2405, + "eval_samples_per_second": 26.853, + "eval_steps_per_second": 1.128, + "step": 1410 + }, + { + "epoch": 1.5463013698630137, + "grad_norm": 2.2672812938690186, + "learning_rate": 2.7213647441104797e-06, + "loss": 1.814, + "step": 1411 + }, + { + "epoch": 1.5473972602739727, + "grad_norm": 2.3539397716522217, + "learning_rate": 2.7193338748984565e-06, + "loss": 1.7907, + "step": 1412 + }, + { + "epoch": 1.5484931506849315, + "grad_norm": 2.3229010105133057, + "learning_rate": 2.7173030056864342e-06, + "loss": 1.8061, + "step": 1413 + }, + { + "epoch": 1.5495890410958904, + "grad_norm": 2.482874870300293, + "learning_rate": 2.715272136474411e-06, + "loss": 1.9577, + "step": 1414 + }, + { + "epoch": 1.5506849315068494, + "grad_norm": 2.211798906326294, + "learning_rate": 2.713241267262389e-06, + "loss": 1.7309, + "step": 1415 + }, + { + "epoch": 1.5517808219178082, + "grad_norm": 2.3159289360046387, + "learning_rate": 2.7112103980503656e-06, + "loss": 1.8132, + "step": 1416 + }, + { + "epoch": 1.552876712328767, + "grad_norm": 2.486035108566284, + "learning_rate": 2.7091795288383433e-06, + "loss": 1.7734, + "step": 1417 + }, + { + "epoch": 1.553972602739726, + "grad_norm": 2.3025195598602295, + "learning_rate": 2.70714865962632e-06, + "loss": 1.8232, + "step": 1418 + }, + { + "epoch": 1.555068493150685, + "grad_norm": 2.234581470489502, + "learning_rate": 2.705117790414298e-06, + "loss": 1.7814, + "step": 1419 + }, + { + "epoch": 1.5561643835616439, + "grad_norm": 2.2855048179626465, + "learning_rate": 2.7030869212022747e-06, + "loss": 1.777, + "step": 1420 + }, + { + "epoch": 1.5561643835616439, + "eval_loss": 1.9851160049438477, + "eval_runtime": 37.4606, + "eval_samples_per_second": 26.695, + "eval_steps_per_second": 1.121, + "step": 1420 + }, + { + "epoch": 1.5572602739726027, + "grad_norm": 2.4190328121185303, + "learning_rate": 2.7010560519902516e-06, + "loss": 1.8456, + "step": 1421 + }, + { + "epoch": 1.5583561643835617, + "grad_norm": 2.225623369216919, + "learning_rate": 2.6990251827782293e-06, + "loss": 1.7638, + "step": 1422 + }, + { + "epoch": 1.5594520547945205, + "grad_norm": 2.1516661643981934, + "learning_rate": 2.696994313566206e-06, + "loss": 1.7861, + "step": 1423 + }, + { + "epoch": 1.5605479452054793, + "grad_norm": 2.2461135387420654, + "learning_rate": 2.694963444354184e-06, + "loss": 1.7763, + "step": 1424 + }, + { + "epoch": 1.5616438356164384, + "grad_norm": 2.280869245529175, + "learning_rate": 2.6929325751421607e-06, + "loss": 1.8589, + "step": 1425 + }, + { + "epoch": 1.5627397260273974, + "grad_norm": 2.2034990787506104, + "learning_rate": 2.6909017059301384e-06, + "loss": 1.7705, + "step": 1426 + }, + { + "epoch": 1.563835616438356, + "grad_norm": 2.1346969604492188, + "learning_rate": 2.6888708367181153e-06, + "loss": 1.8318, + "step": 1427 + }, + { + "epoch": 1.564931506849315, + "grad_norm": 2.37400221824646, + "learning_rate": 2.686839967506093e-06, + "loss": 1.746, + "step": 1428 + }, + { + "epoch": 1.566027397260274, + "grad_norm": 2.392561435699463, + "learning_rate": 2.68480909829407e-06, + "loss": 1.8105, + "step": 1429 + }, + { + "epoch": 1.5671232876712329, + "grad_norm": 2.299830198287964, + "learning_rate": 2.6827782290820475e-06, + "loss": 1.7103, + "step": 1430 + }, + { + "epoch": 1.5671232876712329, + "eval_loss": 1.9877182245254517, + "eval_runtime": 37.4813, + "eval_samples_per_second": 26.68, + "eval_steps_per_second": 1.121, + "step": 1430 + }, + { + "epoch": 1.5682191780821917, + "grad_norm": 2.3431026935577393, + "learning_rate": 2.6807473598700244e-06, + "loss": 1.7217, + "step": 1431 + }, + { + "epoch": 1.5693150684931507, + "grad_norm": 2.3474628925323486, + "learning_rate": 2.678716490658002e-06, + "loss": 1.8419, + "step": 1432 + }, + { + "epoch": 1.5704109589041098, + "grad_norm": 2.2877044677734375, + "learning_rate": 2.676685621445979e-06, + "loss": 1.8031, + "step": 1433 + }, + { + "epoch": 1.5715068493150683, + "grad_norm": 2.278613805770874, + "learning_rate": 2.674654752233956e-06, + "loss": 1.8066, + "step": 1434 + }, + { + "epoch": 1.5726027397260274, + "grad_norm": 2.3357837200164795, + "learning_rate": 2.6726238830219335e-06, + "loss": 1.8079, + "step": 1435 + }, + { + "epoch": 1.5736986301369864, + "grad_norm": 2.1932482719421387, + "learning_rate": 2.6705930138099107e-06, + "loss": 1.8032, + "step": 1436 + }, + { + "epoch": 1.5747945205479452, + "grad_norm": 2.298553705215454, + "learning_rate": 2.668562144597888e-06, + "loss": 1.8439, + "step": 1437 + }, + { + "epoch": 1.575890410958904, + "grad_norm": 2.2906084060668945, + "learning_rate": 2.6665312753858653e-06, + "loss": 1.7144, + "step": 1438 + }, + { + "epoch": 1.576986301369863, + "grad_norm": 2.2512989044189453, + "learning_rate": 2.6645004061738426e-06, + "loss": 1.8031, + "step": 1439 + }, + { + "epoch": 1.5780821917808219, + "grad_norm": 2.177103281021118, + "learning_rate": 2.66246953696182e-06, + "loss": 1.8073, + "step": 1440 + }, + { + "epoch": 1.5780821917808219, + "eval_loss": 1.9914618730545044, + "eval_runtime": 37.2706, + "eval_samples_per_second": 26.831, + "eval_steps_per_second": 1.127, + "step": 1440 + }, + { + "epoch": 1.5791780821917807, + "grad_norm": 2.4129281044006348, + "learning_rate": 2.660438667749797e-06, + "loss": 1.7552, + "step": 1441 + }, + { + "epoch": 1.5802739726027397, + "grad_norm": 2.2975244522094727, + "learning_rate": 2.6584077985377744e-06, + "loss": 1.7318, + "step": 1442 + }, + { + "epoch": 1.5813698630136988, + "grad_norm": 2.4061625003814697, + "learning_rate": 2.6563769293257517e-06, + "loss": 1.7839, + "step": 1443 + }, + { + "epoch": 1.5824657534246576, + "grad_norm": 2.236353635787964, + "learning_rate": 2.654346060113729e-06, + "loss": 1.7995, + "step": 1444 + }, + { + "epoch": 1.5835616438356164, + "grad_norm": 2.2464141845703125, + "learning_rate": 2.6523151909017062e-06, + "loss": 1.7496, + "step": 1445 + }, + { + "epoch": 1.5846575342465754, + "grad_norm": 2.2744123935699463, + "learning_rate": 2.6502843216896835e-06, + "loss": 1.7703, + "step": 1446 + }, + { + "epoch": 1.5857534246575342, + "grad_norm": 2.1578054428100586, + "learning_rate": 2.6482534524776603e-06, + "loss": 1.7561, + "step": 1447 + }, + { + "epoch": 1.586849315068493, + "grad_norm": 2.1924850940704346, + "learning_rate": 2.646222583265638e-06, + "loss": 1.867, + "step": 1448 + }, + { + "epoch": 1.587945205479452, + "grad_norm": 2.303420305252075, + "learning_rate": 2.644191714053615e-06, + "loss": 1.8345, + "step": 1449 + }, + { + "epoch": 1.589041095890411, + "grad_norm": 2.308837413787842, + "learning_rate": 2.6421608448415926e-06, + "loss": 1.7979, + "step": 1450 + }, + { + "epoch": 1.589041095890411, + "eval_loss": 1.9836338758468628, + "eval_runtime": 37.253, + "eval_samples_per_second": 26.843, + "eval_steps_per_second": 1.127, + "step": 1450 + }, + { + "epoch": 1.59013698630137, + "grad_norm": 2.2896456718444824, + "learning_rate": 2.6401299756295694e-06, + "loss": 1.7679, + "step": 1451 + }, + { + "epoch": 1.5912328767123287, + "grad_norm": 2.2482378482818604, + "learning_rate": 2.638099106417547e-06, + "loss": 1.7481, + "step": 1452 + }, + { + "epoch": 1.5923287671232877, + "grad_norm": 2.2620620727539062, + "learning_rate": 2.636068237205524e-06, + "loss": 1.801, + "step": 1453 + }, + { + "epoch": 1.5934246575342466, + "grad_norm": 2.216510772705078, + "learning_rate": 2.6340373679935017e-06, + "loss": 1.7485, + "step": 1454 + }, + { + "epoch": 1.5945205479452054, + "grad_norm": 2.354858160018921, + "learning_rate": 2.6320064987814785e-06, + "loss": 1.8636, + "step": 1455 + }, + { + "epoch": 1.5956164383561644, + "grad_norm": 2.251880168914795, + "learning_rate": 2.6299756295694562e-06, + "loss": 1.7087, + "step": 1456 + }, + { + "epoch": 1.5967123287671234, + "grad_norm": 2.2605879306793213, + "learning_rate": 2.627944760357433e-06, + "loss": 1.8141, + "step": 1457 + }, + { + "epoch": 1.5978082191780822, + "grad_norm": 2.3315021991729736, + "learning_rate": 2.6259138911454104e-06, + "loss": 1.8565, + "step": 1458 + }, + { + "epoch": 1.598904109589041, + "grad_norm": 2.1999425888061523, + "learning_rate": 2.6238830219333877e-06, + "loss": 1.8683, + "step": 1459 + }, + { + "epoch": 1.6, + "grad_norm": 2.2652416229248047, + "learning_rate": 2.621852152721365e-06, + "loss": 1.8489, + "step": 1460 + }, + { + "epoch": 1.6, + "eval_loss": 1.9816856384277344, + "eval_runtime": 37.2727, + "eval_samples_per_second": 26.829, + "eval_steps_per_second": 1.127, + "step": 1460 + }, + { + "epoch": 1.601095890410959, + "grad_norm": 2.469440221786499, + "learning_rate": 2.619821283509342e-06, + "loss": 1.7335, + "step": 1461 + }, + { + "epoch": 1.6021917808219177, + "grad_norm": 2.434267044067383, + "learning_rate": 2.6177904142973195e-06, + "loss": 1.8291, + "step": 1462 + }, + { + "epoch": 1.6032876712328767, + "grad_norm": 2.300492286682129, + "learning_rate": 2.6157595450852968e-06, + "loss": 1.7504, + "step": 1463 + }, + { + "epoch": 1.6043835616438358, + "grad_norm": 2.3906843662261963, + "learning_rate": 2.613728675873274e-06, + "loss": 1.8114, + "step": 1464 + }, + { + "epoch": 1.6054794520547944, + "grad_norm": 2.5111773014068604, + "learning_rate": 2.6116978066612513e-06, + "loss": 1.836, + "step": 1465 + }, + { + "epoch": 1.6065753424657534, + "grad_norm": 2.3834846019744873, + "learning_rate": 2.6096669374492286e-06, + "loss": 1.8589, + "step": 1466 + }, + { + "epoch": 1.6076712328767124, + "grad_norm": 2.2709615230560303, + "learning_rate": 2.607636068237206e-06, + "loss": 1.8052, + "step": 1467 + }, + { + "epoch": 1.6087671232876712, + "grad_norm": 2.3351972103118896, + "learning_rate": 2.605605199025183e-06, + "loss": 1.7559, + "step": 1468 + }, + { + "epoch": 1.60986301369863, + "grad_norm": 2.3994698524475098, + "learning_rate": 2.6035743298131604e-06, + "loss": 1.8099, + "step": 1469 + }, + { + "epoch": 1.610958904109589, + "grad_norm": 2.352590322494507, + "learning_rate": 2.6015434606011377e-06, + "loss": 1.8043, + "step": 1470 + }, + { + "epoch": 1.610958904109589, + "eval_loss": 1.9790339469909668, + "eval_runtime": 37.2619, + "eval_samples_per_second": 26.837, + "eval_steps_per_second": 1.127, + "step": 1470 + }, + { + "epoch": 1.612054794520548, + "grad_norm": 2.35754656791687, + "learning_rate": 2.5995125913891145e-06, + "loss": 1.8252, + "step": 1471 + }, + { + "epoch": 1.6131506849315067, + "grad_norm": 2.4011359214782715, + "learning_rate": 2.5974817221770922e-06, + "loss": 1.8005, + "step": 1472 + }, + { + "epoch": 1.6142465753424657, + "grad_norm": 3.143979787826538, + "learning_rate": 2.595450852965069e-06, + "loss": 1.7952, + "step": 1473 + }, + { + "epoch": 1.6153424657534248, + "grad_norm": 2.3928465843200684, + "learning_rate": 2.593419983753047e-06, + "loss": 1.8344, + "step": 1474 + }, + { + "epoch": 1.6164383561643836, + "grad_norm": 2.370368480682373, + "learning_rate": 2.5913891145410236e-06, + "loss": 1.8358, + "step": 1475 + }, + { + "epoch": 1.6175342465753424, + "grad_norm": 2.6163346767425537, + "learning_rate": 2.5893582453290013e-06, + "loss": 1.7734, + "step": 1476 + }, + { + "epoch": 1.6186301369863014, + "grad_norm": 2.370539903640747, + "learning_rate": 2.587327376116978e-06, + "loss": 1.8383, + "step": 1477 + }, + { + "epoch": 1.6197260273972602, + "grad_norm": 2.397007465362549, + "learning_rate": 2.585296506904956e-06, + "loss": 1.7702, + "step": 1478 + }, + { + "epoch": 1.620821917808219, + "grad_norm": 2.4692702293395996, + "learning_rate": 2.5832656376929327e-06, + "loss": 1.7887, + "step": 1479 + }, + { + "epoch": 1.621917808219178, + "grad_norm": 2.5779969692230225, + "learning_rate": 2.5812347684809104e-06, + "loss": 1.7751, + "step": 1480 + }, + { + "epoch": 1.621917808219178, + "eval_loss": 1.9826369285583496, + "eval_runtime": 37.2779, + "eval_samples_per_second": 26.826, + "eval_steps_per_second": 1.127, + "step": 1480 + }, + { + "epoch": 1.6230136986301371, + "grad_norm": 2.3182761669158936, + "learning_rate": 2.5792038992688873e-06, + "loss": 1.7136, + "step": 1481 + }, + { + "epoch": 1.624109589041096, + "grad_norm": 2.4109292030334473, + "learning_rate": 2.577173030056865e-06, + "loss": 1.8804, + "step": 1482 + }, + { + "epoch": 1.6252054794520547, + "grad_norm": 2.426166534423828, + "learning_rate": 2.575142160844842e-06, + "loss": 1.6996, + "step": 1483 + }, + { + "epoch": 1.6263013698630138, + "grad_norm": 2.5459423065185547, + "learning_rate": 2.5731112916328187e-06, + "loss": 1.8186, + "step": 1484 + }, + { + "epoch": 1.6273972602739726, + "grad_norm": 2.583578586578369, + "learning_rate": 2.5710804224207964e-06, + "loss": 1.8275, + "step": 1485 + }, + { + "epoch": 1.6284931506849314, + "grad_norm": 2.442936658859253, + "learning_rate": 2.5690495532087733e-06, + "loss": 1.7553, + "step": 1486 + }, + { + "epoch": 1.6295890410958904, + "grad_norm": 2.587864637374878, + "learning_rate": 2.567018683996751e-06, + "loss": 1.7171, + "step": 1487 + }, + { + "epoch": 1.6306849315068495, + "grad_norm": 2.4119534492492676, + "learning_rate": 2.564987814784728e-06, + "loss": 1.8296, + "step": 1488 + }, + { + "epoch": 1.6317808219178083, + "grad_norm": 2.3582253456115723, + "learning_rate": 2.5629569455727055e-06, + "loss": 1.7595, + "step": 1489 + }, + { + "epoch": 1.632876712328767, + "grad_norm": 2.522674083709717, + "learning_rate": 2.5609260763606824e-06, + "loss": 1.7908, + "step": 1490 + }, + { + "epoch": 1.632876712328767, + "eval_loss": 1.9791436195373535, + "eval_runtime": 37.2967, + "eval_samples_per_second": 26.812, + "eval_steps_per_second": 1.126, + "step": 1490 + }, + { + "epoch": 1.633972602739726, + "grad_norm": 2.2927470207214355, + "learning_rate": 2.55889520714866e-06, + "loss": 1.7789, + "step": 1491 + }, + { + "epoch": 1.635068493150685, + "grad_norm": 2.4469518661499023, + "learning_rate": 2.556864337936637e-06, + "loss": 1.8937, + "step": 1492 + }, + { + "epoch": 1.6361643835616437, + "grad_norm": 2.36860728263855, + "learning_rate": 2.5548334687246146e-06, + "loss": 1.778, + "step": 1493 + }, + { + "epoch": 1.6372602739726028, + "grad_norm": 2.4953925609588623, + "learning_rate": 2.5528025995125915e-06, + "loss": 1.7894, + "step": 1494 + }, + { + "epoch": 1.6383561643835618, + "grad_norm": 2.3478338718414307, + "learning_rate": 2.550771730300569e-06, + "loss": 1.8327, + "step": 1495 + }, + { + "epoch": 1.6394520547945204, + "grad_norm": 2.4399492740631104, + "learning_rate": 2.548740861088546e-06, + "loss": 1.8062, + "step": 1496 + }, + { + "epoch": 1.6405479452054794, + "grad_norm": 2.4457216262817383, + "learning_rate": 2.5467099918765233e-06, + "loss": 1.8268, + "step": 1497 + }, + { + "epoch": 1.6416438356164385, + "grad_norm": 2.365879535675049, + "learning_rate": 2.5446791226645006e-06, + "loss": 1.7288, + "step": 1498 + }, + { + "epoch": 1.6427397260273973, + "grad_norm": 2.4942879676818848, + "learning_rate": 2.542648253452478e-06, + "loss": 1.7111, + "step": 1499 + }, + { + "epoch": 1.643835616438356, + "grad_norm": 2.275982618331909, + "learning_rate": 2.540617384240455e-06, + "loss": 1.8292, + "step": 1500 + }, + { + "epoch": 1.643835616438356, + "eval_loss": 1.9746516942977905, + "eval_runtime": 37.2247, + "eval_samples_per_second": 26.864, + "eval_steps_per_second": 1.128, + "step": 1500 + }, + { + "epoch": 1.644931506849315, + "grad_norm": 2.4430429935455322, + "learning_rate": 2.5385865150284324e-06, + "loss": 1.8034, + "step": 1501 + }, + { + "epoch": 1.6460273972602741, + "grad_norm": 2.2778332233428955, + "learning_rate": 2.5365556458164097e-06, + "loss": 1.7589, + "step": 1502 + }, + { + "epoch": 1.6471232876712327, + "grad_norm": 2.2818830013275146, + "learning_rate": 2.534524776604387e-06, + "loss": 1.7743, + "step": 1503 + }, + { + "epoch": 1.6482191780821918, + "grad_norm": 2.321328639984131, + "learning_rate": 2.5324939073923642e-06, + "loss": 1.7922, + "step": 1504 + }, + { + "epoch": 1.6493150684931508, + "grad_norm": 2.4312188625335693, + "learning_rate": 2.5304630381803415e-06, + "loss": 1.6784, + "step": 1505 + }, + { + "epoch": 1.6504109589041096, + "grad_norm": 2.298741102218628, + "learning_rate": 2.5284321689683188e-06, + "loss": 1.7745, + "step": 1506 + }, + { + "epoch": 1.6515068493150684, + "grad_norm": 2.2882816791534424, + "learning_rate": 2.526401299756296e-06, + "loss": 1.7896, + "step": 1507 + }, + { + "epoch": 1.6526027397260274, + "grad_norm": 2.2336251735687256, + "learning_rate": 2.524370430544273e-06, + "loss": 1.7548, + "step": 1508 + }, + { + "epoch": 1.6536986301369863, + "grad_norm": 2.4369094371795654, + "learning_rate": 2.5223395613322506e-06, + "loss": 1.8049, + "step": 1509 + }, + { + "epoch": 1.654794520547945, + "grad_norm": 2.2978999614715576, + "learning_rate": 2.5203086921202274e-06, + "loss": 1.7489, + "step": 1510 + }, + { + "epoch": 1.654794520547945, + "eval_loss": 1.9760501384735107, + "eval_runtime": 37.3461, + "eval_samples_per_second": 26.777, + "eval_steps_per_second": 1.125, + "step": 1510 + }, + { + "epoch": 1.655890410958904, + "grad_norm": 2.331296920776367, + "learning_rate": 2.518277822908205e-06, + "loss": 1.8293, + "step": 1511 + }, + { + "epoch": 1.6569863013698631, + "grad_norm": 2.4544663429260254, + "learning_rate": 2.516246953696182e-06, + "loss": 1.7617, + "step": 1512 + }, + { + "epoch": 1.658082191780822, + "grad_norm": 2.4682042598724365, + "learning_rate": 2.5142160844841597e-06, + "loss": 1.8054, + "step": 1513 + }, + { + "epoch": 1.6591780821917808, + "grad_norm": 2.2827956676483154, + "learning_rate": 2.5121852152721365e-06, + "loss": 1.8137, + "step": 1514 + }, + { + "epoch": 1.6602739726027398, + "grad_norm": 2.2961761951446533, + "learning_rate": 2.5101543460601142e-06, + "loss": 1.7938, + "step": 1515 + }, + { + "epoch": 1.6613698630136986, + "grad_norm": 2.3464298248291016, + "learning_rate": 2.508123476848091e-06, + "loss": 1.8026, + "step": 1516 + }, + { + "epoch": 1.6624657534246574, + "grad_norm": 2.4975106716156006, + "learning_rate": 2.506092607636069e-06, + "loss": 1.7457, + "step": 1517 + }, + { + "epoch": 1.6635616438356164, + "grad_norm": 2.4823005199432373, + "learning_rate": 2.5040617384240457e-06, + "loss": 1.8282, + "step": 1518 + }, + { + "epoch": 1.6646575342465755, + "grad_norm": 2.2875800132751465, + "learning_rate": 2.5020308692120234e-06, + "loss": 1.7569, + "step": 1519 + }, + { + "epoch": 1.6657534246575343, + "grad_norm": 2.451022148132324, + "learning_rate": 2.5e-06, + "loss": 1.7483, + "step": 1520 + }, + { + "epoch": 1.6657534246575343, + "eval_loss": 1.9748642444610596, + "eval_runtime": 37.4512, + "eval_samples_per_second": 26.701, + "eval_steps_per_second": 1.121, + "step": 1520 + }, + { + "epoch": 1.666849315068493, + "grad_norm": 2.3427352905273438, + "learning_rate": 2.4979691307879775e-06, + "loss": 1.8493, + "step": 1521 + }, + { + "epoch": 1.6679452054794521, + "grad_norm": 2.4496946334838867, + "learning_rate": 2.4959382615759548e-06, + "loss": 1.7569, + "step": 1522 + }, + { + "epoch": 1.669041095890411, + "grad_norm": 2.141963481903076, + "learning_rate": 2.493907392363932e-06, + "loss": 1.7553, + "step": 1523 + }, + { + "epoch": 1.6701369863013698, + "grad_norm": 2.4115428924560547, + "learning_rate": 2.4918765231519093e-06, + "loss": 1.8606, + "step": 1524 + }, + { + "epoch": 1.6712328767123288, + "grad_norm": 2.614616632461548, + "learning_rate": 2.4898456539398866e-06, + "loss": 1.8217, + "step": 1525 + }, + { + "epoch": 1.6723287671232878, + "grad_norm": 2.338557243347168, + "learning_rate": 2.487814784727864e-06, + "loss": 1.7672, + "step": 1526 + }, + { + "epoch": 1.6734246575342466, + "grad_norm": 2.4309446811676025, + "learning_rate": 2.4857839155158407e-06, + "loss": 1.8476, + "step": 1527 + }, + { + "epoch": 1.6745205479452054, + "grad_norm": 2.25883412361145, + "learning_rate": 2.483753046303818e-06, + "loss": 1.7827, + "step": 1528 + }, + { + "epoch": 1.6756164383561645, + "grad_norm": 2.264845371246338, + "learning_rate": 2.4817221770917953e-06, + "loss": 1.7877, + "step": 1529 + }, + { + "epoch": 1.6767123287671233, + "grad_norm": 2.3584723472595215, + "learning_rate": 2.4796913078797725e-06, + "loss": 1.7891, + "step": 1530 + }, + { + "epoch": 1.6767123287671233, + "eval_loss": 1.9751919507980347, + "eval_runtime": 37.558, + "eval_samples_per_second": 26.625, + "eval_steps_per_second": 1.118, + "step": 1530 + }, + { + "epoch": 1.677808219178082, + "grad_norm": 2.3555185794830322, + "learning_rate": 2.47766043866775e-06, + "loss": 1.8235, + "step": 1531 + }, + { + "epoch": 1.6789041095890411, + "grad_norm": 2.3086600303649902, + "learning_rate": 2.475629569455727e-06, + "loss": 1.8036, + "step": 1532 + }, + { + "epoch": 1.6800000000000002, + "grad_norm": 2.286921262741089, + "learning_rate": 2.4735987002437044e-06, + "loss": 1.7885, + "step": 1533 + }, + { + "epoch": 1.6810958904109587, + "grad_norm": 2.2671141624450684, + "learning_rate": 2.4715678310316816e-06, + "loss": 1.758, + "step": 1534 + }, + { + "epoch": 1.6821917808219178, + "grad_norm": 2.4795749187469482, + "learning_rate": 2.469536961819659e-06, + "loss": 1.7791, + "step": 1535 + }, + { + "epoch": 1.6832876712328768, + "grad_norm": 2.201650857925415, + "learning_rate": 2.467506092607636e-06, + "loss": 1.7802, + "step": 1536 + }, + { + "epoch": 1.6843835616438356, + "grad_norm": 2.3804843425750732, + "learning_rate": 2.4654752233956135e-06, + "loss": 1.7739, + "step": 1537 + }, + { + "epoch": 1.6854794520547944, + "grad_norm": 2.322216510772705, + "learning_rate": 2.4634443541835907e-06, + "loss": 1.758, + "step": 1538 + }, + { + "epoch": 1.6865753424657535, + "grad_norm": 2.3582146167755127, + "learning_rate": 2.461413484971568e-06, + "loss": 1.8439, + "step": 1539 + }, + { + "epoch": 1.6876712328767123, + "grad_norm": 2.3597848415374756, + "learning_rate": 2.4593826157595453e-06, + "loss": 1.7772, + "step": 1540 + }, + { + "epoch": 1.6876712328767123, + "eval_loss": 1.9731802940368652, + "eval_runtime": 37.2439, + "eval_samples_per_second": 26.85, + "eval_steps_per_second": 1.128, + "step": 1540 + }, + { + "epoch": 1.688767123287671, + "grad_norm": 2.404569149017334, + "learning_rate": 2.4573517465475226e-06, + "loss": 1.8683, + "step": 1541 + }, + { + "epoch": 1.6898630136986301, + "grad_norm": 2.318737268447876, + "learning_rate": 2.4553208773355e-06, + "loss": 1.7223, + "step": 1542 + }, + { + "epoch": 1.6909589041095892, + "grad_norm": 2.2535340785980225, + "learning_rate": 2.453290008123477e-06, + "loss": 1.7693, + "step": 1543 + }, + { + "epoch": 1.692054794520548, + "grad_norm": 2.2533493041992188, + "learning_rate": 2.4512591389114544e-06, + "loss": 1.7876, + "step": 1544 + }, + { + "epoch": 1.6931506849315068, + "grad_norm": 2.253278970718384, + "learning_rate": 2.4492282696994317e-06, + "loss": 1.788, + "step": 1545 + }, + { + "epoch": 1.6942465753424658, + "grad_norm": 2.296807289123535, + "learning_rate": 2.447197400487409e-06, + "loss": 1.7877, + "step": 1546 + }, + { + "epoch": 1.6953424657534246, + "grad_norm": 2.4022111892700195, + "learning_rate": 2.4451665312753862e-06, + "loss": 1.7986, + "step": 1547 + }, + { + "epoch": 1.6964383561643834, + "grad_norm": 2.2671430110931396, + "learning_rate": 2.4431356620633635e-06, + "loss": 1.8269, + "step": 1548 + }, + { + "epoch": 1.6975342465753425, + "grad_norm": 2.338252067565918, + "learning_rate": 2.4411047928513408e-06, + "loss": 1.8279, + "step": 1549 + }, + { + "epoch": 1.6986301369863015, + "grad_norm": 2.365365505218506, + "learning_rate": 2.439073923639318e-06, + "loss": 1.7895, + "step": 1550 + }, + { + "epoch": 1.6986301369863015, + "eval_loss": 1.9711860418319702, + "eval_runtime": 37.2757, + "eval_samples_per_second": 26.827, + "eval_steps_per_second": 1.127, + "step": 1550 + }, + { + "epoch": 1.6997260273972603, + "grad_norm": 2.407580614089966, + "learning_rate": 2.4370430544272953e-06, + "loss": 1.7985, + "step": 1551 + }, + { + "epoch": 1.7008219178082191, + "grad_norm": 2.3145346641540527, + "learning_rate": 2.435012185215272e-06, + "loss": 1.7589, + "step": 1552 + }, + { + "epoch": 1.7019178082191782, + "grad_norm": 2.3809280395507812, + "learning_rate": 2.4329813160032495e-06, + "loss": 1.7334, + "step": 1553 + }, + { + "epoch": 1.703013698630137, + "grad_norm": 2.4224488735198975, + "learning_rate": 2.4309504467912267e-06, + "loss": 1.7549, + "step": 1554 + }, + { + "epoch": 1.7041095890410958, + "grad_norm": 2.374133348464966, + "learning_rate": 2.428919577579204e-06, + "loss": 1.7923, + "step": 1555 + }, + { + "epoch": 1.7052054794520548, + "grad_norm": 2.258222818374634, + "learning_rate": 2.4268887083671813e-06, + "loss": 1.7683, + "step": 1556 + }, + { + "epoch": 1.7063013698630138, + "grad_norm": 2.4481396675109863, + "learning_rate": 2.4248578391551586e-06, + "loss": 1.8609, + "step": 1557 + }, + { + "epoch": 1.7073972602739727, + "grad_norm": 2.3932790756225586, + "learning_rate": 2.422826969943136e-06, + "loss": 1.7475, + "step": 1558 + }, + { + "epoch": 1.7084931506849315, + "grad_norm": 2.3034355640411377, + "learning_rate": 2.420796100731113e-06, + "loss": 1.7664, + "step": 1559 + }, + { + "epoch": 1.7095890410958905, + "grad_norm": 2.2018799781799316, + "learning_rate": 2.4187652315190904e-06, + "loss": 1.7328, + "step": 1560 + }, + { + "epoch": 1.7095890410958905, + "eval_loss": 1.9715769290924072, + "eval_runtime": 37.2507, + "eval_samples_per_second": 26.845, + "eval_steps_per_second": 1.127, + "step": 1560 + }, + { + "epoch": 1.7106849315068493, + "grad_norm": 2.396462917327881, + "learning_rate": 2.4167343623070677e-06, + "loss": 1.7773, + "step": 1561 + }, + { + "epoch": 1.7117808219178081, + "grad_norm": 2.3212695121765137, + "learning_rate": 2.414703493095045e-06, + "loss": 1.7966, + "step": 1562 + }, + { + "epoch": 1.7128767123287671, + "grad_norm": 2.4157631397247314, + "learning_rate": 2.4126726238830222e-06, + "loss": 1.7797, + "step": 1563 + }, + { + "epoch": 1.7139726027397262, + "grad_norm": 2.404907464981079, + "learning_rate": 2.410641754670999e-06, + "loss": 1.8093, + "step": 1564 + }, + { + "epoch": 1.7150684931506848, + "grad_norm": 2.3798625469207764, + "learning_rate": 2.4086108854589763e-06, + "loss": 1.7635, + "step": 1565 + }, + { + "epoch": 1.7161643835616438, + "grad_norm": 2.325932025909424, + "learning_rate": 2.4065800162469536e-06, + "loss": 1.8281, + "step": 1566 + }, + { + "epoch": 1.7172602739726028, + "grad_norm": 2.413975238800049, + "learning_rate": 2.404549147034931e-06, + "loss": 1.7609, + "step": 1567 + }, + { + "epoch": 1.7183561643835616, + "grad_norm": 2.512775421142578, + "learning_rate": 2.402518277822908e-06, + "loss": 1.7793, + "step": 1568 + }, + { + "epoch": 1.7194520547945205, + "grad_norm": 2.581679582595825, + "learning_rate": 2.4004874086108854e-06, + "loss": 1.8302, + "step": 1569 + }, + { + "epoch": 1.7205479452054795, + "grad_norm": 2.389822244644165, + "learning_rate": 2.3984565393988627e-06, + "loss": 1.7573, + "step": 1570 + }, + { + "epoch": 1.7205479452054795, + "eval_loss": 1.9670331478118896, + "eval_runtime": 37.2901, + "eval_samples_per_second": 26.817, + "eval_steps_per_second": 1.126, + "step": 1570 + }, + { + "epoch": 1.7216438356164385, + "grad_norm": 2.6191556453704834, + "learning_rate": 2.39642567018684e-06, + "loss": 1.8252, + "step": 1571 + }, + { + "epoch": 1.7227397260273971, + "grad_norm": 2.394347906112671, + "learning_rate": 2.3943948009748173e-06, + "loss": 1.7561, + "step": 1572 + }, + { + "epoch": 1.7238356164383561, + "grad_norm": 2.3016326427459717, + "learning_rate": 2.3923639317627945e-06, + "loss": 1.7619, + "step": 1573 + }, + { + "epoch": 1.7249315068493152, + "grad_norm": 2.3663971424102783, + "learning_rate": 2.390333062550772e-06, + "loss": 1.7922, + "step": 1574 + }, + { + "epoch": 1.726027397260274, + "grad_norm": 2.3328938484191895, + "learning_rate": 2.388302193338749e-06, + "loss": 1.7374, + "step": 1575 + }, + { + "epoch": 1.7271232876712328, + "grad_norm": 2.313525676727295, + "learning_rate": 2.3862713241267264e-06, + "loss": 1.6585, + "step": 1576 + }, + { + "epoch": 1.7282191780821918, + "grad_norm": 2.4609642028808594, + "learning_rate": 2.3842404549147037e-06, + "loss": 1.8757, + "step": 1577 + }, + { + "epoch": 1.7293150684931506, + "grad_norm": 2.3502120971679688, + "learning_rate": 2.382209585702681e-06, + "loss": 1.7786, + "step": 1578 + }, + { + "epoch": 1.7304109589041095, + "grad_norm": 2.383025884628296, + "learning_rate": 2.380178716490658e-06, + "loss": 1.7502, + "step": 1579 + }, + { + "epoch": 1.7315068493150685, + "grad_norm": 2.280890464782715, + "learning_rate": 2.3781478472786355e-06, + "loss": 1.8192, + "step": 1580 + }, + { + "epoch": 1.7315068493150685, + "eval_loss": 1.9658632278442383, + "eval_runtime": 37.3714, + "eval_samples_per_second": 26.758, + "eval_steps_per_second": 1.124, + "step": 1580 + }, + { + "epoch": 1.7326027397260275, + "grad_norm": 2.342240333557129, + "learning_rate": 2.3761169780666128e-06, + "loss": 1.7786, + "step": 1581 + }, + { + "epoch": 1.7336986301369863, + "grad_norm": 2.2916715145111084, + "learning_rate": 2.37408610885459e-06, + "loss": 1.8161, + "step": 1582 + }, + { + "epoch": 1.7347945205479451, + "grad_norm": 2.24537992477417, + "learning_rate": 2.3720552396425673e-06, + "loss": 1.8442, + "step": 1583 + }, + { + "epoch": 1.7358904109589042, + "grad_norm": 2.3003411293029785, + "learning_rate": 2.3700243704305446e-06, + "loss": 1.7775, + "step": 1584 + }, + { + "epoch": 1.736986301369863, + "grad_norm": 2.3062400817871094, + "learning_rate": 2.367993501218522e-06, + "loss": 1.8374, + "step": 1585 + }, + { + "epoch": 1.7380821917808218, + "grad_norm": 2.3534083366394043, + "learning_rate": 2.365962632006499e-06, + "loss": 1.7829, + "step": 1586 + }, + { + "epoch": 1.7391780821917808, + "grad_norm": 2.2612221240997314, + "learning_rate": 2.3639317627944764e-06, + "loss": 1.7934, + "step": 1587 + }, + { + "epoch": 1.7402739726027399, + "grad_norm": 2.4197304248809814, + "learning_rate": 2.3619008935824537e-06, + "loss": 1.8226, + "step": 1588 + }, + { + "epoch": 1.7413698630136987, + "grad_norm": 2.435908079147339, + "learning_rate": 2.3598700243704305e-06, + "loss": 1.8148, + "step": 1589 + }, + { + "epoch": 1.7424657534246575, + "grad_norm": 2.3617560863494873, + "learning_rate": 2.357839155158408e-06, + "loss": 1.8085, + "step": 1590 + }, + { + "epoch": 1.7424657534246575, + "eval_loss": 1.9636695384979248, + "eval_runtime": 37.2968, + "eval_samples_per_second": 26.812, + "eval_steps_per_second": 1.126, + "step": 1590 + }, + { + "epoch": 1.7435616438356165, + "grad_norm": 2.295182228088379, + "learning_rate": 2.355808285946385e-06, + "loss": 1.8091, + "step": 1591 + }, + { + "epoch": 1.7446575342465753, + "grad_norm": 2.3347294330596924, + "learning_rate": 2.3537774167343624e-06, + "loss": 1.8086, + "step": 1592 + }, + { + "epoch": 1.7457534246575341, + "grad_norm": 2.4760799407958984, + "learning_rate": 2.3517465475223396e-06, + "loss": 1.8227, + "step": 1593 + }, + { + "epoch": 1.7468493150684932, + "grad_norm": 2.4243907928466797, + "learning_rate": 2.349715678310317e-06, + "loss": 1.7921, + "step": 1594 + }, + { + "epoch": 1.7479452054794522, + "grad_norm": 2.271735668182373, + "learning_rate": 2.347684809098294e-06, + "loss": 1.7987, + "step": 1595 + }, + { + "epoch": 1.749041095890411, + "grad_norm": 2.5412354469299316, + "learning_rate": 2.3456539398862715e-06, + "loss": 1.7558, + "step": 1596 + }, + { + "epoch": 1.7501369863013698, + "grad_norm": 2.464944362640381, + "learning_rate": 2.3436230706742487e-06, + "loss": 1.8087, + "step": 1597 + }, + { + "epoch": 1.7512328767123289, + "grad_norm": 2.4210381507873535, + "learning_rate": 2.341592201462226e-06, + "loss": 1.756, + "step": 1598 + }, + { + "epoch": 1.7523287671232877, + "grad_norm": 2.284945011138916, + "learning_rate": 2.3395613322502033e-06, + "loss": 1.7909, + "step": 1599 + }, + { + "epoch": 1.7534246575342465, + "grad_norm": 2.3740732669830322, + "learning_rate": 2.3375304630381806e-06, + "loss": 1.7683, + "step": 1600 + }, + { + "epoch": 1.7534246575342465, + "eval_loss": 1.9646408557891846, + "eval_runtime": 37.3266, + "eval_samples_per_second": 26.791, + "eval_steps_per_second": 1.125, + "step": 1600 + }, + { + "epoch": 1.7545205479452055, + "grad_norm": 2.399583578109741, + "learning_rate": 2.335499593826158e-06, + "loss": 1.7342, + "step": 1601 + }, + { + "epoch": 1.7556164383561645, + "grad_norm": 2.4731597900390625, + "learning_rate": 2.333468724614135e-06, + "loss": 1.7507, + "step": 1602 + }, + { + "epoch": 1.7567123287671231, + "grad_norm": 2.365433692932129, + "learning_rate": 2.3314378554021124e-06, + "loss": 1.7944, + "step": 1603 + }, + { + "epoch": 1.7578082191780822, + "grad_norm": 2.4500601291656494, + "learning_rate": 2.3294069861900897e-06, + "loss": 1.822, + "step": 1604 + }, + { + "epoch": 1.7589041095890412, + "grad_norm": 2.3339014053344727, + "learning_rate": 2.327376116978067e-06, + "loss": 1.7464, + "step": 1605 + }, + { + "epoch": 1.76, + "grad_norm": 2.343973159790039, + "learning_rate": 2.3253452477660442e-06, + "loss": 1.7408, + "step": 1606 + }, + { + "epoch": 1.7610958904109588, + "grad_norm": 2.453462839126587, + "learning_rate": 2.3233143785540215e-06, + "loss": 1.8261, + "step": 1607 + }, + { + "epoch": 1.7621917808219179, + "grad_norm": 2.357698678970337, + "learning_rate": 2.3212835093419988e-06, + "loss": 1.762, + "step": 1608 + }, + { + "epoch": 1.7632876712328767, + "grad_norm": 2.353257417678833, + "learning_rate": 2.319252640129976e-06, + "loss": 1.8076, + "step": 1609 + }, + { + "epoch": 1.7643835616438355, + "grad_norm": 2.3269948959350586, + "learning_rate": 2.3172217709179533e-06, + "loss": 1.7516, + "step": 1610 + }, + { + "epoch": 1.7643835616438355, + "eval_loss": 1.9678465127944946, + "eval_runtime": 37.2052, + "eval_samples_per_second": 26.878, + "eval_steps_per_second": 1.129, + "step": 1610 + }, + { + "epoch": 1.7654794520547945, + "grad_norm": 2.4023501873016357, + "learning_rate": 2.3151909017059306e-06, + "loss": 1.8477, + "step": 1611 + }, + { + "epoch": 1.7665753424657535, + "grad_norm": 2.3356990814208984, + "learning_rate": 2.313160032493908e-06, + "loss": 1.8236, + "step": 1612 + }, + { + "epoch": 1.7676712328767124, + "grad_norm": 2.3571391105651855, + "learning_rate": 2.311129163281885e-06, + "loss": 1.7335, + "step": 1613 + }, + { + "epoch": 1.7687671232876712, + "grad_norm": 2.3510208129882812, + "learning_rate": 2.309098294069862e-06, + "loss": 1.8088, + "step": 1614 + }, + { + "epoch": 1.7698630136986302, + "grad_norm": 2.3852577209472656, + "learning_rate": 2.3070674248578393e-06, + "loss": 1.7312, + "step": 1615 + }, + { + "epoch": 1.770958904109589, + "grad_norm": 2.3085618019104004, + "learning_rate": 2.3050365556458166e-06, + "loss": 1.7379, + "step": 1616 + }, + { + "epoch": 1.7720547945205478, + "grad_norm": 2.3454065322875977, + "learning_rate": 2.303005686433794e-06, + "loss": 1.8054, + "step": 1617 + }, + { + "epoch": 1.7731506849315068, + "grad_norm": 2.4034247398376465, + "learning_rate": 2.300974817221771e-06, + "loss": 1.7955, + "step": 1618 + }, + { + "epoch": 1.7742465753424659, + "grad_norm": 2.4181671142578125, + "learning_rate": 2.2989439480097484e-06, + "loss": 1.7732, + "step": 1619 + }, + { + "epoch": 1.7753424657534247, + "grad_norm": 2.3408010005950928, + "learning_rate": 2.2969130787977257e-06, + "loss": 1.7231, + "step": 1620 + }, + { + "epoch": 1.7753424657534247, + "eval_loss": 1.9655163288116455, + "eval_runtime": 37.4878, + "eval_samples_per_second": 26.675, + "eval_steps_per_second": 1.12, + "step": 1620 + }, + { + "epoch": 1.7764383561643835, + "grad_norm": 2.3113718032836914, + "learning_rate": 2.294882209585703e-06, + "loss": 1.7904, + "step": 1621 + }, + { + "epoch": 1.7775342465753425, + "grad_norm": 2.3037896156311035, + "learning_rate": 2.2928513403736802e-06, + "loss": 1.8069, + "step": 1622 + }, + { + "epoch": 1.7786301369863013, + "grad_norm": 2.3962693214416504, + "learning_rate": 2.2908204711616575e-06, + "loss": 1.7354, + "step": 1623 + }, + { + "epoch": 1.7797260273972602, + "grad_norm": 2.3499252796173096, + "learning_rate": 2.2887896019496348e-06, + "loss": 1.779, + "step": 1624 + }, + { + "epoch": 1.7808219178082192, + "grad_norm": 2.390958070755005, + "learning_rate": 2.286758732737612e-06, + "loss": 1.8257, + "step": 1625 + }, + { + "epoch": 1.7819178082191782, + "grad_norm": 2.365626573562622, + "learning_rate": 2.284727863525589e-06, + "loss": 1.7618, + "step": 1626 + }, + { + "epoch": 1.783013698630137, + "grad_norm": 2.3558530807495117, + "learning_rate": 2.282696994313566e-06, + "loss": 1.7729, + "step": 1627 + }, + { + "epoch": 1.7841095890410958, + "grad_norm": 2.3276045322418213, + "learning_rate": 2.2806661251015434e-06, + "loss": 1.8231, + "step": 1628 + }, + { + "epoch": 1.7852054794520549, + "grad_norm": 2.3369953632354736, + "learning_rate": 2.2786352558895207e-06, + "loss": 1.8011, + "step": 1629 + }, + { + "epoch": 1.7863013698630137, + "grad_norm": 2.3588311672210693, + "learning_rate": 2.276604386677498e-06, + "loss": 1.7682, + "step": 1630 + }, + { + "epoch": 1.7863013698630137, + "eval_loss": 1.963529348373413, + "eval_runtime": 37.5589, + "eval_samples_per_second": 26.625, + "eval_steps_per_second": 1.118, + "step": 1630 + }, + { + "epoch": 1.7873972602739725, + "grad_norm": 2.2918803691864014, + "learning_rate": 2.2745735174654753e-06, + "loss": 1.7256, + "step": 1631 + }, + { + "epoch": 1.7884931506849315, + "grad_norm": 2.423738479614258, + "learning_rate": 2.2725426482534526e-06, + "loss": 1.7582, + "step": 1632 + }, + { + "epoch": 1.7895890410958906, + "grad_norm": 2.3500850200653076, + "learning_rate": 2.27051177904143e-06, + "loss": 1.8255, + "step": 1633 + }, + { + "epoch": 1.7906849315068492, + "grad_norm": 2.3280837535858154, + "learning_rate": 2.268480909829407e-06, + "loss": 1.7845, + "step": 1634 + }, + { + "epoch": 1.7917808219178082, + "grad_norm": 2.5055465698242188, + "learning_rate": 2.2664500406173844e-06, + "loss": 1.78, + "step": 1635 + }, + { + "epoch": 1.7928767123287672, + "grad_norm": 2.3132143020629883, + "learning_rate": 2.2644191714053617e-06, + "loss": 1.7814, + "step": 1636 + }, + { + "epoch": 1.793972602739726, + "grad_norm": 2.316361427307129, + "learning_rate": 2.262388302193339e-06, + "loss": 1.8161, + "step": 1637 + }, + { + "epoch": 1.7950684931506848, + "grad_norm": 2.337318181991577, + "learning_rate": 2.260357432981316e-06, + "loss": 1.7942, + "step": 1638 + }, + { + "epoch": 1.7961643835616439, + "grad_norm": 2.362031936645508, + "learning_rate": 2.2583265637692935e-06, + "loss": 1.7771, + "step": 1639 + }, + { + "epoch": 1.7972602739726027, + "grad_norm": 2.3260183334350586, + "learning_rate": 2.2562956945572708e-06, + "loss": 1.7975, + "step": 1640 + }, + { + "epoch": 1.7972602739726027, + "eval_loss": 1.9619382619857788, + "eval_runtime": 37.2606, + "eval_samples_per_second": 26.838, + "eval_steps_per_second": 1.127, + "step": 1640 + }, + { + "epoch": 1.7983561643835615, + "grad_norm": 2.4260714054107666, + "learning_rate": 2.254264825345248e-06, + "loss": 1.7654, + "step": 1641 + }, + { + "epoch": 1.7994520547945205, + "grad_norm": 2.297428846359253, + "learning_rate": 2.2522339561332253e-06, + "loss": 1.775, + "step": 1642 + }, + { + "epoch": 1.8005479452054796, + "grad_norm": 2.5153353214263916, + "learning_rate": 2.2502030869212026e-06, + "loss": 1.7796, + "step": 1643 + }, + { + "epoch": 1.8016438356164384, + "grad_norm": 2.5102577209472656, + "learning_rate": 2.24817221770918e-06, + "loss": 1.7917, + "step": 1644 + }, + { + "epoch": 1.8027397260273972, + "grad_norm": 2.3693106174468994, + "learning_rate": 2.246141348497157e-06, + "loss": 1.8459, + "step": 1645 + }, + { + "epoch": 1.8038356164383562, + "grad_norm": 2.2353153228759766, + "learning_rate": 2.2441104792851344e-06, + "loss": 1.8095, + "step": 1646 + }, + { + "epoch": 1.804931506849315, + "grad_norm": 2.265465497970581, + "learning_rate": 2.2420796100731117e-06, + "loss": 1.7623, + "step": 1647 + }, + { + "epoch": 1.8060273972602738, + "grad_norm": 2.2820940017700195, + "learning_rate": 2.240048740861089e-06, + "loss": 1.7946, + "step": 1648 + }, + { + "epoch": 1.8071232876712329, + "grad_norm": 2.378035068511963, + "learning_rate": 2.2380178716490662e-06, + "loss": 1.8059, + "step": 1649 + }, + { + "epoch": 1.808219178082192, + "grad_norm": 2.328341245651245, + "learning_rate": 2.2359870024370435e-06, + "loss": 1.6611, + "step": 1650 + }, + { + "epoch": 1.808219178082192, + "eval_loss": 1.9600144624710083, + "eval_runtime": 37.2601, + "eval_samples_per_second": 26.838, + "eval_steps_per_second": 1.127, + "step": 1650 + }, + { + "epoch": 1.8093150684931507, + "grad_norm": 2.2819113731384277, + "learning_rate": 2.2339561332250204e-06, + "loss": 1.8058, + "step": 1651 + }, + { + "epoch": 1.8104109589041095, + "grad_norm": 2.4178121089935303, + "learning_rate": 2.2319252640129976e-06, + "loss": 1.8204, + "step": 1652 + }, + { + "epoch": 1.8115068493150686, + "grad_norm": 2.497008800506592, + "learning_rate": 2.229894394800975e-06, + "loss": 1.7968, + "step": 1653 + }, + { + "epoch": 1.8126027397260274, + "grad_norm": 2.4681501388549805, + "learning_rate": 2.227863525588952e-06, + "loss": 1.7292, + "step": 1654 + }, + { + "epoch": 1.8136986301369862, + "grad_norm": 2.2983527183532715, + "learning_rate": 2.2258326563769295e-06, + "loss": 1.7097, + "step": 1655 + }, + { + "epoch": 1.8147945205479452, + "grad_norm": 2.349454164505005, + "learning_rate": 2.2238017871649067e-06, + "loss": 1.7256, + "step": 1656 + }, + { + "epoch": 1.8158904109589042, + "grad_norm": 2.4247677326202393, + "learning_rate": 2.221770917952884e-06, + "loss": 1.7165, + "step": 1657 + }, + { + "epoch": 1.816986301369863, + "grad_norm": 2.3035342693328857, + "learning_rate": 2.2197400487408613e-06, + "loss": 1.7811, + "step": 1658 + }, + { + "epoch": 1.8180821917808219, + "grad_norm": 2.318772554397583, + "learning_rate": 2.2177091795288386e-06, + "loss": 1.8082, + "step": 1659 + }, + { + "epoch": 1.819178082191781, + "grad_norm": 2.2477188110351562, + "learning_rate": 2.215678310316816e-06, + "loss": 1.7206, + "step": 1660 + }, + { + "epoch": 1.819178082191781, + "eval_loss": 1.962982416152954, + "eval_runtime": 37.2709, + "eval_samples_per_second": 26.831, + "eval_steps_per_second": 1.127, + "step": 1660 + }, + { + "epoch": 1.8202739726027397, + "grad_norm": 2.355264186859131, + "learning_rate": 2.213647441104793e-06, + "loss": 1.7634, + "step": 1661 + }, + { + "epoch": 1.8213698630136985, + "grad_norm": 2.267458200454712, + "learning_rate": 2.2116165718927704e-06, + "loss": 1.7819, + "step": 1662 + }, + { + "epoch": 1.8224657534246576, + "grad_norm": 2.2487895488739014, + "learning_rate": 2.2095857026807473e-06, + "loss": 1.759, + "step": 1663 + }, + { + "epoch": 1.8235616438356166, + "grad_norm": 2.4253578186035156, + "learning_rate": 2.2075548334687245e-06, + "loss": 1.7937, + "step": 1664 + }, + { + "epoch": 1.8246575342465754, + "grad_norm": 2.478050470352173, + "learning_rate": 2.205523964256702e-06, + "loss": 1.7241, + "step": 1665 + }, + { + "epoch": 1.8257534246575342, + "grad_norm": 2.289905309677124, + "learning_rate": 2.203493095044679e-06, + "loss": 1.7531, + "step": 1666 + }, + { + "epoch": 1.8268493150684932, + "grad_norm": 2.303421974182129, + "learning_rate": 2.2014622258326564e-06, + "loss": 1.7711, + "step": 1667 + }, + { + "epoch": 1.827945205479452, + "grad_norm": 2.3670248985290527, + "learning_rate": 2.1994313566206336e-06, + "loss": 1.8093, + "step": 1668 + }, + { + "epoch": 1.8290410958904109, + "grad_norm": 2.3721048831939697, + "learning_rate": 2.197400487408611e-06, + "loss": 1.8273, + "step": 1669 + }, + { + "epoch": 1.83013698630137, + "grad_norm": 2.3000190258026123, + "learning_rate": 2.195369618196588e-06, + "loss": 1.75, + "step": 1670 + }, + { + "epoch": 1.83013698630137, + "eval_loss": 1.9577263593673706, + "eval_runtime": 37.2953, + "eval_samples_per_second": 26.813, + "eval_steps_per_second": 1.126, + "step": 1670 + }, + { + "epoch": 1.831232876712329, + "grad_norm": 2.303584575653076, + "learning_rate": 2.1933387489845655e-06, + "loss": 1.7278, + "step": 1671 + }, + { + "epoch": 1.8323287671232875, + "grad_norm": 2.260579824447632, + "learning_rate": 2.1913078797725427e-06, + "loss": 1.7524, + "step": 1672 + }, + { + "epoch": 1.8334246575342465, + "grad_norm": 2.311967372894287, + "learning_rate": 2.18927701056052e-06, + "loss": 1.7799, + "step": 1673 + }, + { + "epoch": 1.8345205479452056, + "grad_norm": 2.338989496231079, + "learning_rate": 2.1872461413484973e-06, + "loss": 1.8468, + "step": 1674 + }, + { + "epoch": 1.8356164383561644, + "grad_norm": 2.338989496231079, + "learning_rate": 2.1872461413484973e-06, + "loss": 1.7853, + "step": 1675 + }, + { + "epoch": 1.8367123287671232, + "grad_norm": 2.526724100112915, + "learning_rate": 2.1852152721364746e-06, + "loss": 1.78, + "step": 1676 + }, + { + "epoch": 1.8378082191780822, + "grad_norm": 2.4637107849121094, + "learning_rate": 2.183184402924452e-06, + "loss": 1.7328, + "step": 1677 + }, + { + "epoch": 1.838904109589041, + "grad_norm": 2.34454345703125, + "learning_rate": 2.181153533712429e-06, + "loss": 1.8058, + "step": 1678 + }, + { + "epoch": 1.8399999999999999, + "grad_norm": 2.5518760681152344, + "learning_rate": 2.1791226645004064e-06, + "loss": 1.7746, + "step": 1679 + }, + { + "epoch": 1.841095890410959, + "grad_norm": 2.437443971633911, + "learning_rate": 2.1770917952883837e-06, + "loss": 1.7517, + "step": 1680 + }, + { + "epoch": 1.841095890410959, + "eval_loss": 1.9572285413742065, + "eval_runtime": 37.3402, + "eval_samples_per_second": 26.781, + "eval_steps_per_second": 1.125, + "step": 1680 + }, + { + "epoch": 1.842191780821918, + "grad_norm": 2.344287157058716, + "learning_rate": 2.175060926076361e-06, + "loss": 1.8229, + "step": 1681 + }, + { + "epoch": 1.8432876712328767, + "grad_norm": 2.333840847015381, + "learning_rate": 2.1730300568643382e-06, + "loss": 1.8534, + "step": 1682 + }, + { + "epoch": 1.8443835616438355, + "grad_norm": 2.372926950454712, + "learning_rate": 2.1709991876523155e-06, + "loss": 1.7609, + "step": 1683 + }, + { + "epoch": 1.8454794520547946, + "grad_norm": 2.2858645915985107, + "learning_rate": 2.1689683184402928e-06, + "loss": 1.7508, + "step": 1684 + }, + { + "epoch": 1.8465753424657534, + "grad_norm": 2.326586961746216, + "learning_rate": 2.16693744922827e-06, + "loss": 1.8122, + "step": 1685 + }, + { + "epoch": 1.8476712328767122, + "grad_norm": 2.3605616092681885, + "learning_rate": 2.1649065800162473e-06, + "loss": 1.7794, + "step": 1686 + }, + { + "epoch": 1.8487671232876712, + "grad_norm": 2.4188039302825928, + "learning_rate": 2.1628757108042246e-06, + "loss": 1.7552, + "step": 1687 + }, + { + "epoch": 1.8498630136986303, + "grad_norm": 2.335617780685425, + "learning_rate": 2.160844841592202e-06, + "loss": 1.7259, + "step": 1688 + }, + { + "epoch": 1.850958904109589, + "grad_norm": 2.245882272720337, + "learning_rate": 2.1588139723801787e-06, + "loss": 1.7259, + "step": 1689 + }, + { + "epoch": 1.8520547945205479, + "grad_norm": 2.4447500705718994, + "learning_rate": 2.156783103168156e-06, + "loss": 1.7605, + "step": 1690 + }, + { + "epoch": 1.8520547945205479, + "eval_loss": 1.9608982801437378, + "eval_runtime": 37.3756, + "eval_samples_per_second": 26.755, + "eval_steps_per_second": 1.124, + "step": 1690 + }, + { + "epoch": 1.853150684931507, + "grad_norm": 2.363443613052368, + "learning_rate": 2.1547522339561333e-06, + "loss": 1.7336, + "step": 1691 + }, + { + "epoch": 1.8542465753424657, + "grad_norm": 2.3573989868164062, + "learning_rate": 2.1527213647441106e-06, + "loss": 1.7953, + "step": 1692 + }, + { + "epoch": 1.8553424657534245, + "grad_norm": 2.4292593002319336, + "learning_rate": 2.150690495532088e-06, + "loss": 1.8276, + "step": 1693 + }, + { + "epoch": 1.8564383561643836, + "grad_norm": 2.374563694000244, + "learning_rate": 2.148659626320065e-06, + "loss": 1.7975, + "step": 1694 + }, + { + "epoch": 1.8575342465753426, + "grad_norm": 2.3935306072235107, + "learning_rate": 2.1466287571080424e-06, + "loss": 1.7824, + "step": 1695 + }, + { + "epoch": 1.8586301369863014, + "grad_norm": 2.343773365020752, + "learning_rate": 2.1445978878960197e-06, + "loss": 1.7574, + "step": 1696 + }, + { + "epoch": 1.8597260273972602, + "grad_norm": 2.2765426635742188, + "learning_rate": 2.142567018683997e-06, + "loss": 1.7682, + "step": 1697 + }, + { + "epoch": 1.8608219178082193, + "grad_norm": 2.339470863342285, + "learning_rate": 2.140536149471974e-06, + "loss": 1.8229, + "step": 1698 + }, + { + "epoch": 1.861917808219178, + "grad_norm": 2.341775894165039, + "learning_rate": 2.1385052802599515e-06, + "loss": 1.7421, + "step": 1699 + }, + { + "epoch": 1.8630136986301369, + "grad_norm": 2.343876838684082, + "learning_rate": 2.1364744110479288e-06, + "loss": 1.7519, + "step": 1700 + }, + { + "epoch": 1.8630136986301369, + "eval_loss": 1.9587067365646362, + "eval_runtime": 37.4192, + "eval_samples_per_second": 26.724, + "eval_steps_per_second": 1.122, + "step": 1700 + }, + { + "epoch": 1.864109589041096, + "grad_norm": 2.388373613357544, + "learning_rate": 2.134443541835906e-06, + "loss": 1.8039, + "step": 1701 + }, + { + "epoch": 1.865205479452055, + "grad_norm": 2.379742383956909, + "learning_rate": 2.132412672623883e-06, + "loss": 1.7549, + "step": 1702 + }, + { + "epoch": 1.8663013698630135, + "grad_norm": 2.25833797454834, + "learning_rate": 2.13038180341186e-06, + "loss": 1.7349, + "step": 1703 + }, + { + "epoch": 1.8673972602739726, + "grad_norm": 2.4541401863098145, + "learning_rate": 2.1283509341998374e-06, + "loss": 1.7785, + "step": 1704 + }, + { + "epoch": 1.8684931506849316, + "grad_norm": 2.338456392288208, + "learning_rate": 2.1263200649878147e-06, + "loss": 1.8272, + "step": 1705 + }, + { + "epoch": 1.8695890410958904, + "grad_norm": 2.474292755126953, + "learning_rate": 2.124289195775792e-06, + "loss": 1.7497, + "step": 1706 + }, + { + "epoch": 1.8706849315068492, + "grad_norm": 2.206749439239502, + "learning_rate": 2.1222583265637693e-06, + "loss": 1.7697, + "step": 1707 + }, + { + "epoch": 1.8717808219178083, + "grad_norm": 2.378093957901001, + "learning_rate": 2.1202274573517465e-06, + "loss": 1.6851, + "step": 1708 + }, + { + "epoch": 1.872876712328767, + "grad_norm": 2.2442147731781006, + "learning_rate": 2.118196588139724e-06, + "loss": 1.7268, + "step": 1709 + }, + { + "epoch": 1.8739726027397259, + "grad_norm": 2.324852228164673, + "learning_rate": 2.116165718927701e-06, + "loss": 1.7369, + "step": 1710 + }, + { + "epoch": 1.8739726027397259, + "eval_loss": 1.9555498361587524, + "eval_runtime": 37.2599, + "eval_samples_per_second": 26.838, + "eval_steps_per_second": 1.127, + "step": 1710 + }, + { + "epoch": 1.875068493150685, + "grad_norm": 2.3466620445251465, + "learning_rate": 2.1141348497156784e-06, + "loss": 1.7642, + "step": 1711 + }, + { + "epoch": 1.876164383561644, + "grad_norm": 2.303654432296753, + "learning_rate": 2.1121039805036556e-06, + "loss": 1.7406, + "step": 1712 + }, + { + "epoch": 1.8772602739726028, + "grad_norm": 2.242719888687134, + "learning_rate": 2.110073111291633e-06, + "loss": 1.7269, + "step": 1713 + }, + { + "epoch": 1.8783561643835616, + "grad_norm": 2.4076626300811768, + "learning_rate": 2.10804224207961e-06, + "loss": 1.7749, + "step": 1714 + }, + { + "epoch": 1.8794520547945206, + "grad_norm": 2.510887861251831, + "learning_rate": 2.1060113728675875e-06, + "loss": 1.8056, + "step": 1715 + }, + { + "epoch": 1.8805479452054794, + "grad_norm": 2.4038093090057373, + "learning_rate": 2.1039805036555647e-06, + "loss": 1.7915, + "step": 1716 + }, + { + "epoch": 1.8816438356164382, + "grad_norm": 2.4485573768615723, + "learning_rate": 2.101949634443542e-06, + "loss": 1.6845, + "step": 1717 + }, + { + "epoch": 1.8827397260273973, + "grad_norm": 2.4619247913360596, + "learning_rate": 2.0999187652315193e-06, + "loss": 1.7597, + "step": 1718 + }, + { + "epoch": 1.8838356164383563, + "grad_norm": 2.5079665184020996, + "learning_rate": 2.0978878960194966e-06, + "loss": 1.7207, + "step": 1719 + }, + { + "epoch": 1.884931506849315, + "grad_norm": 2.38696551322937, + "learning_rate": 2.095857026807474e-06, + "loss": 1.7855, + "step": 1720 + }, + { + "epoch": 1.884931506849315, + "eval_loss": 1.9538354873657227, + "eval_runtime": 37.5269, + "eval_samples_per_second": 26.648, + "eval_steps_per_second": 1.119, + "step": 1720 + }, + { + "epoch": 1.886027397260274, + "grad_norm": 2.2652602195739746, + "learning_rate": 2.093826157595451e-06, + "loss": 1.7806, + "step": 1721 + }, + { + "epoch": 1.887123287671233, + "grad_norm": 2.4074227809906006, + "learning_rate": 2.0917952883834284e-06, + "loss": 1.7397, + "step": 1722 + }, + { + "epoch": 1.8882191780821918, + "grad_norm": 2.4302561283111572, + "learning_rate": 2.0897644191714057e-06, + "loss": 1.7883, + "step": 1723 + }, + { + "epoch": 1.8893150684931506, + "grad_norm": 2.311933755874634, + "learning_rate": 2.087733549959383e-06, + "loss": 1.8186, + "step": 1724 + }, + { + "epoch": 1.8904109589041096, + "grad_norm": 2.4028313159942627, + "learning_rate": 2.0857026807473602e-06, + "loss": 1.7208, + "step": 1725 + }, + { + "epoch": 1.8915068493150686, + "grad_norm": 2.2129130363464355, + "learning_rate": 2.0836718115353375e-06, + "loss": 1.7007, + "step": 1726 + }, + { + "epoch": 1.8926027397260274, + "grad_norm": 2.3893744945526123, + "learning_rate": 2.0816409423233144e-06, + "loss": 1.8281, + "step": 1727 + }, + { + "epoch": 1.8936986301369862, + "grad_norm": 2.2396240234375, + "learning_rate": 2.0796100731112916e-06, + "loss": 1.7252, + "step": 1728 + }, + { + "epoch": 1.8947945205479453, + "grad_norm": 2.4073901176452637, + "learning_rate": 2.077579203899269e-06, + "loss": 1.7533, + "step": 1729 + }, + { + "epoch": 1.895890410958904, + "grad_norm": 2.3669373989105225, + "learning_rate": 2.075548334687246e-06, + "loss": 1.7013, + "step": 1730 + }, + { + "epoch": 1.895890410958904, + "eval_loss": 1.95528244972229, + "eval_runtime": 37.6174, + "eval_samples_per_second": 26.583, + "eval_steps_per_second": 1.117, + "step": 1730 + }, + { + "epoch": 1.896986301369863, + "grad_norm": 2.5254933834075928, + "learning_rate": 2.0735174654752235e-06, + "loss": 1.7848, + "step": 1731 + }, + { + "epoch": 1.898082191780822, + "grad_norm": 2.4138967990875244, + "learning_rate": 2.0714865962632007e-06, + "loss": 1.7204, + "step": 1732 + }, + { + "epoch": 1.899178082191781, + "grad_norm": 2.417165994644165, + "learning_rate": 2.069455727051178e-06, + "loss": 1.7741, + "step": 1733 + }, + { + "epoch": 1.9002739726027398, + "grad_norm": 2.4026505947113037, + "learning_rate": 2.0674248578391553e-06, + "loss": 1.7456, + "step": 1734 + }, + { + "epoch": 1.9013698630136986, + "grad_norm": 2.4637675285339355, + "learning_rate": 2.0653939886271326e-06, + "loss": 1.7679, + "step": 1735 + }, + { + "epoch": 1.9024657534246576, + "grad_norm": 2.269869089126587, + "learning_rate": 2.06336311941511e-06, + "loss": 1.8059, + "step": 1736 + }, + { + "epoch": 1.9035616438356164, + "grad_norm": 2.321094512939453, + "learning_rate": 2.061332250203087e-06, + "loss": 1.743, + "step": 1737 + }, + { + "epoch": 1.9046575342465752, + "grad_norm": 2.477288246154785, + "learning_rate": 2.0593013809910644e-06, + "loss": 1.7143, + "step": 1738 + }, + { + "epoch": 1.9057534246575343, + "grad_norm": 2.3924875259399414, + "learning_rate": 2.0572705117790417e-06, + "loss": 1.75, + "step": 1739 + }, + { + "epoch": 1.9068493150684933, + "grad_norm": 2.4452881813049316, + "learning_rate": 2.055239642567019e-06, + "loss": 1.7831, + "step": 1740 + }, + { + "epoch": 1.9068493150684933, + "eval_loss": 1.9599454402923584, + "eval_runtime": 37.3026, + "eval_samples_per_second": 26.808, + "eval_steps_per_second": 1.126, + "step": 1740 + }, + { + "epoch": 1.907945205479452, + "grad_norm": 2.2778351306915283, + "learning_rate": 2.0532087733549962e-06, + "loss": 1.7633, + "step": 1741 + }, + { + "epoch": 1.909041095890411, + "grad_norm": 2.3171589374542236, + "learning_rate": 2.0511779041429735e-06, + "loss": 1.7726, + "step": 1742 + }, + { + "epoch": 1.91013698630137, + "grad_norm": 2.286094903945923, + "learning_rate": 2.0491470349309508e-06, + "loss": 1.7458, + "step": 1743 + }, + { + "epoch": 1.9112328767123288, + "grad_norm": 2.5381743907928467, + "learning_rate": 2.047116165718928e-06, + "loss": 1.777, + "step": 1744 + }, + { + "epoch": 1.9123287671232876, + "grad_norm": 2.3261141777038574, + "learning_rate": 2.0450852965069053e-06, + "loss": 1.7715, + "step": 1745 + }, + { + "epoch": 1.9134246575342466, + "grad_norm": 2.351914167404175, + "learning_rate": 2.0430544272948826e-06, + "loss": 1.8214, + "step": 1746 + }, + { + "epoch": 1.9145205479452054, + "grad_norm": 2.274226427078247, + "learning_rate": 2.04102355808286e-06, + "loss": 1.763, + "step": 1747 + }, + { + "epoch": 1.9156164383561642, + "grad_norm": 2.4025075435638428, + "learning_rate": 2.038992688870837e-06, + "loss": 1.7072, + "step": 1748 + }, + { + "epoch": 1.9167123287671233, + "grad_norm": 2.334024429321289, + "learning_rate": 2.0369618196588144e-06, + "loss": 1.7381, + "step": 1749 + }, + { + "epoch": 1.9178082191780823, + "grad_norm": 2.3011622428894043, + "learning_rate": 2.0349309504467917e-06, + "loss": 1.7432, + "step": 1750 + }, + { + "epoch": 1.9178082191780823, + "eval_loss": 1.958349347114563, + "eval_runtime": 37.2591, + "eval_samples_per_second": 26.839, + "eval_steps_per_second": 1.127, + "step": 1750 + }, + { + "epoch": 1.9189041095890411, + "grad_norm": 2.643063545227051, + "learning_rate": 2.032900081234769e-06, + "loss": 1.7931, + "step": 1751 + }, + { + "epoch": 1.92, + "grad_norm": 2.3357834815979004, + "learning_rate": 2.030869212022746e-06, + "loss": 1.7933, + "step": 1752 + }, + { + "epoch": 1.921095890410959, + "grad_norm": 2.2614479064941406, + "learning_rate": 2.028838342810723e-06, + "loss": 1.7611, + "step": 1753 + }, + { + "epoch": 1.9221917808219178, + "grad_norm": 2.3057546615600586, + "learning_rate": 2.0268074735987004e-06, + "loss": 1.7406, + "step": 1754 + }, + { + "epoch": 1.9232876712328766, + "grad_norm": 2.6138038635253906, + "learning_rate": 2.0247766043866777e-06, + "loss": 1.7303, + "step": 1755 + }, + { + "epoch": 1.9243835616438356, + "grad_norm": 2.3803296089172363, + "learning_rate": 2.022745735174655e-06, + "loss": 1.8023, + "step": 1756 + }, + { + "epoch": 1.9254794520547946, + "grad_norm": 2.3436431884765625, + "learning_rate": 2.020714865962632e-06, + "loss": 1.7654, + "step": 1757 + }, + { + "epoch": 1.9265753424657535, + "grad_norm": 2.4103317260742188, + "learning_rate": 2.0186839967506095e-06, + "loss": 1.7932, + "step": 1758 + }, + { + "epoch": 1.9276712328767123, + "grad_norm": 2.4537675380706787, + "learning_rate": 2.0166531275385868e-06, + "loss": 1.761, + "step": 1759 + }, + { + "epoch": 1.9287671232876713, + "grad_norm": 2.3816988468170166, + "learning_rate": 2.014622258326564e-06, + "loss": 1.7276, + "step": 1760 + }, + { + "epoch": 1.9287671232876713, + "eval_loss": 1.9558053016662598, + "eval_runtime": 37.2417, + "eval_samples_per_second": 26.852, + "eval_steps_per_second": 1.128, + "step": 1760 + }, + { + "epoch": 1.9298630136986301, + "grad_norm": 2.37233304977417, + "learning_rate": 2.0125913891145413e-06, + "loss": 1.7674, + "step": 1761 + }, + { + "epoch": 1.930958904109589, + "grad_norm": 2.4043285846710205, + "learning_rate": 2.0105605199025186e-06, + "loss": 1.7638, + "step": 1762 + }, + { + "epoch": 1.932054794520548, + "grad_norm": 2.3402836322784424, + "learning_rate": 2.008529650690496e-06, + "loss": 1.7269, + "step": 1763 + }, + { + "epoch": 1.933150684931507, + "grad_norm": 2.42535400390625, + "learning_rate": 2.0064987814784727e-06, + "loss": 1.7965, + "step": 1764 + }, + { + "epoch": 1.9342465753424658, + "grad_norm": 2.361659288406372, + "learning_rate": 2.00446791226645e-06, + "loss": 1.7467, + "step": 1765 + }, + { + "epoch": 1.9353424657534246, + "grad_norm": 2.455253839492798, + "learning_rate": 2.0024370430544273e-06, + "loss": 1.7359, + "step": 1766 + }, + { + "epoch": 1.9364383561643836, + "grad_norm": 2.3062705993652344, + "learning_rate": 2.0004061738424045e-06, + "loss": 1.751, + "step": 1767 + }, + { + "epoch": 1.9375342465753425, + "grad_norm": 2.3683881759643555, + "learning_rate": 1.998375304630382e-06, + "loss": 1.8032, + "step": 1768 + }, + { + "epoch": 1.9386301369863013, + "grad_norm": 2.2175045013427734, + "learning_rate": 1.996344435418359e-06, + "loss": 1.7629, + "step": 1769 + }, + { + "epoch": 1.9397260273972603, + "grad_norm": 2.4331724643707275, + "learning_rate": 1.9943135662063364e-06, + "loss": 1.7899, + "step": 1770 + }, + { + "epoch": 1.9397260273972603, + "eval_loss": 1.9554719924926758, + "eval_runtime": 37.2636, + "eval_samples_per_second": 26.836, + "eval_steps_per_second": 1.127, + "step": 1770 + }, + { + "epoch": 1.9408219178082193, + "grad_norm": 2.370785713195801, + "learning_rate": 1.9922826969943136e-06, + "loss": 1.7687, + "step": 1771 + }, + { + "epoch": 1.941917808219178, + "grad_norm": 2.4047751426696777, + "learning_rate": 1.990251827782291e-06, + "loss": 1.804, + "step": 1772 + }, + { + "epoch": 1.943013698630137, + "grad_norm": 2.4597785472869873, + "learning_rate": 1.988220958570268e-06, + "loss": 1.7731, + "step": 1773 + }, + { + "epoch": 1.944109589041096, + "grad_norm": 2.388749837875366, + "learning_rate": 1.9861900893582455e-06, + "loss": 1.7945, + "step": 1774 + }, + { + "epoch": 1.9452054794520548, + "grad_norm": 2.3419551849365234, + "learning_rate": 1.9841592201462227e-06, + "loss": 1.7987, + "step": 1775 + }, + { + "epoch": 1.9463013698630136, + "grad_norm": 2.43770432472229, + "learning_rate": 1.9821283509342e-06, + "loss": 1.7183, + "step": 1776 + }, + { + "epoch": 1.9473972602739726, + "grad_norm": 2.333385467529297, + "learning_rate": 1.9800974817221773e-06, + "loss": 1.7283, + "step": 1777 + }, + { + "epoch": 1.9484931506849315, + "grad_norm": 2.391052722930908, + "learning_rate": 1.9780666125101546e-06, + "loss": 1.776, + "step": 1778 + }, + { + "epoch": 1.9495890410958903, + "grad_norm": 2.375338077545166, + "learning_rate": 1.976035743298132e-06, + "loss": 1.7495, + "step": 1779 + }, + { + "epoch": 1.9506849315068493, + "grad_norm": 2.4482579231262207, + "learning_rate": 1.974004874086109e-06, + "loss": 1.8256, + "step": 1780 + }, + { + "epoch": 1.9506849315068493, + "eval_loss": 1.9600462913513184, + "eval_runtime": 37.2638, + "eval_samples_per_second": 26.836, + "eval_steps_per_second": 1.127, + "step": 1780 + }, + { + "epoch": 1.9517808219178083, + "grad_norm": 2.296039342880249, + "learning_rate": 1.9719740048740864e-06, + "loss": 1.6996, + "step": 1781 + }, + { + "epoch": 1.9528767123287671, + "grad_norm": 2.2494969367980957, + "learning_rate": 1.9699431356620637e-06, + "loss": 1.7401, + "step": 1782 + }, + { + "epoch": 1.953972602739726, + "grad_norm": 2.4078574180603027, + "learning_rate": 1.967912266450041e-06, + "loss": 1.7374, + "step": 1783 + }, + { + "epoch": 1.955068493150685, + "grad_norm": 2.4018476009368896, + "learning_rate": 1.9658813972380182e-06, + "loss": 1.731, + "step": 1784 + }, + { + "epoch": 1.9561643835616438, + "grad_norm": 2.3429362773895264, + "learning_rate": 1.9638505280259955e-06, + "loss": 1.8531, + "step": 1785 + }, + { + "epoch": 1.9572602739726026, + "grad_norm": 2.4287197589874268, + "learning_rate": 1.9618196588139728e-06, + "loss": 1.7271, + "step": 1786 + }, + { + "epoch": 1.9583561643835616, + "grad_norm": 2.338541030883789, + "learning_rate": 1.95978878960195e-06, + "loss": 1.7594, + "step": 1787 + }, + { + "epoch": 1.9594520547945207, + "grad_norm": 2.531425952911377, + "learning_rate": 1.9577579203899273e-06, + "loss": 1.8374, + "step": 1788 + }, + { + "epoch": 1.9605479452054795, + "grad_norm": 2.413638114929199, + "learning_rate": 1.955727051177904e-06, + "loss": 1.8178, + "step": 1789 + }, + { + "epoch": 1.9616438356164383, + "grad_norm": 2.303318977355957, + "learning_rate": 1.9536961819658815e-06, + "loss": 1.7429, + "step": 1790 + }, + { + "epoch": 1.9616438356164383, + "eval_loss": 1.9622509479522705, + "eval_runtime": 37.2699, + "eval_samples_per_second": 26.831, + "eval_steps_per_second": 1.127, + "step": 1790 + }, + { + "epoch": 1.9627397260273973, + "grad_norm": 2.400160074234009, + "learning_rate": 1.9516653127538587e-06, + "loss": 1.7136, + "step": 1791 + }, + { + "epoch": 1.9638356164383561, + "grad_norm": 2.3861589431762695, + "learning_rate": 1.949634443541836e-06, + "loss": 1.7146, + "step": 1792 + }, + { + "epoch": 1.964931506849315, + "grad_norm": 2.272711753845215, + "learning_rate": 1.9476035743298133e-06, + "loss": 1.7177, + "step": 1793 + }, + { + "epoch": 1.966027397260274, + "grad_norm": 2.4745068550109863, + "learning_rate": 1.9455727051177906e-06, + "loss": 1.7938, + "step": 1794 + }, + { + "epoch": 1.967123287671233, + "grad_norm": 2.4643466472625732, + "learning_rate": 1.943541835905768e-06, + "loss": 1.7601, + "step": 1795 + }, + { + "epoch": 1.9682191780821918, + "grad_norm": 2.5346457958221436, + "learning_rate": 1.941510966693745e-06, + "loss": 1.8132, + "step": 1796 + }, + { + "epoch": 1.9693150684931506, + "grad_norm": 2.280719041824341, + "learning_rate": 1.9394800974817224e-06, + "loss": 1.7101, + "step": 1797 + }, + { + "epoch": 1.9704109589041097, + "grad_norm": 2.3304555416107178, + "learning_rate": 1.9374492282696997e-06, + "loss": 1.7809, + "step": 1798 + }, + { + "epoch": 1.9715068493150685, + "grad_norm": 2.409796714782715, + "learning_rate": 1.935418359057677e-06, + "loss": 1.8127, + "step": 1799 + }, + { + "epoch": 1.9726027397260273, + "grad_norm": 2.208312749862671, + "learning_rate": 1.9333874898456542e-06, + "loss": 1.7752, + "step": 1800 + }, + { + "epoch": 1.9726027397260273, + "eval_loss": 1.96150541305542, + "eval_runtime": 37.3221, + "eval_samples_per_second": 26.794, + "eval_steps_per_second": 1.125, + "step": 1800 + }, + { + "epoch": 1.9736986301369863, + "grad_norm": 2.374393939971924, + "learning_rate": 1.931356620633631e-06, + "loss": 1.8188, + "step": 1801 + }, + { + "epoch": 1.9747945205479454, + "grad_norm": 2.3343780040740967, + "learning_rate": 1.9293257514216083e-06, + "loss": 1.7035, + "step": 1802 + }, + { + "epoch": 1.9758904109589042, + "grad_norm": 2.307894229888916, + "learning_rate": 1.9272948822095856e-06, + "loss": 1.7651, + "step": 1803 + }, + { + "epoch": 1.976986301369863, + "grad_norm": 2.387434244155884, + "learning_rate": 1.925264012997563e-06, + "loss": 1.7021, + "step": 1804 + }, + { + "epoch": 1.978082191780822, + "grad_norm": 2.3444128036499023, + "learning_rate": 1.92323314378554e-06, + "loss": 1.7585, + "step": 1805 + }, + { + "epoch": 1.9791780821917808, + "grad_norm": 2.371931791305542, + "learning_rate": 1.9212022745735174e-06, + "loss": 1.7932, + "step": 1806 + }, + { + "epoch": 1.9802739726027396, + "grad_norm": 2.3159632682800293, + "learning_rate": 1.9191714053614947e-06, + "loss": 1.722, + "step": 1807 + }, + { + "epoch": 1.9813698630136987, + "grad_norm": 2.4669857025146484, + "learning_rate": 1.917140536149472e-06, + "loss": 1.7326, + "step": 1808 + }, + { + "epoch": 1.9824657534246577, + "grad_norm": 2.365138292312622, + "learning_rate": 1.9151096669374493e-06, + "loss": 1.7013, + "step": 1809 + }, + { + "epoch": 1.9835616438356163, + "grad_norm": 2.539698362350464, + "learning_rate": 1.9130787977254266e-06, + "loss": 1.8082, + "step": 1810 + }, + { + "epoch": 1.9835616438356163, + "eval_loss": 1.9535884857177734, + "eval_runtime": 37.3497, + "eval_samples_per_second": 26.774, + "eval_steps_per_second": 1.125, + "step": 1810 + }, + { + "epoch": 1.9846575342465753, + "grad_norm": 2.386599540710449, + "learning_rate": 1.911047928513404e-06, + "loss": 1.7835, + "step": 1811 + }, + { + "epoch": 1.9857534246575343, + "grad_norm": 2.376086711883545, + "learning_rate": 1.909017059301381e-06, + "loss": 1.7161, + "step": 1812 + }, + { + "epoch": 1.9868493150684932, + "grad_norm": 2.5276663303375244, + "learning_rate": 1.9069861900893586e-06, + "loss": 1.7222, + "step": 1813 + }, + { + "epoch": 1.987945205479452, + "grad_norm": 2.466137170791626, + "learning_rate": 1.9049553208773354e-06, + "loss": 1.7841, + "step": 1814 + }, + { + "epoch": 1.989041095890411, + "grad_norm": 2.361128091812134, + "learning_rate": 1.9029244516653127e-06, + "loss": 1.7354, + "step": 1815 + }, + { + "epoch": 1.9901369863013698, + "grad_norm": 2.4348530769348145, + "learning_rate": 1.90089358245329e-06, + "loss": 1.6832, + "step": 1816 + }, + { + "epoch": 1.9912328767123286, + "grad_norm": 2.342285633087158, + "learning_rate": 1.8988627132412673e-06, + "loss": 1.7351, + "step": 1817 + }, + { + "epoch": 1.9923287671232877, + "grad_norm": 2.521137237548828, + "learning_rate": 1.8968318440292445e-06, + "loss": 1.794, + "step": 1818 + }, + { + "epoch": 1.9934246575342467, + "grad_norm": 2.5968360900878906, + "learning_rate": 1.8948009748172218e-06, + "loss": 1.6976, + "step": 1819 + }, + { + "epoch": 1.9945205479452055, + "grad_norm": 2.734095335006714, + "learning_rate": 1.892770105605199e-06, + "loss": 1.7919, + "step": 1820 + }, + { + "epoch": 1.9945205479452055, + "eval_loss": 1.9536023139953613, + "eval_runtime": 37.5805, + "eval_samples_per_second": 26.61, + "eval_steps_per_second": 1.118, + "step": 1820 + }, + { + "epoch": 1.9956164383561643, + "grad_norm": 2.327333927154541, + "learning_rate": 1.8907392363931764e-06, + "loss": 1.7891, + "step": 1821 + }, + { + "epoch": 1.9967123287671233, + "grad_norm": 2.2902543544769287, + "learning_rate": 1.8887083671811536e-06, + "loss": 1.7552, + "step": 1822 + }, + { + "epoch": 1.9978082191780822, + "grad_norm": 2.3587660789489746, + "learning_rate": 1.886677497969131e-06, + "loss": 1.7641, + "step": 1823 + }, + { + "epoch": 1.998904109589041, + "grad_norm": 2.4068408012390137, + "learning_rate": 1.8846466287571082e-06, + "loss": 1.6622, + "step": 1824 + }, + { + "epoch": 2.0, + "grad_norm": 2.453514575958252, + "learning_rate": 1.8826157595450855e-06, + "loss": 1.7962, + "step": 1825 + }, + { + "epoch": 2.001095890410959, + "grad_norm": 2.4694175720214844, + "learning_rate": 1.8805848903330625e-06, + "loss": 1.6823, + "step": 1826 + }, + { + "epoch": 2.0021917808219176, + "grad_norm": 2.594013214111328, + "learning_rate": 1.8785540211210398e-06, + "loss": 1.8453, + "step": 1827 + }, + { + "epoch": 2.0032876712328767, + "grad_norm": 2.455796718597412, + "learning_rate": 1.876523151909017e-06, + "loss": 1.7604, + "step": 1828 + }, + { + "epoch": 2.0043835616438357, + "grad_norm": 2.819857120513916, + "learning_rate": 1.8744922826969944e-06, + "loss": 1.7177, + "step": 1829 + }, + { + "epoch": 2.0054794520547947, + "grad_norm": 2.845775842666626, + "learning_rate": 1.8724614134849716e-06, + "loss": 1.7007, + "step": 1830 + }, + { + "epoch": 2.0054794520547947, + "eval_loss": 1.9594684839248657, + "eval_runtime": 37.6856, + "eval_samples_per_second": 26.535, + "eval_steps_per_second": 1.114, + "step": 1830 + }, + { + "epoch": 2.0065753424657533, + "grad_norm": 2.6129424571990967, + "learning_rate": 1.870430544272949e-06, + "loss": 1.7301, + "step": 1831 + }, + { + "epoch": 2.0076712328767123, + "grad_norm": 2.762805223464966, + "learning_rate": 1.8683996750609262e-06, + "loss": 1.8294, + "step": 1832 + }, + { + "epoch": 2.0087671232876714, + "grad_norm": 2.6397714614868164, + "learning_rate": 1.8663688058489035e-06, + "loss": 1.7176, + "step": 1833 + }, + { + "epoch": 2.00986301369863, + "grad_norm": 2.6637110710144043, + "learning_rate": 1.8643379366368807e-06, + "loss": 1.7363, + "step": 1834 + }, + { + "epoch": 2.010958904109589, + "grad_norm": 2.732534408569336, + "learning_rate": 1.862307067424858e-06, + "loss": 1.7496, + "step": 1835 + }, + { + "epoch": 2.012054794520548, + "grad_norm": 2.7163467407226562, + "learning_rate": 1.8602761982128353e-06, + "loss": 1.7551, + "step": 1836 + }, + { + "epoch": 2.013150684931507, + "grad_norm": 2.442484140396118, + "learning_rate": 1.8582453290008126e-06, + "loss": 1.7761, + "step": 1837 + }, + { + "epoch": 2.0142465753424657, + "grad_norm": 2.544834852218628, + "learning_rate": 1.8562144597887898e-06, + "loss": 1.7251, + "step": 1838 + }, + { + "epoch": 2.0153424657534247, + "grad_norm": 2.6588494777679443, + "learning_rate": 1.854183590576767e-06, + "loss": 1.7381, + "step": 1839 + }, + { + "epoch": 2.0164383561643837, + "grad_norm": 2.5374786853790283, + "learning_rate": 1.8521527213647442e-06, + "loss": 1.7831, + "step": 1840 + }, + { + "epoch": 2.0164383561643837, + "eval_loss": 1.9550480842590332, + "eval_runtime": 37.3312, + "eval_samples_per_second": 26.787, + "eval_steps_per_second": 1.125, + "step": 1840 + }, + { + "epoch": 2.0175342465753423, + "grad_norm": 2.4791345596313477, + "learning_rate": 1.8501218521527215e-06, + "loss": 1.6937, + "step": 1841 + }, + { + "epoch": 2.0186301369863013, + "grad_norm": 2.7243971824645996, + "learning_rate": 1.8480909829406987e-06, + "loss": 1.6674, + "step": 1842 + }, + { + "epoch": 2.0197260273972604, + "grad_norm": 2.618680477142334, + "learning_rate": 1.846060113728676e-06, + "loss": 1.6994, + "step": 1843 + }, + { + "epoch": 2.020821917808219, + "grad_norm": 2.597929000854492, + "learning_rate": 1.8440292445166533e-06, + "loss": 1.7375, + "step": 1844 + }, + { + "epoch": 2.021917808219178, + "grad_norm": 2.860177993774414, + "learning_rate": 1.8419983753046306e-06, + "loss": 1.7128, + "step": 1845 + }, + { + "epoch": 2.023013698630137, + "grad_norm": 2.606898307800293, + "learning_rate": 1.8399675060926078e-06, + "loss": 1.7735, + "step": 1846 + }, + { + "epoch": 2.024109589041096, + "grad_norm": 2.5152604579925537, + "learning_rate": 1.8379366368805851e-06, + "loss": 1.75, + "step": 1847 + }, + { + "epoch": 2.0252054794520546, + "grad_norm": 2.678071975708008, + "learning_rate": 1.8359057676685624e-06, + "loss": 1.7114, + "step": 1848 + }, + { + "epoch": 2.0263013698630137, + "grad_norm": 2.673515558242798, + "learning_rate": 1.8338748984565397e-06, + "loss": 1.6992, + "step": 1849 + }, + { + "epoch": 2.0273972602739727, + "grad_norm": 2.5902516841888428, + "learning_rate": 1.831844029244517e-06, + "loss": 1.7195, + "step": 1850 + }, + { + "epoch": 2.0273972602739727, + "eval_loss": 1.953579068183899, + "eval_runtime": 37.304, + "eval_samples_per_second": 26.807, + "eval_steps_per_second": 1.126, + "step": 1850 + }, + { + "epoch": 2.0284931506849313, + "grad_norm": 2.4912915229797363, + "learning_rate": 1.829813160032494e-06, + "loss": 1.6955, + "step": 1851 + }, + { + "epoch": 2.0295890410958903, + "grad_norm": 2.683821678161621, + "learning_rate": 1.8277822908204713e-06, + "loss": 1.7057, + "step": 1852 + }, + { + "epoch": 2.0306849315068494, + "grad_norm": 2.9541101455688477, + "learning_rate": 1.8257514216084486e-06, + "loss": 1.7377, + "step": 1853 + }, + { + "epoch": 2.0317808219178084, + "grad_norm": 2.783851146697998, + "learning_rate": 1.8237205523964258e-06, + "loss": 1.7197, + "step": 1854 + }, + { + "epoch": 2.032876712328767, + "grad_norm": 2.5224950313568115, + "learning_rate": 1.8216896831844031e-06, + "loss": 1.7396, + "step": 1855 + }, + { + "epoch": 2.033972602739726, + "grad_norm": 2.627767562866211, + "learning_rate": 1.8196588139723804e-06, + "loss": 1.7773, + "step": 1856 + }, + { + "epoch": 2.035068493150685, + "grad_norm": 2.443873405456543, + "learning_rate": 1.8176279447603577e-06, + "loss": 1.6656, + "step": 1857 + }, + { + "epoch": 2.0361643835616436, + "grad_norm": 2.648056745529175, + "learning_rate": 1.815597075548335e-06, + "loss": 1.6954, + "step": 1858 + }, + { + "epoch": 2.0372602739726027, + "grad_norm": 2.536935329437256, + "learning_rate": 1.8135662063363122e-06, + "loss": 1.7427, + "step": 1859 + }, + { + "epoch": 2.0383561643835617, + "grad_norm": 2.42183256149292, + "learning_rate": 1.8115353371242895e-06, + "loss": 1.7634, + "step": 1860 + }, + { + "epoch": 2.0383561643835617, + "eval_loss": 1.953397512435913, + "eval_runtime": 37.2886, + "eval_samples_per_second": 26.818, + "eval_steps_per_second": 1.126, + "step": 1860 + }, + { + "epoch": 2.0394520547945207, + "grad_norm": 2.538804769515991, + "learning_rate": 1.8095044679122668e-06, + "loss": 1.6712, + "step": 1861 + }, + { + "epoch": 2.0405479452054793, + "grad_norm": 2.5820095539093018, + "learning_rate": 1.807473598700244e-06, + "loss": 1.7215, + "step": 1862 + }, + { + "epoch": 2.0416438356164384, + "grad_norm": 2.6997382640838623, + "learning_rate": 1.805442729488221e-06, + "loss": 1.7117, + "step": 1863 + }, + { + "epoch": 2.0427397260273974, + "grad_norm": 2.775996685028076, + "learning_rate": 1.8034118602761982e-06, + "loss": 1.7355, + "step": 1864 + }, + { + "epoch": 2.043835616438356, + "grad_norm": 2.6333167552948, + "learning_rate": 1.8013809910641754e-06, + "loss": 1.7343, + "step": 1865 + }, + { + "epoch": 2.044931506849315, + "grad_norm": 2.5830278396606445, + "learning_rate": 1.7993501218521527e-06, + "loss": 1.7339, + "step": 1866 + }, + { + "epoch": 2.046027397260274, + "grad_norm": 2.6145055294036865, + "learning_rate": 1.79731925264013e-06, + "loss": 1.6835, + "step": 1867 + }, + { + "epoch": 2.047123287671233, + "grad_norm": 2.568159580230713, + "learning_rate": 1.7952883834281073e-06, + "loss": 1.7574, + "step": 1868 + }, + { + "epoch": 2.0482191780821917, + "grad_norm": 2.5235254764556885, + "learning_rate": 1.7932575142160846e-06, + "loss": 1.7067, + "step": 1869 + }, + { + "epoch": 2.0493150684931507, + "grad_norm": 2.464035987854004, + "learning_rate": 1.7912266450040618e-06, + "loss": 1.7451, + "step": 1870 + }, + { + "epoch": 2.0493150684931507, + "eval_loss": 1.9485031366348267, + "eval_runtime": 37.2648, + "eval_samples_per_second": 26.835, + "eval_steps_per_second": 1.127, + "step": 1870 + }, + { + "epoch": 2.0504109589041097, + "grad_norm": 2.4328689575195312, + "learning_rate": 1.789195775792039e-06, + "loss": 1.7508, + "step": 1871 + }, + { + "epoch": 2.0515068493150683, + "grad_norm": 2.5483932495117188, + "learning_rate": 1.7871649065800164e-06, + "loss": 1.6488, + "step": 1872 + }, + { + "epoch": 2.0526027397260274, + "grad_norm": 2.5290846824645996, + "learning_rate": 1.7851340373679937e-06, + "loss": 1.7939, + "step": 1873 + }, + { + "epoch": 2.0536986301369864, + "grad_norm": 2.506943702697754, + "learning_rate": 1.783103168155971e-06, + "loss": 1.6813, + "step": 1874 + }, + { + "epoch": 2.0547945205479454, + "grad_norm": 2.5863115787506104, + "learning_rate": 1.7810722989439482e-06, + "loss": 1.7104, + "step": 1875 + }, + { + "epoch": 2.055890410958904, + "grad_norm": 2.6402666568756104, + "learning_rate": 1.7790414297319253e-06, + "loss": 1.7311, + "step": 1876 + }, + { + "epoch": 2.056986301369863, + "grad_norm": 2.6700708866119385, + "learning_rate": 1.7770105605199025e-06, + "loss": 1.695, + "step": 1877 + }, + { + "epoch": 2.058082191780822, + "grad_norm": 2.580693244934082, + "learning_rate": 1.7749796913078798e-06, + "loss": 1.6763, + "step": 1878 + }, + { + "epoch": 2.0591780821917807, + "grad_norm": 2.6035709381103516, + "learning_rate": 1.772948822095857e-06, + "loss": 1.7389, + "step": 1879 + }, + { + "epoch": 2.0602739726027397, + "grad_norm": 2.508784532546997, + "learning_rate": 1.7709179528838344e-06, + "loss": 1.7762, + "step": 1880 + }, + { + "epoch": 2.0602739726027397, + "eval_loss": 1.9516879320144653, + "eval_runtime": 37.2284, + "eval_samples_per_second": 26.861, + "eval_steps_per_second": 1.128, + "step": 1880 + }, + { + "epoch": 2.0613698630136987, + "grad_norm": 2.47988224029541, + "learning_rate": 1.7688870836718116e-06, + "loss": 1.7659, + "step": 1881 + }, + { + "epoch": 2.0624657534246573, + "grad_norm": 2.5782723426818848, + "learning_rate": 1.766856214459789e-06, + "loss": 1.7179, + "step": 1882 + }, + { + "epoch": 2.0635616438356164, + "grad_norm": 2.5616347789764404, + "learning_rate": 1.7648253452477662e-06, + "loss": 1.7163, + "step": 1883 + }, + { + "epoch": 2.0646575342465754, + "grad_norm": 2.4464340209960938, + "learning_rate": 1.7627944760357435e-06, + "loss": 1.7033, + "step": 1884 + }, + { + "epoch": 2.0657534246575344, + "grad_norm": 2.6022887229919434, + "learning_rate": 1.7607636068237208e-06, + "loss": 1.7823, + "step": 1885 + }, + { + "epoch": 2.066849315068493, + "grad_norm": 2.5734951496124268, + "learning_rate": 1.758732737611698e-06, + "loss": 1.7486, + "step": 1886 + }, + { + "epoch": 2.067945205479452, + "grad_norm": 2.6162502765655518, + "learning_rate": 1.7567018683996753e-06, + "loss": 1.7285, + "step": 1887 + }, + { + "epoch": 2.069041095890411, + "grad_norm": 2.654008388519287, + "learning_rate": 1.7546709991876524e-06, + "loss": 1.7195, + "step": 1888 + }, + { + "epoch": 2.0701369863013697, + "grad_norm": 2.64469575881958, + "learning_rate": 1.7526401299756296e-06, + "loss": 1.7247, + "step": 1889 + }, + { + "epoch": 2.0712328767123287, + "grad_norm": 2.5627946853637695, + "learning_rate": 1.750609260763607e-06, + "loss": 1.7254, + "step": 1890 + }, + { + "epoch": 2.0712328767123287, + "eval_loss": 1.951324224472046, + "eval_runtime": 37.2566, + "eval_samples_per_second": 26.841, + "eval_steps_per_second": 1.127, + "step": 1890 + }, + { + "epoch": 2.0723287671232877, + "grad_norm": 2.656520128250122, + "learning_rate": 1.7485783915515842e-06, + "loss": 1.7236, + "step": 1891 + }, + { + "epoch": 2.0734246575342468, + "grad_norm": 2.6355814933776855, + "learning_rate": 1.7465475223395615e-06, + "loss": 1.7118, + "step": 1892 + }, + { + "epoch": 2.0745205479452054, + "grad_norm": 2.5709362030029297, + "learning_rate": 1.7445166531275387e-06, + "loss": 1.685, + "step": 1893 + }, + { + "epoch": 2.0756164383561644, + "grad_norm": 2.555422782897949, + "learning_rate": 1.742485783915516e-06, + "loss": 1.8356, + "step": 1894 + }, + { + "epoch": 2.0767123287671234, + "grad_norm": 2.6279077529907227, + "learning_rate": 1.7404549147034933e-06, + "loss": 1.8079, + "step": 1895 + }, + { + "epoch": 2.077808219178082, + "grad_norm": 2.5886247158050537, + "learning_rate": 1.7384240454914706e-06, + "loss": 1.639, + "step": 1896 + }, + { + "epoch": 2.078904109589041, + "grad_norm": 2.562755584716797, + "learning_rate": 1.7363931762794478e-06, + "loss": 1.7345, + "step": 1897 + }, + { + "epoch": 2.08, + "grad_norm": 2.670344591140747, + "learning_rate": 1.7343623070674251e-06, + "loss": 1.7594, + "step": 1898 + }, + { + "epoch": 2.081095890410959, + "grad_norm": 2.567506790161133, + "learning_rate": 1.7323314378554024e-06, + "loss": 1.7161, + "step": 1899 + }, + { + "epoch": 2.0821917808219177, + "grad_norm": 2.562826156616211, + "learning_rate": 1.7303005686433797e-06, + "loss": 1.7275, + "step": 1900 + }, + { + "epoch": 2.0821917808219177, + "eval_loss": 1.9498006105422974, + "eval_runtime": 37.3069, + "eval_samples_per_second": 26.805, + "eval_steps_per_second": 1.126, + "step": 1900 + }, + { + "epoch": 2.0832876712328767, + "grad_norm": 2.538393974304199, + "learning_rate": 1.7282696994313567e-06, + "loss": 1.7187, + "step": 1901 + }, + { + "epoch": 2.0843835616438358, + "grad_norm": 2.687603235244751, + "learning_rate": 1.726238830219334e-06, + "loss": 1.7349, + "step": 1902 + }, + { + "epoch": 2.0854794520547943, + "grad_norm": 2.5806703567504883, + "learning_rate": 1.7242079610073113e-06, + "loss": 1.7135, + "step": 1903 + }, + { + "epoch": 2.0865753424657534, + "grad_norm": 2.468123435974121, + "learning_rate": 1.7221770917952886e-06, + "loss": 1.7946, + "step": 1904 + }, + { + "epoch": 2.0876712328767124, + "grad_norm": 2.514617443084717, + "learning_rate": 1.7201462225832658e-06, + "loss": 1.7403, + "step": 1905 + }, + { + "epoch": 2.0887671232876714, + "grad_norm": 2.7283825874328613, + "learning_rate": 1.7181153533712431e-06, + "loss": 1.7369, + "step": 1906 + }, + { + "epoch": 2.08986301369863, + "grad_norm": 2.5634899139404297, + "learning_rate": 1.7160844841592204e-06, + "loss": 1.6337, + "step": 1907 + }, + { + "epoch": 2.090958904109589, + "grad_norm": 2.635281801223755, + "learning_rate": 1.7140536149471977e-06, + "loss": 1.6836, + "step": 1908 + }, + { + "epoch": 2.092054794520548, + "grad_norm": 2.550048351287842, + "learning_rate": 1.712022745735175e-06, + "loss": 1.7171, + "step": 1909 + }, + { + "epoch": 2.0931506849315067, + "grad_norm": 2.62712025642395, + "learning_rate": 1.7099918765231522e-06, + "loss": 1.7027, + "step": 1910 + }, + { + "epoch": 2.0931506849315067, + "eval_loss": 1.9503322839736938, + "eval_runtime": 37.3435, + "eval_samples_per_second": 26.778, + "eval_steps_per_second": 1.125, + "step": 1910 + }, + { + "epoch": 2.0942465753424657, + "grad_norm": 2.540708541870117, + "learning_rate": 1.7079610073111295e-06, + "loss": 1.7358, + "step": 1911 + }, + { + "epoch": 2.0953424657534248, + "grad_norm": 2.6046950817108154, + "learning_rate": 1.7059301380991068e-06, + "loss": 1.6538, + "step": 1912 + }, + { + "epoch": 2.0964383561643833, + "grad_norm": 2.4760568141937256, + "learning_rate": 1.7038992688870836e-06, + "loss": 1.7207, + "step": 1913 + }, + { + "epoch": 2.0975342465753424, + "grad_norm": 2.428621292114258, + "learning_rate": 1.701868399675061e-06, + "loss": 1.7092, + "step": 1914 + }, + { + "epoch": 2.0986301369863014, + "grad_norm": 2.6177377700805664, + "learning_rate": 1.6998375304630382e-06, + "loss": 1.7429, + "step": 1915 + }, + { + "epoch": 2.0997260273972604, + "grad_norm": 2.4981932640075684, + "learning_rate": 1.6978066612510155e-06, + "loss": 1.7996, + "step": 1916 + }, + { + "epoch": 2.100821917808219, + "grad_norm": 2.617943286895752, + "learning_rate": 1.6957757920389927e-06, + "loss": 1.7716, + "step": 1917 + }, + { + "epoch": 2.101917808219178, + "grad_norm": 2.7581987380981445, + "learning_rate": 1.69374492282697e-06, + "loss": 1.7194, + "step": 1918 + }, + { + "epoch": 2.103013698630137, + "grad_norm": 2.622145652770996, + "learning_rate": 1.6917140536149473e-06, + "loss": 1.6887, + "step": 1919 + }, + { + "epoch": 2.1041095890410957, + "grad_norm": 2.742100477218628, + "learning_rate": 1.6896831844029246e-06, + "loss": 1.7787, + "step": 1920 + }, + { + "epoch": 2.1041095890410957, + "eval_loss": 1.9521653652191162, + "eval_runtime": 37.5529, + "eval_samples_per_second": 26.629, + "eval_steps_per_second": 1.118, + "step": 1920 + }, + { + "epoch": 2.1052054794520547, + "grad_norm": 2.5809876918792725, + "learning_rate": 1.6876523151909018e-06, + "loss": 1.7392, + "step": 1921 + }, + { + "epoch": 2.1063013698630138, + "grad_norm": 2.7006125450134277, + "learning_rate": 1.6856214459788791e-06, + "loss": 1.7442, + "step": 1922 + }, + { + "epoch": 2.107397260273973, + "grad_norm": 2.566088914871216, + "learning_rate": 1.6835905767668564e-06, + "loss": 1.6862, + "step": 1923 + }, + { + "epoch": 2.1084931506849314, + "grad_norm": 2.508967638015747, + "learning_rate": 1.6815597075548337e-06, + "loss": 1.695, + "step": 1924 + }, + { + "epoch": 2.1095890410958904, + "grad_norm": 2.6175615787506104, + "learning_rate": 1.679528838342811e-06, + "loss": 1.785, + "step": 1925 + }, + { + "epoch": 2.1106849315068494, + "grad_norm": 2.6141958236694336, + "learning_rate": 1.677497969130788e-06, + "loss": 1.6952, + "step": 1926 + }, + { + "epoch": 2.111780821917808, + "grad_norm": 2.4894959926605225, + "learning_rate": 1.6754670999187653e-06, + "loss": 1.6947, + "step": 1927 + }, + { + "epoch": 2.112876712328767, + "grad_norm": 2.5664544105529785, + "learning_rate": 1.6734362307067426e-06, + "loss": 1.731, + "step": 1928 + }, + { + "epoch": 2.113972602739726, + "grad_norm": 2.5422523021698, + "learning_rate": 1.6714053614947198e-06, + "loss": 1.6765, + "step": 1929 + }, + { + "epoch": 2.115068493150685, + "grad_norm": 2.6428608894348145, + "learning_rate": 1.669374492282697e-06, + "loss": 1.6834, + "step": 1930 + }, + { + "epoch": 2.115068493150685, + "eval_loss": 1.952999472618103, + "eval_runtime": 37.6765, + "eval_samples_per_second": 26.542, + "eval_steps_per_second": 1.115, + "step": 1930 + }, + { + "epoch": 2.1161643835616437, + "grad_norm": 2.6088430881500244, + "learning_rate": 1.6673436230706744e-06, + "loss": 1.7713, + "step": 1931 + }, + { + "epoch": 2.1172602739726027, + "grad_norm": 2.49733304977417, + "learning_rate": 1.6653127538586517e-06, + "loss": 1.7461, + "step": 1932 + }, + { + "epoch": 2.118356164383562, + "grad_norm": 2.5760915279388428, + "learning_rate": 1.663281884646629e-06, + "loss": 1.6837, + "step": 1933 + }, + { + "epoch": 2.1194520547945204, + "grad_norm": 2.4771194458007812, + "learning_rate": 1.6612510154346062e-06, + "loss": 1.7191, + "step": 1934 + }, + { + "epoch": 2.1205479452054794, + "grad_norm": 2.6267881393432617, + "learning_rate": 1.6592201462225835e-06, + "loss": 1.7819, + "step": 1935 + }, + { + "epoch": 2.1216438356164384, + "grad_norm": 2.6205623149871826, + "learning_rate": 1.6571892770105608e-06, + "loss": 1.7443, + "step": 1936 + }, + { + "epoch": 2.1227397260273975, + "grad_norm": 2.660942792892456, + "learning_rate": 1.655158407798538e-06, + "loss": 1.783, + "step": 1937 + }, + { + "epoch": 2.123835616438356, + "grad_norm": 2.5452628135681152, + "learning_rate": 1.653127538586515e-06, + "loss": 1.7316, + "step": 1938 + }, + { + "epoch": 2.124931506849315, + "grad_norm": 2.542837142944336, + "learning_rate": 1.6510966693744924e-06, + "loss": 1.7624, + "step": 1939 + }, + { + "epoch": 2.126027397260274, + "grad_norm": 2.535205841064453, + "learning_rate": 1.6490658001624696e-06, + "loss": 1.7664, + "step": 1940 + }, + { + "epoch": 2.126027397260274, + "eval_loss": 1.9492406845092773, + "eval_runtime": 37.3497, + "eval_samples_per_second": 26.774, + "eval_steps_per_second": 1.125, + "step": 1940 + }, + { + "epoch": 2.1271232876712327, + "grad_norm": 2.7622485160827637, + "learning_rate": 1.647034930950447e-06, + "loss": 1.7603, + "step": 1941 + }, + { + "epoch": 2.1282191780821917, + "grad_norm": 2.629058599472046, + "learning_rate": 1.6450040617384242e-06, + "loss": 1.7036, + "step": 1942 + }, + { + "epoch": 2.1293150684931508, + "grad_norm": 2.543135643005371, + "learning_rate": 1.6429731925264015e-06, + "loss": 1.7172, + "step": 1943 + }, + { + "epoch": 2.1304109589041094, + "grad_norm": 2.792086601257324, + "learning_rate": 1.6409423233143788e-06, + "loss": 1.7486, + "step": 1944 + }, + { + "epoch": 2.1315068493150684, + "grad_norm": 2.7788681983947754, + "learning_rate": 1.638911454102356e-06, + "loss": 1.7136, + "step": 1945 + }, + { + "epoch": 2.1326027397260274, + "grad_norm": 2.5364747047424316, + "learning_rate": 1.6368805848903333e-06, + "loss": 1.6766, + "step": 1946 + }, + { + "epoch": 2.1336986301369865, + "grad_norm": 2.5749800205230713, + "learning_rate": 1.6348497156783106e-06, + "loss": 1.7627, + "step": 1947 + }, + { + "epoch": 2.134794520547945, + "grad_norm": 2.4931676387786865, + "learning_rate": 1.6328188464662879e-06, + "loss": 1.7049, + "step": 1948 + }, + { + "epoch": 2.135890410958904, + "grad_norm": 2.547396183013916, + "learning_rate": 1.6307879772542651e-06, + "loss": 1.7165, + "step": 1949 + }, + { + "epoch": 2.136986301369863, + "grad_norm": 2.5881359577178955, + "learning_rate": 1.628757108042242e-06, + "loss": 1.7508, + "step": 1950 + }, + { + "epoch": 2.136986301369863, + "eval_loss": 1.9464337825775146, + "eval_runtime": 37.284, + "eval_samples_per_second": 26.821, + "eval_steps_per_second": 1.126, + "step": 1950 + }, + { + "epoch": 2.138082191780822, + "grad_norm": 2.6131961345672607, + "learning_rate": 1.6267262388302193e-06, + "loss": 1.7717, + "step": 1951 + }, + { + "epoch": 2.1391780821917807, + "grad_norm": 2.5644335746765137, + "learning_rate": 1.6246953696181965e-06, + "loss": 1.7139, + "step": 1952 + }, + { + "epoch": 2.1402739726027398, + "grad_norm": 2.5851385593414307, + "learning_rate": 1.6226645004061738e-06, + "loss": 1.7473, + "step": 1953 + }, + { + "epoch": 2.141369863013699, + "grad_norm": 2.493952989578247, + "learning_rate": 1.620633631194151e-06, + "loss": 1.8209, + "step": 1954 + }, + { + "epoch": 2.1424657534246574, + "grad_norm": 2.5686917304992676, + "learning_rate": 1.6186027619821284e-06, + "loss": 1.7527, + "step": 1955 + }, + { + "epoch": 2.1435616438356164, + "grad_norm": 2.6349050998687744, + "learning_rate": 1.6165718927701056e-06, + "loss": 1.7033, + "step": 1956 + }, + { + "epoch": 2.1446575342465755, + "grad_norm": 2.7004778385162354, + "learning_rate": 1.614541023558083e-06, + "loss": 1.7498, + "step": 1957 + }, + { + "epoch": 2.145753424657534, + "grad_norm": 2.5406992435455322, + "learning_rate": 1.6125101543460602e-06, + "loss": 1.7605, + "step": 1958 + }, + { + "epoch": 2.146849315068493, + "grad_norm": 2.796125650405884, + "learning_rate": 1.6104792851340375e-06, + "loss": 1.7087, + "step": 1959 + }, + { + "epoch": 2.147945205479452, + "grad_norm": 2.703362226486206, + "learning_rate": 1.6084484159220147e-06, + "loss": 1.751, + "step": 1960 + }, + { + "epoch": 2.147945205479452, + "eval_loss": 1.9470759630203247, + "eval_runtime": 37.2721, + "eval_samples_per_second": 26.83, + "eval_steps_per_second": 1.127, + "step": 1960 + }, + { + "epoch": 2.149041095890411, + "grad_norm": 2.6464223861694336, + "learning_rate": 1.606417546709992e-06, + "loss": 1.7147, + "step": 1961 + }, + { + "epoch": 2.1501369863013697, + "grad_norm": 2.6499273777008057, + "learning_rate": 1.6043866774979693e-06, + "loss": 1.7015, + "step": 1962 + }, + { + "epoch": 2.1512328767123288, + "grad_norm": 2.583768844604492, + "learning_rate": 1.6023558082859464e-06, + "loss": 1.7158, + "step": 1963 + }, + { + "epoch": 2.152328767123288, + "grad_norm": 2.6245381832122803, + "learning_rate": 1.6003249390739236e-06, + "loss": 1.6824, + "step": 1964 + }, + { + "epoch": 2.1534246575342464, + "grad_norm": 2.497676134109497, + "learning_rate": 1.598294069861901e-06, + "loss": 1.7345, + "step": 1965 + }, + { + "epoch": 2.1545205479452054, + "grad_norm": 2.7218856811523438, + "learning_rate": 1.5962632006498782e-06, + "loss": 1.6988, + "step": 1966 + }, + { + "epoch": 2.1556164383561645, + "grad_norm": 2.7654666900634766, + "learning_rate": 1.5942323314378555e-06, + "loss": 1.7635, + "step": 1967 + }, + { + "epoch": 2.1567123287671235, + "grad_norm": 2.598526954650879, + "learning_rate": 1.5922014622258327e-06, + "loss": 1.6769, + "step": 1968 + }, + { + "epoch": 2.157808219178082, + "grad_norm": 2.5995986461639404, + "learning_rate": 1.59017059301381e-06, + "loss": 1.6928, + "step": 1969 + }, + { + "epoch": 2.158904109589041, + "grad_norm": 2.5692837238311768, + "learning_rate": 1.5881397238017873e-06, + "loss": 1.7415, + "step": 1970 + }, + { + "epoch": 2.158904109589041, + "eval_loss": 1.9472142457962036, + "eval_runtime": 37.258, + "eval_samples_per_second": 26.84, + "eval_steps_per_second": 1.127, + "step": 1970 + }, + { + "epoch": 2.16, + "grad_norm": 2.622732162475586, + "learning_rate": 1.5861088545897646e-06, + "loss": 1.7383, + "step": 1971 + }, + { + "epoch": 2.1610958904109587, + "grad_norm": 2.573005437850952, + "learning_rate": 1.5840779853777418e-06, + "loss": 1.6696, + "step": 1972 + }, + { + "epoch": 2.1621917808219178, + "grad_norm": 2.7022411823272705, + "learning_rate": 1.5820471161657191e-06, + "loss": 1.7231, + "step": 1973 + }, + { + "epoch": 2.163287671232877, + "grad_norm": 2.739145278930664, + "learning_rate": 1.5800162469536964e-06, + "loss": 1.7012, + "step": 1974 + }, + { + "epoch": 2.1643835616438354, + "grad_norm": 2.7034459114074707, + "learning_rate": 1.5779853777416735e-06, + "loss": 1.769, + "step": 1975 + }, + { + "epoch": 2.1654794520547944, + "grad_norm": 2.745359182357788, + "learning_rate": 1.5759545085296507e-06, + "loss": 1.7424, + "step": 1976 + }, + { + "epoch": 2.1665753424657535, + "grad_norm": 2.722721576690674, + "learning_rate": 1.573923639317628e-06, + "loss": 1.733, + "step": 1977 + }, + { + "epoch": 2.1676712328767125, + "grad_norm": 2.689074754714966, + "learning_rate": 1.5718927701056053e-06, + "loss": 1.7468, + "step": 1978 + }, + { + "epoch": 2.168767123287671, + "grad_norm": 2.6735317707061768, + "learning_rate": 1.5698619008935826e-06, + "loss": 1.7312, + "step": 1979 + }, + { + "epoch": 2.16986301369863, + "grad_norm": 2.6908175945281982, + "learning_rate": 1.5678310316815598e-06, + "loss": 1.7502, + "step": 1980 + }, + { + "epoch": 2.16986301369863, + "eval_loss": 1.947867512702942, + "eval_runtime": 37.1858, + "eval_samples_per_second": 26.892, + "eval_steps_per_second": 1.129, + "step": 1980 + }, + { + "epoch": 2.170958904109589, + "grad_norm": 2.524951934814453, + "learning_rate": 1.5658001624695371e-06, + "loss": 1.6863, + "step": 1981 + }, + { + "epoch": 2.172054794520548, + "grad_norm": 2.746180295944214, + "learning_rate": 1.5637692932575144e-06, + "loss": 1.7621, + "step": 1982 + }, + { + "epoch": 2.1731506849315068, + "grad_norm": 2.4799301624298096, + "learning_rate": 1.5617384240454917e-06, + "loss": 1.7121, + "step": 1983 + }, + { + "epoch": 2.174246575342466, + "grad_norm": 2.6821515560150146, + "learning_rate": 1.559707554833469e-06, + "loss": 1.7271, + "step": 1984 + }, + { + "epoch": 2.175342465753425, + "grad_norm": 2.572122812271118, + "learning_rate": 1.5576766856214462e-06, + "loss": 1.7671, + "step": 1985 + }, + { + "epoch": 2.1764383561643834, + "grad_norm": 2.614806652069092, + "learning_rate": 1.5556458164094235e-06, + "loss": 1.6411, + "step": 1986 + }, + { + "epoch": 2.1775342465753424, + "grad_norm": 2.602883815765381, + "learning_rate": 1.5536149471974008e-06, + "loss": 1.7143, + "step": 1987 + }, + { + "epoch": 2.1786301369863015, + "grad_norm": 2.5893354415893555, + "learning_rate": 1.5515840779853778e-06, + "loss": 1.6733, + "step": 1988 + }, + { + "epoch": 2.17972602739726, + "grad_norm": 2.652308702468872, + "learning_rate": 1.549553208773355e-06, + "loss": 1.7311, + "step": 1989 + }, + { + "epoch": 2.180821917808219, + "grad_norm": 2.7129695415496826, + "learning_rate": 1.5475223395613324e-06, + "loss": 1.7001, + "step": 1990 + }, + { + "epoch": 2.180821917808219, + "eval_loss": 1.9512702226638794, + "eval_runtime": 37.225, + "eval_samples_per_second": 26.864, + "eval_steps_per_second": 1.128, + "step": 1990 + }, + { + "epoch": 2.181917808219178, + "grad_norm": 2.563638210296631, + "learning_rate": 1.5454914703493097e-06, + "loss": 1.7118, + "step": 1991 + }, + { + "epoch": 2.183013698630137, + "grad_norm": 2.629056453704834, + "learning_rate": 1.543460601137287e-06, + "loss": 1.8041, + "step": 1992 + }, + { + "epoch": 2.1841095890410958, + "grad_norm": 2.6072700023651123, + "learning_rate": 1.5414297319252642e-06, + "loss": 1.7568, + "step": 1993 + }, + { + "epoch": 2.185205479452055, + "grad_norm": 2.525784730911255, + "learning_rate": 1.5393988627132415e-06, + "loss": 1.748, + "step": 1994 + }, + { + "epoch": 2.186301369863014, + "grad_norm": 2.564699649810791, + "learning_rate": 1.5373679935012188e-06, + "loss": 1.7148, + "step": 1995 + }, + { + "epoch": 2.1873972602739724, + "grad_norm": 2.7365031242370605, + "learning_rate": 1.535337124289196e-06, + "loss": 1.6714, + "step": 1996 + }, + { + "epoch": 2.1884931506849314, + "grad_norm": 2.719954252243042, + "learning_rate": 1.5333062550771733e-06, + "loss": 1.7301, + "step": 1997 + }, + { + "epoch": 2.1895890410958905, + "grad_norm": 2.7979655265808105, + "learning_rate": 1.5312753858651506e-06, + "loss": 1.8163, + "step": 1998 + }, + { + "epoch": 2.1906849315068495, + "grad_norm": 2.6182446479797363, + "learning_rate": 1.5292445166531279e-06, + "loss": 1.713, + "step": 1999 + }, + { + "epoch": 2.191780821917808, + "grad_norm": 2.7218992710113525, + "learning_rate": 1.5272136474411047e-06, + "loss": 1.7355, + "step": 2000 + }, + { + "epoch": 2.191780821917808, + "eval_loss": 1.950900673866272, + "eval_runtime": 37.2927, + "eval_samples_per_second": 26.815, + "eval_steps_per_second": 1.126, + "step": 2000 + }, + { + "epoch": 2.192876712328767, + "grad_norm": 2.7384393215179443, + "learning_rate": 1.525182778229082e-06, + "loss": 1.7777, + "step": 2001 + }, + { + "epoch": 2.193972602739726, + "grad_norm": 2.7223575115203857, + "learning_rate": 1.5231519090170593e-06, + "loss": 1.7136, + "step": 2002 + }, + { + "epoch": 2.1950684931506848, + "grad_norm": 2.5952577590942383, + "learning_rate": 1.5211210398050365e-06, + "loss": 1.6964, + "step": 2003 + }, + { + "epoch": 2.196164383561644, + "grad_norm": 2.626406669616699, + "learning_rate": 1.5190901705930138e-06, + "loss": 1.6996, + "step": 2004 + }, + { + "epoch": 2.197260273972603, + "grad_norm": 2.6563360691070557, + "learning_rate": 1.517059301380991e-06, + "loss": 1.7428, + "step": 2005 + }, + { + "epoch": 2.198356164383562, + "grad_norm": 2.647664785385132, + "learning_rate": 1.5150284321689684e-06, + "loss": 1.6953, + "step": 2006 + }, + { + "epoch": 2.1994520547945204, + "grad_norm": 2.5670430660247803, + "learning_rate": 1.5129975629569456e-06, + "loss": 1.7215, + "step": 2007 + }, + { + "epoch": 2.2005479452054795, + "grad_norm": 2.6148288249969482, + "learning_rate": 1.510966693744923e-06, + "loss": 1.724, + "step": 2008 + }, + { + "epoch": 2.2016438356164385, + "grad_norm": 2.8430588245391846, + "learning_rate": 1.5089358245329002e-06, + "loss": 1.7255, + "step": 2009 + }, + { + "epoch": 2.202739726027397, + "grad_norm": 2.5937509536743164, + "learning_rate": 1.5069049553208775e-06, + "loss": 1.7055, + "step": 2010 + }, + { + "epoch": 2.202739726027397, + "eval_loss": 1.9469048976898193, + "eval_runtime": 37.2173, + "eval_samples_per_second": 26.869, + "eval_steps_per_second": 1.129, + "step": 2010 + }, + { + "epoch": 2.203835616438356, + "grad_norm": 2.7167937755584717, + "learning_rate": 1.5048740861088547e-06, + "loss": 1.7474, + "step": 2011 + }, + { + "epoch": 2.204931506849315, + "grad_norm": 2.53837251663208, + "learning_rate": 1.502843216896832e-06, + "loss": 1.74, + "step": 2012 + }, + { + "epoch": 2.206027397260274, + "grad_norm": 2.604931592941284, + "learning_rate": 1.500812347684809e-06, + "loss": 1.6573, + "step": 2013 + }, + { + "epoch": 2.207123287671233, + "grad_norm": 2.6670846939086914, + "learning_rate": 1.4987814784727864e-06, + "loss": 1.7208, + "step": 2014 + }, + { + "epoch": 2.208219178082192, + "grad_norm": 2.677746534347534, + "learning_rate": 1.4967506092607636e-06, + "loss": 1.728, + "step": 2015 + }, + { + "epoch": 2.209315068493151, + "grad_norm": 2.6490612030029297, + "learning_rate": 1.494719740048741e-06, + "loss": 1.6728, + "step": 2016 + }, + { + "epoch": 2.2104109589041094, + "grad_norm": 2.6571807861328125, + "learning_rate": 1.4926888708367182e-06, + "loss": 1.7528, + "step": 2017 + }, + { + "epoch": 2.2115068493150685, + "grad_norm": 2.5933258533477783, + "learning_rate": 1.4906580016246955e-06, + "loss": 1.757, + "step": 2018 + }, + { + "epoch": 2.2126027397260275, + "grad_norm": 2.6451070308685303, + "learning_rate": 1.4886271324126727e-06, + "loss": 1.763, + "step": 2019 + }, + { + "epoch": 2.213698630136986, + "grad_norm": 2.5259275436401367, + "learning_rate": 1.48659626320065e-06, + "loss": 1.7979, + "step": 2020 + }, + { + "epoch": 2.213698630136986, + "eval_loss": 1.9480665922164917, + "eval_runtime": 37.5325, + "eval_samples_per_second": 26.644, + "eval_steps_per_second": 1.119, + "step": 2020 + }, + { + "epoch": 2.214794520547945, + "grad_norm": 2.6959381103515625, + "learning_rate": 1.4845653939886273e-06, + "loss": 1.6376, + "step": 2021 + }, + { + "epoch": 2.215890410958904, + "grad_norm": 2.6179497241973877, + "learning_rate": 1.4825345247766046e-06, + "loss": 1.6575, + "step": 2022 + }, + { + "epoch": 2.216986301369863, + "grad_norm": 2.5618393421173096, + "learning_rate": 1.4805036555645818e-06, + "loss": 1.6486, + "step": 2023 + }, + { + "epoch": 2.2180821917808218, + "grad_norm": 2.545051336288452, + "learning_rate": 1.4784727863525591e-06, + "loss": 1.7218, + "step": 2024 + }, + { + "epoch": 2.219178082191781, + "grad_norm": 2.6248512268066406, + "learning_rate": 1.4764419171405362e-06, + "loss": 1.691, + "step": 2025 + }, + { + "epoch": 2.22027397260274, + "grad_norm": 2.65645170211792, + "learning_rate": 1.4744110479285135e-06, + "loss": 1.7602, + "step": 2026 + }, + { + "epoch": 2.2213698630136984, + "grad_norm": 2.6771113872528076, + "learning_rate": 1.4723801787164907e-06, + "loss": 1.7573, + "step": 2027 + }, + { + "epoch": 2.2224657534246575, + "grad_norm": 2.5466039180755615, + "learning_rate": 1.470349309504468e-06, + "loss": 1.7604, + "step": 2028 + }, + { + "epoch": 2.2235616438356165, + "grad_norm": 2.685079574584961, + "learning_rate": 1.4683184402924453e-06, + "loss": 1.7341, + "step": 2029 + }, + { + "epoch": 2.2246575342465755, + "grad_norm": 2.6854352951049805, + "learning_rate": 1.4662875710804226e-06, + "loss": 1.7427, + "step": 2030 + }, + { + "epoch": 2.2246575342465755, + "eval_loss": 1.9459377527236938, + "eval_runtime": 37.5255, + "eval_samples_per_second": 26.649, + "eval_steps_per_second": 1.119, + "step": 2030 + }, + { + "epoch": 2.225753424657534, + "grad_norm": 2.667675733566284, + "learning_rate": 1.4642567018683998e-06, + "loss": 1.8328, + "step": 2031 + }, + { + "epoch": 2.226849315068493, + "grad_norm": 2.603386640548706, + "learning_rate": 1.4622258326563771e-06, + "loss": 1.7193, + "step": 2032 + }, + { + "epoch": 2.227945205479452, + "grad_norm": 2.4960451126098633, + "learning_rate": 1.4601949634443544e-06, + "loss": 1.695, + "step": 2033 + }, + { + "epoch": 2.2290410958904108, + "grad_norm": 2.755707025527954, + "learning_rate": 1.4581640942323317e-06, + "loss": 1.7475, + "step": 2034 + }, + { + "epoch": 2.23013698630137, + "grad_norm": 2.5849006175994873, + "learning_rate": 1.456133225020309e-06, + "loss": 1.8106, + "step": 2035 + }, + { + "epoch": 2.231232876712329, + "grad_norm": 2.690680980682373, + "learning_rate": 1.4541023558082862e-06, + "loss": 1.7368, + "step": 2036 + }, + { + "epoch": 2.232328767123288, + "grad_norm": 2.6760425567626953, + "learning_rate": 1.4520714865962635e-06, + "loss": 1.722, + "step": 2037 + }, + { + "epoch": 2.2334246575342465, + "grad_norm": 2.656379461288452, + "learning_rate": 1.4500406173842406e-06, + "loss": 1.7623, + "step": 2038 + }, + { + "epoch": 2.2345205479452055, + "grad_norm": 2.8213517665863037, + "learning_rate": 1.4480097481722178e-06, + "loss": 1.7344, + "step": 2039 + }, + { + "epoch": 2.2356164383561645, + "grad_norm": 2.877163887023926, + "learning_rate": 1.4459788789601951e-06, + "loss": 1.7184, + "step": 2040 + }, + { + "epoch": 2.2356164383561645, + "eval_loss": 1.9430633783340454, + "eval_runtime": 37.3102, + "eval_samples_per_second": 26.802, + "eval_steps_per_second": 1.126, + "step": 2040 + }, + { + "epoch": 2.236712328767123, + "grad_norm": 2.7285237312316895, + "learning_rate": 1.4439480097481724e-06, + "loss": 1.6826, + "step": 2041 + }, + { + "epoch": 2.237808219178082, + "grad_norm": 2.5208237171173096, + "learning_rate": 1.4419171405361497e-06, + "loss": 1.6788, + "step": 2042 + }, + { + "epoch": 2.238904109589041, + "grad_norm": 2.7415900230407715, + "learning_rate": 1.439886271324127e-06, + "loss": 1.7423, + "step": 2043 + }, + { + "epoch": 2.24, + "grad_norm": 2.760897397994995, + "learning_rate": 1.4378554021121042e-06, + "loss": 1.7341, + "step": 2044 + }, + { + "epoch": 2.241095890410959, + "grad_norm": 2.5650556087493896, + "learning_rate": 1.4358245329000815e-06, + "loss": 1.7036, + "step": 2045 + }, + { + "epoch": 2.242191780821918, + "grad_norm": 2.60780930519104, + "learning_rate": 1.4337936636880588e-06, + "loss": 1.7719, + "step": 2046 + }, + { + "epoch": 2.243287671232877, + "grad_norm": 2.592473268508911, + "learning_rate": 1.431762794476036e-06, + "loss": 1.7421, + "step": 2047 + }, + { + "epoch": 2.2443835616438355, + "grad_norm": 2.6672041416168213, + "learning_rate": 1.4297319252640133e-06, + "loss": 1.6865, + "step": 2048 + }, + { + "epoch": 2.2454794520547945, + "grad_norm": 2.6182847023010254, + "learning_rate": 1.4277010560519906e-06, + "loss": 1.731, + "step": 2049 + }, + { + "epoch": 2.2465753424657535, + "grad_norm": 2.8213913440704346, + "learning_rate": 1.4256701868399674e-06, + "loss": 1.6845, + "step": 2050 + }, + { + "epoch": 2.2465753424657535, + "eval_loss": 1.9434115886688232, + "eval_runtime": 37.3048, + "eval_samples_per_second": 26.806, + "eval_steps_per_second": 1.126, + "step": 2050 + }, + { + "epoch": 2.247671232876712, + "grad_norm": 2.7278084754943848, + "learning_rate": 1.4236393176279447e-06, + "loss": 1.8077, + "step": 2051 + }, + { + "epoch": 2.248767123287671, + "grad_norm": 2.5665392875671387, + "learning_rate": 1.421608448415922e-06, + "loss": 1.7235, + "step": 2052 + }, + { + "epoch": 2.24986301369863, + "grad_norm": 2.680154323577881, + "learning_rate": 1.4195775792038993e-06, + "loss": 1.8164, + "step": 2053 + }, + { + "epoch": 2.250958904109589, + "grad_norm": 2.770051956176758, + "learning_rate": 1.4175467099918765e-06, + "loss": 1.6948, + "step": 2054 + }, + { + "epoch": 2.252054794520548, + "grad_norm": 2.6189041137695312, + "learning_rate": 1.4155158407798538e-06, + "loss": 1.7369, + "step": 2055 + }, + { + "epoch": 2.253150684931507, + "grad_norm": 2.6655054092407227, + "learning_rate": 1.413484971567831e-06, + "loss": 1.7456, + "step": 2056 + }, + { + "epoch": 2.254246575342466, + "grad_norm": 2.6520164012908936, + "learning_rate": 1.4114541023558084e-06, + "loss": 1.6719, + "step": 2057 + }, + { + "epoch": 2.255342465753425, + "grad_norm": 2.594533681869507, + "learning_rate": 1.4094232331437856e-06, + "loss": 1.7078, + "step": 2058 + }, + { + "epoch": 2.2564383561643835, + "grad_norm": 2.592308759689331, + "learning_rate": 1.407392363931763e-06, + "loss": 1.7517, + "step": 2059 + }, + { + "epoch": 2.2575342465753425, + "grad_norm": 2.5815367698669434, + "learning_rate": 1.4053614947197402e-06, + "loss": 1.8262, + "step": 2060 + }, + { + "epoch": 2.2575342465753425, + "eval_loss": 1.945846438407898, + "eval_runtime": 37.186, + "eval_samples_per_second": 26.892, + "eval_steps_per_second": 1.129, + "step": 2060 + }, + { + "epoch": 2.2586301369863016, + "grad_norm": 2.643691301345825, + "learning_rate": 1.4033306255077175e-06, + "loss": 1.7559, + "step": 2061 + }, + { + "epoch": 2.25972602739726, + "grad_norm": 2.6091558933258057, + "learning_rate": 1.4012997562956945e-06, + "loss": 1.6433, + "step": 2062 + }, + { + "epoch": 2.260821917808219, + "grad_norm": 2.7172136306762695, + "learning_rate": 1.3992688870836718e-06, + "loss": 1.7735, + "step": 2063 + }, + { + "epoch": 2.261917808219178, + "grad_norm": 2.5673139095306396, + "learning_rate": 1.397238017871649e-06, + "loss": 1.7402, + "step": 2064 + }, + { + "epoch": 2.263013698630137, + "grad_norm": 2.711242198944092, + "learning_rate": 1.3952071486596264e-06, + "loss": 1.7183, + "step": 2065 + }, + { + "epoch": 2.264109589041096, + "grad_norm": 2.6999948024749756, + "learning_rate": 1.3931762794476036e-06, + "loss": 1.6856, + "step": 2066 + }, + { + "epoch": 2.265205479452055, + "grad_norm": 2.670964002609253, + "learning_rate": 1.391145410235581e-06, + "loss": 1.78, + "step": 2067 + }, + { + "epoch": 2.266301369863014, + "grad_norm": 2.4599525928497314, + "learning_rate": 1.3891145410235582e-06, + "loss": 1.7143, + "step": 2068 + }, + { + "epoch": 2.2673972602739725, + "grad_norm": 2.552266836166382, + "learning_rate": 1.3870836718115355e-06, + "loss": 1.6503, + "step": 2069 + }, + { + "epoch": 2.2684931506849315, + "grad_norm": 2.5764405727386475, + "learning_rate": 1.3850528025995127e-06, + "loss": 1.7046, + "step": 2070 + }, + { + "epoch": 2.2684931506849315, + "eval_loss": 1.9457523822784424, + "eval_runtime": 37.217, + "eval_samples_per_second": 26.869, + "eval_steps_per_second": 1.129, + "step": 2070 + }, + { + "epoch": 2.2695890410958905, + "grad_norm": 2.524228572845459, + "learning_rate": 1.38302193338749e-06, + "loss": 1.7078, + "step": 2071 + }, + { + "epoch": 2.270684931506849, + "grad_norm": 2.5531418323516846, + "learning_rate": 1.3809910641754673e-06, + "loss": 1.646, + "step": 2072 + }, + { + "epoch": 2.271780821917808, + "grad_norm": 2.705824613571167, + "learning_rate": 1.3789601949634446e-06, + "loss": 1.7407, + "step": 2073 + }, + { + "epoch": 2.272876712328767, + "grad_norm": 2.5266356468200684, + "learning_rate": 1.3769293257514218e-06, + "loss": 1.6472, + "step": 2074 + }, + { + "epoch": 2.2739726027397262, + "grad_norm": 2.5785422325134277, + "learning_rate": 1.374898456539399e-06, + "loss": 1.7561, + "step": 2075 + }, + { + "epoch": 2.275068493150685, + "grad_norm": 2.663120985031128, + "learning_rate": 1.3728675873273762e-06, + "loss": 1.6901, + "step": 2076 + }, + { + "epoch": 2.276164383561644, + "grad_norm": 2.6085517406463623, + "learning_rate": 1.3708367181153535e-06, + "loss": 1.7829, + "step": 2077 + }, + { + "epoch": 2.277260273972603, + "grad_norm": 2.6439929008483887, + "learning_rate": 1.3688058489033307e-06, + "loss": 1.7684, + "step": 2078 + }, + { + "epoch": 2.2783561643835615, + "grad_norm": 2.6614694595336914, + "learning_rate": 1.366774979691308e-06, + "loss": 1.726, + "step": 2079 + }, + { + "epoch": 2.2794520547945205, + "grad_norm": 2.7339539527893066, + "learning_rate": 1.3647441104792853e-06, + "loss": 1.7577, + "step": 2080 + }, + { + "epoch": 2.2794520547945205, + "eval_loss": 1.9456545114517212, + "eval_runtime": 37.174, + "eval_samples_per_second": 26.901, + "eval_steps_per_second": 1.13, + "step": 2080 + }, + { + "epoch": 2.2805479452054795, + "grad_norm": 2.603545665740967, + "learning_rate": 1.3627132412672626e-06, + "loss": 1.7297, + "step": 2081 + }, + { + "epoch": 2.281643835616438, + "grad_norm": 2.548466682434082, + "learning_rate": 1.3606823720552398e-06, + "loss": 1.6701, + "step": 2082 + }, + { + "epoch": 2.282739726027397, + "grad_norm": 2.6043875217437744, + "learning_rate": 1.3586515028432171e-06, + "loss": 1.7369, + "step": 2083 + }, + { + "epoch": 2.283835616438356, + "grad_norm": 2.7412586212158203, + "learning_rate": 1.3566206336311944e-06, + "loss": 1.654, + "step": 2084 + }, + { + "epoch": 2.2849315068493152, + "grad_norm": 2.669410228729248, + "learning_rate": 1.3545897644191717e-06, + "loss": 1.7186, + "step": 2085 + }, + { + "epoch": 2.286027397260274, + "grad_norm": 2.7052674293518066, + "learning_rate": 1.352558895207149e-06, + "loss": 1.7005, + "step": 2086 + }, + { + "epoch": 2.287123287671233, + "grad_norm": 2.5840418338775635, + "learning_rate": 1.3505280259951258e-06, + "loss": 1.7459, + "step": 2087 + }, + { + "epoch": 2.288219178082192, + "grad_norm": 2.6355416774749756, + "learning_rate": 1.348497156783103e-06, + "loss": 1.7243, + "step": 2088 + }, + { + "epoch": 2.289315068493151, + "grad_norm": 2.6697628498077393, + "learning_rate": 1.3464662875710804e-06, + "loss": 1.7597, + "step": 2089 + }, + { + "epoch": 2.2904109589041095, + "grad_norm": 2.679842233657837, + "learning_rate": 1.3444354183590576e-06, + "loss": 1.6557, + "step": 2090 + }, + { + "epoch": 2.2904109589041095, + "eval_loss": 1.9469218254089355, + "eval_runtime": 37.1615, + "eval_samples_per_second": 26.91, + "eval_steps_per_second": 1.13, + "step": 2090 + }, + { + "epoch": 2.2915068493150685, + "grad_norm": 2.8354430198669434, + "learning_rate": 1.342404549147035e-06, + "loss": 1.7496, + "step": 2091 + }, + { + "epoch": 2.2926027397260276, + "grad_norm": 2.6957387924194336, + "learning_rate": 1.3403736799350122e-06, + "loss": 1.6872, + "step": 2092 + }, + { + "epoch": 2.293698630136986, + "grad_norm": 2.5815868377685547, + "learning_rate": 1.3383428107229895e-06, + "loss": 1.7497, + "step": 2093 + }, + { + "epoch": 2.294794520547945, + "grad_norm": 2.727085828781128, + "learning_rate": 1.3363119415109667e-06, + "loss": 1.7723, + "step": 2094 + }, + { + "epoch": 2.2958904109589042, + "grad_norm": 2.6119000911712646, + "learning_rate": 1.334281072298944e-06, + "loss": 1.7414, + "step": 2095 + }, + { + "epoch": 2.296986301369863, + "grad_norm": 2.6403088569641113, + "learning_rate": 1.3322502030869213e-06, + "loss": 1.7398, + "step": 2096 + }, + { + "epoch": 2.298082191780822, + "grad_norm": 2.5066816806793213, + "learning_rate": 1.3302193338748986e-06, + "loss": 1.6974, + "step": 2097 + }, + { + "epoch": 2.299178082191781, + "grad_norm": 2.6601762771606445, + "learning_rate": 1.3281884646628758e-06, + "loss": 1.7578, + "step": 2098 + }, + { + "epoch": 2.30027397260274, + "grad_norm": 2.571499824523926, + "learning_rate": 1.3261575954508531e-06, + "loss": 1.7573, + "step": 2099 + }, + { + "epoch": 2.3013698630136985, + "grad_norm": 2.533543825149536, + "learning_rate": 1.3241267262388302e-06, + "loss": 1.7566, + "step": 2100 + }, + { + "epoch": 2.3013698630136985, + "eval_loss": 1.9441688060760498, + "eval_runtime": 37.2158, + "eval_samples_per_second": 26.87, + "eval_steps_per_second": 1.129, + "step": 2100 + }, + { + "epoch": 2.3024657534246575, + "grad_norm": 2.782416582107544, + "learning_rate": 1.3220958570268074e-06, + "loss": 1.714, + "step": 2101 + }, + { + "epoch": 2.3035616438356166, + "grad_norm": 2.742504358291626, + "learning_rate": 1.3200649878147847e-06, + "loss": 1.6677, + "step": 2102 + }, + { + "epoch": 2.304657534246575, + "grad_norm": 2.7778940200805664, + "learning_rate": 1.318034118602762e-06, + "loss": 1.7178, + "step": 2103 + }, + { + "epoch": 2.305753424657534, + "grad_norm": 2.7003014087677, + "learning_rate": 1.3160032493907393e-06, + "loss": 1.7691, + "step": 2104 + }, + { + "epoch": 2.3068493150684932, + "grad_norm": 2.6644296646118164, + "learning_rate": 1.3139723801787166e-06, + "loss": 1.7285, + "step": 2105 + }, + { + "epoch": 2.3079452054794523, + "grad_norm": 2.7859749794006348, + "learning_rate": 1.3119415109666938e-06, + "loss": 1.8026, + "step": 2106 + }, + { + "epoch": 2.309041095890411, + "grad_norm": 2.6862375736236572, + "learning_rate": 1.309910641754671e-06, + "loss": 1.7238, + "step": 2107 + }, + { + "epoch": 2.31013698630137, + "grad_norm": 2.749908685684204, + "learning_rate": 1.3078797725426484e-06, + "loss": 1.676, + "step": 2108 + }, + { + "epoch": 2.311232876712329, + "grad_norm": 2.679384469985962, + "learning_rate": 1.3058489033306257e-06, + "loss": 1.6899, + "step": 2109 + }, + { + "epoch": 2.3123287671232875, + "grad_norm": 2.5621979236602783, + "learning_rate": 1.303818034118603e-06, + "loss": 1.6963, + "step": 2110 + }, + { + "epoch": 2.3123287671232875, + "eval_loss": 1.9453113079071045, + "eval_runtime": 37.2803, + "eval_samples_per_second": 26.824, + "eval_steps_per_second": 1.127, + "step": 2110 + }, + { + "epoch": 2.3134246575342465, + "grad_norm": 2.6183035373687744, + "learning_rate": 1.3017871649065802e-06, + "loss": 1.7272, + "step": 2111 + }, + { + "epoch": 2.3145205479452056, + "grad_norm": 2.5777511596679688, + "learning_rate": 1.2997562956945573e-06, + "loss": 1.7041, + "step": 2112 + }, + { + "epoch": 2.315616438356164, + "grad_norm": 2.784181594848633, + "learning_rate": 1.2977254264825345e-06, + "loss": 1.6835, + "step": 2113 + }, + { + "epoch": 2.316712328767123, + "grad_norm": 2.679591417312622, + "learning_rate": 1.2956945572705118e-06, + "loss": 1.7242, + "step": 2114 + }, + { + "epoch": 2.317808219178082, + "grad_norm": 2.561269521713257, + "learning_rate": 1.293663688058489e-06, + "loss": 1.7455, + "step": 2115 + }, + { + "epoch": 2.3189041095890413, + "grad_norm": 2.5087854862213135, + "learning_rate": 1.2916328188464664e-06, + "loss": 1.6824, + "step": 2116 + }, + { + "epoch": 2.32, + "grad_norm": 2.7247722148895264, + "learning_rate": 1.2896019496344436e-06, + "loss": 1.6268, + "step": 2117 + }, + { + "epoch": 2.321095890410959, + "grad_norm": 2.603994846343994, + "learning_rate": 1.287571080422421e-06, + "loss": 1.6925, + "step": 2118 + }, + { + "epoch": 2.322191780821918, + "grad_norm": 2.5525155067443848, + "learning_rate": 1.2855402112103982e-06, + "loss": 1.7643, + "step": 2119 + }, + { + "epoch": 2.323287671232877, + "grad_norm": 2.5990493297576904, + "learning_rate": 1.2835093419983755e-06, + "loss": 1.6712, + "step": 2120 + }, + { + "epoch": 2.323287671232877, + "eval_loss": 1.9467085599899292, + "eval_runtime": 37.5353, + "eval_samples_per_second": 26.642, + "eval_steps_per_second": 1.119, + "step": 2120 + }, + { + "epoch": 2.3243835616438355, + "grad_norm": 2.764711618423462, + "learning_rate": 1.2814784727863528e-06, + "loss": 1.71, + "step": 2121 + }, + { + "epoch": 2.3254794520547946, + "grad_norm": 2.6009247303009033, + "learning_rate": 1.27944760357433e-06, + "loss": 1.6956, + "step": 2122 + }, + { + "epoch": 2.3265753424657536, + "grad_norm": 2.6900346279144287, + "learning_rate": 1.2774167343623073e-06, + "loss": 1.7529, + "step": 2123 + }, + { + "epoch": 2.327671232876712, + "grad_norm": 2.6602327823638916, + "learning_rate": 1.2753858651502846e-06, + "loss": 1.7588, + "step": 2124 + }, + { + "epoch": 2.328767123287671, + "grad_norm": 2.7727952003479004, + "learning_rate": 1.2733549959382616e-06, + "loss": 1.7499, + "step": 2125 + }, + { + "epoch": 2.3298630136986302, + "grad_norm": 2.703307867050171, + "learning_rate": 1.271324126726239e-06, + "loss": 1.7169, + "step": 2126 + }, + { + "epoch": 2.330958904109589, + "grad_norm": 2.7290568351745605, + "learning_rate": 1.2692932575142162e-06, + "loss": 1.7076, + "step": 2127 + }, + { + "epoch": 2.332054794520548, + "grad_norm": 2.664865255355835, + "learning_rate": 1.2672623883021935e-06, + "loss": 1.6655, + "step": 2128 + }, + { + "epoch": 2.333150684931507, + "grad_norm": 2.6096653938293457, + "learning_rate": 1.2652315190901707e-06, + "loss": 1.6644, + "step": 2129 + }, + { + "epoch": 2.334246575342466, + "grad_norm": 2.7201335430145264, + "learning_rate": 1.263200649878148e-06, + "loss": 1.7924, + "step": 2130 + }, + { + "epoch": 2.334246575342466, + "eval_loss": 1.9422904253005981, + "eval_runtime": 37.5421, + "eval_samples_per_second": 26.637, + "eval_steps_per_second": 1.119, + "step": 2130 + }, + { + "epoch": 2.3353424657534245, + "grad_norm": 2.519702434539795, + "learning_rate": 1.2611697806661253e-06, + "loss": 1.7108, + "step": 2131 + }, + { + "epoch": 2.3364383561643836, + "grad_norm": 2.6535212993621826, + "learning_rate": 1.2591389114541026e-06, + "loss": 1.7243, + "step": 2132 + }, + { + "epoch": 2.3375342465753426, + "grad_norm": 2.651458501815796, + "learning_rate": 1.2571080422420798e-06, + "loss": 1.7158, + "step": 2133 + }, + { + "epoch": 2.338630136986301, + "grad_norm": 2.6745946407318115, + "learning_rate": 1.2550771730300571e-06, + "loss": 1.7014, + "step": 2134 + }, + { + "epoch": 2.33972602739726, + "grad_norm": 2.6088216304779053, + "learning_rate": 1.2530463038180344e-06, + "loss": 1.7724, + "step": 2135 + }, + { + "epoch": 2.3408219178082192, + "grad_norm": 2.703014373779297, + "learning_rate": 1.2510154346060117e-06, + "loss": 1.7104, + "step": 2136 + }, + { + "epoch": 2.3419178082191783, + "grad_norm": 2.57769513130188, + "learning_rate": 1.2489845653939887e-06, + "loss": 1.7259, + "step": 2137 + }, + { + "epoch": 2.343013698630137, + "grad_norm": 2.596695899963379, + "learning_rate": 1.246953696181966e-06, + "loss": 1.7571, + "step": 2138 + }, + { + "epoch": 2.344109589041096, + "grad_norm": 2.6988871097564697, + "learning_rate": 1.2449228269699433e-06, + "loss": 1.8002, + "step": 2139 + }, + { + "epoch": 2.345205479452055, + "grad_norm": 2.6352357864379883, + "learning_rate": 1.2428919577579204e-06, + "loss": 1.7213, + "step": 2140 + }, + { + "epoch": 2.345205479452055, + "eval_loss": 1.9401096105575562, + "eval_runtime": 37.4404, + "eval_samples_per_second": 26.709, + "eval_steps_per_second": 1.122, + "step": 2140 + }, + { + "epoch": 2.3463013698630135, + "grad_norm": 2.7089645862579346, + "learning_rate": 1.2408610885458976e-06, + "loss": 1.6737, + "step": 2141 + }, + { + "epoch": 2.3473972602739726, + "grad_norm": 2.6364030838012695, + "learning_rate": 1.238830219333875e-06, + "loss": 1.7186, + "step": 2142 + }, + { + "epoch": 2.3484931506849316, + "grad_norm": 2.5563573837280273, + "learning_rate": 1.2367993501218522e-06, + "loss": 1.6956, + "step": 2143 + }, + { + "epoch": 2.34958904109589, + "grad_norm": 2.638512372970581, + "learning_rate": 1.2347684809098295e-06, + "loss": 1.7693, + "step": 2144 + }, + { + "epoch": 2.350684931506849, + "grad_norm": 2.733036518096924, + "learning_rate": 1.2327376116978067e-06, + "loss": 1.7327, + "step": 2145 + }, + { + "epoch": 2.3517808219178082, + "grad_norm": 2.6013307571411133, + "learning_rate": 1.230706742485784e-06, + "loss": 1.6539, + "step": 2146 + }, + { + "epoch": 2.3528767123287673, + "grad_norm": 2.7629592418670654, + "learning_rate": 1.2286758732737613e-06, + "loss": 1.666, + "step": 2147 + }, + { + "epoch": 2.353972602739726, + "grad_norm": 2.6855108737945557, + "learning_rate": 1.2266450040617386e-06, + "loss": 1.8044, + "step": 2148 + }, + { + "epoch": 2.355068493150685, + "grad_norm": 2.607609748840332, + "learning_rate": 1.2246141348497158e-06, + "loss": 1.8022, + "step": 2149 + }, + { + "epoch": 2.356164383561644, + "grad_norm": 2.5609383583068848, + "learning_rate": 1.2225832656376931e-06, + "loss": 1.73, + "step": 2150 + }, + { + "epoch": 2.356164383561644, + "eval_loss": 1.9387279748916626, + "eval_runtime": 37.2896, + "eval_samples_per_second": 26.817, + "eval_steps_per_second": 1.126, + "step": 2150 + }, + { + "epoch": 2.357260273972603, + "grad_norm": 2.523988723754883, + "learning_rate": 1.2205523964256704e-06, + "loss": 1.7289, + "step": 2151 + }, + { + "epoch": 2.3583561643835615, + "grad_norm": 2.675995111465454, + "learning_rate": 1.2185215272136477e-06, + "loss": 1.7036, + "step": 2152 + }, + { + "epoch": 2.3594520547945206, + "grad_norm": 2.535367012023926, + "learning_rate": 1.2164906580016247e-06, + "loss": 1.7309, + "step": 2153 + }, + { + "epoch": 2.3605479452054796, + "grad_norm": 2.6444637775421143, + "learning_rate": 1.214459788789602e-06, + "loss": 1.7192, + "step": 2154 + }, + { + "epoch": 2.361643835616438, + "grad_norm": 2.5746796131134033, + "learning_rate": 1.2124289195775793e-06, + "loss": 1.6343, + "step": 2155 + }, + { + "epoch": 2.3627397260273972, + "grad_norm": 2.612849712371826, + "learning_rate": 1.2103980503655566e-06, + "loss": 1.6686, + "step": 2156 + }, + { + "epoch": 2.3638356164383563, + "grad_norm": 2.6836912631988525, + "learning_rate": 1.2083671811535338e-06, + "loss": 1.7173, + "step": 2157 + }, + { + "epoch": 2.364931506849315, + "grad_norm": 2.716177225112915, + "learning_rate": 1.2063363119415111e-06, + "loss": 1.695, + "step": 2158 + }, + { + "epoch": 2.366027397260274, + "grad_norm": 2.6918227672576904, + "learning_rate": 1.2043054427294882e-06, + "loss": 1.684, + "step": 2159 + }, + { + "epoch": 2.367123287671233, + "grad_norm": 2.6265008449554443, + "learning_rate": 1.2022745735174654e-06, + "loss": 1.6959, + "step": 2160 + }, + { + "epoch": 2.367123287671233, + "eval_loss": 1.939719319343567, + "eval_runtime": 37.2557, + "eval_samples_per_second": 26.842, + "eval_steps_per_second": 1.127, + "step": 2160 + }, + { + "epoch": 2.368219178082192, + "grad_norm": 2.638829469680786, + "learning_rate": 1.2002437043054427e-06, + "loss": 1.7362, + "step": 2161 + }, + { + "epoch": 2.3693150684931505, + "grad_norm": 2.6369569301605225, + "learning_rate": 1.19821283509342e-06, + "loss": 1.7765, + "step": 2162 + }, + { + "epoch": 2.3704109589041096, + "grad_norm": 2.6272799968719482, + "learning_rate": 1.1961819658813973e-06, + "loss": 1.7695, + "step": 2163 + }, + { + "epoch": 2.3715068493150686, + "grad_norm": 2.601388931274414, + "learning_rate": 1.1941510966693746e-06, + "loss": 1.7259, + "step": 2164 + }, + { + "epoch": 2.3726027397260276, + "grad_norm": 2.5515146255493164, + "learning_rate": 1.1921202274573518e-06, + "loss": 1.7048, + "step": 2165 + }, + { + "epoch": 2.3736986301369862, + "grad_norm": 2.6139824390411377, + "learning_rate": 1.190089358245329e-06, + "loss": 1.6951, + "step": 2166 + }, + { + "epoch": 2.3747945205479453, + "grad_norm": 2.7428653240203857, + "learning_rate": 1.1880584890333064e-06, + "loss": 1.7771, + "step": 2167 + }, + { + "epoch": 2.3758904109589043, + "grad_norm": 2.690804958343506, + "learning_rate": 1.1860276198212837e-06, + "loss": 1.7073, + "step": 2168 + }, + { + "epoch": 2.376986301369863, + "grad_norm": 2.64790415763855, + "learning_rate": 1.183996750609261e-06, + "loss": 1.7344, + "step": 2169 + }, + { + "epoch": 2.378082191780822, + "grad_norm": 2.555832624435425, + "learning_rate": 1.1819658813972382e-06, + "loss": 1.7452, + "step": 2170 + }, + { + "epoch": 2.378082191780822, + "eval_loss": 1.9401137828826904, + "eval_runtime": 37.2362, + "eval_samples_per_second": 26.856, + "eval_steps_per_second": 1.128, + "step": 2170 + }, + { + "epoch": 2.379178082191781, + "grad_norm": 2.604097366333008, + "learning_rate": 1.1799350121852153e-06, + "loss": 1.7176, + "step": 2171 + }, + { + "epoch": 2.3802739726027395, + "grad_norm": 2.579312324523926, + "learning_rate": 1.1779041429731925e-06, + "loss": 1.7588, + "step": 2172 + }, + { + "epoch": 2.3813698630136986, + "grad_norm": 2.555727481842041, + "learning_rate": 1.1758732737611698e-06, + "loss": 1.7612, + "step": 2173 + }, + { + "epoch": 2.3824657534246576, + "grad_norm": 2.6929352283477783, + "learning_rate": 1.173842404549147e-06, + "loss": 1.6707, + "step": 2174 + }, + { + "epoch": 2.383561643835616, + "grad_norm": 2.6732795238494873, + "learning_rate": 1.1718115353371244e-06, + "loss": 1.7743, + "step": 2175 + }, + { + "epoch": 2.3846575342465752, + "grad_norm": 2.673490047454834, + "learning_rate": 1.1697806661251016e-06, + "loss": 1.6968, + "step": 2176 + }, + { + "epoch": 2.3857534246575343, + "grad_norm": 2.657837152481079, + "learning_rate": 1.167749796913079e-06, + "loss": 1.7633, + "step": 2177 + }, + { + "epoch": 2.3868493150684933, + "grad_norm": 2.6677606105804443, + "learning_rate": 1.1657189277010562e-06, + "loss": 1.7646, + "step": 2178 + }, + { + "epoch": 2.387945205479452, + "grad_norm": 2.63075852394104, + "learning_rate": 1.1636880584890335e-06, + "loss": 1.7019, + "step": 2179 + }, + { + "epoch": 2.389041095890411, + "grad_norm": 2.874946117401123, + "learning_rate": 1.1616571892770108e-06, + "loss": 1.691, + "step": 2180 + }, + { + "epoch": 2.389041095890411, + "eval_loss": 1.9399209022521973, + "eval_runtime": 37.1991, + "eval_samples_per_second": 26.882, + "eval_steps_per_second": 1.129, + "step": 2180 + }, + { + "epoch": 2.39013698630137, + "grad_norm": 2.734825611114502, + "learning_rate": 1.159626320064988e-06, + "loss": 1.7551, + "step": 2181 + }, + { + "epoch": 2.391232876712329, + "grad_norm": 2.5722298622131348, + "learning_rate": 1.1575954508529653e-06, + "loss": 1.7279, + "step": 2182 + }, + { + "epoch": 2.3923287671232876, + "grad_norm": 2.6449334621429443, + "learning_rate": 1.1555645816409426e-06, + "loss": 1.7591, + "step": 2183 + }, + { + "epoch": 2.3934246575342466, + "grad_norm": 2.774374485015869, + "learning_rate": 1.1535337124289196e-06, + "loss": 1.7243, + "step": 2184 + }, + { + "epoch": 2.3945205479452056, + "grad_norm": 2.554065227508545, + "learning_rate": 1.151502843216897e-06, + "loss": 1.6767, + "step": 2185 + }, + { + "epoch": 2.3956164383561642, + "grad_norm": 2.5883984565734863, + "learning_rate": 1.1494719740048742e-06, + "loss": 1.6416, + "step": 2186 + }, + { + "epoch": 2.3967123287671233, + "grad_norm": 2.5799498558044434, + "learning_rate": 1.1474411047928515e-06, + "loss": 1.7162, + "step": 2187 + }, + { + "epoch": 2.3978082191780823, + "grad_norm": 2.64442777633667, + "learning_rate": 1.1454102355808287e-06, + "loss": 1.709, + "step": 2188 + }, + { + "epoch": 2.398904109589041, + "grad_norm": 2.6813411712646484, + "learning_rate": 1.143379366368806e-06, + "loss": 1.6903, + "step": 2189 + }, + { + "epoch": 2.4, + "grad_norm": 2.6276469230651855, + "learning_rate": 1.141348497156783e-06, + "loss": 1.6734, + "step": 2190 + }, + { + "epoch": 2.4, + "eval_loss": 1.9423893690109253, + "eval_runtime": 37.1963, + "eval_samples_per_second": 26.884, + "eval_steps_per_second": 1.129, + "step": 2190 + }, + { + "epoch": 2.401095890410959, + "grad_norm": 2.6805641651153564, + "learning_rate": 1.1393176279447604e-06, + "loss": 1.8038, + "step": 2191 + }, + { + "epoch": 2.402191780821918, + "grad_norm": 2.813957691192627, + "learning_rate": 1.1372867587327376e-06, + "loss": 1.7895, + "step": 2192 + }, + { + "epoch": 2.4032876712328766, + "grad_norm": 2.505695343017578, + "learning_rate": 1.135255889520715e-06, + "loss": 1.6398, + "step": 2193 + }, + { + "epoch": 2.4043835616438356, + "grad_norm": 2.7426211833953857, + "learning_rate": 1.1332250203086922e-06, + "loss": 1.7472, + "step": 2194 + }, + { + "epoch": 2.4054794520547946, + "grad_norm": 2.627737522125244, + "learning_rate": 1.1311941510966695e-06, + "loss": 1.7272, + "step": 2195 + }, + { + "epoch": 2.4065753424657537, + "grad_norm": 2.615032911300659, + "learning_rate": 1.1291632818846467e-06, + "loss": 1.7004, + "step": 2196 + }, + { + "epoch": 2.4076712328767123, + "grad_norm": 2.5147581100463867, + "learning_rate": 1.127132412672624e-06, + "loss": 1.6894, + "step": 2197 + }, + { + "epoch": 2.4087671232876713, + "grad_norm": 2.7482388019561768, + "learning_rate": 1.1251015434606013e-06, + "loss": 1.6738, + "step": 2198 + }, + { + "epoch": 2.4098630136986303, + "grad_norm": 2.7309365272521973, + "learning_rate": 1.1230706742485786e-06, + "loss": 1.7222, + "step": 2199 + }, + { + "epoch": 2.410958904109589, + "grad_norm": 2.8322935104370117, + "learning_rate": 1.1210398050365558e-06, + "loss": 1.7762, + "step": 2200 + }, + { + "epoch": 2.410958904109589, + "eval_loss": 1.9395875930786133, + "eval_runtime": 37.2409, + "eval_samples_per_second": 26.852, + "eval_steps_per_second": 1.128, + "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/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-2200/zero_to_fp32.py b/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-2200/zero_to_fp32.py new file mode 100644 index 0000000000000000000000000000000000000000..24cc342e78d1a006c782b3a4cd68d9ce786d8fd8 --- /dev/null +++ b/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_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/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-2300/config.json b/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-2300/config.json new file mode 100644 index 0000000000000000000000000000000000000000..fb0f9a53a507ffcdb60410deb71d60ef801bf350 --- /dev/null +++ b/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_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/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-2300/generation_config.json b/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-2300/generation_config.json new file mode 100644 index 0000000000000000000000000000000000000000..9c389b841a9a29ddd904e02d1eb0e08dcce82ad9 --- /dev/null +++ b/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_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/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-2300/latest b/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-2300/latest new file mode 100644 index 0000000000000000000000000000000000000000..78dc01ce4ce1498b753887dfb7af92409705a69b --- /dev/null +++ b/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-2300/latest @@ -0,0 +1 @@ +global_step2300 \ No newline at end of file diff --git a/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-2300/model.safetensors.index.json b/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-2300/model.safetensors.index.json new file mode 100644 index 0000000000000000000000000000000000000000..ed64de846d720b9a7859dc20575fea8e8ca51940 --- /dev/null +++ b/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_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/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-2300/special_tokens_map.json b/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-2300/special_tokens_map.json new file mode 100644 index 0000000000000000000000000000000000000000..6949c5975ee0e961ef61cf31010dce04df0a03f8 --- /dev/null +++ b/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_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/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-2300/tokenizer_config.json b/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-2300/tokenizer_config.json new file mode 100644 index 0000000000000000000000000000000000000000..5684b14e2756b56aa397c1324e239b442162b34a --- /dev/null +++ b/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_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/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-2300/trainer_state.json b/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-2300/trainer_state.json new file mode 100644 index 0000000000000000000000000000000000000000..632ba33fe6ea0f916eaa46c088c79944d974d1cd --- /dev/null +++ b/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_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": 3.0476, + "step": 1 + }, + { + "epoch": 0.002191780821917808, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.0289, + "step": 2 + }, + { + "epoch": 0.0032876712328767125, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 2.9935, + "step": 3 + }, + { + "epoch": 0.004383561643835616, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.0139, + "step": 4 + }, + { + "epoch": 0.005479452054794521, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 2.9892, + "step": 5 + }, + { + "epoch": 0.006575342465753425, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.1252, + "step": 6 + }, + { + "epoch": 0.007671232876712329, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.0291, + "step": 7 + }, + { + "epoch": 0.008767123287671232, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.0607, + "step": 8 + }, + { + "epoch": 0.009863013698630137, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.0306, + "step": 9 + }, + { + "epoch": 0.010958904109589041, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.0811, + "step": 10 + }, + { + "epoch": 0.010958904109589041, + "eval_loss": 3.0768930912017822, + "eval_runtime": 38.0377, + "eval_samples_per_second": 26.29, + "eval_steps_per_second": 1.104, + "step": 10 + }, + { + "epoch": 0.012054794520547946, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.0693, + "step": 11 + }, + { + "epoch": 0.01315068493150685, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 2.9668, + "step": 12 + }, + { + "epoch": 0.014246575342465753, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.052, + "step": 13 + }, + { + "epoch": 0.015342465753424657, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.0363, + "step": 14 + }, + { + "epoch": 0.01643835616438356, + "grad_norm": 5.58175802230835, + "learning_rate": 1.8248175182481753e-08, + "loss": 3.0778, + "step": 15 + }, + { + "epoch": 0.017534246575342465, + "grad_norm": 6.360531806945801, + "learning_rate": 3.6496350364963505e-08, + "loss": 3.0257, + "step": 16 + }, + { + "epoch": 0.01863013698630137, + "grad_norm": 6.453113555908203, + "learning_rate": 5.4744525547445264e-08, + "loss": 3.0262, + "step": 17 + }, + { + "epoch": 0.019726027397260273, + "grad_norm": 6.789405345916748, + "learning_rate": 7.299270072992701e-08, + "loss": 3.0671, + "step": 18 + }, + { + "epoch": 0.020821917808219178, + "grad_norm": 6.185514450073242, + "learning_rate": 9.124087591240876e-08, + "loss": 3.0077, + "step": 19 + }, + { + "epoch": 0.021917808219178082, + "grad_norm": 5.719798564910889, + "learning_rate": 1.0948905109489053e-07, + "loss": 3.089, + "step": 20 + }, + { + "epoch": 0.021917808219178082, + "eval_loss": 3.07546329498291, + "eval_runtime": 37.5318, + "eval_samples_per_second": 26.644, + "eval_steps_per_second": 1.119, + "step": 20 + }, + { + "epoch": 0.023013698630136987, + "grad_norm": 6.307768821716309, + "learning_rate": 1.2773722627737227e-07, + "loss": 3.0641, + "step": 21 + }, + { + "epoch": 0.02410958904109589, + "grad_norm": 6.255112171173096, + "learning_rate": 1.4598540145985402e-07, + "loss": 2.9849, + "step": 22 + }, + { + "epoch": 0.025205479452054796, + "grad_norm": 5.812101364135742, + "learning_rate": 1.642335766423358e-07, + "loss": 3.0211, + "step": 23 + }, + { + "epoch": 0.0263013698630137, + "grad_norm": 6.262075901031494, + "learning_rate": 1.8248175182481753e-07, + "loss": 3.0654, + "step": 24 + }, + { + "epoch": 0.0273972602739726, + "grad_norm": 5.419229507446289, + "learning_rate": 2.0072992700729928e-07, + "loss": 3.0661, + "step": 25 + }, + { + "epoch": 0.028493150684931506, + "grad_norm": 4.750737190246582, + "learning_rate": 2.1897810218978106e-07, + "loss": 3.0494, + "step": 26 + }, + { + "epoch": 0.02958904109589041, + "grad_norm": 4.732361316680908, + "learning_rate": 2.3722627737226278e-07, + "loss": 3.0471, + "step": 27 + }, + { + "epoch": 0.030684931506849315, + "grad_norm": 4.749786853790283, + "learning_rate": 2.5547445255474454e-07, + "loss": 3.0107, + "step": 28 + }, + { + "epoch": 0.03178082191780822, + "grad_norm": 4.451244831085205, + "learning_rate": 2.737226277372263e-07, + "loss": 2.9987, + "step": 29 + }, + { + "epoch": 0.03287671232876712, + "grad_norm": 4.5158257484436035, + "learning_rate": 2.9197080291970804e-07, + "loss": 2.9686, + "step": 30 + }, + { + "epoch": 0.03287671232876712, + "eval_loss": 3.031872034072876, + "eval_runtime": 37.2608, + "eval_samples_per_second": 26.838, + "eval_steps_per_second": 1.127, + "step": 30 + }, + { + "epoch": 0.03397260273972603, + "grad_norm": 3.8239498138427734, + "learning_rate": 3.102189781021898e-07, + "loss": 3.0057, + "step": 31 + }, + { + "epoch": 0.03506849315068493, + "grad_norm": 4.037254810333252, + "learning_rate": 3.284671532846716e-07, + "loss": 2.9733, + "step": 32 + }, + { + "epoch": 0.03616438356164384, + "grad_norm": 4.283931732177734, + "learning_rate": 3.467153284671533e-07, + "loss": 2.9515, + "step": 33 + }, + { + "epoch": 0.03726027397260274, + "grad_norm": 4.1784987449646, + "learning_rate": 3.6496350364963505e-07, + "loss": 2.9483, + "step": 34 + }, + { + "epoch": 0.038356164383561646, + "grad_norm": 3.9210753440856934, + "learning_rate": 3.8321167883211686e-07, + "loss": 2.9679, + "step": 35 + }, + { + "epoch": 0.03945205479452055, + "grad_norm": 3.72884202003479, + "learning_rate": 4.0145985401459856e-07, + "loss": 2.9989, + "step": 36 + }, + { + "epoch": 0.040547945205479455, + "grad_norm": 3.2882330417633057, + "learning_rate": 4.197080291970803e-07, + "loss": 2.9384, + "step": 37 + }, + { + "epoch": 0.041643835616438356, + "grad_norm": 3.515293836593628, + "learning_rate": 4.379562043795621e-07, + "loss": 2.8986, + "step": 38 + }, + { + "epoch": 0.042739726027397264, + "grad_norm": 3.568453073501587, + "learning_rate": 4.562043795620438e-07, + "loss": 2.9854, + "step": 39 + }, + { + "epoch": 0.043835616438356165, + "grad_norm": 3.389789342880249, + "learning_rate": 4.7445255474452557e-07, + "loss": 2.8891, + "step": 40 + }, + { + "epoch": 0.043835616438356165, + "eval_loss": 2.964235544204712, + "eval_runtime": 37.3406, + "eval_samples_per_second": 26.781, + "eval_steps_per_second": 1.125, + "step": 40 + }, + { + "epoch": 0.044931506849315066, + "grad_norm": 3.1826283931732178, + "learning_rate": 4.927007299270073e-07, + "loss": 2.9271, + "step": 41 + }, + { + "epoch": 0.046027397260273974, + "grad_norm": 2.941539764404297, + "learning_rate": 5.109489051094891e-07, + "loss": 2.9486, + "step": 42 + }, + { + "epoch": 0.047123287671232875, + "grad_norm": 3.0076773166656494, + "learning_rate": 5.291970802919708e-07, + "loss": 2.9252, + "step": 43 + }, + { + "epoch": 0.04821917808219178, + "grad_norm": 2.79028582572937, + "learning_rate": 5.474452554744526e-07, + "loss": 2.9277, + "step": 44 + }, + { + "epoch": 0.049315068493150684, + "grad_norm": 2.603456735610962, + "learning_rate": 5.656934306569343e-07, + "loss": 2.9144, + "step": 45 + }, + { + "epoch": 0.05041095890410959, + "grad_norm": 2.674888849258423, + "learning_rate": 5.839416058394161e-07, + "loss": 2.8732, + "step": 46 + }, + { + "epoch": 0.05150684931506849, + "grad_norm": 2.972093343734741, + "learning_rate": 6.021897810218979e-07, + "loss": 2.8623, + "step": 47 + }, + { + "epoch": 0.0526027397260274, + "grad_norm": 2.757766008377075, + "learning_rate": 6.204379562043796e-07, + "loss": 2.8941, + "step": 48 + }, + { + "epoch": 0.0536986301369863, + "grad_norm": 2.910634994506836, + "learning_rate": 6.386861313868614e-07, + "loss": 2.8101, + "step": 49 + }, + { + "epoch": 0.0547945205479452, + "grad_norm": 2.5542638301849365, + "learning_rate": 6.569343065693432e-07, + "loss": 2.9262, + "step": 50 + }, + { + "epoch": 0.0547945205479452, + "eval_loss": 2.883579969406128, + "eval_runtime": 37.326, + "eval_samples_per_second": 26.791, + "eval_steps_per_second": 1.125, + "step": 50 + }, + { + "epoch": 0.05589041095890411, + "grad_norm": 2.333974599838257, + "learning_rate": 6.751824817518248e-07, + "loss": 2.8217, + "step": 51 + }, + { + "epoch": 0.05698630136986301, + "grad_norm": 2.181022882461548, + "learning_rate": 6.934306569343066e-07, + "loss": 2.8483, + "step": 52 + }, + { + "epoch": 0.05808219178082192, + "grad_norm": 2.12200927734375, + "learning_rate": 7.116788321167883e-07, + "loss": 2.7788, + "step": 53 + }, + { + "epoch": 0.05917808219178082, + "grad_norm": 2.052626371383667, + "learning_rate": 7.299270072992701e-07, + "loss": 2.7774, + "step": 54 + }, + { + "epoch": 0.06027397260273973, + "grad_norm": 2.11487078666687, + "learning_rate": 7.48175182481752e-07, + "loss": 2.8045, + "step": 55 + }, + { + "epoch": 0.06136986301369863, + "grad_norm": 2.083588123321533, + "learning_rate": 7.664233576642337e-07, + "loss": 2.8119, + "step": 56 + }, + { + "epoch": 0.06246575342465754, + "grad_norm": 2.2079808712005615, + "learning_rate": 7.846715328467154e-07, + "loss": 2.82, + "step": 57 + }, + { + "epoch": 0.06356164383561644, + "grad_norm": 2.100802421569824, + "learning_rate": 8.029197080291971e-07, + "loss": 2.794, + "step": 58 + }, + { + "epoch": 0.06465753424657535, + "grad_norm": 1.8322066068649292, + "learning_rate": 8.211678832116789e-07, + "loss": 2.7444, + "step": 59 + }, + { + "epoch": 0.06575342465753424, + "grad_norm": 2.077005386352539, + "learning_rate": 8.394160583941606e-07, + "loss": 2.7093, + "step": 60 + }, + { + "epoch": 0.06575342465753424, + "eval_loss": 2.8177874088287354, + "eval_runtime": 37.2903, + "eval_samples_per_second": 26.817, + "eval_steps_per_second": 1.126, + "step": 60 + }, + { + "epoch": 0.06684931506849315, + "grad_norm": 2.1688592433929443, + "learning_rate": 8.576642335766425e-07, + "loss": 2.7446, + "step": 61 + }, + { + "epoch": 0.06794520547945206, + "grad_norm": 2.1709237098693848, + "learning_rate": 8.759124087591242e-07, + "loss": 2.736, + "step": 62 + }, + { + "epoch": 0.06904109589041096, + "grad_norm": 2.3578364849090576, + "learning_rate": 8.94160583941606e-07, + "loss": 2.7583, + "step": 63 + }, + { + "epoch": 0.07013698630136986, + "grad_norm": 1.9025483131408691, + "learning_rate": 9.124087591240876e-07, + "loss": 2.7651, + "step": 64 + }, + { + "epoch": 0.07123287671232877, + "grad_norm": 1.9450963735580444, + "learning_rate": 9.306569343065694e-07, + "loss": 2.7186, + "step": 65 + }, + { + "epoch": 0.07232876712328767, + "grad_norm": 1.957723617553711, + "learning_rate": 9.489051094890511e-07, + "loss": 2.7421, + "step": 66 + }, + { + "epoch": 0.07342465753424658, + "grad_norm": 1.9823631048202515, + "learning_rate": 9.67153284671533e-07, + "loss": 2.7258, + "step": 67 + }, + { + "epoch": 0.07452054794520548, + "grad_norm": 1.945495843887329, + "learning_rate": 9.854014598540146e-07, + "loss": 2.7733, + "step": 68 + }, + { + "epoch": 0.07561643835616438, + "grad_norm": 1.9684759378433228, + "learning_rate": 1.0036496350364965e-06, + "loss": 2.7041, + "step": 69 + }, + { + "epoch": 0.07671232876712329, + "grad_norm": 2.087336540222168, + "learning_rate": 1.0218978102189781e-06, + "loss": 2.6662, + "step": 70 + }, + { + "epoch": 0.07671232876712329, + "eval_loss": 2.7529091835021973, + "eval_runtime": 37.3122, + "eval_samples_per_second": 26.801, + "eval_steps_per_second": 1.126, + "step": 70 + }, + { + "epoch": 0.07780821917808219, + "grad_norm": 1.7790117263793945, + "learning_rate": 1.04014598540146e-06, + "loss": 2.7379, + "step": 71 + }, + { + "epoch": 0.0789041095890411, + "grad_norm": 2.2780861854553223, + "learning_rate": 1.0583941605839416e-06, + "loss": 2.6234, + "step": 72 + }, + { + "epoch": 0.08, + "grad_norm": 2.1144700050354004, + "learning_rate": 1.0766423357664233e-06, + "loss": 2.6248, + "step": 73 + }, + { + "epoch": 0.08109589041095891, + "grad_norm": 2.0247490406036377, + "learning_rate": 1.0948905109489052e-06, + "loss": 2.6456, + "step": 74 + }, + { + "epoch": 0.0821917808219178, + "grad_norm": 2.702547311782837, + "learning_rate": 1.113138686131387e-06, + "loss": 2.6759, + "step": 75 + }, + { + "epoch": 0.08328767123287671, + "grad_norm": 1.8208750486373901, + "learning_rate": 1.1313868613138687e-06, + "loss": 2.6845, + "step": 76 + }, + { + "epoch": 0.08438356164383562, + "grad_norm": 2.3829221725463867, + "learning_rate": 1.1496350364963505e-06, + "loss": 2.61, + "step": 77 + }, + { + "epoch": 0.08547945205479453, + "grad_norm": 2.580885887145996, + "learning_rate": 1.1678832116788322e-06, + "loss": 2.5896, + "step": 78 + }, + { + "epoch": 0.08657534246575342, + "grad_norm": 2.2477641105651855, + "learning_rate": 1.1861313868613138e-06, + "loss": 2.596, + "step": 79 + }, + { + "epoch": 0.08767123287671233, + "grad_norm": 2.1895623207092285, + "learning_rate": 1.2043795620437959e-06, + "loss": 2.5906, + "step": 80 + }, + { + "epoch": 0.08767123287671233, + "eval_loss": 2.6815176010131836, + "eval_runtime": 37.3342, + "eval_samples_per_second": 26.785, + "eval_steps_per_second": 1.125, + "step": 80 + }, + { + "epoch": 0.08876712328767124, + "grad_norm": 2.217090606689453, + "learning_rate": 1.2226277372262775e-06, + "loss": 2.6005, + "step": 81 + }, + { + "epoch": 0.08986301369863013, + "grad_norm": 2.4679362773895264, + "learning_rate": 1.2408759124087592e-06, + "loss": 2.6333, + "step": 82 + }, + { + "epoch": 0.09095890410958904, + "grad_norm": 2.2590575218200684, + "learning_rate": 1.259124087591241e-06, + "loss": 2.5814, + "step": 83 + }, + { + "epoch": 0.09205479452054795, + "grad_norm": 1.877977728843689, + "learning_rate": 1.2773722627737229e-06, + "loss": 2.5948, + "step": 84 + }, + { + "epoch": 0.09315068493150686, + "grad_norm": 2.1535534858703613, + "learning_rate": 1.2956204379562045e-06, + "loss": 2.5781, + "step": 85 + }, + { + "epoch": 0.09424657534246575, + "grad_norm": 2.5001144409179688, + "learning_rate": 1.3138686131386864e-06, + "loss": 2.5339, + "step": 86 + }, + { + "epoch": 0.09534246575342466, + "grad_norm": 2.3279049396514893, + "learning_rate": 1.332116788321168e-06, + "loss": 2.545, + "step": 87 + }, + { + "epoch": 0.09643835616438357, + "grad_norm": 2.2230734825134277, + "learning_rate": 1.3503649635036497e-06, + "loss": 2.4745, + "step": 88 + }, + { + "epoch": 0.09753424657534246, + "grad_norm": 2.8762636184692383, + "learning_rate": 1.3686131386861315e-06, + "loss": 2.516, + "step": 89 + }, + { + "epoch": 0.09863013698630137, + "grad_norm": 2.153506278991699, + "learning_rate": 1.3868613138686132e-06, + "loss": 2.4547, + "step": 90 + }, + { + "epoch": 0.09863013698630137, + "eval_loss": 2.603238344192505, + "eval_runtime": 37.4162, + "eval_samples_per_second": 26.726, + "eval_steps_per_second": 1.123, + "step": 90 + }, + { + "epoch": 0.09972602739726028, + "grad_norm": 2.4143614768981934, + "learning_rate": 1.405109489051095e-06, + "loss": 2.547, + "step": 91 + }, + { + "epoch": 0.10082191780821918, + "grad_norm": 2.354632616043091, + "learning_rate": 1.4233576642335767e-06, + "loss": 2.5009, + "step": 92 + }, + { + "epoch": 0.10191780821917808, + "grad_norm": 2.6605563163757324, + "learning_rate": 1.4416058394160583e-06, + "loss": 2.5304, + "step": 93 + }, + { + "epoch": 0.10301369863013699, + "grad_norm": 2.3724122047424316, + "learning_rate": 1.4598540145985402e-06, + "loss": 2.4925, + "step": 94 + }, + { + "epoch": 0.10410958904109589, + "grad_norm": 3.6323022842407227, + "learning_rate": 1.478102189781022e-06, + "loss": 2.4592, + "step": 95 + }, + { + "epoch": 0.1052054794520548, + "grad_norm": 2.3407089710235596, + "learning_rate": 1.496350364963504e-06, + "loss": 2.5223, + "step": 96 + }, + { + "epoch": 0.1063013698630137, + "grad_norm": 2.8238413333892822, + "learning_rate": 1.5145985401459856e-06, + "loss": 2.489, + "step": 97 + }, + { + "epoch": 0.1073972602739726, + "grad_norm": 2.6433658599853516, + "learning_rate": 1.5328467153284674e-06, + "loss": 2.5101, + "step": 98 + }, + { + "epoch": 0.10849315068493151, + "grad_norm": 2.562761068344116, + "learning_rate": 1.551094890510949e-06, + "loss": 2.4642, + "step": 99 + }, + { + "epoch": 0.1095890410958904, + "grad_norm": 2.4501149654388428, + "learning_rate": 1.5693430656934307e-06, + "loss": 2.4102, + "step": 100 + }, + { + "epoch": 0.1095890410958904, + "eval_loss": 2.5452804565429688, + "eval_runtime": 37.4224, + "eval_samples_per_second": 26.722, + "eval_steps_per_second": 1.122, + "step": 100 + }, + { + "epoch": 0.11068493150684931, + "grad_norm": 2.475864887237549, + "learning_rate": 1.5875912408759126e-06, + "loss": 2.4147, + "step": 101 + }, + { + "epoch": 0.11178082191780822, + "grad_norm": 2.558875322341919, + "learning_rate": 1.6058394160583942e-06, + "loss": 2.4694, + "step": 102 + }, + { + "epoch": 0.11287671232876713, + "grad_norm": 2.1085734367370605, + "learning_rate": 1.624087591240876e-06, + "loss": 2.5099, + "step": 103 + }, + { + "epoch": 0.11397260273972602, + "grad_norm": 2.410731792449951, + "learning_rate": 1.6423357664233577e-06, + "loss": 2.432, + "step": 104 + }, + { + "epoch": 0.11506849315068493, + "grad_norm": 2.2800745964050293, + "learning_rate": 1.6605839416058394e-06, + "loss": 2.3669, + "step": 105 + }, + { + "epoch": 0.11616438356164384, + "grad_norm": 2.2234933376312256, + "learning_rate": 1.6788321167883212e-06, + "loss": 2.4145, + "step": 106 + }, + { + "epoch": 0.11726027397260275, + "grad_norm": 2.3367040157318115, + "learning_rate": 1.6970802919708029e-06, + "loss": 2.4272, + "step": 107 + }, + { + "epoch": 0.11835616438356164, + "grad_norm": 2.7002811431884766, + "learning_rate": 1.715328467153285e-06, + "loss": 2.3478, + "step": 108 + }, + { + "epoch": 0.11945205479452055, + "grad_norm": 2.6628332138061523, + "learning_rate": 1.7335766423357666e-06, + "loss": 2.4096, + "step": 109 + }, + { + "epoch": 0.12054794520547946, + "grad_norm": 2.4174230098724365, + "learning_rate": 1.7518248175182485e-06, + "loss": 2.407, + "step": 110 + }, + { + "epoch": 0.12054794520547946, + "eval_loss": 2.491746664047241, + "eval_runtime": 37.2555, + "eval_samples_per_second": 26.842, + "eval_steps_per_second": 1.127, + "step": 110 + }, + { + "epoch": 0.12164383561643835, + "grad_norm": 2.24796986579895, + "learning_rate": 1.77007299270073e-06, + "loss": 2.4369, + "step": 111 + }, + { + "epoch": 0.12273972602739726, + "grad_norm": 2.720939874649048, + "learning_rate": 1.788321167883212e-06, + "loss": 2.4116, + "step": 112 + }, + { + "epoch": 0.12383561643835617, + "grad_norm": 2.6907670497894287, + "learning_rate": 1.8065693430656936e-06, + "loss": 2.4577, + "step": 113 + }, + { + "epoch": 0.12493150684931507, + "grad_norm": 2.559636354446411, + "learning_rate": 1.8248175182481753e-06, + "loss": 2.3999, + "step": 114 + }, + { + "epoch": 0.12602739726027398, + "grad_norm": 2.6730551719665527, + "learning_rate": 1.8430656934306571e-06, + "loss": 2.3428, + "step": 115 + }, + { + "epoch": 0.12712328767123288, + "grad_norm": 3.172179937362671, + "learning_rate": 1.8613138686131388e-06, + "loss": 2.3722, + "step": 116 + }, + { + "epoch": 0.12821917808219177, + "grad_norm": 2.6757171154022217, + "learning_rate": 1.8795620437956206e-06, + "loss": 2.3572, + "step": 117 + }, + { + "epoch": 0.1293150684931507, + "grad_norm": 2.5840766429901123, + "learning_rate": 1.8978102189781023e-06, + "loss": 2.353, + "step": 118 + }, + { + "epoch": 0.1304109589041096, + "grad_norm": 2.6248581409454346, + "learning_rate": 1.916058394160584e-06, + "loss": 2.36, + "step": 119 + }, + { + "epoch": 0.13150684931506848, + "grad_norm": 2.428516387939453, + "learning_rate": 1.934306569343066e-06, + "loss": 2.3821, + "step": 120 + }, + { + "epoch": 0.13150684931506848, + "eval_loss": 2.451565980911255, + "eval_runtime": 37.3669, + "eval_samples_per_second": 26.762, + "eval_steps_per_second": 1.124, + "step": 120 + }, + { + "epoch": 0.1326027397260274, + "grad_norm": 2.9956233501434326, + "learning_rate": 1.952554744525548e-06, + "loss": 2.2778, + "step": 121 + }, + { + "epoch": 0.1336986301369863, + "grad_norm": 2.4120893478393555, + "learning_rate": 1.9708029197080293e-06, + "loss": 2.325, + "step": 122 + }, + { + "epoch": 0.13479452054794522, + "grad_norm": 2.1410024166107178, + "learning_rate": 1.989051094890511e-06, + "loss": 2.3248, + "step": 123 + }, + { + "epoch": 0.1358904109589041, + "grad_norm": 2.923799514770508, + "learning_rate": 2.007299270072993e-06, + "loss": 2.327, + "step": 124 + }, + { + "epoch": 0.136986301369863, + "grad_norm": 3.0770339965820312, + "learning_rate": 2.025547445255475e-06, + "loss": 2.3847, + "step": 125 + }, + { + "epoch": 0.13808219178082193, + "grad_norm": 2.6408181190490723, + "learning_rate": 2.0437956204379563e-06, + "loss": 2.3196, + "step": 126 + }, + { + "epoch": 0.13917808219178082, + "grad_norm": 2.9127118587493896, + "learning_rate": 2.062043795620438e-06, + "loss": 2.3287, + "step": 127 + }, + { + "epoch": 0.14027397260273972, + "grad_norm": 2.9805052280426025, + "learning_rate": 2.08029197080292e-06, + "loss": 2.3243, + "step": 128 + }, + { + "epoch": 0.14136986301369864, + "grad_norm": 3.1342010498046875, + "learning_rate": 2.0985401459854014e-06, + "loss": 2.38, + "step": 129 + }, + { + "epoch": 0.14246575342465753, + "grad_norm": 2.678422451019287, + "learning_rate": 2.1167883211678833e-06, + "loss": 2.2727, + "step": 130 + }, + { + "epoch": 0.14246575342465753, + "eval_loss": 2.4187021255493164, + "eval_runtime": 37.3117, + "eval_samples_per_second": 26.801, + "eval_steps_per_second": 1.126, + "step": 130 + }, + { + "epoch": 0.14356164383561643, + "grad_norm": 2.9281492233276367, + "learning_rate": 2.135036496350365e-06, + "loss": 2.3327, + "step": 131 + }, + { + "epoch": 0.14465753424657535, + "grad_norm": 2.9592933654785156, + "learning_rate": 2.1532846715328466e-06, + "loss": 2.3674, + "step": 132 + }, + { + "epoch": 0.14575342465753424, + "grad_norm": 3.404761552810669, + "learning_rate": 2.171532846715329e-06, + "loss": 2.2679, + "step": 133 + }, + { + "epoch": 0.14684931506849316, + "grad_norm": 3.635571241378784, + "learning_rate": 2.1897810218978103e-06, + "loss": 2.2799, + "step": 134 + }, + { + "epoch": 0.14794520547945206, + "grad_norm": 3.9130170345306396, + "learning_rate": 2.208029197080292e-06, + "loss": 2.3263, + "step": 135 + }, + { + "epoch": 0.14904109589041095, + "grad_norm": 4.300438404083252, + "learning_rate": 2.226277372262774e-06, + "loss": 2.2969, + "step": 136 + }, + { + "epoch": 0.15013698630136987, + "grad_norm": 2.863511085510254, + "learning_rate": 2.244525547445256e-06, + "loss": 2.297, + "step": 137 + }, + { + "epoch": 0.15123287671232877, + "grad_norm": 4.910321235656738, + "learning_rate": 2.2627737226277373e-06, + "loss": 2.3269, + "step": 138 + }, + { + "epoch": 0.15232876712328766, + "grad_norm": 3.3323862552642822, + "learning_rate": 2.281021897810219e-06, + "loss": 2.3195, + "step": 139 + }, + { + "epoch": 0.15342465753424658, + "grad_norm": 3.2672693729400635, + "learning_rate": 2.299270072992701e-06, + "loss": 2.2797, + "step": 140 + }, + { + "epoch": 0.15342465753424658, + "eval_loss": 2.3943886756896973, + "eval_runtime": 37.3137, + "eval_samples_per_second": 26.8, + "eval_steps_per_second": 1.126, + "step": 140 + }, + { + "epoch": 0.15452054794520548, + "grad_norm": 3.9229774475097656, + "learning_rate": 2.3175182481751825e-06, + "loss": 2.2718, + "step": 141 + }, + { + "epoch": 0.15561643835616437, + "grad_norm": 2.6409823894500732, + "learning_rate": 2.3357664233576643e-06, + "loss": 2.2828, + "step": 142 + }, + { + "epoch": 0.1567123287671233, + "grad_norm": 4.6122612953186035, + "learning_rate": 2.354014598540146e-06, + "loss": 2.247, + "step": 143 + }, + { + "epoch": 0.1578082191780822, + "grad_norm": 3.391357898712158, + "learning_rate": 2.3722627737226276e-06, + "loss": 2.2863, + "step": 144 + }, + { + "epoch": 0.1589041095890411, + "grad_norm": 3.3550853729248047, + "learning_rate": 2.39051094890511e-06, + "loss": 2.2789, + "step": 145 + }, + { + "epoch": 0.16, + "grad_norm": 3.7376883029937744, + "learning_rate": 2.4087591240875918e-06, + "loss": 2.2297, + "step": 146 + }, + { + "epoch": 0.1610958904109589, + "grad_norm": 3.005157709121704, + "learning_rate": 2.427007299270073e-06, + "loss": 2.2467, + "step": 147 + }, + { + "epoch": 0.16219178082191782, + "grad_norm": 3.4616103172302246, + "learning_rate": 2.445255474452555e-06, + "loss": 2.2612, + "step": 148 + }, + { + "epoch": 0.1632876712328767, + "grad_norm": 3.3595101833343506, + "learning_rate": 2.463503649635037e-06, + "loss": 2.3021, + "step": 149 + }, + { + "epoch": 0.1643835616438356, + "grad_norm": 4.831241607666016, + "learning_rate": 2.4817518248175183e-06, + "loss": 2.2174, + "step": 150 + }, + { + "epoch": 0.1643835616438356, + "eval_loss": 2.3677749633789062, + "eval_runtime": 37.3092, + "eval_samples_per_second": 26.803, + "eval_steps_per_second": 1.126, + "step": 150 + }, + { + "epoch": 0.16547945205479453, + "grad_norm": 2.778820037841797, + "learning_rate": 2.5e-06, + "loss": 2.1856, + "step": 151 + }, + { + "epoch": 0.16657534246575342, + "grad_norm": 4.566643714904785, + "learning_rate": 2.518248175182482e-06, + "loss": 2.3186, + "step": 152 + }, + { + "epoch": 0.16767123287671232, + "grad_norm": 2.9908089637756348, + "learning_rate": 2.5364963503649635e-06, + "loss": 2.2127, + "step": 153 + }, + { + "epoch": 0.16876712328767124, + "grad_norm": 2.9702792167663574, + "learning_rate": 2.5547445255474458e-06, + "loss": 2.2071, + "step": 154 + }, + { + "epoch": 0.16986301369863013, + "grad_norm": 2.850044012069702, + "learning_rate": 2.5729927007299272e-06, + "loss": 2.2585, + "step": 155 + }, + { + "epoch": 0.17095890410958905, + "grad_norm": 2.926091432571411, + "learning_rate": 2.591240875912409e-06, + "loss": 2.2397, + "step": 156 + }, + { + "epoch": 0.17205479452054795, + "grad_norm": 2.8720169067382812, + "learning_rate": 2.6094890510948905e-06, + "loss": 2.2306, + "step": 157 + }, + { + "epoch": 0.17315068493150684, + "grad_norm": 2.8837177753448486, + "learning_rate": 2.627737226277373e-06, + "loss": 2.2666, + "step": 158 + }, + { + "epoch": 0.17424657534246576, + "grad_norm": 3.119621753692627, + "learning_rate": 2.6459854014598542e-06, + "loss": 2.2722, + "step": 159 + }, + { + "epoch": 0.17534246575342466, + "grad_norm": 2.7085700035095215, + "learning_rate": 2.664233576642336e-06, + "loss": 2.2346, + "step": 160 + }, + { + "epoch": 0.17534246575342466, + "eval_loss": 2.352083921432495, + "eval_runtime": 37.245, + "eval_samples_per_second": 26.849, + "eval_steps_per_second": 1.128, + "step": 160 + }, + { + "epoch": 0.17643835616438355, + "grad_norm": 2.9964346885681152, + "learning_rate": 2.6824817518248175e-06, + "loss": 2.2155, + "step": 161 + }, + { + "epoch": 0.17753424657534247, + "grad_norm": 3.1535208225250244, + "learning_rate": 2.7007299270072994e-06, + "loss": 2.235, + "step": 162 + }, + { + "epoch": 0.17863013698630137, + "grad_norm": 2.7240943908691406, + "learning_rate": 2.718978102189781e-06, + "loss": 2.2095, + "step": 163 + }, + { + "epoch": 0.17972602739726026, + "grad_norm": 3.122654676437378, + "learning_rate": 2.737226277372263e-06, + "loss": 2.2667, + "step": 164 + }, + { + "epoch": 0.18082191780821918, + "grad_norm": 2.9718410968780518, + "learning_rate": 2.755474452554745e-06, + "loss": 2.2499, + "step": 165 + }, + { + "epoch": 0.18191780821917808, + "grad_norm": 3.0081896781921387, + "learning_rate": 2.7737226277372264e-06, + "loss": 2.2089, + "step": 166 + }, + { + "epoch": 0.18301369863013697, + "grad_norm": 2.829509973526001, + "learning_rate": 2.7919708029197082e-06, + "loss": 2.3139, + "step": 167 + }, + { + "epoch": 0.1841095890410959, + "grad_norm": 3.4480397701263428, + "learning_rate": 2.81021897810219e-06, + "loss": 2.2705, + "step": 168 + }, + { + "epoch": 0.1852054794520548, + "grad_norm": 3.1107025146484375, + "learning_rate": 2.828467153284672e-06, + "loss": 2.1895, + "step": 169 + }, + { + "epoch": 0.1863013698630137, + "grad_norm": 2.916790008544922, + "learning_rate": 2.8467153284671534e-06, + "loss": 2.2317, + "step": 170 + }, + { + "epoch": 0.1863013698630137, + "eval_loss": 2.336153030395508, + "eval_runtime": 37.2401, + "eval_samples_per_second": 26.853, + "eval_steps_per_second": 1.128, + "step": 170 + }, + { + "epoch": 0.1873972602739726, + "grad_norm": 3.2582287788391113, + "learning_rate": 2.8649635036496353e-06, + "loss": 2.143, + "step": 171 + }, + { + "epoch": 0.1884931506849315, + "grad_norm": 2.430973768234253, + "learning_rate": 2.8832116788321167e-06, + "loss": 2.1977, + "step": 172 + }, + { + "epoch": 0.18958904109589042, + "grad_norm": 2.8763599395751953, + "learning_rate": 2.901459854014599e-06, + "loss": 2.1551, + "step": 173 + }, + { + "epoch": 0.19068493150684931, + "grad_norm": 2.9242653846740723, + "learning_rate": 2.9197080291970804e-06, + "loss": 2.2569, + "step": 174 + }, + { + "epoch": 0.1917808219178082, + "grad_norm": 2.623551607131958, + "learning_rate": 2.9379562043795623e-06, + "loss": 2.2329, + "step": 175 + }, + { + "epoch": 0.19287671232876713, + "grad_norm": 2.9511337280273438, + "learning_rate": 2.956204379562044e-06, + "loss": 2.2154, + "step": 176 + }, + { + "epoch": 0.19397260273972602, + "grad_norm": 2.415118932723999, + "learning_rate": 2.9744525547445256e-06, + "loss": 2.2637, + "step": 177 + }, + { + "epoch": 0.19506849315068492, + "grad_norm": 2.9664924144744873, + "learning_rate": 2.992700729927008e-06, + "loss": 2.2263, + "step": 178 + }, + { + "epoch": 0.19616438356164384, + "grad_norm": 2.7214572429656982, + "learning_rate": 3.0109489051094893e-06, + "loss": 2.1667, + "step": 179 + }, + { + "epoch": 0.19726027397260273, + "grad_norm": 2.8712823390960693, + "learning_rate": 3.029197080291971e-06, + "loss": 2.1661, + "step": 180 + }, + { + "epoch": 0.19726027397260273, + "eval_loss": 2.3191800117492676, + "eval_runtime": 37.24, + "eval_samples_per_second": 26.853, + "eval_steps_per_second": 1.128, + "step": 180 + }, + { + "epoch": 0.19835616438356166, + "grad_norm": 2.959709405899048, + "learning_rate": 3.0474452554744526e-06, + "loss": 2.2483, + "step": 181 + }, + { + "epoch": 0.19945205479452055, + "grad_norm": 2.567089319229126, + "learning_rate": 3.065693430656935e-06, + "loss": 2.2382, + "step": 182 + }, + { + "epoch": 0.20054794520547944, + "grad_norm": 2.47188401222229, + "learning_rate": 3.0839416058394163e-06, + "loss": 2.2148, + "step": 183 + }, + { + "epoch": 0.20164383561643837, + "grad_norm": 2.3435261249542236, + "learning_rate": 3.102189781021898e-06, + "loss": 2.2334, + "step": 184 + }, + { + "epoch": 0.20273972602739726, + "grad_norm": 2.5280468463897705, + "learning_rate": 3.1204379562043796e-06, + "loss": 2.2214, + "step": 185 + }, + { + "epoch": 0.20383561643835615, + "grad_norm": 2.4808971881866455, + "learning_rate": 3.1386861313868614e-06, + "loss": 2.1554, + "step": 186 + }, + { + "epoch": 0.20493150684931508, + "grad_norm": 2.9937026500701904, + "learning_rate": 3.156934306569343e-06, + "loss": 2.1389, + "step": 187 + }, + { + "epoch": 0.20602739726027397, + "grad_norm": 2.9220476150512695, + "learning_rate": 3.175182481751825e-06, + "loss": 2.2515, + "step": 188 + }, + { + "epoch": 0.20712328767123286, + "grad_norm": 3.0449001789093018, + "learning_rate": 3.193430656934307e-06, + "loss": 2.2016, + "step": 189 + }, + { + "epoch": 0.20821917808219179, + "grad_norm": 2.8027074337005615, + "learning_rate": 3.2116788321167884e-06, + "loss": 2.1823, + "step": 190 + }, + { + "epoch": 0.20821917808219179, + "eval_loss": 2.309677839279175, + "eval_runtime": 37.2442, + "eval_samples_per_second": 26.85, + "eval_steps_per_second": 1.128, + "step": 190 + }, + { + "epoch": 0.20931506849315068, + "grad_norm": 2.9158318042755127, + "learning_rate": 3.2299270072992707e-06, + "loss": 2.1928, + "step": 191 + }, + { + "epoch": 0.2104109589041096, + "grad_norm": 2.5328080654144287, + "learning_rate": 3.248175182481752e-06, + "loss": 2.1828, + "step": 192 + }, + { + "epoch": 0.2115068493150685, + "grad_norm": 3.1709704399108887, + "learning_rate": 3.266423357664234e-06, + "loss": 2.1052, + "step": 193 + }, + { + "epoch": 0.2126027397260274, + "grad_norm": 2.9156367778778076, + "learning_rate": 3.2846715328467155e-06, + "loss": 2.1943, + "step": 194 + }, + { + "epoch": 0.2136986301369863, + "grad_norm": 2.879408836364746, + "learning_rate": 3.3029197080291973e-06, + "loss": 2.1782, + "step": 195 + }, + { + "epoch": 0.2147945205479452, + "grad_norm": 3.748483180999756, + "learning_rate": 3.3211678832116788e-06, + "loss": 2.2107, + "step": 196 + }, + { + "epoch": 0.2158904109589041, + "grad_norm": 3.197510004043579, + "learning_rate": 3.339416058394161e-06, + "loss": 2.1845, + "step": 197 + }, + { + "epoch": 0.21698630136986302, + "grad_norm": 3.2596049308776855, + "learning_rate": 3.3576642335766425e-06, + "loss": 2.1816, + "step": 198 + }, + { + "epoch": 0.21808219178082192, + "grad_norm": 3.3329780101776123, + "learning_rate": 3.3759124087591243e-06, + "loss": 2.181, + "step": 199 + }, + { + "epoch": 0.2191780821917808, + "grad_norm": 2.7302327156066895, + "learning_rate": 3.3941605839416058e-06, + "loss": 2.1761, + "step": 200 + }, + { + "epoch": 0.2191780821917808, + "eval_loss": 2.2909653186798096, + "eval_runtime": 37.284, + "eval_samples_per_second": 26.821, + "eval_steps_per_second": 1.126, + "step": 200 + }, + { + "epoch": 0.22027397260273973, + "grad_norm": 3.002255916595459, + "learning_rate": 3.412408759124088e-06, + "loss": 2.1868, + "step": 201 + }, + { + "epoch": 0.22136986301369863, + "grad_norm": 2.6694891452789307, + "learning_rate": 3.43065693430657e-06, + "loss": 2.192, + "step": 202 + }, + { + "epoch": 0.22246575342465755, + "grad_norm": 2.9225986003875732, + "learning_rate": 3.4489051094890513e-06, + "loss": 2.1303, + "step": 203 + }, + { + "epoch": 0.22356164383561644, + "grad_norm": 2.7844080924987793, + "learning_rate": 3.467153284671533e-06, + "loss": 2.1788, + "step": 204 + }, + { + "epoch": 0.22465753424657534, + "grad_norm": 2.726123332977295, + "learning_rate": 3.4854014598540146e-06, + "loss": 2.1193, + "step": 205 + }, + { + "epoch": 0.22575342465753426, + "grad_norm": 2.838040590286255, + "learning_rate": 3.503649635036497e-06, + "loss": 2.1674, + "step": 206 + }, + { + "epoch": 0.22684931506849315, + "grad_norm": 2.6863937377929688, + "learning_rate": 3.5218978102189783e-06, + "loss": 2.161, + "step": 207 + }, + { + "epoch": 0.22794520547945205, + "grad_norm": 2.5116159915924072, + "learning_rate": 3.54014598540146e-06, + "loss": 2.179, + "step": 208 + }, + { + "epoch": 0.22904109589041097, + "grad_norm": 3.381863594055176, + "learning_rate": 3.5583941605839416e-06, + "loss": 2.2035, + "step": 209 + }, + { + "epoch": 0.23013698630136986, + "grad_norm": 3.2741382122039795, + "learning_rate": 3.576642335766424e-06, + "loss": 2.2518, + "step": 210 + }, + { + "epoch": 0.23013698630136986, + "eval_loss": 2.2773635387420654, + "eval_runtime": 37.2658, + "eval_samples_per_second": 26.834, + "eval_steps_per_second": 1.127, + "step": 210 + }, + { + "epoch": 0.23123287671232876, + "grad_norm": 2.917370319366455, + "learning_rate": 3.5948905109489054e-06, + "loss": 2.2265, + "step": 211 + }, + { + "epoch": 0.23232876712328768, + "grad_norm": 2.8392834663391113, + "learning_rate": 3.6131386861313872e-06, + "loss": 2.0709, + "step": 212 + }, + { + "epoch": 0.23342465753424657, + "grad_norm": 2.753612518310547, + "learning_rate": 3.6313868613138687e-06, + "loss": 2.0752, + "step": 213 + }, + { + "epoch": 0.2345205479452055, + "grad_norm": 3.1477746963500977, + "learning_rate": 3.6496350364963505e-06, + "loss": 2.1169, + "step": 214 + }, + { + "epoch": 0.2356164383561644, + "grad_norm": 2.5448100566864014, + "learning_rate": 3.667883211678833e-06, + "loss": 2.1338, + "step": 215 + }, + { + "epoch": 0.23671232876712328, + "grad_norm": 3.1622560024261475, + "learning_rate": 3.6861313868613142e-06, + "loss": 2.1065, + "step": 216 + }, + { + "epoch": 0.2378082191780822, + "grad_norm": 2.7054336071014404, + "learning_rate": 3.704379562043796e-06, + "loss": 2.1104, + "step": 217 + }, + { + "epoch": 0.2389041095890411, + "grad_norm": 2.964141845703125, + "learning_rate": 3.7226277372262775e-06, + "loss": 2.1949, + "step": 218 + }, + { + "epoch": 0.24, + "grad_norm": 2.7586112022399902, + "learning_rate": 3.7408759124087594e-06, + "loss": 2.165, + "step": 219 + }, + { + "epoch": 0.2410958904109589, + "grad_norm": 2.9035964012145996, + "learning_rate": 3.7591240875912412e-06, + "loss": 2.1192, + "step": 220 + }, + { + "epoch": 0.2410958904109589, + "eval_loss": 2.2649471759796143, + "eval_runtime": 37.4311, + "eval_samples_per_second": 26.716, + "eval_steps_per_second": 1.122, + "step": 220 + }, + { + "epoch": 0.2421917808219178, + "grad_norm": 2.4212076663970947, + "learning_rate": 3.777372262773723e-06, + "loss": 2.1389, + "step": 221 + }, + { + "epoch": 0.2432876712328767, + "grad_norm": 2.973227024078369, + "learning_rate": 3.7956204379562045e-06, + "loss": 2.132, + "step": 222 + }, + { + "epoch": 0.24438356164383562, + "grad_norm": 2.9831430912017822, + "learning_rate": 3.8138686131386864e-06, + "loss": 2.1142, + "step": 223 + }, + { + "epoch": 0.24547945205479452, + "grad_norm": 3.2880353927612305, + "learning_rate": 3.832116788321168e-06, + "loss": 2.168, + "step": 224 + }, + { + "epoch": 0.2465753424657534, + "grad_norm": 3.3920998573303223, + "learning_rate": 3.85036496350365e-06, + "loss": 2.1277, + "step": 225 + }, + { + "epoch": 0.24767123287671233, + "grad_norm": 2.7170026302337646, + "learning_rate": 3.868613138686132e-06, + "loss": 2.0936, + "step": 226 + }, + { + "epoch": 0.24876712328767123, + "grad_norm": 3.6220011711120605, + "learning_rate": 3.886861313868613e-06, + "loss": 2.1259, + "step": 227 + }, + { + "epoch": 0.24986301369863015, + "grad_norm": 3.4229273796081543, + "learning_rate": 3.905109489051096e-06, + "loss": 2.1421, + "step": 228 + }, + { + "epoch": 0.25095890410958904, + "grad_norm": 3.0358338356018066, + "learning_rate": 3.923357664233577e-06, + "loss": 2.1747, + "step": 229 + }, + { + "epoch": 0.25205479452054796, + "grad_norm": 2.611771821975708, + "learning_rate": 3.9416058394160585e-06, + "loss": 2.1282, + "step": 230 + }, + { + "epoch": 0.25205479452054796, + "eval_loss": 2.254887580871582, + "eval_runtime": 37.5943, + "eval_samples_per_second": 26.6, + "eval_steps_per_second": 1.117, + "step": 230 + }, + { + "epoch": 0.25315068493150683, + "grad_norm": 3.846412181854248, + "learning_rate": 3.95985401459854e-06, + "loss": 2.1892, + "step": 231 + }, + { + "epoch": 0.25424657534246575, + "grad_norm": 2.488525390625, + "learning_rate": 3.978102189781022e-06, + "loss": 2.2064, + "step": 232 + }, + { + "epoch": 0.2553424657534247, + "grad_norm": 3.114964485168457, + "learning_rate": 3.996350364963504e-06, + "loss": 2.0653, + "step": 233 + }, + { + "epoch": 0.25643835616438354, + "grad_norm": 2.5559768676757812, + "learning_rate": 4.014598540145986e-06, + "loss": 2.1503, + "step": 234 + }, + { + "epoch": 0.25753424657534246, + "grad_norm": 3.2347540855407715, + "learning_rate": 4.032846715328467e-06, + "loss": 2.0523, + "step": 235 + }, + { + "epoch": 0.2586301369863014, + "grad_norm": 2.8086111545562744, + "learning_rate": 4.05109489051095e-06, + "loss": 2.1206, + "step": 236 + }, + { + "epoch": 0.25972602739726025, + "grad_norm": 2.909231185913086, + "learning_rate": 4.069343065693431e-06, + "loss": 2.139, + "step": 237 + }, + { + "epoch": 0.2608219178082192, + "grad_norm": 3.2171082496643066, + "learning_rate": 4.0875912408759126e-06, + "loss": 2.0908, + "step": 238 + }, + { + "epoch": 0.2619178082191781, + "grad_norm": 2.3851265907287598, + "learning_rate": 4.105839416058395e-06, + "loss": 2.1133, + "step": 239 + }, + { + "epoch": 0.26301369863013696, + "grad_norm": 3.3912737369537354, + "learning_rate": 4.124087591240876e-06, + "loss": 2.1911, + "step": 240 + }, + { + "epoch": 0.26301369863013696, + "eval_loss": 2.2436957359313965, + "eval_runtime": 37.2947, + "eval_samples_per_second": 26.813, + "eval_steps_per_second": 1.126, + "step": 240 + }, + { + "epoch": 0.2641095890410959, + "grad_norm": 2.82906174659729, + "learning_rate": 4.1423357664233586e-06, + "loss": 2.11, + "step": 241 + }, + { + "epoch": 0.2652054794520548, + "grad_norm": 3.0317952632904053, + "learning_rate": 4.16058394160584e-06, + "loss": 2.0953, + "step": 242 + }, + { + "epoch": 0.2663013698630137, + "grad_norm": 3.6438329219818115, + "learning_rate": 4.1788321167883214e-06, + "loss": 2.1746, + "step": 243 + }, + { + "epoch": 0.2673972602739726, + "grad_norm": 2.6690592765808105, + "learning_rate": 4.197080291970803e-06, + "loss": 2.0555, + "step": 244 + }, + { + "epoch": 0.2684931506849315, + "grad_norm": 3.135376453399658, + "learning_rate": 4.215328467153285e-06, + "loss": 2.1394, + "step": 245 + }, + { + "epoch": 0.26958904109589044, + "grad_norm": 2.5474987030029297, + "learning_rate": 4.233576642335767e-06, + "loss": 2.0822, + "step": 246 + }, + { + "epoch": 0.2706849315068493, + "grad_norm": 2.883108377456665, + "learning_rate": 4.251824817518249e-06, + "loss": 2.0869, + "step": 247 + }, + { + "epoch": 0.2717808219178082, + "grad_norm": 2.9801037311553955, + "learning_rate": 4.27007299270073e-06, + "loss": 2.1194, + "step": 248 + }, + { + "epoch": 0.27287671232876715, + "grad_norm": 2.7688217163085938, + "learning_rate": 4.288321167883212e-06, + "loss": 2.0908, + "step": 249 + }, + { + "epoch": 0.273972602739726, + "grad_norm": 3.205148458480835, + "learning_rate": 4.306569343065693e-06, + "loss": 2.0762, + "step": 250 + }, + { + "epoch": 0.273972602739726, + "eval_loss": 2.238699197769165, + "eval_runtime": 37.2428, + "eval_samples_per_second": 26.851, + "eval_steps_per_second": 1.128, + "step": 250 + }, + { + "epoch": 0.27506849315068493, + "grad_norm": 2.8262946605682373, + "learning_rate": 4.3248175182481755e-06, + "loss": 2.0509, + "step": 251 + }, + { + "epoch": 0.27616438356164386, + "grad_norm": 3.156261682510376, + "learning_rate": 4.343065693430658e-06, + "loss": 2.1213, + "step": 252 + }, + { + "epoch": 0.2772602739726027, + "grad_norm": 2.6367998123168945, + "learning_rate": 4.361313868613139e-06, + "loss": 2.1071, + "step": 253 + }, + { + "epoch": 0.27835616438356164, + "grad_norm": 2.871534824371338, + "learning_rate": 4.379562043795621e-06, + "loss": 2.1008, + "step": 254 + }, + { + "epoch": 0.27945205479452057, + "grad_norm": 2.839432716369629, + "learning_rate": 4.397810218978102e-06, + "loss": 2.1293, + "step": 255 + }, + { + "epoch": 0.28054794520547943, + "grad_norm": 2.627143383026123, + "learning_rate": 4.416058394160584e-06, + "loss": 2.1141, + "step": 256 + }, + { + "epoch": 0.28164383561643835, + "grad_norm": 2.8292996883392334, + "learning_rate": 4.434306569343066e-06, + "loss": 2.0981, + "step": 257 + }, + { + "epoch": 0.2827397260273973, + "grad_norm": 2.5711069107055664, + "learning_rate": 4.452554744525548e-06, + "loss": 2.0707, + "step": 258 + }, + { + "epoch": 0.28383561643835614, + "grad_norm": 2.488954544067383, + "learning_rate": 4.4708029197080295e-06, + "loss": 2.1172, + "step": 259 + }, + { + "epoch": 0.28493150684931506, + "grad_norm": 2.442209005355835, + "learning_rate": 4.489051094890512e-06, + "loss": 2.1119, + "step": 260 + }, + { + "epoch": 0.28493150684931506, + "eval_loss": 2.2306911945343018, + "eval_runtime": 37.2412, + "eval_samples_per_second": 26.852, + "eval_steps_per_second": 1.128, + "step": 260 + }, + { + "epoch": 0.286027397260274, + "grad_norm": 2.9147934913635254, + "learning_rate": 4.507299270072993e-06, + "loss": 2.1517, + "step": 261 + }, + { + "epoch": 0.28712328767123285, + "grad_norm": 2.549833297729492, + "learning_rate": 4.525547445255475e-06, + "loss": 2.1047, + "step": 262 + }, + { + "epoch": 0.2882191780821918, + "grad_norm": 3.3405206203460693, + "learning_rate": 4.543795620437956e-06, + "loss": 2.0975, + "step": 263 + }, + { + "epoch": 0.2893150684931507, + "grad_norm": 2.554431676864624, + "learning_rate": 4.562043795620438e-06, + "loss": 2.0622, + "step": 264 + }, + { + "epoch": 0.29041095890410956, + "grad_norm": 3.5529873371124268, + "learning_rate": 4.580291970802921e-06, + "loss": 2.1567, + "step": 265 + }, + { + "epoch": 0.2915068493150685, + "grad_norm": 2.421682119369507, + "learning_rate": 4.598540145985402e-06, + "loss": 2.1619, + "step": 266 + }, + { + "epoch": 0.2926027397260274, + "grad_norm": 2.918288230895996, + "learning_rate": 4.6167883211678835e-06, + "loss": 2.1238, + "step": 267 + }, + { + "epoch": 0.2936986301369863, + "grad_norm": 3.0737621784210205, + "learning_rate": 4.635036496350365e-06, + "loss": 2.0728, + "step": 268 + }, + { + "epoch": 0.2947945205479452, + "grad_norm": 2.5862410068511963, + "learning_rate": 4.653284671532847e-06, + "loss": 2.0675, + "step": 269 + }, + { + "epoch": 0.2958904109589041, + "grad_norm": 2.9004979133605957, + "learning_rate": 4.671532846715329e-06, + "loss": 2.1248, + "step": 270 + }, + { + "epoch": 0.2958904109589041, + "eval_loss": 2.2179365158081055, + "eval_runtime": 37.2352, + "eval_samples_per_second": 26.856, + "eval_steps_per_second": 1.128, + "step": 270 + }, + { + "epoch": 0.29698630136986304, + "grad_norm": 2.4784982204437256, + "learning_rate": 4.689781021897811e-06, + "loss": 2.1276, + "step": 271 + }, + { + "epoch": 0.2980821917808219, + "grad_norm": 2.7971739768981934, + "learning_rate": 4.708029197080292e-06, + "loss": 2.0931, + "step": 272 + }, + { + "epoch": 0.2991780821917808, + "grad_norm": 2.5839481353759766, + "learning_rate": 4.726277372262774e-06, + "loss": 2.1371, + "step": 273 + }, + { + "epoch": 0.30027397260273975, + "grad_norm": 2.7168564796447754, + "learning_rate": 4.744525547445255e-06, + "loss": 2.1086, + "step": 274 + }, + { + "epoch": 0.3013698630136986, + "grad_norm": 2.6566879749298096, + "learning_rate": 4.7627737226277375e-06, + "loss": 2.0368, + "step": 275 + }, + { + "epoch": 0.30246575342465754, + "grad_norm": 2.5068323612213135, + "learning_rate": 4.78102189781022e-06, + "loss": 2.1077, + "step": 276 + }, + { + "epoch": 0.30356164383561646, + "grad_norm": 2.631408214569092, + "learning_rate": 4.799270072992701e-06, + "loss": 2.1013, + "step": 277 + }, + { + "epoch": 0.3046575342465753, + "grad_norm": 2.5642812252044678, + "learning_rate": 4.8175182481751835e-06, + "loss": 2.0739, + "step": 278 + }, + { + "epoch": 0.30575342465753425, + "grad_norm": 2.3326497077941895, + "learning_rate": 4.835766423357665e-06, + "loss": 2.0971, + "step": 279 + }, + { + "epoch": 0.30684931506849317, + "grad_norm": 2.70701265335083, + "learning_rate": 4.854014598540146e-06, + "loss": 2.0904, + "step": 280 + }, + { + "epoch": 0.30684931506849317, + "eval_loss": 2.215972900390625, + "eval_runtime": 37.2196, + "eval_samples_per_second": 26.868, + "eval_steps_per_second": 1.128, + "step": 280 + }, + { + "epoch": 0.30794520547945203, + "grad_norm": 2.615201234817505, + "learning_rate": 4.872262773722628e-06, + "loss": 2.039, + "step": 281 + }, + { + "epoch": 0.30904109589041096, + "grad_norm": 2.2821998596191406, + "learning_rate": 4.89051094890511e-06, + "loss": 2.0529, + "step": 282 + }, + { + "epoch": 0.3101369863013699, + "grad_norm": 2.767153024673462, + "learning_rate": 4.9087591240875915e-06, + "loss": 2.0351, + "step": 283 + }, + { + "epoch": 0.31123287671232874, + "grad_norm": 2.537121295928955, + "learning_rate": 4.927007299270074e-06, + "loss": 2.0752, + "step": 284 + }, + { + "epoch": 0.31232876712328766, + "grad_norm": 2.835052013397217, + "learning_rate": 4.945255474452555e-06, + "loss": 2.0892, + "step": 285 + }, + { + "epoch": 0.3134246575342466, + "grad_norm": 2.5154924392700195, + "learning_rate": 4.963503649635037e-06, + "loss": 2.0569, + "step": 286 + }, + { + "epoch": 0.31452054794520545, + "grad_norm": 2.8212568759918213, + "learning_rate": 4.981751824817518e-06, + "loss": 2.1588, + "step": 287 + }, + { + "epoch": 0.3156164383561644, + "grad_norm": 2.3840994834899902, + "learning_rate": 5e-06, + "loss": 2.1409, + "step": 288 + }, + { + "epoch": 0.3167123287671233, + "grad_norm": 2.9641315937042236, + "learning_rate": 4.997969130787978e-06, + "loss": 2.0892, + "step": 289 + }, + { + "epoch": 0.3178082191780822, + "grad_norm": 2.4516749382019043, + "learning_rate": 4.995938261575955e-06, + "loss": 2.1105, + "step": 290 + }, + { + "epoch": 0.3178082191780822, + "eval_loss": 2.207075834274292, + "eval_runtime": 37.1304, + "eval_samples_per_second": 26.932, + "eval_steps_per_second": 1.131, + "step": 290 + }, + { + "epoch": 0.3189041095890411, + "grad_norm": 2.768869400024414, + "learning_rate": 4.993907392363932e-06, + "loss": 2.0231, + "step": 291 + }, + { + "epoch": 0.32, + "grad_norm": 2.771475315093994, + "learning_rate": 4.9918765231519095e-06, + "loss": 2.1018, + "step": 292 + }, + { + "epoch": 0.32109589041095893, + "grad_norm": 2.773983955383301, + "learning_rate": 4.989845653939887e-06, + "loss": 2.0317, + "step": 293 + }, + { + "epoch": 0.3221917808219178, + "grad_norm": 2.4447343349456787, + "learning_rate": 4.987814784727864e-06, + "loss": 2.0739, + "step": 294 + }, + { + "epoch": 0.3232876712328767, + "grad_norm": 2.444261312484741, + "learning_rate": 4.985783915515841e-06, + "loss": 2.073, + "step": 295 + }, + { + "epoch": 0.32438356164383564, + "grad_norm": 2.871792793273926, + "learning_rate": 4.983753046303819e-06, + "loss": 2.058, + "step": 296 + }, + { + "epoch": 0.3254794520547945, + "grad_norm": 2.3651537895202637, + "learning_rate": 4.981722177091796e-06, + "loss": 2.0749, + "step": 297 + }, + { + "epoch": 0.3265753424657534, + "grad_norm": 2.3395845890045166, + "learning_rate": 4.979691307879773e-06, + "loss": 2.065, + "step": 298 + }, + { + "epoch": 0.32767123287671235, + "grad_norm": 2.6275217533111572, + "learning_rate": 4.9776604386677504e-06, + "loss": 2.0945, + "step": 299 + }, + { + "epoch": 0.3287671232876712, + "grad_norm": 2.866285562515259, + "learning_rate": 4.975629569455728e-06, + "loss": 2.0702, + "step": 300 + }, + { + "epoch": 0.3287671232876712, + "eval_loss": 2.200012683868408, + "eval_runtime": 37.2082, + "eval_samples_per_second": 26.876, + "eval_steps_per_second": 1.129, + "step": 300 + }, + { + "epoch": 0.32986301369863014, + "grad_norm": 2.268401861190796, + "learning_rate": 4.973598700243705e-06, + "loss": 2.1257, + "step": 301 + }, + { + "epoch": 0.33095890410958906, + "grad_norm": 2.6197733879089355, + "learning_rate": 4.971567831031681e-06, + "loss": 2.0347, + "step": 302 + }, + { + "epoch": 0.3320547945205479, + "grad_norm": 2.4738848209381104, + "learning_rate": 4.9695369618196595e-06, + "loss": 2.0368, + "step": 303 + }, + { + "epoch": 0.33315068493150685, + "grad_norm": 2.705670118331909, + "learning_rate": 4.967506092607636e-06, + "loss": 1.9256, + "step": 304 + }, + { + "epoch": 0.33424657534246577, + "grad_norm": 2.813279628753662, + "learning_rate": 4.965475223395614e-06, + "loss": 2.0214, + "step": 305 + }, + { + "epoch": 0.33534246575342463, + "grad_norm": 2.320960283279419, + "learning_rate": 4.9634443541835905e-06, + "loss": 2.0342, + "step": 306 + }, + { + "epoch": 0.33643835616438356, + "grad_norm": 2.4495227336883545, + "learning_rate": 4.961413484971569e-06, + "loss": 2.0421, + "step": 307 + }, + { + "epoch": 0.3375342465753425, + "grad_norm": 2.5384018421173096, + "learning_rate": 4.959382615759545e-06, + "loss": 2.0572, + "step": 308 + }, + { + "epoch": 0.33863013698630134, + "grad_norm": 2.4018006324768066, + "learning_rate": 4.957351746547523e-06, + "loss": 2.0617, + "step": 309 + }, + { + "epoch": 0.33972602739726027, + "grad_norm": 2.552424907684326, + "learning_rate": 4.9553208773355e-06, + "loss": 2.0384, + "step": 310 + }, + { + "epoch": 0.33972602739726027, + "eval_loss": 2.1974849700927734, + "eval_runtime": 37.2308, + "eval_samples_per_second": 26.859, + "eval_steps_per_second": 1.128, + "step": 310 + }, + { + "epoch": 0.3408219178082192, + "grad_norm": 2.4587671756744385, + "learning_rate": 4.953290008123478e-06, + "loss": 2.1701, + "step": 311 + }, + { + "epoch": 0.3419178082191781, + "grad_norm": 2.4486496448516846, + "learning_rate": 4.951259138911454e-06, + "loss": 2.0654, + "step": 312 + }, + { + "epoch": 0.343013698630137, + "grad_norm": 2.123568058013916, + "learning_rate": 4.9492282696994315e-06, + "loss": 1.9961, + "step": 313 + }, + { + "epoch": 0.3441095890410959, + "grad_norm": 2.549788475036621, + "learning_rate": 4.947197400487409e-06, + "loss": 2.1265, + "step": 314 + }, + { + "epoch": 0.3452054794520548, + "grad_norm": 2.3453609943389893, + "learning_rate": 4.945166531275386e-06, + "loss": 2.0501, + "step": 315 + }, + { + "epoch": 0.3463013698630137, + "grad_norm": 2.3584256172180176, + "learning_rate": 4.943135662063363e-06, + "loss": 2.0402, + "step": 316 + }, + { + "epoch": 0.3473972602739726, + "grad_norm": 3.3100943565368652, + "learning_rate": 4.9411047928513406e-06, + "loss": 2.0545, + "step": 317 + }, + { + "epoch": 0.34849315068493153, + "grad_norm": 2.4266037940979004, + "learning_rate": 4.939073923639318e-06, + "loss": 2.0926, + "step": 318 + }, + { + "epoch": 0.3495890410958904, + "grad_norm": 3.101090908050537, + "learning_rate": 4.937043054427295e-06, + "loss": 2.0987, + "step": 319 + }, + { + "epoch": 0.3506849315068493, + "grad_norm": 2.743619203567505, + "learning_rate": 4.935012185215272e-06, + "loss": 2.0315, + "step": 320 + }, + { + "epoch": 0.3506849315068493, + "eval_loss": 2.188877820968628, + "eval_runtime": 37.4609, + "eval_samples_per_second": 26.694, + "eval_steps_per_second": 1.121, + "step": 320 + }, + { + "epoch": 0.35178082191780824, + "grad_norm": 2.6065986156463623, + "learning_rate": 4.93298131600325e-06, + "loss": 2.0166, + "step": 321 + }, + { + "epoch": 0.3528767123287671, + "grad_norm": 2.7554283142089844, + "learning_rate": 4.930950446791227e-06, + "loss": 2.0862, + "step": 322 + }, + { + "epoch": 0.353972602739726, + "grad_norm": 2.4773144721984863, + "learning_rate": 4.928919577579204e-06, + "loss": 2.0472, + "step": 323 + }, + { + "epoch": 0.35506849315068495, + "grad_norm": 2.695289134979248, + "learning_rate": 4.9268887083671815e-06, + "loss": 2.0567, + "step": 324 + }, + { + "epoch": 0.3561643835616438, + "grad_norm": 2.408220052719116, + "learning_rate": 4.924857839155159e-06, + "loss": 2.046, + "step": 325 + }, + { + "epoch": 0.35726027397260274, + "grad_norm": 2.3851559162139893, + "learning_rate": 4.922826969943136e-06, + "loss": 2.054, + "step": 326 + }, + { + "epoch": 0.35835616438356166, + "grad_norm": 2.8667280673980713, + "learning_rate": 4.920796100731113e-06, + "loss": 2.0717, + "step": 327 + }, + { + "epoch": 0.3594520547945205, + "grad_norm": 2.3814942836761475, + "learning_rate": 4.918765231519091e-06, + "loss": 1.9884, + "step": 328 + }, + { + "epoch": 0.36054794520547945, + "grad_norm": 2.341076612472534, + "learning_rate": 4.916734362307068e-06, + "loss": 2.0474, + "step": 329 + }, + { + "epoch": 0.36164383561643837, + "grad_norm": 2.3571853637695312, + "learning_rate": 4.914703493095045e-06, + "loss": 2.0009, + "step": 330 + }, + { + "epoch": 0.36164383561643837, + "eval_loss": 2.186048746109009, + "eval_runtime": 37.5746, + "eval_samples_per_second": 26.614, + "eval_steps_per_second": 1.118, + "step": 330 + }, + { + "epoch": 0.36273972602739724, + "grad_norm": 2.3770196437835693, + "learning_rate": 4.912672623883022e-06, + "loss": 2.0167, + "step": 331 + }, + { + "epoch": 0.36383561643835616, + "grad_norm": 2.3357701301574707, + "learning_rate": 4.910641754671e-06, + "loss": 1.9972, + "step": 332 + }, + { + "epoch": 0.3649315068493151, + "grad_norm": 2.4100499153137207, + "learning_rate": 4.908610885458977e-06, + "loss": 2.0319, + "step": 333 + }, + { + "epoch": 0.36602739726027395, + "grad_norm": 2.193018913269043, + "learning_rate": 4.906580016246954e-06, + "loss": 2.0285, + "step": 334 + }, + { + "epoch": 0.36712328767123287, + "grad_norm": 2.275637149810791, + "learning_rate": 4.9045491470349315e-06, + "loss": 2.0749, + "step": 335 + }, + { + "epoch": 0.3682191780821918, + "grad_norm": 2.347121238708496, + "learning_rate": 4.902518277822909e-06, + "loss": 2.0502, + "step": 336 + }, + { + "epoch": 0.3693150684931507, + "grad_norm": 2.210817575454712, + "learning_rate": 4.900487408610886e-06, + "loss": 2.0612, + "step": 337 + }, + { + "epoch": 0.3704109589041096, + "grad_norm": 2.15594482421875, + "learning_rate": 4.898456539398863e-06, + "loss": 2.0653, + "step": 338 + }, + { + "epoch": 0.3715068493150685, + "grad_norm": 2.3996052742004395, + "learning_rate": 4.89642567018684e-06, + "loss": 2.074, + "step": 339 + }, + { + "epoch": 0.3726027397260274, + "grad_norm": 2.384172201156616, + "learning_rate": 4.894394800974818e-06, + "loss": 2.1533, + "step": 340 + }, + { + "epoch": 0.3726027397260274, + "eval_loss": 2.175034284591675, + "eval_runtime": 37.2455, + "eval_samples_per_second": 26.849, + "eval_steps_per_second": 1.128, + "step": 340 + }, + { + "epoch": 0.3736986301369863, + "grad_norm": 2.3431336879730225, + "learning_rate": 4.892363931762794e-06, + "loss": 1.9919, + "step": 341 + }, + { + "epoch": 0.3747945205479452, + "grad_norm": 2.1828389167785645, + "learning_rate": 4.8903330625507725e-06, + "loss": 2.0703, + "step": 342 + }, + { + "epoch": 0.37589041095890413, + "grad_norm": 2.2117249965667725, + "learning_rate": 4.888302193338749e-06, + "loss": 2.0071, + "step": 343 + }, + { + "epoch": 0.376986301369863, + "grad_norm": 2.3595967292785645, + "learning_rate": 4.886271324126727e-06, + "loss": 1.9879, + "step": 344 + }, + { + "epoch": 0.3780821917808219, + "grad_norm": 2.5211799144744873, + "learning_rate": 4.8842404549147034e-06, + "loss": 2.0861, + "step": 345 + }, + { + "epoch": 0.37917808219178084, + "grad_norm": 2.1551895141601562, + "learning_rate": 4.8822095857026816e-06, + "loss": 2.0683, + "step": 346 + }, + { + "epoch": 0.3802739726027397, + "grad_norm": 2.4145829677581787, + "learning_rate": 4.880178716490658e-06, + "loss": 2.043, + "step": 347 + }, + { + "epoch": 0.38136986301369863, + "grad_norm": 2.4273521900177, + "learning_rate": 4.878147847278636e-06, + "loss": 1.9928, + "step": 348 + }, + { + "epoch": 0.38246575342465755, + "grad_norm": 2.261054754257202, + "learning_rate": 4.8761169780666125e-06, + "loss": 2.0065, + "step": 349 + }, + { + "epoch": 0.3835616438356164, + "grad_norm": 2.4703147411346436, + "learning_rate": 4.874086108854591e-06, + "loss": 1.9766, + "step": 350 + }, + { + "epoch": 0.3835616438356164, + "eval_loss": 2.176915168762207, + "eval_runtime": 37.3223, + "eval_samples_per_second": 26.794, + "eval_steps_per_second": 1.125, + "step": 350 + }, + { + "epoch": 0.38465753424657534, + "grad_norm": 2.3353374004364014, + "learning_rate": 4.872055239642567e-06, + "loss": 1.9874, + "step": 351 + }, + { + "epoch": 0.38575342465753426, + "grad_norm": 2.227834463119507, + "learning_rate": 4.870024370430544e-06, + "loss": 1.9644, + "step": 352 + }, + { + "epoch": 0.3868493150684931, + "grad_norm": 2.3155159950256348, + "learning_rate": 4.867993501218522e-06, + "loss": 2.0227, + "step": 353 + }, + { + "epoch": 0.38794520547945205, + "grad_norm": 2.2456047534942627, + "learning_rate": 4.865962632006499e-06, + "loss": 2.0566, + "step": 354 + }, + { + "epoch": 0.38904109589041097, + "grad_norm": 2.3364651203155518, + "learning_rate": 4.863931762794476e-06, + "loss": 1.9941, + "step": 355 + }, + { + "epoch": 0.39013698630136984, + "grad_norm": 2.2886059284210205, + "learning_rate": 4.8619008935824535e-06, + "loss": 2.0163, + "step": 356 + }, + { + "epoch": 0.39123287671232876, + "grad_norm": 2.3796911239624023, + "learning_rate": 4.859870024370431e-06, + "loss": 2.0525, + "step": 357 + }, + { + "epoch": 0.3923287671232877, + "grad_norm": 2.381986379623413, + "learning_rate": 4.857839155158408e-06, + "loss": 2.0682, + "step": 358 + }, + { + "epoch": 0.3934246575342466, + "grad_norm": 2.2048425674438477, + "learning_rate": 4.855808285946385e-06, + "loss": 2.0748, + "step": 359 + }, + { + "epoch": 0.39452054794520547, + "grad_norm": 2.162846088409424, + "learning_rate": 4.8537774167343626e-06, + "loss": 2.0512, + "step": 360 + }, + { + "epoch": 0.39452054794520547, + "eval_loss": 2.1645994186401367, + "eval_runtime": 37.3126, + "eval_samples_per_second": 26.801, + "eval_steps_per_second": 1.126, + "step": 360 + }, + { + "epoch": 0.3956164383561644, + "grad_norm": 2.4070873260498047, + "learning_rate": 4.85174654752234e-06, + "loss": 2.042, + "step": 361 + }, + { + "epoch": 0.3967123287671233, + "grad_norm": 2.11716628074646, + "learning_rate": 4.849715678310317e-06, + "loss": 2.0398, + "step": 362 + }, + { + "epoch": 0.3978082191780822, + "grad_norm": 2.265479803085327, + "learning_rate": 4.847684809098294e-06, + "loss": 2.0334, + "step": 363 + }, + { + "epoch": 0.3989041095890411, + "grad_norm": 2.2270994186401367, + "learning_rate": 4.845653939886272e-06, + "loss": 2.0292, + "step": 364 + }, + { + "epoch": 0.4, + "grad_norm": 2.406064033508301, + "learning_rate": 4.843623070674249e-06, + "loss": 2.0351, + "step": 365 + }, + { + "epoch": 0.4010958904109589, + "grad_norm": 2.216654062271118, + "learning_rate": 4.841592201462226e-06, + "loss": 2.005, + "step": 366 + }, + { + "epoch": 0.4021917808219178, + "grad_norm": 2.4241385459899902, + "learning_rate": 4.8395613322502035e-06, + "loss": 2.0647, + "step": 367 + }, + { + "epoch": 0.40328767123287673, + "grad_norm": 2.4315145015716553, + "learning_rate": 4.837530463038181e-06, + "loss": 1.9679, + "step": 368 + }, + { + "epoch": 0.4043835616438356, + "grad_norm": 2.3034439086914062, + "learning_rate": 4.835499593826158e-06, + "loss": 2.0259, + "step": 369 + }, + { + "epoch": 0.4054794520547945, + "grad_norm": 2.493436098098755, + "learning_rate": 4.833468724614135e-06, + "loss": 2.0468, + "step": 370 + }, + { + "epoch": 0.4054794520547945, + "eval_loss": 2.161508083343506, + "eval_runtime": 37.2692, + "eval_samples_per_second": 26.832, + "eval_steps_per_second": 1.127, + "step": 370 + }, + { + "epoch": 0.40657534246575344, + "grad_norm": 2.3887157440185547, + "learning_rate": 4.831437855402113e-06, + "loss": 2.0313, + "step": 371 + }, + { + "epoch": 0.4076712328767123, + "grad_norm": 2.563095808029175, + "learning_rate": 4.82940698619009e-06, + "loss": 1.9288, + "step": 372 + }, + { + "epoch": 0.40876712328767123, + "grad_norm": 2.2609968185424805, + "learning_rate": 4.827376116978067e-06, + "loss": 1.9975, + "step": 373 + }, + { + "epoch": 0.40986301369863015, + "grad_norm": 2.624257802963257, + "learning_rate": 4.8253452477660444e-06, + "loss": 2.0639, + "step": 374 + }, + { + "epoch": 0.410958904109589, + "grad_norm": 2.502197265625, + "learning_rate": 4.823314378554022e-06, + "loss": 2.0424, + "step": 375 + }, + { + "epoch": 0.41205479452054794, + "grad_norm": 2.36297345161438, + "learning_rate": 4.821283509341998e-06, + "loss": 2.027, + "step": 376 + }, + { + "epoch": 0.41315068493150686, + "grad_norm": 2.388918161392212, + "learning_rate": 4.819252640129976e-06, + "loss": 2.0002, + "step": 377 + }, + { + "epoch": 0.41424657534246573, + "grad_norm": 2.4262335300445557, + "learning_rate": 4.817221770917953e-06, + "loss": 2.0215, + "step": 378 + }, + { + "epoch": 0.41534246575342465, + "grad_norm": 2.8953094482421875, + "learning_rate": 4.815190901705931e-06, + "loss": 1.988, + "step": 379 + }, + { + "epoch": 0.41643835616438357, + "grad_norm": 2.3092124462127686, + "learning_rate": 4.813160032493907e-06, + "loss": 2.0199, + "step": 380 + }, + { + "epoch": 0.41643835616438357, + "eval_loss": 2.1589064598083496, + "eval_runtime": 37.2649, + "eval_samples_per_second": 26.835, + "eval_steps_per_second": 1.127, + "step": 380 + }, + { + "epoch": 0.41753424657534244, + "grad_norm": 3.0842413902282715, + "learning_rate": 4.811129163281885e-06, + "loss": 2.0134, + "step": 381 + }, + { + "epoch": 0.41863013698630136, + "grad_norm": 2.2015092372894287, + "learning_rate": 4.809098294069862e-06, + "loss": 2.0447, + "step": 382 + }, + { + "epoch": 0.4197260273972603, + "grad_norm": 2.93471097946167, + "learning_rate": 4.80706742485784e-06, + "loss": 2.0329, + "step": 383 + }, + { + "epoch": 0.4208219178082192, + "grad_norm": 2.3277227878570557, + "learning_rate": 4.805036555645816e-06, + "loss": 2.0805, + "step": 384 + }, + { + "epoch": 0.42191780821917807, + "grad_norm": 2.357966184616089, + "learning_rate": 4.8030056864337945e-06, + "loss": 2.0008, + "step": 385 + }, + { + "epoch": 0.423013698630137, + "grad_norm": 2.7505595684051514, + "learning_rate": 4.800974817221771e-06, + "loss": 2.059, + "step": 386 + }, + { + "epoch": 0.4241095890410959, + "grad_norm": 2.144338369369507, + "learning_rate": 4.798943948009749e-06, + "loss": 2.0125, + "step": 387 + }, + { + "epoch": 0.4252054794520548, + "grad_norm": 2.608407735824585, + "learning_rate": 4.7969130787977254e-06, + "loss": 2.0263, + "step": 388 + }, + { + "epoch": 0.4263013698630137, + "grad_norm": 2.4293761253356934, + "learning_rate": 4.794882209585703e-06, + "loss": 2.0107, + "step": 389 + }, + { + "epoch": 0.4273972602739726, + "grad_norm": 2.2136595249176025, + "learning_rate": 4.79285134037368e-06, + "loss": 2.0421, + "step": 390 + }, + { + "epoch": 0.4273972602739726, + "eval_loss": 2.1473641395568848, + "eval_runtime": 37.2895, + "eval_samples_per_second": 26.817, + "eval_steps_per_second": 1.126, + "step": 390 + }, + { + "epoch": 0.4284931506849315, + "grad_norm": 2.3706328868865967, + "learning_rate": 4.790820471161657e-06, + "loss": 1.9828, + "step": 391 + }, + { + "epoch": 0.4295890410958904, + "grad_norm": 2.1628777980804443, + "learning_rate": 4.7887896019496345e-06, + "loss": 2.0006, + "step": 392 + }, + { + "epoch": 0.43068493150684933, + "grad_norm": 2.5572149753570557, + "learning_rate": 4.786758732737612e-06, + "loss": 2.0046, + "step": 393 + }, + { + "epoch": 0.4317808219178082, + "grad_norm": 2.44661808013916, + "learning_rate": 4.784727863525589e-06, + "loss": 2.0513, + "step": 394 + }, + { + "epoch": 0.4328767123287671, + "grad_norm": 2.429487466812134, + "learning_rate": 4.782696994313566e-06, + "loss": 1.9619, + "step": 395 + }, + { + "epoch": 0.43397260273972604, + "grad_norm": 2.233353614807129, + "learning_rate": 4.780666125101544e-06, + "loss": 2.0141, + "step": 396 + }, + { + "epoch": 0.4350684931506849, + "grad_norm": 2.137549638748169, + "learning_rate": 4.778635255889521e-06, + "loss": 1.9512, + "step": 397 + }, + { + "epoch": 0.43616438356164383, + "grad_norm": 2.187638759613037, + "learning_rate": 4.776604386677498e-06, + "loss": 1.9802, + "step": 398 + }, + { + "epoch": 0.43726027397260275, + "grad_norm": 2.3084137439727783, + "learning_rate": 4.7745735174654755e-06, + "loss": 1.9887, + "step": 399 + }, + { + "epoch": 0.4383561643835616, + "grad_norm": 2.16654109954834, + "learning_rate": 4.772542648253453e-06, + "loss": 2.0535, + "step": 400 + }, + { + "epoch": 0.4383561643835616, + "eval_loss": 2.1444413661956787, + "eval_runtime": 37.2158, + "eval_samples_per_second": 26.87, + "eval_steps_per_second": 1.129, + "step": 400 + }, + { + "epoch": 0.43945205479452054, + "grad_norm": 2.0973072052001953, + "learning_rate": 4.77051177904143e-06, + "loss": 2.0529, + "step": 401 + }, + { + "epoch": 0.44054794520547946, + "grad_norm": 2.421881675720215, + "learning_rate": 4.768480909829407e-06, + "loss": 1.9893, + "step": 402 + }, + { + "epoch": 0.44164383561643833, + "grad_norm": 2.349296808242798, + "learning_rate": 4.766450040617385e-06, + "loss": 2.0237, + "step": 403 + }, + { + "epoch": 0.44273972602739725, + "grad_norm": 2.2154288291931152, + "learning_rate": 4.764419171405362e-06, + "loss": 1.996, + "step": 404 + }, + { + "epoch": 0.4438356164383562, + "grad_norm": 2.6355648040771484, + "learning_rate": 4.762388302193339e-06, + "loss": 2.0345, + "step": 405 + }, + { + "epoch": 0.4449315068493151, + "grad_norm": 2.352238655090332, + "learning_rate": 4.760357432981316e-06, + "loss": 2.0219, + "step": 406 + }, + { + "epoch": 0.44602739726027396, + "grad_norm": 2.1886510848999023, + "learning_rate": 4.758326563769294e-06, + "loss": 1.9923, + "step": 407 + }, + { + "epoch": 0.4471232876712329, + "grad_norm": 2.6483092308044434, + "learning_rate": 4.756295694557271e-06, + "loss": 2.0236, + "step": 408 + }, + { + "epoch": 0.4482191780821918, + "grad_norm": 2.382389545440674, + "learning_rate": 4.754264825345248e-06, + "loss": 2.0249, + "step": 409 + }, + { + "epoch": 0.44931506849315067, + "grad_norm": 2.4824411869049072, + "learning_rate": 4.7522339561332255e-06, + "loss": 1.9571, + "step": 410 + }, + { + "epoch": 0.44931506849315067, + "eval_loss": 2.1502621173858643, + "eval_runtime": 37.1789, + "eval_samples_per_second": 26.897, + "eval_steps_per_second": 1.13, + "step": 410 + }, + { + "epoch": 0.4504109589041096, + "grad_norm": 2.325589656829834, + "learning_rate": 4.750203086921203e-06, + "loss": 1.9753, + "step": 411 + }, + { + "epoch": 0.4515068493150685, + "grad_norm": 2.3618829250335693, + "learning_rate": 4.74817221770918e-06, + "loss": 1.9397, + "step": 412 + }, + { + "epoch": 0.4526027397260274, + "grad_norm": 2.3922839164733887, + "learning_rate": 4.7461413484971565e-06, + "loss": 2.0119, + "step": 413 + }, + { + "epoch": 0.4536986301369863, + "grad_norm": 2.403475046157837, + "learning_rate": 4.744110479285135e-06, + "loss": 2.0664, + "step": 414 + }, + { + "epoch": 0.4547945205479452, + "grad_norm": 2.225860834121704, + "learning_rate": 4.742079610073111e-06, + "loss": 1.9525, + "step": 415 + }, + { + "epoch": 0.4558904109589041, + "grad_norm": 2.344074010848999, + "learning_rate": 4.740048740861089e-06, + "loss": 1.9785, + "step": 416 + }, + { + "epoch": 0.456986301369863, + "grad_norm": 2.255382776260376, + "learning_rate": 4.738017871649066e-06, + "loss": 1.9902, + "step": 417 + }, + { + "epoch": 0.45808219178082193, + "grad_norm": 2.160628318786621, + "learning_rate": 4.735987002437044e-06, + "loss": 2.0075, + "step": 418 + }, + { + "epoch": 0.4591780821917808, + "grad_norm": 2.5667307376861572, + "learning_rate": 4.73395613322502e-06, + "loss": 2.0625, + "step": 419 + }, + { + "epoch": 0.4602739726027397, + "grad_norm": 2.3771040439605713, + "learning_rate": 4.731925264012998e-06, + "loss": 2.0724, + "step": 420 + }, + { + "epoch": 0.4602739726027397, + "eval_loss": 2.138829231262207, + "eval_runtime": 37.4361, + "eval_samples_per_second": 26.712, + "eval_steps_per_second": 1.122, + "step": 420 + }, + { + "epoch": 0.46136986301369864, + "grad_norm": 2.336482048034668, + "learning_rate": 4.729894394800975e-06, + "loss": 2.0234, + "step": 421 + }, + { + "epoch": 0.4624657534246575, + "grad_norm": 2.4206933975219727, + "learning_rate": 4.727863525588953e-06, + "loss": 1.9693, + "step": 422 + }, + { + "epoch": 0.46356164383561643, + "grad_norm": 2.4845645427703857, + "learning_rate": 4.725832656376929e-06, + "loss": 1.9239, + "step": 423 + }, + { + "epoch": 0.46465753424657535, + "grad_norm": 2.5199646949768066, + "learning_rate": 4.723801787164907e-06, + "loss": 1.9801, + "step": 424 + }, + { + "epoch": 0.4657534246575342, + "grad_norm": 2.2969748973846436, + "learning_rate": 4.721770917952885e-06, + "loss": 2.0192, + "step": 425 + }, + { + "epoch": 0.46684931506849314, + "grad_norm": 2.423503875732422, + "learning_rate": 4.719740048740861e-06, + "loss": 1.9684, + "step": 426 + }, + { + "epoch": 0.46794520547945206, + "grad_norm": 2.3681061267852783, + "learning_rate": 4.717709179528839e-06, + "loss": 1.9934, + "step": 427 + }, + { + "epoch": 0.469041095890411, + "grad_norm": 2.277965784072876, + "learning_rate": 4.715678310316816e-06, + "loss": 2.0166, + "step": 428 + }, + { + "epoch": 0.47013698630136985, + "grad_norm": 2.2730283737182617, + "learning_rate": 4.713647441104794e-06, + "loss": 2.011, + "step": 429 + }, + { + "epoch": 0.4712328767123288, + "grad_norm": 2.6694860458374023, + "learning_rate": 4.71161657189277e-06, + "loss": 1.9645, + "step": 430 + }, + { + "epoch": 0.4712328767123288, + "eval_loss": 2.1322011947631836, + "eval_runtime": 37.5723, + "eval_samples_per_second": 26.615, + "eval_steps_per_second": 1.118, + "step": 430 + }, + { + "epoch": 0.4723287671232877, + "grad_norm": 2.4100632667541504, + "learning_rate": 4.709585702680748e-06, + "loss": 1.9478, + "step": 431 + }, + { + "epoch": 0.47342465753424656, + "grad_norm": 2.2779541015625, + "learning_rate": 4.707554833468725e-06, + "loss": 2.0274, + "step": 432 + }, + { + "epoch": 0.4745205479452055, + "grad_norm": 2.55690336227417, + "learning_rate": 4.705523964256703e-06, + "loss": 2.0831, + "step": 433 + }, + { + "epoch": 0.4756164383561644, + "grad_norm": 2.455623149871826, + "learning_rate": 4.703493095044679e-06, + "loss": 1.9673, + "step": 434 + }, + { + "epoch": 0.4767123287671233, + "grad_norm": 2.2613699436187744, + "learning_rate": 4.701462225832657e-06, + "loss": 1.992, + "step": 435 + }, + { + "epoch": 0.4778082191780822, + "grad_norm": 2.4467933177948, + "learning_rate": 4.699431356620634e-06, + "loss": 1.9105, + "step": 436 + }, + { + "epoch": 0.4789041095890411, + "grad_norm": 2.4194023609161377, + "learning_rate": 4.697400487408612e-06, + "loss": 2.074, + "step": 437 + }, + { + "epoch": 0.48, + "grad_norm": 2.2165699005126953, + "learning_rate": 4.695369618196588e-06, + "loss": 2.0152, + "step": 438 + }, + { + "epoch": 0.4810958904109589, + "grad_norm": 2.1709225177764893, + "learning_rate": 4.693338748984566e-06, + "loss": 1.9888, + "step": 439 + }, + { + "epoch": 0.4821917808219178, + "grad_norm": 2.2325828075408936, + "learning_rate": 4.691307879772543e-06, + "loss": 1.9838, + "step": 440 + }, + { + "epoch": 0.4821917808219178, + "eval_loss": 2.128732919692993, + "eval_runtime": 37.205, + "eval_samples_per_second": 26.878, + "eval_steps_per_second": 1.129, + "step": 440 + }, + { + "epoch": 0.4832876712328767, + "grad_norm": 2.446331739425659, + "learning_rate": 4.68927701056052e-06, + "loss": 1.9295, + "step": 441 + }, + { + "epoch": 0.4843835616438356, + "grad_norm": 2.5459415912628174, + "learning_rate": 4.6872461413484975e-06, + "loss": 2.0351, + "step": 442 + }, + { + "epoch": 0.48547945205479454, + "grad_norm": 2.2877118587493896, + "learning_rate": 4.685215272136475e-06, + "loss": 2.0, + "step": 443 + }, + { + "epoch": 0.4865753424657534, + "grad_norm": 2.346625328063965, + "learning_rate": 4.683184402924452e-06, + "loss": 1.9928, + "step": 444 + }, + { + "epoch": 0.4876712328767123, + "grad_norm": 2.268467664718628, + "learning_rate": 4.681153533712429e-06, + "loss": 1.9969, + "step": 445 + }, + { + "epoch": 0.48876712328767125, + "grad_norm": 2.3699793815612793, + "learning_rate": 4.679122664500407e-06, + "loss": 2.0636, + "step": 446 + }, + { + "epoch": 0.4898630136986301, + "grad_norm": 2.2871041297912598, + "learning_rate": 4.677091795288384e-06, + "loss": 1.9535, + "step": 447 + }, + { + "epoch": 0.49095890410958903, + "grad_norm": 2.249393939971924, + "learning_rate": 4.675060926076361e-06, + "loss": 1.9358, + "step": 448 + }, + { + "epoch": 0.49205479452054796, + "grad_norm": 2.299863815307617, + "learning_rate": 4.673030056864338e-06, + "loss": 2.0299, + "step": 449 + }, + { + "epoch": 0.4931506849315068, + "grad_norm": 2.31239652633667, + "learning_rate": 4.670999187652316e-06, + "loss": 1.9372, + "step": 450 + }, + { + "epoch": 0.4931506849315068, + "eval_loss": 2.1267426013946533, + "eval_runtime": 37.1884, + "eval_samples_per_second": 26.89, + "eval_steps_per_second": 1.129, + "step": 450 + }, + { + "epoch": 0.49424657534246574, + "grad_norm": 2.2824623584747314, + "learning_rate": 4.668968318440293e-06, + "loss": 1.9435, + "step": 451 + }, + { + "epoch": 0.49534246575342467, + "grad_norm": 2.285339593887329, + "learning_rate": 4.66693744922827e-06, + "loss": 1.9455, + "step": 452 + }, + { + "epoch": 0.4964383561643836, + "grad_norm": 2.346592903137207, + "learning_rate": 4.6649065800162475e-06, + "loss": 2.0212, + "step": 453 + }, + { + "epoch": 0.49753424657534245, + "grad_norm": 2.280669689178467, + "learning_rate": 4.662875710804225e-06, + "loss": 1.9136, + "step": 454 + }, + { + "epoch": 0.4986301369863014, + "grad_norm": 2.178434133529663, + "learning_rate": 4.660844841592202e-06, + "loss": 1.9952, + "step": 455 + }, + { + "epoch": 0.4997260273972603, + "grad_norm": 2.108492851257324, + "learning_rate": 4.658813972380179e-06, + "loss": 1.9022, + "step": 456 + }, + { + "epoch": 0.5008219178082192, + "grad_norm": 2.3862147331237793, + "learning_rate": 4.656783103168157e-06, + "loss": 2.0332, + "step": 457 + }, + { + "epoch": 0.5019178082191781, + "grad_norm": 2.2017860412597656, + "learning_rate": 4.654752233956134e-06, + "loss": 1.8998, + "step": 458 + }, + { + "epoch": 0.503013698630137, + "grad_norm": 2.222134828567505, + "learning_rate": 4.652721364744111e-06, + "loss": 1.9836, + "step": 459 + }, + { + "epoch": 0.5041095890410959, + "grad_norm": 2.2974510192871094, + "learning_rate": 4.6506904955320885e-06, + "loss": 1.9626, + "step": 460 + }, + { + "epoch": 0.5041095890410959, + "eval_loss": 2.1345443725585938, + "eval_runtime": 37.2785, + "eval_samples_per_second": 26.825, + "eval_steps_per_second": 1.127, + "step": 460 + }, + { + "epoch": 0.5052054794520547, + "grad_norm": 2.107480764389038, + "learning_rate": 4.648659626320066e-06, + "loss": 2.0061, + "step": 461 + }, + { + "epoch": 0.5063013698630137, + "grad_norm": 2.3541181087493896, + "learning_rate": 4.646628757108043e-06, + "loss": 2.0133, + "step": 462 + }, + { + "epoch": 0.5073972602739726, + "grad_norm": 2.215571880340576, + "learning_rate": 4.6445978878960194e-06, + "loss": 1.9727, + "step": 463 + }, + { + "epoch": 0.5084931506849315, + "grad_norm": 2.2263615131378174, + "learning_rate": 4.6425670186839976e-06, + "loss": 1.9477, + "step": 464 + }, + { + "epoch": 0.5095890410958904, + "grad_norm": 2.178900957107544, + "learning_rate": 4.640536149471974e-06, + "loss": 1.9842, + "step": 465 + }, + { + "epoch": 0.5106849315068493, + "grad_norm": 2.3703105449676514, + "learning_rate": 4.638505280259952e-06, + "loss": 2.0082, + "step": 466 + }, + { + "epoch": 0.5117808219178083, + "grad_norm": 2.1587533950805664, + "learning_rate": 4.6364744110479285e-06, + "loss": 2.0271, + "step": 467 + }, + { + "epoch": 0.5128767123287671, + "grad_norm": 2.1626017093658447, + "learning_rate": 4.634443541835907e-06, + "loss": 1.9858, + "step": 468 + }, + { + "epoch": 0.513972602739726, + "grad_norm": 2.326869487762451, + "learning_rate": 4.632412672623883e-06, + "loss": 2.0413, + "step": 469 + }, + { + "epoch": 0.5150684931506849, + "grad_norm": 2.266502857208252, + "learning_rate": 4.630381803411861e-06, + "loss": 1.9584, + "step": 470 + }, + { + "epoch": 0.5150684931506849, + "eval_loss": 2.1273915767669678, + "eval_runtime": 37.2571, + "eval_samples_per_second": 26.841, + "eval_steps_per_second": 1.127, + "step": 470 + }, + { + "epoch": 0.5161643835616438, + "grad_norm": 2.131348133087158, + "learning_rate": 4.628350934199838e-06, + "loss": 1.863, + "step": 471 + }, + { + "epoch": 0.5172602739726028, + "grad_norm": 2.2635257244110107, + "learning_rate": 4.626320064987816e-06, + "loss": 1.9699, + "step": 472 + }, + { + "epoch": 0.5183561643835617, + "grad_norm": 2.1303744316101074, + "learning_rate": 4.624289195775792e-06, + "loss": 2.0001, + "step": 473 + }, + { + "epoch": 0.5194520547945205, + "grad_norm": 2.1736788749694824, + "learning_rate": 4.62225832656377e-06, + "loss": 1.9684, + "step": 474 + }, + { + "epoch": 0.5205479452054794, + "grad_norm": 2.1996278762817383, + "learning_rate": 4.620227457351747e-06, + "loss": 1.9739, + "step": 475 + }, + { + "epoch": 0.5216438356164383, + "grad_norm": 2.225456953048706, + "learning_rate": 4.618196588139724e-06, + "loss": 1.9649, + "step": 476 + }, + { + "epoch": 0.5227397260273973, + "grad_norm": 2.118537187576294, + "learning_rate": 4.616165718927701e-06, + "loss": 1.9911, + "step": 477 + }, + { + "epoch": 0.5238356164383562, + "grad_norm": 2.232940435409546, + "learning_rate": 4.6141348497156786e-06, + "loss": 1.944, + "step": 478 + }, + { + "epoch": 0.5249315068493151, + "grad_norm": 2.1776034832000732, + "learning_rate": 4.612103980503656e-06, + "loss": 2.0342, + "step": 479 + }, + { + "epoch": 0.5260273972602739, + "grad_norm": 2.042645215988159, + "learning_rate": 4.610073111291633e-06, + "loss": 2.0124, + "step": 480 + }, + { + "epoch": 0.5260273972602739, + "eval_loss": 2.118508815765381, + "eval_runtime": 37.2764, + "eval_samples_per_second": 26.827, + "eval_steps_per_second": 1.127, + "step": 480 + }, + { + "epoch": 0.5271232876712328, + "grad_norm": 2.302661895751953, + "learning_rate": 4.60804224207961e-06, + "loss": 1.9874, + "step": 481 + }, + { + "epoch": 0.5282191780821918, + "grad_norm": 2.1251118183135986, + "learning_rate": 4.606011372867588e-06, + "loss": 1.9594, + "step": 482 + }, + { + "epoch": 0.5293150684931507, + "grad_norm": 2.157043933868408, + "learning_rate": 4.603980503655565e-06, + "loss": 1.9955, + "step": 483 + }, + { + "epoch": 0.5304109589041096, + "grad_norm": 2.291428804397583, + "learning_rate": 4.601949634443542e-06, + "loss": 1.9695, + "step": 484 + }, + { + "epoch": 0.5315068493150685, + "grad_norm": 2.2341537475585938, + "learning_rate": 4.5999187652315195e-06, + "loss": 1.899, + "step": 485 + }, + { + "epoch": 0.5326027397260275, + "grad_norm": 2.1881206035614014, + "learning_rate": 4.597887896019497e-06, + "loss": 1.9066, + "step": 486 + }, + { + "epoch": 0.5336986301369863, + "grad_norm": 2.3057658672332764, + "learning_rate": 4.595857026807474e-06, + "loss": 1.9646, + "step": 487 + }, + { + "epoch": 0.5347945205479452, + "grad_norm": 2.2318356037139893, + "learning_rate": 4.593826157595451e-06, + "loss": 1.9662, + "step": 488 + }, + { + "epoch": 0.5358904109589041, + "grad_norm": 2.170509099960327, + "learning_rate": 4.591795288383429e-06, + "loss": 1.9522, + "step": 489 + }, + { + "epoch": 0.536986301369863, + "grad_norm": 2.101646661758423, + "learning_rate": 4.589764419171406e-06, + "loss": 2.0315, + "step": 490 + }, + { + "epoch": 0.536986301369863, + "eval_loss": 2.113968849182129, + "eval_runtime": 37.2751, + "eval_samples_per_second": 26.828, + "eval_steps_per_second": 1.127, + "step": 490 + }, + { + "epoch": 0.538082191780822, + "grad_norm": 2.1563706398010254, + "learning_rate": 4.587733549959383e-06, + "loss": 1.9843, + "step": 491 + }, + { + "epoch": 0.5391780821917809, + "grad_norm": 2.295722246170044, + "learning_rate": 4.5857026807473604e-06, + "loss": 1.9499, + "step": 492 + }, + { + "epoch": 0.5402739726027397, + "grad_norm": 1.974424123764038, + "learning_rate": 4.583671811535338e-06, + "loss": 1.9346, + "step": 493 + }, + { + "epoch": 0.5413698630136986, + "grad_norm": 2.4402668476104736, + "learning_rate": 4.581640942323315e-06, + "loss": 1.9053, + "step": 494 + }, + { + "epoch": 0.5424657534246575, + "grad_norm": 2.269000768661499, + "learning_rate": 4.579610073111292e-06, + "loss": 2.0108, + "step": 495 + }, + { + "epoch": 0.5435616438356164, + "grad_norm": 2.1182594299316406, + "learning_rate": 4.5775792038992695e-06, + "loss": 2.0155, + "step": 496 + }, + { + "epoch": 0.5446575342465754, + "grad_norm": 2.2181780338287354, + "learning_rate": 4.575548334687247e-06, + "loss": 2.0112, + "step": 497 + }, + { + "epoch": 0.5457534246575343, + "grad_norm": 2.2806756496429443, + "learning_rate": 4.573517465475224e-06, + "loss": 2.0418, + "step": 498 + }, + { + "epoch": 0.5468493150684931, + "grad_norm": 2.2690632343292236, + "learning_rate": 4.571486596263201e-06, + "loss": 1.9457, + "step": 499 + }, + { + "epoch": 0.547945205479452, + "grad_norm": 2.361335515975952, + "learning_rate": 4.569455727051178e-06, + "loss": 1.9949, + "step": 500 + }, + { + "epoch": 0.547945205479452, + "eval_loss": 2.1165614128112793, + "eval_runtime": 37.2664, + "eval_samples_per_second": 26.834, + "eval_steps_per_second": 1.127, + "step": 500 + }, + { + "epoch": 0.549041095890411, + "grad_norm": 2.3833138942718506, + "learning_rate": 4.567424857839156e-06, + "loss": 1.9932, + "step": 501 + }, + { + "epoch": 0.5501369863013699, + "grad_norm": 2.328500509262085, + "learning_rate": 4.565393988627132e-06, + "loss": 1.9773, + "step": 502 + }, + { + "epoch": 0.5512328767123288, + "grad_norm": 2.4191694259643555, + "learning_rate": 4.5633631194151105e-06, + "loss": 2.0389, + "step": 503 + }, + { + "epoch": 0.5523287671232877, + "grad_norm": 2.4106884002685547, + "learning_rate": 4.561332250203087e-06, + "loss": 1.965, + "step": 504 + }, + { + "epoch": 0.5534246575342465, + "grad_norm": 2.1452605724334717, + "learning_rate": 4.559301380991065e-06, + "loss": 1.9713, + "step": 505 + }, + { + "epoch": 0.5545205479452054, + "grad_norm": 2.2753357887268066, + "learning_rate": 4.5572705117790414e-06, + "loss": 1.9116, + "step": 506 + }, + { + "epoch": 0.5556164383561644, + "grad_norm": 2.6925032138824463, + "learning_rate": 4.5552396425670196e-06, + "loss": 1.9593, + "step": 507 + }, + { + "epoch": 0.5567123287671233, + "grad_norm": 2.3136658668518066, + "learning_rate": 4.553208773354996e-06, + "loss": 2.0072, + "step": 508 + }, + { + "epoch": 0.5578082191780822, + "grad_norm": 2.254849672317505, + "learning_rate": 4.551177904142974e-06, + "loss": 1.9193, + "step": 509 + }, + { + "epoch": 0.5589041095890411, + "grad_norm": 2.352755546569824, + "learning_rate": 4.5491470349309505e-06, + "loss": 2.0228, + "step": 510 + }, + { + "epoch": 0.5589041095890411, + "eval_loss": 2.114323377609253, + "eval_runtime": 37.3629, + "eval_samples_per_second": 26.764, + "eval_steps_per_second": 1.124, + "step": 510 + }, + { + "epoch": 0.56, + "grad_norm": 2.31339955329895, + "learning_rate": 4.547116165718929e-06, + "loss": 1.9416, + "step": 511 + }, + { + "epoch": 0.5610958904109589, + "grad_norm": 2.191164970397949, + "learning_rate": 4.545085296506905e-06, + "loss": 1.9709, + "step": 512 + }, + { + "epoch": 0.5621917808219178, + "grad_norm": 2.193551778793335, + "learning_rate": 4.543054427294882e-06, + "loss": 1.9475, + "step": 513 + }, + { + "epoch": 0.5632876712328767, + "grad_norm": 2.222710371017456, + "learning_rate": 4.54102355808286e-06, + "loss": 1.9283, + "step": 514 + }, + { + "epoch": 0.5643835616438356, + "grad_norm": 2.078907012939453, + "learning_rate": 4.538992688870837e-06, + "loss": 1.929, + "step": 515 + }, + { + "epoch": 0.5654794520547946, + "grad_norm": 2.2043654918670654, + "learning_rate": 4.536961819658814e-06, + "loss": 1.9591, + "step": 516 + }, + { + "epoch": 0.5665753424657535, + "grad_norm": 2.192481517791748, + "learning_rate": 4.5349309504467915e-06, + "loss": 1.9618, + "step": 517 + }, + { + "epoch": 0.5676712328767123, + "grad_norm": 2.3509292602539062, + "learning_rate": 4.532900081234769e-06, + "loss": 1.9782, + "step": 518 + }, + { + "epoch": 0.5687671232876712, + "grad_norm": 2.2415924072265625, + "learning_rate": 4.530869212022746e-06, + "loss": 1.9456, + "step": 519 + }, + { + "epoch": 0.5698630136986301, + "grad_norm": 2.2135009765625, + "learning_rate": 4.528838342810723e-06, + "loss": 1.9671, + "step": 520 + }, + { + "epoch": 0.5698630136986301, + "eval_loss": 2.109738826751709, + "eval_runtime": 37.5253, + "eval_samples_per_second": 26.649, + "eval_steps_per_second": 1.119, + "step": 520 + }, + { + "epoch": 0.570958904109589, + "grad_norm": 2.3513574600219727, + "learning_rate": 4.526807473598701e-06, + "loss": 1.9625, + "step": 521 + }, + { + "epoch": 0.572054794520548, + "grad_norm": 1.9949424266815186, + "learning_rate": 4.524776604386678e-06, + "loss": 1.9621, + "step": 522 + }, + { + "epoch": 0.5731506849315069, + "grad_norm": 2.199488401412964, + "learning_rate": 4.522745735174655e-06, + "loss": 1.9389, + "step": 523 + }, + { + "epoch": 0.5742465753424657, + "grad_norm": 2.2119452953338623, + "learning_rate": 4.520714865962632e-06, + "loss": 1.9936, + "step": 524 + }, + { + "epoch": 0.5753424657534246, + "grad_norm": 2.242812156677246, + "learning_rate": 4.51868399675061e-06, + "loss": 1.9728, + "step": 525 + }, + { + "epoch": 0.5764383561643835, + "grad_norm": 2.1713850498199463, + "learning_rate": 4.516653127538587e-06, + "loss": 1.98, + "step": 526 + }, + { + "epoch": 0.5775342465753425, + "grad_norm": 2.261742353439331, + "learning_rate": 4.514622258326564e-06, + "loss": 1.9412, + "step": 527 + }, + { + "epoch": 0.5786301369863014, + "grad_norm": 2.222663402557373, + "learning_rate": 4.5125913891145415e-06, + "loss": 1.9266, + "step": 528 + }, + { + "epoch": 0.5797260273972603, + "grad_norm": 2.1108906269073486, + "learning_rate": 4.510560519902519e-06, + "loss": 1.9539, + "step": 529 + }, + { + "epoch": 0.5808219178082191, + "grad_norm": 1.9594205617904663, + "learning_rate": 4.508529650690496e-06, + "loss": 1.9842, + "step": 530 + }, + { + "epoch": 0.5808219178082191, + "eval_loss": 2.1073970794677734, + "eval_runtime": 37.6167, + "eval_samples_per_second": 26.584, + "eval_steps_per_second": 1.117, + "step": 530 + }, + { + "epoch": 0.581917808219178, + "grad_norm": 2.1339495182037354, + "learning_rate": 4.506498781478473e-06, + "loss": 1.9095, + "step": 531 + }, + { + "epoch": 0.583013698630137, + "grad_norm": 2.116248607635498, + "learning_rate": 4.504467912266451e-06, + "loss": 1.9091, + "step": 532 + }, + { + "epoch": 0.5841095890410959, + "grad_norm": 2.063962459564209, + "learning_rate": 4.502437043054428e-06, + "loss": 1.9537, + "step": 533 + }, + { + "epoch": 0.5852054794520548, + "grad_norm": 2.162569522857666, + "learning_rate": 4.500406173842405e-06, + "loss": 2.0253, + "step": 534 + }, + { + "epoch": 0.5863013698630137, + "grad_norm": 2.1783549785614014, + "learning_rate": 4.4983753046303824e-06, + "loss": 1.9714, + "step": 535 + }, + { + "epoch": 0.5873972602739727, + "grad_norm": 2.0791046619415283, + "learning_rate": 4.49634443541836e-06, + "loss": 1.9629, + "step": 536 + }, + { + "epoch": 0.5884931506849315, + "grad_norm": 2.3923251628875732, + "learning_rate": 4.494313566206336e-06, + "loss": 1.9445, + "step": 537 + }, + { + "epoch": 0.5895890410958904, + "grad_norm": 2.1848347187042236, + "learning_rate": 4.492282696994314e-06, + "loss": 2.007, + "step": 538 + }, + { + "epoch": 0.5906849315068493, + "grad_norm": 2.0824925899505615, + "learning_rate": 4.490251827782291e-06, + "loss": 1.9409, + "step": 539 + }, + { + "epoch": 0.5917808219178082, + "grad_norm": 2.057790756225586, + "learning_rate": 4.488220958570269e-06, + "loss": 1.9061, + "step": 540 + }, + { + "epoch": 0.5917808219178082, + "eval_loss": 2.1101415157318115, + "eval_runtime": 37.2235, + "eval_samples_per_second": 26.865, + "eval_steps_per_second": 1.128, + "step": 540 + }, + { + "epoch": 0.5928767123287672, + "grad_norm": 2.3652148246765137, + "learning_rate": 4.486190089358245e-06, + "loss": 1.9379, + "step": 541 + }, + { + "epoch": 0.5939726027397261, + "grad_norm": 2.195535659790039, + "learning_rate": 4.484159220146223e-06, + "loss": 1.8881, + "step": 542 + }, + { + "epoch": 0.5950684931506849, + "grad_norm": 2.258000135421753, + "learning_rate": 4.4821283509342e-06, + "loss": 1.9923, + "step": 543 + }, + { + "epoch": 0.5961643835616438, + "grad_norm": 2.097041606903076, + "learning_rate": 4.480097481722178e-06, + "loss": 1.9715, + "step": 544 + }, + { + "epoch": 0.5972602739726027, + "grad_norm": 2.1741156578063965, + "learning_rate": 4.478066612510154e-06, + "loss": 1.9487, + "step": 545 + }, + { + "epoch": 0.5983561643835617, + "grad_norm": 2.135338306427002, + "learning_rate": 4.4760357432981325e-06, + "loss": 1.9172, + "step": 546 + }, + { + "epoch": 0.5994520547945206, + "grad_norm": 2.3435418605804443, + "learning_rate": 4.474004874086109e-06, + "loss": 1.945, + "step": 547 + }, + { + "epoch": 0.6005479452054795, + "grad_norm": 2.2242488861083984, + "learning_rate": 4.471974004874087e-06, + "loss": 2.0717, + "step": 548 + }, + { + "epoch": 0.6016438356164383, + "grad_norm": 2.3371622562408447, + "learning_rate": 4.4699431356620635e-06, + "loss": 1.9082, + "step": 549 + }, + { + "epoch": 0.6027397260273972, + "grad_norm": 2.176064968109131, + "learning_rate": 4.467912266450041e-06, + "loss": 2.0154, + "step": 550 + }, + { + "epoch": 0.6027397260273972, + "eval_loss": 2.105236291885376, + "eval_runtime": 37.1885, + "eval_samples_per_second": 26.89, + "eval_steps_per_second": 1.129, + "step": 550 + }, + { + "epoch": 0.6038356164383561, + "grad_norm": 2.131166934967041, + "learning_rate": 4.465881397238018e-06, + "loss": 1.9575, + "step": 551 + }, + { + "epoch": 0.6049315068493151, + "grad_norm": 2.094514846801758, + "learning_rate": 4.463850528025995e-06, + "loss": 1.9544, + "step": 552 + }, + { + "epoch": 0.606027397260274, + "grad_norm": 2.0827765464782715, + "learning_rate": 4.4618196588139726e-06, + "loss": 1.9539, + "step": 553 + }, + { + "epoch": 0.6071232876712329, + "grad_norm": 2.225749969482422, + "learning_rate": 4.45978878960195e-06, + "loss": 1.983, + "step": 554 + }, + { + "epoch": 0.6082191780821918, + "grad_norm": 2.1363680362701416, + "learning_rate": 4.457757920389927e-06, + "loss": 1.9769, + "step": 555 + }, + { + "epoch": 0.6093150684931506, + "grad_norm": 2.238170623779297, + "learning_rate": 4.455727051177904e-06, + "loss": 1.9094, + "step": 556 + }, + { + "epoch": 0.6104109589041096, + "grad_norm": 2.228384494781494, + "learning_rate": 4.453696181965882e-06, + "loss": 1.9464, + "step": 557 + }, + { + "epoch": 0.6115068493150685, + "grad_norm": 2.479428768157959, + "learning_rate": 4.451665312753859e-06, + "loss": 2.0017, + "step": 558 + }, + { + "epoch": 0.6126027397260274, + "grad_norm": 2.2331902980804443, + "learning_rate": 4.449634443541836e-06, + "loss": 1.9728, + "step": 559 + }, + { + "epoch": 0.6136986301369863, + "grad_norm": 2.1728556156158447, + "learning_rate": 4.4476035743298135e-06, + "loss": 2.0153, + "step": 560 + }, + { + "epoch": 0.6136986301369863, + "eval_loss": 2.0978519916534424, + "eval_runtime": 37.233, + "eval_samples_per_second": 26.858, + "eval_steps_per_second": 1.128, + "step": 560 + }, + { + "epoch": 0.6147945205479453, + "grad_norm": 2.377814292907715, + "learning_rate": 4.445572705117791e-06, + "loss": 1.9242, + "step": 561 + }, + { + "epoch": 0.6158904109589041, + "grad_norm": 2.190436601638794, + "learning_rate": 4.443541835905768e-06, + "loss": 1.9454, + "step": 562 + }, + { + "epoch": 0.616986301369863, + "grad_norm": 2.052201986312866, + "learning_rate": 4.441510966693745e-06, + "loss": 2.0048, + "step": 563 + }, + { + "epoch": 0.6180821917808219, + "grad_norm": 2.2706589698791504, + "learning_rate": 4.439480097481723e-06, + "loss": 1.9426, + "step": 564 + }, + { + "epoch": 0.6191780821917808, + "grad_norm": 2.3478105068206787, + "learning_rate": 4.4374492282697e-06, + "loss": 2.0047, + "step": 565 + }, + { + "epoch": 0.6202739726027398, + "grad_norm": 2.465078115463257, + "learning_rate": 4.435418359057677e-06, + "loss": 1.8739, + "step": 566 + }, + { + "epoch": 0.6213698630136987, + "grad_norm": 2.0979490280151367, + "learning_rate": 4.433387489845654e-06, + "loss": 1.9588, + "step": 567 + }, + { + "epoch": 0.6224657534246575, + "grad_norm": 2.1239588260650635, + "learning_rate": 4.431356620633632e-06, + "loss": 1.9695, + "step": 568 + }, + { + "epoch": 0.6235616438356164, + "grad_norm": 2.014007806777954, + "learning_rate": 4.429325751421609e-06, + "loss": 1.9232, + "step": 569 + }, + { + "epoch": 0.6246575342465753, + "grad_norm": 2.212414026260376, + "learning_rate": 4.427294882209586e-06, + "loss": 1.928, + "step": 570 + }, + { + "epoch": 0.6246575342465753, + "eval_loss": 2.103149890899658, + "eval_runtime": 37.2641, + "eval_samples_per_second": 26.835, + "eval_steps_per_second": 1.127, + "step": 570 + }, + { + "epoch": 0.6257534246575343, + "grad_norm": 2.0944411754608154, + "learning_rate": 4.4252640129975635e-06, + "loss": 1.9568, + "step": 571 + }, + { + "epoch": 0.6268493150684932, + "grad_norm": 2.1068055629730225, + "learning_rate": 4.423233143785541e-06, + "loss": 1.8917, + "step": 572 + }, + { + "epoch": 0.6279452054794521, + "grad_norm": 2.1992151737213135, + "learning_rate": 4.421202274573518e-06, + "loss": 1.9536, + "step": 573 + }, + { + "epoch": 0.6290410958904109, + "grad_norm": 2.2079834938049316, + "learning_rate": 4.4191714053614945e-06, + "loss": 1.9603, + "step": 574 + }, + { + "epoch": 0.6301369863013698, + "grad_norm": 2.1230316162109375, + "learning_rate": 4.417140536149473e-06, + "loss": 1.8529, + "step": 575 + }, + { + "epoch": 0.6312328767123287, + "grad_norm": 2.1069579124450684, + "learning_rate": 4.415109666937449e-06, + "loss": 1.8997, + "step": 576 + }, + { + "epoch": 0.6323287671232877, + "grad_norm": 2.1683409214019775, + "learning_rate": 4.413078797725427e-06, + "loss": 1.971, + "step": 577 + }, + { + "epoch": 0.6334246575342466, + "grad_norm": 2.0599024295806885, + "learning_rate": 4.411047928513404e-06, + "loss": 1.9154, + "step": 578 + }, + { + "epoch": 0.6345205479452055, + "grad_norm": 2.1605849266052246, + "learning_rate": 4.409017059301382e-06, + "loss": 1.9562, + "step": 579 + }, + { + "epoch": 0.6356164383561644, + "grad_norm": 2.151249408721924, + "learning_rate": 4.406986190089358e-06, + "loss": 1.9571, + "step": 580 + }, + { + "epoch": 0.6356164383561644, + "eval_loss": 2.098177671432495, + "eval_runtime": 37.2578, + "eval_samples_per_second": 26.84, + "eval_steps_per_second": 1.127, + "step": 580 + }, + { + "epoch": 0.6367123287671232, + "grad_norm": 2.1819236278533936, + "learning_rate": 4.404955320877336e-06, + "loss": 1.9816, + "step": 581 + }, + { + "epoch": 0.6378082191780822, + "grad_norm": 2.1131205558776855, + "learning_rate": 4.402924451665313e-06, + "loss": 1.9738, + "step": 582 + }, + { + "epoch": 0.6389041095890411, + "grad_norm": 2.0852456092834473, + "learning_rate": 4.400893582453291e-06, + "loss": 1.924, + "step": 583 + }, + { + "epoch": 0.64, + "grad_norm": 2.0997941493988037, + "learning_rate": 4.398862713241267e-06, + "loss": 1.9067, + "step": 584 + }, + { + "epoch": 0.6410958904109589, + "grad_norm": 2.0848538875579834, + "learning_rate": 4.396831844029245e-06, + "loss": 1.9393, + "step": 585 + }, + { + "epoch": 0.6421917808219179, + "grad_norm": 2.2121758460998535, + "learning_rate": 4.394800974817222e-06, + "loss": 1.9458, + "step": 586 + }, + { + "epoch": 0.6432876712328767, + "grad_norm": 2.194462299346924, + "learning_rate": 4.392770105605199e-06, + "loss": 1.9528, + "step": 587 + }, + { + "epoch": 0.6443835616438356, + "grad_norm": 2.1936254501342773, + "learning_rate": 4.390739236393176e-06, + "loss": 1.9584, + "step": 588 + }, + { + "epoch": 0.6454794520547945, + "grad_norm": 2.1685783863067627, + "learning_rate": 4.388708367181154e-06, + "loss": 1.9666, + "step": 589 + }, + { + "epoch": 0.6465753424657534, + "grad_norm": 2.334571123123169, + "learning_rate": 4.386677497969131e-06, + "loss": 1.9748, + "step": 590 + }, + { + "epoch": 0.6465753424657534, + "eval_loss": 2.0937106609344482, + "eval_runtime": 37.2877, + "eval_samples_per_second": 26.818, + "eval_steps_per_second": 1.126, + "step": 590 + }, + { + "epoch": 0.6476712328767124, + "grad_norm": 2.2901110649108887, + "learning_rate": 4.384646628757108e-06, + "loss": 1.9597, + "step": 591 + }, + { + "epoch": 0.6487671232876713, + "grad_norm": 2.2364838123321533, + "learning_rate": 4.3826157595450855e-06, + "loss": 2.004, + "step": 592 + }, + { + "epoch": 0.6498630136986301, + "grad_norm": 2.1088364124298096, + "learning_rate": 4.380584890333063e-06, + "loss": 1.9319, + "step": 593 + }, + { + "epoch": 0.650958904109589, + "grad_norm": 2.164210319519043, + "learning_rate": 4.37855402112104e-06, + "loss": 1.8906, + "step": 594 + }, + { + "epoch": 0.6520547945205479, + "grad_norm": 2.205200672149658, + "learning_rate": 4.376523151909017e-06, + "loss": 1.8474, + "step": 595 + }, + { + "epoch": 0.6531506849315069, + "grad_norm": 2.239530086517334, + "learning_rate": 4.3744922826969946e-06, + "loss": 2.0033, + "step": 596 + }, + { + "epoch": 0.6542465753424658, + "grad_norm": 2.214747190475464, + "learning_rate": 4.372461413484972e-06, + "loss": 1.9258, + "step": 597 + }, + { + "epoch": 0.6553424657534247, + "grad_norm": 2.1746792793273926, + "learning_rate": 4.370430544272949e-06, + "loss": 1.9479, + "step": 598 + }, + { + "epoch": 0.6564383561643835, + "grad_norm": 2.194131851196289, + "learning_rate": 4.368399675060926e-06, + "loss": 1.9825, + "step": 599 + }, + { + "epoch": 0.6575342465753424, + "grad_norm": 2.3334717750549316, + "learning_rate": 4.366368805848904e-06, + "loss": 1.9831, + "step": 600 + }, + { + "epoch": 0.6575342465753424, + "eval_loss": 2.0881216526031494, + "eval_runtime": 37.3226, + "eval_samples_per_second": 26.793, + "eval_steps_per_second": 1.125, + "step": 600 + }, + { + "epoch": 0.6586301369863014, + "grad_norm": 2.269484758377075, + "learning_rate": 4.364337936636881e-06, + "loss": 1.9407, + "step": 601 + }, + { + "epoch": 0.6597260273972603, + "grad_norm": 2.470773935317993, + "learning_rate": 4.362307067424858e-06, + "loss": 1.8905, + "step": 602 + }, + { + "epoch": 0.6608219178082192, + "grad_norm": 2.485574960708618, + "learning_rate": 4.3602761982128355e-06, + "loss": 1.953, + "step": 603 + }, + { + "epoch": 0.6619178082191781, + "grad_norm": 2.2443618774414062, + "learning_rate": 4.358245329000813e-06, + "loss": 1.9244, + "step": 604 + }, + { + "epoch": 0.663013698630137, + "grad_norm": 2.039600372314453, + "learning_rate": 4.35621445978879e-06, + "loss": 1.9364, + "step": 605 + }, + { + "epoch": 0.6641095890410958, + "grad_norm": 2.093001127243042, + "learning_rate": 4.354183590576767e-06, + "loss": 1.9439, + "step": 606 + }, + { + "epoch": 0.6652054794520548, + "grad_norm": 2.3433260917663574, + "learning_rate": 4.352152721364745e-06, + "loss": 1.9322, + "step": 607 + }, + { + "epoch": 0.6663013698630137, + "grad_norm": 2.2403812408447266, + "learning_rate": 4.350121852152722e-06, + "loss": 1.8829, + "step": 608 + }, + { + "epoch": 0.6673972602739726, + "grad_norm": 2.1542811393737793, + "learning_rate": 4.348090982940699e-06, + "loss": 1.9549, + "step": 609 + }, + { + "epoch": 0.6684931506849315, + "grad_norm": 2.1526548862457275, + "learning_rate": 4.3460601137286764e-06, + "loss": 1.9433, + "step": 610 + }, + { + "epoch": 0.6684931506849315, + "eval_loss": 2.083638906478882, + "eval_runtime": 37.2677, + "eval_samples_per_second": 26.833, + "eval_steps_per_second": 1.127, + "step": 610 + }, + { + "epoch": 0.6695890410958905, + "grad_norm": 2.093421459197998, + "learning_rate": 4.344029244516654e-06, + "loss": 1.9643, + "step": 611 + }, + { + "epoch": 0.6706849315068493, + "grad_norm": 2.2205088138580322, + "learning_rate": 4.341998375304631e-06, + "loss": 1.9309, + "step": 612 + }, + { + "epoch": 0.6717808219178082, + "grad_norm": 2.059763193130493, + "learning_rate": 4.339967506092607e-06, + "loss": 2.0064, + "step": 613 + }, + { + "epoch": 0.6728767123287671, + "grad_norm": 2.0340707302093506, + "learning_rate": 4.3379366368805855e-06, + "loss": 1.9394, + "step": 614 + }, + { + "epoch": 0.673972602739726, + "grad_norm": 2.0716137886047363, + "learning_rate": 4.335905767668562e-06, + "loss": 1.9421, + "step": 615 + }, + { + "epoch": 0.675068493150685, + "grad_norm": 2.176809549331665, + "learning_rate": 4.33387489845654e-06, + "loss": 1.9261, + "step": 616 + }, + { + "epoch": 0.6761643835616439, + "grad_norm": 2.144385814666748, + "learning_rate": 4.3318440292445165e-06, + "loss": 1.9635, + "step": 617 + }, + { + "epoch": 0.6772602739726027, + "grad_norm": 2.0949580669403076, + "learning_rate": 4.329813160032495e-06, + "loss": 1.9417, + "step": 618 + }, + { + "epoch": 0.6783561643835616, + "grad_norm": 2.1727259159088135, + "learning_rate": 4.327782290820471e-06, + "loss": 1.9041, + "step": 619 + }, + { + "epoch": 0.6794520547945205, + "grad_norm": 2.012868881225586, + "learning_rate": 4.325751421608449e-06, + "loss": 1.8973, + "step": 620 + }, + { + "epoch": 0.6794520547945205, + "eval_loss": 2.0839755535125732, + "eval_runtime": 37.5302, + "eval_samples_per_second": 26.645, + "eval_steps_per_second": 1.119, + "step": 620 + }, + { + "epoch": 0.6805479452054795, + "grad_norm": 2.204153299331665, + "learning_rate": 4.323720552396426e-06, + "loss": 1.9719, + "step": 621 + }, + { + "epoch": 0.6816438356164384, + "grad_norm": 2.2421770095825195, + "learning_rate": 4.321689683184404e-06, + "loss": 1.9207, + "step": 622 + }, + { + "epoch": 0.6827397260273973, + "grad_norm": 2.2263410091400146, + "learning_rate": 4.31965881397238e-06, + "loss": 1.9146, + "step": 623 + }, + { + "epoch": 0.6838356164383562, + "grad_norm": 2.2015221118927, + "learning_rate": 4.3176279447603574e-06, + "loss": 1.9311, + "step": 624 + }, + { + "epoch": 0.684931506849315, + "grad_norm": 2.1459643840789795, + "learning_rate": 4.315597075548335e-06, + "loss": 1.9673, + "step": 625 + }, + { + "epoch": 0.686027397260274, + "grad_norm": 2.278928756713867, + "learning_rate": 4.313566206336312e-06, + "loss": 1.954, + "step": 626 + }, + { + "epoch": 0.6871232876712329, + "grad_norm": 2.05305552482605, + "learning_rate": 4.311535337124289e-06, + "loss": 1.9789, + "step": 627 + }, + { + "epoch": 0.6882191780821918, + "grad_norm": 2.3979289531707764, + "learning_rate": 4.3095044679122665e-06, + "loss": 1.926, + "step": 628 + }, + { + "epoch": 0.6893150684931507, + "grad_norm": 2.1635828018188477, + "learning_rate": 4.307473598700244e-06, + "loss": 1.9316, + "step": 629 + }, + { + "epoch": 0.6904109589041096, + "grad_norm": 2.237713098526001, + "learning_rate": 4.305442729488221e-06, + "loss": 1.9409, + "step": 630 + }, + { + "epoch": 0.6904109589041096, + "eval_loss": 2.084226131439209, + "eval_runtime": 37.6039, + "eval_samples_per_second": 26.593, + "eval_steps_per_second": 1.117, + "step": 630 + }, + { + "epoch": 0.6915068493150685, + "grad_norm": 2.157435894012451, + "learning_rate": 4.303411860276198e-06, + "loss": 1.8777, + "step": 631 + }, + { + "epoch": 0.6926027397260274, + "grad_norm": 2.20766019821167, + "learning_rate": 4.301380991064176e-06, + "loss": 1.9803, + "step": 632 + }, + { + "epoch": 0.6936986301369863, + "grad_norm": 2.173079490661621, + "learning_rate": 4.299350121852153e-06, + "loss": 1.9541, + "step": 633 + }, + { + "epoch": 0.6947945205479452, + "grad_norm": 1.9727106094360352, + "learning_rate": 4.29731925264013e-06, + "loss": 1.9766, + "step": 634 + }, + { + "epoch": 0.6958904109589041, + "grad_norm": 2.268538475036621, + "learning_rate": 4.2952883834281075e-06, + "loss": 1.9737, + "step": 635 + }, + { + "epoch": 0.6969863013698631, + "grad_norm": 2.2118406295776367, + "learning_rate": 4.293257514216085e-06, + "loss": 1.9291, + "step": 636 + }, + { + "epoch": 0.6980821917808219, + "grad_norm": 2.2939369678497314, + "learning_rate": 4.291226645004062e-06, + "loss": 1.8868, + "step": 637 + }, + { + "epoch": 0.6991780821917808, + "grad_norm": 2.4940083026885986, + "learning_rate": 4.289195775792039e-06, + "loss": 1.9694, + "step": 638 + }, + { + "epoch": 0.7002739726027397, + "grad_norm": 2.1988351345062256, + "learning_rate": 4.287164906580017e-06, + "loss": 1.9781, + "step": 639 + }, + { + "epoch": 0.7013698630136986, + "grad_norm": 2.099578619003296, + "learning_rate": 4.285134037367994e-06, + "loss": 1.9145, + "step": 640 + }, + { + "epoch": 0.7013698630136986, + "eval_loss": 2.07913875579834, + "eval_runtime": 37.3458, + "eval_samples_per_second": 26.777, + "eval_steps_per_second": 1.125, + "step": 640 + }, + { + "epoch": 0.7024657534246576, + "grad_norm": 2.2437307834625244, + "learning_rate": 4.283103168155971e-06, + "loss": 1.9075, + "step": 641 + }, + { + "epoch": 0.7035616438356165, + "grad_norm": 2.320880174636841, + "learning_rate": 4.281072298943948e-06, + "loss": 1.8929, + "step": 642 + }, + { + "epoch": 0.7046575342465753, + "grad_norm": 2.1174306869506836, + "learning_rate": 4.279041429731926e-06, + "loss": 1.8872, + "step": 643 + }, + { + "epoch": 0.7057534246575342, + "grad_norm": 2.2826361656188965, + "learning_rate": 4.277010560519903e-06, + "loss": 1.9287, + "step": 644 + }, + { + "epoch": 0.7068493150684931, + "grad_norm": 2.7543559074401855, + "learning_rate": 4.27497969130788e-06, + "loss": 1.9119, + "step": 645 + }, + { + "epoch": 0.707945205479452, + "grad_norm": 2.2582201957702637, + "learning_rate": 4.2729488220958575e-06, + "loss": 1.9118, + "step": 646 + }, + { + "epoch": 0.709041095890411, + "grad_norm": 2.3121371269226074, + "learning_rate": 4.270917952883835e-06, + "loss": 1.959, + "step": 647 + }, + { + "epoch": 0.7101369863013699, + "grad_norm": 2.419102430343628, + "learning_rate": 4.268887083671812e-06, + "loss": 1.9679, + "step": 648 + }, + { + "epoch": 0.7112328767123288, + "grad_norm": 2.285771369934082, + "learning_rate": 4.266856214459789e-06, + "loss": 1.8939, + "step": 649 + }, + { + "epoch": 0.7123287671232876, + "grad_norm": 2.193603754043579, + "learning_rate": 4.264825345247766e-06, + "loss": 2.0097, + "step": 650 + }, + { + "epoch": 0.7123287671232876, + "eval_loss": 2.0785164833068848, + "eval_runtime": 37.2595, + "eval_samples_per_second": 26.839, + "eval_steps_per_second": 1.127, + "step": 650 + }, + { + "epoch": 0.7134246575342466, + "grad_norm": 2.2645511627197266, + "learning_rate": 4.262794476035744e-06, + "loss": 1.8865, + "step": 651 + }, + { + "epoch": 0.7145205479452055, + "grad_norm": 2.5553910732269287, + "learning_rate": 4.26076360682372e-06, + "loss": 1.9331, + "step": 652 + }, + { + "epoch": 0.7156164383561644, + "grad_norm": 2.2770845890045166, + "learning_rate": 4.2587327376116984e-06, + "loss": 1.847, + "step": 653 + }, + { + "epoch": 0.7167123287671233, + "grad_norm": 2.1486358642578125, + "learning_rate": 4.256701868399675e-06, + "loss": 1.9058, + "step": 654 + }, + { + "epoch": 0.7178082191780822, + "grad_norm": 2.168975353240967, + "learning_rate": 4.254670999187653e-06, + "loss": 1.9408, + "step": 655 + }, + { + "epoch": 0.718904109589041, + "grad_norm": 2.2156167030334473, + "learning_rate": 4.252640129975629e-06, + "loss": 1.932, + "step": 656 + }, + { + "epoch": 0.72, + "grad_norm": 2.148214101791382, + "learning_rate": 4.2506092607636075e-06, + "loss": 1.9327, + "step": 657 + }, + { + "epoch": 0.7210958904109589, + "grad_norm": 2.3860998153686523, + "learning_rate": 4.248578391551584e-06, + "loss": 1.9191, + "step": 658 + }, + { + "epoch": 0.7221917808219178, + "grad_norm": 2.3917558193206787, + "learning_rate": 4.246547522339562e-06, + "loss": 1.8607, + "step": 659 + }, + { + "epoch": 0.7232876712328767, + "grad_norm": 2.3024795055389404, + "learning_rate": 4.2445166531275385e-06, + "loss": 1.9083, + "step": 660 + }, + { + "epoch": 0.7232876712328767, + "eval_loss": 2.07985782623291, + "eval_runtime": 37.1583, + "eval_samples_per_second": 26.912, + "eval_steps_per_second": 1.13, + "step": 660 + }, + { + "epoch": 0.7243835616438357, + "grad_norm": 2.261615037918091, + "learning_rate": 4.242485783915517e-06, + "loss": 1.9155, + "step": 661 + }, + { + "epoch": 0.7254794520547945, + "grad_norm": 2.190124034881592, + "learning_rate": 4.240454914703493e-06, + "loss": 1.858, + "step": 662 + }, + { + "epoch": 0.7265753424657534, + "grad_norm": 2.3396036624908447, + "learning_rate": 4.23842404549147e-06, + "loss": 1.878, + "step": 663 + }, + { + "epoch": 0.7276712328767123, + "grad_norm": 2.198410987854004, + "learning_rate": 4.236393176279448e-06, + "loss": 1.8755, + "step": 664 + }, + { + "epoch": 0.7287671232876712, + "grad_norm": 2.108875274658203, + "learning_rate": 4.234362307067425e-06, + "loss": 1.8709, + "step": 665 + }, + { + "epoch": 0.7298630136986302, + "grad_norm": 2.0908193588256836, + "learning_rate": 4.232331437855402e-06, + "loss": 1.9395, + "step": 666 + }, + { + "epoch": 0.7309589041095891, + "grad_norm": 2.208134651184082, + "learning_rate": 4.2303005686433795e-06, + "loss": 1.9675, + "step": 667 + }, + { + "epoch": 0.7320547945205479, + "grad_norm": 2.2779526710510254, + "learning_rate": 4.228269699431357e-06, + "loss": 1.9661, + "step": 668 + }, + { + "epoch": 0.7331506849315068, + "grad_norm": 2.053279161453247, + "learning_rate": 4.226238830219334e-06, + "loss": 1.9244, + "step": 669 + }, + { + "epoch": 0.7342465753424657, + "grad_norm": 2.2060821056365967, + "learning_rate": 4.224207961007311e-06, + "loss": 1.9252, + "step": 670 + }, + { + "epoch": 0.7342465753424657, + "eval_loss": 2.0804648399353027, + "eval_runtime": 37.1852, + "eval_samples_per_second": 26.892, + "eval_steps_per_second": 1.129, + "step": 670 + }, + { + "epoch": 0.7353424657534247, + "grad_norm": 2.1996090412139893, + "learning_rate": 4.2221770917952886e-06, + "loss": 1.926, + "step": 671 + }, + { + "epoch": 0.7364383561643836, + "grad_norm": 2.0817034244537354, + "learning_rate": 4.220146222583266e-06, + "loss": 1.9123, + "step": 672 + }, + { + "epoch": 0.7375342465753425, + "grad_norm": 2.2505035400390625, + "learning_rate": 4.218115353371243e-06, + "loss": 1.9369, + "step": 673 + }, + { + "epoch": 0.7386301369863014, + "grad_norm": 2.2053000926971436, + "learning_rate": 4.21608448415922e-06, + "loss": 1.9225, + "step": 674 + }, + { + "epoch": 0.7397260273972602, + "grad_norm": 2.039905071258545, + "learning_rate": 4.214053614947198e-06, + "loss": 1.8876, + "step": 675 + }, + { + "epoch": 0.7408219178082192, + "grad_norm": 2.047445297241211, + "learning_rate": 4.212022745735175e-06, + "loss": 1.9724, + "step": 676 + }, + { + "epoch": 0.7419178082191781, + "grad_norm": 2.277803897857666, + "learning_rate": 4.209991876523152e-06, + "loss": 1.8687, + "step": 677 + }, + { + "epoch": 0.743013698630137, + "grad_norm": 2.0048105716705322, + "learning_rate": 4.2079610073111295e-06, + "loss": 1.9349, + "step": 678 + }, + { + "epoch": 0.7441095890410959, + "grad_norm": 2.121614456176758, + "learning_rate": 4.205930138099107e-06, + "loss": 1.8723, + "step": 679 + }, + { + "epoch": 0.7452054794520548, + "grad_norm": 2.169165849685669, + "learning_rate": 4.203899268887084e-06, + "loss": 1.9713, + "step": 680 + }, + { + "epoch": 0.7452054794520548, + "eval_loss": 2.0704824924468994, + "eval_runtime": 37.2015, + "eval_samples_per_second": 26.881, + "eval_steps_per_second": 1.129, + "step": 680 + }, + { + "epoch": 0.7463013698630137, + "grad_norm": 2.004974603652954, + "learning_rate": 4.201868399675061e-06, + "loss": 1.9237, + "step": 681 + }, + { + "epoch": 0.7473972602739726, + "grad_norm": 2.042402505874634, + "learning_rate": 4.199837530463039e-06, + "loss": 1.9686, + "step": 682 + }, + { + "epoch": 0.7484931506849315, + "grad_norm": 2.152040719985962, + "learning_rate": 4.197806661251016e-06, + "loss": 1.8498, + "step": 683 + }, + { + "epoch": 0.7495890410958904, + "grad_norm": 2.116483211517334, + "learning_rate": 4.195775792038993e-06, + "loss": 1.9241, + "step": 684 + }, + { + "epoch": 0.7506849315068493, + "grad_norm": 2.1628384590148926, + "learning_rate": 4.19374492282697e-06, + "loss": 1.9019, + "step": 685 + }, + { + "epoch": 0.7517808219178083, + "grad_norm": 2.1851415634155273, + "learning_rate": 4.191714053614948e-06, + "loss": 1.836, + "step": 686 + }, + { + "epoch": 0.7528767123287671, + "grad_norm": 2.173215866088867, + "learning_rate": 4.189683184402924e-06, + "loss": 1.9412, + "step": 687 + }, + { + "epoch": 0.753972602739726, + "grad_norm": 2.1002466678619385, + "learning_rate": 4.187652315190902e-06, + "loss": 1.8871, + "step": 688 + }, + { + "epoch": 0.7550684931506849, + "grad_norm": 2.1932122707366943, + "learning_rate": 4.185621445978879e-06, + "loss": 1.9671, + "step": 689 + }, + { + "epoch": 0.7561643835616438, + "grad_norm": 2.2915258407592773, + "learning_rate": 4.183590576766857e-06, + "loss": 1.8957, + "step": 690 + }, + { + "epoch": 0.7561643835616438, + "eval_loss": 2.067124128341675, + "eval_runtime": 37.2362, + "eval_samples_per_second": 26.856, + "eval_steps_per_second": 1.128, + "step": 690 + }, + { + "epoch": 0.7572602739726028, + "grad_norm": 2.060879707336426, + "learning_rate": 4.181559707554833e-06, + "loss": 1.9091, + "step": 691 + }, + { + "epoch": 0.7583561643835617, + "grad_norm": 2.031156539916992, + "learning_rate": 4.179528838342811e-06, + "loss": 1.9606, + "step": 692 + }, + { + "epoch": 0.7594520547945205, + "grad_norm": 2.121628999710083, + "learning_rate": 4.177497969130788e-06, + "loss": 1.88, + "step": 693 + }, + { + "epoch": 0.7605479452054794, + "grad_norm": 2.2009618282318115, + "learning_rate": 4.175467099918766e-06, + "loss": 1.8916, + "step": 694 + }, + { + "epoch": 0.7616438356164383, + "grad_norm": 2.124082326889038, + "learning_rate": 4.173436230706742e-06, + "loss": 1.9805, + "step": 695 + }, + { + "epoch": 0.7627397260273973, + "grad_norm": 2.3914906978607178, + "learning_rate": 4.1714053614947205e-06, + "loss": 1.9052, + "step": 696 + }, + { + "epoch": 0.7638356164383562, + "grad_norm": 2.2398853302001953, + "learning_rate": 4.169374492282697e-06, + "loss": 1.8965, + "step": 697 + }, + { + "epoch": 0.7649315068493151, + "grad_norm": 2.224860191345215, + "learning_rate": 4.167343623070675e-06, + "loss": 1.8962, + "step": 698 + }, + { + "epoch": 0.766027397260274, + "grad_norm": 2.3356244564056396, + "learning_rate": 4.165312753858652e-06, + "loss": 1.9543, + "step": 699 + }, + { + "epoch": 0.7671232876712328, + "grad_norm": 2.1308751106262207, + "learning_rate": 4.163281884646629e-06, + "loss": 1.9095, + "step": 700 + }, + { + "epoch": 0.7671232876712328, + "eval_loss": 2.066145896911621, + "eval_runtime": 37.2629, + "eval_samples_per_second": 26.836, + "eval_steps_per_second": 1.127, + "step": 700 + }, + { + "epoch": 0.7682191780821918, + "grad_norm": 2.1655938625335693, + "learning_rate": 4.161251015434607e-06, + "loss": 1.8923, + "step": 701 + }, + { + "epoch": 0.7693150684931507, + "grad_norm": 2.1985316276550293, + "learning_rate": 4.159220146222583e-06, + "loss": 1.9216, + "step": 702 + }, + { + "epoch": 0.7704109589041096, + "grad_norm": 2.1537251472473145, + "learning_rate": 4.157189277010561e-06, + "loss": 1.879, + "step": 703 + }, + { + "epoch": 0.7715068493150685, + "grad_norm": 2.1867849826812744, + "learning_rate": 4.155158407798538e-06, + "loss": 1.905, + "step": 704 + }, + { + "epoch": 0.7726027397260274, + "grad_norm": 2.342320203781128, + "learning_rate": 4.153127538586516e-06, + "loss": 1.9403, + "step": 705 + }, + { + "epoch": 0.7736986301369863, + "grad_norm": 2.2250113487243652, + "learning_rate": 4.151096669374492e-06, + "loss": 1.9109, + "step": 706 + }, + { + "epoch": 0.7747945205479452, + "grad_norm": 2.179103374481201, + "learning_rate": 4.1490658001624705e-06, + "loss": 1.9288, + "step": 707 + }, + { + "epoch": 0.7758904109589041, + "grad_norm": 2.3618645668029785, + "learning_rate": 4.147034930950447e-06, + "loss": 1.8982, + "step": 708 + }, + { + "epoch": 0.776986301369863, + "grad_norm": 2.150402307510376, + "learning_rate": 4.145004061738425e-06, + "loss": 1.9, + "step": 709 + }, + { + "epoch": 0.7780821917808219, + "grad_norm": 2.295571804046631, + "learning_rate": 4.1429731925264015e-06, + "loss": 1.9003, + "step": 710 + }, + { + "epoch": 0.7780821917808219, + "eval_loss": 2.067948341369629, + "eval_runtime": 37.2379, + "eval_samples_per_second": 26.854, + "eval_steps_per_second": 1.128, + "step": 710 + }, + { + "epoch": 0.7791780821917809, + "grad_norm": 2.3279545307159424, + "learning_rate": 4.140942323314379e-06, + "loss": 1.8307, + "step": 711 + }, + { + "epoch": 0.7802739726027397, + "grad_norm": 2.3993380069732666, + "learning_rate": 4.138911454102356e-06, + "loss": 1.8755, + "step": 712 + }, + { + "epoch": 0.7813698630136986, + "grad_norm": 2.139653444290161, + "learning_rate": 4.136880584890333e-06, + "loss": 1.8914, + "step": 713 + }, + { + "epoch": 0.7824657534246575, + "grad_norm": 2.239544153213501, + "learning_rate": 4.1348497156783106e-06, + "loss": 1.9289, + "step": 714 + }, + { + "epoch": 0.7835616438356164, + "grad_norm": 2.2039642333984375, + "learning_rate": 4.132818846466288e-06, + "loss": 1.9237, + "step": 715 + }, + { + "epoch": 0.7846575342465754, + "grad_norm": 2.1453804969787598, + "learning_rate": 4.130787977254265e-06, + "loss": 1.9291, + "step": 716 + }, + { + "epoch": 0.7857534246575343, + "grad_norm": 2.2350032329559326, + "learning_rate": 4.128757108042242e-06, + "loss": 1.9413, + "step": 717 + }, + { + "epoch": 0.7868493150684932, + "grad_norm": 2.1718318462371826, + "learning_rate": 4.12672623883022e-06, + "loss": 1.9318, + "step": 718 + }, + { + "epoch": 0.787945205479452, + "grad_norm": 2.1860601902008057, + "learning_rate": 4.124695369618197e-06, + "loss": 1.9934, + "step": 719 + }, + { + "epoch": 0.7890410958904109, + "grad_norm": 2.205166816711426, + "learning_rate": 4.122664500406174e-06, + "loss": 1.9038, + "step": 720 + }, + { + "epoch": 0.7890410958904109, + "eval_loss": 2.06229829788208, + "eval_runtime": 37.6121, + "eval_samples_per_second": 26.587, + "eval_steps_per_second": 1.117, + "step": 720 + }, + { + "epoch": 0.7901369863013699, + "grad_norm": 2.2094533443450928, + "learning_rate": 4.1206336311941515e-06, + "loss": 1.906, + "step": 721 + }, + { + "epoch": 0.7912328767123288, + "grad_norm": 2.178891658782959, + "learning_rate": 4.118602761982129e-06, + "loss": 1.936, + "step": 722 + }, + { + "epoch": 0.7923287671232877, + "grad_norm": 2.051077365875244, + "learning_rate": 4.116571892770106e-06, + "loss": 1.8736, + "step": 723 + }, + { + "epoch": 0.7934246575342466, + "grad_norm": 2.2430427074432373, + "learning_rate": 4.114541023558083e-06, + "loss": 1.8396, + "step": 724 + }, + { + "epoch": 0.7945205479452054, + "grad_norm": 2.2255303859710693, + "learning_rate": 4.112510154346061e-06, + "loss": 1.9668, + "step": 725 + }, + { + "epoch": 0.7956164383561644, + "grad_norm": 2.280313491821289, + "learning_rate": 4.110479285134038e-06, + "loss": 1.8945, + "step": 726 + }, + { + "epoch": 0.7967123287671233, + "grad_norm": 2.2516801357269287, + "learning_rate": 4.108448415922015e-06, + "loss": 1.9157, + "step": 727 + }, + { + "epoch": 0.7978082191780822, + "grad_norm": 2.1379809379577637, + "learning_rate": 4.1064175467099924e-06, + "loss": 1.935, + "step": 728 + }, + { + "epoch": 0.7989041095890411, + "grad_norm": 2.149425983428955, + "learning_rate": 4.10438667749797e-06, + "loss": 1.935, + "step": 729 + }, + { + "epoch": 0.8, + "grad_norm": 2.275270700454712, + "learning_rate": 4.102355808285947e-06, + "loss": 1.7889, + "step": 730 + }, + { + "epoch": 0.8, + "eval_loss": 2.0633974075317383, + "eval_runtime": 37.6895, + "eval_samples_per_second": 26.533, + "eval_steps_per_second": 1.114, + "step": 730 + }, + { + "epoch": 0.8010958904109589, + "grad_norm": 2.1550278663635254, + "learning_rate": 4.100324939073924e-06, + "loss": 1.9833, + "step": 731 + }, + { + "epoch": 0.8021917808219178, + "grad_norm": 2.3050060272216797, + "learning_rate": 4.0982940698619015e-06, + "loss": 1.8969, + "step": 732 + }, + { + "epoch": 0.8032876712328767, + "grad_norm": 2.4548778533935547, + "learning_rate": 4.096263200649879e-06, + "loss": 1.8945, + "step": 733 + }, + { + "epoch": 0.8043835616438356, + "grad_norm": 2.2611639499664307, + "learning_rate": 4.094232331437856e-06, + "loss": 1.9071, + "step": 734 + }, + { + "epoch": 0.8054794520547945, + "grad_norm": 2.1806774139404297, + "learning_rate": 4.092201462225833e-06, + "loss": 1.8787, + "step": 735 + }, + { + "epoch": 0.8065753424657535, + "grad_norm": 2.4374301433563232, + "learning_rate": 4.090170593013811e-06, + "loss": 1.88, + "step": 736 + }, + { + "epoch": 0.8076712328767123, + "grad_norm": 2.328333854675293, + "learning_rate": 4.088139723801787e-06, + "loss": 1.9178, + "step": 737 + }, + { + "epoch": 0.8087671232876712, + "grad_norm": 2.33552885055542, + "learning_rate": 4.086108854589765e-06, + "loss": 1.8587, + "step": 738 + }, + { + "epoch": 0.8098630136986301, + "grad_norm": 2.3472790718078613, + "learning_rate": 4.084077985377742e-06, + "loss": 1.877, + "step": 739 + }, + { + "epoch": 0.810958904109589, + "grad_norm": 2.307123899459839, + "learning_rate": 4.08204711616572e-06, + "loss": 1.8888, + "step": 740 + }, + { + "epoch": 0.810958904109589, + "eval_loss": 2.065826892852783, + "eval_runtime": 37.2747, + "eval_samples_per_second": 26.828, + "eval_steps_per_second": 1.127, + "step": 740 + }, + { + "epoch": 0.812054794520548, + "grad_norm": 2.3688747882843018, + "learning_rate": 4.080016246953696e-06, + "loss": 1.9571, + "step": 741 + }, + { + "epoch": 0.8131506849315069, + "grad_norm": 2.385007381439209, + "learning_rate": 4.077985377741674e-06, + "loss": 1.8657, + "step": 742 + }, + { + "epoch": 0.8142465753424658, + "grad_norm": 2.156766414642334, + "learning_rate": 4.075954508529651e-06, + "loss": 1.9421, + "step": 743 + }, + { + "epoch": 0.8153424657534246, + "grad_norm": 2.2074286937713623, + "learning_rate": 4.073923639317629e-06, + "loss": 1.9396, + "step": 744 + }, + { + "epoch": 0.8164383561643835, + "grad_norm": 2.185458183288574, + "learning_rate": 4.071892770105605e-06, + "loss": 1.9612, + "step": 745 + }, + { + "epoch": 0.8175342465753425, + "grad_norm": 2.324209451675415, + "learning_rate": 4.069861900893583e-06, + "loss": 1.892, + "step": 746 + }, + { + "epoch": 0.8186301369863014, + "grad_norm": 2.1312127113342285, + "learning_rate": 4.06783103168156e-06, + "loss": 1.8522, + "step": 747 + }, + { + "epoch": 0.8197260273972603, + "grad_norm": 2.0959460735321045, + "learning_rate": 4.065800162469538e-06, + "loss": 1.8732, + "step": 748 + }, + { + "epoch": 0.8208219178082192, + "grad_norm": 2.085519552230835, + "learning_rate": 4.063769293257514e-06, + "loss": 1.8832, + "step": 749 + }, + { + "epoch": 0.821917808219178, + "grad_norm": 2.310145854949951, + "learning_rate": 4.061738424045492e-06, + "loss": 1.919, + "step": 750 + }, + { + "epoch": 0.821917808219178, + "eval_loss": 2.0637967586517334, + "eval_runtime": 37.2246, + "eval_samples_per_second": 26.864, + "eval_steps_per_second": 1.128, + "step": 750 + }, + { + "epoch": 0.823013698630137, + "grad_norm": 2.3171160221099854, + "learning_rate": 4.059707554833469e-06, + "loss": 1.8874, + "step": 751 + }, + { + "epoch": 0.8241095890410959, + "grad_norm": 2.1065480709075928, + "learning_rate": 4.057676685621446e-06, + "loss": 1.9064, + "step": 752 + }, + { + "epoch": 0.8252054794520548, + "grad_norm": 2.0539677143096924, + "learning_rate": 4.0556458164094235e-06, + "loss": 1.9376, + "step": 753 + }, + { + "epoch": 0.8263013698630137, + "grad_norm": 2.4440360069274902, + "learning_rate": 4.053614947197401e-06, + "loss": 1.9129, + "step": 754 + }, + { + "epoch": 0.8273972602739726, + "grad_norm": 2.2324604988098145, + "learning_rate": 4.051584077985378e-06, + "loss": 1.8281, + "step": 755 + }, + { + "epoch": 0.8284931506849315, + "grad_norm": 2.262627363204956, + "learning_rate": 4.049553208773355e-06, + "loss": 1.9036, + "step": 756 + }, + { + "epoch": 0.8295890410958904, + "grad_norm": 2.2473418712615967, + "learning_rate": 4.047522339561333e-06, + "loss": 1.9455, + "step": 757 + }, + { + "epoch": 0.8306849315068493, + "grad_norm": 2.473444938659668, + "learning_rate": 4.04549147034931e-06, + "loss": 1.9003, + "step": 758 + }, + { + "epoch": 0.8317808219178082, + "grad_norm": 2.3454110622406006, + "learning_rate": 4.043460601137287e-06, + "loss": 1.8941, + "step": 759 + }, + { + "epoch": 0.8328767123287671, + "grad_norm": 2.4391231536865234, + "learning_rate": 4.041429731925264e-06, + "loss": 2.0146, + "step": 760 + }, + { + "epoch": 0.8328767123287671, + "eval_loss": 2.0618460178375244, + "eval_runtime": 37.1704, + "eval_samples_per_second": 26.903, + "eval_steps_per_second": 1.13, + "step": 760 + }, + { + "epoch": 0.8339726027397261, + "grad_norm": 2.3129498958587646, + "learning_rate": 4.039398862713242e-06, + "loss": 1.9, + "step": 761 + }, + { + "epoch": 0.8350684931506849, + "grad_norm": 2.3766934871673584, + "learning_rate": 4.037367993501219e-06, + "loss": 1.9017, + "step": 762 + }, + { + "epoch": 0.8361643835616438, + "grad_norm": 2.1698460578918457, + "learning_rate": 4.035337124289196e-06, + "loss": 1.9657, + "step": 763 + }, + { + "epoch": 0.8372602739726027, + "grad_norm": 2.384904146194458, + "learning_rate": 4.0333062550771735e-06, + "loss": 1.8867, + "step": 764 + }, + { + "epoch": 0.8383561643835616, + "grad_norm": 2.572391986846924, + "learning_rate": 4.031275385865151e-06, + "loss": 1.8561, + "step": 765 + }, + { + "epoch": 0.8394520547945206, + "grad_norm": 2.216606616973877, + "learning_rate": 4.029244516653128e-06, + "loss": 1.8926, + "step": 766 + }, + { + "epoch": 0.8405479452054795, + "grad_norm": 2.215756893157959, + "learning_rate": 4.027213647441105e-06, + "loss": 1.928, + "step": 767 + }, + { + "epoch": 0.8416438356164384, + "grad_norm": 2.2713284492492676, + "learning_rate": 4.025182778229083e-06, + "loss": 1.8286, + "step": 768 + }, + { + "epoch": 0.8427397260273972, + "grad_norm": 2.341747283935547, + "learning_rate": 4.02315190901706e-06, + "loss": 1.8789, + "step": 769 + }, + { + "epoch": 0.8438356164383561, + "grad_norm": 2.1485848426818848, + "learning_rate": 4.021121039805037e-06, + "loss": 1.8998, + "step": 770 + }, + { + "epoch": 0.8438356164383561, + "eval_loss": 2.0596234798431396, + "eval_runtime": 37.1854, + "eval_samples_per_second": 26.892, + "eval_steps_per_second": 1.129, + "step": 770 + }, + { + "epoch": 0.8449315068493151, + "grad_norm": 2.234192132949829, + "learning_rate": 4.0190901705930144e-06, + "loss": 1.8995, + "step": 771 + }, + { + "epoch": 0.846027397260274, + "grad_norm": 2.289328098297119, + "learning_rate": 4.017059301380992e-06, + "loss": 1.9882, + "step": 772 + }, + { + "epoch": 0.8471232876712329, + "grad_norm": 2.1001455783843994, + "learning_rate": 4.015028432168969e-06, + "loss": 1.8706, + "step": 773 + }, + { + "epoch": 0.8482191780821918, + "grad_norm": 2.2200522422790527, + "learning_rate": 4.012997562956945e-06, + "loss": 1.9443, + "step": 774 + }, + { + "epoch": 0.8493150684931506, + "grad_norm": 2.2540805339813232, + "learning_rate": 4.0109666937449235e-06, + "loss": 1.8552, + "step": 775 + }, + { + "epoch": 0.8504109589041096, + "grad_norm": 2.1214053630828857, + "learning_rate": 4.0089358245329e-06, + "loss": 1.8741, + "step": 776 + }, + { + "epoch": 0.8515068493150685, + "grad_norm": 2.0221869945526123, + "learning_rate": 4.006904955320878e-06, + "loss": 1.8798, + "step": 777 + }, + { + "epoch": 0.8526027397260274, + "grad_norm": 2.217261791229248, + "learning_rate": 4.0048740861088545e-06, + "loss": 1.8542, + "step": 778 + }, + { + "epoch": 0.8536986301369863, + "grad_norm": 2.099202871322632, + "learning_rate": 4.002843216896833e-06, + "loss": 1.9342, + "step": 779 + }, + { + "epoch": 0.8547945205479452, + "grad_norm": 2.139526844024658, + "learning_rate": 4.000812347684809e-06, + "loss": 1.9252, + "step": 780 + }, + { + "epoch": 0.8547945205479452, + "eval_loss": 2.054051399230957, + "eval_runtime": 37.1125, + "eval_samples_per_second": 26.945, + "eval_steps_per_second": 1.132, + "step": 780 + }, + { + "epoch": 0.8558904109589041, + "grad_norm": 2.2231476306915283, + "learning_rate": 3.998781478472787e-06, + "loss": 1.8671, + "step": 781 + }, + { + "epoch": 0.856986301369863, + "grad_norm": 2.143979072570801, + "learning_rate": 3.996750609260764e-06, + "loss": 1.8759, + "step": 782 + }, + { + "epoch": 0.8580821917808219, + "grad_norm": 2.1353135108947754, + "learning_rate": 3.994719740048742e-06, + "loss": 1.8559, + "step": 783 + }, + { + "epoch": 0.8591780821917808, + "grad_norm": 2.222564697265625, + "learning_rate": 3.992688870836718e-06, + "loss": 1.999, + "step": 784 + }, + { + "epoch": 0.8602739726027397, + "grad_norm": 1.958031177520752, + "learning_rate": 3.990658001624696e-06, + "loss": 1.9019, + "step": 785 + }, + { + "epoch": 0.8613698630136987, + "grad_norm": 2.2048962116241455, + "learning_rate": 3.988627132412673e-06, + "loss": 1.8561, + "step": 786 + }, + { + "epoch": 0.8624657534246576, + "grad_norm": 2.150332450866699, + "learning_rate": 3.98659626320065e-06, + "loss": 1.9399, + "step": 787 + }, + { + "epoch": 0.8635616438356164, + "grad_norm": 2.182742118835449, + "learning_rate": 3.984565393988627e-06, + "loss": 1.8933, + "step": 788 + }, + { + "epoch": 0.8646575342465753, + "grad_norm": 2.19435715675354, + "learning_rate": 3.9825345247766046e-06, + "loss": 1.8918, + "step": 789 + }, + { + "epoch": 0.8657534246575342, + "grad_norm": 2.3382859230041504, + "learning_rate": 3.980503655564582e-06, + "loss": 1.8937, + "step": 790 + }, + { + "epoch": 0.8657534246575342, + "eval_loss": 2.051997423171997, + "eval_runtime": 37.1866, + "eval_samples_per_second": 26.891, + "eval_steps_per_second": 1.129, + "step": 790 + }, + { + "epoch": 0.8668493150684932, + "grad_norm": 2.2700726985931396, + "learning_rate": 3.978472786352559e-06, + "loss": 1.9148, + "step": 791 + }, + { + "epoch": 0.8679452054794521, + "grad_norm": 2.2998759746551514, + "learning_rate": 3.976441917140536e-06, + "loss": 1.9268, + "step": 792 + }, + { + "epoch": 0.869041095890411, + "grad_norm": 2.2367377281188965, + "learning_rate": 3.974411047928514e-06, + "loss": 1.901, + "step": 793 + }, + { + "epoch": 0.8701369863013698, + "grad_norm": 2.0868773460388184, + "learning_rate": 3.972380178716491e-06, + "loss": 1.9677, + "step": 794 + }, + { + "epoch": 0.8712328767123287, + "grad_norm": 2.183732509613037, + "learning_rate": 3.970349309504468e-06, + "loss": 1.9271, + "step": 795 + }, + { + "epoch": 0.8723287671232877, + "grad_norm": 2.187282085418701, + "learning_rate": 3.9683184402924455e-06, + "loss": 1.9437, + "step": 796 + }, + { + "epoch": 0.8734246575342466, + "grad_norm": 2.1906070709228516, + "learning_rate": 3.966287571080423e-06, + "loss": 1.8437, + "step": 797 + }, + { + "epoch": 0.8745205479452055, + "grad_norm": 2.16979718208313, + "learning_rate": 3.9642567018684e-06, + "loss": 1.829, + "step": 798 + }, + { + "epoch": 0.8756164383561644, + "grad_norm": 2.2387869358062744, + "learning_rate": 3.962225832656377e-06, + "loss": 1.9143, + "step": 799 + }, + { + "epoch": 0.8767123287671232, + "grad_norm": 2.208221197128296, + "learning_rate": 3.960194963444355e-06, + "loss": 1.8894, + "step": 800 + }, + { + "epoch": 0.8767123287671232, + "eval_loss": 2.047945976257324, + "eval_runtime": 37.1778, + "eval_samples_per_second": 26.898, + "eval_steps_per_second": 1.13, + "step": 800 + }, + { + "epoch": 0.8778082191780822, + "grad_norm": 2.335108518600464, + "learning_rate": 3.958164094232332e-06, + "loss": 1.8155, + "step": 801 + }, + { + "epoch": 0.8789041095890411, + "grad_norm": 2.168006181716919, + "learning_rate": 3.956133225020309e-06, + "loss": 1.8919, + "step": 802 + }, + { + "epoch": 0.88, + "grad_norm": 2.3029897212982178, + "learning_rate": 3.954102355808286e-06, + "loss": 1.8996, + "step": 803 + }, + { + "epoch": 0.8810958904109589, + "grad_norm": 2.156141519546509, + "learning_rate": 3.952071486596264e-06, + "loss": 1.8965, + "step": 804 + }, + { + "epoch": 0.8821917808219178, + "grad_norm": 2.1522068977355957, + "learning_rate": 3.950040617384241e-06, + "loss": 1.8922, + "step": 805 + }, + { + "epoch": 0.8832876712328767, + "grad_norm": 2.153430700302124, + "learning_rate": 3.948009748172218e-06, + "loss": 1.8839, + "step": 806 + }, + { + "epoch": 0.8843835616438356, + "grad_norm": 2.1891674995422363, + "learning_rate": 3.9459788789601955e-06, + "loss": 1.8756, + "step": 807 + }, + { + "epoch": 0.8854794520547945, + "grad_norm": 2.1092092990875244, + "learning_rate": 3.943948009748173e-06, + "loss": 1.9452, + "step": 808 + }, + { + "epoch": 0.8865753424657534, + "grad_norm": 2.0465025901794434, + "learning_rate": 3.94191714053615e-06, + "loss": 1.9067, + "step": 809 + }, + { + "epoch": 0.8876712328767123, + "grad_norm": 2.1150474548339844, + "learning_rate": 3.939886271324127e-06, + "loss": 1.8874, + "step": 810 + }, + { + "epoch": 0.8876712328767123, + "eval_loss": 2.046292304992676, + "eval_runtime": 37.2421, + "eval_samples_per_second": 26.851, + "eval_steps_per_second": 1.128, + "step": 810 + }, + { + "epoch": 0.8887671232876713, + "grad_norm": 2.039989948272705, + "learning_rate": 3.937855402112104e-06, + "loss": 1.9212, + "step": 811 + }, + { + "epoch": 0.8898630136986302, + "grad_norm": 2.17354679107666, + "learning_rate": 3.935824532900082e-06, + "loss": 1.8473, + "step": 812 + }, + { + "epoch": 0.890958904109589, + "grad_norm": 2.3101940155029297, + "learning_rate": 3.933793663688058e-06, + "loss": 1.8413, + "step": 813 + }, + { + "epoch": 0.8920547945205479, + "grad_norm": 2.1300928592681885, + "learning_rate": 3.9317627944760365e-06, + "loss": 1.9126, + "step": 814 + }, + { + "epoch": 0.8931506849315068, + "grad_norm": 2.2317357063293457, + "learning_rate": 3.929731925264013e-06, + "loss": 1.938, + "step": 815 + }, + { + "epoch": 0.8942465753424658, + "grad_norm": 2.1097419261932373, + "learning_rate": 3.927701056051991e-06, + "loss": 1.8426, + "step": 816 + }, + { + "epoch": 0.8953424657534247, + "grad_norm": 2.1709141731262207, + "learning_rate": 3.9256701868399674e-06, + "loss": 1.8779, + "step": 817 + }, + { + "epoch": 0.8964383561643836, + "grad_norm": 2.104193925857544, + "learning_rate": 3.9236393176279456e-06, + "loss": 1.8736, + "step": 818 + }, + { + "epoch": 0.8975342465753424, + "grad_norm": 2.094637632369995, + "learning_rate": 3.921608448415922e-06, + "loss": 1.8556, + "step": 819 + }, + { + "epoch": 0.8986301369863013, + "grad_norm": 2.217918872833252, + "learning_rate": 3.9195775792039e-06, + "loss": 1.8429, + "step": 820 + }, + { + "epoch": 0.8986301369863013, + "eval_loss": 2.046095848083496, + "eval_runtime": 37.5659, + "eval_samples_per_second": 26.62, + "eval_steps_per_second": 1.118, + "step": 820 + }, + { + "epoch": 0.8997260273972603, + "grad_norm": 2.220355272293091, + "learning_rate": 3.9175467099918765e-06, + "loss": 1.8973, + "step": 821 + }, + { + "epoch": 0.9008219178082192, + "grad_norm": 2.0766689777374268, + "learning_rate": 3.915515840779855e-06, + "loss": 1.8728, + "step": 822 + }, + { + "epoch": 0.9019178082191781, + "grad_norm": 2.2046821117401123, + "learning_rate": 3.913484971567831e-06, + "loss": 1.9211, + "step": 823 + }, + { + "epoch": 0.903013698630137, + "grad_norm": 2.286024332046509, + "learning_rate": 3.911454102355808e-06, + "loss": 1.913, + "step": 824 + }, + { + "epoch": 0.9041095890410958, + "grad_norm": 2.2495479583740234, + "learning_rate": 3.909423233143786e-06, + "loss": 1.8214, + "step": 825 + }, + { + "epoch": 0.9052054794520548, + "grad_norm": 2.1300203800201416, + "learning_rate": 3.907392363931763e-06, + "loss": 1.9033, + "step": 826 + }, + { + "epoch": 0.9063013698630137, + "grad_norm": 2.092094898223877, + "learning_rate": 3.90536149471974e-06, + "loss": 1.9007, + "step": 827 + }, + { + "epoch": 0.9073972602739726, + "grad_norm": 2.1571428775787354, + "learning_rate": 3.9033306255077175e-06, + "loss": 1.9606, + "step": 828 + }, + { + "epoch": 0.9084931506849315, + "grad_norm": 2.1052281856536865, + "learning_rate": 3.901299756295695e-06, + "loss": 1.9916, + "step": 829 + }, + { + "epoch": 0.9095890410958904, + "grad_norm": 2.113666534423828, + "learning_rate": 3.899268887083672e-06, + "loss": 2.0136, + "step": 830 + }, + { + "epoch": 0.9095890410958904, + "eval_loss": 2.0450072288513184, + "eval_runtime": 37.6746, + "eval_samples_per_second": 26.543, + "eval_steps_per_second": 1.115, + "step": 830 + }, + { + "epoch": 0.9106849315068493, + "grad_norm": 2.112591028213501, + "learning_rate": 3.897238017871649e-06, + "loss": 1.8784, + "step": 831 + }, + { + "epoch": 0.9117808219178082, + "grad_norm": 2.1584672927856445, + "learning_rate": 3.8952071486596266e-06, + "loss": 1.8876, + "step": 832 + }, + { + "epoch": 0.9128767123287671, + "grad_norm": 2.2744317054748535, + "learning_rate": 3.893176279447604e-06, + "loss": 1.8981, + "step": 833 + }, + { + "epoch": 0.913972602739726, + "grad_norm": 2.21824312210083, + "learning_rate": 3.891145410235581e-06, + "loss": 1.85, + "step": 834 + }, + { + "epoch": 0.915068493150685, + "grad_norm": 2.2203729152679443, + "learning_rate": 3.889114541023558e-06, + "loss": 1.9649, + "step": 835 + }, + { + "epoch": 0.9161643835616439, + "grad_norm": 2.1767687797546387, + "learning_rate": 3.887083671811536e-06, + "loss": 1.9044, + "step": 836 + }, + { + "epoch": 0.9172602739726028, + "grad_norm": 2.0724353790283203, + "learning_rate": 3.885052802599513e-06, + "loss": 1.8787, + "step": 837 + }, + { + "epoch": 0.9183561643835616, + "grad_norm": 2.1283740997314453, + "learning_rate": 3.88302193338749e-06, + "loss": 1.8284, + "step": 838 + }, + { + "epoch": 0.9194520547945205, + "grad_norm": 2.5087451934814453, + "learning_rate": 3.8809910641754675e-06, + "loss": 1.9546, + "step": 839 + }, + { + "epoch": 0.9205479452054794, + "grad_norm": 2.0633344650268555, + "learning_rate": 3.878960194963445e-06, + "loss": 1.8969, + "step": 840 + }, + { + "epoch": 0.9205479452054794, + "eval_loss": 2.047938346862793, + "eval_runtime": 37.2293, + "eval_samples_per_second": 26.861, + "eval_steps_per_second": 1.128, + "step": 840 + }, + { + "epoch": 0.9216438356164384, + "grad_norm": 2.178361654281616, + "learning_rate": 3.876929325751422e-06, + "loss": 1.8855, + "step": 841 + }, + { + "epoch": 0.9227397260273973, + "grad_norm": 2.3223721981048584, + "learning_rate": 3.874898456539399e-06, + "loss": 1.9363, + "step": 842 + }, + { + "epoch": 0.9238356164383562, + "grad_norm": 2.3330960273742676, + "learning_rate": 3.872867587327377e-06, + "loss": 1.8374, + "step": 843 + }, + { + "epoch": 0.924931506849315, + "grad_norm": 2.0584356784820557, + "learning_rate": 3.870836718115354e-06, + "loss": 1.911, + "step": 844 + }, + { + "epoch": 0.9260273972602739, + "grad_norm": 2.1834609508514404, + "learning_rate": 3.868805848903331e-06, + "loss": 1.8645, + "step": 845 + }, + { + "epoch": 0.9271232876712329, + "grad_norm": 2.2576122283935547, + "learning_rate": 3.8667749796913084e-06, + "loss": 1.8662, + "step": 846 + }, + { + "epoch": 0.9282191780821918, + "grad_norm": 2.1271183490753174, + "learning_rate": 3.864744110479286e-06, + "loss": 1.8935, + "step": 847 + }, + { + "epoch": 0.9293150684931507, + "grad_norm": 2.226727247238159, + "learning_rate": 3.862713241267262e-06, + "loss": 1.8969, + "step": 848 + }, + { + "epoch": 0.9304109589041096, + "grad_norm": 2.3301939964294434, + "learning_rate": 3.86068237205524e-06, + "loss": 1.9393, + "step": 849 + }, + { + "epoch": 0.9315068493150684, + "grad_norm": 2.180250883102417, + "learning_rate": 3.858651502843217e-06, + "loss": 1.9331, + "step": 850 + }, + { + "epoch": 0.9315068493150684, + "eval_loss": 2.0509681701660156, + "eval_runtime": 37.2111, + "eval_samples_per_second": 26.874, + "eval_steps_per_second": 1.129, + "step": 850 + }, + { + "epoch": 0.9326027397260274, + "grad_norm": 2.223548173904419, + "learning_rate": 3.856620633631195e-06, + "loss": 1.894, + "step": 851 + }, + { + "epoch": 0.9336986301369863, + "grad_norm": 2.1904265880584717, + "learning_rate": 3.854589764419171e-06, + "loss": 1.8265, + "step": 852 + }, + { + "epoch": 0.9347945205479452, + "grad_norm": 2.1374316215515137, + "learning_rate": 3.852558895207149e-06, + "loss": 1.8755, + "step": 853 + }, + { + "epoch": 0.9358904109589041, + "grad_norm": 2.2175488471984863, + "learning_rate": 3.850528025995126e-06, + "loss": 1.9649, + "step": 854 + }, + { + "epoch": 0.936986301369863, + "grad_norm": 2.192126989364624, + "learning_rate": 3.848497156783104e-06, + "loss": 1.919, + "step": 855 + }, + { + "epoch": 0.938082191780822, + "grad_norm": 2.219954013824463, + "learning_rate": 3.84646628757108e-06, + "loss": 1.8651, + "step": 856 + }, + { + "epoch": 0.9391780821917808, + "grad_norm": 2.2897071838378906, + "learning_rate": 3.8444354183590585e-06, + "loss": 1.8744, + "step": 857 + }, + { + "epoch": 0.9402739726027397, + "grad_norm": 2.1656205654144287, + "learning_rate": 3.842404549147035e-06, + "loss": 1.8622, + "step": 858 + }, + { + "epoch": 0.9413698630136986, + "grad_norm": 2.20072603225708, + "learning_rate": 3.840373679935013e-06, + "loss": 1.8678, + "step": 859 + }, + { + "epoch": 0.9424657534246575, + "grad_norm": 2.1474406719207764, + "learning_rate": 3.8383428107229894e-06, + "loss": 1.8462, + "step": 860 + }, + { + "epoch": 0.9424657534246575, + "eval_loss": 2.0451602935791016, + "eval_runtime": 37.1383, + "eval_samples_per_second": 26.926, + "eval_steps_per_second": 1.131, + "step": 860 + }, + { + "epoch": 0.9435616438356165, + "grad_norm": 1.9542036056518555, + "learning_rate": 3.836311941510967e-06, + "loss": 1.8739, + "step": 861 + }, + { + "epoch": 0.9446575342465754, + "grad_norm": 2.3951354026794434, + "learning_rate": 3.834281072298944e-06, + "loss": 1.8565, + "step": 862 + }, + { + "epoch": 0.9457534246575342, + "grad_norm": 2.4916961193084717, + "learning_rate": 3.832250203086921e-06, + "loss": 1.9138, + "step": 863 + }, + { + "epoch": 0.9468493150684931, + "grad_norm": 2.1650452613830566, + "learning_rate": 3.8302193338748985e-06, + "loss": 1.8499, + "step": 864 + }, + { + "epoch": 0.947945205479452, + "grad_norm": 2.2215073108673096, + "learning_rate": 3.828188464662876e-06, + "loss": 1.9376, + "step": 865 + }, + { + "epoch": 0.949041095890411, + "grad_norm": 2.3827078342437744, + "learning_rate": 3.826157595450853e-06, + "loss": 1.9726, + "step": 866 + }, + { + "epoch": 0.9501369863013699, + "grad_norm": 2.37194561958313, + "learning_rate": 3.82412672623883e-06, + "loss": 1.9009, + "step": 867 + }, + { + "epoch": 0.9512328767123288, + "grad_norm": 2.116562604904175, + "learning_rate": 3.822095857026808e-06, + "loss": 1.8342, + "step": 868 + }, + { + "epoch": 0.9523287671232876, + "grad_norm": 2.3410849571228027, + "learning_rate": 3.820064987814785e-06, + "loss": 1.9035, + "step": 869 + }, + { + "epoch": 0.9534246575342465, + "grad_norm": 2.207545280456543, + "learning_rate": 3.818034118602762e-06, + "loss": 1.8539, + "step": 870 + }, + { + "epoch": 0.9534246575342465, + "eval_loss": 2.0433826446533203, + "eval_runtime": 37.114, + "eval_samples_per_second": 26.944, + "eval_steps_per_second": 1.132, + "step": 870 + }, + { + "epoch": 0.9545205479452055, + "grad_norm": 2.1716196537017822, + "learning_rate": 3.8160032493907395e-06, + "loss": 1.8557, + "step": 871 + }, + { + "epoch": 0.9556164383561644, + "grad_norm": 2.213124990463257, + "learning_rate": 3.813972380178717e-06, + "loss": 1.8638, + "step": 872 + }, + { + "epoch": 0.9567123287671233, + "grad_norm": 2.3185672760009766, + "learning_rate": 3.811941510966694e-06, + "loss": 1.8645, + "step": 873 + }, + { + "epoch": 0.9578082191780822, + "grad_norm": 2.453956127166748, + "learning_rate": 3.809910641754671e-06, + "loss": 1.9183, + "step": 874 + }, + { + "epoch": 0.958904109589041, + "grad_norm": 2.1971592903137207, + "learning_rate": 3.8078797725426486e-06, + "loss": 1.8445, + "step": 875 + }, + { + "epoch": 0.96, + "grad_norm": 2.273247241973877, + "learning_rate": 3.8058489033306254e-06, + "loss": 1.8024, + "step": 876 + }, + { + "epoch": 0.9610958904109589, + "grad_norm": 2.1687841415405273, + "learning_rate": 3.803818034118603e-06, + "loss": 1.8931, + "step": 877 + }, + { + "epoch": 0.9621917808219178, + "grad_norm": 2.326498508453369, + "learning_rate": 3.80178716490658e-06, + "loss": 1.8587, + "step": 878 + }, + { + "epoch": 0.9632876712328767, + "grad_norm": 2.0317249298095703, + "learning_rate": 3.7997562956945577e-06, + "loss": 1.9394, + "step": 879 + }, + { + "epoch": 0.9643835616438357, + "grad_norm": 2.3171892166137695, + "learning_rate": 3.7977254264825345e-06, + "loss": 1.92, + "step": 880 + }, + { + "epoch": 0.9643835616438357, + "eval_loss": 2.0431556701660156, + "eval_runtime": 37.1174, + "eval_samples_per_second": 26.942, + "eval_steps_per_second": 1.132, + "step": 880 + }, + { + "epoch": 0.9654794520547946, + "grad_norm": 2.2215981483459473, + "learning_rate": 3.7956945572705122e-06, + "loss": 1.8658, + "step": 881 + }, + { + "epoch": 0.9665753424657534, + "grad_norm": 2.2940685749053955, + "learning_rate": 3.793663688058489e-06, + "loss": 1.8528, + "step": 882 + }, + { + "epoch": 0.9676712328767123, + "grad_norm": 2.2218081951141357, + "learning_rate": 3.791632818846467e-06, + "loss": 1.9446, + "step": 883 + }, + { + "epoch": 0.9687671232876712, + "grad_norm": 2.2608420848846436, + "learning_rate": 3.7896019496344436e-06, + "loss": 1.9433, + "step": 884 + }, + { + "epoch": 0.9698630136986301, + "grad_norm": 2.0247788429260254, + "learning_rate": 3.787571080422421e-06, + "loss": 1.9286, + "step": 885 + }, + { + "epoch": 0.9709589041095891, + "grad_norm": 2.164071798324585, + "learning_rate": 3.785540211210398e-06, + "loss": 1.9349, + "step": 886 + }, + { + "epoch": 0.972054794520548, + "grad_norm": 2.089215040206909, + "learning_rate": 3.7835093419983755e-06, + "loss": 1.9077, + "step": 887 + }, + { + "epoch": 0.9731506849315068, + "grad_norm": 2.245256185531616, + "learning_rate": 3.7814784727863527e-06, + "loss": 1.8911, + "step": 888 + }, + { + "epoch": 0.9742465753424657, + "grad_norm": 2.1227362155914307, + "learning_rate": 3.77944760357433e-06, + "loss": 1.8761, + "step": 889 + }, + { + "epoch": 0.9753424657534246, + "grad_norm": 2.1797730922698975, + "learning_rate": 3.7774167343623073e-06, + "loss": 1.9954, + "step": 890 + }, + { + "epoch": 0.9753424657534246, + "eval_loss": 2.034895896911621, + "eval_runtime": 37.1682, + "eval_samples_per_second": 26.905, + "eval_steps_per_second": 1.13, + "step": 890 + }, + { + "epoch": 0.9764383561643836, + "grad_norm": 2.171062707901001, + "learning_rate": 3.7753858651502846e-06, + "loss": 1.9287, + "step": 891 + }, + { + "epoch": 0.9775342465753425, + "grad_norm": 2.1950814723968506, + "learning_rate": 3.773354995938262e-06, + "loss": 1.8587, + "step": 892 + }, + { + "epoch": 0.9786301369863014, + "grad_norm": 2.2582995891571045, + "learning_rate": 3.771324126726239e-06, + "loss": 1.9269, + "step": 893 + }, + { + "epoch": 0.9797260273972602, + "grad_norm": 2.218113660812378, + "learning_rate": 3.7692932575142164e-06, + "loss": 1.9007, + "step": 894 + }, + { + "epoch": 0.9808219178082191, + "grad_norm": 2.187178134918213, + "learning_rate": 3.7672623883021937e-06, + "loss": 1.9326, + "step": 895 + }, + { + "epoch": 0.9819178082191781, + "grad_norm": 2.20451283454895, + "learning_rate": 3.765231519090171e-06, + "loss": 1.8644, + "step": 896 + }, + { + "epoch": 0.983013698630137, + "grad_norm": 2.1399290561676025, + "learning_rate": 3.7632006498781482e-06, + "loss": 1.7875, + "step": 897 + }, + { + "epoch": 0.9841095890410959, + "grad_norm": 2.049473524093628, + "learning_rate": 3.761169780666125e-06, + "loss": 1.8987, + "step": 898 + }, + { + "epoch": 0.9852054794520548, + "grad_norm": 2.154444456100464, + "learning_rate": 3.7591389114541028e-06, + "loss": 1.8475, + "step": 899 + }, + { + "epoch": 0.9863013698630136, + "grad_norm": 2.1840224266052246, + "learning_rate": 3.7571080422420796e-06, + "loss": 1.8966, + "step": 900 + }, + { + "epoch": 0.9863013698630136, + "eval_loss": 2.0326430797576904, + "eval_runtime": 37.1551, + "eval_samples_per_second": 26.914, + "eval_steps_per_second": 1.13, + "step": 900 + }, + { + "epoch": 0.9873972602739726, + "grad_norm": 2.0891942977905273, + "learning_rate": 3.7550771730300573e-06, + "loss": 1.9113, + "step": 901 + }, + { + "epoch": 0.9884931506849315, + "grad_norm": 2.2646753787994385, + "learning_rate": 3.753046303818034e-06, + "loss": 1.8624, + "step": 902 + }, + { + "epoch": 0.9895890410958904, + "grad_norm": 2.15173602104187, + "learning_rate": 3.751015434606012e-06, + "loss": 1.933, + "step": 903 + }, + { + "epoch": 0.9906849315068493, + "grad_norm": 2.1751456260681152, + "learning_rate": 3.7489845653939887e-06, + "loss": 1.9114, + "step": 904 + }, + { + "epoch": 0.9917808219178083, + "grad_norm": 2.161860942840576, + "learning_rate": 3.7469536961819664e-06, + "loss": 1.9508, + "step": 905 + }, + { + "epoch": 0.9928767123287672, + "grad_norm": 2.0921733379364014, + "learning_rate": 3.7449228269699433e-06, + "loss": 1.8482, + "step": 906 + }, + { + "epoch": 0.993972602739726, + "grad_norm": 2.0823893547058105, + "learning_rate": 3.742891957757921e-06, + "loss": 1.8744, + "step": 907 + }, + { + "epoch": 0.9950684931506849, + "grad_norm": 2.136939525604248, + "learning_rate": 3.740861088545898e-06, + "loss": 1.9037, + "step": 908 + }, + { + "epoch": 0.9961643835616438, + "grad_norm": 2.0600836277008057, + "learning_rate": 3.7388302193338755e-06, + "loss": 1.8922, + "step": 909 + }, + { + "epoch": 0.9972602739726028, + "grad_norm": 2.235780715942383, + "learning_rate": 3.7367993501218524e-06, + "loss": 1.9222, + "step": 910 + }, + { + "epoch": 0.9972602739726028, + "eval_loss": 2.034031629562378, + "eval_runtime": 37.2911, + "eval_samples_per_second": 26.816, + "eval_steps_per_second": 1.126, + "step": 910 + }, + { + "epoch": 0.9983561643835617, + "grad_norm": 2.124662160873413, + "learning_rate": 3.7347684809098297e-06, + "loss": 1.8584, + "step": 911 + }, + { + "epoch": 0.9994520547945206, + "grad_norm": 2.1237940788269043, + "learning_rate": 3.732737611697807e-06, + "loss": 1.8579, + "step": 912 + }, + { + "epoch": 1.0005479452054795, + "grad_norm": 2.1175124645233154, + "learning_rate": 3.7307067424857842e-06, + "loss": 1.8348, + "step": 913 + }, + { + "epoch": 1.0016438356164383, + "grad_norm": 2.349686861038208, + "learning_rate": 3.7286758732737615e-06, + "loss": 1.7822, + "step": 914 + }, + { + "epoch": 1.0027397260273974, + "grad_norm": 2.3278446197509766, + "learning_rate": 3.7266450040617388e-06, + "loss": 1.8346, + "step": 915 + }, + { + "epoch": 1.0038356164383562, + "grad_norm": 2.1633124351501465, + "learning_rate": 3.724614134849716e-06, + "loss": 1.8137, + "step": 916 + }, + { + "epoch": 1.004931506849315, + "grad_norm": 2.5075809955596924, + "learning_rate": 3.7225832656376933e-06, + "loss": 1.955, + "step": 917 + }, + { + "epoch": 1.006027397260274, + "grad_norm": 2.3370068073272705, + "learning_rate": 3.7205523964256706e-06, + "loss": 1.7985, + "step": 918 + }, + { + "epoch": 1.0071232876712328, + "grad_norm": 2.288229465484619, + "learning_rate": 3.718521527213648e-06, + "loss": 1.923, + "step": 919 + }, + { + "epoch": 1.0082191780821919, + "grad_norm": 2.1989684104919434, + "learning_rate": 3.716490658001625e-06, + "loss": 1.8876, + "step": 920 + }, + { + "epoch": 1.0082191780821919, + "eval_loss": 2.03442120552063, + "eval_runtime": 37.5659, + "eval_samples_per_second": 26.62, + "eval_steps_per_second": 1.118, + "step": 920 + }, + { + "epoch": 1.0093150684931507, + "grad_norm": 2.202122211456299, + "learning_rate": 3.7144597887896024e-06, + "loss": 1.879, + "step": 921 + }, + { + "epoch": 1.0104109589041095, + "grad_norm": 2.325223445892334, + "learning_rate": 3.7124289195775797e-06, + "loss": 1.7632, + "step": 922 + }, + { + "epoch": 1.0115068493150685, + "grad_norm": 2.2964577674865723, + "learning_rate": 3.710398050365557e-06, + "loss": 1.8983, + "step": 923 + }, + { + "epoch": 1.0126027397260273, + "grad_norm": 2.548537015914917, + "learning_rate": 3.708367181153534e-06, + "loss": 1.8666, + "step": 924 + }, + { + "epoch": 1.0136986301369864, + "grad_norm": 2.229745626449585, + "learning_rate": 3.7063363119415115e-06, + "loss": 1.8395, + "step": 925 + }, + { + "epoch": 1.0147945205479452, + "grad_norm": 2.3639235496520996, + "learning_rate": 3.7043054427294884e-06, + "loss": 1.8864, + "step": 926 + }, + { + "epoch": 1.0158904109589042, + "grad_norm": 2.3403055667877197, + "learning_rate": 3.702274573517466e-06, + "loss": 1.8385, + "step": 927 + }, + { + "epoch": 1.016986301369863, + "grad_norm": 2.2838001251220703, + "learning_rate": 3.700243704305443e-06, + "loss": 1.8135, + "step": 928 + }, + { + "epoch": 1.0180821917808218, + "grad_norm": 2.4730923175811768, + "learning_rate": 3.6982128350934206e-06, + "loss": 1.837, + "step": 929 + }, + { + "epoch": 1.0191780821917809, + "grad_norm": 2.5711073875427246, + "learning_rate": 3.6961819658813975e-06, + "loss": 1.8276, + "step": 930 + }, + { + "epoch": 1.0191780821917809, + "eval_loss": 2.0346033573150635, + "eval_runtime": 37.6741, + "eval_samples_per_second": 26.543, + "eval_steps_per_second": 1.115, + "step": 930 + }, + { + "epoch": 1.0202739726027397, + "grad_norm": 2.296823740005493, + "learning_rate": 3.694151096669375e-06, + "loss": 1.8054, + "step": 931 + }, + { + "epoch": 1.0213698630136987, + "grad_norm": 2.321107864379883, + "learning_rate": 3.692120227457352e-06, + "loss": 1.9296, + "step": 932 + }, + { + "epoch": 1.0224657534246575, + "grad_norm": 2.3478338718414307, + "learning_rate": 3.6900893582453297e-06, + "loss": 1.8503, + "step": 933 + }, + { + "epoch": 1.0235616438356165, + "grad_norm": 2.3698856830596924, + "learning_rate": 3.6880584890333066e-06, + "loss": 1.8766, + "step": 934 + }, + { + "epoch": 1.0246575342465754, + "grad_norm": 2.2751212120056152, + "learning_rate": 3.6860276198212834e-06, + "loss": 1.7575, + "step": 935 + }, + { + "epoch": 1.0257534246575342, + "grad_norm": 2.368741035461426, + "learning_rate": 3.683996750609261e-06, + "loss": 1.8555, + "step": 936 + }, + { + "epoch": 1.0268493150684932, + "grad_norm": 2.4507620334625244, + "learning_rate": 3.681965881397238e-06, + "loss": 1.8765, + "step": 937 + }, + { + "epoch": 1.027945205479452, + "grad_norm": 2.1907308101654053, + "learning_rate": 3.6799350121852157e-06, + "loss": 1.8652, + "step": 938 + }, + { + "epoch": 1.029041095890411, + "grad_norm": 2.2696914672851562, + "learning_rate": 3.6779041429731925e-06, + "loss": 1.8854, + "step": 939 + }, + { + "epoch": 1.0301369863013699, + "grad_norm": 2.3696796894073486, + "learning_rate": 3.6758732737611702e-06, + "loss": 1.8303, + "step": 940 + }, + { + "epoch": 1.0301369863013699, + "eval_loss": 2.030118703842163, + "eval_runtime": 37.1755, + "eval_samples_per_second": 26.899, + "eval_steps_per_second": 1.13, + "step": 940 + }, + { + "epoch": 1.0312328767123287, + "grad_norm": 2.440284013748169, + "learning_rate": 3.673842404549147e-06, + "loss": 1.8498, + "step": 941 + }, + { + "epoch": 1.0323287671232877, + "grad_norm": 2.250570297241211, + "learning_rate": 3.671811535337125e-06, + "loss": 1.8361, + "step": 942 + }, + { + "epoch": 1.0334246575342465, + "grad_norm": 2.5550990104675293, + "learning_rate": 3.6697806661251016e-06, + "loss": 1.8464, + "step": 943 + }, + { + "epoch": 1.0345205479452055, + "grad_norm": 2.502474784851074, + "learning_rate": 3.6677497969130793e-06, + "loss": 1.8471, + "step": 944 + }, + { + "epoch": 1.0356164383561643, + "grad_norm": 2.3815231323242188, + "learning_rate": 3.665718927701056e-06, + "loss": 1.9786, + "step": 945 + }, + { + "epoch": 1.0367123287671234, + "grad_norm": 2.3053736686706543, + "learning_rate": 3.663688058489034e-06, + "loss": 1.8625, + "step": 946 + }, + { + "epoch": 1.0378082191780822, + "grad_norm": 2.2623229026794434, + "learning_rate": 3.6616571892770107e-06, + "loss": 1.9206, + "step": 947 + }, + { + "epoch": 1.038904109589041, + "grad_norm": 2.634559392929077, + "learning_rate": 3.659626320064988e-06, + "loss": 1.8475, + "step": 948 + }, + { + "epoch": 1.04, + "grad_norm": 2.264615297317505, + "learning_rate": 3.6575954508529653e-06, + "loss": 1.7857, + "step": 949 + }, + { + "epoch": 1.0410958904109588, + "grad_norm": 2.4125282764434814, + "learning_rate": 3.6555645816409426e-06, + "loss": 1.841, + "step": 950 + }, + { + "epoch": 1.0410958904109588, + "eval_loss": 2.0278165340423584, + "eval_runtime": 37.1503, + "eval_samples_per_second": 26.918, + "eval_steps_per_second": 1.131, + "step": 950 + }, + { + "epoch": 1.0421917808219179, + "grad_norm": 2.365743637084961, + "learning_rate": 3.65353371242892e-06, + "loss": 1.8352, + "step": 951 + }, + { + "epoch": 1.0432876712328767, + "grad_norm": 2.6469883918762207, + "learning_rate": 3.651502843216897e-06, + "loss": 1.8814, + "step": 952 + }, + { + "epoch": 1.0443835616438357, + "grad_norm": 2.3646206855773926, + "learning_rate": 3.6494719740048744e-06, + "loss": 1.7282, + "step": 953 + }, + { + "epoch": 1.0454794520547945, + "grad_norm": 2.250856876373291, + "learning_rate": 3.6474411047928517e-06, + "loss": 1.8647, + "step": 954 + }, + { + "epoch": 1.0465753424657533, + "grad_norm": 2.5404837131500244, + "learning_rate": 3.645410235580829e-06, + "loss": 1.8175, + "step": 955 + }, + { + "epoch": 1.0476712328767124, + "grad_norm": 2.4388837814331055, + "learning_rate": 3.6433793663688062e-06, + "loss": 1.8643, + "step": 956 + }, + { + "epoch": 1.0487671232876712, + "grad_norm": 2.285959005355835, + "learning_rate": 3.6413484971567835e-06, + "loss": 1.8367, + "step": 957 + }, + { + "epoch": 1.0498630136986302, + "grad_norm": 2.1190242767333984, + "learning_rate": 3.6393176279447608e-06, + "loss": 1.8242, + "step": 958 + }, + { + "epoch": 1.050958904109589, + "grad_norm": 2.397578716278076, + "learning_rate": 3.637286758732738e-06, + "loss": 1.8681, + "step": 959 + }, + { + "epoch": 1.0520547945205478, + "grad_norm": 2.510599374771118, + "learning_rate": 3.6352558895207153e-06, + "loss": 1.8804, + "step": 960 + }, + { + "epoch": 1.0520547945205478, + "eval_loss": 2.0267090797424316, + "eval_runtime": 37.1226, + "eval_samples_per_second": 26.938, + "eval_steps_per_second": 1.131, + "step": 960 + }, + { + "epoch": 1.0531506849315069, + "grad_norm": 2.4743072986602783, + "learning_rate": 3.633225020308692e-06, + "loss": 1.8895, + "step": 961 + }, + { + "epoch": 1.0542465753424657, + "grad_norm": 2.3666775226593018, + "learning_rate": 3.63119415109667e-06, + "loss": 1.8836, + "step": 962 + }, + { + "epoch": 1.0553424657534247, + "grad_norm": 2.301119327545166, + "learning_rate": 3.6291632818846467e-06, + "loss": 1.8541, + "step": 963 + }, + { + "epoch": 1.0564383561643835, + "grad_norm": 2.482811212539673, + "learning_rate": 3.6271324126726244e-06, + "loss": 1.88, + "step": 964 + }, + { + "epoch": 1.0575342465753426, + "grad_norm": 2.2962276935577393, + "learning_rate": 3.6251015434606013e-06, + "loss": 1.7834, + "step": 965 + }, + { + "epoch": 1.0586301369863014, + "grad_norm": 2.365130662918091, + "learning_rate": 3.623070674248579e-06, + "loss": 1.7946, + "step": 966 + }, + { + "epoch": 1.0597260273972602, + "grad_norm": 2.4644675254821777, + "learning_rate": 3.621039805036556e-06, + "loss": 1.8379, + "step": 967 + }, + { + "epoch": 1.0608219178082192, + "grad_norm": 2.31742787361145, + "learning_rate": 3.6190089358245335e-06, + "loss": 1.8192, + "step": 968 + }, + { + "epoch": 1.061917808219178, + "grad_norm": 2.1614131927490234, + "learning_rate": 3.6169780666125104e-06, + "loss": 1.8794, + "step": 969 + }, + { + "epoch": 1.063013698630137, + "grad_norm": 2.4367032051086426, + "learning_rate": 3.614947197400488e-06, + "loss": 1.8673, + "step": 970 + }, + { + "epoch": 1.063013698630137, + "eval_loss": 2.0261292457580566, + "eval_runtime": 37.1116, + "eval_samples_per_second": 26.946, + "eval_steps_per_second": 1.132, + "step": 970 + }, + { + "epoch": 1.0641095890410959, + "grad_norm": 2.262789726257324, + "learning_rate": 3.612916328188465e-06, + "loss": 1.8808, + "step": 971 + }, + { + "epoch": 1.0652054794520547, + "grad_norm": 2.316316843032837, + "learning_rate": 3.610885458976442e-06, + "loss": 1.7803, + "step": 972 + }, + { + "epoch": 1.0663013698630137, + "grad_norm": 2.2936418056488037, + "learning_rate": 3.6088545897644195e-06, + "loss": 1.8783, + "step": 973 + }, + { + "epoch": 1.0673972602739725, + "grad_norm": 2.667069673538208, + "learning_rate": 3.6068237205523963e-06, + "loss": 1.8748, + "step": 974 + }, + { + "epoch": 1.0684931506849316, + "grad_norm": 2.31111216545105, + "learning_rate": 3.604792851340374e-06, + "loss": 1.8119, + "step": 975 + }, + { + "epoch": 1.0695890410958904, + "grad_norm": 2.268258571624756, + "learning_rate": 3.602761982128351e-06, + "loss": 1.8055, + "step": 976 + }, + { + "epoch": 1.0706849315068494, + "grad_norm": 2.307137966156006, + "learning_rate": 3.6007311129163286e-06, + "loss": 1.8976, + "step": 977 + }, + { + "epoch": 1.0717808219178082, + "grad_norm": 2.406036853790283, + "learning_rate": 3.5987002437043054e-06, + "loss": 1.8608, + "step": 978 + }, + { + "epoch": 1.072876712328767, + "grad_norm": 2.3303062915802, + "learning_rate": 3.596669374492283e-06, + "loss": 1.8038, + "step": 979 + }, + { + "epoch": 1.073972602739726, + "grad_norm": 2.2280168533325195, + "learning_rate": 3.59463850528026e-06, + "loss": 1.8342, + "step": 980 + }, + { + "epoch": 1.073972602739726, + "eval_loss": 2.0273854732513428, + "eval_runtime": 37.1546, + "eval_samples_per_second": 26.915, + "eval_steps_per_second": 1.13, + "step": 980 + }, + { + "epoch": 1.0750684931506849, + "grad_norm": 2.4218759536743164, + "learning_rate": 3.5926076360682377e-06, + "loss": 1.9104, + "step": 981 + }, + { + "epoch": 1.076164383561644, + "grad_norm": 2.2140634059906006, + "learning_rate": 3.5905767668562145e-06, + "loss": 1.7927, + "step": 982 + }, + { + "epoch": 1.0772602739726027, + "grad_norm": 2.179877996444702, + "learning_rate": 3.5885458976441922e-06, + "loss": 1.8868, + "step": 983 + }, + { + "epoch": 1.0783561643835617, + "grad_norm": 2.1988537311553955, + "learning_rate": 3.586515028432169e-06, + "loss": 1.8516, + "step": 984 + }, + { + "epoch": 1.0794520547945206, + "grad_norm": 2.1724448204040527, + "learning_rate": 3.5844841592201464e-06, + "loss": 1.8869, + "step": 985 + }, + { + "epoch": 1.0805479452054794, + "grad_norm": 2.2685718536376953, + "learning_rate": 3.5824532900081237e-06, + "loss": 1.8955, + "step": 986 + }, + { + "epoch": 1.0816438356164384, + "grad_norm": 2.39284348487854, + "learning_rate": 3.580422420796101e-06, + "loss": 1.7769, + "step": 987 + }, + { + "epoch": 1.0827397260273972, + "grad_norm": 2.3448750972747803, + "learning_rate": 3.578391551584078e-06, + "loss": 1.8417, + "step": 988 + }, + { + "epoch": 1.0838356164383562, + "grad_norm": 2.3518283367156982, + "learning_rate": 3.5763606823720555e-06, + "loss": 1.8524, + "step": 989 + }, + { + "epoch": 1.084931506849315, + "grad_norm": 2.1696958541870117, + "learning_rate": 3.5743298131600328e-06, + "loss": 1.7668, + "step": 990 + }, + { + "epoch": 1.084931506849315, + "eval_loss": 2.0244593620300293, + "eval_runtime": 37.1929, + "eval_samples_per_second": 26.887, + "eval_steps_per_second": 1.129, + "step": 990 + }, + { + "epoch": 1.086027397260274, + "grad_norm": 2.18300724029541, + "learning_rate": 3.57229894394801e-06, + "loss": 1.852, + "step": 991 + }, + { + "epoch": 1.087123287671233, + "grad_norm": 2.2075037956237793, + "learning_rate": 3.5702680747359873e-06, + "loss": 1.8799, + "step": 992 + }, + { + "epoch": 1.0882191780821917, + "grad_norm": 2.2172231674194336, + "learning_rate": 3.5682372055239646e-06, + "loss": 1.8573, + "step": 993 + }, + { + "epoch": 1.0893150684931507, + "grad_norm": 2.25169038772583, + "learning_rate": 3.566206336311942e-06, + "loss": 1.8125, + "step": 994 + }, + { + "epoch": 1.0904109589041096, + "grad_norm": 2.168386697769165, + "learning_rate": 3.564175467099919e-06, + "loss": 1.857, + "step": 995 + }, + { + "epoch": 1.0915068493150686, + "grad_norm": 2.2610065937042236, + "learning_rate": 3.5621445978878964e-06, + "loss": 1.817, + "step": 996 + }, + { + "epoch": 1.0926027397260274, + "grad_norm": 2.2561864852905273, + "learning_rate": 3.5601137286758737e-06, + "loss": 1.7754, + "step": 997 + }, + { + "epoch": 1.0936986301369862, + "grad_norm": 2.143625497817993, + "learning_rate": 3.5580828594638505e-06, + "loss": 1.8734, + "step": 998 + }, + { + "epoch": 1.0947945205479452, + "grad_norm": 2.433831214904785, + "learning_rate": 3.5560519902518282e-06, + "loss": 1.8248, + "step": 999 + }, + { + "epoch": 1.095890410958904, + "grad_norm": 2.237497091293335, + "learning_rate": 3.554021121039805e-06, + "loss": 1.8543, + "step": 1000 + }, + { + "epoch": 1.095890410958904, + "eval_loss": 2.020202159881592, + "eval_runtime": 37.2413, + "eval_samples_per_second": 26.852, + "eval_steps_per_second": 1.128, + "step": 1000 + }, + { + "epoch": 1.096986301369863, + "grad_norm": 2.253833532333374, + "learning_rate": 3.551990251827783e-06, + "loss": 1.8799, + "step": 1001 + }, + { + "epoch": 1.098082191780822, + "grad_norm": 2.205322504043579, + "learning_rate": 3.5499593826157596e-06, + "loss": 1.834, + "step": 1002 + }, + { + "epoch": 1.099178082191781, + "grad_norm": 2.3098037242889404, + "learning_rate": 3.5479285134037373e-06, + "loss": 1.8184, + "step": 1003 + }, + { + "epoch": 1.1002739726027397, + "grad_norm": 2.361888885498047, + "learning_rate": 3.545897644191714e-06, + "loss": 1.827, + "step": 1004 + }, + { + "epoch": 1.1013698630136985, + "grad_norm": 2.341947555541992, + "learning_rate": 3.543866774979692e-06, + "loss": 1.8184, + "step": 1005 + }, + { + "epoch": 1.1024657534246576, + "grad_norm": 2.2927780151367188, + "learning_rate": 3.5418359057676687e-06, + "loss": 1.805, + "step": 1006 + }, + { + "epoch": 1.1035616438356164, + "grad_norm": 2.432988166809082, + "learning_rate": 3.5398050365556464e-06, + "loss": 1.8465, + "step": 1007 + }, + { + "epoch": 1.1046575342465754, + "grad_norm": 2.428126335144043, + "learning_rate": 3.5377741673436233e-06, + "loss": 1.8115, + "step": 1008 + }, + { + "epoch": 1.1057534246575342, + "grad_norm": 2.32551646232605, + "learning_rate": 3.535743298131601e-06, + "loss": 1.8522, + "step": 1009 + }, + { + "epoch": 1.106849315068493, + "grad_norm": 2.297980308532715, + "learning_rate": 3.533712428919578e-06, + "loss": 1.8137, + "step": 1010 + }, + { + "epoch": 1.106849315068493, + "eval_loss": 2.0190494060516357, + "eval_runtime": 37.2149, + "eval_samples_per_second": 26.871, + "eval_steps_per_second": 1.129, + "step": 1010 + }, + { + "epoch": 1.107945205479452, + "grad_norm": 2.364342451095581, + "learning_rate": 3.5316815597075547e-06, + "loss": 1.8677, + "step": 1011 + }, + { + "epoch": 1.1090410958904109, + "grad_norm": 2.355008125305176, + "learning_rate": 3.5296506904955324e-06, + "loss": 1.8937, + "step": 1012 + }, + { + "epoch": 1.11013698630137, + "grad_norm": 2.2288317680358887, + "learning_rate": 3.5276198212835093e-06, + "loss": 1.8188, + "step": 1013 + }, + { + "epoch": 1.1112328767123287, + "grad_norm": 2.2801361083984375, + "learning_rate": 3.525588952071487e-06, + "loss": 1.8386, + "step": 1014 + }, + { + "epoch": 1.1123287671232878, + "grad_norm": 2.2801361083984375, + "learning_rate": 3.525588952071487e-06, + "loss": 1.8845, + "step": 1015 + }, + { + "epoch": 1.1134246575342466, + "grad_norm": 2.4301352500915527, + "learning_rate": 3.523558082859464e-06, + "loss": 1.8898, + "step": 1016 + }, + { + "epoch": 1.1145205479452054, + "grad_norm": 2.3492274284362793, + "learning_rate": 3.5215272136474415e-06, + "loss": 1.8711, + "step": 1017 + }, + { + "epoch": 1.1156164383561644, + "grad_norm": 2.1788229942321777, + "learning_rate": 3.5194963444354184e-06, + "loss": 1.8177, + "step": 1018 + }, + { + "epoch": 1.1167123287671232, + "grad_norm": 2.269767999649048, + "learning_rate": 3.517465475223396e-06, + "loss": 1.8259, + "step": 1019 + }, + { + "epoch": 1.1178082191780823, + "grad_norm": 2.2890219688415527, + "learning_rate": 3.515434606011373e-06, + "loss": 1.7873, + "step": 1020 + }, + { + "epoch": 1.1178082191780823, + "eval_loss": 2.023195266723633, + "eval_runtime": 37.5701, + "eval_samples_per_second": 26.617, + "eval_steps_per_second": 1.118, + "step": 1020 + }, + { + "epoch": 1.118904109589041, + "grad_norm": 2.5528533458709717, + "learning_rate": 3.5134037367993506e-06, + "loss": 1.8624, + "step": 1021 + }, + { + "epoch": 1.12, + "grad_norm": 2.5109457969665527, + "learning_rate": 3.5113728675873275e-06, + "loss": 1.9178, + "step": 1022 + }, + { + "epoch": 1.121095890410959, + "grad_norm": 2.288877010345459, + "learning_rate": 3.5093419983753047e-06, + "loss": 1.8604, + "step": 1023 + }, + { + "epoch": 1.1221917808219177, + "grad_norm": 2.460153818130493, + "learning_rate": 3.507311129163282e-06, + "loss": 1.8551, + "step": 1024 + }, + { + "epoch": 1.1232876712328768, + "grad_norm": 2.3520290851593018, + "learning_rate": 3.5052802599512593e-06, + "loss": 1.7818, + "step": 1025 + }, + { + "epoch": 1.1243835616438356, + "grad_norm": 2.32812237739563, + "learning_rate": 3.5032493907392366e-06, + "loss": 1.8255, + "step": 1026 + }, + { + "epoch": 1.1254794520547946, + "grad_norm": 2.3118042945861816, + "learning_rate": 3.501218521527214e-06, + "loss": 1.8493, + "step": 1027 + }, + { + "epoch": 1.1265753424657534, + "grad_norm": 2.2750277519226074, + "learning_rate": 3.499187652315191e-06, + "loss": 1.7759, + "step": 1028 + }, + { + "epoch": 1.1276712328767124, + "grad_norm": 2.272120475769043, + "learning_rate": 3.4971567831031684e-06, + "loss": 1.8399, + "step": 1029 + }, + { + "epoch": 1.1287671232876713, + "grad_norm": 2.2075254917144775, + "learning_rate": 3.4951259138911457e-06, + "loss": 1.8774, + "step": 1030 + }, + { + "epoch": 1.1287671232876713, + "eval_loss": 2.0199804306030273, + "eval_runtime": 37.6753, + "eval_samples_per_second": 26.543, + "eval_steps_per_second": 1.115, + "step": 1030 + }, + { + "epoch": 1.12986301369863, + "grad_norm": 2.359877586364746, + "learning_rate": 3.493095044679123e-06, + "loss": 1.7942, + "step": 1031 + }, + { + "epoch": 1.130958904109589, + "grad_norm": 2.3057703971862793, + "learning_rate": 3.4910641754671002e-06, + "loss": 1.8996, + "step": 1032 + }, + { + "epoch": 1.132054794520548, + "grad_norm": 2.3109569549560547, + "learning_rate": 3.4890333062550775e-06, + "loss": 1.7336, + "step": 1033 + }, + { + "epoch": 1.133150684931507, + "grad_norm": 2.3021819591522217, + "learning_rate": 3.4870024370430548e-06, + "loss": 1.869, + "step": 1034 + }, + { + "epoch": 1.1342465753424658, + "grad_norm": 2.4741852283477783, + "learning_rate": 3.484971567831032e-06, + "loss": 1.7652, + "step": 1035 + }, + { + "epoch": 1.1353424657534246, + "grad_norm": 2.326500415802002, + "learning_rate": 3.482940698619009e-06, + "loss": 1.809, + "step": 1036 + }, + { + "epoch": 1.1364383561643836, + "grad_norm": 2.2562193870544434, + "learning_rate": 3.4809098294069866e-06, + "loss": 1.8806, + "step": 1037 + }, + { + "epoch": 1.1375342465753424, + "grad_norm": 2.2849345207214355, + "learning_rate": 3.4788789601949634e-06, + "loss": 1.8329, + "step": 1038 + }, + { + "epoch": 1.1386301369863014, + "grad_norm": 2.333707332611084, + "learning_rate": 3.476848090982941e-06, + "loss": 1.8541, + "step": 1039 + }, + { + "epoch": 1.1397260273972603, + "grad_norm": 2.234175682067871, + "learning_rate": 3.474817221770918e-06, + "loss": 1.8126, + "step": 1040 + }, + { + "epoch": 1.1397260273972603, + "eval_loss": 2.0176141262054443, + "eval_runtime": 37.232, + "eval_samples_per_second": 26.859, + "eval_steps_per_second": 1.128, + "step": 1040 + }, + { + "epoch": 1.140821917808219, + "grad_norm": 2.4212377071380615, + "learning_rate": 3.4727863525588957e-06, + "loss": 1.9325, + "step": 1041 + }, + { + "epoch": 1.141917808219178, + "grad_norm": 2.1526010036468506, + "learning_rate": 3.4707554833468725e-06, + "loss": 1.8674, + "step": 1042 + }, + { + "epoch": 1.143013698630137, + "grad_norm": 2.336186408996582, + "learning_rate": 3.4687246141348502e-06, + "loss": 1.8275, + "step": 1043 + }, + { + "epoch": 1.144109589041096, + "grad_norm": 2.2991745471954346, + "learning_rate": 3.466693744922827e-06, + "loss": 1.8057, + "step": 1044 + }, + { + "epoch": 1.1452054794520548, + "grad_norm": 2.3816370964050293, + "learning_rate": 3.464662875710805e-06, + "loss": 1.8212, + "step": 1045 + }, + { + "epoch": 1.1463013698630138, + "grad_norm": 2.2383193969726562, + "learning_rate": 3.4626320064987817e-06, + "loss": 1.8444, + "step": 1046 + }, + { + "epoch": 1.1473972602739726, + "grad_norm": 2.377574920654297, + "learning_rate": 3.4606011372867594e-06, + "loss": 1.8319, + "step": 1047 + }, + { + "epoch": 1.1484931506849314, + "grad_norm": 2.1803700923919678, + "learning_rate": 3.458570268074736e-06, + "loss": 1.8484, + "step": 1048 + }, + { + "epoch": 1.1495890410958904, + "grad_norm": 2.1810641288757324, + "learning_rate": 3.4565393988627135e-06, + "loss": 1.8187, + "step": 1049 + }, + { + "epoch": 1.1506849315068493, + "grad_norm": 2.1565394401550293, + "learning_rate": 3.4545085296506908e-06, + "loss": 1.8117, + "step": 1050 + }, + { + "epoch": 1.1506849315068493, + "eval_loss": 2.0156562328338623, + "eval_runtime": 37.2072, + "eval_samples_per_second": 26.877, + "eval_steps_per_second": 1.129, + "step": 1050 + }, + { + "epoch": 1.1517808219178083, + "grad_norm": 2.1753759384155273, + "learning_rate": 3.452477660438668e-06, + "loss": 1.7822, + "step": 1051 + }, + { + "epoch": 1.152876712328767, + "grad_norm": 2.250089645385742, + "learning_rate": 3.4504467912266453e-06, + "loss": 1.8634, + "step": 1052 + }, + { + "epoch": 1.1539726027397261, + "grad_norm": 2.3043293952941895, + "learning_rate": 3.4484159220146226e-06, + "loss": 1.7743, + "step": 1053 + }, + { + "epoch": 1.155068493150685, + "grad_norm": 2.2193732261657715, + "learning_rate": 3.4463850528026e-06, + "loss": 1.8328, + "step": 1054 + }, + { + "epoch": 1.1561643835616437, + "grad_norm": 2.4204866886138916, + "learning_rate": 3.444354183590577e-06, + "loss": 1.8133, + "step": 1055 + }, + { + "epoch": 1.1572602739726028, + "grad_norm": 2.3088018894195557, + "learning_rate": 3.4423233143785544e-06, + "loss": 1.8523, + "step": 1056 + }, + { + "epoch": 1.1583561643835616, + "grad_norm": 2.38910174369812, + "learning_rate": 3.4402924451665317e-06, + "loss": 1.7846, + "step": 1057 + }, + { + "epoch": 1.1594520547945206, + "grad_norm": 2.2473702430725098, + "learning_rate": 3.438261575954509e-06, + "loss": 1.7632, + "step": 1058 + }, + { + "epoch": 1.1605479452054794, + "grad_norm": 2.3356645107269287, + "learning_rate": 3.4362307067424862e-06, + "loss": 1.8461, + "step": 1059 + }, + { + "epoch": 1.1616438356164385, + "grad_norm": 2.2460737228393555, + "learning_rate": 3.434199837530463e-06, + "loss": 1.8853, + "step": 1060 + }, + { + "epoch": 1.1616438356164385, + "eval_loss": 2.016077995300293, + "eval_runtime": 37.1943, + "eval_samples_per_second": 26.886, + "eval_steps_per_second": 1.129, + "step": 1060 + }, + { + "epoch": 1.1627397260273973, + "grad_norm": 2.28434157371521, + "learning_rate": 3.432168968318441e-06, + "loss": 1.8451, + "step": 1061 + }, + { + "epoch": 1.163835616438356, + "grad_norm": 2.299049139022827, + "learning_rate": 3.4301380991064176e-06, + "loss": 1.7777, + "step": 1062 + }, + { + "epoch": 1.1649315068493151, + "grad_norm": 2.364074468612671, + "learning_rate": 3.4281072298943953e-06, + "loss": 1.8517, + "step": 1063 + }, + { + "epoch": 1.166027397260274, + "grad_norm": 2.2381322383880615, + "learning_rate": 3.426076360682372e-06, + "loss": 1.889, + "step": 1064 + }, + { + "epoch": 1.167123287671233, + "grad_norm": 2.2984025478363037, + "learning_rate": 3.42404549147035e-06, + "loss": 1.8907, + "step": 1065 + }, + { + "epoch": 1.1682191780821918, + "grad_norm": 2.2215473651885986, + "learning_rate": 3.4220146222583267e-06, + "loss": 1.8313, + "step": 1066 + }, + { + "epoch": 1.1693150684931506, + "grad_norm": 2.3414270877838135, + "learning_rate": 3.4199837530463044e-06, + "loss": 1.783, + "step": 1067 + }, + { + "epoch": 1.1704109589041096, + "grad_norm": 2.0636470317840576, + "learning_rate": 3.4179528838342813e-06, + "loss": 1.8185, + "step": 1068 + }, + { + "epoch": 1.1715068493150684, + "grad_norm": 2.2158873081207275, + "learning_rate": 3.415922014622259e-06, + "loss": 1.9319, + "step": 1069 + }, + { + "epoch": 1.1726027397260275, + "grad_norm": 2.244051694869995, + "learning_rate": 3.413891145410236e-06, + "loss": 1.8732, + "step": 1070 + }, + { + "epoch": 1.1726027397260275, + "eval_loss": 2.0180234909057617, + "eval_runtime": 37.1822, + "eval_samples_per_second": 26.895, + "eval_steps_per_second": 1.13, + "step": 1070 + }, + { + "epoch": 1.1736986301369863, + "grad_norm": 2.1608433723449707, + "learning_rate": 3.4118602761982135e-06, + "loss": 1.8709, + "step": 1071 + }, + { + "epoch": 1.174794520547945, + "grad_norm": 2.2612338066101074, + "learning_rate": 3.4098294069861904e-06, + "loss": 1.8882, + "step": 1072 + }, + { + "epoch": 1.1758904109589041, + "grad_norm": 2.3257741928100586, + "learning_rate": 3.4077985377741673e-06, + "loss": 1.8377, + "step": 1073 + }, + { + "epoch": 1.176986301369863, + "grad_norm": 2.2486586570739746, + "learning_rate": 3.405767668562145e-06, + "loss": 1.8105, + "step": 1074 + }, + { + "epoch": 1.178082191780822, + "grad_norm": 2.2840020656585693, + "learning_rate": 3.403736799350122e-06, + "loss": 1.8868, + "step": 1075 + }, + { + "epoch": 1.1791780821917808, + "grad_norm": 2.3264284133911133, + "learning_rate": 3.4017059301380995e-06, + "loss": 1.8701, + "step": 1076 + }, + { + "epoch": 1.1802739726027398, + "grad_norm": 2.311424493789673, + "learning_rate": 3.3996750609260764e-06, + "loss": 1.801, + "step": 1077 + }, + { + "epoch": 1.1813698630136986, + "grad_norm": 2.2177181243896484, + "learning_rate": 3.397644191714054e-06, + "loss": 1.9188, + "step": 1078 + }, + { + "epoch": 1.1824657534246574, + "grad_norm": 2.413118362426758, + "learning_rate": 3.395613322502031e-06, + "loss": 1.8198, + "step": 1079 + }, + { + "epoch": 1.1835616438356165, + "grad_norm": 2.334688186645508, + "learning_rate": 3.3935824532900086e-06, + "loss": 1.822, + "step": 1080 + }, + { + "epoch": 1.1835616438356165, + "eval_loss": 2.012676954269409, + "eval_runtime": 37.2415, + "eval_samples_per_second": 26.852, + "eval_steps_per_second": 1.128, + "step": 1080 + }, + { + "epoch": 1.1846575342465753, + "grad_norm": 2.3901584148406982, + "learning_rate": 3.3915515840779855e-06, + "loss": 1.791, + "step": 1081 + }, + { + "epoch": 1.1857534246575343, + "grad_norm": 2.2731425762176514, + "learning_rate": 3.389520714865963e-06, + "loss": 1.8367, + "step": 1082 + }, + { + "epoch": 1.1868493150684931, + "grad_norm": 2.311553478240967, + "learning_rate": 3.38748984565394e-06, + "loss": 1.8302, + "step": 1083 + }, + { + "epoch": 1.1879452054794521, + "grad_norm": 2.3030776977539062, + "learning_rate": 3.3854589764419177e-06, + "loss": 1.8652, + "step": 1084 + }, + { + "epoch": 1.189041095890411, + "grad_norm": 2.341188430786133, + "learning_rate": 3.3834281072298946e-06, + "loss": 1.8094, + "step": 1085 + }, + { + "epoch": 1.1901369863013698, + "grad_norm": 2.331962823867798, + "learning_rate": 3.381397238017872e-06, + "loss": 1.827, + "step": 1086 + }, + { + "epoch": 1.1912328767123288, + "grad_norm": 2.275923728942871, + "learning_rate": 3.379366368805849e-06, + "loss": 1.7935, + "step": 1087 + }, + { + "epoch": 1.1923287671232876, + "grad_norm": 2.4608163833618164, + "learning_rate": 3.3773354995938264e-06, + "loss": 1.8965, + "step": 1088 + }, + { + "epoch": 1.1934246575342466, + "grad_norm": 2.2208592891693115, + "learning_rate": 3.3753046303818037e-06, + "loss": 1.8447, + "step": 1089 + }, + { + "epoch": 1.1945205479452055, + "grad_norm": 2.4284770488739014, + "learning_rate": 3.373273761169781e-06, + "loss": 1.8306, + "step": 1090 + }, + { + "epoch": 1.1945205479452055, + "eval_loss": 2.010664939880371, + "eval_runtime": 37.237, + "eval_samples_per_second": 26.855, + "eval_steps_per_second": 1.128, + "step": 1090 + }, + { + "epoch": 1.1956164383561645, + "grad_norm": 2.329601764678955, + "learning_rate": 3.3712428919577582e-06, + "loss": 1.8275, + "step": 1091 + }, + { + "epoch": 1.1967123287671233, + "grad_norm": 2.277939796447754, + "learning_rate": 3.3692120227457355e-06, + "loss": 1.8161, + "step": 1092 + }, + { + "epoch": 1.1978082191780821, + "grad_norm": 2.446780204772949, + "learning_rate": 3.3671811535337128e-06, + "loss": 1.8511, + "step": 1093 + }, + { + "epoch": 1.1989041095890411, + "grad_norm": 2.382477283477783, + "learning_rate": 3.36515028432169e-06, + "loss": 1.7834, + "step": 1094 + }, + { + "epoch": 1.2, + "grad_norm": 2.366361379623413, + "learning_rate": 3.3631194151096673e-06, + "loss": 1.8105, + "step": 1095 + }, + { + "epoch": 1.201095890410959, + "grad_norm": 2.3439717292785645, + "learning_rate": 3.3610885458976446e-06, + "loss": 1.8418, + "step": 1096 + }, + { + "epoch": 1.2021917808219178, + "grad_norm": 2.2739830017089844, + "learning_rate": 3.359057676685622e-06, + "loss": 1.8818, + "step": 1097 + }, + { + "epoch": 1.2032876712328768, + "grad_norm": 2.1636738777160645, + "learning_rate": 3.357026807473599e-06, + "loss": 1.8757, + "step": 1098 + }, + { + "epoch": 1.2043835616438356, + "grad_norm": 2.386441946029663, + "learning_rate": 3.354995938261576e-06, + "loss": 1.756, + "step": 1099 + }, + { + "epoch": 1.2054794520547945, + "grad_norm": 2.2241456508636475, + "learning_rate": 3.3529650690495537e-06, + "loss": 1.8034, + "step": 1100 + }, + { + "epoch": 1.2054794520547945, + "eval_loss": 2.008504867553711, + "eval_runtime": 37.2855, + "eval_samples_per_second": 26.82, + "eval_steps_per_second": 1.126, + "step": 1100 + }, + { + "epoch": 1.2065753424657535, + "grad_norm": 2.196992874145508, + "learning_rate": 3.3509341998375305e-06, + "loss": 1.7722, + "step": 1101 + }, + { + "epoch": 1.2076712328767123, + "grad_norm": 2.3572094440460205, + "learning_rate": 3.3489033306255082e-06, + "loss": 1.8109, + "step": 1102 + }, + { + "epoch": 1.208767123287671, + "grad_norm": 2.443281650543213, + "learning_rate": 3.346872461413485e-06, + "loss": 1.843, + "step": 1103 + }, + { + "epoch": 1.2098630136986301, + "grad_norm": 2.2613234519958496, + "learning_rate": 3.344841592201463e-06, + "loss": 1.8023, + "step": 1104 + }, + { + "epoch": 1.210958904109589, + "grad_norm": 2.2466726303100586, + "learning_rate": 3.3428107229894397e-06, + "loss": 1.7838, + "step": 1105 + }, + { + "epoch": 1.212054794520548, + "grad_norm": 2.2314462661743164, + "learning_rate": 3.3407798537774174e-06, + "loss": 1.8077, + "step": 1106 + }, + { + "epoch": 1.2131506849315068, + "grad_norm": 2.3288400173187256, + "learning_rate": 3.338748984565394e-06, + "loss": 1.8319, + "step": 1107 + }, + { + "epoch": 1.2142465753424658, + "grad_norm": 2.3477249145507812, + "learning_rate": 3.336718115353372e-06, + "loss": 1.8772, + "step": 1108 + }, + { + "epoch": 1.2153424657534246, + "grad_norm": 2.3588409423828125, + "learning_rate": 3.3346872461413488e-06, + "loss": 1.8832, + "step": 1109 + }, + { + "epoch": 1.2164383561643834, + "grad_norm": 2.327796220779419, + "learning_rate": 3.3326563769293256e-06, + "loss": 1.8033, + "step": 1110 + }, + { + "epoch": 1.2164383561643834, + "eval_loss": 2.0094382762908936, + "eval_runtime": 37.2666, + "eval_samples_per_second": 26.834, + "eval_steps_per_second": 1.127, + "step": 1110 + }, + { + "epoch": 1.2175342465753425, + "grad_norm": 2.257244825363159, + "learning_rate": 3.3306255077173033e-06, + "loss": 1.7903, + "step": 1111 + }, + { + "epoch": 1.2186301369863013, + "grad_norm": 2.3321614265441895, + "learning_rate": 3.32859463850528e-06, + "loss": 1.8295, + "step": 1112 + }, + { + "epoch": 1.2197260273972603, + "grad_norm": 2.4073803424835205, + "learning_rate": 3.326563769293258e-06, + "loss": 1.8395, + "step": 1113 + }, + { + "epoch": 1.2208219178082191, + "grad_norm": 2.3631434440612793, + "learning_rate": 3.3245329000812347e-06, + "loss": 1.7548, + "step": 1114 + }, + { + "epoch": 1.2219178082191782, + "grad_norm": 2.275766134262085, + "learning_rate": 3.3225020308692124e-06, + "loss": 1.808, + "step": 1115 + }, + { + "epoch": 1.223013698630137, + "grad_norm": 2.438011407852173, + "learning_rate": 3.3204711616571893e-06, + "loss": 1.8274, + "step": 1116 + }, + { + "epoch": 1.2241095890410958, + "grad_norm": 2.371826171875, + "learning_rate": 3.318440292445167e-06, + "loss": 1.7742, + "step": 1117 + }, + { + "epoch": 1.2252054794520548, + "grad_norm": 2.422004461288452, + "learning_rate": 3.316409423233144e-06, + "loss": 1.8268, + "step": 1118 + }, + { + "epoch": 1.2263013698630136, + "grad_norm": 2.377979278564453, + "learning_rate": 3.3143785540211215e-06, + "loss": 1.7857, + "step": 1119 + }, + { + "epoch": 1.2273972602739727, + "grad_norm": 2.288482904434204, + "learning_rate": 3.3123476848090984e-06, + "loss": 1.876, + "step": 1120 + }, + { + "epoch": 1.2273972602739727, + "eval_loss": 2.0104730129241943, + "eval_runtime": 37.5365, + "eval_samples_per_second": 26.641, + "eval_steps_per_second": 1.119, + "step": 1120 + }, + { + "epoch": 1.2284931506849315, + "grad_norm": 2.451533555984497, + "learning_rate": 3.310316815597076e-06, + "loss": 1.8324, + "step": 1121 + }, + { + "epoch": 1.2295890410958905, + "grad_norm": 2.3785181045532227, + "learning_rate": 3.308285946385053e-06, + "loss": 1.8239, + "step": 1122 + }, + { + "epoch": 1.2306849315068493, + "grad_norm": 2.2276721000671387, + "learning_rate": 3.30625507717303e-06, + "loss": 1.9132, + "step": 1123 + }, + { + "epoch": 1.2317808219178081, + "grad_norm": 2.2575314044952393, + "learning_rate": 3.3042242079610075e-06, + "loss": 1.769, + "step": 1124 + }, + { + "epoch": 1.2328767123287672, + "grad_norm": 2.5311717987060547, + "learning_rate": 3.3021933387489847e-06, + "loss": 1.8605, + "step": 1125 + }, + { + "epoch": 1.233972602739726, + "grad_norm": 2.2395896911621094, + "learning_rate": 3.300162469536962e-06, + "loss": 1.8725, + "step": 1126 + }, + { + "epoch": 1.235068493150685, + "grad_norm": 2.2402031421661377, + "learning_rate": 3.2981316003249393e-06, + "loss": 1.7955, + "step": 1127 + }, + { + "epoch": 1.2361643835616438, + "grad_norm": 2.308950424194336, + "learning_rate": 3.2961007311129166e-06, + "loss": 1.8812, + "step": 1128 + }, + { + "epoch": 1.2372602739726029, + "grad_norm": 2.437767267227173, + "learning_rate": 3.294069861900894e-06, + "loss": 1.8489, + "step": 1129 + }, + { + "epoch": 1.2383561643835617, + "grad_norm": 2.3188693523406982, + "learning_rate": 3.292038992688871e-06, + "loss": 1.9082, + "step": 1130 + }, + { + "epoch": 1.2383561643835617, + "eval_loss": 2.0110676288604736, + "eval_runtime": 37.6437, + "eval_samples_per_second": 26.565, + "eval_steps_per_second": 1.116, + "step": 1130 + }, + { + "epoch": 1.2394520547945205, + "grad_norm": 2.312760591506958, + "learning_rate": 3.2900081234768484e-06, + "loss": 1.8344, + "step": 1131 + }, + { + "epoch": 1.2405479452054795, + "grad_norm": 2.5837206840515137, + "learning_rate": 3.2879772542648257e-06, + "loss": 1.8503, + "step": 1132 + }, + { + "epoch": 1.2416438356164383, + "grad_norm": 2.1864733695983887, + "learning_rate": 3.285946385052803e-06, + "loss": 1.8822, + "step": 1133 + }, + { + "epoch": 1.2427397260273974, + "grad_norm": 2.2644550800323486, + "learning_rate": 3.2839155158407802e-06, + "loss": 1.7541, + "step": 1134 + }, + { + "epoch": 1.2438356164383562, + "grad_norm": 2.292646646499634, + "learning_rate": 3.2818846466287575e-06, + "loss": 1.8575, + "step": 1135 + }, + { + "epoch": 1.244931506849315, + "grad_norm": 2.2162070274353027, + "learning_rate": 3.2798537774167344e-06, + "loss": 1.7525, + "step": 1136 + }, + { + "epoch": 1.246027397260274, + "grad_norm": 2.32688307762146, + "learning_rate": 3.277822908204712e-06, + "loss": 1.8955, + "step": 1137 + }, + { + "epoch": 1.2471232876712328, + "grad_norm": 2.3226473331451416, + "learning_rate": 3.275792038992689e-06, + "loss": 1.8255, + "step": 1138 + }, + { + "epoch": 1.2482191780821918, + "grad_norm": 2.3533706665039062, + "learning_rate": 3.2737611697806666e-06, + "loss": 1.783, + "step": 1139 + }, + { + "epoch": 1.2493150684931507, + "grad_norm": 2.237661123275757, + "learning_rate": 3.2717303005686435e-06, + "loss": 1.8367, + "step": 1140 + }, + { + "epoch": 1.2493150684931507, + "eval_loss": 2.0123534202575684, + "eval_runtime": 37.3573, + "eval_samples_per_second": 26.769, + "eval_steps_per_second": 1.124, + "step": 1140 + }, + { + "epoch": 1.2504109589041095, + "grad_norm": 2.283047676086426, + "learning_rate": 3.269699431356621e-06, + "loss": 1.7987, + "step": 1141 + }, + { + "epoch": 1.2515068493150685, + "grad_norm": 2.2264456748962402, + "learning_rate": 3.267668562144598e-06, + "loss": 1.8307, + "step": 1142 + }, + { + "epoch": 1.2526027397260273, + "grad_norm": 2.2724905014038086, + "learning_rate": 3.2656376929325757e-06, + "loss": 1.8237, + "step": 1143 + }, + { + "epoch": 1.2536986301369863, + "grad_norm": 2.270782947540283, + "learning_rate": 3.2636068237205526e-06, + "loss": 1.7846, + "step": 1144 + }, + { + "epoch": 1.2547945205479452, + "grad_norm": 2.2032551765441895, + "learning_rate": 3.2615759545085303e-06, + "loss": 1.8438, + "step": 1145 + }, + { + "epoch": 1.2558904109589042, + "grad_norm": 2.199150800704956, + "learning_rate": 3.259545085296507e-06, + "loss": 1.8199, + "step": 1146 + }, + { + "epoch": 1.256986301369863, + "grad_norm": 2.4404170513153076, + "learning_rate": 3.257514216084484e-06, + "loss": 1.8817, + "step": 1147 + }, + { + "epoch": 1.2580821917808218, + "grad_norm": 2.281348943710327, + "learning_rate": 3.2554833468724617e-06, + "loss": 1.771, + "step": 1148 + }, + { + "epoch": 1.2591780821917808, + "grad_norm": 2.19482159614563, + "learning_rate": 3.2534524776604385e-06, + "loss": 1.8684, + "step": 1149 + }, + { + "epoch": 1.2602739726027397, + "grad_norm": 2.221618413925171, + "learning_rate": 3.2514216084484162e-06, + "loss": 1.8398, + "step": 1150 + }, + { + "epoch": 1.2602739726027397, + "eval_loss": 2.007969617843628, + "eval_runtime": 37.3679, + "eval_samples_per_second": 26.761, + "eval_steps_per_second": 1.124, + "step": 1150 + }, + { + "epoch": 1.2613698630136987, + "grad_norm": 2.193432569503784, + "learning_rate": 3.249390739236393e-06, + "loss": 1.8122, + "step": 1151 + }, + { + "epoch": 1.2624657534246575, + "grad_norm": 2.246462106704712, + "learning_rate": 3.2473598700243708e-06, + "loss": 1.8447, + "step": 1152 + }, + { + "epoch": 1.2635616438356165, + "grad_norm": 2.3411879539489746, + "learning_rate": 3.2453290008123476e-06, + "loss": 1.7988, + "step": 1153 + }, + { + "epoch": 1.2646575342465753, + "grad_norm": 2.245286464691162, + "learning_rate": 3.2432981316003253e-06, + "loss": 1.8209, + "step": 1154 + }, + { + "epoch": 1.2657534246575342, + "grad_norm": 2.292067527770996, + "learning_rate": 3.241267262388302e-06, + "loss": 1.8285, + "step": 1155 + }, + { + "epoch": 1.2668493150684932, + "grad_norm": 2.3058111667633057, + "learning_rate": 3.23923639317628e-06, + "loss": 1.7842, + "step": 1156 + }, + { + "epoch": 1.267945205479452, + "grad_norm": 2.306243896484375, + "learning_rate": 3.2372055239642567e-06, + "loss": 1.8221, + "step": 1157 + }, + { + "epoch": 1.269041095890411, + "grad_norm": 2.399759531021118, + "learning_rate": 3.2351746547522344e-06, + "loss": 1.7836, + "step": 1158 + }, + { + "epoch": 1.2701369863013698, + "grad_norm": 2.3405187129974365, + "learning_rate": 3.2331437855402113e-06, + "loss": 1.8365, + "step": 1159 + }, + { + "epoch": 1.2712328767123289, + "grad_norm": 2.3855013847351074, + "learning_rate": 3.2311129163281885e-06, + "loss": 1.8973, + "step": 1160 + }, + { + "epoch": 1.2712328767123289, + "eval_loss": 2.005983591079712, + "eval_runtime": 37.3061, + "eval_samples_per_second": 26.805, + "eval_steps_per_second": 1.126, + "step": 1160 + }, + { + "epoch": 1.2723287671232877, + "grad_norm": 2.2767140865325928, + "learning_rate": 3.229082047116166e-06, + "loss": 1.7791, + "step": 1161 + }, + { + "epoch": 1.2734246575342465, + "grad_norm": 2.289494276046753, + "learning_rate": 3.227051177904143e-06, + "loss": 1.8508, + "step": 1162 + }, + { + "epoch": 1.2745205479452055, + "grad_norm": 2.1836023330688477, + "learning_rate": 3.2250203086921204e-06, + "loss": 1.8275, + "step": 1163 + }, + { + "epoch": 1.2756164383561643, + "grad_norm": 2.1917505264282227, + "learning_rate": 3.2229894394800977e-06, + "loss": 1.8289, + "step": 1164 + }, + { + "epoch": 1.2767123287671232, + "grad_norm": 2.315983295440674, + "learning_rate": 3.220958570268075e-06, + "loss": 1.8655, + "step": 1165 + }, + { + "epoch": 1.2778082191780822, + "grad_norm": 2.3628008365631104, + "learning_rate": 3.218927701056052e-06, + "loss": 1.7396, + "step": 1166 + }, + { + "epoch": 1.2789041095890412, + "grad_norm": 2.3200488090515137, + "learning_rate": 3.2168968318440295e-06, + "loss": 1.8196, + "step": 1167 + }, + { + "epoch": 1.28, + "grad_norm": 2.2591991424560547, + "learning_rate": 3.2148659626320068e-06, + "loss": 1.7323, + "step": 1168 + }, + { + "epoch": 1.2810958904109588, + "grad_norm": 2.2458770275115967, + "learning_rate": 3.212835093419984e-06, + "loss": 1.8881, + "step": 1169 + }, + { + "epoch": 1.2821917808219179, + "grad_norm": 2.202010154724121, + "learning_rate": 3.2108042242079613e-06, + "loss": 1.7776, + "step": 1170 + }, + { + "epoch": 1.2821917808219179, + "eval_loss": 2.005174398422241, + "eval_runtime": 37.3597, + "eval_samples_per_second": 26.767, + "eval_steps_per_second": 1.124, + "step": 1170 + }, + { + "epoch": 1.2832876712328767, + "grad_norm": 2.2375359535217285, + "learning_rate": 3.2087733549959386e-06, + "loss": 1.7994, + "step": 1171 + }, + { + "epoch": 1.2843835616438355, + "grad_norm": 2.3152217864990234, + "learning_rate": 3.206742485783916e-06, + "loss": 1.8767, + "step": 1172 + }, + { + "epoch": 1.2854794520547945, + "grad_norm": 2.3317618370056152, + "learning_rate": 3.2047116165718927e-06, + "loss": 1.772, + "step": 1173 + }, + { + "epoch": 1.2865753424657536, + "grad_norm": 2.249943971633911, + "learning_rate": 3.2026807473598704e-06, + "loss": 1.7793, + "step": 1174 + }, + { + "epoch": 1.2876712328767124, + "grad_norm": 2.320481538772583, + "learning_rate": 3.2006498781478473e-06, + "loss": 1.8275, + "step": 1175 + }, + { + "epoch": 1.2887671232876712, + "grad_norm": 2.21785569190979, + "learning_rate": 3.198619008935825e-06, + "loss": 1.8449, + "step": 1176 + }, + { + "epoch": 1.2898630136986302, + "grad_norm": 2.469965934753418, + "learning_rate": 3.196588139723802e-06, + "loss": 1.779, + "step": 1177 + }, + { + "epoch": 1.290958904109589, + "grad_norm": 2.1989049911499023, + "learning_rate": 3.1945572705117795e-06, + "loss": 1.8188, + "step": 1178 + }, + { + "epoch": 1.2920547945205478, + "grad_norm": 2.293513298034668, + "learning_rate": 3.1925264012997564e-06, + "loss": 1.8377, + "step": 1179 + }, + { + "epoch": 1.2931506849315069, + "grad_norm": 2.3462443351745605, + "learning_rate": 3.190495532087734e-06, + "loss": 1.8418, + "step": 1180 + }, + { + "epoch": 1.2931506849315069, + "eval_loss": 2.0061001777648926, + "eval_runtime": 37.2951, + "eval_samples_per_second": 26.813, + "eval_steps_per_second": 1.126, + "step": 1180 + }, + { + "epoch": 1.2942465753424657, + "grad_norm": 2.2071478366851807, + "learning_rate": 3.188464662875711e-06, + "loss": 1.8419, + "step": 1181 + }, + { + "epoch": 1.2953424657534247, + "grad_norm": 2.233333110809326, + "learning_rate": 3.1864337936636886e-06, + "loss": 1.8931, + "step": 1182 + }, + { + "epoch": 1.2964383561643835, + "grad_norm": 2.383085250854492, + "learning_rate": 3.1844029244516655e-06, + "loss": 1.8047, + "step": 1183 + }, + { + "epoch": 1.2975342465753426, + "grad_norm": 2.473808765411377, + "learning_rate": 3.182372055239643e-06, + "loss": 1.7794, + "step": 1184 + }, + { + "epoch": 1.2986301369863014, + "grad_norm": 2.3357670307159424, + "learning_rate": 3.18034118602762e-06, + "loss": 1.7717, + "step": 1185 + }, + { + "epoch": 1.2997260273972602, + "grad_norm": 2.376835823059082, + "learning_rate": 3.1783103168155973e-06, + "loss": 1.8534, + "step": 1186 + }, + { + "epoch": 1.3008219178082192, + "grad_norm": 2.447415590286255, + "learning_rate": 3.1762794476035746e-06, + "loss": 1.8251, + "step": 1187 + }, + { + "epoch": 1.301917808219178, + "grad_norm": 2.5145719051361084, + "learning_rate": 3.174248578391552e-06, + "loss": 1.8117, + "step": 1188 + }, + { + "epoch": 1.303013698630137, + "grad_norm": 2.3134653568267822, + "learning_rate": 3.172217709179529e-06, + "loss": 1.8111, + "step": 1189 + }, + { + "epoch": 1.3041095890410959, + "grad_norm": 2.293311834335327, + "learning_rate": 3.1701868399675064e-06, + "loss": 1.791, + "step": 1190 + }, + { + "epoch": 1.3041095890410959, + "eval_loss": 2.0038933753967285, + "eval_runtime": 37.2282, + "eval_samples_per_second": 26.861, + "eval_steps_per_second": 1.128, + "step": 1190 + }, + { + "epoch": 1.305205479452055, + "grad_norm": 2.298020124435425, + "learning_rate": 3.1681559707554837e-06, + "loss": 1.8301, + "step": 1191 + }, + { + "epoch": 1.3063013698630137, + "grad_norm": 2.5261130332946777, + "learning_rate": 3.166125101543461e-06, + "loss": 1.7714, + "step": 1192 + }, + { + "epoch": 1.3073972602739725, + "grad_norm": 2.270820379257202, + "learning_rate": 3.1640942323314382e-06, + "loss": 1.8246, + "step": 1193 + }, + { + "epoch": 1.3084931506849315, + "grad_norm": 2.373260021209717, + "learning_rate": 3.1620633631194155e-06, + "loss": 1.8316, + "step": 1194 + }, + { + "epoch": 1.3095890410958904, + "grad_norm": 2.3265364170074463, + "learning_rate": 3.1600324939073928e-06, + "loss": 1.7679, + "step": 1195 + }, + { + "epoch": 1.3106849315068494, + "grad_norm": 2.2378532886505127, + "learning_rate": 3.15800162469537e-06, + "loss": 1.8149, + "step": 1196 + }, + { + "epoch": 1.3117808219178082, + "grad_norm": 2.2459309101104736, + "learning_rate": 3.155970755483347e-06, + "loss": 1.8086, + "step": 1197 + }, + { + "epoch": 1.3128767123287672, + "grad_norm": 2.2769811153411865, + "learning_rate": 3.1539398862713246e-06, + "loss": 1.8109, + "step": 1198 + }, + { + "epoch": 1.313972602739726, + "grad_norm": 2.310483694076538, + "learning_rate": 3.1519090170593015e-06, + "loss": 1.8297, + "step": 1199 + }, + { + "epoch": 1.3150684931506849, + "grad_norm": 2.302478313446045, + "learning_rate": 3.149878147847279e-06, + "loss": 1.8237, + "step": 1200 + }, + { + "epoch": 1.3150684931506849, + "eval_loss": 2.0011062622070312, + "eval_runtime": 37.2077, + "eval_samples_per_second": 26.876, + "eval_steps_per_second": 1.129, + "step": 1200 + }, + { + "epoch": 1.316164383561644, + "grad_norm": 2.2240796089172363, + "learning_rate": 3.147847278635256e-06, + "loss": 1.8745, + "step": 1201 + }, + { + "epoch": 1.3172602739726027, + "grad_norm": 2.303537130355835, + "learning_rate": 3.1458164094232337e-06, + "loss": 1.9051, + "step": 1202 + }, + { + "epoch": 1.3183561643835615, + "grad_norm": 2.4038636684417725, + "learning_rate": 3.1437855402112106e-06, + "loss": 1.7974, + "step": 1203 + }, + { + "epoch": 1.3194520547945205, + "grad_norm": 2.3458874225616455, + "learning_rate": 3.1417546709991883e-06, + "loss": 1.7601, + "step": 1204 + }, + { + "epoch": 1.3205479452054796, + "grad_norm": 2.3399100303649902, + "learning_rate": 3.139723801787165e-06, + "loss": 1.8314, + "step": 1205 + }, + { + "epoch": 1.3216438356164384, + "grad_norm": 2.293229818344116, + "learning_rate": 3.137692932575143e-06, + "loss": 1.8117, + "step": 1206 + }, + { + "epoch": 1.3227397260273972, + "grad_norm": 2.5378005504608154, + "learning_rate": 3.1356620633631197e-06, + "loss": 1.8614, + "step": 1207 + }, + { + "epoch": 1.3238356164383562, + "grad_norm": 2.3644745349884033, + "learning_rate": 3.1336311941510974e-06, + "loss": 1.7843, + "step": 1208 + }, + { + "epoch": 1.324931506849315, + "grad_norm": 2.1592626571655273, + "learning_rate": 3.1316003249390742e-06, + "loss": 1.8128, + "step": 1209 + }, + { + "epoch": 1.3260273972602739, + "grad_norm": 2.155649185180664, + "learning_rate": 3.129569455727051e-06, + "loss": 1.8636, + "step": 1210 + }, + { + "epoch": 1.3260273972602739, + "eval_loss": 1.9952574968338013, + "eval_runtime": 37.3004, + "eval_samples_per_second": 26.809, + "eval_steps_per_second": 1.126, + "step": 1210 + }, + { + "epoch": 1.3271232876712329, + "grad_norm": 2.1906397342681885, + "learning_rate": 3.1275385865150288e-06, + "loss": 1.8583, + "step": 1211 + }, + { + "epoch": 1.3282191780821917, + "grad_norm": 2.2708194255828857, + "learning_rate": 3.1255077173030056e-06, + "loss": 1.8363, + "step": 1212 + }, + { + "epoch": 1.3293150684931507, + "grad_norm": 2.1753969192504883, + "learning_rate": 3.1234768480909833e-06, + "loss": 1.8372, + "step": 1213 + }, + { + "epoch": 1.3304109589041095, + "grad_norm": 2.229363441467285, + "learning_rate": 3.12144597887896e-06, + "loss": 1.8179, + "step": 1214 + }, + { + "epoch": 1.3315068493150686, + "grad_norm": 2.362152338027954, + "learning_rate": 3.119415109666938e-06, + "loss": 1.8596, + "step": 1215 + }, + { + "epoch": 1.3326027397260274, + "grad_norm": 2.3278255462646484, + "learning_rate": 3.1173842404549147e-06, + "loss": 1.8778, + "step": 1216 + }, + { + "epoch": 1.3336986301369862, + "grad_norm": 2.477710247039795, + "learning_rate": 3.1153533712428924e-06, + "loss": 1.8297, + "step": 1217 + }, + { + "epoch": 1.3347945205479452, + "grad_norm": 2.2611043453216553, + "learning_rate": 3.1133225020308693e-06, + "loss": 1.8374, + "step": 1218 + }, + { + "epoch": 1.335890410958904, + "grad_norm": 2.198275566101074, + "learning_rate": 3.111291632818847e-06, + "loss": 1.8686, + "step": 1219 + }, + { + "epoch": 1.336986301369863, + "grad_norm": 2.227508068084717, + "learning_rate": 3.109260763606824e-06, + "loss": 1.8642, + "step": 1220 + }, + { + "epoch": 1.336986301369863, + "eval_loss": 1.9998422861099243, + "eval_runtime": 37.5152, + "eval_samples_per_second": 26.656, + "eval_steps_per_second": 1.12, + "step": 1220 + }, + { + "epoch": 1.3380821917808219, + "grad_norm": 2.386310338973999, + "learning_rate": 3.1072298943948015e-06, + "loss": 1.8347, + "step": 1221 + }, + { + "epoch": 1.339178082191781, + "grad_norm": 2.290313482284546, + "learning_rate": 3.1051990251827784e-06, + "loss": 1.8455, + "step": 1222 + }, + { + "epoch": 1.3402739726027397, + "grad_norm": 2.3063457012176514, + "learning_rate": 3.1031681559707557e-06, + "loss": 1.811, + "step": 1223 + }, + { + "epoch": 1.3413698630136985, + "grad_norm": 2.297840118408203, + "learning_rate": 3.101137286758733e-06, + "loss": 1.8558, + "step": 1224 + }, + { + "epoch": 1.3424657534246576, + "grad_norm": 2.40864896774292, + "learning_rate": 3.09910641754671e-06, + "loss": 1.8604, + "step": 1225 + }, + { + "epoch": 1.3435616438356164, + "grad_norm": 2.397599458694458, + "learning_rate": 3.0970755483346875e-06, + "loss": 1.8556, + "step": 1226 + }, + { + "epoch": 1.3446575342465754, + "grad_norm": 2.409029245376587, + "learning_rate": 3.0950446791226648e-06, + "loss": 1.8424, + "step": 1227 + }, + { + "epoch": 1.3457534246575342, + "grad_norm": 2.2789814472198486, + "learning_rate": 3.093013809910642e-06, + "loss": 1.8945, + "step": 1228 + }, + { + "epoch": 1.3468493150684933, + "grad_norm": 2.3749048709869385, + "learning_rate": 3.0909829406986193e-06, + "loss": 1.8387, + "step": 1229 + }, + { + "epoch": 1.347945205479452, + "grad_norm": 2.44486141204834, + "learning_rate": 3.0889520714865966e-06, + "loss": 1.8838, + "step": 1230 + }, + { + "epoch": 1.347945205479452, + "eval_loss": 1.9961321353912354, + "eval_runtime": 37.6273, + "eval_samples_per_second": 26.576, + "eval_steps_per_second": 1.116, + "step": 1230 + }, + { + "epoch": 1.3490410958904109, + "grad_norm": 2.2852954864501953, + "learning_rate": 3.086921202274574e-06, + "loss": 1.8551, + "step": 1231 + }, + { + "epoch": 1.35013698630137, + "grad_norm": 2.3131206035614014, + "learning_rate": 3.084890333062551e-06, + "loss": 1.8183, + "step": 1232 + }, + { + "epoch": 1.3512328767123287, + "grad_norm": 2.4095048904418945, + "learning_rate": 3.0828594638505284e-06, + "loss": 1.7946, + "step": 1233 + }, + { + "epoch": 1.3523287671232875, + "grad_norm": 2.429971933364868, + "learning_rate": 3.0808285946385057e-06, + "loss": 1.8349, + "step": 1234 + }, + { + "epoch": 1.3534246575342466, + "grad_norm": 2.2360565662384033, + "learning_rate": 3.078797725426483e-06, + "loss": 1.794, + "step": 1235 + }, + { + "epoch": 1.3545205479452056, + "grad_norm": 2.3630571365356445, + "learning_rate": 3.07676685621446e-06, + "loss": 1.7377, + "step": 1236 + }, + { + "epoch": 1.3556164383561644, + "grad_norm": 2.2604660987854004, + "learning_rate": 3.0747359870024375e-06, + "loss": 1.8239, + "step": 1237 + }, + { + "epoch": 1.3567123287671232, + "grad_norm": 2.209717035293579, + "learning_rate": 3.0727051177904144e-06, + "loss": 1.7835, + "step": 1238 + }, + { + "epoch": 1.3578082191780823, + "grad_norm": 2.3221793174743652, + "learning_rate": 3.070674248578392e-06, + "loss": 1.8192, + "step": 1239 + }, + { + "epoch": 1.358904109589041, + "grad_norm": 2.357313394546509, + "learning_rate": 3.068643379366369e-06, + "loss": 1.8246, + "step": 1240 + }, + { + "epoch": 1.358904109589041, + "eval_loss": 1.9935922622680664, + "eval_runtime": 37.2832, + "eval_samples_per_second": 26.822, + "eval_steps_per_second": 1.127, + "step": 1240 + }, + { + "epoch": 1.3599999999999999, + "grad_norm": 2.2563390731811523, + "learning_rate": 3.0666125101543466e-06, + "loss": 1.8175, + "step": 1241 + }, + { + "epoch": 1.361095890410959, + "grad_norm": 2.2531445026397705, + "learning_rate": 3.0645816409423235e-06, + "loss": 1.7713, + "step": 1242 + }, + { + "epoch": 1.362191780821918, + "grad_norm": 2.2369470596313477, + "learning_rate": 3.062550771730301e-06, + "loss": 1.8377, + "step": 1243 + }, + { + "epoch": 1.3632876712328768, + "grad_norm": 2.4063796997070312, + "learning_rate": 3.060519902518278e-06, + "loss": 1.8116, + "step": 1244 + }, + { + "epoch": 1.3643835616438356, + "grad_norm": 2.4628145694732666, + "learning_rate": 3.0584890333062557e-06, + "loss": 1.7998, + "step": 1245 + }, + { + "epoch": 1.3654794520547946, + "grad_norm": 2.3018743991851807, + "learning_rate": 3.0564581640942326e-06, + "loss": 1.7413, + "step": 1246 + }, + { + "epoch": 1.3665753424657534, + "grad_norm": 2.2432262897491455, + "learning_rate": 3.0544272948822094e-06, + "loss": 1.82, + "step": 1247 + }, + { + "epoch": 1.3676712328767122, + "grad_norm": 2.368435859680176, + "learning_rate": 3.052396425670187e-06, + "loss": 1.7976, + "step": 1248 + }, + { + "epoch": 1.3687671232876713, + "grad_norm": 2.2668025493621826, + "learning_rate": 3.050365556458164e-06, + "loss": 1.8379, + "step": 1249 + }, + { + "epoch": 1.36986301369863, + "grad_norm": 2.2622179985046387, + "learning_rate": 3.0483346872461417e-06, + "loss": 1.794, + "step": 1250 + }, + { + "epoch": 1.36986301369863, + "eval_loss": 1.9959099292755127, + "eval_runtime": 37.3358, + "eval_samples_per_second": 26.784, + "eval_steps_per_second": 1.125, + "step": 1250 + }, + { + "epoch": 1.370958904109589, + "grad_norm": 2.3414738178253174, + "learning_rate": 3.0463038180341185e-06, + "loss": 1.7993, + "step": 1251 + }, + { + "epoch": 1.372054794520548, + "grad_norm": 2.2678418159484863, + "learning_rate": 3.0442729488220962e-06, + "loss": 1.8581, + "step": 1252 + }, + { + "epoch": 1.373150684931507, + "grad_norm": 2.268195867538452, + "learning_rate": 3.042242079610073e-06, + "loss": 1.8486, + "step": 1253 + }, + { + "epoch": 1.3742465753424657, + "grad_norm": 2.230173110961914, + "learning_rate": 3.0402112103980508e-06, + "loss": 1.8468, + "step": 1254 + }, + { + "epoch": 1.3753424657534246, + "grad_norm": 2.3020238876342773, + "learning_rate": 3.0381803411860276e-06, + "loss": 1.7623, + "step": 1255 + }, + { + "epoch": 1.3764383561643836, + "grad_norm": 2.385343313217163, + "learning_rate": 3.0361494719740053e-06, + "loss": 1.7976, + "step": 1256 + }, + { + "epoch": 1.3775342465753424, + "grad_norm": 2.344768524169922, + "learning_rate": 3.034118602761982e-06, + "loss": 1.807, + "step": 1257 + }, + { + "epoch": 1.3786301369863014, + "grad_norm": 2.411760091781616, + "learning_rate": 3.03208773354996e-06, + "loss": 1.8079, + "step": 1258 + }, + { + "epoch": 1.3797260273972602, + "grad_norm": 2.3160128593444824, + "learning_rate": 3.0300568643379367e-06, + "loss": 1.7868, + "step": 1259 + }, + { + "epoch": 1.3808219178082193, + "grad_norm": 2.334632635116577, + "learning_rate": 3.028025995125914e-06, + "loss": 1.8466, + "step": 1260 + }, + { + "epoch": 1.3808219178082193, + "eval_loss": 1.9903327226638794, + "eval_runtime": 37.2256, + "eval_samples_per_second": 26.863, + "eval_steps_per_second": 1.128, + "step": 1260 + }, + { + "epoch": 1.381917808219178, + "grad_norm": 2.3443334102630615, + "learning_rate": 3.0259951259138913e-06, + "loss": 1.8302, + "step": 1261 + }, + { + "epoch": 1.383013698630137, + "grad_norm": 2.311563014984131, + "learning_rate": 3.0239642567018686e-06, + "loss": 1.7727, + "step": 1262 + }, + { + "epoch": 1.384109589041096, + "grad_norm": 2.2501776218414307, + "learning_rate": 3.021933387489846e-06, + "loss": 1.8058, + "step": 1263 + }, + { + "epoch": 1.3852054794520547, + "grad_norm": 2.259157657623291, + "learning_rate": 3.019902518277823e-06, + "loss": 1.7627, + "step": 1264 + }, + { + "epoch": 1.3863013698630138, + "grad_norm": 2.4380314350128174, + "learning_rate": 3.0178716490658004e-06, + "loss": 1.8237, + "step": 1265 + }, + { + "epoch": 1.3873972602739726, + "grad_norm": 2.3203837871551514, + "learning_rate": 3.0158407798537777e-06, + "loss": 1.8557, + "step": 1266 + }, + { + "epoch": 1.3884931506849316, + "grad_norm": 2.220996141433716, + "learning_rate": 3.013809910641755e-06, + "loss": 1.7641, + "step": 1267 + }, + { + "epoch": 1.3895890410958904, + "grad_norm": 2.190006971359253, + "learning_rate": 3.0117790414297322e-06, + "loss": 1.7705, + "step": 1268 + }, + { + "epoch": 1.3906849315068492, + "grad_norm": 2.229748249053955, + "learning_rate": 3.0097481722177095e-06, + "loss": 1.7709, + "step": 1269 + }, + { + "epoch": 1.3917808219178083, + "grad_norm": 2.3755619525909424, + "learning_rate": 3.0077173030056868e-06, + "loss": 1.8172, + "step": 1270 + }, + { + "epoch": 1.3917808219178083, + "eval_loss": 1.992952585220337, + "eval_runtime": 37.2403, + "eval_samples_per_second": 26.853, + "eval_steps_per_second": 1.128, + "step": 1270 + }, + { + "epoch": 1.392876712328767, + "grad_norm": 2.2514004707336426, + "learning_rate": 3.005686433793664e-06, + "loss": 1.7719, + "step": 1271 + }, + { + "epoch": 1.393972602739726, + "grad_norm": 2.3010191917419434, + "learning_rate": 3.0036555645816413e-06, + "loss": 1.8369, + "step": 1272 + }, + { + "epoch": 1.395068493150685, + "grad_norm": 2.68790602684021, + "learning_rate": 3.001624695369618e-06, + "loss": 1.7974, + "step": 1273 + }, + { + "epoch": 1.396164383561644, + "grad_norm": 2.454904794692993, + "learning_rate": 2.999593826157596e-06, + "loss": 1.8509, + "step": 1274 + }, + { + "epoch": 1.3972602739726028, + "grad_norm": 2.251556396484375, + "learning_rate": 2.9975629569455727e-06, + "loss": 1.8011, + "step": 1275 + }, + { + "epoch": 1.3983561643835616, + "grad_norm": 2.3789098262786865, + "learning_rate": 2.9955320877335504e-06, + "loss": 1.8488, + "step": 1276 + }, + { + "epoch": 1.3994520547945206, + "grad_norm": 2.2659196853637695, + "learning_rate": 2.9935012185215273e-06, + "loss": 1.7946, + "step": 1277 + }, + { + "epoch": 1.4005479452054794, + "grad_norm": 2.2148971557617188, + "learning_rate": 2.991470349309505e-06, + "loss": 1.8388, + "step": 1278 + }, + { + "epoch": 1.4016438356164382, + "grad_norm": 2.1883089542388916, + "learning_rate": 2.989439480097482e-06, + "loss": 1.7925, + "step": 1279 + }, + { + "epoch": 1.4027397260273973, + "grad_norm": 2.290208339691162, + "learning_rate": 2.9874086108854595e-06, + "loss": 1.8099, + "step": 1280 + }, + { + "epoch": 1.4027397260273973, + "eval_loss": 1.9924782514572144, + "eval_runtime": 37.1891, + "eval_samples_per_second": 26.89, + "eval_steps_per_second": 1.129, + "step": 1280 + }, + { + "epoch": 1.403835616438356, + "grad_norm": 2.121649980545044, + "learning_rate": 2.9853777416734364e-06, + "loss": 1.7991, + "step": 1281 + }, + { + "epoch": 1.4049315068493151, + "grad_norm": 2.3528690338134766, + "learning_rate": 2.983346872461414e-06, + "loss": 1.8487, + "step": 1282 + }, + { + "epoch": 1.406027397260274, + "grad_norm": 2.391491174697876, + "learning_rate": 2.981316003249391e-06, + "loss": 1.8564, + "step": 1283 + }, + { + "epoch": 1.407123287671233, + "grad_norm": 2.336615800857544, + "learning_rate": 2.9792851340373678e-06, + "loss": 1.8216, + "step": 1284 + }, + { + "epoch": 1.4082191780821918, + "grad_norm": 2.2958548069000244, + "learning_rate": 2.9772542648253455e-06, + "loss": 1.8003, + "step": 1285 + }, + { + "epoch": 1.4093150684931506, + "grad_norm": 2.2062184810638428, + "learning_rate": 2.9752233956133223e-06, + "loss": 1.8488, + "step": 1286 + }, + { + "epoch": 1.4104109589041096, + "grad_norm": 2.486133575439453, + "learning_rate": 2.9731925264013e-06, + "loss": 1.7914, + "step": 1287 + }, + { + "epoch": 1.4115068493150684, + "grad_norm": 2.2933201789855957, + "learning_rate": 2.971161657189277e-06, + "loss": 1.8164, + "step": 1288 + }, + { + "epoch": 1.4126027397260275, + "grad_norm": 2.3301920890808105, + "learning_rate": 2.9691307879772546e-06, + "loss": 1.7883, + "step": 1289 + }, + { + "epoch": 1.4136986301369863, + "grad_norm": 2.248832941055298, + "learning_rate": 2.9670999187652314e-06, + "loss": 1.7916, + "step": 1290 + }, + { + "epoch": 1.4136986301369863, + "eval_loss": 1.9919097423553467, + "eval_runtime": 37.1283, + "eval_samples_per_second": 26.934, + "eval_steps_per_second": 1.131, + "step": 1290 + }, + { + "epoch": 1.4147945205479453, + "grad_norm": 2.37760853767395, + "learning_rate": 2.965069049553209e-06, + "loss": 1.7854, + "step": 1291 + }, + { + "epoch": 1.415890410958904, + "grad_norm": 2.327718734741211, + "learning_rate": 2.963038180341186e-06, + "loss": 1.874, + "step": 1292 + }, + { + "epoch": 1.416986301369863, + "grad_norm": 2.2413878440856934, + "learning_rate": 2.9610073111291637e-06, + "loss": 1.7591, + "step": 1293 + }, + { + "epoch": 1.418082191780822, + "grad_norm": 2.2203164100646973, + "learning_rate": 2.9589764419171405e-06, + "loss": 1.8466, + "step": 1294 + }, + { + "epoch": 1.4191780821917808, + "grad_norm": 2.2070505619049072, + "learning_rate": 2.9569455727051182e-06, + "loss": 1.8689, + "step": 1295 + }, + { + "epoch": 1.4202739726027398, + "grad_norm": 2.2122223377227783, + "learning_rate": 2.954914703493095e-06, + "loss": 1.7689, + "step": 1296 + }, + { + "epoch": 1.4213698630136986, + "grad_norm": 2.331303358078003, + "learning_rate": 2.9528838342810724e-06, + "loss": 1.8089, + "step": 1297 + }, + { + "epoch": 1.4224657534246576, + "grad_norm": 2.408184289932251, + "learning_rate": 2.9508529650690496e-06, + "loss": 1.7847, + "step": 1298 + }, + { + "epoch": 1.4235616438356165, + "grad_norm": 2.212766170501709, + "learning_rate": 2.948822095857027e-06, + "loss": 1.7899, + "step": 1299 + }, + { + "epoch": 1.4246575342465753, + "grad_norm": 2.259953737258911, + "learning_rate": 2.946791226645004e-06, + "loss": 1.8435, + "step": 1300 + }, + { + "epoch": 1.4246575342465753, + "eval_loss": 1.9893293380737305, + "eval_runtime": 37.1675, + "eval_samples_per_second": 26.905, + "eval_steps_per_second": 1.13, + "step": 1300 + }, + { + "epoch": 1.4257534246575343, + "grad_norm": 2.342400312423706, + "learning_rate": 2.9447603574329815e-06, + "loss": 1.7779, + "step": 1301 + }, + { + "epoch": 1.426849315068493, + "grad_norm": 2.237618923187256, + "learning_rate": 2.9427294882209587e-06, + "loss": 1.7509, + "step": 1302 + }, + { + "epoch": 1.427945205479452, + "grad_norm": 2.3418633937835693, + "learning_rate": 2.940698619008936e-06, + "loss": 1.7963, + "step": 1303 + }, + { + "epoch": 1.429041095890411, + "grad_norm": 2.470827579498291, + "learning_rate": 2.9386677497969133e-06, + "loss": 1.8045, + "step": 1304 + }, + { + "epoch": 1.43013698630137, + "grad_norm": 2.508345365524292, + "learning_rate": 2.9366368805848906e-06, + "loss": 1.7926, + "step": 1305 + }, + { + "epoch": 1.4312328767123288, + "grad_norm": 2.261319160461426, + "learning_rate": 2.934606011372868e-06, + "loss": 1.831, + "step": 1306 + }, + { + "epoch": 1.4323287671232876, + "grad_norm": 2.2703914642333984, + "learning_rate": 2.932575142160845e-06, + "loss": 1.8352, + "step": 1307 + }, + { + "epoch": 1.4334246575342466, + "grad_norm": 2.243626117706299, + "learning_rate": 2.9305442729488224e-06, + "loss": 1.8162, + "step": 1308 + }, + { + "epoch": 1.4345205479452054, + "grad_norm": 2.3859238624572754, + "learning_rate": 2.9285134037367997e-06, + "loss": 1.8171, + "step": 1309 + }, + { + "epoch": 1.4356164383561643, + "grad_norm": 2.3350934982299805, + "learning_rate": 2.9264825345247765e-06, + "loss": 1.8066, + "step": 1310 + }, + { + "epoch": 1.4356164383561643, + "eval_loss": 1.9905259609222412, + "eval_runtime": 37.2009, + "eval_samples_per_second": 26.881, + "eval_steps_per_second": 1.129, + "step": 1310 + }, + { + "epoch": 1.4367123287671233, + "grad_norm": 2.1943347454071045, + "learning_rate": 2.9244516653127542e-06, + "loss": 1.8435, + "step": 1311 + }, + { + "epoch": 1.4378082191780823, + "grad_norm": 2.429128885269165, + "learning_rate": 2.922420796100731e-06, + "loss": 1.8179, + "step": 1312 + }, + { + "epoch": 1.4389041095890411, + "grad_norm": 2.3677234649658203, + "learning_rate": 2.9203899268887088e-06, + "loss": 1.7245, + "step": 1313 + }, + { + "epoch": 1.44, + "grad_norm": 2.255570888519287, + "learning_rate": 2.9183590576766856e-06, + "loss": 1.8545, + "step": 1314 + }, + { + "epoch": 1.441095890410959, + "grad_norm": 2.5176682472229004, + "learning_rate": 2.9163281884646633e-06, + "loss": 1.7509, + "step": 1315 + }, + { + "epoch": 1.4421917808219178, + "grad_norm": 2.29972767829895, + "learning_rate": 2.91429731925264e-06, + "loss": 1.8116, + "step": 1316 + }, + { + "epoch": 1.4432876712328766, + "grad_norm": 2.2853286266326904, + "learning_rate": 2.912266450040618e-06, + "loss": 1.8552, + "step": 1317 + }, + { + "epoch": 1.4443835616438356, + "grad_norm": 2.505075216293335, + "learning_rate": 2.9102355808285947e-06, + "loss": 1.811, + "step": 1318 + }, + { + "epoch": 1.4454794520547944, + "grad_norm": 2.229064464569092, + "learning_rate": 2.9082047116165724e-06, + "loss": 1.8068, + "step": 1319 + }, + { + "epoch": 1.4465753424657535, + "grad_norm": 2.199187994003296, + "learning_rate": 2.9061738424045493e-06, + "loss": 1.7825, + "step": 1320 + }, + { + "epoch": 1.4465753424657535, + "eval_loss": 1.9851222038269043, + "eval_runtime": 37.4614, + "eval_samples_per_second": 26.694, + "eval_steps_per_second": 1.121, + "step": 1320 + }, + { + "epoch": 1.4476712328767123, + "grad_norm": 2.4318959712982178, + "learning_rate": 2.904142973192527e-06, + "loss": 1.77, + "step": 1321 + }, + { + "epoch": 1.4487671232876713, + "grad_norm": 2.480703592300415, + "learning_rate": 2.902112103980504e-06, + "loss": 1.8774, + "step": 1322 + }, + { + "epoch": 1.4498630136986301, + "grad_norm": 2.367624521255493, + "learning_rate": 2.900081234768481e-06, + "loss": 1.818, + "step": 1323 + }, + { + "epoch": 1.450958904109589, + "grad_norm": 2.245974540710449, + "learning_rate": 2.8980503655564584e-06, + "loss": 1.8054, + "step": 1324 + }, + { + "epoch": 1.452054794520548, + "grad_norm": 2.352687120437622, + "learning_rate": 2.8960194963444357e-06, + "loss": 1.7922, + "step": 1325 + }, + { + "epoch": 1.4531506849315068, + "grad_norm": 2.308253526687622, + "learning_rate": 2.893988627132413e-06, + "loss": 1.7637, + "step": 1326 + }, + { + "epoch": 1.4542465753424658, + "grad_norm": 2.309495687484741, + "learning_rate": 2.8919577579203902e-06, + "loss": 1.7905, + "step": 1327 + }, + { + "epoch": 1.4553424657534246, + "grad_norm": 2.354365348815918, + "learning_rate": 2.8899268887083675e-06, + "loss": 1.7321, + "step": 1328 + }, + { + "epoch": 1.4564383561643837, + "grad_norm": 2.3886210918426514, + "learning_rate": 2.8878960194963448e-06, + "loss": 1.7932, + "step": 1329 + }, + { + "epoch": 1.4575342465753425, + "grad_norm": 2.395714044570923, + "learning_rate": 2.885865150284322e-06, + "loss": 1.813, + "step": 1330 + }, + { + "epoch": 1.4575342465753425, + "eval_loss": 1.9851990938186646, + "eval_runtime": 37.2966, + "eval_samples_per_second": 26.812, + "eval_steps_per_second": 1.126, + "step": 1330 + }, + { + "epoch": 1.4586301369863013, + "grad_norm": 2.433069944381714, + "learning_rate": 2.8838342810722993e-06, + "loss": 1.8346, + "step": 1331 + }, + { + "epoch": 1.4597260273972603, + "grad_norm": 2.295894145965576, + "learning_rate": 2.8818034118602766e-06, + "loss": 1.7855, + "step": 1332 + }, + { + "epoch": 1.4608219178082191, + "grad_norm": 2.495742082595825, + "learning_rate": 2.879772542648254e-06, + "loss": 1.7601, + "step": 1333 + }, + { + "epoch": 1.4619178082191782, + "grad_norm": 2.35984206199646, + "learning_rate": 2.8777416734362307e-06, + "loss": 1.8329, + "step": 1334 + }, + { + "epoch": 1.463013698630137, + "grad_norm": 2.39632248878479, + "learning_rate": 2.8757108042242084e-06, + "loss": 1.7426, + "step": 1335 + }, + { + "epoch": 1.464109589041096, + "grad_norm": 2.4151153564453125, + "learning_rate": 2.8736799350121853e-06, + "loss": 1.8899, + "step": 1336 + }, + { + "epoch": 1.4652054794520548, + "grad_norm": 2.43914532661438, + "learning_rate": 2.871649065800163e-06, + "loss": 1.8464, + "step": 1337 + }, + { + "epoch": 1.4663013698630136, + "grad_norm": 2.44490122795105, + "learning_rate": 2.86961819658814e-06, + "loss": 1.8643, + "step": 1338 + }, + { + "epoch": 1.4673972602739727, + "grad_norm": 2.2639644145965576, + "learning_rate": 2.8675873273761175e-06, + "loss": 1.792, + "step": 1339 + }, + { + "epoch": 1.4684931506849315, + "grad_norm": 2.2803847789764404, + "learning_rate": 2.8655564581640944e-06, + "loss": 1.8109, + "step": 1340 + }, + { + "epoch": 1.4684931506849315, + "eval_loss": 1.9856781959533691, + "eval_runtime": 37.2519, + "eval_samples_per_second": 26.844, + "eval_steps_per_second": 1.127, + "step": 1340 + }, + { + "epoch": 1.4695890410958903, + "grad_norm": 2.494727849960327, + "learning_rate": 2.863525588952072e-06, + "loss": 1.8032, + "step": 1341 + }, + { + "epoch": 1.4706849315068493, + "grad_norm": 2.239595413208008, + "learning_rate": 2.861494719740049e-06, + "loss": 1.8659, + "step": 1342 + }, + { + "epoch": 1.4717808219178083, + "grad_norm": 2.2922589778900146, + "learning_rate": 2.8594638505280266e-06, + "loss": 1.7587, + "step": 1343 + }, + { + "epoch": 1.4728767123287672, + "grad_norm": 2.345224142074585, + "learning_rate": 2.8574329813160035e-06, + "loss": 1.7602, + "step": 1344 + }, + { + "epoch": 1.473972602739726, + "grad_norm": 2.3415307998657227, + "learning_rate": 2.855402112103981e-06, + "loss": 1.8044, + "step": 1345 + }, + { + "epoch": 1.475068493150685, + "grad_norm": 2.331557035446167, + "learning_rate": 2.853371242891958e-06, + "loss": 1.7808, + "step": 1346 + }, + { + "epoch": 1.4761643835616438, + "grad_norm": 2.293572425842285, + "learning_rate": 2.851340373679935e-06, + "loss": 1.8017, + "step": 1347 + }, + { + "epoch": 1.4772602739726026, + "grad_norm": 2.3477773666381836, + "learning_rate": 2.8493095044679126e-06, + "loss": 1.8113, + "step": 1348 + }, + { + "epoch": 1.4783561643835617, + "grad_norm": 2.3740994930267334, + "learning_rate": 2.8472786352558894e-06, + "loss": 1.8801, + "step": 1349 + }, + { + "epoch": 1.4794520547945205, + "grad_norm": 2.4202206134796143, + "learning_rate": 2.845247766043867e-06, + "loss": 1.821, + "step": 1350 + }, + { + "epoch": 1.4794520547945205, + "eval_loss": 1.9867810010910034, + "eval_runtime": 37.2597, + "eval_samples_per_second": 26.839, + "eval_steps_per_second": 1.127, + "step": 1350 + }, + { + "epoch": 1.4805479452054795, + "grad_norm": 2.2323198318481445, + "learning_rate": 2.843216896831844e-06, + "loss": 1.7574, + "step": 1351 + }, + { + "epoch": 1.4816438356164383, + "grad_norm": 2.2192625999450684, + "learning_rate": 2.8411860276198217e-06, + "loss": 1.7662, + "step": 1352 + }, + { + "epoch": 1.4827397260273973, + "grad_norm": 2.276507616043091, + "learning_rate": 2.8391551584077985e-06, + "loss": 1.7748, + "step": 1353 + }, + { + "epoch": 1.4838356164383562, + "grad_norm": 2.292773723602295, + "learning_rate": 2.8371242891957762e-06, + "loss": 1.8388, + "step": 1354 + }, + { + "epoch": 1.484931506849315, + "grad_norm": 2.3127009868621826, + "learning_rate": 2.835093419983753e-06, + "loss": 1.861, + "step": 1355 + }, + { + "epoch": 1.486027397260274, + "grad_norm": 2.2752184867858887, + "learning_rate": 2.833062550771731e-06, + "loss": 1.7883, + "step": 1356 + }, + { + "epoch": 1.4871232876712328, + "grad_norm": 2.288158416748047, + "learning_rate": 2.8310316815597076e-06, + "loss": 1.7963, + "step": 1357 + }, + { + "epoch": 1.4882191780821918, + "grad_norm": 2.3287482261657715, + "learning_rate": 2.8290008123476853e-06, + "loss": 1.7534, + "step": 1358 + }, + { + "epoch": 1.4893150684931507, + "grad_norm": 2.441279888153076, + "learning_rate": 2.826969943135662e-06, + "loss": 1.8073, + "step": 1359 + }, + { + "epoch": 1.4904109589041097, + "grad_norm": 2.3399839401245117, + "learning_rate": 2.8249390739236395e-06, + "loss": 1.7876, + "step": 1360 + }, + { + "epoch": 1.4904109589041097, + "eval_loss": 1.9868030548095703, + "eval_runtime": 37.2891, + "eval_samples_per_second": 26.817, + "eval_steps_per_second": 1.126, + "step": 1360 + }, + { + "epoch": 1.4915068493150685, + "grad_norm": 2.234246253967285, + "learning_rate": 2.8229082047116167e-06, + "loss": 1.8031, + "step": 1361 + }, + { + "epoch": 1.4926027397260273, + "grad_norm": 2.3136820793151855, + "learning_rate": 2.820877335499594e-06, + "loss": 1.7669, + "step": 1362 + }, + { + "epoch": 1.4936986301369863, + "grad_norm": 2.389420747756958, + "learning_rate": 2.8188464662875713e-06, + "loss": 1.7879, + "step": 1363 + }, + { + "epoch": 1.4947945205479451, + "grad_norm": 2.222306489944458, + "learning_rate": 2.8168155970755486e-06, + "loss": 1.7412, + "step": 1364 + }, + { + "epoch": 1.4958904109589042, + "grad_norm": 2.1689319610595703, + "learning_rate": 2.814784727863526e-06, + "loss": 1.7651, + "step": 1365 + }, + { + "epoch": 1.496986301369863, + "grad_norm": 2.262047529220581, + "learning_rate": 2.812753858651503e-06, + "loss": 1.7798, + "step": 1366 + }, + { + "epoch": 1.498082191780822, + "grad_norm": 2.266894817352295, + "learning_rate": 2.8107229894394804e-06, + "loss": 1.8172, + "step": 1367 + }, + { + "epoch": 1.4991780821917808, + "grad_norm": 2.6400444507598877, + "learning_rate": 2.8086921202274577e-06, + "loss": 1.7871, + "step": 1368 + }, + { + "epoch": 1.5002739726027396, + "grad_norm": 2.3868865966796875, + "learning_rate": 2.806661251015435e-06, + "loss": 1.7427, + "step": 1369 + }, + { + "epoch": 1.5013698630136987, + "grad_norm": 2.5206542015075684, + "learning_rate": 2.8046303818034122e-06, + "loss": 1.817, + "step": 1370 + }, + { + "epoch": 1.5013698630136987, + "eval_loss": 1.9870396852493286, + "eval_runtime": 37.2372, + "eval_samples_per_second": 26.855, + "eval_steps_per_second": 1.128, + "step": 1370 + }, + { + "epoch": 1.5024657534246575, + "grad_norm": 2.5236382484436035, + "learning_rate": 2.802599512591389e-06, + "loss": 1.8662, + "step": 1371 + }, + { + "epoch": 1.5035616438356163, + "grad_norm": 2.5095479488372803, + "learning_rate": 2.8005686433793668e-06, + "loss": 1.9168, + "step": 1372 + }, + { + "epoch": 1.5046575342465753, + "grad_norm": 2.307508707046509, + "learning_rate": 2.7985377741673436e-06, + "loss": 1.799, + "step": 1373 + }, + { + "epoch": 1.5057534246575344, + "grad_norm": 2.410879373550415, + "learning_rate": 2.7965069049553213e-06, + "loss": 1.7447, + "step": 1374 + }, + { + "epoch": 1.5068493150684932, + "grad_norm": 2.738558530807495, + "learning_rate": 2.794476035743298e-06, + "loss": 1.8073, + "step": 1375 + }, + { + "epoch": 1.507945205479452, + "grad_norm": 2.3826518058776855, + "learning_rate": 2.792445166531276e-06, + "loss": 1.8506, + "step": 1376 + }, + { + "epoch": 1.509041095890411, + "grad_norm": 2.334123373031616, + "learning_rate": 2.7904142973192527e-06, + "loss": 1.8209, + "step": 1377 + }, + { + "epoch": 1.5101369863013698, + "grad_norm": 2.668513298034668, + "learning_rate": 2.7883834281072304e-06, + "loss": 1.8285, + "step": 1378 + }, + { + "epoch": 1.5112328767123286, + "grad_norm": 2.471938133239746, + "learning_rate": 2.7863525588952073e-06, + "loss": 1.7668, + "step": 1379 + }, + { + "epoch": 1.5123287671232877, + "grad_norm": 2.3712313175201416, + "learning_rate": 2.784321689683185e-06, + "loss": 1.7565, + "step": 1380 + }, + { + "epoch": 1.5123287671232877, + "eval_loss": 1.9826414585113525, + "eval_runtime": 37.2426, + "eval_samples_per_second": 26.851, + "eval_steps_per_second": 1.128, + "step": 1380 + }, + { + "epoch": 1.5134246575342467, + "grad_norm": 2.3364150524139404, + "learning_rate": 2.782290820471162e-06, + "loss": 1.7596, + "step": 1381 + }, + { + "epoch": 1.5145205479452055, + "grad_norm": 2.3088393211364746, + "learning_rate": 2.7802599512591395e-06, + "loss": 1.7568, + "step": 1382 + }, + { + "epoch": 1.5156164383561643, + "grad_norm": 2.2584595680236816, + "learning_rate": 2.7782290820471164e-06, + "loss": 1.833, + "step": 1383 + }, + { + "epoch": 1.5167123287671234, + "grad_norm": 2.2440273761749268, + "learning_rate": 2.7761982128350932e-06, + "loss": 1.8237, + "step": 1384 + }, + { + "epoch": 1.5178082191780822, + "grad_norm": 2.3552212715148926, + "learning_rate": 2.774167343623071e-06, + "loss": 1.8288, + "step": 1385 + }, + { + "epoch": 1.518904109589041, + "grad_norm": 2.3001790046691895, + "learning_rate": 2.772136474411048e-06, + "loss": 1.8519, + "step": 1386 + }, + { + "epoch": 1.52, + "grad_norm": 2.3691201210021973, + "learning_rate": 2.7701056051990255e-06, + "loss": 1.7759, + "step": 1387 + }, + { + "epoch": 1.521095890410959, + "grad_norm": 2.327988624572754, + "learning_rate": 2.7680747359870023e-06, + "loss": 1.7132, + "step": 1388 + }, + { + "epoch": 1.5221917808219179, + "grad_norm": 2.30245304107666, + "learning_rate": 2.76604386677498e-06, + "loss": 1.8167, + "step": 1389 + }, + { + "epoch": 1.5232876712328767, + "grad_norm": 2.290121078491211, + "learning_rate": 2.764012997562957e-06, + "loss": 1.7913, + "step": 1390 + }, + { + "epoch": 1.5232876712328767, + "eval_loss": 1.98329758644104, + "eval_runtime": 37.2462, + "eval_samples_per_second": 26.848, + "eval_steps_per_second": 1.128, + "step": 1390 + }, + { + "epoch": 1.5243835616438357, + "grad_norm": 2.294076681137085, + "learning_rate": 2.7619821283509346e-06, + "loss": 1.8598, + "step": 1391 + }, + { + "epoch": 1.5254794520547945, + "grad_norm": 2.3325250148773193, + "learning_rate": 2.7599512591389114e-06, + "loss": 1.7392, + "step": 1392 + }, + { + "epoch": 1.5265753424657533, + "grad_norm": 2.2802610397338867, + "learning_rate": 2.757920389926889e-06, + "loss": 1.8332, + "step": 1393 + }, + { + "epoch": 1.5276712328767124, + "grad_norm": 2.284207344055176, + "learning_rate": 2.755889520714866e-06, + "loss": 1.7875, + "step": 1394 + }, + { + "epoch": 1.5287671232876714, + "grad_norm": 2.325998306274414, + "learning_rate": 2.7538586515028437e-06, + "loss": 1.7657, + "step": 1395 + }, + { + "epoch": 1.52986301369863, + "grad_norm": 2.285393476486206, + "learning_rate": 2.7518277822908205e-06, + "loss": 1.7416, + "step": 1396 + }, + { + "epoch": 1.530958904109589, + "grad_norm": 2.3110482692718506, + "learning_rate": 2.749796913078798e-06, + "loss": 1.7931, + "step": 1397 + }, + { + "epoch": 1.532054794520548, + "grad_norm": 2.4234025478363037, + "learning_rate": 2.747766043866775e-06, + "loss": 1.8472, + "step": 1398 + }, + { + "epoch": 1.5331506849315069, + "grad_norm": 2.384767532348633, + "learning_rate": 2.7457351746547524e-06, + "loss": 1.7961, + "step": 1399 + }, + { + "epoch": 1.5342465753424657, + "grad_norm": 2.1857433319091797, + "learning_rate": 2.7437043054427297e-06, + "loss": 1.8654, + "step": 1400 + }, + { + "epoch": 1.5342465753424657, + "eval_loss": 1.9802658557891846, + "eval_runtime": 37.1443, + "eval_samples_per_second": 26.922, + "eval_steps_per_second": 1.131, + "step": 1400 + }, + { + "epoch": 1.5353424657534247, + "grad_norm": 2.1302895545959473, + "learning_rate": 2.741673436230707e-06, + "loss": 1.9051, + "step": 1401 + }, + { + "epoch": 1.5364383561643835, + "grad_norm": 2.377307415008545, + "learning_rate": 2.739642567018684e-06, + "loss": 1.7942, + "step": 1402 + }, + { + "epoch": 1.5375342465753423, + "grad_norm": 2.316549062728882, + "learning_rate": 2.7376116978066615e-06, + "loss": 1.7734, + "step": 1403 + }, + { + "epoch": 1.5386301369863014, + "grad_norm": 2.3452532291412354, + "learning_rate": 2.7355808285946388e-06, + "loss": 1.8243, + "step": 1404 + }, + { + "epoch": 1.5397260273972604, + "grad_norm": 2.3139419555664062, + "learning_rate": 2.733549959382616e-06, + "loss": 1.812, + "step": 1405 + }, + { + "epoch": 1.5408219178082192, + "grad_norm": 2.3135571479797363, + "learning_rate": 2.7315190901705933e-06, + "loss": 1.7751, + "step": 1406 + }, + { + "epoch": 1.541917808219178, + "grad_norm": 2.403231620788574, + "learning_rate": 2.7294882209585706e-06, + "loss": 1.8561, + "step": 1407 + }, + { + "epoch": 1.543013698630137, + "grad_norm": 2.4028689861297607, + "learning_rate": 2.727457351746548e-06, + "loss": 1.8249, + "step": 1408 + }, + { + "epoch": 1.5441095890410959, + "grad_norm": 2.4028851985931396, + "learning_rate": 2.725426482534525e-06, + "loss": 1.779, + "step": 1409 + }, + { + "epoch": 1.5452054794520547, + "grad_norm": 2.270709753036499, + "learning_rate": 2.723395613322502e-06, + "loss": 1.7754, + "step": 1410 + }, + { + "epoch": 1.5452054794520547, + "eval_loss": 1.983272910118103, + "eval_runtime": 37.2405, + "eval_samples_per_second": 26.853, + "eval_steps_per_second": 1.128, + "step": 1410 + }, + { + "epoch": 1.5463013698630137, + "grad_norm": 2.2672812938690186, + "learning_rate": 2.7213647441104797e-06, + "loss": 1.814, + "step": 1411 + }, + { + "epoch": 1.5473972602739727, + "grad_norm": 2.3539397716522217, + "learning_rate": 2.7193338748984565e-06, + "loss": 1.7907, + "step": 1412 + }, + { + "epoch": 1.5484931506849315, + "grad_norm": 2.3229010105133057, + "learning_rate": 2.7173030056864342e-06, + "loss": 1.8061, + "step": 1413 + }, + { + "epoch": 1.5495890410958904, + "grad_norm": 2.482874870300293, + "learning_rate": 2.715272136474411e-06, + "loss": 1.9577, + "step": 1414 + }, + { + "epoch": 1.5506849315068494, + "grad_norm": 2.211798906326294, + "learning_rate": 2.713241267262389e-06, + "loss": 1.7309, + "step": 1415 + }, + { + "epoch": 1.5517808219178082, + "grad_norm": 2.3159289360046387, + "learning_rate": 2.7112103980503656e-06, + "loss": 1.8132, + "step": 1416 + }, + { + "epoch": 1.552876712328767, + "grad_norm": 2.486035108566284, + "learning_rate": 2.7091795288383433e-06, + "loss": 1.7734, + "step": 1417 + }, + { + "epoch": 1.553972602739726, + "grad_norm": 2.3025195598602295, + "learning_rate": 2.70714865962632e-06, + "loss": 1.8232, + "step": 1418 + }, + { + "epoch": 1.555068493150685, + "grad_norm": 2.234581470489502, + "learning_rate": 2.705117790414298e-06, + "loss": 1.7814, + "step": 1419 + }, + { + "epoch": 1.5561643835616439, + "grad_norm": 2.2855048179626465, + "learning_rate": 2.7030869212022747e-06, + "loss": 1.777, + "step": 1420 + }, + { + "epoch": 1.5561643835616439, + "eval_loss": 1.9851160049438477, + "eval_runtime": 37.4606, + "eval_samples_per_second": 26.695, + "eval_steps_per_second": 1.121, + "step": 1420 + }, + { + "epoch": 1.5572602739726027, + "grad_norm": 2.4190328121185303, + "learning_rate": 2.7010560519902516e-06, + "loss": 1.8456, + "step": 1421 + }, + { + "epoch": 1.5583561643835617, + "grad_norm": 2.225623369216919, + "learning_rate": 2.6990251827782293e-06, + "loss": 1.7638, + "step": 1422 + }, + { + "epoch": 1.5594520547945205, + "grad_norm": 2.1516661643981934, + "learning_rate": 2.696994313566206e-06, + "loss": 1.7861, + "step": 1423 + }, + { + "epoch": 1.5605479452054793, + "grad_norm": 2.2461135387420654, + "learning_rate": 2.694963444354184e-06, + "loss": 1.7763, + "step": 1424 + }, + { + "epoch": 1.5616438356164384, + "grad_norm": 2.280869245529175, + "learning_rate": 2.6929325751421607e-06, + "loss": 1.8589, + "step": 1425 + }, + { + "epoch": 1.5627397260273974, + "grad_norm": 2.2034990787506104, + "learning_rate": 2.6909017059301384e-06, + "loss": 1.7705, + "step": 1426 + }, + { + "epoch": 1.563835616438356, + "grad_norm": 2.1346969604492188, + "learning_rate": 2.6888708367181153e-06, + "loss": 1.8318, + "step": 1427 + }, + { + "epoch": 1.564931506849315, + "grad_norm": 2.37400221824646, + "learning_rate": 2.686839967506093e-06, + "loss": 1.746, + "step": 1428 + }, + { + "epoch": 1.566027397260274, + "grad_norm": 2.392561435699463, + "learning_rate": 2.68480909829407e-06, + "loss": 1.8105, + "step": 1429 + }, + { + "epoch": 1.5671232876712329, + "grad_norm": 2.299830198287964, + "learning_rate": 2.6827782290820475e-06, + "loss": 1.7103, + "step": 1430 + }, + { + "epoch": 1.5671232876712329, + "eval_loss": 1.9877182245254517, + "eval_runtime": 37.4813, + "eval_samples_per_second": 26.68, + "eval_steps_per_second": 1.121, + "step": 1430 + }, + { + "epoch": 1.5682191780821917, + "grad_norm": 2.3431026935577393, + "learning_rate": 2.6807473598700244e-06, + "loss": 1.7217, + "step": 1431 + }, + { + "epoch": 1.5693150684931507, + "grad_norm": 2.3474628925323486, + "learning_rate": 2.678716490658002e-06, + "loss": 1.8419, + "step": 1432 + }, + { + "epoch": 1.5704109589041098, + "grad_norm": 2.2877044677734375, + "learning_rate": 2.676685621445979e-06, + "loss": 1.8031, + "step": 1433 + }, + { + "epoch": 1.5715068493150683, + "grad_norm": 2.278613805770874, + "learning_rate": 2.674654752233956e-06, + "loss": 1.8066, + "step": 1434 + }, + { + "epoch": 1.5726027397260274, + "grad_norm": 2.3357837200164795, + "learning_rate": 2.6726238830219335e-06, + "loss": 1.8079, + "step": 1435 + }, + { + "epoch": 1.5736986301369864, + "grad_norm": 2.1932482719421387, + "learning_rate": 2.6705930138099107e-06, + "loss": 1.8032, + "step": 1436 + }, + { + "epoch": 1.5747945205479452, + "grad_norm": 2.298553705215454, + "learning_rate": 2.668562144597888e-06, + "loss": 1.8439, + "step": 1437 + }, + { + "epoch": 1.575890410958904, + "grad_norm": 2.2906084060668945, + "learning_rate": 2.6665312753858653e-06, + "loss": 1.7144, + "step": 1438 + }, + { + "epoch": 1.576986301369863, + "grad_norm": 2.2512989044189453, + "learning_rate": 2.6645004061738426e-06, + "loss": 1.8031, + "step": 1439 + }, + { + "epoch": 1.5780821917808219, + "grad_norm": 2.177103281021118, + "learning_rate": 2.66246953696182e-06, + "loss": 1.8073, + "step": 1440 + }, + { + "epoch": 1.5780821917808219, + "eval_loss": 1.9914618730545044, + "eval_runtime": 37.2706, + "eval_samples_per_second": 26.831, + "eval_steps_per_second": 1.127, + "step": 1440 + }, + { + "epoch": 1.5791780821917807, + "grad_norm": 2.4129281044006348, + "learning_rate": 2.660438667749797e-06, + "loss": 1.7552, + "step": 1441 + }, + { + "epoch": 1.5802739726027397, + "grad_norm": 2.2975244522094727, + "learning_rate": 2.6584077985377744e-06, + "loss": 1.7318, + "step": 1442 + }, + { + "epoch": 1.5813698630136988, + "grad_norm": 2.4061625003814697, + "learning_rate": 2.6563769293257517e-06, + "loss": 1.7839, + "step": 1443 + }, + { + "epoch": 1.5824657534246576, + "grad_norm": 2.236353635787964, + "learning_rate": 2.654346060113729e-06, + "loss": 1.7995, + "step": 1444 + }, + { + "epoch": 1.5835616438356164, + "grad_norm": 2.2464141845703125, + "learning_rate": 2.6523151909017062e-06, + "loss": 1.7496, + "step": 1445 + }, + { + "epoch": 1.5846575342465754, + "grad_norm": 2.2744123935699463, + "learning_rate": 2.6502843216896835e-06, + "loss": 1.7703, + "step": 1446 + }, + { + "epoch": 1.5857534246575342, + "grad_norm": 2.1578054428100586, + "learning_rate": 2.6482534524776603e-06, + "loss": 1.7561, + "step": 1447 + }, + { + "epoch": 1.586849315068493, + "grad_norm": 2.1924850940704346, + "learning_rate": 2.646222583265638e-06, + "loss": 1.867, + "step": 1448 + }, + { + "epoch": 1.587945205479452, + "grad_norm": 2.303420305252075, + "learning_rate": 2.644191714053615e-06, + "loss": 1.8345, + "step": 1449 + }, + { + "epoch": 1.589041095890411, + "grad_norm": 2.308837413787842, + "learning_rate": 2.6421608448415926e-06, + "loss": 1.7979, + "step": 1450 + }, + { + "epoch": 1.589041095890411, + "eval_loss": 1.9836338758468628, + "eval_runtime": 37.253, + "eval_samples_per_second": 26.843, + "eval_steps_per_second": 1.127, + "step": 1450 + }, + { + "epoch": 1.59013698630137, + "grad_norm": 2.2896456718444824, + "learning_rate": 2.6401299756295694e-06, + "loss": 1.7679, + "step": 1451 + }, + { + "epoch": 1.5912328767123287, + "grad_norm": 2.2482378482818604, + "learning_rate": 2.638099106417547e-06, + "loss": 1.7481, + "step": 1452 + }, + { + "epoch": 1.5923287671232877, + "grad_norm": 2.2620620727539062, + "learning_rate": 2.636068237205524e-06, + "loss": 1.801, + "step": 1453 + }, + { + "epoch": 1.5934246575342466, + "grad_norm": 2.216510772705078, + "learning_rate": 2.6340373679935017e-06, + "loss": 1.7485, + "step": 1454 + }, + { + "epoch": 1.5945205479452054, + "grad_norm": 2.354858160018921, + "learning_rate": 2.6320064987814785e-06, + "loss": 1.8636, + "step": 1455 + }, + { + "epoch": 1.5956164383561644, + "grad_norm": 2.251880168914795, + "learning_rate": 2.6299756295694562e-06, + "loss": 1.7087, + "step": 1456 + }, + { + "epoch": 1.5967123287671234, + "grad_norm": 2.2605879306793213, + "learning_rate": 2.627944760357433e-06, + "loss": 1.8141, + "step": 1457 + }, + { + "epoch": 1.5978082191780822, + "grad_norm": 2.3315021991729736, + "learning_rate": 2.6259138911454104e-06, + "loss": 1.8565, + "step": 1458 + }, + { + "epoch": 1.598904109589041, + "grad_norm": 2.1999425888061523, + "learning_rate": 2.6238830219333877e-06, + "loss": 1.8683, + "step": 1459 + }, + { + "epoch": 1.6, + "grad_norm": 2.2652416229248047, + "learning_rate": 2.621852152721365e-06, + "loss": 1.8489, + "step": 1460 + }, + { + "epoch": 1.6, + "eval_loss": 1.9816856384277344, + "eval_runtime": 37.2727, + "eval_samples_per_second": 26.829, + "eval_steps_per_second": 1.127, + "step": 1460 + }, + { + "epoch": 1.601095890410959, + "grad_norm": 2.469440221786499, + "learning_rate": 2.619821283509342e-06, + "loss": 1.7335, + "step": 1461 + }, + { + "epoch": 1.6021917808219177, + "grad_norm": 2.434267044067383, + "learning_rate": 2.6177904142973195e-06, + "loss": 1.8291, + "step": 1462 + }, + { + "epoch": 1.6032876712328767, + "grad_norm": 2.300492286682129, + "learning_rate": 2.6157595450852968e-06, + "loss": 1.7504, + "step": 1463 + }, + { + "epoch": 1.6043835616438358, + "grad_norm": 2.3906843662261963, + "learning_rate": 2.613728675873274e-06, + "loss": 1.8114, + "step": 1464 + }, + { + "epoch": 1.6054794520547944, + "grad_norm": 2.5111773014068604, + "learning_rate": 2.6116978066612513e-06, + "loss": 1.836, + "step": 1465 + }, + { + "epoch": 1.6065753424657534, + "grad_norm": 2.3834846019744873, + "learning_rate": 2.6096669374492286e-06, + "loss": 1.8589, + "step": 1466 + }, + { + "epoch": 1.6076712328767124, + "grad_norm": 2.2709615230560303, + "learning_rate": 2.607636068237206e-06, + "loss": 1.8052, + "step": 1467 + }, + { + "epoch": 1.6087671232876712, + "grad_norm": 2.3351972103118896, + "learning_rate": 2.605605199025183e-06, + "loss": 1.7559, + "step": 1468 + }, + { + "epoch": 1.60986301369863, + "grad_norm": 2.3994698524475098, + "learning_rate": 2.6035743298131604e-06, + "loss": 1.8099, + "step": 1469 + }, + { + "epoch": 1.610958904109589, + "grad_norm": 2.352590322494507, + "learning_rate": 2.6015434606011377e-06, + "loss": 1.8043, + "step": 1470 + }, + { + "epoch": 1.610958904109589, + "eval_loss": 1.9790339469909668, + "eval_runtime": 37.2619, + "eval_samples_per_second": 26.837, + "eval_steps_per_second": 1.127, + "step": 1470 + }, + { + "epoch": 1.612054794520548, + "grad_norm": 2.35754656791687, + "learning_rate": 2.5995125913891145e-06, + "loss": 1.8252, + "step": 1471 + }, + { + "epoch": 1.6131506849315067, + "grad_norm": 2.4011359214782715, + "learning_rate": 2.5974817221770922e-06, + "loss": 1.8005, + "step": 1472 + }, + { + "epoch": 1.6142465753424657, + "grad_norm": 3.143979787826538, + "learning_rate": 2.595450852965069e-06, + "loss": 1.7952, + "step": 1473 + }, + { + "epoch": 1.6153424657534248, + "grad_norm": 2.3928465843200684, + "learning_rate": 2.593419983753047e-06, + "loss": 1.8344, + "step": 1474 + }, + { + "epoch": 1.6164383561643836, + "grad_norm": 2.370368480682373, + "learning_rate": 2.5913891145410236e-06, + "loss": 1.8358, + "step": 1475 + }, + { + "epoch": 1.6175342465753424, + "grad_norm": 2.6163346767425537, + "learning_rate": 2.5893582453290013e-06, + "loss": 1.7734, + "step": 1476 + }, + { + "epoch": 1.6186301369863014, + "grad_norm": 2.370539903640747, + "learning_rate": 2.587327376116978e-06, + "loss": 1.8383, + "step": 1477 + }, + { + "epoch": 1.6197260273972602, + "grad_norm": 2.397007465362549, + "learning_rate": 2.585296506904956e-06, + "loss": 1.7702, + "step": 1478 + }, + { + "epoch": 1.620821917808219, + "grad_norm": 2.4692702293395996, + "learning_rate": 2.5832656376929327e-06, + "loss": 1.7887, + "step": 1479 + }, + { + "epoch": 1.621917808219178, + "grad_norm": 2.5779969692230225, + "learning_rate": 2.5812347684809104e-06, + "loss": 1.7751, + "step": 1480 + }, + { + "epoch": 1.621917808219178, + "eval_loss": 1.9826369285583496, + "eval_runtime": 37.2779, + "eval_samples_per_second": 26.826, + "eval_steps_per_second": 1.127, + "step": 1480 + }, + { + "epoch": 1.6230136986301371, + "grad_norm": 2.3182761669158936, + "learning_rate": 2.5792038992688873e-06, + "loss": 1.7136, + "step": 1481 + }, + { + "epoch": 1.624109589041096, + "grad_norm": 2.4109292030334473, + "learning_rate": 2.577173030056865e-06, + "loss": 1.8804, + "step": 1482 + }, + { + "epoch": 1.6252054794520547, + "grad_norm": 2.426166534423828, + "learning_rate": 2.575142160844842e-06, + "loss": 1.6996, + "step": 1483 + }, + { + "epoch": 1.6263013698630138, + "grad_norm": 2.5459423065185547, + "learning_rate": 2.5731112916328187e-06, + "loss": 1.8186, + "step": 1484 + }, + { + "epoch": 1.6273972602739726, + "grad_norm": 2.583578586578369, + "learning_rate": 2.5710804224207964e-06, + "loss": 1.8275, + "step": 1485 + }, + { + "epoch": 1.6284931506849314, + "grad_norm": 2.442936658859253, + "learning_rate": 2.5690495532087733e-06, + "loss": 1.7553, + "step": 1486 + }, + { + "epoch": 1.6295890410958904, + "grad_norm": 2.587864637374878, + "learning_rate": 2.567018683996751e-06, + "loss": 1.7171, + "step": 1487 + }, + { + "epoch": 1.6306849315068495, + "grad_norm": 2.4119534492492676, + "learning_rate": 2.564987814784728e-06, + "loss": 1.8296, + "step": 1488 + }, + { + "epoch": 1.6317808219178083, + "grad_norm": 2.3582253456115723, + "learning_rate": 2.5629569455727055e-06, + "loss": 1.7595, + "step": 1489 + }, + { + "epoch": 1.632876712328767, + "grad_norm": 2.522674083709717, + "learning_rate": 2.5609260763606824e-06, + "loss": 1.7908, + "step": 1490 + }, + { + "epoch": 1.632876712328767, + "eval_loss": 1.9791436195373535, + "eval_runtime": 37.2967, + "eval_samples_per_second": 26.812, + "eval_steps_per_second": 1.126, + "step": 1490 + }, + { + "epoch": 1.633972602739726, + "grad_norm": 2.2927470207214355, + "learning_rate": 2.55889520714866e-06, + "loss": 1.7789, + "step": 1491 + }, + { + "epoch": 1.635068493150685, + "grad_norm": 2.4469518661499023, + "learning_rate": 2.556864337936637e-06, + "loss": 1.8937, + "step": 1492 + }, + { + "epoch": 1.6361643835616437, + "grad_norm": 2.36860728263855, + "learning_rate": 2.5548334687246146e-06, + "loss": 1.778, + "step": 1493 + }, + { + "epoch": 1.6372602739726028, + "grad_norm": 2.4953925609588623, + "learning_rate": 2.5528025995125915e-06, + "loss": 1.7894, + "step": 1494 + }, + { + "epoch": 1.6383561643835618, + "grad_norm": 2.3478338718414307, + "learning_rate": 2.550771730300569e-06, + "loss": 1.8327, + "step": 1495 + }, + { + "epoch": 1.6394520547945204, + "grad_norm": 2.4399492740631104, + "learning_rate": 2.548740861088546e-06, + "loss": 1.8062, + "step": 1496 + }, + { + "epoch": 1.6405479452054794, + "grad_norm": 2.4457216262817383, + "learning_rate": 2.5467099918765233e-06, + "loss": 1.8268, + "step": 1497 + }, + { + "epoch": 1.6416438356164385, + "grad_norm": 2.365879535675049, + "learning_rate": 2.5446791226645006e-06, + "loss": 1.7288, + "step": 1498 + }, + { + "epoch": 1.6427397260273973, + "grad_norm": 2.4942879676818848, + "learning_rate": 2.542648253452478e-06, + "loss": 1.7111, + "step": 1499 + }, + { + "epoch": 1.643835616438356, + "grad_norm": 2.275982618331909, + "learning_rate": 2.540617384240455e-06, + "loss": 1.8292, + "step": 1500 + }, + { + "epoch": 1.643835616438356, + "eval_loss": 1.9746516942977905, + "eval_runtime": 37.2247, + "eval_samples_per_second": 26.864, + "eval_steps_per_second": 1.128, + "step": 1500 + }, + { + "epoch": 1.644931506849315, + "grad_norm": 2.4430429935455322, + "learning_rate": 2.5385865150284324e-06, + "loss": 1.8034, + "step": 1501 + }, + { + "epoch": 1.6460273972602741, + "grad_norm": 2.2778332233428955, + "learning_rate": 2.5365556458164097e-06, + "loss": 1.7589, + "step": 1502 + }, + { + "epoch": 1.6471232876712327, + "grad_norm": 2.2818830013275146, + "learning_rate": 2.534524776604387e-06, + "loss": 1.7743, + "step": 1503 + }, + { + "epoch": 1.6482191780821918, + "grad_norm": 2.321328639984131, + "learning_rate": 2.5324939073923642e-06, + "loss": 1.7922, + "step": 1504 + }, + { + "epoch": 1.6493150684931508, + "grad_norm": 2.4312188625335693, + "learning_rate": 2.5304630381803415e-06, + "loss": 1.6784, + "step": 1505 + }, + { + "epoch": 1.6504109589041096, + "grad_norm": 2.298741102218628, + "learning_rate": 2.5284321689683188e-06, + "loss": 1.7745, + "step": 1506 + }, + { + "epoch": 1.6515068493150684, + "grad_norm": 2.2882816791534424, + "learning_rate": 2.526401299756296e-06, + "loss": 1.7896, + "step": 1507 + }, + { + "epoch": 1.6526027397260274, + "grad_norm": 2.2336251735687256, + "learning_rate": 2.524370430544273e-06, + "loss": 1.7548, + "step": 1508 + }, + { + "epoch": 1.6536986301369863, + "grad_norm": 2.4369094371795654, + "learning_rate": 2.5223395613322506e-06, + "loss": 1.8049, + "step": 1509 + }, + { + "epoch": 1.654794520547945, + "grad_norm": 2.2978999614715576, + "learning_rate": 2.5203086921202274e-06, + "loss": 1.7489, + "step": 1510 + }, + { + "epoch": 1.654794520547945, + "eval_loss": 1.9760501384735107, + "eval_runtime": 37.3461, + "eval_samples_per_second": 26.777, + "eval_steps_per_second": 1.125, + "step": 1510 + }, + { + "epoch": 1.655890410958904, + "grad_norm": 2.331296920776367, + "learning_rate": 2.518277822908205e-06, + "loss": 1.8293, + "step": 1511 + }, + { + "epoch": 1.6569863013698631, + "grad_norm": 2.4544663429260254, + "learning_rate": 2.516246953696182e-06, + "loss": 1.7617, + "step": 1512 + }, + { + "epoch": 1.658082191780822, + "grad_norm": 2.4682042598724365, + "learning_rate": 2.5142160844841597e-06, + "loss": 1.8054, + "step": 1513 + }, + { + "epoch": 1.6591780821917808, + "grad_norm": 2.2827956676483154, + "learning_rate": 2.5121852152721365e-06, + "loss": 1.8137, + "step": 1514 + }, + { + "epoch": 1.6602739726027398, + "grad_norm": 2.2961761951446533, + "learning_rate": 2.5101543460601142e-06, + "loss": 1.7938, + "step": 1515 + }, + { + "epoch": 1.6613698630136986, + "grad_norm": 2.3464298248291016, + "learning_rate": 2.508123476848091e-06, + "loss": 1.8026, + "step": 1516 + }, + { + "epoch": 1.6624657534246574, + "grad_norm": 2.4975106716156006, + "learning_rate": 2.506092607636069e-06, + "loss": 1.7457, + "step": 1517 + }, + { + "epoch": 1.6635616438356164, + "grad_norm": 2.4823005199432373, + "learning_rate": 2.5040617384240457e-06, + "loss": 1.8282, + "step": 1518 + }, + { + "epoch": 1.6646575342465755, + "grad_norm": 2.2875800132751465, + "learning_rate": 2.5020308692120234e-06, + "loss": 1.7569, + "step": 1519 + }, + { + "epoch": 1.6657534246575343, + "grad_norm": 2.451022148132324, + "learning_rate": 2.5e-06, + "loss": 1.7483, + "step": 1520 + }, + { + "epoch": 1.6657534246575343, + "eval_loss": 1.9748642444610596, + "eval_runtime": 37.4512, + "eval_samples_per_second": 26.701, + "eval_steps_per_second": 1.121, + "step": 1520 + }, + { + "epoch": 1.666849315068493, + "grad_norm": 2.3427352905273438, + "learning_rate": 2.4979691307879775e-06, + "loss": 1.8493, + "step": 1521 + }, + { + "epoch": 1.6679452054794521, + "grad_norm": 2.4496946334838867, + "learning_rate": 2.4959382615759548e-06, + "loss": 1.7569, + "step": 1522 + }, + { + "epoch": 1.669041095890411, + "grad_norm": 2.141963481903076, + "learning_rate": 2.493907392363932e-06, + "loss": 1.7553, + "step": 1523 + }, + { + "epoch": 1.6701369863013698, + "grad_norm": 2.4115428924560547, + "learning_rate": 2.4918765231519093e-06, + "loss": 1.8606, + "step": 1524 + }, + { + "epoch": 1.6712328767123288, + "grad_norm": 2.614616632461548, + "learning_rate": 2.4898456539398866e-06, + "loss": 1.8217, + "step": 1525 + }, + { + "epoch": 1.6723287671232878, + "grad_norm": 2.338557243347168, + "learning_rate": 2.487814784727864e-06, + "loss": 1.7672, + "step": 1526 + }, + { + "epoch": 1.6734246575342466, + "grad_norm": 2.4309446811676025, + "learning_rate": 2.4857839155158407e-06, + "loss": 1.8476, + "step": 1527 + }, + { + "epoch": 1.6745205479452054, + "grad_norm": 2.25883412361145, + "learning_rate": 2.483753046303818e-06, + "loss": 1.7827, + "step": 1528 + }, + { + "epoch": 1.6756164383561645, + "grad_norm": 2.264845371246338, + "learning_rate": 2.4817221770917953e-06, + "loss": 1.7877, + "step": 1529 + }, + { + "epoch": 1.6767123287671233, + "grad_norm": 2.3584723472595215, + "learning_rate": 2.4796913078797725e-06, + "loss": 1.7891, + "step": 1530 + }, + { + "epoch": 1.6767123287671233, + "eval_loss": 1.9751919507980347, + "eval_runtime": 37.558, + "eval_samples_per_second": 26.625, + "eval_steps_per_second": 1.118, + "step": 1530 + }, + { + "epoch": 1.677808219178082, + "grad_norm": 2.3555185794830322, + "learning_rate": 2.47766043866775e-06, + "loss": 1.8235, + "step": 1531 + }, + { + "epoch": 1.6789041095890411, + "grad_norm": 2.3086600303649902, + "learning_rate": 2.475629569455727e-06, + "loss": 1.8036, + "step": 1532 + }, + { + "epoch": 1.6800000000000002, + "grad_norm": 2.286921262741089, + "learning_rate": 2.4735987002437044e-06, + "loss": 1.7885, + "step": 1533 + }, + { + "epoch": 1.6810958904109587, + "grad_norm": 2.2671141624450684, + "learning_rate": 2.4715678310316816e-06, + "loss": 1.758, + "step": 1534 + }, + { + "epoch": 1.6821917808219178, + "grad_norm": 2.4795749187469482, + "learning_rate": 2.469536961819659e-06, + "loss": 1.7791, + "step": 1535 + }, + { + "epoch": 1.6832876712328768, + "grad_norm": 2.201650857925415, + "learning_rate": 2.467506092607636e-06, + "loss": 1.7802, + "step": 1536 + }, + { + "epoch": 1.6843835616438356, + "grad_norm": 2.3804843425750732, + "learning_rate": 2.4654752233956135e-06, + "loss": 1.7739, + "step": 1537 + }, + { + "epoch": 1.6854794520547944, + "grad_norm": 2.322216510772705, + "learning_rate": 2.4634443541835907e-06, + "loss": 1.758, + "step": 1538 + }, + { + "epoch": 1.6865753424657535, + "grad_norm": 2.3582146167755127, + "learning_rate": 2.461413484971568e-06, + "loss": 1.8439, + "step": 1539 + }, + { + "epoch": 1.6876712328767123, + "grad_norm": 2.3597848415374756, + "learning_rate": 2.4593826157595453e-06, + "loss": 1.7772, + "step": 1540 + }, + { + "epoch": 1.6876712328767123, + "eval_loss": 1.9731802940368652, + "eval_runtime": 37.2439, + "eval_samples_per_second": 26.85, + "eval_steps_per_second": 1.128, + "step": 1540 + }, + { + "epoch": 1.688767123287671, + "grad_norm": 2.404569149017334, + "learning_rate": 2.4573517465475226e-06, + "loss": 1.8683, + "step": 1541 + }, + { + "epoch": 1.6898630136986301, + "grad_norm": 2.318737268447876, + "learning_rate": 2.4553208773355e-06, + "loss": 1.7223, + "step": 1542 + }, + { + "epoch": 1.6909589041095892, + "grad_norm": 2.2535340785980225, + "learning_rate": 2.453290008123477e-06, + "loss": 1.7693, + "step": 1543 + }, + { + "epoch": 1.692054794520548, + "grad_norm": 2.2533493041992188, + "learning_rate": 2.4512591389114544e-06, + "loss": 1.7876, + "step": 1544 + }, + { + "epoch": 1.6931506849315068, + "grad_norm": 2.253278970718384, + "learning_rate": 2.4492282696994317e-06, + "loss": 1.788, + "step": 1545 + }, + { + "epoch": 1.6942465753424658, + "grad_norm": 2.296807289123535, + "learning_rate": 2.447197400487409e-06, + "loss": 1.7877, + "step": 1546 + }, + { + "epoch": 1.6953424657534246, + "grad_norm": 2.4022111892700195, + "learning_rate": 2.4451665312753862e-06, + "loss": 1.7986, + "step": 1547 + }, + { + "epoch": 1.6964383561643834, + "grad_norm": 2.2671430110931396, + "learning_rate": 2.4431356620633635e-06, + "loss": 1.8269, + "step": 1548 + }, + { + "epoch": 1.6975342465753425, + "grad_norm": 2.338252067565918, + "learning_rate": 2.4411047928513408e-06, + "loss": 1.8279, + "step": 1549 + }, + { + "epoch": 1.6986301369863015, + "grad_norm": 2.365365505218506, + "learning_rate": 2.439073923639318e-06, + "loss": 1.7895, + "step": 1550 + }, + { + "epoch": 1.6986301369863015, + "eval_loss": 1.9711860418319702, + "eval_runtime": 37.2757, + "eval_samples_per_second": 26.827, + "eval_steps_per_second": 1.127, + "step": 1550 + }, + { + "epoch": 1.6997260273972603, + "grad_norm": 2.407580614089966, + "learning_rate": 2.4370430544272953e-06, + "loss": 1.7985, + "step": 1551 + }, + { + "epoch": 1.7008219178082191, + "grad_norm": 2.3145346641540527, + "learning_rate": 2.435012185215272e-06, + "loss": 1.7589, + "step": 1552 + }, + { + "epoch": 1.7019178082191782, + "grad_norm": 2.3809280395507812, + "learning_rate": 2.4329813160032495e-06, + "loss": 1.7334, + "step": 1553 + }, + { + "epoch": 1.703013698630137, + "grad_norm": 2.4224488735198975, + "learning_rate": 2.4309504467912267e-06, + "loss": 1.7549, + "step": 1554 + }, + { + "epoch": 1.7041095890410958, + "grad_norm": 2.374133348464966, + "learning_rate": 2.428919577579204e-06, + "loss": 1.7923, + "step": 1555 + }, + { + "epoch": 1.7052054794520548, + "grad_norm": 2.258222818374634, + "learning_rate": 2.4268887083671813e-06, + "loss": 1.7683, + "step": 1556 + }, + { + "epoch": 1.7063013698630138, + "grad_norm": 2.4481396675109863, + "learning_rate": 2.4248578391551586e-06, + "loss": 1.8609, + "step": 1557 + }, + { + "epoch": 1.7073972602739727, + "grad_norm": 2.3932790756225586, + "learning_rate": 2.422826969943136e-06, + "loss": 1.7475, + "step": 1558 + }, + { + "epoch": 1.7084931506849315, + "grad_norm": 2.3034355640411377, + "learning_rate": 2.420796100731113e-06, + "loss": 1.7664, + "step": 1559 + }, + { + "epoch": 1.7095890410958905, + "grad_norm": 2.2018799781799316, + "learning_rate": 2.4187652315190904e-06, + "loss": 1.7328, + "step": 1560 + }, + { + "epoch": 1.7095890410958905, + "eval_loss": 1.9715769290924072, + "eval_runtime": 37.2507, + "eval_samples_per_second": 26.845, + "eval_steps_per_second": 1.127, + "step": 1560 + }, + { + "epoch": 1.7106849315068493, + "grad_norm": 2.396462917327881, + "learning_rate": 2.4167343623070677e-06, + "loss": 1.7773, + "step": 1561 + }, + { + "epoch": 1.7117808219178081, + "grad_norm": 2.3212695121765137, + "learning_rate": 2.414703493095045e-06, + "loss": 1.7966, + "step": 1562 + }, + { + "epoch": 1.7128767123287671, + "grad_norm": 2.4157631397247314, + "learning_rate": 2.4126726238830222e-06, + "loss": 1.7797, + "step": 1563 + }, + { + "epoch": 1.7139726027397262, + "grad_norm": 2.404907464981079, + "learning_rate": 2.410641754670999e-06, + "loss": 1.8093, + "step": 1564 + }, + { + "epoch": 1.7150684931506848, + "grad_norm": 2.3798625469207764, + "learning_rate": 2.4086108854589763e-06, + "loss": 1.7635, + "step": 1565 + }, + { + "epoch": 1.7161643835616438, + "grad_norm": 2.325932025909424, + "learning_rate": 2.4065800162469536e-06, + "loss": 1.8281, + "step": 1566 + }, + { + "epoch": 1.7172602739726028, + "grad_norm": 2.413975238800049, + "learning_rate": 2.404549147034931e-06, + "loss": 1.7609, + "step": 1567 + }, + { + "epoch": 1.7183561643835616, + "grad_norm": 2.512775421142578, + "learning_rate": 2.402518277822908e-06, + "loss": 1.7793, + "step": 1568 + }, + { + "epoch": 1.7194520547945205, + "grad_norm": 2.581679582595825, + "learning_rate": 2.4004874086108854e-06, + "loss": 1.8302, + "step": 1569 + }, + { + "epoch": 1.7205479452054795, + "grad_norm": 2.389822244644165, + "learning_rate": 2.3984565393988627e-06, + "loss": 1.7573, + "step": 1570 + }, + { + "epoch": 1.7205479452054795, + "eval_loss": 1.9670331478118896, + "eval_runtime": 37.2901, + "eval_samples_per_second": 26.817, + "eval_steps_per_second": 1.126, + "step": 1570 + }, + { + "epoch": 1.7216438356164385, + "grad_norm": 2.6191556453704834, + "learning_rate": 2.39642567018684e-06, + "loss": 1.8252, + "step": 1571 + }, + { + "epoch": 1.7227397260273971, + "grad_norm": 2.394347906112671, + "learning_rate": 2.3943948009748173e-06, + "loss": 1.7561, + "step": 1572 + }, + { + "epoch": 1.7238356164383561, + "grad_norm": 2.3016326427459717, + "learning_rate": 2.3923639317627945e-06, + "loss": 1.7619, + "step": 1573 + }, + { + "epoch": 1.7249315068493152, + "grad_norm": 2.3663971424102783, + "learning_rate": 2.390333062550772e-06, + "loss": 1.7922, + "step": 1574 + }, + { + "epoch": 1.726027397260274, + "grad_norm": 2.3328938484191895, + "learning_rate": 2.388302193338749e-06, + "loss": 1.7374, + "step": 1575 + }, + { + "epoch": 1.7271232876712328, + "grad_norm": 2.313525676727295, + "learning_rate": 2.3862713241267264e-06, + "loss": 1.6585, + "step": 1576 + }, + { + "epoch": 1.7282191780821918, + "grad_norm": 2.4609642028808594, + "learning_rate": 2.3842404549147037e-06, + "loss": 1.8757, + "step": 1577 + }, + { + "epoch": 1.7293150684931506, + "grad_norm": 2.3502120971679688, + "learning_rate": 2.382209585702681e-06, + "loss": 1.7786, + "step": 1578 + }, + { + "epoch": 1.7304109589041095, + "grad_norm": 2.383025884628296, + "learning_rate": 2.380178716490658e-06, + "loss": 1.7502, + "step": 1579 + }, + { + "epoch": 1.7315068493150685, + "grad_norm": 2.280890464782715, + "learning_rate": 2.3781478472786355e-06, + "loss": 1.8192, + "step": 1580 + }, + { + "epoch": 1.7315068493150685, + "eval_loss": 1.9658632278442383, + "eval_runtime": 37.3714, + "eval_samples_per_second": 26.758, + "eval_steps_per_second": 1.124, + "step": 1580 + }, + { + "epoch": 1.7326027397260275, + "grad_norm": 2.342240333557129, + "learning_rate": 2.3761169780666128e-06, + "loss": 1.7786, + "step": 1581 + }, + { + "epoch": 1.7336986301369863, + "grad_norm": 2.2916715145111084, + "learning_rate": 2.37408610885459e-06, + "loss": 1.8161, + "step": 1582 + }, + { + "epoch": 1.7347945205479451, + "grad_norm": 2.24537992477417, + "learning_rate": 2.3720552396425673e-06, + "loss": 1.8442, + "step": 1583 + }, + { + "epoch": 1.7358904109589042, + "grad_norm": 2.3003411293029785, + "learning_rate": 2.3700243704305446e-06, + "loss": 1.7775, + "step": 1584 + }, + { + "epoch": 1.736986301369863, + "grad_norm": 2.3062400817871094, + "learning_rate": 2.367993501218522e-06, + "loss": 1.8374, + "step": 1585 + }, + { + "epoch": 1.7380821917808218, + "grad_norm": 2.3534083366394043, + "learning_rate": 2.365962632006499e-06, + "loss": 1.7829, + "step": 1586 + }, + { + "epoch": 1.7391780821917808, + "grad_norm": 2.2612221240997314, + "learning_rate": 2.3639317627944764e-06, + "loss": 1.7934, + "step": 1587 + }, + { + "epoch": 1.7402739726027399, + "grad_norm": 2.4197304248809814, + "learning_rate": 2.3619008935824537e-06, + "loss": 1.8226, + "step": 1588 + }, + { + "epoch": 1.7413698630136987, + "grad_norm": 2.435908079147339, + "learning_rate": 2.3598700243704305e-06, + "loss": 1.8148, + "step": 1589 + }, + { + "epoch": 1.7424657534246575, + "grad_norm": 2.3617560863494873, + "learning_rate": 2.357839155158408e-06, + "loss": 1.8085, + "step": 1590 + }, + { + "epoch": 1.7424657534246575, + "eval_loss": 1.9636695384979248, + "eval_runtime": 37.2968, + "eval_samples_per_second": 26.812, + "eval_steps_per_second": 1.126, + "step": 1590 + }, + { + "epoch": 1.7435616438356165, + "grad_norm": 2.295182228088379, + "learning_rate": 2.355808285946385e-06, + "loss": 1.8091, + "step": 1591 + }, + { + "epoch": 1.7446575342465753, + "grad_norm": 2.3347294330596924, + "learning_rate": 2.3537774167343624e-06, + "loss": 1.8086, + "step": 1592 + }, + { + "epoch": 1.7457534246575341, + "grad_norm": 2.4760799407958984, + "learning_rate": 2.3517465475223396e-06, + "loss": 1.8227, + "step": 1593 + }, + { + "epoch": 1.7468493150684932, + "grad_norm": 2.4243907928466797, + "learning_rate": 2.349715678310317e-06, + "loss": 1.7921, + "step": 1594 + }, + { + "epoch": 1.7479452054794522, + "grad_norm": 2.271735668182373, + "learning_rate": 2.347684809098294e-06, + "loss": 1.7987, + "step": 1595 + }, + { + "epoch": 1.749041095890411, + "grad_norm": 2.5412354469299316, + "learning_rate": 2.3456539398862715e-06, + "loss": 1.7558, + "step": 1596 + }, + { + "epoch": 1.7501369863013698, + "grad_norm": 2.464944362640381, + "learning_rate": 2.3436230706742487e-06, + "loss": 1.8087, + "step": 1597 + }, + { + "epoch": 1.7512328767123289, + "grad_norm": 2.4210381507873535, + "learning_rate": 2.341592201462226e-06, + "loss": 1.756, + "step": 1598 + }, + { + "epoch": 1.7523287671232877, + "grad_norm": 2.284945011138916, + "learning_rate": 2.3395613322502033e-06, + "loss": 1.7909, + "step": 1599 + }, + { + "epoch": 1.7534246575342465, + "grad_norm": 2.3740732669830322, + "learning_rate": 2.3375304630381806e-06, + "loss": 1.7683, + "step": 1600 + }, + { + "epoch": 1.7534246575342465, + "eval_loss": 1.9646408557891846, + "eval_runtime": 37.3266, + "eval_samples_per_second": 26.791, + "eval_steps_per_second": 1.125, + "step": 1600 + }, + { + "epoch": 1.7545205479452055, + "grad_norm": 2.399583578109741, + "learning_rate": 2.335499593826158e-06, + "loss": 1.7342, + "step": 1601 + }, + { + "epoch": 1.7556164383561645, + "grad_norm": 2.4731597900390625, + "learning_rate": 2.333468724614135e-06, + "loss": 1.7507, + "step": 1602 + }, + { + "epoch": 1.7567123287671231, + "grad_norm": 2.365433692932129, + "learning_rate": 2.3314378554021124e-06, + "loss": 1.7944, + "step": 1603 + }, + { + "epoch": 1.7578082191780822, + "grad_norm": 2.4500601291656494, + "learning_rate": 2.3294069861900897e-06, + "loss": 1.822, + "step": 1604 + }, + { + "epoch": 1.7589041095890412, + "grad_norm": 2.3339014053344727, + "learning_rate": 2.327376116978067e-06, + "loss": 1.7464, + "step": 1605 + }, + { + "epoch": 1.76, + "grad_norm": 2.343973159790039, + "learning_rate": 2.3253452477660442e-06, + "loss": 1.7408, + "step": 1606 + }, + { + "epoch": 1.7610958904109588, + "grad_norm": 2.453462839126587, + "learning_rate": 2.3233143785540215e-06, + "loss": 1.8261, + "step": 1607 + }, + { + "epoch": 1.7621917808219179, + "grad_norm": 2.357698678970337, + "learning_rate": 2.3212835093419988e-06, + "loss": 1.762, + "step": 1608 + }, + { + "epoch": 1.7632876712328767, + "grad_norm": 2.353257417678833, + "learning_rate": 2.319252640129976e-06, + "loss": 1.8076, + "step": 1609 + }, + { + "epoch": 1.7643835616438355, + "grad_norm": 2.3269948959350586, + "learning_rate": 2.3172217709179533e-06, + "loss": 1.7516, + "step": 1610 + }, + { + "epoch": 1.7643835616438355, + "eval_loss": 1.9678465127944946, + "eval_runtime": 37.2052, + "eval_samples_per_second": 26.878, + "eval_steps_per_second": 1.129, + "step": 1610 + }, + { + "epoch": 1.7654794520547945, + "grad_norm": 2.4023501873016357, + "learning_rate": 2.3151909017059306e-06, + "loss": 1.8477, + "step": 1611 + }, + { + "epoch": 1.7665753424657535, + "grad_norm": 2.3356990814208984, + "learning_rate": 2.313160032493908e-06, + "loss": 1.8236, + "step": 1612 + }, + { + "epoch": 1.7676712328767124, + "grad_norm": 2.3571391105651855, + "learning_rate": 2.311129163281885e-06, + "loss": 1.7335, + "step": 1613 + }, + { + "epoch": 1.7687671232876712, + "grad_norm": 2.3510208129882812, + "learning_rate": 2.309098294069862e-06, + "loss": 1.8088, + "step": 1614 + }, + { + "epoch": 1.7698630136986302, + "grad_norm": 2.3852577209472656, + "learning_rate": 2.3070674248578393e-06, + "loss": 1.7312, + "step": 1615 + }, + { + "epoch": 1.770958904109589, + "grad_norm": 2.3085618019104004, + "learning_rate": 2.3050365556458166e-06, + "loss": 1.7379, + "step": 1616 + }, + { + "epoch": 1.7720547945205478, + "grad_norm": 2.3454065322875977, + "learning_rate": 2.303005686433794e-06, + "loss": 1.8054, + "step": 1617 + }, + { + "epoch": 1.7731506849315068, + "grad_norm": 2.4034247398376465, + "learning_rate": 2.300974817221771e-06, + "loss": 1.7955, + "step": 1618 + }, + { + "epoch": 1.7742465753424659, + "grad_norm": 2.4181671142578125, + "learning_rate": 2.2989439480097484e-06, + "loss": 1.7732, + "step": 1619 + }, + { + "epoch": 1.7753424657534247, + "grad_norm": 2.3408010005950928, + "learning_rate": 2.2969130787977257e-06, + "loss": 1.7231, + "step": 1620 + }, + { + "epoch": 1.7753424657534247, + "eval_loss": 1.9655163288116455, + "eval_runtime": 37.4878, + "eval_samples_per_second": 26.675, + "eval_steps_per_second": 1.12, + "step": 1620 + }, + { + "epoch": 1.7764383561643835, + "grad_norm": 2.3113718032836914, + "learning_rate": 2.294882209585703e-06, + "loss": 1.7904, + "step": 1621 + }, + { + "epoch": 1.7775342465753425, + "grad_norm": 2.3037896156311035, + "learning_rate": 2.2928513403736802e-06, + "loss": 1.8069, + "step": 1622 + }, + { + "epoch": 1.7786301369863013, + "grad_norm": 2.3962693214416504, + "learning_rate": 2.2908204711616575e-06, + "loss": 1.7354, + "step": 1623 + }, + { + "epoch": 1.7797260273972602, + "grad_norm": 2.3499252796173096, + "learning_rate": 2.2887896019496348e-06, + "loss": 1.779, + "step": 1624 + }, + { + "epoch": 1.7808219178082192, + "grad_norm": 2.390958070755005, + "learning_rate": 2.286758732737612e-06, + "loss": 1.8257, + "step": 1625 + }, + { + "epoch": 1.7819178082191782, + "grad_norm": 2.365626573562622, + "learning_rate": 2.284727863525589e-06, + "loss": 1.7618, + "step": 1626 + }, + { + "epoch": 1.783013698630137, + "grad_norm": 2.3558530807495117, + "learning_rate": 2.282696994313566e-06, + "loss": 1.7729, + "step": 1627 + }, + { + "epoch": 1.7841095890410958, + "grad_norm": 2.3276045322418213, + "learning_rate": 2.2806661251015434e-06, + "loss": 1.8231, + "step": 1628 + }, + { + "epoch": 1.7852054794520549, + "grad_norm": 2.3369953632354736, + "learning_rate": 2.2786352558895207e-06, + "loss": 1.8011, + "step": 1629 + }, + { + "epoch": 1.7863013698630137, + "grad_norm": 2.3588311672210693, + "learning_rate": 2.276604386677498e-06, + "loss": 1.7682, + "step": 1630 + }, + { + "epoch": 1.7863013698630137, + "eval_loss": 1.963529348373413, + "eval_runtime": 37.5589, + "eval_samples_per_second": 26.625, + "eval_steps_per_second": 1.118, + "step": 1630 + }, + { + "epoch": 1.7873972602739725, + "grad_norm": 2.2918803691864014, + "learning_rate": 2.2745735174654753e-06, + "loss": 1.7256, + "step": 1631 + }, + { + "epoch": 1.7884931506849315, + "grad_norm": 2.423738479614258, + "learning_rate": 2.2725426482534526e-06, + "loss": 1.7582, + "step": 1632 + }, + { + "epoch": 1.7895890410958906, + "grad_norm": 2.3500850200653076, + "learning_rate": 2.27051177904143e-06, + "loss": 1.8255, + "step": 1633 + }, + { + "epoch": 1.7906849315068492, + "grad_norm": 2.3280837535858154, + "learning_rate": 2.268480909829407e-06, + "loss": 1.7845, + "step": 1634 + }, + { + "epoch": 1.7917808219178082, + "grad_norm": 2.5055465698242188, + "learning_rate": 2.2664500406173844e-06, + "loss": 1.78, + "step": 1635 + }, + { + "epoch": 1.7928767123287672, + "grad_norm": 2.3132143020629883, + "learning_rate": 2.2644191714053617e-06, + "loss": 1.7814, + "step": 1636 + }, + { + "epoch": 1.793972602739726, + "grad_norm": 2.316361427307129, + "learning_rate": 2.262388302193339e-06, + "loss": 1.8161, + "step": 1637 + }, + { + "epoch": 1.7950684931506848, + "grad_norm": 2.337318181991577, + "learning_rate": 2.260357432981316e-06, + "loss": 1.7942, + "step": 1638 + }, + { + "epoch": 1.7961643835616439, + "grad_norm": 2.362031936645508, + "learning_rate": 2.2583265637692935e-06, + "loss": 1.7771, + "step": 1639 + }, + { + "epoch": 1.7972602739726027, + "grad_norm": 2.3260183334350586, + "learning_rate": 2.2562956945572708e-06, + "loss": 1.7975, + "step": 1640 + }, + { + "epoch": 1.7972602739726027, + "eval_loss": 1.9619382619857788, + "eval_runtime": 37.2606, + "eval_samples_per_second": 26.838, + "eval_steps_per_second": 1.127, + "step": 1640 + }, + { + "epoch": 1.7983561643835615, + "grad_norm": 2.4260714054107666, + "learning_rate": 2.254264825345248e-06, + "loss": 1.7654, + "step": 1641 + }, + { + "epoch": 1.7994520547945205, + "grad_norm": 2.297428846359253, + "learning_rate": 2.2522339561332253e-06, + "loss": 1.775, + "step": 1642 + }, + { + "epoch": 1.8005479452054796, + "grad_norm": 2.5153353214263916, + "learning_rate": 2.2502030869212026e-06, + "loss": 1.7796, + "step": 1643 + }, + { + "epoch": 1.8016438356164384, + "grad_norm": 2.5102577209472656, + "learning_rate": 2.24817221770918e-06, + "loss": 1.7917, + "step": 1644 + }, + { + "epoch": 1.8027397260273972, + "grad_norm": 2.3693106174468994, + "learning_rate": 2.246141348497157e-06, + "loss": 1.8459, + "step": 1645 + }, + { + "epoch": 1.8038356164383562, + "grad_norm": 2.2353153228759766, + "learning_rate": 2.2441104792851344e-06, + "loss": 1.8095, + "step": 1646 + }, + { + "epoch": 1.804931506849315, + "grad_norm": 2.265465497970581, + "learning_rate": 2.2420796100731117e-06, + "loss": 1.7623, + "step": 1647 + }, + { + "epoch": 1.8060273972602738, + "grad_norm": 2.2820940017700195, + "learning_rate": 2.240048740861089e-06, + "loss": 1.7946, + "step": 1648 + }, + { + "epoch": 1.8071232876712329, + "grad_norm": 2.378035068511963, + "learning_rate": 2.2380178716490662e-06, + "loss": 1.8059, + "step": 1649 + }, + { + "epoch": 1.808219178082192, + "grad_norm": 2.328341245651245, + "learning_rate": 2.2359870024370435e-06, + "loss": 1.6611, + "step": 1650 + }, + { + "epoch": 1.808219178082192, + "eval_loss": 1.9600144624710083, + "eval_runtime": 37.2601, + "eval_samples_per_second": 26.838, + "eval_steps_per_second": 1.127, + "step": 1650 + }, + { + "epoch": 1.8093150684931507, + "grad_norm": 2.2819113731384277, + "learning_rate": 2.2339561332250204e-06, + "loss": 1.8058, + "step": 1651 + }, + { + "epoch": 1.8104109589041095, + "grad_norm": 2.4178121089935303, + "learning_rate": 2.2319252640129976e-06, + "loss": 1.8204, + "step": 1652 + }, + { + "epoch": 1.8115068493150686, + "grad_norm": 2.497008800506592, + "learning_rate": 2.229894394800975e-06, + "loss": 1.7968, + "step": 1653 + }, + { + "epoch": 1.8126027397260274, + "grad_norm": 2.4681501388549805, + "learning_rate": 2.227863525588952e-06, + "loss": 1.7292, + "step": 1654 + }, + { + "epoch": 1.8136986301369862, + "grad_norm": 2.2983527183532715, + "learning_rate": 2.2258326563769295e-06, + "loss": 1.7097, + "step": 1655 + }, + { + "epoch": 1.8147945205479452, + "grad_norm": 2.349454164505005, + "learning_rate": 2.2238017871649067e-06, + "loss": 1.7256, + "step": 1656 + }, + { + "epoch": 1.8158904109589042, + "grad_norm": 2.4247677326202393, + "learning_rate": 2.221770917952884e-06, + "loss": 1.7165, + "step": 1657 + }, + { + "epoch": 1.816986301369863, + "grad_norm": 2.3035342693328857, + "learning_rate": 2.2197400487408613e-06, + "loss": 1.7811, + "step": 1658 + }, + { + "epoch": 1.8180821917808219, + "grad_norm": 2.318772554397583, + "learning_rate": 2.2177091795288386e-06, + "loss": 1.8082, + "step": 1659 + }, + { + "epoch": 1.819178082191781, + "grad_norm": 2.2477188110351562, + "learning_rate": 2.215678310316816e-06, + "loss": 1.7206, + "step": 1660 + }, + { + "epoch": 1.819178082191781, + "eval_loss": 1.962982416152954, + "eval_runtime": 37.2709, + "eval_samples_per_second": 26.831, + "eval_steps_per_second": 1.127, + "step": 1660 + }, + { + "epoch": 1.8202739726027397, + "grad_norm": 2.355264186859131, + "learning_rate": 2.213647441104793e-06, + "loss": 1.7634, + "step": 1661 + }, + { + "epoch": 1.8213698630136985, + "grad_norm": 2.267458200454712, + "learning_rate": 2.2116165718927704e-06, + "loss": 1.7819, + "step": 1662 + }, + { + "epoch": 1.8224657534246576, + "grad_norm": 2.2487895488739014, + "learning_rate": 2.2095857026807473e-06, + "loss": 1.759, + "step": 1663 + }, + { + "epoch": 1.8235616438356166, + "grad_norm": 2.4253578186035156, + "learning_rate": 2.2075548334687245e-06, + "loss": 1.7937, + "step": 1664 + }, + { + "epoch": 1.8246575342465754, + "grad_norm": 2.478050470352173, + "learning_rate": 2.205523964256702e-06, + "loss": 1.7241, + "step": 1665 + }, + { + "epoch": 1.8257534246575342, + "grad_norm": 2.289905309677124, + "learning_rate": 2.203493095044679e-06, + "loss": 1.7531, + "step": 1666 + }, + { + "epoch": 1.8268493150684932, + "grad_norm": 2.303421974182129, + "learning_rate": 2.2014622258326564e-06, + "loss": 1.7711, + "step": 1667 + }, + { + "epoch": 1.827945205479452, + "grad_norm": 2.3670248985290527, + "learning_rate": 2.1994313566206336e-06, + "loss": 1.8093, + "step": 1668 + }, + { + "epoch": 1.8290410958904109, + "grad_norm": 2.3721048831939697, + "learning_rate": 2.197400487408611e-06, + "loss": 1.8273, + "step": 1669 + }, + { + "epoch": 1.83013698630137, + "grad_norm": 2.3000190258026123, + "learning_rate": 2.195369618196588e-06, + "loss": 1.75, + "step": 1670 + }, + { + "epoch": 1.83013698630137, + "eval_loss": 1.9577263593673706, + "eval_runtime": 37.2953, + "eval_samples_per_second": 26.813, + "eval_steps_per_second": 1.126, + "step": 1670 + }, + { + "epoch": 1.831232876712329, + "grad_norm": 2.303584575653076, + "learning_rate": 2.1933387489845655e-06, + "loss": 1.7278, + "step": 1671 + }, + { + "epoch": 1.8323287671232875, + "grad_norm": 2.260579824447632, + "learning_rate": 2.1913078797725427e-06, + "loss": 1.7524, + "step": 1672 + }, + { + "epoch": 1.8334246575342465, + "grad_norm": 2.311967372894287, + "learning_rate": 2.18927701056052e-06, + "loss": 1.7799, + "step": 1673 + }, + { + "epoch": 1.8345205479452056, + "grad_norm": 2.338989496231079, + "learning_rate": 2.1872461413484973e-06, + "loss": 1.8468, + "step": 1674 + }, + { + "epoch": 1.8356164383561644, + "grad_norm": 2.338989496231079, + "learning_rate": 2.1872461413484973e-06, + "loss": 1.7853, + "step": 1675 + }, + { + "epoch": 1.8367123287671232, + "grad_norm": 2.526724100112915, + "learning_rate": 2.1852152721364746e-06, + "loss": 1.78, + "step": 1676 + }, + { + "epoch": 1.8378082191780822, + "grad_norm": 2.4637107849121094, + "learning_rate": 2.183184402924452e-06, + "loss": 1.7328, + "step": 1677 + }, + { + "epoch": 1.838904109589041, + "grad_norm": 2.34454345703125, + "learning_rate": 2.181153533712429e-06, + "loss": 1.8058, + "step": 1678 + }, + { + "epoch": 1.8399999999999999, + "grad_norm": 2.5518760681152344, + "learning_rate": 2.1791226645004064e-06, + "loss": 1.7746, + "step": 1679 + }, + { + "epoch": 1.841095890410959, + "grad_norm": 2.437443971633911, + "learning_rate": 2.1770917952883837e-06, + "loss": 1.7517, + "step": 1680 + }, + { + "epoch": 1.841095890410959, + "eval_loss": 1.9572285413742065, + "eval_runtime": 37.3402, + "eval_samples_per_second": 26.781, + "eval_steps_per_second": 1.125, + "step": 1680 + }, + { + "epoch": 1.842191780821918, + "grad_norm": 2.344287157058716, + "learning_rate": 2.175060926076361e-06, + "loss": 1.8229, + "step": 1681 + }, + { + "epoch": 1.8432876712328767, + "grad_norm": 2.333840847015381, + "learning_rate": 2.1730300568643382e-06, + "loss": 1.8534, + "step": 1682 + }, + { + "epoch": 1.8443835616438355, + "grad_norm": 2.372926950454712, + "learning_rate": 2.1709991876523155e-06, + "loss": 1.7609, + "step": 1683 + }, + { + "epoch": 1.8454794520547946, + "grad_norm": 2.2858645915985107, + "learning_rate": 2.1689683184402928e-06, + "loss": 1.7508, + "step": 1684 + }, + { + "epoch": 1.8465753424657534, + "grad_norm": 2.326586961746216, + "learning_rate": 2.16693744922827e-06, + "loss": 1.8122, + "step": 1685 + }, + { + "epoch": 1.8476712328767122, + "grad_norm": 2.3605616092681885, + "learning_rate": 2.1649065800162473e-06, + "loss": 1.7794, + "step": 1686 + }, + { + "epoch": 1.8487671232876712, + "grad_norm": 2.4188039302825928, + "learning_rate": 2.1628757108042246e-06, + "loss": 1.7552, + "step": 1687 + }, + { + "epoch": 1.8498630136986303, + "grad_norm": 2.335617780685425, + "learning_rate": 2.160844841592202e-06, + "loss": 1.7259, + "step": 1688 + }, + { + "epoch": 1.850958904109589, + "grad_norm": 2.245882272720337, + "learning_rate": 2.1588139723801787e-06, + "loss": 1.7259, + "step": 1689 + }, + { + "epoch": 1.8520547945205479, + "grad_norm": 2.4447500705718994, + "learning_rate": 2.156783103168156e-06, + "loss": 1.7605, + "step": 1690 + }, + { + "epoch": 1.8520547945205479, + "eval_loss": 1.9608982801437378, + "eval_runtime": 37.3756, + "eval_samples_per_second": 26.755, + "eval_steps_per_second": 1.124, + "step": 1690 + }, + { + "epoch": 1.853150684931507, + "grad_norm": 2.363443613052368, + "learning_rate": 2.1547522339561333e-06, + "loss": 1.7336, + "step": 1691 + }, + { + "epoch": 1.8542465753424657, + "grad_norm": 2.3573989868164062, + "learning_rate": 2.1527213647441106e-06, + "loss": 1.7953, + "step": 1692 + }, + { + "epoch": 1.8553424657534245, + "grad_norm": 2.4292593002319336, + "learning_rate": 2.150690495532088e-06, + "loss": 1.8276, + "step": 1693 + }, + { + "epoch": 1.8564383561643836, + "grad_norm": 2.374563694000244, + "learning_rate": 2.148659626320065e-06, + "loss": 1.7975, + "step": 1694 + }, + { + "epoch": 1.8575342465753426, + "grad_norm": 2.3935306072235107, + "learning_rate": 2.1466287571080424e-06, + "loss": 1.7824, + "step": 1695 + }, + { + "epoch": 1.8586301369863014, + "grad_norm": 2.343773365020752, + "learning_rate": 2.1445978878960197e-06, + "loss": 1.7574, + "step": 1696 + }, + { + "epoch": 1.8597260273972602, + "grad_norm": 2.2765426635742188, + "learning_rate": 2.142567018683997e-06, + "loss": 1.7682, + "step": 1697 + }, + { + "epoch": 1.8608219178082193, + "grad_norm": 2.339470863342285, + "learning_rate": 2.140536149471974e-06, + "loss": 1.8229, + "step": 1698 + }, + { + "epoch": 1.861917808219178, + "grad_norm": 2.341775894165039, + "learning_rate": 2.1385052802599515e-06, + "loss": 1.7421, + "step": 1699 + }, + { + "epoch": 1.8630136986301369, + "grad_norm": 2.343876838684082, + "learning_rate": 2.1364744110479288e-06, + "loss": 1.7519, + "step": 1700 + }, + { + "epoch": 1.8630136986301369, + "eval_loss": 1.9587067365646362, + "eval_runtime": 37.4192, + "eval_samples_per_second": 26.724, + "eval_steps_per_second": 1.122, + "step": 1700 + }, + { + "epoch": 1.864109589041096, + "grad_norm": 2.388373613357544, + "learning_rate": 2.134443541835906e-06, + "loss": 1.8039, + "step": 1701 + }, + { + "epoch": 1.865205479452055, + "grad_norm": 2.379742383956909, + "learning_rate": 2.132412672623883e-06, + "loss": 1.7549, + "step": 1702 + }, + { + "epoch": 1.8663013698630135, + "grad_norm": 2.25833797454834, + "learning_rate": 2.13038180341186e-06, + "loss": 1.7349, + "step": 1703 + }, + { + "epoch": 1.8673972602739726, + "grad_norm": 2.4541401863098145, + "learning_rate": 2.1283509341998374e-06, + "loss": 1.7785, + "step": 1704 + }, + { + "epoch": 1.8684931506849316, + "grad_norm": 2.338456392288208, + "learning_rate": 2.1263200649878147e-06, + "loss": 1.8272, + "step": 1705 + }, + { + "epoch": 1.8695890410958904, + "grad_norm": 2.474292755126953, + "learning_rate": 2.124289195775792e-06, + "loss": 1.7497, + "step": 1706 + }, + { + "epoch": 1.8706849315068492, + "grad_norm": 2.206749439239502, + "learning_rate": 2.1222583265637693e-06, + "loss": 1.7697, + "step": 1707 + }, + { + "epoch": 1.8717808219178083, + "grad_norm": 2.378093957901001, + "learning_rate": 2.1202274573517465e-06, + "loss": 1.6851, + "step": 1708 + }, + { + "epoch": 1.872876712328767, + "grad_norm": 2.2442147731781006, + "learning_rate": 2.118196588139724e-06, + "loss": 1.7268, + "step": 1709 + }, + { + "epoch": 1.8739726027397259, + "grad_norm": 2.324852228164673, + "learning_rate": 2.116165718927701e-06, + "loss": 1.7369, + "step": 1710 + }, + { + "epoch": 1.8739726027397259, + "eval_loss": 1.9555498361587524, + "eval_runtime": 37.2599, + "eval_samples_per_second": 26.838, + "eval_steps_per_second": 1.127, + "step": 1710 + }, + { + "epoch": 1.875068493150685, + "grad_norm": 2.3466620445251465, + "learning_rate": 2.1141348497156784e-06, + "loss": 1.7642, + "step": 1711 + }, + { + "epoch": 1.876164383561644, + "grad_norm": 2.303654432296753, + "learning_rate": 2.1121039805036556e-06, + "loss": 1.7406, + "step": 1712 + }, + { + "epoch": 1.8772602739726028, + "grad_norm": 2.242719888687134, + "learning_rate": 2.110073111291633e-06, + "loss": 1.7269, + "step": 1713 + }, + { + "epoch": 1.8783561643835616, + "grad_norm": 2.4076626300811768, + "learning_rate": 2.10804224207961e-06, + "loss": 1.7749, + "step": 1714 + }, + { + "epoch": 1.8794520547945206, + "grad_norm": 2.510887861251831, + "learning_rate": 2.1060113728675875e-06, + "loss": 1.8056, + "step": 1715 + }, + { + "epoch": 1.8805479452054794, + "grad_norm": 2.4038093090057373, + "learning_rate": 2.1039805036555647e-06, + "loss": 1.7915, + "step": 1716 + }, + { + "epoch": 1.8816438356164382, + "grad_norm": 2.4485573768615723, + "learning_rate": 2.101949634443542e-06, + "loss": 1.6845, + "step": 1717 + }, + { + "epoch": 1.8827397260273973, + "grad_norm": 2.4619247913360596, + "learning_rate": 2.0999187652315193e-06, + "loss": 1.7597, + "step": 1718 + }, + { + "epoch": 1.8838356164383563, + "grad_norm": 2.5079665184020996, + "learning_rate": 2.0978878960194966e-06, + "loss": 1.7207, + "step": 1719 + }, + { + "epoch": 1.884931506849315, + "grad_norm": 2.38696551322937, + "learning_rate": 2.095857026807474e-06, + "loss": 1.7855, + "step": 1720 + }, + { + "epoch": 1.884931506849315, + "eval_loss": 1.9538354873657227, + "eval_runtime": 37.5269, + "eval_samples_per_second": 26.648, + "eval_steps_per_second": 1.119, + "step": 1720 + }, + { + "epoch": 1.886027397260274, + "grad_norm": 2.2652602195739746, + "learning_rate": 2.093826157595451e-06, + "loss": 1.7806, + "step": 1721 + }, + { + "epoch": 1.887123287671233, + "grad_norm": 2.4074227809906006, + "learning_rate": 2.0917952883834284e-06, + "loss": 1.7397, + "step": 1722 + }, + { + "epoch": 1.8882191780821918, + "grad_norm": 2.4302561283111572, + "learning_rate": 2.0897644191714057e-06, + "loss": 1.7883, + "step": 1723 + }, + { + "epoch": 1.8893150684931506, + "grad_norm": 2.311933755874634, + "learning_rate": 2.087733549959383e-06, + "loss": 1.8186, + "step": 1724 + }, + { + "epoch": 1.8904109589041096, + "grad_norm": 2.4028313159942627, + "learning_rate": 2.0857026807473602e-06, + "loss": 1.7208, + "step": 1725 + }, + { + "epoch": 1.8915068493150686, + "grad_norm": 2.2129130363464355, + "learning_rate": 2.0836718115353375e-06, + "loss": 1.7007, + "step": 1726 + }, + { + "epoch": 1.8926027397260274, + "grad_norm": 2.3893744945526123, + "learning_rate": 2.0816409423233144e-06, + "loss": 1.8281, + "step": 1727 + }, + { + "epoch": 1.8936986301369862, + "grad_norm": 2.2396240234375, + "learning_rate": 2.0796100731112916e-06, + "loss": 1.7252, + "step": 1728 + }, + { + "epoch": 1.8947945205479453, + "grad_norm": 2.4073901176452637, + "learning_rate": 2.077579203899269e-06, + "loss": 1.7533, + "step": 1729 + }, + { + "epoch": 1.895890410958904, + "grad_norm": 2.3669373989105225, + "learning_rate": 2.075548334687246e-06, + "loss": 1.7013, + "step": 1730 + }, + { + "epoch": 1.895890410958904, + "eval_loss": 1.95528244972229, + "eval_runtime": 37.6174, + "eval_samples_per_second": 26.583, + "eval_steps_per_second": 1.117, + "step": 1730 + }, + { + "epoch": 1.896986301369863, + "grad_norm": 2.5254933834075928, + "learning_rate": 2.0735174654752235e-06, + "loss": 1.7848, + "step": 1731 + }, + { + "epoch": 1.898082191780822, + "grad_norm": 2.4138967990875244, + "learning_rate": 2.0714865962632007e-06, + "loss": 1.7204, + "step": 1732 + }, + { + "epoch": 1.899178082191781, + "grad_norm": 2.417165994644165, + "learning_rate": 2.069455727051178e-06, + "loss": 1.7741, + "step": 1733 + }, + { + "epoch": 1.9002739726027398, + "grad_norm": 2.4026505947113037, + "learning_rate": 2.0674248578391553e-06, + "loss": 1.7456, + "step": 1734 + }, + { + "epoch": 1.9013698630136986, + "grad_norm": 2.4637675285339355, + "learning_rate": 2.0653939886271326e-06, + "loss": 1.7679, + "step": 1735 + }, + { + "epoch": 1.9024657534246576, + "grad_norm": 2.269869089126587, + "learning_rate": 2.06336311941511e-06, + "loss": 1.8059, + "step": 1736 + }, + { + "epoch": 1.9035616438356164, + "grad_norm": 2.321094512939453, + "learning_rate": 2.061332250203087e-06, + "loss": 1.743, + "step": 1737 + }, + { + "epoch": 1.9046575342465752, + "grad_norm": 2.477288246154785, + "learning_rate": 2.0593013809910644e-06, + "loss": 1.7143, + "step": 1738 + }, + { + "epoch": 1.9057534246575343, + "grad_norm": 2.3924875259399414, + "learning_rate": 2.0572705117790417e-06, + "loss": 1.75, + "step": 1739 + }, + { + "epoch": 1.9068493150684933, + "grad_norm": 2.4452881813049316, + "learning_rate": 2.055239642567019e-06, + "loss": 1.7831, + "step": 1740 + }, + { + "epoch": 1.9068493150684933, + "eval_loss": 1.9599454402923584, + "eval_runtime": 37.3026, + "eval_samples_per_second": 26.808, + "eval_steps_per_second": 1.126, + "step": 1740 + }, + { + "epoch": 1.907945205479452, + "grad_norm": 2.2778351306915283, + "learning_rate": 2.0532087733549962e-06, + "loss": 1.7633, + "step": 1741 + }, + { + "epoch": 1.909041095890411, + "grad_norm": 2.3171589374542236, + "learning_rate": 2.0511779041429735e-06, + "loss": 1.7726, + "step": 1742 + }, + { + "epoch": 1.91013698630137, + "grad_norm": 2.286094903945923, + "learning_rate": 2.0491470349309508e-06, + "loss": 1.7458, + "step": 1743 + }, + { + "epoch": 1.9112328767123288, + "grad_norm": 2.5381743907928467, + "learning_rate": 2.047116165718928e-06, + "loss": 1.777, + "step": 1744 + }, + { + "epoch": 1.9123287671232876, + "grad_norm": 2.3261141777038574, + "learning_rate": 2.0450852965069053e-06, + "loss": 1.7715, + "step": 1745 + }, + { + "epoch": 1.9134246575342466, + "grad_norm": 2.351914167404175, + "learning_rate": 2.0430544272948826e-06, + "loss": 1.8214, + "step": 1746 + }, + { + "epoch": 1.9145205479452054, + "grad_norm": 2.274226427078247, + "learning_rate": 2.04102355808286e-06, + "loss": 1.763, + "step": 1747 + }, + { + "epoch": 1.9156164383561642, + "grad_norm": 2.4025075435638428, + "learning_rate": 2.038992688870837e-06, + "loss": 1.7072, + "step": 1748 + }, + { + "epoch": 1.9167123287671233, + "grad_norm": 2.334024429321289, + "learning_rate": 2.0369618196588144e-06, + "loss": 1.7381, + "step": 1749 + }, + { + "epoch": 1.9178082191780823, + "grad_norm": 2.3011622428894043, + "learning_rate": 2.0349309504467917e-06, + "loss": 1.7432, + "step": 1750 + }, + { + "epoch": 1.9178082191780823, + "eval_loss": 1.958349347114563, + "eval_runtime": 37.2591, + "eval_samples_per_second": 26.839, + "eval_steps_per_second": 1.127, + "step": 1750 + }, + { + "epoch": 1.9189041095890411, + "grad_norm": 2.643063545227051, + "learning_rate": 2.032900081234769e-06, + "loss": 1.7931, + "step": 1751 + }, + { + "epoch": 1.92, + "grad_norm": 2.3357834815979004, + "learning_rate": 2.030869212022746e-06, + "loss": 1.7933, + "step": 1752 + }, + { + "epoch": 1.921095890410959, + "grad_norm": 2.2614479064941406, + "learning_rate": 2.028838342810723e-06, + "loss": 1.7611, + "step": 1753 + }, + { + "epoch": 1.9221917808219178, + "grad_norm": 2.3057546615600586, + "learning_rate": 2.0268074735987004e-06, + "loss": 1.7406, + "step": 1754 + }, + { + "epoch": 1.9232876712328766, + "grad_norm": 2.6138038635253906, + "learning_rate": 2.0247766043866777e-06, + "loss": 1.7303, + "step": 1755 + }, + { + "epoch": 1.9243835616438356, + "grad_norm": 2.3803296089172363, + "learning_rate": 2.022745735174655e-06, + "loss": 1.8023, + "step": 1756 + }, + { + "epoch": 1.9254794520547946, + "grad_norm": 2.3436431884765625, + "learning_rate": 2.020714865962632e-06, + "loss": 1.7654, + "step": 1757 + }, + { + "epoch": 1.9265753424657535, + "grad_norm": 2.4103317260742188, + "learning_rate": 2.0186839967506095e-06, + "loss": 1.7932, + "step": 1758 + }, + { + "epoch": 1.9276712328767123, + "grad_norm": 2.4537675380706787, + "learning_rate": 2.0166531275385868e-06, + "loss": 1.761, + "step": 1759 + }, + { + "epoch": 1.9287671232876713, + "grad_norm": 2.3816988468170166, + "learning_rate": 2.014622258326564e-06, + "loss": 1.7276, + "step": 1760 + }, + { + "epoch": 1.9287671232876713, + "eval_loss": 1.9558053016662598, + "eval_runtime": 37.2417, + "eval_samples_per_second": 26.852, + "eval_steps_per_second": 1.128, + "step": 1760 + }, + { + "epoch": 1.9298630136986301, + "grad_norm": 2.37233304977417, + "learning_rate": 2.0125913891145413e-06, + "loss": 1.7674, + "step": 1761 + }, + { + "epoch": 1.930958904109589, + "grad_norm": 2.4043285846710205, + "learning_rate": 2.0105605199025186e-06, + "loss": 1.7638, + "step": 1762 + }, + { + "epoch": 1.932054794520548, + "grad_norm": 2.3402836322784424, + "learning_rate": 2.008529650690496e-06, + "loss": 1.7269, + "step": 1763 + }, + { + "epoch": 1.933150684931507, + "grad_norm": 2.42535400390625, + "learning_rate": 2.0064987814784727e-06, + "loss": 1.7965, + "step": 1764 + }, + { + "epoch": 1.9342465753424658, + "grad_norm": 2.361659288406372, + "learning_rate": 2.00446791226645e-06, + "loss": 1.7467, + "step": 1765 + }, + { + "epoch": 1.9353424657534246, + "grad_norm": 2.455253839492798, + "learning_rate": 2.0024370430544273e-06, + "loss": 1.7359, + "step": 1766 + }, + { + "epoch": 1.9364383561643836, + "grad_norm": 2.3062705993652344, + "learning_rate": 2.0004061738424045e-06, + "loss": 1.751, + "step": 1767 + }, + { + "epoch": 1.9375342465753425, + "grad_norm": 2.3683881759643555, + "learning_rate": 1.998375304630382e-06, + "loss": 1.8032, + "step": 1768 + }, + { + "epoch": 1.9386301369863013, + "grad_norm": 2.2175045013427734, + "learning_rate": 1.996344435418359e-06, + "loss": 1.7629, + "step": 1769 + }, + { + "epoch": 1.9397260273972603, + "grad_norm": 2.4331724643707275, + "learning_rate": 1.9943135662063364e-06, + "loss": 1.7899, + "step": 1770 + }, + { + "epoch": 1.9397260273972603, + "eval_loss": 1.9554719924926758, + "eval_runtime": 37.2636, + "eval_samples_per_second": 26.836, + "eval_steps_per_second": 1.127, + "step": 1770 + }, + { + "epoch": 1.9408219178082193, + "grad_norm": 2.370785713195801, + "learning_rate": 1.9922826969943136e-06, + "loss": 1.7687, + "step": 1771 + }, + { + "epoch": 1.941917808219178, + "grad_norm": 2.4047751426696777, + "learning_rate": 1.990251827782291e-06, + "loss": 1.804, + "step": 1772 + }, + { + "epoch": 1.943013698630137, + "grad_norm": 2.4597785472869873, + "learning_rate": 1.988220958570268e-06, + "loss": 1.7731, + "step": 1773 + }, + { + "epoch": 1.944109589041096, + "grad_norm": 2.388749837875366, + "learning_rate": 1.9861900893582455e-06, + "loss": 1.7945, + "step": 1774 + }, + { + "epoch": 1.9452054794520548, + "grad_norm": 2.3419551849365234, + "learning_rate": 1.9841592201462227e-06, + "loss": 1.7987, + "step": 1775 + }, + { + "epoch": 1.9463013698630136, + "grad_norm": 2.43770432472229, + "learning_rate": 1.9821283509342e-06, + "loss": 1.7183, + "step": 1776 + }, + { + "epoch": 1.9473972602739726, + "grad_norm": 2.333385467529297, + "learning_rate": 1.9800974817221773e-06, + "loss": 1.7283, + "step": 1777 + }, + { + "epoch": 1.9484931506849315, + "grad_norm": 2.391052722930908, + "learning_rate": 1.9780666125101546e-06, + "loss": 1.776, + "step": 1778 + }, + { + "epoch": 1.9495890410958903, + "grad_norm": 2.375338077545166, + "learning_rate": 1.976035743298132e-06, + "loss": 1.7495, + "step": 1779 + }, + { + "epoch": 1.9506849315068493, + "grad_norm": 2.4482579231262207, + "learning_rate": 1.974004874086109e-06, + "loss": 1.8256, + "step": 1780 + }, + { + "epoch": 1.9506849315068493, + "eval_loss": 1.9600462913513184, + "eval_runtime": 37.2638, + "eval_samples_per_second": 26.836, + "eval_steps_per_second": 1.127, + "step": 1780 + }, + { + "epoch": 1.9517808219178083, + "grad_norm": 2.296039342880249, + "learning_rate": 1.9719740048740864e-06, + "loss": 1.6996, + "step": 1781 + }, + { + "epoch": 1.9528767123287671, + "grad_norm": 2.2494969367980957, + "learning_rate": 1.9699431356620637e-06, + "loss": 1.7401, + "step": 1782 + }, + { + "epoch": 1.953972602739726, + "grad_norm": 2.4078574180603027, + "learning_rate": 1.967912266450041e-06, + "loss": 1.7374, + "step": 1783 + }, + { + "epoch": 1.955068493150685, + "grad_norm": 2.4018476009368896, + "learning_rate": 1.9658813972380182e-06, + "loss": 1.731, + "step": 1784 + }, + { + "epoch": 1.9561643835616438, + "grad_norm": 2.3429362773895264, + "learning_rate": 1.9638505280259955e-06, + "loss": 1.8531, + "step": 1785 + }, + { + "epoch": 1.9572602739726026, + "grad_norm": 2.4287197589874268, + "learning_rate": 1.9618196588139728e-06, + "loss": 1.7271, + "step": 1786 + }, + { + "epoch": 1.9583561643835616, + "grad_norm": 2.338541030883789, + "learning_rate": 1.95978878960195e-06, + "loss": 1.7594, + "step": 1787 + }, + { + "epoch": 1.9594520547945207, + "grad_norm": 2.531425952911377, + "learning_rate": 1.9577579203899273e-06, + "loss": 1.8374, + "step": 1788 + }, + { + "epoch": 1.9605479452054795, + "grad_norm": 2.413638114929199, + "learning_rate": 1.955727051177904e-06, + "loss": 1.8178, + "step": 1789 + }, + { + "epoch": 1.9616438356164383, + "grad_norm": 2.303318977355957, + "learning_rate": 1.9536961819658815e-06, + "loss": 1.7429, + "step": 1790 + }, + { + "epoch": 1.9616438356164383, + "eval_loss": 1.9622509479522705, + "eval_runtime": 37.2699, + "eval_samples_per_second": 26.831, + "eval_steps_per_second": 1.127, + "step": 1790 + }, + { + "epoch": 1.9627397260273973, + "grad_norm": 2.400160074234009, + "learning_rate": 1.9516653127538587e-06, + "loss": 1.7136, + "step": 1791 + }, + { + "epoch": 1.9638356164383561, + "grad_norm": 2.3861589431762695, + "learning_rate": 1.949634443541836e-06, + "loss": 1.7146, + "step": 1792 + }, + { + "epoch": 1.964931506849315, + "grad_norm": 2.272711753845215, + "learning_rate": 1.9476035743298133e-06, + "loss": 1.7177, + "step": 1793 + }, + { + "epoch": 1.966027397260274, + "grad_norm": 2.4745068550109863, + "learning_rate": 1.9455727051177906e-06, + "loss": 1.7938, + "step": 1794 + }, + { + "epoch": 1.967123287671233, + "grad_norm": 2.4643466472625732, + "learning_rate": 1.943541835905768e-06, + "loss": 1.7601, + "step": 1795 + }, + { + "epoch": 1.9682191780821918, + "grad_norm": 2.5346457958221436, + "learning_rate": 1.941510966693745e-06, + "loss": 1.8132, + "step": 1796 + }, + { + "epoch": 1.9693150684931506, + "grad_norm": 2.280719041824341, + "learning_rate": 1.9394800974817224e-06, + "loss": 1.7101, + "step": 1797 + }, + { + "epoch": 1.9704109589041097, + "grad_norm": 2.3304555416107178, + "learning_rate": 1.9374492282696997e-06, + "loss": 1.7809, + "step": 1798 + }, + { + "epoch": 1.9715068493150685, + "grad_norm": 2.409796714782715, + "learning_rate": 1.935418359057677e-06, + "loss": 1.8127, + "step": 1799 + }, + { + "epoch": 1.9726027397260273, + "grad_norm": 2.208312749862671, + "learning_rate": 1.9333874898456542e-06, + "loss": 1.7752, + "step": 1800 + }, + { + "epoch": 1.9726027397260273, + "eval_loss": 1.96150541305542, + "eval_runtime": 37.3221, + "eval_samples_per_second": 26.794, + "eval_steps_per_second": 1.125, + "step": 1800 + }, + { + "epoch": 1.9736986301369863, + "grad_norm": 2.374393939971924, + "learning_rate": 1.931356620633631e-06, + "loss": 1.8188, + "step": 1801 + }, + { + "epoch": 1.9747945205479454, + "grad_norm": 2.3343780040740967, + "learning_rate": 1.9293257514216083e-06, + "loss": 1.7035, + "step": 1802 + }, + { + "epoch": 1.9758904109589042, + "grad_norm": 2.307894229888916, + "learning_rate": 1.9272948822095856e-06, + "loss": 1.7651, + "step": 1803 + }, + { + "epoch": 1.976986301369863, + "grad_norm": 2.387434244155884, + "learning_rate": 1.925264012997563e-06, + "loss": 1.7021, + "step": 1804 + }, + { + "epoch": 1.978082191780822, + "grad_norm": 2.3444128036499023, + "learning_rate": 1.92323314378554e-06, + "loss": 1.7585, + "step": 1805 + }, + { + "epoch": 1.9791780821917808, + "grad_norm": 2.371931791305542, + "learning_rate": 1.9212022745735174e-06, + "loss": 1.7932, + "step": 1806 + }, + { + "epoch": 1.9802739726027396, + "grad_norm": 2.3159632682800293, + "learning_rate": 1.9191714053614947e-06, + "loss": 1.722, + "step": 1807 + }, + { + "epoch": 1.9813698630136987, + "grad_norm": 2.4669857025146484, + "learning_rate": 1.917140536149472e-06, + "loss": 1.7326, + "step": 1808 + }, + { + "epoch": 1.9824657534246577, + "grad_norm": 2.365138292312622, + "learning_rate": 1.9151096669374493e-06, + "loss": 1.7013, + "step": 1809 + }, + { + "epoch": 1.9835616438356163, + "grad_norm": 2.539698362350464, + "learning_rate": 1.9130787977254266e-06, + "loss": 1.8082, + "step": 1810 + }, + { + "epoch": 1.9835616438356163, + "eval_loss": 1.9535884857177734, + "eval_runtime": 37.3497, + "eval_samples_per_second": 26.774, + "eval_steps_per_second": 1.125, + "step": 1810 + }, + { + "epoch": 1.9846575342465753, + "grad_norm": 2.386599540710449, + "learning_rate": 1.911047928513404e-06, + "loss": 1.7835, + "step": 1811 + }, + { + "epoch": 1.9857534246575343, + "grad_norm": 2.376086711883545, + "learning_rate": 1.909017059301381e-06, + "loss": 1.7161, + "step": 1812 + }, + { + "epoch": 1.9868493150684932, + "grad_norm": 2.5276663303375244, + "learning_rate": 1.9069861900893586e-06, + "loss": 1.7222, + "step": 1813 + }, + { + "epoch": 1.987945205479452, + "grad_norm": 2.466137170791626, + "learning_rate": 1.9049553208773354e-06, + "loss": 1.7841, + "step": 1814 + }, + { + "epoch": 1.989041095890411, + "grad_norm": 2.361128091812134, + "learning_rate": 1.9029244516653127e-06, + "loss": 1.7354, + "step": 1815 + }, + { + "epoch": 1.9901369863013698, + "grad_norm": 2.4348530769348145, + "learning_rate": 1.90089358245329e-06, + "loss": 1.6832, + "step": 1816 + }, + { + "epoch": 1.9912328767123286, + "grad_norm": 2.342285633087158, + "learning_rate": 1.8988627132412673e-06, + "loss": 1.7351, + "step": 1817 + }, + { + "epoch": 1.9923287671232877, + "grad_norm": 2.521137237548828, + "learning_rate": 1.8968318440292445e-06, + "loss": 1.794, + "step": 1818 + }, + { + "epoch": 1.9934246575342467, + "grad_norm": 2.5968360900878906, + "learning_rate": 1.8948009748172218e-06, + "loss": 1.6976, + "step": 1819 + }, + { + "epoch": 1.9945205479452055, + "grad_norm": 2.734095335006714, + "learning_rate": 1.892770105605199e-06, + "loss": 1.7919, + "step": 1820 + }, + { + "epoch": 1.9945205479452055, + "eval_loss": 1.9536023139953613, + "eval_runtime": 37.5805, + "eval_samples_per_second": 26.61, + "eval_steps_per_second": 1.118, + "step": 1820 + }, + { + "epoch": 1.9956164383561643, + "grad_norm": 2.327333927154541, + "learning_rate": 1.8907392363931764e-06, + "loss": 1.7891, + "step": 1821 + }, + { + "epoch": 1.9967123287671233, + "grad_norm": 2.2902543544769287, + "learning_rate": 1.8887083671811536e-06, + "loss": 1.7552, + "step": 1822 + }, + { + "epoch": 1.9978082191780822, + "grad_norm": 2.3587660789489746, + "learning_rate": 1.886677497969131e-06, + "loss": 1.7641, + "step": 1823 + }, + { + "epoch": 1.998904109589041, + "grad_norm": 2.4068408012390137, + "learning_rate": 1.8846466287571082e-06, + "loss": 1.6622, + "step": 1824 + }, + { + "epoch": 2.0, + "grad_norm": 2.453514575958252, + "learning_rate": 1.8826157595450855e-06, + "loss": 1.7962, + "step": 1825 + }, + { + "epoch": 2.001095890410959, + "grad_norm": 2.4694175720214844, + "learning_rate": 1.8805848903330625e-06, + "loss": 1.6823, + "step": 1826 + }, + { + "epoch": 2.0021917808219176, + "grad_norm": 2.594013214111328, + "learning_rate": 1.8785540211210398e-06, + "loss": 1.8453, + "step": 1827 + }, + { + "epoch": 2.0032876712328767, + "grad_norm": 2.455796718597412, + "learning_rate": 1.876523151909017e-06, + "loss": 1.7604, + "step": 1828 + }, + { + "epoch": 2.0043835616438357, + "grad_norm": 2.819857120513916, + "learning_rate": 1.8744922826969944e-06, + "loss": 1.7177, + "step": 1829 + }, + { + "epoch": 2.0054794520547947, + "grad_norm": 2.845775842666626, + "learning_rate": 1.8724614134849716e-06, + "loss": 1.7007, + "step": 1830 + }, + { + "epoch": 2.0054794520547947, + "eval_loss": 1.9594684839248657, + "eval_runtime": 37.6856, + "eval_samples_per_second": 26.535, + "eval_steps_per_second": 1.114, + "step": 1830 + }, + { + "epoch": 2.0065753424657533, + "grad_norm": 2.6129424571990967, + "learning_rate": 1.870430544272949e-06, + "loss": 1.7301, + "step": 1831 + }, + { + "epoch": 2.0076712328767123, + "grad_norm": 2.762805223464966, + "learning_rate": 1.8683996750609262e-06, + "loss": 1.8294, + "step": 1832 + }, + { + "epoch": 2.0087671232876714, + "grad_norm": 2.6397714614868164, + "learning_rate": 1.8663688058489035e-06, + "loss": 1.7176, + "step": 1833 + }, + { + "epoch": 2.00986301369863, + "grad_norm": 2.6637110710144043, + "learning_rate": 1.8643379366368807e-06, + "loss": 1.7363, + "step": 1834 + }, + { + "epoch": 2.010958904109589, + "grad_norm": 2.732534408569336, + "learning_rate": 1.862307067424858e-06, + "loss": 1.7496, + "step": 1835 + }, + { + "epoch": 2.012054794520548, + "grad_norm": 2.7163467407226562, + "learning_rate": 1.8602761982128353e-06, + "loss": 1.7551, + "step": 1836 + }, + { + "epoch": 2.013150684931507, + "grad_norm": 2.442484140396118, + "learning_rate": 1.8582453290008126e-06, + "loss": 1.7761, + "step": 1837 + }, + { + "epoch": 2.0142465753424657, + "grad_norm": 2.544834852218628, + "learning_rate": 1.8562144597887898e-06, + "loss": 1.7251, + "step": 1838 + }, + { + "epoch": 2.0153424657534247, + "grad_norm": 2.6588494777679443, + "learning_rate": 1.854183590576767e-06, + "loss": 1.7381, + "step": 1839 + }, + { + "epoch": 2.0164383561643837, + "grad_norm": 2.5374786853790283, + "learning_rate": 1.8521527213647442e-06, + "loss": 1.7831, + "step": 1840 + }, + { + "epoch": 2.0164383561643837, + "eval_loss": 1.9550480842590332, + "eval_runtime": 37.3312, + "eval_samples_per_second": 26.787, + "eval_steps_per_second": 1.125, + "step": 1840 + }, + { + "epoch": 2.0175342465753423, + "grad_norm": 2.4791345596313477, + "learning_rate": 1.8501218521527215e-06, + "loss": 1.6937, + "step": 1841 + }, + { + "epoch": 2.0186301369863013, + "grad_norm": 2.7243971824645996, + "learning_rate": 1.8480909829406987e-06, + "loss": 1.6674, + "step": 1842 + }, + { + "epoch": 2.0197260273972604, + "grad_norm": 2.618680477142334, + "learning_rate": 1.846060113728676e-06, + "loss": 1.6994, + "step": 1843 + }, + { + "epoch": 2.020821917808219, + "grad_norm": 2.597929000854492, + "learning_rate": 1.8440292445166533e-06, + "loss": 1.7375, + "step": 1844 + }, + { + "epoch": 2.021917808219178, + "grad_norm": 2.860177993774414, + "learning_rate": 1.8419983753046306e-06, + "loss": 1.7128, + "step": 1845 + }, + { + "epoch": 2.023013698630137, + "grad_norm": 2.606898307800293, + "learning_rate": 1.8399675060926078e-06, + "loss": 1.7735, + "step": 1846 + }, + { + "epoch": 2.024109589041096, + "grad_norm": 2.5152604579925537, + "learning_rate": 1.8379366368805851e-06, + "loss": 1.75, + "step": 1847 + }, + { + "epoch": 2.0252054794520546, + "grad_norm": 2.678071975708008, + "learning_rate": 1.8359057676685624e-06, + "loss": 1.7114, + "step": 1848 + }, + { + "epoch": 2.0263013698630137, + "grad_norm": 2.673515558242798, + "learning_rate": 1.8338748984565397e-06, + "loss": 1.6992, + "step": 1849 + }, + { + "epoch": 2.0273972602739727, + "grad_norm": 2.5902516841888428, + "learning_rate": 1.831844029244517e-06, + "loss": 1.7195, + "step": 1850 + }, + { + "epoch": 2.0273972602739727, + "eval_loss": 1.953579068183899, + "eval_runtime": 37.304, + "eval_samples_per_second": 26.807, + "eval_steps_per_second": 1.126, + "step": 1850 + }, + { + "epoch": 2.0284931506849313, + "grad_norm": 2.4912915229797363, + "learning_rate": 1.829813160032494e-06, + "loss": 1.6955, + "step": 1851 + }, + { + "epoch": 2.0295890410958903, + "grad_norm": 2.683821678161621, + "learning_rate": 1.8277822908204713e-06, + "loss": 1.7057, + "step": 1852 + }, + { + "epoch": 2.0306849315068494, + "grad_norm": 2.9541101455688477, + "learning_rate": 1.8257514216084486e-06, + "loss": 1.7377, + "step": 1853 + }, + { + "epoch": 2.0317808219178084, + "grad_norm": 2.783851146697998, + "learning_rate": 1.8237205523964258e-06, + "loss": 1.7197, + "step": 1854 + }, + { + "epoch": 2.032876712328767, + "grad_norm": 2.5224950313568115, + "learning_rate": 1.8216896831844031e-06, + "loss": 1.7396, + "step": 1855 + }, + { + "epoch": 2.033972602739726, + "grad_norm": 2.627767562866211, + "learning_rate": 1.8196588139723804e-06, + "loss": 1.7773, + "step": 1856 + }, + { + "epoch": 2.035068493150685, + "grad_norm": 2.443873405456543, + "learning_rate": 1.8176279447603577e-06, + "loss": 1.6656, + "step": 1857 + }, + { + "epoch": 2.0361643835616436, + "grad_norm": 2.648056745529175, + "learning_rate": 1.815597075548335e-06, + "loss": 1.6954, + "step": 1858 + }, + { + "epoch": 2.0372602739726027, + "grad_norm": 2.536935329437256, + "learning_rate": 1.8135662063363122e-06, + "loss": 1.7427, + "step": 1859 + }, + { + "epoch": 2.0383561643835617, + "grad_norm": 2.42183256149292, + "learning_rate": 1.8115353371242895e-06, + "loss": 1.7634, + "step": 1860 + }, + { + "epoch": 2.0383561643835617, + "eval_loss": 1.953397512435913, + "eval_runtime": 37.2886, + "eval_samples_per_second": 26.818, + "eval_steps_per_second": 1.126, + "step": 1860 + }, + { + "epoch": 2.0394520547945207, + "grad_norm": 2.538804769515991, + "learning_rate": 1.8095044679122668e-06, + "loss": 1.6712, + "step": 1861 + }, + { + "epoch": 2.0405479452054793, + "grad_norm": 2.5820095539093018, + "learning_rate": 1.807473598700244e-06, + "loss": 1.7215, + "step": 1862 + }, + { + "epoch": 2.0416438356164384, + "grad_norm": 2.6997382640838623, + "learning_rate": 1.805442729488221e-06, + "loss": 1.7117, + "step": 1863 + }, + { + "epoch": 2.0427397260273974, + "grad_norm": 2.775996685028076, + "learning_rate": 1.8034118602761982e-06, + "loss": 1.7355, + "step": 1864 + }, + { + "epoch": 2.043835616438356, + "grad_norm": 2.6333167552948, + "learning_rate": 1.8013809910641754e-06, + "loss": 1.7343, + "step": 1865 + }, + { + "epoch": 2.044931506849315, + "grad_norm": 2.5830278396606445, + "learning_rate": 1.7993501218521527e-06, + "loss": 1.7339, + "step": 1866 + }, + { + "epoch": 2.046027397260274, + "grad_norm": 2.6145055294036865, + "learning_rate": 1.79731925264013e-06, + "loss": 1.6835, + "step": 1867 + }, + { + "epoch": 2.047123287671233, + "grad_norm": 2.568159580230713, + "learning_rate": 1.7952883834281073e-06, + "loss": 1.7574, + "step": 1868 + }, + { + "epoch": 2.0482191780821917, + "grad_norm": 2.5235254764556885, + "learning_rate": 1.7932575142160846e-06, + "loss": 1.7067, + "step": 1869 + }, + { + "epoch": 2.0493150684931507, + "grad_norm": 2.464035987854004, + "learning_rate": 1.7912266450040618e-06, + "loss": 1.7451, + "step": 1870 + }, + { + "epoch": 2.0493150684931507, + "eval_loss": 1.9485031366348267, + "eval_runtime": 37.2648, + "eval_samples_per_second": 26.835, + "eval_steps_per_second": 1.127, + "step": 1870 + }, + { + "epoch": 2.0504109589041097, + "grad_norm": 2.4328689575195312, + "learning_rate": 1.789195775792039e-06, + "loss": 1.7508, + "step": 1871 + }, + { + "epoch": 2.0515068493150683, + "grad_norm": 2.5483932495117188, + "learning_rate": 1.7871649065800164e-06, + "loss": 1.6488, + "step": 1872 + }, + { + "epoch": 2.0526027397260274, + "grad_norm": 2.5290846824645996, + "learning_rate": 1.7851340373679937e-06, + "loss": 1.7939, + "step": 1873 + }, + { + "epoch": 2.0536986301369864, + "grad_norm": 2.506943702697754, + "learning_rate": 1.783103168155971e-06, + "loss": 1.6813, + "step": 1874 + }, + { + "epoch": 2.0547945205479454, + "grad_norm": 2.5863115787506104, + "learning_rate": 1.7810722989439482e-06, + "loss": 1.7104, + "step": 1875 + }, + { + "epoch": 2.055890410958904, + "grad_norm": 2.6402666568756104, + "learning_rate": 1.7790414297319253e-06, + "loss": 1.7311, + "step": 1876 + }, + { + "epoch": 2.056986301369863, + "grad_norm": 2.6700708866119385, + "learning_rate": 1.7770105605199025e-06, + "loss": 1.695, + "step": 1877 + }, + { + "epoch": 2.058082191780822, + "grad_norm": 2.580693244934082, + "learning_rate": 1.7749796913078798e-06, + "loss": 1.6763, + "step": 1878 + }, + { + "epoch": 2.0591780821917807, + "grad_norm": 2.6035709381103516, + "learning_rate": 1.772948822095857e-06, + "loss": 1.7389, + "step": 1879 + }, + { + "epoch": 2.0602739726027397, + "grad_norm": 2.508784532546997, + "learning_rate": 1.7709179528838344e-06, + "loss": 1.7762, + "step": 1880 + }, + { + "epoch": 2.0602739726027397, + "eval_loss": 1.9516879320144653, + "eval_runtime": 37.2284, + "eval_samples_per_second": 26.861, + "eval_steps_per_second": 1.128, + "step": 1880 + }, + { + "epoch": 2.0613698630136987, + "grad_norm": 2.47988224029541, + "learning_rate": 1.7688870836718116e-06, + "loss": 1.7659, + "step": 1881 + }, + { + "epoch": 2.0624657534246573, + "grad_norm": 2.5782723426818848, + "learning_rate": 1.766856214459789e-06, + "loss": 1.7179, + "step": 1882 + }, + { + "epoch": 2.0635616438356164, + "grad_norm": 2.5616347789764404, + "learning_rate": 1.7648253452477662e-06, + "loss": 1.7163, + "step": 1883 + }, + { + "epoch": 2.0646575342465754, + "grad_norm": 2.4464340209960938, + "learning_rate": 1.7627944760357435e-06, + "loss": 1.7033, + "step": 1884 + }, + { + "epoch": 2.0657534246575344, + "grad_norm": 2.6022887229919434, + "learning_rate": 1.7607636068237208e-06, + "loss": 1.7823, + "step": 1885 + }, + { + "epoch": 2.066849315068493, + "grad_norm": 2.5734951496124268, + "learning_rate": 1.758732737611698e-06, + "loss": 1.7486, + "step": 1886 + }, + { + "epoch": 2.067945205479452, + "grad_norm": 2.6162502765655518, + "learning_rate": 1.7567018683996753e-06, + "loss": 1.7285, + "step": 1887 + }, + { + "epoch": 2.069041095890411, + "grad_norm": 2.654008388519287, + "learning_rate": 1.7546709991876524e-06, + "loss": 1.7195, + "step": 1888 + }, + { + "epoch": 2.0701369863013697, + "grad_norm": 2.64469575881958, + "learning_rate": 1.7526401299756296e-06, + "loss": 1.7247, + "step": 1889 + }, + { + "epoch": 2.0712328767123287, + "grad_norm": 2.5627946853637695, + "learning_rate": 1.750609260763607e-06, + "loss": 1.7254, + "step": 1890 + }, + { + "epoch": 2.0712328767123287, + "eval_loss": 1.951324224472046, + "eval_runtime": 37.2566, + "eval_samples_per_second": 26.841, + "eval_steps_per_second": 1.127, + "step": 1890 + }, + { + "epoch": 2.0723287671232877, + "grad_norm": 2.656520128250122, + "learning_rate": 1.7485783915515842e-06, + "loss": 1.7236, + "step": 1891 + }, + { + "epoch": 2.0734246575342468, + "grad_norm": 2.6355814933776855, + "learning_rate": 1.7465475223395615e-06, + "loss": 1.7118, + "step": 1892 + }, + { + "epoch": 2.0745205479452054, + "grad_norm": 2.5709362030029297, + "learning_rate": 1.7445166531275387e-06, + "loss": 1.685, + "step": 1893 + }, + { + "epoch": 2.0756164383561644, + "grad_norm": 2.555422782897949, + "learning_rate": 1.742485783915516e-06, + "loss": 1.8356, + "step": 1894 + }, + { + "epoch": 2.0767123287671234, + "grad_norm": 2.6279077529907227, + "learning_rate": 1.7404549147034933e-06, + "loss": 1.8079, + "step": 1895 + }, + { + "epoch": 2.077808219178082, + "grad_norm": 2.5886247158050537, + "learning_rate": 1.7384240454914706e-06, + "loss": 1.639, + "step": 1896 + }, + { + "epoch": 2.078904109589041, + "grad_norm": 2.562755584716797, + "learning_rate": 1.7363931762794478e-06, + "loss": 1.7345, + "step": 1897 + }, + { + "epoch": 2.08, + "grad_norm": 2.670344591140747, + "learning_rate": 1.7343623070674251e-06, + "loss": 1.7594, + "step": 1898 + }, + { + "epoch": 2.081095890410959, + "grad_norm": 2.567506790161133, + "learning_rate": 1.7323314378554024e-06, + "loss": 1.7161, + "step": 1899 + }, + { + "epoch": 2.0821917808219177, + "grad_norm": 2.562826156616211, + "learning_rate": 1.7303005686433797e-06, + "loss": 1.7275, + "step": 1900 + }, + { + "epoch": 2.0821917808219177, + "eval_loss": 1.9498006105422974, + "eval_runtime": 37.3069, + "eval_samples_per_second": 26.805, + "eval_steps_per_second": 1.126, + "step": 1900 + }, + { + "epoch": 2.0832876712328767, + "grad_norm": 2.538393974304199, + "learning_rate": 1.7282696994313567e-06, + "loss": 1.7187, + "step": 1901 + }, + { + "epoch": 2.0843835616438358, + "grad_norm": 2.687603235244751, + "learning_rate": 1.726238830219334e-06, + "loss": 1.7349, + "step": 1902 + }, + { + "epoch": 2.0854794520547943, + "grad_norm": 2.5806703567504883, + "learning_rate": 1.7242079610073113e-06, + "loss": 1.7135, + "step": 1903 + }, + { + "epoch": 2.0865753424657534, + "grad_norm": 2.468123435974121, + "learning_rate": 1.7221770917952886e-06, + "loss": 1.7946, + "step": 1904 + }, + { + "epoch": 2.0876712328767124, + "grad_norm": 2.514617443084717, + "learning_rate": 1.7201462225832658e-06, + "loss": 1.7403, + "step": 1905 + }, + { + "epoch": 2.0887671232876714, + "grad_norm": 2.7283825874328613, + "learning_rate": 1.7181153533712431e-06, + "loss": 1.7369, + "step": 1906 + }, + { + "epoch": 2.08986301369863, + "grad_norm": 2.5634899139404297, + "learning_rate": 1.7160844841592204e-06, + "loss": 1.6337, + "step": 1907 + }, + { + "epoch": 2.090958904109589, + "grad_norm": 2.635281801223755, + "learning_rate": 1.7140536149471977e-06, + "loss": 1.6836, + "step": 1908 + }, + { + "epoch": 2.092054794520548, + "grad_norm": 2.550048351287842, + "learning_rate": 1.712022745735175e-06, + "loss": 1.7171, + "step": 1909 + }, + { + "epoch": 2.0931506849315067, + "grad_norm": 2.62712025642395, + "learning_rate": 1.7099918765231522e-06, + "loss": 1.7027, + "step": 1910 + }, + { + "epoch": 2.0931506849315067, + "eval_loss": 1.9503322839736938, + "eval_runtime": 37.3435, + "eval_samples_per_second": 26.778, + "eval_steps_per_second": 1.125, + "step": 1910 + }, + { + "epoch": 2.0942465753424657, + "grad_norm": 2.540708541870117, + "learning_rate": 1.7079610073111295e-06, + "loss": 1.7358, + "step": 1911 + }, + { + "epoch": 2.0953424657534248, + "grad_norm": 2.6046950817108154, + "learning_rate": 1.7059301380991068e-06, + "loss": 1.6538, + "step": 1912 + }, + { + "epoch": 2.0964383561643833, + "grad_norm": 2.4760568141937256, + "learning_rate": 1.7038992688870836e-06, + "loss": 1.7207, + "step": 1913 + }, + { + "epoch": 2.0975342465753424, + "grad_norm": 2.428621292114258, + "learning_rate": 1.701868399675061e-06, + "loss": 1.7092, + "step": 1914 + }, + { + "epoch": 2.0986301369863014, + "grad_norm": 2.6177377700805664, + "learning_rate": 1.6998375304630382e-06, + "loss": 1.7429, + "step": 1915 + }, + { + "epoch": 2.0997260273972604, + "grad_norm": 2.4981932640075684, + "learning_rate": 1.6978066612510155e-06, + "loss": 1.7996, + "step": 1916 + }, + { + "epoch": 2.100821917808219, + "grad_norm": 2.617943286895752, + "learning_rate": 1.6957757920389927e-06, + "loss": 1.7716, + "step": 1917 + }, + { + "epoch": 2.101917808219178, + "grad_norm": 2.7581987380981445, + "learning_rate": 1.69374492282697e-06, + "loss": 1.7194, + "step": 1918 + }, + { + "epoch": 2.103013698630137, + "grad_norm": 2.622145652770996, + "learning_rate": 1.6917140536149473e-06, + "loss": 1.6887, + "step": 1919 + }, + { + "epoch": 2.1041095890410957, + "grad_norm": 2.742100477218628, + "learning_rate": 1.6896831844029246e-06, + "loss": 1.7787, + "step": 1920 + }, + { + "epoch": 2.1041095890410957, + "eval_loss": 1.9521653652191162, + "eval_runtime": 37.5529, + "eval_samples_per_second": 26.629, + "eval_steps_per_second": 1.118, + "step": 1920 + }, + { + "epoch": 2.1052054794520547, + "grad_norm": 2.5809876918792725, + "learning_rate": 1.6876523151909018e-06, + "loss": 1.7392, + "step": 1921 + }, + { + "epoch": 2.1063013698630138, + "grad_norm": 2.7006125450134277, + "learning_rate": 1.6856214459788791e-06, + "loss": 1.7442, + "step": 1922 + }, + { + "epoch": 2.107397260273973, + "grad_norm": 2.566088914871216, + "learning_rate": 1.6835905767668564e-06, + "loss": 1.6862, + "step": 1923 + }, + { + "epoch": 2.1084931506849314, + "grad_norm": 2.508967638015747, + "learning_rate": 1.6815597075548337e-06, + "loss": 1.695, + "step": 1924 + }, + { + "epoch": 2.1095890410958904, + "grad_norm": 2.6175615787506104, + "learning_rate": 1.679528838342811e-06, + "loss": 1.785, + "step": 1925 + }, + { + "epoch": 2.1106849315068494, + "grad_norm": 2.6141958236694336, + "learning_rate": 1.677497969130788e-06, + "loss": 1.6952, + "step": 1926 + }, + { + "epoch": 2.111780821917808, + "grad_norm": 2.4894959926605225, + "learning_rate": 1.6754670999187653e-06, + "loss": 1.6947, + "step": 1927 + }, + { + "epoch": 2.112876712328767, + "grad_norm": 2.5664544105529785, + "learning_rate": 1.6734362307067426e-06, + "loss": 1.731, + "step": 1928 + }, + { + "epoch": 2.113972602739726, + "grad_norm": 2.5422523021698, + "learning_rate": 1.6714053614947198e-06, + "loss": 1.6765, + "step": 1929 + }, + { + "epoch": 2.115068493150685, + "grad_norm": 2.6428608894348145, + "learning_rate": 1.669374492282697e-06, + "loss": 1.6834, + "step": 1930 + }, + { + "epoch": 2.115068493150685, + "eval_loss": 1.952999472618103, + "eval_runtime": 37.6765, + "eval_samples_per_second": 26.542, + "eval_steps_per_second": 1.115, + "step": 1930 + }, + { + "epoch": 2.1161643835616437, + "grad_norm": 2.6088430881500244, + "learning_rate": 1.6673436230706744e-06, + "loss": 1.7713, + "step": 1931 + }, + { + "epoch": 2.1172602739726027, + "grad_norm": 2.49733304977417, + "learning_rate": 1.6653127538586517e-06, + "loss": 1.7461, + "step": 1932 + }, + { + "epoch": 2.118356164383562, + "grad_norm": 2.5760915279388428, + "learning_rate": 1.663281884646629e-06, + "loss": 1.6837, + "step": 1933 + }, + { + "epoch": 2.1194520547945204, + "grad_norm": 2.4771194458007812, + "learning_rate": 1.6612510154346062e-06, + "loss": 1.7191, + "step": 1934 + }, + { + "epoch": 2.1205479452054794, + "grad_norm": 2.6267881393432617, + "learning_rate": 1.6592201462225835e-06, + "loss": 1.7819, + "step": 1935 + }, + { + "epoch": 2.1216438356164384, + "grad_norm": 2.6205623149871826, + "learning_rate": 1.6571892770105608e-06, + "loss": 1.7443, + "step": 1936 + }, + { + "epoch": 2.1227397260273975, + "grad_norm": 2.660942792892456, + "learning_rate": 1.655158407798538e-06, + "loss": 1.783, + "step": 1937 + }, + { + "epoch": 2.123835616438356, + "grad_norm": 2.5452628135681152, + "learning_rate": 1.653127538586515e-06, + "loss": 1.7316, + "step": 1938 + }, + { + "epoch": 2.124931506849315, + "grad_norm": 2.542837142944336, + "learning_rate": 1.6510966693744924e-06, + "loss": 1.7624, + "step": 1939 + }, + { + "epoch": 2.126027397260274, + "grad_norm": 2.535205841064453, + "learning_rate": 1.6490658001624696e-06, + "loss": 1.7664, + "step": 1940 + }, + { + "epoch": 2.126027397260274, + "eval_loss": 1.9492406845092773, + "eval_runtime": 37.3497, + "eval_samples_per_second": 26.774, + "eval_steps_per_second": 1.125, + "step": 1940 + }, + { + "epoch": 2.1271232876712327, + "grad_norm": 2.7622485160827637, + "learning_rate": 1.647034930950447e-06, + "loss": 1.7603, + "step": 1941 + }, + { + "epoch": 2.1282191780821917, + "grad_norm": 2.629058599472046, + "learning_rate": 1.6450040617384242e-06, + "loss": 1.7036, + "step": 1942 + }, + { + "epoch": 2.1293150684931508, + "grad_norm": 2.543135643005371, + "learning_rate": 1.6429731925264015e-06, + "loss": 1.7172, + "step": 1943 + }, + { + "epoch": 2.1304109589041094, + "grad_norm": 2.792086601257324, + "learning_rate": 1.6409423233143788e-06, + "loss": 1.7486, + "step": 1944 + }, + { + "epoch": 2.1315068493150684, + "grad_norm": 2.7788681983947754, + "learning_rate": 1.638911454102356e-06, + "loss": 1.7136, + "step": 1945 + }, + { + "epoch": 2.1326027397260274, + "grad_norm": 2.5364747047424316, + "learning_rate": 1.6368805848903333e-06, + "loss": 1.6766, + "step": 1946 + }, + { + "epoch": 2.1336986301369865, + "grad_norm": 2.5749800205230713, + "learning_rate": 1.6348497156783106e-06, + "loss": 1.7627, + "step": 1947 + }, + { + "epoch": 2.134794520547945, + "grad_norm": 2.4931676387786865, + "learning_rate": 1.6328188464662879e-06, + "loss": 1.7049, + "step": 1948 + }, + { + "epoch": 2.135890410958904, + "grad_norm": 2.547396183013916, + "learning_rate": 1.6307879772542651e-06, + "loss": 1.7165, + "step": 1949 + }, + { + "epoch": 2.136986301369863, + "grad_norm": 2.5881359577178955, + "learning_rate": 1.628757108042242e-06, + "loss": 1.7508, + "step": 1950 + }, + { + "epoch": 2.136986301369863, + "eval_loss": 1.9464337825775146, + "eval_runtime": 37.284, + "eval_samples_per_second": 26.821, + "eval_steps_per_second": 1.126, + "step": 1950 + }, + { + "epoch": 2.138082191780822, + "grad_norm": 2.6131961345672607, + "learning_rate": 1.6267262388302193e-06, + "loss": 1.7717, + "step": 1951 + }, + { + "epoch": 2.1391780821917807, + "grad_norm": 2.5644335746765137, + "learning_rate": 1.6246953696181965e-06, + "loss": 1.7139, + "step": 1952 + }, + { + "epoch": 2.1402739726027398, + "grad_norm": 2.5851385593414307, + "learning_rate": 1.6226645004061738e-06, + "loss": 1.7473, + "step": 1953 + }, + { + "epoch": 2.141369863013699, + "grad_norm": 2.493952989578247, + "learning_rate": 1.620633631194151e-06, + "loss": 1.8209, + "step": 1954 + }, + { + "epoch": 2.1424657534246574, + "grad_norm": 2.5686917304992676, + "learning_rate": 1.6186027619821284e-06, + "loss": 1.7527, + "step": 1955 + }, + { + "epoch": 2.1435616438356164, + "grad_norm": 2.6349050998687744, + "learning_rate": 1.6165718927701056e-06, + "loss": 1.7033, + "step": 1956 + }, + { + "epoch": 2.1446575342465755, + "grad_norm": 2.7004778385162354, + "learning_rate": 1.614541023558083e-06, + "loss": 1.7498, + "step": 1957 + }, + { + "epoch": 2.145753424657534, + "grad_norm": 2.5406992435455322, + "learning_rate": 1.6125101543460602e-06, + "loss": 1.7605, + "step": 1958 + }, + { + "epoch": 2.146849315068493, + "grad_norm": 2.796125650405884, + "learning_rate": 1.6104792851340375e-06, + "loss": 1.7087, + "step": 1959 + }, + { + "epoch": 2.147945205479452, + "grad_norm": 2.703362226486206, + "learning_rate": 1.6084484159220147e-06, + "loss": 1.751, + "step": 1960 + }, + { + "epoch": 2.147945205479452, + "eval_loss": 1.9470759630203247, + "eval_runtime": 37.2721, + "eval_samples_per_second": 26.83, + "eval_steps_per_second": 1.127, + "step": 1960 + }, + { + "epoch": 2.149041095890411, + "grad_norm": 2.6464223861694336, + "learning_rate": 1.606417546709992e-06, + "loss": 1.7147, + "step": 1961 + }, + { + "epoch": 2.1501369863013697, + "grad_norm": 2.6499273777008057, + "learning_rate": 1.6043866774979693e-06, + "loss": 1.7015, + "step": 1962 + }, + { + "epoch": 2.1512328767123288, + "grad_norm": 2.583768844604492, + "learning_rate": 1.6023558082859464e-06, + "loss": 1.7158, + "step": 1963 + }, + { + "epoch": 2.152328767123288, + "grad_norm": 2.6245381832122803, + "learning_rate": 1.6003249390739236e-06, + "loss": 1.6824, + "step": 1964 + }, + { + "epoch": 2.1534246575342464, + "grad_norm": 2.497676134109497, + "learning_rate": 1.598294069861901e-06, + "loss": 1.7345, + "step": 1965 + }, + { + "epoch": 2.1545205479452054, + "grad_norm": 2.7218856811523438, + "learning_rate": 1.5962632006498782e-06, + "loss": 1.6988, + "step": 1966 + }, + { + "epoch": 2.1556164383561645, + "grad_norm": 2.7654666900634766, + "learning_rate": 1.5942323314378555e-06, + "loss": 1.7635, + "step": 1967 + }, + { + "epoch": 2.1567123287671235, + "grad_norm": 2.598526954650879, + "learning_rate": 1.5922014622258327e-06, + "loss": 1.6769, + "step": 1968 + }, + { + "epoch": 2.157808219178082, + "grad_norm": 2.5995986461639404, + "learning_rate": 1.59017059301381e-06, + "loss": 1.6928, + "step": 1969 + }, + { + "epoch": 2.158904109589041, + "grad_norm": 2.5692837238311768, + "learning_rate": 1.5881397238017873e-06, + "loss": 1.7415, + "step": 1970 + }, + { + "epoch": 2.158904109589041, + "eval_loss": 1.9472142457962036, + "eval_runtime": 37.258, + "eval_samples_per_second": 26.84, + "eval_steps_per_second": 1.127, + "step": 1970 + }, + { + "epoch": 2.16, + "grad_norm": 2.622732162475586, + "learning_rate": 1.5861088545897646e-06, + "loss": 1.7383, + "step": 1971 + }, + { + "epoch": 2.1610958904109587, + "grad_norm": 2.573005437850952, + "learning_rate": 1.5840779853777418e-06, + "loss": 1.6696, + "step": 1972 + }, + { + "epoch": 2.1621917808219178, + "grad_norm": 2.7022411823272705, + "learning_rate": 1.5820471161657191e-06, + "loss": 1.7231, + "step": 1973 + }, + { + "epoch": 2.163287671232877, + "grad_norm": 2.739145278930664, + "learning_rate": 1.5800162469536964e-06, + "loss": 1.7012, + "step": 1974 + }, + { + "epoch": 2.1643835616438354, + "grad_norm": 2.7034459114074707, + "learning_rate": 1.5779853777416735e-06, + "loss": 1.769, + "step": 1975 + }, + { + "epoch": 2.1654794520547944, + "grad_norm": 2.745359182357788, + "learning_rate": 1.5759545085296507e-06, + "loss": 1.7424, + "step": 1976 + }, + { + "epoch": 2.1665753424657535, + "grad_norm": 2.722721576690674, + "learning_rate": 1.573923639317628e-06, + "loss": 1.733, + "step": 1977 + }, + { + "epoch": 2.1676712328767125, + "grad_norm": 2.689074754714966, + "learning_rate": 1.5718927701056053e-06, + "loss": 1.7468, + "step": 1978 + }, + { + "epoch": 2.168767123287671, + "grad_norm": 2.6735317707061768, + "learning_rate": 1.5698619008935826e-06, + "loss": 1.7312, + "step": 1979 + }, + { + "epoch": 2.16986301369863, + "grad_norm": 2.6908175945281982, + "learning_rate": 1.5678310316815598e-06, + "loss": 1.7502, + "step": 1980 + }, + { + "epoch": 2.16986301369863, + "eval_loss": 1.947867512702942, + "eval_runtime": 37.1858, + "eval_samples_per_second": 26.892, + "eval_steps_per_second": 1.129, + "step": 1980 + }, + { + "epoch": 2.170958904109589, + "grad_norm": 2.524951934814453, + "learning_rate": 1.5658001624695371e-06, + "loss": 1.6863, + "step": 1981 + }, + { + "epoch": 2.172054794520548, + "grad_norm": 2.746180295944214, + "learning_rate": 1.5637692932575144e-06, + "loss": 1.7621, + "step": 1982 + }, + { + "epoch": 2.1731506849315068, + "grad_norm": 2.4799301624298096, + "learning_rate": 1.5617384240454917e-06, + "loss": 1.7121, + "step": 1983 + }, + { + "epoch": 2.174246575342466, + "grad_norm": 2.6821515560150146, + "learning_rate": 1.559707554833469e-06, + "loss": 1.7271, + "step": 1984 + }, + { + "epoch": 2.175342465753425, + "grad_norm": 2.572122812271118, + "learning_rate": 1.5576766856214462e-06, + "loss": 1.7671, + "step": 1985 + }, + { + "epoch": 2.1764383561643834, + "grad_norm": 2.614806652069092, + "learning_rate": 1.5556458164094235e-06, + "loss": 1.6411, + "step": 1986 + }, + { + "epoch": 2.1775342465753424, + "grad_norm": 2.602883815765381, + "learning_rate": 1.5536149471974008e-06, + "loss": 1.7143, + "step": 1987 + }, + { + "epoch": 2.1786301369863015, + "grad_norm": 2.5893354415893555, + "learning_rate": 1.5515840779853778e-06, + "loss": 1.6733, + "step": 1988 + }, + { + "epoch": 2.17972602739726, + "grad_norm": 2.652308702468872, + "learning_rate": 1.549553208773355e-06, + "loss": 1.7311, + "step": 1989 + }, + { + "epoch": 2.180821917808219, + "grad_norm": 2.7129695415496826, + "learning_rate": 1.5475223395613324e-06, + "loss": 1.7001, + "step": 1990 + }, + { + "epoch": 2.180821917808219, + "eval_loss": 1.9512702226638794, + "eval_runtime": 37.225, + "eval_samples_per_second": 26.864, + "eval_steps_per_second": 1.128, + "step": 1990 + }, + { + "epoch": 2.181917808219178, + "grad_norm": 2.563638210296631, + "learning_rate": 1.5454914703493097e-06, + "loss": 1.7118, + "step": 1991 + }, + { + "epoch": 2.183013698630137, + "grad_norm": 2.629056453704834, + "learning_rate": 1.543460601137287e-06, + "loss": 1.8041, + "step": 1992 + }, + { + "epoch": 2.1841095890410958, + "grad_norm": 2.6072700023651123, + "learning_rate": 1.5414297319252642e-06, + "loss": 1.7568, + "step": 1993 + }, + { + "epoch": 2.185205479452055, + "grad_norm": 2.525784730911255, + "learning_rate": 1.5393988627132415e-06, + "loss": 1.748, + "step": 1994 + }, + { + "epoch": 2.186301369863014, + "grad_norm": 2.564699649810791, + "learning_rate": 1.5373679935012188e-06, + "loss": 1.7148, + "step": 1995 + }, + { + "epoch": 2.1873972602739724, + "grad_norm": 2.7365031242370605, + "learning_rate": 1.535337124289196e-06, + "loss": 1.6714, + "step": 1996 + }, + { + "epoch": 2.1884931506849314, + "grad_norm": 2.719954252243042, + "learning_rate": 1.5333062550771733e-06, + "loss": 1.7301, + "step": 1997 + }, + { + "epoch": 2.1895890410958905, + "grad_norm": 2.7979655265808105, + "learning_rate": 1.5312753858651506e-06, + "loss": 1.8163, + "step": 1998 + }, + { + "epoch": 2.1906849315068495, + "grad_norm": 2.6182446479797363, + "learning_rate": 1.5292445166531279e-06, + "loss": 1.713, + "step": 1999 + }, + { + "epoch": 2.191780821917808, + "grad_norm": 2.7218992710113525, + "learning_rate": 1.5272136474411047e-06, + "loss": 1.7355, + "step": 2000 + }, + { + "epoch": 2.191780821917808, + "eval_loss": 1.950900673866272, + "eval_runtime": 37.2927, + "eval_samples_per_second": 26.815, + "eval_steps_per_second": 1.126, + "step": 2000 + }, + { + "epoch": 2.192876712328767, + "grad_norm": 2.7384393215179443, + "learning_rate": 1.525182778229082e-06, + "loss": 1.7777, + "step": 2001 + }, + { + "epoch": 2.193972602739726, + "grad_norm": 2.7223575115203857, + "learning_rate": 1.5231519090170593e-06, + "loss": 1.7136, + "step": 2002 + }, + { + "epoch": 2.1950684931506848, + "grad_norm": 2.5952577590942383, + "learning_rate": 1.5211210398050365e-06, + "loss": 1.6964, + "step": 2003 + }, + { + "epoch": 2.196164383561644, + "grad_norm": 2.626406669616699, + "learning_rate": 1.5190901705930138e-06, + "loss": 1.6996, + "step": 2004 + }, + { + "epoch": 2.197260273972603, + "grad_norm": 2.6563360691070557, + "learning_rate": 1.517059301380991e-06, + "loss": 1.7428, + "step": 2005 + }, + { + "epoch": 2.198356164383562, + "grad_norm": 2.647664785385132, + "learning_rate": 1.5150284321689684e-06, + "loss": 1.6953, + "step": 2006 + }, + { + "epoch": 2.1994520547945204, + "grad_norm": 2.5670430660247803, + "learning_rate": 1.5129975629569456e-06, + "loss": 1.7215, + "step": 2007 + }, + { + "epoch": 2.2005479452054795, + "grad_norm": 2.6148288249969482, + "learning_rate": 1.510966693744923e-06, + "loss": 1.724, + "step": 2008 + }, + { + "epoch": 2.2016438356164385, + "grad_norm": 2.8430588245391846, + "learning_rate": 1.5089358245329002e-06, + "loss": 1.7255, + "step": 2009 + }, + { + "epoch": 2.202739726027397, + "grad_norm": 2.5937509536743164, + "learning_rate": 1.5069049553208775e-06, + "loss": 1.7055, + "step": 2010 + }, + { + "epoch": 2.202739726027397, + "eval_loss": 1.9469048976898193, + "eval_runtime": 37.2173, + "eval_samples_per_second": 26.869, + "eval_steps_per_second": 1.129, + "step": 2010 + }, + { + "epoch": 2.203835616438356, + "grad_norm": 2.7167937755584717, + "learning_rate": 1.5048740861088547e-06, + "loss": 1.7474, + "step": 2011 + }, + { + "epoch": 2.204931506849315, + "grad_norm": 2.53837251663208, + "learning_rate": 1.502843216896832e-06, + "loss": 1.74, + "step": 2012 + }, + { + "epoch": 2.206027397260274, + "grad_norm": 2.604931592941284, + "learning_rate": 1.500812347684809e-06, + "loss": 1.6573, + "step": 2013 + }, + { + "epoch": 2.207123287671233, + "grad_norm": 2.6670846939086914, + "learning_rate": 1.4987814784727864e-06, + "loss": 1.7208, + "step": 2014 + }, + { + "epoch": 2.208219178082192, + "grad_norm": 2.677746534347534, + "learning_rate": 1.4967506092607636e-06, + "loss": 1.728, + "step": 2015 + }, + { + "epoch": 2.209315068493151, + "grad_norm": 2.6490612030029297, + "learning_rate": 1.494719740048741e-06, + "loss": 1.6728, + "step": 2016 + }, + { + "epoch": 2.2104109589041094, + "grad_norm": 2.6571807861328125, + "learning_rate": 1.4926888708367182e-06, + "loss": 1.7528, + "step": 2017 + }, + { + "epoch": 2.2115068493150685, + "grad_norm": 2.5933258533477783, + "learning_rate": 1.4906580016246955e-06, + "loss": 1.757, + "step": 2018 + }, + { + "epoch": 2.2126027397260275, + "grad_norm": 2.6451070308685303, + "learning_rate": 1.4886271324126727e-06, + "loss": 1.763, + "step": 2019 + }, + { + "epoch": 2.213698630136986, + "grad_norm": 2.5259275436401367, + "learning_rate": 1.48659626320065e-06, + "loss": 1.7979, + "step": 2020 + }, + { + "epoch": 2.213698630136986, + "eval_loss": 1.9480665922164917, + "eval_runtime": 37.5325, + "eval_samples_per_second": 26.644, + "eval_steps_per_second": 1.119, + "step": 2020 + }, + { + "epoch": 2.214794520547945, + "grad_norm": 2.6959381103515625, + "learning_rate": 1.4845653939886273e-06, + "loss": 1.6376, + "step": 2021 + }, + { + "epoch": 2.215890410958904, + "grad_norm": 2.6179497241973877, + "learning_rate": 1.4825345247766046e-06, + "loss": 1.6575, + "step": 2022 + }, + { + "epoch": 2.216986301369863, + "grad_norm": 2.5618393421173096, + "learning_rate": 1.4805036555645818e-06, + "loss": 1.6486, + "step": 2023 + }, + { + "epoch": 2.2180821917808218, + "grad_norm": 2.545051336288452, + "learning_rate": 1.4784727863525591e-06, + "loss": 1.7218, + "step": 2024 + }, + { + "epoch": 2.219178082191781, + "grad_norm": 2.6248512268066406, + "learning_rate": 1.4764419171405362e-06, + "loss": 1.691, + "step": 2025 + }, + { + "epoch": 2.22027397260274, + "grad_norm": 2.65645170211792, + "learning_rate": 1.4744110479285135e-06, + "loss": 1.7602, + "step": 2026 + }, + { + "epoch": 2.2213698630136984, + "grad_norm": 2.6771113872528076, + "learning_rate": 1.4723801787164907e-06, + "loss": 1.7573, + "step": 2027 + }, + { + "epoch": 2.2224657534246575, + "grad_norm": 2.5466039180755615, + "learning_rate": 1.470349309504468e-06, + "loss": 1.7604, + "step": 2028 + }, + { + "epoch": 2.2235616438356165, + "grad_norm": 2.685079574584961, + "learning_rate": 1.4683184402924453e-06, + "loss": 1.7341, + "step": 2029 + }, + { + "epoch": 2.2246575342465755, + "grad_norm": 2.6854352951049805, + "learning_rate": 1.4662875710804226e-06, + "loss": 1.7427, + "step": 2030 + }, + { + "epoch": 2.2246575342465755, + "eval_loss": 1.9459377527236938, + "eval_runtime": 37.5255, + "eval_samples_per_second": 26.649, + "eval_steps_per_second": 1.119, + "step": 2030 + }, + { + "epoch": 2.225753424657534, + "grad_norm": 2.667675733566284, + "learning_rate": 1.4642567018683998e-06, + "loss": 1.8328, + "step": 2031 + }, + { + "epoch": 2.226849315068493, + "grad_norm": 2.603386640548706, + "learning_rate": 1.4622258326563771e-06, + "loss": 1.7193, + "step": 2032 + }, + { + "epoch": 2.227945205479452, + "grad_norm": 2.4960451126098633, + "learning_rate": 1.4601949634443544e-06, + "loss": 1.695, + "step": 2033 + }, + { + "epoch": 2.2290410958904108, + "grad_norm": 2.755707025527954, + "learning_rate": 1.4581640942323317e-06, + "loss": 1.7475, + "step": 2034 + }, + { + "epoch": 2.23013698630137, + "grad_norm": 2.5849006175994873, + "learning_rate": 1.456133225020309e-06, + "loss": 1.8106, + "step": 2035 + }, + { + "epoch": 2.231232876712329, + "grad_norm": 2.690680980682373, + "learning_rate": 1.4541023558082862e-06, + "loss": 1.7368, + "step": 2036 + }, + { + "epoch": 2.232328767123288, + "grad_norm": 2.6760425567626953, + "learning_rate": 1.4520714865962635e-06, + "loss": 1.722, + "step": 2037 + }, + { + "epoch": 2.2334246575342465, + "grad_norm": 2.656379461288452, + "learning_rate": 1.4500406173842406e-06, + "loss": 1.7623, + "step": 2038 + }, + { + "epoch": 2.2345205479452055, + "grad_norm": 2.8213517665863037, + "learning_rate": 1.4480097481722178e-06, + "loss": 1.7344, + "step": 2039 + }, + { + "epoch": 2.2356164383561645, + "grad_norm": 2.877163887023926, + "learning_rate": 1.4459788789601951e-06, + "loss": 1.7184, + "step": 2040 + }, + { + "epoch": 2.2356164383561645, + "eval_loss": 1.9430633783340454, + "eval_runtime": 37.3102, + "eval_samples_per_second": 26.802, + "eval_steps_per_second": 1.126, + "step": 2040 + }, + { + "epoch": 2.236712328767123, + "grad_norm": 2.7285237312316895, + "learning_rate": 1.4439480097481724e-06, + "loss": 1.6826, + "step": 2041 + }, + { + "epoch": 2.237808219178082, + "grad_norm": 2.5208237171173096, + "learning_rate": 1.4419171405361497e-06, + "loss": 1.6788, + "step": 2042 + }, + { + "epoch": 2.238904109589041, + "grad_norm": 2.7415900230407715, + "learning_rate": 1.439886271324127e-06, + "loss": 1.7423, + "step": 2043 + }, + { + "epoch": 2.24, + "grad_norm": 2.760897397994995, + "learning_rate": 1.4378554021121042e-06, + "loss": 1.7341, + "step": 2044 + }, + { + "epoch": 2.241095890410959, + "grad_norm": 2.5650556087493896, + "learning_rate": 1.4358245329000815e-06, + "loss": 1.7036, + "step": 2045 + }, + { + "epoch": 2.242191780821918, + "grad_norm": 2.60780930519104, + "learning_rate": 1.4337936636880588e-06, + "loss": 1.7719, + "step": 2046 + }, + { + "epoch": 2.243287671232877, + "grad_norm": 2.592473268508911, + "learning_rate": 1.431762794476036e-06, + "loss": 1.7421, + "step": 2047 + }, + { + "epoch": 2.2443835616438355, + "grad_norm": 2.6672041416168213, + "learning_rate": 1.4297319252640133e-06, + "loss": 1.6865, + "step": 2048 + }, + { + "epoch": 2.2454794520547945, + "grad_norm": 2.6182847023010254, + "learning_rate": 1.4277010560519906e-06, + "loss": 1.731, + "step": 2049 + }, + { + "epoch": 2.2465753424657535, + "grad_norm": 2.8213913440704346, + "learning_rate": 1.4256701868399674e-06, + "loss": 1.6845, + "step": 2050 + }, + { + "epoch": 2.2465753424657535, + "eval_loss": 1.9434115886688232, + "eval_runtime": 37.3048, + "eval_samples_per_second": 26.806, + "eval_steps_per_second": 1.126, + "step": 2050 + }, + { + "epoch": 2.247671232876712, + "grad_norm": 2.7278084754943848, + "learning_rate": 1.4236393176279447e-06, + "loss": 1.8077, + "step": 2051 + }, + { + "epoch": 2.248767123287671, + "grad_norm": 2.5665392875671387, + "learning_rate": 1.421608448415922e-06, + "loss": 1.7235, + "step": 2052 + }, + { + "epoch": 2.24986301369863, + "grad_norm": 2.680154323577881, + "learning_rate": 1.4195775792038993e-06, + "loss": 1.8164, + "step": 2053 + }, + { + "epoch": 2.250958904109589, + "grad_norm": 2.770051956176758, + "learning_rate": 1.4175467099918765e-06, + "loss": 1.6948, + "step": 2054 + }, + { + "epoch": 2.252054794520548, + "grad_norm": 2.6189041137695312, + "learning_rate": 1.4155158407798538e-06, + "loss": 1.7369, + "step": 2055 + }, + { + "epoch": 2.253150684931507, + "grad_norm": 2.6655054092407227, + "learning_rate": 1.413484971567831e-06, + "loss": 1.7456, + "step": 2056 + }, + { + "epoch": 2.254246575342466, + "grad_norm": 2.6520164012908936, + "learning_rate": 1.4114541023558084e-06, + "loss": 1.6719, + "step": 2057 + }, + { + "epoch": 2.255342465753425, + "grad_norm": 2.594533681869507, + "learning_rate": 1.4094232331437856e-06, + "loss": 1.7078, + "step": 2058 + }, + { + "epoch": 2.2564383561643835, + "grad_norm": 2.592308759689331, + "learning_rate": 1.407392363931763e-06, + "loss": 1.7517, + "step": 2059 + }, + { + "epoch": 2.2575342465753425, + "grad_norm": 2.5815367698669434, + "learning_rate": 1.4053614947197402e-06, + "loss": 1.8262, + "step": 2060 + }, + { + "epoch": 2.2575342465753425, + "eval_loss": 1.945846438407898, + "eval_runtime": 37.186, + "eval_samples_per_second": 26.892, + "eval_steps_per_second": 1.129, + "step": 2060 + }, + { + "epoch": 2.2586301369863016, + "grad_norm": 2.643691301345825, + "learning_rate": 1.4033306255077175e-06, + "loss": 1.7559, + "step": 2061 + }, + { + "epoch": 2.25972602739726, + "grad_norm": 2.6091558933258057, + "learning_rate": 1.4012997562956945e-06, + "loss": 1.6433, + "step": 2062 + }, + { + "epoch": 2.260821917808219, + "grad_norm": 2.7172136306762695, + "learning_rate": 1.3992688870836718e-06, + "loss": 1.7735, + "step": 2063 + }, + { + "epoch": 2.261917808219178, + "grad_norm": 2.5673139095306396, + "learning_rate": 1.397238017871649e-06, + "loss": 1.7402, + "step": 2064 + }, + { + "epoch": 2.263013698630137, + "grad_norm": 2.711242198944092, + "learning_rate": 1.3952071486596264e-06, + "loss": 1.7183, + "step": 2065 + }, + { + "epoch": 2.264109589041096, + "grad_norm": 2.6999948024749756, + "learning_rate": 1.3931762794476036e-06, + "loss": 1.6856, + "step": 2066 + }, + { + "epoch": 2.265205479452055, + "grad_norm": 2.670964002609253, + "learning_rate": 1.391145410235581e-06, + "loss": 1.78, + "step": 2067 + }, + { + "epoch": 2.266301369863014, + "grad_norm": 2.4599525928497314, + "learning_rate": 1.3891145410235582e-06, + "loss": 1.7143, + "step": 2068 + }, + { + "epoch": 2.2673972602739725, + "grad_norm": 2.552266836166382, + "learning_rate": 1.3870836718115355e-06, + "loss": 1.6503, + "step": 2069 + }, + { + "epoch": 2.2684931506849315, + "grad_norm": 2.5764405727386475, + "learning_rate": 1.3850528025995127e-06, + "loss": 1.7046, + "step": 2070 + }, + { + "epoch": 2.2684931506849315, + "eval_loss": 1.9457523822784424, + "eval_runtime": 37.217, + "eval_samples_per_second": 26.869, + "eval_steps_per_second": 1.129, + "step": 2070 + }, + { + "epoch": 2.2695890410958905, + "grad_norm": 2.524228572845459, + "learning_rate": 1.38302193338749e-06, + "loss": 1.7078, + "step": 2071 + }, + { + "epoch": 2.270684931506849, + "grad_norm": 2.5531418323516846, + "learning_rate": 1.3809910641754673e-06, + "loss": 1.646, + "step": 2072 + }, + { + "epoch": 2.271780821917808, + "grad_norm": 2.705824613571167, + "learning_rate": 1.3789601949634446e-06, + "loss": 1.7407, + "step": 2073 + }, + { + "epoch": 2.272876712328767, + "grad_norm": 2.5266356468200684, + "learning_rate": 1.3769293257514218e-06, + "loss": 1.6472, + "step": 2074 + }, + { + "epoch": 2.2739726027397262, + "grad_norm": 2.5785422325134277, + "learning_rate": 1.374898456539399e-06, + "loss": 1.7561, + "step": 2075 + }, + { + "epoch": 2.275068493150685, + "grad_norm": 2.663120985031128, + "learning_rate": 1.3728675873273762e-06, + "loss": 1.6901, + "step": 2076 + }, + { + "epoch": 2.276164383561644, + "grad_norm": 2.6085517406463623, + "learning_rate": 1.3708367181153535e-06, + "loss": 1.7829, + "step": 2077 + }, + { + "epoch": 2.277260273972603, + "grad_norm": 2.6439929008483887, + "learning_rate": 1.3688058489033307e-06, + "loss": 1.7684, + "step": 2078 + }, + { + "epoch": 2.2783561643835615, + "grad_norm": 2.6614694595336914, + "learning_rate": 1.366774979691308e-06, + "loss": 1.726, + "step": 2079 + }, + { + "epoch": 2.2794520547945205, + "grad_norm": 2.7339539527893066, + "learning_rate": 1.3647441104792853e-06, + "loss": 1.7577, + "step": 2080 + }, + { + "epoch": 2.2794520547945205, + "eval_loss": 1.9456545114517212, + "eval_runtime": 37.174, + "eval_samples_per_second": 26.901, + "eval_steps_per_second": 1.13, + "step": 2080 + }, + { + "epoch": 2.2805479452054795, + "grad_norm": 2.603545665740967, + "learning_rate": 1.3627132412672626e-06, + "loss": 1.7297, + "step": 2081 + }, + { + "epoch": 2.281643835616438, + "grad_norm": 2.548466682434082, + "learning_rate": 1.3606823720552398e-06, + "loss": 1.6701, + "step": 2082 + }, + { + "epoch": 2.282739726027397, + "grad_norm": 2.6043875217437744, + "learning_rate": 1.3586515028432171e-06, + "loss": 1.7369, + "step": 2083 + }, + { + "epoch": 2.283835616438356, + "grad_norm": 2.7412586212158203, + "learning_rate": 1.3566206336311944e-06, + "loss": 1.654, + "step": 2084 + }, + { + "epoch": 2.2849315068493152, + "grad_norm": 2.669410228729248, + "learning_rate": 1.3545897644191717e-06, + "loss": 1.7186, + "step": 2085 + }, + { + "epoch": 2.286027397260274, + "grad_norm": 2.7052674293518066, + "learning_rate": 1.352558895207149e-06, + "loss": 1.7005, + "step": 2086 + }, + { + "epoch": 2.287123287671233, + "grad_norm": 2.5840418338775635, + "learning_rate": 1.3505280259951258e-06, + "loss": 1.7459, + "step": 2087 + }, + { + "epoch": 2.288219178082192, + "grad_norm": 2.6355416774749756, + "learning_rate": 1.348497156783103e-06, + "loss": 1.7243, + "step": 2088 + }, + { + "epoch": 2.289315068493151, + "grad_norm": 2.6697628498077393, + "learning_rate": 1.3464662875710804e-06, + "loss": 1.7597, + "step": 2089 + }, + { + "epoch": 2.2904109589041095, + "grad_norm": 2.679842233657837, + "learning_rate": 1.3444354183590576e-06, + "loss": 1.6557, + "step": 2090 + }, + { + "epoch": 2.2904109589041095, + "eval_loss": 1.9469218254089355, + "eval_runtime": 37.1615, + "eval_samples_per_second": 26.91, + "eval_steps_per_second": 1.13, + "step": 2090 + }, + { + "epoch": 2.2915068493150685, + "grad_norm": 2.8354430198669434, + "learning_rate": 1.342404549147035e-06, + "loss": 1.7496, + "step": 2091 + }, + { + "epoch": 2.2926027397260276, + "grad_norm": 2.6957387924194336, + "learning_rate": 1.3403736799350122e-06, + "loss": 1.6872, + "step": 2092 + }, + { + "epoch": 2.293698630136986, + "grad_norm": 2.5815868377685547, + "learning_rate": 1.3383428107229895e-06, + "loss": 1.7497, + "step": 2093 + }, + { + "epoch": 2.294794520547945, + "grad_norm": 2.727085828781128, + "learning_rate": 1.3363119415109667e-06, + "loss": 1.7723, + "step": 2094 + }, + { + "epoch": 2.2958904109589042, + "grad_norm": 2.6119000911712646, + "learning_rate": 1.334281072298944e-06, + "loss": 1.7414, + "step": 2095 + }, + { + "epoch": 2.296986301369863, + "grad_norm": 2.6403088569641113, + "learning_rate": 1.3322502030869213e-06, + "loss": 1.7398, + "step": 2096 + }, + { + "epoch": 2.298082191780822, + "grad_norm": 2.5066816806793213, + "learning_rate": 1.3302193338748986e-06, + "loss": 1.6974, + "step": 2097 + }, + { + "epoch": 2.299178082191781, + "grad_norm": 2.6601762771606445, + "learning_rate": 1.3281884646628758e-06, + "loss": 1.7578, + "step": 2098 + }, + { + "epoch": 2.30027397260274, + "grad_norm": 2.571499824523926, + "learning_rate": 1.3261575954508531e-06, + "loss": 1.7573, + "step": 2099 + }, + { + "epoch": 2.3013698630136985, + "grad_norm": 2.533543825149536, + "learning_rate": 1.3241267262388302e-06, + "loss": 1.7566, + "step": 2100 + }, + { + "epoch": 2.3013698630136985, + "eval_loss": 1.9441688060760498, + "eval_runtime": 37.2158, + "eval_samples_per_second": 26.87, + "eval_steps_per_second": 1.129, + "step": 2100 + }, + { + "epoch": 2.3024657534246575, + "grad_norm": 2.782416582107544, + "learning_rate": 1.3220958570268074e-06, + "loss": 1.714, + "step": 2101 + }, + { + "epoch": 2.3035616438356166, + "grad_norm": 2.742504358291626, + "learning_rate": 1.3200649878147847e-06, + "loss": 1.6677, + "step": 2102 + }, + { + "epoch": 2.304657534246575, + "grad_norm": 2.7778940200805664, + "learning_rate": 1.318034118602762e-06, + "loss": 1.7178, + "step": 2103 + }, + { + "epoch": 2.305753424657534, + "grad_norm": 2.7003014087677, + "learning_rate": 1.3160032493907393e-06, + "loss": 1.7691, + "step": 2104 + }, + { + "epoch": 2.3068493150684932, + "grad_norm": 2.6644296646118164, + "learning_rate": 1.3139723801787166e-06, + "loss": 1.7285, + "step": 2105 + }, + { + "epoch": 2.3079452054794523, + "grad_norm": 2.7859749794006348, + "learning_rate": 1.3119415109666938e-06, + "loss": 1.8026, + "step": 2106 + }, + { + "epoch": 2.309041095890411, + "grad_norm": 2.6862375736236572, + "learning_rate": 1.309910641754671e-06, + "loss": 1.7238, + "step": 2107 + }, + { + "epoch": 2.31013698630137, + "grad_norm": 2.749908685684204, + "learning_rate": 1.3078797725426484e-06, + "loss": 1.676, + "step": 2108 + }, + { + "epoch": 2.311232876712329, + "grad_norm": 2.679384469985962, + "learning_rate": 1.3058489033306257e-06, + "loss": 1.6899, + "step": 2109 + }, + { + "epoch": 2.3123287671232875, + "grad_norm": 2.5621979236602783, + "learning_rate": 1.303818034118603e-06, + "loss": 1.6963, + "step": 2110 + }, + { + "epoch": 2.3123287671232875, + "eval_loss": 1.9453113079071045, + "eval_runtime": 37.2803, + "eval_samples_per_second": 26.824, + "eval_steps_per_second": 1.127, + "step": 2110 + }, + { + "epoch": 2.3134246575342465, + "grad_norm": 2.6183035373687744, + "learning_rate": 1.3017871649065802e-06, + "loss": 1.7272, + "step": 2111 + }, + { + "epoch": 2.3145205479452056, + "grad_norm": 2.5777511596679688, + "learning_rate": 1.2997562956945573e-06, + "loss": 1.7041, + "step": 2112 + }, + { + "epoch": 2.315616438356164, + "grad_norm": 2.784181594848633, + "learning_rate": 1.2977254264825345e-06, + "loss": 1.6835, + "step": 2113 + }, + { + "epoch": 2.316712328767123, + "grad_norm": 2.679591417312622, + "learning_rate": 1.2956945572705118e-06, + "loss": 1.7242, + "step": 2114 + }, + { + "epoch": 2.317808219178082, + "grad_norm": 2.561269521713257, + "learning_rate": 1.293663688058489e-06, + "loss": 1.7455, + "step": 2115 + }, + { + "epoch": 2.3189041095890413, + "grad_norm": 2.5087854862213135, + "learning_rate": 1.2916328188464664e-06, + "loss": 1.6824, + "step": 2116 + }, + { + "epoch": 2.32, + "grad_norm": 2.7247722148895264, + "learning_rate": 1.2896019496344436e-06, + "loss": 1.6268, + "step": 2117 + }, + { + "epoch": 2.321095890410959, + "grad_norm": 2.603994846343994, + "learning_rate": 1.287571080422421e-06, + "loss": 1.6925, + "step": 2118 + }, + { + "epoch": 2.322191780821918, + "grad_norm": 2.5525155067443848, + "learning_rate": 1.2855402112103982e-06, + "loss": 1.7643, + "step": 2119 + }, + { + "epoch": 2.323287671232877, + "grad_norm": 2.5990493297576904, + "learning_rate": 1.2835093419983755e-06, + "loss": 1.6712, + "step": 2120 + }, + { + "epoch": 2.323287671232877, + "eval_loss": 1.9467085599899292, + "eval_runtime": 37.5353, + "eval_samples_per_second": 26.642, + "eval_steps_per_second": 1.119, + "step": 2120 + }, + { + "epoch": 2.3243835616438355, + "grad_norm": 2.764711618423462, + "learning_rate": 1.2814784727863528e-06, + "loss": 1.71, + "step": 2121 + }, + { + "epoch": 2.3254794520547946, + "grad_norm": 2.6009247303009033, + "learning_rate": 1.27944760357433e-06, + "loss": 1.6956, + "step": 2122 + }, + { + "epoch": 2.3265753424657536, + "grad_norm": 2.6900346279144287, + "learning_rate": 1.2774167343623073e-06, + "loss": 1.7529, + "step": 2123 + }, + { + "epoch": 2.327671232876712, + "grad_norm": 2.6602327823638916, + "learning_rate": 1.2753858651502846e-06, + "loss": 1.7588, + "step": 2124 + }, + { + "epoch": 2.328767123287671, + "grad_norm": 2.7727952003479004, + "learning_rate": 1.2733549959382616e-06, + "loss": 1.7499, + "step": 2125 + }, + { + "epoch": 2.3298630136986302, + "grad_norm": 2.703307867050171, + "learning_rate": 1.271324126726239e-06, + "loss": 1.7169, + "step": 2126 + }, + { + "epoch": 2.330958904109589, + "grad_norm": 2.7290568351745605, + "learning_rate": 1.2692932575142162e-06, + "loss": 1.7076, + "step": 2127 + }, + { + "epoch": 2.332054794520548, + "grad_norm": 2.664865255355835, + "learning_rate": 1.2672623883021935e-06, + "loss": 1.6655, + "step": 2128 + }, + { + "epoch": 2.333150684931507, + "grad_norm": 2.6096653938293457, + "learning_rate": 1.2652315190901707e-06, + "loss": 1.6644, + "step": 2129 + }, + { + "epoch": 2.334246575342466, + "grad_norm": 2.7201335430145264, + "learning_rate": 1.263200649878148e-06, + "loss": 1.7924, + "step": 2130 + }, + { + "epoch": 2.334246575342466, + "eval_loss": 1.9422904253005981, + "eval_runtime": 37.5421, + "eval_samples_per_second": 26.637, + "eval_steps_per_second": 1.119, + "step": 2130 + }, + { + "epoch": 2.3353424657534245, + "grad_norm": 2.519702434539795, + "learning_rate": 1.2611697806661253e-06, + "loss": 1.7108, + "step": 2131 + }, + { + "epoch": 2.3364383561643836, + "grad_norm": 2.6535212993621826, + "learning_rate": 1.2591389114541026e-06, + "loss": 1.7243, + "step": 2132 + }, + { + "epoch": 2.3375342465753426, + "grad_norm": 2.651458501815796, + "learning_rate": 1.2571080422420798e-06, + "loss": 1.7158, + "step": 2133 + }, + { + "epoch": 2.338630136986301, + "grad_norm": 2.6745946407318115, + "learning_rate": 1.2550771730300571e-06, + "loss": 1.7014, + "step": 2134 + }, + { + "epoch": 2.33972602739726, + "grad_norm": 2.6088216304779053, + "learning_rate": 1.2530463038180344e-06, + "loss": 1.7724, + "step": 2135 + }, + { + "epoch": 2.3408219178082192, + "grad_norm": 2.703014373779297, + "learning_rate": 1.2510154346060117e-06, + "loss": 1.7104, + "step": 2136 + }, + { + "epoch": 2.3419178082191783, + "grad_norm": 2.57769513130188, + "learning_rate": 1.2489845653939887e-06, + "loss": 1.7259, + "step": 2137 + }, + { + "epoch": 2.343013698630137, + "grad_norm": 2.596695899963379, + "learning_rate": 1.246953696181966e-06, + "loss": 1.7571, + "step": 2138 + }, + { + "epoch": 2.344109589041096, + "grad_norm": 2.6988871097564697, + "learning_rate": 1.2449228269699433e-06, + "loss": 1.8002, + "step": 2139 + }, + { + "epoch": 2.345205479452055, + "grad_norm": 2.6352357864379883, + "learning_rate": 1.2428919577579204e-06, + "loss": 1.7213, + "step": 2140 + }, + { + "epoch": 2.345205479452055, + "eval_loss": 1.9401096105575562, + "eval_runtime": 37.4404, + "eval_samples_per_second": 26.709, + "eval_steps_per_second": 1.122, + "step": 2140 + }, + { + "epoch": 2.3463013698630135, + "grad_norm": 2.7089645862579346, + "learning_rate": 1.2408610885458976e-06, + "loss": 1.6737, + "step": 2141 + }, + { + "epoch": 2.3473972602739726, + "grad_norm": 2.6364030838012695, + "learning_rate": 1.238830219333875e-06, + "loss": 1.7186, + "step": 2142 + }, + { + "epoch": 2.3484931506849316, + "grad_norm": 2.5563573837280273, + "learning_rate": 1.2367993501218522e-06, + "loss": 1.6956, + "step": 2143 + }, + { + "epoch": 2.34958904109589, + "grad_norm": 2.638512372970581, + "learning_rate": 1.2347684809098295e-06, + "loss": 1.7693, + "step": 2144 + }, + { + "epoch": 2.350684931506849, + "grad_norm": 2.733036518096924, + "learning_rate": 1.2327376116978067e-06, + "loss": 1.7327, + "step": 2145 + }, + { + "epoch": 2.3517808219178082, + "grad_norm": 2.6013307571411133, + "learning_rate": 1.230706742485784e-06, + "loss": 1.6539, + "step": 2146 + }, + { + "epoch": 2.3528767123287673, + "grad_norm": 2.7629592418670654, + "learning_rate": 1.2286758732737613e-06, + "loss": 1.666, + "step": 2147 + }, + { + "epoch": 2.353972602739726, + "grad_norm": 2.6855108737945557, + "learning_rate": 1.2266450040617386e-06, + "loss": 1.8044, + "step": 2148 + }, + { + "epoch": 2.355068493150685, + "grad_norm": 2.607609748840332, + "learning_rate": 1.2246141348497158e-06, + "loss": 1.8022, + "step": 2149 + }, + { + "epoch": 2.356164383561644, + "grad_norm": 2.5609383583068848, + "learning_rate": 1.2225832656376931e-06, + "loss": 1.73, + "step": 2150 + }, + { + "epoch": 2.356164383561644, + "eval_loss": 1.9387279748916626, + "eval_runtime": 37.2896, + "eval_samples_per_second": 26.817, + "eval_steps_per_second": 1.126, + "step": 2150 + }, + { + "epoch": 2.357260273972603, + "grad_norm": 2.523988723754883, + "learning_rate": 1.2205523964256704e-06, + "loss": 1.7289, + "step": 2151 + }, + { + "epoch": 2.3583561643835615, + "grad_norm": 2.675995111465454, + "learning_rate": 1.2185215272136477e-06, + "loss": 1.7036, + "step": 2152 + }, + { + "epoch": 2.3594520547945206, + "grad_norm": 2.535367012023926, + "learning_rate": 1.2164906580016247e-06, + "loss": 1.7309, + "step": 2153 + }, + { + "epoch": 2.3605479452054796, + "grad_norm": 2.6444637775421143, + "learning_rate": 1.214459788789602e-06, + "loss": 1.7192, + "step": 2154 + }, + { + "epoch": 2.361643835616438, + "grad_norm": 2.5746796131134033, + "learning_rate": 1.2124289195775793e-06, + "loss": 1.6343, + "step": 2155 + }, + { + "epoch": 2.3627397260273972, + "grad_norm": 2.612849712371826, + "learning_rate": 1.2103980503655566e-06, + "loss": 1.6686, + "step": 2156 + }, + { + "epoch": 2.3638356164383563, + "grad_norm": 2.6836912631988525, + "learning_rate": 1.2083671811535338e-06, + "loss": 1.7173, + "step": 2157 + }, + { + "epoch": 2.364931506849315, + "grad_norm": 2.716177225112915, + "learning_rate": 1.2063363119415111e-06, + "loss": 1.695, + "step": 2158 + }, + { + "epoch": 2.366027397260274, + "grad_norm": 2.6918227672576904, + "learning_rate": 1.2043054427294882e-06, + "loss": 1.684, + "step": 2159 + }, + { + "epoch": 2.367123287671233, + "grad_norm": 2.6265008449554443, + "learning_rate": 1.2022745735174654e-06, + "loss": 1.6959, + "step": 2160 + }, + { + "epoch": 2.367123287671233, + "eval_loss": 1.939719319343567, + "eval_runtime": 37.2557, + "eval_samples_per_second": 26.842, + "eval_steps_per_second": 1.127, + "step": 2160 + }, + { + "epoch": 2.368219178082192, + "grad_norm": 2.638829469680786, + "learning_rate": 1.2002437043054427e-06, + "loss": 1.7362, + "step": 2161 + }, + { + "epoch": 2.3693150684931505, + "grad_norm": 2.6369569301605225, + "learning_rate": 1.19821283509342e-06, + "loss": 1.7765, + "step": 2162 + }, + { + "epoch": 2.3704109589041096, + "grad_norm": 2.6272799968719482, + "learning_rate": 1.1961819658813973e-06, + "loss": 1.7695, + "step": 2163 + }, + { + "epoch": 2.3715068493150686, + "grad_norm": 2.601388931274414, + "learning_rate": 1.1941510966693746e-06, + "loss": 1.7259, + "step": 2164 + }, + { + "epoch": 2.3726027397260276, + "grad_norm": 2.5515146255493164, + "learning_rate": 1.1921202274573518e-06, + "loss": 1.7048, + "step": 2165 + }, + { + "epoch": 2.3736986301369862, + "grad_norm": 2.6139824390411377, + "learning_rate": 1.190089358245329e-06, + "loss": 1.6951, + "step": 2166 + }, + { + "epoch": 2.3747945205479453, + "grad_norm": 2.7428653240203857, + "learning_rate": 1.1880584890333064e-06, + "loss": 1.7771, + "step": 2167 + }, + { + "epoch": 2.3758904109589043, + "grad_norm": 2.690804958343506, + "learning_rate": 1.1860276198212837e-06, + "loss": 1.7073, + "step": 2168 + }, + { + "epoch": 2.376986301369863, + "grad_norm": 2.64790415763855, + "learning_rate": 1.183996750609261e-06, + "loss": 1.7344, + "step": 2169 + }, + { + "epoch": 2.378082191780822, + "grad_norm": 2.555832624435425, + "learning_rate": 1.1819658813972382e-06, + "loss": 1.7452, + "step": 2170 + }, + { + "epoch": 2.378082191780822, + "eval_loss": 1.9401137828826904, + "eval_runtime": 37.2362, + "eval_samples_per_second": 26.856, + "eval_steps_per_second": 1.128, + "step": 2170 + }, + { + "epoch": 2.379178082191781, + "grad_norm": 2.604097366333008, + "learning_rate": 1.1799350121852153e-06, + "loss": 1.7176, + "step": 2171 + }, + { + "epoch": 2.3802739726027395, + "grad_norm": 2.579312324523926, + "learning_rate": 1.1779041429731925e-06, + "loss": 1.7588, + "step": 2172 + }, + { + "epoch": 2.3813698630136986, + "grad_norm": 2.555727481842041, + "learning_rate": 1.1758732737611698e-06, + "loss": 1.7612, + "step": 2173 + }, + { + "epoch": 2.3824657534246576, + "grad_norm": 2.6929352283477783, + "learning_rate": 1.173842404549147e-06, + "loss": 1.6707, + "step": 2174 + }, + { + "epoch": 2.383561643835616, + "grad_norm": 2.6732795238494873, + "learning_rate": 1.1718115353371244e-06, + "loss": 1.7743, + "step": 2175 + }, + { + "epoch": 2.3846575342465752, + "grad_norm": 2.673490047454834, + "learning_rate": 1.1697806661251016e-06, + "loss": 1.6968, + "step": 2176 + }, + { + "epoch": 2.3857534246575343, + "grad_norm": 2.657837152481079, + "learning_rate": 1.167749796913079e-06, + "loss": 1.7633, + "step": 2177 + }, + { + "epoch": 2.3868493150684933, + "grad_norm": 2.6677606105804443, + "learning_rate": 1.1657189277010562e-06, + "loss": 1.7646, + "step": 2178 + }, + { + "epoch": 2.387945205479452, + "grad_norm": 2.63075852394104, + "learning_rate": 1.1636880584890335e-06, + "loss": 1.7019, + "step": 2179 + }, + { + "epoch": 2.389041095890411, + "grad_norm": 2.874946117401123, + "learning_rate": 1.1616571892770108e-06, + "loss": 1.691, + "step": 2180 + }, + { + "epoch": 2.389041095890411, + "eval_loss": 1.9399209022521973, + "eval_runtime": 37.1991, + "eval_samples_per_second": 26.882, + "eval_steps_per_second": 1.129, + "step": 2180 + }, + { + "epoch": 2.39013698630137, + "grad_norm": 2.734825611114502, + "learning_rate": 1.159626320064988e-06, + "loss": 1.7551, + "step": 2181 + }, + { + "epoch": 2.391232876712329, + "grad_norm": 2.5722298622131348, + "learning_rate": 1.1575954508529653e-06, + "loss": 1.7279, + "step": 2182 + }, + { + "epoch": 2.3923287671232876, + "grad_norm": 2.6449334621429443, + "learning_rate": 1.1555645816409426e-06, + "loss": 1.7591, + "step": 2183 + }, + { + "epoch": 2.3934246575342466, + "grad_norm": 2.774374485015869, + "learning_rate": 1.1535337124289196e-06, + "loss": 1.7243, + "step": 2184 + }, + { + "epoch": 2.3945205479452056, + "grad_norm": 2.554065227508545, + "learning_rate": 1.151502843216897e-06, + "loss": 1.6767, + "step": 2185 + }, + { + "epoch": 2.3956164383561642, + "grad_norm": 2.5883984565734863, + "learning_rate": 1.1494719740048742e-06, + "loss": 1.6416, + "step": 2186 + }, + { + "epoch": 2.3967123287671233, + "grad_norm": 2.5799498558044434, + "learning_rate": 1.1474411047928515e-06, + "loss": 1.7162, + "step": 2187 + }, + { + "epoch": 2.3978082191780823, + "grad_norm": 2.64442777633667, + "learning_rate": 1.1454102355808287e-06, + "loss": 1.709, + "step": 2188 + }, + { + "epoch": 2.398904109589041, + "grad_norm": 2.6813411712646484, + "learning_rate": 1.143379366368806e-06, + "loss": 1.6903, + "step": 2189 + }, + { + "epoch": 2.4, + "grad_norm": 2.6276469230651855, + "learning_rate": 1.141348497156783e-06, + "loss": 1.6734, + "step": 2190 + }, + { + "epoch": 2.4, + "eval_loss": 1.9423893690109253, + "eval_runtime": 37.1963, + "eval_samples_per_second": 26.884, + "eval_steps_per_second": 1.129, + "step": 2190 + }, + { + "epoch": 2.401095890410959, + "grad_norm": 2.6805641651153564, + "learning_rate": 1.1393176279447604e-06, + "loss": 1.8038, + "step": 2191 + }, + { + "epoch": 2.402191780821918, + "grad_norm": 2.813957691192627, + "learning_rate": 1.1372867587327376e-06, + "loss": 1.7895, + "step": 2192 + }, + { + "epoch": 2.4032876712328766, + "grad_norm": 2.505695343017578, + "learning_rate": 1.135255889520715e-06, + "loss": 1.6398, + "step": 2193 + }, + { + "epoch": 2.4043835616438356, + "grad_norm": 2.7426211833953857, + "learning_rate": 1.1332250203086922e-06, + "loss": 1.7472, + "step": 2194 + }, + { + "epoch": 2.4054794520547946, + "grad_norm": 2.627737522125244, + "learning_rate": 1.1311941510966695e-06, + "loss": 1.7272, + "step": 2195 + }, + { + "epoch": 2.4065753424657537, + "grad_norm": 2.615032911300659, + "learning_rate": 1.1291632818846467e-06, + "loss": 1.7004, + "step": 2196 + }, + { + "epoch": 2.4076712328767123, + "grad_norm": 2.5147581100463867, + "learning_rate": 1.127132412672624e-06, + "loss": 1.6894, + "step": 2197 + }, + { + "epoch": 2.4087671232876713, + "grad_norm": 2.7482388019561768, + "learning_rate": 1.1251015434606013e-06, + "loss": 1.6738, + "step": 2198 + }, + { + "epoch": 2.4098630136986303, + "grad_norm": 2.7309365272521973, + "learning_rate": 1.1230706742485786e-06, + "loss": 1.7222, + "step": 2199 + }, + { + "epoch": 2.410958904109589, + "grad_norm": 2.8322935104370117, + "learning_rate": 1.1210398050365558e-06, + "loss": 1.7762, + "step": 2200 + }, + { + "epoch": 2.410958904109589, + "eval_loss": 1.9395875930786133, + "eval_runtime": 37.2409, + "eval_samples_per_second": 26.852, + "eval_steps_per_second": 1.128, + "step": 2200 + }, + { + "epoch": 2.412054794520548, + "grad_norm": 2.7829575538635254, + "learning_rate": 1.1190089358245331e-06, + "loss": 1.7092, + "step": 2201 + }, + { + "epoch": 2.413150684931507, + "grad_norm": 2.518369197845459, + "learning_rate": 1.1169780666125102e-06, + "loss": 1.655, + "step": 2202 + }, + { + "epoch": 2.4142465753424656, + "grad_norm": 2.715089797973633, + "learning_rate": 1.1149471974004875e-06, + "loss": 1.7461, + "step": 2203 + }, + { + "epoch": 2.4153424657534246, + "grad_norm": 2.7772066593170166, + "learning_rate": 1.1129163281884647e-06, + "loss": 1.6931, + "step": 2204 + }, + { + "epoch": 2.4164383561643836, + "grad_norm": 2.813166379928589, + "learning_rate": 1.110885458976442e-06, + "loss": 1.7514, + "step": 2205 + }, + { + "epoch": 2.417534246575342, + "grad_norm": 2.6345536708831787, + "learning_rate": 1.1088545897644193e-06, + "loss": 1.6865, + "step": 2206 + }, + { + "epoch": 2.4186301369863012, + "grad_norm": 2.6846344470977783, + "learning_rate": 1.1068237205523966e-06, + "loss": 1.7564, + "step": 2207 + }, + { + "epoch": 2.4197260273972603, + "grad_norm": 2.6230814456939697, + "learning_rate": 1.1047928513403736e-06, + "loss": 1.7499, + "step": 2208 + }, + { + "epoch": 2.4208219178082193, + "grad_norm": 2.6820080280303955, + "learning_rate": 1.102761982128351e-06, + "loss": 1.7494, + "step": 2209 + }, + { + "epoch": 2.421917808219178, + "grad_norm": 2.572378396987915, + "learning_rate": 1.1007311129163282e-06, + "loss": 1.6585, + "step": 2210 + }, + { + "epoch": 2.421917808219178, + "eval_loss": 1.9389609098434448, + "eval_runtime": 37.2665, + "eval_samples_per_second": 26.834, + "eval_steps_per_second": 1.127, + "step": 2210 + }, + { + "epoch": 2.423013698630137, + "grad_norm": 2.7218871116638184, + "learning_rate": 1.0987002437043055e-06, + "loss": 1.7187, + "step": 2211 + }, + { + "epoch": 2.424109589041096, + "grad_norm": 2.7090065479278564, + "learning_rate": 1.0966693744922827e-06, + "loss": 1.6983, + "step": 2212 + }, + { + "epoch": 2.425205479452055, + "grad_norm": 2.6419243812561035, + "learning_rate": 1.09463850528026e-06, + "loss": 1.6485, + "step": 2213 + }, + { + "epoch": 2.4263013698630136, + "grad_norm": 2.635533571243286, + "learning_rate": 1.0926076360682373e-06, + "loss": 1.7023, + "step": 2214 + }, + { + "epoch": 2.4273972602739726, + "grad_norm": 2.644904375076294, + "learning_rate": 1.0905767668562146e-06, + "loss": 1.6774, + "step": 2215 + }, + { + "epoch": 2.4284931506849317, + "grad_norm": 2.6821508407592773, + "learning_rate": 1.0885458976441918e-06, + "loss": 1.6995, + "step": 2216 + }, + { + "epoch": 2.4295890410958902, + "grad_norm": 2.674916982650757, + "learning_rate": 1.0865150284321691e-06, + "loss": 1.7888, + "step": 2217 + }, + { + "epoch": 2.4306849315068493, + "grad_norm": 2.6966474056243896, + "learning_rate": 1.0844841592201464e-06, + "loss": 1.7303, + "step": 2218 + }, + { + "epoch": 2.4317808219178083, + "grad_norm": 2.6421585083007812, + "learning_rate": 1.0824532900081237e-06, + "loss": 1.7497, + "step": 2219 + }, + { + "epoch": 2.432876712328767, + "grad_norm": 2.8147826194763184, + "learning_rate": 1.080422420796101e-06, + "loss": 1.7088, + "step": 2220 + }, + { + "epoch": 2.432876712328767, + "eval_loss": 1.9437190294265747, + "eval_runtime": 37.5268, + "eval_samples_per_second": 26.648, + "eval_steps_per_second": 1.119, + "step": 2220 + }, + { + "epoch": 2.433972602739726, + "grad_norm": 2.6919329166412354, + "learning_rate": 1.078391551584078e-06, + "loss": 1.7168, + "step": 2221 + }, + { + "epoch": 2.435068493150685, + "grad_norm": 2.6179897785186768, + "learning_rate": 1.0763606823720553e-06, + "loss": 1.7477, + "step": 2222 + }, + { + "epoch": 2.436164383561644, + "grad_norm": 2.7324624061584473, + "learning_rate": 1.0743298131600326e-06, + "loss": 1.7921, + "step": 2223 + }, + { + "epoch": 2.4372602739726026, + "grad_norm": 2.6431500911712646, + "learning_rate": 1.0722989439480098e-06, + "loss": 1.7432, + "step": 2224 + }, + { + "epoch": 2.4383561643835616, + "grad_norm": 2.593528985977173, + "learning_rate": 1.070268074735987e-06, + "loss": 1.7097, + "step": 2225 + }, + { + "epoch": 2.4394520547945207, + "grad_norm": 2.6277310848236084, + "learning_rate": 1.0682372055239644e-06, + "loss": 1.6911, + "step": 2226 + }, + { + "epoch": 2.4405479452054797, + "grad_norm": 2.5935580730438232, + "learning_rate": 1.0662063363119414e-06, + "loss": 1.6933, + "step": 2227 + }, + { + "epoch": 2.4416438356164383, + "grad_norm": 2.672347068786621, + "learning_rate": 1.0641754670999187e-06, + "loss": 1.6826, + "step": 2228 + }, + { + "epoch": 2.4427397260273973, + "grad_norm": 2.7214746475219727, + "learning_rate": 1.062144597887896e-06, + "loss": 1.7152, + "step": 2229 + }, + { + "epoch": 2.4438356164383563, + "grad_norm": 2.7052741050720215, + "learning_rate": 1.0601137286758733e-06, + "loss": 1.7119, + "step": 2230 + }, + { + "epoch": 2.4438356164383563, + "eval_loss": 1.9452862739562988, + "eval_runtime": 37.6622, + "eval_samples_per_second": 26.552, + "eval_steps_per_second": 1.115, + "step": 2230 + }, + { + "epoch": 2.444931506849315, + "grad_norm": 2.7374205589294434, + "learning_rate": 1.0580828594638505e-06, + "loss": 1.7904, + "step": 2231 + }, + { + "epoch": 2.446027397260274, + "grad_norm": 2.6822078227996826, + "learning_rate": 1.0560519902518278e-06, + "loss": 1.6238, + "step": 2232 + }, + { + "epoch": 2.447123287671233, + "grad_norm": 2.6262941360473633, + "learning_rate": 1.054021121039805e-06, + "loss": 1.7501, + "step": 2233 + }, + { + "epoch": 2.4482191780821916, + "grad_norm": 2.689777374267578, + "learning_rate": 1.0519902518277824e-06, + "loss": 1.6604, + "step": 2234 + }, + { + "epoch": 2.4493150684931506, + "grad_norm": 2.6919682025909424, + "learning_rate": 1.0499593826157596e-06, + "loss": 1.6493, + "step": 2235 + }, + { + "epoch": 2.4504109589041096, + "grad_norm": 2.732959270477295, + "learning_rate": 1.047928513403737e-06, + "loss": 1.7429, + "step": 2236 + }, + { + "epoch": 2.4515068493150687, + "grad_norm": 2.6201255321502686, + "learning_rate": 1.0458976441917142e-06, + "loss": 1.7131, + "step": 2237 + }, + { + "epoch": 2.4526027397260273, + "grad_norm": 2.655271291732788, + "learning_rate": 1.0438667749796915e-06, + "loss": 1.634, + "step": 2238 + }, + { + "epoch": 2.4536986301369863, + "grad_norm": 2.6041507720947266, + "learning_rate": 1.0418359057676688e-06, + "loss": 1.6384, + "step": 2239 + }, + { + "epoch": 2.4547945205479453, + "grad_norm": 2.582871198654175, + "learning_rate": 1.0398050365556458e-06, + "loss": 1.6903, + "step": 2240 + }, + { + "epoch": 2.4547945205479453, + "eval_loss": 1.9449145793914795, + "eval_runtime": 37.4116, + "eval_samples_per_second": 26.73, + "eval_steps_per_second": 1.123, + "step": 2240 + }, + { + "epoch": 2.455890410958904, + "grad_norm": 2.647946357727051, + "learning_rate": 1.037774167343623e-06, + "loss": 1.752, + "step": 2241 + }, + { + "epoch": 2.456986301369863, + "grad_norm": 2.6174609661102295, + "learning_rate": 1.0357432981316004e-06, + "loss": 1.6745, + "step": 2242 + }, + { + "epoch": 2.458082191780822, + "grad_norm": 2.7191693782806396, + "learning_rate": 1.0337124289195776e-06, + "loss": 1.7291, + "step": 2243 + }, + { + "epoch": 2.459178082191781, + "grad_norm": 2.6612565517425537, + "learning_rate": 1.031681559707555e-06, + "loss": 1.7523, + "step": 2244 + }, + { + "epoch": 2.4602739726027396, + "grad_norm": 2.617875337600708, + "learning_rate": 1.0296506904955322e-06, + "loss": 1.7597, + "step": 2245 + }, + { + "epoch": 2.4613698630136986, + "grad_norm": 2.730677843093872, + "learning_rate": 1.0276198212835095e-06, + "loss": 1.7146, + "step": 2246 + }, + { + "epoch": 2.4624657534246577, + "grad_norm": 2.6297593116760254, + "learning_rate": 1.0255889520714867e-06, + "loss": 1.7183, + "step": 2247 + }, + { + "epoch": 2.4635616438356163, + "grad_norm": 2.600847005844116, + "learning_rate": 1.023558082859464e-06, + "loss": 1.7086, + "step": 2248 + }, + { + "epoch": 2.4646575342465753, + "grad_norm": 2.773501396179199, + "learning_rate": 1.0215272136474413e-06, + "loss": 1.6708, + "step": 2249 + }, + { + "epoch": 2.4657534246575343, + "grad_norm": 2.6344449520111084, + "learning_rate": 1.0194963444354186e-06, + "loss": 1.7818, + "step": 2250 + }, + { + "epoch": 2.4657534246575343, + "eval_loss": 1.940114974975586, + "eval_runtime": 37.3629, + "eval_samples_per_second": 26.764, + "eval_steps_per_second": 1.124, + "step": 2250 + }, + { + "epoch": 2.466849315068493, + "grad_norm": 2.6345012187957764, + "learning_rate": 1.0174654752233958e-06, + "loss": 1.6525, + "step": 2251 + }, + { + "epoch": 2.467945205479452, + "grad_norm": 2.6919827461242676, + "learning_rate": 1.015434606011373e-06, + "loss": 1.7457, + "step": 2252 + }, + { + "epoch": 2.469041095890411, + "grad_norm": 2.6676719188690186, + "learning_rate": 1.0134037367993502e-06, + "loss": 1.6934, + "step": 2253 + }, + { + "epoch": 2.47013698630137, + "grad_norm": 2.623911142349243, + "learning_rate": 1.0113728675873275e-06, + "loss": 1.7089, + "step": 2254 + }, + { + "epoch": 2.4712328767123286, + "grad_norm": 2.6342320442199707, + "learning_rate": 1.0093419983753047e-06, + "loss": 1.7123, + "step": 2255 + }, + { + "epoch": 2.4723287671232876, + "grad_norm": 2.8066930770874023, + "learning_rate": 1.007311129163282e-06, + "loss": 1.7095, + "step": 2256 + }, + { + "epoch": 2.4734246575342467, + "grad_norm": 2.6137008666992188, + "learning_rate": 1.0052802599512593e-06, + "loss": 1.6602, + "step": 2257 + }, + { + "epoch": 2.4745205479452057, + "grad_norm": 2.6158881187438965, + "learning_rate": 1.0032493907392364e-06, + "loss": 1.6659, + "step": 2258 + }, + { + "epoch": 2.4756164383561643, + "grad_norm": 2.7409005165100098, + "learning_rate": 1.0012185215272136e-06, + "loss": 1.7194, + "step": 2259 + }, + { + "epoch": 2.4767123287671233, + "grad_norm": 2.783179759979248, + "learning_rate": 9.99187652315191e-07, + "loss": 1.658, + "step": 2260 + }, + { + "epoch": 2.4767123287671233, + "eval_loss": 1.9401936531066895, + "eval_runtime": 37.2693, + "eval_samples_per_second": 26.832, + "eval_steps_per_second": 1.127, + "step": 2260 + }, + { + "epoch": 2.4778082191780824, + "grad_norm": 2.7862110137939453, + "learning_rate": 9.971567831031682e-07, + "loss": 1.7535, + "step": 2261 + }, + { + "epoch": 2.478904109589041, + "grad_norm": 2.623033046722412, + "learning_rate": 9.951259138911455e-07, + "loss": 1.7285, + "step": 2262 + }, + { + "epoch": 2.48, + "grad_norm": 2.5325446128845215, + "learning_rate": 9.930950446791227e-07, + "loss": 1.6486, + "step": 2263 + }, + { + "epoch": 2.481095890410959, + "grad_norm": 2.6013123989105225, + "learning_rate": 9.910641754671e-07, + "loss": 1.7677, + "step": 2264 + }, + { + "epoch": 2.4821917808219176, + "grad_norm": 2.706526279449463, + "learning_rate": 9.890333062550773e-07, + "loss": 1.81, + "step": 2265 + }, + { + "epoch": 2.4832876712328766, + "grad_norm": 2.6389000415802, + "learning_rate": 9.870024370430546e-07, + "loss": 1.7104, + "step": 2266 + }, + { + "epoch": 2.4843835616438357, + "grad_norm": 2.8673291206359863, + "learning_rate": 9.849715678310318e-07, + "loss": 1.7439, + "step": 2267 + }, + { + "epoch": 2.4854794520547947, + "grad_norm": 2.574463367462158, + "learning_rate": 9.829406986190091e-07, + "loss": 1.7134, + "step": 2268 + }, + { + "epoch": 2.4865753424657533, + "grad_norm": 2.6605708599090576, + "learning_rate": 9.809098294069864e-07, + "loss": 1.7266, + "step": 2269 + }, + { + "epoch": 2.4876712328767123, + "grad_norm": 2.6304521560668945, + "learning_rate": 9.788789601949637e-07, + "loss": 1.6388, + "step": 2270 + }, + { + "epoch": 2.4876712328767123, + "eval_loss": 1.9415092468261719, + "eval_runtime": 37.2274, + "eval_samples_per_second": 26.862, + "eval_steps_per_second": 1.128, + "step": 2270 + }, + { + "epoch": 2.4887671232876714, + "grad_norm": 2.6577560901641846, + "learning_rate": 9.768480909829407e-07, + "loss": 1.7252, + "step": 2271 + }, + { + "epoch": 2.48986301369863, + "grad_norm": 2.801530122756958, + "learning_rate": 9.74817221770918e-07, + "loss": 1.7774, + "step": 2272 + }, + { + "epoch": 2.490958904109589, + "grad_norm": 2.761235237121582, + "learning_rate": 9.727863525588953e-07, + "loss": 1.6889, + "step": 2273 + }, + { + "epoch": 2.492054794520548, + "grad_norm": 2.719083547592163, + "learning_rate": 9.707554833468726e-07, + "loss": 1.7231, + "step": 2274 + }, + { + "epoch": 2.493150684931507, + "grad_norm": 2.5667307376861572, + "learning_rate": 9.687246141348498e-07, + "loss": 1.7301, + "step": 2275 + }, + { + "epoch": 2.4942465753424656, + "grad_norm": 2.6383678913116455, + "learning_rate": 9.666937449228271e-07, + "loss": 1.6999, + "step": 2276 + }, + { + "epoch": 2.4953424657534247, + "grad_norm": 2.7380318641662598, + "learning_rate": 9.646628757108042e-07, + "loss": 1.7202, + "step": 2277 + }, + { + "epoch": 2.4964383561643837, + "grad_norm": 2.715460777282715, + "learning_rate": 9.626320064987814e-07, + "loss": 1.6626, + "step": 2278 + }, + { + "epoch": 2.4975342465753423, + "grad_norm": 2.5860559940338135, + "learning_rate": 9.606011372867587e-07, + "loss": 1.7345, + "step": 2279 + }, + { + "epoch": 2.4986301369863013, + "grad_norm": 2.614272356033325, + "learning_rate": 9.58570268074736e-07, + "loss": 1.6728, + "step": 2280 + }, + { + "epoch": 2.4986301369863013, + "eval_loss": 1.9393996000289917, + "eval_runtime": 37.2508, + "eval_samples_per_second": 26.845, + "eval_steps_per_second": 1.127, + "step": 2280 + }, + { + "epoch": 2.4997260273972604, + "grad_norm": 2.6709139347076416, + "learning_rate": 9.565393988627133e-07, + "loss": 1.7094, + "step": 2281 + }, + { + "epoch": 2.500821917808219, + "grad_norm": 2.468263626098633, + "learning_rate": 9.545085296506906e-07, + "loss": 1.7044, + "step": 2282 + }, + { + "epoch": 2.501917808219178, + "grad_norm": 2.646242141723633, + "learning_rate": 9.524776604386677e-07, + "loss": 1.7144, + "step": 2283 + }, + { + "epoch": 2.503013698630137, + "grad_norm": 2.6524763107299805, + "learning_rate": 9.50446791226645e-07, + "loss": 1.7221, + "step": 2284 + }, + { + "epoch": 2.504109589041096, + "grad_norm": 2.681304693222046, + "learning_rate": 9.484159220146223e-07, + "loss": 1.6446, + "step": 2285 + }, + { + "epoch": 2.5052054794520546, + "grad_norm": 2.5855886936187744, + "learning_rate": 9.463850528025995e-07, + "loss": 1.757, + "step": 2286 + }, + { + "epoch": 2.5063013698630137, + "grad_norm": 2.5410566329956055, + "learning_rate": 9.443541835905768e-07, + "loss": 1.7487, + "step": 2287 + }, + { + "epoch": 2.5073972602739727, + "grad_norm": 2.726349115371704, + "learning_rate": 9.423233143785541e-07, + "loss": 1.7841, + "step": 2288 + }, + { + "epoch": 2.5084931506849317, + "grad_norm": 2.7225544452667236, + "learning_rate": 9.402924451665313e-07, + "loss": 1.7921, + "step": 2289 + }, + { + "epoch": 2.5095890410958903, + "grad_norm": 2.7081549167633057, + "learning_rate": 9.382615759545085e-07, + "loss": 1.6726, + "step": 2290 + }, + { + "epoch": 2.5095890410958903, + "eval_loss": 1.9398577213287354, + "eval_runtime": 37.2122, + "eval_samples_per_second": 26.873, + "eval_steps_per_second": 1.129, + "step": 2290 + }, + { + "epoch": 2.5106849315068493, + "grad_norm": 2.6577272415161133, + "learning_rate": 9.362307067424858e-07, + "loss": 1.7571, + "step": 2291 + }, + { + "epoch": 2.5117808219178084, + "grad_norm": 2.659438133239746, + "learning_rate": 9.341998375304631e-07, + "loss": 1.7113, + "step": 2292 + }, + { + "epoch": 2.512876712328767, + "grad_norm": 2.6679470539093018, + "learning_rate": 9.321689683184404e-07, + "loss": 1.7718, + "step": 2293 + }, + { + "epoch": 2.513972602739726, + "grad_norm": 2.703716993331909, + "learning_rate": 9.301380991064176e-07, + "loss": 1.6617, + "step": 2294 + }, + { + "epoch": 2.515068493150685, + "grad_norm": 2.66267728805542, + "learning_rate": 9.281072298943949e-07, + "loss": 1.6479, + "step": 2295 + }, + { + "epoch": 2.5161643835616436, + "grad_norm": 2.7160699367523193, + "learning_rate": 9.260763606823721e-07, + "loss": 1.6931, + "step": 2296 + }, + { + "epoch": 2.5172602739726027, + "grad_norm": 2.7481086254119873, + "learning_rate": 9.240454914703494e-07, + "loss": 1.73, + "step": 2297 + }, + { + "epoch": 2.5183561643835617, + "grad_norm": 2.702737331390381, + "learning_rate": 9.220146222583266e-07, + "loss": 1.7414, + "step": 2298 + }, + { + "epoch": 2.5194520547945203, + "grad_norm": 2.78155517578125, + "learning_rate": 9.199837530463039e-07, + "loss": 1.684, + "step": 2299 + }, + { + "epoch": 2.5205479452054793, + "grad_norm": 2.5439021587371826, + "learning_rate": 9.179528838342812e-07, + "loss": 1.7116, + "step": 2300 + }, + { + "epoch": 2.5205479452054793, + "eval_loss": 1.9369380474090576, + "eval_runtime": 37.2139, + "eval_samples_per_second": 26.872, + "eval_steps_per_second": 1.129, + "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/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-2300/zero_to_fp32.py b/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-2300/zero_to_fp32.py new file mode 100644 index 0000000000000000000000000000000000000000..24cc342e78d1a006c782b3a4cd68d9ce786d8fd8 --- /dev/null +++ b/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_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/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-2400/config.json b/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-2400/config.json new file mode 100644 index 0000000000000000000000000000000000000000..fb0f9a53a507ffcdb60410deb71d60ef801bf350 --- /dev/null +++ b/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-2400/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/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-2400/generation_config.json b/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-2400/generation_config.json new file mode 100644 index 0000000000000000000000000000000000000000..9c389b841a9a29ddd904e02d1eb0e08dcce82ad9 --- /dev/null +++ b/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-2400/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/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-2400/latest b/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-2400/latest new file mode 100644 index 0000000000000000000000000000000000000000..361f8a1a19370f3a8801a82732f02b0e9e639e14 --- /dev/null +++ b/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-2400/latest @@ -0,0 +1 @@ +global_step2400 \ No newline at end of file diff --git a/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-2400/model.safetensors.index.json b/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-2400/model.safetensors.index.json new file mode 100644 index 0000000000000000000000000000000000000000..ed64de846d720b9a7859dc20575fea8e8ca51940 --- /dev/null +++ b/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-2400/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/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-2400/special_tokens_map.json b/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-2400/special_tokens_map.json new file mode 100644 index 0000000000000000000000000000000000000000..6949c5975ee0e961ef61cf31010dce04df0a03f8 --- /dev/null +++ b/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-2400/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/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-2400/tokenizer_config.json b/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-2400/tokenizer_config.json new file mode 100644 index 0000000000000000000000000000000000000000..5684b14e2756b56aa397c1324e239b442162b34a --- /dev/null +++ b/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-2400/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/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-2400/trainer_state.json b/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-2400/trainer_state.json new file mode 100644 index 0000000000000000000000000000000000000000..3bfb690191194b8e576c5705ad3f5a19e0320e46 --- /dev/null +++ b/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-2400/trainer_state.json @@ -0,0 +1,18753 @@ +{ + "best_metric": null, + "best_model_checkpoint": null, + "epoch": 2.6301369863013697, + "eval_steps": 10, + "global_step": 2400, + "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": 3.0476, + "step": 1 + }, + { + "epoch": 0.002191780821917808, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.0289, + "step": 2 + }, + { + "epoch": 0.0032876712328767125, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 2.9935, + "step": 3 + }, + { + "epoch": 0.004383561643835616, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.0139, + "step": 4 + }, + { + "epoch": 0.005479452054794521, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 2.9892, + "step": 5 + }, + { + "epoch": 0.006575342465753425, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.1252, + "step": 6 + }, + { + "epoch": 0.007671232876712329, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.0291, + "step": 7 + }, + { + "epoch": 0.008767123287671232, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.0607, + "step": 8 + }, + { + "epoch": 0.009863013698630137, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.0306, + "step": 9 + }, + { + "epoch": 0.010958904109589041, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.0811, + "step": 10 + }, + { + "epoch": 0.010958904109589041, + "eval_loss": 3.0768930912017822, + "eval_runtime": 38.0377, + "eval_samples_per_second": 26.29, + "eval_steps_per_second": 1.104, + "step": 10 + }, + { + "epoch": 0.012054794520547946, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.0693, + "step": 11 + }, + { + "epoch": 0.01315068493150685, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 2.9668, + "step": 12 + }, + { + "epoch": 0.014246575342465753, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.052, + "step": 13 + }, + { + "epoch": 0.015342465753424657, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.0363, + "step": 14 + }, + { + "epoch": 0.01643835616438356, + "grad_norm": 5.58175802230835, + "learning_rate": 1.8248175182481753e-08, + "loss": 3.0778, + "step": 15 + }, + { + "epoch": 0.017534246575342465, + "grad_norm": 6.360531806945801, + "learning_rate": 3.6496350364963505e-08, + "loss": 3.0257, + "step": 16 + }, + { + "epoch": 0.01863013698630137, + "grad_norm": 6.453113555908203, + "learning_rate": 5.4744525547445264e-08, + "loss": 3.0262, + "step": 17 + }, + { + "epoch": 0.019726027397260273, + "grad_norm": 6.789405345916748, + "learning_rate": 7.299270072992701e-08, + "loss": 3.0671, + "step": 18 + }, + { + "epoch": 0.020821917808219178, + "grad_norm": 6.185514450073242, + "learning_rate": 9.124087591240876e-08, + "loss": 3.0077, + "step": 19 + }, + { + "epoch": 0.021917808219178082, + "grad_norm": 5.719798564910889, + "learning_rate": 1.0948905109489053e-07, + "loss": 3.089, + "step": 20 + }, + { + "epoch": 0.021917808219178082, + "eval_loss": 3.07546329498291, + "eval_runtime": 37.5318, + "eval_samples_per_second": 26.644, + "eval_steps_per_second": 1.119, + "step": 20 + }, + { + "epoch": 0.023013698630136987, + "grad_norm": 6.307768821716309, + "learning_rate": 1.2773722627737227e-07, + "loss": 3.0641, + "step": 21 + }, + { + "epoch": 0.02410958904109589, + "grad_norm": 6.255112171173096, + "learning_rate": 1.4598540145985402e-07, + "loss": 2.9849, + "step": 22 + }, + { + "epoch": 0.025205479452054796, + "grad_norm": 5.812101364135742, + "learning_rate": 1.642335766423358e-07, + "loss": 3.0211, + "step": 23 + }, + { + "epoch": 0.0263013698630137, + "grad_norm": 6.262075901031494, + "learning_rate": 1.8248175182481753e-07, + "loss": 3.0654, + "step": 24 + }, + { + "epoch": 0.0273972602739726, + "grad_norm": 5.419229507446289, + "learning_rate": 2.0072992700729928e-07, + "loss": 3.0661, + "step": 25 + }, + { + "epoch": 0.028493150684931506, + "grad_norm": 4.750737190246582, + "learning_rate": 2.1897810218978106e-07, + "loss": 3.0494, + "step": 26 + }, + { + "epoch": 0.02958904109589041, + "grad_norm": 4.732361316680908, + "learning_rate": 2.3722627737226278e-07, + "loss": 3.0471, + "step": 27 + }, + { + "epoch": 0.030684931506849315, + "grad_norm": 4.749786853790283, + "learning_rate": 2.5547445255474454e-07, + "loss": 3.0107, + "step": 28 + }, + { + "epoch": 0.03178082191780822, + "grad_norm": 4.451244831085205, + "learning_rate": 2.737226277372263e-07, + "loss": 2.9987, + "step": 29 + }, + { + "epoch": 0.03287671232876712, + "grad_norm": 4.5158257484436035, + "learning_rate": 2.9197080291970804e-07, + "loss": 2.9686, + "step": 30 + }, + { + "epoch": 0.03287671232876712, + "eval_loss": 3.031872034072876, + "eval_runtime": 37.2608, + "eval_samples_per_second": 26.838, + "eval_steps_per_second": 1.127, + "step": 30 + }, + { + "epoch": 0.03397260273972603, + "grad_norm": 3.8239498138427734, + "learning_rate": 3.102189781021898e-07, + "loss": 3.0057, + "step": 31 + }, + { + "epoch": 0.03506849315068493, + "grad_norm": 4.037254810333252, + "learning_rate": 3.284671532846716e-07, + "loss": 2.9733, + "step": 32 + }, + { + "epoch": 0.03616438356164384, + "grad_norm": 4.283931732177734, + "learning_rate": 3.467153284671533e-07, + "loss": 2.9515, + "step": 33 + }, + { + "epoch": 0.03726027397260274, + "grad_norm": 4.1784987449646, + "learning_rate": 3.6496350364963505e-07, + "loss": 2.9483, + "step": 34 + }, + { + "epoch": 0.038356164383561646, + "grad_norm": 3.9210753440856934, + "learning_rate": 3.8321167883211686e-07, + "loss": 2.9679, + "step": 35 + }, + { + "epoch": 0.03945205479452055, + "grad_norm": 3.72884202003479, + "learning_rate": 4.0145985401459856e-07, + "loss": 2.9989, + "step": 36 + }, + { + "epoch": 0.040547945205479455, + "grad_norm": 3.2882330417633057, + "learning_rate": 4.197080291970803e-07, + "loss": 2.9384, + "step": 37 + }, + { + "epoch": 0.041643835616438356, + "grad_norm": 3.515293836593628, + "learning_rate": 4.379562043795621e-07, + "loss": 2.8986, + "step": 38 + }, + { + "epoch": 0.042739726027397264, + "grad_norm": 3.568453073501587, + "learning_rate": 4.562043795620438e-07, + "loss": 2.9854, + "step": 39 + }, + { + "epoch": 0.043835616438356165, + "grad_norm": 3.389789342880249, + "learning_rate": 4.7445255474452557e-07, + "loss": 2.8891, + "step": 40 + }, + { + "epoch": 0.043835616438356165, + "eval_loss": 2.964235544204712, + "eval_runtime": 37.3406, + "eval_samples_per_second": 26.781, + "eval_steps_per_second": 1.125, + "step": 40 + }, + { + "epoch": 0.044931506849315066, + "grad_norm": 3.1826283931732178, + "learning_rate": 4.927007299270073e-07, + "loss": 2.9271, + "step": 41 + }, + { + "epoch": 0.046027397260273974, + "grad_norm": 2.941539764404297, + "learning_rate": 5.109489051094891e-07, + "loss": 2.9486, + "step": 42 + }, + { + "epoch": 0.047123287671232875, + "grad_norm": 3.0076773166656494, + "learning_rate": 5.291970802919708e-07, + "loss": 2.9252, + "step": 43 + }, + { + "epoch": 0.04821917808219178, + "grad_norm": 2.79028582572937, + "learning_rate": 5.474452554744526e-07, + "loss": 2.9277, + "step": 44 + }, + { + "epoch": 0.049315068493150684, + "grad_norm": 2.603456735610962, + "learning_rate": 5.656934306569343e-07, + "loss": 2.9144, + "step": 45 + }, + { + "epoch": 0.05041095890410959, + "grad_norm": 2.674888849258423, + "learning_rate": 5.839416058394161e-07, + "loss": 2.8732, + "step": 46 + }, + { + "epoch": 0.05150684931506849, + "grad_norm": 2.972093343734741, + "learning_rate": 6.021897810218979e-07, + "loss": 2.8623, + "step": 47 + }, + { + "epoch": 0.0526027397260274, + "grad_norm": 2.757766008377075, + "learning_rate": 6.204379562043796e-07, + "loss": 2.8941, + "step": 48 + }, + { + "epoch": 0.0536986301369863, + "grad_norm": 2.910634994506836, + "learning_rate": 6.386861313868614e-07, + "loss": 2.8101, + "step": 49 + }, + { + "epoch": 0.0547945205479452, + "grad_norm": 2.5542638301849365, + "learning_rate": 6.569343065693432e-07, + "loss": 2.9262, + "step": 50 + }, + { + "epoch": 0.0547945205479452, + "eval_loss": 2.883579969406128, + "eval_runtime": 37.326, + "eval_samples_per_second": 26.791, + "eval_steps_per_second": 1.125, + "step": 50 + }, + { + "epoch": 0.05589041095890411, + "grad_norm": 2.333974599838257, + "learning_rate": 6.751824817518248e-07, + "loss": 2.8217, + "step": 51 + }, + { + "epoch": 0.05698630136986301, + "grad_norm": 2.181022882461548, + "learning_rate": 6.934306569343066e-07, + "loss": 2.8483, + "step": 52 + }, + { + "epoch": 0.05808219178082192, + "grad_norm": 2.12200927734375, + "learning_rate": 7.116788321167883e-07, + "loss": 2.7788, + "step": 53 + }, + { + "epoch": 0.05917808219178082, + "grad_norm": 2.052626371383667, + "learning_rate": 7.299270072992701e-07, + "loss": 2.7774, + "step": 54 + }, + { + "epoch": 0.06027397260273973, + "grad_norm": 2.11487078666687, + "learning_rate": 7.48175182481752e-07, + "loss": 2.8045, + "step": 55 + }, + { + "epoch": 0.06136986301369863, + "grad_norm": 2.083588123321533, + "learning_rate": 7.664233576642337e-07, + "loss": 2.8119, + "step": 56 + }, + { + "epoch": 0.06246575342465754, + "grad_norm": 2.2079808712005615, + "learning_rate": 7.846715328467154e-07, + "loss": 2.82, + "step": 57 + }, + { + "epoch": 0.06356164383561644, + "grad_norm": 2.100802421569824, + "learning_rate": 8.029197080291971e-07, + "loss": 2.794, + "step": 58 + }, + { + "epoch": 0.06465753424657535, + "grad_norm": 1.8322066068649292, + "learning_rate": 8.211678832116789e-07, + "loss": 2.7444, + "step": 59 + }, + { + "epoch": 0.06575342465753424, + "grad_norm": 2.077005386352539, + "learning_rate": 8.394160583941606e-07, + "loss": 2.7093, + "step": 60 + }, + { + "epoch": 0.06575342465753424, + "eval_loss": 2.8177874088287354, + "eval_runtime": 37.2903, + "eval_samples_per_second": 26.817, + "eval_steps_per_second": 1.126, + "step": 60 + }, + { + "epoch": 0.06684931506849315, + "grad_norm": 2.1688592433929443, + "learning_rate": 8.576642335766425e-07, + "loss": 2.7446, + "step": 61 + }, + { + "epoch": 0.06794520547945206, + "grad_norm": 2.1709237098693848, + "learning_rate": 8.759124087591242e-07, + "loss": 2.736, + "step": 62 + }, + { + "epoch": 0.06904109589041096, + "grad_norm": 2.3578364849090576, + "learning_rate": 8.94160583941606e-07, + "loss": 2.7583, + "step": 63 + }, + { + "epoch": 0.07013698630136986, + "grad_norm": 1.9025483131408691, + "learning_rate": 9.124087591240876e-07, + "loss": 2.7651, + "step": 64 + }, + { + "epoch": 0.07123287671232877, + "grad_norm": 1.9450963735580444, + "learning_rate": 9.306569343065694e-07, + "loss": 2.7186, + "step": 65 + }, + { + "epoch": 0.07232876712328767, + "grad_norm": 1.957723617553711, + "learning_rate": 9.489051094890511e-07, + "loss": 2.7421, + "step": 66 + }, + { + "epoch": 0.07342465753424658, + "grad_norm": 1.9823631048202515, + "learning_rate": 9.67153284671533e-07, + "loss": 2.7258, + "step": 67 + }, + { + "epoch": 0.07452054794520548, + "grad_norm": 1.945495843887329, + "learning_rate": 9.854014598540146e-07, + "loss": 2.7733, + "step": 68 + }, + { + "epoch": 0.07561643835616438, + "grad_norm": 1.9684759378433228, + "learning_rate": 1.0036496350364965e-06, + "loss": 2.7041, + "step": 69 + }, + { + "epoch": 0.07671232876712329, + "grad_norm": 2.087336540222168, + "learning_rate": 1.0218978102189781e-06, + "loss": 2.6662, + "step": 70 + }, + { + "epoch": 0.07671232876712329, + "eval_loss": 2.7529091835021973, + "eval_runtime": 37.3122, + "eval_samples_per_second": 26.801, + "eval_steps_per_second": 1.126, + "step": 70 + }, + { + "epoch": 0.07780821917808219, + "grad_norm": 1.7790117263793945, + "learning_rate": 1.04014598540146e-06, + "loss": 2.7379, + "step": 71 + }, + { + "epoch": 0.0789041095890411, + "grad_norm": 2.2780861854553223, + "learning_rate": 1.0583941605839416e-06, + "loss": 2.6234, + "step": 72 + }, + { + "epoch": 0.08, + "grad_norm": 2.1144700050354004, + "learning_rate": 1.0766423357664233e-06, + "loss": 2.6248, + "step": 73 + }, + { + "epoch": 0.08109589041095891, + "grad_norm": 2.0247490406036377, + "learning_rate": 1.0948905109489052e-06, + "loss": 2.6456, + "step": 74 + }, + { + "epoch": 0.0821917808219178, + "grad_norm": 2.702547311782837, + "learning_rate": 1.113138686131387e-06, + "loss": 2.6759, + "step": 75 + }, + { + "epoch": 0.08328767123287671, + "grad_norm": 1.8208750486373901, + "learning_rate": 1.1313868613138687e-06, + "loss": 2.6845, + "step": 76 + }, + { + "epoch": 0.08438356164383562, + "grad_norm": 2.3829221725463867, + "learning_rate": 1.1496350364963505e-06, + "loss": 2.61, + "step": 77 + }, + { + "epoch": 0.08547945205479453, + "grad_norm": 2.580885887145996, + "learning_rate": 1.1678832116788322e-06, + "loss": 2.5896, + "step": 78 + }, + { + "epoch": 0.08657534246575342, + "grad_norm": 2.2477641105651855, + "learning_rate": 1.1861313868613138e-06, + "loss": 2.596, + "step": 79 + }, + { + "epoch": 0.08767123287671233, + "grad_norm": 2.1895623207092285, + "learning_rate": 1.2043795620437959e-06, + "loss": 2.5906, + "step": 80 + }, + { + "epoch": 0.08767123287671233, + "eval_loss": 2.6815176010131836, + "eval_runtime": 37.3342, + "eval_samples_per_second": 26.785, + "eval_steps_per_second": 1.125, + "step": 80 + }, + { + "epoch": 0.08876712328767124, + "grad_norm": 2.217090606689453, + "learning_rate": 1.2226277372262775e-06, + "loss": 2.6005, + "step": 81 + }, + { + "epoch": 0.08986301369863013, + "grad_norm": 2.4679362773895264, + "learning_rate": 1.2408759124087592e-06, + "loss": 2.6333, + "step": 82 + }, + { + "epoch": 0.09095890410958904, + "grad_norm": 2.2590575218200684, + "learning_rate": 1.259124087591241e-06, + "loss": 2.5814, + "step": 83 + }, + { + "epoch": 0.09205479452054795, + "grad_norm": 1.877977728843689, + "learning_rate": 1.2773722627737229e-06, + "loss": 2.5948, + "step": 84 + }, + { + "epoch": 0.09315068493150686, + "grad_norm": 2.1535534858703613, + "learning_rate": 1.2956204379562045e-06, + "loss": 2.5781, + "step": 85 + }, + { + "epoch": 0.09424657534246575, + "grad_norm": 2.5001144409179688, + "learning_rate": 1.3138686131386864e-06, + "loss": 2.5339, + "step": 86 + }, + { + "epoch": 0.09534246575342466, + "grad_norm": 2.3279049396514893, + "learning_rate": 1.332116788321168e-06, + "loss": 2.545, + "step": 87 + }, + { + "epoch": 0.09643835616438357, + "grad_norm": 2.2230734825134277, + "learning_rate": 1.3503649635036497e-06, + "loss": 2.4745, + "step": 88 + }, + { + "epoch": 0.09753424657534246, + "grad_norm": 2.8762636184692383, + "learning_rate": 1.3686131386861315e-06, + "loss": 2.516, + "step": 89 + }, + { + "epoch": 0.09863013698630137, + "grad_norm": 2.153506278991699, + "learning_rate": 1.3868613138686132e-06, + "loss": 2.4547, + "step": 90 + }, + { + "epoch": 0.09863013698630137, + "eval_loss": 2.603238344192505, + "eval_runtime": 37.4162, + "eval_samples_per_second": 26.726, + "eval_steps_per_second": 1.123, + "step": 90 + }, + { + "epoch": 0.09972602739726028, + "grad_norm": 2.4143614768981934, + "learning_rate": 1.405109489051095e-06, + "loss": 2.547, + "step": 91 + }, + { + "epoch": 0.10082191780821918, + "grad_norm": 2.354632616043091, + "learning_rate": 1.4233576642335767e-06, + "loss": 2.5009, + "step": 92 + }, + { + "epoch": 0.10191780821917808, + "grad_norm": 2.6605563163757324, + "learning_rate": 1.4416058394160583e-06, + "loss": 2.5304, + "step": 93 + }, + { + "epoch": 0.10301369863013699, + "grad_norm": 2.3724122047424316, + "learning_rate": 1.4598540145985402e-06, + "loss": 2.4925, + "step": 94 + }, + { + "epoch": 0.10410958904109589, + "grad_norm": 3.6323022842407227, + "learning_rate": 1.478102189781022e-06, + "loss": 2.4592, + "step": 95 + }, + { + "epoch": 0.1052054794520548, + "grad_norm": 2.3407089710235596, + "learning_rate": 1.496350364963504e-06, + "loss": 2.5223, + "step": 96 + }, + { + "epoch": 0.1063013698630137, + "grad_norm": 2.8238413333892822, + "learning_rate": 1.5145985401459856e-06, + "loss": 2.489, + "step": 97 + }, + { + "epoch": 0.1073972602739726, + "grad_norm": 2.6433658599853516, + "learning_rate": 1.5328467153284674e-06, + "loss": 2.5101, + "step": 98 + }, + { + "epoch": 0.10849315068493151, + "grad_norm": 2.562761068344116, + "learning_rate": 1.551094890510949e-06, + "loss": 2.4642, + "step": 99 + }, + { + "epoch": 0.1095890410958904, + "grad_norm": 2.4501149654388428, + "learning_rate": 1.5693430656934307e-06, + "loss": 2.4102, + "step": 100 + }, + { + "epoch": 0.1095890410958904, + "eval_loss": 2.5452804565429688, + "eval_runtime": 37.4224, + "eval_samples_per_second": 26.722, + "eval_steps_per_second": 1.122, + "step": 100 + }, + { + "epoch": 0.11068493150684931, + "grad_norm": 2.475864887237549, + "learning_rate": 1.5875912408759126e-06, + "loss": 2.4147, + "step": 101 + }, + { + "epoch": 0.11178082191780822, + "grad_norm": 2.558875322341919, + "learning_rate": 1.6058394160583942e-06, + "loss": 2.4694, + "step": 102 + }, + { + "epoch": 0.11287671232876713, + "grad_norm": 2.1085734367370605, + "learning_rate": 1.624087591240876e-06, + "loss": 2.5099, + "step": 103 + }, + { + "epoch": 0.11397260273972602, + "grad_norm": 2.410731792449951, + "learning_rate": 1.6423357664233577e-06, + "loss": 2.432, + "step": 104 + }, + { + "epoch": 0.11506849315068493, + "grad_norm": 2.2800745964050293, + "learning_rate": 1.6605839416058394e-06, + "loss": 2.3669, + "step": 105 + }, + { + "epoch": 0.11616438356164384, + "grad_norm": 2.2234933376312256, + "learning_rate": 1.6788321167883212e-06, + "loss": 2.4145, + "step": 106 + }, + { + "epoch": 0.11726027397260275, + "grad_norm": 2.3367040157318115, + "learning_rate": 1.6970802919708029e-06, + "loss": 2.4272, + "step": 107 + }, + { + "epoch": 0.11835616438356164, + "grad_norm": 2.7002811431884766, + "learning_rate": 1.715328467153285e-06, + "loss": 2.3478, + "step": 108 + }, + { + "epoch": 0.11945205479452055, + "grad_norm": 2.6628332138061523, + "learning_rate": 1.7335766423357666e-06, + "loss": 2.4096, + "step": 109 + }, + { + "epoch": 0.12054794520547946, + "grad_norm": 2.4174230098724365, + "learning_rate": 1.7518248175182485e-06, + "loss": 2.407, + "step": 110 + }, + { + "epoch": 0.12054794520547946, + "eval_loss": 2.491746664047241, + "eval_runtime": 37.2555, + "eval_samples_per_second": 26.842, + "eval_steps_per_second": 1.127, + "step": 110 + }, + { + "epoch": 0.12164383561643835, + "grad_norm": 2.24796986579895, + "learning_rate": 1.77007299270073e-06, + "loss": 2.4369, + "step": 111 + }, + { + "epoch": 0.12273972602739726, + "grad_norm": 2.720939874649048, + "learning_rate": 1.788321167883212e-06, + "loss": 2.4116, + "step": 112 + }, + { + "epoch": 0.12383561643835617, + "grad_norm": 2.6907670497894287, + "learning_rate": 1.8065693430656936e-06, + "loss": 2.4577, + "step": 113 + }, + { + "epoch": 0.12493150684931507, + "grad_norm": 2.559636354446411, + "learning_rate": 1.8248175182481753e-06, + "loss": 2.3999, + "step": 114 + }, + { + "epoch": 0.12602739726027398, + "grad_norm": 2.6730551719665527, + "learning_rate": 1.8430656934306571e-06, + "loss": 2.3428, + "step": 115 + }, + { + "epoch": 0.12712328767123288, + "grad_norm": 3.172179937362671, + "learning_rate": 1.8613138686131388e-06, + "loss": 2.3722, + "step": 116 + }, + { + "epoch": 0.12821917808219177, + "grad_norm": 2.6757171154022217, + "learning_rate": 1.8795620437956206e-06, + "loss": 2.3572, + "step": 117 + }, + { + "epoch": 0.1293150684931507, + "grad_norm": 2.5840766429901123, + "learning_rate": 1.8978102189781023e-06, + "loss": 2.353, + "step": 118 + }, + { + "epoch": 0.1304109589041096, + "grad_norm": 2.6248581409454346, + "learning_rate": 1.916058394160584e-06, + "loss": 2.36, + "step": 119 + }, + { + "epoch": 0.13150684931506848, + "grad_norm": 2.428516387939453, + "learning_rate": 1.934306569343066e-06, + "loss": 2.3821, + "step": 120 + }, + { + "epoch": 0.13150684931506848, + "eval_loss": 2.451565980911255, + "eval_runtime": 37.3669, + "eval_samples_per_second": 26.762, + "eval_steps_per_second": 1.124, + "step": 120 + }, + { + "epoch": 0.1326027397260274, + "grad_norm": 2.9956233501434326, + "learning_rate": 1.952554744525548e-06, + "loss": 2.2778, + "step": 121 + }, + { + "epoch": 0.1336986301369863, + "grad_norm": 2.4120893478393555, + "learning_rate": 1.9708029197080293e-06, + "loss": 2.325, + "step": 122 + }, + { + "epoch": 0.13479452054794522, + "grad_norm": 2.1410024166107178, + "learning_rate": 1.989051094890511e-06, + "loss": 2.3248, + "step": 123 + }, + { + "epoch": 0.1358904109589041, + "grad_norm": 2.923799514770508, + "learning_rate": 2.007299270072993e-06, + "loss": 2.327, + "step": 124 + }, + { + "epoch": 0.136986301369863, + "grad_norm": 3.0770339965820312, + "learning_rate": 2.025547445255475e-06, + "loss": 2.3847, + "step": 125 + }, + { + "epoch": 0.13808219178082193, + "grad_norm": 2.6408181190490723, + "learning_rate": 2.0437956204379563e-06, + "loss": 2.3196, + "step": 126 + }, + { + "epoch": 0.13917808219178082, + "grad_norm": 2.9127118587493896, + "learning_rate": 2.062043795620438e-06, + "loss": 2.3287, + "step": 127 + }, + { + "epoch": 0.14027397260273972, + "grad_norm": 2.9805052280426025, + "learning_rate": 2.08029197080292e-06, + "loss": 2.3243, + "step": 128 + }, + { + "epoch": 0.14136986301369864, + "grad_norm": 3.1342010498046875, + "learning_rate": 2.0985401459854014e-06, + "loss": 2.38, + "step": 129 + }, + { + "epoch": 0.14246575342465753, + "grad_norm": 2.678422451019287, + "learning_rate": 2.1167883211678833e-06, + "loss": 2.2727, + "step": 130 + }, + { + "epoch": 0.14246575342465753, + "eval_loss": 2.4187021255493164, + "eval_runtime": 37.3117, + "eval_samples_per_second": 26.801, + "eval_steps_per_second": 1.126, + "step": 130 + }, + { + "epoch": 0.14356164383561643, + "grad_norm": 2.9281492233276367, + "learning_rate": 2.135036496350365e-06, + "loss": 2.3327, + "step": 131 + }, + { + "epoch": 0.14465753424657535, + "grad_norm": 2.9592933654785156, + "learning_rate": 2.1532846715328466e-06, + "loss": 2.3674, + "step": 132 + }, + { + "epoch": 0.14575342465753424, + "grad_norm": 3.404761552810669, + "learning_rate": 2.171532846715329e-06, + "loss": 2.2679, + "step": 133 + }, + { + "epoch": 0.14684931506849316, + "grad_norm": 3.635571241378784, + "learning_rate": 2.1897810218978103e-06, + "loss": 2.2799, + "step": 134 + }, + { + "epoch": 0.14794520547945206, + "grad_norm": 3.9130170345306396, + "learning_rate": 2.208029197080292e-06, + "loss": 2.3263, + "step": 135 + }, + { + "epoch": 0.14904109589041095, + "grad_norm": 4.300438404083252, + "learning_rate": 2.226277372262774e-06, + "loss": 2.2969, + "step": 136 + }, + { + "epoch": 0.15013698630136987, + "grad_norm": 2.863511085510254, + "learning_rate": 2.244525547445256e-06, + "loss": 2.297, + "step": 137 + }, + { + "epoch": 0.15123287671232877, + "grad_norm": 4.910321235656738, + "learning_rate": 2.2627737226277373e-06, + "loss": 2.3269, + "step": 138 + }, + { + "epoch": 0.15232876712328766, + "grad_norm": 3.3323862552642822, + "learning_rate": 2.281021897810219e-06, + "loss": 2.3195, + "step": 139 + }, + { + "epoch": 0.15342465753424658, + "grad_norm": 3.2672693729400635, + "learning_rate": 2.299270072992701e-06, + "loss": 2.2797, + "step": 140 + }, + { + "epoch": 0.15342465753424658, + "eval_loss": 2.3943886756896973, + "eval_runtime": 37.3137, + "eval_samples_per_second": 26.8, + "eval_steps_per_second": 1.126, + "step": 140 + }, + { + "epoch": 0.15452054794520548, + "grad_norm": 3.9229774475097656, + "learning_rate": 2.3175182481751825e-06, + "loss": 2.2718, + "step": 141 + }, + { + "epoch": 0.15561643835616437, + "grad_norm": 2.6409823894500732, + "learning_rate": 2.3357664233576643e-06, + "loss": 2.2828, + "step": 142 + }, + { + "epoch": 0.1567123287671233, + "grad_norm": 4.6122612953186035, + "learning_rate": 2.354014598540146e-06, + "loss": 2.247, + "step": 143 + }, + { + "epoch": 0.1578082191780822, + "grad_norm": 3.391357898712158, + "learning_rate": 2.3722627737226276e-06, + "loss": 2.2863, + "step": 144 + }, + { + "epoch": 0.1589041095890411, + "grad_norm": 3.3550853729248047, + "learning_rate": 2.39051094890511e-06, + "loss": 2.2789, + "step": 145 + }, + { + "epoch": 0.16, + "grad_norm": 3.7376883029937744, + "learning_rate": 2.4087591240875918e-06, + "loss": 2.2297, + "step": 146 + }, + { + "epoch": 0.1610958904109589, + "grad_norm": 3.005157709121704, + "learning_rate": 2.427007299270073e-06, + "loss": 2.2467, + "step": 147 + }, + { + "epoch": 0.16219178082191782, + "grad_norm": 3.4616103172302246, + "learning_rate": 2.445255474452555e-06, + "loss": 2.2612, + "step": 148 + }, + { + "epoch": 0.1632876712328767, + "grad_norm": 3.3595101833343506, + "learning_rate": 2.463503649635037e-06, + "loss": 2.3021, + "step": 149 + }, + { + "epoch": 0.1643835616438356, + "grad_norm": 4.831241607666016, + "learning_rate": 2.4817518248175183e-06, + "loss": 2.2174, + "step": 150 + }, + { + "epoch": 0.1643835616438356, + "eval_loss": 2.3677749633789062, + "eval_runtime": 37.3092, + "eval_samples_per_second": 26.803, + "eval_steps_per_second": 1.126, + "step": 150 + }, + { + "epoch": 0.16547945205479453, + "grad_norm": 2.778820037841797, + "learning_rate": 2.5e-06, + "loss": 2.1856, + "step": 151 + }, + { + "epoch": 0.16657534246575342, + "grad_norm": 4.566643714904785, + "learning_rate": 2.518248175182482e-06, + "loss": 2.3186, + "step": 152 + }, + { + "epoch": 0.16767123287671232, + "grad_norm": 2.9908089637756348, + "learning_rate": 2.5364963503649635e-06, + "loss": 2.2127, + "step": 153 + }, + { + "epoch": 0.16876712328767124, + "grad_norm": 2.9702792167663574, + "learning_rate": 2.5547445255474458e-06, + "loss": 2.2071, + "step": 154 + }, + { + "epoch": 0.16986301369863013, + "grad_norm": 2.850044012069702, + "learning_rate": 2.5729927007299272e-06, + "loss": 2.2585, + "step": 155 + }, + { + "epoch": 0.17095890410958905, + "grad_norm": 2.926091432571411, + "learning_rate": 2.591240875912409e-06, + "loss": 2.2397, + "step": 156 + }, + { + "epoch": 0.17205479452054795, + "grad_norm": 2.8720169067382812, + "learning_rate": 2.6094890510948905e-06, + "loss": 2.2306, + "step": 157 + }, + { + "epoch": 0.17315068493150684, + "grad_norm": 2.8837177753448486, + "learning_rate": 2.627737226277373e-06, + "loss": 2.2666, + "step": 158 + }, + { + "epoch": 0.17424657534246576, + "grad_norm": 3.119621753692627, + "learning_rate": 2.6459854014598542e-06, + "loss": 2.2722, + "step": 159 + }, + { + "epoch": 0.17534246575342466, + "grad_norm": 2.7085700035095215, + "learning_rate": 2.664233576642336e-06, + "loss": 2.2346, + "step": 160 + }, + { + "epoch": 0.17534246575342466, + "eval_loss": 2.352083921432495, + "eval_runtime": 37.245, + "eval_samples_per_second": 26.849, + "eval_steps_per_second": 1.128, + "step": 160 + }, + { + "epoch": 0.17643835616438355, + "grad_norm": 2.9964346885681152, + "learning_rate": 2.6824817518248175e-06, + "loss": 2.2155, + "step": 161 + }, + { + "epoch": 0.17753424657534247, + "grad_norm": 3.1535208225250244, + "learning_rate": 2.7007299270072994e-06, + "loss": 2.235, + "step": 162 + }, + { + "epoch": 0.17863013698630137, + "grad_norm": 2.7240943908691406, + "learning_rate": 2.718978102189781e-06, + "loss": 2.2095, + "step": 163 + }, + { + "epoch": 0.17972602739726026, + "grad_norm": 3.122654676437378, + "learning_rate": 2.737226277372263e-06, + "loss": 2.2667, + "step": 164 + }, + { + "epoch": 0.18082191780821918, + "grad_norm": 2.9718410968780518, + "learning_rate": 2.755474452554745e-06, + "loss": 2.2499, + "step": 165 + }, + { + "epoch": 0.18191780821917808, + "grad_norm": 3.0081896781921387, + "learning_rate": 2.7737226277372264e-06, + "loss": 2.2089, + "step": 166 + }, + { + "epoch": 0.18301369863013697, + "grad_norm": 2.829509973526001, + "learning_rate": 2.7919708029197082e-06, + "loss": 2.3139, + "step": 167 + }, + { + "epoch": 0.1841095890410959, + "grad_norm": 3.4480397701263428, + "learning_rate": 2.81021897810219e-06, + "loss": 2.2705, + "step": 168 + }, + { + "epoch": 0.1852054794520548, + "grad_norm": 3.1107025146484375, + "learning_rate": 2.828467153284672e-06, + "loss": 2.1895, + "step": 169 + }, + { + "epoch": 0.1863013698630137, + "grad_norm": 2.916790008544922, + "learning_rate": 2.8467153284671534e-06, + "loss": 2.2317, + "step": 170 + }, + { + "epoch": 0.1863013698630137, + "eval_loss": 2.336153030395508, + "eval_runtime": 37.2401, + "eval_samples_per_second": 26.853, + "eval_steps_per_second": 1.128, + "step": 170 + }, + { + "epoch": 0.1873972602739726, + "grad_norm": 3.2582287788391113, + "learning_rate": 2.8649635036496353e-06, + "loss": 2.143, + "step": 171 + }, + { + "epoch": 0.1884931506849315, + "grad_norm": 2.430973768234253, + "learning_rate": 2.8832116788321167e-06, + "loss": 2.1977, + "step": 172 + }, + { + "epoch": 0.18958904109589042, + "grad_norm": 2.8763599395751953, + "learning_rate": 2.901459854014599e-06, + "loss": 2.1551, + "step": 173 + }, + { + "epoch": 0.19068493150684931, + "grad_norm": 2.9242653846740723, + "learning_rate": 2.9197080291970804e-06, + "loss": 2.2569, + "step": 174 + }, + { + "epoch": 0.1917808219178082, + "grad_norm": 2.623551607131958, + "learning_rate": 2.9379562043795623e-06, + "loss": 2.2329, + "step": 175 + }, + { + "epoch": 0.19287671232876713, + "grad_norm": 2.9511337280273438, + "learning_rate": 2.956204379562044e-06, + "loss": 2.2154, + "step": 176 + }, + { + "epoch": 0.19397260273972602, + "grad_norm": 2.415118932723999, + "learning_rate": 2.9744525547445256e-06, + "loss": 2.2637, + "step": 177 + }, + { + "epoch": 0.19506849315068492, + "grad_norm": 2.9664924144744873, + "learning_rate": 2.992700729927008e-06, + "loss": 2.2263, + "step": 178 + }, + { + "epoch": 0.19616438356164384, + "grad_norm": 2.7214572429656982, + "learning_rate": 3.0109489051094893e-06, + "loss": 2.1667, + "step": 179 + }, + { + "epoch": 0.19726027397260273, + "grad_norm": 2.8712823390960693, + "learning_rate": 3.029197080291971e-06, + "loss": 2.1661, + "step": 180 + }, + { + "epoch": 0.19726027397260273, + "eval_loss": 2.3191800117492676, + "eval_runtime": 37.24, + "eval_samples_per_second": 26.853, + "eval_steps_per_second": 1.128, + "step": 180 + }, + { + "epoch": 0.19835616438356166, + "grad_norm": 2.959709405899048, + "learning_rate": 3.0474452554744526e-06, + "loss": 2.2483, + "step": 181 + }, + { + "epoch": 0.19945205479452055, + "grad_norm": 2.567089319229126, + "learning_rate": 3.065693430656935e-06, + "loss": 2.2382, + "step": 182 + }, + { + "epoch": 0.20054794520547944, + "grad_norm": 2.47188401222229, + "learning_rate": 3.0839416058394163e-06, + "loss": 2.2148, + "step": 183 + }, + { + "epoch": 0.20164383561643837, + "grad_norm": 2.3435261249542236, + "learning_rate": 3.102189781021898e-06, + "loss": 2.2334, + "step": 184 + }, + { + "epoch": 0.20273972602739726, + "grad_norm": 2.5280468463897705, + "learning_rate": 3.1204379562043796e-06, + "loss": 2.2214, + "step": 185 + }, + { + "epoch": 0.20383561643835615, + "grad_norm": 2.4808971881866455, + "learning_rate": 3.1386861313868614e-06, + "loss": 2.1554, + "step": 186 + }, + { + "epoch": 0.20493150684931508, + "grad_norm": 2.9937026500701904, + "learning_rate": 3.156934306569343e-06, + "loss": 2.1389, + "step": 187 + }, + { + "epoch": 0.20602739726027397, + "grad_norm": 2.9220476150512695, + "learning_rate": 3.175182481751825e-06, + "loss": 2.2515, + "step": 188 + }, + { + "epoch": 0.20712328767123286, + "grad_norm": 3.0449001789093018, + "learning_rate": 3.193430656934307e-06, + "loss": 2.2016, + "step": 189 + }, + { + "epoch": 0.20821917808219179, + "grad_norm": 2.8027074337005615, + "learning_rate": 3.2116788321167884e-06, + "loss": 2.1823, + "step": 190 + }, + { + "epoch": 0.20821917808219179, + "eval_loss": 2.309677839279175, + "eval_runtime": 37.2442, + "eval_samples_per_second": 26.85, + "eval_steps_per_second": 1.128, + "step": 190 + }, + { + "epoch": 0.20931506849315068, + "grad_norm": 2.9158318042755127, + "learning_rate": 3.2299270072992707e-06, + "loss": 2.1928, + "step": 191 + }, + { + "epoch": 0.2104109589041096, + "grad_norm": 2.5328080654144287, + "learning_rate": 3.248175182481752e-06, + "loss": 2.1828, + "step": 192 + }, + { + "epoch": 0.2115068493150685, + "grad_norm": 3.1709704399108887, + "learning_rate": 3.266423357664234e-06, + "loss": 2.1052, + "step": 193 + }, + { + "epoch": 0.2126027397260274, + "grad_norm": 2.9156367778778076, + "learning_rate": 3.2846715328467155e-06, + "loss": 2.1943, + "step": 194 + }, + { + "epoch": 0.2136986301369863, + "grad_norm": 2.879408836364746, + "learning_rate": 3.3029197080291973e-06, + "loss": 2.1782, + "step": 195 + }, + { + "epoch": 0.2147945205479452, + "grad_norm": 3.748483180999756, + "learning_rate": 3.3211678832116788e-06, + "loss": 2.2107, + "step": 196 + }, + { + "epoch": 0.2158904109589041, + "grad_norm": 3.197510004043579, + "learning_rate": 3.339416058394161e-06, + "loss": 2.1845, + "step": 197 + }, + { + "epoch": 0.21698630136986302, + "grad_norm": 3.2596049308776855, + "learning_rate": 3.3576642335766425e-06, + "loss": 2.1816, + "step": 198 + }, + { + "epoch": 0.21808219178082192, + "grad_norm": 3.3329780101776123, + "learning_rate": 3.3759124087591243e-06, + "loss": 2.181, + "step": 199 + }, + { + "epoch": 0.2191780821917808, + "grad_norm": 2.7302327156066895, + "learning_rate": 3.3941605839416058e-06, + "loss": 2.1761, + "step": 200 + }, + { + "epoch": 0.2191780821917808, + "eval_loss": 2.2909653186798096, + "eval_runtime": 37.284, + "eval_samples_per_second": 26.821, + "eval_steps_per_second": 1.126, + "step": 200 + }, + { + "epoch": 0.22027397260273973, + "grad_norm": 3.002255916595459, + "learning_rate": 3.412408759124088e-06, + "loss": 2.1868, + "step": 201 + }, + { + "epoch": 0.22136986301369863, + "grad_norm": 2.6694891452789307, + "learning_rate": 3.43065693430657e-06, + "loss": 2.192, + "step": 202 + }, + { + "epoch": 0.22246575342465755, + "grad_norm": 2.9225986003875732, + "learning_rate": 3.4489051094890513e-06, + "loss": 2.1303, + "step": 203 + }, + { + "epoch": 0.22356164383561644, + "grad_norm": 2.7844080924987793, + "learning_rate": 3.467153284671533e-06, + "loss": 2.1788, + "step": 204 + }, + { + "epoch": 0.22465753424657534, + "grad_norm": 2.726123332977295, + "learning_rate": 3.4854014598540146e-06, + "loss": 2.1193, + "step": 205 + }, + { + "epoch": 0.22575342465753426, + "grad_norm": 2.838040590286255, + "learning_rate": 3.503649635036497e-06, + "loss": 2.1674, + "step": 206 + }, + { + "epoch": 0.22684931506849315, + "grad_norm": 2.6863937377929688, + "learning_rate": 3.5218978102189783e-06, + "loss": 2.161, + "step": 207 + }, + { + "epoch": 0.22794520547945205, + "grad_norm": 2.5116159915924072, + "learning_rate": 3.54014598540146e-06, + "loss": 2.179, + "step": 208 + }, + { + "epoch": 0.22904109589041097, + "grad_norm": 3.381863594055176, + "learning_rate": 3.5583941605839416e-06, + "loss": 2.2035, + "step": 209 + }, + { + "epoch": 0.23013698630136986, + "grad_norm": 3.2741382122039795, + "learning_rate": 3.576642335766424e-06, + "loss": 2.2518, + "step": 210 + }, + { + "epoch": 0.23013698630136986, + "eval_loss": 2.2773635387420654, + "eval_runtime": 37.2658, + "eval_samples_per_second": 26.834, + "eval_steps_per_second": 1.127, + "step": 210 + }, + { + "epoch": 0.23123287671232876, + "grad_norm": 2.917370319366455, + "learning_rate": 3.5948905109489054e-06, + "loss": 2.2265, + "step": 211 + }, + { + "epoch": 0.23232876712328768, + "grad_norm": 2.8392834663391113, + "learning_rate": 3.6131386861313872e-06, + "loss": 2.0709, + "step": 212 + }, + { + "epoch": 0.23342465753424657, + "grad_norm": 2.753612518310547, + "learning_rate": 3.6313868613138687e-06, + "loss": 2.0752, + "step": 213 + }, + { + "epoch": 0.2345205479452055, + "grad_norm": 3.1477746963500977, + "learning_rate": 3.6496350364963505e-06, + "loss": 2.1169, + "step": 214 + }, + { + "epoch": 0.2356164383561644, + "grad_norm": 2.5448100566864014, + "learning_rate": 3.667883211678833e-06, + "loss": 2.1338, + "step": 215 + }, + { + "epoch": 0.23671232876712328, + "grad_norm": 3.1622560024261475, + "learning_rate": 3.6861313868613142e-06, + "loss": 2.1065, + "step": 216 + }, + { + "epoch": 0.2378082191780822, + "grad_norm": 2.7054336071014404, + "learning_rate": 3.704379562043796e-06, + "loss": 2.1104, + "step": 217 + }, + { + "epoch": 0.2389041095890411, + "grad_norm": 2.964141845703125, + "learning_rate": 3.7226277372262775e-06, + "loss": 2.1949, + "step": 218 + }, + { + "epoch": 0.24, + "grad_norm": 2.7586112022399902, + "learning_rate": 3.7408759124087594e-06, + "loss": 2.165, + "step": 219 + }, + { + "epoch": 0.2410958904109589, + "grad_norm": 2.9035964012145996, + "learning_rate": 3.7591240875912412e-06, + "loss": 2.1192, + "step": 220 + }, + { + "epoch": 0.2410958904109589, + "eval_loss": 2.2649471759796143, + "eval_runtime": 37.4311, + "eval_samples_per_second": 26.716, + "eval_steps_per_second": 1.122, + "step": 220 + }, + { + "epoch": 0.2421917808219178, + "grad_norm": 2.4212076663970947, + "learning_rate": 3.777372262773723e-06, + "loss": 2.1389, + "step": 221 + }, + { + "epoch": 0.2432876712328767, + "grad_norm": 2.973227024078369, + "learning_rate": 3.7956204379562045e-06, + "loss": 2.132, + "step": 222 + }, + { + "epoch": 0.24438356164383562, + "grad_norm": 2.9831430912017822, + "learning_rate": 3.8138686131386864e-06, + "loss": 2.1142, + "step": 223 + }, + { + "epoch": 0.24547945205479452, + "grad_norm": 3.2880353927612305, + "learning_rate": 3.832116788321168e-06, + "loss": 2.168, + "step": 224 + }, + { + "epoch": 0.2465753424657534, + "grad_norm": 3.3920998573303223, + "learning_rate": 3.85036496350365e-06, + "loss": 2.1277, + "step": 225 + }, + { + "epoch": 0.24767123287671233, + "grad_norm": 2.7170026302337646, + "learning_rate": 3.868613138686132e-06, + "loss": 2.0936, + "step": 226 + }, + { + "epoch": 0.24876712328767123, + "grad_norm": 3.6220011711120605, + "learning_rate": 3.886861313868613e-06, + "loss": 2.1259, + "step": 227 + }, + { + "epoch": 0.24986301369863015, + "grad_norm": 3.4229273796081543, + "learning_rate": 3.905109489051096e-06, + "loss": 2.1421, + "step": 228 + }, + { + "epoch": 0.25095890410958904, + "grad_norm": 3.0358338356018066, + "learning_rate": 3.923357664233577e-06, + "loss": 2.1747, + "step": 229 + }, + { + "epoch": 0.25205479452054796, + "grad_norm": 2.611771821975708, + "learning_rate": 3.9416058394160585e-06, + "loss": 2.1282, + "step": 230 + }, + { + "epoch": 0.25205479452054796, + "eval_loss": 2.254887580871582, + "eval_runtime": 37.5943, + "eval_samples_per_second": 26.6, + "eval_steps_per_second": 1.117, + "step": 230 + }, + { + "epoch": 0.25315068493150683, + "grad_norm": 3.846412181854248, + "learning_rate": 3.95985401459854e-06, + "loss": 2.1892, + "step": 231 + }, + { + "epoch": 0.25424657534246575, + "grad_norm": 2.488525390625, + "learning_rate": 3.978102189781022e-06, + "loss": 2.2064, + "step": 232 + }, + { + "epoch": 0.2553424657534247, + "grad_norm": 3.114964485168457, + "learning_rate": 3.996350364963504e-06, + "loss": 2.0653, + "step": 233 + }, + { + "epoch": 0.25643835616438354, + "grad_norm": 2.5559768676757812, + "learning_rate": 4.014598540145986e-06, + "loss": 2.1503, + "step": 234 + }, + { + "epoch": 0.25753424657534246, + "grad_norm": 3.2347540855407715, + "learning_rate": 4.032846715328467e-06, + "loss": 2.0523, + "step": 235 + }, + { + "epoch": 0.2586301369863014, + "grad_norm": 2.8086111545562744, + "learning_rate": 4.05109489051095e-06, + "loss": 2.1206, + "step": 236 + }, + { + "epoch": 0.25972602739726025, + "grad_norm": 2.909231185913086, + "learning_rate": 4.069343065693431e-06, + "loss": 2.139, + "step": 237 + }, + { + "epoch": 0.2608219178082192, + "grad_norm": 3.2171082496643066, + "learning_rate": 4.0875912408759126e-06, + "loss": 2.0908, + "step": 238 + }, + { + "epoch": 0.2619178082191781, + "grad_norm": 2.3851265907287598, + "learning_rate": 4.105839416058395e-06, + "loss": 2.1133, + "step": 239 + }, + { + "epoch": 0.26301369863013696, + "grad_norm": 3.3912737369537354, + "learning_rate": 4.124087591240876e-06, + "loss": 2.1911, + "step": 240 + }, + { + "epoch": 0.26301369863013696, + "eval_loss": 2.2436957359313965, + "eval_runtime": 37.2947, + "eval_samples_per_second": 26.813, + "eval_steps_per_second": 1.126, + "step": 240 + }, + { + "epoch": 0.2641095890410959, + "grad_norm": 2.82906174659729, + "learning_rate": 4.1423357664233586e-06, + "loss": 2.11, + "step": 241 + }, + { + "epoch": 0.2652054794520548, + "grad_norm": 3.0317952632904053, + "learning_rate": 4.16058394160584e-06, + "loss": 2.0953, + "step": 242 + }, + { + "epoch": 0.2663013698630137, + "grad_norm": 3.6438329219818115, + "learning_rate": 4.1788321167883214e-06, + "loss": 2.1746, + "step": 243 + }, + { + "epoch": 0.2673972602739726, + "grad_norm": 2.6690592765808105, + "learning_rate": 4.197080291970803e-06, + "loss": 2.0555, + "step": 244 + }, + { + "epoch": 0.2684931506849315, + "grad_norm": 3.135376453399658, + "learning_rate": 4.215328467153285e-06, + "loss": 2.1394, + "step": 245 + }, + { + "epoch": 0.26958904109589044, + "grad_norm": 2.5474987030029297, + "learning_rate": 4.233576642335767e-06, + "loss": 2.0822, + "step": 246 + }, + { + "epoch": 0.2706849315068493, + "grad_norm": 2.883108377456665, + "learning_rate": 4.251824817518249e-06, + "loss": 2.0869, + "step": 247 + }, + { + "epoch": 0.2717808219178082, + "grad_norm": 2.9801037311553955, + "learning_rate": 4.27007299270073e-06, + "loss": 2.1194, + "step": 248 + }, + { + "epoch": 0.27287671232876715, + "grad_norm": 2.7688217163085938, + "learning_rate": 4.288321167883212e-06, + "loss": 2.0908, + "step": 249 + }, + { + "epoch": 0.273972602739726, + "grad_norm": 3.205148458480835, + "learning_rate": 4.306569343065693e-06, + "loss": 2.0762, + "step": 250 + }, + { + "epoch": 0.273972602739726, + "eval_loss": 2.238699197769165, + "eval_runtime": 37.2428, + "eval_samples_per_second": 26.851, + "eval_steps_per_second": 1.128, + "step": 250 + }, + { + "epoch": 0.27506849315068493, + "grad_norm": 2.8262946605682373, + "learning_rate": 4.3248175182481755e-06, + "loss": 2.0509, + "step": 251 + }, + { + "epoch": 0.27616438356164386, + "grad_norm": 3.156261682510376, + "learning_rate": 4.343065693430658e-06, + "loss": 2.1213, + "step": 252 + }, + { + "epoch": 0.2772602739726027, + "grad_norm": 2.6367998123168945, + "learning_rate": 4.361313868613139e-06, + "loss": 2.1071, + "step": 253 + }, + { + "epoch": 0.27835616438356164, + "grad_norm": 2.871534824371338, + "learning_rate": 4.379562043795621e-06, + "loss": 2.1008, + "step": 254 + }, + { + "epoch": 0.27945205479452057, + "grad_norm": 2.839432716369629, + "learning_rate": 4.397810218978102e-06, + "loss": 2.1293, + "step": 255 + }, + { + "epoch": 0.28054794520547943, + "grad_norm": 2.627143383026123, + "learning_rate": 4.416058394160584e-06, + "loss": 2.1141, + "step": 256 + }, + { + "epoch": 0.28164383561643835, + "grad_norm": 2.8292996883392334, + "learning_rate": 4.434306569343066e-06, + "loss": 2.0981, + "step": 257 + }, + { + "epoch": 0.2827397260273973, + "grad_norm": 2.5711069107055664, + "learning_rate": 4.452554744525548e-06, + "loss": 2.0707, + "step": 258 + }, + { + "epoch": 0.28383561643835614, + "grad_norm": 2.488954544067383, + "learning_rate": 4.4708029197080295e-06, + "loss": 2.1172, + "step": 259 + }, + { + "epoch": 0.28493150684931506, + "grad_norm": 2.442209005355835, + "learning_rate": 4.489051094890512e-06, + "loss": 2.1119, + "step": 260 + }, + { + "epoch": 0.28493150684931506, + "eval_loss": 2.2306911945343018, + "eval_runtime": 37.2412, + "eval_samples_per_second": 26.852, + "eval_steps_per_second": 1.128, + "step": 260 + }, + { + "epoch": 0.286027397260274, + "grad_norm": 2.9147934913635254, + "learning_rate": 4.507299270072993e-06, + "loss": 2.1517, + "step": 261 + }, + { + "epoch": 0.28712328767123285, + "grad_norm": 2.549833297729492, + "learning_rate": 4.525547445255475e-06, + "loss": 2.1047, + "step": 262 + }, + { + "epoch": 0.2882191780821918, + "grad_norm": 3.3405206203460693, + "learning_rate": 4.543795620437956e-06, + "loss": 2.0975, + "step": 263 + }, + { + "epoch": 0.2893150684931507, + "grad_norm": 2.554431676864624, + "learning_rate": 4.562043795620438e-06, + "loss": 2.0622, + "step": 264 + }, + { + "epoch": 0.29041095890410956, + "grad_norm": 3.5529873371124268, + "learning_rate": 4.580291970802921e-06, + "loss": 2.1567, + "step": 265 + }, + { + "epoch": 0.2915068493150685, + "grad_norm": 2.421682119369507, + "learning_rate": 4.598540145985402e-06, + "loss": 2.1619, + "step": 266 + }, + { + "epoch": 0.2926027397260274, + "grad_norm": 2.918288230895996, + "learning_rate": 4.6167883211678835e-06, + "loss": 2.1238, + "step": 267 + }, + { + "epoch": 0.2936986301369863, + "grad_norm": 3.0737621784210205, + "learning_rate": 4.635036496350365e-06, + "loss": 2.0728, + "step": 268 + }, + { + "epoch": 0.2947945205479452, + "grad_norm": 2.5862410068511963, + "learning_rate": 4.653284671532847e-06, + "loss": 2.0675, + "step": 269 + }, + { + "epoch": 0.2958904109589041, + "grad_norm": 2.9004979133605957, + "learning_rate": 4.671532846715329e-06, + "loss": 2.1248, + "step": 270 + }, + { + "epoch": 0.2958904109589041, + "eval_loss": 2.2179365158081055, + "eval_runtime": 37.2352, + "eval_samples_per_second": 26.856, + "eval_steps_per_second": 1.128, + "step": 270 + }, + { + "epoch": 0.29698630136986304, + "grad_norm": 2.4784982204437256, + "learning_rate": 4.689781021897811e-06, + "loss": 2.1276, + "step": 271 + }, + { + "epoch": 0.2980821917808219, + "grad_norm": 2.7971739768981934, + "learning_rate": 4.708029197080292e-06, + "loss": 2.0931, + "step": 272 + }, + { + "epoch": 0.2991780821917808, + "grad_norm": 2.5839481353759766, + "learning_rate": 4.726277372262774e-06, + "loss": 2.1371, + "step": 273 + }, + { + "epoch": 0.30027397260273975, + "grad_norm": 2.7168564796447754, + "learning_rate": 4.744525547445255e-06, + "loss": 2.1086, + "step": 274 + }, + { + "epoch": 0.3013698630136986, + "grad_norm": 2.6566879749298096, + "learning_rate": 4.7627737226277375e-06, + "loss": 2.0368, + "step": 275 + }, + { + "epoch": 0.30246575342465754, + "grad_norm": 2.5068323612213135, + "learning_rate": 4.78102189781022e-06, + "loss": 2.1077, + "step": 276 + }, + { + "epoch": 0.30356164383561646, + "grad_norm": 2.631408214569092, + "learning_rate": 4.799270072992701e-06, + "loss": 2.1013, + "step": 277 + }, + { + "epoch": 0.3046575342465753, + "grad_norm": 2.5642812252044678, + "learning_rate": 4.8175182481751835e-06, + "loss": 2.0739, + "step": 278 + }, + { + "epoch": 0.30575342465753425, + "grad_norm": 2.3326497077941895, + "learning_rate": 4.835766423357665e-06, + "loss": 2.0971, + "step": 279 + }, + { + "epoch": 0.30684931506849317, + "grad_norm": 2.70701265335083, + "learning_rate": 4.854014598540146e-06, + "loss": 2.0904, + "step": 280 + }, + { + "epoch": 0.30684931506849317, + "eval_loss": 2.215972900390625, + "eval_runtime": 37.2196, + "eval_samples_per_second": 26.868, + "eval_steps_per_second": 1.128, + "step": 280 + }, + { + "epoch": 0.30794520547945203, + "grad_norm": 2.615201234817505, + "learning_rate": 4.872262773722628e-06, + "loss": 2.039, + "step": 281 + }, + { + "epoch": 0.30904109589041096, + "grad_norm": 2.2821998596191406, + "learning_rate": 4.89051094890511e-06, + "loss": 2.0529, + "step": 282 + }, + { + "epoch": 0.3101369863013699, + "grad_norm": 2.767153024673462, + "learning_rate": 4.9087591240875915e-06, + "loss": 2.0351, + "step": 283 + }, + { + "epoch": 0.31123287671232874, + "grad_norm": 2.537121295928955, + "learning_rate": 4.927007299270074e-06, + "loss": 2.0752, + "step": 284 + }, + { + "epoch": 0.31232876712328766, + "grad_norm": 2.835052013397217, + "learning_rate": 4.945255474452555e-06, + "loss": 2.0892, + "step": 285 + }, + { + "epoch": 0.3134246575342466, + "grad_norm": 2.5154924392700195, + "learning_rate": 4.963503649635037e-06, + "loss": 2.0569, + "step": 286 + }, + { + "epoch": 0.31452054794520545, + "grad_norm": 2.8212568759918213, + "learning_rate": 4.981751824817518e-06, + "loss": 2.1588, + "step": 287 + }, + { + "epoch": 0.3156164383561644, + "grad_norm": 2.3840994834899902, + "learning_rate": 5e-06, + "loss": 2.1409, + "step": 288 + }, + { + "epoch": 0.3167123287671233, + "grad_norm": 2.9641315937042236, + "learning_rate": 4.997969130787978e-06, + "loss": 2.0892, + "step": 289 + }, + { + "epoch": 0.3178082191780822, + "grad_norm": 2.4516749382019043, + "learning_rate": 4.995938261575955e-06, + "loss": 2.1105, + "step": 290 + }, + { + "epoch": 0.3178082191780822, + "eval_loss": 2.207075834274292, + "eval_runtime": 37.1304, + "eval_samples_per_second": 26.932, + "eval_steps_per_second": 1.131, + "step": 290 + }, + { + "epoch": 0.3189041095890411, + "grad_norm": 2.768869400024414, + "learning_rate": 4.993907392363932e-06, + "loss": 2.0231, + "step": 291 + }, + { + "epoch": 0.32, + "grad_norm": 2.771475315093994, + "learning_rate": 4.9918765231519095e-06, + "loss": 2.1018, + "step": 292 + }, + { + "epoch": 0.32109589041095893, + "grad_norm": 2.773983955383301, + "learning_rate": 4.989845653939887e-06, + "loss": 2.0317, + "step": 293 + }, + { + "epoch": 0.3221917808219178, + "grad_norm": 2.4447343349456787, + "learning_rate": 4.987814784727864e-06, + "loss": 2.0739, + "step": 294 + }, + { + "epoch": 0.3232876712328767, + "grad_norm": 2.444261312484741, + "learning_rate": 4.985783915515841e-06, + "loss": 2.073, + "step": 295 + }, + { + "epoch": 0.32438356164383564, + "grad_norm": 2.871792793273926, + "learning_rate": 4.983753046303819e-06, + "loss": 2.058, + "step": 296 + }, + { + "epoch": 0.3254794520547945, + "grad_norm": 2.3651537895202637, + "learning_rate": 4.981722177091796e-06, + "loss": 2.0749, + "step": 297 + }, + { + "epoch": 0.3265753424657534, + "grad_norm": 2.3395845890045166, + "learning_rate": 4.979691307879773e-06, + "loss": 2.065, + "step": 298 + }, + { + "epoch": 0.32767123287671235, + "grad_norm": 2.6275217533111572, + "learning_rate": 4.9776604386677504e-06, + "loss": 2.0945, + "step": 299 + }, + { + "epoch": 0.3287671232876712, + "grad_norm": 2.866285562515259, + "learning_rate": 4.975629569455728e-06, + "loss": 2.0702, + "step": 300 + }, + { + "epoch": 0.3287671232876712, + "eval_loss": 2.200012683868408, + "eval_runtime": 37.2082, + "eval_samples_per_second": 26.876, + "eval_steps_per_second": 1.129, + "step": 300 + }, + { + "epoch": 0.32986301369863014, + "grad_norm": 2.268401861190796, + "learning_rate": 4.973598700243705e-06, + "loss": 2.1257, + "step": 301 + }, + { + "epoch": 0.33095890410958906, + "grad_norm": 2.6197733879089355, + "learning_rate": 4.971567831031681e-06, + "loss": 2.0347, + "step": 302 + }, + { + "epoch": 0.3320547945205479, + "grad_norm": 2.4738848209381104, + "learning_rate": 4.9695369618196595e-06, + "loss": 2.0368, + "step": 303 + }, + { + "epoch": 0.33315068493150685, + "grad_norm": 2.705670118331909, + "learning_rate": 4.967506092607636e-06, + "loss": 1.9256, + "step": 304 + }, + { + "epoch": 0.33424657534246577, + "grad_norm": 2.813279628753662, + "learning_rate": 4.965475223395614e-06, + "loss": 2.0214, + "step": 305 + }, + { + "epoch": 0.33534246575342463, + "grad_norm": 2.320960283279419, + "learning_rate": 4.9634443541835905e-06, + "loss": 2.0342, + "step": 306 + }, + { + "epoch": 0.33643835616438356, + "grad_norm": 2.4495227336883545, + "learning_rate": 4.961413484971569e-06, + "loss": 2.0421, + "step": 307 + }, + { + "epoch": 0.3375342465753425, + "grad_norm": 2.5384018421173096, + "learning_rate": 4.959382615759545e-06, + "loss": 2.0572, + "step": 308 + }, + { + "epoch": 0.33863013698630134, + "grad_norm": 2.4018006324768066, + "learning_rate": 4.957351746547523e-06, + "loss": 2.0617, + "step": 309 + }, + { + "epoch": 0.33972602739726027, + "grad_norm": 2.552424907684326, + "learning_rate": 4.9553208773355e-06, + "loss": 2.0384, + "step": 310 + }, + { + "epoch": 0.33972602739726027, + "eval_loss": 2.1974849700927734, + "eval_runtime": 37.2308, + "eval_samples_per_second": 26.859, + "eval_steps_per_second": 1.128, + "step": 310 + }, + { + "epoch": 0.3408219178082192, + "grad_norm": 2.4587671756744385, + "learning_rate": 4.953290008123478e-06, + "loss": 2.1701, + "step": 311 + }, + { + "epoch": 0.3419178082191781, + "grad_norm": 2.4486496448516846, + "learning_rate": 4.951259138911454e-06, + "loss": 2.0654, + "step": 312 + }, + { + "epoch": 0.343013698630137, + "grad_norm": 2.123568058013916, + "learning_rate": 4.9492282696994315e-06, + "loss": 1.9961, + "step": 313 + }, + { + "epoch": 0.3441095890410959, + "grad_norm": 2.549788475036621, + "learning_rate": 4.947197400487409e-06, + "loss": 2.1265, + "step": 314 + }, + { + "epoch": 0.3452054794520548, + "grad_norm": 2.3453609943389893, + "learning_rate": 4.945166531275386e-06, + "loss": 2.0501, + "step": 315 + }, + { + "epoch": 0.3463013698630137, + "grad_norm": 2.3584256172180176, + "learning_rate": 4.943135662063363e-06, + "loss": 2.0402, + "step": 316 + }, + { + "epoch": 0.3473972602739726, + "grad_norm": 3.3100943565368652, + "learning_rate": 4.9411047928513406e-06, + "loss": 2.0545, + "step": 317 + }, + { + "epoch": 0.34849315068493153, + "grad_norm": 2.4266037940979004, + "learning_rate": 4.939073923639318e-06, + "loss": 2.0926, + "step": 318 + }, + { + "epoch": 0.3495890410958904, + "grad_norm": 3.101090908050537, + "learning_rate": 4.937043054427295e-06, + "loss": 2.0987, + "step": 319 + }, + { + "epoch": 0.3506849315068493, + "grad_norm": 2.743619203567505, + "learning_rate": 4.935012185215272e-06, + "loss": 2.0315, + "step": 320 + }, + { + "epoch": 0.3506849315068493, + "eval_loss": 2.188877820968628, + "eval_runtime": 37.4609, + "eval_samples_per_second": 26.694, + "eval_steps_per_second": 1.121, + "step": 320 + }, + { + "epoch": 0.35178082191780824, + "grad_norm": 2.6065986156463623, + "learning_rate": 4.93298131600325e-06, + "loss": 2.0166, + "step": 321 + }, + { + "epoch": 0.3528767123287671, + "grad_norm": 2.7554283142089844, + "learning_rate": 4.930950446791227e-06, + "loss": 2.0862, + "step": 322 + }, + { + "epoch": 0.353972602739726, + "grad_norm": 2.4773144721984863, + "learning_rate": 4.928919577579204e-06, + "loss": 2.0472, + "step": 323 + }, + { + "epoch": 0.35506849315068495, + "grad_norm": 2.695289134979248, + "learning_rate": 4.9268887083671815e-06, + "loss": 2.0567, + "step": 324 + }, + { + "epoch": 0.3561643835616438, + "grad_norm": 2.408220052719116, + "learning_rate": 4.924857839155159e-06, + "loss": 2.046, + "step": 325 + }, + { + "epoch": 0.35726027397260274, + "grad_norm": 2.3851559162139893, + "learning_rate": 4.922826969943136e-06, + "loss": 2.054, + "step": 326 + }, + { + "epoch": 0.35835616438356166, + "grad_norm": 2.8667280673980713, + "learning_rate": 4.920796100731113e-06, + "loss": 2.0717, + "step": 327 + }, + { + "epoch": 0.3594520547945205, + "grad_norm": 2.3814942836761475, + "learning_rate": 4.918765231519091e-06, + "loss": 1.9884, + "step": 328 + }, + { + "epoch": 0.36054794520547945, + "grad_norm": 2.341076612472534, + "learning_rate": 4.916734362307068e-06, + "loss": 2.0474, + "step": 329 + }, + { + "epoch": 0.36164383561643837, + "grad_norm": 2.3571853637695312, + "learning_rate": 4.914703493095045e-06, + "loss": 2.0009, + "step": 330 + }, + { + "epoch": 0.36164383561643837, + "eval_loss": 2.186048746109009, + "eval_runtime": 37.5746, + "eval_samples_per_second": 26.614, + "eval_steps_per_second": 1.118, + "step": 330 + }, + { + "epoch": 0.36273972602739724, + "grad_norm": 2.3770196437835693, + "learning_rate": 4.912672623883022e-06, + "loss": 2.0167, + "step": 331 + }, + { + "epoch": 0.36383561643835616, + "grad_norm": 2.3357701301574707, + "learning_rate": 4.910641754671e-06, + "loss": 1.9972, + "step": 332 + }, + { + "epoch": 0.3649315068493151, + "grad_norm": 2.4100499153137207, + "learning_rate": 4.908610885458977e-06, + "loss": 2.0319, + "step": 333 + }, + { + "epoch": 0.36602739726027395, + "grad_norm": 2.193018913269043, + "learning_rate": 4.906580016246954e-06, + "loss": 2.0285, + "step": 334 + }, + { + "epoch": 0.36712328767123287, + "grad_norm": 2.275637149810791, + "learning_rate": 4.9045491470349315e-06, + "loss": 2.0749, + "step": 335 + }, + { + "epoch": 0.3682191780821918, + "grad_norm": 2.347121238708496, + "learning_rate": 4.902518277822909e-06, + "loss": 2.0502, + "step": 336 + }, + { + "epoch": 0.3693150684931507, + "grad_norm": 2.210817575454712, + "learning_rate": 4.900487408610886e-06, + "loss": 2.0612, + "step": 337 + }, + { + "epoch": 0.3704109589041096, + "grad_norm": 2.15594482421875, + "learning_rate": 4.898456539398863e-06, + "loss": 2.0653, + "step": 338 + }, + { + "epoch": 0.3715068493150685, + "grad_norm": 2.3996052742004395, + "learning_rate": 4.89642567018684e-06, + "loss": 2.074, + "step": 339 + }, + { + "epoch": 0.3726027397260274, + "grad_norm": 2.384172201156616, + "learning_rate": 4.894394800974818e-06, + "loss": 2.1533, + "step": 340 + }, + { + "epoch": 0.3726027397260274, + "eval_loss": 2.175034284591675, + "eval_runtime": 37.2455, + "eval_samples_per_second": 26.849, + "eval_steps_per_second": 1.128, + "step": 340 + }, + { + "epoch": 0.3736986301369863, + "grad_norm": 2.3431336879730225, + "learning_rate": 4.892363931762794e-06, + "loss": 1.9919, + "step": 341 + }, + { + "epoch": 0.3747945205479452, + "grad_norm": 2.1828389167785645, + "learning_rate": 4.8903330625507725e-06, + "loss": 2.0703, + "step": 342 + }, + { + "epoch": 0.37589041095890413, + "grad_norm": 2.2117249965667725, + "learning_rate": 4.888302193338749e-06, + "loss": 2.0071, + "step": 343 + }, + { + "epoch": 0.376986301369863, + "grad_norm": 2.3595967292785645, + "learning_rate": 4.886271324126727e-06, + "loss": 1.9879, + "step": 344 + }, + { + "epoch": 0.3780821917808219, + "grad_norm": 2.5211799144744873, + "learning_rate": 4.8842404549147034e-06, + "loss": 2.0861, + "step": 345 + }, + { + "epoch": 0.37917808219178084, + "grad_norm": 2.1551895141601562, + "learning_rate": 4.8822095857026816e-06, + "loss": 2.0683, + "step": 346 + }, + { + "epoch": 0.3802739726027397, + "grad_norm": 2.4145829677581787, + "learning_rate": 4.880178716490658e-06, + "loss": 2.043, + "step": 347 + }, + { + "epoch": 0.38136986301369863, + "grad_norm": 2.4273521900177, + "learning_rate": 4.878147847278636e-06, + "loss": 1.9928, + "step": 348 + }, + { + "epoch": 0.38246575342465755, + "grad_norm": 2.261054754257202, + "learning_rate": 4.8761169780666125e-06, + "loss": 2.0065, + "step": 349 + }, + { + "epoch": 0.3835616438356164, + "grad_norm": 2.4703147411346436, + "learning_rate": 4.874086108854591e-06, + "loss": 1.9766, + "step": 350 + }, + { + "epoch": 0.3835616438356164, + "eval_loss": 2.176915168762207, + "eval_runtime": 37.3223, + "eval_samples_per_second": 26.794, + "eval_steps_per_second": 1.125, + "step": 350 + }, + { + "epoch": 0.38465753424657534, + "grad_norm": 2.3353374004364014, + "learning_rate": 4.872055239642567e-06, + "loss": 1.9874, + "step": 351 + }, + { + "epoch": 0.38575342465753426, + "grad_norm": 2.227834463119507, + "learning_rate": 4.870024370430544e-06, + "loss": 1.9644, + "step": 352 + }, + { + "epoch": 0.3868493150684931, + "grad_norm": 2.3155159950256348, + "learning_rate": 4.867993501218522e-06, + "loss": 2.0227, + "step": 353 + }, + { + "epoch": 0.38794520547945205, + "grad_norm": 2.2456047534942627, + "learning_rate": 4.865962632006499e-06, + "loss": 2.0566, + "step": 354 + }, + { + "epoch": 0.38904109589041097, + "grad_norm": 2.3364651203155518, + "learning_rate": 4.863931762794476e-06, + "loss": 1.9941, + "step": 355 + }, + { + "epoch": 0.39013698630136984, + "grad_norm": 2.2886059284210205, + "learning_rate": 4.8619008935824535e-06, + "loss": 2.0163, + "step": 356 + }, + { + "epoch": 0.39123287671232876, + "grad_norm": 2.3796911239624023, + "learning_rate": 4.859870024370431e-06, + "loss": 2.0525, + "step": 357 + }, + { + "epoch": 0.3923287671232877, + "grad_norm": 2.381986379623413, + "learning_rate": 4.857839155158408e-06, + "loss": 2.0682, + "step": 358 + }, + { + "epoch": 0.3934246575342466, + "grad_norm": 2.2048425674438477, + "learning_rate": 4.855808285946385e-06, + "loss": 2.0748, + "step": 359 + }, + { + "epoch": 0.39452054794520547, + "grad_norm": 2.162846088409424, + "learning_rate": 4.8537774167343626e-06, + "loss": 2.0512, + "step": 360 + }, + { + "epoch": 0.39452054794520547, + "eval_loss": 2.1645994186401367, + "eval_runtime": 37.3126, + "eval_samples_per_second": 26.801, + "eval_steps_per_second": 1.126, + "step": 360 + }, + { + "epoch": 0.3956164383561644, + "grad_norm": 2.4070873260498047, + "learning_rate": 4.85174654752234e-06, + "loss": 2.042, + "step": 361 + }, + { + "epoch": 0.3967123287671233, + "grad_norm": 2.11716628074646, + "learning_rate": 4.849715678310317e-06, + "loss": 2.0398, + "step": 362 + }, + { + "epoch": 0.3978082191780822, + "grad_norm": 2.265479803085327, + "learning_rate": 4.847684809098294e-06, + "loss": 2.0334, + "step": 363 + }, + { + "epoch": 0.3989041095890411, + "grad_norm": 2.2270994186401367, + "learning_rate": 4.845653939886272e-06, + "loss": 2.0292, + "step": 364 + }, + { + "epoch": 0.4, + "grad_norm": 2.406064033508301, + "learning_rate": 4.843623070674249e-06, + "loss": 2.0351, + "step": 365 + }, + { + "epoch": 0.4010958904109589, + "grad_norm": 2.216654062271118, + "learning_rate": 4.841592201462226e-06, + "loss": 2.005, + "step": 366 + }, + { + "epoch": 0.4021917808219178, + "grad_norm": 2.4241385459899902, + "learning_rate": 4.8395613322502035e-06, + "loss": 2.0647, + "step": 367 + }, + { + "epoch": 0.40328767123287673, + "grad_norm": 2.4315145015716553, + "learning_rate": 4.837530463038181e-06, + "loss": 1.9679, + "step": 368 + }, + { + "epoch": 0.4043835616438356, + "grad_norm": 2.3034439086914062, + "learning_rate": 4.835499593826158e-06, + "loss": 2.0259, + "step": 369 + }, + { + "epoch": 0.4054794520547945, + "grad_norm": 2.493436098098755, + "learning_rate": 4.833468724614135e-06, + "loss": 2.0468, + "step": 370 + }, + { + "epoch": 0.4054794520547945, + "eval_loss": 2.161508083343506, + "eval_runtime": 37.2692, + "eval_samples_per_second": 26.832, + "eval_steps_per_second": 1.127, + "step": 370 + }, + { + "epoch": 0.40657534246575344, + "grad_norm": 2.3887157440185547, + "learning_rate": 4.831437855402113e-06, + "loss": 2.0313, + "step": 371 + }, + { + "epoch": 0.4076712328767123, + "grad_norm": 2.563095808029175, + "learning_rate": 4.82940698619009e-06, + "loss": 1.9288, + "step": 372 + }, + { + "epoch": 0.40876712328767123, + "grad_norm": 2.2609968185424805, + "learning_rate": 4.827376116978067e-06, + "loss": 1.9975, + "step": 373 + }, + { + "epoch": 0.40986301369863015, + "grad_norm": 2.624257802963257, + "learning_rate": 4.8253452477660444e-06, + "loss": 2.0639, + "step": 374 + }, + { + "epoch": 0.410958904109589, + "grad_norm": 2.502197265625, + "learning_rate": 4.823314378554022e-06, + "loss": 2.0424, + "step": 375 + }, + { + "epoch": 0.41205479452054794, + "grad_norm": 2.36297345161438, + "learning_rate": 4.821283509341998e-06, + "loss": 2.027, + "step": 376 + }, + { + "epoch": 0.41315068493150686, + "grad_norm": 2.388918161392212, + "learning_rate": 4.819252640129976e-06, + "loss": 2.0002, + "step": 377 + }, + { + "epoch": 0.41424657534246573, + "grad_norm": 2.4262335300445557, + "learning_rate": 4.817221770917953e-06, + "loss": 2.0215, + "step": 378 + }, + { + "epoch": 0.41534246575342465, + "grad_norm": 2.8953094482421875, + "learning_rate": 4.815190901705931e-06, + "loss": 1.988, + "step": 379 + }, + { + "epoch": 0.41643835616438357, + "grad_norm": 2.3092124462127686, + "learning_rate": 4.813160032493907e-06, + "loss": 2.0199, + "step": 380 + }, + { + "epoch": 0.41643835616438357, + "eval_loss": 2.1589064598083496, + "eval_runtime": 37.2649, + "eval_samples_per_second": 26.835, + "eval_steps_per_second": 1.127, + "step": 380 + }, + { + "epoch": 0.41753424657534244, + "grad_norm": 3.0842413902282715, + "learning_rate": 4.811129163281885e-06, + "loss": 2.0134, + "step": 381 + }, + { + "epoch": 0.41863013698630136, + "grad_norm": 2.2015092372894287, + "learning_rate": 4.809098294069862e-06, + "loss": 2.0447, + "step": 382 + }, + { + "epoch": 0.4197260273972603, + "grad_norm": 2.93471097946167, + "learning_rate": 4.80706742485784e-06, + "loss": 2.0329, + "step": 383 + }, + { + "epoch": 0.4208219178082192, + "grad_norm": 2.3277227878570557, + "learning_rate": 4.805036555645816e-06, + "loss": 2.0805, + "step": 384 + }, + { + "epoch": 0.42191780821917807, + "grad_norm": 2.357966184616089, + "learning_rate": 4.8030056864337945e-06, + "loss": 2.0008, + "step": 385 + }, + { + "epoch": 0.423013698630137, + "grad_norm": 2.7505595684051514, + "learning_rate": 4.800974817221771e-06, + "loss": 2.059, + "step": 386 + }, + { + "epoch": 0.4241095890410959, + "grad_norm": 2.144338369369507, + "learning_rate": 4.798943948009749e-06, + "loss": 2.0125, + "step": 387 + }, + { + "epoch": 0.4252054794520548, + "grad_norm": 2.608407735824585, + "learning_rate": 4.7969130787977254e-06, + "loss": 2.0263, + "step": 388 + }, + { + "epoch": 0.4263013698630137, + "grad_norm": 2.4293761253356934, + "learning_rate": 4.794882209585703e-06, + "loss": 2.0107, + "step": 389 + }, + { + "epoch": 0.4273972602739726, + "grad_norm": 2.2136595249176025, + "learning_rate": 4.79285134037368e-06, + "loss": 2.0421, + "step": 390 + }, + { + "epoch": 0.4273972602739726, + "eval_loss": 2.1473641395568848, + "eval_runtime": 37.2895, + "eval_samples_per_second": 26.817, + "eval_steps_per_second": 1.126, + "step": 390 + }, + { + "epoch": 0.4284931506849315, + "grad_norm": 2.3706328868865967, + "learning_rate": 4.790820471161657e-06, + "loss": 1.9828, + "step": 391 + }, + { + "epoch": 0.4295890410958904, + "grad_norm": 2.1628777980804443, + "learning_rate": 4.7887896019496345e-06, + "loss": 2.0006, + "step": 392 + }, + { + "epoch": 0.43068493150684933, + "grad_norm": 2.5572149753570557, + "learning_rate": 4.786758732737612e-06, + "loss": 2.0046, + "step": 393 + }, + { + "epoch": 0.4317808219178082, + "grad_norm": 2.44661808013916, + "learning_rate": 4.784727863525589e-06, + "loss": 2.0513, + "step": 394 + }, + { + "epoch": 0.4328767123287671, + "grad_norm": 2.429487466812134, + "learning_rate": 4.782696994313566e-06, + "loss": 1.9619, + "step": 395 + }, + { + "epoch": 0.43397260273972604, + "grad_norm": 2.233353614807129, + "learning_rate": 4.780666125101544e-06, + "loss": 2.0141, + "step": 396 + }, + { + "epoch": 0.4350684931506849, + "grad_norm": 2.137549638748169, + "learning_rate": 4.778635255889521e-06, + "loss": 1.9512, + "step": 397 + }, + { + "epoch": 0.43616438356164383, + "grad_norm": 2.187638759613037, + "learning_rate": 4.776604386677498e-06, + "loss": 1.9802, + "step": 398 + }, + { + "epoch": 0.43726027397260275, + "grad_norm": 2.3084137439727783, + "learning_rate": 4.7745735174654755e-06, + "loss": 1.9887, + "step": 399 + }, + { + "epoch": 0.4383561643835616, + "grad_norm": 2.16654109954834, + "learning_rate": 4.772542648253453e-06, + "loss": 2.0535, + "step": 400 + }, + { + "epoch": 0.4383561643835616, + "eval_loss": 2.1444413661956787, + "eval_runtime": 37.2158, + "eval_samples_per_second": 26.87, + "eval_steps_per_second": 1.129, + "step": 400 + }, + { + "epoch": 0.43945205479452054, + "grad_norm": 2.0973072052001953, + "learning_rate": 4.77051177904143e-06, + "loss": 2.0529, + "step": 401 + }, + { + "epoch": 0.44054794520547946, + "grad_norm": 2.421881675720215, + "learning_rate": 4.768480909829407e-06, + "loss": 1.9893, + "step": 402 + }, + { + "epoch": 0.44164383561643833, + "grad_norm": 2.349296808242798, + "learning_rate": 4.766450040617385e-06, + "loss": 2.0237, + "step": 403 + }, + { + "epoch": 0.44273972602739725, + "grad_norm": 2.2154288291931152, + "learning_rate": 4.764419171405362e-06, + "loss": 1.996, + "step": 404 + }, + { + "epoch": 0.4438356164383562, + "grad_norm": 2.6355648040771484, + "learning_rate": 4.762388302193339e-06, + "loss": 2.0345, + "step": 405 + }, + { + "epoch": 0.4449315068493151, + "grad_norm": 2.352238655090332, + "learning_rate": 4.760357432981316e-06, + "loss": 2.0219, + "step": 406 + }, + { + "epoch": 0.44602739726027396, + "grad_norm": 2.1886510848999023, + "learning_rate": 4.758326563769294e-06, + "loss": 1.9923, + "step": 407 + }, + { + "epoch": 0.4471232876712329, + "grad_norm": 2.6483092308044434, + "learning_rate": 4.756295694557271e-06, + "loss": 2.0236, + "step": 408 + }, + { + "epoch": 0.4482191780821918, + "grad_norm": 2.382389545440674, + "learning_rate": 4.754264825345248e-06, + "loss": 2.0249, + "step": 409 + }, + { + "epoch": 0.44931506849315067, + "grad_norm": 2.4824411869049072, + "learning_rate": 4.7522339561332255e-06, + "loss": 1.9571, + "step": 410 + }, + { + "epoch": 0.44931506849315067, + "eval_loss": 2.1502621173858643, + "eval_runtime": 37.1789, + "eval_samples_per_second": 26.897, + "eval_steps_per_second": 1.13, + "step": 410 + }, + { + "epoch": 0.4504109589041096, + "grad_norm": 2.325589656829834, + "learning_rate": 4.750203086921203e-06, + "loss": 1.9753, + "step": 411 + }, + { + "epoch": 0.4515068493150685, + "grad_norm": 2.3618829250335693, + "learning_rate": 4.74817221770918e-06, + "loss": 1.9397, + "step": 412 + }, + { + "epoch": 0.4526027397260274, + "grad_norm": 2.3922839164733887, + "learning_rate": 4.7461413484971565e-06, + "loss": 2.0119, + "step": 413 + }, + { + "epoch": 0.4536986301369863, + "grad_norm": 2.403475046157837, + "learning_rate": 4.744110479285135e-06, + "loss": 2.0664, + "step": 414 + }, + { + "epoch": 0.4547945205479452, + "grad_norm": 2.225860834121704, + "learning_rate": 4.742079610073111e-06, + "loss": 1.9525, + "step": 415 + }, + { + "epoch": 0.4558904109589041, + "grad_norm": 2.344074010848999, + "learning_rate": 4.740048740861089e-06, + "loss": 1.9785, + "step": 416 + }, + { + "epoch": 0.456986301369863, + "grad_norm": 2.255382776260376, + "learning_rate": 4.738017871649066e-06, + "loss": 1.9902, + "step": 417 + }, + { + "epoch": 0.45808219178082193, + "grad_norm": 2.160628318786621, + "learning_rate": 4.735987002437044e-06, + "loss": 2.0075, + "step": 418 + }, + { + "epoch": 0.4591780821917808, + "grad_norm": 2.5667307376861572, + "learning_rate": 4.73395613322502e-06, + "loss": 2.0625, + "step": 419 + }, + { + "epoch": 0.4602739726027397, + "grad_norm": 2.3771040439605713, + "learning_rate": 4.731925264012998e-06, + "loss": 2.0724, + "step": 420 + }, + { + "epoch": 0.4602739726027397, + "eval_loss": 2.138829231262207, + "eval_runtime": 37.4361, + "eval_samples_per_second": 26.712, + "eval_steps_per_second": 1.122, + "step": 420 + }, + { + "epoch": 0.46136986301369864, + "grad_norm": 2.336482048034668, + "learning_rate": 4.729894394800975e-06, + "loss": 2.0234, + "step": 421 + }, + { + "epoch": 0.4624657534246575, + "grad_norm": 2.4206933975219727, + "learning_rate": 4.727863525588953e-06, + "loss": 1.9693, + "step": 422 + }, + { + "epoch": 0.46356164383561643, + "grad_norm": 2.4845645427703857, + "learning_rate": 4.725832656376929e-06, + "loss": 1.9239, + "step": 423 + }, + { + "epoch": 0.46465753424657535, + "grad_norm": 2.5199646949768066, + "learning_rate": 4.723801787164907e-06, + "loss": 1.9801, + "step": 424 + }, + { + "epoch": 0.4657534246575342, + "grad_norm": 2.2969748973846436, + "learning_rate": 4.721770917952885e-06, + "loss": 2.0192, + "step": 425 + }, + { + "epoch": 0.46684931506849314, + "grad_norm": 2.423503875732422, + "learning_rate": 4.719740048740861e-06, + "loss": 1.9684, + "step": 426 + }, + { + "epoch": 0.46794520547945206, + "grad_norm": 2.3681061267852783, + "learning_rate": 4.717709179528839e-06, + "loss": 1.9934, + "step": 427 + }, + { + "epoch": 0.469041095890411, + "grad_norm": 2.277965784072876, + "learning_rate": 4.715678310316816e-06, + "loss": 2.0166, + "step": 428 + }, + { + "epoch": 0.47013698630136985, + "grad_norm": 2.2730283737182617, + "learning_rate": 4.713647441104794e-06, + "loss": 2.011, + "step": 429 + }, + { + "epoch": 0.4712328767123288, + "grad_norm": 2.6694860458374023, + "learning_rate": 4.71161657189277e-06, + "loss": 1.9645, + "step": 430 + }, + { + "epoch": 0.4712328767123288, + "eval_loss": 2.1322011947631836, + "eval_runtime": 37.5723, + "eval_samples_per_second": 26.615, + "eval_steps_per_second": 1.118, + "step": 430 + }, + { + "epoch": 0.4723287671232877, + "grad_norm": 2.4100632667541504, + "learning_rate": 4.709585702680748e-06, + "loss": 1.9478, + "step": 431 + }, + { + "epoch": 0.47342465753424656, + "grad_norm": 2.2779541015625, + "learning_rate": 4.707554833468725e-06, + "loss": 2.0274, + "step": 432 + }, + { + "epoch": 0.4745205479452055, + "grad_norm": 2.55690336227417, + "learning_rate": 4.705523964256703e-06, + "loss": 2.0831, + "step": 433 + }, + { + "epoch": 0.4756164383561644, + "grad_norm": 2.455623149871826, + "learning_rate": 4.703493095044679e-06, + "loss": 1.9673, + "step": 434 + }, + { + "epoch": 0.4767123287671233, + "grad_norm": 2.2613699436187744, + "learning_rate": 4.701462225832657e-06, + "loss": 1.992, + "step": 435 + }, + { + "epoch": 0.4778082191780822, + "grad_norm": 2.4467933177948, + "learning_rate": 4.699431356620634e-06, + "loss": 1.9105, + "step": 436 + }, + { + "epoch": 0.4789041095890411, + "grad_norm": 2.4194023609161377, + "learning_rate": 4.697400487408612e-06, + "loss": 2.074, + "step": 437 + }, + { + "epoch": 0.48, + "grad_norm": 2.2165699005126953, + "learning_rate": 4.695369618196588e-06, + "loss": 2.0152, + "step": 438 + }, + { + "epoch": 0.4810958904109589, + "grad_norm": 2.1709225177764893, + "learning_rate": 4.693338748984566e-06, + "loss": 1.9888, + "step": 439 + }, + { + "epoch": 0.4821917808219178, + "grad_norm": 2.2325828075408936, + "learning_rate": 4.691307879772543e-06, + "loss": 1.9838, + "step": 440 + }, + { + "epoch": 0.4821917808219178, + "eval_loss": 2.128732919692993, + "eval_runtime": 37.205, + "eval_samples_per_second": 26.878, + "eval_steps_per_second": 1.129, + "step": 440 + }, + { + "epoch": 0.4832876712328767, + "grad_norm": 2.446331739425659, + "learning_rate": 4.68927701056052e-06, + "loss": 1.9295, + "step": 441 + }, + { + "epoch": 0.4843835616438356, + "grad_norm": 2.5459415912628174, + "learning_rate": 4.6872461413484975e-06, + "loss": 2.0351, + "step": 442 + }, + { + "epoch": 0.48547945205479454, + "grad_norm": 2.2877118587493896, + "learning_rate": 4.685215272136475e-06, + "loss": 2.0, + "step": 443 + }, + { + "epoch": 0.4865753424657534, + "grad_norm": 2.346625328063965, + "learning_rate": 4.683184402924452e-06, + "loss": 1.9928, + "step": 444 + }, + { + "epoch": 0.4876712328767123, + "grad_norm": 2.268467664718628, + "learning_rate": 4.681153533712429e-06, + "loss": 1.9969, + "step": 445 + }, + { + "epoch": 0.48876712328767125, + "grad_norm": 2.3699793815612793, + "learning_rate": 4.679122664500407e-06, + "loss": 2.0636, + "step": 446 + }, + { + "epoch": 0.4898630136986301, + "grad_norm": 2.2871041297912598, + "learning_rate": 4.677091795288384e-06, + "loss": 1.9535, + "step": 447 + }, + { + "epoch": 0.49095890410958903, + "grad_norm": 2.249393939971924, + "learning_rate": 4.675060926076361e-06, + "loss": 1.9358, + "step": 448 + }, + { + "epoch": 0.49205479452054796, + "grad_norm": 2.299863815307617, + "learning_rate": 4.673030056864338e-06, + "loss": 2.0299, + "step": 449 + }, + { + "epoch": 0.4931506849315068, + "grad_norm": 2.31239652633667, + "learning_rate": 4.670999187652316e-06, + "loss": 1.9372, + "step": 450 + }, + { + "epoch": 0.4931506849315068, + "eval_loss": 2.1267426013946533, + "eval_runtime": 37.1884, + "eval_samples_per_second": 26.89, + "eval_steps_per_second": 1.129, + "step": 450 + }, + { + "epoch": 0.49424657534246574, + "grad_norm": 2.2824623584747314, + "learning_rate": 4.668968318440293e-06, + "loss": 1.9435, + "step": 451 + }, + { + "epoch": 0.49534246575342467, + "grad_norm": 2.285339593887329, + "learning_rate": 4.66693744922827e-06, + "loss": 1.9455, + "step": 452 + }, + { + "epoch": 0.4964383561643836, + "grad_norm": 2.346592903137207, + "learning_rate": 4.6649065800162475e-06, + "loss": 2.0212, + "step": 453 + }, + { + "epoch": 0.49753424657534245, + "grad_norm": 2.280669689178467, + "learning_rate": 4.662875710804225e-06, + "loss": 1.9136, + "step": 454 + }, + { + "epoch": 0.4986301369863014, + "grad_norm": 2.178434133529663, + "learning_rate": 4.660844841592202e-06, + "loss": 1.9952, + "step": 455 + }, + { + "epoch": 0.4997260273972603, + "grad_norm": 2.108492851257324, + "learning_rate": 4.658813972380179e-06, + "loss": 1.9022, + "step": 456 + }, + { + "epoch": 0.5008219178082192, + "grad_norm": 2.3862147331237793, + "learning_rate": 4.656783103168157e-06, + "loss": 2.0332, + "step": 457 + }, + { + "epoch": 0.5019178082191781, + "grad_norm": 2.2017860412597656, + "learning_rate": 4.654752233956134e-06, + "loss": 1.8998, + "step": 458 + }, + { + "epoch": 0.503013698630137, + "grad_norm": 2.222134828567505, + "learning_rate": 4.652721364744111e-06, + "loss": 1.9836, + "step": 459 + }, + { + "epoch": 0.5041095890410959, + "grad_norm": 2.2974510192871094, + "learning_rate": 4.6506904955320885e-06, + "loss": 1.9626, + "step": 460 + }, + { + "epoch": 0.5041095890410959, + "eval_loss": 2.1345443725585938, + "eval_runtime": 37.2785, + "eval_samples_per_second": 26.825, + "eval_steps_per_second": 1.127, + "step": 460 + }, + { + "epoch": 0.5052054794520547, + "grad_norm": 2.107480764389038, + "learning_rate": 4.648659626320066e-06, + "loss": 2.0061, + "step": 461 + }, + { + "epoch": 0.5063013698630137, + "grad_norm": 2.3541181087493896, + "learning_rate": 4.646628757108043e-06, + "loss": 2.0133, + "step": 462 + }, + { + "epoch": 0.5073972602739726, + "grad_norm": 2.215571880340576, + "learning_rate": 4.6445978878960194e-06, + "loss": 1.9727, + "step": 463 + }, + { + "epoch": 0.5084931506849315, + "grad_norm": 2.2263615131378174, + "learning_rate": 4.6425670186839976e-06, + "loss": 1.9477, + "step": 464 + }, + { + "epoch": 0.5095890410958904, + "grad_norm": 2.178900957107544, + "learning_rate": 4.640536149471974e-06, + "loss": 1.9842, + "step": 465 + }, + { + "epoch": 0.5106849315068493, + "grad_norm": 2.3703105449676514, + "learning_rate": 4.638505280259952e-06, + "loss": 2.0082, + "step": 466 + }, + { + "epoch": 0.5117808219178083, + "grad_norm": 2.1587533950805664, + "learning_rate": 4.6364744110479285e-06, + "loss": 2.0271, + "step": 467 + }, + { + "epoch": 0.5128767123287671, + "grad_norm": 2.1626017093658447, + "learning_rate": 4.634443541835907e-06, + "loss": 1.9858, + "step": 468 + }, + { + "epoch": 0.513972602739726, + "grad_norm": 2.326869487762451, + "learning_rate": 4.632412672623883e-06, + "loss": 2.0413, + "step": 469 + }, + { + "epoch": 0.5150684931506849, + "grad_norm": 2.266502857208252, + "learning_rate": 4.630381803411861e-06, + "loss": 1.9584, + "step": 470 + }, + { + "epoch": 0.5150684931506849, + "eval_loss": 2.1273915767669678, + "eval_runtime": 37.2571, + "eval_samples_per_second": 26.841, + "eval_steps_per_second": 1.127, + "step": 470 + }, + { + "epoch": 0.5161643835616438, + "grad_norm": 2.131348133087158, + "learning_rate": 4.628350934199838e-06, + "loss": 1.863, + "step": 471 + }, + { + "epoch": 0.5172602739726028, + "grad_norm": 2.2635257244110107, + "learning_rate": 4.626320064987816e-06, + "loss": 1.9699, + "step": 472 + }, + { + "epoch": 0.5183561643835617, + "grad_norm": 2.1303744316101074, + "learning_rate": 4.624289195775792e-06, + "loss": 2.0001, + "step": 473 + }, + { + "epoch": 0.5194520547945205, + "grad_norm": 2.1736788749694824, + "learning_rate": 4.62225832656377e-06, + "loss": 1.9684, + "step": 474 + }, + { + "epoch": 0.5205479452054794, + "grad_norm": 2.1996278762817383, + "learning_rate": 4.620227457351747e-06, + "loss": 1.9739, + "step": 475 + }, + { + "epoch": 0.5216438356164383, + "grad_norm": 2.225456953048706, + "learning_rate": 4.618196588139724e-06, + "loss": 1.9649, + "step": 476 + }, + { + "epoch": 0.5227397260273973, + "grad_norm": 2.118537187576294, + "learning_rate": 4.616165718927701e-06, + "loss": 1.9911, + "step": 477 + }, + { + "epoch": 0.5238356164383562, + "grad_norm": 2.232940435409546, + "learning_rate": 4.6141348497156786e-06, + "loss": 1.944, + "step": 478 + }, + { + "epoch": 0.5249315068493151, + "grad_norm": 2.1776034832000732, + "learning_rate": 4.612103980503656e-06, + "loss": 2.0342, + "step": 479 + }, + { + "epoch": 0.5260273972602739, + "grad_norm": 2.042645215988159, + "learning_rate": 4.610073111291633e-06, + "loss": 2.0124, + "step": 480 + }, + { + "epoch": 0.5260273972602739, + "eval_loss": 2.118508815765381, + "eval_runtime": 37.2764, + "eval_samples_per_second": 26.827, + "eval_steps_per_second": 1.127, + "step": 480 + }, + { + "epoch": 0.5271232876712328, + "grad_norm": 2.302661895751953, + "learning_rate": 4.60804224207961e-06, + "loss": 1.9874, + "step": 481 + }, + { + "epoch": 0.5282191780821918, + "grad_norm": 2.1251118183135986, + "learning_rate": 4.606011372867588e-06, + "loss": 1.9594, + "step": 482 + }, + { + "epoch": 0.5293150684931507, + "grad_norm": 2.157043933868408, + "learning_rate": 4.603980503655565e-06, + "loss": 1.9955, + "step": 483 + }, + { + "epoch": 0.5304109589041096, + "grad_norm": 2.291428804397583, + "learning_rate": 4.601949634443542e-06, + "loss": 1.9695, + "step": 484 + }, + { + "epoch": 0.5315068493150685, + "grad_norm": 2.2341537475585938, + "learning_rate": 4.5999187652315195e-06, + "loss": 1.899, + "step": 485 + }, + { + "epoch": 0.5326027397260275, + "grad_norm": 2.1881206035614014, + "learning_rate": 4.597887896019497e-06, + "loss": 1.9066, + "step": 486 + }, + { + "epoch": 0.5336986301369863, + "grad_norm": 2.3057658672332764, + "learning_rate": 4.595857026807474e-06, + "loss": 1.9646, + "step": 487 + }, + { + "epoch": 0.5347945205479452, + "grad_norm": 2.2318356037139893, + "learning_rate": 4.593826157595451e-06, + "loss": 1.9662, + "step": 488 + }, + { + "epoch": 0.5358904109589041, + "grad_norm": 2.170509099960327, + "learning_rate": 4.591795288383429e-06, + "loss": 1.9522, + "step": 489 + }, + { + "epoch": 0.536986301369863, + "grad_norm": 2.101646661758423, + "learning_rate": 4.589764419171406e-06, + "loss": 2.0315, + "step": 490 + }, + { + "epoch": 0.536986301369863, + "eval_loss": 2.113968849182129, + "eval_runtime": 37.2751, + "eval_samples_per_second": 26.828, + "eval_steps_per_second": 1.127, + "step": 490 + }, + { + "epoch": 0.538082191780822, + "grad_norm": 2.1563706398010254, + "learning_rate": 4.587733549959383e-06, + "loss": 1.9843, + "step": 491 + }, + { + "epoch": 0.5391780821917809, + "grad_norm": 2.295722246170044, + "learning_rate": 4.5857026807473604e-06, + "loss": 1.9499, + "step": 492 + }, + { + "epoch": 0.5402739726027397, + "grad_norm": 1.974424123764038, + "learning_rate": 4.583671811535338e-06, + "loss": 1.9346, + "step": 493 + }, + { + "epoch": 0.5413698630136986, + "grad_norm": 2.4402668476104736, + "learning_rate": 4.581640942323315e-06, + "loss": 1.9053, + "step": 494 + }, + { + "epoch": 0.5424657534246575, + "grad_norm": 2.269000768661499, + "learning_rate": 4.579610073111292e-06, + "loss": 2.0108, + "step": 495 + }, + { + "epoch": 0.5435616438356164, + "grad_norm": 2.1182594299316406, + "learning_rate": 4.5775792038992695e-06, + "loss": 2.0155, + "step": 496 + }, + { + "epoch": 0.5446575342465754, + "grad_norm": 2.2181780338287354, + "learning_rate": 4.575548334687247e-06, + "loss": 2.0112, + "step": 497 + }, + { + "epoch": 0.5457534246575343, + "grad_norm": 2.2806756496429443, + "learning_rate": 4.573517465475224e-06, + "loss": 2.0418, + "step": 498 + }, + { + "epoch": 0.5468493150684931, + "grad_norm": 2.2690632343292236, + "learning_rate": 4.571486596263201e-06, + "loss": 1.9457, + "step": 499 + }, + { + "epoch": 0.547945205479452, + "grad_norm": 2.361335515975952, + "learning_rate": 4.569455727051178e-06, + "loss": 1.9949, + "step": 500 + }, + { + "epoch": 0.547945205479452, + "eval_loss": 2.1165614128112793, + "eval_runtime": 37.2664, + "eval_samples_per_second": 26.834, + "eval_steps_per_second": 1.127, + "step": 500 + }, + { + "epoch": 0.549041095890411, + "grad_norm": 2.3833138942718506, + "learning_rate": 4.567424857839156e-06, + "loss": 1.9932, + "step": 501 + }, + { + "epoch": 0.5501369863013699, + "grad_norm": 2.328500509262085, + "learning_rate": 4.565393988627132e-06, + "loss": 1.9773, + "step": 502 + }, + { + "epoch": 0.5512328767123288, + "grad_norm": 2.4191694259643555, + "learning_rate": 4.5633631194151105e-06, + "loss": 2.0389, + "step": 503 + }, + { + "epoch": 0.5523287671232877, + "grad_norm": 2.4106884002685547, + "learning_rate": 4.561332250203087e-06, + "loss": 1.965, + "step": 504 + }, + { + "epoch": 0.5534246575342465, + "grad_norm": 2.1452605724334717, + "learning_rate": 4.559301380991065e-06, + "loss": 1.9713, + "step": 505 + }, + { + "epoch": 0.5545205479452054, + "grad_norm": 2.2753357887268066, + "learning_rate": 4.5572705117790414e-06, + "loss": 1.9116, + "step": 506 + }, + { + "epoch": 0.5556164383561644, + "grad_norm": 2.6925032138824463, + "learning_rate": 4.5552396425670196e-06, + "loss": 1.9593, + "step": 507 + }, + { + "epoch": 0.5567123287671233, + "grad_norm": 2.3136658668518066, + "learning_rate": 4.553208773354996e-06, + "loss": 2.0072, + "step": 508 + }, + { + "epoch": 0.5578082191780822, + "grad_norm": 2.254849672317505, + "learning_rate": 4.551177904142974e-06, + "loss": 1.9193, + "step": 509 + }, + { + "epoch": 0.5589041095890411, + "grad_norm": 2.352755546569824, + "learning_rate": 4.5491470349309505e-06, + "loss": 2.0228, + "step": 510 + }, + { + "epoch": 0.5589041095890411, + "eval_loss": 2.114323377609253, + "eval_runtime": 37.3629, + "eval_samples_per_second": 26.764, + "eval_steps_per_second": 1.124, + "step": 510 + }, + { + "epoch": 0.56, + "grad_norm": 2.31339955329895, + "learning_rate": 4.547116165718929e-06, + "loss": 1.9416, + "step": 511 + }, + { + "epoch": 0.5610958904109589, + "grad_norm": 2.191164970397949, + "learning_rate": 4.545085296506905e-06, + "loss": 1.9709, + "step": 512 + }, + { + "epoch": 0.5621917808219178, + "grad_norm": 2.193551778793335, + "learning_rate": 4.543054427294882e-06, + "loss": 1.9475, + "step": 513 + }, + { + "epoch": 0.5632876712328767, + "grad_norm": 2.222710371017456, + "learning_rate": 4.54102355808286e-06, + "loss": 1.9283, + "step": 514 + }, + { + "epoch": 0.5643835616438356, + "grad_norm": 2.078907012939453, + "learning_rate": 4.538992688870837e-06, + "loss": 1.929, + "step": 515 + }, + { + "epoch": 0.5654794520547946, + "grad_norm": 2.2043654918670654, + "learning_rate": 4.536961819658814e-06, + "loss": 1.9591, + "step": 516 + }, + { + "epoch": 0.5665753424657535, + "grad_norm": 2.192481517791748, + "learning_rate": 4.5349309504467915e-06, + "loss": 1.9618, + "step": 517 + }, + { + "epoch": 0.5676712328767123, + "grad_norm": 2.3509292602539062, + "learning_rate": 4.532900081234769e-06, + "loss": 1.9782, + "step": 518 + }, + { + "epoch": 0.5687671232876712, + "grad_norm": 2.2415924072265625, + "learning_rate": 4.530869212022746e-06, + "loss": 1.9456, + "step": 519 + }, + { + "epoch": 0.5698630136986301, + "grad_norm": 2.2135009765625, + "learning_rate": 4.528838342810723e-06, + "loss": 1.9671, + "step": 520 + }, + { + "epoch": 0.5698630136986301, + "eval_loss": 2.109738826751709, + "eval_runtime": 37.5253, + "eval_samples_per_second": 26.649, + "eval_steps_per_second": 1.119, + "step": 520 + }, + { + "epoch": 0.570958904109589, + "grad_norm": 2.3513574600219727, + "learning_rate": 4.526807473598701e-06, + "loss": 1.9625, + "step": 521 + }, + { + "epoch": 0.572054794520548, + "grad_norm": 1.9949424266815186, + "learning_rate": 4.524776604386678e-06, + "loss": 1.9621, + "step": 522 + }, + { + "epoch": 0.5731506849315069, + "grad_norm": 2.199488401412964, + "learning_rate": 4.522745735174655e-06, + "loss": 1.9389, + "step": 523 + }, + { + "epoch": 0.5742465753424657, + "grad_norm": 2.2119452953338623, + "learning_rate": 4.520714865962632e-06, + "loss": 1.9936, + "step": 524 + }, + { + "epoch": 0.5753424657534246, + "grad_norm": 2.242812156677246, + "learning_rate": 4.51868399675061e-06, + "loss": 1.9728, + "step": 525 + }, + { + "epoch": 0.5764383561643835, + "grad_norm": 2.1713850498199463, + "learning_rate": 4.516653127538587e-06, + "loss": 1.98, + "step": 526 + }, + { + "epoch": 0.5775342465753425, + "grad_norm": 2.261742353439331, + "learning_rate": 4.514622258326564e-06, + "loss": 1.9412, + "step": 527 + }, + { + "epoch": 0.5786301369863014, + "grad_norm": 2.222663402557373, + "learning_rate": 4.5125913891145415e-06, + "loss": 1.9266, + "step": 528 + }, + { + "epoch": 0.5797260273972603, + "grad_norm": 2.1108906269073486, + "learning_rate": 4.510560519902519e-06, + "loss": 1.9539, + "step": 529 + }, + { + "epoch": 0.5808219178082191, + "grad_norm": 1.9594205617904663, + "learning_rate": 4.508529650690496e-06, + "loss": 1.9842, + "step": 530 + }, + { + "epoch": 0.5808219178082191, + "eval_loss": 2.1073970794677734, + "eval_runtime": 37.6167, + "eval_samples_per_second": 26.584, + "eval_steps_per_second": 1.117, + "step": 530 + }, + { + "epoch": 0.581917808219178, + "grad_norm": 2.1339495182037354, + "learning_rate": 4.506498781478473e-06, + "loss": 1.9095, + "step": 531 + }, + { + "epoch": 0.583013698630137, + "grad_norm": 2.116248607635498, + "learning_rate": 4.504467912266451e-06, + "loss": 1.9091, + "step": 532 + }, + { + "epoch": 0.5841095890410959, + "grad_norm": 2.063962459564209, + "learning_rate": 4.502437043054428e-06, + "loss": 1.9537, + "step": 533 + }, + { + "epoch": 0.5852054794520548, + "grad_norm": 2.162569522857666, + "learning_rate": 4.500406173842405e-06, + "loss": 2.0253, + "step": 534 + }, + { + "epoch": 0.5863013698630137, + "grad_norm": 2.1783549785614014, + "learning_rate": 4.4983753046303824e-06, + "loss": 1.9714, + "step": 535 + }, + { + "epoch": 0.5873972602739727, + "grad_norm": 2.0791046619415283, + "learning_rate": 4.49634443541836e-06, + "loss": 1.9629, + "step": 536 + }, + { + "epoch": 0.5884931506849315, + "grad_norm": 2.3923251628875732, + "learning_rate": 4.494313566206336e-06, + "loss": 1.9445, + "step": 537 + }, + { + "epoch": 0.5895890410958904, + "grad_norm": 2.1848347187042236, + "learning_rate": 4.492282696994314e-06, + "loss": 2.007, + "step": 538 + }, + { + "epoch": 0.5906849315068493, + "grad_norm": 2.0824925899505615, + "learning_rate": 4.490251827782291e-06, + "loss": 1.9409, + "step": 539 + }, + { + "epoch": 0.5917808219178082, + "grad_norm": 2.057790756225586, + "learning_rate": 4.488220958570269e-06, + "loss": 1.9061, + "step": 540 + }, + { + "epoch": 0.5917808219178082, + "eval_loss": 2.1101415157318115, + "eval_runtime": 37.2235, + "eval_samples_per_second": 26.865, + "eval_steps_per_second": 1.128, + "step": 540 + }, + { + "epoch": 0.5928767123287672, + "grad_norm": 2.3652148246765137, + "learning_rate": 4.486190089358245e-06, + "loss": 1.9379, + "step": 541 + }, + { + "epoch": 0.5939726027397261, + "grad_norm": 2.195535659790039, + "learning_rate": 4.484159220146223e-06, + "loss": 1.8881, + "step": 542 + }, + { + "epoch": 0.5950684931506849, + "grad_norm": 2.258000135421753, + "learning_rate": 4.4821283509342e-06, + "loss": 1.9923, + "step": 543 + }, + { + "epoch": 0.5961643835616438, + "grad_norm": 2.097041606903076, + "learning_rate": 4.480097481722178e-06, + "loss": 1.9715, + "step": 544 + }, + { + "epoch": 0.5972602739726027, + "grad_norm": 2.1741156578063965, + "learning_rate": 4.478066612510154e-06, + "loss": 1.9487, + "step": 545 + }, + { + "epoch": 0.5983561643835617, + "grad_norm": 2.135338306427002, + "learning_rate": 4.4760357432981325e-06, + "loss": 1.9172, + "step": 546 + }, + { + "epoch": 0.5994520547945206, + "grad_norm": 2.3435418605804443, + "learning_rate": 4.474004874086109e-06, + "loss": 1.945, + "step": 547 + }, + { + "epoch": 0.6005479452054795, + "grad_norm": 2.2242488861083984, + "learning_rate": 4.471974004874087e-06, + "loss": 2.0717, + "step": 548 + }, + { + "epoch": 0.6016438356164383, + "grad_norm": 2.3371622562408447, + "learning_rate": 4.4699431356620635e-06, + "loss": 1.9082, + "step": 549 + }, + { + "epoch": 0.6027397260273972, + "grad_norm": 2.176064968109131, + "learning_rate": 4.467912266450041e-06, + "loss": 2.0154, + "step": 550 + }, + { + "epoch": 0.6027397260273972, + "eval_loss": 2.105236291885376, + "eval_runtime": 37.1885, + "eval_samples_per_second": 26.89, + "eval_steps_per_second": 1.129, + "step": 550 + }, + { + "epoch": 0.6038356164383561, + "grad_norm": 2.131166934967041, + "learning_rate": 4.465881397238018e-06, + "loss": 1.9575, + "step": 551 + }, + { + "epoch": 0.6049315068493151, + "grad_norm": 2.094514846801758, + "learning_rate": 4.463850528025995e-06, + "loss": 1.9544, + "step": 552 + }, + { + "epoch": 0.606027397260274, + "grad_norm": 2.0827765464782715, + "learning_rate": 4.4618196588139726e-06, + "loss": 1.9539, + "step": 553 + }, + { + "epoch": 0.6071232876712329, + "grad_norm": 2.225749969482422, + "learning_rate": 4.45978878960195e-06, + "loss": 1.983, + "step": 554 + }, + { + "epoch": 0.6082191780821918, + "grad_norm": 2.1363680362701416, + "learning_rate": 4.457757920389927e-06, + "loss": 1.9769, + "step": 555 + }, + { + "epoch": 0.6093150684931506, + "grad_norm": 2.238170623779297, + "learning_rate": 4.455727051177904e-06, + "loss": 1.9094, + "step": 556 + }, + { + "epoch": 0.6104109589041096, + "grad_norm": 2.228384494781494, + "learning_rate": 4.453696181965882e-06, + "loss": 1.9464, + "step": 557 + }, + { + "epoch": 0.6115068493150685, + "grad_norm": 2.479428768157959, + "learning_rate": 4.451665312753859e-06, + "loss": 2.0017, + "step": 558 + }, + { + "epoch": 0.6126027397260274, + "grad_norm": 2.2331902980804443, + "learning_rate": 4.449634443541836e-06, + "loss": 1.9728, + "step": 559 + }, + { + "epoch": 0.6136986301369863, + "grad_norm": 2.1728556156158447, + "learning_rate": 4.4476035743298135e-06, + "loss": 2.0153, + "step": 560 + }, + { + "epoch": 0.6136986301369863, + "eval_loss": 2.0978519916534424, + "eval_runtime": 37.233, + "eval_samples_per_second": 26.858, + "eval_steps_per_second": 1.128, + "step": 560 + }, + { + "epoch": 0.6147945205479453, + "grad_norm": 2.377814292907715, + "learning_rate": 4.445572705117791e-06, + "loss": 1.9242, + "step": 561 + }, + { + "epoch": 0.6158904109589041, + "grad_norm": 2.190436601638794, + "learning_rate": 4.443541835905768e-06, + "loss": 1.9454, + "step": 562 + }, + { + "epoch": 0.616986301369863, + "grad_norm": 2.052201986312866, + "learning_rate": 4.441510966693745e-06, + "loss": 2.0048, + "step": 563 + }, + { + "epoch": 0.6180821917808219, + "grad_norm": 2.2706589698791504, + "learning_rate": 4.439480097481723e-06, + "loss": 1.9426, + "step": 564 + }, + { + "epoch": 0.6191780821917808, + "grad_norm": 2.3478105068206787, + "learning_rate": 4.4374492282697e-06, + "loss": 2.0047, + "step": 565 + }, + { + "epoch": 0.6202739726027398, + "grad_norm": 2.465078115463257, + "learning_rate": 4.435418359057677e-06, + "loss": 1.8739, + "step": 566 + }, + { + "epoch": 0.6213698630136987, + "grad_norm": 2.0979490280151367, + "learning_rate": 4.433387489845654e-06, + "loss": 1.9588, + "step": 567 + }, + { + "epoch": 0.6224657534246575, + "grad_norm": 2.1239588260650635, + "learning_rate": 4.431356620633632e-06, + "loss": 1.9695, + "step": 568 + }, + { + "epoch": 0.6235616438356164, + "grad_norm": 2.014007806777954, + "learning_rate": 4.429325751421609e-06, + "loss": 1.9232, + "step": 569 + }, + { + "epoch": 0.6246575342465753, + "grad_norm": 2.212414026260376, + "learning_rate": 4.427294882209586e-06, + "loss": 1.928, + "step": 570 + }, + { + "epoch": 0.6246575342465753, + "eval_loss": 2.103149890899658, + "eval_runtime": 37.2641, + "eval_samples_per_second": 26.835, + "eval_steps_per_second": 1.127, + "step": 570 + }, + { + "epoch": 0.6257534246575343, + "grad_norm": 2.0944411754608154, + "learning_rate": 4.4252640129975635e-06, + "loss": 1.9568, + "step": 571 + }, + { + "epoch": 0.6268493150684932, + "grad_norm": 2.1068055629730225, + "learning_rate": 4.423233143785541e-06, + "loss": 1.8917, + "step": 572 + }, + { + "epoch": 0.6279452054794521, + "grad_norm": 2.1992151737213135, + "learning_rate": 4.421202274573518e-06, + "loss": 1.9536, + "step": 573 + }, + { + "epoch": 0.6290410958904109, + "grad_norm": 2.2079834938049316, + "learning_rate": 4.4191714053614945e-06, + "loss": 1.9603, + "step": 574 + }, + { + "epoch": 0.6301369863013698, + "grad_norm": 2.1230316162109375, + "learning_rate": 4.417140536149473e-06, + "loss": 1.8529, + "step": 575 + }, + { + "epoch": 0.6312328767123287, + "grad_norm": 2.1069579124450684, + "learning_rate": 4.415109666937449e-06, + "loss": 1.8997, + "step": 576 + }, + { + "epoch": 0.6323287671232877, + "grad_norm": 2.1683409214019775, + "learning_rate": 4.413078797725427e-06, + "loss": 1.971, + "step": 577 + }, + { + "epoch": 0.6334246575342466, + "grad_norm": 2.0599024295806885, + "learning_rate": 4.411047928513404e-06, + "loss": 1.9154, + "step": 578 + }, + { + "epoch": 0.6345205479452055, + "grad_norm": 2.1605849266052246, + "learning_rate": 4.409017059301382e-06, + "loss": 1.9562, + "step": 579 + }, + { + "epoch": 0.6356164383561644, + "grad_norm": 2.151249408721924, + "learning_rate": 4.406986190089358e-06, + "loss": 1.9571, + "step": 580 + }, + { + "epoch": 0.6356164383561644, + "eval_loss": 2.098177671432495, + "eval_runtime": 37.2578, + "eval_samples_per_second": 26.84, + "eval_steps_per_second": 1.127, + "step": 580 + }, + { + "epoch": 0.6367123287671232, + "grad_norm": 2.1819236278533936, + "learning_rate": 4.404955320877336e-06, + "loss": 1.9816, + "step": 581 + }, + { + "epoch": 0.6378082191780822, + "grad_norm": 2.1131205558776855, + "learning_rate": 4.402924451665313e-06, + "loss": 1.9738, + "step": 582 + }, + { + "epoch": 0.6389041095890411, + "grad_norm": 2.0852456092834473, + "learning_rate": 4.400893582453291e-06, + "loss": 1.924, + "step": 583 + }, + { + "epoch": 0.64, + "grad_norm": 2.0997941493988037, + "learning_rate": 4.398862713241267e-06, + "loss": 1.9067, + "step": 584 + }, + { + "epoch": 0.6410958904109589, + "grad_norm": 2.0848538875579834, + "learning_rate": 4.396831844029245e-06, + "loss": 1.9393, + "step": 585 + }, + { + "epoch": 0.6421917808219179, + "grad_norm": 2.2121758460998535, + "learning_rate": 4.394800974817222e-06, + "loss": 1.9458, + "step": 586 + }, + { + "epoch": 0.6432876712328767, + "grad_norm": 2.194462299346924, + "learning_rate": 4.392770105605199e-06, + "loss": 1.9528, + "step": 587 + }, + { + "epoch": 0.6443835616438356, + "grad_norm": 2.1936254501342773, + "learning_rate": 4.390739236393176e-06, + "loss": 1.9584, + "step": 588 + }, + { + "epoch": 0.6454794520547945, + "grad_norm": 2.1685783863067627, + "learning_rate": 4.388708367181154e-06, + "loss": 1.9666, + "step": 589 + }, + { + "epoch": 0.6465753424657534, + "grad_norm": 2.334571123123169, + "learning_rate": 4.386677497969131e-06, + "loss": 1.9748, + "step": 590 + }, + { + "epoch": 0.6465753424657534, + "eval_loss": 2.0937106609344482, + "eval_runtime": 37.2877, + "eval_samples_per_second": 26.818, + "eval_steps_per_second": 1.126, + "step": 590 + }, + { + "epoch": 0.6476712328767124, + "grad_norm": 2.2901110649108887, + "learning_rate": 4.384646628757108e-06, + "loss": 1.9597, + "step": 591 + }, + { + "epoch": 0.6487671232876713, + "grad_norm": 2.2364838123321533, + "learning_rate": 4.3826157595450855e-06, + "loss": 2.004, + "step": 592 + }, + { + "epoch": 0.6498630136986301, + "grad_norm": 2.1088364124298096, + "learning_rate": 4.380584890333063e-06, + "loss": 1.9319, + "step": 593 + }, + { + "epoch": 0.650958904109589, + "grad_norm": 2.164210319519043, + "learning_rate": 4.37855402112104e-06, + "loss": 1.8906, + "step": 594 + }, + { + "epoch": 0.6520547945205479, + "grad_norm": 2.205200672149658, + "learning_rate": 4.376523151909017e-06, + "loss": 1.8474, + "step": 595 + }, + { + "epoch": 0.6531506849315069, + "grad_norm": 2.239530086517334, + "learning_rate": 4.3744922826969946e-06, + "loss": 2.0033, + "step": 596 + }, + { + "epoch": 0.6542465753424658, + "grad_norm": 2.214747190475464, + "learning_rate": 4.372461413484972e-06, + "loss": 1.9258, + "step": 597 + }, + { + "epoch": 0.6553424657534247, + "grad_norm": 2.1746792793273926, + "learning_rate": 4.370430544272949e-06, + "loss": 1.9479, + "step": 598 + }, + { + "epoch": 0.6564383561643835, + "grad_norm": 2.194131851196289, + "learning_rate": 4.368399675060926e-06, + "loss": 1.9825, + "step": 599 + }, + { + "epoch": 0.6575342465753424, + "grad_norm": 2.3334717750549316, + "learning_rate": 4.366368805848904e-06, + "loss": 1.9831, + "step": 600 + }, + { + "epoch": 0.6575342465753424, + "eval_loss": 2.0881216526031494, + "eval_runtime": 37.3226, + "eval_samples_per_second": 26.793, + "eval_steps_per_second": 1.125, + "step": 600 + }, + { + "epoch": 0.6586301369863014, + "grad_norm": 2.269484758377075, + "learning_rate": 4.364337936636881e-06, + "loss": 1.9407, + "step": 601 + }, + { + "epoch": 0.6597260273972603, + "grad_norm": 2.470773935317993, + "learning_rate": 4.362307067424858e-06, + "loss": 1.8905, + "step": 602 + }, + { + "epoch": 0.6608219178082192, + "grad_norm": 2.485574960708618, + "learning_rate": 4.3602761982128355e-06, + "loss": 1.953, + "step": 603 + }, + { + "epoch": 0.6619178082191781, + "grad_norm": 2.2443618774414062, + "learning_rate": 4.358245329000813e-06, + "loss": 1.9244, + "step": 604 + }, + { + "epoch": 0.663013698630137, + "grad_norm": 2.039600372314453, + "learning_rate": 4.35621445978879e-06, + "loss": 1.9364, + "step": 605 + }, + { + "epoch": 0.6641095890410958, + "grad_norm": 2.093001127243042, + "learning_rate": 4.354183590576767e-06, + "loss": 1.9439, + "step": 606 + }, + { + "epoch": 0.6652054794520548, + "grad_norm": 2.3433260917663574, + "learning_rate": 4.352152721364745e-06, + "loss": 1.9322, + "step": 607 + }, + { + "epoch": 0.6663013698630137, + "grad_norm": 2.2403812408447266, + "learning_rate": 4.350121852152722e-06, + "loss": 1.8829, + "step": 608 + }, + { + "epoch": 0.6673972602739726, + "grad_norm": 2.1542811393737793, + "learning_rate": 4.348090982940699e-06, + "loss": 1.9549, + "step": 609 + }, + { + "epoch": 0.6684931506849315, + "grad_norm": 2.1526548862457275, + "learning_rate": 4.3460601137286764e-06, + "loss": 1.9433, + "step": 610 + }, + { + "epoch": 0.6684931506849315, + "eval_loss": 2.083638906478882, + "eval_runtime": 37.2677, + "eval_samples_per_second": 26.833, + "eval_steps_per_second": 1.127, + "step": 610 + }, + { + "epoch": 0.6695890410958905, + "grad_norm": 2.093421459197998, + "learning_rate": 4.344029244516654e-06, + "loss": 1.9643, + "step": 611 + }, + { + "epoch": 0.6706849315068493, + "grad_norm": 2.2205088138580322, + "learning_rate": 4.341998375304631e-06, + "loss": 1.9309, + "step": 612 + }, + { + "epoch": 0.6717808219178082, + "grad_norm": 2.059763193130493, + "learning_rate": 4.339967506092607e-06, + "loss": 2.0064, + "step": 613 + }, + { + "epoch": 0.6728767123287671, + "grad_norm": 2.0340707302093506, + "learning_rate": 4.3379366368805855e-06, + "loss": 1.9394, + "step": 614 + }, + { + "epoch": 0.673972602739726, + "grad_norm": 2.0716137886047363, + "learning_rate": 4.335905767668562e-06, + "loss": 1.9421, + "step": 615 + }, + { + "epoch": 0.675068493150685, + "grad_norm": 2.176809549331665, + "learning_rate": 4.33387489845654e-06, + "loss": 1.9261, + "step": 616 + }, + { + "epoch": 0.6761643835616439, + "grad_norm": 2.144385814666748, + "learning_rate": 4.3318440292445165e-06, + "loss": 1.9635, + "step": 617 + }, + { + "epoch": 0.6772602739726027, + "grad_norm": 2.0949580669403076, + "learning_rate": 4.329813160032495e-06, + "loss": 1.9417, + "step": 618 + }, + { + "epoch": 0.6783561643835616, + "grad_norm": 2.1727259159088135, + "learning_rate": 4.327782290820471e-06, + "loss": 1.9041, + "step": 619 + }, + { + "epoch": 0.6794520547945205, + "grad_norm": 2.012868881225586, + "learning_rate": 4.325751421608449e-06, + "loss": 1.8973, + "step": 620 + }, + { + "epoch": 0.6794520547945205, + "eval_loss": 2.0839755535125732, + "eval_runtime": 37.5302, + "eval_samples_per_second": 26.645, + "eval_steps_per_second": 1.119, + "step": 620 + }, + { + "epoch": 0.6805479452054795, + "grad_norm": 2.204153299331665, + "learning_rate": 4.323720552396426e-06, + "loss": 1.9719, + "step": 621 + }, + { + "epoch": 0.6816438356164384, + "grad_norm": 2.2421770095825195, + "learning_rate": 4.321689683184404e-06, + "loss": 1.9207, + "step": 622 + }, + { + "epoch": 0.6827397260273973, + "grad_norm": 2.2263410091400146, + "learning_rate": 4.31965881397238e-06, + "loss": 1.9146, + "step": 623 + }, + { + "epoch": 0.6838356164383562, + "grad_norm": 2.2015221118927, + "learning_rate": 4.3176279447603574e-06, + "loss": 1.9311, + "step": 624 + }, + { + "epoch": 0.684931506849315, + "grad_norm": 2.1459643840789795, + "learning_rate": 4.315597075548335e-06, + "loss": 1.9673, + "step": 625 + }, + { + "epoch": 0.686027397260274, + "grad_norm": 2.278928756713867, + "learning_rate": 4.313566206336312e-06, + "loss": 1.954, + "step": 626 + }, + { + "epoch": 0.6871232876712329, + "grad_norm": 2.05305552482605, + "learning_rate": 4.311535337124289e-06, + "loss": 1.9789, + "step": 627 + }, + { + "epoch": 0.6882191780821918, + "grad_norm": 2.3979289531707764, + "learning_rate": 4.3095044679122665e-06, + "loss": 1.926, + "step": 628 + }, + { + "epoch": 0.6893150684931507, + "grad_norm": 2.1635828018188477, + "learning_rate": 4.307473598700244e-06, + "loss": 1.9316, + "step": 629 + }, + { + "epoch": 0.6904109589041096, + "grad_norm": 2.237713098526001, + "learning_rate": 4.305442729488221e-06, + "loss": 1.9409, + "step": 630 + }, + { + "epoch": 0.6904109589041096, + "eval_loss": 2.084226131439209, + "eval_runtime": 37.6039, + "eval_samples_per_second": 26.593, + "eval_steps_per_second": 1.117, + "step": 630 + }, + { + "epoch": 0.6915068493150685, + "grad_norm": 2.157435894012451, + "learning_rate": 4.303411860276198e-06, + "loss": 1.8777, + "step": 631 + }, + { + "epoch": 0.6926027397260274, + "grad_norm": 2.20766019821167, + "learning_rate": 4.301380991064176e-06, + "loss": 1.9803, + "step": 632 + }, + { + "epoch": 0.6936986301369863, + "grad_norm": 2.173079490661621, + "learning_rate": 4.299350121852153e-06, + "loss": 1.9541, + "step": 633 + }, + { + "epoch": 0.6947945205479452, + "grad_norm": 1.9727106094360352, + "learning_rate": 4.29731925264013e-06, + "loss": 1.9766, + "step": 634 + }, + { + "epoch": 0.6958904109589041, + "grad_norm": 2.268538475036621, + "learning_rate": 4.2952883834281075e-06, + "loss": 1.9737, + "step": 635 + }, + { + "epoch": 0.6969863013698631, + "grad_norm": 2.2118406295776367, + "learning_rate": 4.293257514216085e-06, + "loss": 1.9291, + "step": 636 + }, + { + "epoch": 0.6980821917808219, + "grad_norm": 2.2939369678497314, + "learning_rate": 4.291226645004062e-06, + "loss": 1.8868, + "step": 637 + }, + { + "epoch": 0.6991780821917808, + "grad_norm": 2.4940083026885986, + "learning_rate": 4.289195775792039e-06, + "loss": 1.9694, + "step": 638 + }, + { + "epoch": 0.7002739726027397, + "grad_norm": 2.1988351345062256, + "learning_rate": 4.287164906580017e-06, + "loss": 1.9781, + "step": 639 + }, + { + "epoch": 0.7013698630136986, + "grad_norm": 2.099578619003296, + "learning_rate": 4.285134037367994e-06, + "loss": 1.9145, + "step": 640 + }, + { + "epoch": 0.7013698630136986, + "eval_loss": 2.07913875579834, + "eval_runtime": 37.3458, + "eval_samples_per_second": 26.777, + "eval_steps_per_second": 1.125, + "step": 640 + }, + { + "epoch": 0.7024657534246576, + "grad_norm": 2.2437307834625244, + "learning_rate": 4.283103168155971e-06, + "loss": 1.9075, + "step": 641 + }, + { + "epoch": 0.7035616438356165, + "grad_norm": 2.320880174636841, + "learning_rate": 4.281072298943948e-06, + "loss": 1.8929, + "step": 642 + }, + { + "epoch": 0.7046575342465753, + "grad_norm": 2.1174306869506836, + "learning_rate": 4.279041429731926e-06, + "loss": 1.8872, + "step": 643 + }, + { + "epoch": 0.7057534246575342, + "grad_norm": 2.2826361656188965, + "learning_rate": 4.277010560519903e-06, + "loss": 1.9287, + "step": 644 + }, + { + "epoch": 0.7068493150684931, + "grad_norm": 2.7543559074401855, + "learning_rate": 4.27497969130788e-06, + "loss": 1.9119, + "step": 645 + }, + { + "epoch": 0.707945205479452, + "grad_norm": 2.2582201957702637, + "learning_rate": 4.2729488220958575e-06, + "loss": 1.9118, + "step": 646 + }, + { + "epoch": 0.709041095890411, + "grad_norm": 2.3121371269226074, + "learning_rate": 4.270917952883835e-06, + "loss": 1.959, + "step": 647 + }, + { + "epoch": 0.7101369863013699, + "grad_norm": 2.419102430343628, + "learning_rate": 4.268887083671812e-06, + "loss": 1.9679, + "step": 648 + }, + { + "epoch": 0.7112328767123288, + "grad_norm": 2.285771369934082, + "learning_rate": 4.266856214459789e-06, + "loss": 1.8939, + "step": 649 + }, + { + "epoch": 0.7123287671232876, + "grad_norm": 2.193603754043579, + "learning_rate": 4.264825345247766e-06, + "loss": 2.0097, + "step": 650 + }, + { + "epoch": 0.7123287671232876, + "eval_loss": 2.0785164833068848, + "eval_runtime": 37.2595, + "eval_samples_per_second": 26.839, + "eval_steps_per_second": 1.127, + "step": 650 + }, + { + "epoch": 0.7134246575342466, + "grad_norm": 2.2645511627197266, + "learning_rate": 4.262794476035744e-06, + "loss": 1.8865, + "step": 651 + }, + { + "epoch": 0.7145205479452055, + "grad_norm": 2.5553910732269287, + "learning_rate": 4.26076360682372e-06, + "loss": 1.9331, + "step": 652 + }, + { + "epoch": 0.7156164383561644, + "grad_norm": 2.2770845890045166, + "learning_rate": 4.2587327376116984e-06, + "loss": 1.847, + "step": 653 + }, + { + "epoch": 0.7167123287671233, + "grad_norm": 2.1486358642578125, + "learning_rate": 4.256701868399675e-06, + "loss": 1.9058, + "step": 654 + }, + { + "epoch": 0.7178082191780822, + "grad_norm": 2.168975353240967, + "learning_rate": 4.254670999187653e-06, + "loss": 1.9408, + "step": 655 + }, + { + "epoch": 0.718904109589041, + "grad_norm": 2.2156167030334473, + "learning_rate": 4.252640129975629e-06, + "loss": 1.932, + "step": 656 + }, + { + "epoch": 0.72, + "grad_norm": 2.148214101791382, + "learning_rate": 4.2506092607636075e-06, + "loss": 1.9327, + "step": 657 + }, + { + "epoch": 0.7210958904109589, + "grad_norm": 2.3860998153686523, + "learning_rate": 4.248578391551584e-06, + "loss": 1.9191, + "step": 658 + }, + { + "epoch": 0.7221917808219178, + "grad_norm": 2.3917558193206787, + "learning_rate": 4.246547522339562e-06, + "loss": 1.8607, + "step": 659 + }, + { + "epoch": 0.7232876712328767, + "grad_norm": 2.3024795055389404, + "learning_rate": 4.2445166531275385e-06, + "loss": 1.9083, + "step": 660 + }, + { + "epoch": 0.7232876712328767, + "eval_loss": 2.07985782623291, + "eval_runtime": 37.1583, + "eval_samples_per_second": 26.912, + "eval_steps_per_second": 1.13, + "step": 660 + }, + { + "epoch": 0.7243835616438357, + "grad_norm": 2.261615037918091, + "learning_rate": 4.242485783915517e-06, + "loss": 1.9155, + "step": 661 + }, + { + "epoch": 0.7254794520547945, + "grad_norm": 2.190124034881592, + "learning_rate": 4.240454914703493e-06, + "loss": 1.858, + "step": 662 + }, + { + "epoch": 0.7265753424657534, + "grad_norm": 2.3396036624908447, + "learning_rate": 4.23842404549147e-06, + "loss": 1.878, + "step": 663 + }, + { + "epoch": 0.7276712328767123, + "grad_norm": 2.198410987854004, + "learning_rate": 4.236393176279448e-06, + "loss": 1.8755, + "step": 664 + }, + { + "epoch": 0.7287671232876712, + "grad_norm": 2.108875274658203, + "learning_rate": 4.234362307067425e-06, + "loss": 1.8709, + "step": 665 + }, + { + "epoch": 0.7298630136986302, + "grad_norm": 2.0908193588256836, + "learning_rate": 4.232331437855402e-06, + "loss": 1.9395, + "step": 666 + }, + { + "epoch": 0.7309589041095891, + "grad_norm": 2.208134651184082, + "learning_rate": 4.2303005686433795e-06, + "loss": 1.9675, + "step": 667 + }, + { + "epoch": 0.7320547945205479, + "grad_norm": 2.2779526710510254, + "learning_rate": 4.228269699431357e-06, + "loss": 1.9661, + "step": 668 + }, + { + "epoch": 0.7331506849315068, + "grad_norm": 2.053279161453247, + "learning_rate": 4.226238830219334e-06, + "loss": 1.9244, + "step": 669 + }, + { + "epoch": 0.7342465753424657, + "grad_norm": 2.2060821056365967, + "learning_rate": 4.224207961007311e-06, + "loss": 1.9252, + "step": 670 + }, + { + "epoch": 0.7342465753424657, + "eval_loss": 2.0804648399353027, + "eval_runtime": 37.1852, + "eval_samples_per_second": 26.892, + "eval_steps_per_second": 1.129, + "step": 670 + }, + { + "epoch": 0.7353424657534247, + "grad_norm": 2.1996090412139893, + "learning_rate": 4.2221770917952886e-06, + "loss": 1.926, + "step": 671 + }, + { + "epoch": 0.7364383561643836, + "grad_norm": 2.0817034244537354, + "learning_rate": 4.220146222583266e-06, + "loss": 1.9123, + "step": 672 + }, + { + "epoch": 0.7375342465753425, + "grad_norm": 2.2505035400390625, + "learning_rate": 4.218115353371243e-06, + "loss": 1.9369, + "step": 673 + }, + { + "epoch": 0.7386301369863014, + "grad_norm": 2.2053000926971436, + "learning_rate": 4.21608448415922e-06, + "loss": 1.9225, + "step": 674 + }, + { + "epoch": 0.7397260273972602, + "grad_norm": 2.039905071258545, + "learning_rate": 4.214053614947198e-06, + "loss": 1.8876, + "step": 675 + }, + { + "epoch": 0.7408219178082192, + "grad_norm": 2.047445297241211, + "learning_rate": 4.212022745735175e-06, + "loss": 1.9724, + "step": 676 + }, + { + "epoch": 0.7419178082191781, + "grad_norm": 2.277803897857666, + "learning_rate": 4.209991876523152e-06, + "loss": 1.8687, + "step": 677 + }, + { + "epoch": 0.743013698630137, + "grad_norm": 2.0048105716705322, + "learning_rate": 4.2079610073111295e-06, + "loss": 1.9349, + "step": 678 + }, + { + "epoch": 0.7441095890410959, + "grad_norm": 2.121614456176758, + "learning_rate": 4.205930138099107e-06, + "loss": 1.8723, + "step": 679 + }, + { + "epoch": 0.7452054794520548, + "grad_norm": 2.169165849685669, + "learning_rate": 4.203899268887084e-06, + "loss": 1.9713, + "step": 680 + }, + { + "epoch": 0.7452054794520548, + "eval_loss": 2.0704824924468994, + "eval_runtime": 37.2015, + "eval_samples_per_second": 26.881, + "eval_steps_per_second": 1.129, + "step": 680 + }, + { + "epoch": 0.7463013698630137, + "grad_norm": 2.004974603652954, + "learning_rate": 4.201868399675061e-06, + "loss": 1.9237, + "step": 681 + }, + { + "epoch": 0.7473972602739726, + "grad_norm": 2.042402505874634, + "learning_rate": 4.199837530463039e-06, + "loss": 1.9686, + "step": 682 + }, + { + "epoch": 0.7484931506849315, + "grad_norm": 2.152040719985962, + "learning_rate": 4.197806661251016e-06, + "loss": 1.8498, + "step": 683 + }, + { + "epoch": 0.7495890410958904, + "grad_norm": 2.116483211517334, + "learning_rate": 4.195775792038993e-06, + "loss": 1.9241, + "step": 684 + }, + { + "epoch": 0.7506849315068493, + "grad_norm": 2.1628384590148926, + "learning_rate": 4.19374492282697e-06, + "loss": 1.9019, + "step": 685 + }, + { + "epoch": 0.7517808219178083, + "grad_norm": 2.1851415634155273, + "learning_rate": 4.191714053614948e-06, + "loss": 1.836, + "step": 686 + }, + { + "epoch": 0.7528767123287671, + "grad_norm": 2.173215866088867, + "learning_rate": 4.189683184402924e-06, + "loss": 1.9412, + "step": 687 + }, + { + "epoch": 0.753972602739726, + "grad_norm": 2.1002466678619385, + "learning_rate": 4.187652315190902e-06, + "loss": 1.8871, + "step": 688 + }, + { + "epoch": 0.7550684931506849, + "grad_norm": 2.1932122707366943, + "learning_rate": 4.185621445978879e-06, + "loss": 1.9671, + "step": 689 + }, + { + "epoch": 0.7561643835616438, + "grad_norm": 2.2915258407592773, + "learning_rate": 4.183590576766857e-06, + "loss": 1.8957, + "step": 690 + }, + { + "epoch": 0.7561643835616438, + "eval_loss": 2.067124128341675, + "eval_runtime": 37.2362, + "eval_samples_per_second": 26.856, + "eval_steps_per_second": 1.128, + "step": 690 + }, + { + "epoch": 0.7572602739726028, + "grad_norm": 2.060879707336426, + "learning_rate": 4.181559707554833e-06, + "loss": 1.9091, + "step": 691 + }, + { + "epoch": 0.7583561643835617, + "grad_norm": 2.031156539916992, + "learning_rate": 4.179528838342811e-06, + "loss": 1.9606, + "step": 692 + }, + { + "epoch": 0.7594520547945205, + "grad_norm": 2.121628999710083, + "learning_rate": 4.177497969130788e-06, + "loss": 1.88, + "step": 693 + }, + { + "epoch": 0.7605479452054794, + "grad_norm": 2.2009618282318115, + "learning_rate": 4.175467099918766e-06, + "loss": 1.8916, + "step": 694 + }, + { + "epoch": 0.7616438356164383, + "grad_norm": 2.124082326889038, + "learning_rate": 4.173436230706742e-06, + "loss": 1.9805, + "step": 695 + }, + { + "epoch": 0.7627397260273973, + "grad_norm": 2.3914906978607178, + "learning_rate": 4.1714053614947205e-06, + "loss": 1.9052, + "step": 696 + }, + { + "epoch": 0.7638356164383562, + "grad_norm": 2.2398853302001953, + "learning_rate": 4.169374492282697e-06, + "loss": 1.8965, + "step": 697 + }, + { + "epoch": 0.7649315068493151, + "grad_norm": 2.224860191345215, + "learning_rate": 4.167343623070675e-06, + "loss": 1.8962, + "step": 698 + }, + { + "epoch": 0.766027397260274, + "grad_norm": 2.3356244564056396, + "learning_rate": 4.165312753858652e-06, + "loss": 1.9543, + "step": 699 + }, + { + "epoch": 0.7671232876712328, + "grad_norm": 2.1308751106262207, + "learning_rate": 4.163281884646629e-06, + "loss": 1.9095, + "step": 700 + }, + { + "epoch": 0.7671232876712328, + "eval_loss": 2.066145896911621, + "eval_runtime": 37.2629, + "eval_samples_per_second": 26.836, + "eval_steps_per_second": 1.127, + "step": 700 + }, + { + "epoch": 0.7682191780821918, + "grad_norm": 2.1655938625335693, + "learning_rate": 4.161251015434607e-06, + "loss": 1.8923, + "step": 701 + }, + { + "epoch": 0.7693150684931507, + "grad_norm": 2.1985316276550293, + "learning_rate": 4.159220146222583e-06, + "loss": 1.9216, + "step": 702 + }, + { + "epoch": 0.7704109589041096, + "grad_norm": 2.1537251472473145, + "learning_rate": 4.157189277010561e-06, + "loss": 1.879, + "step": 703 + }, + { + "epoch": 0.7715068493150685, + "grad_norm": 2.1867849826812744, + "learning_rate": 4.155158407798538e-06, + "loss": 1.905, + "step": 704 + }, + { + "epoch": 0.7726027397260274, + "grad_norm": 2.342320203781128, + "learning_rate": 4.153127538586516e-06, + "loss": 1.9403, + "step": 705 + }, + { + "epoch": 0.7736986301369863, + "grad_norm": 2.2250113487243652, + "learning_rate": 4.151096669374492e-06, + "loss": 1.9109, + "step": 706 + }, + { + "epoch": 0.7747945205479452, + "grad_norm": 2.179103374481201, + "learning_rate": 4.1490658001624705e-06, + "loss": 1.9288, + "step": 707 + }, + { + "epoch": 0.7758904109589041, + "grad_norm": 2.3618645668029785, + "learning_rate": 4.147034930950447e-06, + "loss": 1.8982, + "step": 708 + }, + { + "epoch": 0.776986301369863, + "grad_norm": 2.150402307510376, + "learning_rate": 4.145004061738425e-06, + "loss": 1.9, + "step": 709 + }, + { + "epoch": 0.7780821917808219, + "grad_norm": 2.295571804046631, + "learning_rate": 4.1429731925264015e-06, + "loss": 1.9003, + "step": 710 + }, + { + "epoch": 0.7780821917808219, + "eval_loss": 2.067948341369629, + "eval_runtime": 37.2379, + "eval_samples_per_second": 26.854, + "eval_steps_per_second": 1.128, + "step": 710 + }, + { + "epoch": 0.7791780821917809, + "grad_norm": 2.3279545307159424, + "learning_rate": 4.140942323314379e-06, + "loss": 1.8307, + "step": 711 + }, + { + "epoch": 0.7802739726027397, + "grad_norm": 2.3993380069732666, + "learning_rate": 4.138911454102356e-06, + "loss": 1.8755, + "step": 712 + }, + { + "epoch": 0.7813698630136986, + "grad_norm": 2.139653444290161, + "learning_rate": 4.136880584890333e-06, + "loss": 1.8914, + "step": 713 + }, + { + "epoch": 0.7824657534246575, + "grad_norm": 2.239544153213501, + "learning_rate": 4.1348497156783106e-06, + "loss": 1.9289, + "step": 714 + }, + { + "epoch": 0.7835616438356164, + "grad_norm": 2.2039642333984375, + "learning_rate": 4.132818846466288e-06, + "loss": 1.9237, + "step": 715 + }, + { + "epoch": 0.7846575342465754, + "grad_norm": 2.1453804969787598, + "learning_rate": 4.130787977254265e-06, + "loss": 1.9291, + "step": 716 + }, + { + "epoch": 0.7857534246575343, + "grad_norm": 2.2350032329559326, + "learning_rate": 4.128757108042242e-06, + "loss": 1.9413, + "step": 717 + }, + { + "epoch": 0.7868493150684932, + "grad_norm": 2.1718318462371826, + "learning_rate": 4.12672623883022e-06, + "loss": 1.9318, + "step": 718 + }, + { + "epoch": 0.787945205479452, + "grad_norm": 2.1860601902008057, + "learning_rate": 4.124695369618197e-06, + "loss": 1.9934, + "step": 719 + }, + { + "epoch": 0.7890410958904109, + "grad_norm": 2.205166816711426, + "learning_rate": 4.122664500406174e-06, + "loss": 1.9038, + "step": 720 + }, + { + "epoch": 0.7890410958904109, + "eval_loss": 2.06229829788208, + "eval_runtime": 37.6121, + "eval_samples_per_second": 26.587, + "eval_steps_per_second": 1.117, + "step": 720 + }, + { + "epoch": 0.7901369863013699, + "grad_norm": 2.2094533443450928, + "learning_rate": 4.1206336311941515e-06, + "loss": 1.906, + "step": 721 + }, + { + "epoch": 0.7912328767123288, + "grad_norm": 2.178891658782959, + "learning_rate": 4.118602761982129e-06, + "loss": 1.936, + "step": 722 + }, + { + "epoch": 0.7923287671232877, + "grad_norm": 2.051077365875244, + "learning_rate": 4.116571892770106e-06, + "loss": 1.8736, + "step": 723 + }, + { + "epoch": 0.7934246575342466, + "grad_norm": 2.2430427074432373, + "learning_rate": 4.114541023558083e-06, + "loss": 1.8396, + "step": 724 + }, + { + "epoch": 0.7945205479452054, + "grad_norm": 2.2255303859710693, + "learning_rate": 4.112510154346061e-06, + "loss": 1.9668, + "step": 725 + }, + { + "epoch": 0.7956164383561644, + "grad_norm": 2.280313491821289, + "learning_rate": 4.110479285134038e-06, + "loss": 1.8945, + "step": 726 + }, + { + "epoch": 0.7967123287671233, + "grad_norm": 2.2516801357269287, + "learning_rate": 4.108448415922015e-06, + "loss": 1.9157, + "step": 727 + }, + { + "epoch": 0.7978082191780822, + "grad_norm": 2.1379809379577637, + "learning_rate": 4.1064175467099924e-06, + "loss": 1.935, + "step": 728 + }, + { + "epoch": 0.7989041095890411, + "grad_norm": 2.149425983428955, + "learning_rate": 4.10438667749797e-06, + "loss": 1.935, + "step": 729 + }, + { + "epoch": 0.8, + "grad_norm": 2.275270700454712, + "learning_rate": 4.102355808285947e-06, + "loss": 1.7889, + "step": 730 + }, + { + "epoch": 0.8, + "eval_loss": 2.0633974075317383, + "eval_runtime": 37.6895, + "eval_samples_per_second": 26.533, + "eval_steps_per_second": 1.114, + "step": 730 + }, + { + "epoch": 0.8010958904109589, + "grad_norm": 2.1550278663635254, + "learning_rate": 4.100324939073924e-06, + "loss": 1.9833, + "step": 731 + }, + { + "epoch": 0.8021917808219178, + "grad_norm": 2.3050060272216797, + "learning_rate": 4.0982940698619015e-06, + "loss": 1.8969, + "step": 732 + }, + { + "epoch": 0.8032876712328767, + "grad_norm": 2.4548778533935547, + "learning_rate": 4.096263200649879e-06, + "loss": 1.8945, + "step": 733 + }, + { + "epoch": 0.8043835616438356, + "grad_norm": 2.2611639499664307, + "learning_rate": 4.094232331437856e-06, + "loss": 1.9071, + "step": 734 + }, + { + "epoch": 0.8054794520547945, + "grad_norm": 2.1806774139404297, + "learning_rate": 4.092201462225833e-06, + "loss": 1.8787, + "step": 735 + }, + { + "epoch": 0.8065753424657535, + "grad_norm": 2.4374301433563232, + "learning_rate": 4.090170593013811e-06, + "loss": 1.88, + "step": 736 + }, + { + "epoch": 0.8076712328767123, + "grad_norm": 2.328333854675293, + "learning_rate": 4.088139723801787e-06, + "loss": 1.9178, + "step": 737 + }, + { + "epoch": 0.8087671232876712, + "grad_norm": 2.33552885055542, + "learning_rate": 4.086108854589765e-06, + "loss": 1.8587, + "step": 738 + }, + { + "epoch": 0.8098630136986301, + "grad_norm": 2.3472790718078613, + "learning_rate": 4.084077985377742e-06, + "loss": 1.877, + "step": 739 + }, + { + "epoch": 0.810958904109589, + "grad_norm": 2.307123899459839, + "learning_rate": 4.08204711616572e-06, + "loss": 1.8888, + "step": 740 + }, + { + "epoch": 0.810958904109589, + "eval_loss": 2.065826892852783, + "eval_runtime": 37.2747, + "eval_samples_per_second": 26.828, + "eval_steps_per_second": 1.127, + "step": 740 + }, + { + "epoch": 0.812054794520548, + "grad_norm": 2.3688747882843018, + "learning_rate": 4.080016246953696e-06, + "loss": 1.9571, + "step": 741 + }, + { + "epoch": 0.8131506849315069, + "grad_norm": 2.385007381439209, + "learning_rate": 4.077985377741674e-06, + "loss": 1.8657, + "step": 742 + }, + { + "epoch": 0.8142465753424658, + "grad_norm": 2.156766414642334, + "learning_rate": 4.075954508529651e-06, + "loss": 1.9421, + "step": 743 + }, + { + "epoch": 0.8153424657534246, + "grad_norm": 2.2074286937713623, + "learning_rate": 4.073923639317629e-06, + "loss": 1.9396, + "step": 744 + }, + { + "epoch": 0.8164383561643835, + "grad_norm": 2.185458183288574, + "learning_rate": 4.071892770105605e-06, + "loss": 1.9612, + "step": 745 + }, + { + "epoch": 0.8175342465753425, + "grad_norm": 2.324209451675415, + "learning_rate": 4.069861900893583e-06, + "loss": 1.892, + "step": 746 + }, + { + "epoch": 0.8186301369863014, + "grad_norm": 2.1312127113342285, + "learning_rate": 4.06783103168156e-06, + "loss": 1.8522, + "step": 747 + }, + { + "epoch": 0.8197260273972603, + "grad_norm": 2.0959460735321045, + "learning_rate": 4.065800162469538e-06, + "loss": 1.8732, + "step": 748 + }, + { + "epoch": 0.8208219178082192, + "grad_norm": 2.085519552230835, + "learning_rate": 4.063769293257514e-06, + "loss": 1.8832, + "step": 749 + }, + { + "epoch": 0.821917808219178, + "grad_norm": 2.310145854949951, + "learning_rate": 4.061738424045492e-06, + "loss": 1.919, + "step": 750 + }, + { + "epoch": 0.821917808219178, + "eval_loss": 2.0637967586517334, + "eval_runtime": 37.2246, + "eval_samples_per_second": 26.864, + "eval_steps_per_second": 1.128, + "step": 750 + }, + { + "epoch": 0.823013698630137, + "grad_norm": 2.3171160221099854, + "learning_rate": 4.059707554833469e-06, + "loss": 1.8874, + "step": 751 + }, + { + "epoch": 0.8241095890410959, + "grad_norm": 2.1065480709075928, + "learning_rate": 4.057676685621446e-06, + "loss": 1.9064, + "step": 752 + }, + { + "epoch": 0.8252054794520548, + "grad_norm": 2.0539677143096924, + "learning_rate": 4.0556458164094235e-06, + "loss": 1.9376, + "step": 753 + }, + { + "epoch": 0.8263013698630137, + "grad_norm": 2.4440360069274902, + "learning_rate": 4.053614947197401e-06, + "loss": 1.9129, + "step": 754 + }, + { + "epoch": 0.8273972602739726, + "grad_norm": 2.2324604988098145, + "learning_rate": 4.051584077985378e-06, + "loss": 1.8281, + "step": 755 + }, + { + "epoch": 0.8284931506849315, + "grad_norm": 2.262627363204956, + "learning_rate": 4.049553208773355e-06, + "loss": 1.9036, + "step": 756 + }, + { + "epoch": 0.8295890410958904, + "grad_norm": 2.2473418712615967, + "learning_rate": 4.047522339561333e-06, + "loss": 1.9455, + "step": 757 + }, + { + "epoch": 0.8306849315068493, + "grad_norm": 2.473444938659668, + "learning_rate": 4.04549147034931e-06, + "loss": 1.9003, + "step": 758 + }, + { + "epoch": 0.8317808219178082, + "grad_norm": 2.3454110622406006, + "learning_rate": 4.043460601137287e-06, + "loss": 1.8941, + "step": 759 + }, + { + "epoch": 0.8328767123287671, + "grad_norm": 2.4391231536865234, + "learning_rate": 4.041429731925264e-06, + "loss": 2.0146, + "step": 760 + }, + { + "epoch": 0.8328767123287671, + "eval_loss": 2.0618460178375244, + "eval_runtime": 37.1704, + "eval_samples_per_second": 26.903, + "eval_steps_per_second": 1.13, + "step": 760 + }, + { + "epoch": 0.8339726027397261, + "grad_norm": 2.3129498958587646, + "learning_rate": 4.039398862713242e-06, + "loss": 1.9, + "step": 761 + }, + { + "epoch": 0.8350684931506849, + "grad_norm": 2.3766934871673584, + "learning_rate": 4.037367993501219e-06, + "loss": 1.9017, + "step": 762 + }, + { + "epoch": 0.8361643835616438, + "grad_norm": 2.1698460578918457, + "learning_rate": 4.035337124289196e-06, + "loss": 1.9657, + "step": 763 + }, + { + "epoch": 0.8372602739726027, + "grad_norm": 2.384904146194458, + "learning_rate": 4.0333062550771735e-06, + "loss": 1.8867, + "step": 764 + }, + { + "epoch": 0.8383561643835616, + "grad_norm": 2.572391986846924, + "learning_rate": 4.031275385865151e-06, + "loss": 1.8561, + "step": 765 + }, + { + "epoch": 0.8394520547945206, + "grad_norm": 2.216606616973877, + "learning_rate": 4.029244516653128e-06, + "loss": 1.8926, + "step": 766 + }, + { + "epoch": 0.8405479452054795, + "grad_norm": 2.215756893157959, + "learning_rate": 4.027213647441105e-06, + "loss": 1.928, + "step": 767 + }, + { + "epoch": 0.8416438356164384, + "grad_norm": 2.2713284492492676, + "learning_rate": 4.025182778229083e-06, + "loss": 1.8286, + "step": 768 + }, + { + "epoch": 0.8427397260273972, + "grad_norm": 2.341747283935547, + "learning_rate": 4.02315190901706e-06, + "loss": 1.8789, + "step": 769 + }, + { + "epoch": 0.8438356164383561, + "grad_norm": 2.1485848426818848, + "learning_rate": 4.021121039805037e-06, + "loss": 1.8998, + "step": 770 + }, + { + "epoch": 0.8438356164383561, + "eval_loss": 2.0596234798431396, + "eval_runtime": 37.1854, + "eval_samples_per_second": 26.892, + "eval_steps_per_second": 1.129, + "step": 770 + }, + { + "epoch": 0.8449315068493151, + "grad_norm": 2.234192132949829, + "learning_rate": 4.0190901705930144e-06, + "loss": 1.8995, + "step": 771 + }, + { + "epoch": 0.846027397260274, + "grad_norm": 2.289328098297119, + "learning_rate": 4.017059301380992e-06, + "loss": 1.9882, + "step": 772 + }, + { + "epoch": 0.8471232876712329, + "grad_norm": 2.1001455783843994, + "learning_rate": 4.015028432168969e-06, + "loss": 1.8706, + "step": 773 + }, + { + "epoch": 0.8482191780821918, + "grad_norm": 2.2200522422790527, + "learning_rate": 4.012997562956945e-06, + "loss": 1.9443, + "step": 774 + }, + { + "epoch": 0.8493150684931506, + "grad_norm": 2.2540805339813232, + "learning_rate": 4.0109666937449235e-06, + "loss": 1.8552, + "step": 775 + }, + { + "epoch": 0.8504109589041096, + "grad_norm": 2.1214053630828857, + "learning_rate": 4.0089358245329e-06, + "loss": 1.8741, + "step": 776 + }, + { + "epoch": 0.8515068493150685, + "grad_norm": 2.0221869945526123, + "learning_rate": 4.006904955320878e-06, + "loss": 1.8798, + "step": 777 + }, + { + "epoch": 0.8526027397260274, + "grad_norm": 2.217261791229248, + "learning_rate": 4.0048740861088545e-06, + "loss": 1.8542, + "step": 778 + }, + { + "epoch": 0.8536986301369863, + "grad_norm": 2.099202871322632, + "learning_rate": 4.002843216896833e-06, + "loss": 1.9342, + "step": 779 + }, + { + "epoch": 0.8547945205479452, + "grad_norm": 2.139526844024658, + "learning_rate": 4.000812347684809e-06, + "loss": 1.9252, + "step": 780 + }, + { + "epoch": 0.8547945205479452, + "eval_loss": 2.054051399230957, + "eval_runtime": 37.1125, + "eval_samples_per_second": 26.945, + "eval_steps_per_second": 1.132, + "step": 780 + }, + { + "epoch": 0.8558904109589041, + "grad_norm": 2.2231476306915283, + "learning_rate": 3.998781478472787e-06, + "loss": 1.8671, + "step": 781 + }, + { + "epoch": 0.856986301369863, + "grad_norm": 2.143979072570801, + "learning_rate": 3.996750609260764e-06, + "loss": 1.8759, + "step": 782 + }, + { + "epoch": 0.8580821917808219, + "grad_norm": 2.1353135108947754, + "learning_rate": 3.994719740048742e-06, + "loss": 1.8559, + "step": 783 + }, + { + "epoch": 0.8591780821917808, + "grad_norm": 2.222564697265625, + "learning_rate": 3.992688870836718e-06, + "loss": 1.999, + "step": 784 + }, + { + "epoch": 0.8602739726027397, + "grad_norm": 1.958031177520752, + "learning_rate": 3.990658001624696e-06, + "loss": 1.9019, + "step": 785 + }, + { + "epoch": 0.8613698630136987, + "grad_norm": 2.2048962116241455, + "learning_rate": 3.988627132412673e-06, + "loss": 1.8561, + "step": 786 + }, + { + "epoch": 0.8624657534246576, + "grad_norm": 2.150332450866699, + "learning_rate": 3.98659626320065e-06, + "loss": 1.9399, + "step": 787 + }, + { + "epoch": 0.8635616438356164, + "grad_norm": 2.182742118835449, + "learning_rate": 3.984565393988627e-06, + "loss": 1.8933, + "step": 788 + }, + { + "epoch": 0.8646575342465753, + "grad_norm": 2.19435715675354, + "learning_rate": 3.9825345247766046e-06, + "loss": 1.8918, + "step": 789 + }, + { + "epoch": 0.8657534246575342, + "grad_norm": 2.3382859230041504, + "learning_rate": 3.980503655564582e-06, + "loss": 1.8937, + "step": 790 + }, + { + "epoch": 0.8657534246575342, + "eval_loss": 2.051997423171997, + "eval_runtime": 37.1866, + "eval_samples_per_second": 26.891, + "eval_steps_per_second": 1.129, + "step": 790 + }, + { + "epoch": 0.8668493150684932, + "grad_norm": 2.2700726985931396, + "learning_rate": 3.978472786352559e-06, + "loss": 1.9148, + "step": 791 + }, + { + "epoch": 0.8679452054794521, + "grad_norm": 2.2998759746551514, + "learning_rate": 3.976441917140536e-06, + "loss": 1.9268, + "step": 792 + }, + { + "epoch": 0.869041095890411, + "grad_norm": 2.2367377281188965, + "learning_rate": 3.974411047928514e-06, + "loss": 1.901, + "step": 793 + }, + { + "epoch": 0.8701369863013698, + "grad_norm": 2.0868773460388184, + "learning_rate": 3.972380178716491e-06, + "loss": 1.9677, + "step": 794 + }, + { + "epoch": 0.8712328767123287, + "grad_norm": 2.183732509613037, + "learning_rate": 3.970349309504468e-06, + "loss": 1.9271, + "step": 795 + }, + { + "epoch": 0.8723287671232877, + "grad_norm": 2.187282085418701, + "learning_rate": 3.9683184402924455e-06, + "loss": 1.9437, + "step": 796 + }, + { + "epoch": 0.8734246575342466, + "grad_norm": 2.1906070709228516, + "learning_rate": 3.966287571080423e-06, + "loss": 1.8437, + "step": 797 + }, + { + "epoch": 0.8745205479452055, + "grad_norm": 2.16979718208313, + "learning_rate": 3.9642567018684e-06, + "loss": 1.829, + "step": 798 + }, + { + "epoch": 0.8756164383561644, + "grad_norm": 2.2387869358062744, + "learning_rate": 3.962225832656377e-06, + "loss": 1.9143, + "step": 799 + }, + { + "epoch": 0.8767123287671232, + "grad_norm": 2.208221197128296, + "learning_rate": 3.960194963444355e-06, + "loss": 1.8894, + "step": 800 + }, + { + "epoch": 0.8767123287671232, + "eval_loss": 2.047945976257324, + "eval_runtime": 37.1778, + "eval_samples_per_second": 26.898, + "eval_steps_per_second": 1.13, + "step": 800 + }, + { + "epoch": 0.8778082191780822, + "grad_norm": 2.335108518600464, + "learning_rate": 3.958164094232332e-06, + "loss": 1.8155, + "step": 801 + }, + { + "epoch": 0.8789041095890411, + "grad_norm": 2.168006181716919, + "learning_rate": 3.956133225020309e-06, + "loss": 1.8919, + "step": 802 + }, + { + "epoch": 0.88, + "grad_norm": 2.3029897212982178, + "learning_rate": 3.954102355808286e-06, + "loss": 1.8996, + "step": 803 + }, + { + "epoch": 0.8810958904109589, + "grad_norm": 2.156141519546509, + "learning_rate": 3.952071486596264e-06, + "loss": 1.8965, + "step": 804 + }, + { + "epoch": 0.8821917808219178, + "grad_norm": 2.1522068977355957, + "learning_rate": 3.950040617384241e-06, + "loss": 1.8922, + "step": 805 + }, + { + "epoch": 0.8832876712328767, + "grad_norm": 2.153430700302124, + "learning_rate": 3.948009748172218e-06, + "loss": 1.8839, + "step": 806 + }, + { + "epoch": 0.8843835616438356, + "grad_norm": 2.1891674995422363, + "learning_rate": 3.9459788789601955e-06, + "loss": 1.8756, + "step": 807 + }, + { + "epoch": 0.8854794520547945, + "grad_norm": 2.1092092990875244, + "learning_rate": 3.943948009748173e-06, + "loss": 1.9452, + "step": 808 + }, + { + "epoch": 0.8865753424657534, + "grad_norm": 2.0465025901794434, + "learning_rate": 3.94191714053615e-06, + "loss": 1.9067, + "step": 809 + }, + { + "epoch": 0.8876712328767123, + "grad_norm": 2.1150474548339844, + "learning_rate": 3.939886271324127e-06, + "loss": 1.8874, + "step": 810 + }, + { + "epoch": 0.8876712328767123, + "eval_loss": 2.046292304992676, + "eval_runtime": 37.2421, + "eval_samples_per_second": 26.851, + "eval_steps_per_second": 1.128, + "step": 810 + }, + { + "epoch": 0.8887671232876713, + "grad_norm": 2.039989948272705, + "learning_rate": 3.937855402112104e-06, + "loss": 1.9212, + "step": 811 + }, + { + "epoch": 0.8898630136986302, + "grad_norm": 2.17354679107666, + "learning_rate": 3.935824532900082e-06, + "loss": 1.8473, + "step": 812 + }, + { + "epoch": 0.890958904109589, + "grad_norm": 2.3101940155029297, + "learning_rate": 3.933793663688058e-06, + "loss": 1.8413, + "step": 813 + }, + { + "epoch": 0.8920547945205479, + "grad_norm": 2.1300928592681885, + "learning_rate": 3.9317627944760365e-06, + "loss": 1.9126, + "step": 814 + }, + { + "epoch": 0.8931506849315068, + "grad_norm": 2.2317357063293457, + "learning_rate": 3.929731925264013e-06, + "loss": 1.938, + "step": 815 + }, + { + "epoch": 0.8942465753424658, + "grad_norm": 2.1097419261932373, + "learning_rate": 3.927701056051991e-06, + "loss": 1.8426, + "step": 816 + }, + { + "epoch": 0.8953424657534247, + "grad_norm": 2.1709141731262207, + "learning_rate": 3.9256701868399674e-06, + "loss": 1.8779, + "step": 817 + }, + { + "epoch": 0.8964383561643836, + "grad_norm": 2.104193925857544, + "learning_rate": 3.9236393176279456e-06, + "loss": 1.8736, + "step": 818 + }, + { + "epoch": 0.8975342465753424, + "grad_norm": 2.094637632369995, + "learning_rate": 3.921608448415922e-06, + "loss": 1.8556, + "step": 819 + }, + { + "epoch": 0.8986301369863013, + "grad_norm": 2.217918872833252, + "learning_rate": 3.9195775792039e-06, + "loss": 1.8429, + "step": 820 + }, + { + "epoch": 0.8986301369863013, + "eval_loss": 2.046095848083496, + "eval_runtime": 37.5659, + "eval_samples_per_second": 26.62, + "eval_steps_per_second": 1.118, + "step": 820 + }, + { + "epoch": 0.8997260273972603, + "grad_norm": 2.220355272293091, + "learning_rate": 3.9175467099918765e-06, + "loss": 1.8973, + "step": 821 + }, + { + "epoch": 0.9008219178082192, + "grad_norm": 2.0766689777374268, + "learning_rate": 3.915515840779855e-06, + "loss": 1.8728, + "step": 822 + }, + { + "epoch": 0.9019178082191781, + "grad_norm": 2.2046821117401123, + "learning_rate": 3.913484971567831e-06, + "loss": 1.9211, + "step": 823 + }, + { + "epoch": 0.903013698630137, + "grad_norm": 2.286024332046509, + "learning_rate": 3.911454102355808e-06, + "loss": 1.913, + "step": 824 + }, + { + "epoch": 0.9041095890410958, + "grad_norm": 2.2495479583740234, + "learning_rate": 3.909423233143786e-06, + "loss": 1.8214, + "step": 825 + }, + { + "epoch": 0.9052054794520548, + "grad_norm": 2.1300203800201416, + "learning_rate": 3.907392363931763e-06, + "loss": 1.9033, + "step": 826 + }, + { + "epoch": 0.9063013698630137, + "grad_norm": 2.092094898223877, + "learning_rate": 3.90536149471974e-06, + "loss": 1.9007, + "step": 827 + }, + { + "epoch": 0.9073972602739726, + "grad_norm": 2.1571428775787354, + "learning_rate": 3.9033306255077175e-06, + "loss": 1.9606, + "step": 828 + }, + { + "epoch": 0.9084931506849315, + "grad_norm": 2.1052281856536865, + "learning_rate": 3.901299756295695e-06, + "loss": 1.9916, + "step": 829 + }, + { + "epoch": 0.9095890410958904, + "grad_norm": 2.113666534423828, + "learning_rate": 3.899268887083672e-06, + "loss": 2.0136, + "step": 830 + }, + { + "epoch": 0.9095890410958904, + "eval_loss": 2.0450072288513184, + "eval_runtime": 37.6746, + "eval_samples_per_second": 26.543, + "eval_steps_per_second": 1.115, + "step": 830 + }, + { + "epoch": 0.9106849315068493, + "grad_norm": 2.112591028213501, + "learning_rate": 3.897238017871649e-06, + "loss": 1.8784, + "step": 831 + }, + { + "epoch": 0.9117808219178082, + "grad_norm": 2.1584672927856445, + "learning_rate": 3.8952071486596266e-06, + "loss": 1.8876, + "step": 832 + }, + { + "epoch": 0.9128767123287671, + "grad_norm": 2.2744317054748535, + "learning_rate": 3.893176279447604e-06, + "loss": 1.8981, + "step": 833 + }, + { + "epoch": 0.913972602739726, + "grad_norm": 2.21824312210083, + "learning_rate": 3.891145410235581e-06, + "loss": 1.85, + "step": 834 + }, + { + "epoch": 0.915068493150685, + "grad_norm": 2.2203729152679443, + "learning_rate": 3.889114541023558e-06, + "loss": 1.9649, + "step": 835 + }, + { + "epoch": 0.9161643835616439, + "grad_norm": 2.1767687797546387, + "learning_rate": 3.887083671811536e-06, + "loss": 1.9044, + "step": 836 + }, + { + "epoch": 0.9172602739726028, + "grad_norm": 2.0724353790283203, + "learning_rate": 3.885052802599513e-06, + "loss": 1.8787, + "step": 837 + }, + { + "epoch": 0.9183561643835616, + "grad_norm": 2.1283740997314453, + "learning_rate": 3.88302193338749e-06, + "loss": 1.8284, + "step": 838 + }, + { + "epoch": 0.9194520547945205, + "grad_norm": 2.5087451934814453, + "learning_rate": 3.8809910641754675e-06, + "loss": 1.9546, + "step": 839 + }, + { + "epoch": 0.9205479452054794, + "grad_norm": 2.0633344650268555, + "learning_rate": 3.878960194963445e-06, + "loss": 1.8969, + "step": 840 + }, + { + "epoch": 0.9205479452054794, + "eval_loss": 2.047938346862793, + "eval_runtime": 37.2293, + "eval_samples_per_second": 26.861, + "eval_steps_per_second": 1.128, + "step": 840 + }, + { + "epoch": 0.9216438356164384, + "grad_norm": 2.178361654281616, + "learning_rate": 3.876929325751422e-06, + "loss": 1.8855, + "step": 841 + }, + { + "epoch": 0.9227397260273973, + "grad_norm": 2.3223721981048584, + "learning_rate": 3.874898456539399e-06, + "loss": 1.9363, + "step": 842 + }, + { + "epoch": 0.9238356164383562, + "grad_norm": 2.3330960273742676, + "learning_rate": 3.872867587327377e-06, + "loss": 1.8374, + "step": 843 + }, + { + "epoch": 0.924931506849315, + "grad_norm": 2.0584356784820557, + "learning_rate": 3.870836718115354e-06, + "loss": 1.911, + "step": 844 + }, + { + "epoch": 0.9260273972602739, + "grad_norm": 2.1834609508514404, + "learning_rate": 3.868805848903331e-06, + "loss": 1.8645, + "step": 845 + }, + { + "epoch": 0.9271232876712329, + "grad_norm": 2.2576122283935547, + "learning_rate": 3.8667749796913084e-06, + "loss": 1.8662, + "step": 846 + }, + { + "epoch": 0.9282191780821918, + "grad_norm": 2.1271183490753174, + "learning_rate": 3.864744110479286e-06, + "loss": 1.8935, + "step": 847 + }, + { + "epoch": 0.9293150684931507, + "grad_norm": 2.226727247238159, + "learning_rate": 3.862713241267262e-06, + "loss": 1.8969, + "step": 848 + }, + { + "epoch": 0.9304109589041096, + "grad_norm": 2.3301939964294434, + "learning_rate": 3.86068237205524e-06, + "loss": 1.9393, + "step": 849 + }, + { + "epoch": 0.9315068493150684, + "grad_norm": 2.180250883102417, + "learning_rate": 3.858651502843217e-06, + "loss": 1.9331, + "step": 850 + }, + { + "epoch": 0.9315068493150684, + "eval_loss": 2.0509681701660156, + "eval_runtime": 37.2111, + "eval_samples_per_second": 26.874, + "eval_steps_per_second": 1.129, + "step": 850 + }, + { + "epoch": 0.9326027397260274, + "grad_norm": 2.223548173904419, + "learning_rate": 3.856620633631195e-06, + "loss": 1.894, + "step": 851 + }, + { + "epoch": 0.9336986301369863, + "grad_norm": 2.1904265880584717, + "learning_rate": 3.854589764419171e-06, + "loss": 1.8265, + "step": 852 + }, + { + "epoch": 0.9347945205479452, + "grad_norm": 2.1374316215515137, + "learning_rate": 3.852558895207149e-06, + "loss": 1.8755, + "step": 853 + }, + { + "epoch": 0.9358904109589041, + "grad_norm": 2.2175488471984863, + "learning_rate": 3.850528025995126e-06, + "loss": 1.9649, + "step": 854 + }, + { + "epoch": 0.936986301369863, + "grad_norm": 2.192126989364624, + "learning_rate": 3.848497156783104e-06, + "loss": 1.919, + "step": 855 + }, + { + "epoch": 0.938082191780822, + "grad_norm": 2.219954013824463, + "learning_rate": 3.84646628757108e-06, + "loss": 1.8651, + "step": 856 + }, + { + "epoch": 0.9391780821917808, + "grad_norm": 2.2897071838378906, + "learning_rate": 3.8444354183590585e-06, + "loss": 1.8744, + "step": 857 + }, + { + "epoch": 0.9402739726027397, + "grad_norm": 2.1656205654144287, + "learning_rate": 3.842404549147035e-06, + "loss": 1.8622, + "step": 858 + }, + { + "epoch": 0.9413698630136986, + "grad_norm": 2.20072603225708, + "learning_rate": 3.840373679935013e-06, + "loss": 1.8678, + "step": 859 + }, + { + "epoch": 0.9424657534246575, + "grad_norm": 2.1474406719207764, + "learning_rate": 3.8383428107229894e-06, + "loss": 1.8462, + "step": 860 + }, + { + "epoch": 0.9424657534246575, + "eval_loss": 2.0451602935791016, + "eval_runtime": 37.1383, + "eval_samples_per_second": 26.926, + "eval_steps_per_second": 1.131, + "step": 860 + }, + { + "epoch": 0.9435616438356165, + "grad_norm": 1.9542036056518555, + "learning_rate": 3.836311941510967e-06, + "loss": 1.8739, + "step": 861 + }, + { + "epoch": 0.9446575342465754, + "grad_norm": 2.3951354026794434, + "learning_rate": 3.834281072298944e-06, + "loss": 1.8565, + "step": 862 + }, + { + "epoch": 0.9457534246575342, + "grad_norm": 2.4916961193084717, + "learning_rate": 3.832250203086921e-06, + "loss": 1.9138, + "step": 863 + }, + { + "epoch": 0.9468493150684931, + "grad_norm": 2.1650452613830566, + "learning_rate": 3.8302193338748985e-06, + "loss": 1.8499, + "step": 864 + }, + { + "epoch": 0.947945205479452, + "grad_norm": 2.2215073108673096, + "learning_rate": 3.828188464662876e-06, + "loss": 1.9376, + "step": 865 + }, + { + "epoch": 0.949041095890411, + "grad_norm": 2.3827078342437744, + "learning_rate": 3.826157595450853e-06, + "loss": 1.9726, + "step": 866 + }, + { + "epoch": 0.9501369863013699, + "grad_norm": 2.37194561958313, + "learning_rate": 3.82412672623883e-06, + "loss": 1.9009, + "step": 867 + }, + { + "epoch": 0.9512328767123288, + "grad_norm": 2.116562604904175, + "learning_rate": 3.822095857026808e-06, + "loss": 1.8342, + "step": 868 + }, + { + "epoch": 0.9523287671232876, + "grad_norm": 2.3410849571228027, + "learning_rate": 3.820064987814785e-06, + "loss": 1.9035, + "step": 869 + }, + { + "epoch": 0.9534246575342465, + "grad_norm": 2.207545280456543, + "learning_rate": 3.818034118602762e-06, + "loss": 1.8539, + "step": 870 + }, + { + "epoch": 0.9534246575342465, + "eval_loss": 2.0433826446533203, + "eval_runtime": 37.114, + "eval_samples_per_second": 26.944, + "eval_steps_per_second": 1.132, + "step": 870 + }, + { + "epoch": 0.9545205479452055, + "grad_norm": 2.1716196537017822, + "learning_rate": 3.8160032493907395e-06, + "loss": 1.8557, + "step": 871 + }, + { + "epoch": 0.9556164383561644, + "grad_norm": 2.213124990463257, + "learning_rate": 3.813972380178717e-06, + "loss": 1.8638, + "step": 872 + }, + { + "epoch": 0.9567123287671233, + "grad_norm": 2.3185672760009766, + "learning_rate": 3.811941510966694e-06, + "loss": 1.8645, + "step": 873 + }, + { + "epoch": 0.9578082191780822, + "grad_norm": 2.453956127166748, + "learning_rate": 3.809910641754671e-06, + "loss": 1.9183, + "step": 874 + }, + { + "epoch": 0.958904109589041, + "grad_norm": 2.1971592903137207, + "learning_rate": 3.8078797725426486e-06, + "loss": 1.8445, + "step": 875 + }, + { + "epoch": 0.96, + "grad_norm": 2.273247241973877, + "learning_rate": 3.8058489033306254e-06, + "loss": 1.8024, + "step": 876 + }, + { + "epoch": 0.9610958904109589, + "grad_norm": 2.1687841415405273, + "learning_rate": 3.803818034118603e-06, + "loss": 1.8931, + "step": 877 + }, + { + "epoch": 0.9621917808219178, + "grad_norm": 2.326498508453369, + "learning_rate": 3.80178716490658e-06, + "loss": 1.8587, + "step": 878 + }, + { + "epoch": 0.9632876712328767, + "grad_norm": 2.0317249298095703, + "learning_rate": 3.7997562956945577e-06, + "loss": 1.9394, + "step": 879 + }, + { + "epoch": 0.9643835616438357, + "grad_norm": 2.3171892166137695, + "learning_rate": 3.7977254264825345e-06, + "loss": 1.92, + "step": 880 + }, + { + "epoch": 0.9643835616438357, + "eval_loss": 2.0431556701660156, + "eval_runtime": 37.1174, + "eval_samples_per_second": 26.942, + "eval_steps_per_second": 1.132, + "step": 880 + }, + { + "epoch": 0.9654794520547946, + "grad_norm": 2.2215981483459473, + "learning_rate": 3.7956945572705122e-06, + "loss": 1.8658, + "step": 881 + }, + { + "epoch": 0.9665753424657534, + "grad_norm": 2.2940685749053955, + "learning_rate": 3.793663688058489e-06, + "loss": 1.8528, + "step": 882 + }, + { + "epoch": 0.9676712328767123, + "grad_norm": 2.2218081951141357, + "learning_rate": 3.791632818846467e-06, + "loss": 1.9446, + "step": 883 + }, + { + "epoch": 0.9687671232876712, + "grad_norm": 2.2608420848846436, + "learning_rate": 3.7896019496344436e-06, + "loss": 1.9433, + "step": 884 + }, + { + "epoch": 0.9698630136986301, + "grad_norm": 2.0247788429260254, + "learning_rate": 3.787571080422421e-06, + "loss": 1.9286, + "step": 885 + }, + { + "epoch": 0.9709589041095891, + "grad_norm": 2.164071798324585, + "learning_rate": 3.785540211210398e-06, + "loss": 1.9349, + "step": 886 + }, + { + "epoch": 0.972054794520548, + "grad_norm": 2.089215040206909, + "learning_rate": 3.7835093419983755e-06, + "loss": 1.9077, + "step": 887 + }, + { + "epoch": 0.9731506849315068, + "grad_norm": 2.245256185531616, + "learning_rate": 3.7814784727863527e-06, + "loss": 1.8911, + "step": 888 + }, + { + "epoch": 0.9742465753424657, + "grad_norm": 2.1227362155914307, + "learning_rate": 3.77944760357433e-06, + "loss": 1.8761, + "step": 889 + }, + { + "epoch": 0.9753424657534246, + "grad_norm": 2.1797730922698975, + "learning_rate": 3.7774167343623073e-06, + "loss": 1.9954, + "step": 890 + }, + { + "epoch": 0.9753424657534246, + "eval_loss": 2.034895896911621, + "eval_runtime": 37.1682, + "eval_samples_per_second": 26.905, + "eval_steps_per_second": 1.13, + "step": 890 + }, + { + "epoch": 0.9764383561643836, + "grad_norm": 2.171062707901001, + "learning_rate": 3.7753858651502846e-06, + "loss": 1.9287, + "step": 891 + }, + { + "epoch": 0.9775342465753425, + "grad_norm": 2.1950814723968506, + "learning_rate": 3.773354995938262e-06, + "loss": 1.8587, + "step": 892 + }, + { + "epoch": 0.9786301369863014, + "grad_norm": 2.2582995891571045, + "learning_rate": 3.771324126726239e-06, + "loss": 1.9269, + "step": 893 + }, + { + "epoch": 0.9797260273972602, + "grad_norm": 2.218113660812378, + "learning_rate": 3.7692932575142164e-06, + "loss": 1.9007, + "step": 894 + }, + { + "epoch": 0.9808219178082191, + "grad_norm": 2.187178134918213, + "learning_rate": 3.7672623883021937e-06, + "loss": 1.9326, + "step": 895 + }, + { + "epoch": 0.9819178082191781, + "grad_norm": 2.20451283454895, + "learning_rate": 3.765231519090171e-06, + "loss": 1.8644, + "step": 896 + }, + { + "epoch": 0.983013698630137, + "grad_norm": 2.1399290561676025, + "learning_rate": 3.7632006498781482e-06, + "loss": 1.7875, + "step": 897 + }, + { + "epoch": 0.9841095890410959, + "grad_norm": 2.049473524093628, + "learning_rate": 3.761169780666125e-06, + "loss": 1.8987, + "step": 898 + }, + { + "epoch": 0.9852054794520548, + "grad_norm": 2.154444456100464, + "learning_rate": 3.7591389114541028e-06, + "loss": 1.8475, + "step": 899 + }, + { + "epoch": 0.9863013698630136, + "grad_norm": 2.1840224266052246, + "learning_rate": 3.7571080422420796e-06, + "loss": 1.8966, + "step": 900 + }, + { + "epoch": 0.9863013698630136, + "eval_loss": 2.0326430797576904, + "eval_runtime": 37.1551, + "eval_samples_per_second": 26.914, + "eval_steps_per_second": 1.13, + "step": 900 + }, + { + "epoch": 0.9873972602739726, + "grad_norm": 2.0891942977905273, + "learning_rate": 3.7550771730300573e-06, + "loss": 1.9113, + "step": 901 + }, + { + "epoch": 0.9884931506849315, + "grad_norm": 2.2646753787994385, + "learning_rate": 3.753046303818034e-06, + "loss": 1.8624, + "step": 902 + }, + { + "epoch": 0.9895890410958904, + "grad_norm": 2.15173602104187, + "learning_rate": 3.751015434606012e-06, + "loss": 1.933, + "step": 903 + }, + { + "epoch": 0.9906849315068493, + "grad_norm": 2.1751456260681152, + "learning_rate": 3.7489845653939887e-06, + "loss": 1.9114, + "step": 904 + }, + { + "epoch": 0.9917808219178083, + "grad_norm": 2.161860942840576, + "learning_rate": 3.7469536961819664e-06, + "loss": 1.9508, + "step": 905 + }, + { + "epoch": 0.9928767123287672, + "grad_norm": 2.0921733379364014, + "learning_rate": 3.7449228269699433e-06, + "loss": 1.8482, + "step": 906 + }, + { + "epoch": 0.993972602739726, + "grad_norm": 2.0823893547058105, + "learning_rate": 3.742891957757921e-06, + "loss": 1.8744, + "step": 907 + }, + { + "epoch": 0.9950684931506849, + "grad_norm": 2.136939525604248, + "learning_rate": 3.740861088545898e-06, + "loss": 1.9037, + "step": 908 + }, + { + "epoch": 0.9961643835616438, + "grad_norm": 2.0600836277008057, + "learning_rate": 3.7388302193338755e-06, + "loss": 1.8922, + "step": 909 + }, + { + "epoch": 0.9972602739726028, + "grad_norm": 2.235780715942383, + "learning_rate": 3.7367993501218524e-06, + "loss": 1.9222, + "step": 910 + }, + { + "epoch": 0.9972602739726028, + "eval_loss": 2.034031629562378, + "eval_runtime": 37.2911, + "eval_samples_per_second": 26.816, + "eval_steps_per_second": 1.126, + "step": 910 + }, + { + "epoch": 0.9983561643835617, + "grad_norm": 2.124662160873413, + "learning_rate": 3.7347684809098297e-06, + "loss": 1.8584, + "step": 911 + }, + { + "epoch": 0.9994520547945206, + "grad_norm": 2.1237940788269043, + "learning_rate": 3.732737611697807e-06, + "loss": 1.8579, + "step": 912 + }, + { + "epoch": 1.0005479452054795, + "grad_norm": 2.1175124645233154, + "learning_rate": 3.7307067424857842e-06, + "loss": 1.8348, + "step": 913 + }, + { + "epoch": 1.0016438356164383, + "grad_norm": 2.349686861038208, + "learning_rate": 3.7286758732737615e-06, + "loss": 1.7822, + "step": 914 + }, + { + "epoch": 1.0027397260273974, + "grad_norm": 2.3278446197509766, + "learning_rate": 3.7266450040617388e-06, + "loss": 1.8346, + "step": 915 + }, + { + "epoch": 1.0038356164383562, + "grad_norm": 2.1633124351501465, + "learning_rate": 3.724614134849716e-06, + "loss": 1.8137, + "step": 916 + }, + { + "epoch": 1.004931506849315, + "grad_norm": 2.5075809955596924, + "learning_rate": 3.7225832656376933e-06, + "loss": 1.955, + "step": 917 + }, + { + "epoch": 1.006027397260274, + "grad_norm": 2.3370068073272705, + "learning_rate": 3.7205523964256706e-06, + "loss": 1.7985, + "step": 918 + }, + { + "epoch": 1.0071232876712328, + "grad_norm": 2.288229465484619, + "learning_rate": 3.718521527213648e-06, + "loss": 1.923, + "step": 919 + }, + { + "epoch": 1.0082191780821919, + "grad_norm": 2.1989684104919434, + "learning_rate": 3.716490658001625e-06, + "loss": 1.8876, + "step": 920 + }, + { + "epoch": 1.0082191780821919, + "eval_loss": 2.03442120552063, + "eval_runtime": 37.5659, + "eval_samples_per_second": 26.62, + "eval_steps_per_second": 1.118, + "step": 920 + }, + { + "epoch": 1.0093150684931507, + "grad_norm": 2.202122211456299, + "learning_rate": 3.7144597887896024e-06, + "loss": 1.879, + "step": 921 + }, + { + "epoch": 1.0104109589041095, + "grad_norm": 2.325223445892334, + "learning_rate": 3.7124289195775797e-06, + "loss": 1.7632, + "step": 922 + }, + { + "epoch": 1.0115068493150685, + "grad_norm": 2.2964577674865723, + "learning_rate": 3.710398050365557e-06, + "loss": 1.8983, + "step": 923 + }, + { + "epoch": 1.0126027397260273, + "grad_norm": 2.548537015914917, + "learning_rate": 3.708367181153534e-06, + "loss": 1.8666, + "step": 924 + }, + { + "epoch": 1.0136986301369864, + "grad_norm": 2.229745626449585, + "learning_rate": 3.7063363119415115e-06, + "loss": 1.8395, + "step": 925 + }, + { + "epoch": 1.0147945205479452, + "grad_norm": 2.3639235496520996, + "learning_rate": 3.7043054427294884e-06, + "loss": 1.8864, + "step": 926 + }, + { + "epoch": 1.0158904109589042, + "grad_norm": 2.3403055667877197, + "learning_rate": 3.702274573517466e-06, + "loss": 1.8385, + "step": 927 + }, + { + "epoch": 1.016986301369863, + "grad_norm": 2.2838001251220703, + "learning_rate": 3.700243704305443e-06, + "loss": 1.8135, + "step": 928 + }, + { + "epoch": 1.0180821917808218, + "grad_norm": 2.4730923175811768, + "learning_rate": 3.6982128350934206e-06, + "loss": 1.837, + "step": 929 + }, + { + "epoch": 1.0191780821917809, + "grad_norm": 2.5711073875427246, + "learning_rate": 3.6961819658813975e-06, + "loss": 1.8276, + "step": 930 + }, + { + "epoch": 1.0191780821917809, + "eval_loss": 2.0346033573150635, + "eval_runtime": 37.6741, + "eval_samples_per_second": 26.543, + "eval_steps_per_second": 1.115, + "step": 930 + }, + { + "epoch": 1.0202739726027397, + "grad_norm": 2.296823740005493, + "learning_rate": 3.694151096669375e-06, + "loss": 1.8054, + "step": 931 + }, + { + "epoch": 1.0213698630136987, + "grad_norm": 2.321107864379883, + "learning_rate": 3.692120227457352e-06, + "loss": 1.9296, + "step": 932 + }, + { + "epoch": 1.0224657534246575, + "grad_norm": 2.3478338718414307, + "learning_rate": 3.6900893582453297e-06, + "loss": 1.8503, + "step": 933 + }, + { + "epoch": 1.0235616438356165, + "grad_norm": 2.3698856830596924, + "learning_rate": 3.6880584890333066e-06, + "loss": 1.8766, + "step": 934 + }, + { + "epoch": 1.0246575342465754, + "grad_norm": 2.2751212120056152, + "learning_rate": 3.6860276198212834e-06, + "loss": 1.7575, + "step": 935 + }, + { + "epoch": 1.0257534246575342, + "grad_norm": 2.368741035461426, + "learning_rate": 3.683996750609261e-06, + "loss": 1.8555, + "step": 936 + }, + { + "epoch": 1.0268493150684932, + "grad_norm": 2.4507620334625244, + "learning_rate": 3.681965881397238e-06, + "loss": 1.8765, + "step": 937 + }, + { + "epoch": 1.027945205479452, + "grad_norm": 2.1907308101654053, + "learning_rate": 3.6799350121852157e-06, + "loss": 1.8652, + "step": 938 + }, + { + "epoch": 1.029041095890411, + "grad_norm": 2.2696914672851562, + "learning_rate": 3.6779041429731925e-06, + "loss": 1.8854, + "step": 939 + }, + { + "epoch": 1.0301369863013699, + "grad_norm": 2.3696796894073486, + "learning_rate": 3.6758732737611702e-06, + "loss": 1.8303, + "step": 940 + }, + { + "epoch": 1.0301369863013699, + "eval_loss": 2.030118703842163, + "eval_runtime": 37.1755, + "eval_samples_per_second": 26.899, + "eval_steps_per_second": 1.13, + "step": 940 + }, + { + "epoch": 1.0312328767123287, + "grad_norm": 2.440284013748169, + "learning_rate": 3.673842404549147e-06, + "loss": 1.8498, + "step": 941 + }, + { + "epoch": 1.0323287671232877, + "grad_norm": 2.250570297241211, + "learning_rate": 3.671811535337125e-06, + "loss": 1.8361, + "step": 942 + }, + { + "epoch": 1.0334246575342465, + "grad_norm": 2.5550990104675293, + "learning_rate": 3.6697806661251016e-06, + "loss": 1.8464, + "step": 943 + }, + { + "epoch": 1.0345205479452055, + "grad_norm": 2.502474784851074, + "learning_rate": 3.6677497969130793e-06, + "loss": 1.8471, + "step": 944 + }, + { + "epoch": 1.0356164383561643, + "grad_norm": 2.3815231323242188, + "learning_rate": 3.665718927701056e-06, + "loss": 1.9786, + "step": 945 + }, + { + "epoch": 1.0367123287671234, + "grad_norm": 2.3053736686706543, + "learning_rate": 3.663688058489034e-06, + "loss": 1.8625, + "step": 946 + }, + { + "epoch": 1.0378082191780822, + "grad_norm": 2.2623229026794434, + "learning_rate": 3.6616571892770107e-06, + "loss": 1.9206, + "step": 947 + }, + { + "epoch": 1.038904109589041, + "grad_norm": 2.634559392929077, + "learning_rate": 3.659626320064988e-06, + "loss": 1.8475, + "step": 948 + }, + { + "epoch": 1.04, + "grad_norm": 2.264615297317505, + "learning_rate": 3.6575954508529653e-06, + "loss": 1.7857, + "step": 949 + }, + { + "epoch": 1.0410958904109588, + "grad_norm": 2.4125282764434814, + "learning_rate": 3.6555645816409426e-06, + "loss": 1.841, + "step": 950 + }, + { + "epoch": 1.0410958904109588, + "eval_loss": 2.0278165340423584, + "eval_runtime": 37.1503, + "eval_samples_per_second": 26.918, + "eval_steps_per_second": 1.131, + "step": 950 + }, + { + "epoch": 1.0421917808219179, + "grad_norm": 2.365743637084961, + "learning_rate": 3.65353371242892e-06, + "loss": 1.8352, + "step": 951 + }, + { + "epoch": 1.0432876712328767, + "grad_norm": 2.6469883918762207, + "learning_rate": 3.651502843216897e-06, + "loss": 1.8814, + "step": 952 + }, + { + "epoch": 1.0443835616438357, + "grad_norm": 2.3646206855773926, + "learning_rate": 3.6494719740048744e-06, + "loss": 1.7282, + "step": 953 + }, + { + "epoch": 1.0454794520547945, + "grad_norm": 2.250856876373291, + "learning_rate": 3.6474411047928517e-06, + "loss": 1.8647, + "step": 954 + }, + { + "epoch": 1.0465753424657533, + "grad_norm": 2.5404837131500244, + "learning_rate": 3.645410235580829e-06, + "loss": 1.8175, + "step": 955 + }, + { + "epoch": 1.0476712328767124, + "grad_norm": 2.4388837814331055, + "learning_rate": 3.6433793663688062e-06, + "loss": 1.8643, + "step": 956 + }, + { + "epoch": 1.0487671232876712, + "grad_norm": 2.285959005355835, + "learning_rate": 3.6413484971567835e-06, + "loss": 1.8367, + "step": 957 + }, + { + "epoch": 1.0498630136986302, + "grad_norm": 2.1190242767333984, + "learning_rate": 3.6393176279447608e-06, + "loss": 1.8242, + "step": 958 + }, + { + "epoch": 1.050958904109589, + "grad_norm": 2.397578716278076, + "learning_rate": 3.637286758732738e-06, + "loss": 1.8681, + "step": 959 + }, + { + "epoch": 1.0520547945205478, + "grad_norm": 2.510599374771118, + "learning_rate": 3.6352558895207153e-06, + "loss": 1.8804, + "step": 960 + }, + { + "epoch": 1.0520547945205478, + "eval_loss": 2.0267090797424316, + "eval_runtime": 37.1226, + "eval_samples_per_second": 26.938, + "eval_steps_per_second": 1.131, + "step": 960 + }, + { + "epoch": 1.0531506849315069, + "grad_norm": 2.4743072986602783, + "learning_rate": 3.633225020308692e-06, + "loss": 1.8895, + "step": 961 + }, + { + "epoch": 1.0542465753424657, + "grad_norm": 2.3666775226593018, + "learning_rate": 3.63119415109667e-06, + "loss": 1.8836, + "step": 962 + }, + { + "epoch": 1.0553424657534247, + "grad_norm": 2.301119327545166, + "learning_rate": 3.6291632818846467e-06, + "loss": 1.8541, + "step": 963 + }, + { + "epoch": 1.0564383561643835, + "grad_norm": 2.482811212539673, + "learning_rate": 3.6271324126726244e-06, + "loss": 1.88, + "step": 964 + }, + { + "epoch": 1.0575342465753426, + "grad_norm": 2.2962276935577393, + "learning_rate": 3.6251015434606013e-06, + "loss": 1.7834, + "step": 965 + }, + { + "epoch": 1.0586301369863014, + "grad_norm": 2.365130662918091, + "learning_rate": 3.623070674248579e-06, + "loss": 1.7946, + "step": 966 + }, + { + "epoch": 1.0597260273972602, + "grad_norm": 2.4644675254821777, + "learning_rate": 3.621039805036556e-06, + "loss": 1.8379, + "step": 967 + }, + { + "epoch": 1.0608219178082192, + "grad_norm": 2.31742787361145, + "learning_rate": 3.6190089358245335e-06, + "loss": 1.8192, + "step": 968 + }, + { + "epoch": 1.061917808219178, + "grad_norm": 2.1614131927490234, + "learning_rate": 3.6169780666125104e-06, + "loss": 1.8794, + "step": 969 + }, + { + "epoch": 1.063013698630137, + "grad_norm": 2.4367032051086426, + "learning_rate": 3.614947197400488e-06, + "loss": 1.8673, + "step": 970 + }, + { + "epoch": 1.063013698630137, + "eval_loss": 2.0261292457580566, + "eval_runtime": 37.1116, + "eval_samples_per_second": 26.946, + "eval_steps_per_second": 1.132, + "step": 970 + }, + { + "epoch": 1.0641095890410959, + "grad_norm": 2.262789726257324, + "learning_rate": 3.612916328188465e-06, + "loss": 1.8808, + "step": 971 + }, + { + "epoch": 1.0652054794520547, + "grad_norm": 2.316316843032837, + "learning_rate": 3.610885458976442e-06, + "loss": 1.7803, + "step": 972 + }, + { + "epoch": 1.0663013698630137, + "grad_norm": 2.2936418056488037, + "learning_rate": 3.6088545897644195e-06, + "loss": 1.8783, + "step": 973 + }, + { + "epoch": 1.0673972602739725, + "grad_norm": 2.667069673538208, + "learning_rate": 3.6068237205523963e-06, + "loss": 1.8748, + "step": 974 + }, + { + "epoch": 1.0684931506849316, + "grad_norm": 2.31111216545105, + "learning_rate": 3.604792851340374e-06, + "loss": 1.8119, + "step": 975 + }, + { + "epoch": 1.0695890410958904, + "grad_norm": 2.268258571624756, + "learning_rate": 3.602761982128351e-06, + "loss": 1.8055, + "step": 976 + }, + { + "epoch": 1.0706849315068494, + "grad_norm": 2.307137966156006, + "learning_rate": 3.6007311129163286e-06, + "loss": 1.8976, + "step": 977 + }, + { + "epoch": 1.0717808219178082, + "grad_norm": 2.406036853790283, + "learning_rate": 3.5987002437043054e-06, + "loss": 1.8608, + "step": 978 + }, + { + "epoch": 1.072876712328767, + "grad_norm": 2.3303062915802, + "learning_rate": 3.596669374492283e-06, + "loss": 1.8038, + "step": 979 + }, + { + "epoch": 1.073972602739726, + "grad_norm": 2.2280168533325195, + "learning_rate": 3.59463850528026e-06, + "loss": 1.8342, + "step": 980 + }, + { + "epoch": 1.073972602739726, + "eval_loss": 2.0273854732513428, + "eval_runtime": 37.1546, + "eval_samples_per_second": 26.915, + "eval_steps_per_second": 1.13, + "step": 980 + }, + { + "epoch": 1.0750684931506849, + "grad_norm": 2.4218759536743164, + "learning_rate": 3.5926076360682377e-06, + "loss": 1.9104, + "step": 981 + }, + { + "epoch": 1.076164383561644, + "grad_norm": 2.2140634059906006, + "learning_rate": 3.5905767668562145e-06, + "loss": 1.7927, + "step": 982 + }, + { + "epoch": 1.0772602739726027, + "grad_norm": 2.179877996444702, + "learning_rate": 3.5885458976441922e-06, + "loss": 1.8868, + "step": 983 + }, + { + "epoch": 1.0783561643835617, + "grad_norm": 2.1988537311553955, + "learning_rate": 3.586515028432169e-06, + "loss": 1.8516, + "step": 984 + }, + { + "epoch": 1.0794520547945206, + "grad_norm": 2.1724448204040527, + "learning_rate": 3.5844841592201464e-06, + "loss": 1.8869, + "step": 985 + }, + { + "epoch": 1.0805479452054794, + "grad_norm": 2.2685718536376953, + "learning_rate": 3.5824532900081237e-06, + "loss": 1.8955, + "step": 986 + }, + { + "epoch": 1.0816438356164384, + "grad_norm": 2.39284348487854, + "learning_rate": 3.580422420796101e-06, + "loss": 1.7769, + "step": 987 + }, + { + "epoch": 1.0827397260273972, + "grad_norm": 2.3448750972747803, + "learning_rate": 3.578391551584078e-06, + "loss": 1.8417, + "step": 988 + }, + { + "epoch": 1.0838356164383562, + "grad_norm": 2.3518283367156982, + "learning_rate": 3.5763606823720555e-06, + "loss": 1.8524, + "step": 989 + }, + { + "epoch": 1.084931506849315, + "grad_norm": 2.1696958541870117, + "learning_rate": 3.5743298131600328e-06, + "loss": 1.7668, + "step": 990 + }, + { + "epoch": 1.084931506849315, + "eval_loss": 2.0244593620300293, + "eval_runtime": 37.1929, + "eval_samples_per_second": 26.887, + "eval_steps_per_second": 1.129, + "step": 990 + }, + { + "epoch": 1.086027397260274, + "grad_norm": 2.18300724029541, + "learning_rate": 3.57229894394801e-06, + "loss": 1.852, + "step": 991 + }, + { + "epoch": 1.087123287671233, + "grad_norm": 2.2075037956237793, + "learning_rate": 3.5702680747359873e-06, + "loss": 1.8799, + "step": 992 + }, + { + "epoch": 1.0882191780821917, + "grad_norm": 2.2172231674194336, + "learning_rate": 3.5682372055239646e-06, + "loss": 1.8573, + "step": 993 + }, + { + "epoch": 1.0893150684931507, + "grad_norm": 2.25169038772583, + "learning_rate": 3.566206336311942e-06, + "loss": 1.8125, + "step": 994 + }, + { + "epoch": 1.0904109589041096, + "grad_norm": 2.168386697769165, + "learning_rate": 3.564175467099919e-06, + "loss": 1.857, + "step": 995 + }, + { + "epoch": 1.0915068493150686, + "grad_norm": 2.2610065937042236, + "learning_rate": 3.5621445978878964e-06, + "loss": 1.817, + "step": 996 + }, + { + "epoch": 1.0926027397260274, + "grad_norm": 2.2561864852905273, + "learning_rate": 3.5601137286758737e-06, + "loss": 1.7754, + "step": 997 + }, + { + "epoch": 1.0936986301369862, + "grad_norm": 2.143625497817993, + "learning_rate": 3.5580828594638505e-06, + "loss": 1.8734, + "step": 998 + }, + { + "epoch": 1.0947945205479452, + "grad_norm": 2.433831214904785, + "learning_rate": 3.5560519902518282e-06, + "loss": 1.8248, + "step": 999 + }, + { + "epoch": 1.095890410958904, + "grad_norm": 2.237497091293335, + "learning_rate": 3.554021121039805e-06, + "loss": 1.8543, + "step": 1000 + }, + { + "epoch": 1.095890410958904, + "eval_loss": 2.020202159881592, + "eval_runtime": 37.2413, + "eval_samples_per_second": 26.852, + "eval_steps_per_second": 1.128, + "step": 1000 + }, + { + "epoch": 1.096986301369863, + "grad_norm": 2.253833532333374, + "learning_rate": 3.551990251827783e-06, + "loss": 1.8799, + "step": 1001 + }, + { + "epoch": 1.098082191780822, + "grad_norm": 2.205322504043579, + "learning_rate": 3.5499593826157596e-06, + "loss": 1.834, + "step": 1002 + }, + { + "epoch": 1.099178082191781, + "grad_norm": 2.3098037242889404, + "learning_rate": 3.5479285134037373e-06, + "loss": 1.8184, + "step": 1003 + }, + { + "epoch": 1.1002739726027397, + "grad_norm": 2.361888885498047, + "learning_rate": 3.545897644191714e-06, + "loss": 1.827, + "step": 1004 + }, + { + "epoch": 1.1013698630136985, + "grad_norm": 2.341947555541992, + "learning_rate": 3.543866774979692e-06, + "loss": 1.8184, + "step": 1005 + }, + { + "epoch": 1.1024657534246576, + "grad_norm": 2.2927780151367188, + "learning_rate": 3.5418359057676687e-06, + "loss": 1.805, + "step": 1006 + }, + { + "epoch": 1.1035616438356164, + "grad_norm": 2.432988166809082, + "learning_rate": 3.5398050365556464e-06, + "loss": 1.8465, + "step": 1007 + }, + { + "epoch": 1.1046575342465754, + "grad_norm": 2.428126335144043, + "learning_rate": 3.5377741673436233e-06, + "loss": 1.8115, + "step": 1008 + }, + { + "epoch": 1.1057534246575342, + "grad_norm": 2.32551646232605, + "learning_rate": 3.535743298131601e-06, + "loss": 1.8522, + "step": 1009 + }, + { + "epoch": 1.106849315068493, + "grad_norm": 2.297980308532715, + "learning_rate": 3.533712428919578e-06, + "loss": 1.8137, + "step": 1010 + }, + { + "epoch": 1.106849315068493, + "eval_loss": 2.0190494060516357, + "eval_runtime": 37.2149, + "eval_samples_per_second": 26.871, + "eval_steps_per_second": 1.129, + "step": 1010 + }, + { + "epoch": 1.107945205479452, + "grad_norm": 2.364342451095581, + "learning_rate": 3.5316815597075547e-06, + "loss": 1.8677, + "step": 1011 + }, + { + "epoch": 1.1090410958904109, + "grad_norm": 2.355008125305176, + "learning_rate": 3.5296506904955324e-06, + "loss": 1.8937, + "step": 1012 + }, + { + "epoch": 1.11013698630137, + "grad_norm": 2.2288317680358887, + "learning_rate": 3.5276198212835093e-06, + "loss": 1.8188, + "step": 1013 + }, + { + "epoch": 1.1112328767123287, + "grad_norm": 2.2801361083984375, + "learning_rate": 3.525588952071487e-06, + "loss": 1.8386, + "step": 1014 + }, + { + "epoch": 1.1123287671232878, + "grad_norm": 2.2801361083984375, + "learning_rate": 3.525588952071487e-06, + "loss": 1.8845, + "step": 1015 + }, + { + "epoch": 1.1134246575342466, + "grad_norm": 2.4301352500915527, + "learning_rate": 3.523558082859464e-06, + "loss": 1.8898, + "step": 1016 + }, + { + "epoch": 1.1145205479452054, + "grad_norm": 2.3492274284362793, + "learning_rate": 3.5215272136474415e-06, + "loss": 1.8711, + "step": 1017 + }, + { + "epoch": 1.1156164383561644, + "grad_norm": 2.1788229942321777, + "learning_rate": 3.5194963444354184e-06, + "loss": 1.8177, + "step": 1018 + }, + { + "epoch": 1.1167123287671232, + "grad_norm": 2.269767999649048, + "learning_rate": 3.517465475223396e-06, + "loss": 1.8259, + "step": 1019 + }, + { + "epoch": 1.1178082191780823, + "grad_norm": 2.2890219688415527, + "learning_rate": 3.515434606011373e-06, + "loss": 1.7873, + "step": 1020 + }, + { + "epoch": 1.1178082191780823, + "eval_loss": 2.023195266723633, + "eval_runtime": 37.5701, + "eval_samples_per_second": 26.617, + "eval_steps_per_second": 1.118, + "step": 1020 + }, + { + "epoch": 1.118904109589041, + "grad_norm": 2.5528533458709717, + "learning_rate": 3.5134037367993506e-06, + "loss": 1.8624, + "step": 1021 + }, + { + "epoch": 1.12, + "grad_norm": 2.5109457969665527, + "learning_rate": 3.5113728675873275e-06, + "loss": 1.9178, + "step": 1022 + }, + { + "epoch": 1.121095890410959, + "grad_norm": 2.288877010345459, + "learning_rate": 3.5093419983753047e-06, + "loss": 1.8604, + "step": 1023 + }, + { + "epoch": 1.1221917808219177, + "grad_norm": 2.460153818130493, + "learning_rate": 3.507311129163282e-06, + "loss": 1.8551, + "step": 1024 + }, + { + "epoch": 1.1232876712328768, + "grad_norm": 2.3520290851593018, + "learning_rate": 3.5052802599512593e-06, + "loss": 1.7818, + "step": 1025 + }, + { + "epoch": 1.1243835616438356, + "grad_norm": 2.32812237739563, + "learning_rate": 3.5032493907392366e-06, + "loss": 1.8255, + "step": 1026 + }, + { + "epoch": 1.1254794520547946, + "grad_norm": 2.3118042945861816, + "learning_rate": 3.501218521527214e-06, + "loss": 1.8493, + "step": 1027 + }, + { + "epoch": 1.1265753424657534, + "grad_norm": 2.2750277519226074, + "learning_rate": 3.499187652315191e-06, + "loss": 1.7759, + "step": 1028 + }, + { + "epoch": 1.1276712328767124, + "grad_norm": 2.272120475769043, + "learning_rate": 3.4971567831031684e-06, + "loss": 1.8399, + "step": 1029 + }, + { + "epoch": 1.1287671232876713, + "grad_norm": 2.2075254917144775, + "learning_rate": 3.4951259138911457e-06, + "loss": 1.8774, + "step": 1030 + }, + { + "epoch": 1.1287671232876713, + "eval_loss": 2.0199804306030273, + "eval_runtime": 37.6753, + "eval_samples_per_second": 26.543, + "eval_steps_per_second": 1.115, + "step": 1030 + }, + { + "epoch": 1.12986301369863, + "grad_norm": 2.359877586364746, + "learning_rate": 3.493095044679123e-06, + "loss": 1.7942, + "step": 1031 + }, + { + "epoch": 1.130958904109589, + "grad_norm": 2.3057703971862793, + "learning_rate": 3.4910641754671002e-06, + "loss": 1.8996, + "step": 1032 + }, + { + "epoch": 1.132054794520548, + "grad_norm": 2.3109569549560547, + "learning_rate": 3.4890333062550775e-06, + "loss": 1.7336, + "step": 1033 + }, + { + "epoch": 1.133150684931507, + "grad_norm": 2.3021819591522217, + "learning_rate": 3.4870024370430548e-06, + "loss": 1.869, + "step": 1034 + }, + { + "epoch": 1.1342465753424658, + "grad_norm": 2.4741852283477783, + "learning_rate": 3.484971567831032e-06, + "loss": 1.7652, + "step": 1035 + }, + { + "epoch": 1.1353424657534246, + "grad_norm": 2.326500415802002, + "learning_rate": 3.482940698619009e-06, + "loss": 1.809, + "step": 1036 + }, + { + "epoch": 1.1364383561643836, + "grad_norm": 2.2562193870544434, + "learning_rate": 3.4809098294069866e-06, + "loss": 1.8806, + "step": 1037 + }, + { + "epoch": 1.1375342465753424, + "grad_norm": 2.2849345207214355, + "learning_rate": 3.4788789601949634e-06, + "loss": 1.8329, + "step": 1038 + }, + { + "epoch": 1.1386301369863014, + "grad_norm": 2.333707332611084, + "learning_rate": 3.476848090982941e-06, + "loss": 1.8541, + "step": 1039 + }, + { + "epoch": 1.1397260273972603, + "grad_norm": 2.234175682067871, + "learning_rate": 3.474817221770918e-06, + "loss": 1.8126, + "step": 1040 + }, + { + "epoch": 1.1397260273972603, + "eval_loss": 2.0176141262054443, + "eval_runtime": 37.232, + "eval_samples_per_second": 26.859, + "eval_steps_per_second": 1.128, + "step": 1040 + }, + { + "epoch": 1.140821917808219, + "grad_norm": 2.4212377071380615, + "learning_rate": 3.4727863525588957e-06, + "loss": 1.9325, + "step": 1041 + }, + { + "epoch": 1.141917808219178, + "grad_norm": 2.1526010036468506, + "learning_rate": 3.4707554833468725e-06, + "loss": 1.8674, + "step": 1042 + }, + { + "epoch": 1.143013698630137, + "grad_norm": 2.336186408996582, + "learning_rate": 3.4687246141348502e-06, + "loss": 1.8275, + "step": 1043 + }, + { + "epoch": 1.144109589041096, + "grad_norm": 2.2991745471954346, + "learning_rate": 3.466693744922827e-06, + "loss": 1.8057, + "step": 1044 + }, + { + "epoch": 1.1452054794520548, + "grad_norm": 2.3816370964050293, + "learning_rate": 3.464662875710805e-06, + "loss": 1.8212, + "step": 1045 + }, + { + "epoch": 1.1463013698630138, + "grad_norm": 2.2383193969726562, + "learning_rate": 3.4626320064987817e-06, + "loss": 1.8444, + "step": 1046 + }, + { + "epoch": 1.1473972602739726, + "grad_norm": 2.377574920654297, + "learning_rate": 3.4606011372867594e-06, + "loss": 1.8319, + "step": 1047 + }, + { + "epoch": 1.1484931506849314, + "grad_norm": 2.1803700923919678, + "learning_rate": 3.458570268074736e-06, + "loss": 1.8484, + "step": 1048 + }, + { + "epoch": 1.1495890410958904, + "grad_norm": 2.1810641288757324, + "learning_rate": 3.4565393988627135e-06, + "loss": 1.8187, + "step": 1049 + }, + { + "epoch": 1.1506849315068493, + "grad_norm": 2.1565394401550293, + "learning_rate": 3.4545085296506908e-06, + "loss": 1.8117, + "step": 1050 + }, + { + "epoch": 1.1506849315068493, + "eval_loss": 2.0156562328338623, + "eval_runtime": 37.2072, + "eval_samples_per_second": 26.877, + "eval_steps_per_second": 1.129, + "step": 1050 + }, + { + "epoch": 1.1517808219178083, + "grad_norm": 2.1753759384155273, + "learning_rate": 3.452477660438668e-06, + "loss": 1.7822, + "step": 1051 + }, + { + "epoch": 1.152876712328767, + "grad_norm": 2.250089645385742, + "learning_rate": 3.4504467912266453e-06, + "loss": 1.8634, + "step": 1052 + }, + { + "epoch": 1.1539726027397261, + "grad_norm": 2.3043293952941895, + "learning_rate": 3.4484159220146226e-06, + "loss": 1.7743, + "step": 1053 + }, + { + "epoch": 1.155068493150685, + "grad_norm": 2.2193732261657715, + "learning_rate": 3.4463850528026e-06, + "loss": 1.8328, + "step": 1054 + }, + { + "epoch": 1.1561643835616437, + "grad_norm": 2.4204866886138916, + "learning_rate": 3.444354183590577e-06, + "loss": 1.8133, + "step": 1055 + }, + { + "epoch": 1.1572602739726028, + "grad_norm": 2.3088018894195557, + "learning_rate": 3.4423233143785544e-06, + "loss": 1.8523, + "step": 1056 + }, + { + "epoch": 1.1583561643835616, + "grad_norm": 2.38910174369812, + "learning_rate": 3.4402924451665317e-06, + "loss": 1.7846, + "step": 1057 + }, + { + "epoch": 1.1594520547945206, + "grad_norm": 2.2473702430725098, + "learning_rate": 3.438261575954509e-06, + "loss": 1.7632, + "step": 1058 + }, + { + "epoch": 1.1605479452054794, + "grad_norm": 2.3356645107269287, + "learning_rate": 3.4362307067424862e-06, + "loss": 1.8461, + "step": 1059 + }, + { + "epoch": 1.1616438356164385, + "grad_norm": 2.2460737228393555, + "learning_rate": 3.434199837530463e-06, + "loss": 1.8853, + "step": 1060 + }, + { + "epoch": 1.1616438356164385, + "eval_loss": 2.016077995300293, + "eval_runtime": 37.1943, + "eval_samples_per_second": 26.886, + "eval_steps_per_second": 1.129, + "step": 1060 + }, + { + "epoch": 1.1627397260273973, + "grad_norm": 2.28434157371521, + "learning_rate": 3.432168968318441e-06, + "loss": 1.8451, + "step": 1061 + }, + { + "epoch": 1.163835616438356, + "grad_norm": 2.299049139022827, + "learning_rate": 3.4301380991064176e-06, + "loss": 1.7777, + "step": 1062 + }, + { + "epoch": 1.1649315068493151, + "grad_norm": 2.364074468612671, + "learning_rate": 3.4281072298943953e-06, + "loss": 1.8517, + "step": 1063 + }, + { + "epoch": 1.166027397260274, + "grad_norm": 2.2381322383880615, + "learning_rate": 3.426076360682372e-06, + "loss": 1.889, + "step": 1064 + }, + { + "epoch": 1.167123287671233, + "grad_norm": 2.2984025478363037, + "learning_rate": 3.42404549147035e-06, + "loss": 1.8907, + "step": 1065 + }, + { + "epoch": 1.1682191780821918, + "grad_norm": 2.2215473651885986, + "learning_rate": 3.4220146222583267e-06, + "loss": 1.8313, + "step": 1066 + }, + { + "epoch": 1.1693150684931506, + "grad_norm": 2.3414270877838135, + "learning_rate": 3.4199837530463044e-06, + "loss": 1.783, + "step": 1067 + }, + { + "epoch": 1.1704109589041096, + "grad_norm": 2.0636470317840576, + "learning_rate": 3.4179528838342813e-06, + "loss": 1.8185, + "step": 1068 + }, + { + "epoch": 1.1715068493150684, + "grad_norm": 2.2158873081207275, + "learning_rate": 3.415922014622259e-06, + "loss": 1.9319, + "step": 1069 + }, + { + "epoch": 1.1726027397260275, + "grad_norm": 2.244051694869995, + "learning_rate": 3.413891145410236e-06, + "loss": 1.8732, + "step": 1070 + }, + { + "epoch": 1.1726027397260275, + "eval_loss": 2.0180234909057617, + "eval_runtime": 37.1822, + "eval_samples_per_second": 26.895, + "eval_steps_per_second": 1.13, + "step": 1070 + }, + { + "epoch": 1.1736986301369863, + "grad_norm": 2.1608433723449707, + "learning_rate": 3.4118602761982135e-06, + "loss": 1.8709, + "step": 1071 + }, + { + "epoch": 1.174794520547945, + "grad_norm": 2.2612338066101074, + "learning_rate": 3.4098294069861904e-06, + "loss": 1.8882, + "step": 1072 + }, + { + "epoch": 1.1758904109589041, + "grad_norm": 2.3257741928100586, + "learning_rate": 3.4077985377741673e-06, + "loss": 1.8377, + "step": 1073 + }, + { + "epoch": 1.176986301369863, + "grad_norm": 2.2486586570739746, + "learning_rate": 3.405767668562145e-06, + "loss": 1.8105, + "step": 1074 + }, + { + "epoch": 1.178082191780822, + "grad_norm": 2.2840020656585693, + "learning_rate": 3.403736799350122e-06, + "loss": 1.8868, + "step": 1075 + }, + { + "epoch": 1.1791780821917808, + "grad_norm": 2.3264284133911133, + "learning_rate": 3.4017059301380995e-06, + "loss": 1.8701, + "step": 1076 + }, + { + "epoch": 1.1802739726027398, + "grad_norm": 2.311424493789673, + "learning_rate": 3.3996750609260764e-06, + "loss": 1.801, + "step": 1077 + }, + { + "epoch": 1.1813698630136986, + "grad_norm": 2.2177181243896484, + "learning_rate": 3.397644191714054e-06, + "loss": 1.9188, + "step": 1078 + }, + { + "epoch": 1.1824657534246574, + "grad_norm": 2.413118362426758, + "learning_rate": 3.395613322502031e-06, + "loss": 1.8198, + "step": 1079 + }, + { + "epoch": 1.1835616438356165, + "grad_norm": 2.334688186645508, + "learning_rate": 3.3935824532900086e-06, + "loss": 1.822, + "step": 1080 + }, + { + "epoch": 1.1835616438356165, + "eval_loss": 2.012676954269409, + "eval_runtime": 37.2415, + "eval_samples_per_second": 26.852, + "eval_steps_per_second": 1.128, + "step": 1080 + }, + { + "epoch": 1.1846575342465753, + "grad_norm": 2.3901584148406982, + "learning_rate": 3.3915515840779855e-06, + "loss": 1.791, + "step": 1081 + }, + { + "epoch": 1.1857534246575343, + "grad_norm": 2.2731425762176514, + "learning_rate": 3.389520714865963e-06, + "loss": 1.8367, + "step": 1082 + }, + { + "epoch": 1.1868493150684931, + "grad_norm": 2.311553478240967, + "learning_rate": 3.38748984565394e-06, + "loss": 1.8302, + "step": 1083 + }, + { + "epoch": 1.1879452054794521, + "grad_norm": 2.3030776977539062, + "learning_rate": 3.3854589764419177e-06, + "loss": 1.8652, + "step": 1084 + }, + { + "epoch": 1.189041095890411, + "grad_norm": 2.341188430786133, + "learning_rate": 3.3834281072298946e-06, + "loss": 1.8094, + "step": 1085 + }, + { + "epoch": 1.1901369863013698, + "grad_norm": 2.331962823867798, + "learning_rate": 3.381397238017872e-06, + "loss": 1.827, + "step": 1086 + }, + { + "epoch": 1.1912328767123288, + "grad_norm": 2.275923728942871, + "learning_rate": 3.379366368805849e-06, + "loss": 1.7935, + "step": 1087 + }, + { + "epoch": 1.1923287671232876, + "grad_norm": 2.4608163833618164, + "learning_rate": 3.3773354995938264e-06, + "loss": 1.8965, + "step": 1088 + }, + { + "epoch": 1.1934246575342466, + "grad_norm": 2.2208592891693115, + "learning_rate": 3.3753046303818037e-06, + "loss": 1.8447, + "step": 1089 + }, + { + "epoch": 1.1945205479452055, + "grad_norm": 2.4284770488739014, + "learning_rate": 3.373273761169781e-06, + "loss": 1.8306, + "step": 1090 + }, + { + "epoch": 1.1945205479452055, + "eval_loss": 2.010664939880371, + "eval_runtime": 37.237, + "eval_samples_per_second": 26.855, + "eval_steps_per_second": 1.128, + "step": 1090 + }, + { + "epoch": 1.1956164383561645, + "grad_norm": 2.329601764678955, + "learning_rate": 3.3712428919577582e-06, + "loss": 1.8275, + "step": 1091 + }, + { + "epoch": 1.1967123287671233, + "grad_norm": 2.277939796447754, + "learning_rate": 3.3692120227457355e-06, + "loss": 1.8161, + "step": 1092 + }, + { + "epoch": 1.1978082191780821, + "grad_norm": 2.446780204772949, + "learning_rate": 3.3671811535337128e-06, + "loss": 1.8511, + "step": 1093 + }, + { + "epoch": 1.1989041095890411, + "grad_norm": 2.382477283477783, + "learning_rate": 3.36515028432169e-06, + "loss": 1.7834, + "step": 1094 + }, + { + "epoch": 1.2, + "grad_norm": 2.366361379623413, + "learning_rate": 3.3631194151096673e-06, + "loss": 1.8105, + "step": 1095 + }, + { + "epoch": 1.201095890410959, + "grad_norm": 2.3439717292785645, + "learning_rate": 3.3610885458976446e-06, + "loss": 1.8418, + "step": 1096 + }, + { + "epoch": 1.2021917808219178, + "grad_norm": 2.2739830017089844, + "learning_rate": 3.359057676685622e-06, + "loss": 1.8818, + "step": 1097 + }, + { + "epoch": 1.2032876712328768, + "grad_norm": 2.1636738777160645, + "learning_rate": 3.357026807473599e-06, + "loss": 1.8757, + "step": 1098 + }, + { + "epoch": 1.2043835616438356, + "grad_norm": 2.386441946029663, + "learning_rate": 3.354995938261576e-06, + "loss": 1.756, + "step": 1099 + }, + { + "epoch": 1.2054794520547945, + "grad_norm": 2.2241456508636475, + "learning_rate": 3.3529650690495537e-06, + "loss": 1.8034, + "step": 1100 + }, + { + "epoch": 1.2054794520547945, + "eval_loss": 2.008504867553711, + "eval_runtime": 37.2855, + "eval_samples_per_second": 26.82, + "eval_steps_per_second": 1.126, + "step": 1100 + }, + { + "epoch": 1.2065753424657535, + "grad_norm": 2.196992874145508, + "learning_rate": 3.3509341998375305e-06, + "loss": 1.7722, + "step": 1101 + }, + { + "epoch": 1.2076712328767123, + "grad_norm": 2.3572094440460205, + "learning_rate": 3.3489033306255082e-06, + "loss": 1.8109, + "step": 1102 + }, + { + "epoch": 1.208767123287671, + "grad_norm": 2.443281650543213, + "learning_rate": 3.346872461413485e-06, + "loss": 1.843, + "step": 1103 + }, + { + "epoch": 1.2098630136986301, + "grad_norm": 2.2613234519958496, + "learning_rate": 3.344841592201463e-06, + "loss": 1.8023, + "step": 1104 + }, + { + "epoch": 1.210958904109589, + "grad_norm": 2.2466726303100586, + "learning_rate": 3.3428107229894397e-06, + "loss": 1.7838, + "step": 1105 + }, + { + "epoch": 1.212054794520548, + "grad_norm": 2.2314462661743164, + "learning_rate": 3.3407798537774174e-06, + "loss": 1.8077, + "step": 1106 + }, + { + "epoch": 1.2131506849315068, + "grad_norm": 2.3288400173187256, + "learning_rate": 3.338748984565394e-06, + "loss": 1.8319, + "step": 1107 + }, + { + "epoch": 1.2142465753424658, + "grad_norm": 2.3477249145507812, + "learning_rate": 3.336718115353372e-06, + "loss": 1.8772, + "step": 1108 + }, + { + "epoch": 1.2153424657534246, + "grad_norm": 2.3588409423828125, + "learning_rate": 3.3346872461413488e-06, + "loss": 1.8832, + "step": 1109 + }, + { + "epoch": 1.2164383561643834, + "grad_norm": 2.327796220779419, + "learning_rate": 3.3326563769293256e-06, + "loss": 1.8033, + "step": 1110 + }, + { + "epoch": 1.2164383561643834, + "eval_loss": 2.0094382762908936, + "eval_runtime": 37.2666, + "eval_samples_per_second": 26.834, + "eval_steps_per_second": 1.127, + "step": 1110 + }, + { + "epoch": 1.2175342465753425, + "grad_norm": 2.257244825363159, + "learning_rate": 3.3306255077173033e-06, + "loss": 1.7903, + "step": 1111 + }, + { + "epoch": 1.2186301369863013, + "grad_norm": 2.3321614265441895, + "learning_rate": 3.32859463850528e-06, + "loss": 1.8295, + "step": 1112 + }, + { + "epoch": 1.2197260273972603, + "grad_norm": 2.4073803424835205, + "learning_rate": 3.326563769293258e-06, + "loss": 1.8395, + "step": 1113 + }, + { + "epoch": 1.2208219178082191, + "grad_norm": 2.3631434440612793, + "learning_rate": 3.3245329000812347e-06, + "loss": 1.7548, + "step": 1114 + }, + { + "epoch": 1.2219178082191782, + "grad_norm": 2.275766134262085, + "learning_rate": 3.3225020308692124e-06, + "loss": 1.808, + "step": 1115 + }, + { + "epoch": 1.223013698630137, + "grad_norm": 2.438011407852173, + "learning_rate": 3.3204711616571893e-06, + "loss": 1.8274, + "step": 1116 + }, + { + "epoch": 1.2241095890410958, + "grad_norm": 2.371826171875, + "learning_rate": 3.318440292445167e-06, + "loss": 1.7742, + "step": 1117 + }, + { + "epoch": 1.2252054794520548, + "grad_norm": 2.422004461288452, + "learning_rate": 3.316409423233144e-06, + "loss": 1.8268, + "step": 1118 + }, + { + "epoch": 1.2263013698630136, + "grad_norm": 2.377979278564453, + "learning_rate": 3.3143785540211215e-06, + "loss": 1.7857, + "step": 1119 + }, + { + "epoch": 1.2273972602739727, + "grad_norm": 2.288482904434204, + "learning_rate": 3.3123476848090984e-06, + "loss": 1.876, + "step": 1120 + }, + { + "epoch": 1.2273972602739727, + "eval_loss": 2.0104730129241943, + "eval_runtime": 37.5365, + "eval_samples_per_second": 26.641, + "eval_steps_per_second": 1.119, + "step": 1120 + }, + { + "epoch": 1.2284931506849315, + "grad_norm": 2.451533555984497, + "learning_rate": 3.310316815597076e-06, + "loss": 1.8324, + "step": 1121 + }, + { + "epoch": 1.2295890410958905, + "grad_norm": 2.3785181045532227, + "learning_rate": 3.308285946385053e-06, + "loss": 1.8239, + "step": 1122 + }, + { + "epoch": 1.2306849315068493, + "grad_norm": 2.2276721000671387, + "learning_rate": 3.30625507717303e-06, + "loss": 1.9132, + "step": 1123 + }, + { + "epoch": 1.2317808219178081, + "grad_norm": 2.2575314044952393, + "learning_rate": 3.3042242079610075e-06, + "loss": 1.769, + "step": 1124 + }, + { + "epoch": 1.2328767123287672, + "grad_norm": 2.5311717987060547, + "learning_rate": 3.3021933387489847e-06, + "loss": 1.8605, + "step": 1125 + }, + { + "epoch": 1.233972602739726, + "grad_norm": 2.2395896911621094, + "learning_rate": 3.300162469536962e-06, + "loss": 1.8725, + "step": 1126 + }, + { + "epoch": 1.235068493150685, + "grad_norm": 2.2402031421661377, + "learning_rate": 3.2981316003249393e-06, + "loss": 1.7955, + "step": 1127 + }, + { + "epoch": 1.2361643835616438, + "grad_norm": 2.308950424194336, + "learning_rate": 3.2961007311129166e-06, + "loss": 1.8812, + "step": 1128 + }, + { + "epoch": 1.2372602739726029, + "grad_norm": 2.437767267227173, + "learning_rate": 3.294069861900894e-06, + "loss": 1.8489, + "step": 1129 + }, + { + "epoch": 1.2383561643835617, + "grad_norm": 2.3188693523406982, + "learning_rate": 3.292038992688871e-06, + "loss": 1.9082, + "step": 1130 + }, + { + "epoch": 1.2383561643835617, + "eval_loss": 2.0110676288604736, + "eval_runtime": 37.6437, + "eval_samples_per_second": 26.565, + "eval_steps_per_second": 1.116, + "step": 1130 + }, + { + "epoch": 1.2394520547945205, + "grad_norm": 2.312760591506958, + "learning_rate": 3.2900081234768484e-06, + "loss": 1.8344, + "step": 1131 + }, + { + "epoch": 1.2405479452054795, + "grad_norm": 2.5837206840515137, + "learning_rate": 3.2879772542648257e-06, + "loss": 1.8503, + "step": 1132 + }, + { + "epoch": 1.2416438356164383, + "grad_norm": 2.1864733695983887, + "learning_rate": 3.285946385052803e-06, + "loss": 1.8822, + "step": 1133 + }, + { + "epoch": 1.2427397260273974, + "grad_norm": 2.2644550800323486, + "learning_rate": 3.2839155158407802e-06, + "loss": 1.7541, + "step": 1134 + }, + { + "epoch": 1.2438356164383562, + "grad_norm": 2.292646646499634, + "learning_rate": 3.2818846466287575e-06, + "loss": 1.8575, + "step": 1135 + }, + { + "epoch": 1.244931506849315, + "grad_norm": 2.2162070274353027, + "learning_rate": 3.2798537774167344e-06, + "loss": 1.7525, + "step": 1136 + }, + { + "epoch": 1.246027397260274, + "grad_norm": 2.32688307762146, + "learning_rate": 3.277822908204712e-06, + "loss": 1.8955, + "step": 1137 + }, + { + "epoch": 1.2471232876712328, + "grad_norm": 2.3226473331451416, + "learning_rate": 3.275792038992689e-06, + "loss": 1.8255, + "step": 1138 + }, + { + "epoch": 1.2482191780821918, + "grad_norm": 2.3533706665039062, + "learning_rate": 3.2737611697806666e-06, + "loss": 1.783, + "step": 1139 + }, + { + "epoch": 1.2493150684931507, + "grad_norm": 2.237661123275757, + "learning_rate": 3.2717303005686435e-06, + "loss": 1.8367, + "step": 1140 + }, + { + "epoch": 1.2493150684931507, + "eval_loss": 2.0123534202575684, + "eval_runtime": 37.3573, + "eval_samples_per_second": 26.769, + "eval_steps_per_second": 1.124, + "step": 1140 + }, + { + "epoch": 1.2504109589041095, + "grad_norm": 2.283047676086426, + "learning_rate": 3.269699431356621e-06, + "loss": 1.7987, + "step": 1141 + }, + { + "epoch": 1.2515068493150685, + "grad_norm": 2.2264456748962402, + "learning_rate": 3.267668562144598e-06, + "loss": 1.8307, + "step": 1142 + }, + { + "epoch": 1.2526027397260273, + "grad_norm": 2.2724905014038086, + "learning_rate": 3.2656376929325757e-06, + "loss": 1.8237, + "step": 1143 + }, + { + "epoch": 1.2536986301369863, + "grad_norm": 2.270782947540283, + "learning_rate": 3.2636068237205526e-06, + "loss": 1.7846, + "step": 1144 + }, + { + "epoch": 1.2547945205479452, + "grad_norm": 2.2032551765441895, + "learning_rate": 3.2615759545085303e-06, + "loss": 1.8438, + "step": 1145 + }, + { + "epoch": 1.2558904109589042, + "grad_norm": 2.199150800704956, + "learning_rate": 3.259545085296507e-06, + "loss": 1.8199, + "step": 1146 + }, + { + "epoch": 1.256986301369863, + "grad_norm": 2.4404170513153076, + "learning_rate": 3.257514216084484e-06, + "loss": 1.8817, + "step": 1147 + }, + { + "epoch": 1.2580821917808218, + "grad_norm": 2.281348943710327, + "learning_rate": 3.2554833468724617e-06, + "loss": 1.771, + "step": 1148 + }, + { + "epoch": 1.2591780821917808, + "grad_norm": 2.19482159614563, + "learning_rate": 3.2534524776604385e-06, + "loss": 1.8684, + "step": 1149 + }, + { + "epoch": 1.2602739726027397, + "grad_norm": 2.221618413925171, + "learning_rate": 3.2514216084484162e-06, + "loss": 1.8398, + "step": 1150 + }, + { + "epoch": 1.2602739726027397, + "eval_loss": 2.007969617843628, + "eval_runtime": 37.3679, + "eval_samples_per_second": 26.761, + "eval_steps_per_second": 1.124, + "step": 1150 + }, + { + "epoch": 1.2613698630136987, + "grad_norm": 2.193432569503784, + "learning_rate": 3.249390739236393e-06, + "loss": 1.8122, + "step": 1151 + }, + { + "epoch": 1.2624657534246575, + "grad_norm": 2.246462106704712, + "learning_rate": 3.2473598700243708e-06, + "loss": 1.8447, + "step": 1152 + }, + { + "epoch": 1.2635616438356165, + "grad_norm": 2.3411879539489746, + "learning_rate": 3.2453290008123476e-06, + "loss": 1.7988, + "step": 1153 + }, + { + "epoch": 1.2646575342465753, + "grad_norm": 2.245286464691162, + "learning_rate": 3.2432981316003253e-06, + "loss": 1.8209, + "step": 1154 + }, + { + "epoch": 1.2657534246575342, + "grad_norm": 2.292067527770996, + "learning_rate": 3.241267262388302e-06, + "loss": 1.8285, + "step": 1155 + }, + { + "epoch": 1.2668493150684932, + "grad_norm": 2.3058111667633057, + "learning_rate": 3.23923639317628e-06, + "loss": 1.7842, + "step": 1156 + }, + { + "epoch": 1.267945205479452, + "grad_norm": 2.306243896484375, + "learning_rate": 3.2372055239642567e-06, + "loss": 1.8221, + "step": 1157 + }, + { + "epoch": 1.269041095890411, + "grad_norm": 2.399759531021118, + "learning_rate": 3.2351746547522344e-06, + "loss": 1.7836, + "step": 1158 + }, + { + "epoch": 1.2701369863013698, + "grad_norm": 2.3405187129974365, + "learning_rate": 3.2331437855402113e-06, + "loss": 1.8365, + "step": 1159 + }, + { + "epoch": 1.2712328767123289, + "grad_norm": 2.3855013847351074, + "learning_rate": 3.2311129163281885e-06, + "loss": 1.8973, + "step": 1160 + }, + { + "epoch": 1.2712328767123289, + "eval_loss": 2.005983591079712, + "eval_runtime": 37.3061, + "eval_samples_per_second": 26.805, + "eval_steps_per_second": 1.126, + "step": 1160 + }, + { + "epoch": 1.2723287671232877, + "grad_norm": 2.2767140865325928, + "learning_rate": 3.229082047116166e-06, + "loss": 1.7791, + "step": 1161 + }, + { + "epoch": 1.2734246575342465, + "grad_norm": 2.289494276046753, + "learning_rate": 3.227051177904143e-06, + "loss": 1.8508, + "step": 1162 + }, + { + "epoch": 1.2745205479452055, + "grad_norm": 2.1836023330688477, + "learning_rate": 3.2250203086921204e-06, + "loss": 1.8275, + "step": 1163 + }, + { + "epoch": 1.2756164383561643, + "grad_norm": 2.1917505264282227, + "learning_rate": 3.2229894394800977e-06, + "loss": 1.8289, + "step": 1164 + }, + { + "epoch": 1.2767123287671232, + "grad_norm": 2.315983295440674, + "learning_rate": 3.220958570268075e-06, + "loss": 1.8655, + "step": 1165 + }, + { + "epoch": 1.2778082191780822, + "grad_norm": 2.3628008365631104, + "learning_rate": 3.218927701056052e-06, + "loss": 1.7396, + "step": 1166 + }, + { + "epoch": 1.2789041095890412, + "grad_norm": 2.3200488090515137, + "learning_rate": 3.2168968318440295e-06, + "loss": 1.8196, + "step": 1167 + }, + { + "epoch": 1.28, + "grad_norm": 2.2591991424560547, + "learning_rate": 3.2148659626320068e-06, + "loss": 1.7323, + "step": 1168 + }, + { + "epoch": 1.2810958904109588, + "grad_norm": 2.2458770275115967, + "learning_rate": 3.212835093419984e-06, + "loss": 1.8881, + "step": 1169 + }, + { + "epoch": 1.2821917808219179, + "grad_norm": 2.202010154724121, + "learning_rate": 3.2108042242079613e-06, + "loss": 1.7776, + "step": 1170 + }, + { + "epoch": 1.2821917808219179, + "eval_loss": 2.005174398422241, + "eval_runtime": 37.3597, + "eval_samples_per_second": 26.767, + "eval_steps_per_second": 1.124, + "step": 1170 + }, + { + "epoch": 1.2832876712328767, + "grad_norm": 2.2375359535217285, + "learning_rate": 3.2087733549959386e-06, + "loss": 1.7994, + "step": 1171 + }, + { + "epoch": 1.2843835616438355, + "grad_norm": 2.3152217864990234, + "learning_rate": 3.206742485783916e-06, + "loss": 1.8767, + "step": 1172 + }, + { + "epoch": 1.2854794520547945, + "grad_norm": 2.3317618370056152, + "learning_rate": 3.2047116165718927e-06, + "loss": 1.772, + "step": 1173 + }, + { + "epoch": 1.2865753424657536, + "grad_norm": 2.249943971633911, + "learning_rate": 3.2026807473598704e-06, + "loss": 1.7793, + "step": 1174 + }, + { + "epoch": 1.2876712328767124, + "grad_norm": 2.320481538772583, + "learning_rate": 3.2006498781478473e-06, + "loss": 1.8275, + "step": 1175 + }, + { + "epoch": 1.2887671232876712, + "grad_norm": 2.21785569190979, + "learning_rate": 3.198619008935825e-06, + "loss": 1.8449, + "step": 1176 + }, + { + "epoch": 1.2898630136986302, + "grad_norm": 2.469965934753418, + "learning_rate": 3.196588139723802e-06, + "loss": 1.779, + "step": 1177 + }, + { + "epoch": 1.290958904109589, + "grad_norm": 2.1989049911499023, + "learning_rate": 3.1945572705117795e-06, + "loss": 1.8188, + "step": 1178 + }, + { + "epoch": 1.2920547945205478, + "grad_norm": 2.293513298034668, + "learning_rate": 3.1925264012997564e-06, + "loss": 1.8377, + "step": 1179 + }, + { + "epoch": 1.2931506849315069, + "grad_norm": 2.3462443351745605, + "learning_rate": 3.190495532087734e-06, + "loss": 1.8418, + "step": 1180 + }, + { + "epoch": 1.2931506849315069, + "eval_loss": 2.0061001777648926, + "eval_runtime": 37.2951, + "eval_samples_per_second": 26.813, + "eval_steps_per_second": 1.126, + "step": 1180 + }, + { + "epoch": 1.2942465753424657, + "grad_norm": 2.2071478366851807, + "learning_rate": 3.188464662875711e-06, + "loss": 1.8419, + "step": 1181 + }, + { + "epoch": 1.2953424657534247, + "grad_norm": 2.233333110809326, + "learning_rate": 3.1864337936636886e-06, + "loss": 1.8931, + "step": 1182 + }, + { + "epoch": 1.2964383561643835, + "grad_norm": 2.383085250854492, + "learning_rate": 3.1844029244516655e-06, + "loss": 1.8047, + "step": 1183 + }, + { + "epoch": 1.2975342465753426, + "grad_norm": 2.473808765411377, + "learning_rate": 3.182372055239643e-06, + "loss": 1.7794, + "step": 1184 + }, + { + "epoch": 1.2986301369863014, + "grad_norm": 2.3357670307159424, + "learning_rate": 3.18034118602762e-06, + "loss": 1.7717, + "step": 1185 + }, + { + "epoch": 1.2997260273972602, + "grad_norm": 2.376835823059082, + "learning_rate": 3.1783103168155973e-06, + "loss": 1.8534, + "step": 1186 + }, + { + "epoch": 1.3008219178082192, + "grad_norm": 2.447415590286255, + "learning_rate": 3.1762794476035746e-06, + "loss": 1.8251, + "step": 1187 + }, + { + "epoch": 1.301917808219178, + "grad_norm": 2.5145719051361084, + "learning_rate": 3.174248578391552e-06, + "loss": 1.8117, + "step": 1188 + }, + { + "epoch": 1.303013698630137, + "grad_norm": 2.3134653568267822, + "learning_rate": 3.172217709179529e-06, + "loss": 1.8111, + "step": 1189 + }, + { + "epoch": 1.3041095890410959, + "grad_norm": 2.293311834335327, + "learning_rate": 3.1701868399675064e-06, + "loss": 1.791, + "step": 1190 + }, + { + "epoch": 1.3041095890410959, + "eval_loss": 2.0038933753967285, + "eval_runtime": 37.2282, + "eval_samples_per_second": 26.861, + "eval_steps_per_second": 1.128, + "step": 1190 + }, + { + "epoch": 1.305205479452055, + "grad_norm": 2.298020124435425, + "learning_rate": 3.1681559707554837e-06, + "loss": 1.8301, + "step": 1191 + }, + { + "epoch": 1.3063013698630137, + "grad_norm": 2.5261130332946777, + "learning_rate": 3.166125101543461e-06, + "loss": 1.7714, + "step": 1192 + }, + { + "epoch": 1.3073972602739725, + "grad_norm": 2.270820379257202, + "learning_rate": 3.1640942323314382e-06, + "loss": 1.8246, + "step": 1193 + }, + { + "epoch": 1.3084931506849315, + "grad_norm": 2.373260021209717, + "learning_rate": 3.1620633631194155e-06, + "loss": 1.8316, + "step": 1194 + }, + { + "epoch": 1.3095890410958904, + "grad_norm": 2.3265364170074463, + "learning_rate": 3.1600324939073928e-06, + "loss": 1.7679, + "step": 1195 + }, + { + "epoch": 1.3106849315068494, + "grad_norm": 2.2378532886505127, + "learning_rate": 3.15800162469537e-06, + "loss": 1.8149, + "step": 1196 + }, + { + "epoch": 1.3117808219178082, + "grad_norm": 2.2459309101104736, + "learning_rate": 3.155970755483347e-06, + "loss": 1.8086, + "step": 1197 + }, + { + "epoch": 1.3128767123287672, + "grad_norm": 2.2769811153411865, + "learning_rate": 3.1539398862713246e-06, + "loss": 1.8109, + "step": 1198 + }, + { + "epoch": 1.313972602739726, + "grad_norm": 2.310483694076538, + "learning_rate": 3.1519090170593015e-06, + "loss": 1.8297, + "step": 1199 + }, + { + "epoch": 1.3150684931506849, + "grad_norm": 2.302478313446045, + "learning_rate": 3.149878147847279e-06, + "loss": 1.8237, + "step": 1200 + }, + { + "epoch": 1.3150684931506849, + "eval_loss": 2.0011062622070312, + "eval_runtime": 37.2077, + "eval_samples_per_second": 26.876, + "eval_steps_per_second": 1.129, + "step": 1200 + }, + { + "epoch": 1.316164383561644, + "grad_norm": 2.2240796089172363, + "learning_rate": 3.147847278635256e-06, + "loss": 1.8745, + "step": 1201 + }, + { + "epoch": 1.3172602739726027, + "grad_norm": 2.303537130355835, + "learning_rate": 3.1458164094232337e-06, + "loss": 1.9051, + "step": 1202 + }, + { + "epoch": 1.3183561643835615, + "grad_norm": 2.4038636684417725, + "learning_rate": 3.1437855402112106e-06, + "loss": 1.7974, + "step": 1203 + }, + { + "epoch": 1.3194520547945205, + "grad_norm": 2.3458874225616455, + "learning_rate": 3.1417546709991883e-06, + "loss": 1.7601, + "step": 1204 + }, + { + "epoch": 1.3205479452054796, + "grad_norm": 2.3399100303649902, + "learning_rate": 3.139723801787165e-06, + "loss": 1.8314, + "step": 1205 + }, + { + "epoch": 1.3216438356164384, + "grad_norm": 2.293229818344116, + "learning_rate": 3.137692932575143e-06, + "loss": 1.8117, + "step": 1206 + }, + { + "epoch": 1.3227397260273972, + "grad_norm": 2.5378005504608154, + "learning_rate": 3.1356620633631197e-06, + "loss": 1.8614, + "step": 1207 + }, + { + "epoch": 1.3238356164383562, + "grad_norm": 2.3644745349884033, + "learning_rate": 3.1336311941510974e-06, + "loss": 1.7843, + "step": 1208 + }, + { + "epoch": 1.324931506849315, + "grad_norm": 2.1592626571655273, + "learning_rate": 3.1316003249390742e-06, + "loss": 1.8128, + "step": 1209 + }, + { + "epoch": 1.3260273972602739, + "grad_norm": 2.155649185180664, + "learning_rate": 3.129569455727051e-06, + "loss": 1.8636, + "step": 1210 + }, + { + "epoch": 1.3260273972602739, + "eval_loss": 1.9952574968338013, + "eval_runtime": 37.3004, + "eval_samples_per_second": 26.809, + "eval_steps_per_second": 1.126, + "step": 1210 + }, + { + "epoch": 1.3271232876712329, + "grad_norm": 2.1906397342681885, + "learning_rate": 3.1275385865150288e-06, + "loss": 1.8583, + "step": 1211 + }, + { + "epoch": 1.3282191780821917, + "grad_norm": 2.2708194255828857, + "learning_rate": 3.1255077173030056e-06, + "loss": 1.8363, + "step": 1212 + }, + { + "epoch": 1.3293150684931507, + "grad_norm": 2.1753969192504883, + "learning_rate": 3.1234768480909833e-06, + "loss": 1.8372, + "step": 1213 + }, + { + "epoch": 1.3304109589041095, + "grad_norm": 2.229363441467285, + "learning_rate": 3.12144597887896e-06, + "loss": 1.8179, + "step": 1214 + }, + { + "epoch": 1.3315068493150686, + "grad_norm": 2.362152338027954, + "learning_rate": 3.119415109666938e-06, + "loss": 1.8596, + "step": 1215 + }, + { + "epoch": 1.3326027397260274, + "grad_norm": 2.3278255462646484, + "learning_rate": 3.1173842404549147e-06, + "loss": 1.8778, + "step": 1216 + }, + { + "epoch": 1.3336986301369862, + "grad_norm": 2.477710247039795, + "learning_rate": 3.1153533712428924e-06, + "loss": 1.8297, + "step": 1217 + }, + { + "epoch": 1.3347945205479452, + "grad_norm": 2.2611043453216553, + "learning_rate": 3.1133225020308693e-06, + "loss": 1.8374, + "step": 1218 + }, + { + "epoch": 1.335890410958904, + "grad_norm": 2.198275566101074, + "learning_rate": 3.111291632818847e-06, + "loss": 1.8686, + "step": 1219 + }, + { + "epoch": 1.336986301369863, + "grad_norm": 2.227508068084717, + "learning_rate": 3.109260763606824e-06, + "loss": 1.8642, + "step": 1220 + }, + { + "epoch": 1.336986301369863, + "eval_loss": 1.9998422861099243, + "eval_runtime": 37.5152, + "eval_samples_per_second": 26.656, + "eval_steps_per_second": 1.12, + "step": 1220 + }, + { + "epoch": 1.3380821917808219, + "grad_norm": 2.386310338973999, + "learning_rate": 3.1072298943948015e-06, + "loss": 1.8347, + "step": 1221 + }, + { + "epoch": 1.339178082191781, + "grad_norm": 2.290313482284546, + "learning_rate": 3.1051990251827784e-06, + "loss": 1.8455, + "step": 1222 + }, + { + "epoch": 1.3402739726027397, + "grad_norm": 2.3063457012176514, + "learning_rate": 3.1031681559707557e-06, + "loss": 1.811, + "step": 1223 + }, + { + "epoch": 1.3413698630136985, + "grad_norm": 2.297840118408203, + "learning_rate": 3.101137286758733e-06, + "loss": 1.8558, + "step": 1224 + }, + { + "epoch": 1.3424657534246576, + "grad_norm": 2.40864896774292, + "learning_rate": 3.09910641754671e-06, + "loss": 1.8604, + "step": 1225 + }, + { + "epoch": 1.3435616438356164, + "grad_norm": 2.397599458694458, + "learning_rate": 3.0970755483346875e-06, + "loss": 1.8556, + "step": 1226 + }, + { + "epoch": 1.3446575342465754, + "grad_norm": 2.409029245376587, + "learning_rate": 3.0950446791226648e-06, + "loss": 1.8424, + "step": 1227 + }, + { + "epoch": 1.3457534246575342, + "grad_norm": 2.2789814472198486, + "learning_rate": 3.093013809910642e-06, + "loss": 1.8945, + "step": 1228 + }, + { + "epoch": 1.3468493150684933, + "grad_norm": 2.3749048709869385, + "learning_rate": 3.0909829406986193e-06, + "loss": 1.8387, + "step": 1229 + }, + { + "epoch": 1.347945205479452, + "grad_norm": 2.44486141204834, + "learning_rate": 3.0889520714865966e-06, + "loss": 1.8838, + "step": 1230 + }, + { + "epoch": 1.347945205479452, + "eval_loss": 1.9961321353912354, + "eval_runtime": 37.6273, + "eval_samples_per_second": 26.576, + "eval_steps_per_second": 1.116, + "step": 1230 + }, + { + "epoch": 1.3490410958904109, + "grad_norm": 2.2852954864501953, + "learning_rate": 3.086921202274574e-06, + "loss": 1.8551, + "step": 1231 + }, + { + "epoch": 1.35013698630137, + "grad_norm": 2.3131206035614014, + "learning_rate": 3.084890333062551e-06, + "loss": 1.8183, + "step": 1232 + }, + { + "epoch": 1.3512328767123287, + "grad_norm": 2.4095048904418945, + "learning_rate": 3.0828594638505284e-06, + "loss": 1.7946, + "step": 1233 + }, + { + "epoch": 1.3523287671232875, + "grad_norm": 2.429971933364868, + "learning_rate": 3.0808285946385057e-06, + "loss": 1.8349, + "step": 1234 + }, + { + "epoch": 1.3534246575342466, + "grad_norm": 2.2360565662384033, + "learning_rate": 3.078797725426483e-06, + "loss": 1.794, + "step": 1235 + }, + { + "epoch": 1.3545205479452056, + "grad_norm": 2.3630571365356445, + "learning_rate": 3.07676685621446e-06, + "loss": 1.7377, + "step": 1236 + }, + { + "epoch": 1.3556164383561644, + "grad_norm": 2.2604660987854004, + "learning_rate": 3.0747359870024375e-06, + "loss": 1.8239, + "step": 1237 + }, + { + "epoch": 1.3567123287671232, + "grad_norm": 2.209717035293579, + "learning_rate": 3.0727051177904144e-06, + "loss": 1.7835, + "step": 1238 + }, + { + "epoch": 1.3578082191780823, + "grad_norm": 2.3221793174743652, + "learning_rate": 3.070674248578392e-06, + "loss": 1.8192, + "step": 1239 + }, + { + "epoch": 1.358904109589041, + "grad_norm": 2.357313394546509, + "learning_rate": 3.068643379366369e-06, + "loss": 1.8246, + "step": 1240 + }, + { + "epoch": 1.358904109589041, + "eval_loss": 1.9935922622680664, + "eval_runtime": 37.2832, + "eval_samples_per_second": 26.822, + "eval_steps_per_second": 1.127, + "step": 1240 + }, + { + "epoch": 1.3599999999999999, + "grad_norm": 2.2563390731811523, + "learning_rate": 3.0666125101543466e-06, + "loss": 1.8175, + "step": 1241 + }, + { + "epoch": 1.361095890410959, + "grad_norm": 2.2531445026397705, + "learning_rate": 3.0645816409423235e-06, + "loss": 1.7713, + "step": 1242 + }, + { + "epoch": 1.362191780821918, + "grad_norm": 2.2369470596313477, + "learning_rate": 3.062550771730301e-06, + "loss": 1.8377, + "step": 1243 + }, + { + "epoch": 1.3632876712328768, + "grad_norm": 2.4063796997070312, + "learning_rate": 3.060519902518278e-06, + "loss": 1.8116, + "step": 1244 + }, + { + "epoch": 1.3643835616438356, + "grad_norm": 2.4628145694732666, + "learning_rate": 3.0584890333062557e-06, + "loss": 1.7998, + "step": 1245 + }, + { + "epoch": 1.3654794520547946, + "grad_norm": 2.3018743991851807, + "learning_rate": 3.0564581640942326e-06, + "loss": 1.7413, + "step": 1246 + }, + { + "epoch": 1.3665753424657534, + "grad_norm": 2.2432262897491455, + "learning_rate": 3.0544272948822094e-06, + "loss": 1.82, + "step": 1247 + }, + { + "epoch": 1.3676712328767122, + "grad_norm": 2.368435859680176, + "learning_rate": 3.052396425670187e-06, + "loss": 1.7976, + "step": 1248 + }, + { + "epoch": 1.3687671232876713, + "grad_norm": 2.2668025493621826, + "learning_rate": 3.050365556458164e-06, + "loss": 1.8379, + "step": 1249 + }, + { + "epoch": 1.36986301369863, + "grad_norm": 2.2622179985046387, + "learning_rate": 3.0483346872461417e-06, + "loss": 1.794, + "step": 1250 + }, + { + "epoch": 1.36986301369863, + "eval_loss": 1.9959099292755127, + "eval_runtime": 37.3358, + "eval_samples_per_second": 26.784, + "eval_steps_per_second": 1.125, + "step": 1250 + }, + { + "epoch": 1.370958904109589, + "grad_norm": 2.3414738178253174, + "learning_rate": 3.0463038180341185e-06, + "loss": 1.7993, + "step": 1251 + }, + { + "epoch": 1.372054794520548, + "grad_norm": 2.2678418159484863, + "learning_rate": 3.0442729488220962e-06, + "loss": 1.8581, + "step": 1252 + }, + { + "epoch": 1.373150684931507, + "grad_norm": 2.268195867538452, + "learning_rate": 3.042242079610073e-06, + "loss": 1.8486, + "step": 1253 + }, + { + "epoch": 1.3742465753424657, + "grad_norm": 2.230173110961914, + "learning_rate": 3.0402112103980508e-06, + "loss": 1.8468, + "step": 1254 + }, + { + "epoch": 1.3753424657534246, + "grad_norm": 2.3020238876342773, + "learning_rate": 3.0381803411860276e-06, + "loss": 1.7623, + "step": 1255 + }, + { + "epoch": 1.3764383561643836, + "grad_norm": 2.385343313217163, + "learning_rate": 3.0361494719740053e-06, + "loss": 1.7976, + "step": 1256 + }, + { + "epoch": 1.3775342465753424, + "grad_norm": 2.344768524169922, + "learning_rate": 3.034118602761982e-06, + "loss": 1.807, + "step": 1257 + }, + { + "epoch": 1.3786301369863014, + "grad_norm": 2.411760091781616, + "learning_rate": 3.03208773354996e-06, + "loss": 1.8079, + "step": 1258 + }, + { + "epoch": 1.3797260273972602, + "grad_norm": 2.3160128593444824, + "learning_rate": 3.0300568643379367e-06, + "loss": 1.7868, + "step": 1259 + }, + { + "epoch": 1.3808219178082193, + "grad_norm": 2.334632635116577, + "learning_rate": 3.028025995125914e-06, + "loss": 1.8466, + "step": 1260 + }, + { + "epoch": 1.3808219178082193, + "eval_loss": 1.9903327226638794, + "eval_runtime": 37.2256, + "eval_samples_per_second": 26.863, + "eval_steps_per_second": 1.128, + "step": 1260 + }, + { + "epoch": 1.381917808219178, + "grad_norm": 2.3443334102630615, + "learning_rate": 3.0259951259138913e-06, + "loss": 1.8302, + "step": 1261 + }, + { + "epoch": 1.383013698630137, + "grad_norm": 2.311563014984131, + "learning_rate": 3.0239642567018686e-06, + "loss": 1.7727, + "step": 1262 + }, + { + "epoch": 1.384109589041096, + "grad_norm": 2.2501776218414307, + "learning_rate": 3.021933387489846e-06, + "loss": 1.8058, + "step": 1263 + }, + { + "epoch": 1.3852054794520547, + "grad_norm": 2.259157657623291, + "learning_rate": 3.019902518277823e-06, + "loss": 1.7627, + "step": 1264 + }, + { + "epoch": 1.3863013698630138, + "grad_norm": 2.4380314350128174, + "learning_rate": 3.0178716490658004e-06, + "loss": 1.8237, + "step": 1265 + }, + { + "epoch": 1.3873972602739726, + "grad_norm": 2.3203837871551514, + "learning_rate": 3.0158407798537777e-06, + "loss": 1.8557, + "step": 1266 + }, + { + "epoch": 1.3884931506849316, + "grad_norm": 2.220996141433716, + "learning_rate": 3.013809910641755e-06, + "loss": 1.7641, + "step": 1267 + }, + { + "epoch": 1.3895890410958904, + "grad_norm": 2.190006971359253, + "learning_rate": 3.0117790414297322e-06, + "loss": 1.7705, + "step": 1268 + }, + { + "epoch": 1.3906849315068492, + "grad_norm": 2.229748249053955, + "learning_rate": 3.0097481722177095e-06, + "loss": 1.7709, + "step": 1269 + }, + { + "epoch": 1.3917808219178083, + "grad_norm": 2.3755619525909424, + "learning_rate": 3.0077173030056868e-06, + "loss": 1.8172, + "step": 1270 + }, + { + "epoch": 1.3917808219178083, + "eval_loss": 1.992952585220337, + "eval_runtime": 37.2403, + "eval_samples_per_second": 26.853, + "eval_steps_per_second": 1.128, + "step": 1270 + }, + { + "epoch": 1.392876712328767, + "grad_norm": 2.2514004707336426, + "learning_rate": 3.005686433793664e-06, + "loss": 1.7719, + "step": 1271 + }, + { + "epoch": 1.393972602739726, + "grad_norm": 2.3010191917419434, + "learning_rate": 3.0036555645816413e-06, + "loss": 1.8369, + "step": 1272 + }, + { + "epoch": 1.395068493150685, + "grad_norm": 2.68790602684021, + "learning_rate": 3.001624695369618e-06, + "loss": 1.7974, + "step": 1273 + }, + { + "epoch": 1.396164383561644, + "grad_norm": 2.454904794692993, + "learning_rate": 2.999593826157596e-06, + "loss": 1.8509, + "step": 1274 + }, + { + "epoch": 1.3972602739726028, + "grad_norm": 2.251556396484375, + "learning_rate": 2.9975629569455727e-06, + "loss": 1.8011, + "step": 1275 + }, + { + "epoch": 1.3983561643835616, + "grad_norm": 2.3789098262786865, + "learning_rate": 2.9955320877335504e-06, + "loss": 1.8488, + "step": 1276 + }, + { + "epoch": 1.3994520547945206, + "grad_norm": 2.2659196853637695, + "learning_rate": 2.9935012185215273e-06, + "loss": 1.7946, + "step": 1277 + }, + { + "epoch": 1.4005479452054794, + "grad_norm": 2.2148971557617188, + "learning_rate": 2.991470349309505e-06, + "loss": 1.8388, + "step": 1278 + }, + { + "epoch": 1.4016438356164382, + "grad_norm": 2.1883089542388916, + "learning_rate": 2.989439480097482e-06, + "loss": 1.7925, + "step": 1279 + }, + { + "epoch": 1.4027397260273973, + "grad_norm": 2.290208339691162, + "learning_rate": 2.9874086108854595e-06, + "loss": 1.8099, + "step": 1280 + }, + { + "epoch": 1.4027397260273973, + "eval_loss": 1.9924782514572144, + "eval_runtime": 37.1891, + "eval_samples_per_second": 26.89, + "eval_steps_per_second": 1.129, + "step": 1280 + }, + { + "epoch": 1.403835616438356, + "grad_norm": 2.121649980545044, + "learning_rate": 2.9853777416734364e-06, + "loss": 1.7991, + "step": 1281 + }, + { + "epoch": 1.4049315068493151, + "grad_norm": 2.3528690338134766, + "learning_rate": 2.983346872461414e-06, + "loss": 1.8487, + "step": 1282 + }, + { + "epoch": 1.406027397260274, + "grad_norm": 2.391491174697876, + "learning_rate": 2.981316003249391e-06, + "loss": 1.8564, + "step": 1283 + }, + { + "epoch": 1.407123287671233, + "grad_norm": 2.336615800857544, + "learning_rate": 2.9792851340373678e-06, + "loss": 1.8216, + "step": 1284 + }, + { + "epoch": 1.4082191780821918, + "grad_norm": 2.2958548069000244, + "learning_rate": 2.9772542648253455e-06, + "loss": 1.8003, + "step": 1285 + }, + { + "epoch": 1.4093150684931506, + "grad_norm": 2.2062184810638428, + "learning_rate": 2.9752233956133223e-06, + "loss": 1.8488, + "step": 1286 + }, + { + "epoch": 1.4104109589041096, + "grad_norm": 2.486133575439453, + "learning_rate": 2.9731925264013e-06, + "loss": 1.7914, + "step": 1287 + }, + { + "epoch": 1.4115068493150684, + "grad_norm": 2.2933201789855957, + "learning_rate": 2.971161657189277e-06, + "loss": 1.8164, + "step": 1288 + }, + { + "epoch": 1.4126027397260275, + "grad_norm": 2.3301920890808105, + "learning_rate": 2.9691307879772546e-06, + "loss": 1.7883, + "step": 1289 + }, + { + "epoch": 1.4136986301369863, + "grad_norm": 2.248832941055298, + "learning_rate": 2.9670999187652314e-06, + "loss": 1.7916, + "step": 1290 + }, + { + "epoch": 1.4136986301369863, + "eval_loss": 1.9919097423553467, + "eval_runtime": 37.1283, + "eval_samples_per_second": 26.934, + "eval_steps_per_second": 1.131, + "step": 1290 + }, + { + "epoch": 1.4147945205479453, + "grad_norm": 2.37760853767395, + "learning_rate": 2.965069049553209e-06, + "loss": 1.7854, + "step": 1291 + }, + { + "epoch": 1.415890410958904, + "grad_norm": 2.327718734741211, + "learning_rate": 2.963038180341186e-06, + "loss": 1.874, + "step": 1292 + }, + { + "epoch": 1.416986301369863, + "grad_norm": 2.2413878440856934, + "learning_rate": 2.9610073111291637e-06, + "loss": 1.7591, + "step": 1293 + }, + { + "epoch": 1.418082191780822, + "grad_norm": 2.2203164100646973, + "learning_rate": 2.9589764419171405e-06, + "loss": 1.8466, + "step": 1294 + }, + { + "epoch": 1.4191780821917808, + "grad_norm": 2.2070505619049072, + "learning_rate": 2.9569455727051182e-06, + "loss": 1.8689, + "step": 1295 + }, + { + "epoch": 1.4202739726027398, + "grad_norm": 2.2122223377227783, + "learning_rate": 2.954914703493095e-06, + "loss": 1.7689, + "step": 1296 + }, + { + "epoch": 1.4213698630136986, + "grad_norm": 2.331303358078003, + "learning_rate": 2.9528838342810724e-06, + "loss": 1.8089, + "step": 1297 + }, + { + "epoch": 1.4224657534246576, + "grad_norm": 2.408184289932251, + "learning_rate": 2.9508529650690496e-06, + "loss": 1.7847, + "step": 1298 + }, + { + "epoch": 1.4235616438356165, + "grad_norm": 2.212766170501709, + "learning_rate": 2.948822095857027e-06, + "loss": 1.7899, + "step": 1299 + }, + { + "epoch": 1.4246575342465753, + "grad_norm": 2.259953737258911, + "learning_rate": 2.946791226645004e-06, + "loss": 1.8435, + "step": 1300 + }, + { + "epoch": 1.4246575342465753, + "eval_loss": 1.9893293380737305, + "eval_runtime": 37.1675, + "eval_samples_per_second": 26.905, + "eval_steps_per_second": 1.13, + "step": 1300 + }, + { + "epoch": 1.4257534246575343, + "grad_norm": 2.342400312423706, + "learning_rate": 2.9447603574329815e-06, + "loss": 1.7779, + "step": 1301 + }, + { + "epoch": 1.426849315068493, + "grad_norm": 2.237618923187256, + "learning_rate": 2.9427294882209587e-06, + "loss": 1.7509, + "step": 1302 + }, + { + "epoch": 1.427945205479452, + "grad_norm": 2.3418633937835693, + "learning_rate": 2.940698619008936e-06, + "loss": 1.7963, + "step": 1303 + }, + { + "epoch": 1.429041095890411, + "grad_norm": 2.470827579498291, + "learning_rate": 2.9386677497969133e-06, + "loss": 1.8045, + "step": 1304 + }, + { + "epoch": 1.43013698630137, + "grad_norm": 2.508345365524292, + "learning_rate": 2.9366368805848906e-06, + "loss": 1.7926, + "step": 1305 + }, + { + "epoch": 1.4312328767123288, + "grad_norm": 2.261319160461426, + "learning_rate": 2.934606011372868e-06, + "loss": 1.831, + "step": 1306 + }, + { + "epoch": 1.4323287671232876, + "grad_norm": 2.2703914642333984, + "learning_rate": 2.932575142160845e-06, + "loss": 1.8352, + "step": 1307 + }, + { + "epoch": 1.4334246575342466, + "grad_norm": 2.243626117706299, + "learning_rate": 2.9305442729488224e-06, + "loss": 1.8162, + "step": 1308 + }, + { + "epoch": 1.4345205479452054, + "grad_norm": 2.3859238624572754, + "learning_rate": 2.9285134037367997e-06, + "loss": 1.8171, + "step": 1309 + }, + { + "epoch": 1.4356164383561643, + "grad_norm": 2.3350934982299805, + "learning_rate": 2.9264825345247765e-06, + "loss": 1.8066, + "step": 1310 + }, + { + "epoch": 1.4356164383561643, + "eval_loss": 1.9905259609222412, + "eval_runtime": 37.2009, + "eval_samples_per_second": 26.881, + "eval_steps_per_second": 1.129, + "step": 1310 + }, + { + "epoch": 1.4367123287671233, + "grad_norm": 2.1943347454071045, + "learning_rate": 2.9244516653127542e-06, + "loss": 1.8435, + "step": 1311 + }, + { + "epoch": 1.4378082191780823, + "grad_norm": 2.429128885269165, + "learning_rate": 2.922420796100731e-06, + "loss": 1.8179, + "step": 1312 + }, + { + "epoch": 1.4389041095890411, + "grad_norm": 2.3677234649658203, + "learning_rate": 2.9203899268887088e-06, + "loss": 1.7245, + "step": 1313 + }, + { + "epoch": 1.44, + "grad_norm": 2.255570888519287, + "learning_rate": 2.9183590576766856e-06, + "loss": 1.8545, + "step": 1314 + }, + { + "epoch": 1.441095890410959, + "grad_norm": 2.5176682472229004, + "learning_rate": 2.9163281884646633e-06, + "loss": 1.7509, + "step": 1315 + }, + { + "epoch": 1.4421917808219178, + "grad_norm": 2.29972767829895, + "learning_rate": 2.91429731925264e-06, + "loss": 1.8116, + "step": 1316 + }, + { + "epoch": 1.4432876712328766, + "grad_norm": 2.2853286266326904, + "learning_rate": 2.912266450040618e-06, + "loss": 1.8552, + "step": 1317 + }, + { + "epoch": 1.4443835616438356, + "grad_norm": 2.505075216293335, + "learning_rate": 2.9102355808285947e-06, + "loss": 1.811, + "step": 1318 + }, + { + "epoch": 1.4454794520547944, + "grad_norm": 2.229064464569092, + "learning_rate": 2.9082047116165724e-06, + "loss": 1.8068, + "step": 1319 + }, + { + "epoch": 1.4465753424657535, + "grad_norm": 2.199187994003296, + "learning_rate": 2.9061738424045493e-06, + "loss": 1.7825, + "step": 1320 + }, + { + "epoch": 1.4465753424657535, + "eval_loss": 1.9851222038269043, + "eval_runtime": 37.4614, + "eval_samples_per_second": 26.694, + "eval_steps_per_second": 1.121, + "step": 1320 + }, + { + "epoch": 1.4476712328767123, + "grad_norm": 2.4318959712982178, + "learning_rate": 2.904142973192527e-06, + "loss": 1.77, + "step": 1321 + }, + { + "epoch": 1.4487671232876713, + "grad_norm": 2.480703592300415, + "learning_rate": 2.902112103980504e-06, + "loss": 1.8774, + "step": 1322 + }, + { + "epoch": 1.4498630136986301, + "grad_norm": 2.367624521255493, + "learning_rate": 2.900081234768481e-06, + "loss": 1.818, + "step": 1323 + }, + { + "epoch": 1.450958904109589, + "grad_norm": 2.245974540710449, + "learning_rate": 2.8980503655564584e-06, + "loss": 1.8054, + "step": 1324 + }, + { + "epoch": 1.452054794520548, + "grad_norm": 2.352687120437622, + "learning_rate": 2.8960194963444357e-06, + "loss": 1.7922, + "step": 1325 + }, + { + "epoch": 1.4531506849315068, + "grad_norm": 2.308253526687622, + "learning_rate": 2.893988627132413e-06, + "loss": 1.7637, + "step": 1326 + }, + { + "epoch": 1.4542465753424658, + "grad_norm": 2.309495687484741, + "learning_rate": 2.8919577579203902e-06, + "loss": 1.7905, + "step": 1327 + }, + { + "epoch": 1.4553424657534246, + "grad_norm": 2.354365348815918, + "learning_rate": 2.8899268887083675e-06, + "loss": 1.7321, + "step": 1328 + }, + { + "epoch": 1.4564383561643837, + "grad_norm": 2.3886210918426514, + "learning_rate": 2.8878960194963448e-06, + "loss": 1.7932, + "step": 1329 + }, + { + "epoch": 1.4575342465753425, + "grad_norm": 2.395714044570923, + "learning_rate": 2.885865150284322e-06, + "loss": 1.813, + "step": 1330 + }, + { + "epoch": 1.4575342465753425, + "eval_loss": 1.9851990938186646, + "eval_runtime": 37.2966, + "eval_samples_per_second": 26.812, + "eval_steps_per_second": 1.126, + "step": 1330 + }, + { + "epoch": 1.4586301369863013, + "grad_norm": 2.433069944381714, + "learning_rate": 2.8838342810722993e-06, + "loss": 1.8346, + "step": 1331 + }, + { + "epoch": 1.4597260273972603, + "grad_norm": 2.295894145965576, + "learning_rate": 2.8818034118602766e-06, + "loss": 1.7855, + "step": 1332 + }, + { + "epoch": 1.4608219178082191, + "grad_norm": 2.495742082595825, + "learning_rate": 2.879772542648254e-06, + "loss": 1.7601, + "step": 1333 + }, + { + "epoch": 1.4619178082191782, + "grad_norm": 2.35984206199646, + "learning_rate": 2.8777416734362307e-06, + "loss": 1.8329, + "step": 1334 + }, + { + "epoch": 1.463013698630137, + "grad_norm": 2.39632248878479, + "learning_rate": 2.8757108042242084e-06, + "loss": 1.7426, + "step": 1335 + }, + { + "epoch": 1.464109589041096, + "grad_norm": 2.4151153564453125, + "learning_rate": 2.8736799350121853e-06, + "loss": 1.8899, + "step": 1336 + }, + { + "epoch": 1.4652054794520548, + "grad_norm": 2.43914532661438, + "learning_rate": 2.871649065800163e-06, + "loss": 1.8464, + "step": 1337 + }, + { + "epoch": 1.4663013698630136, + "grad_norm": 2.44490122795105, + "learning_rate": 2.86961819658814e-06, + "loss": 1.8643, + "step": 1338 + }, + { + "epoch": 1.4673972602739727, + "grad_norm": 2.2639644145965576, + "learning_rate": 2.8675873273761175e-06, + "loss": 1.792, + "step": 1339 + }, + { + "epoch": 1.4684931506849315, + "grad_norm": 2.2803847789764404, + "learning_rate": 2.8655564581640944e-06, + "loss": 1.8109, + "step": 1340 + }, + { + "epoch": 1.4684931506849315, + "eval_loss": 1.9856781959533691, + "eval_runtime": 37.2519, + "eval_samples_per_second": 26.844, + "eval_steps_per_second": 1.127, + "step": 1340 + }, + { + "epoch": 1.4695890410958903, + "grad_norm": 2.494727849960327, + "learning_rate": 2.863525588952072e-06, + "loss": 1.8032, + "step": 1341 + }, + { + "epoch": 1.4706849315068493, + "grad_norm": 2.239595413208008, + "learning_rate": 2.861494719740049e-06, + "loss": 1.8659, + "step": 1342 + }, + { + "epoch": 1.4717808219178083, + "grad_norm": 2.2922589778900146, + "learning_rate": 2.8594638505280266e-06, + "loss": 1.7587, + "step": 1343 + }, + { + "epoch": 1.4728767123287672, + "grad_norm": 2.345224142074585, + "learning_rate": 2.8574329813160035e-06, + "loss": 1.7602, + "step": 1344 + }, + { + "epoch": 1.473972602739726, + "grad_norm": 2.3415307998657227, + "learning_rate": 2.855402112103981e-06, + "loss": 1.8044, + "step": 1345 + }, + { + "epoch": 1.475068493150685, + "grad_norm": 2.331557035446167, + "learning_rate": 2.853371242891958e-06, + "loss": 1.7808, + "step": 1346 + }, + { + "epoch": 1.4761643835616438, + "grad_norm": 2.293572425842285, + "learning_rate": 2.851340373679935e-06, + "loss": 1.8017, + "step": 1347 + }, + { + "epoch": 1.4772602739726026, + "grad_norm": 2.3477773666381836, + "learning_rate": 2.8493095044679126e-06, + "loss": 1.8113, + "step": 1348 + }, + { + "epoch": 1.4783561643835617, + "grad_norm": 2.3740994930267334, + "learning_rate": 2.8472786352558894e-06, + "loss": 1.8801, + "step": 1349 + }, + { + "epoch": 1.4794520547945205, + "grad_norm": 2.4202206134796143, + "learning_rate": 2.845247766043867e-06, + "loss": 1.821, + "step": 1350 + }, + { + "epoch": 1.4794520547945205, + "eval_loss": 1.9867810010910034, + "eval_runtime": 37.2597, + "eval_samples_per_second": 26.839, + "eval_steps_per_second": 1.127, + "step": 1350 + }, + { + "epoch": 1.4805479452054795, + "grad_norm": 2.2323198318481445, + "learning_rate": 2.843216896831844e-06, + "loss": 1.7574, + "step": 1351 + }, + { + "epoch": 1.4816438356164383, + "grad_norm": 2.2192625999450684, + "learning_rate": 2.8411860276198217e-06, + "loss": 1.7662, + "step": 1352 + }, + { + "epoch": 1.4827397260273973, + "grad_norm": 2.276507616043091, + "learning_rate": 2.8391551584077985e-06, + "loss": 1.7748, + "step": 1353 + }, + { + "epoch": 1.4838356164383562, + "grad_norm": 2.292773723602295, + "learning_rate": 2.8371242891957762e-06, + "loss": 1.8388, + "step": 1354 + }, + { + "epoch": 1.484931506849315, + "grad_norm": 2.3127009868621826, + "learning_rate": 2.835093419983753e-06, + "loss": 1.861, + "step": 1355 + }, + { + "epoch": 1.486027397260274, + "grad_norm": 2.2752184867858887, + "learning_rate": 2.833062550771731e-06, + "loss": 1.7883, + "step": 1356 + }, + { + "epoch": 1.4871232876712328, + "grad_norm": 2.288158416748047, + "learning_rate": 2.8310316815597076e-06, + "loss": 1.7963, + "step": 1357 + }, + { + "epoch": 1.4882191780821918, + "grad_norm": 2.3287482261657715, + "learning_rate": 2.8290008123476853e-06, + "loss": 1.7534, + "step": 1358 + }, + { + "epoch": 1.4893150684931507, + "grad_norm": 2.441279888153076, + "learning_rate": 2.826969943135662e-06, + "loss": 1.8073, + "step": 1359 + }, + { + "epoch": 1.4904109589041097, + "grad_norm": 2.3399839401245117, + "learning_rate": 2.8249390739236395e-06, + "loss": 1.7876, + "step": 1360 + }, + { + "epoch": 1.4904109589041097, + "eval_loss": 1.9868030548095703, + "eval_runtime": 37.2891, + "eval_samples_per_second": 26.817, + "eval_steps_per_second": 1.126, + "step": 1360 + }, + { + "epoch": 1.4915068493150685, + "grad_norm": 2.234246253967285, + "learning_rate": 2.8229082047116167e-06, + "loss": 1.8031, + "step": 1361 + }, + { + "epoch": 1.4926027397260273, + "grad_norm": 2.3136820793151855, + "learning_rate": 2.820877335499594e-06, + "loss": 1.7669, + "step": 1362 + }, + { + "epoch": 1.4936986301369863, + "grad_norm": 2.389420747756958, + "learning_rate": 2.8188464662875713e-06, + "loss": 1.7879, + "step": 1363 + }, + { + "epoch": 1.4947945205479451, + "grad_norm": 2.222306489944458, + "learning_rate": 2.8168155970755486e-06, + "loss": 1.7412, + "step": 1364 + }, + { + "epoch": 1.4958904109589042, + "grad_norm": 2.1689319610595703, + "learning_rate": 2.814784727863526e-06, + "loss": 1.7651, + "step": 1365 + }, + { + "epoch": 1.496986301369863, + "grad_norm": 2.262047529220581, + "learning_rate": 2.812753858651503e-06, + "loss": 1.7798, + "step": 1366 + }, + { + "epoch": 1.498082191780822, + "grad_norm": 2.266894817352295, + "learning_rate": 2.8107229894394804e-06, + "loss": 1.8172, + "step": 1367 + }, + { + "epoch": 1.4991780821917808, + "grad_norm": 2.6400444507598877, + "learning_rate": 2.8086921202274577e-06, + "loss": 1.7871, + "step": 1368 + }, + { + "epoch": 1.5002739726027396, + "grad_norm": 2.3868865966796875, + "learning_rate": 2.806661251015435e-06, + "loss": 1.7427, + "step": 1369 + }, + { + "epoch": 1.5013698630136987, + "grad_norm": 2.5206542015075684, + "learning_rate": 2.8046303818034122e-06, + "loss": 1.817, + "step": 1370 + }, + { + "epoch": 1.5013698630136987, + "eval_loss": 1.9870396852493286, + "eval_runtime": 37.2372, + "eval_samples_per_second": 26.855, + "eval_steps_per_second": 1.128, + "step": 1370 + }, + { + "epoch": 1.5024657534246575, + "grad_norm": 2.5236382484436035, + "learning_rate": 2.802599512591389e-06, + "loss": 1.8662, + "step": 1371 + }, + { + "epoch": 1.5035616438356163, + "grad_norm": 2.5095479488372803, + "learning_rate": 2.8005686433793668e-06, + "loss": 1.9168, + "step": 1372 + }, + { + "epoch": 1.5046575342465753, + "grad_norm": 2.307508707046509, + "learning_rate": 2.7985377741673436e-06, + "loss": 1.799, + "step": 1373 + }, + { + "epoch": 1.5057534246575344, + "grad_norm": 2.410879373550415, + "learning_rate": 2.7965069049553213e-06, + "loss": 1.7447, + "step": 1374 + }, + { + "epoch": 1.5068493150684932, + "grad_norm": 2.738558530807495, + "learning_rate": 2.794476035743298e-06, + "loss": 1.8073, + "step": 1375 + }, + { + "epoch": 1.507945205479452, + "grad_norm": 2.3826518058776855, + "learning_rate": 2.792445166531276e-06, + "loss": 1.8506, + "step": 1376 + }, + { + "epoch": 1.509041095890411, + "grad_norm": 2.334123373031616, + "learning_rate": 2.7904142973192527e-06, + "loss": 1.8209, + "step": 1377 + }, + { + "epoch": 1.5101369863013698, + "grad_norm": 2.668513298034668, + "learning_rate": 2.7883834281072304e-06, + "loss": 1.8285, + "step": 1378 + }, + { + "epoch": 1.5112328767123286, + "grad_norm": 2.471938133239746, + "learning_rate": 2.7863525588952073e-06, + "loss": 1.7668, + "step": 1379 + }, + { + "epoch": 1.5123287671232877, + "grad_norm": 2.3712313175201416, + "learning_rate": 2.784321689683185e-06, + "loss": 1.7565, + "step": 1380 + }, + { + "epoch": 1.5123287671232877, + "eval_loss": 1.9826414585113525, + "eval_runtime": 37.2426, + "eval_samples_per_second": 26.851, + "eval_steps_per_second": 1.128, + "step": 1380 + }, + { + "epoch": 1.5134246575342467, + "grad_norm": 2.3364150524139404, + "learning_rate": 2.782290820471162e-06, + "loss": 1.7596, + "step": 1381 + }, + { + "epoch": 1.5145205479452055, + "grad_norm": 2.3088393211364746, + "learning_rate": 2.7802599512591395e-06, + "loss": 1.7568, + "step": 1382 + }, + { + "epoch": 1.5156164383561643, + "grad_norm": 2.2584595680236816, + "learning_rate": 2.7782290820471164e-06, + "loss": 1.833, + "step": 1383 + }, + { + "epoch": 1.5167123287671234, + "grad_norm": 2.2440273761749268, + "learning_rate": 2.7761982128350932e-06, + "loss": 1.8237, + "step": 1384 + }, + { + "epoch": 1.5178082191780822, + "grad_norm": 2.3552212715148926, + "learning_rate": 2.774167343623071e-06, + "loss": 1.8288, + "step": 1385 + }, + { + "epoch": 1.518904109589041, + "grad_norm": 2.3001790046691895, + "learning_rate": 2.772136474411048e-06, + "loss": 1.8519, + "step": 1386 + }, + { + "epoch": 1.52, + "grad_norm": 2.3691201210021973, + "learning_rate": 2.7701056051990255e-06, + "loss": 1.7759, + "step": 1387 + }, + { + "epoch": 1.521095890410959, + "grad_norm": 2.327988624572754, + "learning_rate": 2.7680747359870023e-06, + "loss": 1.7132, + "step": 1388 + }, + { + "epoch": 1.5221917808219179, + "grad_norm": 2.30245304107666, + "learning_rate": 2.76604386677498e-06, + "loss": 1.8167, + "step": 1389 + }, + { + "epoch": 1.5232876712328767, + "grad_norm": 2.290121078491211, + "learning_rate": 2.764012997562957e-06, + "loss": 1.7913, + "step": 1390 + }, + { + "epoch": 1.5232876712328767, + "eval_loss": 1.98329758644104, + "eval_runtime": 37.2462, + "eval_samples_per_second": 26.848, + "eval_steps_per_second": 1.128, + "step": 1390 + }, + { + "epoch": 1.5243835616438357, + "grad_norm": 2.294076681137085, + "learning_rate": 2.7619821283509346e-06, + "loss": 1.8598, + "step": 1391 + }, + { + "epoch": 1.5254794520547945, + "grad_norm": 2.3325250148773193, + "learning_rate": 2.7599512591389114e-06, + "loss": 1.7392, + "step": 1392 + }, + { + "epoch": 1.5265753424657533, + "grad_norm": 2.2802610397338867, + "learning_rate": 2.757920389926889e-06, + "loss": 1.8332, + "step": 1393 + }, + { + "epoch": 1.5276712328767124, + "grad_norm": 2.284207344055176, + "learning_rate": 2.755889520714866e-06, + "loss": 1.7875, + "step": 1394 + }, + { + "epoch": 1.5287671232876714, + "grad_norm": 2.325998306274414, + "learning_rate": 2.7538586515028437e-06, + "loss": 1.7657, + "step": 1395 + }, + { + "epoch": 1.52986301369863, + "grad_norm": 2.285393476486206, + "learning_rate": 2.7518277822908205e-06, + "loss": 1.7416, + "step": 1396 + }, + { + "epoch": 1.530958904109589, + "grad_norm": 2.3110482692718506, + "learning_rate": 2.749796913078798e-06, + "loss": 1.7931, + "step": 1397 + }, + { + "epoch": 1.532054794520548, + "grad_norm": 2.4234025478363037, + "learning_rate": 2.747766043866775e-06, + "loss": 1.8472, + "step": 1398 + }, + { + "epoch": 1.5331506849315069, + "grad_norm": 2.384767532348633, + "learning_rate": 2.7457351746547524e-06, + "loss": 1.7961, + "step": 1399 + }, + { + "epoch": 1.5342465753424657, + "grad_norm": 2.1857433319091797, + "learning_rate": 2.7437043054427297e-06, + "loss": 1.8654, + "step": 1400 + }, + { + "epoch": 1.5342465753424657, + "eval_loss": 1.9802658557891846, + "eval_runtime": 37.1443, + "eval_samples_per_second": 26.922, + "eval_steps_per_second": 1.131, + "step": 1400 + }, + { + "epoch": 1.5353424657534247, + "grad_norm": 2.1302895545959473, + "learning_rate": 2.741673436230707e-06, + "loss": 1.9051, + "step": 1401 + }, + { + "epoch": 1.5364383561643835, + "grad_norm": 2.377307415008545, + "learning_rate": 2.739642567018684e-06, + "loss": 1.7942, + "step": 1402 + }, + { + "epoch": 1.5375342465753423, + "grad_norm": 2.316549062728882, + "learning_rate": 2.7376116978066615e-06, + "loss": 1.7734, + "step": 1403 + }, + { + "epoch": 1.5386301369863014, + "grad_norm": 2.3452532291412354, + "learning_rate": 2.7355808285946388e-06, + "loss": 1.8243, + "step": 1404 + }, + { + "epoch": 1.5397260273972604, + "grad_norm": 2.3139419555664062, + "learning_rate": 2.733549959382616e-06, + "loss": 1.812, + "step": 1405 + }, + { + "epoch": 1.5408219178082192, + "grad_norm": 2.3135571479797363, + "learning_rate": 2.7315190901705933e-06, + "loss": 1.7751, + "step": 1406 + }, + { + "epoch": 1.541917808219178, + "grad_norm": 2.403231620788574, + "learning_rate": 2.7294882209585706e-06, + "loss": 1.8561, + "step": 1407 + }, + { + "epoch": 1.543013698630137, + "grad_norm": 2.4028689861297607, + "learning_rate": 2.727457351746548e-06, + "loss": 1.8249, + "step": 1408 + }, + { + "epoch": 1.5441095890410959, + "grad_norm": 2.4028851985931396, + "learning_rate": 2.725426482534525e-06, + "loss": 1.779, + "step": 1409 + }, + { + "epoch": 1.5452054794520547, + "grad_norm": 2.270709753036499, + "learning_rate": 2.723395613322502e-06, + "loss": 1.7754, + "step": 1410 + }, + { + "epoch": 1.5452054794520547, + "eval_loss": 1.983272910118103, + "eval_runtime": 37.2405, + "eval_samples_per_second": 26.853, + "eval_steps_per_second": 1.128, + "step": 1410 + }, + { + "epoch": 1.5463013698630137, + "grad_norm": 2.2672812938690186, + "learning_rate": 2.7213647441104797e-06, + "loss": 1.814, + "step": 1411 + }, + { + "epoch": 1.5473972602739727, + "grad_norm": 2.3539397716522217, + "learning_rate": 2.7193338748984565e-06, + "loss": 1.7907, + "step": 1412 + }, + { + "epoch": 1.5484931506849315, + "grad_norm": 2.3229010105133057, + "learning_rate": 2.7173030056864342e-06, + "loss": 1.8061, + "step": 1413 + }, + { + "epoch": 1.5495890410958904, + "grad_norm": 2.482874870300293, + "learning_rate": 2.715272136474411e-06, + "loss": 1.9577, + "step": 1414 + }, + { + "epoch": 1.5506849315068494, + "grad_norm": 2.211798906326294, + "learning_rate": 2.713241267262389e-06, + "loss": 1.7309, + "step": 1415 + }, + { + "epoch": 1.5517808219178082, + "grad_norm": 2.3159289360046387, + "learning_rate": 2.7112103980503656e-06, + "loss": 1.8132, + "step": 1416 + }, + { + "epoch": 1.552876712328767, + "grad_norm": 2.486035108566284, + "learning_rate": 2.7091795288383433e-06, + "loss": 1.7734, + "step": 1417 + }, + { + "epoch": 1.553972602739726, + "grad_norm": 2.3025195598602295, + "learning_rate": 2.70714865962632e-06, + "loss": 1.8232, + "step": 1418 + }, + { + "epoch": 1.555068493150685, + "grad_norm": 2.234581470489502, + "learning_rate": 2.705117790414298e-06, + "loss": 1.7814, + "step": 1419 + }, + { + "epoch": 1.5561643835616439, + "grad_norm": 2.2855048179626465, + "learning_rate": 2.7030869212022747e-06, + "loss": 1.777, + "step": 1420 + }, + { + "epoch": 1.5561643835616439, + "eval_loss": 1.9851160049438477, + "eval_runtime": 37.4606, + "eval_samples_per_second": 26.695, + "eval_steps_per_second": 1.121, + "step": 1420 + }, + { + "epoch": 1.5572602739726027, + "grad_norm": 2.4190328121185303, + "learning_rate": 2.7010560519902516e-06, + "loss": 1.8456, + "step": 1421 + }, + { + "epoch": 1.5583561643835617, + "grad_norm": 2.225623369216919, + "learning_rate": 2.6990251827782293e-06, + "loss": 1.7638, + "step": 1422 + }, + { + "epoch": 1.5594520547945205, + "grad_norm": 2.1516661643981934, + "learning_rate": 2.696994313566206e-06, + "loss": 1.7861, + "step": 1423 + }, + { + "epoch": 1.5605479452054793, + "grad_norm": 2.2461135387420654, + "learning_rate": 2.694963444354184e-06, + "loss": 1.7763, + "step": 1424 + }, + { + "epoch": 1.5616438356164384, + "grad_norm": 2.280869245529175, + "learning_rate": 2.6929325751421607e-06, + "loss": 1.8589, + "step": 1425 + }, + { + "epoch": 1.5627397260273974, + "grad_norm": 2.2034990787506104, + "learning_rate": 2.6909017059301384e-06, + "loss": 1.7705, + "step": 1426 + }, + { + "epoch": 1.563835616438356, + "grad_norm": 2.1346969604492188, + "learning_rate": 2.6888708367181153e-06, + "loss": 1.8318, + "step": 1427 + }, + { + "epoch": 1.564931506849315, + "grad_norm": 2.37400221824646, + "learning_rate": 2.686839967506093e-06, + "loss": 1.746, + "step": 1428 + }, + { + "epoch": 1.566027397260274, + "grad_norm": 2.392561435699463, + "learning_rate": 2.68480909829407e-06, + "loss": 1.8105, + "step": 1429 + }, + { + "epoch": 1.5671232876712329, + "grad_norm": 2.299830198287964, + "learning_rate": 2.6827782290820475e-06, + "loss": 1.7103, + "step": 1430 + }, + { + "epoch": 1.5671232876712329, + "eval_loss": 1.9877182245254517, + "eval_runtime": 37.4813, + "eval_samples_per_second": 26.68, + "eval_steps_per_second": 1.121, + "step": 1430 + }, + { + "epoch": 1.5682191780821917, + "grad_norm": 2.3431026935577393, + "learning_rate": 2.6807473598700244e-06, + "loss": 1.7217, + "step": 1431 + }, + { + "epoch": 1.5693150684931507, + "grad_norm": 2.3474628925323486, + "learning_rate": 2.678716490658002e-06, + "loss": 1.8419, + "step": 1432 + }, + { + "epoch": 1.5704109589041098, + "grad_norm": 2.2877044677734375, + "learning_rate": 2.676685621445979e-06, + "loss": 1.8031, + "step": 1433 + }, + { + "epoch": 1.5715068493150683, + "grad_norm": 2.278613805770874, + "learning_rate": 2.674654752233956e-06, + "loss": 1.8066, + "step": 1434 + }, + { + "epoch": 1.5726027397260274, + "grad_norm": 2.3357837200164795, + "learning_rate": 2.6726238830219335e-06, + "loss": 1.8079, + "step": 1435 + }, + { + "epoch": 1.5736986301369864, + "grad_norm": 2.1932482719421387, + "learning_rate": 2.6705930138099107e-06, + "loss": 1.8032, + "step": 1436 + }, + { + "epoch": 1.5747945205479452, + "grad_norm": 2.298553705215454, + "learning_rate": 2.668562144597888e-06, + "loss": 1.8439, + "step": 1437 + }, + { + "epoch": 1.575890410958904, + "grad_norm": 2.2906084060668945, + "learning_rate": 2.6665312753858653e-06, + "loss": 1.7144, + "step": 1438 + }, + { + "epoch": 1.576986301369863, + "grad_norm": 2.2512989044189453, + "learning_rate": 2.6645004061738426e-06, + "loss": 1.8031, + "step": 1439 + }, + { + "epoch": 1.5780821917808219, + "grad_norm": 2.177103281021118, + "learning_rate": 2.66246953696182e-06, + "loss": 1.8073, + "step": 1440 + }, + { + "epoch": 1.5780821917808219, + "eval_loss": 1.9914618730545044, + "eval_runtime": 37.2706, + "eval_samples_per_second": 26.831, + "eval_steps_per_second": 1.127, + "step": 1440 + }, + { + "epoch": 1.5791780821917807, + "grad_norm": 2.4129281044006348, + "learning_rate": 2.660438667749797e-06, + "loss": 1.7552, + "step": 1441 + }, + { + "epoch": 1.5802739726027397, + "grad_norm": 2.2975244522094727, + "learning_rate": 2.6584077985377744e-06, + "loss": 1.7318, + "step": 1442 + }, + { + "epoch": 1.5813698630136988, + "grad_norm": 2.4061625003814697, + "learning_rate": 2.6563769293257517e-06, + "loss": 1.7839, + "step": 1443 + }, + { + "epoch": 1.5824657534246576, + "grad_norm": 2.236353635787964, + "learning_rate": 2.654346060113729e-06, + "loss": 1.7995, + "step": 1444 + }, + { + "epoch": 1.5835616438356164, + "grad_norm": 2.2464141845703125, + "learning_rate": 2.6523151909017062e-06, + "loss": 1.7496, + "step": 1445 + }, + { + "epoch": 1.5846575342465754, + "grad_norm": 2.2744123935699463, + "learning_rate": 2.6502843216896835e-06, + "loss": 1.7703, + "step": 1446 + }, + { + "epoch": 1.5857534246575342, + "grad_norm": 2.1578054428100586, + "learning_rate": 2.6482534524776603e-06, + "loss": 1.7561, + "step": 1447 + }, + { + "epoch": 1.586849315068493, + "grad_norm": 2.1924850940704346, + "learning_rate": 2.646222583265638e-06, + "loss": 1.867, + "step": 1448 + }, + { + "epoch": 1.587945205479452, + "grad_norm": 2.303420305252075, + "learning_rate": 2.644191714053615e-06, + "loss": 1.8345, + "step": 1449 + }, + { + "epoch": 1.589041095890411, + "grad_norm": 2.308837413787842, + "learning_rate": 2.6421608448415926e-06, + "loss": 1.7979, + "step": 1450 + }, + { + "epoch": 1.589041095890411, + "eval_loss": 1.9836338758468628, + "eval_runtime": 37.253, + "eval_samples_per_second": 26.843, + "eval_steps_per_second": 1.127, + "step": 1450 + }, + { + "epoch": 1.59013698630137, + "grad_norm": 2.2896456718444824, + "learning_rate": 2.6401299756295694e-06, + "loss": 1.7679, + "step": 1451 + }, + { + "epoch": 1.5912328767123287, + "grad_norm": 2.2482378482818604, + "learning_rate": 2.638099106417547e-06, + "loss": 1.7481, + "step": 1452 + }, + { + "epoch": 1.5923287671232877, + "grad_norm": 2.2620620727539062, + "learning_rate": 2.636068237205524e-06, + "loss": 1.801, + "step": 1453 + }, + { + "epoch": 1.5934246575342466, + "grad_norm": 2.216510772705078, + "learning_rate": 2.6340373679935017e-06, + "loss": 1.7485, + "step": 1454 + }, + { + "epoch": 1.5945205479452054, + "grad_norm": 2.354858160018921, + "learning_rate": 2.6320064987814785e-06, + "loss": 1.8636, + "step": 1455 + }, + { + "epoch": 1.5956164383561644, + "grad_norm": 2.251880168914795, + "learning_rate": 2.6299756295694562e-06, + "loss": 1.7087, + "step": 1456 + }, + { + "epoch": 1.5967123287671234, + "grad_norm": 2.2605879306793213, + "learning_rate": 2.627944760357433e-06, + "loss": 1.8141, + "step": 1457 + }, + { + "epoch": 1.5978082191780822, + "grad_norm": 2.3315021991729736, + "learning_rate": 2.6259138911454104e-06, + "loss": 1.8565, + "step": 1458 + }, + { + "epoch": 1.598904109589041, + "grad_norm": 2.1999425888061523, + "learning_rate": 2.6238830219333877e-06, + "loss": 1.8683, + "step": 1459 + }, + { + "epoch": 1.6, + "grad_norm": 2.2652416229248047, + "learning_rate": 2.621852152721365e-06, + "loss": 1.8489, + "step": 1460 + }, + { + "epoch": 1.6, + "eval_loss": 1.9816856384277344, + "eval_runtime": 37.2727, + "eval_samples_per_second": 26.829, + "eval_steps_per_second": 1.127, + "step": 1460 + }, + { + "epoch": 1.601095890410959, + "grad_norm": 2.469440221786499, + "learning_rate": 2.619821283509342e-06, + "loss": 1.7335, + "step": 1461 + }, + { + "epoch": 1.6021917808219177, + "grad_norm": 2.434267044067383, + "learning_rate": 2.6177904142973195e-06, + "loss": 1.8291, + "step": 1462 + }, + { + "epoch": 1.6032876712328767, + "grad_norm": 2.300492286682129, + "learning_rate": 2.6157595450852968e-06, + "loss": 1.7504, + "step": 1463 + }, + { + "epoch": 1.6043835616438358, + "grad_norm": 2.3906843662261963, + "learning_rate": 2.613728675873274e-06, + "loss": 1.8114, + "step": 1464 + }, + { + "epoch": 1.6054794520547944, + "grad_norm": 2.5111773014068604, + "learning_rate": 2.6116978066612513e-06, + "loss": 1.836, + "step": 1465 + }, + { + "epoch": 1.6065753424657534, + "grad_norm": 2.3834846019744873, + "learning_rate": 2.6096669374492286e-06, + "loss": 1.8589, + "step": 1466 + }, + { + "epoch": 1.6076712328767124, + "grad_norm": 2.2709615230560303, + "learning_rate": 2.607636068237206e-06, + "loss": 1.8052, + "step": 1467 + }, + { + "epoch": 1.6087671232876712, + "grad_norm": 2.3351972103118896, + "learning_rate": 2.605605199025183e-06, + "loss": 1.7559, + "step": 1468 + }, + { + "epoch": 1.60986301369863, + "grad_norm": 2.3994698524475098, + "learning_rate": 2.6035743298131604e-06, + "loss": 1.8099, + "step": 1469 + }, + { + "epoch": 1.610958904109589, + "grad_norm": 2.352590322494507, + "learning_rate": 2.6015434606011377e-06, + "loss": 1.8043, + "step": 1470 + }, + { + "epoch": 1.610958904109589, + "eval_loss": 1.9790339469909668, + "eval_runtime": 37.2619, + "eval_samples_per_second": 26.837, + "eval_steps_per_second": 1.127, + "step": 1470 + }, + { + "epoch": 1.612054794520548, + "grad_norm": 2.35754656791687, + "learning_rate": 2.5995125913891145e-06, + "loss": 1.8252, + "step": 1471 + }, + { + "epoch": 1.6131506849315067, + "grad_norm": 2.4011359214782715, + "learning_rate": 2.5974817221770922e-06, + "loss": 1.8005, + "step": 1472 + }, + { + "epoch": 1.6142465753424657, + "grad_norm": 3.143979787826538, + "learning_rate": 2.595450852965069e-06, + "loss": 1.7952, + "step": 1473 + }, + { + "epoch": 1.6153424657534248, + "grad_norm": 2.3928465843200684, + "learning_rate": 2.593419983753047e-06, + "loss": 1.8344, + "step": 1474 + }, + { + "epoch": 1.6164383561643836, + "grad_norm": 2.370368480682373, + "learning_rate": 2.5913891145410236e-06, + "loss": 1.8358, + "step": 1475 + }, + { + "epoch": 1.6175342465753424, + "grad_norm": 2.6163346767425537, + "learning_rate": 2.5893582453290013e-06, + "loss": 1.7734, + "step": 1476 + }, + { + "epoch": 1.6186301369863014, + "grad_norm": 2.370539903640747, + "learning_rate": 2.587327376116978e-06, + "loss": 1.8383, + "step": 1477 + }, + { + "epoch": 1.6197260273972602, + "grad_norm": 2.397007465362549, + "learning_rate": 2.585296506904956e-06, + "loss": 1.7702, + "step": 1478 + }, + { + "epoch": 1.620821917808219, + "grad_norm": 2.4692702293395996, + "learning_rate": 2.5832656376929327e-06, + "loss": 1.7887, + "step": 1479 + }, + { + "epoch": 1.621917808219178, + "grad_norm": 2.5779969692230225, + "learning_rate": 2.5812347684809104e-06, + "loss": 1.7751, + "step": 1480 + }, + { + "epoch": 1.621917808219178, + "eval_loss": 1.9826369285583496, + "eval_runtime": 37.2779, + "eval_samples_per_second": 26.826, + "eval_steps_per_second": 1.127, + "step": 1480 + }, + { + "epoch": 1.6230136986301371, + "grad_norm": 2.3182761669158936, + "learning_rate": 2.5792038992688873e-06, + "loss": 1.7136, + "step": 1481 + }, + { + "epoch": 1.624109589041096, + "grad_norm": 2.4109292030334473, + "learning_rate": 2.577173030056865e-06, + "loss": 1.8804, + "step": 1482 + }, + { + "epoch": 1.6252054794520547, + "grad_norm": 2.426166534423828, + "learning_rate": 2.575142160844842e-06, + "loss": 1.6996, + "step": 1483 + }, + { + "epoch": 1.6263013698630138, + "grad_norm": 2.5459423065185547, + "learning_rate": 2.5731112916328187e-06, + "loss": 1.8186, + "step": 1484 + }, + { + "epoch": 1.6273972602739726, + "grad_norm": 2.583578586578369, + "learning_rate": 2.5710804224207964e-06, + "loss": 1.8275, + "step": 1485 + }, + { + "epoch": 1.6284931506849314, + "grad_norm": 2.442936658859253, + "learning_rate": 2.5690495532087733e-06, + "loss": 1.7553, + "step": 1486 + }, + { + "epoch": 1.6295890410958904, + "grad_norm": 2.587864637374878, + "learning_rate": 2.567018683996751e-06, + "loss": 1.7171, + "step": 1487 + }, + { + "epoch": 1.6306849315068495, + "grad_norm": 2.4119534492492676, + "learning_rate": 2.564987814784728e-06, + "loss": 1.8296, + "step": 1488 + }, + { + "epoch": 1.6317808219178083, + "grad_norm": 2.3582253456115723, + "learning_rate": 2.5629569455727055e-06, + "loss": 1.7595, + "step": 1489 + }, + { + "epoch": 1.632876712328767, + "grad_norm": 2.522674083709717, + "learning_rate": 2.5609260763606824e-06, + "loss": 1.7908, + "step": 1490 + }, + { + "epoch": 1.632876712328767, + "eval_loss": 1.9791436195373535, + "eval_runtime": 37.2967, + "eval_samples_per_second": 26.812, + "eval_steps_per_second": 1.126, + "step": 1490 + }, + { + "epoch": 1.633972602739726, + "grad_norm": 2.2927470207214355, + "learning_rate": 2.55889520714866e-06, + "loss": 1.7789, + "step": 1491 + }, + { + "epoch": 1.635068493150685, + "grad_norm": 2.4469518661499023, + "learning_rate": 2.556864337936637e-06, + "loss": 1.8937, + "step": 1492 + }, + { + "epoch": 1.6361643835616437, + "grad_norm": 2.36860728263855, + "learning_rate": 2.5548334687246146e-06, + "loss": 1.778, + "step": 1493 + }, + { + "epoch": 1.6372602739726028, + "grad_norm": 2.4953925609588623, + "learning_rate": 2.5528025995125915e-06, + "loss": 1.7894, + "step": 1494 + }, + { + "epoch": 1.6383561643835618, + "grad_norm": 2.3478338718414307, + "learning_rate": 2.550771730300569e-06, + "loss": 1.8327, + "step": 1495 + }, + { + "epoch": 1.6394520547945204, + "grad_norm": 2.4399492740631104, + "learning_rate": 2.548740861088546e-06, + "loss": 1.8062, + "step": 1496 + }, + { + "epoch": 1.6405479452054794, + "grad_norm": 2.4457216262817383, + "learning_rate": 2.5467099918765233e-06, + "loss": 1.8268, + "step": 1497 + }, + { + "epoch": 1.6416438356164385, + "grad_norm": 2.365879535675049, + "learning_rate": 2.5446791226645006e-06, + "loss": 1.7288, + "step": 1498 + }, + { + "epoch": 1.6427397260273973, + "grad_norm": 2.4942879676818848, + "learning_rate": 2.542648253452478e-06, + "loss": 1.7111, + "step": 1499 + }, + { + "epoch": 1.643835616438356, + "grad_norm": 2.275982618331909, + "learning_rate": 2.540617384240455e-06, + "loss": 1.8292, + "step": 1500 + }, + { + "epoch": 1.643835616438356, + "eval_loss": 1.9746516942977905, + "eval_runtime": 37.2247, + "eval_samples_per_second": 26.864, + "eval_steps_per_second": 1.128, + "step": 1500 + }, + { + "epoch": 1.644931506849315, + "grad_norm": 2.4430429935455322, + "learning_rate": 2.5385865150284324e-06, + "loss": 1.8034, + "step": 1501 + }, + { + "epoch": 1.6460273972602741, + "grad_norm": 2.2778332233428955, + "learning_rate": 2.5365556458164097e-06, + "loss": 1.7589, + "step": 1502 + }, + { + "epoch": 1.6471232876712327, + "grad_norm": 2.2818830013275146, + "learning_rate": 2.534524776604387e-06, + "loss": 1.7743, + "step": 1503 + }, + { + "epoch": 1.6482191780821918, + "grad_norm": 2.321328639984131, + "learning_rate": 2.5324939073923642e-06, + "loss": 1.7922, + "step": 1504 + }, + { + "epoch": 1.6493150684931508, + "grad_norm": 2.4312188625335693, + "learning_rate": 2.5304630381803415e-06, + "loss": 1.6784, + "step": 1505 + }, + { + "epoch": 1.6504109589041096, + "grad_norm": 2.298741102218628, + "learning_rate": 2.5284321689683188e-06, + "loss": 1.7745, + "step": 1506 + }, + { + "epoch": 1.6515068493150684, + "grad_norm": 2.2882816791534424, + "learning_rate": 2.526401299756296e-06, + "loss": 1.7896, + "step": 1507 + }, + { + "epoch": 1.6526027397260274, + "grad_norm": 2.2336251735687256, + "learning_rate": 2.524370430544273e-06, + "loss": 1.7548, + "step": 1508 + }, + { + "epoch": 1.6536986301369863, + "grad_norm": 2.4369094371795654, + "learning_rate": 2.5223395613322506e-06, + "loss": 1.8049, + "step": 1509 + }, + { + "epoch": 1.654794520547945, + "grad_norm": 2.2978999614715576, + "learning_rate": 2.5203086921202274e-06, + "loss": 1.7489, + "step": 1510 + }, + { + "epoch": 1.654794520547945, + "eval_loss": 1.9760501384735107, + "eval_runtime": 37.3461, + "eval_samples_per_second": 26.777, + "eval_steps_per_second": 1.125, + "step": 1510 + }, + { + "epoch": 1.655890410958904, + "grad_norm": 2.331296920776367, + "learning_rate": 2.518277822908205e-06, + "loss": 1.8293, + "step": 1511 + }, + { + "epoch": 1.6569863013698631, + "grad_norm": 2.4544663429260254, + "learning_rate": 2.516246953696182e-06, + "loss": 1.7617, + "step": 1512 + }, + { + "epoch": 1.658082191780822, + "grad_norm": 2.4682042598724365, + "learning_rate": 2.5142160844841597e-06, + "loss": 1.8054, + "step": 1513 + }, + { + "epoch": 1.6591780821917808, + "grad_norm": 2.2827956676483154, + "learning_rate": 2.5121852152721365e-06, + "loss": 1.8137, + "step": 1514 + }, + { + "epoch": 1.6602739726027398, + "grad_norm": 2.2961761951446533, + "learning_rate": 2.5101543460601142e-06, + "loss": 1.7938, + "step": 1515 + }, + { + "epoch": 1.6613698630136986, + "grad_norm": 2.3464298248291016, + "learning_rate": 2.508123476848091e-06, + "loss": 1.8026, + "step": 1516 + }, + { + "epoch": 1.6624657534246574, + "grad_norm": 2.4975106716156006, + "learning_rate": 2.506092607636069e-06, + "loss": 1.7457, + "step": 1517 + }, + { + "epoch": 1.6635616438356164, + "grad_norm": 2.4823005199432373, + "learning_rate": 2.5040617384240457e-06, + "loss": 1.8282, + "step": 1518 + }, + { + "epoch": 1.6646575342465755, + "grad_norm": 2.2875800132751465, + "learning_rate": 2.5020308692120234e-06, + "loss": 1.7569, + "step": 1519 + }, + { + "epoch": 1.6657534246575343, + "grad_norm": 2.451022148132324, + "learning_rate": 2.5e-06, + "loss": 1.7483, + "step": 1520 + }, + { + "epoch": 1.6657534246575343, + "eval_loss": 1.9748642444610596, + "eval_runtime": 37.4512, + "eval_samples_per_second": 26.701, + "eval_steps_per_second": 1.121, + "step": 1520 + }, + { + "epoch": 1.666849315068493, + "grad_norm": 2.3427352905273438, + "learning_rate": 2.4979691307879775e-06, + "loss": 1.8493, + "step": 1521 + }, + { + "epoch": 1.6679452054794521, + "grad_norm": 2.4496946334838867, + "learning_rate": 2.4959382615759548e-06, + "loss": 1.7569, + "step": 1522 + }, + { + "epoch": 1.669041095890411, + "grad_norm": 2.141963481903076, + "learning_rate": 2.493907392363932e-06, + "loss": 1.7553, + "step": 1523 + }, + { + "epoch": 1.6701369863013698, + "grad_norm": 2.4115428924560547, + "learning_rate": 2.4918765231519093e-06, + "loss": 1.8606, + "step": 1524 + }, + { + "epoch": 1.6712328767123288, + "grad_norm": 2.614616632461548, + "learning_rate": 2.4898456539398866e-06, + "loss": 1.8217, + "step": 1525 + }, + { + "epoch": 1.6723287671232878, + "grad_norm": 2.338557243347168, + "learning_rate": 2.487814784727864e-06, + "loss": 1.7672, + "step": 1526 + }, + { + "epoch": 1.6734246575342466, + "grad_norm": 2.4309446811676025, + "learning_rate": 2.4857839155158407e-06, + "loss": 1.8476, + "step": 1527 + }, + { + "epoch": 1.6745205479452054, + "grad_norm": 2.25883412361145, + "learning_rate": 2.483753046303818e-06, + "loss": 1.7827, + "step": 1528 + }, + { + "epoch": 1.6756164383561645, + "grad_norm": 2.264845371246338, + "learning_rate": 2.4817221770917953e-06, + "loss": 1.7877, + "step": 1529 + }, + { + "epoch": 1.6767123287671233, + "grad_norm": 2.3584723472595215, + "learning_rate": 2.4796913078797725e-06, + "loss": 1.7891, + "step": 1530 + }, + { + "epoch": 1.6767123287671233, + "eval_loss": 1.9751919507980347, + "eval_runtime": 37.558, + "eval_samples_per_second": 26.625, + "eval_steps_per_second": 1.118, + "step": 1530 + }, + { + "epoch": 1.677808219178082, + "grad_norm": 2.3555185794830322, + "learning_rate": 2.47766043866775e-06, + "loss": 1.8235, + "step": 1531 + }, + { + "epoch": 1.6789041095890411, + "grad_norm": 2.3086600303649902, + "learning_rate": 2.475629569455727e-06, + "loss": 1.8036, + "step": 1532 + }, + { + "epoch": 1.6800000000000002, + "grad_norm": 2.286921262741089, + "learning_rate": 2.4735987002437044e-06, + "loss": 1.7885, + "step": 1533 + }, + { + "epoch": 1.6810958904109587, + "grad_norm": 2.2671141624450684, + "learning_rate": 2.4715678310316816e-06, + "loss": 1.758, + "step": 1534 + }, + { + "epoch": 1.6821917808219178, + "grad_norm": 2.4795749187469482, + "learning_rate": 2.469536961819659e-06, + "loss": 1.7791, + "step": 1535 + }, + { + "epoch": 1.6832876712328768, + "grad_norm": 2.201650857925415, + "learning_rate": 2.467506092607636e-06, + "loss": 1.7802, + "step": 1536 + }, + { + "epoch": 1.6843835616438356, + "grad_norm": 2.3804843425750732, + "learning_rate": 2.4654752233956135e-06, + "loss": 1.7739, + "step": 1537 + }, + { + "epoch": 1.6854794520547944, + "grad_norm": 2.322216510772705, + "learning_rate": 2.4634443541835907e-06, + "loss": 1.758, + "step": 1538 + }, + { + "epoch": 1.6865753424657535, + "grad_norm": 2.3582146167755127, + "learning_rate": 2.461413484971568e-06, + "loss": 1.8439, + "step": 1539 + }, + { + "epoch": 1.6876712328767123, + "grad_norm": 2.3597848415374756, + "learning_rate": 2.4593826157595453e-06, + "loss": 1.7772, + "step": 1540 + }, + { + "epoch": 1.6876712328767123, + "eval_loss": 1.9731802940368652, + "eval_runtime": 37.2439, + "eval_samples_per_second": 26.85, + "eval_steps_per_second": 1.128, + "step": 1540 + }, + { + "epoch": 1.688767123287671, + "grad_norm": 2.404569149017334, + "learning_rate": 2.4573517465475226e-06, + "loss": 1.8683, + "step": 1541 + }, + { + "epoch": 1.6898630136986301, + "grad_norm": 2.318737268447876, + "learning_rate": 2.4553208773355e-06, + "loss": 1.7223, + "step": 1542 + }, + { + "epoch": 1.6909589041095892, + "grad_norm": 2.2535340785980225, + "learning_rate": 2.453290008123477e-06, + "loss": 1.7693, + "step": 1543 + }, + { + "epoch": 1.692054794520548, + "grad_norm": 2.2533493041992188, + "learning_rate": 2.4512591389114544e-06, + "loss": 1.7876, + "step": 1544 + }, + { + "epoch": 1.6931506849315068, + "grad_norm": 2.253278970718384, + "learning_rate": 2.4492282696994317e-06, + "loss": 1.788, + "step": 1545 + }, + { + "epoch": 1.6942465753424658, + "grad_norm": 2.296807289123535, + "learning_rate": 2.447197400487409e-06, + "loss": 1.7877, + "step": 1546 + }, + { + "epoch": 1.6953424657534246, + "grad_norm": 2.4022111892700195, + "learning_rate": 2.4451665312753862e-06, + "loss": 1.7986, + "step": 1547 + }, + { + "epoch": 1.6964383561643834, + "grad_norm": 2.2671430110931396, + "learning_rate": 2.4431356620633635e-06, + "loss": 1.8269, + "step": 1548 + }, + { + "epoch": 1.6975342465753425, + "grad_norm": 2.338252067565918, + "learning_rate": 2.4411047928513408e-06, + "loss": 1.8279, + "step": 1549 + }, + { + "epoch": 1.6986301369863015, + "grad_norm": 2.365365505218506, + "learning_rate": 2.439073923639318e-06, + "loss": 1.7895, + "step": 1550 + }, + { + "epoch": 1.6986301369863015, + "eval_loss": 1.9711860418319702, + "eval_runtime": 37.2757, + "eval_samples_per_second": 26.827, + "eval_steps_per_second": 1.127, + "step": 1550 + }, + { + "epoch": 1.6997260273972603, + "grad_norm": 2.407580614089966, + "learning_rate": 2.4370430544272953e-06, + "loss": 1.7985, + "step": 1551 + }, + { + "epoch": 1.7008219178082191, + "grad_norm": 2.3145346641540527, + "learning_rate": 2.435012185215272e-06, + "loss": 1.7589, + "step": 1552 + }, + { + "epoch": 1.7019178082191782, + "grad_norm": 2.3809280395507812, + "learning_rate": 2.4329813160032495e-06, + "loss": 1.7334, + "step": 1553 + }, + { + "epoch": 1.703013698630137, + "grad_norm": 2.4224488735198975, + "learning_rate": 2.4309504467912267e-06, + "loss": 1.7549, + "step": 1554 + }, + { + "epoch": 1.7041095890410958, + "grad_norm": 2.374133348464966, + "learning_rate": 2.428919577579204e-06, + "loss": 1.7923, + "step": 1555 + }, + { + "epoch": 1.7052054794520548, + "grad_norm": 2.258222818374634, + "learning_rate": 2.4268887083671813e-06, + "loss": 1.7683, + "step": 1556 + }, + { + "epoch": 1.7063013698630138, + "grad_norm": 2.4481396675109863, + "learning_rate": 2.4248578391551586e-06, + "loss": 1.8609, + "step": 1557 + }, + { + "epoch": 1.7073972602739727, + "grad_norm": 2.3932790756225586, + "learning_rate": 2.422826969943136e-06, + "loss": 1.7475, + "step": 1558 + }, + { + "epoch": 1.7084931506849315, + "grad_norm": 2.3034355640411377, + "learning_rate": 2.420796100731113e-06, + "loss": 1.7664, + "step": 1559 + }, + { + "epoch": 1.7095890410958905, + "grad_norm": 2.2018799781799316, + "learning_rate": 2.4187652315190904e-06, + "loss": 1.7328, + "step": 1560 + }, + { + "epoch": 1.7095890410958905, + "eval_loss": 1.9715769290924072, + "eval_runtime": 37.2507, + "eval_samples_per_second": 26.845, + "eval_steps_per_second": 1.127, + "step": 1560 + }, + { + "epoch": 1.7106849315068493, + "grad_norm": 2.396462917327881, + "learning_rate": 2.4167343623070677e-06, + "loss": 1.7773, + "step": 1561 + }, + { + "epoch": 1.7117808219178081, + "grad_norm": 2.3212695121765137, + "learning_rate": 2.414703493095045e-06, + "loss": 1.7966, + "step": 1562 + }, + { + "epoch": 1.7128767123287671, + "grad_norm": 2.4157631397247314, + "learning_rate": 2.4126726238830222e-06, + "loss": 1.7797, + "step": 1563 + }, + { + "epoch": 1.7139726027397262, + "grad_norm": 2.404907464981079, + "learning_rate": 2.410641754670999e-06, + "loss": 1.8093, + "step": 1564 + }, + { + "epoch": 1.7150684931506848, + "grad_norm": 2.3798625469207764, + "learning_rate": 2.4086108854589763e-06, + "loss": 1.7635, + "step": 1565 + }, + { + "epoch": 1.7161643835616438, + "grad_norm": 2.325932025909424, + "learning_rate": 2.4065800162469536e-06, + "loss": 1.8281, + "step": 1566 + }, + { + "epoch": 1.7172602739726028, + "grad_norm": 2.413975238800049, + "learning_rate": 2.404549147034931e-06, + "loss": 1.7609, + "step": 1567 + }, + { + "epoch": 1.7183561643835616, + "grad_norm": 2.512775421142578, + "learning_rate": 2.402518277822908e-06, + "loss": 1.7793, + "step": 1568 + }, + { + "epoch": 1.7194520547945205, + "grad_norm": 2.581679582595825, + "learning_rate": 2.4004874086108854e-06, + "loss": 1.8302, + "step": 1569 + }, + { + "epoch": 1.7205479452054795, + "grad_norm": 2.389822244644165, + "learning_rate": 2.3984565393988627e-06, + "loss": 1.7573, + "step": 1570 + }, + { + "epoch": 1.7205479452054795, + "eval_loss": 1.9670331478118896, + "eval_runtime": 37.2901, + "eval_samples_per_second": 26.817, + "eval_steps_per_second": 1.126, + "step": 1570 + }, + { + "epoch": 1.7216438356164385, + "grad_norm": 2.6191556453704834, + "learning_rate": 2.39642567018684e-06, + "loss": 1.8252, + "step": 1571 + }, + { + "epoch": 1.7227397260273971, + "grad_norm": 2.394347906112671, + "learning_rate": 2.3943948009748173e-06, + "loss": 1.7561, + "step": 1572 + }, + { + "epoch": 1.7238356164383561, + "grad_norm": 2.3016326427459717, + "learning_rate": 2.3923639317627945e-06, + "loss": 1.7619, + "step": 1573 + }, + { + "epoch": 1.7249315068493152, + "grad_norm": 2.3663971424102783, + "learning_rate": 2.390333062550772e-06, + "loss": 1.7922, + "step": 1574 + }, + { + "epoch": 1.726027397260274, + "grad_norm": 2.3328938484191895, + "learning_rate": 2.388302193338749e-06, + "loss": 1.7374, + "step": 1575 + }, + { + "epoch": 1.7271232876712328, + "grad_norm": 2.313525676727295, + "learning_rate": 2.3862713241267264e-06, + "loss": 1.6585, + "step": 1576 + }, + { + "epoch": 1.7282191780821918, + "grad_norm": 2.4609642028808594, + "learning_rate": 2.3842404549147037e-06, + "loss": 1.8757, + "step": 1577 + }, + { + "epoch": 1.7293150684931506, + "grad_norm": 2.3502120971679688, + "learning_rate": 2.382209585702681e-06, + "loss": 1.7786, + "step": 1578 + }, + { + "epoch": 1.7304109589041095, + "grad_norm": 2.383025884628296, + "learning_rate": 2.380178716490658e-06, + "loss": 1.7502, + "step": 1579 + }, + { + "epoch": 1.7315068493150685, + "grad_norm": 2.280890464782715, + "learning_rate": 2.3781478472786355e-06, + "loss": 1.8192, + "step": 1580 + }, + { + "epoch": 1.7315068493150685, + "eval_loss": 1.9658632278442383, + "eval_runtime": 37.3714, + "eval_samples_per_second": 26.758, + "eval_steps_per_second": 1.124, + "step": 1580 + }, + { + "epoch": 1.7326027397260275, + "grad_norm": 2.342240333557129, + "learning_rate": 2.3761169780666128e-06, + "loss": 1.7786, + "step": 1581 + }, + { + "epoch": 1.7336986301369863, + "grad_norm": 2.2916715145111084, + "learning_rate": 2.37408610885459e-06, + "loss": 1.8161, + "step": 1582 + }, + { + "epoch": 1.7347945205479451, + "grad_norm": 2.24537992477417, + "learning_rate": 2.3720552396425673e-06, + "loss": 1.8442, + "step": 1583 + }, + { + "epoch": 1.7358904109589042, + "grad_norm": 2.3003411293029785, + "learning_rate": 2.3700243704305446e-06, + "loss": 1.7775, + "step": 1584 + }, + { + "epoch": 1.736986301369863, + "grad_norm": 2.3062400817871094, + "learning_rate": 2.367993501218522e-06, + "loss": 1.8374, + "step": 1585 + }, + { + "epoch": 1.7380821917808218, + "grad_norm": 2.3534083366394043, + "learning_rate": 2.365962632006499e-06, + "loss": 1.7829, + "step": 1586 + }, + { + "epoch": 1.7391780821917808, + "grad_norm": 2.2612221240997314, + "learning_rate": 2.3639317627944764e-06, + "loss": 1.7934, + "step": 1587 + }, + { + "epoch": 1.7402739726027399, + "grad_norm": 2.4197304248809814, + "learning_rate": 2.3619008935824537e-06, + "loss": 1.8226, + "step": 1588 + }, + { + "epoch": 1.7413698630136987, + "grad_norm": 2.435908079147339, + "learning_rate": 2.3598700243704305e-06, + "loss": 1.8148, + "step": 1589 + }, + { + "epoch": 1.7424657534246575, + "grad_norm": 2.3617560863494873, + "learning_rate": 2.357839155158408e-06, + "loss": 1.8085, + "step": 1590 + }, + { + "epoch": 1.7424657534246575, + "eval_loss": 1.9636695384979248, + "eval_runtime": 37.2968, + "eval_samples_per_second": 26.812, + "eval_steps_per_second": 1.126, + "step": 1590 + }, + { + "epoch": 1.7435616438356165, + "grad_norm": 2.295182228088379, + "learning_rate": 2.355808285946385e-06, + "loss": 1.8091, + "step": 1591 + }, + { + "epoch": 1.7446575342465753, + "grad_norm": 2.3347294330596924, + "learning_rate": 2.3537774167343624e-06, + "loss": 1.8086, + "step": 1592 + }, + { + "epoch": 1.7457534246575341, + "grad_norm": 2.4760799407958984, + "learning_rate": 2.3517465475223396e-06, + "loss": 1.8227, + "step": 1593 + }, + { + "epoch": 1.7468493150684932, + "grad_norm": 2.4243907928466797, + "learning_rate": 2.349715678310317e-06, + "loss": 1.7921, + "step": 1594 + }, + { + "epoch": 1.7479452054794522, + "grad_norm": 2.271735668182373, + "learning_rate": 2.347684809098294e-06, + "loss": 1.7987, + "step": 1595 + }, + { + "epoch": 1.749041095890411, + "grad_norm": 2.5412354469299316, + "learning_rate": 2.3456539398862715e-06, + "loss": 1.7558, + "step": 1596 + }, + { + "epoch": 1.7501369863013698, + "grad_norm": 2.464944362640381, + "learning_rate": 2.3436230706742487e-06, + "loss": 1.8087, + "step": 1597 + }, + { + "epoch": 1.7512328767123289, + "grad_norm": 2.4210381507873535, + "learning_rate": 2.341592201462226e-06, + "loss": 1.756, + "step": 1598 + }, + { + "epoch": 1.7523287671232877, + "grad_norm": 2.284945011138916, + "learning_rate": 2.3395613322502033e-06, + "loss": 1.7909, + "step": 1599 + }, + { + "epoch": 1.7534246575342465, + "grad_norm": 2.3740732669830322, + "learning_rate": 2.3375304630381806e-06, + "loss": 1.7683, + "step": 1600 + }, + { + "epoch": 1.7534246575342465, + "eval_loss": 1.9646408557891846, + "eval_runtime": 37.3266, + "eval_samples_per_second": 26.791, + "eval_steps_per_second": 1.125, + "step": 1600 + }, + { + "epoch": 1.7545205479452055, + "grad_norm": 2.399583578109741, + "learning_rate": 2.335499593826158e-06, + "loss": 1.7342, + "step": 1601 + }, + { + "epoch": 1.7556164383561645, + "grad_norm": 2.4731597900390625, + "learning_rate": 2.333468724614135e-06, + "loss": 1.7507, + "step": 1602 + }, + { + "epoch": 1.7567123287671231, + "grad_norm": 2.365433692932129, + "learning_rate": 2.3314378554021124e-06, + "loss": 1.7944, + "step": 1603 + }, + { + "epoch": 1.7578082191780822, + "grad_norm": 2.4500601291656494, + "learning_rate": 2.3294069861900897e-06, + "loss": 1.822, + "step": 1604 + }, + { + "epoch": 1.7589041095890412, + "grad_norm": 2.3339014053344727, + "learning_rate": 2.327376116978067e-06, + "loss": 1.7464, + "step": 1605 + }, + { + "epoch": 1.76, + "grad_norm": 2.343973159790039, + "learning_rate": 2.3253452477660442e-06, + "loss": 1.7408, + "step": 1606 + }, + { + "epoch": 1.7610958904109588, + "grad_norm": 2.453462839126587, + "learning_rate": 2.3233143785540215e-06, + "loss": 1.8261, + "step": 1607 + }, + { + "epoch": 1.7621917808219179, + "grad_norm": 2.357698678970337, + "learning_rate": 2.3212835093419988e-06, + "loss": 1.762, + "step": 1608 + }, + { + "epoch": 1.7632876712328767, + "grad_norm": 2.353257417678833, + "learning_rate": 2.319252640129976e-06, + "loss": 1.8076, + "step": 1609 + }, + { + "epoch": 1.7643835616438355, + "grad_norm": 2.3269948959350586, + "learning_rate": 2.3172217709179533e-06, + "loss": 1.7516, + "step": 1610 + }, + { + "epoch": 1.7643835616438355, + "eval_loss": 1.9678465127944946, + "eval_runtime": 37.2052, + "eval_samples_per_second": 26.878, + "eval_steps_per_second": 1.129, + "step": 1610 + }, + { + "epoch": 1.7654794520547945, + "grad_norm": 2.4023501873016357, + "learning_rate": 2.3151909017059306e-06, + "loss": 1.8477, + "step": 1611 + }, + { + "epoch": 1.7665753424657535, + "grad_norm": 2.3356990814208984, + "learning_rate": 2.313160032493908e-06, + "loss": 1.8236, + "step": 1612 + }, + { + "epoch": 1.7676712328767124, + "grad_norm": 2.3571391105651855, + "learning_rate": 2.311129163281885e-06, + "loss": 1.7335, + "step": 1613 + }, + { + "epoch": 1.7687671232876712, + "grad_norm": 2.3510208129882812, + "learning_rate": 2.309098294069862e-06, + "loss": 1.8088, + "step": 1614 + }, + { + "epoch": 1.7698630136986302, + "grad_norm": 2.3852577209472656, + "learning_rate": 2.3070674248578393e-06, + "loss": 1.7312, + "step": 1615 + }, + { + "epoch": 1.770958904109589, + "grad_norm": 2.3085618019104004, + "learning_rate": 2.3050365556458166e-06, + "loss": 1.7379, + "step": 1616 + }, + { + "epoch": 1.7720547945205478, + "grad_norm": 2.3454065322875977, + "learning_rate": 2.303005686433794e-06, + "loss": 1.8054, + "step": 1617 + }, + { + "epoch": 1.7731506849315068, + "grad_norm": 2.4034247398376465, + "learning_rate": 2.300974817221771e-06, + "loss": 1.7955, + "step": 1618 + }, + { + "epoch": 1.7742465753424659, + "grad_norm": 2.4181671142578125, + "learning_rate": 2.2989439480097484e-06, + "loss": 1.7732, + "step": 1619 + }, + { + "epoch": 1.7753424657534247, + "grad_norm": 2.3408010005950928, + "learning_rate": 2.2969130787977257e-06, + "loss": 1.7231, + "step": 1620 + }, + { + "epoch": 1.7753424657534247, + "eval_loss": 1.9655163288116455, + "eval_runtime": 37.4878, + "eval_samples_per_second": 26.675, + "eval_steps_per_second": 1.12, + "step": 1620 + }, + { + "epoch": 1.7764383561643835, + "grad_norm": 2.3113718032836914, + "learning_rate": 2.294882209585703e-06, + "loss": 1.7904, + "step": 1621 + }, + { + "epoch": 1.7775342465753425, + "grad_norm": 2.3037896156311035, + "learning_rate": 2.2928513403736802e-06, + "loss": 1.8069, + "step": 1622 + }, + { + "epoch": 1.7786301369863013, + "grad_norm": 2.3962693214416504, + "learning_rate": 2.2908204711616575e-06, + "loss": 1.7354, + "step": 1623 + }, + { + "epoch": 1.7797260273972602, + "grad_norm": 2.3499252796173096, + "learning_rate": 2.2887896019496348e-06, + "loss": 1.779, + "step": 1624 + }, + { + "epoch": 1.7808219178082192, + "grad_norm": 2.390958070755005, + "learning_rate": 2.286758732737612e-06, + "loss": 1.8257, + "step": 1625 + }, + { + "epoch": 1.7819178082191782, + "grad_norm": 2.365626573562622, + "learning_rate": 2.284727863525589e-06, + "loss": 1.7618, + "step": 1626 + }, + { + "epoch": 1.783013698630137, + "grad_norm": 2.3558530807495117, + "learning_rate": 2.282696994313566e-06, + "loss": 1.7729, + "step": 1627 + }, + { + "epoch": 1.7841095890410958, + "grad_norm": 2.3276045322418213, + "learning_rate": 2.2806661251015434e-06, + "loss": 1.8231, + "step": 1628 + }, + { + "epoch": 1.7852054794520549, + "grad_norm": 2.3369953632354736, + "learning_rate": 2.2786352558895207e-06, + "loss": 1.8011, + "step": 1629 + }, + { + "epoch": 1.7863013698630137, + "grad_norm": 2.3588311672210693, + "learning_rate": 2.276604386677498e-06, + "loss": 1.7682, + "step": 1630 + }, + { + "epoch": 1.7863013698630137, + "eval_loss": 1.963529348373413, + "eval_runtime": 37.5589, + "eval_samples_per_second": 26.625, + "eval_steps_per_second": 1.118, + "step": 1630 + }, + { + "epoch": 1.7873972602739725, + "grad_norm": 2.2918803691864014, + "learning_rate": 2.2745735174654753e-06, + "loss": 1.7256, + "step": 1631 + }, + { + "epoch": 1.7884931506849315, + "grad_norm": 2.423738479614258, + "learning_rate": 2.2725426482534526e-06, + "loss": 1.7582, + "step": 1632 + }, + { + "epoch": 1.7895890410958906, + "grad_norm": 2.3500850200653076, + "learning_rate": 2.27051177904143e-06, + "loss": 1.8255, + "step": 1633 + }, + { + "epoch": 1.7906849315068492, + "grad_norm": 2.3280837535858154, + "learning_rate": 2.268480909829407e-06, + "loss": 1.7845, + "step": 1634 + }, + { + "epoch": 1.7917808219178082, + "grad_norm": 2.5055465698242188, + "learning_rate": 2.2664500406173844e-06, + "loss": 1.78, + "step": 1635 + }, + { + "epoch": 1.7928767123287672, + "grad_norm": 2.3132143020629883, + "learning_rate": 2.2644191714053617e-06, + "loss": 1.7814, + "step": 1636 + }, + { + "epoch": 1.793972602739726, + "grad_norm": 2.316361427307129, + "learning_rate": 2.262388302193339e-06, + "loss": 1.8161, + "step": 1637 + }, + { + "epoch": 1.7950684931506848, + "grad_norm": 2.337318181991577, + "learning_rate": 2.260357432981316e-06, + "loss": 1.7942, + "step": 1638 + }, + { + "epoch": 1.7961643835616439, + "grad_norm": 2.362031936645508, + "learning_rate": 2.2583265637692935e-06, + "loss": 1.7771, + "step": 1639 + }, + { + "epoch": 1.7972602739726027, + "grad_norm": 2.3260183334350586, + "learning_rate": 2.2562956945572708e-06, + "loss": 1.7975, + "step": 1640 + }, + { + "epoch": 1.7972602739726027, + "eval_loss": 1.9619382619857788, + "eval_runtime": 37.2606, + "eval_samples_per_second": 26.838, + "eval_steps_per_second": 1.127, + "step": 1640 + }, + { + "epoch": 1.7983561643835615, + "grad_norm": 2.4260714054107666, + "learning_rate": 2.254264825345248e-06, + "loss": 1.7654, + "step": 1641 + }, + { + "epoch": 1.7994520547945205, + "grad_norm": 2.297428846359253, + "learning_rate": 2.2522339561332253e-06, + "loss": 1.775, + "step": 1642 + }, + { + "epoch": 1.8005479452054796, + "grad_norm": 2.5153353214263916, + "learning_rate": 2.2502030869212026e-06, + "loss": 1.7796, + "step": 1643 + }, + { + "epoch": 1.8016438356164384, + "grad_norm": 2.5102577209472656, + "learning_rate": 2.24817221770918e-06, + "loss": 1.7917, + "step": 1644 + }, + { + "epoch": 1.8027397260273972, + "grad_norm": 2.3693106174468994, + "learning_rate": 2.246141348497157e-06, + "loss": 1.8459, + "step": 1645 + }, + { + "epoch": 1.8038356164383562, + "grad_norm": 2.2353153228759766, + "learning_rate": 2.2441104792851344e-06, + "loss": 1.8095, + "step": 1646 + }, + { + "epoch": 1.804931506849315, + "grad_norm": 2.265465497970581, + "learning_rate": 2.2420796100731117e-06, + "loss": 1.7623, + "step": 1647 + }, + { + "epoch": 1.8060273972602738, + "grad_norm": 2.2820940017700195, + "learning_rate": 2.240048740861089e-06, + "loss": 1.7946, + "step": 1648 + }, + { + "epoch": 1.8071232876712329, + "grad_norm": 2.378035068511963, + "learning_rate": 2.2380178716490662e-06, + "loss": 1.8059, + "step": 1649 + }, + { + "epoch": 1.808219178082192, + "grad_norm": 2.328341245651245, + "learning_rate": 2.2359870024370435e-06, + "loss": 1.6611, + "step": 1650 + }, + { + "epoch": 1.808219178082192, + "eval_loss": 1.9600144624710083, + "eval_runtime": 37.2601, + "eval_samples_per_second": 26.838, + "eval_steps_per_second": 1.127, + "step": 1650 + }, + { + "epoch": 1.8093150684931507, + "grad_norm": 2.2819113731384277, + "learning_rate": 2.2339561332250204e-06, + "loss": 1.8058, + "step": 1651 + }, + { + "epoch": 1.8104109589041095, + "grad_norm": 2.4178121089935303, + "learning_rate": 2.2319252640129976e-06, + "loss": 1.8204, + "step": 1652 + }, + { + "epoch": 1.8115068493150686, + "grad_norm": 2.497008800506592, + "learning_rate": 2.229894394800975e-06, + "loss": 1.7968, + "step": 1653 + }, + { + "epoch": 1.8126027397260274, + "grad_norm": 2.4681501388549805, + "learning_rate": 2.227863525588952e-06, + "loss": 1.7292, + "step": 1654 + }, + { + "epoch": 1.8136986301369862, + "grad_norm": 2.2983527183532715, + "learning_rate": 2.2258326563769295e-06, + "loss": 1.7097, + "step": 1655 + }, + { + "epoch": 1.8147945205479452, + "grad_norm": 2.349454164505005, + "learning_rate": 2.2238017871649067e-06, + "loss": 1.7256, + "step": 1656 + }, + { + "epoch": 1.8158904109589042, + "grad_norm": 2.4247677326202393, + "learning_rate": 2.221770917952884e-06, + "loss": 1.7165, + "step": 1657 + }, + { + "epoch": 1.816986301369863, + "grad_norm": 2.3035342693328857, + "learning_rate": 2.2197400487408613e-06, + "loss": 1.7811, + "step": 1658 + }, + { + "epoch": 1.8180821917808219, + "grad_norm": 2.318772554397583, + "learning_rate": 2.2177091795288386e-06, + "loss": 1.8082, + "step": 1659 + }, + { + "epoch": 1.819178082191781, + "grad_norm": 2.2477188110351562, + "learning_rate": 2.215678310316816e-06, + "loss": 1.7206, + "step": 1660 + }, + { + "epoch": 1.819178082191781, + "eval_loss": 1.962982416152954, + "eval_runtime": 37.2709, + "eval_samples_per_second": 26.831, + "eval_steps_per_second": 1.127, + "step": 1660 + }, + { + "epoch": 1.8202739726027397, + "grad_norm": 2.355264186859131, + "learning_rate": 2.213647441104793e-06, + "loss": 1.7634, + "step": 1661 + }, + { + "epoch": 1.8213698630136985, + "grad_norm": 2.267458200454712, + "learning_rate": 2.2116165718927704e-06, + "loss": 1.7819, + "step": 1662 + }, + { + "epoch": 1.8224657534246576, + "grad_norm": 2.2487895488739014, + "learning_rate": 2.2095857026807473e-06, + "loss": 1.759, + "step": 1663 + }, + { + "epoch": 1.8235616438356166, + "grad_norm": 2.4253578186035156, + "learning_rate": 2.2075548334687245e-06, + "loss": 1.7937, + "step": 1664 + }, + { + "epoch": 1.8246575342465754, + "grad_norm": 2.478050470352173, + "learning_rate": 2.205523964256702e-06, + "loss": 1.7241, + "step": 1665 + }, + { + "epoch": 1.8257534246575342, + "grad_norm": 2.289905309677124, + "learning_rate": 2.203493095044679e-06, + "loss": 1.7531, + "step": 1666 + }, + { + "epoch": 1.8268493150684932, + "grad_norm": 2.303421974182129, + "learning_rate": 2.2014622258326564e-06, + "loss": 1.7711, + "step": 1667 + }, + { + "epoch": 1.827945205479452, + "grad_norm": 2.3670248985290527, + "learning_rate": 2.1994313566206336e-06, + "loss": 1.8093, + "step": 1668 + }, + { + "epoch": 1.8290410958904109, + "grad_norm": 2.3721048831939697, + "learning_rate": 2.197400487408611e-06, + "loss": 1.8273, + "step": 1669 + }, + { + "epoch": 1.83013698630137, + "grad_norm": 2.3000190258026123, + "learning_rate": 2.195369618196588e-06, + "loss": 1.75, + "step": 1670 + }, + { + "epoch": 1.83013698630137, + "eval_loss": 1.9577263593673706, + "eval_runtime": 37.2953, + "eval_samples_per_second": 26.813, + "eval_steps_per_second": 1.126, + "step": 1670 + }, + { + "epoch": 1.831232876712329, + "grad_norm": 2.303584575653076, + "learning_rate": 2.1933387489845655e-06, + "loss": 1.7278, + "step": 1671 + }, + { + "epoch": 1.8323287671232875, + "grad_norm": 2.260579824447632, + "learning_rate": 2.1913078797725427e-06, + "loss": 1.7524, + "step": 1672 + }, + { + "epoch": 1.8334246575342465, + "grad_norm": 2.311967372894287, + "learning_rate": 2.18927701056052e-06, + "loss": 1.7799, + "step": 1673 + }, + { + "epoch": 1.8345205479452056, + "grad_norm": 2.338989496231079, + "learning_rate": 2.1872461413484973e-06, + "loss": 1.8468, + "step": 1674 + }, + { + "epoch": 1.8356164383561644, + "grad_norm": 2.338989496231079, + "learning_rate": 2.1872461413484973e-06, + "loss": 1.7853, + "step": 1675 + }, + { + "epoch": 1.8367123287671232, + "grad_norm": 2.526724100112915, + "learning_rate": 2.1852152721364746e-06, + "loss": 1.78, + "step": 1676 + }, + { + "epoch": 1.8378082191780822, + "grad_norm": 2.4637107849121094, + "learning_rate": 2.183184402924452e-06, + "loss": 1.7328, + "step": 1677 + }, + { + "epoch": 1.838904109589041, + "grad_norm": 2.34454345703125, + "learning_rate": 2.181153533712429e-06, + "loss": 1.8058, + "step": 1678 + }, + { + "epoch": 1.8399999999999999, + "grad_norm": 2.5518760681152344, + "learning_rate": 2.1791226645004064e-06, + "loss": 1.7746, + "step": 1679 + }, + { + "epoch": 1.841095890410959, + "grad_norm": 2.437443971633911, + "learning_rate": 2.1770917952883837e-06, + "loss": 1.7517, + "step": 1680 + }, + { + "epoch": 1.841095890410959, + "eval_loss": 1.9572285413742065, + "eval_runtime": 37.3402, + "eval_samples_per_second": 26.781, + "eval_steps_per_second": 1.125, + "step": 1680 + }, + { + "epoch": 1.842191780821918, + "grad_norm": 2.344287157058716, + "learning_rate": 2.175060926076361e-06, + "loss": 1.8229, + "step": 1681 + }, + { + "epoch": 1.8432876712328767, + "grad_norm": 2.333840847015381, + "learning_rate": 2.1730300568643382e-06, + "loss": 1.8534, + "step": 1682 + }, + { + "epoch": 1.8443835616438355, + "grad_norm": 2.372926950454712, + "learning_rate": 2.1709991876523155e-06, + "loss": 1.7609, + "step": 1683 + }, + { + "epoch": 1.8454794520547946, + "grad_norm": 2.2858645915985107, + "learning_rate": 2.1689683184402928e-06, + "loss": 1.7508, + "step": 1684 + }, + { + "epoch": 1.8465753424657534, + "grad_norm": 2.326586961746216, + "learning_rate": 2.16693744922827e-06, + "loss": 1.8122, + "step": 1685 + }, + { + "epoch": 1.8476712328767122, + "grad_norm": 2.3605616092681885, + "learning_rate": 2.1649065800162473e-06, + "loss": 1.7794, + "step": 1686 + }, + { + "epoch": 1.8487671232876712, + "grad_norm": 2.4188039302825928, + "learning_rate": 2.1628757108042246e-06, + "loss": 1.7552, + "step": 1687 + }, + { + "epoch": 1.8498630136986303, + "grad_norm": 2.335617780685425, + "learning_rate": 2.160844841592202e-06, + "loss": 1.7259, + "step": 1688 + }, + { + "epoch": 1.850958904109589, + "grad_norm": 2.245882272720337, + "learning_rate": 2.1588139723801787e-06, + "loss": 1.7259, + "step": 1689 + }, + { + "epoch": 1.8520547945205479, + "grad_norm": 2.4447500705718994, + "learning_rate": 2.156783103168156e-06, + "loss": 1.7605, + "step": 1690 + }, + { + "epoch": 1.8520547945205479, + "eval_loss": 1.9608982801437378, + "eval_runtime": 37.3756, + "eval_samples_per_second": 26.755, + "eval_steps_per_second": 1.124, + "step": 1690 + }, + { + "epoch": 1.853150684931507, + "grad_norm": 2.363443613052368, + "learning_rate": 2.1547522339561333e-06, + "loss": 1.7336, + "step": 1691 + }, + { + "epoch": 1.8542465753424657, + "grad_norm": 2.3573989868164062, + "learning_rate": 2.1527213647441106e-06, + "loss": 1.7953, + "step": 1692 + }, + { + "epoch": 1.8553424657534245, + "grad_norm": 2.4292593002319336, + "learning_rate": 2.150690495532088e-06, + "loss": 1.8276, + "step": 1693 + }, + { + "epoch": 1.8564383561643836, + "grad_norm": 2.374563694000244, + "learning_rate": 2.148659626320065e-06, + "loss": 1.7975, + "step": 1694 + }, + { + "epoch": 1.8575342465753426, + "grad_norm": 2.3935306072235107, + "learning_rate": 2.1466287571080424e-06, + "loss": 1.7824, + "step": 1695 + }, + { + "epoch": 1.8586301369863014, + "grad_norm": 2.343773365020752, + "learning_rate": 2.1445978878960197e-06, + "loss": 1.7574, + "step": 1696 + }, + { + "epoch": 1.8597260273972602, + "grad_norm": 2.2765426635742188, + "learning_rate": 2.142567018683997e-06, + "loss": 1.7682, + "step": 1697 + }, + { + "epoch": 1.8608219178082193, + "grad_norm": 2.339470863342285, + "learning_rate": 2.140536149471974e-06, + "loss": 1.8229, + "step": 1698 + }, + { + "epoch": 1.861917808219178, + "grad_norm": 2.341775894165039, + "learning_rate": 2.1385052802599515e-06, + "loss": 1.7421, + "step": 1699 + }, + { + "epoch": 1.8630136986301369, + "grad_norm": 2.343876838684082, + "learning_rate": 2.1364744110479288e-06, + "loss": 1.7519, + "step": 1700 + }, + { + "epoch": 1.8630136986301369, + "eval_loss": 1.9587067365646362, + "eval_runtime": 37.4192, + "eval_samples_per_second": 26.724, + "eval_steps_per_second": 1.122, + "step": 1700 + }, + { + "epoch": 1.864109589041096, + "grad_norm": 2.388373613357544, + "learning_rate": 2.134443541835906e-06, + "loss": 1.8039, + "step": 1701 + }, + { + "epoch": 1.865205479452055, + "grad_norm": 2.379742383956909, + "learning_rate": 2.132412672623883e-06, + "loss": 1.7549, + "step": 1702 + }, + { + "epoch": 1.8663013698630135, + "grad_norm": 2.25833797454834, + "learning_rate": 2.13038180341186e-06, + "loss": 1.7349, + "step": 1703 + }, + { + "epoch": 1.8673972602739726, + "grad_norm": 2.4541401863098145, + "learning_rate": 2.1283509341998374e-06, + "loss": 1.7785, + "step": 1704 + }, + { + "epoch": 1.8684931506849316, + "grad_norm": 2.338456392288208, + "learning_rate": 2.1263200649878147e-06, + "loss": 1.8272, + "step": 1705 + }, + { + "epoch": 1.8695890410958904, + "grad_norm": 2.474292755126953, + "learning_rate": 2.124289195775792e-06, + "loss": 1.7497, + "step": 1706 + }, + { + "epoch": 1.8706849315068492, + "grad_norm": 2.206749439239502, + "learning_rate": 2.1222583265637693e-06, + "loss": 1.7697, + "step": 1707 + }, + { + "epoch": 1.8717808219178083, + "grad_norm": 2.378093957901001, + "learning_rate": 2.1202274573517465e-06, + "loss": 1.6851, + "step": 1708 + }, + { + "epoch": 1.872876712328767, + "grad_norm": 2.2442147731781006, + "learning_rate": 2.118196588139724e-06, + "loss": 1.7268, + "step": 1709 + }, + { + "epoch": 1.8739726027397259, + "grad_norm": 2.324852228164673, + "learning_rate": 2.116165718927701e-06, + "loss": 1.7369, + "step": 1710 + }, + { + "epoch": 1.8739726027397259, + "eval_loss": 1.9555498361587524, + "eval_runtime": 37.2599, + "eval_samples_per_second": 26.838, + "eval_steps_per_second": 1.127, + "step": 1710 + }, + { + "epoch": 1.875068493150685, + "grad_norm": 2.3466620445251465, + "learning_rate": 2.1141348497156784e-06, + "loss": 1.7642, + "step": 1711 + }, + { + "epoch": 1.876164383561644, + "grad_norm": 2.303654432296753, + "learning_rate": 2.1121039805036556e-06, + "loss": 1.7406, + "step": 1712 + }, + { + "epoch": 1.8772602739726028, + "grad_norm": 2.242719888687134, + "learning_rate": 2.110073111291633e-06, + "loss": 1.7269, + "step": 1713 + }, + { + "epoch": 1.8783561643835616, + "grad_norm": 2.4076626300811768, + "learning_rate": 2.10804224207961e-06, + "loss": 1.7749, + "step": 1714 + }, + { + "epoch": 1.8794520547945206, + "grad_norm": 2.510887861251831, + "learning_rate": 2.1060113728675875e-06, + "loss": 1.8056, + "step": 1715 + }, + { + "epoch": 1.8805479452054794, + "grad_norm": 2.4038093090057373, + "learning_rate": 2.1039805036555647e-06, + "loss": 1.7915, + "step": 1716 + }, + { + "epoch": 1.8816438356164382, + "grad_norm": 2.4485573768615723, + "learning_rate": 2.101949634443542e-06, + "loss": 1.6845, + "step": 1717 + }, + { + "epoch": 1.8827397260273973, + "grad_norm": 2.4619247913360596, + "learning_rate": 2.0999187652315193e-06, + "loss": 1.7597, + "step": 1718 + }, + { + "epoch": 1.8838356164383563, + "grad_norm": 2.5079665184020996, + "learning_rate": 2.0978878960194966e-06, + "loss": 1.7207, + "step": 1719 + }, + { + "epoch": 1.884931506849315, + "grad_norm": 2.38696551322937, + "learning_rate": 2.095857026807474e-06, + "loss": 1.7855, + "step": 1720 + }, + { + "epoch": 1.884931506849315, + "eval_loss": 1.9538354873657227, + "eval_runtime": 37.5269, + "eval_samples_per_second": 26.648, + "eval_steps_per_second": 1.119, + "step": 1720 + }, + { + "epoch": 1.886027397260274, + "grad_norm": 2.2652602195739746, + "learning_rate": 2.093826157595451e-06, + "loss": 1.7806, + "step": 1721 + }, + { + "epoch": 1.887123287671233, + "grad_norm": 2.4074227809906006, + "learning_rate": 2.0917952883834284e-06, + "loss": 1.7397, + "step": 1722 + }, + { + "epoch": 1.8882191780821918, + "grad_norm": 2.4302561283111572, + "learning_rate": 2.0897644191714057e-06, + "loss": 1.7883, + "step": 1723 + }, + { + "epoch": 1.8893150684931506, + "grad_norm": 2.311933755874634, + "learning_rate": 2.087733549959383e-06, + "loss": 1.8186, + "step": 1724 + }, + { + "epoch": 1.8904109589041096, + "grad_norm": 2.4028313159942627, + "learning_rate": 2.0857026807473602e-06, + "loss": 1.7208, + "step": 1725 + }, + { + "epoch": 1.8915068493150686, + "grad_norm": 2.2129130363464355, + "learning_rate": 2.0836718115353375e-06, + "loss": 1.7007, + "step": 1726 + }, + { + "epoch": 1.8926027397260274, + "grad_norm": 2.3893744945526123, + "learning_rate": 2.0816409423233144e-06, + "loss": 1.8281, + "step": 1727 + }, + { + "epoch": 1.8936986301369862, + "grad_norm": 2.2396240234375, + "learning_rate": 2.0796100731112916e-06, + "loss": 1.7252, + "step": 1728 + }, + { + "epoch": 1.8947945205479453, + "grad_norm": 2.4073901176452637, + "learning_rate": 2.077579203899269e-06, + "loss": 1.7533, + "step": 1729 + }, + { + "epoch": 1.895890410958904, + "grad_norm": 2.3669373989105225, + "learning_rate": 2.075548334687246e-06, + "loss": 1.7013, + "step": 1730 + }, + { + "epoch": 1.895890410958904, + "eval_loss": 1.95528244972229, + "eval_runtime": 37.6174, + "eval_samples_per_second": 26.583, + "eval_steps_per_second": 1.117, + "step": 1730 + }, + { + "epoch": 1.896986301369863, + "grad_norm": 2.5254933834075928, + "learning_rate": 2.0735174654752235e-06, + "loss": 1.7848, + "step": 1731 + }, + { + "epoch": 1.898082191780822, + "grad_norm": 2.4138967990875244, + "learning_rate": 2.0714865962632007e-06, + "loss": 1.7204, + "step": 1732 + }, + { + "epoch": 1.899178082191781, + "grad_norm": 2.417165994644165, + "learning_rate": 2.069455727051178e-06, + "loss": 1.7741, + "step": 1733 + }, + { + "epoch": 1.9002739726027398, + "grad_norm": 2.4026505947113037, + "learning_rate": 2.0674248578391553e-06, + "loss": 1.7456, + "step": 1734 + }, + { + "epoch": 1.9013698630136986, + "grad_norm": 2.4637675285339355, + "learning_rate": 2.0653939886271326e-06, + "loss": 1.7679, + "step": 1735 + }, + { + "epoch": 1.9024657534246576, + "grad_norm": 2.269869089126587, + "learning_rate": 2.06336311941511e-06, + "loss": 1.8059, + "step": 1736 + }, + { + "epoch": 1.9035616438356164, + "grad_norm": 2.321094512939453, + "learning_rate": 2.061332250203087e-06, + "loss": 1.743, + "step": 1737 + }, + { + "epoch": 1.9046575342465752, + "grad_norm": 2.477288246154785, + "learning_rate": 2.0593013809910644e-06, + "loss": 1.7143, + "step": 1738 + }, + { + "epoch": 1.9057534246575343, + "grad_norm": 2.3924875259399414, + "learning_rate": 2.0572705117790417e-06, + "loss": 1.75, + "step": 1739 + }, + { + "epoch": 1.9068493150684933, + "grad_norm": 2.4452881813049316, + "learning_rate": 2.055239642567019e-06, + "loss": 1.7831, + "step": 1740 + }, + { + "epoch": 1.9068493150684933, + "eval_loss": 1.9599454402923584, + "eval_runtime": 37.3026, + "eval_samples_per_second": 26.808, + "eval_steps_per_second": 1.126, + "step": 1740 + }, + { + "epoch": 1.907945205479452, + "grad_norm": 2.2778351306915283, + "learning_rate": 2.0532087733549962e-06, + "loss": 1.7633, + "step": 1741 + }, + { + "epoch": 1.909041095890411, + "grad_norm": 2.3171589374542236, + "learning_rate": 2.0511779041429735e-06, + "loss": 1.7726, + "step": 1742 + }, + { + "epoch": 1.91013698630137, + "grad_norm": 2.286094903945923, + "learning_rate": 2.0491470349309508e-06, + "loss": 1.7458, + "step": 1743 + }, + { + "epoch": 1.9112328767123288, + "grad_norm": 2.5381743907928467, + "learning_rate": 2.047116165718928e-06, + "loss": 1.777, + "step": 1744 + }, + { + "epoch": 1.9123287671232876, + "grad_norm": 2.3261141777038574, + "learning_rate": 2.0450852965069053e-06, + "loss": 1.7715, + "step": 1745 + }, + { + "epoch": 1.9134246575342466, + "grad_norm": 2.351914167404175, + "learning_rate": 2.0430544272948826e-06, + "loss": 1.8214, + "step": 1746 + }, + { + "epoch": 1.9145205479452054, + "grad_norm": 2.274226427078247, + "learning_rate": 2.04102355808286e-06, + "loss": 1.763, + "step": 1747 + }, + { + "epoch": 1.9156164383561642, + "grad_norm": 2.4025075435638428, + "learning_rate": 2.038992688870837e-06, + "loss": 1.7072, + "step": 1748 + }, + { + "epoch": 1.9167123287671233, + "grad_norm": 2.334024429321289, + "learning_rate": 2.0369618196588144e-06, + "loss": 1.7381, + "step": 1749 + }, + { + "epoch": 1.9178082191780823, + "grad_norm": 2.3011622428894043, + "learning_rate": 2.0349309504467917e-06, + "loss": 1.7432, + "step": 1750 + }, + { + "epoch": 1.9178082191780823, + "eval_loss": 1.958349347114563, + "eval_runtime": 37.2591, + "eval_samples_per_second": 26.839, + "eval_steps_per_second": 1.127, + "step": 1750 + }, + { + "epoch": 1.9189041095890411, + "grad_norm": 2.643063545227051, + "learning_rate": 2.032900081234769e-06, + "loss": 1.7931, + "step": 1751 + }, + { + "epoch": 1.92, + "grad_norm": 2.3357834815979004, + "learning_rate": 2.030869212022746e-06, + "loss": 1.7933, + "step": 1752 + }, + { + "epoch": 1.921095890410959, + "grad_norm": 2.2614479064941406, + "learning_rate": 2.028838342810723e-06, + "loss": 1.7611, + "step": 1753 + }, + { + "epoch": 1.9221917808219178, + "grad_norm": 2.3057546615600586, + "learning_rate": 2.0268074735987004e-06, + "loss": 1.7406, + "step": 1754 + }, + { + "epoch": 1.9232876712328766, + "grad_norm": 2.6138038635253906, + "learning_rate": 2.0247766043866777e-06, + "loss": 1.7303, + "step": 1755 + }, + { + "epoch": 1.9243835616438356, + "grad_norm": 2.3803296089172363, + "learning_rate": 2.022745735174655e-06, + "loss": 1.8023, + "step": 1756 + }, + { + "epoch": 1.9254794520547946, + "grad_norm": 2.3436431884765625, + "learning_rate": 2.020714865962632e-06, + "loss": 1.7654, + "step": 1757 + }, + { + "epoch": 1.9265753424657535, + "grad_norm": 2.4103317260742188, + "learning_rate": 2.0186839967506095e-06, + "loss": 1.7932, + "step": 1758 + }, + { + "epoch": 1.9276712328767123, + "grad_norm": 2.4537675380706787, + "learning_rate": 2.0166531275385868e-06, + "loss": 1.761, + "step": 1759 + }, + { + "epoch": 1.9287671232876713, + "grad_norm": 2.3816988468170166, + "learning_rate": 2.014622258326564e-06, + "loss": 1.7276, + "step": 1760 + }, + { + "epoch": 1.9287671232876713, + "eval_loss": 1.9558053016662598, + "eval_runtime": 37.2417, + "eval_samples_per_second": 26.852, + "eval_steps_per_second": 1.128, + "step": 1760 + }, + { + "epoch": 1.9298630136986301, + "grad_norm": 2.37233304977417, + "learning_rate": 2.0125913891145413e-06, + "loss": 1.7674, + "step": 1761 + }, + { + "epoch": 1.930958904109589, + "grad_norm": 2.4043285846710205, + "learning_rate": 2.0105605199025186e-06, + "loss": 1.7638, + "step": 1762 + }, + { + "epoch": 1.932054794520548, + "grad_norm": 2.3402836322784424, + "learning_rate": 2.008529650690496e-06, + "loss": 1.7269, + "step": 1763 + }, + { + "epoch": 1.933150684931507, + "grad_norm": 2.42535400390625, + "learning_rate": 2.0064987814784727e-06, + "loss": 1.7965, + "step": 1764 + }, + { + "epoch": 1.9342465753424658, + "grad_norm": 2.361659288406372, + "learning_rate": 2.00446791226645e-06, + "loss": 1.7467, + "step": 1765 + }, + { + "epoch": 1.9353424657534246, + "grad_norm": 2.455253839492798, + "learning_rate": 2.0024370430544273e-06, + "loss": 1.7359, + "step": 1766 + }, + { + "epoch": 1.9364383561643836, + "grad_norm": 2.3062705993652344, + "learning_rate": 2.0004061738424045e-06, + "loss": 1.751, + "step": 1767 + }, + { + "epoch": 1.9375342465753425, + "grad_norm": 2.3683881759643555, + "learning_rate": 1.998375304630382e-06, + "loss": 1.8032, + "step": 1768 + }, + { + "epoch": 1.9386301369863013, + "grad_norm": 2.2175045013427734, + "learning_rate": 1.996344435418359e-06, + "loss": 1.7629, + "step": 1769 + }, + { + "epoch": 1.9397260273972603, + "grad_norm": 2.4331724643707275, + "learning_rate": 1.9943135662063364e-06, + "loss": 1.7899, + "step": 1770 + }, + { + "epoch": 1.9397260273972603, + "eval_loss": 1.9554719924926758, + "eval_runtime": 37.2636, + "eval_samples_per_second": 26.836, + "eval_steps_per_second": 1.127, + "step": 1770 + }, + { + "epoch": 1.9408219178082193, + "grad_norm": 2.370785713195801, + "learning_rate": 1.9922826969943136e-06, + "loss": 1.7687, + "step": 1771 + }, + { + "epoch": 1.941917808219178, + "grad_norm": 2.4047751426696777, + "learning_rate": 1.990251827782291e-06, + "loss": 1.804, + "step": 1772 + }, + { + "epoch": 1.943013698630137, + "grad_norm": 2.4597785472869873, + "learning_rate": 1.988220958570268e-06, + "loss": 1.7731, + "step": 1773 + }, + { + "epoch": 1.944109589041096, + "grad_norm": 2.388749837875366, + "learning_rate": 1.9861900893582455e-06, + "loss": 1.7945, + "step": 1774 + }, + { + "epoch": 1.9452054794520548, + "grad_norm": 2.3419551849365234, + "learning_rate": 1.9841592201462227e-06, + "loss": 1.7987, + "step": 1775 + }, + { + "epoch": 1.9463013698630136, + "grad_norm": 2.43770432472229, + "learning_rate": 1.9821283509342e-06, + "loss": 1.7183, + "step": 1776 + }, + { + "epoch": 1.9473972602739726, + "grad_norm": 2.333385467529297, + "learning_rate": 1.9800974817221773e-06, + "loss": 1.7283, + "step": 1777 + }, + { + "epoch": 1.9484931506849315, + "grad_norm": 2.391052722930908, + "learning_rate": 1.9780666125101546e-06, + "loss": 1.776, + "step": 1778 + }, + { + "epoch": 1.9495890410958903, + "grad_norm": 2.375338077545166, + "learning_rate": 1.976035743298132e-06, + "loss": 1.7495, + "step": 1779 + }, + { + "epoch": 1.9506849315068493, + "grad_norm": 2.4482579231262207, + "learning_rate": 1.974004874086109e-06, + "loss": 1.8256, + "step": 1780 + }, + { + "epoch": 1.9506849315068493, + "eval_loss": 1.9600462913513184, + "eval_runtime": 37.2638, + "eval_samples_per_second": 26.836, + "eval_steps_per_second": 1.127, + "step": 1780 + }, + { + "epoch": 1.9517808219178083, + "grad_norm": 2.296039342880249, + "learning_rate": 1.9719740048740864e-06, + "loss": 1.6996, + "step": 1781 + }, + { + "epoch": 1.9528767123287671, + "grad_norm": 2.2494969367980957, + "learning_rate": 1.9699431356620637e-06, + "loss": 1.7401, + "step": 1782 + }, + { + "epoch": 1.953972602739726, + "grad_norm": 2.4078574180603027, + "learning_rate": 1.967912266450041e-06, + "loss": 1.7374, + "step": 1783 + }, + { + "epoch": 1.955068493150685, + "grad_norm": 2.4018476009368896, + "learning_rate": 1.9658813972380182e-06, + "loss": 1.731, + "step": 1784 + }, + { + "epoch": 1.9561643835616438, + "grad_norm": 2.3429362773895264, + "learning_rate": 1.9638505280259955e-06, + "loss": 1.8531, + "step": 1785 + }, + { + "epoch": 1.9572602739726026, + "grad_norm": 2.4287197589874268, + "learning_rate": 1.9618196588139728e-06, + "loss": 1.7271, + "step": 1786 + }, + { + "epoch": 1.9583561643835616, + "grad_norm": 2.338541030883789, + "learning_rate": 1.95978878960195e-06, + "loss": 1.7594, + "step": 1787 + }, + { + "epoch": 1.9594520547945207, + "grad_norm": 2.531425952911377, + "learning_rate": 1.9577579203899273e-06, + "loss": 1.8374, + "step": 1788 + }, + { + "epoch": 1.9605479452054795, + "grad_norm": 2.413638114929199, + "learning_rate": 1.955727051177904e-06, + "loss": 1.8178, + "step": 1789 + }, + { + "epoch": 1.9616438356164383, + "grad_norm": 2.303318977355957, + "learning_rate": 1.9536961819658815e-06, + "loss": 1.7429, + "step": 1790 + }, + { + "epoch": 1.9616438356164383, + "eval_loss": 1.9622509479522705, + "eval_runtime": 37.2699, + "eval_samples_per_second": 26.831, + "eval_steps_per_second": 1.127, + "step": 1790 + }, + { + "epoch": 1.9627397260273973, + "grad_norm": 2.400160074234009, + "learning_rate": 1.9516653127538587e-06, + "loss": 1.7136, + "step": 1791 + }, + { + "epoch": 1.9638356164383561, + "grad_norm": 2.3861589431762695, + "learning_rate": 1.949634443541836e-06, + "loss": 1.7146, + "step": 1792 + }, + { + "epoch": 1.964931506849315, + "grad_norm": 2.272711753845215, + "learning_rate": 1.9476035743298133e-06, + "loss": 1.7177, + "step": 1793 + }, + { + "epoch": 1.966027397260274, + "grad_norm": 2.4745068550109863, + "learning_rate": 1.9455727051177906e-06, + "loss": 1.7938, + "step": 1794 + }, + { + "epoch": 1.967123287671233, + "grad_norm": 2.4643466472625732, + "learning_rate": 1.943541835905768e-06, + "loss": 1.7601, + "step": 1795 + }, + { + "epoch": 1.9682191780821918, + "grad_norm": 2.5346457958221436, + "learning_rate": 1.941510966693745e-06, + "loss": 1.8132, + "step": 1796 + }, + { + "epoch": 1.9693150684931506, + "grad_norm": 2.280719041824341, + "learning_rate": 1.9394800974817224e-06, + "loss": 1.7101, + "step": 1797 + }, + { + "epoch": 1.9704109589041097, + "grad_norm": 2.3304555416107178, + "learning_rate": 1.9374492282696997e-06, + "loss": 1.7809, + "step": 1798 + }, + { + "epoch": 1.9715068493150685, + "grad_norm": 2.409796714782715, + "learning_rate": 1.935418359057677e-06, + "loss": 1.8127, + "step": 1799 + }, + { + "epoch": 1.9726027397260273, + "grad_norm": 2.208312749862671, + "learning_rate": 1.9333874898456542e-06, + "loss": 1.7752, + "step": 1800 + }, + { + "epoch": 1.9726027397260273, + "eval_loss": 1.96150541305542, + "eval_runtime": 37.3221, + "eval_samples_per_second": 26.794, + "eval_steps_per_second": 1.125, + "step": 1800 + }, + { + "epoch": 1.9736986301369863, + "grad_norm": 2.374393939971924, + "learning_rate": 1.931356620633631e-06, + "loss": 1.8188, + "step": 1801 + }, + { + "epoch": 1.9747945205479454, + "grad_norm": 2.3343780040740967, + "learning_rate": 1.9293257514216083e-06, + "loss": 1.7035, + "step": 1802 + }, + { + "epoch": 1.9758904109589042, + "grad_norm": 2.307894229888916, + "learning_rate": 1.9272948822095856e-06, + "loss": 1.7651, + "step": 1803 + }, + { + "epoch": 1.976986301369863, + "grad_norm": 2.387434244155884, + "learning_rate": 1.925264012997563e-06, + "loss": 1.7021, + "step": 1804 + }, + { + "epoch": 1.978082191780822, + "grad_norm": 2.3444128036499023, + "learning_rate": 1.92323314378554e-06, + "loss": 1.7585, + "step": 1805 + }, + { + "epoch": 1.9791780821917808, + "grad_norm": 2.371931791305542, + "learning_rate": 1.9212022745735174e-06, + "loss": 1.7932, + "step": 1806 + }, + { + "epoch": 1.9802739726027396, + "grad_norm": 2.3159632682800293, + "learning_rate": 1.9191714053614947e-06, + "loss": 1.722, + "step": 1807 + }, + { + "epoch": 1.9813698630136987, + "grad_norm": 2.4669857025146484, + "learning_rate": 1.917140536149472e-06, + "loss": 1.7326, + "step": 1808 + }, + { + "epoch": 1.9824657534246577, + "grad_norm": 2.365138292312622, + "learning_rate": 1.9151096669374493e-06, + "loss": 1.7013, + "step": 1809 + }, + { + "epoch": 1.9835616438356163, + "grad_norm": 2.539698362350464, + "learning_rate": 1.9130787977254266e-06, + "loss": 1.8082, + "step": 1810 + }, + { + "epoch": 1.9835616438356163, + "eval_loss": 1.9535884857177734, + "eval_runtime": 37.3497, + "eval_samples_per_second": 26.774, + "eval_steps_per_second": 1.125, + "step": 1810 + }, + { + "epoch": 1.9846575342465753, + "grad_norm": 2.386599540710449, + "learning_rate": 1.911047928513404e-06, + "loss": 1.7835, + "step": 1811 + }, + { + "epoch": 1.9857534246575343, + "grad_norm": 2.376086711883545, + "learning_rate": 1.909017059301381e-06, + "loss": 1.7161, + "step": 1812 + }, + { + "epoch": 1.9868493150684932, + "grad_norm": 2.5276663303375244, + "learning_rate": 1.9069861900893586e-06, + "loss": 1.7222, + "step": 1813 + }, + { + "epoch": 1.987945205479452, + "grad_norm": 2.466137170791626, + "learning_rate": 1.9049553208773354e-06, + "loss": 1.7841, + "step": 1814 + }, + { + "epoch": 1.989041095890411, + "grad_norm": 2.361128091812134, + "learning_rate": 1.9029244516653127e-06, + "loss": 1.7354, + "step": 1815 + }, + { + "epoch": 1.9901369863013698, + "grad_norm": 2.4348530769348145, + "learning_rate": 1.90089358245329e-06, + "loss": 1.6832, + "step": 1816 + }, + { + "epoch": 1.9912328767123286, + "grad_norm": 2.342285633087158, + "learning_rate": 1.8988627132412673e-06, + "loss": 1.7351, + "step": 1817 + }, + { + "epoch": 1.9923287671232877, + "grad_norm": 2.521137237548828, + "learning_rate": 1.8968318440292445e-06, + "loss": 1.794, + "step": 1818 + }, + { + "epoch": 1.9934246575342467, + "grad_norm": 2.5968360900878906, + "learning_rate": 1.8948009748172218e-06, + "loss": 1.6976, + "step": 1819 + }, + { + "epoch": 1.9945205479452055, + "grad_norm": 2.734095335006714, + "learning_rate": 1.892770105605199e-06, + "loss": 1.7919, + "step": 1820 + }, + { + "epoch": 1.9945205479452055, + "eval_loss": 1.9536023139953613, + "eval_runtime": 37.5805, + "eval_samples_per_second": 26.61, + "eval_steps_per_second": 1.118, + "step": 1820 + }, + { + "epoch": 1.9956164383561643, + "grad_norm": 2.327333927154541, + "learning_rate": 1.8907392363931764e-06, + "loss": 1.7891, + "step": 1821 + }, + { + "epoch": 1.9967123287671233, + "grad_norm": 2.2902543544769287, + "learning_rate": 1.8887083671811536e-06, + "loss": 1.7552, + "step": 1822 + }, + { + "epoch": 1.9978082191780822, + "grad_norm": 2.3587660789489746, + "learning_rate": 1.886677497969131e-06, + "loss": 1.7641, + "step": 1823 + }, + { + "epoch": 1.998904109589041, + "grad_norm": 2.4068408012390137, + "learning_rate": 1.8846466287571082e-06, + "loss": 1.6622, + "step": 1824 + }, + { + "epoch": 2.0, + "grad_norm": 2.453514575958252, + "learning_rate": 1.8826157595450855e-06, + "loss": 1.7962, + "step": 1825 + }, + { + "epoch": 2.001095890410959, + "grad_norm": 2.4694175720214844, + "learning_rate": 1.8805848903330625e-06, + "loss": 1.6823, + "step": 1826 + }, + { + "epoch": 2.0021917808219176, + "grad_norm": 2.594013214111328, + "learning_rate": 1.8785540211210398e-06, + "loss": 1.8453, + "step": 1827 + }, + { + "epoch": 2.0032876712328767, + "grad_norm": 2.455796718597412, + "learning_rate": 1.876523151909017e-06, + "loss": 1.7604, + "step": 1828 + }, + { + "epoch": 2.0043835616438357, + "grad_norm": 2.819857120513916, + "learning_rate": 1.8744922826969944e-06, + "loss": 1.7177, + "step": 1829 + }, + { + "epoch": 2.0054794520547947, + "grad_norm": 2.845775842666626, + "learning_rate": 1.8724614134849716e-06, + "loss": 1.7007, + "step": 1830 + }, + { + "epoch": 2.0054794520547947, + "eval_loss": 1.9594684839248657, + "eval_runtime": 37.6856, + "eval_samples_per_second": 26.535, + "eval_steps_per_second": 1.114, + "step": 1830 + }, + { + "epoch": 2.0065753424657533, + "grad_norm": 2.6129424571990967, + "learning_rate": 1.870430544272949e-06, + "loss": 1.7301, + "step": 1831 + }, + { + "epoch": 2.0076712328767123, + "grad_norm": 2.762805223464966, + "learning_rate": 1.8683996750609262e-06, + "loss": 1.8294, + "step": 1832 + }, + { + "epoch": 2.0087671232876714, + "grad_norm": 2.6397714614868164, + "learning_rate": 1.8663688058489035e-06, + "loss": 1.7176, + "step": 1833 + }, + { + "epoch": 2.00986301369863, + "grad_norm": 2.6637110710144043, + "learning_rate": 1.8643379366368807e-06, + "loss": 1.7363, + "step": 1834 + }, + { + "epoch": 2.010958904109589, + "grad_norm": 2.732534408569336, + "learning_rate": 1.862307067424858e-06, + "loss": 1.7496, + "step": 1835 + }, + { + "epoch": 2.012054794520548, + "grad_norm": 2.7163467407226562, + "learning_rate": 1.8602761982128353e-06, + "loss": 1.7551, + "step": 1836 + }, + { + "epoch": 2.013150684931507, + "grad_norm": 2.442484140396118, + "learning_rate": 1.8582453290008126e-06, + "loss": 1.7761, + "step": 1837 + }, + { + "epoch": 2.0142465753424657, + "grad_norm": 2.544834852218628, + "learning_rate": 1.8562144597887898e-06, + "loss": 1.7251, + "step": 1838 + }, + { + "epoch": 2.0153424657534247, + "grad_norm": 2.6588494777679443, + "learning_rate": 1.854183590576767e-06, + "loss": 1.7381, + "step": 1839 + }, + { + "epoch": 2.0164383561643837, + "grad_norm": 2.5374786853790283, + "learning_rate": 1.8521527213647442e-06, + "loss": 1.7831, + "step": 1840 + }, + { + "epoch": 2.0164383561643837, + "eval_loss": 1.9550480842590332, + "eval_runtime": 37.3312, + "eval_samples_per_second": 26.787, + "eval_steps_per_second": 1.125, + "step": 1840 + }, + { + "epoch": 2.0175342465753423, + "grad_norm": 2.4791345596313477, + "learning_rate": 1.8501218521527215e-06, + "loss": 1.6937, + "step": 1841 + }, + { + "epoch": 2.0186301369863013, + "grad_norm": 2.7243971824645996, + "learning_rate": 1.8480909829406987e-06, + "loss": 1.6674, + "step": 1842 + }, + { + "epoch": 2.0197260273972604, + "grad_norm": 2.618680477142334, + "learning_rate": 1.846060113728676e-06, + "loss": 1.6994, + "step": 1843 + }, + { + "epoch": 2.020821917808219, + "grad_norm": 2.597929000854492, + "learning_rate": 1.8440292445166533e-06, + "loss": 1.7375, + "step": 1844 + }, + { + "epoch": 2.021917808219178, + "grad_norm": 2.860177993774414, + "learning_rate": 1.8419983753046306e-06, + "loss": 1.7128, + "step": 1845 + }, + { + "epoch": 2.023013698630137, + "grad_norm": 2.606898307800293, + "learning_rate": 1.8399675060926078e-06, + "loss": 1.7735, + "step": 1846 + }, + { + "epoch": 2.024109589041096, + "grad_norm": 2.5152604579925537, + "learning_rate": 1.8379366368805851e-06, + "loss": 1.75, + "step": 1847 + }, + { + "epoch": 2.0252054794520546, + "grad_norm": 2.678071975708008, + "learning_rate": 1.8359057676685624e-06, + "loss": 1.7114, + "step": 1848 + }, + { + "epoch": 2.0263013698630137, + "grad_norm": 2.673515558242798, + "learning_rate": 1.8338748984565397e-06, + "loss": 1.6992, + "step": 1849 + }, + { + "epoch": 2.0273972602739727, + "grad_norm": 2.5902516841888428, + "learning_rate": 1.831844029244517e-06, + "loss": 1.7195, + "step": 1850 + }, + { + "epoch": 2.0273972602739727, + "eval_loss": 1.953579068183899, + "eval_runtime": 37.304, + "eval_samples_per_second": 26.807, + "eval_steps_per_second": 1.126, + "step": 1850 + }, + { + "epoch": 2.0284931506849313, + "grad_norm": 2.4912915229797363, + "learning_rate": 1.829813160032494e-06, + "loss": 1.6955, + "step": 1851 + }, + { + "epoch": 2.0295890410958903, + "grad_norm": 2.683821678161621, + "learning_rate": 1.8277822908204713e-06, + "loss": 1.7057, + "step": 1852 + }, + { + "epoch": 2.0306849315068494, + "grad_norm": 2.9541101455688477, + "learning_rate": 1.8257514216084486e-06, + "loss": 1.7377, + "step": 1853 + }, + { + "epoch": 2.0317808219178084, + "grad_norm": 2.783851146697998, + "learning_rate": 1.8237205523964258e-06, + "loss": 1.7197, + "step": 1854 + }, + { + "epoch": 2.032876712328767, + "grad_norm": 2.5224950313568115, + "learning_rate": 1.8216896831844031e-06, + "loss": 1.7396, + "step": 1855 + }, + { + "epoch": 2.033972602739726, + "grad_norm": 2.627767562866211, + "learning_rate": 1.8196588139723804e-06, + "loss": 1.7773, + "step": 1856 + }, + { + "epoch": 2.035068493150685, + "grad_norm": 2.443873405456543, + "learning_rate": 1.8176279447603577e-06, + "loss": 1.6656, + "step": 1857 + }, + { + "epoch": 2.0361643835616436, + "grad_norm": 2.648056745529175, + "learning_rate": 1.815597075548335e-06, + "loss": 1.6954, + "step": 1858 + }, + { + "epoch": 2.0372602739726027, + "grad_norm": 2.536935329437256, + "learning_rate": 1.8135662063363122e-06, + "loss": 1.7427, + "step": 1859 + }, + { + "epoch": 2.0383561643835617, + "grad_norm": 2.42183256149292, + "learning_rate": 1.8115353371242895e-06, + "loss": 1.7634, + "step": 1860 + }, + { + "epoch": 2.0383561643835617, + "eval_loss": 1.953397512435913, + "eval_runtime": 37.2886, + "eval_samples_per_second": 26.818, + "eval_steps_per_second": 1.126, + "step": 1860 + }, + { + "epoch": 2.0394520547945207, + "grad_norm": 2.538804769515991, + "learning_rate": 1.8095044679122668e-06, + "loss": 1.6712, + "step": 1861 + }, + { + "epoch": 2.0405479452054793, + "grad_norm": 2.5820095539093018, + "learning_rate": 1.807473598700244e-06, + "loss": 1.7215, + "step": 1862 + }, + { + "epoch": 2.0416438356164384, + "grad_norm": 2.6997382640838623, + "learning_rate": 1.805442729488221e-06, + "loss": 1.7117, + "step": 1863 + }, + { + "epoch": 2.0427397260273974, + "grad_norm": 2.775996685028076, + "learning_rate": 1.8034118602761982e-06, + "loss": 1.7355, + "step": 1864 + }, + { + "epoch": 2.043835616438356, + "grad_norm": 2.6333167552948, + "learning_rate": 1.8013809910641754e-06, + "loss": 1.7343, + "step": 1865 + }, + { + "epoch": 2.044931506849315, + "grad_norm": 2.5830278396606445, + "learning_rate": 1.7993501218521527e-06, + "loss": 1.7339, + "step": 1866 + }, + { + "epoch": 2.046027397260274, + "grad_norm": 2.6145055294036865, + "learning_rate": 1.79731925264013e-06, + "loss": 1.6835, + "step": 1867 + }, + { + "epoch": 2.047123287671233, + "grad_norm": 2.568159580230713, + "learning_rate": 1.7952883834281073e-06, + "loss": 1.7574, + "step": 1868 + }, + { + "epoch": 2.0482191780821917, + "grad_norm": 2.5235254764556885, + "learning_rate": 1.7932575142160846e-06, + "loss": 1.7067, + "step": 1869 + }, + { + "epoch": 2.0493150684931507, + "grad_norm": 2.464035987854004, + "learning_rate": 1.7912266450040618e-06, + "loss": 1.7451, + "step": 1870 + }, + { + "epoch": 2.0493150684931507, + "eval_loss": 1.9485031366348267, + "eval_runtime": 37.2648, + "eval_samples_per_second": 26.835, + "eval_steps_per_second": 1.127, + "step": 1870 + }, + { + "epoch": 2.0504109589041097, + "grad_norm": 2.4328689575195312, + "learning_rate": 1.789195775792039e-06, + "loss": 1.7508, + "step": 1871 + }, + { + "epoch": 2.0515068493150683, + "grad_norm": 2.5483932495117188, + "learning_rate": 1.7871649065800164e-06, + "loss": 1.6488, + "step": 1872 + }, + { + "epoch": 2.0526027397260274, + "grad_norm": 2.5290846824645996, + "learning_rate": 1.7851340373679937e-06, + "loss": 1.7939, + "step": 1873 + }, + { + "epoch": 2.0536986301369864, + "grad_norm": 2.506943702697754, + "learning_rate": 1.783103168155971e-06, + "loss": 1.6813, + "step": 1874 + }, + { + "epoch": 2.0547945205479454, + "grad_norm": 2.5863115787506104, + "learning_rate": 1.7810722989439482e-06, + "loss": 1.7104, + "step": 1875 + }, + { + "epoch": 2.055890410958904, + "grad_norm": 2.6402666568756104, + "learning_rate": 1.7790414297319253e-06, + "loss": 1.7311, + "step": 1876 + }, + { + "epoch": 2.056986301369863, + "grad_norm": 2.6700708866119385, + "learning_rate": 1.7770105605199025e-06, + "loss": 1.695, + "step": 1877 + }, + { + "epoch": 2.058082191780822, + "grad_norm": 2.580693244934082, + "learning_rate": 1.7749796913078798e-06, + "loss": 1.6763, + "step": 1878 + }, + { + "epoch": 2.0591780821917807, + "grad_norm": 2.6035709381103516, + "learning_rate": 1.772948822095857e-06, + "loss": 1.7389, + "step": 1879 + }, + { + "epoch": 2.0602739726027397, + "grad_norm": 2.508784532546997, + "learning_rate": 1.7709179528838344e-06, + "loss": 1.7762, + "step": 1880 + }, + { + "epoch": 2.0602739726027397, + "eval_loss": 1.9516879320144653, + "eval_runtime": 37.2284, + "eval_samples_per_second": 26.861, + "eval_steps_per_second": 1.128, + "step": 1880 + }, + { + "epoch": 2.0613698630136987, + "grad_norm": 2.47988224029541, + "learning_rate": 1.7688870836718116e-06, + "loss": 1.7659, + "step": 1881 + }, + { + "epoch": 2.0624657534246573, + "grad_norm": 2.5782723426818848, + "learning_rate": 1.766856214459789e-06, + "loss": 1.7179, + "step": 1882 + }, + { + "epoch": 2.0635616438356164, + "grad_norm": 2.5616347789764404, + "learning_rate": 1.7648253452477662e-06, + "loss": 1.7163, + "step": 1883 + }, + { + "epoch": 2.0646575342465754, + "grad_norm": 2.4464340209960938, + "learning_rate": 1.7627944760357435e-06, + "loss": 1.7033, + "step": 1884 + }, + { + "epoch": 2.0657534246575344, + "grad_norm": 2.6022887229919434, + "learning_rate": 1.7607636068237208e-06, + "loss": 1.7823, + "step": 1885 + }, + { + "epoch": 2.066849315068493, + "grad_norm": 2.5734951496124268, + "learning_rate": 1.758732737611698e-06, + "loss": 1.7486, + "step": 1886 + }, + { + "epoch": 2.067945205479452, + "grad_norm": 2.6162502765655518, + "learning_rate": 1.7567018683996753e-06, + "loss": 1.7285, + "step": 1887 + }, + { + "epoch": 2.069041095890411, + "grad_norm": 2.654008388519287, + "learning_rate": 1.7546709991876524e-06, + "loss": 1.7195, + "step": 1888 + }, + { + "epoch": 2.0701369863013697, + "grad_norm": 2.64469575881958, + "learning_rate": 1.7526401299756296e-06, + "loss": 1.7247, + "step": 1889 + }, + { + "epoch": 2.0712328767123287, + "grad_norm": 2.5627946853637695, + "learning_rate": 1.750609260763607e-06, + "loss": 1.7254, + "step": 1890 + }, + { + "epoch": 2.0712328767123287, + "eval_loss": 1.951324224472046, + "eval_runtime": 37.2566, + "eval_samples_per_second": 26.841, + "eval_steps_per_second": 1.127, + "step": 1890 + }, + { + "epoch": 2.0723287671232877, + "grad_norm": 2.656520128250122, + "learning_rate": 1.7485783915515842e-06, + "loss": 1.7236, + "step": 1891 + }, + { + "epoch": 2.0734246575342468, + "grad_norm": 2.6355814933776855, + "learning_rate": 1.7465475223395615e-06, + "loss": 1.7118, + "step": 1892 + }, + { + "epoch": 2.0745205479452054, + "grad_norm": 2.5709362030029297, + "learning_rate": 1.7445166531275387e-06, + "loss": 1.685, + "step": 1893 + }, + { + "epoch": 2.0756164383561644, + "grad_norm": 2.555422782897949, + "learning_rate": 1.742485783915516e-06, + "loss": 1.8356, + "step": 1894 + }, + { + "epoch": 2.0767123287671234, + "grad_norm": 2.6279077529907227, + "learning_rate": 1.7404549147034933e-06, + "loss": 1.8079, + "step": 1895 + }, + { + "epoch": 2.077808219178082, + "grad_norm": 2.5886247158050537, + "learning_rate": 1.7384240454914706e-06, + "loss": 1.639, + "step": 1896 + }, + { + "epoch": 2.078904109589041, + "grad_norm": 2.562755584716797, + "learning_rate": 1.7363931762794478e-06, + "loss": 1.7345, + "step": 1897 + }, + { + "epoch": 2.08, + "grad_norm": 2.670344591140747, + "learning_rate": 1.7343623070674251e-06, + "loss": 1.7594, + "step": 1898 + }, + { + "epoch": 2.081095890410959, + "grad_norm": 2.567506790161133, + "learning_rate": 1.7323314378554024e-06, + "loss": 1.7161, + "step": 1899 + }, + { + "epoch": 2.0821917808219177, + "grad_norm": 2.562826156616211, + "learning_rate": 1.7303005686433797e-06, + "loss": 1.7275, + "step": 1900 + }, + { + "epoch": 2.0821917808219177, + "eval_loss": 1.9498006105422974, + "eval_runtime": 37.3069, + "eval_samples_per_second": 26.805, + "eval_steps_per_second": 1.126, + "step": 1900 + }, + { + "epoch": 2.0832876712328767, + "grad_norm": 2.538393974304199, + "learning_rate": 1.7282696994313567e-06, + "loss": 1.7187, + "step": 1901 + }, + { + "epoch": 2.0843835616438358, + "grad_norm": 2.687603235244751, + "learning_rate": 1.726238830219334e-06, + "loss": 1.7349, + "step": 1902 + }, + { + "epoch": 2.0854794520547943, + "grad_norm": 2.5806703567504883, + "learning_rate": 1.7242079610073113e-06, + "loss": 1.7135, + "step": 1903 + }, + { + "epoch": 2.0865753424657534, + "grad_norm": 2.468123435974121, + "learning_rate": 1.7221770917952886e-06, + "loss": 1.7946, + "step": 1904 + }, + { + "epoch": 2.0876712328767124, + "grad_norm": 2.514617443084717, + "learning_rate": 1.7201462225832658e-06, + "loss": 1.7403, + "step": 1905 + }, + { + "epoch": 2.0887671232876714, + "grad_norm": 2.7283825874328613, + "learning_rate": 1.7181153533712431e-06, + "loss": 1.7369, + "step": 1906 + }, + { + "epoch": 2.08986301369863, + "grad_norm": 2.5634899139404297, + "learning_rate": 1.7160844841592204e-06, + "loss": 1.6337, + "step": 1907 + }, + { + "epoch": 2.090958904109589, + "grad_norm": 2.635281801223755, + "learning_rate": 1.7140536149471977e-06, + "loss": 1.6836, + "step": 1908 + }, + { + "epoch": 2.092054794520548, + "grad_norm": 2.550048351287842, + "learning_rate": 1.712022745735175e-06, + "loss": 1.7171, + "step": 1909 + }, + { + "epoch": 2.0931506849315067, + "grad_norm": 2.62712025642395, + "learning_rate": 1.7099918765231522e-06, + "loss": 1.7027, + "step": 1910 + }, + { + "epoch": 2.0931506849315067, + "eval_loss": 1.9503322839736938, + "eval_runtime": 37.3435, + "eval_samples_per_second": 26.778, + "eval_steps_per_second": 1.125, + "step": 1910 + }, + { + "epoch": 2.0942465753424657, + "grad_norm": 2.540708541870117, + "learning_rate": 1.7079610073111295e-06, + "loss": 1.7358, + "step": 1911 + }, + { + "epoch": 2.0953424657534248, + "grad_norm": 2.6046950817108154, + "learning_rate": 1.7059301380991068e-06, + "loss": 1.6538, + "step": 1912 + }, + { + "epoch": 2.0964383561643833, + "grad_norm": 2.4760568141937256, + "learning_rate": 1.7038992688870836e-06, + "loss": 1.7207, + "step": 1913 + }, + { + "epoch": 2.0975342465753424, + "grad_norm": 2.428621292114258, + "learning_rate": 1.701868399675061e-06, + "loss": 1.7092, + "step": 1914 + }, + { + "epoch": 2.0986301369863014, + "grad_norm": 2.6177377700805664, + "learning_rate": 1.6998375304630382e-06, + "loss": 1.7429, + "step": 1915 + }, + { + "epoch": 2.0997260273972604, + "grad_norm": 2.4981932640075684, + "learning_rate": 1.6978066612510155e-06, + "loss": 1.7996, + "step": 1916 + }, + { + "epoch": 2.100821917808219, + "grad_norm": 2.617943286895752, + "learning_rate": 1.6957757920389927e-06, + "loss": 1.7716, + "step": 1917 + }, + { + "epoch": 2.101917808219178, + "grad_norm": 2.7581987380981445, + "learning_rate": 1.69374492282697e-06, + "loss": 1.7194, + "step": 1918 + }, + { + "epoch": 2.103013698630137, + "grad_norm": 2.622145652770996, + "learning_rate": 1.6917140536149473e-06, + "loss": 1.6887, + "step": 1919 + }, + { + "epoch": 2.1041095890410957, + "grad_norm": 2.742100477218628, + "learning_rate": 1.6896831844029246e-06, + "loss": 1.7787, + "step": 1920 + }, + { + "epoch": 2.1041095890410957, + "eval_loss": 1.9521653652191162, + "eval_runtime": 37.5529, + "eval_samples_per_second": 26.629, + "eval_steps_per_second": 1.118, + "step": 1920 + }, + { + "epoch": 2.1052054794520547, + "grad_norm": 2.5809876918792725, + "learning_rate": 1.6876523151909018e-06, + "loss": 1.7392, + "step": 1921 + }, + { + "epoch": 2.1063013698630138, + "grad_norm": 2.7006125450134277, + "learning_rate": 1.6856214459788791e-06, + "loss": 1.7442, + "step": 1922 + }, + { + "epoch": 2.107397260273973, + "grad_norm": 2.566088914871216, + "learning_rate": 1.6835905767668564e-06, + "loss": 1.6862, + "step": 1923 + }, + { + "epoch": 2.1084931506849314, + "grad_norm": 2.508967638015747, + "learning_rate": 1.6815597075548337e-06, + "loss": 1.695, + "step": 1924 + }, + { + "epoch": 2.1095890410958904, + "grad_norm": 2.6175615787506104, + "learning_rate": 1.679528838342811e-06, + "loss": 1.785, + "step": 1925 + }, + { + "epoch": 2.1106849315068494, + "grad_norm": 2.6141958236694336, + "learning_rate": 1.677497969130788e-06, + "loss": 1.6952, + "step": 1926 + }, + { + "epoch": 2.111780821917808, + "grad_norm": 2.4894959926605225, + "learning_rate": 1.6754670999187653e-06, + "loss": 1.6947, + "step": 1927 + }, + { + "epoch": 2.112876712328767, + "grad_norm": 2.5664544105529785, + "learning_rate": 1.6734362307067426e-06, + "loss": 1.731, + "step": 1928 + }, + { + "epoch": 2.113972602739726, + "grad_norm": 2.5422523021698, + "learning_rate": 1.6714053614947198e-06, + "loss": 1.6765, + "step": 1929 + }, + { + "epoch": 2.115068493150685, + "grad_norm": 2.6428608894348145, + "learning_rate": 1.669374492282697e-06, + "loss": 1.6834, + "step": 1930 + }, + { + "epoch": 2.115068493150685, + "eval_loss": 1.952999472618103, + "eval_runtime": 37.6765, + "eval_samples_per_second": 26.542, + "eval_steps_per_second": 1.115, + "step": 1930 + }, + { + "epoch": 2.1161643835616437, + "grad_norm": 2.6088430881500244, + "learning_rate": 1.6673436230706744e-06, + "loss": 1.7713, + "step": 1931 + }, + { + "epoch": 2.1172602739726027, + "grad_norm": 2.49733304977417, + "learning_rate": 1.6653127538586517e-06, + "loss": 1.7461, + "step": 1932 + }, + { + "epoch": 2.118356164383562, + "grad_norm": 2.5760915279388428, + "learning_rate": 1.663281884646629e-06, + "loss": 1.6837, + "step": 1933 + }, + { + "epoch": 2.1194520547945204, + "grad_norm": 2.4771194458007812, + "learning_rate": 1.6612510154346062e-06, + "loss": 1.7191, + "step": 1934 + }, + { + "epoch": 2.1205479452054794, + "grad_norm": 2.6267881393432617, + "learning_rate": 1.6592201462225835e-06, + "loss": 1.7819, + "step": 1935 + }, + { + "epoch": 2.1216438356164384, + "grad_norm": 2.6205623149871826, + "learning_rate": 1.6571892770105608e-06, + "loss": 1.7443, + "step": 1936 + }, + { + "epoch": 2.1227397260273975, + "grad_norm": 2.660942792892456, + "learning_rate": 1.655158407798538e-06, + "loss": 1.783, + "step": 1937 + }, + { + "epoch": 2.123835616438356, + "grad_norm": 2.5452628135681152, + "learning_rate": 1.653127538586515e-06, + "loss": 1.7316, + "step": 1938 + }, + { + "epoch": 2.124931506849315, + "grad_norm": 2.542837142944336, + "learning_rate": 1.6510966693744924e-06, + "loss": 1.7624, + "step": 1939 + }, + { + "epoch": 2.126027397260274, + "grad_norm": 2.535205841064453, + "learning_rate": 1.6490658001624696e-06, + "loss": 1.7664, + "step": 1940 + }, + { + "epoch": 2.126027397260274, + "eval_loss": 1.9492406845092773, + "eval_runtime": 37.3497, + "eval_samples_per_second": 26.774, + "eval_steps_per_second": 1.125, + "step": 1940 + }, + { + "epoch": 2.1271232876712327, + "grad_norm": 2.7622485160827637, + "learning_rate": 1.647034930950447e-06, + "loss": 1.7603, + "step": 1941 + }, + { + "epoch": 2.1282191780821917, + "grad_norm": 2.629058599472046, + "learning_rate": 1.6450040617384242e-06, + "loss": 1.7036, + "step": 1942 + }, + { + "epoch": 2.1293150684931508, + "grad_norm": 2.543135643005371, + "learning_rate": 1.6429731925264015e-06, + "loss": 1.7172, + "step": 1943 + }, + { + "epoch": 2.1304109589041094, + "grad_norm": 2.792086601257324, + "learning_rate": 1.6409423233143788e-06, + "loss": 1.7486, + "step": 1944 + }, + { + "epoch": 2.1315068493150684, + "grad_norm": 2.7788681983947754, + "learning_rate": 1.638911454102356e-06, + "loss": 1.7136, + "step": 1945 + }, + { + "epoch": 2.1326027397260274, + "grad_norm": 2.5364747047424316, + "learning_rate": 1.6368805848903333e-06, + "loss": 1.6766, + "step": 1946 + }, + { + "epoch": 2.1336986301369865, + "grad_norm": 2.5749800205230713, + "learning_rate": 1.6348497156783106e-06, + "loss": 1.7627, + "step": 1947 + }, + { + "epoch": 2.134794520547945, + "grad_norm": 2.4931676387786865, + "learning_rate": 1.6328188464662879e-06, + "loss": 1.7049, + "step": 1948 + }, + { + "epoch": 2.135890410958904, + "grad_norm": 2.547396183013916, + "learning_rate": 1.6307879772542651e-06, + "loss": 1.7165, + "step": 1949 + }, + { + "epoch": 2.136986301369863, + "grad_norm": 2.5881359577178955, + "learning_rate": 1.628757108042242e-06, + "loss": 1.7508, + "step": 1950 + }, + { + "epoch": 2.136986301369863, + "eval_loss": 1.9464337825775146, + "eval_runtime": 37.284, + "eval_samples_per_second": 26.821, + "eval_steps_per_second": 1.126, + "step": 1950 + }, + { + "epoch": 2.138082191780822, + "grad_norm": 2.6131961345672607, + "learning_rate": 1.6267262388302193e-06, + "loss": 1.7717, + "step": 1951 + }, + { + "epoch": 2.1391780821917807, + "grad_norm": 2.5644335746765137, + "learning_rate": 1.6246953696181965e-06, + "loss": 1.7139, + "step": 1952 + }, + { + "epoch": 2.1402739726027398, + "grad_norm": 2.5851385593414307, + "learning_rate": 1.6226645004061738e-06, + "loss": 1.7473, + "step": 1953 + }, + { + "epoch": 2.141369863013699, + "grad_norm": 2.493952989578247, + "learning_rate": 1.620633631194151e-06, + "loss": 1.8209, + "step": 1954 + }, + { + "epoch": 2.1424657534246574, + "grad_norm": 2.5686917304992676, + "learning_rate": 1.6186027619821284e-06, + "loss": 1.7527, + "step": 1955 + }, + { + "epoch": 2.1435616438356164, + "grad_norm": 2.6349050998687744, + "learning_rate": 1.6165718927701056e-06, + "loss": 1.7033, + "step": 1956 + }, + { + "epoch": 2.1446575342465755, + "grad_norm": 2.7004778385162354, + "learning_rate": 1.614541023558083e-06, + "loss": 1.7498, + "step": 1957 + }, + { + "epoch": 2.145753424657534, + "grad_norm": 2.5406992435455322, + "learning_rate": 1.6125101543460602e-06, + "loss": 1.7605, + "step": 1958 + }, + { + "epoch": 2.146849315068493, + "grad_norm": 2.796125650405884, + "learning_rate": 1.6104792851340375e-06, + "loss": 1.7087, + "step": 1959 + }, + { + "epoch": 2.147945205479452, + "grad_norm": 2.703362226486206, + "learning_rate": 1.6084484159220147e-06, + "loss": 1.751, + "step": 1960 + }, + { + "epoch": 2.147945205479452, + "eval_loss": 1.9470759630203247, + "eval_runtime": 37.2721, + "eval_samples_per_second": 26.83, + "eval_steps_per_second": 1.127, + "step": 1960 + }, + { + "epoch": 2.149041095890411, + "grad_norm": 2.6464223861694336, + "learning_rate": 1.606417546709992e-06, + "loss": 1.7147, + "step": 1961 + }, + { + "epoch": 2.1501369863013697, + "grad_norm": 2.6499273777008057, + "learning_rate": 1.6043866774979693e-06, + "loss": 1.7015, + "step": 1962 + }, + { + "epoch": 2.1512328767123288, + "grad_norm": 2.583768844604492, + "learning_rate": 1.6023558082859464e-06, + "loss": 1.7158, + "step": 1963 + }, + { + "epoch": 2.152328767123288, + "grad_norm": 2.6245381832122803, + "learning_rate": 1.6003249390739236e-06, + "loss": 1.6824, + "step": 1964 + }, + { + "epoch": 2.1534246575342464, + "grad_norm": 2.497676134109497, + "learning_rate": 1.598294069861901e-06, + "loss": 1.7345, + "step": 1965 + }, + { + "epoch": 2.1545205479452054, + "grad_norm": 2.7218856811523438, + "learning_rate": 1.5962632006498782e-06, + "loss": 1.6988, + "step": 1966 + }, + { + "epoch": 2.1556164383561645, + "grad_norm": 2.7654666900634766, + "learning_rate": 1.5942323314378555e-06, + "loss": 1.7635, + "step": 1967 + }, + { + "epoch": 2.1567123287671235, + "grad_norm": 2.598526954650879, + "learning_rate": 1.5922014622258327e-06, + "loss": 1.6769, + "step": 1968 + }, + { + "epoch": 2.157808219178082, + "grad_norm": 2.5995986461639404, + "learning_rate": 1.59017059301381e-06, + "loss": 1.6928, + "step": 1969 + }, + { + "epoch": 2.158904109589041, + "grad_norm": 2.5692837238311768, + "learning_rate": 1.5881397238017873e-06, + "loss": 1.7415, + "step": 1970 + }, + { + "epoch": 2.158904109589041, + "eval_loss": 1.9472142457962036, + "eval_runtime": 37.258, + "eval_samples_per_second": 26.84, + "eval_steps_per_second": 1.127, + "step": 1970 + }, + { + "epoch": 2.16, + "grad_norm": 2.622732162475586, + "learning_rate": 1.5861088545897646e-06, + "loss": 1.7383, + "step": 1971 + }, + { + "epoch": 2.1610958904109587, + "grad_norm": 2.573005437850952, + "learning_rate": 1.5840779853777418e-06, + "loss": 1.6696, + "step": 1972 + }, + { + "epoch": 2.1621917808219178, + "grad_norm": 2.7022411823272705, + "learning_rate": 1.5820471161657191e-06, + "loss": 1.7231, + "step": 1973 + }, + { + "epoch": 2.163287671232877, + "grad_norm": 2.739145278930664, + "learning_rate": 1.5800162469536964e-06, + "loss": 1.7012, + "step": 1974 + }, + { + "epoch": 2.1643835616438354, + "grad_norm": 2.7034459114074707, + "learning_rate": 1.5779853777416735e-06, + "loss": 1.769, + "step": 1975 + }, + { + "epoch": 2.1654794520547944, + "grad_norm": 2.745359182357788, + "learning_rate": 1.5759545085296507e-06, + "loss": 1.7424, + "step": 1976 + }, + { + "epoch": 2.1665753424657535, + "grad_norm": 2.722721576690674, + "learning_rate": 1.573923639317628e-06, + "loss": 1.733, + "step": 1977 + }, + { + "epoch": 2.1676712328767125, + "grad_norm": 2.689074754714966, + "learning_rate": 1.5718927701056053e-06, + "loss": 1.7468, + "step": 1978 + }, + { + "epoch": 2.168767123287671, + "grad_norm": 2.6735317707061768, + "learning_rate": 1.5698619008935826e-06, + "loss": 1.7312, + "step": 1979 + }, + { + "epoch": 2.16986301369863, + "grad_norm": 2.6908175945281982, + "learning_rate": 1.5678310316815598e-06, + "loss": 1.7502, + "step": 1980 + }, + { + "epoch": 2.16986301369863, + "eval_loss": 1.947867512702942, + "eval_runtime": 37.1858, + "eval_samples_per_second": 26.892, + "eval_steps_per_second": 1.129, + "step": 1980 + }, + { + "epoch": 2.170958904109589, + "grad_norm": 2.524951934814453, + "learning_rate": 1.5658001624695371e-06, + "loss": 1.6863, + "step": 1981 + }, + { + "epoch": 2.172054794520548, + "grad_norm": 2.746180295944214, + "learning_rate": 1.5637692932575144e-06, + "loss": 1.7621, + "step": 1982 + }, + { + "epoch": 2.1731506849315068, + "grad_norm": 2.4799301624298096, + "learning_rate": 1.5617384240454917e-06, + "loss": 1.7121, + "step": 1983 + }, + { + "epoch": 2.174246575342466, + "grad_norm": 2.6821515560150146, + "learning_rate": 1.559707554833469e-06, + "loss": 1.7271, + "step": 1984 + }, + { + "epoch": 2.175342465753425, + "grad_norm": 2.572122812271118, + "learning_rate": 1.5576766856214462e-06, + "loss": 1.7671, + "step": 1985 + }, + { + "epoch": 2.1764383561643834, + "grad_norm": 2.614806652069092, + "learning_rate": 1.5556458164094235e-06, + "loss": 1.6411, + "step": 1986 + }, + { + "epoch": 2.1775342465753424, + "grad_norm": 2.602883815765381, + "learning_rate": 1.5536149471974008e-06, + "loss": 1.7143, + "step": 1987 + }, + { + "epoch": 2.1786301369863015, + "grad_norm": 2.5893354415893555, + "learning_rate": 1.5515840779853778e-06, + "loss": 1.6733, + "step": 1988 + }, + { + "epoch": 2.17972602739726, + "grad_norm": 2.652308702468872, + "learning_rate": 1.549553208773355e-06, + "loss": 1.7311, + "step": 1989 + }, + { + "epoch": 2.180821917808219, + "grad_norm": 2.7129695415496826, + "learning_rate": 1.5475223395613324e-06, + "loss": 1.7001, + "step": 1990 + }, + { + "epoch": 2.180821917808219, + "eval_loss": 1.9512702226638794, + "eval_runtime": 37.225, + "eval_samples_per_second": 26.864, + "eval_steps_per_second": 1.128, + "step": 1990 + }, + { + "epoch": 2.181917808219178, + "grad_norm": 2.563638210296631, + "learning_rate": 1.5454914703493097e-06, + "loss": 1.7118, + "step": 1991 + }, + { + "epoch": 2.183013698630137, + "grad_norm": 2.629056453704834, + "learning_rate": 1.543460601137287e-06, + "loss": 1.8041, + "step": 1992 + }, + { + "epoch": 2.1841095890410958, + "grad_norm": 2.6072700023651123, + "learning_rate": 1.5414297319252642e-06, + "loss": 1.7568, + "step": 1993 + }, + { + "epoch": 2.185205479452055, + "grad_norm": 2.525784730911255, + "learning_rate": 1.5393988627132415e-06, + "loss": 1.748, + "step": 1994 + }, + { + "epoch": 2.186301369863014, + "grad_norm": 2.564699649810791, + "learning_rate": 1.5373679935012188e-06, + "loss": 1.7148, + "step": 1995 + }, + { + "epoch": 2.1873972602739724, + "grad_norm": 2.7365031242370605, + "learning_rate": 1.535337124289196e-06, + "loss": 1.6714, + "step": 1996 + }, + { + "epoch": 2.1884931506849314, + "grad_norm": 2.719954252243042, + "learning_rate": 1.5333062550771733e-06, + "loss": 1.7301, + "step": 1997 + }, + { + "epoch": 2.1895890410958905, + "grad_norm": 2.7979655265808105, + "learning_rate": 1.5312753858651506e-06, + "loss": 1.8163, + "step": 1998 + }, + { + "epoch": 2.1906849315068495, + "grad_norm": 2.6182446479797363, + "learning_rate": 1.5292445166531279e-06, + "loss": 1.713, + "step": 1999 + }, + { + "epoch": 2.191780821917808, + "grad_norm": 2.7218992710113525, + "learning_rate": 1.5272136474411047e-06, + "loss": 1.7355, + "step": 2000 + }, + { + "epoch": 2.191780821917808, + "eval_loss": 1.950900673866272, + "eval_runtime": 37.2927, + "eval_samples_per_second": 26.815, + "eval_steps_per_second": 1.126, + "step": 2000 + }, + { + "epoch": 2.192876712328767, + "grad_norm": 2.7384393215179443, + "learning_rate": 1.525182778229082e-06, + "loss": 1.7777, + "step": 2001 + }, + { + "epoch": 2.193972602739726, + "grad_norm": 2.7223575115203857, + "learning_rate": 1.5231519090170593e-06, + "loss": 1.7136, + "step": 2002 + }, + { + "epoch": 2.1950684931506848, + "grad_norm": 2.5952577590942383, + "learning_rate": 1.5211210398050365e-06, + "loss": 1.6964, + "step": 2003 + }, + { + "epoch": 2.196164383561644, + "grad_norm": 2.626406669616699, + "learning_rate": 1.5190901705930138e-06, + "loss": 1.6996, + "step": 2004 + }, + { + "epoch": 2.197260273972603, + "grad_norm": 2.6563360691070557, + "learning_rate": 1.517059301380991e-06, + "loss": 1.7428, + "step": 2005 + }, + { + "epoch": 2.198356164383562, + "grad_norm": 2.647664785385132, + "learning_rate": 1.5150284321689684e-06, + "loss": 1.6953, + "step": 2006 + }, + { + "epoch": 2.1994520547945204, + "grad_norm": 2.5670430660247803, + "learning_rate": 1.5129975629569456e-06, + "loss": 1.7215, + "step": 2007 + }, + { + "epoch": 2.2005479452054795, + "grad_norm": 2.6148288249969482, + "learning_rate": 1.510966693744923e-06, + "loss": 1.724, + "step": 2008 + }, + { + "epoch": 2.2016438356164385, + "grad_norm": 2.8430588245391846, + "learning_rate": 1.5089358245329002e-06, + "loss": 1.7255, + "step": 2009 + }, + { + "epoch": 2.202739726027397, + "grad_norm": 2.5937509536743164, + "learning_rate": 1.5069049553208775e-06, + "loss": 1.7055, + "step": 2010 + }, + { + "epoch": 2.202739726027397, + "eval_loss": 1.9469048976898193, + "eval_runtime": 37.2173, + "eval_samples_per_second": 26.869, + "eval_steps_per_second": 1.129, + "step": 2010 + }, + { + "epoch": 2.203835616438356, + "grad_norm": 2.7167937755584717, + "learning_rate": 1.5048740861088547e-06, + "loss": 1.7474, + "step": 2011 + }, + { + "epoch": 2.204931506849315, + "grad_norm": 2.53837251663208, + "learning_rate": 1.502843216896832e-06, + "loss": 1.74, + "step": 2012 + }, + { + "epoch": 2.206027397260274, + "grad_norm": 2.604931592941284, + "learning_rate": 1.500812347684809e-06, + "loss": 1.6573, + "step": 2013 + }, + { + "epoch": 2.207123287671233, + "grad_norm": 2.6670846939086914, + "learning_rate": 1.4987814784727864e-06, + "loss": 1.7208, + "step": 2014 + }, + { + "epoch": 2.208219178082192, + "grad_norm": 2.677746534347534, + "learning_rate": 1.4967506092607636e-06, + "loss": 1.728, + "step": 2015 + }, + { + "epoch": 2.209315068493151, + "grad_norm": 2.6490612030029297, + "learning_rate": 1.494719740048741e-06, + "loss": 1.6728, + "step": 2016 + }, + { + "epoch": 2.2104109589041094, + "grad_norm": 2.6571807861328125, + "learning_rate": 1.4926888708367182e-06, + "loss": 1.7528, + "step": 2017 + }, + { + "epoch": 2.2115068493150685, + "grad_norm": 2.5933258533477783, + "learning_rate": 1.4906580016246955e-06, + "loss": 1.757, + "step": 2018 + }, + { + "epoch": 2.2126027397260275, + "grad_norm": 2.6451070308685303, + "learning_rate": 1.4886271324126727e-06, + "loss": 1.763, + "step": 2019 + }, + { + "epoch": 2.213698630136986, + "grad_norm": 2.5259275436401367, + "learning_rate": 1.48659626320065e-06, + "loss": 1.7979, + "step": 2020 + }, + { + "epoch": 2.213698630136986, + "eval_loss": 1.9480665922164917, + "eval_runtime": 37.5325, + "eval_samples_per_second": 26.644, + "eval_steps_per_second": 1.119, + "step": 2020 + }, + { + "epoch": 2.214794520547945, + "grad_norm": 2.6959381103515625, + "learning_rate": 1.4845653939886273e-06, + "loss": 1.6376, + "step": 2021 + }, + { + "epoch": 2.215890410958904, + "grad_norm": 2.6179497241973877, + "learning_rate": 1.4825345247766046e-06, + "loss": 1.6575, + "step": 2022 + }, + { + "epoch": 2.216986301369863, + "grad_norm": 2.5618393421173096, + "learning_rate": 1.4805036555645818e-06, + "loss": 1.6486, + "step": 2023 + }, + { + "epoch": 2.2180821917808218, + "grad_norm": 2.545051336288452, + "learning_rate": 1.4784727863525591e-06, + "loss": 1.7218, + "step": 2024 + }, + { + "epoch": 2.219178082191781, + "grad_norm": 2.6248512268066406, + "learning_rate": 1.4764419171405362e-06, + "loss": 1.691, + "step": 2025 + }, + { + "epoch": 2.22027397260274, + "grad_norm": 2.65645170211792, + "learning_rate": 1.4744110479285135e-06, + "loss": 1.7602, + "step": 2026 + }, + { + "epoch": 2.2213698630136984, + "grad_norm": 2.6771113872528076, + "learning_rate": 1.4723801787164907e-06, + "loss": 1.7573, + "step": 2027 + }, + { + "epoch": 2.2224657534246575, + "grad_norm": 2.5466039180755615, + "learning_rate": 1.470349309504468e-06, + "loss": 1.7604, + "step": 2028 + }, + { + "epoch": 2.2235616438356165, + "grad_norm": 2.685079574584961, + "learning_rate": 1.4683184402924453e-06, + "loss": 1.7341, + "step": 2029 + }, + { + "epoch": 2.2246575342465755, + "grad_norm": 2.6854352951049805, + "learning_rate": 1.4662875710804226e-06, + "loss": 1.7427, + "step": 2030 + }, + { + "epoch": 2.2246575342465755, + "eval_loss": 1.9459377527236938, + "eval_runtime": 37.5255, + "eval_samples_per_second": 26.649, + "eval_steps_per_second": 1.119, + "step": 2030 + }, + { + "epoch": 2.225753424657534, + "grad_norm": 2.667675733566284, + "learning_rate": 1.4642567018683998e-06, + "loss": 1.8328, + "step": 2031 + }, + { + "epoch": 2.226849315068493, + "grad_norm": 2.603386640548706, + "learning_rate": 1.4622258326563771e-06, + "loss": 1.7193, + "step": 2032 + }, + { + "epoch": 2.227945205479452, + "grad_norm": 2.4960451126098633, + "learning_rate": 1.4601949634443544e-06, + "loss": 1.695, + "step": 2033 + }, + { + "epoch": 2.2290410958904108, + "grad_norm": 2.755707025527954, + "learning_rate": 1.4581640942323317e-06, + "loss": 1.7475, + "step": 2034 + }, + { + "epoch": 2.23013698630137, + "grad_norm": 2.5849006175994873, + "learning_rate": 1.456133225020309e-06, + "loss": 1.8106, + "step": 2035 + }, + { + "epoch": 2.231232876712329, + "grad_norm": 2.690680980682373, + "learning_rate": 1.4541023558082862e-06, + "loss": 1.7368, + "step": 2036 + }, + { + "epoch": 2.232328767123288, + "grad_norm": 2.6760425567626953, + "learning_rate": 1.4520714865962635e-06, + "loss": 1.722, + "step": 2037 + }, + { + "epoch": 2.2334246575342465, + "grad_norm": 2.656379461288452, + "learning_rate": 1.4500406173842406e-06, + "loss": 1.7623, + "step": 2038 + }, + { + "epoch": 2.2345205479452055, + "grad_norm": 2.8213517665863037, + "learning_rate": 1.4480097481722178e-06, + "loss": 1.7344, + "step": 2039 + }, + { + "epoch": 2.2356164383561645, + "grad_norm": 2.877163887023926, + "learning_rate": 1.4459788789601951e-06, + "loss": 1.7184, + "step": 2040 + }, + { + "epoch": 2.2356164383561645, + "eval_loss": 1.9430633783340454, + "eval_runtime": 37.3102, + "eval_samples_per_second": 26.802, + "eval_steps_per_second": 1.126, + "step": 2040 + }, + { + "epoch": 2.236712328767123, + "grad_norm": 2.7285237312316895, + "learning_rate": 1.4439480097481724e-06, + "loss": 1.6826, + "step": 2041 + }, + { + "epoch": 2.237808219178082, + "grad_norm": 2.5208237171173096, + "learning_rate": 1.4419171405361497e-06, + "loss": 1.6788, + "step": 2042 + }, + { + "epoch": 2.238904109589041, + "grad_norm": 2.7415900230407715, + "learning_rate": 1.439886271324127e-06, + "loss": 1.7423, + "step": 2043 + }, + { + "epoch": 2.24, + "grad_norm": 2.760897397994995, + "learning_rate": 1.4378554021121042e-06, + "loss": 1.7341, + "step": 2044 + }, + { + "epoch": 2.241095890410959, + "grad_norm": 2.5650556087493896, + "learning_rate": 1.4358245329000815e-06, + "loss": 1.7036, + "step": 2045 + }, + { + "epoch": 2.242191780821918, + "grad_norm": 2.60780930519104, + "learning_rate": 1.4337936636880588e-06, + "loss": 1.7719, + "step": 2046 + }, + { + "epoch": 2.243287671232877, + "grad_norm": 2.592473268508911, + "learning_rate": 1.431762794476036e-06, + "loss": 1.7421, + "step": 2047 + }, + { + "epoch": 2.2443835616438355, + "grad_norm": 2.6672041416168213, + "learning_rate": 1.4297319252640133e-06, + "loss": 1.6865, + "step": 2048 + }, + { + "epoch": 2.2454794520547945, + "grad_norm": 2.6182847023010254, + "learning_rate": 1.4277010560519906e-06, + "loss": 1.731, + "step": 2049 + }, + { + "epoch": 2.2465753424657535, + "grad_norm": 2.8213913440704346, + "learning_rate": 1.4256701868399674e-06, + "loss": 1.6845, + "step": 2050 + }, + { + "epoch": 2.2465753424657535, + "eval_loss": 1.9434115886688232, + "eval_runtime": 37.3048, + "eval_samples_per_second": 26.806, + "eval_steps_per_second": 1.126, + "step": 2050 + }, + { + "epoch": 2.247671232876712, + "grad_norm": 2.7278084754943848, + "learning_rate": 1.4236393176279447e-06, + "loss": 1.8077, + "step": 2051 + }, + { + "epoch": 2.248767123287671, + "grad_norm": 2.5665392875671387, + "learning_rate": 1.421608448415922e-06, + "loss": 1.7235, + "step": 2052 + }, + { + "epoch": 2.24986301369863, + "grad_norm": 2.680154323577881, + "learning_rate": 1.4195775792038993e-06, + "loss": 1.8164, + "step": 2053 + }, + { + "epoch": 2.250958904109589, + "grad_norm": 2.770051956176758, + "learning_rate": 1.4175467099918765e-06, + "loss": 1.6948, + "step": 2054 + }, + { + "epoch": 2.252054794520548, + "grad_norm": 2.6189041137695312, + "learning_rate": 1.4155158407798538e-06, + "loss": 1.7369, + "step": 2055 + }, + { + "epoch": 2.253150684931507, + "grad_norm": 2.6655054092407227, + "learning_rate": 1.413484971567831e-06, + "loss": 1.7456, + "step": 2056 + }, + { + "epoch": 2.254246575342466, + "grad_norm": 2.6520164012908936, + "learning_rate": 1.4114541023558084e-06, + "loss": 1.6719, + "step": 2057 + }, + { + "epoch": 2.255342465753425, + "grad_norm": 2.594533681869507, + "learning_rate": 1.4094232331437856e-06, + "loss": 1.7078, + "step": 2058 + }, + { + "epoch": 2.2564383561643835, + "grad_norm": 2.592308759689331, + "learning_rate": 1.407392363931763e-06, + "loss": 1.7517, + "step": 2059 + }, + { + "epoch": 2.2575342465753425, + "grad_norm": 2.5815367698669434, + "learning_rate": 1.4053614947197402e-06, + "loss": 1.8262, + "step": 2060 + }, + { + "epoch": 2.2575342465753425, + "eval_loss": 1.945846438407898, + "eval_runtime": 37.186, + "eval_samples_per_second": 26.892, + "eval_steps_per_second": 1.129, + "step": 2060 + }, + { + "epoch": 2.2586301369863016, + "grad_norm": 2.643691301345825, + "learning_rate": 1.4033306255077175e-06, + "loss": 1.7559, + "step": 2061 + }, + { + "epoch": 2.25972602739726, + "grad_norm": 2.6091558933258057, + "learning_rate": 1.4012997562956945e-06, + "loss": 1.6433, + "step": 2062 + }, + { + "epoch": 2.260821917808219, + "grad_norm": 2.7172136306762695, + "learning_rate": 1.3992688870836718e-06, + "loss": 1.7735, + "step": 2063 + }, + { + "epoch": 2.261917808219178, + "grad_norm": 2.5673139095306396, + "learning_rate": 1.397238017871649e-06, + "loss": 1.7402, + "step": 2064 + }, + { + "epoch": 2.263013698630137, + "grad_norm": 2.711242198944092, + "learning_rate": 1.3952071486596264e-06, + "loss": 1.7183, + "step": 2065 + }, + { + "epoch": 2.264109589041096, + "grad_norm": 2.6999948024749756, + "learning_rate": 1.3931762794476036e-06, + "loss": 1.6856, + "step": 2066 + }, + { + "epoch": 2.265205479452055, + "grad_norm": 2.670964002609253, + "learning_rate": 1.391145410235581e-06, + "loss": 1.78, + "step": 2067 + }, + { + "epoch": 2.266301369863014, + "grad_norm": 2.4599525928497314, + "learning_rate": 1.3891145410235582e-06, + "loss": 1.7143, + "step": 2068 + }, + { + "epoch": 2.2673972602739725, + "grad_norm": 2.552266836166382, + "learning_rate": 1.3870836718115355e-06, + "loss": 1.6503, + "step": 2069 + }, + { + "epoch": 2.2684931506849315, + "grad_norm": 2.5764405727386475, + "learning_rate": 1.3850528025995127e-06, + "loss": 1.7046, + "step": 2070 + }, + { + "epoch": 2.2684931506849315, + "eval_loss": 1.9457523822784424, + "eval_runtime": 37.217, + "eval_samples_per_second": 26.869, + "eval_steps_per_second": 1.129, + "step": 2070 + }, + { + "epoch": 2.2695890410958905, + "grad_norm": 2.524228572845459, + "learning_rate": 1.38302193338749e-06, + "loss": 1.7078, + "step": 2071 + }, + { + "epoch": 2.270684931506849, + "grad_norm": 2.5531418323516846, + "learning_rate": 1.3809910641754673e-06, + "loss": 1.646, + "step": 2072 + }, + { + "epoch": 2.271780821917808, + "grad_norm": 2.705824613571167, + "learning_rate": 1.3789601949634446e-06, + "loss": 1.7407, + "step": 2073 + }, + { + "epoch": 2.272876712328767, + "grad_norm": 2.5266356468200684, + "learning_rate": 1.3769293257514218e-06, + "loss": 1.6472, + "step": 2074 + }, + { + "epoch": 2.2739726027397262, + "grad_norm": 2.5785422325134277, + "learning_rate": 1.374898456539399e-06, + "loss": 1.7561, + "step": 2075 + }, + { + "epoch": 2.275068493150685, + "grad_norm": 2.663120985031128, + "learning_rate": 1.3728675873273762e-06, + "loss": 1.6901, + "step": 2076 + }, + { + "epoch": 2.276164383561644, + "grad_norm": 2.6085517406463623, + "learning_rate": 1.3708367181153535e-06, + "loss": 1.7829, + "step": 2077 + }, + { + "epoch": 2.277260273972603, + "grad_norm": 2.6439929008483887, + "learning_rate": 1.3688058489033307e-06, + "loss": 1.7684, + "step": 2078 + }, + { + "epoch": 2.2783561643835615, + "grad_norm": 2.6614694595336914, + "learning_rate": 1.366774979691308e-06, + "loss": 1.726, + "step": 2079 + }, + { + "epoch": 2.2794520547945205, + "grad_norm": 2.7339539527893066, + "learning_rate": 1.3647441104792853e-06, + "loss": 1.7577, + "step": 2080 + }, + { + "epoch": 2.2794520547945205, + "eval_loss": 1.9456545114517212, + "eval_runtime": 37.174, + "eval_samples_per_second": 26.901, + "eval_steps_per_second": 1.13, + "step": 2080 + }, + { + "epoch": 2.2805479452054795, + "grad_norm": 2.603545665740967, + "learning_rate": 1.3627132412672626e-06, + "loss": 1.7297, + "step": 2081 + }, + { + "epoch": 2.281643835616438, + "grad_norm": 2.548466682434082, + "learning_rate": 1.3606823720552398e-06, + "loss": 1.6701, + "step": 2082 + }, + { + "epoch": 2.282739726027397, + "grad_norm": 2.6043875217437744, + "learning_rate": 1.3586515028432171e-06, + "loss": 1.7369, + "step": 2083 + }, + { + "epoch": 2.283835616438356, + "grad_norm": 2.7412586212158203, + "learning_rate": 1.3566206336311944e-06, + "loss": 1.654, + "step": 2084 + }, + { + "epoch": 2.2849315068493152, + "grad_norm": 2.669410228729248, + "learning_rate": 1.3545897644191717e-06, + "loss": 1.7186, + "step": 2085 + }, + { + "epoch": 2.286027397260274, + "grad_norm": 2.7052674293518066, + "learning_rate": 1.352558895207149e-06, + "loss": 1.7005, + "step": 2086 + }, + { + "epoch": 2.287123287671233, + "grad_norm": 2.5840418338775635, + "learning_rate": 1.3505280259951258e-06, + "loss": 1.7459, + "step": 2087 + }, + { + "epoch": 2.288219178082192, + "grad_norm": 2.6355416774749756, + "learning_rate": 1.348497156783103e-06, + "loss": 1.7243, + "step": 2088 + }, + { + "epoch": 2.289315068493151, + "grad_norm": 2.6697628498077393, + "learning_rate": 1.3464662875710804e-06, + "loss": 1.7597, + "step": 2089 + }, + { + "epoch": 2.2904109589041095, + "grad_norm": 2.679842233657837, + "learning_rate": 1.3444354183590576e-06, + "loss": 1.6557, + "step": 2090 + }, + { + "epoch": 2.2904109589041095, + "eval_loss": 1.9469218254089355, + "eval_runtime": 37.1615, + "eval_samples_per_second": 26.91, + "eval_steps_per_second": 1.13, + "step": 2090 + }, + { + "epoch": 2.2915068493150685, + "grad_norm": 2.8354430198669434, + "learning_rate": 1.342404549147035e-06, + "loss": 1.7496, + "step": 2091 + }, + { + "epoch": 2.2926027397260276, + "grad_norm": 2.6957387924194336, + "learning_rate": 1.3403736799350122e-06, + "loss": 1.6872, + "step": 2092 + }, + { + "epoch": 2.293698630136986, + "grad_norm": 2.5815868377685547, + "learning_rate": 1.3383428107229895e-06, + "loss": 1.7497, + "step": 2093 + }, + { + "epoch": 2.294794520547945, + "grad_norm": 2.727085828781128, + "learning_rate": 1.3363119415109667e-06, + "loss": 1.7723, + "step": 2094 + }, + { + "epoch": 2.2958904109589042, + "grad_norm": 2.6119000911712646, + "learning_rate": 1.334281072298944e-06, + "loss": 1.7414, + "step": 2095 + }, + { + "epoch": 2.296986301369863, + "grad_norm": 2.6403088569641113, + "learning_rate": 1.3322502030869213e-06, + "loss": 1.7398, + "step": 2096 + }, + { + "epoch": 2.298082191780822, + "grad_norm": 2.5066816806793213, + "learning_rate": 1.3302193338748986e-06, + "loss": 1.6974, + "step": 2097 + }, + { + "epoch": 2.299178082191781, + "grad_norm": 2.6601762771606445, + "learning_rate": 1.3281884646628758e-06, + "loss": 1.7578, + "step": 2098 + }, + { + "epoch": 2.30027397260274, + "grad_norm": 2.571499824523926, + "learning_rate": 1.3261575954508531e-06, + "loss": 1.7573, + "step": 2099 + }, + { + "epoch": 2.3013698630136985, + "grad_norm": 2.533543825149536, + "learning_rate": 1.3241267262388302e-06, + "loss": 1.7566, + "step": 2100 + }, + { + "epoch": 2.3013698630136985, + "eval_loss": 1.9441688060760498, + "eval_runtime": 37.2158, + "eval_samples_per_second": 26.87, + "eval_steps_per_second": 1.129, + "step": 2100 + }, + { + "epoch": 2.3024657534246575, + "grad_norm": 2.782416582107544, + "learning_rate": 1.3220958570268074e-06, + "loss": 1.714, + "step": 2101 + }, + { + "epoch": 2.3035616438356166, + "grad_norm": 2.742504358291626, + "learning_rate": 1.3200649878147847e-06, + "loss": 1.6677, + "step": 2102 + }, + { + "epoch": 2.304657534246575, + "grad_norm": 2.7778940200805664, + "learning_rate": 1.318034118602762e-06, + "loss": 1.7178, + "step": 2103 + }, + { + "epoch": 2.305753424657534, + "grad_norm": 2.7003014087677, + "learning_rate": 1.3160032493907393e-06, + "loss": 1.7691, + "step": 2104 + }, + { + "epoch": 2.3068493150684932, + "grad_norm": 2.6644296646118164, + "learning_rate": 1.3139723801787166e-06, + "loss": 1.7285, + "step": 2105 + }, + { + "epoch": 2.3079452054794523, + "grad_norm": 2.7859749794006348, + "learning_rate": 1.3119415109666938e-06, + "loss": 1.8026, + "step": 2106 + }, + { + "epoch": 2.309041095890411, + "grad_norm": 2.6862375736236572, + "learning_rate": 1.309910641754671e-06, + "loss": 1.7238, + "step": 2107 + }, + { + "epoch": 2.31013698630137, + "grad_norm": 2.749908685684204, + "learning_rate": 1.3078797725426484e-06, + "loss": 1.676, + "step": 2108 + }, + { + "epoch": 2.311232876712329, + "grad_norm": 2.679384469985962, + "learning_rate": 1.3058489033306257e-06, + "loss": 1.6899, + "step": 2109 + }, + { + "epoch": 2.3123287671232875, + "grad_norm": 2.5621979236602783, + "learning_rate": 1.303818034118603e-06, + "loss": 1.6963, + "step": 2110 + }, + { + "epoch": 2.3123287671232875, + "eval_loss": 1.9453113079071045, + "eval_runtime": 37.2803, + "eval_samples_per_second": 26.824, + "eval_steps_per_second": 1.127, + "step": 2110 + }, + { + "epoch": 2.3134246575342465, + "grad_norm": 2.6183035373687744, + "learning_rate": 1.3017871649065802e-06, + "loss": 1.7272, + "step": 2111 + }, + { + "epoch": 2.3145205479452056, + "grad_norm": 2.5777511596679688, + "learning_rate": 1.2997562956945573e-06, + "loss": 1.7041, + "step": 2112 + }, + { + "epoch": 2.315616438356164, + "grad_norm": 2.784181594848633, + "learning_rate": 1.2977254264825345e-06, + "loss": 1.6835, + "step": 2113 + }, + { + "epoch": 2.316712328767123, + "grad_norm": 2.679591417312622, + "learning_rate": 1.2956945572705118e-06, + "loss": 1.7242, + "step": 2114 + }, + { + "epoch": 2.317808219178082, + "grad_norm": 2.561269521713257, + "learning_rate": 1.293663688058489e-06, + "loss": 1.7455, + "step": 2115 + }, + { + "epoch": 2.3189041095890413, + "grad_norm": 2.5087854862213135, + "learning_rate": 1.2916328188464664e-06, + "loss": 1.6824, + "step": 2116 + }, + { + "epoch": 2.32, + "grad_norm": 2.7247722148895264, + "learning_rate": 1.2896019496344436e-06, + "loss": 1.6268, + "step": 2117 + }, + { + "epoch": 2.321095890410959, + "grad_norm": 2.603994846343994, + "learning_rate": 1.287571080422421e-06, + "loss": 1.6925, + "step": 2118 + }, + { + "epoch": 2.322191780821918, + "grad_norm": 2.5525155067443848, + "learning_rate": 1.2855402112103982e-06, + "loss": 1.7643, + "step": 2119 + }, + { + "epoch": 2.323287671232877, + "grad_norm": 2.5990493297576904, + "learning_rate": 1.2835093419983755e-06, + "loss": 1.6712, + "step": 2120 + }, + { + "epoch": 2.323287671232877, + "eval_loss": 1.9467085599899292, + "eval_runtime": 37.5353, + "eval_samples_per_second": 26.642, + "eval_steps_per_second": 1.119, + "step": 2120 + }, + { + "epoch": 2.3243835616438355, + "grad_norm": 2.764711618423462, + "learning_rate": 1.2814784727863528e-06, + "loss": 1.71, + "step": 2121 + }, + { + "epoch": 2.3254794520547946, + "grad_norm": 2.6009247303009033, + "learning_rate": 1.27944760357433e-06, + "loss": 1.6956, + "step": 2122 + }, + { + "epoch": 2.3265753424657536, + "grad_norm": 2.6900346279144287, + "learning_rate": 1.2774167343623073e-06, + "loss": 1.7529, + "step": 2123 + }, + { + "epoch": 2.327671232876712, + "grad_norm": 2.6602327823638916, + "learning_rate": 1.2753858651502846e-06, + "loss": 1.7588, + "step": 2124 + }, + { + "epoch": 2.328767123287671, + "grad_norm": 2.7727952003479004, + "learning_rate": 1.2733549959382616e-06, + "loss": 1.7499, + "step": 2125 + }, + { + "epoch": 2.3298630136986302, + "grad_norm": 2.703307867050171, + "learning_rate": 1.271324126726239e-06, + "loss": 1.7169, + "step": 2126 + }, + { + "epoch": 2.330958904109589, + "grad_norm": 2.7290568351745605, + "learning_rate": 1.2692932575142162e-06, + "loss": 1.7076, + "step": 2127 + }, + { + "epoch": 2.332054794520548, + "grad_norm": 2.664865255355835, + "learning_rate": 1.2672623883021935e-06, + "loss": 1.6655, + "step": 2128 + }, + { + "epoch": 2.333150684931507, + "grad_norm": 2.6096653938293457, + "learning_rate": 1.2652315190901707e-06, + "loss": 1.6644, + "step": 2129 + }, + { + "epoch": 2.334246575342466, + "grad_norm": 2.7201335430145264, + "learning_rate": 1.263200649878148e-06, + "loss": 1.7924, + "step": 2130 + }, + { + "epoch": 2.334246575342466, + "eval_loss": 1.9422904253005981, + "eval_runtime": 37.5421, + "eval_samples_per_second": 26.637, + "eval_steps_per_second": 1.119, + "step": 2130 + }, + { + "epoch": 2.3353424657534245, + "grad_norm": 2.519702434539795, + "learning_rate": 1.2611697806661253e-06, + "loss": 1.7108, + "step": 2131 + }, + { + "epoch": 2.3364383561643836, + "grad_norm": 2.6535212993621826, + "learning_rate": 1.2591389114541026e-06, + "loss": 1.7243, + "step": 2132 + }, + { + "epoch": 2.3375342465753426, + "grad_norm": 2.651458501815796, + "learning_rate": 1.2571080422420798e-06, + "loss": 1.7158, + "step": 2133 + }, + { + "epoch": 2.338630136986301, + "grad_norm": 2.6745946407318115, + "learning_rate": 1.2550771730300571e-06, + "loss": 1.7014, + "step": 2134 + }, + { + "epoch": 2.33972602739726, + "grad_norm": 2.6088216304779053, + "learning_rate": 1.2530463038180344e-06, + "loss": 1.7724, + "step": 2135 + }, + { + "epoch": 2.3408219178082192, + "grad_norm": 2.703014373779297, + "learning_rate": 1.2510154346060117e-06, + "loss": 1.7104, + "step": 2136 + }, + { + "epoch": 2.3419178082191783, + "grad_norm": 2.57769513130188, + "learning_rate": 1.2489845653939887e-06, + "loss": 1.7259, + "step": 2137 + }, + { + "epoch": 2.343013698630137, + "grad_norm": 2.596695899963379, + "learning_rate": 1.246953696181966e-06, + "loss": 1.7571, + "step": 2138 + }, + { + "epoch": 2.344109589041096, + "grad_norm": 2.6988871097564697, + "learning_rate": 1.2449228269699433e-06, + "loss": 1.8002, + "step": 2139 + }, + { + "epoch": 2.345205479452055, + "grad_norm": 2.6352357864379883, + "learning_rate": 1.2428919577579204e-06, + "loss": 1.7213, + "step": 2140 + }, + { + "epoch": 2.345205479452055, + "eval_loss": 1.9401096105575562, + "eval_runtime": 37.4404, + "eval_samples_per_second": 26.709, + "eval_steps_per_second": 1.122, + "step": 2140 + }, + { + "epoch": 2.3463013698630135, + "grad_norm": 2.7089645862579346, + "learning_rate": 1.2408610885458976e-06, + "loss": 1.6737, + "step": 2141 + }, + { + "epoch": 2.3473972602739726, + "grad_norm": 2.6364030838012695, + "learning_rate": 1.238830219333875e-06, + "loss": 1.7186, + "step": 2142 + }, + { + "epoch": 2.3484931506849316, + "grad_norm": 2.5563573837280273, + "learning_rate": 1.2367993501218522e-06, + "loss": 1.6956, + "step": 2143 + }, + { + "epoch": 2.34958904109589, + "grad_norm": 2.638512372970581, + "learning_rate": 1.2347684809098295e-06, + "loss": 1.7693, + "step": 2144 + }, + { + "epoch": 2.350684931506849, + "grad_norm": 2.733036518096924, + "learning_rate": 1.2327376116978067e-06, + "loss": 1.7327, + "step": 2145 + }, + { + "epoch": 2.3517808219178082, + "grad_norm": 2.6013307571411133, + "learning_rate": 1.230706742485784e-06, + "loss": 1.6539, + "step": 2146 + }, + { + "epoch": 2.3528767123287673, + "grad_norm": 2.7629592418670654, + "learning_rate": 1.2286758732737613e-06, + "loss": 1.666, + "step": 2147 + }, + { + "epoch": 2.353972602739726, + "grad_norm": 2.6855108737945557, + "learning_rate": 1.2266450040617386e-06, + "loss": 1.8044, + "step": 2148 + }, + { + "epoch": 2.355068493150685, + "grad_norm": 2.607609748840332, + "learning_rate": 1.2246141348497158e-06, + "loss": 1.8022, + "step": 2149 + }, + { + "epoch": 2.356164383561644, + "grad_norm": 2.5609383583068848, + "learning_rate": 1.2225832656376931e-06, + "loss": 1.73, + "step": 2150 + }, + { + "epoch": 2.356164383561644, + "eval_loss": 1.9387279748916626, + "eval_runtime": 37.2896, + "eval_samples_per_second": 26.817, + "eval_steps_per_second": 1.126, + "step": 2150 + }, + { + "epoch": 2.357260273972603, + "grad_norm": 2.523988723754883, + "learning_rate": 1.2205523964256704e-06, + "loss": 1.7289, + "step": 2151 + }, + { + "epoch": 2.3583561643835615, + "grad_norm": 2.675995111465454, + "learning_rate": 1.2185215272136477e-06, + "loss": 1.7036, + "step": 2152 + }, + { + "epoch": 2.3594520547945206, + "grad_norm": 2.535367012023926, + "learning_rate": 1.2164906580016247e-06, + "loss": 1.7309, + "step": 2153 + }, + { + "epoch": 2.3605479452054796, + "grad_norm": 2.6444637775421143, + "learning_rate": 1.214459788789602e-06, + "loss": 1.7192, + "step": 2154 + }, + { + "epoch": 2.361643835616438, + "grad_norm": 2.5746796131134033, + "learning_rate": 1.2124289195775793e-06, + "loss": 1.6343, + "step": 2155 + }, + { + "epoch": 2.3627397260273972, + "grad_norm": 2.612849712371826, + "learning_rate": 1.2103980503655566e-06, + "loss": 1.6686, + "step": 2156 + }, + { + "epoch": 2.3638356164383563, + "grad_norm": 2.6836912631988525, + "learning_rate": 1.2083671811535338e-06, + "loss": 1.7173, + "step": 2157 + }, + { + "epoch": 2.364931506849315, + "grad_norm": 2.716177225112915, + "learning_rate": 1.2063363119415111e-06, + "loss": 1.695, + "step": 2158 + }, + { + "epoch": 2.366027397260274, + "grad_norm": 2.6918227672576904, + "learning_rate": 1.2043054427294882e-06, + "loss": 1.684, + "step": 2159 + }, + { + "epoch": 2.367123287671233, + "grad_norm": 2.6265008449554443, + "learning_rate": 1.2022745735174654e-06, + "loss": 1.6959, + "step": 2160 + }, + { + "epoch": 2.367123287671233, + "eval_loss": 1.939719319343567, + "eval_runtime": 37.2557, + "eval_samples_per_second": 26.842, + "eval_steps_per_second": 1.127, + "step": 2160 + }, + { + "epoch": 2.368219178082192, + "grad_norm": 2.638829469680786, + "learning_rate": 1.2002437043054427e-06, + "loss": 1.7362, + "step": 2161 + }, + { + "epoch": 2.3693150684931505, + "grad_norm": 2.6369569301605225, + "learning_rate": 1.19821283509342e-06, + "loss": 1.7765, + "step": 2162 + }, + { + "epoch": 2.3704109589041096, + "grad_norm": 2.6272799968719482, + "learning_rate": 1.1961819658813973e-06, + "loss": 1.7695, + "step": 2163 + }, + { + "epoch": 2.3715068493150686, + "grad_norm": 2.601388931274414, + "learning_rate": 1.1941510966693746e-06, + "loss": 1.7259, + "step": 2164 + }, + { + "epoch": 2.3726027397260276, + "grad_norm": 2.5515146255493164, + "learning_rate": 1.1921202274573518e-06, + "loss": 1.7048, + "step": 2165 + }, + { + "epoch": 2.3736986301369862, + "grad_norm": 2.6139824390411377, + "learning_rate": 1.190089358245329e-06, + "loss": 1.6951, + "step": 2166 + }, + { + "epoch": 2.3747945205479453, + "grad_norm": 2.7428653240203857, + "learning_rate": 1.1880584890333064e-06, + "loss": 1.7771, + "step": 2167 + }, + { + "epoch": 2.3758904109589043, + "grad_norm": 2.690804958343506, + "learning_rate": 1.1860276198212837e-06, + "loss": 1.7073, + "step": 2168 + }, + { + "epoch": 2.376986301369863, + "grad_norm": 2.64790415763855, + "learning_rate": 1.183996750609261e-06, + "loss": 1.7344, + "step": 2169 + }, + { + "epoch": 2.378082191780822, + "grad_norm": 2.555832624435425, + "learning_rate": 1.1819658813972382e-06, + "loss": 1.7452, + "step": 2170 + }, + { + "epoch": 2.378082191780822, + "eval_loss": 1.9401137828826904, + "eval_runtime": 37.2362, + "eval_samples_per_second": 26.856, + "eval_steps_per_second": 1.128, + "step": 2170 + }, + { + "epoch": 2.379178082191781, + "grad_norm": 2.604097366333008, + "learning_rate": 1.1799350121852153e-06, + "loss": 1.7176, + "step": 2171 + }, + { + "epoch": 2.3802739726027395, + "grad_norm": 2.579312324523926, + "learning_rate": 1.1779041429731925e-06, + "loss": 1.7588, + "step": 2172 + }, + { + "epoch": 2.3813698630136986, + "grad_norm": 2.555727481842041, + "learning_rate": 1.1758732737611698e-06, + "loss": 1.7612, + "step": 2173 + }, + { + "epoch": 2.3824657534246576, + "grad_norm": 2.6929352283477783, + "learning_rate": 1.173842404549147e-06, + "loss": 1.6707, + "step": 2174 + }, + { + "epoch": 2.383561643835616, + "grad_norm": 2.6732795238494873, + "learning_rate": 1.1718115353371244e-06, + "loss": 1.7743, + "step": 2175 + }, + { + "epoch": 2.3846575342465752, + "grad_norm": 2.673490047454834, + "learning_rate": 1.1697806661251016e-06, + "loss": 1.6968, + "step": 2176 + }, + { + "epoch": 2.3857534246575343, + "grad_norm": 2.657837152481079, + "learning_rate": 1.167749796913079e-06, + "loss": 1.7633, + "step": 2177 + }, + { + "epoch": 2.3868493150684933, + "grad_norm": 2.6677606105804443, + "learning_rate": 1.1657189277010562e-06, + "loss": 1.7646, + "step": 2178 + }, + { + "epoch": 2.387945205479452, + "grad_norm": 2.63075852394104, + "learning_rate": 1.1636880584890335e-06, + "loss": 1.7019, + "step": 2179 + }, + { + "epoch": 2.389041095890411, + "grad_norm": 2.874946117401123, + "learning_rate": 1.1616571892770108e-06, + "loss": 1.691, + "step": 2180 + }, + { + "epoch": 2.389041095890411, + "eval_loss": 1.9399209022521973, + "eval_runtime": 37.1991, + "eval_samples_per_second": 26.882, + "eval_steps_per_second": 1.129, + "step": 2180 + }, + { + "epoch": 2.39013698630137, + "grad_norm": 2.734825611114502, + "learning_rate": 1.159626320064988e-06, + "loss": 1.7551, + "step": 2181 + }, + { + "epoch": 2.391232876712329, + "grad_norm": 2.5722298622131348, + "learning_rate": 1.1575954508529653e-06, + "loss": 1.7279, + "step": 2182 + }, + { + "epoch": 2.3923287671232876, + "grad_norm": 2.6449334621429443, + "learning_rate": 1.1555645816409426e-06, + "loss": 1.7591, + "step": 2183 + }, + { + "epoch": 2.3934246575342466, + "grad_norm": 2.774374485015869, + "learning_rate": 1.1535337124289196e-06, + "loss": 1.7243, + "step": 2184 + }, + { + "epoch": 2.3945205479452056, + "grad_norm": 2.554065227508545, + "learning_rate": 1.151502843216897e-06, + "loss": 1.6767, + "step": 2185 + }, + { + "epoch": 2.3956164383561642, + "grad_norm": 2.5883984565734863, + "learning_rate": 1.1494719740048742e-06, + "loss": 1.6416, + "step": 2186 + }, + { + "epoch": 2.3967123287671233, + "grad_norm": 2.5799498558044434, + "learning_rate": 1.1474411047928515e-06, + "loss": 1.7162, + "step": 2187 + }, + { + "epoch": 2.3978082191780823, + "grad_norm": 2.64442777633667, + "learning_rate": 1.1454102355808287e-06, + "loss": 1.709, + "step": 2188 + }, + { + "epoch": 2.398904109589041, + "grad_norm": 2.6813411712646484, + "learning_rate": 1.143379366368806e-06, + "loss": 1.6903, + "step": 2189 + }, + { + "epoch": 2.4, + "grad_norm": 2.6276469230651855, + "learning_rate": 1.141348497156783e-06, + "loss": 1.6734, + "step": 2190 + }, + { + "epoch": 2.4, + "eval_loss": 1.9423893690109253, + "eval_runtime": 37.1963, + "eval_samples_per_second": 26.884, + "eval_steps_per_second": 1.129, + "step": 2190 + }, + { + "epoch": 2.401095890410959, + "grad_norm": 2.6805641651153564, + "learning_rate": 1.1393176279447604e-06, + "loss": 1.8038, + "step": 2191 + }, + { + "epoch": 2.402191780821918, + "grad_norm": 2.813957691192627, + "learning_rate": 1.1372867587327376e-06, + "loss": 1.7895, + "step": 2192 + }, + { + "epoch": 2.4032876712328766, + "grad_norm": 2.505695343017578, + "learning_rate": 1.135255889520715e-06, + "loss": 1.6398, + "step": 2193 + }, + { + "epoch": 2.4043835616438356, + "grad_norm": 2.7426211833953857, + "learning_rate": 1.1332250203086922e-06, + "loss": 1.7472, + "step": 2194 + }, + { + "epoch": 2.4054794520547946, + "grad_norm": 2.627737522125244, + "learning_rate": 1.1311941510966695e-06, + "loss": 1.7272, + "step": 2195 + }, + { + "epoch": 2.4065753424657537, + "grad_norm": 2.615032911300659, + "learning_rate": 1.1291632818846467e-06, + "loss": 1.7004, + "step": 2196 + }, + { + "epoch": 2.4076712328767123, + "grad_norm": 2.5147581100463867, + "learning_rate": 1.127132412672624e-06, + "loss": 1.6894, + "step": 2197 + }, + { + "epoch": 2.4087671232876713, + "grad_norm": 2.7482388019561768, + "learning_rate": 1.1251015434606013e-06, + "loss": 1.6738, + "step": 2198 + }, + { + "epoch": 2.4098630136986303, + "grad_norm": 2.7309365272521973, + "learning_rate": 1.1230706742485786e-06, + "loss": 1.7222, + "step": 2199 + }, + { + "epoch": 2.410958904109589, + "grad_norm": 2.8322935104370117, + "learning_rate": 1.1210398050365558e-06, + "loss": 1.7762, + "step": 2200 + }, + { + "epoch": 2.410958904109589, + "eval_loss": 1.9395875930786133, + "eval_runtime": 37.2409, + "eval_samples_per_second": 26.852, + "eval_steps_per_second": 1.128, + "step": 2200 + }, + { + "epoch": 2.412054794520548, + "grad_norm": 2.7829575538635254, + "learning_rate": 1.1190089358245331e-06, + "loss": 1.7092, + "step": 2201 + }, + { + "epoch": 2.413150684931507, + "grad_norm": 2.518369197845459, + "learning_rate": 1.1169780666125102e-06, + "loss": 1.655, + "step": 2202 + }, + { + "epoch": 2.4142465753424656, + "grad_norm": 2.715089797973633, + "learning_rate": 1.1149471974004875e-06, + "loss": 1.7461, + "step": 2203 + }, + { + "epoch": 2.4153424657534246, + "grad_norm": 2.7772066593170166, + "learning_rate": 1.1129163281884647e-06, + "loss": 1.6931, + "step": 2204 + }, + { + "epoch": 2.4164383561643836, + "grad_norm": 2.813166379928589, + "learning_rate": 1.110885458976442e-06, + "loss": 1.7514, + "step": 2205 + }, + { + "epoch": 2.417534246575342, + "grad_norm": 2.6345536708831787, + "learning_rate": 1.1088545897644193e-06, + "loss": 1.6865, + "step": 2206 + }, + { + "epoch": 2.4186301369863012, + "grad_norm": 2.6846344470977783, + "learning_rate": 1.1068237205523966e-06, + "loss": 1.7564, + "step": 2207 + }, + { + "epoch": 2.4197260273972603, + "grad_norm": 2.6230814456939697, + "learning_rate": 1.1047928513403736e-06, + "loss": 1.7499, + "step": 2208 + }, + { + "epoch": 2.4208219178082193, + "grad_norm": 2.6820080280303955, + "learning_rate": 1.102761982128351e-06, + "loss": 1.7494, + "step": 2209 + }, + { + "epoch": 2.421917808219178, + "grad_norm": 2.572378396987915, + "learning_rate": 1.1007311129163282e-06, + "loss": 1.6585, + "step": 2210 + }, + { + "epoch": 2.421917808219178, + "eval_loss": 1.9389609098434448, + "eval_runtime": 37.2665, + "eval_samples_per_second": 26.834, + "eval_steps_per_second": 1.127, + "step": 2210 + }, + { + "epoch": 2.423013698630137, + "grad_norm": 2.7218871116638184, + "learning_rate": 1.0987002437043055e-06, + "loss": 1.7187, + "step": 2211 + }, + { + "epoch": 2.424109589041096, + "grad_norm": 2.7090065479278564, + "learning_rate": 1.0966693744922827e-06, + "loss": 1.6983, + "step": 2212 + }, + { + "epoch": 2.425205479452055, + "grad_norm": 2.6419243812561035, + "learning_rate": 1.09463850528026e-06, + "loss": 1.6485, + "step": 2213 + }, + { + "epoch": 2.4263013698630136, + "grad_norm": 2.635533571243286, + "learning_rate": 1.0926076360682373e-06, + "loss": 1.7023, + "step": 2214 + }, + { + "epoch": 2.4273972602739726, + "grad_norm": 2.644904375076294, + "learning_rate": 1.0905767668562146e-06, + "loss": 1.6774, + "step": 2215 + }, + { + "epoch": 2.4284931506849317, + "grad_norm": 2.6821508407592773, + "learning_rate": 1.0885458976441918e-06, + "loss": 1.6995, + "step": 2216 + }, + { + "epoch": 2.4295890410958902, + "grad_norm": 2.674916982650757, + "learning_rate": 1.0865150284321691e-06, + "loss": 1.7888, + "step": 2217 + }, + { + "epoch": 2.4306849315068493, + "grad_norm": 2.6966474056243896, + "learning_rate": 1.0844841592201464e-06, + "loss": 1.7303, + "step": 2218 + }, + { + "epoch": 2.4317808219178083, + "grad_norm": 2.6421585083007812, + "learning_rate": 1.0824532900081237e-06, + "loss": 1.7497, + "step": 2219 + }, + { + "epoch": 2.432876712328767, + "grad_norm": 2.8147826194763184, + "learning_rate": 1.080422420796101e-06, + "loss": 1.7088, + "step": 2220 + }, + { + "epoch": 2.432876712328767, + "eval_loss": 1.9437190294265747, + "eval_runtime": 37.5268, + "eval_samples_per_second": 26.648, + "eval_steps_per_second": 1.119, + "step": 2220 + }, + { + "epoch": 2.433972602739726, + "grad_norm": 2.6919329166412354, + "learning_rate": 1.078391551584078e-06, + "loss": 1.7168, + "step": 2221 + }, + { + "epoch": 2.435068493150685, + "grad_norm": 2.6179897785186768, + "learning_rate": 1.0763606823720553e-06, + "loss": 1.7477, + "step": 2222 + }, + { + "epoch": 2.436164383561644, + "grad_norm": 2.7324624061584473, + "learning_rate": 1.0743298131600326e-06, + "loss": 1.7921, + "step": 2223 + }, + { + "epoch": 2.4372602739726026, + "grad_norm": 2.6431500911712646, + "learning_rate": 1.0722989439480098e-06, + "loss": 1.7432, + "step": 2224 + }, + { + "epoch": 2.4383561643835616, + "grad_norm": 2.593528985977173, + "learning_rate": 1.070268074735987e-06, + "loss": 1.7097, + "step": 2225 + }, + { + "epoch": 2.4394520547945207, + "grad_norm": 2.6277310848236084, + "learning_rate": 1.0682372055239644e-06, + "loss": 1.6911, + "step": 2226 + }, + { + "epoch": 2.4405479452054797, + "grad_norm": 2.5935580730438232, + "learning_rate": 1.0662063363119414e-06, + "loss": 1.6933, + "step": 2227 + }, + { + "epoch": 2.4416438356164383, + "grad_norm": 2.672347068786621, + "learning_rate": 1.0641754670999187e-06, + "loss": 1.6826, + "step": 2228 + }, + { + "epoch": 2.4427397260273973, + "grad_norm": 2.7214746475219727, + "learning_rate": 1.062144597887896e-06, + "loss": 1.7152, + "step": 2229 + }, + { + "epoch": 2.4438356164383563, + "grad_norm": 2.7052741050720215, + "learning_rate": 1.0601137286758733e-06, + "loss": 1.7119, + "step": 2230 + }, + { + "epoch": 2.4438356164383563, + "eval_loss": 1.9452862739562988, + "eval_runtime": 37.6622, + "eval_samples_per_second": 26.552, + "eval_steps_per_second": 1.115, + "step": 2230 + }, + { + "epoch": 2.444931506849315, + "grad_norm": 2.7374205589294434, + "learning_rate": 1.0580828594638505e-06, + "loss": 1.7904, + "step": 2231 + }, + { + "epoch": 2.446027397260274, + "grad_norm": 2.6822078227996826, + "learning_rate": 1.0560519902518278e-06, + "loss": 1.6238, + "step": 2232 + }, + { + "epoch": 2.447123287671233, + "grad_norm": 2.6262941360473633, + "learning_rate": 1.054021121039805e-06, + "loss": 1.7501, + "step": 2233 + }, + { + "epoch": 2.4482191780821916, + "grad_norm": 2.689777374267578, + "learning_rate": 1.0519902518277824e-06, + "loss": 1.6604, + "step": 2234 + }, + { + "epoch": 2.4493150684931506, + "grad_norm": 2.6919682025909424, + "learning_rate": 1.0499593826157596e-06, + "loss": 1.6493, + "step": 2235 + }, + { + "epoch": 2.4504109589041096, + "grad_norm": 2.732959270477295, + "learning_rate": 1.047928513403737e-06, + "loss": 1.7429, + "step": 2236 + }, + { + "epoch": 2.4515068493150687, + "grad_norm": 2.6201255321502686, + "learning_rate": 1.0458976441917142e-06, + "loss": 1.7131, + "step": 2237 + }, + { + "epoch": 2.4526027397260273, + "grad_norm": 2.655271291732788, + "learning_rate": 1.0438667749796915e-06, + "loss": 1.634, + "step": 2238 + }, + { + "epoch": 2.4536986301369863, + "grad_norm": 2.6041507720947266, + "learning_rate": 1.0418359057676688e-06, + "loss": 1.6384, + "step": 2239 + }, + { + "epoch": 2.4547945205479453, + "grad_norm": 2.582871198654175, + "learning_rate": 1.0398050365556458e-06, + "loss": 1.6903, + "step": 2240 + }, + { + "epoch": 2.4547945205479453, + "eval_loss": 1.9449145793914795, + "eval_runtime": 37.4116, + "eval_samples_per_second": 26.73, + "eval_steps_per_second": 1.123, + "step": 2240 + }, + { + "epoch": 2.455890410958904, + "grad_norm": 2.647946357727051, + "learning_rate": 1.037774167343623e-06, + "loss": 1.752, + "step": 2241 + }, + { + "epoch": 2.456986301369863, + "grad_norm": 2.6174609661102295, + "learning_rate": 1.0357432981316004e-06, + "loss": 1.6745, + "step": 2242 + }, + { + "epoch": 2.458082191780822, + "grad_norm": 2.7191693782806396, + "learning_rate": 1.0337124289195776e-06, + "loss": 1.7291, + "step": 2243 + }, + { + "epoch": 2.459178082191781, + "grad_norm": 2.6612565517425537, + "learning_rate": 1.031681559707555e-06, + "loss": 1.7523, + "step": 2244 + }, + { + "epoch": 2.4602739726027396, + "grad_norm": 2.617875337600708, + "learning_rate": 1.0296506904955322e-06, + "loss": 1.7597, + "step": 2245 + }, + { + "epoch": 2.4613698630136986, + "grad_norm": 2.730677843093872, + "learning_rate": 1.0276198212835095e-06, + "loss": 1.7146, + "step": 2246 + }, + { + "epoch": 2.4624657534246577, + "grad_norm": 2.6297593116760254, + "learning_rate": 1.0255889520714867e-06, + "loss": 1.7183, + "step": 2247 + }, + { + "epoch": 2.4635616438356163, + "grad_norm": 2.600847005844116, + "learning_rate": 1.023558082859464e-06, + "loss": 1.7086, + "step": 2248 + }, + { + "epoch": 2.4646575342465753, + "grad_norm": 2.773501396179199, + "learning_rate": 1.0215272136474413e-06, + "loss": 1.6708, + "step": 2249 + }, + { + "epoch": 2.4657534246575343, + "grad_norm": 2.6344449520111084, + "learning_rate": 1.0194963444354186e-06, + "loss": 1.7818, + "step": 2250 + }, + { + "epoch": 2.4657534246575343, + "eval_loss": 1.940114974975586, + "eval_runtime": 37.3629, + "eval_samples_per_second": 26.764, + "eval_steps_per_second": 1.124, + "step": 2250 + }, + { + "epoch": 2.466849315068493, + "grad_norm": 2.6345012187957764, + "learning_rate": 1.0174654752233958e-06, + "loss": 1.6525, + "step": 2251 + }, + { + "epoch": 2.467945205479452, + "grad_norm": 2.6919827461242676, + "learning_rate": 1.015434606011373e-06, + "loss": 1.7457, + "step": 2252 + }, + { + "epoch": 2.469041095890411, + "grad_norm": 2.6676719188690186, + "learning_rate": 1.0134037367993502e-06, + "loss": 1.6934, + "step": 2253 + }, + { + "epoch": 2.47013698630137, + "grad_norm": 2.623911142349243, + "learning_rate": 1.0113728675873275e-06, + "loss": 1.7089, + "step": 2254 + }, + { + "epoch": 2.4712328767123286, + "grad_norm": 2.6342320442199707, + "learning_rate": 1.0093419983753047e-06, + "loss": 1.7123, + "step": 2255 + }, + { + "epoch": 2.4723287671232876, + "grad_norm": 2.8066930770874023, + "learning_rate": 1.007311129163282e-06, + "loss": 1.7095, + "step": 2256 + }, + { + "epoch": 2.4734246575342467, + "grad_norm": 2.6137008666992188, + "learning_rate": 1.0052802599512593e-06, + "loss": 1.6602, + "step": 2257 + }, + { + "epoch": 2.4745205479452057, + "grad_norm": 2.6158881187438965, + "learning_rate": 1.0032493907392364e-06, + "loss": 1.6659, + "step": 2258 + }, + { + "epoch": 2.4756164383561643, + "grad_norm": 2.7409005165100098, + "learning_rate": 1.0012185215272136e-06, + "loss": 1.7194, + "step": 2259 + }, + { + "epoch": 2.4767123287671233, + "grad_norm": 2.783179759979248, + "learning_rate": 9.99187652315191e-07, + "loss": 1.658, + "step": 2260 + }, + { + "epoch": 2.4767123287671233, + "eval_loss": 1.9401936531066895, + "eval_runtime": 37.2693, + "eval_samples_per_second": 26.832, + "eval_steps_per_second": 1.127, + "step": 2260 + }, + { + "epoch": 2.4778082191780824, + "grad_norm": 2.7862110137939453, + "learning_rate": 9.971567831031682e-07, + "loss": 1.7535, + "step": 2261 + }, + { + "epoch": 2.478904109589041, + "grad_norm": 2.623033046722412, + "learning_rate": 9.951259138911455e-07, + "loss": 1.7285, + "step": 2262 + }, + { + "epoch": 2.48, + "grad_norm": 2.5325446128845215, + "learning_rate": 9.930950446791227e-07, + "loss": 1.6486, + "step": 2263 + }, + { + "epoch": 2.481095890410959, + "grad_norm": 2.6013123989105225, + "learning_rate": 9.910641754671e-07, + "loss": 1.7677, + "step": 2264 + }, + { + "epoch": 2.4821917808219176, + "grad_norm": 2.706526279449463, + "learning_rate": 9.890333062550773e-07, + "loss": 1.81, + "step": 2265 + }, + { + "epoch": 2.4832876712328766, + "grad_norm": 2.6389000415802, + "learning_rate": 9.870024370430546e-07, + "loss": 1.7104, + "step": 2266 + }, + { + "epoch": 2.4843835616438357, + "grad_norm": 2.8673291206359863, + "learning_rate": 9.849715678310318e-07, + "loss": 1.7439, + "step": 2267 + }, + { + "epoch": 2.4854794520547947, + "grad_norm": 2.574463367462158, + "learning_rate": 9.829406986190091e-07, + "loss": 1.7134, + "step": 2268 + }, + { + "epoch": 2.4865753424657533, + "grad_norm": 2.6605708599090576, + "learning_rate": 9.809098294069864e-07, + "loss": 1.7266, + "step": 2269 + }, + { + "epoch": 2.4876712328767123, + "grad_norm": 2.6304521560668945, + "learning_rate": 9.788789601949637e-07, + "loss": 1.6388, + "step": 2270 + }, + { + "epoch": 2.4876712328767123, + "eval_loss": 1.9415092468261719, + "eval_runtime": 37.2274, + "eval_samples_per_second": 26.862, + "eval_steps_per_second": 1.128, + "step": 2270 + }, + { + "epoch": 2.4887671232876714, + "grad_norm": 2.6577560901641846, + "learning_rate": 9.768480909829407e-07, + "loss": 1.7252, + "step": 2271 + }, + { + "epoch": 2.48986301369863, + "grad_norm": 2.801530122756958, + "learning_rate": 9.74817221770918e-07, + "loss": 1.7774, + "step": 2272 + }, + { + "epoch": 2.490958904109589, + "grad_norm": 2.761235237121582, + "learning_rate": 9.727863525588953e-07, + "loss": 1.6889, + "step": 2273 + }, + { + "epoch": 2.492054794520548, + "grad_norm": 2.719083547592163, + "learning_rate": 9.707554833468726e-07, + "loss": 1.7231, + "step": 2274 + }, + { + "epoch": 2.493150684931507, + "grad_norm": 2.5667307376861572, + "learning_rate": 9.687246141348498e-07, + "loss": 1.7301, + "step": 2275 + }, + { + "epoch": 2.4942465753424656, + "grad_norm": 2.6383678913116455, + "learning_rate": 9.666937449228271e-07, + "loss": 1.6999, + "step": 2276 + }, + { + "epoch": 2.4953424657534247, + "grad_norm": 2.7380318641662598, + "learning_rate": 9.646628757108042e-07, + "loss": 1.7202, + "step": 2277 + }, + { + "epoch": 2.4964383561643837, + "grad_norm": 2.715460777282715, + "learning_rate": 9.626320064987814e-07, + "loss": 1.6626, + "step": 2278 + }, + { + "epoch": 2.4975342465753423, + "grad_norm": 2.5860559940338135, + "learning_rate": 9.606011372867587e-07, + "loss": 1.7345, + "step": 2279 + }, + { + "epoch": 2.4986301369863013, + "grad_norm": 2.614272356033325, + "learning_rate": 9.58570268074736e-07, + "loss": 1.6728, + "step": 2280 + }, + { + "epoch": 2.4986301369863013, + "eval_loss": 1.9393996000289917, + "eval_runtime": 37.2508, + "eval_samples_per_second": 26.845, + "eval_steps_per_second": 1.127, + "step": 2280 + }, + { + "epoch": 2.4997260273972604, + "grad_norm": 2.6709139347076416, + "learning_rate": 9.565393988627133e-07, + "loss": 1.7094, + "step": 2281 + }, + { + "epoch": 2.500821917808219, + "grad_norm": 2.468263626098633, + "learning_rate": 9.545085296506906e-07, + "loss": 1.7044, + "step": 2282 + }, + { + "epoch": 2.501917808219178, + "grad_norm": 2.646242141723633, + "learning_rate": 9.524776604386677e-07, + "loss": 1.7144, + "step": 2283 + }, + { + "epoch": 2.503013698630137, + "grad_norm": 2.6524763107299805, + "learning_rate": 9.50446791226645e-07, + "loss": 1.7221, + "step": 2284 + }, + { + "epoch": 2.504109589041096, + "grad_norm": 2.681304693222046, + "learning_rate": 9.484159220146223e-07, + "loss": 1.6446, + "step": 2285 + }, + { + "epoch": 2.5052054794520546, + "grad_norm": 2.5855886936187744, + "learning_rate": 9.463850528025995e-07, + "loss": 1.757, + "step": 2286 + }, + { + "epoch": 2.5063013698630137, + "grad_norm": 2.5410566329956055, + "learning_rate": 9.443541835905768e-07, + "loss": 1.7487, + "step": 2287 + }, + { + "epoch": 2.5073972602739727, + "grad_norm": 2.726349115371704, + "learning_rate": 9.423233143785541e-07, + "loss": 1.7841, + "step": 2288 + }, + { + "epoch": 2.5084931506849317, + "grad_norm": 2.7225544452667236, + "learning_rate": 9.402924451665313e-07, + "loss": 1.7921, + "step": 2289 + }, + { + "epoch": 2.5095890410958903, + "grad_norm": 2.7081549167633057, + "learning_rate": 9.382615759545085e-07, + "loss": 1.6726, + "step": 2290 + }, + { + "epoch": 2.5095890410958903, + "eval_loss": 1.9398577213287354, + "eval_runtime": 37.2122, + "eval_samples_per_second": 26.873, + "eval_steps_per_second": 1.129, + "step": 2290 + }, + { + "epoch": 2.5106849315068493, + "grad_norm": 2.6577272415161133, + "learning_rate": 9.362307067424858e-07, + "loss": 1.7571, + "step": 2291 + }, + { + "epoch": 2.5117808219178084, + "grad_norm": 2.659438133239746, + "learning_rate": 9.341998375304631e-07, + "loss": 1.7113, + "step": 2292 + }, + { + "epoch": 2.512876712328767, + "grad_norm": 2.6679470539093018, + "learning_rate": 9.321689683184404e-07, + "loss": 1.7718, + "step": 2293 + }, + { + "epoch": 2.513972602739726, + "grad_norm": 2.703716993331909, + "learning_rate": 9.301380991064176e-07, + "loss": 1.6617, + "step": 2294 + }, + { + "epoch": 2.515068493150685, + "grad_norm": 2.66267728805542, + "learning_rate": 9.281072298943949e-07, + "loss": 1.6479, + "step": 2295 + }, + { + "epoch": 2.5161643835616436, + "grad_norm": 2.7160699367523193, + "learning_rate": 9.260763606823721e-07, + "loss": 1.6931, + "step": 2296 + }, + { + "epoch": 2.5172602739726027, + "grad_norm": 2.7481086254119873, + "learning_rate": 9.240454914703494e-07, + "loss": 1.73, + "step": 2297 + }, + { + "epoch": 2.5183561643835617, + "grad_norm": 2.702737331390381, + "learning_rate": 9.220146222583266e-07, + "loss": 1.7414, + "step": 2298 + }, + { + "epoch": 2.5194520547945203, + "grad_norm": 2.78155517578125, + "learning_rate": 9.199837530463039e-07, + "loss": 1.684, + "step": 2299 + }, + { + "epoch": 2.5205479452054793, + "grad_norm": 2.5439021587371826, + "learning_rate": 9.179528838342812e-07, + "loss": 1.7116, + "step": 2300 + }, + { + "epoch": 2.5205479452054793, + "eval_loss": 1.9369380474090576, + "eval_runtime": 37.2139, + "eval_samples_per_second": 26.872, + "eval_steps_per_second": 1.129, + "step": 2300 + }, + { + "epoch": 2.5216438356164383, + "grad_norm": 2.5379269123077393, + "learning_rate": 9.159220146222585e-07, + "loss": 1.6888, + "step": 2301 + }, + { + "epoch": 2.5227397260273974, + "grad_norm": 2.6183183193206787, + "learning_rate": 9.138911454102356e-07, + "loss": 1.7297, + "step": 2302 + }, + { + "epoch": 2.5238356164383564, + "grad_norm": 2.6776952743530273, + "learning_rate": 9.118602761982129e-07, + "loss": 1.5998, + "step": 2303 + }, + { + "epoch": 2.524931506849315, + "grad_norm": 2.71006441116333, + "learning_rate": 9.098294069861902e-07, + "loss": 1.7272, + "step": 2304 + }, + { + "epoch": 2.526027397260274, + "grad_norm": 2.6362080574035645, + "learning_rate": 9.077985377741675e-07, + "loss": 1.7147, + "step": 2305 + }, + { + "epoch": 2.527123287671233, + "grad_norm": 2.629568576812744, + "learning_rate": 9.057676685621447e-07, + "loss": 1.6762, + "step": 2306 + }, + { + "epoch": 2.5282191780821917, + "grad_norm": 2.4955074787139893, + "learning_rate": 9.03736799350122e-07, + "loss": 1.642, + "step": 2307 + }, + { + "epoch": 2.5293150684931507, + "grad_norm": 2.610558271408081, + "learning_rate": 9.017059301380991e-07, + "loss": 1.7773, + "step": 2308 + }, + { + "epoch": 2.5304109589041097, + "grad_norm": 2.5644302368164062, + "learning_rate": 8.996750609260764e-07, + "loss": 1.7524, + "step": 2309 + }, + { + "epoch": 2.5315068493150683, + "grad_norm": 2.6172549724578857, + "learning_rate": 8.976441917140536e-07, + "loss": 1.7865, + "step": 2310 + }, + { + "epoch": 2.5315068493150683, + "eval_loss": 1.9368717670440674, + "eval_runtime": 37.2323, + "eval_samples_per_second": 26.858, + "eval_steps_per_second": 1.128, + "step": 2310 + }, + { + "epoch": 2.5326027397260273, + "grad_norm": 2.6722280979156494, + "learning_rate": 8.956133225020309e-07, + "loss": 1.7449, + "step": 2311 + }, + { + "epoch": 2.5336986301369864, + "grad_norm": 2.6534717082977295, + "learning_rate": 8.935824532900082e-07, + "loss": 1.7844, + "step": 2312 + }, + { + "epoch": 2.534794520547945, + "grad_norm": 2.566446542739868, + "learning_rate": 8.915515840779855e-07, + "loss": 1.6826, + "step": 2313 + }, + { + "epoch": 2.535890410958904, + "grad_norm": 2.6202712059020996, + "learning_rate": 8.895207148659626e-07, + "loss": 1.6535, + "step": 2314 + }, + { + "epoch": 2.536986301369863, + "grad_norm": 2.6147568225860596, + "learning_rate": 8.874898456539399e-07, + "loss": 1.7338, + "step": 2315 + }, + { + "epoch": 2.538082191780822, + "grad_norm": 2.6190552711486816, + "learning_rate": 8.854589764419172e-07, + "loss": 1.7297, + "step": 2316 + }, + { + "epoch": 2.539178082191781, + "grad_norm": 2.7838358879089355, + "learning_rate": 8.834281072298945e-07, + "loss": 1.7372, + "step": 2317 + }, + { + "epoch": 2.5402739726027397, + "grad_norm": 2.6310222148895264, + "learning_rate": 8.813972380178717e-07, + "loss": 1.6706, + "step": 2318 + }, + { + "epoch": 2.5413698630136987, + "grad_norm": 2.7847073078155518, + "learning_rate": 8.79366368805849e-07, + "loss": 1.6967, + "step": 2319 + }, + { + "epoch": 2.5424657534246577, + "grad_norm": 2.6233162879943848, + "learning_rate": 8.773354995938262e-07, + "loss": 1.7215, + "step": 2320 + }, + { + "epoch": 2.5424657534246577, + "eval_loss": 1.9374815225601196, + "eval_runtime": 37.5549, + "eval_samples_per_second": 26.628, + "eval_steps_per_second": 1.118, + "step": 2320 + }, + { + "epoch": 2.5435616438356163, + "grad_norm": 2.6680665016174316, + "learning_rate": 8.753046303818035e-07, + "loss": 1.6705, + "step": 2321 + }, + { + "epoch": 2.5446575342465754, + "grad_norm": 2.570737600326538, + "learning_rate": 8.732737611697807e-07, + "loss": 1.6617, + "step": 2322 + }, + { + "epoch": 2.5457534246575344, + "grad_norm": 2.6836602687835693, + "learning_rate": 8.71242891957758e-07, + "loss": 1.6989, + "step": 2323 + }, + { + "epoch": 2.546849315068493, + "grad_norm": 2.637993812561035, + "learning_rate": 8.692120227457353e-07, + "loss": 1.7281, + "step": 2324 + }, + { + "epoch": 2.547945205479452, + "grad_norm": 2.61149001121521, + "learning_rate": 8.671811535337126e-07, + "loss": 1.7459, + "step": 2325 + }, + { + "epoch": 2.549041095890411, + "grad_norm": 2.6190171241760254, + "learning_rate": 8.651502843216898e-07, + "loss": 1.6925, + "step": 2326 + }, + { + "epoch": 2.5501369863013696, + "grad_norm": 2.6211154460906982, + "learning_rate": 8.63119415109667e-07, + "loss": 1.7999, + "step": 2327 + }, + { + "epoch": 2.5512328767123287, + "grad_norm": 2.6447153091430664, + "learning_rate": 8.610885458976443e-07, + "loss": 1.7356, + "step": 2328 + }, + { + "epoch": 2.5523287671232877, + "grad_norm": 2.6078338623046875, + "learning_rate": 8.590576766856216e-07, + "loss": 1.7281, + "step": 2329 + }, + { + "epoch": 2.5534246575342463, + "grad_norm": 2.5882375240325928, + "learning_rate": 8.570268074735988e-07, + "loss": 1.6858, + "step": 2330 + }, + { + "epoch": 2.5534246575342463, + "eval_loss": 1.9356836080551147, + "eval_runtime": 37.5283, + "eval_samples_per_second": 26.647, + "eval_steps_per_second": 1.119, + "step": 2330 + }, + { + "epoch": 2.5545205479452053, + "grad_norm": 2.6218528747558594, + "learning_rate": 8.549959382615761e-07, + "loss": 1.6712, + "step": 2331 + }, + { + "epoch": 2.5556164383561644, + "grad_norm": 2.60943341255188, + "learning_rate": 8.529650690495534e-07, + "loss": 1.6924, + "step": 2332 + }, + { + "epoch": 2.5567123287671234, + "grad_norm": 2.8169174194335938, + "learning_rate": 8.509341998375305e-07, + "loss": 1.715, + "step": 2333 + }, + { + "epoch": 2.5578082191780824, + "grad_norm": 2.601630210876465, + "learning_rate": 8.489033306255077e-07, + "loss": 1.6473, + "step": 2334 + }, + { + "epoch": 2.558904109589041, + "grad_norm": 2.622537612915039, + "learning_rate": 8.46872461413485e-07, + "loss": 1.6131, + "step": 2335 + }, + { + "epoch": 2.56, + "grad_norm": 2.57839035987854, + "learning_rate": 8.448415922014623e-07, + "loss": 1.6713, + "step": 2336 + }, + { + "epoch": 2.561095890410959, + "grad_norm": 2.7201271057128906, + "learning_rate": 8.428107229894396e-07, + "loss": 1.6825, + "step": 2337 + }, + { + "epoch": 2.5621917808219177, + "grad_norm": 2.57549786567688, + "learning_rate": 8.407798537774168e-07, + "loss": 1.7242, + "step": 2338 + }, + { + "epoch": 2.5632876712328767, + "grad_norm": 2.8006980419158936, + "learning_rate": 8.38748984565394e-07, + "loss": 1.7053, + "step": 2339 + }, + { + "epoch": 2.5643835616438357, + "grad_norm": 2.7859253883361816, + "learning_rate": 8.367181153533713e-07, + "loss": 1.6751, + "step": 2340 + }, + { + "epoch": 2.5643835616438357, + "eval_loss": 1.9371570348739624, + "eval_runtime": 37.3729, + "eval_samples_per_second": 26.757, + "eval_steps_per_second": 1.124, + "step": 2340 + }, + { + "epoch": 2.5654794520547943, + "grad_norm": 2.664045572280884, + "learning_rate": 8.346872461413486e-07, + "loss": 1.6977, + "step": 2341 + }, + { + "epoch": 2.5665753424657534, + "grad_norm": 2.7048933506011963, + "learning_rate": 8.326563769293258e-07, + "loss": 1.691, + "step": 2342 + }, + { + "epoch": 2.5676712328767124, + "grad_norm": 2.6367626190185547, + "learning_rate": 8.306255077173031e-07, + "loss": 1.6999, + "step": 2343 + }, + { + "epoch": 2.568767123287671, + "grad_norm": 2.588322162628174, + "learning_rate": 8.285946385052804e-07, + "loss": 1.7915, + "step": 2344 + }, + { + "epoch": 2.56986301369863, + "grad_norm": 2.6249992847442627, + "learning_rate": 8.265637692932575e-07, + "loss": 1.6534, + "step": 2345 + }, + { + "epoch": 2.570958904109589, + "grad_norm": 2.6095449924468994, + "learning_rate": 8.245329000812348e-07, + "loss": 1.7339, + "step": 2346 + }, + { + "epoch": 2.572054794520548, + "grad_norm": 2.6527903079986572, + "learning_rate": 8.225020308692121e-07, + "loss": 1.7199, + "step": 2347 + }, + { + "epoch": 2.573150684931507, + "grad_norm": 2.6879372596740723, + "learning_rate": 8.204711616571894e-07, + "loss": 1.6669, + "step": 2348 + }, + { + "epoch": 2.5742465753424657, + "grad_norm": 2.6295018196105957, + "learning_rate": 8.184402924451667e-07, + "loss": 1.7001, + "step": 2349 + }, + { + "epoch": 2.5753424657534247, + "grad_norm": 2.669123888015747, + "learning_rate": 8.164094232331439e-07, + "loss": 1.6573, + "step": 2350 + }, + { + "epoch": 2.5753424657534247, + "eval_loss": 1.9378976821899414, + "eval_runtime": 37.3297, + "eval_samples_per_second": 26.788, + "eval_steps_per_second": 1.125, + "step": 2350 + }, + { + "epoch": 2.5764383561643838, + "grad_norm": 2.5638668537139893, + "learning_rate": 8.14378554021121e-07, + "loss": 1.7396, + "step": 2351 + }, + { + "epoch": 2.5775342465753424, + "grad_norm": 2.613858461380005, + "learning_rate": 8.123476848090983e-07, + "loss": 1.7081, + "step": 2352 + }, + { + "epoch": 2.5786301369863014, + "grad_norm": 2.633657693862915, + "learning_rate": 8.103168155970755e-07, + "loss": 1.6905, + "step": 2353 + }, + { + "epoch": 2.5797260273972604, + "grad_norm": 2.600965738296509, + "learning_rate": 8.082859463850528e-07, + "loss": 1.6979, + "step": 2354 + }, + { + "epoch": 2.580821917808219, + "grad_norm": 2.7222094535827637, + "learning_rate": 8.062550771730301e-07, + "loss": 1.7267, + "step": 2355 + }, + { + "epoch": 2.581917808219178, + "grad_norm": 2.783076047897339, + "learning_rate": 8.042242079610074e-07, + "loss": 1.6906, + "step": 2356 + }, + { + "epoch": 2.583013698630137, + "grad_norm": 2.671109199523926, + "learning_rate": 8.021933387489846e-07, + "loss": 1.6822, + "step": 2357 + }, + { + "epoch": 2.5841095890410957, + "grad_norm": 2.602358818054199, + "learning_rate": 8.001624695369618e-07, + "loss": 1.6834, + "step": 2358 + }, + { + "epoch": 2.5852054794520547, + "grad_norm": 2.665668249130249, + "learning_rate": 7.981316003249391e-07, + "loss": 1.7003, + "step": 2359 + }, + { + "epoch": 2.5863013698630137, + "grad_norm": 2.827331781387329, + "learning_rate": 7.961007311129164e-07, + "loss": 1.7663, + "step": 2360 + }, + { + "epoch": 2.5863013698630137, + "eval_loss": 1.9384022951126099, + "eval_runtime": 37.2505, + "eval_samples_per_second": 26.845, + "eval_steps_per_second": 1.128, + "step": 2360 + }, + { + "epoch": 2.5873972602739728, + "grad_norm": 2.603614568710327, + "learning_rate": 7.940698619008936e-07, + "loss": 1.6496, + "step": 2361 + }, + { + "epoch": 2.5884931506849314, + "grad_norm": 2.695448875427246, + "learning_rate": 7.920389926888709e-07, + "loss": 1.7325, + "step": 2362 + }, + { + "epoch": 2.5895890410958904, + "grad_norm": 2.6337499618530273, + "learning_rate": 7.900081234768482e-07, + "loss": 1.8151, + "step": 2363 + }, + { + "epoch": 2.5906849315068494, + "grad_norm": 2.6526639461517334, + "learning_rate": 7.879772542648254e-07, + "loss": 1.6684, + "step": 2364 + }, + { + "epoch": 2.5917808219178085, + "grad_norm": 2.5932109355926514, + "learning_rate": 7.859463850528026e-07, + "loss": 1.7238, + "step": 2365 + }, + { + "epoch": 2.592876712328767, + "grad_norm": 2.6547865867614746, + "learning_rate": 7.839155158407799e-07, + "loss": 1.7204, + "step": 2366 + }, + { + "epoch": 2.593972602739726, + "grad_norm": 2.6299901008605957, + "learning_rate": 7.818846466287572e-07, + "loss": 1.72, + "step": 2367 + }, + { + "epoch": 2.595068493150685, + "grad_norm": 2.69770884513855, + "learning_rate": 7.798537774167345e-07, + "loss": 1.7157, + "step": 2368 + }, + { + "epoch": 2.5961643835616437, + "grad_norm": 2.721971273422241, + "learning_rate": 7.778229082047117e-07, + "loss": 1.7384, + "step": 2369 + }, + { + "epoch": 2.5972602739726027, + "grad_norm": 2.6927170753479004, + "learning_rate": 7.757920389926889e-07, + "loss": 1.6823, + "step": 2370 + }, + { + "epoch": 2.5972602739726027, + "eval_loss": 1.935421109199524, + "eval_runtime": 37.2369, + "eval_samples_per_second": 26.855, + "eval_steps_per_second": 1.128, + "step": 2370 + }, + { + "epoch": 2.5983561643835618, + "grad_norm": 2.6370513439178467, + "learning_rate": 7.737611697806662e-07, + "loss": 1.7193, + "step": 2371 + }, + { + "epoch": 2.5994520547945204, + "grad_norm": 2.6238462924957275, + "learning_rate": 7.717303005686435e-07, + "loss": 1.685, + "step": 2372 + }, + { + "epoch": 2.6005479452054794, + "grad_norm": 2.643711566925049, + "learning_rate": 7.696994313566207e-07, + "loss": 1.7318, + "step": 2373 + }, + { + "epoch": 2.6016438356164384, + "grad_norm": 2.6582396030426025, + "learning_rate": 7.67668562144598e-07, + "loss": 1.7782, + "step": 2374 + }, + { + "epoch": 2.602739726027397, + "grad_norm": 2.6774542331695557, + "learning_rate": 7.656376929325753e-07, + "loss": 1.6587, + "step": 2375 + }, + { + "epoch": 2.603835616438356, + "grad_norm": 2.611940622329712, + "learning_rate": 7.636068237205524e-07, + "loss": 1.6779, + "step": 2376 + }, + { + "epoch": 2.604931506849315, + "grad_norm": 2.6520285606384277, + "learning_rate": 7.615759545085296e-07, + "loss": 1.7437, + "step": 2377 + }, + { + "epoch": 2.606027397260274, + "grad_norm": 2.666217565536499, + "learning_rate": 7.595450852965069e-07, + "loss": 1.6655, + "step": 2378 + }, + { + "epoch": 2.607123287671233, + "grad_norm": 2.695321559906006, + "learning_rate": 7.575142160844842e-07, + "loss": 1.6835, + "step": 2379 + }, + { + "epoch": 2.6082191780821917, + "grad_norm": 2.6860222816467285, + "learning_rate": 7.554833468724615e-07, + "loss": 1.7316, + "step": 2380 + }, + { + "epoch": 2.6082191780821917, + "eval_loss": 1.9367016553878784, + "eval_runtime": 37.1903, + "eval_samples_per_second": 26.889, + "eval_steps_per_second": 1.129, + "step": 2380 + }, + { + "epoch": 2.6093150684931508, + "grad_norm": 2.6422245502471924, + "learning_rate": 7.534524776604387e-07, + "loss": 1.6533, + "step": 2381 + }, + { + "epoch": 2.61041095890411, + "grad_norm": 2.6544158458709717, + "learning_rate": 7.51421608448416e-07, + "loss": 1.6734, + "step": 2382 + }, + { + "epoch": 2.6115068493150684, + "grad_norm": 2.648153305053711, + "learning_rate": 7.493907392363932e-07, + "loss": 1.7374, + "step": 2383 + }, + { + "epoch": 2.6126027397260274, + "grad_norm": 2.646836042404175, + "learning_rate": 7.473598700243705e-07, + "loss": 1.6794, + "step": 2384 + }, + { + "epoch": 2.6136986301369864, + "grad_norm": 2.633972644805908, + "learning_rate": 7.453290008123477e-07, + "loss": 1.633, + "step": 2385 + }, + { + "epoch": 2.614794520547945, + "grad_norm": 2.580915927886963, + "learning_rate": 7.43298131600325e-07, + "loss": 1.6981, + "step": 2386 + }, + { + "epoch": 2.615890410958904, + "grad_norm": 2.605191946029663, + "learning_rate": 7.412672623883023e-07, + "loss": 1.711, + "step": 2387 + }, + { + "epoch": 2.616986301369863, + "grad_norm": 2.7410099506378174, + "learning_rate": 7.392363931762796e-07, + "loss": 1.7175, + "step": 2388 + }, + { + "epoch": 2.6180821917808217, + "grad_norm": 2.688540458679199, + "learning_rate": 7.372055239642567e-07, + "loss": 1.6447, + "step": 2389 + }, + { + "epoch": 2.6191780821917807, + "grad_norm": 2.719954252243042, + "learning_rate": 7.35174654752234e-07, + "loss": 1.7034, + "step": 2390 + }, + { + "epoch": 2.6191780821917807, + "eval_loss": 1.934234380722046, + "eval_runtime": 37.1917, + "eval_samples_per_second": 26.888, + "eval_steps_per_second": 1.129, + "step": 2390 + }, + { + "epoch": 2.6202739726027398, + "grad_norm": 2.7232038974761963, + "learning_rate": 7.331437855402113e-07, + "loss": 1.7318, + "step": 2391 + }, + { + "epoch": 2.621369863013699, + "grad_norm": 2.586639404296875, + "learning_rate": 7.311129163281886e-07, + "loss": 1.7529, + "step": 2392 + }, + { + "epoch": 2.6224657534246574, + "grad_norm": 2.643853187561035, + "learning_rate": 7.290820471161658e-07, + "loss": 1.6809, + "step": 2393 + }, + { + "epoch": 2.6235616438356164, + "grad_norm": 2.6359288692474365, + "learning_rate": 7.270511779041431e-07, + "loss": 1.7261, + "step": 2394 + }, + { + "epoch": 2.6246575342465754, + "grad_norm": 2.6692726612091064, + "learning_rate": 7.250203086921203e-07, + "loss": 1.7311, + "step": 2395 + }, + { + "epoch": 2.6257534246575345, + "grad_norm": 2.6712090969085693, + "learning_rate": 7.229894394800976e-07, + "loss": 1.6367, + "step": 2396 + }, + { + "epoch": 2.626849315068493, + "grad_norm": 2.618422031402588, + "learning_rate": 7.209585702680748e-07, + "loss": 1.7017, + "step": 2397 + }, + { + "epoch": 2.627945205479452, + "grad_norm": 2.6770448684692383, + "learning_rate": 7.189277010560521e-07, + "loss": 1.7193, + "step": 2398 + }, + { + "epoch": 2.629041095890411, + "grad_norm": 2.605620861053467, + "learning_rate": 7.168968318440294e-07, + "loss": 1.6515, + "step": 2399 + }, + { + "epoch": 2.6301369863013697, + "grad_norm": 2.616734266281128, + "learning_rate": 7.148659626320067e-07, + "loss": 1.685, + "step": 2400 + }, + { + "epoch": 2.6301369863013697, + "eval_loss": 1.9330971240997314, + "eval_runtime": 37.1928, + "eval_samples_per_second": 26.887, + "eval_steps_per_second": 1.129, + "step": 2400 + } + ], + "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": 7.481542582468608e+17, + "train_batch_size": 3, + "trial_name": null, + "trial_params": null +} diff --git a/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-2400/zero_to_fp32.py b/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-2400/zero_to_fp32.py new file mode 100644 index 0000000000000000000000000000000000000000..24cc342e78d1a006c782b3a4cd68d9ce786d8fd8 --- /dev/null +++ b/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-2400/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/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-2500/latest b/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-2500/latest new file mode 100644 index 0000000000000000000000000000000000000000..98f8bed9a5485ee900d9931cc06950de69499848 --- /dev/null +++ b/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-2500/latest @@ -0,0 +1 @@ +global_step2500 \ No newline at end of file diff --git a/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-2500/zero_to_fp32.py b/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-2500/zero_to_fp32.py new file mode 100644 index 0000000000000000000000000000000000000000..24cc342e78d1a006c782b3a4cd68d9ce786d8fd8 --- /dev/null +++ b/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-2500/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/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-2600/config.json b/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-2600/config.json new file mode 100644 index 0000000000000000000000000000000000000000..fb0f9a53a507ffcdb60410deb71d60ef801bf350 --- /dev/null +++ b/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-2600/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/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-2600/generation_config.json b/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-2600/generation_config.json new file mode 100644 index 0000000000000000000000000000000000000000..9c389b841a9a29ddd904e02d1eb0e08dcce82ad9 --- /dev/null +++ b/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-2600/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/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-2600/latest b/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-2600/latest new file mode 100644 index 0000000000000000000000000000000000000000..cfe584fa09082a54d48545d05e97aee6fd00673d --- /dev/null +++ b/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-2600/latest @@ -0,0 +1 @@ +global_step2600 \ No newline at end of file diff --git a/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-2600/model.safetensors.index.json b/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-2600/model.safetensors.index.json new file mode 100644 index 0000000000000000000000000000000000000000..ed64de846d720b9a7859dc20575fea8e8ca51940 --- /dev/null +++ b/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-2600/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/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-2600/special_tokens_map.json b/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-2600/special_tokens_map.json new file mode 100644 index 0000000000000000000000000000000000000000..6949c5975ee0e961ef61cf31010dce04df0a03f8 --- /dev/null +++ b/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-2600/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/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-2600/tokenizer_config.json b/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-2600/tokenizer_config.json new file mode 100644 index 0000000000000000000000000000000000000000..5684b14e2756b56aa397c1324e239b442162b34a --- /dev/null +++ b/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-2600/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/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-2600/trainer_state.json b/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-2600/trainer_state.json new file mode 100644 index 0000000000000000000000000000000000000000..d4a59a415aba69c9dc4ce93487ca48dea4e95e5c --- /dev/null +++ b/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-2600/trainer_state.json @@ -0,0 +1,20313 @@ +{ + "best_metric": null, + "best_model_checkpoint": null, + "epoch": 2.8493150684931505, + "eval_steps": 10, + "global_step": 2600, + "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": 3.0476, + "step": 1 + }, + { + "epoch": 0.002191780821917808, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.0289, + "step": 2 + }, + { + "epoch": 0.0032876712328767125, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 2.9935, + "step": 3 + }, + { + "epoch": 0.004383561643835616, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.0139, + "step": 4 + }, + { + "epoch": 0.005479452054794521, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 2.9892, + "step": 5 + }, + { + "epoch": 0.006575342465753425, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.1252, + "step": 6 + }, + { + "epoch": 0.007671232876712329, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.0291, + "step": 7 + }, + { + "epoch": 0.008767123287671232, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.0607, + "step": 8 + }, + { + "epoch": 0.009863013698630137, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.0306, + "step": 9 + }, + { + "epoch": 0.010958904109589041, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.0811, + "step": 10 + }, + { + "epoch": 0.010958904109589041, + "eval_loss": 3.0768930912017822, + "eval_runtime": 38.0377, + "eval_samples_per_second": 26.29, + "eval_steps_per_second": 1.104, + "step": 10 + }, + { + "epoch": 0.012054794520547946, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.0693, + "step": 11 + }, + { + "epoch": 0.01315068493150685, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 2.9668, + "step": 12 + }, + { + "epoch": 0.014246575342465753, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.052, + "step": 13 + }, + { + "epoch": 0.015342465753424657, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.0363, + "step": 14 + }, + { + "epoch": 0.01643835616438356, + "grad_norm": 5.58175802230835, + "learning_rate": 1.8248175182481753e-08, + "loss": 3.0778, + "step": 15 + }, + { + "epoch": 0.017534246575342465, + "grad_norm": 6.360531806945801, + "learning_rate": 3.6496350364963505e-08, + "loss": 3.0257, + "step": 16 + }, + { + "epoch": 0.01863013698630137, + "grad_norm": 6.453113555908203, + "learning_rate": 5.4744525547445264e-08, + "loss": 3.0262, + "step": 17 + }, + { + "epoch": 0.019726027397260273, + "grad_norm": 6.789405345916748, + "learning_rate": 7.299270072992701e-08, + "loss": 3.0671, + "step": 18 + }, + { + "epoch": 0.020821917808219178, + "grad_norm": 6.185514450073242, + "learning_rate": 9.124087591240876e-08, + "loss": 3.0077, + "step": 19 + }, + { + "epoch": 0.021917808219178082, + "grad_norm": 5.719798564910889, + "learning_rate": 1.0948905109489053e-07, + "loss": 3.089, + "step": 20 + }, + { + "epoch": 0.021917808219178082, + "eval_loss": 3.07546329498291, + "eval_runtime": 37.5318, + "eval_samples_per_second": 26.644, + "eval_steps_per_second": 1.119, + "step": 20 + }, + { + "epoch": 0.023013698630136987, + "grad_norm": 6.307768821716309, + "learning_rate": 1.2773722627737227e-07, + "loss": 3.0641, + "step": 21 + }, + { + "epoch": 0.02410958904109589, + "grad_norm": 6.255112171173096, + "learning_rate": 1.4598540145985402e-07, + "loss": 2.9849, + "step": 22 + }, + { + "epoch": 0.025205479452054796, + "grad_norm": 5.812101364135742, + "learning_rate": 1.642335766423358e-07, + "loss": 3.0211, + "step": 23 + }, + { + "epoch": 0.0263013698630137, + "grad_norm": 6.262075901031494, + "learning_rate": 1.8248175182481753e-07, + "loss": 3.0654, + "step": 24 + }, + { + "epoch": 0.0273972602739726, + "grad_norm": 5.419229507446289, + "learning_rate": 2.0072992700729928e-07, + "loss": 3.0661, + "step": 25 + }, + { + "epoch": 0.028493150684931506, + "grad_norm": 4.750737190246582, + "learning_rate": 2.1897810218978106e-07, + "loss": 3.0494, + "step": 26 + }, + { + "epoch": 0.02958904109589041, + "grad_norm": 4.732361316680908, + "learning_rate": 2.3722627737226278e-07, + "loss": 3.0471, + "step": 27 + }, + { + "epoch": 0.030684931506849315, + "grad_norm": 4.749786853790283, + "learning_rate": 2.5547445255474454e-07, + "loss": 3.0107, + "step": 28 + }, + { + "epoch": 0.03178082191780822, + "grad_norm": 4.451244831085205, + "learning_rate": 2.737226277372263e-07, + "loss": 2.9987, + "step": 29 + }, + { + "epoch": 0.03287671232876712, + "grad_norm": 4.5158257484436035, + "learning_rate": 2.9197080291970804e-07, + "loss": 2.9686, + "step": 30 + }, + { + "epoch": 0.03287671232876712, + "eval_loss": 3.031872034072876, + "eval_runtime": 37.2608, + "eval_samples_per_second": 26.838, + "eval_steps_per_second": 1.127, + "step": 30 + }, + { + "epoch": 0.03397260273972603, + "grad_norm": 3.8239498138427734, + "learning_rate": 3.102189781021898e-07, + "loss": 3.0057, + "step": 31 + }, + { + "epoch": 0.03506849315068493, + "grad_norm": 4.037254810333252, + "learning_rate": 3.284671532846716e-07, + "loss": 2.9733, + "step": 32 + }, + { + "epoch": 0.03616438356164384, + "grad_norm": 4.283931732177734, + "learning_rate": 3.467153284671533e-07, + "loss": 2.9515, + "step": 33 + }, + { + "epoch": 0.03726027397260274, + "grad_norm": 4.1784987449646, + "learning_rate": 3.6496350364963505e-07, + "loss": 2.9483, + "step": 34 + }, + { + "epoch": 0.038356164383561646, + "grad_norm": 3.9210753440856934, + "learning_rate": 3.8321167883211686e-07, + "loss": 2.9679, + "step": 35 + }, + { + "epoch": 0.03945205479452055, + "grad_norm": 3.72884202003479, + "learning_rate": 4.0145985401459856e-07, + "loss": 2.9989, + "step": 36 + }, + { + "epoch": 0.040547945205479455, + "grad_norm": 3.2882330417633057, + "learning_rate": 4.197080291970803e-07, + "loss": 2.9384, + "step": 37 + }, + { + "epoch": 0.041643835616438356, + "grad_norm": 3.515293836593628, + "learning_rate": 4.379562043795621e-07, + "loss": 2.8986, + "step": 38 + }, + { + "epoch": 0.042739726027397264, + "grad_norm": 3.568453073501587, + "learning_rate": 4.562043795620438e-07, + "loss": 2.9854, + "step": 39 + }, + { + "epoch": 0.043835616438356165, + "grad_norm": 3.389789342880249, + "learning_rate": 4.7445255474452557e-07, + "loss": 2.8891, + "step": 40 + }, + { + "epoch": 0.043835616438356165, + "eval_loss": 2.964235544204712, + "eval_runtime": 37.3406, + "eval_samples_per_second": 26.781, + "eval_steps_per_second": 1.125, + "step": 40 + }, + { + "epoch": 0.044931506849315066, + "grad_norm": 3.1826283931732178, + "learning_rate": 4.927007299270073e-07, + "loss": 2.9271, + "step": 41 + }, + { + "epoch": 0.046027397260273974, + "grad_norm": 2.941539764404297, + "learning_rate": 5.109489051094891e-07, + "loss": 2.9486, + "step": 42 + }, + { + "epoch": 0.047123287671232875, + "grad_norm": 3.0076773166656494, + "learning_rate": 5.291970802919708e-07, + "loss": 2.9252, + "step": 43 + }, + { + "epoch": 0.04821917808219178, + "grad_norm": 2.79028582572937, + "learning_rate": 5.474452554744526e-07, + "loss": 2.9277, + "step": 44 + }, + { + "epoch": 0.049315068493150684, + "grad_norm": 2.603456735610962, + "learning_rate": 5.656934306569343e-07, + "loss": 2.9144, + "step": 45 + }, + { + "epoch": 0.05041095890410959, + "grad_norm": 2.674888849258423, + "learning_rate": 5.839416058394161e-07, + "loss": 2.8732, + "step": 46 + }, + { + "epoch": 0.05150684931506849, + "grad_norm": 2.972093343734741, + "learning_rate": 6.021897810218979e-07, + "loss": 2.8623, + "step": 47 + }, + { + "epoch": 0.0526027397260274, + "grad_norm": 2.757766008377075, + "learning_rate": 6.204379562043796e-07, + "loss": 2.8941, + "step": 48 + }, + { + "epoch": 0.0536986301369863, + "grad_norm": 2.910634994506836, + "learning_rate": 6.386861313868614e-07, + "loss": 2.8101, + "step": 49 + }, + { + "epoch": 0.0547945205479452, + "grad_norm": 2.5542638301849365, + "learning_rate": 6.569343065693432e-07, + "loss": 2.9262, + "step": 50 + }, + { + "epoch": 0.0547945205479452, + "eval_loss": 2.883579969406128, + "eval_runtime": 37.326, + "eval_samples_per_second": 26.791, + "eval_steps_per_second": 1.125, + "step": 50 + }, + { + "epoch": 0.05589041095890411, + "grad_norm": 2.333974599838257, + "learning_rate": 6.751824817518248e-07, + "loss": 2.8217, + "step": 51 + }, + { + "epoch": 0.05698630136986301, + "grad_norm": 2.181022882461548, + "learning_rate": 6.934306569343066e-07, + "loss": 2.8483, + "step": 52 + }, + { + "epoch": 0.05808219178082192, + "grad_norm": 2.12200927734375, + "learning_rate": 7.116788321167883e-07, + "loss": 2.7788, + "step": 53 + }, + { + "epoch": 0.05917808219178082, + "grad_norm": 2.052626371383667, + "learning_rate": 7.299270072992701e-07, + "loss": 2.7774, + "step": 54 + }, + { + "epoch": 0.06027397260273973, + "grad_norm": 2.11487078666687, + "learning_rate": 7.48175182481752e-07, + "loss": 2.8045, + "step": 55 + }, + { + "epoch": 0.06136986301369863, + "grad_norm": 2.083588123321533, + "learning_rate": 7.664233576642337e-07, + "loss": 2.8119, + "step": 56 + }, + { + "epoch": 0.06246575342465754, + "grad_norm": 2.2079808712005615, + "learning_rate": 7.846715328467154e-07, + "loss": 2.82, + "step": 57 + }, + { + "epoch": 0.06356164383561644, + "grad_norm": 2.100802421569824, + "learning_rate": 8.029197080291971e-07, + "loss": 2.794, + "step": 58 + }, + { + "epoch": 0.06465753424657535, + "grad_norm": 1.8322066068649292, + "learning_rate": 8.211678832116789e-07, + "loss": 2.7444, + "step": 59 + }, + { + "epoch": 0.06575342465753424, + "grad_norm": 2.077005386352539, + "learning_rate": 8.394160583941606e-07, + "loss": 2.7093, + "step": 60 + }, + { + "epoch": 0.06575342465753424, + "eval_loss": 2.8177874088287354, + "eval_runtime": 37.2903, + "eval_samples_per_second": 26.817, + "eval_steps_per_second": 1.126, + "step": 60 + }, + { + "epoch": 0.06684931506849315, + "grad_norm": 2.1688592433929443, + "learning_rate": 8.576642335766425e-07, + "loss": 2.7446, + "step": 61 + }, + { + "epoch": 0.06794520547945206, + "grad_norm": 2.1709237098693848, + "learning_rate": 8.759124087591242e-07, + "loss": 2.736, + "step": 62 + }, + { + "epoch": 0.06904109589041096, + "grad_norm": 2.3578364849090576, + "learning_rate": 8.94160583941606e-07, + "loss": 2.7583, + "step": 63 + }, + { + "epoch": 0.07013698630136986, + "grad_norm": 1.9025483131408691, + "learning_rate": 9.124087591240876e-07, + "loss": 2.7651, + "step": 64 + }, + { + "epoch": 0.07123287671232877, + "grad_norm": 1.9450963735580444, + "learning_rate": 9.306569343065694e-07, + "loss": 2.7186, + "step": 65 + }, + { + "epoch": 0.07232876712328767, + "grad_norm": 1.957723617553711, + "learning_rate": 9.489051094890511e-07, + "loss": 2.7421, + "step": 66 + }, + { + "epoch": 0.07342465753424658, + "grad_norm": 1.9823631048202515, + "learning_rate": 9.67153284671533e-07, + "loss": 2.7258, + "step": 67 + }, + { + "epoch": 0.07452054794520548, + "grad_norm": 1.945495843887329, + "learning_rate": 9.854014598540146e-07, + "loss": 2.7733, + "step": 68 + }, + { + "epoch": 0.07561643835616438, + "grad_norm": 1.9684759378433228, + "learning_rate": 1.0036496350364965e-06, + "loss": 2.7041, + "step": 69 + }, + { + "epoch": 0.07671232876712329, + "grad_norm": 2.087336540222168, + "learning_rate": 1.0218978102189781e-06, + "loss": 2.6662, + "step": 70 + }, + { + "epoch": 0.07671232876712329, + "eval_loss": 2.7529091835021973, + "eval_runtime": 37.3122, + "eval_samples_per_second": 26.801, + "eval_steps_per_second": 1.126, + "step": 70 + }, + { + "epoch": 0.07780821917808219, + "grad_norm": 1.7790117263793945, + "learning_rate": 1.04014598540146e-06, + "loss": 2.7379, + "step": 71 + }, + { + "epoch": 0.0789041095890411, + "grad_norm": 2.2780861854553223, + "learning_rate": 1.0583941605839416e-06, + "loss": 2.6234, + "step": 72 + }, + { + "epoch": 0.08, + "grad_norm": 2.1144700050354004, + "learning_rate": 1.0766423357664233e-06, + "loss": 2.6248, + "step": 73 + }, + { + "epoch": 0.08109589041095891, + "grad_norm": 2.0247490406036377, + "learning_rate": 1.0948905109489052e-06, + "loss": 2.6456, + "step": 74 + }, + { + "epoch": 0.0821917808219178, + "grad_norm": 2.702547311782837, + "learning_rate": 1.113138686131387e-06, + "loss": 2.6759, + "step": 75 + }, + { + "epoch": 0.08328767123287671, + "grad_norm": 1.8208750486373901, + "learning_rate": 1.1313868613138687e-06, + "loss": 2.6845, + "step": 76 + }, + { + "epoch": 0.08438356164383562, + "grad_norm": 2.3829221725463867, + "learning_rate": 1.1496350364963505e-06, + "loss": 2.61, + "step": 77 + }, + { + "epoch": 0.08547945205479453, + "grad_norm": 2.580885887145996, + "learning_rate": 1.1678832116788322e-06, + "loss": 2.5896, + "step": 78 + }, + { + "epoch": 0.08657534246575342, + "grad_norm": 2.2477641105651855, + "learning_rate": 1.1861313868613138e-06, + "loss": 2.596, + "step": 79 + }, + { + "epoch": 0.08767123287671233, + "grad_norm": 2.1895623207092285, + "learning_rate": 1.2043795620437959e-06, + "loss": 2.5906, + "step": 80 + }, + { + "epoch": 0.08767123287671233, + "eval_loss": 2.6815176010131836, + "eval_runtime": 37.3342, + "eval_samples_per_second": 26.785, + "eval_steps_per_second": 1.125, + "step": 80 + }, + { + "epoch": 0.08876712328767124, + "grad_norm": 2.217090606689453, + "learning_rate": 1.2226277372262775e-06, + "loss": 2.6005, + "step": 81 + }, + { + "epoch": 0.08986301369863013, + "grad_norm": 2.4679362773895264, + "learning_rate": 1.2408759124087592e-06, + "loss": 2.6333, + "step": 82 + }, + { + "epoch": 0.09095890410958904, + "grad_norm": 2.2590575218200684, + "learning_rate": 1.259124087591241e-06, + "loss": 2.5814, + "step": 83 + }, + { + "epoch": 0.09205479452054795, + "grad_norm": 1.877977728843689, + "learning_rate": 1.2773722627737229e-06, + "loss": 2.5948, + "step": 84 + }, + { + "epoch": 0.09315068493150686, + "grad_norm": 2.1535534858703613, + "learning_rate": 1.2956204379562045e-06, + "loss": 2.5781, + "step": 85 + }, + { + "epoch": 0.09424657534246575, + "grad_norm": 2.5001144409179688, + "learning_rate": 1.3138686131386864e-06, + "loss": 2.5339, + "step": 86 + }, + { + "epoch": 0.09534246575342466, + "grad_norm": 2.3279049396514893, + "learning_rate": 1.332116788321168e-06, + "loss": 2.545, + "step": 87 + }, + { + "epoch": 0.09643835616438357, + "grad_norm": 2.2230734825134277, + "learning_rate": 1.3503649635036497e-06, + "loss": 2.4745, + "step": 88 + }, + { + "epoch": 0.09753424657534246, + "grad_norm": 2.8762636184692383, + "learning_rate": 1.3686131386861315e-06, + "loss": 2.516, + "step": 89 + }, + { + "epoch": 0.09863013698630137, + "grad_norm": 2.153506278991699, + "learning_rate": 1.3868613138686132e-06, + "loss": 2.4547, + "step": 90 + }, + { + "epoch": 0.09863013698630137, + "eval_loss": 2.603238344192505, + "eval_runtime": 37.4162, + "eval_samples_per_second": 26.726, + "eval_steps_per_second": 1.123, + "step": 90 + }, + { + "epoch": 0.09972602739726028, + "grad_norm": 2.4143614768981934, + "learning_rate": 1.405109489051095e-06, + "loss": 2.547, + "step": 91 + }, + { + "epoch": 0.10082191780821918, + "grad_norm": 2.354632616043091, + "learning_rate": 1.4233576642335767e-06, + "loss": 2.5009, + "step": 92 + }, + { + "epoch": 0.10191780821917808, + "grad_norm": 2.6605563163757324, + "learning_rate": 1.4416058394160583e-06, + "loss": 2.5304, + "step": 93 + }, + { + "epoch": 0.10301369863013699, + "grad_norm": 2.3724122047424316, + "learning_rate": 1.4598540145985402e-06, + "loss": 2.4925, + "step": 94 + }, + { + "epoch": 0.10410958904109589, + "grad_norm": 3.6323022842407227, + "learning_rate": 1.478102189781022e-06, + "loss": 2.4592, + "step": 95 + }, + { + "epoch": 0.1052054794520548, + "grad_norm": 2.3407089710235596, + "learning_rate": 1.496350364963504e-06, + "loss": 2.5223, + "step": 96 + }, + { + "epoch": 0.1063013698630137, + "grad_norm": 2.8238413333892822, + "learning_rate": 1.5145985401459856e-06, + "loss": 2.489, + "step": 97 + }, + { + "epoch": 0.1073972602739726, + "grad_norm": 2.6433658599853516, + "learning_rate": 1.5328467153284674e-06, + "loss": 2.5101, + "step": 98 + }, + { + "epoch": 0.10849315068493151, + "grad_norm": 2.562761068344116, + "learning_rate": 1.551094890510949e-06, + "loss": 2.4642, + "step": 99 + }, + { + "epoch": 0.1095890410958904, + "grad_norm": 2.4501149654388428, + "learning_rate": 1.5693430656934307e-06, + "loss": 2.4102, + "step": 100 + }, + { + "epoch": 0.1095890410958904, + "eval_loss": 2.5452804565429688, + "eval_runtime": 37.4224, + "eval_samples_per_second": 26.722, + "eval_steps_per_second": 1.122, + "step": 100 + }, + { + "epoch": 0.11068493150684931, + "grad_norm": 2.475864887237549, + "learning_rate": 1.5875912408759126e-06, + "loss": 2.4147, + "step": 101 + }, + { + "epoch": 0.11178082191780822, + "grad_norm": 2.558875322341919, + "learning_rate": 1.6058394160583942e-06, + "loss": 2.4694, + "step": 102 + }, + { + "epoch": 0.11287671232876713, + "grad_norm": 2.1085734367370605, + "learning_rate": 1.624087591240876e-06, + "loss": 2.5099, + "step": 103 + }, + { + "epoch": 0.11397260273972602, + "grad_norm": 2.410731792449951, + "learning_rate": 1.6423357664233577e-06, + "loss": 2.432, + "step": 104 + }, + { + "epoch": 0.11506849315068493, + "grad_norm": 2.2800745964050293, + "learning_rate": 1.6605839416058394e-06, + "loss": 2.3669, + "step": 105 + }, + { + "epoch": 0.11616438356164384, + "grad_norm": 2.2234933376312256, + "learning_rate": 1.6788321167883212e-06, + "loss": 2.4145, + "step": 106 + }, + { + "epoch": 0.11726027397260275, + "grad_norm": 2.3367040157318115, + "learning_rate": 1.6970802919708029e-06, + "loss": 2.4272, + "step": 107 + }, + { + "epoch": 0.11835616438356164, + "grad_norm": 2.7002811431884766, + "learning_rate": 1.715328467153285e-06, + "loss": 2.3478, + "step": 108 + }, + { + "epoch": 0.11945205479452055, + "grad_norm": 2.6628332138061523, + "learning_rate": 1.7335766423357666e-06, + "loss": 2.4096, + "step": 109 + }, + { + "epoch": 0.12054794520547946, + "grad_norm": 2.4174230098724365, + "learning_rate": 1.7518248175182485e-06, + "loss": 2.407, + "step": 110 + }, + { + "epoch": 0.12054794520547946, + "eval_loss": 2.491746664047241, + "eval_runtime": 37.2555, + "eval_samples_per_second": 26.842, + "eval_steps_per_second": 1.127, + "step": 110 + }, + { + "epoch": 0.12164383561643835, + "grad_norm": 2.24796986579895, + "learning_rate": 1.77007299270073e-06, + "loss": 2.4369, + "step": 111 + }, + { + "epoch": 0.12273972602739726, + "grad_norm": 2.720939874649048, + "learning_rate": 1.788321167883212e-06, + "loss": 2.4116, + "step": 112 + }, + { + "epoch": 0.12383561643835617, + "grad_norm": 2.6907670497894287, + "learning_rate": 1.8065693430656936e-06, + "loss": 2.4577, + "step": 113 + }, + { + "epoch": 0.12493150684931507, + "grad_norm": 2.559636354446411, + "learning_rate": 1.8248175182481753e-06, + "loss": 2.3999, + "step": 114 + }, + { + "epoch": 0.12602739726027398, + "grad_norm": 2.6730551719665527, + "learning_rate": 1.8430656934306571e-06, + "loss": 2.3428, + "step": 115 + }, + { + "epoch": 0.12712328767123288, + "grad_norm": 3.172179937362671, + "learning_rate": 1.8613138686131388e-06, + "loss": 2.3722, + "step": 116 + }, + { + "epoch": 0.12821917808219177, + "grad_norm": 2.6757171154022217, + "learning_rate": 1.8795620437956206e-06, + "loss": 2.3572, + "step": 117 + }, + { + "epoch": 0.1293150684931507, + "grad_norm": 2.5840766429901123, + "learning_rate": 1.8978102189781023e-06, + "loss": 2.353, + "step": 118 + }, + { + "epoch": 0.1304109589041096, + "grad_norm": 2.6248581409454346, + "learning_rate": 1.916058394160584e-06, + "loss": 2.36, + "step": 119 + }, + { + "epoch": 0.13150684931506848, + "grad_norm": 2.428516387939453, + "learning_rate": 1.934306569343066e-06, + "loss": 2.3821, + "step": 120 + }, + { + "epoch": 0.13150684931506848, + "eval_loss": 2.451565980911255, + "eval_runtime": 37.3669, + "eval_samples_per_second": 26.762, + "eval_steps_per_second": 1.124, + "step": 120 + }, + { + "epoch": 0.1326027397260274, + "grad_norm": 2.9956233501434326, + "learning_rate": 1.952554744525548e-06, + "loss": 2.2778, + "step": 121 + }, + { + "epoch": 0.1336986301369863, + "grad_norm": 2.4120893478393555, + "learning_rate": 1.9708029197080293e-06, + "loss": 2.325, + "step": 122 + }, + { + "epoch": 0.13479452054794522, + "grad_norm": 2.1410024166107178, + "learning_rate": 1.989051094890511e-06, + "loss": 2.3248, + "step": 123 + }, + { + "epoch": 0.1358904109589041, + "grad_norm": 2.923799514770508, + "learning_rate": 2.007299270072993e-06, + "loss": 2.327, + "step": 124 + }, + { + "epoch": 0.136986301369863, + "grad_norm": 3.0770339965820312, + "learning_rate": 2.025547445255475e-06, + "loss": 2.3847, + "step": 125 + }, + { + "epoch": 0.13808219178082193, + "grad_norm": 2.6408181190490723, + "learning_rate": 2.0437956204379563e-06, + "loss": 2.3196, + "step": 126 + }, + { + "epoch": 0.13917808219178082, + "grad_norm": 2.9127118587493896, + "learning_rate": 2.062043795620438e-06, + "loss": 2.3287, + "step": 127 + }, + { + "epoch": 0.14027397260273972, + "grad_norm": 2.9805052280426025, + "learning_rate": 2.08029197080292e-06, + "loss": 2.3243, + "step": 128 + }, + { + "epoch": 0.14136986301369864, + "grad_norm": 3.1342010498046875, + "learning_rate": 2.0985401459854014e-06, + "loss": 2.38, + "step": 129 + }, + { + "epoch": 0.14246575342465753, + "grad_norm": 2.678422451019287, + "learning_rate": 2.1167883211678833e-06, + "loss": 2.2727, + "step": 130 + }, + { + "epoch": 0.14246575342465753, + "eval_loss": 2.4187021255493164, + "eval_runtime": 37.3117, + "eval_samples_per_second": 26.801, + "eval_steps_per_second": 1.126, + "step": 130 + }, + { + "epoch": 0.14356164383561643, + "grad_norm": 2.9281492233276367, + "learning_rate": 2.135036496350365e-06, + "loss": 2.3327, + "step": 131 + }, + { + "epoch": 0.14465753424657535, + "grad_norm": 2.9592933654785156, + "learning_rate": 2.1532846715328466e-06, + "loss": 2.3674, + "step": 132 + }, + { + "epoch": 0.14575342465753424, + "grad_norm": 3.404761552810669, + "learning_rate": 2.171532846715329e-06, + "loss": 2.2679, + "step": 133 + }, + { + "epoch": 0.14684931506849316, + "grad_norm": 3.635571241378784, + "learning_rate": 2.1897810218978103e-06, + "loss": 2.2799, + "step": 134 + }, + { + "epoch": 0.14794520547945206, + "grad_norm": 3.9130170345306396, + "learning_rate": 2.208029197080292e-06, + "loss": 2.3263, + "step": 135 + }, + { + "epoch": 0.14904109589041095, + "grad_norm": 4.300438404083252, + "learning_rate": 2.226277372262774e-06, + "loss": 2.2969, + "step": 136 + }, + { + "epoch": 0.15013698630136987, + "grad_norm": 2.863511085510254, + "learning_rate": 2.244525547445256e-06, + "loss": 2.297, + "step": 137 + }, + { + "epoch": 0.15123287671232877, + "grad_norm": 4.910321235656738, + "learning_rate": 2.2627737226277373e-06, + "loss": 2.3269, + "step": 138 + }, + { + "epoch": 0.15232876712328766, + "grad_norm": 3.3323862552642822, + "learning_rate": 2.281021897810219e-06, + "loss": 2.3195, + "step": 139 + }, + { + "epoch": 0.15342465753424658, + "grad_norm": 3.2672693729400635, + "learning_rate": 2.299270072992701e-06, + "loss": 2.2797, + "step": 140 + }, + { + "epoch": 0.15342465753424658, + "eval_loss": 2.3943886756896973, + "eval_runtime": 37.3137, + "eval_samples_per_second": 26.8, + "eval_steps_per_second": 1.126, + "step": 140 + }, + { + "epoch": 0.15452054794520548, + "grad_norm": 3.9229774475097656, + "learning_rate": 2.3175182481751825e-06, + "loss": 2.2718, + "step": 141 + }, + { + "epoch": 0.15561643835616437, + "grad_norm": 2.6409823894500732, + "learning_rate": 2.3357664233576643e-06, + "loss": 2.2828, + "step": 142 + }, + { + "epoch": 0.1567123287671233, + "grad_norm": 4.6122612953186035, + "learning_rate": 2.354014598540146e-06, + "loss": 2.247, + "step": 143 + }, + { + "epoch": 0.1578082191780822, + "grad_norm": 3.391357898712158, + "learning_rate": 2.3722627737226276e-06, + "loss": 2.2863, + "step": 144 + }, + { + "epoch": 0.1589041095890411, + "grad_norm": 3.3550853729248047, + "learning_rate": 2.39051094890511e-06, + "loss": 2.2789, + "step": 145 + }, + { + "epoch": 0.16, + "grad_norm": 3.7376883029937744, + "learning_rate": 2.4087591240875918e-06, + "loss": 2.2297, + "step": 146 + }, + { + "epoch": 0.1610958904109589, + "grad_norm": 3.005157709121704, + "learning_rate": 2.427007299270073e-06, + "loss": 2.2467, + "step": 147 + }, + { + "epoch": 0.16219178082191782, + "grad_norm": 3.4616103172302246, + "learning_rate": 2.445255474452555e-06, + "loss": 2.2612, + "step": 148 + }, + { + "epoch": 0.1632876712328767, + "grad_norm": 3.3595101833343506, + "learning_rate": 2.463503649635037e-06, + "loss": 2.3021, + "step": 149 + }, + { + "epoch": 0.1643835616438356, + "grad_norm": 4.831241607666016, + "learning_rate": 2.4817518248175183e-06, + "loss": 2.2174, + "step": 150 + }, + { + "epoch": 0.1643835616438356, + "eval_loss": 2.3677749633789062, + "eval_runtime": 37.3092, + "eval_samples_per_second": 26.803, + "eval_steps_per_second": 1.126, + "step": 150 + }, + { + "epoch": 0.16547945205479453, + "grad_norm": 2.778820037841797, + "learning_rate": 2.5e-06, + "loss": 2.1856, + "step": 151 + }, + { + "epoch": 0.16657534246575342, + "grad_norm": 4.566643714904785, + "learning_rate": 2.518248175182482e-06, + "loss": 2.3186, + "step": 152 + }, + { + "epoch": 0.16767123287671232, + "grad_norm": 2.9908089637756348, + "learning_rate": 2.5364963503649635e-06, + "loss": 2.2127, + "step": 153 + }, + { + "epoch": 0.16876712328767124, + "grad_norm": 2.9702792167663574, + "learning_rate": 2.5547445255474458e-06, + "loss": 2.2071, + "step": 154 + }, + { + "epoch": 0.16986301369863013, + "grad_norm": 2.850044012069702, + "learning_rate": 2.5729927007299272e-06, + "loss": 2.2585, + "step": 155 + }, + { + "epoch": 0.17095890410958905, + "grad_norm": 2.926091432571411, + "learning_rate": 2.591240875912409e-06, + "loss": 2.2397, + "step": 156 + }, + { + "epoch": 0.17205479452054795, + "grad_norm": 2.8720169067382812, + "learning_rate": 2.6094890510948905e-06, + "loss": 2.2306, + "step": 157 + }, + { + "epoch": 0.17315068493150684, + "grad_norm": 2.8837177753448486, + "learning_rate": 2.627737226277373e-06, + "loss": 2.2666, + "step": 158 + }, + { + "epoch": 0.17424657534246576, + "grad_norm": 3.119621753692627, + "learning_rate": 2.6459854014598542e-06, + "loss": 2.2722, + "step": 159 + }, + { + "epoch": 0.17534246575342466, + "grad_norm": 2.7085700035095215, + "learning_rate": 2.664233576642336e-06, + "loss": 2.2346, + "step": 160 + }, + { + "epoch": 0.17534246575342466, + "eval_loss": 2.352083921432495, + "eval_runtime": 37.245, + "eval_samples_per_second": 26.849, + "eval_steps_per_second": 1.128, + "step": 160 + }, + { + "epoch": 0.17643835616438355, + "grad_norm": 2.9964346885681152, + "learning_rate": 2.6824817518248175e-06, + "loss": 2.2155, + "step": 161 + }, + { + "epoch": 0.17753424657534247, + "grad_norm": 3.1535208225250244, + "learning_rate": 2.7007299270072994e-06, + "loss": 2.235, + "step": 162 + }, + { + "epoch": 0.17863013698630137, + "grad_norm": 2.7240943908691406, + "learning_rate": 2.718978102189781e-06, + "loss": 2.2095, + "step": 163 + }, + { + "epoch": 0.17972602739726026, + "grad_norm": 3.122654676437378, + "learning_rate": 2.737226277372263e-06, + "loss": 2.2667, + "step": 164 + }, + { + "epoch": 0.18082191780821918, + "grad_norm": 2.9718410968780518, + "learning_rate": 2.755474452554745e-06, + "loss": 2.2499, + "step": 165 + }, + { + "epoch": 0.18191780821917808, + "grad_norm": 3.0081896781921387, + "learning_rate": 2.7737226277372264e-06, + "loss": 2.2089, + "step": 166 + }, + { + "epoch": 0.18301369863013697, + "grad_norm": 2.829509973526001, + "learning_rate": 2.7919708029197082e-06, + "loss": 2.3139, + "step": 167 + }, + { + "epoch": 0.1841095890410959, + "grad_norm": 3.4480397701263428, + "learning_rate": 2.81021897810219e-06, + "loss": 2.2705, + "step": 168 + }, + { + "epoch": 0.1852054794520548, + "grad_norm": 3.1107025146484375, + "learning_rate": 2.828467153284672e-06, + "loss": 2.1895, + "step": 169 + }, + { + "epoch": 0.1863013698630137, + "grad_norm": 2.916790008544922, + "learning_rate": 2.8467153284671534e-06, + "loss": 2.2317, + "step": 170 + }, + { + "epoch": 0.1863013698630137, + "eval_loss": 2.336153030395508, + "eval_runtime": 37.2401, + "eval_samples_per_second": 26.853, + "eval_steps_per_second": 1.128, + "step": 170 + }, + { + "epoch": 0.1873972602739726, + "grad_norm": 3.2582287788391113, + "learning_rate": 2.8649635036496353e-06, + "loss": 2.143, + "step": 171 + }, + { + "epoch": 0.1884931506849315, + "grad_norm": 2.430973768234253, + "learning_rate": 2.8832116788321167e-06, + "loss": 2.1977, + "step": 172 + }, + { + "epoch": 0.18958904109589042, + "grad_norm": 2.8763599395751953, + "learning_rate": 2.901459854014599e-06, + "loss": 2.1551, + "step": 173 + }, + { + "epoch": 0.19068493150684931, + "grad_norm": 2.9242653846740723, + "learning_rate": 2.9197080291970804e-06, + "loss": 2.2569, + "step": 174 + }, + { + "epoch": 0.1917808219178082, + "grad_norm": 2.623551607131958, + "learning_rate": 2.9379562043795623e-06, + "loss": 2.2329, + "step": 175 + }, + { + "epoch": 0.19287671232876713, + "grad_norm": 2.9511337280273438, + "learning_rate": 2.956204379562044e-06, + "loss": 2.2154, + "step": 176 + }, + { + "epoch": 0.19397260273972602, + "grad_norm": 2.415118932723999, + "learning_rate": 2.9744525547445256e-06, + "loss": 2.2637, + "step": 177 + }, + { + "epoch": 0.19506849315068492, + "grad_norm": 2.9664924144744873, + "learning_rate": 2.992700729927008e-06, + "loss": 2.2263, + "step": 178 + }, + { + "epoch": 0.19616438356164384, + "grad_norm": 2.7214572429656982, + "learning_rate": 3.0109489051094893e-06, + "loss": 2.1667, + "step": 179 + }, + { + "epoch": 0.19726027397260273, + "grad_norm": 2.8712823390960693, + "learning_rate": 3.029197080291971e-06, + "loss": 2.1661, + "step": 180 + }, + { + "epoch": 0.19726027397260273, + "eval_loss": 2.3191800117492676, + "eval_runtime": 37.24, + "eval_samples_per_second": 26.853, + "eval_steps_per_second": 1.128, + "step": 180 + }, + { + "epoch": 0.19835616438356166, + "grad_norm": 2.959709405899048, + "learning_rate": 3.0474452554744526e-06, + "loss": 2.2483, + "step": 181 + }, + { + "epoch": 0.19945205479452055, + "grad_norm": 2.567089319229126, + "learning_rate": 3.065693430656935e-06, + "loss": 2.2382, + "step": 182 + }, + { + "epoch": 0.20054794520547944, + "grad_norm": 2.47188401222229, + "learning_rate": 3.0839416058394163e-06, + "loss": 2.2148, + "step": 183 + }, + { + "epoch": 0.20164383561643837, + "grad_norm": 2.3435261249542236, + "learning_rate": 3.102189781021898e-06, + "loss": 2.2334, + "step": 184 + }, + { + "epoch": 0.20273972602739726, + "grad_norm": 2.5280468463897705, + "learning_rate": 3.1204379562043796e-06, + "loss": 2.2214, + "step": 185 + }, + { + "epoch": 0.20383561643835615, + "grad_norm": 2.4808971881866455, + "learning_rate": 3.1386861313868614e-06, + "loss": 2.1554, + "step": 186 + }, + { + "epoch": 0.20493150684931508, + "grad_norm": 2.9937026500701904, + "learning_rate": 3.156934306569343e-06, + "loss": 2.1389, + "step": 187 + }, + { + "epoch": 0.20602739726027397, + "grad_norm": 2.9220476150512695, + "learning_rate": 3.175182481751825e-06, + "loss": 2.2515, + "step": 188 + }, + { + "epoch": 0.20712328767123286, + "grad_norm": 3.0449001789093018, + "learning_rate": 3.193430656934307e-06, + "loss": 2.2016, + "step": 189 + }, + { + "epoch": 0.20821917808219179, + "grad_norm": 2.8027074337005615, + "learning_rate": 3.2116788321167884e-06, + "loss": 2.1823, + "step": 190 + }, + { + "epoch": 0.20821917808219179, + "eval_loss": 2.309677839279175, + "eval_runtime": 37.2442, + "eval_samples_per_second": 26.85, + "eval_steps_per_second": 1.128, + "step": 190 + }, + { + "epoch": 0.20931506849315068, + "grad_norm": 2.9158318042755127, + "learning_rate": 3.2299270072992707e-06, + "loss": 2.1928, + "step": 191 + }, + { + "epoch": 0.2104109589041096, + "grad_norm": 2.5328080654144287, + "learning_rate": 3.248175182481752e-06, + "loss": 2.1828, + "step": 192 + }, + { + "epoch": 0.2115068493150685, + "grad_norm": 3.1709704399108887, + "learning_rate": 3.266423357664234e-06, + "loss": 2.1052, + "step": 193 + }, + { + "epoch": 0.2126027397260274, + "grad_norm": 2.9156367778778076, + "learning_rate": 3.2846715328467155e-06, + "loss": 2.1943, + "step": 194 + }, + { + "epoch": 0.2136986301369863, + "grad_norm": 2.879408836364746, + "learning_rate": 3.3029197080291973e-06, + "loss": 2.1782, + "step": 195 + }, + { + "epoch": 0.2147945205479452, + "grad_norm": 3.748483180999756, + "learning_rate": 3.3211678832116788e-06, + "loss": 2.2107, + "step": 196 + }, + { + "epoch": 0.2158904109589041, + "grad_norm": 3.197510004043579, + "learning_rate": 3.339416058394161e-06, + "loss": 2.1845, + "step": 197 + }, + { + "epoch": 0.21698630136986302, + "grad_norm": 3.2596049308776855, + "learning_rate": 3.3576642335766425e-06, + "loss": 2.1816, + "step": 198 + }, + { + "epoch": 0.21808219178082192, + "grad_norm": 3.3329780101776123, + "learning_rate": 3.3759124087591243e-06, + "loss": 2.181, + "step": 199 + }, + { + "epoch": 0.2191780821917808, + "grad_norm": 2.7302327156066895, + "learning_rate": 3.3941605839416058e-06, + "loss": 2.1761, + "step": 200 + }, + { + "epoch": 0.2191780821917808, + "eval_loss": 2.2909653186798096, + "eval_runtime": 37.284, + "eval_samples_per_second": 26.821, + "eval_steps_per_second": 1.126, + "step": 200 + }, + { + "epoch": 0.22027397260273973, + "grad_norm": 3.002255916595459, + "learning_rate": 3.412408759124088e-06, + "loss": 2.1868, + "step": 201 + }, + { + "epoch": 0.22136986301369863, + "grad_norm": 2.6694891452789307, + "learning_rate": 3.43065693430657e-06, + "loss": 2.192, + "step": 202 + }, + { + "epoch": 0.22246575342465755, + "grad_norm": 2.9225986003875732, + "learning_rate": 3.4489051094890513e-06, + "loss": 2.1303, + "step": 203 + }, + { + "epoch": 0.22356164383561644, + "grad_norm": 2.7844080924987793, + "learning_rate": 3.467153284671533e-06, + "loss": 2.1788, + "step": 204 + }, + { + "epoch": 0.22465753424657534, + "grad_norm": 2.726123332977295, + "learning_rate": 3.4854014598540146e-06, + "loss": 2.1193, + "step": 205 + }, + { + "epoch": 0.22575342465753426, + "grad_norm": 2.838040590286255, + "learning_rate": 3.503649635036497e-06, + "loss": 2.1674, + "step": 206 + }, + { + "epoch": 0.22684931506849315, + "grad_norm": 2.6863937377929688, + "learning_rate": 3.5218978102189783e-06, + "loss": 2.161, + "step": 207 + }, + { + "epoch": 0.22794520547945205, + "grad_norm": 2.5116159915924072, + "learning_rate": 3.54014598540146e-06, + "loss": 2.179, + "step": 208 + }, + { + "epoch": 0.22904109589041097, + "grad_norm": 3.381863594055176, + "learning_rate": 3.5583941605839416e-06, + "loss": 2.2035, + "step": 209 + }, + { + "epoch": 0.23013698630136986, + "grad_norm": 3.2741382122039795, + "learning_rate": 3.576642335766424e-06, + "loss": 2.2518, + "step": 210 + }, + { + "epoch": 0.23013698630136986, + "eval_loss": 2.2773635387420654, + "eval_runtime": 37.2658, + "eval_samples_per_second": 26.834, + "eval_steps_per_second": 1.127, + "step": 210 + }, + { + "epoch": 0.23123287671232876, + "grad_norm": 2.917370319366455, + "learning_rate": 3.5948905109489054e-06, + "loss": 2.2265, + "step": 211 + }, + { + "epoch": 0.23232876712328768, + "grad_norm": 2.8392834663391113, + "learning_rate": 3.6131386861313872e-06, + "loss": 2.0709, + "step": 212 + }, + { + "epoch": 0.23342465753424657, + "grad_norm": 2.753612518310547, + "learning_rate": 3.6313868613138687e-06, + "loss": 2.0752, + "step": 213 + }, + { + "epoch": 0.2345205479452055, + "grad_norm": 3.1477746963500977, + "learning_rate": 3.6496350364963505e-06, + "loss": 2.1169, + "step": 214 + }, + { + "epoch": 0.2356164383561644, + "grad_norm": 2.5448100566864014, + "learning_rate": 3.667883211678833e-06, + "loss": 2.1338, + "step": 215 + }, + { + "epoch": 0.23671232876712328, + "grad_norm": 3.1622560024261475, + "learning_rate": 3.6861313868613142e-06, + "loss": 2.1065, + "step": 216 + }, + { + "epoch": 0.2378082191780822, + "grad_norm": 2.7054336071014404, + "learning_rate": 3.704379562043796e-06, + "loss": 2.1104, + "step": 217 + }, + { + "epoch": 0.2389041095890411, + "grad_norm": 2.964141845703125, + "learning_rate": 3.7226277372262775e-06, + "loss": 2.1949, + "step": 218 + }, + { + "epoch": 0.24, + "grad_norm": 2.7586112022399902, + "learning_rate": 3.7408759124087594e-06, + "loss": 2.165, + "step": 219 + }, + { + "epoch": 0.2410958904109589, + "grad_norm": 2.9035964012145996, + "learning_rate": 3.7591240875912412e-06, + "loss": 2.1192, + "step": 220 + }, + { + "epoch": 0.2410958904109589, + "eval_loss": 2.2649471759796143, + "eval_runtime": 37.4311, + "eval_samples_per_second": 26.716, + "eval_steps_per_second": 1.122, + "step": 220 + }, + { + "epoch": 0.2421917808219178, + "grad_norm": 2.4212076663970947, + "learning_rate": 3.777372262773723e-06, + "loss": 2.1389, + "step": 221 + }, + { + "epoch": 0.2432876712328767, + "grad_norm": 2.973227024078369, + "learning_rate": 3.7956204379562045e-06, + "loss": 2.132, + "step": 222 + }, + { + "epoch": 0.24438356164383562, + "grad_norm": 2.9831430912017822, + "learning_rate": 3.8138686131386864e-06, + "loss": 2.1142, + "step": 223 + }, + { + "epoch": 0.24547945205479452, + "grad_norm": 3.2880353927612305, + "learning_rate": 3.832116788321168e-06, + "loss": 2.168, + "step": 224 + }, + { + "epoch": 0.2465753424657534, + "grad_norm": 3.3920998573303223, + "learning_rate": 3.85036496350365e-06, + "loss": 2.1277, + "step": 225 + }, + { + "epoch": 0.24767123287671233, + "grad_norm": 2.7170026302337646, + "learning_rate": 3.868613138686132e-06, + "loss": 2.0936, + "step": 226 + }, + { + "epoch": 0.24876712328767123, + "grad_norm": 3.6220011711120605, + "learning_rate": 3.886861313868613e-06, + "loss": 2.1259, + "step": 227 + }, + { + "epoch": 0.24986301369863015, + "grad_norm": 3.4229273796081543, + "learning_rate": 3.905109489051096e-06, + "loss": 2.1421, + "step": 228 + }, + { + "epoch": 0.25095890410958904, + "grad_norm": 3.0358338356018066, + "learning_rate": 3.923357664233577e-06, + "loss": 2.1747, + "step": 229 + }, + { + "epoch": 0.25205479452054796, + "grad_norm": 2.611771821975708, + "learning_rate": 3.9416058394160585e-06, + "loss": 2.1282, + "step": 230 + }, + { + "epoch": 0.25205479452054796, + "eval_loss": 2.254887580871582, + "eval_runtime": 37.5943, + "eval_samples_per_second": 26.6, + "eval_steps_per_second": 1.117, + "step": 230 + }, + { + "epoch": 0.25315068493150683, + "grad_norm": 3.846412181854248, + "learning_rate": 3.95985401459854e-06, + "loss": 2.1892, + "step": 231 + }, + { + "epoch": 0.25424657534246575, + "grad_norm": 2.488525390625, + "learning_rate": 3.978102189781022e-06, + "loss": 2.2064, + "step": 232 + }, + { + "epoch": 0.2553424657534247, + "grad_norm": 3.114964485168457, + "learning_rate": 3.996350364963504e-06, + "loss": 2.0653, + "step": 233 + }, + { + "epoch": 0.25643835616438354, + "grad_norm": 2.5559768676757812, + "learning_rate": 4.014598540145986e-06, + "loss": 2.1503, + "step": 234 + }, + { + "epoch": 0.25753424657534246, + "grad_norm": 3.2347540855407715, + "learning_rate": 4.032846715328467e-06, + "loss": 2.0523, + "step": 235 + }, + { + "epoch": 0.2586301369863014, + "grad_norm": 2.8086111545562744, + "learning_rate": 4.05109489051095e-06, + "loss": 2.1206, + "step": 236 + }, + { + "epoch": 0.25972602739726025, + "grad_norm": 2.909231185913086, + "learning_rate": 4.069343065693431e-06, + "loss": 2.139, + "step": 237 + }, + { + "epoch": 0.2608219178082192, + "grad_norm": 3.2171082496643066, + "learning_rate": 4.0875912408759126e-06, + "loss": 2.0908, + "step": 238 + }, + { + "epoch": 0.2619178082191781, + "grad_norm": 2.3851265907287598, + "learning_rate": 4.105839416058395e-06, + "loss": 2.1133, + "step": 239 + }, + { + "epoch": 0.26301369863013696, + "grad_norm": 3.3912737369537354, + "learning_rate": 4.124087591240876e-06, + "loss": 2.1911, + "step": 240 + }, + { + "epoch": 0.26301369863013696, + "eval_loss": 2.2436957359313965, + "eval_runtime": 37.2947, + "eval_samples_per_second": 26.813, + "eval_steps_per_second": 1.126, + "step": 240 + }, + { + "epoch": 0.2641095890410959, + "grad_norm": 2.82906174659729, + "learning_rate": 4.1423357664233586e-06, + "loss": 2.11, + "step": 241 + }, + { + "epoch": 0.2652054794520548, + "grad_norm": 3.0317952632904053, + "learning_rate": 4.16058394160584e-06, + "loss": 2.0953, + "step": 242 + }, + { + "epoch": 0.2663013698630137, + "grad_norm": 3.6438329219818115, + "learning_rate": 4.1788321167883214e-06, + "loss": 2.1746, + "step": 243 + }, + { + "epoch": 0.2673972602739726, + "grad_norm": 2.6690592765808105, + "learning_rate": 4.197080291970803e-06, + "loss": 2.0555, + "step": 244 + }, + { + "epoch": 0.2684931506849315, + "grad_norm": 3.135376453399658, + "learning_rate": 4.215328467153285e-06, + "loss": 2.1394, + "step": 245 + }, + { + "epoch": 0.26958904109589044, + "grad_norm": 2.5474987030029297, + "learning_rate": 4.233576642335767e-06, + "loss": 2.0822, + "step": 246 + }, + { + "epoch": 0.2706849315068493, + "grad_norm": 2.883108377456665, + "learning_rate": 4.251824817518249e-06, + "loss": 2.0869, + "step": 247 + }, + { + "epoch": 0.2717808219178082, + "grad_norm": 2.9801037311553955, + "learning_rate": 4.27007299270073e-06, + "loss": 2.1194, + "step": 248 + }, + { + "epoch": 0.27287671232876715, + "grad_norm": 2.7688217163085938, + "learning_rate": 4.288321167883212e-06, + "loss": 2.0908, + "step": 249 + }, + { + "epoch": 0.273972602739726, + "grad_norm": 3.205148458480835, + "learning_rate": 4.306569343065693e-06, + "loss": 2.0762, + "step": 250 + }, + { + "epoch": 0.273972602739726, + "eval_loss": 2.238699197769165, + "eval_runtime": 37.2428, + "eval_samples_per_second": 26.851, + "eval_steps_per_second": 1.128, + "step": 250 + }, + { + "epoch": 0.27506849315068493, + "grad_norm": 2.8262946605682373, + "learning_rate": 4.3248175182481755e-06, + "loss": 2.0509, + "step": 251 + }, + { + "epoch": 0.27616438356164386, + "grad_norm": 3.156261682510376, + "learning_rate": 4.343065693430658e-06, + "loss": 2.1213, + "step": 252 + }, + { + "epoch": 0.2772602739726027, + "grad_norm": 2.6367998123168945, + "learning_rate": 4.361313868613139e-06, + "loss": 2.1071, + "step": 253 + }, + { + "epoch": 0.27835616438356164, + "grad_norm": 2.871534824371338, + "learning_rate": 4.379562043795621e-06, + "loss": 2.1008, + "step": 254 + }, + { + "epoch": 0.27945205479452057, + "grad_norm": 2.839432716369629, + "learning_rate": 4.397810218978102e-06, + "loss": 2.1293, + "step": 255 + }, + { + "epoch": 0.28054794520547943, + "grad_norm": 2.627143383026123, + "learning_rate": 4.416058394160584e-06, + "loss": 2.1141, + "step": 256 + }, + { + "epoch": 0.28164383561643835, + "grad_norm": 2.8292996883392334, + "learning_rate": 4.434306569343066e-06, + "loss": 2.0981, + "step": 257 + }, + { + "epoch": 0.2827397260273973, + "grad_norm": 2.5711069107055664, + "learning_rate": 4.452554744525548e-06, + "loss": 2.0707, + "step": 258 + }, + { + "epoch": 0.28383561643835614, + "grad_norm": 2.488954544067383, + "learning_rate": 4.4708029197080295e-06, + "loss": 2.1172, + "step": 259 + }, + { + "epoch": 0.28493150684931506, + "grad_norm": 2.442209005355835, + "learning_rate": 4.489051094890512e-06, + "loss": 2.1119, + "step": 260 + }, + { + "epoch": 0.28493150684931506, + "eval_loss": 2.2306911945343018, + "eval_runtime": 37.2412, + "eval_samples_per_second": 26.852, + "eval_steps_per_second": 1.128, + "step": 260 + }, + { + "epoch": 0.286027397260274, + "grad_norm": 2.9147934913635254, + "learning_rate": 4.507299270072993e-06, + "loss": 2.1517, + "step": 261 + }, + { + "epoch": 0.28712328767123285, + "grad_norm": 2.549833297729492, + "learning_rate": 4.525547445255475e-06, + "loss": 2.1047, + "step": 262 + }, + { + "epoch": 0.2882191780821918, + "grad_norm": 3.3405206203460693, + "learning_rate": 4.543795620437956e-06, + "loss": 2.0975, + "step": 263 + }, + { + "epoch": 0.2893150684931507, + "grad_norm": 2.554431676864624, + "learning_rate": 4.562043795620438e-06, + "loss": 2.0622, + "step": 264 + }, + { + "epoch": 0.29041095890410956, + "grad_norm": 3.5529873371124268, + "learning_rate": 4.580291970802921e-06, + "loss": 2.1567, + "step": 265 + }, + { + "epoch": 0.2915068493150685, + "grad_norm": 2.421682119369507, + "learning_rate": 4.598540145985402e-06, + "loss": 2.1619, + "step": 266 + }, + { + "epoch": 0.2926027397260274, + "grad_norm": 2.918288230895996, + "learning_rate": 4.6167883211678835e-06, + "loss": 2.1238, + "step": 267 + }, + { + "epoch": 0.2936986301369863, + "grad_norm": 3.0737621784210205, + "learning_rate": 4.635036496350365e-06, + "loss": 2.0728, + "step": 268 + }, + { + "epoch": 0.2947945205479452, + "grad_norm": 2.5862410068511963, + "learning_rate": 4.653284671532847e-06, + "loss": 2.0675, + "step": 269 + }, + { + "epoch": 0.2958904109589041, + "grad_norm": 2.9004979133605957, + "learning_rate": 4.671532846715329e-06, + "loss": 2.1248, + "step": 270 + }, + { + "epoch": 0.2958904109589041, + "eval_loss": 2.2179365158081055, + "eval_runtime": 37.2352, + "eval_samples_per_second": 26.856, + "eval_steps_per_second": 1.128, + "step": 270 + }, + { + "epoch": 0.29698630136986304, + "grad_norm": 2.4784982204437256, + "learning_rate": 4.689781021897811e-06, + "loss": 2.1276, + "step": 271 + }, + { + "epoch": 0.2980821917808219, + "grad_norm": 2.7971739768981934, + "learning_rate": 4.708029197080292e-06, + "loss": 2.0931, + "step": 272 + }, + { + "epoch": 0.2991780821917808, + "grad_norm": 2.5839481353759766, + "learning_rate": 4.726277372262774e-06, + "loss": 2.1371, + "step": 273 + }, + { + "epoch": 0.30027397260273975, + "grad_norm": 2.7168564796447754, + "learning_rate": 4.744525547445255e-06, + "loss": 2.1086, + "step": 274 + }, + { + "epoch": 0.3013698630136986, + "grad_norm": 2.6566879749298096, + "learning_rate": 4.7627737226277375e-06, + "loss": 2.0368, + "step": 275 + }, + { + "epoch": 0.30246575342465754, + "grad_norm": 2.5068323612213135, + "learning_rate": 4.78102189781022e-06, + "loss": 2.1077, + "step": 276 + }, + { + "epoch": 0.30356164383561646, + "grad_norm": 2.631408214569092, + "learning_rate": 4.799270072992701e-06, + "loss": 2.1013, + "step": 277 + }, + { + "epoch": 0.3046575342465753, + "grad_norm": 2.5642812252044678, + "learning_rate": 4.8175182481751835e-06, + "loss": 2.0739, + "step": 278 + }, + { + "epoch": 0.30575342465753425, + "grad_norm": 2.3326497077941895, + "learning_rate": 4.835766423357665e-06, + "loss": 2.0971, + "step": 279 + }, + { + "epoch": 0.30684931506849317, + "grad_norm": 2.70701265335083, + "learning_rate": 4.854014598540146e-06, + "loss": 2.0904, + "step": 280 + }, + { + "epoch": 0.30684931506849317, + "eval_loss": 2.215972900390625, + "eval_runtime": 37.2196, + "eval_samples_per_second": 26.868, + "eval_steps_per_second": 1.128, + "step": 280 + }, + { + "epoch": 0.30794520547945203, + "grad_norm": 2.615201234817505, + "learning_rate": 4.872262773722628e-06, + "loss": 2.039, + "step": 281 + }, + { + "epoch": 0.30904109589041096, + "grad_norm": 2.2821998596191406, + "learning_rate": 4.89051094890511e-06, + "loss": 2.0529, + "step": 282 + }, + { + "epoch": 0.3101369863013699, + "grad_norm": 2.767153024673462, + "learning_rate": 4.9087591240875915e-06, + "loss": 2.0351, + "step": 283 + }, + { + "epoch": 0.31123287671232874, + "grad_norm": 2.537121295928955, + "learning_rate": 4.927007299270074e-06, + "loss": 2.0752, + "step": 284 + }, + { + "epoch": 0.31232876712328766, + "grad_norm": 2.835052013397217, + "learning_rate": 4.945255474452555e-06, + "loss": 2.0892, + "step": 285 + }, + { + "epoch": 0.3134246575342466, + "grad_norm": 2.5154924392700195, + "learning_rate": 4.963503649635037e-06, + "loss": 2.0569, + "step": 286 + }, + { + "epoch": 0.31452054794520545, + "grad_norm": 2.8212568759918213, + "learning_rate": 4.981751824817518e-06, + "loss": 2.1588, + "step": 287 + }, + { + "epoch": 0.3156164383561644, + "grad_norm": 2.3840994834899902, + "learning_rate": 5e-06, + "loss": 2.1409, + "step": 288 + }, + { + "epoch": 0.3167123287671233, + "grad_norm": 2.9641315937042236, + "learning_rate": 4.997969130787978e-06, + "loss": 2.0892, + "step": 289 + }, + { + "epoch": 0.3178082191780822, + "grad_norm": 2.4516749382019043, + "learning_rate": 4.995938261575955e-06, + "loss": 2.1105, + "step": 290 + }, + { + "epoch": 0.3178082191780822, + "eval_loss": 2.207075834274292, + "eval_runtime": 37.1304, + "eval_samples_per_second": 26.932, + "eval_steps_per_second": 1.131, + "step": 290 + }, + { + "epoch": 0.3189041095890411, + "grad_norm": 2.768869400024414, + "learning_rate": 4.993907392363932e-06, + "loss": 2.0231, + "step": 291 + }, + { + "epoch": 0.32, + "grad_norm": 2.771475315093994, + "learning_rate": 4.9918765231519095e-06, + "loss": 2.1018, + "step": 292 + }, + { + "epoch": 0.32109589041095893, + "grad_norm": 2.773983955383301, + "learning_rate": 4.989845653939887e-06, + "loss": 2.0317, + "step": 293 + }, + { + "epoch": 0.3221917808219178, + "grad_norm": 2.4447343349456787, + "learning_rate": 4.987814784727864e-06, + "loss": 2.0739, + "step": 294 + }, + { + "epoch": 0.3232876712328767, + "grad_norm": 2.444261312484741, + "learning_rate": 4.985783915515841e-06, + "loss": 2.073, + "step": 295 + }, + { + "epoch": 0.32438356164383564, + "grad_norm": 2.871792793273926, + "learning_rate": 4.983753046303819e-06, + "loss": 2.058, + "step": 296 + }, + { + "epoch": 0.3254794520547945, + "grad_norm": 2.3651537895202637, + "learning_rate": 4.981722177091796e-06, + "loss": 2.0749, + "step": 297 + }, + { + "epoch": 0.3265753424657534, + "grad_norm": 2.3395845890045166, + "learning_rate": 4.979691307879773e-06, + "loss": 2.065, + "step": 298 + }, + { + "epoch": 0.32767123287671235, + "grad_norm": 2.6275217533111572, + "learning_rate": 4.9776604386677504e-06, + "loss": 2.0945, + "step": 299 + }, + { + "epoch": 0.3287671232876712, + "grad_norm": 2.866285562515259, + "learning_rate": 4.975629569455728e-06, + "loss": 2.0702, + "step": 300 + }, + { + "epoch": 0.3287671232876712, + "eval_loss": 2.200012683868408, + "eval_runtime": 37.2082, + "eval_samples_per_second": 26.876, + "eval_steps_per_second": 1.129, + "step": 300 + }, + { + "epoch": 0.32986301369863014, + "grad_norm": 2.268401861190796, + "learning_rate": 4.973598700243705e-06, + "loss": 2.1257, + "step": 301 + }, + { + "epoch": 0.33095890410958906, + "grad_norm": 2.6197733879089355, + "learning_rate": 4.971567831031681e-06, + "loss": 2.0347, + "step": 302 + }, + { + "epoch": 0.3320547945205479, + "grad_norm": 2.4738848209381104, + "learning_rate": 4.9695369618196595e-06, + "loss": 2.0368, + "step": 303 + }, + { + "epoch": 0.33315068493150685, + "grad_norm": 2.705670118331909, + "learning_rate": 4.967506092607636e-06, + "loss": 1.9256, + "step": 304 + }, + { + "epoch": 0.33424657534246577, + "grad_norm": 2.813279628753662, + "learning_rate": 4.965475223395614e-06, + "loss": 2.0214, + "step": 305 + }, + { + "epoch": 0.33534246575342463, + "grad_norm": 2.320960283279419, + "learning_rate": 4.9634443541835905e-06, + "loss": 2.0342, + "step": 306 + }, + { + "epoch": 0.33643835616438356, + "grad_norm": 2.4495227336883545, + "learning_rate": 4.961413484971569e-06, + "loss": 2.0421, + "step": 307 + }, + { + "epoch": 0.3375342465753425, + "grad_norm": 2.5384018421173096, + "learning_rate": 4.959382615759545e-06, + "loss": 2.0572, + "step": 308 + }, + { + "epoch": 0.33863013698630134, + "grad_norm": 2.4018006324768066, + "learning_rate": 4.957351746547523e-06, + "loss": 2.0617, + "step": 309 + }, + { + "epoch": 0.33972602739726027, + "grad_norm": 2.552424907684326, + "learning_rate": 4.9553208773355e-06, + "loss": 2.0384, + "step": 310 + }, + { + "epoch": 0.33972602739726027, + "eval_loss": 2.1974849700927734, + "eval_runtime": 37.2308, + "eval_samples_per_second": 26.859, + "eval_steps_per_second": 1.128, + "step": 310 + }, + { + "epoch": 0.3408219178082192, + "grad_norm": 2.4587671756744385, + "learning_rate": 4.953290008123478e-06, + "loss": 2.1701, + "step": 311 + }, + { + "epoch": 0.3419178082191781, + "grad_norm": 2.4486496448516846, + "learning_rate": 4.951259138911454e-06, + "loss": 2.0654, + "step": 312 + }, + { + "epoch": 0.343013698630137, + "grad_norm": 2.123568058013916, + "learning_rate": 4.9492282696994315e-06, + "loss": 1.9961, + "step": 313 + }, + { + "epoch": 0.3441095890410959, + "grad_norm": 2.549788475036621, + "learning_rate": 4.947197400487409e-06, + "loss": 2.1265, + "step": 314 + }, + { + "epoch": 0.3452054794520548, + "grad_norm": 2.3453609943389893, + "learning_rate": 4.945166531275386e-06, + "loss": 2.0501, + "step": 315 + }, + { + "epoch": 0.3463013698630137, + "grad_norm": 2.3584256172180176, + "learning_rate": 4.943135662063363e-06, + "loss": 2.0402, + "step": 316 + }, + { + "epoch": 0.3473972602739726, + "grad_norm": 3.3100943565368652, + "learning_rate": 4.9411047928513406e-06, + "loss": 2.0545, + "step": 317 + }, + { + "epoch": 0.34849315068493153, + "grad_norm": 2.4266037940979004, + "learning_rate": 4.939073923639318e-06, + "loss": 2.0926, + "step": 318 + }, + { + "epoch": 0.3495890410958904, + "grad_norm": 3.101090908050537, + "learning_rate": 4.937043054427295e-06, + "loss": 2.0987, + "step": 319 + }, + { + "epoch": 0.3506849315068493, + "grad_norm": 2.743619203567505, + "learning_rate": 4.935012185215272e-06, + "loss": 2.0315, + "step": 320 + }, + { + "epoch": 0.3506849315068493, + "eval_loss": 2.188877820968628, + "eval_runtime": 37.4609, + "eval_samples_per_second": 26.694, + "eval_steps_per_second": 1.121, + "step": 320 + }, + { + "epoch": 0.35178082191780824, + "grad_norm": 2.6065986156463623, + "learning_rate": 4.93298131600325e-06, + "loss": 2.0166, + "step": 321 + }, + { + "epoch": 0.3528767123287671, + "grad_norm": 2.7554283142089844, + "learning_rate": 4.930950446791227e-06, + "loss": 2.0862, + "step": 322 + }, + { + "epoch": 0.353972602739726, + "grad_norm": 2.4773144721984863, + "learning_rate": 4.928919577579204e-06, + "loss": 2.0472, + "step": 323 + }, + { + "epoch": 0.35506849315068495, + "grad_norm": 2.695289134979248, + "learning_rate": 4.9268887083671815e-06, + "loss": 2.0567, + "step": 324 + }, + { + "epoch": 0.3561643835616438, + "grad_norm": 2.408220052719116, + "learning_rate": 4.924857839155159e-06, + "loss": 2.046, + "step": 325 + }, + { + "epoch": 0.35726027397260274, + "grad_norm": 2.3851559162139893, + "learning_rate": 4.922826969943136e-06, + "loss": 2.054, + "step": 326 + }, + { + "epoch": 0.35835616438356166, + "grad_norm": 2.8667280673980713, + "learning_rate": 4.920796100731113e-06, + "loss": 2.0717, + "step": 327 + }, + { + "epoch": 0.3594520547945205, + "grad_norm": 2.3814942836761475, + "learning_rate": 4.918765231519091e-06, + "loss": 1.9884, + "step": 328 + }, + { + "epoch": 0.36054794520547945, + "grad_norm": 2.341076612472534, + "learning_rate": 4.916734362307068e-06, + "loss": 2.0474, + "step": 329 + }, + { + "epoch": 0.36164383561643837, + "grad_norm": 2.3571853637695312, + "learning_rate": 4.914703493095045e-06, + "loss": 2.0009, + "step": 330 + }, + { + "epoch": 0.36164383561643837, + "eval_loss": 2.186048746109009, + "eval_runtime": 37.5746, + "eval_samples_per_second": 26.614, + "eval_steps_per_second": 1.118, + "step": 330 + }, + { + "epoch": 0.36273972602739724, + "grad_norm": 2.3770196437835693, + "learning_rate": 4.912672623883022e-06, + "loss": 2.0167, + "step": 331 + }, + { + "epoch": 0.36383561643835616, + "grad_norm": 2.3357701301574707, + "learning_rate": 4.910641754671e-06, + "loss": 1.9972, + "step": 332 + }, + { + "epoch": 0.3649315068493151, + "grad_norm": 2.4100499153137207, + "learning_rate": 4.908610885458977e-06, + "loss": 2.0319, + "step": 333 + }, + { + "epoch": 0.36602739726027395, + "grad_norm": 2.193018913269043, + "learning_rate": 4.906580016246954e-06, + "loss": 2.0285, + "step": 334 + }, + { + "epoch": 0.36712328767123287, + "grad_norm": 2.275637149810791, + "learning_rate": 4.9045491470349315e-06, + "loss": 2.0749, + "step": 335 + }, + { + "epoch": 0.3682191780821918, + "grad_norm": 2.347121238708496, + "learning_rate": 4.902518277822909e-06, + "loss": 2.0502, + "step": 336 + }, + { + "epoch": 0.3693150684931507, + "grad_norm": 2.210817575454712, + "learning_rate": 4.900487408610886e-06, + "loss": 2.0612, + "step": 337 + }, + { + "epoch": 0.3704109589041096, + "grad_norm": 2.15594482421875, + "learning_rate": 4.898456539398863e-06, + "loss": 2.0653, + "step": 338 + }, + { + "epoch": 0.3715068493150685, + "grad_norm": 2.3996052742004395, + "learning_rate": 4.89642567018684e-06, + "loss": 2.074, + "step": 339 + }, + { + "epoch": 0.3726027397260274, + "grad_norm": 2.384172201156616, + "learning_rate": 4.894394800974818e-06, + "loss": 2.1533, + "step": 340 + }, + { + "epoch": 0.3726027397260274, + "eval_loss": 2.175034284591675, + "eval_runtime": 37.2455, + "eval_samples_per_second": 26.849, + "eval_steps_per_second": 1.128, + "step": 340 + }, + { + "epoch": 0.3736986301369863, + "grad_norm": 2.3431336879730225, + "learning_rate": 4.892363931762794e-06, + "loss": 1.9919, + "step": 341 + }, + { + "epoch": 0.3747945205479452, + "grad_norm": 2.1828389167785645, + "learning_rate": 4.8903330625507725e-06, + "loss": 2.0703, + "step": 342 + }, + { + "epoch": 0.37589041095890413, + "grad_norm": 2.2117249965667725, + "learning_rate": 4.888302193338749e-06, + "loss": 2.0071, + "step": 343 + }, + { + "epoch": 0.376986301369863, + "grad_norm": 2.3595967292785645, + "learning_rate": 4.886271324126727e-06, + "loss": 1.9879, + "step": 344 + }, + { + "epoch": 0.3780821917808219, + "grad_norm": 2.5211799144744873, + "learning_rate": 4.8842404549147034e-06, + "loss": 2.0861, + "step": 345 + }, + { + "epoch": 0.37917808219178084, + "grad_norm": 2.1551895141601562, + "learning_rate": 4.8822095857026816e-06, + "loss": 2.0683, + "step": 346 + }, + { + "epoch": 0.3802739726027397, + "grad_norm": 2.4145829677581787, + "learning_rate": 4.880178716490658e-06, + "loss": 2.043, + "step": 347 + }, + { + "epoch": 0.38136986301369863, + "grad_norm": 2.4273521900177, + "learning_rate": 4.878147847278636e-06, + "loss": 1.9928, + "step": 348 + }, + { + "epoch": 0.38246575342465755, + "grad_norm": 2.261054754257202, + "learning_rate": 4.8761169780666125e-06, + "loss": 2.0065, + "step": 349 + }, + { + "epoch": 0.3835616438356164, + "grad_norm": 2.4703147411346436, + "learning_rate": 4.874086108854591e-06, + "loss": 1.9766, + "step": 350 + }, + { + "epoch": 0.3835616438356164, + "eval_loss": 2.176915168762207, + "eval_runtime": 37.3223, + "eval_samples_per_second": 26.794, + "eval_steps_per_second": 1.125, + "step": 350 + }, + { + "epoch": 0.38465753424657534, + "grad_norm": 2.3353374004364014, + "learning_rate": 4.872055239642567e-06, + "loss": 1.9874, + "step": 351 + }, + { + "epoch": 0.38575342465753426, + "grad_norm": 2.227834463119507, + "learning_rate": 4.870024370430544e-06, + "loss": 1.9644, + "step": 352 + }, + { + "epoch": 0.3868493150684931, + "grad_norm": 2.3155159950256348, + "learning_rate": 4.867993501218522e-06, + "loss": 2.0227, + "step": 353 + }, + { + "epoch": 0.38794520547945205, + "grad_norm": 2.2456047534942627, + "learning_rate": 4.865962632006499e-06, + "loss": 2.0566, + "step": 354 + }, + { + "epoch": 0.38904109589041097, + "grad_norm": 2.3364651203155518, + "learning_rate": 4.863931762794476e-06, + "loss": 1.9941, + "step": 355 + }, + { + "epoch": 0.39013698630136984, + "grad_norm": 2.2886059284210205, + "learning_rate": 4.8619008935824535e-06, + "loss": 2.0163, + "step": 356 + }, + { + "epoch": 0.39123287671232876, + "grad_norm": 2.3796911239624023, + "learning_rate": 4.859870024370431e-06, + "loss": 2.0525, + "step": 357 + }, + { + "epoch": 0.3923287671232877, + "grad_norm": 2.381986379623413, + "learning_rate": 4.857839155158408e-06, + "loss": 2.0682, + "step": 358 + }, + { + "epoch": 0.3934246575342466, + "grad_norm": 2.2048425674438477, + "learning_rate": 4.855808285946385e-06, + "loss": 2.0748, + "step": 359 + }, + { + "epoch": 0.39452054794520547, + "grad_norm": 2.162846088409424, + "learning_rate": 4.8537774167343626e-06, + "loss": 2.0512, + "step": 360 + }, + { + "epoch": 0.39452054794520547, + "eval_loss": 2.1645994186401367, + "eval_runtime": 37.3126, + "eval_samples_per_second": 26.801, + "eval_steps_per_second": 1.126, + "step": 360 + }, + { + "epoch": 0.3956164383561644, + "grad_norm": 2.4070873260498047, + "learning_rate": 4.85174654752234e-06, + "loss": 2.042, + "step": 361 + }, + { + "epoch": 0.3967123287671233, + "grad_norm": 2.11716628074646, + "learning_rate": 4.849715678310317e-06, + "loss": 2.0398, + "step": 362 + }, + { + "epoch": 0.3978082191780822, + "grad_norm": 2.265479803085327, + "learning_rate": 4.847684809098294e-06, + "loss": 2.0334, + "step": 363 + }, + { + "epoch": 0.3989041095890411, + "grad_norm": 2.2270994186401367, + "learning_rate": 4.845653939886272e-06, + "loss": 2.0292, + "step": 364 + }, + { + "epoch": 0.4, + "grad_norm": 2.406064033508301, + "learning_rate": 4.843623070674249e-06, + "loss": 2.0351, + "step": 365 + }, + { + "epoch": 0.4010958904109589, + "grad_norm": 2.216654062271118, + "learning_rate": 4.841592201462226e-06, + "loss": 2.005, + "step": 366 + }, + { + "epoch": 0.4021917808219178, + "grad_norm": 2.4241385459899902, + "learning_rate": 4.8395613322502035e-06, + "loss": 2.0647, + "step": 367 + }, + { + "epoch": 0.40328767123287673, + "grad_norm": 2.4315145015716553, + "learning_rate": 4.837530463038181e-06, + "loss": 1.9679, + "step": 368 + }, + { + "epoch": 0.4043835616438356, + "grad_norm": 2.3034439086914062, + "learning_rate": 4.835499593826158e-06, + "loss": 2.0259, + "step": 369 + }, + { + "epoch": 0.4054794520547945, + "grad_norm": 2.493436098098755, + "learning_rate": 4.833468724614135e-06, + "loss": 2.0468, + "step": 370 + }, + { + "epoch": 0.4054794520547945, + "eval_loss": 2.161508083343506, + "eval_runtime": 37.2692, + "eval_samples_per_second": 26.832, + "eval_steps_per_second": 1.127, + "step": 370 + }, + { + "epoch": 0.40657534246575344, + "grad_norm": 2.3887157440185547, + "learning_rate": 4.831437855402113e-06, + "loss": 2.0313, + "step": 371 + }, + { + "epoch": 0.4076712328767123, + "grad_norm": 2.563095808029175, + "learning_rate": 4.82940698619009e-06, + "loss": 1.9288, + "step": 372 + }, + { + "epoch": 0.40876712328767123, + "grad_norm": 2.2609968185424805, + "learning_rate": 4.827376116978067e-06, + "loss": 1.9975, + "step": 373 + }, + { + "epoch": 0.40986301369863015, + "grad_norm": 2.624257802963257, + "learning_rate": 4.8253452477660444e-06, + "loss": 2.0639, + "step": 374 + }, + { + "epoch": 0.410958904109589, + "grad_norm": 2.502197265625, + "learning_rate": 4.823314378554022e-06, + "loss": 2.0424, + "step": 375 + }, + { + "epoch": 0.41205479452054794, + "grad_norm": 2.36297345161438, + "learning_rate": 4.821283509341998e-06, + "loss": 2.027, + "step": 376 + }, + { + "epoch": 0.41315068493150686, + "grad_norm": 2.388918161392212, + "learning_rate": 4.819252640129976e-06, + "loss": 2.0002, + "step": 377 + }, + { + "epoch": 0.41424657534246573, + "grad_norm": 2.4262335300445557, + "learning_rate": 4.817221770917953e-06, + "loss": 2.0215, + "step": 378 + }, + { + "epoch": 0.41534246575342465, + "grad_norm": 2.8953094482421875, + "learning_rate": 4.815190901705931e-06, + "loss": 1.988, + "step": 379 + }, + { + "epoch": 0.41643835616438357, + "grad_norm": 2.3092124462127686, + "learning_rate": 4.813160032493907e-06, + "loss": 2.0199, + "step": 380 + }, + { + "epoch": 0.41643835616438357, + "eval_loss": 2.1589064598083496, + "eval_runtime": 37.2649, + "eval_samples_per_second": 26.835, + "eval_steps_per_second": 1.127, + "step": 380 + }, + { + "epoch": 0.41753424657534244, + "grad_norm": 3.0842413902282715, + "learning_rate": 4.811129163281885e-06, + "loss": 2.0134, + "step": 381 + }, + { + "epoch": 0.41863013698630136, + "grad_norm": 2.2015092372894287, + "learning_rate": 4.809098294069862e-06, + "loss": 2.0447, + "step": 382 + }, + { + "epoch": 0.4197260273972603, + "grad_norm": 2.93471097946167, + "learning_rate": 4.80706742485784e-06, + "loss": 2.0329, + "step": 383 + }, + { + "epoch": 0.4208219178082192, + "grad_norm": 2.3277227878570557, + "learning_rate": 4.805036555645816e-06, + "loss": 2.0805, + "step": 384 + }, + { + "epoch": 0.42191780821917807, + "grad_norm": 2.357966184616089, + "learning_rate": 4.8030056864337945e-06, + "loss": 2.0008, + "step": 385 + }, + { + "epoch": 0.423013698630137, + "grad_norm": 2.7505595684051514, + "learning_rate": 4.800974817221771e-06, + "loss": 2.059, + "step": 386 + }, + { + "epoch": 0.4241095890410959, + "grad_norm": 2.144338369369507, + "learning_rate": 4.798943948009749e-06, + "loss": 2.0125, + "step": 387 + }, + { + "epoch": 0.4252054794520548, + "grad_norm": 2.608407735824585, + "learning_rate": 4.7969130787977254e-06, + "loss": 2.0263, + "step": 388 + }, + { + "epoch": 0.4263013698630137, + "grad_norm": 2.4293761253356934, + "learning_rate": 4.794882209585703e-06, + "loss": 2.0107, + "step": 389 + }, + { + "epoch": 0.4273972602739726, + "grad_norm": 2.2136595249176025, + "learning_rate": 4.79285134037368e-06, + "loss": 2.0421, + "step": 390 + }, + { + "epoch": 0.4273972602739726, + "eval_loss": 2.1473641395568848, + "eval_runtime": 37.2895, + "eval_samples_per_second": 26.817, + "eval_steps_per_second": 1.126, + "step": 390 + }, + { + "epoch": 0.4284931506849315, + "grad_norm": 2.3706328868865967, + "learning_rate": 4.790820471161657e-06, + "loss": 1.9828, + "step": 391 + }, + { + "epoch": 0.4295890410958904, + "grad_norm": 2.1628777980804443, + "learning_rate": 4.7887896019496345e-06, + "loss": 2.0006, + "step": 392 + }, + { + "epoch": 0.43068493150684933, + "grad_norm": 2.5572149753570557, + "learning_rate": 4.786758732737612e-06, + "loss": 2.0046, + "step": 393 + }, + { + "epoch": 0.4317808219178082, + "grad_norm": 2.44661808013916, + "learning_rate": 4.784727863525589e-06, + "loss": 2.0513, + "step": 394 + }, + { + "epoch": 0.4328767123287671, + "grad_norm": 2.429487466812134, + "learning_rate": 4.782696994313566e-06, + "loss": 1.9619, + "step": 395 + }, + { + "epoch": 0.43397260273972604, + "grad_norm": 2.233353614807129, + "learning_rate": 4.780666125101544e-06, + "loss": 2.0141, + "step": 396 + }, + { + "epoch": 0.4350684931506849, + "grad_norm": 2.137549638748169, + "learning_rate": 4.778635255889521e-06, + "loss": 1.9512, + "step": 397 + }, + { + "epoch": 0.43616438356164383, + "grad_norm": 2.187638759613037, + "learning_rate": 4.776604386677498e-06, + "loss": 1.9802, + "step": 398 + }, + { + "epoch": 0.43726027397260275, + "grad_norm": 2.3084137439727783, + "learning_rate": 4.7745735174654755e-06, + "loss": 1.9887, + "step": 399 + }, + { + "epoch": 0.4383561643835616, + "grad_norm": 2.16654109954834, + "learning_rate": 4.772542648253453e-06, + "loss": 2.0535, + "step": 400 + }, + { + "epoch": 0.4383561643835616, + "eval_loss": 2.1444413661956787, + "eval_runtime": 37.2158, + "eval_samples_per_second": 26.87, + "eval_steps_per_second": 1.129, + "step": 400 + }, + { + "epoch": 0.43945205479452054, + "grad_norm": 2.0973072052001953, + "learning_rate": 4.77051177904143e-06, + "loss": 2.0529, + "step": 401 + }, + { + "epoch": 0.44054794520547946, + "grad_norm": 2.421881675720215, + "learning_rate": 4.768480909829407e-06, + "loss": 1.9893, + "step": 402 + }, + { + "epoch": 0.44164383561643833, + "grad_norm": 2.349296808242798, + "learning_rate": 4.766450040617385e-06, + "loss": 2.0237, + "step": 403 + }, + { + "epoch": 0.44273972602739725, + "grad_norm": 2.2154288291931152, + "learning_rate": 4.764419171405362e-06, + "loss": 1.996, + "step": 404 + }, + { + "epoch": 0.4438356164383562, + "grad_norm": 2.6355648040771484, + "learning_rate": 4.762388302193339e-06, + "loss": 2.0345, + "step": 405 + }, + { + "epoch": 0.4449315068493151, + "grad_norm": 2.352238655090332, + "learning_rate": 4.760357432981316e-06, + "loss": 2.0219, + "step": 406 + }, + { + "epoch": 0.44602739726027396, + "grad_norm": 2.1886510848999023, + "learning_rate": 4.758326563769294e-06, + "loss": 1.9923, + "step": 407 + }, + { + "epoch": 0.4471232876712329, + "grad_norm": 2.6483092308044434, + "learning_rate": 4.756295694557271e-06, + "loss": 2.0236, + "step": 408 + }, + { + "epoch": 0.4482191780821918, + "grad_norm": 2.382389545440674, + "learning_rate": 4.754264825345248e-06, + "loss": 2.0249, + "step": 409 + }, + { + "epoch": 0.44931506849315067, + "grad_norm": 2.4824411869049072, + "learning_rate": 4.7522339561332255e-06, + "loss": 1.9571, + "step": 410 + }, + { + "epoch": 0.44931506849315067, + "eval_loss": 2.1502621173858643, + "eval_runtime": 37.1789, + "eval_samples_per_second": 26.897, + "eval_steps_per_second": 1.13, + "step": 410 + }, + { + "epoch": 0.4504109589041096, + "grad_norm": 2.325589656829834, + "learning_rate": 4.750203086921203e-06, + "loss": 1.9753, + "step": 411 + }, + { + "epoch": 0.4515068493150685, + "grad_norm": 2.3618829250335693, + "learning_rate": 4.74817221770918e-06, + "loss": 1.9397, + "step": 412 + }, + { + "epoch": 0.4526027397260274, + "grad_norm": 2.3922839164733887, + "learning_rate": 4.7461413484971565e-06, + "loss": 2.0119, + "step": 413 + }, + { + "epoch": 0.4536986301369863, + "grad_norm": 2.403475046157837, + "learning_rate": 4.744110479285135e-06, + "loss": 2.0664, + "step": 414 + }, + { + "epoch": 0.4547945205479452, + "grad_norm": 2.225860834121704, + "learning_rate": 4.742079610073111e-06, + "loss": 1.9525, + "step": 415 + }, + { + "epoch": 0.4558904109589041, + "grad_norm": 2.344074010848999, + "learning_rate": 4.740048740861089e-06, + "loss": 1.9785, + "step": 416 + }, + { + "epoch": 0.456986301369863, + "grad_norm": 2.255382776260376, + "learning_rate": 4.738017871649066e-06, + "loss": 1.9902, + "step": 417 + }, + { + "epoch": 0.45808219178082193, + "grad_norm": 2.160628318786621, + "learning_rate": 4.735987002437044e-06, + "loss": 2.0075, + "step": 418 + }, + { + "epoch": 0.4591780821917808, + "grad_norm": 2.5667307376861572, + "learning_rate": 4.73395613322502e-06, + "loss": 2.0625, + "step": 419 + }, + { + "epoch": 0.4602739726027397, + "grad_norm": 2.3771040439605713, + "learning_rate": 4.731925264012998e-06, + "loss": 2.0724, + "step": 420 + }, + { + "epoch": 0.4602739726027397, + "eval_loss": 2.138829231262207, + "eval_runtime": 37.4361, + "eval_samples_per_second": 26.712, + "eval_steps_per_second": 1.122, + "step": 420 + }, + { + "epoch": 0.46136986301369864, + "grad_norm": 2.336482048034668, + "learning_rate": 4.729894394800975e-06, + "loss": 2.0234, + "step": 421 + }, + { + "epoch": 0.4624657534246575, + "grad_norm": 2.4206933975219727, + "learning_rate": 4.727863525588953e-06, + "loss": 1.9693, + "step": 422 + }, + { + "epoch": 0.46356164383561643, + "grad_norm": 2.4845645427703857, + "learning_rate": 4.725832656376929e-06, + "loss": 1.9239, + "step": 423 + }, + { + "epoch": 0.46465753424657535, + "grad_norm": 2.5199646949768066, + "learning_rate": 4.723801787164907e-06, + "loss": 1.9801, + "step": 424 + }, + { + "epoch": 0.4657534246575342, + "grad_norm": 2.2969748973846436, + "learning_rate": 4.721770917952885e-06, + "loss": 2.0192, + "step": 425 + }, + { + "epoch": 0.46684931506849314, + "grad_norm": 2.423503875732422, + "learning_rate": 4.719740048740861e-06, + "loss": 1.9684, + "step": 426 + }, + { + "epoch": 0.46794520547945206, + "grad_norm": 2.3681061267852783, + "learning_rate": 4.717709179528839e-06, + "loss": 1.9934, + "step": 427 + }, + { + "epoch": 0.469041095890411, + "grad_norm": 2.277965784072876, + "learning_rate": 4.715678310316816e-06, + "loss": 2.0166, + "step": 428 + }, + { + "epoch": 0.47013698630136985, + "grad_norm": 2.2730283737182617, + "learning_rate": 4.713647441104794e-06, + "loss": 2.011, + "step": 429 + }, + { + "epoch": 0.4712328767123288, + "grad_norm": 2.6694860458374023, + "learning_rate": 4.71161657189277e-06, + "loss": 1.9645, + "step": 430 + }, + { + "epoch": 0.4712328767123288, + "eval_loss": 2.1322011947631836, + "eval_runtime": 37.5723, + "eval_samples_per_second": 26.615, + "eval_steps_per_second": 1.118, + "step": 430 + }, + { + "epoch": 0.4723287671232877, + "grad_norm": 2.4100632667541504, + "learning_rate": 4.709585702680748e-06, + "loss": 1.9478, + "step": 431 + }, + { + "epoch": 0.47342465753424656, + "grad_norm": 2.2779541015625, + "learning_rate": 4.707554833468725e-06, + "loss": 2.0274, + "step": 432 + }, + { + "epoch": 0.4745205479452055, + "grad_norm": 2.55690336227417, + "learning_rate": 4.705523964256703e-06, + "loss": 2.0831, + "step": 433 + }, + { + "epoch": 0.4756164383561644, + "grad_norm": 2.455623149871826, + "learning_rate": 4.703493095044679e-06, + "loss": 1.9673, + "step": 434 + }, + { + "epoch": 0.4767123287671233, + "grad_norm": 2.2613699436187744, + "learning_rate": 4.701462225832657e-06, + "loss": 1.992, + "step": 435 + }, + { + "epoch": 0.4778082191780822, + "grad_norm": 2.4467933177948, + "learning_rate": 4.699431356620634e-06, + "loss": 1.9105, + "step": 436 + }, + { + "epoch": 0.4789041095890411, + "grad_norm": 2.4194023609161377, + "learning_rate": 4.697400487408612e-06, + "loss": 2.074, + "step": 437 + }, + { + "epoch": 0.48, + "grad_norm": 2.2165699005126953, + "learning_rate": 4.695369618196588e-06, + "loss": 2.0152, + "step": 438 + }, + { + "epoch": 0.4810958904109589, + "grad_norm": 2.1709225177764893, + "learning_rate": 4.693338748984566e-06, + "loss": 1.9888, + "step": 439 + }, + { + "epoch": 0.4821917808219178, + "grad_norm": 2.2325828075408936, + "learning_rate": 4.691307879772543e-06, + "loss": 1.9838, + "step": 440 + }, + { + "epoch": 0.4821917808219178, + "eval_loss": 2.128732919692993, + "eval_runtime": 37.205, + "eval_samples_per_second": 26.878, + "eval_steps_per_second": 1.129, + "step": 440 + }, + { + "epoch": 0.4832876712328767, + "grad_norm": 2.446331739425659, + "learning_rate": 4.68927701056052e-06, + "loss": 1.9295, + "step": 441 + }, + { + "epoch": 0.4843835616438356, + "grad_norm": 2.5459415912628174, + "learning_rate": 4.6872461413484975e-06, + "loss": 2.0351, + "step": 442 + }, + { + "epoch": 0.48547945205479454, + "grad_norm": 2.2877118587493896, + "learning_rate": 4.685215272136475e-06, + "loss": 2.0, + "step": 443 + }, + { + "epoch": 0.4865753424657534, + "grad_norm": 2.346625328063965, + "learning_rate": 4.683184402924452e-06, + "loss": 1.9928, + "step": 444 + }, + { + "epoch": 0.4876712328767123, + "grad_norm": 2.268467664718628, + "learning_rate": 4.681153533712429e-06, + "loss": 1.9969, + "step": 445 + }, + { + "epoch": 0.48876712328767125, + "grad_norm": 2.3699793815612793, + "learning_rate": 4.679122664500407e-06, + "loss": 2.0636, + "step": 446 + }, + { + "epoch": 0.4898630136986301, + "grad_norm": 2.2871041297912598, + "learning_rate": 4.677091795288384e-06, + "loss": 1.9535, + "step": 447 + }, + { + "epoch": 0.49095890410958903, + "grad_norm": 2.249393939971924, + "learning_rate": 4.675060926076361e-06, + "loss": 1.9358, + "step": 448 + }, + { + "epoch": 0.49205479452054796, + "grad_norm": 2.299863815307617, + "learning_rate": 4.673030056864338e-06, + "loss": 2.0299, + "step": 449 + }, + { + "epoch": 0.4931506849315068, + "grad_norm": 2.31239652633667, + "learning_rate": 4.670999187652316e-06, + "loss": 1.9372, + "step": 450 + }, + { + "epoch": 0.4931506849315068, + "eval_loss": 2.1267426013946533, + "eval_runtime": 37.1884, + "eval_samples_per_second": 26.89, + "eval_steps_per_second": 1.129, + "step": 450 + }, + { + "epoch": 0.49424657534246574, + "grad_norm": 2.2824623584747314, + "learning_rate": 4.668968318440293e-06, + "loss": 1.9435, + "step": 451 + }, + { + "epoch": 0.49534246575342467, + "grad_norm": 2.285339593887329, + "learning_rate": 4.66693744922827e-06, + "loss": 1.9455, + "step": 452 + }, + { + "epoch": 0.4964383561643836, + "grad_norm": 2.346592903137207, + "learning_rate": 4.6649065800162475e-06, + "loss": 2.0212, + "step": 453 + }, + { + "epoch": 0.49753424657534245, + "grad_norm": 2.280669689178467, + "learning_rate": 4.662875710804225e-06, + "loss": 1.9136, + "step": 454 + }, + { + "epoch": 0.4986301369863014, + "grad_norm": 2.178434133529663, + "learning_rate": 4.660844841592202e-06, + "loss": 1.9952, + "step": 455 + }, + { + "epoch": 0.4997260273972603, + "grad_norm": 2.108492851257324, + "learning_rate": 4.658813972380179e-06, + "loss": 1.9022, + "step": 456 + }, + { + "epoch": 0.5008219178082192, + "grad_norm": 2.3862147331237793, + "learning_rate": 4.656783103168157e-06, + "loss": 2.0332, + "step": 457 + }, + { + "epoch": 0.5019178082191781, + "grad_norm": 2.2017860412597656, + "learning_rate": 4.654752233956134e-06, + "loss": 1.8998, + "step": 458 + }, + { + "epoch": 0.503013698630137, + "grad_norm": 2.222134828567505, + "learning_rate": 4.652721364744111e-06, + "loss": 1.9836, + "step": 459 + }, + { + "epoch": 0.5041095890410959, + "grad_norm": 2.2974510192871094, + "learning_rate": 4.6506904955320885e-06, + "loss": 1.9626, + "step": 460 + }, + { + "epoch": 0.5041095890410959, + "eval_loss": 2.1345443725585938, + "eval_runtime": 37.2785, + "eval_samples_per_second": 26.825, + "eval_steps_per_second": 1.127, + "step": 460 + }, + { + "epoch": 0.5052054794520547, + "grad_norm": 2.107480764389038, + "learning_rate": 4.648659626320066e-06, + "loss": 2.0061, + "step": 461 + }, + { + "epoch": 0.5063013698630137, + "grad_norm": 2.3541181087493896, + "learning_rate": 4.646628757108043e-06, + "loss": 2.0133, + "step": 462 + }, + { + "epoch": 0.5073972602739726, + "grad_norm": 2.215571880340576, + "learning_rate": 4.6445978878960194e-06, + "loss": 1.9727, + "step": 463 + }, + { + "epoch": 0.5084931506849315, + "grad_norm": 2.2263615131378174, + "learning_rate": 4.6425670186839976e-06, + "loss": 1.9477, + "step": 464 + }, + { + "epoch": 0.5095890410958904, + "grad_norm": 2.178900957107544, + "learning_rate": 4.640536149471974e-06, + "loss": 1.9842, + "step": 465 + }, + { + "epoch": 0.5106849315068493, + "grad_norm": 2.3703105449676514, + "learning_rate": 4.638505280259952e-06, + "loss": 2.0082, + "step": 466 + }, + { + "epoch": 0.5117808219178083, + "grad_norm": 2.1587533950805664, + "learning_rate": 4.6364744110479285e-06, + "loss": 2.0271, + "step": 467 + }, + { + "epoch": 0.5128767123287671, + "grad_norm": 2.1626017093658447, + "learning_rate": 4.634443541835907e-06, + "loss": 1.9858, + "step": 468 + }, + { + "epoch": 0.513972602739726, + "grad_norm": 2.326869487762451, + "learning_rate": 4.632412672623883e-06, + "loss": 2.0413, + "step": 469 + }, + { + "epoch": 0.5150684931506849, + "grad_norm": 2.266502857208252, + "learning_rate": 4.630381803411861e-06, + "loss": 1.9584, + "step": 470 + }, + { + "epoch": 0.5150684931506849, + "eval_loss": 2.1273915767669678, + "eval_runtime": 37.2571, + "eval_samples_per_second": 26.841, + "eval_steps_per_second": 1.127, + "step": 470 + }, + { + "epoch": 0.5161643835616438, + "grad_norm": 2.131348133087158, + "learning_rate": 4.628350934199838e-06, + "loss": 1.863, + "step": 471 + }, + { + "epoch": 0.5172602739726028, + "grad_norm": 2.2635257244110107, + "learning_rate": 4.626320064987816e-06, + "loss": 1.9699, + "step": 472 + }, + { + "epoch": 0.5183561643835617, + "grad_norm": 2.1303744316101074, + "learning_rate": 4.624289195775792e-06, + "loss": 2.0001, + "step": 473 + }, + { + "epoch": 0.5194520547945205, + "grad_norm": 2.1736788749694824, + "learning_rate": 4.62225832656377e-06, + "loss": 1.9684, + "step": 474 + }, + { + "epoch": 0.5205479452054794, + "grad_norm": 2.1996278762817383, + "learning_rate": 4.620227457351747e-06, + "loss": 1.9739, + "step": 475 + }, + { + "epoch": 0.5216438356164383, + "grad_norm": 2.225456953048706, + "learning_rate": 4.618196588139724e-06, + "loss": 1.9649, + "step": 476 + }, + { + "epoch": 0.5227397260273973, + "grad_norm": 2.118537187576294, + "learning_rate": 4.616165718927701e-06, + "loss": 1.9911, + "step": 477 + }, + { + "epoch": 0.5238356164383562, + "grad_norm": 2.232940435409546, + "learning_rate": 4.6141348497156786e-06, + "loss": 1.944, + "step": 478 + }, + { + "epoch": 0.5249315068493151, + "grad_norm": 2.1776034832000732, + "learning_rate": 4.612103980503656e-06, + "loss": 2.0342, + "step": 479 + }, + { + "epoch": 0.5260273972602739, + "grad_norm": 2.042645215988159, + "learning_rate": 4.610073111291633e-06, + "loss": 2.0124, + "step": 480 + }, + { + "epoch": 0.5260273972602739, + "eval_loss": 2.118508815765381, + "eval_runtime": 37.2764, + "eval_samples_per_second": 26.827, + "eval_steps_per_second": 1.127, + "step": 480 + }, + { + "epoch": 0.5271232876712328, + "grad_norm": 2.302661895751953, + "learning_rate": 4.60804224207961e-06, + "loss": 1.9874, + "step": 481 + }, + { + "epoch": 0.5282191780821918, + "grad_norm": 2.1251118183135986, + "learning_rate": 4.606011372867588e-06, + "loss": 1.9594, + "step": 482 + }, + { + "epoch": 0.5293150684931507, + "grad_norm": 2.157043933868408, + "learning_rate": 4.603980503655565e-06, + "loss": 1.9955, + "step": 483 + }, + { + "epoch": 0.5304109589041096, + "grad_norm": 2.291428804397583, + "learning_rate": 4.601949634443542e-06, + "loss": 1.9695, + "step": 484 + }, + { + "epoch": 0.5315068493150685, + "grad_norm": 2.2341537475585938, + "learning_rate": 4.5999187652315195e-06, + "loss": 1.899, + "step": 485 + }, + { + "epoch": 0.5326027397260275, + "grad_norm": 2.1881206035614014, + "learning_rate": 4.597887896019497e-06, + "loss": 1.9066, + "step": 486 + }, + { + "epoch": 0.5336986301369863, + "grad_norm": 2.3057658672332764, + "learning_rate": 4.595857026807474e-06, + "loss": 1.9646, + "step": 487 + }, + { + "epoch": 0.5347945205479452, + "grad_norm": 2.2318356037139893, + "learning_rate": 4.593826157595451e-06, + "loss": 1.9662, + "step": 488 + }, + { + "epoch": 0.5358904109589041, + "grad_norm": 2.170509099960327, + "learning_rate": 4.591795288383429e-06, + "loss": 1.9522, + "step": 489 + }, + { + "epoch": 0.536986301369863, + "grad_norm": 2.101646661758423, + "learning_rate": 4.589764419171406e-06, + "loss": 2.0315, + "step": 490 + }, + { + "epoch": 0.536986301369863, + "eval_loss": 2.113968849182129, + "eval_runtime": 37.2751, + "eval_samples_per_second": 26.828, + "eval_steps_per_second": 1.127, + "step": 490 + }, + { + "epoch": 0.538082191780822, + "grad_norm": 2.1563706398010254, + "learning_rate": 4.587733549959383e-06, + "loss": 1.9843, + "step": 491 + }, + { + "epoch": 0.5391780821917809, + "grad_norm": 2.295722246170044, + "learning_rate": 4.5857026807473604e-06, + "loss": 1.9499, + "step": 492 + }, + { + "epoch": 0.5402739726027397, + "grad_norm": 1.974424123764038, + "learning_rate": 4.583671811535338e-06, + "loss": 1.9346, + "step": 493 + }, + { + "epoch": 0.5413698630136986, + "grad_norm": 2.4402668476104736, + "learning_rate": 4.581640942323315e-06, + "loss": 1.9053, + "step": 494 + }, + { + "epoch": 0.5424657534246575, + "grad_norm": 2.269000768661499, + "learning_rate": 4.579610073111292e-06, + "loss": 2.0108, + "step": 495 + }, + { + "epoch": 0.5435616438356164, + "grad_norm": 2.1182594299316406, + "learning_rate": 4.5775792038992695e-06, + "loss": 2.0155, + "step": 496 + }, + { + "epoch": 0.5446575342465754, + "grad_norm": 2.2181780338287354, + "learning_rate": 4.575548334687247e-06, + "loss": 2.0112, + "step": 497 + }, + { + "epoch": 0.5457534246575343, + "grad_norm": 2.2806756496429443, + "learning_rate": 4.573517465475224e-06, + "loss": 2.0418, + "step": 498 + }, + { + "epoch": 0.5468493150684931, + "grad_norm": 2.2690632343292236, + "learning_rate": 4.571486596263201e-06, + "loss": 1.9457, + "step": 499 + }, + { + "epoch": 0.547945205479452, + "grad_norm": 2.361335515975952, + "learning_rate": 4.569455727051178e-06, + "loss": 1.9949, + "step": 500 + }, + { + "epoch": 0.547945205479452, + "eval_loss": 2.1165614128112793, + "eval_runtime": 37.2664, + "eval_samples_per_second": 26.834, + "eval_steps_per_second": 1.127, + "step": 500 + }, + { + "epoch": 0.549041095890411, + "grad_norm": 2.3833138942718506, + "learning_rate": 4.567424857839156e-06, + "loss": 1.9932, + "step": 501 + }, + { + "epoch": 0.5501369863013699, + "grad_norm": 2.328500509262085, + "learning_rate": 4.565393988627132e-06, + "loss": 1.9773, + "step": 502 + }, + { + "epoch": 0.5512328767123288, + "grad_norm": 2.4191694259643555, + "learning_rate": 4.5633631194151105e-06, + "loss": 2.0389, + "step": 503 + }, + { + "epoch": 0.5523287671232877, + "grad_norm": 2.4106884002685547, + "learning_rate": 4.561332250203087e-06, + "loss": 1.965, + "step": 504 + }, + { + "epoch": 0.5534246575342465, + "grad_norm": 2.1452605724334717, + "learning_rate": 4.559301380991065e-06, + "loss": 1.9713, + "step": 505 + }, + { + "epoch": 0.5545205479452054, + "grad_norm": 2.2753357887268066, + "learning_rate": 4.5572705117790414e-06, + "loss": 1.9116, + "step": 506 + }, + { + "epoch": 0.5556164383561644, + "grad_norm": 2.6925032138824463, + "learning_rate": 4.5552396425670196e-06, + "loss": 1.9593, + "step": 507 + }, + { + "epoch": 0.5567123287671233, + "grad_norm": 2.3136658668518066, + "learning_rate": 4.553208773354996e-06, + "loss": 2.0072, + "step": 508 + }, + { + "epoch": 0.5578082191780822, + "grad_norm": 2.254849672317505, + "learning_rate": 4.551177904142974e-06, + "loss": 1.9193, + "step": 509 + }, + { + "epoch": 0.5589041095890411, + "grad_norm": 2.352755546569824, + "learning_rate": 4.5491470349309505e-06, + "loss": 2.0228, + "step": 510 + }, + { + "epoch": 0.5589041095890411, + "eval_loss": 2.114323377609253, + "eval_runtime": 37.3629, + "eval_samples_per_second": 26.764, + "eval_steps_per_second": 1.124, + "step": 510 + }, + { + "epoch": 0.56, + "grad_norm": 2.31339955329895, + "learning_rate": 4.547116165718929e-06, + "loss": 1.9416, + "step": 511 + }, + { + "epoch": 0.5610958904109589, + "grad_norm": 2.191164970397949, + "learning_rate": 4.545085296506905e-06, + "loss": 1.9709, + "step": 512 + }, + { + "epoch": 0.5621917808219178, + "grad_norm": 2.193551778793335, + "learning_rate": 4.543054427294882e-06, + "loss": 1.9475, + "step": 513 + }, + { + "epoch": 0.5632876712328767, + "grad_norm": 2.222710371017456, + "learning_rate": 4.54102355808286e-06, + "loss": 1.9283, + "step": 514 + }, + { + "epoch": 0.5643835616438356, + "grad_norm": 2.078907012939453, + "learning_rate": 4.538992688870837e-06, + "loss": 1.929, + "step": 515 + }, + { + "epoch": 0.5654794520547946, + "grad_norm": 2.2043654918670654, + "learning_rate": 4.536961819658814e-06, + "loss": 1.9591, + "step": 516 + }, + { + "epoch": 0.5665753424657535, + "grad_norm": 2.192481517791748, + "learning_rate": 4.5349309504467915e-06, + "loss": 1.9618, + "step": 517 + }, + { + "epoch": 0.5676712328767123, + "grad_norm": 2.3509292602539062, + "learning_rate": 4.532900081234769e-06, + "loss": 1.9782, + "step": 518 + }, + { + "epoch": 0.5687671232876712, + "grad_norm": 2.2415924072265625, + "learning_rate": 4.530869212022746e-06, + "loss": 1.9456, + "step": 519 + }, + { + "epoch": 0.5698630136986301, + "grad_norm": 2.2135009765625, + "learning_rate": 4.528838342810723e-06, + "loss": 1.9671, + "step": 520 + }, + { + "epoch": 0.5698630136986301, + "eval_loss": 2.109738826751709, + "eval_runtime": 37.5253, + "eval_samples_per_second": 26.649, + "eval_steps_per_second": 1.119, + "step": 520 + }, + { + "epoch": 0.570958904109589, + "grad_norm": 2.3513574600219727, + "learning_rate": 4.526807473598701e-06, + "loss": 1.9625, + "step": 521 + }, + { + "epoch": 0.572054794520548, + "grad_norm": 1.9949424266815186, + "learning_rate": 4.524776604386678e-06, + "loss": 1.9621, + "step": 522 + }, + { + "epoch": 0.5731506849315069, + "grad_norm": 2.199488401412964, + "learning_rate": 4.522745735174655e-06, + "loss": 1.9389, + "step": 523 + }, + { + "epoch": 0.5742465753424657, + "grad_norm": 2.2119452953338623, + "learning_rate": 4.520714865962632e-06, + "loss": 1.9936, + "step": 524 + }, + { + "epoch": 0.5753424657534246, + "grad_norm": 2.242812156677246, + "learning_rate": 4.51868399675061e-06, + "loss": 1.9728, + "step": 525 + }, + { + "epoch": 0.5764383561643835, + "grad_norm": 2.1713850498199463, + "learning_rate": 4.516653127538587e-06, + "loss": 1.98, + "step": 526 + }, + { + "epoch": 0.5775342465753425, + "grad_norm": 2.261742353439331, + "learning_rate": 4.514622258326564e-06, + "loss": 1.9412, + "step": 527 + }, + { + "epoch": 0.5786301369863014, + "grad_norm": 2.222663402557373, + "learning_rate": 4.5125913891145415e-06, + "loss": 1.9266, + "step": 528 + }, + { + "epoch": 0.5797260273972603, + "grad_norm": 2.1108906269073486, + "learning_rate": 4.510560519902519e-06, + "loss": 1.9539, + "step": 529 + }, + { + "epoch": 0.5808219178082191, + "grad_norm": 1.9594205617904663, + "learning_rate": 4.508529650690496e-06, + "loss": 1.9842, + "step": 530 + }, + { + "epoch": 0.5808219178082191, + "eval_loss": 2.1073970794677734, + "eval_runtime": 37.6167, + "eval_samples_per_second": 26.584, + "eval_steps_per_second": 1.117, + "step": 530 + }, + { + "epoch": 0.581917808219178, + "grad_norm": 2.1339495182037354, + "learning_rate": 4.506498781478473e-06, + "loss": 1.9095, + "step": 531 + }, + { + "epoch": 0.583013698630137, + "grad_norm": 2.116248607635498, + "learning_rate": 4.504467912266451e-06, + "loss": 1.9091, + "step": 532 + }, + { + "epoch": 0.5841095890410959, + "grad_norm": 2.063962459564209, + "learning_rate": 4.502437043054428e-06, + "loss": 1.9537, + "step": 533 + }, + { + "epoch": 0.5852054794520548, + "grad_norm": 2.162569522857666, + "learning_rate": 4.500406173842405e-06, + "loss": 2.0253, + "step": 534 + }, + { + "epoch": 0.5863013698630137, + "grad_norm": 2.1783549785614014, + "learning_rate": 4.4983753046303824e-06, + "loss": 1.9714, + "step": 535 + }, + { + "epoch": 0.5873972602739727, + "grad_norm": 2.0791046619415283, + "learning_rate": 4.49634443541836e-06, + "loss": 1.9629, + "step": 536 + }, + { + "epoch": 0.5884931506849315, + "grad_norm": 2.3923251628875732, + "learning_rate": 4.494313566206336e-06, + "loss": 1.9445, + "step": 537 + }, + { + "epoch": 0.5895890410958904, + "grad_norm": 2.1848347187042236, + "learning_rate": 4.492282696994314e-06, + "loss": 2.007, + "step": 538 + }, + { + "epoch": 0.5906849315068493, + "grad_norm": 2.0824925899505615, + "learning_rate": 4.490251827782291e-06, + "loss": 1.9409, + "step": 539 + }, + { + "epoch": 0.5917808219178082, + "grad_norm": 2.057790756225586, + "learning_rate": 4.488220958570269e-06, + "loss": 1.9061, + "step": 540 + }, + { + "epoch": 0.5917808219178082, + "eval_loss": 2.1101415157318115, + "eval_runtime": 37.2235, + "eval_samples_per_second": 26.865, + "eval_steps_per_second": 1.128, + "step": 540 + }, + { + "epoch": 0.5928767123287672, + "grad_norm": 2.3652148246765137, + "learning_rate": 4.486190089358245e-06, + "loss": 1.9379, + "step": 541 + }, + { + "epoch": 0.5939726027397261, + "grad_norm": 2.195535659790039, + "learning_rate": 4.484159220146223e-06, + "loss": 1.8881, + "step": 542 + }, + { + "epoch": 0.5950684931506849, + "grad_norm": 2.258000135421753, + "learning_rate": 4.4821283509342e-06, + "loss": 1.9923, + "step": 543 + }, + { + "epoch": 0.5961643835616438, + "grad_norm": 2.097041606903076, + "learning_rate": 4.480097481722178e-06, + "loss": 1.9715, + "step": 544 + }, + { + "epoch": 0.5972602739726027, + "grad_norm": 2.1741156578063965, + "learning_rate": 4.478066612510154e-06, + "loss": 1.9487, + "step": 545 + }, + { + "epoch": 0.5983561643835617, + "grad_norm": 2.135338306427002, + "learning_rate": 4.4760357432981325e-06, + "loss": 1.9172, + "step": 546 + }, + { + "epoch": 0.5994520547945206, + "grad_norm": 2.3435418605804443, + "learning_rate": 4.474004874086109e-06, + "loss": 1.945, + "step": 547 + }, + { + "epoch": 0.6005479452054795, + "grad_norm": 2.2242488861083984, + "learning_rate": 4.471974004874087e-06, + "loss": 2.0717, + "step": 548 + }, + { + "epoch": 0.6016438356164383, + "grad_norm": 2.3371622562408447, + "learning_rate": 4.4699431356620635e-06, + "loss": 1.9082, + "step": 549 + }, + { + "epoch": 0.6027397260273972, + "grad_norm": 2.176064968109131, + "learning_rate": 4.467912266450041e-06, + "loss": 2.0154, + "step": 550 + }, + { + "epoch": 0.6027397260273972, + "eval_loss": 2.105236291885376, + "eval_runtime": 37.1885, + "eval_samples_per_second": 26.89, + "eval_steps_per_second": 1.129, + "step": 550 + }, + { + "epoch": 0.6038356164383561, + "grad_norm": 2.131166934967041, + "learning_rate": 4.465881397238018e-06, + "loss": 1.9575, + "step": 551 + }, + { + "epoch": 0.6049315068493151, + "grad_norm": 2.094514846801758, + "learning_rate": 4.463850528025995e-06, + "loss": 1.9544, + "step": 552 + }, + { + "epoch": 0.606027397260274, + "grad_norm": 2.0827765464782715, + "learning_rate": 4.4618196588139726e-06, + "loss": 1.9539, + "step": 553 + }, + { + "epoch": 0.6071232876712329, + "grad_norm": 2.225749969482422, + "learning_rate": 4.45978878960195e-06, + "loss": 1.983, + "step": 554 + }, + { + "epoch": 0.6082191780821918, + "grad_norm": 2.1363680362701416, + "learning_rate": 4.457757920389927e-06, + "loss": 1.9769, + "step": 555 + }, + { + "epoch": 0.6093150684931506, + "grad_norm": 2.238170623779297, + "learning_rate": 4.455727051177904e-06, + "loss": 1.9094, + "step": 556 + }, + { + "epoch": 0.6104109589041096, + "grad_norm": 2.228384494781494, + "learning_rate": 4.453696181965882e-06, + "loss": 1.9464, + "step": 557 + }, + { + "epoch": 0.6115068493150685, + "grad_norm": 2.479428768157959, + "learning_rate": 4.451665312753859e-06, + "loss": 2.0017, + "step": 558 + }, + { + "epoch": 0.6126027397260274, + "grad_norm": 2.2331902980804443, + "learning_rate": 4.449634443541836e-06, + "loss": 1.9728, + "step": 559 + }, + { + "epoch": 0.6136986301369863, + "grad_norm": 2.1728556156158447, + "learning_rate": 4.4476035743298135e-06, + "loss": 2.0153, + "step": 560 + }, + { + "epoch": 0.6136986301369863, + "eval_loss": 2.0978519916534424, + "eval_runtime": 37.233, + "eval_samples_per_second": 26.858, + "eval_steps_per_second": 1.128, + "step": 560 + }, + { + "epoch": 0.6147945205479453, + "grad_norm": 2.377814292907715, + "learning_rate": 4.445572705117791e-06, + "loss": 1.9242, + "step": 561 + }, + { + "epoch": 0.6158904109589041, + "grad_norm": 2.190436601638794, + "learning_rate": 4.443541835905768e-06, + "loss": 1.9454, + "step": 562 + }, + { + "epoch": 0.616986301369863, + "grad_norm": 2.052201986312866, + "learning_rate": 4.441510966693745e-06, + "loss": 2.0048, + "step": 563 + }, + { + "epoch": 0.6180821917808219, + "grad_norm": 2.2706589698791504, + "learning_rate": 4.439480097481723e-06, + "loss": 1.9426, + "step": 564 + }, + { + "epoch": 0.6191780821917808, + "grad_norm": 2.3478105068206787, + "learning_rate": 4.4374492282697e-06, + "loss": 2.0047, + "step": 565 + }, + { + "epoch": 0.6202739726027398, + "grad_norm": 2.465078115463257, + "learning_rate": 4.435418359057677e-06, + "loss": 1.8739, + "step": 566 + }, + { + "epoch": 0.6213698630136987, + "grad_norm": 2.0979490280151367, + "learning_rate": 4.433387489845654e-06, + "loss": 1.9588, + "step": 567 + }, + { + "epoch": 0.6224657534246575, + "grad_norm": 2.1239588260650635, + "learning_rate": 4.431356620633632e-06, + "loss": 1.9695, + "step": 568 + }, + { + "epoch": 0.6235616438356164, + "grad_norm": 2.014007806777954, + "learning_rate": 4.429325751421609e-06, + "loss": 1.9232, + "step": 569 + }, + { + "epoch": 0.6246575342465753, + "grad_norm": 2.212414026260376, + "learning_rate": 4.427294882209586e-06, + "loss": 1.928, + "step": 570 + }, + { + "epoch": 0.6246575342465753, + "eval_loss": 2.103149890899658, + "eval_runtime": 37.2641, + "eval_samples_per_second": 26.835, + "eval_steps_per_second": 1.127, + "step": 570 + }, + { + "epoch": 0.6257534246575343, + "grad_norm": 2.0944411754608154, + "learning_rate": 4.4252640129975635e-06, + "loss": 1.9568, + "step": 571 + }, + { + "epoch": 0.6268493150684932, + "grad_norm": 2.1068055629730225, + "learning_rate": 4.423233143785541e-06, + "loss": 1.8917, + "step": 572 + }, + { + "epoch": 0.6279452054794521, + "grad_norm": 2.1992151737213135, + "learning_rate": 4.421202274573518e-06, + "loss": 1.9536, + "step": 573 + }, + { + "epoch": 0.6290410958904109, + "grad_norm": 2.2079834938049316, + "learning_rate": 4.4191714053614945e-06, + "loss": 1.9603, + "step": 574 + }, + { + "epoch": 0.6301369863013698, + "grad_norm": 2.1230316162109375, + "learning_rate": 4.417140536149473e-06, + "loss": 1.8529, + "step": 575 + }, + { + "epoch": 0.6312328767123287, + "grad_norm": 2.1069579124450684, + "learning_rate": 4.415109666937449e-06, + "loss": 1.8997, + "step": 576 + }, + { + "epoch": 0.6323287671232877, + "grad_norm": 2.1683409214019775, + "learning_rate": 4.413078797725427e-06, + "loss": 1.971, + "step": 577 + }, + { + "epoch": 0.6334246575342466, + "grad_norm": 2.0599024295806885, + "learning_rate": 4.411047928513404e-06, + "loss": 1.9154, + "step": 578 + }, + { + "epoch": 0.6345205479452055, + "grad_norm": 2.1605849266052246, + "learning_rate": 4.409017059301382e-06, + "loss": 1.9562, + "step": 579 + }, + { + "epoch": 0.6356164383561644, + "grad_norm": 2.151249408721924, + "learning_rate": 4.406986190089358e-06, + "loss": 1.9571, + "step": 580 + }, + { + "epoch": 0.6356164383561644, + "eval_loss": 2.098177671432495, + "eval_runtime": 37.2578, + "eval_samples_per_second": 26.84, + "eval_steps_per_second": 1.127, + "step": 580 + }, + { + "epoch": 0.6367123287671232, + "grad_norm": 2.1819236278533936, + "learning_rate": 4.404955320877336e-06, + "loss": 1.9816, + "step": 581 + }, + { + "epoch": 0.6378082191780822, + "grad_norm": 2.1131205558776855, + "learning_rate": 4.402924451665313e-06, + "loss": 1.9738, + "step": 582 + }, + { + "epoch": 0.6389041095890411, + "grad_norm": 2.0852456092834473, + "learning_rate": 4.400893582453291e-06, + "loss": 1.924, + "step": 583 + }, + { + "epoch": 0.64, + "grad_norm": 2.0997941493988037, + "learning_rate": 4.398862713241267e-06, + "loss": 1.9067, + "step": 584 + }, + { + "epoch": 0.6410958904109589, + "grad_norm": 2.0848538875579834, + "learning_rate": 4.396831844029245e-06, + "loss": 1.9393, + "step": 585 + }, + { + "epoch": 0.6421917808219179, + "grad_norm": 2.2121758460998535, + "learning_rate": 4.394800974817222e-06, + "loss": 1.9458, + "step": 586 + }, + { + "epoch": 0.6432876712328767, + "grad_norm": 2.194462299346924, + "learning_rate": 4.392770105605199e-06, + "loss": 1.9528, + "step": 587 + }, + { + "epoch": 0.6443835616438356, + "grad_norm": 2.1936254501342773, + "learning_rate": 4.390739236393176e-06, + "loss": 1.9584, + "step": 588 + }, + { + "epoch": 0.6454794520547945, + "grad_norm": 2.1685783863067627, + "learning_rate": 4.388708367181154e-06, + "loss": 1.9666, + "step": 589 + }, + { + "epoch": 0.6465753424657534, + "grad_norm": 2.334571123123169, + "learning_rate": 4.386677497969131e-06, + "loss": 1.9748, + "step": 590 + }, + { + "epoch": 0.6465753424657534, + "eval_loss": 2.0937106609344482, + "eval_runtime": 37.2877, + "eval_samples_per_second": 26.818, + "eval_steps_per_second": 1.126, + "step": 590 + }, + { + "epoch": 0.6476712328767124, + "grad_norm": 2.2901110649108887, + "learning_rate": 4.384646628757108e-06, + "loss": 1.9597, + "step": 591 + }, + { + "epoch": 0.6487671232876713, + "grad_norm": 2.2364838123321533, + "learning_rate": 4.3826157595450855e-06, + "loss": 2.004, + "step": 592 + }, + { + "epoch": 0.6498630136986301, + "grad_norm": 2.1088364124298096, + "learning_rate": 4.380584890333063e-06, + "loss": 1.9319, + "step": 593 + }, + { + "epoch": 0.650958904109589, + "grad_norm": 2.164210319519043, + "learning_rate": 4.37855402112104e-06, + "loss": 1.8906, + "step": 594 + }, + { + "epoch": 0.6520547945205479, + "grad_norm": 2.205200672149658, + "learning_rate": 4.376523151909017e-06, + "loss": 1.8474, + "step": 595 + }, + { + "epoch": 0.6531506849315069, + "grad_norm": 2.239530086517334, + "learning_rate": 4.3744922826969946e-06, + "loss": 2.0033, + "step": 596 + }, + { + "epoch": 0.6542465753424658, + "grad_norm": 2.214747190475464, + "learning_rate": 4.372461413484972e-06, + "loss": 1.9258, + "step": 597 + }, + { + "epoch": 0.6553424657534247, + "grad_norm": 2.1746792793273926, + "learning_rate": 4.370430544272949e-06, + "loss": 1.9479, + "step": 598 + }, + { + "epoch": 0.6564383561643835, + "grad_norm": 2.194131851196289, + "learning_rate": 4.368399675060926e-06, + "loss": 1.9825, + "step": 599 + }, + { + "epoch": 0.6575342465753424, + "grad_norm": 2.3334717750549316, + "learning_rate": 4.366368805848904e-06, + "loss": 1.9831, + "step": 600 + }, + { + "epoch": 0.6575342465753424, + "eval_loss": 2.0881216526031494, + "eval_runtime": 37.3226, + "eval_samples_per_second": 26.793, + "eval_steps_per_second": 1.125, + "step": 600 + }, + { + "epoch": 0.6586301369863014, + "grad_norm": 2.269484758377075, + "learning_rate": 4.364337936636881e-06, + "loss": 1.9407, + "step": 601 + }, + { + "epoch": 0.6597260273972603, + "grad_norm": 2.470773935317993, + "learning_rate": 4.362307067424858e-06, + "loss": 1.8905, + "step": 602 + }, + { + "epoch": 0.6608219178082192, + "grad_norm": 2.485574960708618, + "learning_rate": 4.3602761982128355e-06, + "loss": 1.953, + "step": 603 + }, + { + "epoch": 0.6619178082191781, + "grad_norm": 2.2443618774414062, + "learning_rate": 4.358245329000813e-06, + "loss": 1.9244, + "step": 604 + }, + { + "epoch": 0.663013698630137, + "grad_norm": 2.039600372314453, + "learning_rate": 4.35621445978879e-06, + "loss": 1.9364, + "step": 605 + }, + { + "epoch": 0.6641095890410958, + "grad_norm": 2.093001127243042, + "learning_rate": 4.354183590576767e-06, + "loss": 1.9439, + "step": 606 + }, + { + "epoch": 0.6652054794520548, + "grad_norm": 2.3433260917663574, + "learning_rate": 4.352152721364745e-06, + "loss": 1.9322, + "step": 607 + }, + { + "epoch": 0.6663013698630137, + "grad_norm": 2.2403812408447266, + "learning_rate": 4.350121852152722e-06, + "loss": 1.8829, + "step": 608 + }, + { + "epoch": 0.6673972602739726, + "grad_norm": 2.1542811393737793, + "learning_rate": 4.348090982940699e-06, + "loss": 1.9549, + "step": 609 + }, + { + "epoch": 0.6684931506849315, + "grad_norm": 2.1526548862457275, + "learning_rate": 4.3460601137286764e-06, + "loss": 1.9433, + "step": 610 + }, + { + "epoch": 0.6684931506849315, + "eval_loss": 2.083638906478882, + "eval_runtime": 37.2677, + "eval_samples_per_second": 26.833, + "eval_steps_per_second": 1.127, + "step": 610 + }, + { + "epoch": 0.6695890410958905, + "grad_norm": 2.093421459197998, + "learning_rate": 4.344029244516654e-06, + "loss": 1.9643, + "step": 611 + }, + { + "epoch": 0.6706849315068493, + "grad_norm": 2.2205088138580322, + "learning_rate": 4.341998375304631e-06, + "loss": 1.9309, + "step": 612 + }, + { + "epoch": 0.6717808219178082, + "grad_norm": 2.059763193130493, + "learning_rate": 4.339967506092607e-06, + "loss": 2.0064, + "step": 613 + }, + { + "epoch": 0.6728767123287671, + "grad_norm": 2.0340707302093506, + "learning_rate": 4.3379366368805855e-06, + "loss": 1.9394, + "step": 614 + }, + { + "epoch": 0.673972602739726, + "grad_norm": 2.0716137886047363, + "learning_rate": 4.335905767668562e-06, + "loss": 1.9421, + "step": 615 + }, + { + "epoch": 0.675068493150685, + "grad_norm": 2.176809549331665, + "learning_rate": 4.33387489845654e-06, + "loss": 1.9261, + "step": 616 + }, + { + "epoch": 0.6761643835616439, + "grad_norm": 2.144385814666748, + "learning_rate": 4.3318440292445165e-06, + "loss": 1.9635, + "step": 617 + }, + { + "epoch": 0.6772602739726027, + "grad_norm": 2.0949580669403076, + "learning_rate": 4.329813160032495e-06, + "loss": 1.9417, + "step": 618 + }, + { + "epoch": 0.6783561643835616, + "grad_norm": 2.1727259159088135, + "learning_rate": 4.327782290820471e-06, + "loss": 1.9041, + "step": 619 + }, + { + "epoch": 0.6794520547945205, + "grad_norm": 2.012868881225586, + "learning_rate": 4.325751421608449e-06, + "loss": 1.8973, + "step": 620 + }, + { + "epoch": 0.6794520547945205, + "eval_loss": 2.0839755535125732, + "eval_runtime": 37.5302, + "eval_samples_per_second": 26.645, + "eval_steps_per_second": 1.119, + "step": 620 + }, + { + "epoch": 0.6805479452054795, + "grad_norm": 2.204153299331665, + "learning_rate": 4.323720552396426e-06, + "loss": 1.9719, + "step": 621 + }, + { + "epoch": 0.6816438356164384, + "grad_norm": 2.2421770095825195, + "learning_rate": 4.321689683184404e-06, + "loss": 1.9207, + "step": 622 + }, + { + "epoch": 0.6827397260273973, + "grad_norm": 2.2263410091400146, + "learning_rate": 4.31965881397238e-06, + "loss": 1.9146, + "step": 623 + }, + { + "epoch": 0.6838356164383562, + "grad_norm": 2.2015221118927, + "learning_rate": 4.3176279447603574e-06, + "loss": 1.9311, + "step": 624 + }, + { + "epoch": 0.684931506849315, + "grad_norm": 2.1459643840789795, + "learning_rate": 4.315597075548335e-06, + "loss": 1.9673, + "step": 625 + }, + { + "epoch": 0.686027397260274, + "grad_norm": 2.278928756713867, + "learning_rate": 4.313566206336312e-06, + "loss": 1.954, + "step": 626 + }, + { + "epoch": 0.6871232876712329, + "grad_norm": 2.05305552482605, + "learning_rate": 4.311535337124289e-06, + "loss": 1.9789, + "step": 627 + }, + { + "epoch": 0.6882191780821918, + "grad_norm": 2.3979289531707764, + "learning_rate": 4.3095044679122665e-06, + "loss": 1.926, + "step": 628 + }, + { + "epoch": 0.6893150684931507, + "grad_norm": 2.1635828018188477, + "learning_rate": 4.307473598700244e-06, + "loss": 1.9316, + "step": 629 + }, + { + "epoch": 0.6904109589041096, + "grad_norm": 2.237713098526001, + "learning_rate": 4.305442729488221e-06, + "loss": 1.9409, + "step": 630 + }, + { + "epoch": 0.6904109589041096, + "eval_loss": 2.084226131439209, + "eval_runtime": 37.6039, + "eval_samples_per_second": 26.593, + "eval_steps_per_second": 1.117, + "step": 630 + }, + { + "epoch": 0.6915068493150685, + "grad_norm": 2.157435894012451, + "learning_rate": 4.303411860276198e-06, + "loss": 1.8777, + "step": 631 + }, + { + "epoch": 0.6926027397260274, + "grad_norm": 2.20766019821167, + "learning_rate": 4.301380991064176e-06, + "loss": 1.9803, + "step": 632 + }, + { + "epoch": 0.6936986301369863, + "grad_norm": 2.173079490661621, + "learning_rate": 4.299350121852153e-06, + "loss": 1.9541, + "step": 633 + }, + { + "epoch": 0.6947945205479452, + "grad_norm": 1.9727106094360352, + "learning_rate": 4.29731925264013e-06, + "loss": 1.9766, + "step": 634 + }, + { + "epoch": 0.6958904109589041, + "grad_norm": 2.268538475036621, + "learning_rate": 4.2952883834281075e-06, + "loss": 1.9737, + "step": 635 + }, + { + "epoch": 0.6969863013698631, + "grad_norm": 2.2118406295776367, + "learning_rate": 4.293257514216085e-06, + "loss": 1.9291, + "step": 636 + }, + { + "epoch": 0.6980821917808219, + "grad_norm": 2.2939369678497314, + "learning_rate": 4.291226645004062e-06, + "loss": 1.8868, + "step": 637 + }, + { + "epoch": 0.6991780821917808, + "grad_norm": 2.4940083026885986, + "learning_rate": 4.289195775792039e-06, + "loss": 1.9694, + "step": 638 + }, + { + "epoch": 0.7002739726027397, + "grad_norm": 2.1988351345062256, + "learning_rate": 4.287164906580017e-06, + "loss": 1.9781, + "step": 639 + }, + { + "epoch": 0.7013698630136986, + "grad_norm": 2.099578619003296, + "learning_rate": 4.285134037367994e-06, + "loss": 1.9145, + "step": 640 + }, + { + "epoch": 0.7013698630136986, + "eval_loss": 2.07913875579834, + "eval_runtime": 37.3458, + "eval_samples_per_second": 26.777, + "eval_steps_per_second": 1.125, + "step": 640 + }, + { + "epoch": 0.7024657534246576, + "grad_norm": 2.2437307834625244, + "learning_rate": 4.283103168155971e-06, + "loss": 1.9075, + "step": 641 + }, + { + "epoch": 0.7035616438356165, + "grad_norm": 2.320880174636841, + "learning_rate": 4.281072298943948e-06, + "loss": 1.8929, + "step": 642 + }, + { + "epoch": 0.7046575342465753, + "grad_norm": 2.1174306869506836, + "learning_rate": 4.279041429731926e-06, + "loss": 1.8872, + "step": 643 + }, + { + "epoch": 0.7057534246575342, + "grad_norm": 2.2826361656188965, + "learning_rate": 4.277010560519903e-06, + "loss": 1.9287, + "step": 644 + }, + { + "epoch": 0.7068493150684931, + "grad_norm": 2.7543559074401855, + "learning_rate": 4.27497969130788e-06, + "loss": 1.9119, + "step": 645 + }, + { + "epoch": 0.707945205479452, + "grad_norm": 2.2582201957702637, + "learning_rate": 4.2729488220958575e-06, + "loss": 1.9118, + "step": 646 + }, + { + "epoch": 0.709041095890411, + "grad_norm": 2.3121371269226074, + "learning_rate": 4.270917952883835e-06, + "loss": 1.959, + "step": 647 + }, + { + "epoch": 0.7101369863013699, + "grad_norm": 2.419102430343628, + "learning_rate": 4.268887083671812e-06, + "loss": 1.9679, + "step": 648 + }, + { + "epoch": 0.7112328767123288, + "grad_norm": 2.285771369934082, + "learning_rate": 4.266856214459789e-06, + "loss": 1.8939, + "step": 649 + }, + { + "epoch": 0.7123287671232876, + "grad_norm": 2.193603754043579, + "learning_rate": 4.264825345247766e-06, + "loss": 2.0097, + "step": 650 + }, + { + "epoch": 0.7123287671232876, + "eval_loss": 2.0785164833068848, + "eval_runtime": 37.2595, + "eval_samples_per_second": 26.839, + "eval_steps_per_second": 1.127, + "step": 650 + }, + { + "epoch": 0.7134246575342466, + "grad_norm": 2.2645511627197266, + "learning_rate": 4.262794476035744e-06, + "loss": 1.8865, + "step": 651 + }, + { + "epoch": 0.7145205479452055, + "grad_norm": 2.5553910732269287, + "learning_rate": 4.26076360682372e-06, + "loss": 1.9331, + "step": 652 + }, + { + "epoch": 0.7156164383561644, + "grad_norm": 2.2770845890045166, + "learning_rate": 4.2587327376116984e-06, + "loss": 1.847, + "step": 653 + }, + { + "epoch": 0.7167123287671233, + "grad_norm": 2.1486358642578125, + "learning_rate": 4.256701868399675e-06, + "loss": 1.9058, + "step": 654 + }, + { + "epoch": 0.7178082191780822, + "grad_norm": 2.168975353240967, + "learning_rate": 4.254670999187653e-06, + "loss": 1.9408, + "step": 655 + }, + { + "epoch": 0.718904109589041, + "grad_norm": 2.2156167030334473, + "learning_rate": 4.252640129975629e-06, + "loss": 1.932, + "step": 656 + }, + { + "epoch": 0.72, + "grad_norm": 2.148214101791382, + "learning_rate": 4.2506092607636075e-06, + "loss": 1.9327, + "step": 657 + }, + { + "epoch": 0.7210958904109589, + "grad_norm": 2.3860998153686523, + "learning_rate": 4.248578391551584e-06, + "loss": 1.9191, + "step": 658 + }, + { + "epoch": 0.7221917808219178, + "grad_norm": 2.3917558193206787, + "learning_rate": 4.246547522339562e-06, + "loss": 1.8607, + "step": 659 + }, + { + "epoch": 0.7232876712328767, + "grad_norm": 2.3024795055389404, + "learning_rate": 4.2445166531275385e-06, + "loss": 1.9083, + "step": 660 + }, + { + "epoch": 0.7232876712328767, + "eval_loss": 2.07985782623291, + "eval_runtime": 37.1583, + "eval_samples_per_second": 26.912, + "eval_steps_per_second": 1.13, + "step": 660 + }, + { + "epoch": 0.7243835616438357, + "grad_norm": 2.261615037918091, + "learning_rate": 4.242485783915517e-06, + "loss": 1.9155, + "step": 661 + }, + { + "epoch": 0.7254794520547945, + "grad_norm": 2.190124034881592, + "learning_rate": 4.240454914703493e-06, + "loss": 1.858, + "step": 662 + }, + { + "epoch": 0.7265753424657534, + "grad_norm": 2.3396036624908447, + "learning_rate": 4.23842404549147e-06, + "loss": 1.878, + "step": 663 + }, + { + "epoch": 0.7276712328767123, + "grad_norm": 2.198410987854004, + "learning_rate": 4.236393176279448e-06, + "loss": 1.8755, + "step": 664 + }, + { + "epoch": 0.7287671232876712, + "grad_norm": 2.108875274658203, + "learning_rate": 4.234362307067425e-06, + "loss": 1.8709, + "step": 665 + }, + { + "epoch": 0.7298630136986302, + "grad_norm": 2.0908193588256836, + "learning_rate": 4.232331437855402e-06, + "loss": 1.9395, + "step": 666 + }, + { + "epoch": 0.7309589041095891, + "grad_norm": 2.208134651184082, + "learning_rate": 4.2303005686433795e-06, + "loss": 1.9675, + "step": 667 + }, + { + "epoch": 0.7320547945205479, + "grad_norm": 2.2779526710510254, + "learning_rate": 4.228269699431357e-06, + "loss": 1.9661, + "step": 668 + }, + { + "epoch": 0.7331506849315068, + "grad_norm": 2.053279161453247, + "learning_rate": 4.226238830219334e-06, + "loss": 1.9244, + "step": 669 + }, + { + "epoch": 0.7342465753424657, + "grad_norm": 2.2060821056365967, + "learning_rate": 4.224207961007311e-06, + "loss": 1.9252, + "step": 670 + }, + { + "epoch": 0.7342465753424657, + "eval_loss": 2.0804648399353027, + "eval_runtime": 37.1852, + "eval_samples_per_second": 26.892, + "eval_steps_per_second": 1.129, + "step": 670 + }, + { + "epoch": 0.7353424657534247, + "grad_norm": 2.1996090412139893, + "learning_rate": 4.2221770917952886e-06, + "loss": 1.926, + "step": 671 + }, + { + "epoch": 0.7364383561643836, + "grad_norm": 2.0817034244537354, + "learning_rate": 4.220146222583266e-06, + "loss": 1.9123, + "step": 672 + }, + { + "epoch": 0.7375342465753425, + "grad_norm": 2.2505035400390625, + "learning_rate": 4.218115353371243e-06, + "loss": 1.9369, + "step": 673 + }, + { + "epoch": 0.7386301369863014, + "grad_norm": 2.2053000926971436, + "learning_rate": 4.21608448415922e-06, + "loss": 1.9225, + "step": 674 + }, + { + "epoch": 0.7397260273972602, + "grad_norm": 2.039905071258545, + "learning_rate": 4.214053614947198e-06, + "loss": 1.8876, + "step": 675 + }, + { + "epoch": 0.7408219178082192, + "grad_norm": 2.047445297241211, + "learning_rate": 4.212022745735175e-06, + "loss": 1.9724, + "step": 676 + }, + { + "epoch": 0.7419178082191781, + "grad_norm": 2.277803897857666, + "learning_rate": 4.209991876523152e-06, + "loss": 1.8687, + "step": 677 + }, + { + "epoch": 0.743013698630137, + "grad_norm": 2.0048105716705322, + "learning_rate": 4.2079610073111295e-06, + "loss": 1.9349, + "step": 678 + }, + { + "epoch": 0.7441095890410959, + "grad_norm": 2.121614456176758, + "learning_rate": 4.205930138099107e-06, + "loss": 1.8723, + "step": 679 + }, + { + "epoch": 0.7452054794520548, + "grad_norm": 2.169165849685669, + "learning_rate": 4.203899268887084e-06, + "loss": 1.9713, + "step": 680 + }, + { + "epoch": 0.7452054794520548, + "eval_loss": 2.0704824924468994, + "eval_runtime": 37.2015, + "eval_samples_per_second": 26.881, + "eval_steps_per_second": 1.129, + "step": 680 + }, + { + "epoch": 0.7463013698630137, + "grad_norm": 2.004974603652954, + "learning_rate": 4.201868399675061e-06, + "loss": 1.9237, + "step": 681 + }, + { + "epoch": 0.7473972602739726, + "grad_norm": 2.042402505874634, + "learning_rate": 4.199837530463039e-06, + "loss": 1.9686, + "step": 682 + }, + { + "epoch": 0.7484931506849315, + "grad_norm": 2.152040719985962, + "learning_rate": 4.197806661251016e-06, + "loss": 1.8498, + "step": 683 + }, + { + "epoch": 0.7495890410958904, + "grad_norm": 2.116483211517334, + "learning_rate": 4.195775792038993e-06, + "loss": 1.9241, + "step": 684 + }, + { + "epoch": 0.7506849315068493, + "grad_norm": 2.1628384590148926, + "learning_rate": 4.19374492282697e-06, + "loss": 1.9019, + "step": 685 + }, + { + "epoch": 0.7517808219178083, + "grad_norm": 2.1851415634155273, + "learning_rate": 4.191714053614948e-06, + "loss": 1.836, + "step": 686 + }, + { + "epoch": 0.7528767123287671, + "grad_norm": 2.173215866088867, + "learning_rate": 4.189683184402924e-06, + "loss": 1.9412, + "step": 687 + }, + { + "epoch": 0.753972602739726, + "grad_norm": 2.1002466678619385, + "learning_rate": 4.187652315190902e-06, + "loss": 1.8871, + "step": 688 + }, + { + "epoch": 0.7550684931506849, + "grad_norm": 2.1932122707366943, + "learning_rate": 4.185621445978879e-06, + "loss": 1.9671, + "step": 689 + }, + { + "epoch": 0.7561643835616438, + "grad_norm": 2.2915258407592773, + "learning_rate": 4.183590576766857e-06, + "loss": 1.8957, + "step": 690 + }, + { + "epoch": 0.7561643835616438, + "eval_loss": 2.067124128341675, + "eval_runtime": 37.2362, + "eval_samples_per_second": 26.856, + "eval_steps_per_second": 1.128, + "step": 690 + }, + { + "epoch": 0.7572602739726028, + "grad_norm": 2.060879707336426, + "learning_rate": 4.181559707554833e-06, + "loss": 1.9091, + "step": 691 + }, + { + "epoch": 0.7583561643835617, + "grad_norm": 2.031156539916992, + "learning_rate": 4.179528838342811e-06, + "loss": 1.9606, + "step": 692 + }, + { + "epoch": 0.7594520547945205, + "grad_norm": 2.121628999710083, + "learning_rate": 4.177497969130788e-06, + "loss": 1.88, + "step": 693 + }, + { + "epoch": 0.7605479452054794, + "grad_norm": 2.2009618282318115, + "learning_rate": 4.175467099918766e-06, + "loss": 1.8916, + "step": 694 + }, + { + "epoch": 0.7616438356164383, + "grad_norm": 2.124082326889038, + "learning_rate": 4.173436230706742e-06, + "loss": 1.9805, + "step": 695 + }, + { + "epoch": 0.7627397260273973, + "grad_norm": 2.3914906978607178, + "learning_rate": 4.1714053614947205e-06, + "loss": 1.9052, + "step": 696 + }, + { + "epoch": 0.7638356164383562, + "grad_norm": 2.2398853302001953, + "learning_rate": 4.169374492282697e-06, + "loss": 1.8965, + "step": 697 + }, + { + "epoch": 0.7649315068493151, + "grad_norm": 2.224860191345215, + "learning_rate": 4.167343623070675e-06, + "loss": 1.8962, + "step": 698 + }, + { + "epoch": 0.766027397260274, + "grad_norm": 2.3356244564056396, + "learning_rate": 4.165312753858652e-06, + "loss": 1.9543, + "step": 699 + }, + { + "epoch": 0.7671232876712328, + "grad_norm": 2.1308751106262207, + "learning_rate": 4.163281884646629e-06, + "loss": 1.9095, + "step": 700 + }, + { + "epoch": 0.7671232876712328, + "eval_loss": 2.066145896911621, + "eval_runtime": 37.2629, + "eval_samples_per_second": 26.836, + "eval_steps_per_second": 1.127, + "step": 700 + }, + { + "epoch": 0.7682191780821918, + "grad_norm": 2.1655938625335693, + "learning_rate": 4.161251015434607e-06, + "loss": 1.8923, + "step": 701 + }, + { + "epoch": 0.7693150684931507, + "grad_norm": 2.1985316276550293, + "learning_rate": 4.159220146222583e-06, + "loss": 1.9216, + "step": 702 + }, + { + "epoch": 0.7704109589041096, + "grad_norm": 2.1537251472473145, + "learning_rate": 4.157189277010561e-06, + "loss": 1.879, + "step": 703 + }, + { + "epoch": 0.7715068493150685, + "grad_norm": 2.1867849826812744, + "learning_rate": 4.155158407798538e-06, + "loss": 1.905, + "step": 704 + }, + { + "epoch": 0.7726027397260274, + "grad_norm": 2.342320203781128, + "learning_rate": 4.153127538586516e-06, + "loss": 1.9403, + "step": 705 + }, + { + "epoch": 0.7736986301369863, + "grad_norm": 2.2250113487243652, + "learning_rate": 4.151096669374492e-06, + "loss": 1.9109, + "step": 706 + }, + { + "epoch": 0.7747945205479452, + "grad_norm": 2.179103374481201, + "learning_rate": 4.1490658001624705e-06, + "loss": 1.9288, + "step": 707 + }, + { + "epoch": 0.7758904109589041, + "grad_norm": 2.3618645668029785, + "learning_rate": 4.147034930950447e-06, + "loss": 1.8982, + "step": 708 + }, + { + "epoch": 0.776986301369863, + "grad_norm": 2.150402307510376, + "learning_rate": 4.145004061738425e-06, + "loss": 1.9, + "step": 709 + }, + { + "epoch": 0.7780821917808219, + "grad_norm": 2.295571804046631, + "learning_rate": 4.1429731925264015e-06, + "loss": 1.9003, + "step": 710 + }, + { + "epoch": 0.7780821917808219, + "eval_loss": 2.067948341369629, + "eval_runtime": 37.2379, + "eval_samples_per_second": 26.854, + "eval_steps_per_second": 1.128, + "step": 710 + }, + { + "epoch": 0.7791780821917809, + "grad_norm": 2.3279545307159424, + "learning_rate": 4.140942323314379e-06, + "loss": 1.8307, + "step": 711 + }, + { + "epoch": 0.7802739726027397, + "grad_norm": 2.3993380069732666, + "learning_rate": 4.138911454102356e-06, + "loss": 1.8755, + "step": 712 + }, + { + "epoch": 0.7813698630136986, + "grad_norm": 2.139653444290161, + "learning_rate": 4.136880584890333e-06, + "loss": 1.8914, + "step": 713 + }, + { + "epoch": 0.7824657534246575, + "grad_norm": 2.239544153213501, + "learning_rate": 4.1348497156783106e-06, + "loss": 1.9289, + "step": 714 + }, + { + "epoch": 0.7835616438356164, + "grad_norm": 2.2039642333984375, + "learning_rate": 4.132818846466288e-06, + "loss": 1.9237, + "step": 715 + }, + { + "epoch": 0.7846575342465754, + "grad_norm": 2.1453804969787598, + "learning_rate": 4.130787977254265e-06, + "loss": 1.9291, + "step": 716 + }, + { + "epoch": 0.7857534246575343, + "grad_norm": 2.2350032329559326, + "learning_rate": 4.128757108042242e-06, + "loss": 1.9413, + "step": 717 + }, + { + "epoch": 0.7868493150684932, + "grad_norm": 2.1718318462371826, + "learning_rate": 4.12672623883022e-06, + "loss": 1.9318, + "step": 718 + }, + { + "epoch": 0.787945205479452, + "grad_norm": 2.1860601902008057, + "learning_rate": 4.124695369618197e-06, + "loss": 1.9934, + "step": 719 + }, + { + "epoch": 0.7890410958904109, + "grad_norm": 2.205166816711426, + "learning_rate": 4.122664500406174e-06, + "loss": 1.9038, + "step": 720 + }, + { + "epoch": 0.7890410958904109, + "eval_loss": 2.06229829788208, + "eval_runtime": 37.6121, + "eval_samples_per_second": 26.587, + "eval_steps_per_second": 1.117, + "step": 720 + }, + { + "epoch": 0.7901369863013699, + "grad_norm": 2.2094533443450928, + "learning_rate": 4.1206336311941515e-06, + "loss": 1.906, + "step": 721 + }, + { + "epoch": 0.7912328767123288, + "grad_norm": 2.178891658782959, + "learning_rate": 4.118602761982129e-06, + "loss": 1.936, + "step": 722 + }, + { + "epoch": 0.7923287671232877, + "grad_norm": 2.051077365875244, + "learning_rate": 4.116571892770106e-06, + "loss": 1.8736, + "step": 723 + }, + { + "epoch": 0.7934246575342466, + "grad_norm": 2.2430427074432373, + "learning_rate": 4.114541023558083e-06, + "loss": 1.8396, + "step": 724 + }, + { + "epoch": 0.7945205479452054, + "grad_norm": 2.2255303859710693, + "learning_rate": 4.112510154346061e-06, + "loss": 1.9668, + "step": 725 + }, + { + "epoch": 0.7956164383561644, + "grad_norm": 2.280313491821289, + "learning_rate": 4.110479285134038e-06, + "loss": 1.8945, + "step": 726 + }, + { + "epoch": 0.7967123287671233, + "grad_norm": 2.2516801357269287, + "learning_rate": 4.108448415922015e-06, + "loss": 1.9157, + "step": 727 + }, + { + "epoch": 0.7978082191780822, + "grad_norm": 2.1379809379577637, + "learning_rate": 4.1064175467099924e-06, + "loss": 1.935, + "step": 728 + }, + { + "epoch": 0.7989041095890411, + "grad_norm": 2.149425983428955, + "learning_rate": 4.10438667749797e-06, + "loss": 1.935, + "step": 729 + }, + { + "epoch": 0.8, + "grad_norm": 2.275270700454712, + "learning_rate": 4.102355808285947e-06, + "loss": 1.7889, + "step": 730 + }, + { + "epoch": 0.8, + "eval_loss": 2.0633974075317383, + "eval_runtime": 37.6895, + "eval_samples_per_second": 26.533, + "eval_steps_per_second": 1.114, + "step": 730 + }, + { + "epoch": 0.8010958904109589, + "grad_norm": 2.1550278663635254, + "learning_rate": 4.100324939073924e-06, + "loss": 1.9833, + "step": 731 + }, + { + "epoch": 0.8021917808219178, + "grad_norm": 2.3050060272216797, + "learning_rate": 4.0982940698619015e-06, + "loss": 1.8969, + "step": 732 + }, + { + "epoch": 0.8032876712328767, + "grad_norm": 2.4548778533935547, + "learning_rate": 4.096263200649879e-06, + "loss": 1.8945, + "step": 733 + }, + { + "epoch": 0.8043835616438356, + "grad_norm": 2.2611639499664307, + "learning_rate": 4.094232331437856e-06, + "loss": 1.9071, + "step": 734 + }, + { + "epoch": 0.8054794520547945, + "grad_norm": 2.1806774139404297, + "learning_rate": 4.092201462225833e-06, + "loss": 1.8787, + "step": 735 + }, + { + "epoch": 0.8065753424657535, + "grad_norm": 2.4374301433563232, + "learning_rate": 4.090170593013811e-06, + "loss": 1.88, + "step": 736 + }, + { + "epoch": 0.8076712328767123, + "grad_norm": 2.328333854675293, + "learning_rate": 4.088139723801787e-06, + "loss": 1.9178, + "step": 737 + }, + { + "epoch": 0.8087671232876712, + "grad_norm": 2.33552885055542, + "learning_rate": 4.086108854589765e-06, + "loss": 1.8587, + "step": 738 + }, + { + "epoch": 0.8098630136986301, + "grad_norm": 2.3472790718078613, + "learning_rate": 4.084077985377742e-06, + "loss": 1.877, + "step": 739 + }, + { + "epoch": 0.810958904109589, + "grad_norm": 2.307123899459839, + "learning_rate": 4.08204711616572e-06, + "loss": 1.8888, + "step": 740 + }, + { + "epoch": 0.810958904109589, + "eval_loss": 2.065826892852783, + "eval_runtime": 37.2747, + "eval_samples_per_second": 26.828, + "eval_steps_per_second": 1.127, + "step": 740 + }, + { + "epoch": 0.812054794520548, + "grad_norm": 2.3688747882843018, + "learning_rate": 4.080016246953696e-06, + "loss": 1.9571, + "step": 741 + }, + { + "epoch": 0.8131506849315069, + "grad_norm": 2.385007381439209, + "learning_rate": 4.077985377741674e-06, + "loss": 1.8657, + "step": 742 + }, + { + "epoch": 0.8142465753424658, + "grad_norm": 2.156766414642334, + "learning_rate": 4.075954508529651e-06, + "loss": 1.9421, + "step": 743 + }, + { + "epoch": 0.8153424657534246, + "grad_norm": 2.2074286937713623, + "learning_rate": 4.073923639317629e-06, + "loss": 1.9396, + "step": 744 + }, + { + "epoch": 0.8164383561643835, + "grad_norm": 2.185458183288574, + "learning_rate": 4.071892770105605e-06, + "loss": 1.9612, + "step": 745 + }, + { + "epoch": 0.8175342465753425, + "grad_norm": 2.324209451675415, + "learning_rate": 4.069861900893583e-06, + "loss": 1.892, + "step": 746 + }, + { + "epoch": 0.8186301369863014, + "grad_norm": 2.1312127113342285, + "learning_rate": 4.06783103168156e-06, + "loss": 1.8522, + "step": 747 + }, + { + "epoch": 0.8197260273972603, + "grad_norm": 2.0959460735321045, + "learning_rate": 4.065800162469538e-06, + "loss": 1.8732, + "step": 748 + }, + { + "epoch": 0.8208219178082192, + "grad_norm": 2.085519552230835, + "learning_rate": 4.063769293257514e-06, + "loss": 1.8832, + "step": 749 + }, + { + "epoch": 0.821917808219178, + "grad_norm": 2.310145854949951, + "learning_rate": 4.061738424045492e-06, + "loss": 1.919, + "step": 750 + }, + { + "epoch": 0.821917808219178, + "eval_loss": 2.0637967586517334, + "eval_runtime": 37.2246, + "eval_samples_per_second": 26.864, + "eval_steps_per_second": 1.128, + "step": 750 + }, + { + "epoch": 0.823013698630137, + "grad_norm": 2.3171160221099854, + "learning_rate": 4.059707554833469e-06, + "loss": 1.8874, + "step": 751 + }, + { + "epoch": 0.8241095890410959, + "grad_norm": 2.1065480709075928, + "learning_rate": 4.057676685621446e-06, + "loss": 1.9064, + "step": 752 + }, + { + "epoch": 0.8252054794520548, + "grad_norm": 2.0539677143096924, + "learning_rate": 4.0556458164094235e-06, + "loss": 1.9376, + "step": 753 + }, + { + "epoch": 0.8263013698630137, + "grad_norm": 2.4440360069274902, + "learning_rate": 4.053614947197401e-06, + "loss": 1.9129, + "step": 754 + }, + { + "epoch": 0.8273972602739726, + "grad_norm": 2.2324604988098145, + "learning_rate": 4.051584077985378e-06, + "loss": 1.8281, + "step": 755 + }, + { + "epoch": 0.8284931506849315, + "grad_norm": 2.262627363204956, + "learning_rate": 4.049553208773355e-06, + "loss": 1.9036, + "step": 756 + }, + { + "epoch": 0.8295890410958904, + "grad_norm": 2.2473418712615967, + "learning_rate": 4.047522339561333e-06, + "loss": 1.9455, + "step": 757 + }, + { + "epoch": 0.8306849315068493, + "grad_norm": 2.473444938659668, + "learning_rate": 4.04549147034931e-06, + "loss": 1.9003, + "step": 758 + }, + { + "epoch": 0.8317808219178082, + "grad_norm": 2.3454110622406006, + "learning_rate": 4.043460601137287e-06, + "loss": 1.8941, + "step": 759 + }, + { + "epoch": 0.8328767123287671, + "grad_norm": 2.4391231536865234, + "learning_rate": 4.041429731925264e-06, + "loss": 2.0146, + "step": 760 + }, + { + "epoch": 0.8328767123287671, + "eval_loss": 2.0618460178375244, + "eval_runtime": 37.1704, + "eval_samples_per_second": 26.903, + "eval_steps_per_second": 1.13, + "step": 760 + }, + { + "epoch": 0.8339726027397261, + "grad_norm": 2.3129498958587646, + "learning_rate": 4.039398862713242e-06, + "loss": 1.9, + "step": 761 + }, + { + "epoch": 0.8350684931506849, + "grad_norm": 2.3766934871673584, + "learning_rate": 4.037367993501219e-06, + "loss": 1.9017, + "step": 762 + }, + { + "epoch": 0.8361643835616438, + "grad_norm": 2.1698460578918457, + "learning_rate": 4.035337124289196e-06, + "loss": 1.9657, + "step": 763 + }, + { + "epoch": 0.8372602739726027, + "grad_norm": 2.384904146194458, + "learning_rate": 4.0333062550771735e-06, + "loss": 1.8867, + "step": 764 + }, + { + "epoch": 0.8383561643835616, + "grad_norm": 2.572391986846924, + "learning_rate": 4.031275385865151e-06, + "loss": 1.8561, + "step": 765 + }, + { + "epoch": 0.8394520547945206, + "grad_norm": 2.216606616973877, + "learning_rate": 4.029244516653128e-06, + "loss": 1.8926, + "step": 766 + }, + { + "epoch": 0.8405479452054795, + "grad_norm": 2.215756893157959, + "learning_rate": 4.027213647441105e-06, + "loss": 1.928, + "step": 767 + }, + { + "epoch": 0.8416438356164384, + "grad_norm": 2.2713284492492676, + "learning_rate": 4.025182778229083e-06, + "loss": 1.8286, + "step": 768 + }, + { + "epoch": 0.8427397260273972, + "grad_norm": 2.341747283935547, + "learning_rate": 4.02315190901706e-06, + "loss": 1.8789, + "step": 769 + }, + { + "epoch": 0.8438356164383561, + "grad_norm": 2.1485848426818848, + "learning_rate": 4.021121039805037e-06, + "loss": 1.8998, + "step": 770 + }, + { + "epoch": 0.8438356164383561, + "eval_loss": 2.0596234798431396, + "eval_runtime": 37.1854, + "eval_samples_per_second": 26.892, + "eval_steps_per_second": 1.129, + "step": 770 + }, + { + "epoch": 0.8449315068493151, + "grad_norm": 2.234192132949829, + "learning_rate": 4.0190901705930144e-06, + "loss": 1.8995, + "step": 771 + }, + { + "epoch": 0.846027397260274, + "grad_norm": 2.289328098297119, + "learning_rate": 4.017059301380992e-06, + "loss": 1.9882, + "step": 772 + }, + { + "epoch": 0.8471232876712329, + "grad_norm": 2.1001455783843994, + "learning_rate": 4.015028432168969e-06, + "loss": 1.8706, + "step": 773 + }, + { + "epoch": 0.8482191780821918, + "grad_norm": 2.2200522422790527, + "learning_rate": 4.012997562956945e-06, + "loss": 1.9443, + "step": 774 + }, + { + "epoch": 0.8493150684931506, + "grad_norm": 2.2540805339813232, + "learning_rate": 4.0109666937449235e-06, + "loss": 1.8552, + "step": 775 + }, + { + "epoch": 0.8504109589041096, + "grad_norm": 2.1214053630828857, + "learning_rate": 4.0089358245329e-06, + "loss": 1.8741, + "step": 776 + }, + { + "epoch": 0.8515068493150685, + "grad_norm": 2.0221869945526123, + "learning_rate": 4.006904955320878e-06, + "loss": 1.8798, + "step": 777 + }, + { + "epoch": 0.8526027397260274, + "grad_norm": 2.217261791229248, + "learning_rate": 4.0048740861088545e-06, + "loss": 1.8542, + "step": 778 + }, + { + "epoch": 0.8536986301369863, + "grad_norm": 2.099202871322632, + "learning_rate": 4.002843216896833e-06, + "loss": 1.9342, + "step": 779 + }, + { + "epoch": 0.8547945205479452, + "grad_norm": 2.139526844024658, + "learning_rate": 4.000812347684809e-06, + "loss": 1.9252, + "step": 780 + }, + { + "epoch": 0.8547945205479452, + "eval_loss": 2.054051399230957, + "eval_runtime": 37.1125, + "eval_samples_per_second": 26.945, + "eval_steps_per_second": 1.132, + "step": 780 + }, + { + "epoch": 0.8558904109589041, + "grad_norm": 2.2231476306915283, + "learning_rate": 3.998781478472787e-06, + "loss": 1.8671, + "step": 781 + }, + { + "epoch": 0.856986301369863, + "grad_norm": 2.143979072570801, + "learning_rate": 3.996750609260764e-06, + "loss": 1.8759, + "step": 782 + }, + { + "epoch": 0.8580821917808219, + "grad_norm": 2.1353135108947754, + "learning_rate": 3.994719740048742e-06, + "loss": 1.8559, + "step": 783 + }, + { + "epoch": 0.8591780821917808, + "grad_norm": 2.222564697265625, + "learning_rate": 3.992688870836718e-06, + "loss": 1.999, + "step": 784 + }, + { + "epoch": 0.8602739726027397, + "grad_norm": 1.958031177520752, + "learning_rate": 3.990658001624696e-06, + "loss": 1.9019, + "step": 785 + }, + { + "epoch": 0.8613698630136987, + "grad_norm": 2.2048962116241455, + "learning_rate": 3.988627132412673e-06, + "loss": 1.8561, + "step": 786 + }, + { + "epoch": 0.8624657534246576, + "grad_norm": 2.150332450866699, + "learning_rate": 3.98659626320065e-06, + "loss": 1.9399, + "step": 787 + }, + { + "epoch": 0.8635616438356164, + "grad_norm": 2.182742118835449, + "learning_rate": 3.984565393988627e-06, + "loss": 1.8933, + "step": 788 + }, + { + "epoch": 0.8646575342465753, + "grad_norm": 2.19435715675354, + "learning_rate": 3.9825345247766046e-06, + "loss": 1.8918, + "step": 789 + }, + { + "epoch": 0.8657534246575342, + "grad_norm": 2.3382859230041504, + "learning_rate": 3.980503655564582e-06, + "loss": 1.8937, + "step": 790 + }, + { + "epoch": 0.8657534246575342, + "eval_loss": 2.051997423171997, + "eval_runtime": 37.1866, + "eval_samples_per_second": 26.891, + "eval_steps_per_second": 1.129, + "step": 790 + }, + { + "epoch": 0.8668493150684932, + "grad_norm": 2.2700726985931396, + "learning_rate": 3.978472786352559e-06, + "loss": 1.9148, + "step": 791 + }, + { + "epoch": 0.8679452054794521, + "grad_norm": 2.2998759746551514, + "learning_rate": 3.976441917140536e-06, + "loss": 1.9268, + "step": 792 + }, + { + "epoch": 0.869041095890411, + "grad_norm": 2.2367377281188965, + "learning_rate": 3.974411047928514e-06, + "loss": 1.901, + "step": 793 + }, + { + "epoch": 0.8701369863013698, + "grad_norm": 2.0868773460388184, + "learning_rate": 3.972380178716491e-06, + "loss": 1.9677, + "step": 794 + }, + { + "epoch": 0.8712328767123287, + "grad_norm": 2.183732509613037, + "learning_rate": 3.970349309504468e-06, + "loss": 1.9271, + "step": 795 + }, + { + "epoch": 0.8723287671232877, + "grad_norm": 2.187282085418701, + "learning_rate": 3.9683184402924455e-06, + "loss": 1.9437, + "step": 796 + }, + { + "epoch": 0.8734246575342466, + "grad_norm": 2.1906070709228516, + "learning_rate": 3.966287571080423e-06, + "loss": 1.8437, + "step": 797 + }, + { + "epoch": 0.8745205479452055, + "grad_norm": 2.16979718208313, + "learning_rate": 3.9642567018684e-06, + "loss": 1.829, + "step": 798 + }, + { + "epoch": 0.8756164383561644, + "grad_norm": 2.2387869358062744, + "learning_rate": 3.962225832656377e-06, + "loss": 1.9143, + "step": 799 + }, + { + "epoch": 0.8767123287671232, + "grad_norm": 2.208221197128296, + "learning_rate": 3.960194963444355e-06, + "loss": 1.8894, + "step": 800 + }, + { + "epoch": 0.8767123287671232, + "eval_loss": 2.047945976257324, + "eval_runtime": 37.1778, + "eval_samples_per_second": 26.898, + "eval_steps_per_second": 1.13, + "step": 800 + }, + { + "epoch": 0.8778082191780822, + "grad_norm": 2.335108518600464, + "learning_rate": 3.958164094232332e-06, + "loss": 1.8155, + "step": 801 + }, + { + "epoch": 0.8789041095890411, + "grad_norm": 2.168006181716919, + "learning_rate": 3.956133225020309e-06, + "loss": 1.8919, + "step": 802 + }, + { + "epoch": 0.88, + "grad_norm": 2.3029897212982178, + "learning_rate": 3.954102355808286e-06, + "loss": 1.8996, + "step": 803 + }, + { + "epoch": 0.8810958904109589, + "grad_norm": 2.156141519546509, + "learning_rate": 3.952071486596264e-06, + "loss": 1.8965, + "step": 804 + }, + { + "epoch": 0.8821917808219178, + "grad_norm": 2.1522068977355957, + "learning_rate": 3.950040617384241e-06, + "loss": 1.8922, + "step": 805 + }, + { + "epoch": 0.8832876712328767, + "grad_norm": 2.153430700302124, + "learning_rate": 3.948009748172218e-06, + "loss": 1.8839, + "step": 806 + }, + { + "epoch": 0.8843835616438356, + "grad_norm": 2.1891674995422363, + "learning_rate": 3.9459788789601955e-06, + "loss": 1.8756, + "step": 807 + }, + { + "epoch": 0.8854794520547945, + "grad_norm": 2.1092092990875244, + "learning_rate": 3.943948009748173e-06, + "loss": 1.9452, + "step": 808 + }, + { + "epoch": 0.8865753424657534, + "grad_norm": 2.0465025901794434, + "learning_rate": 3.94191714053615e-06, + "loss": 1.9067, + "step": 809 + }, + { + "epoch": 0.8876712328767123, + "grad_norm": 2.1150474548339844, + "learning_rate": 3.939886271324127e-06, + "loss": 1.8874, + "step": 810 + }, + { + "epoch": 0.8876712328767123, + "eval_loss": 2.046292304992676, + "eval_runtime": 37.2421, + "eval_samples_per_second": 26.851, + "eval_steps_per_second": 1.128, + "step": 810 + }, + { + "epoch": 0.8887671232876713, + "grad_norm": 2.039989948272705, + "learning_rate": 3.937855402112104e-06, + "loss": 1.9212, + "step": 811 + }, + { + "epoch": 0.8898630136986302, + "grad_norm": 2.17354679107666, + "learning_rate": 3.935824532900082e-06, + "loss": 1.8473, + "step": 812 + }, + { + "epoch": 0.890958904109589, + "grad_norm": 2.3101940155029297, + "learning_rate": 3.933793663688058e-06, + "loss": 1.8413, + "step": 813 + }, + { + "epoch": 0.8920547945205479, + "grad_norm": 2.1300928592681885, + "learning_rate": 3.9317627944760365e-06, + "loss": 1.9126, + "step": 814 + }, + { + "epoch": 0.8931506849315068, + "grad_norm": 2.2317357063293457, + "learning_rate": 3.929731925264013e-06, + "loss": 1.938, + "step": 815 + }, + { + "epoch": 0.8942465753424658, + "grad_norm": 2.1097419261932373, + "learning_rate": 3.927701056051991e-06, + "loss": 1.8426, + "step": 816 + }, + { + "epoch": 0.8953424657534247, + "grad_norm": 2.1709141731262207, + "learning_rate": 3.9256701868399674e-06, + "loss": 1.8779, + "step": 817 + }, + { + "epoch": 0.8964383561643836, + "grad_norm": 2.104193925857544, + "learning_rate": 3.9236393176279456e-06, + "loss": 1.8736, + "step": 818 + }, + { + "epoch": 0.8975342465753424, + "grad_norm": 2.094637632369995, + "learning_rate": 3.921608448415922e-06, + "loss": 1.8556, + "step": 819 + }, + { + "epoch": 0.8986301369863013, + "grad_norm": 2.217918872833252, + "learning_rate": 3.9195775792039e-06, + "loss": 1.8429, + "step": 820 + }, + { + "epoch": 0.8986301369863013, + "eval_loss": 2.046095848083496, + "eval_runtime": 37.5659, + "eval_samples_per_second": 26.62, + "eval_steps_per_second": 1.118, + "step": 820 + }, + { + "epoch": 0.8997260273972603, + "grad_norm": 2.220355272293091, + "learning_rate": 3.9175467099918765e-06, + "loss": 1.8973, + "step": 821 + }, + { + "epoch": 0.9008219178082192, + "grad_norm": 2.0766689777374268, + "learning_rate": 3.915515840779855e-06, + "loss": 1.8728, + "step": 822 + }, + { + "epoch": 0.9019178082191781, + "grad_norm": 2.2046821117401123, + "learning_rate": 3.913484971567831e-06, + "loss": 1.9211, + "step": 823 + }, + { + "epoch": 0.903013698630137, + "grad_norm": 2.286024332046509, + "learning_rate": 3.911454102355808e-06, + "loss": 1.913, + "step": 824 + }, + { + "epoch": 0.9041095890410958, + "grad_norm": 2.2495479583740234, + "learning_rate": 3.909423233143786e-06, + "loss": 1.8214, + "step": 825 + }, + { + "epoch": 0.9052054794520548, + "grad_norm": 2.1300203800201416, + "learning_rate": 3.907392363931763e-06, + "loss": 1.9033, + "step": 826 + }, + { + "epoch": 0.9063013698630137, + "grad_norm": 2.092094898223877, + "learning_rate": 3.90536149471974e-06, + "loss": 1.9007, + "step": 827 + }, + { + "epoch": 0.9073972602739726, + "grad_norm": 2.1571428775787354, + "learning_rate": 3.9033306255077175e-06, + "loss": 1.9606, + "step": 828 + }, + { + "epoch": 0.9084931506849315, + "grad_norm": 2.1052281856536865, + "learning_rate": 3.901299756295695e-06, + "loss": 1.9916, + "step": 829 + }, + { + "epoch": 0.9095890410958904, + "grad_norm": 2.113666534423828, + "learning_rate": 3.899268887083672e-06, + "loss": 2.0136, + "step": 830 + }, + { + "epoch": 0.9095890410958904, + "eval_loss": 2.0450072288513184, + "eval_runtime": 37.6746, + "eval_samples_per_second": 26.543, + "eval_steps_per_second": 1.115, + "step": 830 + }, + { + "epoch": 0.9106849315068493, + "grad_norm": 2.112591028213501, + "learning_rate": 3.897238017871649e-06, + "loss": 1.8784, + "step": 831 + }, + { + "epoch": 0.9117808219178082, + "grad_norm": 2.1584672927856445, + "learning_rate": 3.8952071486596266e-06, + "loss": 1.8876, + "step": 832 + }, + { + "epoch": 0.9128767123287671, + "grad_norm": 2.2744317054748535, + "learning_rate": 3.893176279447604e-06, + "loss": 1.8981, + "step": 833 + }, + { + "epoch": 0.913972602739726, + "grad_norm": 2.21824312210083, + "learning_rate": 3.891145410235581e-06, + "loss": 1.85, + "step": 834 + }, + { + "epoch": 0.915068493150685, + "grad_norm": 2.2203729152679443, + "learning_rate": 3.889114541023558e-06, + "loss": 1.9649, + "step": 835 + }, + { + "epoch": 0.9161643835616439, + "grad_norm": 2.1767687797546387, + "learning_rate": 3.887083671811536e-06, + "loss": 1.9044, + "step": 836 + }, + { + "epoch": 0.9172602739726028, + "grad_norm": 2.0724353790283203, + "learning_rate": 3.885052802599513e-06, + "loss": 1.8787, + "step": 837 + }, + { + "epoch": 0.9183561643835616, + "grad_norm": 2.1283740997314453, + "learning_rate": 3.88302193338749e-06, + "loss": 1.8284, + "step": 838 + }, + { + "epoch": 0.9194520547945205, + "grad_norm": 2.5087451934814453, + "learning_rate": 3.8809910641754675e-06, + "loss": 1.9546, + "step": 839 + }, + { + "epoch": 0.9205479452054794, + "grad_norm": 2.0633344650268555, + "learning_rate": 3.878960194963445e-06, + "loss": 1.8969, + "step": 840 + }, + { + "epoch": 0.9205479452054794, + "eval_loss": 2.047938346862793, + "eval_runtime": 37.2293, + "eval_samples_per_second": 26.861, + "eval_steps_per_second": 1.128, + "step": 840 + }, + { + "epoch": 0.9216438356164384, + "grad_norm": 2.178361654281616, + "learning_rate": 3.876929325751422e-06, + "loss": 1.8855, + "step": 841 + }, + { + "epoch": 0.9227397260273973, + "grad_norm": 2.3223721981048584, + "learning_rate": 3.874898456539399e-06, + "loss": 1.9363, + "step": 842 + }, + { + "epoch": 0.9238356164383562, + "grad_norm": 2.3330960273742676, + "learning_rate": 3.872867587327377e-06, + "loss": 1.8374, + "step": 843 + }, + { + "epoch": 0.924931506849315, + "grad_norm": 2.0584356784820557, + "learning_rate": 3.870836718115354e-06, + "loss": 1.911, + "step": 844 + }, + { + "epoch": 0.9260273972602739, + "grad_norm": 2.1834609508514404, + "learning_rate": 3.868805848903331e-06, + "loss": 1.8645, + "step": 845 + }, + { + "epoch": 0.9271232876712329, + "grad_norm": 2.2576122283935547, + "learning_rate": 3.8667749796913084e-06, + "loss": 1.8662, + "step": 846 + }, + { + "epoch": 0.9282191780821918, + "grad_norm": 2.1271183490753174, + "learning_rate": 3.864744110479286e-06, + "loss": 1.8935, + "step": 847 + }, + { + "epoch": 0.9293150684931507, + "grad_norm": 2.226727247238159, + "learning_rate": 3.862713241267262e-06, + "loss": 1.8969, + "step": 848 + }, + { + "epoch": 0.9304109589041096, + "grad_norm": 2.3301939964294434, + "learning_rate": 3.86068237205524e-06, + "loss": 1.9393, + "step": 849 + }, + { + "epoch": 0.9315068493150684, + "grad_norm": 2.180250883102417, + "learning_rate": 3.858651502843217e-06, + "loss": 1.9331, + "step": 850 + }, + { + "epoch": 0.9315068493150684, + "eval_loss": 2.0509681701660156, + "eval_runtime": 37.2111, + "eval_samples_per_second": 26.874, + "eval_steps_per_second": 1.129, + "step": 850 + }, + { + "epoch": 0.9326027397260274, + "grad_norm": 2.223548173904419, + "learning_rate": 3.856620633631195e-06, + "loss": 1.894, + "step": 851 + }, + { + "epoch": 0.9336986301369863, + "grad_norm": 2.1904265880584717, + "learning_rate": 3.854589764419171e-06, + "loss": 1.8265, + "step": 852 + }, + { + "epoch": 0.9347945205479452, + "grad_norm": 2.1374316215515137, + "learning_rate": 3.852558895207149e-06, + "loss": 1.8755, + "step": 853 + }, + { + "epoch": 0.9358904109589041, + "grad_norm": 2.2175488471984863, + "learning_rate": 3.850528025995126e-06, + "loss": 1.9649, + "step": 854 + }, + { + "epoch": 0.936986301369863, + "grad_norm": 2.192126989364624, + "learning_rate": 3.848497156783104e-06, + "loss": 1.919, + "step": 855 + }, + { + "epoch": 0.938082191780822, + "grad_norm": 2.219954013824463, + "learning_rate": 3.84646628757108e-06, + "loss": 1.8651, + "step": 856 + }, + { + "epoch": 0.9391780821917808, + "grad_norm": 2.2897071838378906, + "learning_rate": 3.8444354183590585e-06, + "loss": 1.8744, + "step": 857 + }, + { + "epoch": 0.9402739726027397, + "grad_norm": 2.1656205654144287, + "learning_rate": 3.842404549147035e-06, + "loss": 1.8622, + "step": 858 + }, + { + "epoch": 0.9413698630136986, + "grad_norm": 2.20072603225708, + "learning_rate": 3.840373679935013e-06, + "loss": 1.8678, + "step": 859 + }, + { + "epoch": 0.9424657534246575, + "grad_norm": 2.1474406719207764, + "learning_rate": 3.8383428107229894e-06, + "loss": 1.8462, + "step": 860 + }, + { + "epoch": 0.9424657534246575, + "eval_loss": 2.0451602935791016, + "eval_runtime": 37.1383, + "eval_samples_per_second": 26.926, + "eval_steps_per_second": 1.131, + "step": 860 + }, + { + "epoch": 0.9435616438356165, + "grad_norm": 1.9542036056518555, + "learning_rate": 3.836311941510967e-06, + "loss": 1.8739, + "step": 861 + }, + { + "epoch": 0.9446575342465754, + "grad_norm": 2.3951354026794434, + "learning_rate": 3.834281072298944e-06, + "loss": 1.8565, + "step": 862 + }, + { + "epoch": 0.9457534246575342, + "grad_norm": 2.4916961193084717, + "learning_rate": 3.832250203086921e-06, + "loss": 1.9138, + "step": 863 + }, + { + "epoch": 0.9468493150684931, + "grad_norm": 2.1650452613830566, + "learning_rate": 3.8302193338748985e-06, + "loss": 1.8499, + "step": 864 + }, + { + "epoch": 0.947945205479452, + "grad_norm": 2.2215073108673096, + "learning_rate": 3.828188464662876e-06, + "loss": 1.9376, + "step": 865 + }, + { + "epoch": 0.949041095890411, + "grad_norm": 2.3827078342437744, + "learning_rate": 3.826157595450853e-06, + "loss": 1.9726, + "step": 866 + }, + { + "epoch": 0.9501369863013699, + "grad_norm": 2.37194561958313, + "learning_rate": 3.82412672623883e-06, + "loss": 1.9009, + "step": 867 + }, + { + "epoch": 0.9512328767123288, + "grad_norm": 2.116562604904175, + "learning_rate": 3.822095857026808e-06, + "loss": 1.8342, + "step": 868 + }, + { + "epoch": 0.9523287671232876, + "grad_norm": 2.3410849571228027, + "learning_rate": 3.820064987814785e-06, + "loss": 1.9035, + "step": 869 + }, + { + "epoch": 0.9534246575342465, + "grad_norm": 2.207545280456543, + "learning_rate": 3.818034118602762e-06, + "loss": 1.8539, + "step": 870 + }, + { + "epoch": 0.9534246575342465, + "eval_loss": 2.0433826446533203, + "eval_runtime": 37.114, + "eval_samples_per_second": 26.944, + "eval_steps_per_second": 1.132, + "step": 870 + }, + { + "epoch": 0.9545205479452055, + "grad_norm": 2.1716196537017822, + "learning_rate": 3.8160032493907395e-06, + "loss": 1.8557, + "step": 871 + }, + { + "epoch": 0.9556164383561644, + "grad_norm": 2.213124990463257, + "learning_rate": 3.813972380178717e-06, + "loss": 1.8638, + "step": 872 + }, + { + "epoch": 0.9567123287671233, + "grad_norm": 2.3185672760009766, + "learning_rate": 3.811941510966694e-06, + "loss": 1.8645, + "step": 873 + }, + { + "epoch": 0.9578082191780822, + "grad_norm": 2.453956127166748, + "learning_rate": 3.809910641754671e-06, + "loss": 1.9183, + "step": 874 + }, + { + "epoch": 0.958904109589041, + "grad_norm": 2.1971592903137207, + "learning_rate": 3.8078797725426486e-06, + "loss": 1.8445, + "step": 875 + }, + { + "epoch": 0.96, + "grad_norm": 2.273247241973877, + "learning_rate": 3.8058489033306254e-06, + "loss": 1.8024, + "step": 876 + }, + { + "epoch": 0.9610958904109589, + "grad_norm": 2.1687841415405273, + "learning_rate": 3.803818034118603e-06, + "loss": 1.8931, + "step": 877 + }, + { + "epoch": 0.9621917808219178, + "grad_norm": 2.326498508453369, + "learning_rate": 3.80178716490658e-06, + "loss": 1.8587, + "step": 878 + }, + { + "epoch": 0.9632876712328767, + "grad_norm": 2.0317249298095703, + "learning_rate": 3.7997562956945577e-06, + "loss": 1.9394, + "step": 879 + }, + { + "epoch": 0.9643835616438357, + "grad_norm": 2.3171892166137695, + "learning_rate": 3.7977254264825345e-06, + "loss": 1.92, + "step": 880 + }, + { + "epoch": 0.9643835616438357, + "eval_loss": 2.0431556701660156, + "eval_runtime": 37.1174, + "eval_samples_per_second": 26.942, + "eval_steps_per_second": 1.132, + "step": 880 + }, + { + "epoch": 0.9654794520547946, + "grad_norm": 2.2215981483459473, + "learning_rate": 3.7956945572705122e-06, + "loss": 1.8658, + "step": 881 + }, + { + "epoch": 0.9665753424657534, + "grad_norm": 2.2940685749053955, + "learning_rate": 3.793663688058489e-06, + "loss": 1.8528, + "step": 882 + }, + { + "epoch": 0.9676712328767123, + "grad_norm": 2.2218081951141357, + "learning_rate": 3.791632818846467e-06, + "loss": 1.9446, + "step": 883 + }, + { + "epoch": 0.9687671232876712, + "grad_norm": 2.2608420848846436, + "learning_rate": 3.7896019496344436e-06, + "loss": 1.9433, + "step": 884 + }, + { + "epoch": 0.9698630136986301, + "grad_norm": 2.0247788429260254, + "learning_rate": 3.787571080422421e-06, + "loss": 1.9286, + "step": 885 + }, + { + "epoch": 0.9709589041095891, + "grad_norm": 2.164071798324585, + "learning_rate": 3.785540211210398e-06, + "loss": 1.9349, + "step": 886 + }, + { + "epoch": 0.972054794520548, + "grad_norm": 2.089215040206909, + "learning_rate": 3.7835093419983755e-06, + "loss": 1.9077, + "step": 887 + }, + { + "epoch": 0.9731506849315068, + "grad_norm": 2.245256185531616, + "learning_rate": 3.7814784727863527e-06, + "loss": 1.8911, + "step": 888 + }, + { + "epoch": 0.9742465753424657, + "grad_norm": 2.1227362155914307, + "learning_rate": 3.77944760357433e-06, + "loss": 1.8761, + "step": 889 + }, + { + "epoch": 0.9753424657534246, + "grad_norm": 2.1797730922698975, + "learning_rate": 3.7774167343623073e-06, + "loss": 1.9954, + "step": 890 + }, + { + "epoch": 0.9753424657534246, + "eval_loss": 2.034895896911621, + "eval_runtime": 37.1682, + "eval_samples_per_second": 26.905, + "eval_steps_per_second": 1.13, + "step": 890 + }, + { + "epoch": 0.9764383561643836, + "grad_norm": 2.171062707901001, + "learning_rate": 3.7753858651502846e-06, + "loss": 1.9287, + "step": 891 + }, + { + "epoch": 0.9775342465753425, + "grad_norm": 2.1950814723968506, + "learning_rate": 3.773354995938262e-06, + "loss": 1.8587, + "step": 892 + }, + { + "epoch": 0.9786301369863014, + "grad_norm": 2.2582995891571045, + "learning_rate": 3.771324126726239e-06, + "loss": 1.9269, + "step": 893 + }, + { + "epoch": 0.9797260273972602, + "grad_norm": 2.218113660812378, + "learning_rate": 3.7692932575142164e-06, + "loss": 1.9007, + "step": 894 + }, + { + "epoch": 0.9808219178082191, + "grad_norm": 2.187178134918213, + "learning_rate": 3.7672623883021937e-06, + "loss": 1.9326, + "step": 895 + }, + { + "epoch": 0.9819178082191781, + "grad_norm": 2.20451283454895, + "learning_rate": 3.765231519090171e-06, + "loss": 1.8644, + "step": 896 + }, + { + "epoch": 0.983013698630137, + "grad_norm": 2.1399290561676025, + "learning_rate": 3.7632006498781482e-06, + "loss": 1.7875, + "step": 897 + }, + { + "epoch": 0.9841095890410959, + "grad_norm": 2.049473524093628, + "learning_rate": 3.761169780666125e-06, + "loss": 1.8987, + "step": 898 + }, + { + "epoch": 0.9852054794520548, + "grad_norm": 2.154444456100464, + "learning_rate": 3.7591389114541028e-06, + "loss": 1.8475, + "step": 899 + }, + { + "epoch": 0.9863013698630136, + "grad_norm": 2.1840224266052246, + "learning_rate": 3.7571080422420796e-06, + "loss": 1.8966, + "step": 900 + }, + { + "epoch": 0.9863013698630136, + "eval_loss": 2.0326430797576904, + "eval_runtime": 37.1551, + "eval_samples_per_second": 26.914, + "eval_steps_per_second": 1.13, + "step": 900 + }, + { + "epoch": 0.9873972602739726, + "grad_norm": 2.0891942977905273, + "learning_rate": 3.7550771730300573e-06, + "loss": 1.9113, + "step": 901 + }, + { + "epoch": 0.9884931506849315, + "grad_norm": 2.2646753787994385, + "learning_rate": 3.753046303818034e-06, + "loss": 1.8624, + "step": 902 + }, + { + "epoch": 0.9895890410958904, + "grad_norm": 2.15173602104187, + "learning_rate": 3.751015434606012e-06, + "loss": 1.933, + "step": 903 + }, + { + "epoch": 0.9906849315068493, + "grad_norm": 2.1751456260681152, + "learning_rate": 3.7489845653939887e-06, + "loss": 1.9114, + "step": 904 + }, + { + "epoch": 0.9917808219178083, + "grad_norm": 2.161860942840576, + "learning_rate": 3.7469536961819664e-06, + "loss": 1.9508, + "step": 905 + }, + { + "epoch": 0.9928767123287672, + "grad_norm": 2.0921733379364014, + "learning_rate": 3.7449228269699433e-06, + "loss": 1.8482, + "step": 906 + }, + { + "epoch": 0.993972602739726, + "grad_norm": 2.0823893547058105, + "learning_rate": 3.742891957757921e-06, + "loss": 1.8744, + "step": 907 + }, + { + "epoch": 0.9950684931506849, + "grad_norm": 2.136939525604248, + "learning_rate": 3.740861088545898e-06, + "loss": 1.9037, + "step": 908 + }, + { + "epoch": 0.9961643835616438, + "grad_norm": 2.0600836277008057, + "learning_rate": 3.7388302193338755e-06, + "loss": 1.8922, + "step": 909 + }, + { + "epoch": 0.9972602739726028, + "grad_norm": 2.235780715942383, + "learning_rate": 3.7367993501218524e-06, + "loss": 1.9222, + "step": 910 + }, + { + "epoch": 0.9972602739726028, + "eval_loss": 2.034031629562378, + "eval_runtime": 37.2911, + "eval_samples_per_second": 26.816, + "eval_steps_per_second": 1.126, + "step": 910 + }, + { + "epoch": 0.9983561643835617, + "grad_norm": 2.124662160873413, + "learning_rate": 3.7347684809098297e-06, + "loss": 1.8584, + "step": 911 + }, + { + "epoch": 0.9994520547945206, + "grad_norm": 2.1237940788269043, + "learning_rate": 3.732737611697807e-06, + "loss": 1.8579, + "step": 912 + }, + { + "epoch": 1.0005479452054795, + "grad_norm": 2.1175124645233154, + "learning_rate": 3.7307067424857842e-06, + "loss": 1.8348, + "step": 913 + }, + { + "epoch": 1.0016438356164383, + "grad_norm": 2.349686861038208, + "learning_rate": 3.7286758732737615e-06, + "loss": 1.7822, + "step": 914 + }, + { + "epoch": 1.0027397260273974, + "grad_norm": 2.3278446197509766, + "learning_rate": 3.7266450040617388e-06, + "loss": 1.8346, + "step": 915 + }, + { + "epoch": 1.0038356164383562, + "grad_norm": 2.1633124351501465, + "learning_rate": 3.724614134849716e-06, + "loss": 1.8137, + "step": 916 + }, + { + "epoch": 1.004931506849315, + "grad_norm": 2.5075809955596924, + "learning_rate": 3.7225832656376933e-06, + "loss": 1.955, + "step": 917 + }, + { + "epoch": 1.006027397260274, + "grad_norm": 2.3370068073272705, + "learning_rate": 3.7205523964256706e-06, + "loss": 1.7985, + "step": 918 + }, + { + "epoch": 1.0071232876712328, + "grad_norm": 2.288229465484619, + "learning_rate": 3.718521527213648e-06, + "loss": 1.923, + "step": 919 + }, + { + "epoch": 1.0082191780821919, + "grad_norm": 2.1989684104919434, + "learning_rate": 3.716490658001625e-06, + "loss": 1.8876, + "step": 920 + }, + { + "epoch": 1.0082191780821919, + "eval_loss": 2.03442120552063, + "eval_runtime": 37.5659, + "eval_samples_per_second": 26.62, + "eval_steps_per_second": 1.118, + "step": 920 + }, + { + "epoch": 1.0093150684931507, + "grad_norm": 2.202122211456299, + "learning_rate": 3.7144597887896024e-06, + "loss": 1.879, + "step": 921 + }, + { + "epoch": 1.0104109589041095, + "grad_norm": 2.325223445892334, + "learning_rate": 3.7124289195775797e-06, + "loss": 1.7632, + "step": 922 + }, + { + "epoch": 1.0115068493150685, + "grad_norm": 2.2964577674865723, + "learning_rate": 3.710398050365557e-06, + "loss": 1.8983, + "step": 923 + }, + { + "epoch": 1.0126027397260273, + "grad_norm": 2.548537015914917, + "learning_rate": 3.708367181153534e-06, + "loss": 1.8666, + "step": 924 + }, + { + "epoch": 1.0136986301369864, + "grad_norm": 2.229745626449585, + "learning_rate": 3.7063363119415115e-06, + "loss": 1.8395, + "step": 925 + }, + { + "epoch": 1.0147945205479452, + "grad_norm": 2.3639235496520996, + "learning_rate": 3.7043054427294884e-06, + "loss": 1.8864, + "step": 926 + }, + { + "epoch": 1.0158904109589042, + "grad_norm": 2.3403055667877197, + "learning_rate": 3.702274573517466e-06, + "loss": 1.8385, + "step": 927 + }, + { + "epoch": 1.016986301369863, + "grad_norm": 2.2838001251220703, + "learning_rate": 3.700243704305443e-06, + "loss": 1.8135, + "step": 928 + }, + { + "epoch": 1.0180821917808218, + "grad_norm": 2.4730923175811768, + "learning_rate": 3.6982128350934206e-06, + "loss": 1.837, + "step": 929 + }, + { + "epoch": 1.0191780821917809, + "grad_norm": 2.5711073875427246, + "learning_rate": 3.6961819658813975e-06, + "loss": 1.8276, + "step": 930 + }, + { + "epoch": 1.0191780821917809, + "eval_loss": 2.0346033573150635, + "eval_runtime": 37.6741, + "eval_samples_per_second": 26.543, + "eval_steps_per_second": 1.115, + "step": 930 + }, + { + "epoch": 1.0202739726027397, + "grad_norm": 2.296823740005493, + "learning_rate": 3.694151096669375e-06, + "loss": 1.8054, + "step": 931 + }, + { + "epoch": 1.0213698630136987, + "grad_norm": 2.321107864379883, + "learning_rate": 3.692120227457352e-06, + "loss": 1.9296, + "step": 932 + }, + { + "epoch": 1.0224657534246575, + "grad_norm": 2.3478338718414307, + "learning_rate": 3.6900893582453297e-06, + "loss": 1.8503, + "step": 933 + }, + { + "epoch": 1.0235616438356165, + "grad_norm": 2.3698856830596924, + "learning_rate": 3.6880584890333066e-06, + "loss": 1.8766, + "step": 934 + }, + { + "epoch": 1.0246575342465754, + "grad_norm": 2.2751212120056152, + "learning_rate": 3.6860276198212834e-06, + "loss": 1.7575, + "step": 935 + }, + { + "epoch": 1.0257534246575342, + "grad_norm": 2.368741035461426, + "learning_rate": 3.683996750609261e-06, + "loss": 1.8555, + "step": 936 + }, + { + "epoch": 1.0268493150684932, + "grad_norm": 2.4507620334625244, + "learning_rate": 3.681965881397238e-06, + "loss": 1.8765, + "step": 937 + }, + { + "epoch": 1.027945205479452, + "grad_norm": 2.1907308101654053, + "learning_rate": 3.6799350121852157e-06, + "loss": 1.8652, + "step": 938 + }, + { + "epoch": 1.029041095890411, + "grad_norm": 2.2696914672851562, + "learning_rate": 3.6779041429731925e-06, + "loss": 1.8854, + "step": 939 + }, + { + "epoch": 1.0301369863013699, + "grad_norm": 2.3696796894073486, + "learning_rate": 3.6758732737611702e-06, + "loss": 1.8303, + "step": 940 + }, + { + "epoch": 1.0301369863013699, + "eval_loss": 2.030118703842163, + "eval_runtime": 37.1755, + "eval_samples_per_second": 26.899, + "eval_steps_per_second": 1.13, + "step": 940 + }, + { + "epoch": 1.0312328767123287, + "grad_norm": 2.440284013748169, + "learning_rate": 3.673842404549147e-06, + "loss": 1.8498, + "step": 941 + }, + { + "epoch": 1.0323287671232877, + "grad_norm": 2.250570297241211, + "learning_rate": 3.671811535337125e-06, + "loss": 1.8361, + "step": 942 + }, + { + "epoch": 1.0334246575342465, + "grad_norm": 2.5550990104675293, + "learning_rate": 3.6697806661251016e-06, + "loss": 1.8464, + "step": 943 + }, + { + "epoch": 1.0345205479452055, + "grad_norm": 2.502474784851074, + "learning_rate": 3.6677497969130793e-06, + "loss": 1.8471, + "step": 944 + }, + { + "epoch": 1.0356164383561643, + "grad_norm": 2.3815231323242188, + "learning_rate": 3.665718927701056e-06, + "loss": 1.9786, + "step": 945 + }, + { + "epoch": 1.0367123287671234, + "grad_norm": 2.3053736686706543, + "learning_rate": 3.663688058489034e-06, + "loss": 1.8625, + "step": 946 + }, + { + "epoch": 1.0378082191780822, + "grad_norm": 2.2623229026794434, + "learning_rate": 3.6616571892770107e-06, + "loss": 1.9206, + "step": 947 + }, + { + "epoch": 1.038904109589041, + "grad_norm": 2.634559392929077, + "learning_rate": 3.659626320064988e-06, + "loss": 1.8475, + "step": 948 + }, + { + "epoch": 1.04, + "grad_norm": 2.264615297317505, + "learning_rate": 3.6575954508529653e-06, + "loss": 1.7857, + "step": 949 + }, + { + "epoch": 1.0410958904109588, + "grad_norm": 2.4125282764434814, + "learning_rate": 3.6555645816409426e-06, + "loss": 1.841, + "step": 950 + }, + { + "epoch": 1.0410958904109588, + "eval_loss": 2.0278165340423584, + "eval_runtime": 37.1503, + "eval_samples_per_second": 26.918, + "eval_steps_per_second": 1.131, + "step": 950 + }, + { + "epoch": 1.0421917808219179, + "grad_norm": 2.365743637084961, + "learning_rate": 3.65353371242892e-06, + "loss": 1.8352, + "step": 951 + }, + { + "epoch": 1.0432876712328767, + "grad_norm": 2.6469883918762207, + "learning_rate": 3.651502843216897e-06, + "loss": 1.8814, + "step": 952 + }, + { + "epoch": 1.0443835616438357, + "grad_norm": 2.3646206855773926, + "learning_rate": 3.6494719740048744e-06, + "loss": 1.7282, + "step": 953 + }, + { + "epoch": 1.0454794520547945, + "grad_norm": 2.250856876373291, + "learning_rate": 3.6474411047928517e-06, + "loss": 1.8647, + "step": 954 + }, + { + "epoch": 1.0465753424657533, + "grad_norm": 2.5404837131500244, + "learning_rate": 3.645410235580829e-06, + "loss": 1.8175, + "step": 955 + }, + { + "epoch": 1.0476712328767124, + "grad_norm": 2.4388837814331055, + "learning_rate": 3.6433793663688062e-06, + "loss": 1.8643, + "step": 956 + }, + { + "epoch": 1.0487671232876712, + "grad_norm": 2.285959005355835, + "learning_rate": 3.6413484971567835e-06, + "loss": 1.8367, + "step": 957 + }, + { + "epoch": 1.0498630136986302, + "grad_norm": 2.1190242767333984, + "learning_rate": 3.6393176279447608e-06, + "loss": 1.8242, + "step": 958 + }, + { + "epoch": 1.050958904109589, + "grad_norm": 2.397578716278076, + "learning_rate": 3.637286758732738e-06, + "loss": 1.8681, + "step": 959 + }, + { + "epoch": 1.0520547945205478, + "grad_norm": 2.510599374771118, + "learning_rate": 3.6352558895207153e-06, + "loss": 1.8804, + "step": 960 + }, + { + "epoch": 1.0520547945205478, + "eval_loss": 2.0267090797424316, + "eval_runtime": 37.1226, + "eval_samples_per_second": 26.938, + "eval_steps_per_second": 1.131, + "step": 960 + }, + { + "epoch": 1.0531506849315069, + "grad_norm": 2.4743072986602783, + "learning_rate": 3.633225020308692e-06, + "loss": 1.8895, + "step": 961 + }, + { + "epoch": 1.0542465753424657, + "grad_norm": 2.3666775226593018, + "learning_rate": 3.63119415109667e-06, + "loss": 1.8836, + "step": 962 + }, + { + "epoch": 1.0553424657534247, + "grad_norm": 2.301119327545166, + "learning_rate": 3.6291632818846467e-06, + "loss": 1.8541, + "step": 963 + }, + { + "epoch": 1.0564383561643835, + "grad_norm": 2.482811212539673, + "learning_rate": 3.6271324126726244e-06, + "loss": 1.88, + "step": 964 + }, + { + "epoch": 1.0575342465753426, + "grad_norm": 2.2962276935577393, + "learning_rate": 3.6251015434606013e-06, + "loss": 1.7834, + "step": 965 + }, + { + "epoch": 1.0586301369863014, + "grad_norm": 2.365130662918091, + "learning_rate": 3.623070674248579e-06, + "loss": 1.7946, + "step": 966 + }, + { + "epoch": 1.0597260273972602, + "grad_norm": 2.4644675254821777, + "learning_rate": 3.621039805036556e-06, + "loss": 1.8379, + "step": 967 + }, + { + "epoch": 1.0608219178082192, + "grad_norm": 2.31742787361145, + "learning_rate": 3.6190089358245335e-06, + "loss": 1.8192, + "step": 968 + }, + { + "epoch": 1.061917808219178, + "grad_norm": 2.1614131927490234, + "learning_rate": 3.6169780666125104e-06, + "loss": 1.8794, + "step": 969 + }, + { + "epoch": 1.063013698630137, + "grad_norm": 2.4367032051086426, + "learning_rate": 3.614947197400488e-06, + "loss": 1.8673, + "step": 970 + }, + { + "epoch": 1.063013698630137, + "eval_loss": 2.0261292457580566, + "eval_runtime": 37.1116, + "eval_samples_per_second": 26.946, + "eval_steps_per_second": 1.132, + "step": 970 + }, + { + "epoch": 1.0641095890410959, + "grad_norm": 2.262789726257324, + "learning_rate": 3.612916328188465e-06, + "loss": 1.8808, + "step": 971 + }, + { + "epoch": 1.0652054794520547, + "grad_norm": 2.316316843032837, + "learning_rate": 3.610885458976442e-06, + "loss": 1.7803, + "step": 972 + }, + { + "epoch": 1.0663013698630137, + "grad_norm": 2.2936418056488037, + "learning_rate": 3.6088545897644195e-06, + "loss": 1.8783, + "step": 973 + }, + { + "epoch": 1.0673972602739725, + "grad_norm": 2.667069673538208, + "learning_rate": 3.6068237205523963e-06, + "loss": 1.8748, + "step": 974 + }, + { + "epoch": 1.0684931506849316, + "grad_norm": 2.31111216545105, + "learning_rate": 3.604792851340374e-06, + "loss": 1.8119, + "step": 975 + }, + { + "epoch": 1.0695890410958904, + "grad_norm": 2.268258571624756, + "learning_rate": 3.602761982128351e-06, + "loss": 1.8055, + "step": 976 + }, + { + "epoch": 1.0706849315068494, + "grad_norm": 2.307137966156006, + "learning_rate": 3.6007311129163286e-06, + "loss": 1.8976, + "step": 977 + }, + { + "epoch": 1.0717808219178082, + "grad_norm": 2.406036853790283, + "learning_rate": 3.5987002437043054e-06, + "loss": 1.8608, + "step": 978 + }, + { + "epoch": 1.072876712328767, + "grad_norm": 2.3303062915802, + "learning_rate": 3.596669374492283e-06, + "loss": 1.8038, + "step": 979 + }, + { + "epoch": 1.073972602739726, + "grad_norm": 2.2280168533325195, + "learning_rate": 3.59463850528026e-06, + "loss": 1.8342, + "step": 980 + }, + { + "epoch": 1.073972602739726, + "eval_loss": 2.0273854732513428, + "eval_runtime": 37.1546, + "eval_samples_per_second": 26.915, + "eval_steps_per_second": 1.13, + "step": 980 + }, + { + "epoch": 1.0750684931506849, + "grad_norm": 2.4218759536743164, + "learning_rate": 3.5926076360682377e-06, + "loss": 1.9104, + "step": 981 + }, + { + "epoch": 1.076164383561644, + "grad_norm": 2.2140634059906006, + "learning_rate": 3.5905767668562145e-06, + "loss": 1.7927, + "step": 982 + }, + { + "epoch": 1.0772602739726027, + "grad_norm": 2.179877996444702, + "learning_rate": 3.5885458976441922e-06, + "loss": 1.8868, + "step": 983 + }, + { + "epoch": 1.0783561643835617, + "grad_norm": 2.1988537311553955, + "learning_rate": 3.586515028432169e-06, + "loss": 1.8516, + "step": 984 + }, + { + "epoch": 1.0794520547945206, + "grad_norm": 2.1724448204040527, + "learning_rate": 3.5844841592201464e-06, + "loss": 1.8869, + "step": 985 + }, + { + "epoch": 1.0805479452054794, + "grad_norm": 2.2685718536376953, + "learning_rate": 3.5824532900081237e-06, + "loss": 1.8955, + "step": 986 + }, + { + "epoch": 1.0816438356164384, + "grad_norm": 2.39284348487854, + "learning_rate": 3.580422420796101e-06, + "loss": 1.7769, + "step": 987 + }, + { + "epoch": 1.0827397260273972, + "grad_norm": 2.3448750972747803, + "learning_rate": 3.578391551584078e-06, + "loss": 1.8417, + "step": 988 + }, + { + "epoch": 1.0838356164383562, + "grad_norm": 2.3518283367156982, + "learning_rate": 3.5763606823720555e-06, + "loss": 1.8524, + "step": 989 + }, + { + "epoch": 1.084931506849315, + "grad_norm": 2.1696958541870117, + "learning_rate": 3.5743298131600328e-06, + "loss": 1.7668, + "step": 990 + }, + { + "epoch": 1.084931506849315, + "eval_loss": 2.0244593620300293, + "eval_runtime": 37.1929, + "eval_samples_per_second": 26.887, + "eval_steps_per_second": 1.129, + "step": 990 + }, + { + "epoch": 1.086027397260274, + "grad_norm": 2.18300724029541, + "learning_rate": 3.57229894394801e-06, + "loss": 1.852, + "step": 991 + }, + { + "epoch": 1.087123287671233, + "grad_norm": 2.2075037956237793, + "learning_rate": 3.5702680747359873e-06, + "loss": 1.8799, + "step": 992 + }, + { + "epoch": 1.0882191780821917, + "grad_norm": 2.2172231674194336, + "learning_rate": 3.5682372055239646e-06, + "loss": 1.8573, + "step": 993 + }, + { + "epoch": 1.0893150684931507, + "grad_norm": 2.25169038772583, + "learning_rate": 3.566206336311942e-06, + "loss": 1.8125, + "step": 994 + }, + { + "epoch": 1.0904109589041096, + "grad_norm": 2.168386697769165, + "learning_rate": 3.564175467099919e-06, + "loss": 1.857, + "step": 995 + }, + { + "epoch": 1.0915068493150686, + "grad_norm": 2.2610065937042236, + "learning_rate": 3.5621445978878964e-06, + "loss": 1.817, + "step": 996 + }, + { + "epoch": 1.0926027397260274, + "grad_norm": 2.2561864852905273, + "learning_rate": 3.5601137286758737e-06, + "loss": 1.7754, + "step": 997 + }, + { + "epoch": 1.0936986301369862, + "grad_norm": 2.143625497817993, + "learning_rate": 3.5580828594638505e-06, + "loss": 1.8734, + "step": 998 + }, + { + "epoch": 1.0947945205479452, + "grad_norm": 2.433831214904785, + "learning_rate": 3.5560519902518282e-06, + "loss": 1.8248, + "step": 999 + }, + { + "epoch": 1.095890410958904, + "grad_norm": 2.237497091293335, + "learning_rate": 3.554021121039805e-06, + "loss": 1.8543, + "step": 1000 + }, + { + "epoch": 1.095890410958904, + "eval_loss": 2.020202159881592, + "eval_runtime": 37.2413, + "eval_samples_per_second": 26.852, + "eval_steps_per_second": 1.128, + "step": 1000 + }, + { + "epoch": 1.096986301369863, + "grad_norm": 2.253833532333374, + "learning_rate": 3.551990251827783e-06, + "loss": 1.8799, + "step": 1001 + }, + { + "epoch": 1.098082191780822, + "grad_norm": 2.205322504043579, + "learning_rate": 3.5499593826157596e-06, + "loss": 1.834, + "step": 1002 + }, + { + "epoch": 1.099178082191781, + "grad_norm": 2.3098037242889404, + "learning_rate": 3.5479285134037373e-06, + "loss": 1.8184, + "step": 1003 + }, + { + "epoch": 1.1002739726027397, + "grad_norm": 2.361888885498047, + "learning_rate": 3.545897644191714e-06, + "loss": 1.827, + "step": 1004 + }, + { + "epoch": 1.1013698630136985, + "grad_norm": 2.341947555541992, + "learning_rate": 3.543866774979692e-06, + "loss": 1.8184, + "step": 1005 + }, + { + "epoch": 1.1024657534246576, + "grad_norm": 2.2927780151367188, + "learning_rate": 3.5418359057676687e-06, + "loss": 1.805, + "step": 1006 + }, + { + "epoch": 1.1035616438356164, + "grad_norm": 2.432988166809082, + "learning_rate": 3.5398050365556464e-06, + "loss": 1.8465, + "step": 1007 + }, + { + "epoch": 1.1046575342465754, + "grad_norm": 2.428126335144043, + "learning_rate": 3.5377741673436233e-06, + "loss": 1.8115, + "step": 1008 + }, + { + "epoch": 1.1057534246575342, + "grad_norm": 2.32551646232605, + "learning_rate": 3.535743298131601e-06, + "loss": 1.8522, + "step": 1009 + }, + { + "epoch": 1.106849315068493, + "grad_norm": 2.297980308532715, + "learning_rate": 3.533712428919578e-06, + "loss": 1.8137, + "step": 1010 + }, + { + "epoch": 1.106849315068493, + "eval_loss": 2.0190494060516357, + "eval_runtime": 37.2149, + "eval_samples_per_second": 26.871, + "eval_steps_per_second": 1.129, + "step": 1010 + }, + { + "epoch": 1.107945205479452, + "grad_norm": 2.364342451095581, + "learning_rate": 3.5316815597075547e-06, + "loss": 1.8677, + "step": 1011 + }, + { + "epoch": 1.1090410958904109, + "grad_norm": 2.355008125305176, + "learning_rate": 3.5296506904955324e-06, + "loss": 1.8937, + "step": 1012 + }, + { + "epoch": 1.11013698630137, + "grad_norm": 2.2288317680358887, + "learning_rate": 3.5276198212835093e-06, + "loss": 1.8188, + "step": 1013 + }, + { + "epoch": 1.1112328767123287, + "grad_norm": 2.2801361083984375, + "learning_rate": 3.525588952071487e-06, + "loss": 1.8386, + "step": 1014 + }, + { + "epoch": 1.1123287671232878, + "grad_norm": 2.2801361083984375, + "learning_rate": 3.525588952071487e-06, + "loss": 1.8845, + "step": 1015 + }, + { + "epoch": 1.1134246575342466, + "grad_norm": 2.4301352500915527, + "learning_rate": 3.523558082859464e-06, + "loss": 1.8898, + "step": 1016 + }, + { + "epoch": 1.1145205479452054, + "grad_norm": 2.3492274284362793, + "learning_rate": 3.5215272136474415e-06, + "loss": 1.8711, + "step": 1017 + }, + { + "epoch": 1.1156164383561644, + "grad_norm": 2.1788229942321777, + "learning_rate": 3.5194963444354184e-06, + "loss": 1.8177, + "step": 1018 + }, + { + "epoch": 1.1167123287671232, + "grad_norm": 2.269767999649048, + "learning_rate": 3.517465475223396e-06, + "loss": 1.8259, + "step": 1019 + }, + { + "epoch": 1.1178082191780823, + "grad_norm": 2.2890219688415527, + "learning_rate": 3.515434606011373e-06, + "loss": 1.7873, + "step": 1020 + }, + { + "epoch": 1.1178082191780823, + "eval_loss": 2.023195266723633, + "eval_runtime": 37.5701, + "eval_samples_per_second": 26.617, + "eval_steps_per_second": 1.118, + "step": 1020 + }, + { + "epoch": 1.118904109589041, + "grad_norm": 2.5528533458709717, + "learning_rate": 3.5134037367993506e-06, + "loss": 1.8624, + "step": 1021 + }, + { + "epoch": 1.12, + "grad_norm": 2.5109457969665527, + "learning_rate": 3.5113728675873275e-06, + "loss": 1.9178, + "step": 1022 + }, + { + "epoch": 1.121095890410959, + "grad_norm": 2.288877010345459, + "learning_rate": 3.5093419983753047e-06, + "loss": 1.8604, + "step": 1023 + }, + { + "epoch": 1.1221917808219177, + "grad_norm": 2.460153818130493, + "learning_rate": 3.507311129163282e-06, + "loss": 1.8551, + "step": 1024 + }, + { + "epoch": 1.1232876712328768, + "grad_norm": 2.3520290851593018, + "learning_rate": 3.5052802599512593e-06, + "loss": 1.7818, + "step": 1025 + }, + { + "epoch": 1.1243835616438356, + "grad_norm": 2.32812237739563, + "learning_rate": 3.5032493907392366e-06, + "loss": 1.8255, + "step": 1026 + }, + { + "epoch": 1.1254794520547946, + "grad_norm": 2.3118042945861816, + "learning_rate": 3.501218521527214e-06, + "loss": 1.8493, + "step": 1027 + }, + { + "epoch": 1.1265753424657534, + "grad_norm": 2.2750277519226074, + "learning_rate": 3.499187652315191e-06, + "loss": 1.7759, + "step": 1028 + }, + { + "epoch": 1.1276712328767124, + "grad_norm": 2.272120475769043, + "learning_rate": 3.4971567831031684e-06, + "loss": 1.8399, + "step": 1029 + }, + { + "epoch": 1.1287671232876713, + "grad_norm": 2.2075254917144775, + "learning_rate": 3.4951259138911457e-06, + "loss": 1.8774, + "step": 1030 + }, + { + "epoch": 1.1287671232876713, + "eval_loss": 2.0199804306030273, + "eval_runtime": 37.6753, + "eval_samples_per_second": 26.543, + "eval_steps_per_second": 1.115, + "step": 1030 + }, + { + "epoch": 1.12986301369863, + "grad_norm": 2.359877586364746, + "learning_rate": 3.493095044679123e-06, + "loss": 1.7942, + "step": 1031 + }, + { + "epoch": 1.130958904109589, + "grad_norm": 2.3057703971862793, + "learning_rate": 3.4910641754671002e-06, + "loss": 1.8996, + "step": 1032 + }, + { + "epoch": 1.132054794520548, + "grad_norm": 2.3109569549560547, + "learning_rate": 3.4890333062550775e-06, + "loss": 1.7336, + "step": 1033 + }, + { + "epoch": 1.133150684931507, + "grad_norm": 2.3021819591522217, + "learning_rate": 3.4870024370430548e-06, + "loss": 1.869, + "step": 1034 + }, + { + "epoch": 1.1342465753424658, + "grad_norm": 2.4741852283477783, + "learning_rate": 3.484971567831032e-06, + "loss": 1.7652, + "step": 1035 + }, + { + "epoch": 1.1353424657534246, + "grad_norm": 2.326500415802002, + "learning_rate": 3.482940698619009e-06, + "loss": 1.809, + "step": 1036 + }, + { + "epoch": 1.1364383561643836, + "grad_norm": 2.2562193870544434, + "learning_rate": 3.4809098294069866e-06, + "loss": 1.8806, + "step": 1037 + }, + { + "epoch": 1.1375342465753424, + "grad_norm": 2.2849345207214355, + "learning_rate": 3.4788789601949634e-06, + "loss": 1.8329, + "step": 1038 + }, + { + "epoch": 1.1386301369863014, + "grad_norm": 2.333707332611084, + "learning_rate": 3.476848090982941e-06, + "loss": 1.8541, + "step": 1039 + }, + { + "epoch": 1.1397260273972603, + "grad_norm": 2.234175682067871, + "learning_rate": 3.474817221770918e-06, + "loss": 1.8126, + "step": 1040 + }, + { + "epoch": 1.1397260273972603, + "eval_loss": 2.0176141262054443, + "eval_runtime": 37.232, + "eval_samples_per_second": 26.859, + "eval_steps_per_second": 1.128, + "step": 1040 + }, + { + "epoch": 1.140821917808219, + "grad_norm": 2.4212377071380615, + "learning_rate": 3.4727863525588957e-06, + "loss": 1.9325, + "step": 1041 + }, + { + "epoch": 1.141917808219178, + "grad_norm": 2.1526010036468506, + "learning_rate": 3.4707554833468725e-06, + "loss": 1.8674, + "step": 1042 + }, + { + "epoch": 1.143013698630137, + "grad_norm": 2.336186408996582, + "learning_rate": 3.4687246141348502e-06, + "loss": 1.8275, + "step": 1043 + }, + { + "epoch": 1.144109589041096, + "grad_norm": 2.2991745471954346, + "learning_rate": 3.466693744922827e-06, + "loss": 1.8057, + "step": 1044 + }, + { + "epoch": 1.1452054794520548, + "grad_norm": 2.3816370964050293, + "learning_rate": 3.464662875710805e-06, + "loss": 1.8212, + "step": 1045 + }, + { + "epoch": 1.1463013698630138, + "grad_norm": 2.2383193969726562, + "learning_rate": 3.4626320064987817e-06, + "loss": 1.8444, + "step": 1046 + }, + { + "epoch": 1.1473972602739726, + "grad_norm": 2.377574920654297, + "learning_rate": 3.4606011372867594e-06, + "loss": 1.8319, + "step": 1047 + }, + { + "epoch": 1.1484931506849314, + "grad_norm": 2.1803700923919678, + "learning_rate": 3.458570268074736e-06, + "loss": 1.8484, + "step": 1048 + }, + { + "epoch": 1.1495890410958904, + "grad_norm": 2.1810641288757324, + "learning_rate": 3.4565393988627135e-06, + "loss": 1.8187, + "step": 1049 + }, + { + "epoch": 1.1506849315068493, + "grad_norm": 2.1565394401550293, + "learning_rate": 3.4545085296506908e-06, + "loss": 1.8117, + "step": 1050 + }, + { + "epoch": 1.1506849315068493, + "eval_loss": 2.0156562328338623, + "eval_runtime": 37.2072, + "eval_samples_per_second": 26.877, + "eval_steps_per_second": 1.129, + "step": 1050 + }, + { + "epoch": 1.1517808219178083, + "grad_norm": 2.1753759384155273, + "learning_rate": 3.452477660438668e-06, + "loss": 1.7822, + "step": 1051 + }, + { + "epoch": 1.152876712328767, + "grad_norm": 2.250089645385742, + "learning_rate": 3.4504467912266453e-06, + "loss": 1.8634, + "step": 1052 + }, + { + "epoch": 1.1539726027397261, + "grad_norm": 2.3043293952941895, + "learning_rate": 3.4484159220146226e-06, + "loss": 1.7743, + "step": 1053 + }, + { + "epoch": 1.155068493150685, + "grad_norm": 2.2193732261657715, + "learning_rate": 3.4463850528026e-06, + "loss": 1.8328, + "step": 1054 + }, + { + "epoch": 1.1561643835616437, + "grad_norm": 2.4204866886138916, + "learning_rate": 3.444354183590577e-06, + "loss": 1.8133, + "step": 1055 + }, + { + "epoch": 1.1572602739726028, + "grad_norm": 2.3088018894195557, + "learning_rate": 3.4423233143785544e-06, + "loss": 1.8523, + "step": 1056 + }, + { + "epoch": 1.1583561643835616, + "grad_norm": 2.38910174369812, + "learning_rate": 3.4402924451665317e-06, + "loss": 1.7846, + "step": 1057 + }, + { + "epoch": 1.1594520547945206, + "grad_norm": 2.2473702430725098, + "learning_rate": 3.438261575954509e-06, + "loss": 1.7632, + "step": 1058 + }, + { + "epoch": 1.1605479452054794, + "grad_norm": 2.3356645107269287, + "learning_rate": 3.4362307067424862e-06, + "loss": 1.8461, + "step": 1059 + }, + { + "epoch": 1.1616438356164385, + "grad_norm": 2.2460737228393555, + "learning_rate": 3.434199837530463e-06, + "loss": 1.8853, + "step": 1060 + }, + { + "epoch": 1.1616438356164385, + "eval_loss": 2.016077995300293, + "eval_runtime": 37.1943, + "eval_samples_per_second": 26.886, + "eval_steps_per_second": 1.129, + "step": 1060 + }, + { + "epoch": 1.1627397260273973, + "grad_norm": 2.28434157371521, + "learning_rate": 3.432168968318441e-06, + "loss": 1.8451, + "step": 1061 + }, + { + "epoch": 1.163835616438356, + "grad_norm": 2.299049139022827, + "learning_rate": 3.4301380991064176e-06, + "loss": 1.7777, + "step": 1062 + }, + { + "epoch": 1.1649315068493151, + "grad_norm": 2.364074468612671, + "learning_rate": 3.4281072298943953e-06, + "loss": 1.8517, + "step": 1063 + }, + { + "epoch": 1.166027397260274, + "grad_norm": 2.2381322383880615, + "learning_rate": 3.426076360682372e-06, + "loss": 1.889, + "step": 1064 + }, + { + "epoch": 1.167123287671233, + "grad_norm": 2.2984025478363037, + "learning_rate": 3.42404549147035e-06, + "loss": 1.8907, + "step": 1065 + }, + { + "epoch": 1.1682191780821918, + "grad_norm": 2.2215473651885986, + "learning_rate": 3.4220146222583267e-06, + "loss": 1.8313, + "step": 1066 + }, + { + "epoch": 1.1693150684931506, + "grad_norm": 2.3414270877838135, + "learning_rate": 3.4199837530463044e-06, + "loss": 1.783, + "step": 1067 + }, + { + "epoch": 1.1704109589041096, + "grad_norm": 2.0636470317840576, + "learning_rate": 3.4179528838342813e-06, + "loss": 1.8185, + "step": 1068 + }, + { + "epoch": 1.1715068493150684, + "grad_norm": 2.2158873081207275, + "learning_rate": 3.415922014622259e-06, + "loss": 1.9319, + "step": 1069 + }, + { + "epoch": 1.1726027397260275, + "grad_norm": 2.244051694869995, + "learning_rate": 3.413891145410236e-06, + "loss": 1.8732, + "step": 1070 + }, + { + "epoch": 1.1726027397260275, + "eval_loss": 2.0180234909057617, + "eval_runtime": 37.1822, + "eval_samples_per_second": 26.895, + "eval_steps_per_second": 1.13, + "step": 1070 + }, + { + "epoch": 1.1736986301369863, + "grad_norm": 2.1608433723449707, + "learning_rate": 3.4118602761982135e-06, + "loss": 1.8709, + "step": 1071 + }, + { + "epoch": 1.174794520547945, + "grad_norm": 2.2612338066101074, + "learning_rate": 3.4098294069861904e-06, + "loss": 1.8882, + "step": 1072 + }, + { + "epoch": 1.1758904109589041, + "grad_norm": 2.3257741928100586, + "learning_rate": 3.4077985377741673e-06, + "loss": 1.8377, + "step": 1073 + }, + { + "epoch": 1.176986301369863, + "grad_norm": 2.2486586570739746, + "learning_rate": 3.405767668562145e-06, + "loss": 1.8105, + "step": 1074 + }, + { + "epoch": 1.178082191780822, + "grad_norm": 2.2840020656585693, + "learning_rate": 3.403736799350122e-06, + "loss": 1.8868, + "step": 1075 + }, + { + "epoch": 1.1791780821917808, + "grad_norm": 2.3264284133911133, + "learning_rate": 3.4017059301380995e-06, + "loss": 1.8701, + "step": 1076 + }, + { + "epoch": 1.1802739726027398, + "grad_norm": 2.311424493789673, + "learning_rate": 3.3996750609260764e-06, + "loss": 1.801, + "step": 1077 + }, + { + "epoch": 1.1813698630136986, + "grad_norm": 2.2177181243896484, + "learning_rate": 3.397644191714054e-06, + "loss": 1.9188, + "step": 1078 + }, + { + "epoch": 1.1824657534246574, + "grad_norm": 2.413118362426758, + "learning_rate": 3.395613322502031e-06, + "loss": 1.8198, + "step": 1079 + }, + { + "epoch": 1.1835616438356165, + "grad_norm": 2.334688186645508, + "learning_rate": 3.3935824532900086e-06, + "loss": 1.822, + "step": 1080 + }, + { + "epoch": 1.1835616438356165, + "eval_loss": 2.012676954269409, + "eval_runtime": 37.2415, + "eval_samples_per_second": 26.852, + "eval_steps_per_second": 1.128, + "step": 1080 + }, + { + "epoch": 1.1846575342465753, + "grad_norm": 2.3901584148406982, + "learning_rate": 3.3915515840779855e-06, + "loss": 1.791, + "step": 1081 + }, + { + "epoch": 1.1857534246575343, + "grad_norm": 2.2731425762176514, + "learning_rate": 3.389520714865963e-06, + "loss": 1.8367, + "step": 1082 + }, + { + "epoch": 1.1868493150684931, + "grad_norm": 2.311553478240967, + "learning_rate": 3.38748984565394e-06, + "loss": 1.8302, + "step": 1083 + }, + { + "epoch": 1.1879452054794521, + "grad_norm": 2.3030776977539062, + "learning_rate": 3.3854589764419177e-06, + "loss": 1.8652, + "step": 1084 + }, + { + "epoch": 1.189041095890411, + "grad_norm": 2.341188430786133, + "learning_rate": 3.3834281072298946e-06, + "loss": 1.8094, + "step": 1085 + }, + { + "epoch": 1.1901369863013698, + "grad_norm": 2.331962823867798, + "learning_rate": 3.381397238017872e-06, + "loss": 1.827, + "step": 1086 + }, + { + "epoch": 1.1912328767123288, + "grad_norm": 2.275923728942871, + "learning_rate": 3.379366368805849e-06, + "loss": 1.7935, + "step": 1087 + }, + { + "epoch": 1.1923287671232876, + "grad_norm": 2.4608163833618164, + "learning_rate": 3.3773354995938264e-06, + "loss": 1.8965, + "step": 1088 + }, + { + "epoch": 1.1934246575342466, + "grad_norm": 2.2208592891693115, + "learning_rate": 3.3753046303818037e-06, + "loss": 1.8447, + "step": 1089 + }, + { + "epoch": 1.1945205479452055, + "grad_norm": 2.4284770488739014, + "learning_rate": 3.373273761169781e-06, + "loss": 1.8306, + "step": 1090 + }, + { + "epoch": 1.1945205479452055, + "eval_loss": 2.010664939880371, + "eval_runtime": 37.237, + "eval_samples_per_second": 26.855, + "eval_steps_per_second": 1.128, + "step": 1090 + }, + { + "epoch": 1.1956164383561645, + "grad_norm": 2.329601764678955, + "learning_rate": 3.3712428919577582e-06, + "loss": 1.8275, + "step": 1091 + }, + { + "epoch": 1.1967123287671233, + "grad_norm": 2.277939796447754, + "learning_rate": 3.3692120227457355e-06, + "loss": 1.8161, + "step": 1092 + }, + { + "epoch": 1.1978082191780821, + "grad_norm": 2.446780204772949, + "learning_rate": 3.3671811535337128e-06, + "loss": 1.8511, + "step": 1093 + }, + { + "epoch": 1.1989041095890411, + "grad_norm": 2.382477283477783, + "learning_rate": 3.36515028432169e-06, + "loss": 1.7834, + "step": 1094 + }, + { + "epoch": 1.2, + "grad_norm": 2.366361379623413, + "learning_rate": 3.3631194151096673e-06, + "loss": 1.8105, + "step": 1095 + }, + { + "epoch": 1.201095890410959, + "grad_norm": 2.3439717292785645, + "learning_rate": 3.3610885458976446e-06, + "loss": 1.8418, + "step": 1096 + }, + { + "epoch": 1.2021917808219178, + "grad_norm": 2.2739830017089844, + "learning_rate": 3.359057676685622e-06, + "loss": 1.8818, + "step": 1097 + }, + { + "epoch": 1.2032876712328768, + "grad_norm": 2.1636738777160645, + "learning_rate": 3.357026807473599e-06, + "loss": 1.8757, + "step": 1098 + }, + { + "epoch": 1.2043835616438356, + "grad_norm": 2.386441946029663, + "learning_rate": 3.354995938261576e-06, + "loss": 1.756, + "step": 1099 + }, + { + "epoch": 1.2054794520547945, + "grad_norm": 2.2241456508636475, + "learning_rate": 3.3529650690495537e-06, + "loss": 1.8034, + "step": 1100 + }, + { + "epoch": 1.2054794520547945, + "eval_loss": 2.008504867553711, + "eval_runtime": 37.2855, + "eval_samples_per_second": 26.82, + "eval_steps_per_second": 1.126, + "step": 1100 + }, + { + "epoch": 1.2065753424657535, + "grad_norm": 2.196992874145508, + "learning_rate": 3.3509341998375305e-06, + "loss": 1.7722, + "step": 1101 + }, + { + "epoch": 1.2076712328767123, + "grad_norm": 2.3572094440460205, + "learning_rate": 3.3489033306255082e-06, + "loss": 1.8109, + "step": 1102 + }, + { + "epoch": 1.208767123287671, + "grad_norm": 2.443281650543213, + "learning_rate": 3.346872461413485e-06, + "loss": 1.843, + "step": 1103 + }, + { + "epoch": 1.2098630136986301, + "grad_norm": 2.2613234519958496, + "learning_rate": 3.344841592201463e-06, + "loss": 1.8023, + "step": 1104 + }, + { + "epoch": 1.210958904109589, + "grad_norm": 2.2466726303100586, + "learning_rate": 3.3428107229894397e-06, + "loss": 1.7838, + "step": 1105 + }, + { + "epoch": 1.212054794520548, + "grad_norm": 2.2314462661743164, + "learning_rate": 3.3407798537774174e-06, + "loss": 1.8077, + "step": 1106 + }, + { + "epoch": 1.2131506849315068, + "grad_norm": 2.3288400173187256, + "learning_rate": 3.338748984565394e-06, + "loss": 1.8319, + "step": 1107 + }, + { + "epoch": 1.2142465753424658, + "grad_norm": 2.3477249145507812, + "learning_rate": 3.336718115353372e-06, + "loss": 1.8772, + "step": 1108 + }, + { + "epoch": 1.2153424657534246, + "grad_norm": 2.3588409423828125, + "learning_rate": 3.3346872461413488e-06, + "loss": 1.8832, + "step": 1109 + }, + { + "epoch": 1.2164383561643834, + "grad_norm": 2.327796220779419, + "learning_rate": 3.3326563769293256e-06, + "loss": 1.8033, + "step": 1110 + }, + { + "epoch": 1.2164383561643834, + "eval_loss": 2.0094382762908936, + "eval_runtime": 37.2666, + "eval_samples_per_second": 26.834, + "eval_steps_per_second": 1.127, + "step": 1110 + }, + { + "epoch": 1.2175342465753425, + "grad_norm": 2.257244825363159, + "learning_rate": 3.3306255077173033e-06, + "loss": 1.7903, + "step": 1111 + }, + { + "epoch": 1.2186301369863013, + "grad_norm": 2.3321614265441895, + "learning_rate": 3.32859463850528e-06, + "loss": 1.8295, + "step": 1112 + }, + { + "epoch": 1.2197260273972603, + "grad_norm": 2.4073803424835205, + "learning_rate": 3.326563769293258e-06, + "loss": 1.8395, + "step": 1113 + }, + { + "epoch": 1.2208219178082191, + "grad_norm": 2.3631434440612793, + "learning_rate": 3.3245329000812347e-06, + "loss": 1.7548, + "step": 1114 + }, + { + "epoch": 1.2219178082191782, + "grad_norm": 2.275766134262085, + "learning_rate": 3.3225020308692124e-06, + "loss": 1.808, + "step": 1115 + }, + { + "epoch": 1.223013698630137, + "grad_norm": 2.438011407852173, + "learning_rate": 3.3204711616571893e-06, + "loss": 1.8274, + "step": 1116 + }, + { + "epoch": 1.2241095890410958, + "grad_norm": 2.371826171875, + "learning_rate": 3.318440292445167e-06, + "loss": 1.7742, + "step": 1117 + }, + { + "epoch": 1.2252054794520548, + "grad_norm": 2.422004461288452, + "learning_rate": 3.316409423233144e-06, + "loss": 1.8268, + "step": 1118 + }, + { + "epoch": 1.2263013698630136, + "grad_norm": 2.377979278564453, + "learning_rate": 3.3143785540211215e-06, + "loss": 1.7857, + "step": 1119 + }, + { + "epoch": 1.2273972602739727, + "grad_norm": 2.288482904434204, + "learning_rate": 3.3123476848090984e-06, + "loss": 1.876, + "step": 1120 + }, + { + "epoch": 1.2273972602739727, + "eval_loss": 2.0104730129241943, + "eval_runtime": 37.5365, + "eval_samples_per_second": 26.641, + "eval_steps_per_second": 1.119, + "step": 1120 + }, + { + "epoch": 1.2284931506849315, + "grad_norm": 2.451533555984497, + "learning_rate": 3.310316815597076e-06, + "loss": 1.8324, + "step": 1121 + }, + { + "epoch": 1.2295890410958905, + "grad_norm": 2.3785181045532227, + "learning_rate": 3.308285946385053e-06, + "loss": 1.8239, + "step": 1122 + }, + { + "epoch": 1.2306849315068493, + "grad_norm": 2.2276721000671387, + "learning_rate": 3.30625507717303e-06, + "loss": 1.9132, + "step": 1123 + }, + { + "epoch": 1.2317808219178081, + "grad_norm": 2.2575314044952393, + "learning_rate": 3.3042242079610075e-06, + "loss": 1.769, + "step": 1124 + }, + { + "epoch": 1.2328767123287672, + "grad_norm": 2.5311717987060547, + "learning_rate": 3.3021933387489847e-06, + "loss": 1.8605, + "step": 1125 + }, + { + "epoch": 1.233972602739726, + "grad_norm": 2.2395896911621094, + "learning_rate": 3.300162469536962e-06, + "loss": 1.8725, + "step": 1126 + }, + { + "epoch": 1.235068493150685, + "grad_norm": 2.2402031421661377, + "learning_rate": 3.2981316003249393e-06, + "loss": 1.7955, + "step": 1127 + }, + { + "epoch": 1.2361643835616438, + "grad_norm": 2.308950424194336, + "learning_rate": 3.2961007311129166e-06, + "loss": 1.8812, + "step": 1128 + }, + { + "epoch": 1.2372602739726029, + "grad_norm": 2.437767267227173, + "learning_rate": 3.294069861900894e-06, + "loss": 1.8489, + "step": 1129 + }, + { + "epoch": 1.2383561643835617, + "grad_norm": 2.3188693523406982, + "learning_rate": 3.292038992688871e-06, + "loss": 1.9082, + "step": 1130 + }, + { + "epoch": 1.2383561643835617, + "eval_loss": 2.0110676288604736, + "eval_runtime": 37.6437, + "eval_samples_per_second": 26.565, + "eval_steps_per_second": 1.116, + "step": 1130 + }, + { + "epoch": 1.2394520547945205, + "grad_norm": 2.312760591506958, + "learning_rate": 3.2900081234768484e-06, + "loss": 1.8344, + "step": 1131 + }, + { + "epoch": 1.2405479452054795, + "grad_norm": 2.5837206840515137, + "learning_rate": 3.2879772542648257e-06, + "loss": 1.8503, + "step": 1132 + }, + { + "epoch": 1.2416438356164383, + "grad_norm": 2.1864733695983887, + "learning_rate": 3.285946385052803e-06, + "loss": 1.8822, + "step": 1133 + }, + { + "epoch": 1.2427397260273974, + "grad_norm": 2.2644550800323486, + "learning_rate": 3.2839155158407802e-06, + "loss": 1.7541, + "step": 1134 + }, + { + "epoch": 1.2438356164383562, + "grad_norm": 2.292646646499634, + "learning_rate": 3.2818846466287575e-06, + "loss": 1.8575, + "step": 1135 + }, + { + "epoch": 1.244931506849315, + "grad_norm": 2.2162070274353027, + "learning_rate": 3.2798537774167344e-06, + "loss": 1.7525, + "step": 1136 + }, + { + "epoch": 1.246027397260274, + "grad_norm": 2.32688307762146, + "learning_rate": 3.277822908204712e-06, + "loss": 1.8955, + "step": 1137 + }, + { + "epoch": 1.2471232876712328, + "grad_norm": 2.3226473331451416, + "learning_rate": 3.275792038992689e-06, + "loss": 1.8255, + "step": 1138 + }, + { + "epoch": 1.2482191780821918, + "grad_norm": 2.3533706665039062, + "learning_rate": 3.2737611697806666e-06, + "loss": 1.783, + "step": 1139 + }, + { + "epoch": 1.2493150684931507, + "grad_norm": 2.237661123275757, + "learning_rate": 3.2717303005686435e-06, + "loss": 1.8367, + "step": 1140 + }, + { + "epoch": 1.2493150684931507, + "eval_loss": 2.0123534202575684, + "eval_runtime": 37.3573, + "eval_samples_per_second": 26.769, + "eval_steps_per_second": 1.124, + "step": 1140 + }, + { + "epoch": 1.2504109589041095, + "grad_norm": 2.283047676086426, + "learning_rate": 3.269699431356621e-06, + "loss": 1.7987, + "step": 1141 + }, + { + "epoch": 1.2515068493150685, + "grad_norm": 2.2264456748962402, + "learning_rate": 3.267668562144598e-06, + "loss": 1.8307, + "step": 1142 + }, + { + "epoch": 1.2526027397260273, + "grad_norm": 2.2724905014038086, + "learning_rate": 3.2656376929325757e-06, + "loss": 1.8237, + "step": 1143 + }, + { + "epoch": 1.2536986301369863, + "grad_norm": 2.270782947540283, + "learning_rate": 3.2636068237205526e-06, + "loss": 1.7846, + "step": 1144 + }, + { + "epoch": 1.2547945205479452, + "grad_norm": 2.2032551765441895, + "learning_rate": 3.2615759545085303e-06, + "loss": 1.8438, + "step": 1145 + }, + { + "epoch": 1.2558904109589042, + "grad_norm": 2.199150800704956, + "learning_rate": 3.259545085296507e-06, + "loss": 1.8199, + "step": 1146 + }, + { + "epoch": 1.256986301369863, + "grad_norm": 2.4404170513153076, + "learning_rate": 3.257514216084484e-06, + "loss": 1.8817, + "step": 1147 + }, + { + "epoch": 1.2580821917808218, + "grad_norm": 2.281348943710327, + "learning_rate": 3.2554833468724617e-06, + "loss": 1.771, + "step": 1148 + }, + { + "epoch": 1.2591780821917808, + "grad_norm": 2.19482159614563, + "learning_rate": 3.2534524776604385e-06, + "loss": 1.8684, + "step": 1149 + }, + { + "epoch": 1.2602739726027397, + "grad_norm": 2.221618413925171, + "learning_rate": 3.2514216084484162e-06, + "loss": 1.8398, + "step": 1150 + }, + { + "epoch": 1.2602739726027397, + "eval_loss": 2.007969617843628, + "eval_runtime": 37.3679, + "eval_samples_per_second": 26.761, + "eval_steps_per_second": 1.124, + "step": 1150 + }, + { + "epoch": 1.2613698630136987, + "grad_norm": 2.193432569503784, + "learning_rate": 3.249390739236393e-06, + "loss": 1.8122, + "step": 1151 + }, + { + "epoch": 1.2624657534246575, + "grad_norm": 2.246462106704712, + "learning_rate": 3.2473598700243708e-06, + "loss": 1.8447, + "step": 1152 + }, + { + "epoch": 1.2635616438356165, + "grad_norm": 2.3411879539489746, + "learning_rate": 3.2453290008123476e-06, + "loss": 1.7988, + "step": 1153 + }, + { + "epoch": 1.2646575342465753, + "grad_norm": 2.245286464691162, + "learning_rate": 3.2432981316003253e-06, + "loss": 1.8209, + "step": 1154 + }, + { + "epoch": 1.2657534246575342, + "grad_norm": 2.292067527770996, + "learning_rate": 3.241267262388302e-06, + "loss": 1.8285, + "step": 1155 + }, + { + "epoch": 1.2668493150684932, + "grad_norm": 2.3058111667633057, + "learning_rate": 3.23923639317628e-06, + "loss": 1.7842, + "step": 1156 + }, + { + "epoch": 1.267945205479452, + "grad_norm": 2.306243896484375, + "learning_rate": 3.2372055239642567e-06, + "loss": 1.8221, + "step": 1157 + }, + { + "epoch": 1.269041095890411, + "grad_norm": 2.399759531021118, + "learning_rate": 3.2351746547522344e-06, + "loss": 1.7836, + "step": 1158 + }, + { + "epoch": 1.2701369863013698, + "grad_norm": 2.3405187129974365, + "learning_rate": 3.2331437855402113e-06, + "loss": 1.8365, + "step": 1159 + }, + { + "epoch": 1.2712328767123289, + "grad_norm": 2.3855013847351074, + "learning_rate": 3.2311129163281885e-06, + "loss": 1.8973, + "step": 1160 + }, + { + "epoch": 1.2712328767123289, + "eval_loss": 2.005983591079712, + "eval_runtime": 37.3061, + "eval_samples_per_second": 26.805, + "eval_steps_per_second": 1.126, + "step": 1160 + }, + { + "epoch": 1.2723287671232877, + "grad_norm": 2.2767140865325928, + "learning_rate": 3.229082047116166e-06, + "loss": 1.7791, + "step": 1161 + }, + { + "epoch": 1.2734246575342465, + "grad_norm": 2.289494276046753, + "learning_rate": 3.227051177904143e-06, + "loss": 1.8508, + "step": 1162 + }, + { + "epoch": 1.2745205479452055, + "grad_norm": 2.1836023330688477, + "learning_rate": 3.2250203086921204e-06, + "loss": 1.8275, + "step": 1163 + }, + { + "epoch": 1.2756164383561643, + "grad_norm": 2.1917505264282227, + "learning_rate": 3.2229894394800977e-06, + "loss": 1.8289, + "step": 1164 + }, + { + "epoch": 1.2767123287671232, + "grad_norm": 2.315983295440674, + "learning_rate": 3.220958570268075e-06, + "loss": 1.8655, + "step": 1165 + }, + { + "epoch": 1.2778082191780822, + "grad_norm": 2.3628008365631104, + "learning_rate": 3.218927701056052e-06, + "loss": 1.7396, + "step": 1166 + }, + { + "epoch": 1.2789041095890412, + "grad_norm": 2.3200488090515137, + "learning_rate": 3.2168968318440295e-06, + "loss": 1.8196, + "step": 1167 + }, + { + "epoch": 1.28, + "grad_norm": 2.2591991424560547, + "learning_rate": 3.2148659626320068e-06, + "loss": 1.7323, + "step": 1168 + }, + { + "epoch": 1.2810958904109588, + "grad_norm": 2.2458770275115967, + "learning_rate": 3.212835093419984e-06, + "loss": 1.8881, + "step": 1169 + }, + { + "epoch": 1.2821917808219179, + "grad_norm": 2.202010154724121, + "learning_rate": 3.2108042242079613e-06, + "loss": 1.7776, + "step": 1170 + }, + { + "epoch": 1.2821917808219179, + "eval_loss": 2.005174398422241, + "eval_runtime": 37.3597, + "eval_samples_per_second": 26.767, + "eval_steps_per_second": 1.124, + "step": 1170 + }, + { + "epoch": 1.2832876712328767, + "grad_norm": 2.2375359535217285, + "learning_rate": 3.2087733549959386e-06, + "loss": 1.7994, + "step": 1171 + }, + { + "epoch": 1.2843835616438355, + "grad_norm": 2.3152217864990234, + "learning_rate": 3.206742485783916e-06, + "loss": 1.8767, + "step": 1172 + }, + { + "epoch": 1.2854794520547945, + "grad_norm": 2.3317618370056152, + "learning_rate": 3.2047116165718927e-06, + "loss": 1.772, + "step": 1173 + }, + { + "epoch": 1.2865753424657536, + "grad_norm": 2.249943971633911, + "learning_rate": 3.2026807473598704e-06, + "loss": 1.7793, + "step": 1174 + }, + { + "epoch": 1.2876712328767124, + "grad_norm": 2.320481538772583, + "learning_rate": 3.2006498781478473e-06, + "loss": 1.8275, + "step": 1175 + }, + { + "epoch": 1.2887671232876712, + "grad_norm": 2.21785569190979, + "learning_rate": 3.198619008935825e-06, + "loss": 1.8449, + "step": 1176 + }, + { + "epoch": 1.2898630136986302, + "grad_norm": 2.469965934753418, + "learning_rate": 3.196588139723802e-06, + "loss": 1.779, + "step": 1177 + }, + { + "epoch": 1.290958904109589, + "grad_norm": 2.1989049911499023, + "learning_rate": 3.1945572705117795e-06, + "loss": 1.8188, + "step": 1178 + }, + { + "epoch": 1.2920547945205478, + "grad_norm": 2.293513298034668, + "learning_rate": 3.1925264012997564e-06, + "loss": 1.8377, + "step": 1179 + }, + { + "epoch": 1.2931506849315069, + "grad_norm": 2.3462443351745605, + "learning_rate": 3.190495532087734e-06, + "loss": 1.8418, + "step": 1180 + }, + { + "epoch": 1.2931506849315069, + "eval_loss": 2.0061001777648926, + "eval_runtime": 37.2951, + "eval_samples_per_second": 26.813, + "eval_steps_per_second": 1.126, + "step": 1180 + }, + { + "epoch": 1.2942465753424657, + "grad_norm": 2.2071478366851807, + "learning_rate": 3.188464662875711e-06, + "loss": 1.8419, + "step": 1181 + }, + { + "epoch": 1.2953424657534247, + "grad_norm": 2.233333110809326, + "learning_rate": 3.1864337936636886e-06, + "loss": 1.8931, + "step": 1182 + }, + { + "epoch": 1.2964383561643835, + "grad_norm": 2.383085250854492, + "learning_rate": 3.1844029244516655e-06, + "loss": 1.8047, + "step": 1183 + }, + { + "epoch": 1.2975342465753426, + "grad_norm": 2.473808765411377, + "learning_rate": 3.182372055239643e-06, + "loss": 1.7794, + "step": 1184 + }, + { + "epoch": 1.2986301369863014, + "grad_norm": 2.3357670307159424, + "learning_rate": 3.18034118602762e-06, + "loss": 1.7717, + "step": 1185 + }, + { + "epoch": 1.2997260273972602, + "grad_norm": 2.376835823059082, + "learning_rate": 3.1783103168155973e-06, + "loss": 1.8534, + "step": 1186 + }, + { + "epoch": 1.3008219178082192, + "grad_norm": 2.447415590286255, + "learning_rate": 3.1762794476035746e-06, + "loss": 1.8251, + "step": 1187 + }, + { + "epoch": 1.301917808219178, + "grad_norm": 2.5145719051361084, + "learning_rate": 3.174248578391552e-06, + "loss": 1.8117, + "step": 1188 + }, + { + "epoch": 1.303013698630137, + "grad_norm": 2.3134653568267822, + "learning_rate": 3.172217709179529e-06, + "loss": 1.8111, + "step": 1189 + }, + { + "epoch": 1.3041095890410959, + "grad_norm": 2.293311834335327, + "learning_rate": 3.1701868399675064e-06, + "loss": 1.791, + "step": 1190 + }, + { + "epoch": 1.3041095890410959, + "eval_loss": 2.0038933753967285, + "eval_runtime": 37.2282, + "eval_samples_per_second": 26.861, + "eval_steps_per_second": 1.128, + "step": 1190 + }, + { + "epoch": 1.305205479452055, + "grad_norm": 2.298020124435425, + "learning_rate": 3.1681559707554837e-06, + "loss": 1.8301, + "step": 1191 + }, + { + "epoch": 1.3063013698630137, + "grad_norm": 2.5261130332946777, + "learning_rate": 3.166125101543461e-06, + "loss": 1.7714, + "step": 1192 + }, + { + "epoch": 1.3073972602739725, + "grad_norm": 2.270820379257202, + "learning_rate": 3.1640942323314382e-06, + "loss": 1.8246, + "step": 1193 + }, + { + "epoch": 1.3084931506849315, + "grad_norm": 2.373260021209717, + "learning_rate": 3.1620633631194155e-06, + "loss": 1.8316, + "step": 1194 + }, + { + "epoch": 1.3095890410958904, + "grad_norm": 2.3265364170074463, + "learning_rate": 3.1600324939073928e-06, + "loss": 1.7679, + "step": 1195 + }, + { + "epoch": 1.3106849315068494, + "grad_norm": 2.2378532886505127, + "learning_rate": 3.15800162469537e-06, + "loss": 1.8149, + "step": 1196 + }, + { + "epoch": 1.3117808219178082, + "grad_norm": 2.2459309101104736, + "learning_rate": 3.155970755483347e-06, + "loss": 1.8086, + "step": 1197 + }, + { + "epoch": 1.3128767123287672, + "grad_norm": 2.2769811153411865, + "learning_rate": 3.1539398862713246e-06, + "loss": 1.8109, + "step": 1198 + }, + { + "epoch": 1.313972602739726, + "grad_norm": 2.310483694076538, + "learning_rate": 3.1519090170593015e-06, + "loss": 1.8297, + "step": 1199 + }, + { + "epoch": 1.3150684931506849, + "grad_norm": 2.302478313446045, + "learning_rate": 3.149878147847279e-06, + "loss": 1.8237, + "step": 1200 + }, + { + "epoch": 1.3150684931506849, + "eval_loss": 2.0011062622070312, + "eval_runtime": 37.2077, + "eval_samples_per_second": 26.876, + "eval_steps_per_second": 1.129, + "step": 1200 + }, + { + "epoch": 1.316164383561644, + "grad_norm": 2.2240796089172363, + "learning_rate": 3.147847278635256e-06, + "loss": 1.8745, + "step": 1201 + }, + { + "epoch": 1.3172602739726027, + "grad_norm": 2.303537130355835, + "learning_rate": 3.1458164094232337e-06, + "loss": 1.9051, + "step": 1202 + }, + { + "epoch": 1.3183561643835615, + "grad_norm": 2.4038636684417725, + "learning_rate": 3.1437855402112106e-06, + "loss": 1.7974, + "step": 1203 + }, + { + "epoch": 1.3194520547945205, + "grad_norm": 2.3458874225616455, + "learning_rate": 3.1417546709991883e-06, + "loss": 1.7601, + "step": 1204 + }, + { + "epoch": 1.3205479452054796, + "grad_norm": 2.3399100303649902, + "learning_rate": 3.139723801787165e-06, + "loss": 1.8314, + "step": 1205 + }, + { + "epoch": 1.3216438356164384, + "grad_norm": 2.293229818344116, + "learning_rate": 3.137692932575143e-06, + "loss": 1.8117, + "step": 1206 + }, + { + "epoch": 1.3227397260273972, + "grad_norm": 2.5378005504608154, + "learning_rate": 3.1356620633631197e-06, + "loss": 1.8614, + "step": 1207 + }, + { + "epoch": 1.3238356164383562, + "grad_norm": 2.3644745349884033, + "learning_rate": 3.1336311941510974e-06, + "loss": 1.7843, + "step": 1208 + }, + { + "epoch": 1.324931506849315, + "grad_norm": 2.1592626571655273, + "learning_rate": 3.1316003249390742e-06, + "loss": 1.8128, + "step": 1209 + }, + { + "epoch": 1.3260273972602739, + "grad_norm": 2.155649185180664, + "learning_rate": 3.129569455727051e-06, + "loss": 1.8636, + "step": 1210 + }, + { + "epoch": 1.3260273972602739, + "eval_loss": 1.9952574968338013, + "eval_runtime": 37.3004, + "eval_samples_per_second": 26.809, + "eval_steps_per_second": 1.126, + "step": 1210 + }, + { + "epoch": 1.3271232876712329, + "grad_norm": 2.1906397342681885, + "learning_rate": 3.1275385865150288e-06, + "loss": 1.8583, + "step": 1211 + }, + { + "epoch": 1.3282191780821917, + "grad_norm": 2.2708194255828857, + "learning_rate": 3.1255077173030056e-06, + "loss": 1.8363, + "step": 1212 + }, + { + "epoch": 1.3293150684931507, + "grad_norm": 2.1753969192504883, + "learning_rate": 3.1234768480909833e-06, + "loss": 1.8372, + "step": 1213 + }, + { + "epoch": 1.3304109589041095, + "grad_norm": 2.229363441467285, + "learning_rate": 3.12144597887896e-06, + "loss": 1.8179, + "step": 1214 + }, + { + "epoch": 1.3315068493150686, + "grad_norm": 2.362152338027954, + "learning_rate": 3.119415109666938e-06, + "loss": 1.8596, + "step": 1215 + }, + { + "epoch": 1.3326027397260274, + "grad_norm": 2.3278255462646484, + "learning_rate": 3.1173842404549147e-06, + "loss": 1.8778, + "step": 1216 + }, + { + "epoch": 1.3336986301369862, + "grad_norm": 2.477710247039795, + "learning_rate": 3.1153533712428924e-06, + "loss": 1.8297, + "step": 1217 + }, + { + "epoch": 1.3347945205479452, + "grad_norm": 2.2611043453216553, + "learning_rate": 3.1133225020308693e-06, + "loss": 1.8374, + "step": 1218 + }, + { + "epoch": 1.335890410958904, + "grad_norm": 2.198275566101074, + "learning_rate": 3.111291632818847e-06, + "loss": 1.8686, + "step": 1219 + }, + { + "epoch": 1.336986301369863, + "grad_norm": 2.227508068084717, + "learning_rate": 3.109260763606824e-06, + "loss": 1.8642, + "step": 1220 + }, + { + "epoch": 1.336986301369863, + "eval_loss": 1.9998422861099243, + "eval_runtime": 37.5152, + "eval_samples_per_second": 26.656, + "eval_steps_per_second": 1.12, + "step": 1220 + }, + { + "epoch": 1.3380821917808219, + "grad_norm": 2.386310338973999, + "learning_rate": 3.1072298943948015e-06, + "loss": 1.8347, + "step": 1221 + }, + { + "epoch": 1.339178082191781, + "grad_norm": 2.290313482284546, + "learning_rate": 3.1051990251827784e-06, + "loss": 1.8455, + "step": 1222 + }, + { + "epoch": 1.3402739726027397, + "grad_norm": 2.3063457012176514, + "learning_rate": 3.1031681559707557e-06, + "loss": 1.811, + "step": 1223 + }, + { + "epoch": 1.3413698630136985, + "grad_norm": 2.297840118408203, + "learning_rate": 3.101137286758733e-06, + "loss": 1.8558, + "step": 1224 + }, + { + "epoch": 1.3424657534246576, + "grad_norm": 2.40864896774292, + "learning_rate": 3.09910641754671e-06, + "loss": 1.8604, + "step": 1225 + }, + { + "epoch": 1.3435616438356164, + "grad_norm": 2.397599458694458, + "learning_rate": 3.0970755483346875e-06, + "loss": 1.8556, + "step": 1226 + }, + { + "epoch": 1.3446575342465754, + "grad_norm": 2.409029245376587, + "learning_rate": 3.0950446791226648e-06, + "loss": 1.8424, + "step": 1227 + }, + { + "epoch": 1.3457534246575342, + "grad_norm": 2.2789814472198486, + "learning_rate": 3.093013809910642e-06, + "loss": 1.8945, + "step": 1228 + }, + { + "epoch": 1.3468493150684933, + "grad_norm": 2.3749048709869385, + "learning_rate": 3.0909829406986193e-06, + "loss": 1.8387, + "step": 1229 + }, + { + "epoch": 1.347945205479452, + "grad_norm": 2.44486141204834, + "learning_rate": 3.0889520714865966e-06, + "loss": 1.8838, + "step": 1230 + }, + { + "epoch": 1.347945205479452, + "eval_loss": 1.9961321353912354, + "eval_runtime": 37.6273, + "eval_samples_per_second": 26.576, + "eval_steps_per_second": 1.116, + "step": 1230 + }, + { + "epoch": 1.3490410958904109, + "grad_norm": 2.2852954864501953, + "learning_rate": 3.086921202274574e-06, + "loss": 1.8551, + "step": 1231 + }, + { + "epoch": 1.35013698630137, + "grad_norm": 2.3131206035614014, + "learning_rate": 3.084890333062551e-06, + "loss": 1.8183, + "step": 1232 + }, + { + "epoch": 1.3512328767123287, + "grad_norm": 2.4095048904418945, + "learning_rate": 3.0828594638505284e-06, + "loss": 1.7946, + "step": 1233 + }, + { + "epoch": 1.3523287671232875, + "grad_norm": 2.429971933364868, + "learning_rate": 3.0808285946385057e-06, + "loss": 1.8349, + "step": 1234 + }, + { + "epoch": 1.3534246575342466, + "grad_norm": 2.2360565662384033, + "learning_rate": 3.078797725426483e-06, + "loss": 1.794, + "step": 1235 + }, + { + "epoch": 1.3545205479452056, + "grad_norm": 2.3630571365356445, + "learning_rate": 3.07676685621446e-06, + "loss": 1.7377, + "step": 1236 + }, + { + "epoch": 1.3556164383561644, + "grad_norm": 2.2604660987854004, + "learning_rate": 3.0747359870024375e-06, + "loss": 1.8239, + "step": 1237 + }, + { + "epoch": 1.3567123287671232, + "grad_norm": 2.209717035293579, + "learning_rate": 3.0727051177904144e-06, + "loss": 1.7835, + "step": 1238 + }, + { + "epoch": 1.3578082191780823, + "grad_norm": 2.3221793174743652, + "learning_rate": 3.070674248578392e-06, + "loss": 1.8192, + "step": 1239 + }, + { + "epoch": 1.358904109589041, + "grad_norm": 2.357313394546509, + "learning_rate": 3.068643379366369e-06, + "loss": 1.8246, + "step": 1240 + }, + { + "epoch": 1.358904109589041, + "eval_loss": 1.9935922622680664, + "eval_runtime": 37.2832, + "eval_samples_per_second": 26.822, + "eval_steps_per_second": 1.127, + "step": 1240 + }, + { + "epoch": 1.3599999999999999, + "grad_norm": 2.2563390731811523, + "learning_rate": 3.0666125101543466e-06, + "loss": 1.8175, + "step": 1241 + }, + { + "epoch": 1.361095890410959, + "grad_norm": 2.2531445026397705, + "learning_rate": 3.0645816409423235e-06, + "loss": 1.7713, + "step": 1242 + }, + { + "epoch": 1.362191780821918, + "grad_norm": 2.2369470596313477, + "learning_rate": 3.062550771730301e-06, + "loss": 1.8377, + "step": 1243 + }, + { + "epoch": 1.3632876712328768, + "grad_norm": 2.4063796997070312, + "learning_rate": 3.060519902518278e-06, + "loss": 1.8116, + "step": 1244 + }, + { + "epoch": 1.3643835616438356, + "grad_norm": 2.4628145694732666, + "learning_rate": 3.0584890333062557e-06, + "loss": 1.7998, + "step": 1245 + }, + { + "epoch": 1.3654794520547946, + "grad_norm": 2.3018743991851807, + "learning_rate": 3.0564581640942326e-06, + "loss": 1.7413, + "step": 1246 + }, + { + "epoch": 1.3665753424657534, + "grad_norm": 2.2432262897491455, + "learning_rate": 3.0544272948822094e-06, + "loss": 1.82, + "step": 1247 + }, + { + "epoch": 1.3676712328767122, + "grad_norm": 2.368435859680176, + "learning_rate": 3.052396425670187e-06, + "loss": 1.7976, + "step": 1248 + }, + { + "epoch": 1.3687671232876713, + "grad_norm": 2.2668025493621826, + "learning_rate": 3.050365556458164e-06, + "loss": 1.8379, + "step": 1249 + }, + { + "epoch": 1.36986301369863, + "grad_norm": 2.2622179985046387, + "learning_rate": 3.0483346872461417e-06, + "loss": 1.794, + "step": 1250 + }, + { + "epoch": 1.36986301369863, + "eval_loss": 1.9959099292755127, + "eval_runtime": 37.3358, + "eval_samples_per_second": 26.784, + "eval_steps_per_second": 1.125, + "step": 1250 + }, + { + "epoch": 1.370958904109589, + "grad_norm": 2.3414738178253174, + "learning_rate": 3.0463038180341185e-06, + "loss": 1.7993, + "step": 1251 + }, + { + "epoch": 1.372054794520548, + "grad_norm": 2.2678418159484863, + "learning_rate": 3.0442729488220962e-06, + "loss": 1.8581, + "step": 1252 + }, + { + "epoch": 1.373150684931507, + "grad_norm": 2.268195867538452, + "learning_rate": 3.042242079610073e-06, + "loss": 1.8486, + "step": 1253 + }, + { + "epoch": 1.3742465753424657, + "grad_norm": 2.230173110961914, + "learning_rate": 3.0402112103980508e-06, + "loss": 1.8468, + "step": 1254 + }, + { + "epoch": 1.3753424657534246, + "grad_norm": 2.3020238876342773, + "learning_rate": 3.0381803411860276e-06, + "loss": 1.7623, + "step": 1255 + }, + { + "epoch": 1.3764383561643836, + "grad_norm": 2.385343313217163, + "learning_rate": 3.0361494719740053e-06, + "loss": 1.7976, + "step": 1256 + }, + { + "epoch": 1.3775342465753424, + "grad_norm": 2.344768524169922, + "learning_rate": 3.034118602761982e-06, + "loss": 1.807, + "step": 1257 + }, + { + "epoch": 1.3786301369863014, + "grad_norm": 2.411760091781616, + "learning_rate": 3.03208773354996e-06, + "loss": 1.8079, + "step": 1258 + }, + { + "epoch": 1.3797260273972602, + "grad_norm": 2.3160128593444824, + "learning_rate": 3.0300568643379367e-06, + "loss": 1.7868, + "step": 1259 + }, + { + "epoch": 1.3808219178082193, + "grad_norm": 2.334632635116577, + "learning_rate": 3.028025995125914e-06, + "loss": 1.8466, + "step": 1260 + }, + { + "epoch": 1.3808219178082193, + "eval_loss": 1.9903327226638794, + "eval_runtime": 37.2256, + "eval_samples_per_second": 26.863, + "eval_steps_per_second": 1.128, + "step": 1260 + }, + { + "epoch": 1.381917808219178, + "grad_norm": 2.3443334102630615, + "learning_rate": 3.0259951259138913e-06, + "loss": 1.8302, + "step": 1261 + }, + { + "epoch": 1.383013698630137, + "grad_norm": 2.311563014984131, + "learning_rate": 3.0239642567018686e-06, + "loss": 1.7727, + "step": 1262 + }, + { + "epoch": 1.384109589041096, + "grad_norm": 2.2501776218414307, + "learning_rate": 3.021933387489846e-06, + "loss": 1.8058, + "step": 1263 + }, + { + "epoch": 1.3852054794520547, + "grad_norm": 2.259157657623291, + "learning_rate": 3.019902518277823e-06, + "loss": 1.7627, + "step": 1264 + }, + { + "epoch": 1.3863013698630138, + "grad_norm": 2.4380314350128174, + "learning_rate": 3.0178716490658004e-06, + "loss": 1.8237, + "step": 1265 + }, + { + "epoch": 1.3873972602739726, + "grad_norm": 2.3203837871551514, + "learning_rate": 3.0158407798537777e-06, + "loss": 1.8557, + "step": 1266 + }, + { + "epoch": 1.3884931506849316, + "grad_norm": 2.220996141433716, + "learning_rate": 3.013809910641755e-06, + "loss": 1.7641, + "step": 1267 + }, + { + "epoch": 1.3895890410958904, + "grad_norm": 2.190006971359253, + "learning_rate": 3.0117790414297322e-06, + "loss": 1.7705, + "step": 1268 + }, + { + "epoch": 1.3906849315068492, + "grad_norm": 2.229748249053955, + "learning_rate": 3.0097481722177095e-06, + "loss": 1.7709, + "step": 1269 + }, + { + "epoch": 1.3917808219178083, + "grad_norm": 2.3755619525909424, + "learning_rate": 3.0077173030056868e-06, + "loss": 1.8172, + "step": 1270 + }, + { + "epoch": 1.3917808219178083, + "eval_loss": 1.992952585220337, + "eval_runtime": 37.2403, + "eval_samples_per_second": 26.853, + "eval_steps_per_second": 1.128, + "step": 1270 + }, + { + "epoch": 1.392876712328767, + "grad_norm": 2.2514004707336426, + "learning_rate": 3.005686433793664e-06, + "loss": 1.7719, + "step": 1271 + }, + { + "epoch": 1.393972602739726, + "grad_norm": 2.3010191917419434, + "learning_rate": 3.0036555645816413e-06, + "loss": 1.8369, + "step": 1272 + }, + { + "epoch": 1.395068493150685, + "grad_norm": 2.68790602684021, + "learning_rate": 3.001624695369618e-06, + "loss": 1.7974, + "step": 1273 + }, + { + "epoch": 1.396164383561644, + "grad_norm": 2.454904794692993, + "learning_rate": 2.999593826157596e-06, + "loss": 1.8509, + "step": 1274 + }, + { + "epoch": 1.3972602739726028, + "grad_norm": 2.251556396484375, + "learning_rate": 2.9975629569455727e-06, + "loss": 1.8011, + "step": 1275 + }, + { + "epoch": 1.3983561643835616, + "grad_norm": 2.3789098262786865, + "learning_rate": 2.9955320877335504e-06, + "loss": 1.8488, + "step": 1276 + }, + { + "epoch": 1.3994520547945206, + "grad_norm": 2.2659196853637695, + "learning_rate": 2.9935012185215273e-06, + "loss": 1.7946, + "step": 1277 + }, + { + "epoch": 1.4005479452054794, + "grad_norm": 2.2148971557617188, + "learning_rate": 2.991470349309505e-06, + "loss": 1.8388, + "step": 1278 + }, + { + "epoch": 1.4016438356164382, + "grad_norm": 2.1883089542388916, + "learning_rate": 2.989439480097482e-06, + "loss": 1.7925, + "step": 1279 + }, + { + "epoch": 1.4027397260273973, + "grad_norm": 2.290208339691162, + "learning_rate": 2.9874086108854595e-06, + "loss": 1.8099, + "step": 1280 + }, + { + "epoch": 1.4027397260273973, + "eval_loss": 1.9924782514572144, + "eval_runtime": 37.1891, + "eval_samples_per_second": 26.89, + "eval_steps_per_second": 1.129, + "step": 1280 + }, + { + "epoch": 1.403835616438356, + "grad_norm": 2.121649980545044, + "learning_rate": 2.9853777416734364e-06, + "loss": 1.7991, + "step": 1281 + }, + { + "epoch": 1.4049315068493151, + "grad_norm": 2.3528690338134766, + "learning_rate": 2.983346872461414e-06, + "loss": 1.8487, + "step": 1282 + }, + { + "epoch": 1.406027397260274, + "grad_norm": 2.391491174697876, + "learning_rate": 2.981316003249391e-06, + "loss": 1.8564, + "step": 1283 + }, + { + "epoch": 1.407123287671233, + "grad_norm": 2.336615800857544, + "learning_rate": 2.9792851340373678e-06, + "loss": 1.8216, + "step": 1284 + }, + { + "epoch": 1.4082191780821918, + "grad_norm": 2.2958548069000244, + "learning_rate": 2.9772542648253455e-06, + "loss": 1.8003, + "step": 1285 + }, + { + "epoch": 1.4093150684931506, + "grad_norm": 2.2062184810638428, + "learning_rate": 2.9752233956133223e-06, + "loss": 1.8488, + "step": 1286 + }, + { + "epoch": 1.4104109589041096, + "grad_norm": 2.486133575439453, + "learning_rate": 2.9731925264013e-06, + "loss": 1.7914, + "step": 1287 + }, + { + "epoch": 1.4115068493150684, + "grad_norm": 2.2933201789855957, + "learning_rate": 2.971161657189277e-06, + "loss": 1.8164, + "step": 1288 + }, + { + "epoch": 1.4126027397260275, + "grad_norm": 2.3301920890808105, + "learning_rate": 2.9691307879772546e-06, + "loss": 1.7883, + "step": 1289 + }, + { + "epoch": 1.4136986301369863, + "grad_norm": 2.248832941055298, + "learning_rate": 2.9670999187652314e-06, + "loss": 1.7916, + "step": 1290 + }, + { + "epoch": 1.4136986301369863, + "eval_loss": 1.9919097423553467, + "eval_runtime": 37.1283, + "eval_samples_per_second": 26.934, + "eval_steps_per_second": 1.131, + "step": 1290 + }, + { + "epoch": 1.4147945205479453, + "grad_norm": 2.37760853767395, + "learning_rate": 2.965069049553209e-06, + "loss": 1.7854, + "step": 1291 + }, + { + "epoch": 1.415890410958904, + "grad_norm": 2.327718734741211, + "learning_rate": 2.963038180341186e-06, + "loss": 1.874, + "step": 1292 + }, + { + "epoch": 1.416986301369863, + "grad_norm": 2.2413878440856934, + "learning_rate": 2.9610073111291637e-06, + "loss": 1.7591, + "step": 1293 + }, + { + "epoch": 1.418082191780822, + "grad_norm": 2.2203164100646973, + "learning_rate": 2.9589764419171405e-06, + "loss": 1.8466, + "step": 1294 + }, + { + "epoch": 1.4191780821917808, + "grad_norm": 2.2070505619049072, + "learning_rate": 2.9569455727051182e-06, + "loss": 1.8689, + "step": 1295 + }, + { + "epoch": 1.4202739726027398, + "grad_norm": 2.2122223377227783, + "learning_rate": 2.954914703493095e-06, + "loss": 1.7689, + "step": 1296 + }, + { + "epoch": 1.4213698630136986, + "grad_norm": 2.331303358078003, + "learning_rate": 2.9528838342810724e-06, + "loss": 1.8089, + "step": 1297 + }, + { + "epoch": 1.4224657534246576, + "grad_norm": 2.408184289932251, + "learning_rate": 2.9508529650690496e-06, + "loss": 1.7847, + "step": 1298 + }, + { + "epoch": 1.4235616438356165, + "grad_norm": 2.212766170501709, + "learning_rate": 2.948822095857027e-06, + "loss": 1.7899, + "step": 1299 + }, + { + "epoch": 1.4246575342465753, + "grad_norm": 2.259953737258911, + "learning_rate": 2.946791226645004e-06, + "loss": 1.8435, + "step": 1300 + }, + { + "epoch": 1.4246575342465753, + "eval_loss": 1.9893293380737305, + "eval_runtime": 37.1675, + "eval_samples_per_second": 26.905, + "eval_steps_per_second": 1.13, + "step": 1300 + }, + { + "epoch": 1.4257534246575343, + "grad_norm": 2.342400312423706, + "learning_rate": 2.9447603574329815e-06, + "loss": 1.7779, + "step": 1301 + }, + { + "epoch": 1.426849315068493, + "grad_norm": 2.237618923187256, + "learning_rate": 2.9427294882209587e-06, + "loss": 1.7509, + "step": 1302 + }, + { + "epoch": 1.427945205479452, + "grad_norm": 2.3418633937835693, + "learning_rate": 2.940698619008936e-06, + "loss": 1.7963, + "step": 1303 + }, + { + "epoch": 1.429041095890411, + "grad_norm": 2.470827579498291, + "learning_rate": 2.9386677497969133e-06, + "loss": 1.8045, + "step": 1304 + }, + { + "epoch": 1.43013698630137, + "grad_norm": 2.508345365524292, + "learning_rate": 2.9366368805848906e-06, + "loss": 1.7926, + "step": 1305 + }, + { + "epoch": 1.4312328767123288, + "grad_norm": 2.261319160461426, + "learning_rate": 2.934606011372868e-06, + "loss": 1.831, + "step": 1306 + }, + { + "epoch": 1.4323287671232876, + "grad_norm": 2.2703914642333984, + "learning_rate": 2.932575142160845e-06, + "loss": 1.8352, + "step": 1307 + }, + { + "epoch": 1.4334246575342466, + "grad_norm": 2.243626117706299, + "learning_rate": 2.9305442729488224e-06, + "loss": 1.8162, + "step": 1308 + }, + { + "epoch": 1.4345205479452054, + "grad_norm": 2.3859238624572754, + "learning_rate": 2.9285134037367997e-06, + "loss": 1.8171, + "step": 1309 + }, + { + "epoch": 1.4356164383561643, + "grad_norm": 2.3350934982299805, + "learning_rate": 2.9264825345247765e-06, + "loss": 1.8066, + "step": 1310 + }, + { + "epoch": 1.4356164383561643, + "eval_loss": 1.9905259609222412, + "eval_runtime": 37.2009, + "eval_samples_per_second": 26.881, + "eval_steps_per_second": 1.129, + "step": 1310 + }, + { + "epoch": 1.4367123287671233, + "grad_norm": 2.1943347454071045, + "learning_rate": 2.9244516653127542e-06, + "loss": 1.8435, + "step": 1311 + }, + { + "epoch": 1.4378082191780823, + "grad_norm": 2.429128885269165, + "learning_rate": 2.922420796100731e-06, + "loss": 1.8179, + "step": 1312 + }, + { + "epoch": 1.4389041095890411, + "grad_norm": 2.3677234649658203, + "learning_rate": 2.9203899268887088e-06, + "loss": 1.7245, + "step": 1313 + }, + { + "epoch": 1.44, + "grad_norm": 2.255570888519287, + "learning_rate": 2.9183590576766856e-06, + "loss": 1.8545, + "step": 1314 + }, + { + "epoch": 1.441095890410959, + "grad_norm": 2.5176682472229004, + "learning_rate": 2.9163281884646633e-06, + "loss": 1.7509, + "step": 1315 + }, + { + "epoch": 1.4421917808219178, + "grad_norm": 2.29972767829895, + "learning_rate": 2.91429731925264e-06, + "loss": 1.8116, + "step": 1316 + }, + { + "epoch": 1.4432876712328766, + "grad_norm": 2.2853286266326904, + "learning_rate": 2.912266450040618e-06, + "loss": 1.8552, + "step": 1317 + }, + { + "epoch": 1.4443835616438356, + "grad_norm": 2.505075216293335, + "learning_rate": 2.9102355808285947e-06, + "loss": 1.811, + "step": 1318 + }, + { + "epoch": 1.4454794520547944, + "grad_norm": 2.229064464569092, + "learning_rate": 2.9082047116165724e-06, + "loss": 1.8068, + "step": 1319 + }, + { + "epoch": 1.4465753424657535, + "grad_norm": 2.199187994003296, + "learning_rate": 2.9061738424045493e-06, + "loss": 1.7825, + "step": 1320 + }, + { + "epoch": 1.4465753424657535, + "eval_loss": 1.9851222038269043, + "eval_runtime": 37.4614, + "eval_samples_per_second": 26.694, + "eval_steps_per_second": 1.121, + "step": 1320 + }, + { + "epoch": 1.4476712328767123, + "grad_norm": 2.4318959712982178, + "learning_rate": 2.904142973192527e-06, + "loss": 1.77, + "step": 1321 + }, + { + "epoch": 1.4487671232876713, + "grad_norm": 2.480703592300415, + "learning_rate": 2.902112103980504e-06, + "loss": 1.8774, + "step": 1322 + }, + { + "epoch": 1.4498630136986301, + "grad_norm": 2.367624521255493, + "learning_rate": 2.900081234768481e-06, + "loss": 1.818, + "step": 1323 + }, + { + "epoch": 1.450958904109589, + "grad_norm": 2.245974540710449, + "learning_rate": 2.8980503655564584e-06, + "loss": 1.8054, + "step": 1324 + }, + { + "epoch": 1.452054794520548, + "grad_norm": 2.352687120437622, + "learning_rate": 2.8960194963444357e-06, + "loss": 1.7922, + "step": 1325 + }, + { + "epoch": 1.4531506849315068, + "grad_norm": 2.308253526687622, + "learning_rate": 2.893988627132413e-06, + "loss": 1.7637, + "step": 1326 + }, + { + "epoch": 1.4542465753424658, + "grad_norm": 2.309495687484741, + "learning_rate": 2.8919577579203902e-06, + "loss": 1.7905, + "step": 1327 + }, + { + "epoch": 1.4553424657534246, + "grad_norm": 2.354365348815918, + "learning_rate": 2.8899268887083675e-06, + "loss": 1.7321, + "step": 1328 + }, + { + "epoch": 1.4564383561643837, + "grad_norm": 2.3886210918426514, + "learning_rate": 2.8878960194963448e-06, + "loss": 1.7932, + "step": 1329 + }, + { + "epoch": 1.4575342465753425, + "grad_norm": 2.395714044570923, + "learning_rate": 2.885865150284322e-06, + "loss": 1.813, + "step": 1330 + }, + { + "epoch": 1.4575342465753425, + "eval_loss": 1.9851990938186646, + "eval_runtime": 37.2966, + "eval_samples_per_second": 26.812, + "eval_steps_per_second": 1.126, + "step": 1330 + }, + { + "epoch": 1.4586301369863013, + "grad_norm": 2.433069944381714, + "learning_rate": 2.8838342810722993e-06, + "loss": 1.8346, + "step": 1331 + }, + { + "epoch": 1.4597260273972603, + "grad_norm": 2.295894145965576, + "learning_rate": 2.8818034118602766e-06, + "loss": 1.7855, + "step": 1332 + }, + { + "epoch": 1.4608219178082191, + "grad_norm": 2.495742082595825, + "learning_rate": 2.879772542648254e-06, + "loss": 1.7601, + "step": 1333 + }, + { + "epoch": 1.4619178082191782, + "grad_norm": 2.35984206199646, + "learning_rate": 2.8777416734362307e-06, + "loss": 1.8329, + "step": 1334 + }, + { + "epoch": 1.463013698630137, + "grad_norm": 2.39632248878479, + "learning_rate": 2.8757108042242084e-06, + "loss": 1.7426, + "step": 1335 + }, + { + "epoch": 1.464109589041096, + "grad_norm": 2.4151153564453125, + "learning_rate": 2.8736799350121853e-06, + "loss": 1.8899, + "step": 1336 + }, + { + "epoch": 1.4652054794520548, + "grad_norm": 2.43914532661438, + "learning_rate": 2.871649065800163e-06, + "loss": 1.8464, + "step": 1337 + }, + { + "epoch": 1.4663013698630136, + "grad_norm": 2.44490122795105, + "learning_rate": 2.86961819658814e-06, + "loss": 1.8643, + "step": 1338 + }, + { + "epoch": 1.4673972602739727, + "grad_norm": 2.2639644145965576, + "learning_rate": 2.8675873273761175e-06, + "loss": 1.792, + "step": 1339 + }, + { + "epoch": 1.4684931506849315, + "grad_norm": 2.2803847789764404, + "learning_rate": 2.8655564581640944e-06, + "loss": 1.8109, + "step": 1340 + }, + { + "epoch": 1.4684931506849315, + "eval_loss": 1.9856781959533691, + "eval_runtime": 37.2519, + "eval_samples_per_second": 26.844, + "eval_steps_per_second": 1.127, + "step": 1340 + }, + { + "epoch": 1.4695890410958903, + "grad_norm": 2.494727849960327, + "learning_rate": 2.863525588952072e-06, + "loss": 1.8032, + "step": 1341 + }, + { + "epoch": 1.4706849315068493, + "grad_norm": 2.239595413208008, + "learning_rate": 2.861494719740049e-06, + "loss": 1.8659, + "step": 1342 + }, + { + "epoch": 1.4717808219178083, + "grad_norm": 2.2922589778900146, + "learning_rate": 2.8594638505280266e-06, + "loss": 1.7587, + "step": 1343 + }, + { + "epoch": 1.4728767123287672, + "grad_norm": 2.345224142074585, + "learning_rate": 2.8574329813160035e-06, + "loss": 1.7602, + "step": 1344 + }, + { + "epoch": 1.473972602739726, + "grad_norm": 2.3415307998657227, + "learning_rate": 2.855402112103981e-06, + "loss": 1.8044, + "step": 1345 + }, + { + "epoch": 1.475068493150685, + "grad_norm": 2.331557035446167, + "learning_rate": 2.853371242891958e-06, + "loss": 1.7808, + "step": 1346 + }, + { + "epoch": 1.4761643835616438, + "grad_norm": 2.293572425842285, + "learning_rate": 2.851340373679935e-06, + "loss": 1.8017, + "step": 1347 + }, + { + "epoch": 1.4772602739726026, + "grad_norm": 2.3477773666381836, + "learning_rate": 2.8493095044679126e-06, + "loss": 1.8113, + "step": 1348 + }, + { + "epoch": 1.4783561643835617, + "grad_norm": 2.3740994930267334, + "learning_rate": 2.8472786352558894e-06, + "loss": 1.8801, + "step": 1349 + }, + { + "epoch": 1.4794520547945205, + "grad_norm": 2.4202206134796143, + "learning_rate": 2.845247766043867e-06, + "loss": 1.821, + "step": 1350 + }, + { + "epoch": 1.4794520547945205, + "eval_loss": 1.9867810010910034, + "eval_runtime": 37.2597, + "eval_samples_per_second": 26.839, + "eval_steps_per_second": 1.127, + "step": 1350 + }, + { + "epoch": 1.4805479452054795, + "grad_norm": 2.2323198318481445, + "learning_rate": 2.843216896831844e-06, + "loss": 1.7574, + "step": 1351 + }, + { + "epoch": 1.4816438356164383, + "grad_norm": 2.2192625999450684, + "learning_rate": 2.8411860276198217e-06, + "loss": 1.7662, + "step": 1352 + }, + { + "epoch": 1.4827397260273973, + "grad_norm": 2.276507616043091, + "learning_rate": 2.8391551584077985e-06, + "loss": 1.7748, + "step": 1353 + }, + { + "epoch": 1.4838356164383562, + "grad_norm": 2.292773723602295, + "learning_rate": 2.8371242891957762e-06, + "loss": 1.8388, + "step": 1354 + }, + { + "epoch": 1.484931506849315, + "grad_norm": 2.3127009868621826, + "learning_rate": 2.835093419983753e-06, + "loss": 1.861, + "step": 1355 + }, + { + "epoch": 1.486027397260274, + "grad_norm": 2.2752184867858887, + "learning_rate": 2.833062550771731e-06, + "loss": 1.7883, + "step": 1356 + }, + { + "epoch": 1.4871232876712328, + "grad_norm": 2.288158416748047, + "learning_rate": 2.8310316815597076e-06, + "loss": 1.7963, + "step": 1357 + }, + { + "epoch": 1.4882191780821918, + "grad_norm": 2.3287482261657715, + "learning_rate": 2.8290008123476853e-06, + "loss": 1.7534, + "step": 1358 + }, + { + "epoch": 1.4893150684931507, + "grad_norm": 2.441279888153076, + "learning_rate": 2.826969943135662e-06, + "loss": 1.8073, + "step": 1359 + }, + { + "epoch": 1.4904109589041097, + "grad_norm": 2.3399839401245117, + "learning_rate": 2.8249390739236395e-06, + "loss": 1.7876, + "step": 1360 + }, + { + "epoch": 1.4904109589041097, + "eval_loss": 1.9868030548095703, + "eval_runtime": 37.2891, + "eval_samples_per_second": 26.817, + "eval_steps_per_second": 1.126, + "step": 1360 + }, + { + "epoch": 1.4915068493150685, + "grad_norm": 2.234246253967285, + "learning_rate": 2.8229082047116167e-06, + "loss": 1.8031, + "step": 1361 + }, + { + "epoch": 1.4926027397260273, + "grad_norm": 2.3136820793151855, + "learning_rate": 2.820877335499594e-06, + "loss": 1.7669, + "step": 1362 + }, + { + "epoch": 1.4936986301369863, + "grad_norm": 2.389420747756958, + "learning_rate": 2.8188464662875713e-06, + "loss": 1.7879, + "step": 1363 + }, + { + "epoch": 1.4947945205479451, + "grad_norm": 2.222306489944458, + "learning_rate": 2.8168155970755486e-06, + "loss": 1.7412, + "step": 1364 + }, + { + "epoch": 1.4958904109589042, + "grad_norm": 2.1689319610595703, + "learning_rate": 2.814784727863526e-06, + "loss": 1.7651, + "step": 1365 + }, + { + "epoch": 1.496986301369863, + "grad_norm": 2.262047529220581, + "learning_rate": 2.812753858651503e-06, + "loss": 1.7798, + "step": 1366 + }, + { + "epoch": 1.498082191780822, + "grad_norm": 2.266894817352295, + "learning_rate": 2.8107229894394804e-06, + "loss": 1.8172, + "step": 1367 + }, + { + "epoch": 1.4991780821917808, + "grad_norm": 2.6400444507598877, + "learning_rate": 2.8086921202274577e-06, + "loss": 1.7871, + "step": 1368 + }, + { + "epoch": 1.5002739726027396, + "grad_norm": 2.3868865966796875, + "learning_rate": 2.806661251015435e-06, + "loss": 1.7427, + "step": 1369 + }, + { + "epoch": 1.5013698630136987, + "grad_norm": 2.5206542015075684, + "learning_rate": 2.8046303818034122e-06, + "loss": 1.817, + "step": 1370 + }, + { + "epoch": 1.5013698630136987, + "eval_loss": 1.9870396852493286, + "eval_runtime": 37.2372, + "eval_samples_per_second": 26.855, + "eval_steps_per_second": 1.128, + "step": 1370 + }, + { + "epoch": 1.5024657534246575, + "grad_norm": 2.5236382484436035, + "learning_rate": 2.802599512591389e-06, + "loss": 1.8662, + "step": 1371 + }, + { + "epoch": 1.5035616438356163, + "grad_norm": 2.5095479488372803, + "learning_rate": 2.8005686433793668e-06, + "loss": 1.9168, + "step": 1372 + }, + { + "epoch": 1.5046575342465753, + "grad_norm": 2.307508707046509, + "learning_rate": 2.7985377741673436e-06, + "loss": 1.799, + "step": 1373 + }, + { + "epoch": 1.5057534246575344, + "grad_norm": 2.410879373550415, + "learning_rate": 2.7965069049553213e-06, + "loss": 1.7447, + "step": 1374 + }, + { + "epoch": 1.5068493150684932, + "grad_norm": 2.738558530807495, + "learning_rate": 2.794476035743298e-06, + "loss": 1.8073, + "step": 1375 + }, + { + "epoch": 1.507945205479452, + "grad_norm": 2.3826518058776855, + "learning_rate": 2.792445166531276e-06, + "loss": 1.8506, + "step": 1376 + }, + { + "epoch": 1.509041095890411, + "grad_norm": 2.334123373031616, + "learning_rate": 2.7904142973192527e-06, + "loss": 1.8209, + "step": 1377 + }, + { + "epoch": 1.5101369863013698, + "grad_norm": 2.668513298034668, + "learning_rate": 2.7883834281072304e-06, + "loss": 1.8285, + "step": 1378 + }, + { + "epoch": 1.5112328767123286, + "grad_norm": 2.471938133239746, + "learning_rate": 2.7863525588952073e-06, + "loss": 1.7668, + "step": 1379 + }, + { + "epoch": 1.5123287671232877, + "grad_norm": 2.3712313175201416, + "learning_rate": 2.784321689683185e-06, + "loss": 1.7565, + "step": 1380 + }, + { + "epoch": 1.5123287671232877, + "eval_loss": 1.9826414585113525, + "eval_runtime": 37.2426, + "eval_samples_per_second": 26.851, + "eval_steps_per_second": 1.128, + "step": 1380 + }, + { + "epoch": 1.5134246575342467, + "grad_norm": 2.3364150524139404, + "learning_rate": 2.782290820471162e-06, + "loss": 1.7596, + "step": 1381 + }, + { + "epoch": 1.5145205479452055, + "grad_norm": 2.3088393211364746, + "learning_rate": 2.7802599512591395e-06, + "loss": 1.7568, + "step": 1382 + }, + { + "epoch": 1.5156164383561643, + "grad_norm": 2.2584595680236816, + "learning_rate": 2.7782290820471164e-06, + "loss": 1.833, + "step": 1383 + }, + { + "epoch": 1.5167123287671234, + "grad_norm": 2.2440273761749268, + "learning_rate": 2.7761982128350932e-06, + "loss": 1.8237, + "step": 1384 + }, + { + "epoch": 1.5178082191780822, + "grad_norm": 2.3552212715148926, + "learning_rate": 2.774167343623071e-06, + "loss": 1.8288, + "step": 1385 + }, + { + "epoch": 1.518904109589041, + "grad_norm": 2.3001790046691895, + "learning_rate": 2.772136474411048e-06, + "loss": 1.8519, + "step": 1386 + }, + { + "epoch": 1.52, + "grad_norm": 2.3691201210021973, + "learning_rate": 2.7701056051990255e-06, + "loss": 1.7759, + "step": 1387 + }, + { + "epoch": 1.521095890410959, + "grad_norm": 2.327988624572754, + "learning_rate": 2.7680747359870023e-06, + "loss": 1.7132, + "step": 1388 + }, + { + "epoch": 1.5221917808219179, + "grad_norm": 2.30245304107666, + "learning_rate": 2.76604386677498e-06, + "loss": 1.8167, + "step": 1389 + }, + { + "epoch": 1.5232876712328767, + "grad_norm": 2.290121078491211, + "learning_rate": 2.764012997562957e-06, + "loss": 1.7913, + "step": 1390 + }, + { + "epoch": 1.5232876712328767, + "eval_loss": 1.98329758644104, + "eval_runtime": 37.2462, + "eval_samples_per_second": 26.848, + "eval_steps_per_second": 1.128, + "step": 1390 + }, + { + "epoch": 1.5243835616438357, + "grad_norm": 2.294076681137085, + "learning_rate": 2.7619821283509346e-06, + "loss": 1.8598, + "step": 1391 + }, + { + "epoch": 1.5254794520547945, + "grad_norm": 2.3325250148773193, + "learning_rate": 2.7599512591389114e-06, + "loss": 1.7392, + "step": 1392 + }, + { + "epoch": 1.5265753424657533, + "grad_norm": 2.2802610397338867, + "learning_rate": 2.757920389926889e-06, + "loss": 1.8332, + "step": 1393 + }, + { + "epoch": 1.5276712328767124, + "grad_norm": 2.284207344055176, + "learning_rate": 2.755889520714866e-06, + "loss": 1.7875, + "step": 1394 + }, + { + "epoch": 1.5287671232876714, + "grad_norm": 2.325998306274414, + "learning_rate": 2.7538586515028437e-06, + "loss": 1.7657, + "step": 1395 + }, + { + "epoch": 1.52986301369863, + "grad_norm": 2.285393476486206, + "learning_rate": 2.7518277822908205e-06, + "loss": 1.7416, + "step": 1396 + }, + { + "epoch": 1.530958904109589, + "grad_norm": 2.3110482692718506, + "learning_rate": 2.749796913078798e-06, + "loss": 1.7931, + "step": 1397 + }, + { + "epoch": 1.532054794520548, + "grad_norm": 2.4234025478363037, + "learning_rate": 2.747766043866775e-06, + "loss": 1.8472, + "step": 1398 + }, + { + "epoch": 1.5331506849315069, + "grad_norm": 2.384767532348633, + "learning_rate": 2.7457351746547524e-06, + "loss": 1.7961, + "step": 1399 + }, + { + "epoch": 1.5342465753424657, + "grad_norm": 2.1857433319091797, + "learning_rate": 2.7437043054427297e-06, + "loss": 1.8654, + "step": 1400 + }, + { + "epoch": 1.5342465753424657, + "eval_loss": 1.9802658557891846, + "eval_runtime": 37.1443, + "eval_samples_per_second": 26.922, + "eval_steps_per_second": 1.131, + "step": 1400 + }, + { + "epoch": 1.5353424657534247, + "grad_norm": 2.1302895545959473, + "learning_rate": 2.741673436230707e-06, + "loss": 1.9051, + "step": 1401 + }, + { + "epoch": 1.5364383561643835, + "grad_norm": 2.377307415008545, + "learning_rate": 2.739642567018684e-06, + "loss": 1.7942, + "step": 1402 + }, + { + "epoch": 1.5375342465753423, + "grad_norm": 2.316549062728882, + "learning_rate": 2.7376116978066615e-06, + "loss": 1.7734, + "step": 1403 + }, + { + "epoch": 1.5386301369863014, + "grad_norm": 2.3452532291412354, + "learning_rate": 2.7355808285946388e-06, + "loss": 1.8243, + "step": 1404 + }, + { + "epoch": 1.5397260273972604, + "grad_norm": 2.3139419555664062, + "learning_rate": 2.733549959382616e-06, + "loss": 1.812, + "step": 1405 + }, + { + "epoch": 1.5408219178082192, + "grad_norm": 2.3135571479797363, + "learning_rate": 2.7315190901705933e-06, + "loss": 1.7751, + "step": 1406 + }, + { + "epoch": 1.541917808219178, + "grad_norm": 2.403231620788574, + "learning_rate": 2.7294882209585706e-06, + "loss": 1.8561, + "step": 1407 + }, + { + "epoch": 1.543013698630137, + "grad_norm": 2.4028689861297607, + "learning_rate": 2.727457351746548e-06, + "loss": 1.8249, + "step": 1408 + }, + { + "epoch": 1.5441095890410959, + "grad_norm": 2.4028851985931396, + "learning_rate": 2.725426482534525e-06, + "loss": 1.779, + "step": 1409 + }, + { + "epoch": 1.5452054794520547, + "grad_norm": 2.270709753036499, + "learning_rate": 2.723395613322502e-06, + "loss": 1.7754, + "step": 1410 + }, + { + "epoch": 1.5452054794520547, + "eval_loss": 1.983272910118103, + "eval_runtime": 37.2405, + "eval_samples_per_second": 26.853, + "eval_steps_per_second": 1.128, + "step": 1410 + }, + { + "epoch": 1.5463013698630137, + "grad_norm": 2.2672812938690186, + "learning_rate": 2.7213647441104797e-06, + "loss": 1.814, + "step": 1411 + }, + { + "epoch": 1.5473972602739727, + "grad_norm": 2.3539397716522217, + "learning_rate": 2.7193338748984565e-06, + "loss": 1.7907, + "step": 1412 + }, + { + "epoch": 1.5484931506849315, + "grad_norm": 2.3229010105133057, + "learning_rate": 2.7173030056864342e-06, + "loss": 1.8061, + "step": 1413 + }, + { + "epoch": 1.5495890410958904, + "grad_norm": 2.482874870300293, + "learning_rate": 2.715272136474411e-06, + "loss": 1.9577, + "step": 1414 + }, + { + "epoch": 1.5506849315068494, + "grad_norm": 2.211798906326294, + "learning_rate": 2.713241267262389e-06, + "loss": 1.7309, + "step": 1415 + }, + { + "epoch": 1.5517808219178082, + "grad_norm": 2.3159289360046387, + "learning_rate": 2.7112103980503656e-06, + "loss": 1.8132, + "step": 1416 + }, + { + "epoch": 1.552876712328767, + "grad_norm": 2.486035108566284, + "learning_rate": 2.7091795288383433e-06, + "loss": 1.7734, + "step": 1417 + }, + { + "epoch": 1.553972602739726, + "grad_norm": 2.3025195598602295, + "learning_rate": 2.70714865962632e-06, + "loss": 1.8232, + "step": 1418 + }, + { + "epoch": 1.555068493150685, + "grad_norm": 2.234581470489502, + "learning_rate": 2.705117790414298e-06, + "loss": 1.7814, + "step": 1419 + }, + { + "epoch": 1.5561643835616439, + "grad_norm": 2.2855048179626465, + "learning_rate": 2.7030869212022747e-06, + "loss": 1.777, + "step": 1420 + }, + { + "epoch": 1.5561643835616439, + "eval_loss": 1.9851160049438477, + "eval_runtime": 37.4606, + "eval_samples_per_second": 26.695, + "eval_steps_per_second": 1.121, + "step": 1420 + }, + { + "epoch": 1.5572602739726027, + "grad_norm": 2.4190328121185303, + "learning_rate": 2.7010560519902516e-06, + "loss": 1.8456, + "step": 1421 + }, + { + "epoch": 1.5583561643835617, + "grad_norm": 2.225623369216919, + "learning_rate": 2.6990251827782293e-06, + "loss": 1.7638, + "step": 1422 + }, + { + "epoch": 1.5594520547945205, + "grad_norm": 2.1516661643981934, + "learning_rate": 2.696994313566206e-06, + "loss": 1.7861, + "step": 1423 + }, + { + "epoch": 1.5605479452054793, + "grad_norm": 2.2461135387420654, + "learning_rate": 2.694963444354184e-06, + "loss": 1.7763, + "step": 1424 + }, + { + "epoch": 1.5616438356164384, + "grad_norm": 2.280869245529175, + "learning_rate": 2.6929325751421607e-06, + "loss": 1.8589, + "step": 1425 + }, + { + "epoch": 1.5627397260273974, + "grad_norm": 2.2034990787506104, + "learning_rate": 2.6909017059301384e-06, + "loss": 1.7705, + "step": 1426 + }, + { + "epoch": 1.563835616438356, + "grad_norm": 2.1346969604492188, + "learning_rate": 2.6888708367181153e-06, + "loss": 1.8318, + "step": 1427 + }, + { + "epoch": 1.564931506849315, + "grad_norm": 2.37400221824646, + "learning_rate": 2.686839967506093e-06, + "loss": 1.746, + "step": 1428 + }, + { + "epoch": 1.566027397260274, + "grad_norm": 2.392561435699463, + "learning_rate": 2.68480909829407e-06, + "loss": 1.8105, + "step": 1429 + }, + { + "epoch": 1.5671232876712329, + "grad_norm": 2.299830198287964, + "learning_rate": 2.6827782290820475e-06, + "loss": 1.7103, + "step": 1430 + }, + { + "epoch": 1.5671232876712329, + "eval_loss": 1.9877182245254517, + "eval_runtime": 37.4813, + "eval_samples_per_second": 26.68, + "eval_steps_per_second": 1.121, + "step": 1430 + }, + { + "epoch": 1.5682191780821917, + "grad_norm": 2.3431026935577393, + "learning_rate": 2.6807473598700244e-06, + "loss": 1.7217, + "step": 1431 + }, + { + "epoch": 1.5693150684931507, + "grad_norm": 2.3474628925323486, + "learning_rate": 2.678716490658002e-06, + "loss": 1.8419, + "step": 1432 + }, + { + "epoch": 1.5704109589041098, + "grad_norm": 2.2877044677734375, + "learning_rate": 2.676685621445979e-06, + "loss": 1.8031, + "step": 1433 + }, + { + "epoch": 1.5715068493150683, + "grad_norm": 2.278613805770874, + "learning_rate": 2.674654752233956e-06, + "loss": 1.8066, + "step": 1434 + }, + { + "epoch": 1.5726027397260274, + "grad_norm": 2.3357837200164795, + "learning_rate": 2.6726238830219335e-06, + "loss": 1.8079, + "step": 1435 + }, + { + "epoch": 1.5736986301369864, + "grad_norm": 2.1932482719421387, + "learning_rate": 2.6705930138099107e-06, + "loss": 1.8032, + "step": 1436 + }, + { + "epoch": 1.5747945205479452, + "grad_norm": 2.298553705215454, + "learning_rate": 2.668562144597888e-06, + "loss": 1.8439, + "step": 1437 + }, + { + "epoch": 1.575890410958904, + "grad_norm": 2.2906084060668945, + "learning_rate": 2.6665312753858653e-06, + "loss": 1.7144, + "step": 1438 + }, + { + "epoch": 1.576986301369863, + "grad_norm": 2.2512989044189453, + "learning_rate": 2.6645004061738426e-06, + "loss": 1.8031, + "step": 1439 + }, + { + "epoch": 1.5780821917808219, + "grad_norm": 2.177103281021118, + "learning_rate": 2.66246953696182e-06, + "loss": 1.8073, + "step": 1440 + }, + { + "epoch": 1.5780821917808219, + "eval_loss": 1.9914618730545044, + "eval_runtime": 37.2706, + "eval_samples_per_second": 26.831, + "eval_steps_per_second": 1.127, + "step": 1440 + }, + { + "epoch": 1.5791780821917807, + "grad_norm": 2.4129281044006348, + "learning_rate": 2.660438667749797e-06, + "loss": 1.7552, + "step": 1441 + }, + { + "epoch": 1.5802739726027397, + "grad_norm": 2.2975244522094727, + "learning_rate": 2.6584077985377744e-06, + "loss": 1.7318, + "step": 1442 + }, + { + "epoch": 1.5813698630136988, + "grad_norm": 2.4061625003814697, + "learning_rate": 2.6563769293257517e-06, + "loss": 1.7839, + "step": 1443 + }, + { + "epoch": 1.5824657534246576, + "grad_norm": 2.236353635787964, + "learning_rate": 2.654346060113729e-06, + "loss": 1.7995, + "step": 1444 + }, + { + "epoch": 1.5835616438356164, + "grad_norm": 2.2464141845703125, + "learning_rate": 2.6523151909017062e-06, + "loss": 1.7496, + "step": 1445 + }, + { + "epoch": 1.5846575342465754, + "grad_norm": 2.2744123935699463, + "learning_rate": 2.6502843216896835e-06, + "loss": 1.7703, + "step": 1446 + }, + { + "epoch": 1.5857534246575342, + "grad_norm": 2.1578054428100586, + "learning_rate": 2.6482534524776603e-06, + "loss": 1.7561, + "step": 1447 + }, + { + "epoch": 1.586849315068493, + "grad_norm": 2.1924850940704346, + "learning_rate": 2.646222583265638e-06, + "loss": 1.867, + "step": 1448 + }, + { + "epoch": 1.587945205479452, + "grad_norm": 2.303420305252075, + "learning_rate": 2.644191714053615e-06, + "loss": 1.8345, + "step": 1449 + }, + { + "epoch": 1.589041095890411, + "grad_norm": 2.308837413787842, + "learning_rate": 2.6421608448415926e-06, + "loss": 1.7979, + "step": 1450 + }, + { + "epoch": 1.589041095890411, + "eval_loss": 1.9836338758468628, + "eval_runtime": 37.253, + "eval_samples_per_second": 26.843, + "eval_steps_per_second": 1.127, + "step": 1450 + }, + { + "epoch": 1.59013698630137, + "grad_norm": 2.2896456718444824, + "learning_rate": 2.6401299756295694e-06, + "loss": 1.7679, + "step": 1451 + }, + { + "epoch": 1.5912328767123287, + "grad_norm": 2.2482378482818604, + "learning_rate": 2.638099106417547e-06, + "loss": 1.7481, + "step": 1452 + }, + { + "epoch": 1.5923287671232877, + "grad_norm": 2.2620620727539062, + "learning_rate": 2.636068237205524e-06, + "loss": 1.801, + "step": 1453 + }, + { + "epoch": 1.5934246575342466, + "grad_norm": 2.216510772705078, + "learning_rate": 2.6340373679935017e-06, + "loss": 1.7485, + "step": 1454 + }, + { + "epoch": 1.5945205479452054, + "grad_norm": 2.354858160018921, + "learning_rate": 2.6320064987814785e-06, + "loss": 1.8636, + "step": 1455 + }, + { + "epoch": 1.5956164383561644, + "grad_norm": 2.251880168914795, + "learning_rate": 2.6299756295694562e-06, + "loss": 1.7087, + "step": 1456 + }, + { + "epoch": 1.5967123287671234, + "grad_norm": 2.2605879306793213, + "learning_rate": 2.627944760357433e-06, + "loss": 1.8141, + "step": 1457 + }, + { + "epoch": 1.5978082191780822, + "grad_norm": 2.3315021991729736, + "learning_rate": 2.6259138911454104e-06, + "loss": 1.8565, + "step": 1458 + }, + { + "epoch": 1.598904109589041, + "grad_norm": 2.1999425888061523, + "learning_rate": 2.6238830219333877e-06, + "loss": 1.8683, + "step": 1459 + }, + { + "epoch": 1.6, + "grad_norm": 2.2652416229248047, + "learning_rate": 2.621852152721365e-06, + "loss": 1.8489, + "step": 1460 + }, + { + "epoch": 1.6, + "eval_loss": 1.9816856384277344, + "eval_runtime": 37.2727, + "eval_samples_per_second": 26.829, + "eval_steps_per_second": 1.127, + "step": 1460 + }, + { + "epoch": 1.601095890410959, + "grad_norm": 2.469440221786499, + "learning_rate": 2.619821283509342e-06, + "loss": 1.7335, + "step": 1461 + }, + { + "epoch": 1.6021917808219177, + "grad_norm": 2.434267044067383, + "learning_rate": 2.6177904142973195e-06, + "loss": 1.8291, + "step": 1462 + }, + { + "epoch": 1.6032876712328767, + "grad_norm": 2.300492286682129, + "learning_rate": 2.6157595450852968e-06, + "loss": 1.7504, + "step": 1463 + }, + { + "epoch": 1.6043835616438358, + "grad_norm": 2.3906843662261963, + "learning_rate": 2.613728675873274e-06, + "loss": 1.8114, + "step": 1464 + }, + { + "epoch": 1.6054794520547944, + "grad_norm": 2.5111773014068604, + "learning_rate": 2.6116978066612513e-06, + "loss": 1.836, + "step": 1465 + }, + { + "epoch": 1.6065753424657534, + "grad_norm": 2.3834846019744873, + "learning_rate": 2.6096669374492286e-06, + "loss": 1.8589, + "step": 1466 + }, + { + "epoch": 1.6076712328767124, + "grad_norm": 2.2709615230560303, + "learning_rate": 2.607636068237206e-06, + "loss": 1.8052, + "step": 1467 + }, + { + "epoch": 1.6087671232876712, + "grad_norm": 2.3351972103118896, + "learning_rate": 2.605605199025183e-06, + "loss": 1.7559, + "step": 1468 + }, + { + "epoch": 1.60986301369863, + "grad_norm": 2.3994698524475098, + "learning_rate": 2.6035743298131604e-06, + "loss": 1.8099, + "step": 1469 + }, + { + "epoch": 1.610958904109589, + "grad_norm": 2.352590322494507, + "learning_rate": 2.6015434606011377e-06, + "loss": 1.8043, + "step": 1470 + }, + { + "epoch": 1.610958904109589, + "eval_loss": 1.9790339469909668, + "eval_runtime": 37.2619, + "eval_samples_per_second": 26.837, + "eval_steps_per_second": 1.127, + "step": 1470 + }, + { + "epoch": 1.612054794520548, + "grad_norm": 2.35754656791687, + "learning_rate": 2.5995125913891145e-06, + "loss": 1.8252, + "step": 1471 + }, + { + "epoch": 1.6131506849315067, + "grad_norm": 2.4011359214782715, + "learning_rate": 2.5974817221770922e-06, + "loss": 1.8005, + "step": 1472 + }, + { + "epoch": 1.6142465753424657, + "grad_norm": 3.143979787826538, + "learning_rate": 2.595450852965069e-06, + "loss": 1.7952, + "step": 1473 + }, + { + "epoch": 1.6153424657534248, + "grad_norm": 2.3928465843200684, + "learning_rate": 2.593419983753047e-06, + "loss": 1.8344, + "step": 1474 + }, + { + "epoch": 1.6164383561643836, + "grad_norm": 2.370368480682373, + "learning_rate": 2.5913891145410236e-06, + "loss": 1.8358, + "step": 1475 + }, + { + "epoch": 1.6175342465753424, + "grad_norm": 2.6163346767425537, + "learning_rate": 2.5893582453290013e-06, + "loss": 1.7734, + "step": 1476 + }, + { + "epoch": 1.6186301369863014, + "grad_norm": 2.370539903640747, + "learning_rate": 2.587327376116978e-06, + "loss": 1.8383, + "step": 1477 + }, + { + "epoch": 1.6197260273972602, + "grad_norm": 2.397007465362549, + "learning_rate": 2.585296506904956e-06, + "loss": 1.7702, + "step": 1478 + }, + { + "epoch": 1.620821917808219, + "grad_norm": 2.4692702293395996, + "learning_rate": 2.5832656376929327e-06, + "loss": 1.7887, + "step": 1479 + }, + { + "epoch": 1.621917808219178, + "grad_norm": 2.5779969692230225, + "learning_rate": 2.5812347684809104e-06, + "loss": 1.7751, + "step": 1480 + }, + { + "epoch": 1.621917808219178, + "eval_loss": 1.9826369285583496, + "eval_runtime": 37.2779, + "eval_samples_per_second": 26.826, + "eval_steps_per_second": 1.127, + "step": 1480 + }, + { + "epoch": 1.6230136986301371, + "grad_norm": 2.3182761669158936, + "learning_rate": 2.5792038992688873e-06, + "loss": 1.7136, + "step": 1481 + }, + { + "epoch": 1.624109589041096, + "grad_norm": 2.4109292030334473, + "learning_rate": 2.577173030056865e-06, + "loss": 1.8804, + "step": 1482 + }, + { + "epoch": 1.6252054794520547, + "grad_norm": 2.426166534423828, + "learning_rate": 2.575142160844842e-06, + "loss": 1.6996, + "step": 1483 + }, + { + "epoch": 1.6263013698630138, + "grad_norm": 2.5459423065185547, + "learning_rate": 2.5731112916328187e-06, + "loss": 1.8186, + "step": 1484 + }, + { + "epoch": 1.6273972602739726, + "grad_norm": 2.583578586578369, + "learning_rate": 2.5710804224207964e-06, + "loss": 1.8275, + "step": 1485 + }, + { + "epoch": 1.6284931506849314, + "grad_norm": 2.442936658859253, + "learning_rate": 2.5690495532087733e-06, + "loss": 1.7553, + "step": 1486 + }, + { + "epoch": 1.6295890410958904, + "grad_norm": 2.587864637374878, + "learning_rate": 2.567018683996751e-06, + "loss": 1.7171, + "step": 1487 + }, + { + "epoch": 1.6306849315068495, + "grad_norm": 2.4119534492492676, + "learning_rate": 2.564987814784728e-06, + "loss": 1.8296, + "step": 1488 + }, + { + "epoch": 1.6317808219178083, + "grad_norm": 2.3582253456115723, + "learning_rate": 2.5629569455727055e-06, + "loss": 1.7595, + "step": 1489 + }, + { + "epoch": 1.632876712328767, + "grad_norm": 2.522674083709717, + "learning_rate": 2.5609260763606824e-06, + "loss": 1.7908, + "step": 1490 + }, + { + "epoch": 1.632876712328767, + "eval_loss": 1.9791436195373535, + "eval_runtime": 37.2967, + "eval_samples_per_second": 26.812, + "eval_steps_per_second": 1.126, + "step": 1490 + }, + { + "epoch": 1.633972602739726, + "grad_norm": 2.2927470207214355, + "learning_rate": 2.55889520714866e-06, + "loss": 1.7789, + "step": 1491 + }, + { + "epoch": 1.635068493150685, + "grad_norm": 2.4469518661499023, + "learning_rate": 2.556864337936637e-06, + "loss": 1.8937, + "step": 1492 + }, + { + "epoch": 1.6361643835616437, + "grad_norm": 2.36860728263855, + "learning_rate": 2.5548334687246146e-06, + "loss": 1.778, + "step": 1493 + }, + { + "epoch": 1.6372602739726028, + "grad_norm": 2.4953925609588623, + "learning_rate": 2.5528025995125915e-06, + "loss": 1.7894, + "step": 1494 + }, + { + "epoch": 1.6383561643835618, + "grad_norm": 2.3478338718414307, + "learning_rate": 2.550771730300569e-06, + "loss": 1.8327, + "step": 1495 + }, + { + "epoch": 1.6394520547945204, + "grad_norm": 2.4399492740631104, + "learning_rate": 2.548740861088546e-06, + "loss": 1.8062, + "step": 1496 + }, + { + "epoch": 1.6405479452054794, + "grad_norm": 2.4457216262817383, + "learning_rate": 2.5467099918765233e-06, + "loss": 1.8268, + "step": 1497 + }, + { + "epoch": 1.6416438356164385, + "grad_norm": 2.365879535675049, + "learning_rate": 2.5446791226645006e-06, + "loss": 1.7288, + "step": 1498 + }, + { + "epoch": 1.6427397260273973, + "grad_norm": 2.4942879676818848, + "learning_rate": 2.542648253452478e-06, + "loss": 1.7111, + "step": 1499 + }, + { + "epoch": 1.643835616438356, + "grad_norm": 2.275982618331909, + "learning_rate": 2.540617384240455e-06, + "loss": 1.8292, + "step": 1500 + }, + { + "epoch": 1.643835616438356, + "eval_loss": 1.9746516942977905, + "eval_runtime": 37.2247, + "eval_samples_per_second": 26.864, + "eval_steps_per_second": 1.128, + "step": 1500 + }, + { + "epoch": 1.644931506849315, + "grad_norm": 2.4430429935455322, + "learning_rate": 2.5385865150284324e-06, + "loss": 1.8034, + "step": 1501 + }, + { + "epoch": 1.6460273972602741, + "grad_norm": 2.2778332233428955, + "learning_rate": 2.5365556458164097e-06, + "loss": 1.7589, + "step": 1502 + }, + { + "epoch": 1.6471232876712327, + "grad_norm": 2.2818830013275146, + "learning_rate": 2.534524776604387e-06, + "loss": 1.7743, + "step": 1503 + }, + { + "epoch": 1.6482191780821918, + "grad_norm": 2.321328639984131, + "learning_rate": 2.5324939073923642e-06, + "loss": 1.7922, + "step": 1504 + }, + { + "epoch": 1.6493150684931508, + "grad_norm": 2.4312188625335693, + "learning_rate": 2.5304630381803415e-06, + "loss": 1.6784, + "step": 1505 + }, + { + "epoch": 1.6504109589041096, + "grad_norm": 2.298741102218628, + "learning_rate": 2.5284321689683188e-06, + "loss": 1.7745, + "step": 1506 + }, + { + "epoch": 1.6515068493150684, + "grad_norm": 2.2882816791534424, + "learning_rate": 2.526401299756296e-06, + "loss": 1.7896, + "step": 1507 + }, + { + "epoch": 1.6526027397260274, + "grad_norm": 2.2336251735687256, + "learning_rate": 2.524370430544273e-06, + "loss": 1.7548, + "step": 1508 + }, + { + "epoch": 1.6536986301369863, + "grad_norm": 2.4369094371795654, + "learning_rate": 2.5223395613322506e-06, + "loss": 1.8049, + "step": 1509 + }, + { + "epoch": 1.654794520547945, + "grad_norm": 2.2978999614715576, + "learning_rate": 2.5203086921202274e-06, + "loss": 1.7489, + "step": 1510 + }, + { + "epoch": 1.654794520547945, + "eval_loss": 1.9760501384735107, + "eval_runtime": 37.3461, + "eval_samples_per_second": 26.777, + "eval_steps_per_second": 1.125, + "step": 1510 + }, + { + "epoch": 1.655890410958904, + "grad_norm": 2.331296920776367, + "learning_rate": 2.518277822908205e-06, + "loss": 1.8293, + "step": 1511 + }, + { + "epoch": 1.6569863013698631, + "grad_norm": 2.4544663429260254, + "learning_rate": 2.516246953696182e-06, + "loss": 1.7617, + "step": 1512 + }, + { + "epoch": 1.658082191780822, + "grad_norm": 2.4682042598724365, + "learning_rate": 2.5142160844841597e-06, + "loss": 1.8054, + "step": 1513 + }, + { + "epoch": 1.6591780821917808, + "grad_norm": 2.2827956676483154, + "learning_rate": 2.5121852152721365e-06, + "loss": 1.8137, + "step": 1514 + }, + { + "epoch": 1.6602739726027398, + "grad_norm": 2.2961761951446533, + "learning_rate": 2.5101543460601142e-06, + "loss": 1.7938, + "step": 1515 + }, + { + "epoch": 1.6613698630136986, + "grad_norm": 2.3464298248291016, + "learning_rate": 2.508123476848091e-06, + "loss": 1.8026, + "step": 1516 + }, + { + "epoch": 1.6624657534246574, + "grad_norm": 2.4975106716156006, + "learning_rate": 2.506092607636069e-06, + "loss": 1.7457, + "step": 1517 + }, + { + "epoch": 1.6635616438356164, + "grad_norm": 2.4823005199432373, + "learning_rate": 2.5040617384240457e-06, + "loss": 1.8282, + "step": 1518 + }, + { + "epoch": 1.6646575342465755, + "grad_norm": 2.2875800132751465, + "learning_rate": 2.5020308692120234e-06, + "loss": 1.7569, + "step": 1519 + }, + { + "epoch": 1.6657534246575343, + "grad_norm": 2.451022148132324, + "learning_rate": 2.5e-06, + "loss": 1.7483, + "step": 1520 + }, + { + "epoch": 1.6657534246575343, + "eval_loss": 1.9748642444610596, + "eval_runtime": 37.4512, + "eval_samples_per_second": 26.701, + "eval_steps_per_second": 1.121, + "step": 1520 + }, + { + "epoch": 1.666849315068493, + "grad_norm": 2.3427352905273438, + "learning_rate": 2.4979691307879775e-06, + "loss": 1.8493, + "step": 1521 + }, + { + "epoch": 1.6679452054794521, + "grad_norm": 2.4496946334838867, + "learning_rate": 2.4959382615759548e-06, + "loss": 1.7569, + "step": 1522 + }, + { + "epoch": 1.669041095890411, + "grad_norm": 2.141963481903076, + "learning_rate": 2.493907392363932e-06, + "loss": 1.7553, + "step": 1523 + }, + { + "epoch": 1.6701369863013698, + "grad_norm": 2.4115428924560547, + "learning_rate": 2.4918765231519093e-06, + "loss": 1.8606, + "step": 1524 + }, + { + "epoch": 1.6712328767123288, + "grad_norm": 2.614616632461548, + "learning_rate": 2.4898456539398866e-06, + "loss": 1.8217, + "step": 1525 + }, + { + "epoch": 1.6723287671232878, + "grad_norm": 2.338557243347168, + "learning_rate": 2.487814784727864e-06, + "loss": 1.7672, + "step": 1526 + }, + { + "epoch": 1.6734246575342466, + "grad_norm": 2.4309446811676025, + "learning_rate": 2.4857839155158407e-06, + "loss": 1.8476, + "step": 1527 + }, + { + "epoch": 1.6745205479452054, + "grad_norm": 2.25883412361145, + "learning_rate": 2.483753046303818e-06, + "loss": 1.7827, + "step": 1528 + }, + { + "epoch": 1.6756164383561645, + "grad_norm": 2.264845371246338, + "learning_rate": 2.4817221770917953e-06, + "loss": 1.7877, + "step": 1529 + }, + { + "epoch": 1.6767123287671233, + "grad_norm": 2.3584723472595215, + "learning_rate": 2.4796913078797725e-06, + "loss": 1.7891, + "step": 1530 + }, + { + "epoch": 1.6767123287671233, + "eval_loss": 1.9751919507980347, + "eval_runtime": 37.558, + "eval_samples_per_second": 26.625, + "eval_steps_per_second": 1.118, + "step": 1530 + }, + { + "epoch": 1.677808219178082, + "grad_norm": 2.3555185794830322, + "learning_rate": 2.47766043866775e-06, + "loss": 1.8235, + "step": 1531 + }, + { + "epoch": 1.6789041095890411, + "grad_norm": 2.3086600303649902, + "learning_rate": 2.475629569455727e-06, + "loss": 1.8036, + "step": 1532 + }, + { + "epoch": 1.6800000000000002, + "grad_norm": 2.286921262741089, + "learning_rate": 2.4735987002437044e-06, + "loss": 1.7885, + "step": 1533 + }, + { + "epoch": 1.6810958904109587, + "grad_norm": 2.2671141624450684, + "learning_rate": 2.4715678310316816e-06, + "loss": 1.758, + "step": 1534 + }, + { + "epoch": 1.6821917808219178, + "grad_norm": 2.4795749187469482, + "learning_rate": 2.469536961819659e-06, + "loss": 1.7791, + "step": 1535 + }, + { + "epoch": 1.6832876712328768, + "grad_norm": 2.201650857925415, + "learning_rate": 2.467506092607636e-06, + "loss": 1.7802, + "step": 1536 + }, + { + "epoch": 1.6843835616438356, + "grad_norm": 2.3804843425750732, + "learning_rate": 2.4654752233956135e-06, + "loss": 1.7739, + "step": 1537 + }, + { + "epoch": 1.6854794520547944, + "grad_norm": 2.322216510772705, + "learning_rate": 2.4634443541835907e-06, + "loss": 1.758, + "step": 1538 + }, + { + "epoch": 1.6865753424657535, + "grad_norm": 2.3582146167755127, + "learning_rate": 2.461413484971568e-06, + "loss": 1.8439, + "step": 1539 + }, + { + "epoch": 1.6876712328767123, + "grad_norm": 2.3597848415374756, + "learning_rate": 2.4593826157595453e-06, + "loss": 1.7772, + "step": 1540 + }, + { + "epoch": 1.6876712328767123, + "eval_loss": 1.9731802940368652, + "eval_runtime": 37.2439, + "eval_samples_per_second": 26.85, + "eval_steps_per_second": 1.128, + "step": 1540 + }, + { + "epoch": 1.688767123287671, + "grad_norm": 2.404569149017334, + "learning_rate": 2.4573517465475226e-06, + "loss": 1.8683, + "step": 1541 + }, + { + "epoch": 1.6898630136986301, + "grad_norm": 2.318737268447876, + "learning_rate": 2.4553208773355e-06, + "loss": 1.7223, + "step": 1542 + }, + { + "epoch": 1.6909589041095892, + "grad_norm": 2.2535340785980225, + "learning_rate": 2.453290008123477e-06, + "loss": 1.7693, + "step": 1543 + }, + { + "epoch": 1.692054794520548, + "grad_norm": 2.2533493041992188, + "learning_rate": 2.4512591389114544e-06, + "loss": 1.7876, + "step": 1544 + }, + { + "epoch": 1.6931506849315068, + "grad_norm": 2.253278970718384, + "learning_rate": 2.4492282696994317e-06, + "loss": 1.788, + "step": 1545 + }, + { + "epoch": 1.6942465753424658, + "grad_norm": 2.296807289123535, + "learning_rate": 2.447197400487409e-06, + "loss": 1.7877, + "step": 1546 + }, + { + "epoch": 1.6953424657534246, + "grad_norm": 2.4022111892700195, + "learning_rate": 2.4451665312753862e-06, + "loss": 1.7986, + "step": 1547 + }, + { + "epoch": 1.6964383561643834, + "grad_norm": 2.2671430110931396, + "learning_rate": 2.4431356620633635e-06, + "loss": 1.8269, + "step": 1548 + }, + { + "epoch": 1.6975342465753425, + "grad_norm": 2.338252067565918, + "learning_rate": 2.4411047928513408e-06, + "loss": 1.8279, + "step": 1549 + }, + { + "epoch": 1.6986301369863015, + "grad_norm": 2.365365505218506, + "learning_rate": 2.439073923639318e-06, + "loss": 1.7895, + "step": 1550 + }, + { + "epoch": 1.6986301369863015, + "eval_loss": 1.9711860418319702, + "eval_runtime": 37.2757, + "eval_samples_per_second": 26.827, + "eval_steps_per_second": 1.127, + "step": 1550 + }, + { + "epoch": 1.6997260273972603, + "grad_norm": 2.407580614089966, + "learning_rate": 2.4370430544272953e-06, + "loss": 1.7985, + "step": 1551 + }, + { + "epoch": 1.7008219178082191, + "grad_norm": 2.3145346641540527, + "learning_rate": 2.435012185215272e-06, + "loss": 1.7589, + "step": 1552 + }, + { + "epoch": 1.7019178082191782, + "grad_norm": 2.3809280395507812, + "learning_rate": 2.4329813160032495e-06, + "loss": 1.7334, + "step": 1553 + }, + { + "epoch": 1.703013698630137, + "grad_norm": 2.4224488735198975, + "learning_rate": 2.4309504467912267e-06, + "loss": 1.7549, + "step": 1554 + }, + { + "epoch": 1.7041095890410958, + "grad_norm": 2.374133348464966, + "learning_rate": 2.428919577579204e-06, + "loss": 1.7923, + "step": 1555 + }, + { + "epoch": 1.7052054794520548, + "grad_norm": 2.258222818374634, + "learning_rate": 2.4268887083671813e-06, + "loss": 1.7683, + "step": 1556 + }, + { + "epoch": 1.7063013698630138, + "grad_norm": 2.4481396675109863, + "learning_rate": 2.4248578391551586e-06, + "loss": 1.8609, + "step": 1557 + }, + { + "epoch": 1.7073972602739727, + "grad_norm": 2.3932790756225586, + "learning_rate": 2.422826969943136e-06, + "loss": 1.7475, + "step": 1558 + }, + { + "epoch": 1.7084931506849315, + "grad_norm": 2.3034355640411377, + "learning_rate": 2.420796100731113e-06, + "loss": 1.7664, + "step": 1559 + }, + { + "epoch": 1.7095890410958905, + "grad_norm": 2.2018799781799316, + "learning_rate": 2.4187652315190904e-06, + "loss": 1.7328, + "step": 1560 + }, + { + "epoch": 1.7095890410958905, + "eval_loss": 1.9715769290924072, + "eval_runtime": 37.2507, + "eval_samples_per_second": 26.845, + "eval_steps_per_second": 1.127, + "step": 1560 + }, + { + "epoch": 1.7106849315068493, + "grad_norm": 2.396462917327881, + "learning_rate": 2.4167343623070677e-06, + "loss": 1.7773, + "step": 1561 + }, + { + "epoch": 1.7117808219178081, + "grad_norm": 2.3212695121765137, + "learning_rate": 2.414703493095045e-06, + "loss": 1.7966, + "step": 1562 + }, + { + "epoch": 1.7128767123287671, + "grad_norm": 2.4157631397247314, + "learning_rate": 2.4126726238830222e-06, + "loss": 1.7797, + "step": 1563 + }, + { + "epoch": 1.7139726027397262, + "grad_norm": 2.404907464981079, + "learning_rate": 2.410641754670999e-06, + "loss": 1.8093, + "step": 1564 + }, + { + "epoch": 1.7150684931506848, + "grad_norm": 2.3798625469207764, + "learning_rate": 2.4086108854589763e-06, + "loss": 1.7635, + "step": 1565 + }, + { + "epoch": 1.7161643835616438, + "grad_norm": 2.325932025909424, + "learning_rate": 2.4065800162469536e-06, + "loss": 1.8281, + "step": 1566 + }, + { + "epoch": 1.7172602739726028, + "grad_norm": 2.413975238800049, + "learning_rate": 2.404549147034931e-06, + "loss": 1.7609, + "step": 1567 + }, + { + "epoch": 1.7183561643835616, + "grad_norm": 2.512775421142578, + "learning_rate": 2.402518277822908e-06, + "loss": 1.7793, + "step": 1568 + }, + { + "epoch": 1.7194520547945205, + "grad_norm": 2.581679582595825, + "learning_rate": 2.4004874086108854e-06, + "loss": 1.8302, + "step": 1569 + }, + { + "epoch": 1.7205479452054795, + "grad_norm": 2.389822244644165, + "learning_rate": 2.3984565393988627e-06, + "loss": 1.7573, + "step": 1570 + }, + { + "epoch": 1.7205479452054795, + "eval_loss": 1.9670331478118896, + "eval_runtime": 37.2901, + "eval_samples_per_second": 26.817, + "eval_steps_per_second": 1.126, + "step": 1570 + }, + { + "epoch": 1.7216438356164385, + "grad_norm": 2.6191556453704834, + "learning_rate": 2.39642567018684e-06, + "loss": 1.8252, + "step": 1571 + }, + { + "epoch": 1.7227397260273971, + "grad_norm": 2.394347906112671, + "learning_rate": 2.3943948009748173e-06, + "loss": 1.7561, + "step": 1572 + }, + { + "epoch": 1.7238356164383561, + "grad_norm": 2.3016326427459717, + "learning_rate": 2.3923639317627945e-06, + "loss": 1.7619, + "step": 1573 + }, + { + "epoch": 1.7249315068493152, + "grad_norm": 2.3663971424102783, + "learning_rate": 2.390333062550772e-06, + "loss": 1.7922, + "step": 1574 + }, + { + "epoch": 1.726027397260274, + "grad_norm": 2.3328938484191895, + "learning_rate": 2.388302193338749e-06, + "loss": 1.7374, + "step": 1575 + }, + { + "epoch": 1.7271232876712328, + "grad_norm": 2.313525676727295, + "learning_rate": 2.3862713241267264e-06, + "loss": 1.6585, + "step": 1576 + }, + { + "epoch": 1.7282191780821918, + "grad_norm": 2.4609642028808594, + "learning_rate": 2.3842404549147037e-06, + "loss": 1.8757, + "step": 1577 + }, + { + "epoch": 1.7293150684931506, + "grad_norm": 2.3502120971679688, + "learning_rate": 2.382209585702681e-06, + "loss": 1.7786, + "step": 1578 + }, + { + "epoch": 1.7304109589041095, + "grad_norm": 2.383025884628296, + "learning_rate": 2.380178716490658e-06, + "loss": 1.7502, + "step": 1579 + }, + { + "epoch": 1.7315068493150685, + "grad_norm": 2.280890464782715, + "learning_rate": 2.3781478472786355e-06, + "loss": 1.8192, + "step": 1580 + }, + { + "epoch": 1.7315068493150685, + "eval_loss": 1.9658632278442383, + "eval_runtime": 37.3714, + "eval_samples_per_second": 26.758, + "eval_steps_per_second": 1.124, + "step": 1580 + }, + { + "epoch": 1.7326027397260275, + "grad_norm": 2.342240333557129, + "learning_rate": 2.3761169780666128e-06, + "loss": 1.7786, + "step": 1581 + }, + { + "epoch": 1.7336986301369863, + "grad_norm": 2.2916715145111084, + "learning_rate": 2.37408610885459e-06, + "loss": 1.8161, + "step": 1582 + }, + { + "epoch": 1.7347945205479451, + "grad_norm": 2.24537992477417, + "learning_rate": 2.3720552396425673e-06, + "loss": 1.8442, + "step": 1583 + }, + { + "epoch": 1.7358904109589042, + "grad_norm": 2.3003411293029785, + "learning_rate": 2.3700243704305446e-06, + "loss": 1.7775, + "step": 1584 + }, + { + "epoch": 1.736986301369863, + "grad_norm": 2.3062400817871094, + "learning_rate": 2.367993501218522e-06, + "loss": 1.8374, + "step": 1585 + }, + { + "epoch": 1.7380821917808218, + "grad_norm": 2.3534083366394043, + "learning_rate": 2.365962632006499e-06, + "loss": 1.7829, + "step": 1586 + }, + { + "epoch": 1.7391780821917808, + "grad_norm": 2.2612221240997314, + "learning_rate": 2.3639317627944764e-06, + "loss": 1.7934, + "step": 1587 + }, + { + "epoch": 1.7402739726027399, + "grad_norm": 2.4197304248809814, + "learning_rate": 2.3619008935824537e-06, + "loss": 1.8226, + "step": 1588 + }, + { + "epoch": 1.7413698630136987, + "grad_norm": 2.435908079147339, + "learning_rate": 2.3598700243704305e-06, + "loss": 1.8148, + "step": 1589 + }, + { + "epoch": 1.7424657534246575, + "grad_norm": 2.3617560863494873, + "learning_rate": 2.357839155158408e-06, + "loss": 1.8085, + "step": 1590 + }, + { + "epoch": 1.7424657534246575, + "eval_loss": 1.9636695384979248, + "eval_runtime": 37.2968, + "eval_samples_per_second": 26.812, + "eval_steps_per_second": 1.126, + "step": 1590 + }, + { + "epoch": 1.7435616438356165, + "grad_norm": 2.295182228088379, + "learning_rate": 2.355808285946385e-06, + "loss": 1.8091, + "step": 1591 + }, + { + "epoch": 1.7446575342465753, + "grad_norm": 2.3347294330596924, + "learning_rate": 2.3537774167343624e-06, + "loss": 1.8086, + "step": 1592 + }, + { + "epoch": 1.7457534246575341, + "grad_norm": 2.4760799407958984, + "learning_rate": 2.3517465475223396e-06, + "loss": 1.8227, + "step": 1593 + }, + { + "epoch": 1.7468493150684932, + "grad_norm": 2.4243907928466797, + "learning_rate": 2.349715678310317e-06, + "loss": 1.7921, + "step": 1594 + }, + { + "epoch": 1.7479452054794522, + "grad_norm": 2.271735668182373, + "learning_rate": 2.347684809098294e-06, + "loss": 1.7987, + "step": 1595 + }, + { + "epoch": 1.749041095890411, + "grad_norm": 2.5412354469299316, + "learning_rate": 2.3456539398862715e-06, + "loss": 1.7558, + "step": 1596 + }, + { + "epoch": 1.7501369863013698, + "grad_norm": 2.464944362640381, + "learning_rate": 2.3436230706742487e-06, + "loss": 1.8087, + "step": 1597 + }, + { + "epoch": 1.7512328767123289, + "grad_norm": 2.4210381507873535, + "learning_rate": 2.341592201462226e-06, + "loss": 1.756, + "step": 1598 + }, + { + "epoch": 1.7523287671232877, + "grad_norm": 2.284945011138916, + "learning_rate": 2.3395613322502033e-06, + "loss": 1.7909, + "step": 1599 + }, + { + "epoch": 1.7534246575342465, + "grad_norm": 2.3740732669830322, + "learning_rate": 2.3375304630381806e-06, + "loss": 1.7683, + "step": 1600 + }, + { + "epoch": 1.7534246575342465, + "eval_loss": 1.9646408557891846, + "eval_runtime": 37.3266, + "eval_samples_per_second": 26.791, + "eval_steps_per_second": 1.125, + "step": 1600 + }, + { + "epoch": 1.7545205479452055, + "grad_norm": 2.399583578109741, + "learning_rate": 2.335499593826158e-06, + "loss": 1.7342, + "step": 1601 + }, + { + "epoch": 1.7556164383561645, + "grad_norm": 2.4731597900390625, + "learning_rate": 2.333468724614135e-06, + "loss": 1.7507, + "step": 1602 + }, + { + "epoch": 1.7567123287671231, + "grad_norm": 2.365433692932129, + "learning_rate": 2.3314378554021124e-06, + "loss": 1.7944, + "step": 1603 + }, + { + "epoch": 1.7578082191780822, + "grad_norm": 2.4500601291656494, + "learning_rate": 2.3294069861900897e-06, + "loss": 1.822, + "step": 1604 + }, + { + "epoch": 1.7589041095890412, + "grad_norm": 2.3339014053344727, + "learning_rate": 2.327376116978067e-06, + "loss": 1.7464, + "step": 1605 + }, + { + "epoch": 1.76, + "grad_norm": 2.343973159790039, + "learning_rate": 2.3253452477660442e-06, + "loss": 1.7408, + "step": 1606 + }, + { + "epoch": 1.7610958904109588, + "grad_norm": 2.453462839126587, + "learning_rate": 2.3233143785540215e-06, + "loss": 1.8261, + "step": 1607 + }, + { + "epoch": 1.7621917808219179, + "grad_norm": 2.357698678970337, + "learning_rate": 2.3212835093419988e-06, + "loss": 1.762, + "step": 1608 + }, + { + "epoch": 1.7632876712328767, + "grad_norm": 2.353257417678833, + "learning_rate": 2.319252640129976e-06, + "loss": 1.8076, + "step": 1609 + }, + { + "epoch": 1.7643835616438355, + "grad_norm": 2.3269948959350586, + "learning_rate": 2.3172217709179533e-06, + "loss": 1.7516, + "step": 1610 + }, + { + "epoch": 1.7643835616438355, + "eval_loss": 1.9678465127944946, + "eval_runtime": 37.2052, + "eval_samples_per_second": 26.878, + "eval_steps_per_second": 1.129, + "step": 1610 + }, + { + "epoch": 1.7654794520547945, + "grad_norm": 2.4023501873016357, + "learning_rate": 2.3151909017059306e-06, + "loss": 1.8477, + "step": 1611 + }, + { + "epoch": 1.7665753424657535, + "grad_norm": 2.3356990814208984, + "learning_rate": 2.313160032493908e-06, + "loss": 1.8236, + "step": 1612 + }, + { + "epoch": 1.7676712328767124, + "grad_norm": 2.3571391105651855, + "learning_rate": 2.311129163281885e-06, + "loss": 1.7335, + "step": 1613 + }, + { + "epoch": 1.7687671232876712, + "grad_norm": 2.3510208129882812, + "learning_rate": 2.309098294069862e-06, + "loss": 1.8088, + "step": 1614 + }, + { + "epoch": 1.7698630136986302, + "grad_norm": 2.3852577209472656, + "learning_rate": 2.3070674248578393e-06, + "loss": 1.7312, + "step": 1615 + }, + { + "epoch": 1.770958904109589, + "grad_norm": 2.3085618019104004, + "learning_rate": 2.3050365556458166e-06, + "loss": 1.7379, + "step": 1616 + }, + { + "epoch": 1.7720547945205478, + "grad_norm": 2.3454065322875977, + "learning_rate": 2.303005686433794e-06, + "loss": 1.8054, + "step": 1617 + }, + { + "epoch": 1.7731506849315068, + "grad_norm": 2.4034247398376465, + "learning_rate": 2.300974817221771e-06, + "loss": 1.7955, + "step": 1618 + }, + { + "epoch": 1.7742465753424659, + "grad_norm": 2.4181671142578125, + "learning_rate": 2.2989439480097484e-06, + "loss": 1.7732, + "step": 1619 + }, + { + "epoch": 1.7753424657534247, + "grad_norm": 2.3408010005950928, + "learning_rate": 2.2969130787977257e-06, + "loss": 1.7231, + "step": 1620 + }, + { + "epoch": 1.7753424657534247, + "eval_loss": 1.9655163288116455, + "eval_runtime": 37.4878, + "eval_samples_per_second": 26.675, + "eval_steps_per_second": 1.12, + "step": 1620 + }, + { + "epoch": 1.7764383561643835, + "grad_norm": 2.3113718032836914, + "learning_rate": 2.294882209585703e-06, + "loss": 1.7904, + "step": 1621 + }, + { + "epoch": 1.7775342465753425, + "grad_norm": 2.3037896156311035, + "learning_rate": 2.2928513403736802e-06, + "loss": 1.8069, + "step": 1622 + }, + { + "epoch": 1.7786301369863013, + "grad_norm": 2.3962693214416504, + "learning_rate": 2.2908204711616575e-06, + "loss": 1.7354, + "step": 1623 + }, + { + "epoch": 1.7797260273972602, + "grad_norm": 2.3499252796173096, + "learning_rate": 2.2887896019496348e-06, + "loss": 1.779, + "step": 1624 + }, + { + "epoch": 1.7808219178082192, + "grad_norm": 2.390958070755005, + "learning_rate": 2.286758732737612e-06, + "loss": 1.8257, + "step": 1625 + }, + { + "epoch": 1.7819178082191782, + "grad_norm": 2.365626573562622, + "learning_rate": 2.284727863525589e-06, + "loss": 1.7618, + "step": 1626 + }, + { + "epoch": 1.783013698630137, + "grad_norm": 2.3558530807495117, + "learning_rate": 2.282696994313566e-06, + "loss": 1.7729, + "step": 1627 + }, + { + "epoch": 1.7841095890410958, + "grad_norm": 2.3276045322418213, + "learning_rate": 2.2806661251015434e-06, + "loss": 1.8231, + "step": 1628 + }, + { + "epoch": 1.7852054794520549, + "grad_norm": 2.3369953632354736, + "learning_rate": 2.2786352558895207e-06, + "loss": 1.8011, + "step": 1629 + }, + { + "epoch": 1.7863013698630137, + "grad_norm": 2.3588311672210693, + "learning_rate": 2.276604386677498e-06, + "loss": 1.7682, + "step": 1630 + }, + { + "epoch": 1.7863013698630137, + "eval_loss": 1.963529348373413, + "eval_runtime": 37.5589, + "eval_samples_per_second": 26.625, + "eval_steps_per_second": 1.118, + "step": 1630 + }, + { + "epoch": 1.7873972602739725, + "grad_norm": 2.2918803691864014, + "learning_rate": 2.2745735174654753e-06, + "loss": 1.7256, + "step": 1631 + }, + { + "epoch": 1.7884931506849315, + "grad_norm": 2.423738479614258, + "learning_rate": 2.2725426482534526e-06, + "loss": 1.7582, + "step": 1632 + }, + { + "epoch": 1.7895890410958906, + "grad_norm": 2.3500850200653076, + "learning_rate": 2.27051177904143e-06, + "loss": 1.8255, + "step": 1633 + }, + { + "epoch": 1.7906849315068492, + "grad_norm": 2.3280837535858154, + "learning_rate": 2.268480909829407e-06, + "loss": 1.7845, + "step": 1634 + }, + { + "epoch": 1.7917808219178082, + "grad_norm": 2.5055465698242188, + "learning_rate": 2.2664500406173844e-06, + "loss": 1.78, + "step": 1635 + }, + { + "epoch": 1.7928767123287672, + "grad_norm": 2.3132143020629883, + "learning_rate": 2.2644191714053617e-06, + "loss": 1.7814, + "step": 1636 + }, + { + "epoch": 1.793972602739726, + "grad_norm": 2.316361427307129, + "learning_rate": 2.262388302193339e-06, + "loss": 1.8161, + "step": 1637 + }, + { + "epoch": 1.7950684931506848, + "grad_norm": 2.337318181991577, + "learning_rate": 2.260357432981316e-06, + "loss": 1.7942, + "step": 1638 + }, + { + "epoch": 1.7961643835616439, + "grad_norm": 2.362031936645508, + "learning_rate": 2.2583265637692935e-06, + "loss": 1.7771, + "step": 1639 + }, + { + "epoch": 1.7972602739726027, + "grad_norm": 2.3260183334350586, + "learning_rate": 2.2562956945572708e-06, + "loss": 1.7975, + "step": 1640 + }, + { + "epoch": 1.7972602739726027, + "eval_loss": 1.9619382619857788, + "eval_runtime": 37.2606, + "eval_samples_per_second": 26.838, + "eval_steps_per_second": 1.127, + "step": 1640 + }, + { + "epoch": 1.7983561643835615, + "grad_norm": 2.4260714054107666, + "learning_rate": 2.254264825345248e-06, + "loss": 1.7654, + "step": 1641 + }, + { + "epoch": 1.7994520547945205, + "grad_norm": 2.297428846359253, + "learning_rate": 2.2522339561332253e-06, + "loss": 1.775, + "step": 1642 + }, + { + "epoch": 1.8005479452054796, + "grad_norm": 2.5153353214263916, + "learning_rate": 2.2502030869212026e-06, + "loss": 1.7796, + "step": 1643 + }, + { + "epoch": 1.8016438356164384, + "grad_norm": 2.5102577209472656, + "learning_rate": 2.24817221770918e-06, + "loss": 1.7917, + "step": 1644 + }, + { + "epoch": 1.8027397260273972, + "grad_norm": 2.3693106174468994, + "learning_rate": 2.246141348497157e-06, + "loss": 1.8459, + "step": 1645 + }, + { + "epoch": 1.8038356164383562, + "grad_norm": 2.2353153228759766, + "learning_rate": 2.2441104792851344e-06, + "loss": 1.8095, + "step": 1646 + }, + { + "epoch": 1.804931506849315, + "grad_norm": 2.265465497970581, + "learning_rate": 2.2420796100731117e-06, + "loss": 1.7623, + "step": 1647 + }, + { + "epoch": 1.8060273972602738, + "grad_norm": 2.2820940017700195, + "learning_rate": 2.240048740861089e-06, + "loss": 1.7946, + "step": 1648 + }, + { + "epoch": 1.8071232876712329, + "grad_norm": 2.378035068511963, + "learning_rate": 2.2380178716490662e-06, + "loss": 1.8059, + "step": 1649 + }, + { + "epoch": 1.808219178082192, + "grad_norm": 2.328341245651245, + "learning_rate": 2.2359870024370435e-06, + "loss": 1.6611, + "step": 1650 + }, + { + "epoch": 1.808219178082192, + "eval_loss": 1.9600144624710083, + "eval_runtime": 37.2601, + "eval_samples_per_second": 26.838, + "eval_steps_per_second": 1.127, + "step": 1650 + }, + { + "epoch": 1.8093150684931507, + "grad_norm": 2.2819113731384277, + "learning_rate": 2.2339561332250204e-06, + "loss": 1.8058, + "step": 1651 + }, + { + "epoch": 1.8104109589041095, + "grad_norm": 2.4178121089935303, + "learning_rate": 2.2319252640129976e-06, + "loss": 1.8204, + "step": 1652 + }, + { + "epoch": 1.8115068493150686, + "grad_norm": 2.497008800506592, + "learning_rate": 2.229894394800975e-06, + "loss": 1.7968, + "step": 1653 + }, + { + "epoch": 1.8126027397260274, + "grad_norm": 2.4681501388549805, + "learning_rate": 2.227863525588952e-06, + "loss": 1.7292, + "step": 1654 + }, + { + "epoch": 1.8136986301369862, + "grad_norm": 2.2983527183532715, + "learning_rate": 2.2258326563769295e-06, + "loss": 1.7097, + "step": 1655 + }, + { + "epoch": 1.8147945205479452, + "grad_norm": 2.349454164505005, + "learning_rate": 2.2238017871649067e-06, + "loss": 1.7256, + "step": 1656 + }, + { + "epoch": 1.8158904109589042, + "grad_norm": 2.4247677326202393, + "learning_rate": 2.221770917952884e-06, + "loss": 1.7165, + "step": 1657 + }, + { + "epoch": 1.816986301369863, + "grad_norm": 2.3035342693328857, + "learning_rate": 2.2197400487408613e-06, + "loss": 1.7811, + "step": 1658 + }, + { + "epoch": 1.8180821917808219, + "grad_norm": 2.318772554397583, + "learning_rate": 2.2177091795288386e-06, + "loss": 1.8082, + "step": 1659 + }, + { + "epoch": 1.819178082191781, + "grad_norm": 2.2477188110351562, + "learning_rate": 2.215678310316816e-06, + "loss": 1.7206, + "step": 1660 + }, + { + "epoch": 1.819178082191781, + "eval_loss": 1.962982416152954, + "eval_runtime": 37.2709, + "eval_samples_per_second": 26.831, + "eval_steps_per_second": 1.127, + "step": 1660 + }, + { + "epoch": 1.8202739726027397, + "grad_norm": 2.355264186859131, + "learning_rate": 2.213647441104793e-06, + "loss": 1.7634, + "step": 1661 + }, + { + "epoch": 1.8213698630136985, + "grad_norm": 2.267458200454712, + "learning_rate": 2.2116165718927704e-06, + "loss": 1.7819, + "step": 1662 + }, + { + "epoch": 1.8224657534246576, + "grad_norm": 2.2487895488739014, + "learning_rate": 2.2095857026807473e-06, + "loss": 1.759, + "step": 1663 + }, + { + "epoch": 1.8235616438356166, + "grad_norm": 2.4253578186035156, + "learning_rate": 2.2075548334687245e-06, + "loss": 1.7937, + "step": 1664 + }, + { + "epoch": 1.8246575342465754, + "grad_norm": 2.478050470352173, + "learning_rate": 2.205523964256702e-06, + "loss": 1.7241, + "step": 1665 + }, + { + "epoch": 1.8257534246575342, + "grad_norm": 2.289905309677124, + "learning_rate": 2.203493095044679e-06, + "loss": 1.7531, + "step": 1666 + }, + { + "epoch": 1.8268493150684932, + "grad_norm": 2.303421974182129, + "learning_rate": 2.2014622258326564e-06, + "loss": 1.7711, + "step": 1667 + }, + { + "epoch": 1.827945205479452, + "grad_norm": 2.3670248985290527, + "learning_rate": 2.1994313566206336e-06, + "loss": 1.8093, + "step": 1668 + }, + { + "epoch": 1.8290410958904109, + "grad_norm": 2.3721048831939697, + "learning_rate": 2.197400487408611e-06, + "loss": 1.8273, + "step": 1669 + }, + { + "epoch": 1.83013698630137, + "grad_norm": 2.3000190258026123, + "learning_rate": 2.195369618196588e-06, + "loss": 1.75, + "step": 1670 + }, + { + "epoch": 1.83013698630137, + "eval_loss": 1.9577263593673706, + "eval_runtime": 37.2953, + "eval_samples_per_second": 26.813, + "eval_steps_per_second": 1.126, + "step": 1670 + }, + { + "epoch": 1.831232876712329, + "grad_norm": 2.303584575653076, + "learning_rate": 2.1933387489845655e-06, + "loss": 1.7278, + "step": 1671 + }, + { + "epoch": 1.8323287671232875, + "grad_norm": 2.260579824447632, + "learning_rate": 2.1913078797725427e-06, + "loss": 1.7524, + "step": 1672 + }, + { + "epoch": 1.8334246575342465, + "grad_norm": 2.311967372894287, + "learning_rate": 2.18927701056052e-06, + "loss": 1.7799, + "step": 1673 + }, + { + "epoch": 1.8345205479452056, + "grad_norm": 2.338989496231079, + "learning_rate": 2.1872461413484973e-06, + "loss": 1.8468, + "step": 1674 + }, + { + "epoch": 1.8356164383561644, + "grad_norm": 2.338989496231079, + "learning_rate": 2.1872461413484973e-06, + "loss": 1.7853, + "step": 1675 + }, + { + "epoch": 1.8367123287671232, + "grad_norm": 2.526724100112915, + "learning_rate": 2.1852152721364746e-06, + "loss": 1.78, + "step": 1676 + }, + { + "epoch": 1.8378082191780822, + "grad_norm": 2.4637107849121094, + "learning_rate": 2.183184402924452e-06, + "loss": 1.7328, + "step": 1677 + }, + { + "epoch": 1.838904109589041, + "grad_norm": 2.34454345703125, + "learning_rate": 2.181153533712429e-06, + "loss": 1.8058, + "step": 1678 + }, + { + "epoch": 1.8399999999999999, + "grad_norm": 2.5518760681152344, + "learning_rate": 2.1791226645004064e-06, + "loss": 1.7746, + "step": 1679 + }, + { + "epoch": 1.841095890410959, + "grad_norm": 2.437443971633911, + "learning_rate": 2.1770917952883837e-06, + "loss": 1.7517, + "step": 1680 + }, + { + "epoch": 1.841095890410959, + "eval_loss": 1.9572285413742065, + "eval_runtime": 37.3402, + "eval_samples_per_second": 26.781, + "eval_steps_per_second": 1.125, + "step": 1680 + }, + { + "epoch": 1.842191780821918, + "grad_norm": 2.344287157058716, + "learning_rate": 2.175060926076361e-06, + "loss": 1.8229, + "step": 1681 + }, + { + "epoch": 1.8432876712328767, + "grad_norm": 2.333840847015381, + "learning_rate": 2.1730300568643382e-06, + "loss": 1.8534, + "step": 1682 + }, + { + "epoch": 1.8443835616438355, + "grad_norm": 2.372926950454712, + "learning_rate": 2.1709991876523155e-06, + "loss": 1.7609, + "step": 1683 + }, + { + "epoch": 1.8454794520547946, + "grad_norm": 2.2858645915985107, + "learning_rate": 2.1689683184402928e-06, + "loss": 1.7508, + "step": 1684 + }, + { + "epoch": 1.8465753424657534, + "grad_norm": 2.326586961746216, + "learning_rate": 2.16693744922827e-06, + "loss": 1.8122, + "step": 1685 + }, + { + "epoch": 1.8476712328767122, + "grad_norm": 2.3605616092681885, + "learning_rate": 2.1649065800162473e-06, + "loss": 1.7794, + "step": 1686 + }, + { + "epoch": 1.8487671232876712, + "grad_norm": 2.4188039302825928, + "learning_rate": 2.1628757108042246e-06, + "loss": 1.7552, + "step": 1687 + }, + { + "epoch": 1.8498630136986303, + "grad_norm": 2.335617780685425, + "learning_rate": 2.160844841592202e-06, + "loss": 1.7259, + "step": 1688 + }, + { + "epoch": 1.850958904109589, + "grad_norm": 2.245882272720337, + "learning_rate": 2.1588139723801787e-06, + "loss": 1.7259, + "step": 1689 + }, + { + "epoch": 1.8520547945205479, + "grad_norm": 2.4447500705718994, + "learning_rate": 2.156783103168156e-06, + "loss": 1.7605, + "step": 1690 + }, + { + "epoch": 1.8520547945205479, + "eval_loss": 1.9608982801437378, + "eval_runtime": 37.3756, + "eval_samples_per_second": 26.755, + "eval_steps_per_second": 1.124, + "step": 1690 + }, + { + "epoch": 1.853150684931507, + "grad_norm": 2.363443613052368, + "learning_rate": 2.1547522339561333e-06, + "loss": 1.7336, + "step": 1691 + }, + { + "epoch": 1.8542465753424657, + "grad_norm": 2.3573989868164062, + "learning_rate": 2.1527213647441106e-06, + "loss": 1.7953, + "step": 1692 + }, + { + "epoch": 1.8553424657534245, + "grad_norm": 2.4292593002319336, + "learning_rate": 2.150690495532088e-06, + "loss": 1.8276, + "step": 1693 + }, + { + "epoch": 1.8564383561643836, + "grad_norm": 2.374563694000244, + "learning_rate": 2.148659626320065e-06, + "loss": 1.7975, + "step": 1694 + }, + { + "epoch": 1.8575342465753426, + "grad_norm": 2.3935306072235107, + "learning_rate": 2.1466287571080424e-06, + "loss": 1.7824, + "step": 1695 + }, + { + "epoch": 1.8586301369863014, + "grad_norm": 2.343773365020752, + "learning_rate": 2.1445978878960197e-06, + "loss": 1.7574, + "step": 1696 + }, + { + "epoch": 1.8597260273972602, + "grad_norm": 2.2765426635742188, + "learning_rate": 2.142567018683997e-06, + "loss": 1.7682, + "step": 1697 + }, + { + "epoch": 1.8608219178082193, + "grad_norm": 2.339470863342285, + "learning_rate": 2.140536149471974e-06, + "loss": 1.8229, + "step": 1698 + }, + { + "epoch": 1.861917808219178, + "grad_norm": 2.341775894165039, + "learning_rate": 2.1385052802599515e-06, + "loss": 1.7421, + "step": 1699 + }, + { + "epoch": 1.8630136986301369, + "grad_norm": 2.343876838684082, + "learning_rate": 2.1364744110479288e-06, + "loss": 1.7519, + "step": 1700 + }, + { + "epoch": 1.8630136986301369, + "eval_loss": 1.9587067365646362, + "eval_runtime": 37.4192, + "eval_samples_per_second": 26.724, + "eval_steps_per_second": 1.122, + "step": 1700 + }, + { + "epoch": 1.864109589041096, + "grad_norm": 2.388373613357544, + "learning_rate": 2.134443541835906e-06, + "loss": 1.8039, + "step": 1701 + }, + { + "epoch": 1.865205479452055, + "grad_norm": 2.379742383956909, + "learning_rate": 2.132412672623883e-06, + "loss": 1.7549, + "step": 1702 + }, + { + "epoch": 1.8663013698630135, + "grad_norm": 2.25833797454834, + "learning_rate": 2.13038180341186e-06, + "loss": 1.7349, + "step": 1703 + }, + { + "epoch": 1.8673972602739726, + "grad_norm": 2.4541401863098145, + "learning_rate": 2.1283509341998374e-06, + "loss": 1.7785, + "step": 1704 + }, + { + "epoch": 1.8684931506849316, + "grad_norm": 2.338456392288208, + "learning_rate": 2.1263200649878147e-06, + "loss": 1.8272, + "step": 1705 + }, + { + "epoch": 1.8695890410958904, + "grad_norm": 2.474292755126953, + "learning_rate": 2.124289195775792e-06, + "loss": 1.7497, + "step": 1706 + }, + { + "epoch": 1.8706849315068492, + "grad_norm": 2.206749439239502, + "learning_rate": 2.1222583265637693e-06, + "loss": 1.7697, + "step": 1707 + }, + { + "epoch": 1.8717808219178083, + "grad_norm": 2.378093957901001, + "learning_rate": 2.1202274573517465e-06, + "loss": 1.6851, + "step": 1708 + }, + { + "epoch": 1.872876712328767, + "grad_norm": 2.2442147731781006, + "learning_rate": 2.118196588139724e-06, + "loss": 1.7268, + "step": 1709 + }, + { + "epoch": 1.8739726027397259, + "grad_norm": 2.324852228164673, + "learning_rate": 2.116165718927701e-06, + "loss": 1.7369, + "step": 1710 + }, + { + "epoch": 1.8739726027397259, + "eval_loss": 1.9555498361587524, + "eval_runtime": 37.2599, + "eval_samples_per_second": 26.838, + "eval_steps_per_second": 1.127, + "step": 1710 + }, + { + "epoch": 1.875068493150685, + "grad_norm": 2.3466620445251465, + "learning_rate": 2.1141348497156784e-06, + "loss": 1.7642, + "step": 1711 + }, + { + "epoch": 1.876164383561644, + "grad_norm": 2.303654432296753, + "learning_rate": 2.1121039805036556e-06, + "loss": 1.7406, + "step": 1712 + }, + { + "epoch": 1.8772602739726028, + "grad_norm": 2.242719888687134, + "learning_rate": 2.110073111291633e-06, + "loss": 1.7269, + "step": 1713 + }, + { + "epoch": 1.8783561643835616, + "grad_norm": 2.4076626300811768, + "learning_rate": 2.10804224207961e-06, + "loss": 1.7749, + "step": 1714 + }, + { + "epoch": 1.8794520547945206, + "grad_norm": 2.510887861251831, + "learning_rate": 2.1060113728675875e-06, + "loss": 1.8056, + "step": 1715 + }, + { + "epoch": 1.8805479452054794, + "grad_norm": 2.4038093090057373, + "learning_rate": 2.1039805036555647e-06, + "loss": 1.7915, + "step": 1716 + }, + { + "epoch": 1.8816438356164382, + "grad_norm": 2.4485573768615723, + "learning_rate": 2.101949634443542e-06, + "loss": 1.6845, + "step": 1717 + }, + { + "epoch": 1.8827397260273973, + "grad_norm": 2.4619247913360596, + "learning_rate": 2.0999187652315193e-06, + "loss": 1.7597, + "step": 1718 + }, + { + "epoch": 1.8838356164383563, + "grad_norm": 2.5079665184020996, + "learning_rate": 2.0978878960194966e-06, + "loss": 1.7207, + "step": 1719 + }, + { + "epoch": 1.884931506849315, + "grad_norm": 2.38696551322937, + "learning_rate": 2.095857026807474e-06, + "loss": 1.7855, + "step": 1720 + }, + { + "epoch": 1.884931506849315, + "eval_loss": 1.9538354873657227, + "eval_runtime": 37.5269, + "eval_samples_per_second": 26.648, + "eval_steps_per_second": 1.119, + "step": 1720 + }, + { + "epoch": 1.886027397260274, + "grad_norm": 2.2652602195739746, + "learning_rate": 2.093826157595451e-06, + "loss": 1.7806, + "step": 1721 + }, + { + "epoch": 1.887123287671233, + "grad_norm": 2.4074227809906006, + "learning_rate": 2.0917952883834284e-06, + "loss": 1.7397, + "step": 1722 + }, + { + "epoch": 1.8882191780821918, + "grad_norm": 2.4302561283111572, + "learning_rate": 2.0897644191714057e-06, + "loss": 1.7883, + "step": 1723 + }, + { + "epoch": 1.8893150684931506, + "grad_norm": 2.311933755874634, + "learning_rate": 2.087733549959383e-06, + "loss": 1.8186, + "step": 1724 + }, + { + "epoch": 1.8904109589041096, + "grad_norm": 2.4028313159942627, + "learning_rate": 2.0857026807473602e-06, + "loss": 1.7208, + "step": 1725 + }, + { + "epoch": 1.8915068493150686, + "grad_norm": 2.2129130363464355, + "learning_rate": 2.0836718115353375e-06, + "loss": 1.7007, + "step": 1726 + }, + { + "epoch": 1.8926027397260274, + "grad_norm": 2.3893744945526123, + "learning_rate": 2.0816409423233144e-06, + "loss": 1.8281, + "step": 1727 + }, + { + "epoch": 1.8936986301369862, + "grad_norm": 2.2396240234375, + "learning_rate": 2.0796100731112916e-06, + "loss": 1.7252, + "step": 1728 + }, + { + "epoch": 1.8947945205479453, + "grad_norm": 2.4073901176452637, + "learning_rate": 2.077579203899269e-06, + "loss": 1.7533, + "step": 1729 + }, + { + "epoch": 1.895890410958904, + "grad_norm": 2.3669373989105225, + "learning_rate": 2.075548334687246e-06, + "loss": 1.7013, + "step": 1730 + }, + { + "epoch": 1.895890410958904, + "eval_loss": 1.95528244972229, + "eval_runtime": 37.6174, + "eval_samples_per_second": 26.583, + "eval_steps_per_second": 1.117, + "step": 1730 + }, + { + "epoch": 1.896986301369863, + "grad_norm": 2.5254933834075928, + "learning_rate": 2.0735174654752235e-06, + "loss": 1.7848, + "step": 1731 + }, + { + "epoch": 1.898082191780822, + "grad_norm": 2.4138967990875244, + "learning_rate": 2.0714865962632007e-06, + "loss": 1.7204, + "step": 1732 + }, + { + "epoch": 1.899178082191781, + "grad_norm": 2.417165994644165, + "learning_rate": 2.069455727051178e-06, + "loss": 1.7741, + "step": 1733 + }, + { + "epoch": 1.9002739726027398, + "grad_norm": 2.4026505947113037, + "learning_rate": 2.0674248578391553e-06, + "loss": 1.7456, + "step": 1734 + }, + { + "epoch": 1.9013698630136986, + "grad_norm": 2.4637675285339355, + "learning_rate": 2.0653939886271326e-06, + "loss": 1.7679, + "step": 1735 + }, + { + "epoch": 1.9024657534246576, + "grad_norm": 2.269869089126587, + "learning_rate": 2.06336311941511e-06, + "loss": 1.8059, + "step": 1736 + }, + { + "epoch": 1.9035616438356164, + "grad_norm": 2.321094512939453, + "learning_rate": 2.061332250203087e-06, + "loss": 1.743, + "step": 1737 + }, + { + "epoch": 1.9046575342465752, + "grad_norm": 2.477288246154785, + "learning_rate": 2.0593013809910644e-06, + "loss": 1.7143, + "step": 1738 + }, + { + "epoch": 1.9057534246575343, + "grad_norm": 2.3924875259399414, + "learning_rate": 2.0572705117790417e-06, + "loss": 1.75, + "step": 1739 + }, + { + "epoch": 1.9068493150684933, + "grad_norm": 2.4452881813049316, + "learning_rate": 2.055239642567019e-06, + "loss": 1.7831, + "step": 1740 + }, + { + "epoch": 1.9068493150684933, + "eval_loss": 1.9599454402923584, + "eval_runtime": 37.3026, + "eval_samples_per_second": 26.808, + "eval_steps_per_second": 1.126, + "step": 1740 + }, + { + "epoch": 1.907945205479452, + "grad_norm": 2.2778351306915283, + "learning_rate": 2.0532087733549962e-06, + "loss": 1.7633, + "step": 1741 + }, + { + "epoch": 1.909041095890411, + "grad_norm": 2.3171589374542236, + "learning_rate": 2.0511779041429735e-06, + "loss": 1.7726, + "step": 1742 + }, + { + "epoch": 1.91013698630137, + "grad_norm": 2.286094903945923, + "learning_rate": 2.0491470349309508e-06, + "loss": 1.7458, + "step": 1743 + }, + { + "epoch": 1.9112328767123288, + "grad_norm": 2.5381743907928467, + "learning_rate": 2.047116165718928e-06, + "loss": 1.777, + "step": 1744 + }, + { + "epoch": 1.9123287671232876, + "grad_norm": 2.3261141777038574, + "learning_rate": 2.0450852965069053e-06, + "loss": 1.7715, + "step": 1745 + }, + { + "epoch": 1.9134246575342466, + "grad_norm": 2.351914167404175, + "learning_rate": 2.0430544272948826e-06, + "loss": 1.8214, + "step": 1746 + }, + { + "epoch": 1.9145205479452054, + "grad_norm": 2.274226427078247, + "learning_rate": 2.04102355808286e-06, + "loss": 1.763, + "step": 1747 + }, + { + "epoch": 1.9156164383561642, + "grad_norm": 2.4025075435638428, + "learning_rate": 2.038992688870837e-06, + "loss": 1.7072, + "step": 1748 + }, + { + "epoch": 1.9167123287671233, + "grad_norm": 2.334024429321289, + "learning_rate": 2.0369618196588144e-06, + "loss": 1.7381, + "step": 1749 + }, + { + "epoch": 1.9178082191780823, + "grad_norm": 2.3011622428894043, + "learning_rate": 2.0349309504467917e-06, + "loss": 1.7432, + "step": 1750 + }, + { + "epoch": 1.9178082191780823, + "eval_loss": 1.958349347114563, + "eval_runtime": 37.2591, + "eval_samples_per_second": 26.839, + "eval_steps_per_second": 1.127, + "step": 1750 + }, + { + "epoch": 1.9189041095890411, + "grad_norm": 2.643063545227051, + "learning_rate": 2.032900081234769e-06, + "loss": 1.7931, + "step": 1751 + }, + { + "epoch": 1.92, + "grad_norm": 2.3357834815979004, + "learning_rate": 2.030869212022746e-06, + "loss": 1.7933, + "step": 1752 + }, + { + "epoch": 1.921095890410959, + "grad_norm": 2.2614479064941406, + "learning_rate": 2.028838342810723e-06, + "loss": 1.7611, + "step": 1753 + }, + { + "epoch": 1.9221917808219178, + "grad_norm": 2.3057546615600586, + "learning_rate": 2.0268074735987004e-06, + "loss": 1.7406, + "step": 1754 + }, + { + "epoch": 1.9232876712328766, + "grad_norm": 2.6138038635253906, + "learning_rate": 2.0247766043866777e-06, + "loss": 1.7303, + "step": 1755 + }, + { + "epoch": 1.9243835616438356, + "grad_norm": 2.3803296089172363, + "learning_rate": 2.022745735174655e-06, + "loss": 1.8023, + "step": 1756 + }, + { + "epoch": 1.9254794520547946, + "grad_norm": 2.3436431884765625, + "learning_rate": 2.020714865962632e-06, + "loss": 1.7654, + "step": 1757 + }, + { + "epoch": 1.9265753424657535, + "grad_norm": 2.4103317260742188, + "learning_rate": 2.0186839967506095e-06, + "loss": 1.7932, + "step": 1758 + }, + { + "epoch": 1.9276712328767123, + "grad_norm": 2.4537675380706787, + "learning_rate": 2.0166531275385868e-06, + "loss": 1.761, + "step": 1759 + }, + { + "epoch": 1.9287671232876713, + "grad_norm": 2.3816988468170166, + "learning_rate": 2.014622258326564e-06, + "loss": 1.7276, + "step": 1760 + }, + { + "epoch": 1.9287671232876713, + "eval_loss": 1.9558053016662598, + "eval_runtime": 37.2417, + "eval_samples_per_second": 26.852, + "eval_steps_per_second": 1.128, + "step": 1760 + }, + { + "epoch": 1.9298630136986301, + "grad_norm": 2.37233304977417, + "learning_rate": 2.0125913891145413e-06, + "loss": 1.7674, + "step": 1761 + }, + { + "epoch": 1.930958904109589, + "grad_norm": 2.4043285846710205, + "learning_rate": 2.0105605199025186e-06, + "loss": 1.7638, + "step": 1762 + }, + { + "epoch": 1.932054794520548, + "grad_norm": 2.3402836322784424, + "learning_rate": 2.008529650690496e-06, + "loss": 1.7269, + "step": 1763 + }, + { + "epoch": 1.933150684931507, + "grad_norm": 2.42535400390625, + "learning_rate": 2.0064987814784727e-06, + "loss": 1.7965, + "step": 1764 + }, + { + "epoch": 1.9342465753424658, + "grad_norm": 2.361659288406372, + "learning_rate": 2.00446791226645e-06, + "loss": 1.7467, + "step": 1765 + }, + { + "epoch": 1.9353424657534246, + "grad_norm": 2.455253839492798, + "learning_rate": 2.0024370430544273e-06, + "loss": 1.7359, + "step": 1766 + }, + { + "epoch": 1.9364383561643836, + "grad_norm": 2.3062705993652344, + "learning_rate": 2.0004061738424045e-06, + "loss": 1.751, + "step": 1767 + }, + { + "epoch": 1.9375342465753425, + "grad_norm": 2.3683881759643555, + "learning_rate": 1.998375304630382e-06, + "loss": 1.8032, + "step": 1768 + }, + { + "epoch": 1.9386301369863013, + "grad_norm": 2.2175045013427734, + "learning_rate": 1.996344435418359e-06, + "loss": 1.7629, + "step": 1769 + }, + { + "epoch": 1.9397260273972603, + "grad_norm": 2.4331724643707275, + "learning_rate": 1.9943135662063364e-06, + "loss": 1.7899, + "step": 1770 + }, + { + "epoch": 1.9397260273972603, + "eval_loss": 1.9554719924926758, + "eval_runtime": 37.2636, + "eval_samples_per_second": 26.836, + "eval_steps_per_second": 1.127, + "step": 1770 + }, + { + "epoch": 1.9408219178082193, + "grad_norm": 2.370785713195801, + "learning_rate": 1.9922826969943136e-06, + "loss": 1.7687, + "step": 1771 + }, + { + "epoch": 1.941917808219178, + "grad_norm": 2.4047751426696777, + "learning_rate": 1.990251827782291e-06, + "loss": 1.804, + "step": 1772 + }, + { + "epoch": 1.943013698630137, + "grad_norm": 2.4597785472869873, + "learning_rate": 1.988220958570268e-06, + "loss": 1.7731, + "step": 1773 + }, + { + "epoch": 1.944109589041096, + "grad_norm": 2.388749837875366, + "learning_rate": 1.9861900893582455e-06, + "loss": 1.7945, + "step": 1774 + }, + { + "epoch": 1.9452054794520548, + "grad_norm": 2.3419551849365234, + "learning_rate": 1.9841592201462227e-06, + "loss": 1.7987, + "step": 1775 + }, + { + "epoch": 1.9463013698630136, + "grad_norm": 2.43770432472229, + "learning_rate": 1.9821283509342e-06, + "loss": 1.7183, + "step": 1776 + }, + { + "epoch": 1.9473972602739726, + "grad_norm": 2.333385467529297, + "learning_rate": 1.9800974817221773e-06, + "loss": 1.7283, + "step": 1777 + }, + { + "epoch": 1.9484931506849315, + "grad_norm": 2.391052722930908, + "learning_rate": 1.9780666125101546e-06, + "loss": 1.776, + "step": 1778 + }, + { + "epoch": 1.9495890410958903, + "grad_norm": 2.375338077545166, + "learning_rate": 1.976035743298132e-06, + "loss": 1.7495, + "step": 1779 + }, + { + "epoch": 1.9506849315068493, + "grad_norm": 2.4482579231262207, + "learning_rate": 1.974004874086109e-06, + "loss": 1.8256, + "step": 1780 + }, + { + "epoch": 1.9506849315068493, + "eval_loss": 1.9600462913513184, + "eval_runtime": 37.2638, + "eval_samples_per_second": 26.836, + "eval_steps_per_second": 1.127, + "step": 1780 + }, + { + "epoch": 1.9517808219178083, + "grad_norm": 2.296039342880249, + "learning_rate": 1.9719740048740864e-06, + "loss": 1.6996, + "step": 1781 + }, + { + "epoch": 1.9528767123287671, + "grad_norm": 2.2494969367980957, + "learning_rate": 1.9699431356620637e-06, + "loss": 1.7401, + "step": 1782 + }, + { + "epoch": 1.953972602739726, + "grad_norm": 2.4078574180603027, + "learning_rate": 1.967912266450041e-06, + "loss": 1.7374, + "step": 1783 + }, + { + "epoch": 1.955068493150685, + "grad_norm": 2.4018476009368896, + "learning_rate": 1.9658813972380182e-06, + "loss": 1.731, + "step": 1784 + }, + { + "epoch": 1.9561643835616438, + "grad_norm": 2.3429362773895264, + "learning_rate": 1.9638505280259955e-06, + "loss": 1.8531, + "step": 1785 + }, + { + "epoch": 1.9572602739726026, + "grad_norm": 2.4287197589874268, + "learning_rate": 1.9618196588139728e-06, + "loss": 1.7271, + "step": 1786 + }, + { + "epoch": 1.9583561643835616, + "grad_norm": 2.338541030883789, + "learning_rate": 1.95978878960195e-06, + "loss": 1.7594, + "step": 1787 + }, + { + "epoch": 1.9594520547945207, + "grad_norm": 2.531425952911377, + "learning_rate": 1.9577579203899273e-06, + "loss": 1.8374, + "step": 1788 + }, + { + "epoch": 1.9605479452054795, + "grad_norm": 2.413638114929199, + "learning_rate": 1.955727051177904e-06, + "loss": 1.8178, + "step": 1789 + }, + { + "epoch": 1.9616438356164383, + "grad_norm": 2.303318977355957, + "learning_rate": 1.9536961819658815e-06, + "loss": 1.7429, + "step": 1790 + }, + { + "epoch": 1.9616438356164383, + "eval_loss": 1.9622509479522705, + "eval_runtime": 37.2699, + "eval_samples_per_second": 26.831, + "eval_steps_per_second": 1.127, + "step": 1790 + }, + { + "epoch": 1.9627397260273973, + "grad_norm": 2.400160074234009, + "learning_rate": 1.9516653127538587e-06, + "loss": 1.7136, + "step": 1791 + }, + { + "epoch": 1.9638356164383561, + "grad_norm": 2.3861589431762695, + "learning_rate": 1.949634443541836e-06, + "loss": 1.7146, + "step": 1792 + }, + { + "epoch": 1.964931506849315, + "grad_norm": 2.272711753845215, + "learning_rate": 1.9476035743298133e-06, + "loss": 1.7177, + "step": 1793 + }, + { + "epoch": 1.966027397260274, + "grad_norm": 2.4745068550109863, + "learning_rate": 1.9455727051177906e-06, + "loss": 1.7938, + "step": 1794 + }, + { + "epoch": 1.967123287671233, + "grad_norm": 2.4643466472625732, + "learning_rate": 1.943541835905768e-06, + "loss": 1.7601, + "step": 1795 + }, + { + "epoch": 1.9682191780821918, + "grad_norm": 2.5346457958221436, + "learning_rate": 1.941510966693745e-06, + "loss": 1.8132, + "step": 1796 + }, + { + "epoch": 1.9693150684931506, + "grad_norm": 2.280719041824341, + "learning_rate": 1.9394800974817224e-06, + "loss": 1.7101, + "step": 1797 + }, + { + "epoch": 1.9704109589041097, + "grad_norm": 2.3304555416107178, + "learning_rate": 1.9374492282696997e-06, + "loss": 1.7809, + "step": 1798 + }, + { + "epoch": 1.9715068493150685, + "grad_norm": 2.409796714782715, + "learning_rate": 1.935418359057677e-06, + "loss": 1.8127, + "step": 1799 + }, + { + "epoch": 1.9726027397260273, + "grad_norm": 2.208312749862671, + "learning_rate": 1.9333874898456542e-06, + "loss": 1.7752, + "step": 1800 + }, + { + "epoch": 1.9726027397260273, + "eval_loss": 1.96150541305542, + "eval_runtime": 37.3221, + "eval_samples_per_second": 26.794, + "eval_steps_per_second": 1.125, + "step": 1800 + }, + { + "epoch": 1.9736986301369863, + "grad_norm": 2.374393939971924, + "learning_rate": 1.931356620633631e-06, + "loss": 1.8188, + "step": 1801 + }, + { + "epoch": 1.9747945205479454, + "grad_norm": 2.3343780040740967, + "learning_rate": 1.9293257514216083e-06, + "loss": 1.7035, + "step": 1802 + }, + { + "epoch": 1.9758904109589042, + "grad_norm": 2.307894229888916, + "learning_rate": 1.9272948822095856e-06, + "loss": 1.7651, + "step": 1803 + }, + { + "epoch": 1.976986301369863, + "grad_norm": 2.387434244155884, + "learning_rate": 1.925264012997563e-06, + "loss": 1.7021, + "step": 1804 + }, + { + "epoch": 1.978082191780822, + "grad_norm": 2.3444128036499023, + "learning_rate": 1.92323314378554e-06, + "loss": 1.7585, + "step": 1805 + }, + { + "epoch": 1.9791780821917808, + "grad_norm": 2.371931791305542, + "learning_rate": 1.9212022745735174e-06, + "loss": 1.7932, + "step": 1806 + }, + { + "epoch": 1.9802739726027396, + "grad_norm": 2.3159632682800293, + "learning_rate": 1.9191714053614947e-06, + "loss": 1.722, + "step": 1807 + }, + { + "epoch": 1.9813698630136987, + "grad_norm": 2.4669857025146484, + "learning_rate": 1.917140536149472e-06, + "loss": 1.7326, + "step": 1808 + }, + { + "epoch": 1.9824657534246577, + "grad_norm": 2.365138292312622, + "learning_rate": 1.9151096669374493e-06, + "loss": 1.7013, + "step": 1809 + }, + { + "epoch": 1.9835616438356163, + "grad_norm": 2.539698362350464, + "learning_rate": 1.9130787977254266e-06, + "loss": 1.8082, + "step": 1810 + }, + { + "epoch": 1.9835616438356163, + "eval_loss": 1.9535884857177734, + "eval_runtime": 37.3497, + "eval_samples_per_second": 26.774, + "eval_steps_per_second": 1.125, + "step": 1810 + }, + { + "epoch": 1.9846575342465753, + "grad_norm": 2.386599540710449, + "learning_rate": 1.911047928513404e-06, + "loss": 1.7835, + "step": 1811 + }, + { + "epoch": 1.9857534246575343, + "grad_norm": 2.376086711883545, + "learning_rate": 1.909017059301381e-06, + "loss": 1.7161, + "step": 1812 + }, + { + "epoch": 1.9868493150684932, + "grad_norm": 2.5276663303375244, + "learning_rate": 1.9069861900893586e-06, + "loss": 1.7222, + "step": 1813 + }, + { + "epoch": 1.987945205479452, + "grad_norm": 2.466137170791626, + "learning_rate": 1.9049553208773354e-06, + "loss": 1.7841, + "step": 1814 + }, + { + "epoch": 1.989041095890411, + "grad_norm": 2.361128091812134, + "learning_rate": 1.9029244516653127e-06, + "loss": 1.7354, + "step": 1815 + }, + { + "epoch": 1.9901369863013698, + "grad_norm": 2.4348530769348145, + "learning_rate": 1.90089358245329e-06, + "loss": 1.6832, + "step": 1816 + }, + { + "epoch": 1.9912328767123286, + "grad_norm": 2.342285633087158, + "learning_rate": 1.8988627132412673e-06, + "loss": 1.7351, + "step": 1817 + }, + { + "epoch": 1.9923287671232877, + "grad_norm": 2.521137237548828, + "learning_rate": 1.8968318440292445e-06, + "loss": 1.794, + "step": 1818 + }, + { + "epoch": 1.9934246575342467, + "grad_norm": 2.5968360900878906, + "learning_rate": 1.8948009748172218e-06, + "loss": 1.6976, + "step": 1819 + }, + { + "epoch": 1.9945205479452055, + "grad_norm": 2.734095335006714, + "learning_rate": 1.892770105605199e-06, + "loss": 1.7919, + "step": 1820 + }, + { + "epoch": 1.9945205479452055, + "eval_loss": 1.9536023139953613, + "eval_runtime": 37.5805, + "eval_samples_per_second": 26.61, + "eval_steps_per_second": 1.118, + "step": 1820 + }, + { + "epoch": 1.9956164383561643, + "grad_norm": 2.327333927154541, + "learning_rate": 1.8907392363931764e-06, + "loss": 1.7891, + "step": 1821 + }, + { + "epoch": 1.9967123287671233, + "grad_norm": 2.2902543544769287, + "learning_rate": 1.8887083671811536e-06, + "loss": 1.7552, + "step": 1822 + }, + { + "epoch": 1.9978082191780822, + "grad_norm": 2.3587660789489746, + "learning_rate": 1.886677497969131e-06, + "loss": 1.7641, + "step": 1823 + }, + { + "epoch": 1.998904109589041, + "grad_norm": 2.4068408012390137, + "learning_rate": 1.8846466287571082e-06, + "loss": 1.6622, + "step": 1824 + }, + { + "epoch": 2.0, + "grad_norm": 2.453514575958252, + "learning_rate": 1.8826157595450855e-06, + "loss": 1.7962, + "step": 1825 + }, + { + "epoch": 2.001095890410959, + "grad_norm": 2.4694175720214844, + "learning_rate": 1.8805848903330625e-06, + "loss": 1.6823, + "step": 1826 + }, + { + "epoch": 2.0021917808219176, + "grad_norm": 2.594013214111328, + "learning_rate": 1.8785540211210398e-06, + "loss": 1.8453, + "step": 1827 + }, + { + "epoch": 2.0032876712328767, + "grad_norm": 2.455796718597412, + "learning_rate": 1.876523151909017e-06, + "loss": 1.7604, + "step": 1828 + }, + { + "epoch": 2.0043835616438357, + "grad_norm": 2.819857120513916, + "learning_rate": 1.8744922826969944e-06, + "loss": 1.7177, + "step": 1829 + }, + { + "epoch": 2.0054794520547947, + "grad_norm": 2.845775842666626, + "learning_rate": 1.8724614134849716e-06, + "loss": 1.7007, + "step": 1830 + }, + { + "epoch": 2.0054794520547947, + "eval_loss": 1.9594684839248657, + "eval_runtime": 37.6856, + "eval_samples_per_second": 26.535, + "eval_steps_per_second": 1.114, + "step": 1830 + }, + { + "epoch": 2.0065753424657533, + "grad_norm": 2.6129424571990967, + "learning_rate": 1.870430544272949e-06, + "loss": 1.7301, + "step": 1831 + }, + { + "epoch": 2.0076712328767123, + "grad_norm": 2.762805223464966, + "learning_rate": 1.8683996750609262e-06, + "loss": 1.8294, + "step": 1832 + }, + { + "epoch": 2.0087671232876714, + "grad_norm": 2.6397714614868164, + "learning_rate": 1.8663688058489035e-06, + "loss": 1.7176, + "step": 1833 + }, + { + "epoch": 2.00986301369863, + "grad_norm": 2.6637110710144043, + "learning_rate": 1.8643379366368807e-06, + "loss": 1.7363, + "step": 1834 + }, + { + "epoch": 2.010958904109589, + "grad_norm": 2.732534408569336, + "learning_rate": 1.862307067424858e-06, + "loss": 1.7496, + "step": 1835 + }, + { + "epoch": 2.012054794520548, + "grad_norm": 2.7163467407226562, + "learning_rate": 1.8602761982128353e-06, + "loss": 1.7551, + "step": 1836 + }, + { + "epoch": 2.013150684931507, + "grad_norm": 2.442484140396118, + "learning_rate": 1.8582453290008126e-06, + "loss": 1.7761, + "step": 1837 + }, + { + "epoch": 2.0142465753424657, + "grad_norm": 2.544834852218628, + "learning_rate": 1.8562144597887898e-06, + "loss": 1.7251, + "step": 1838 + }, + { + "epoch": 2.0153424657534247, + "grad_norm": 2.6588494777679443, + "learning_rate": 1.854183590576767e-06, + "loss": 1.7381, + "step": 1839 + }, + { + "epoch": 2.0164383561643837, + "grad_norm": 2.5374786853790283, + "learning_rate": 1.8521527213647442e-06, + "loss": 1.7831, + "step": 1840 + }, + { + "epoch": 2.0164383561643837, + "eval_loss": 1.9550480842590332, + "eval_runtime": 37.3312, + "eval_samples_per_second": 26.787, + "eval_steps_per_second": 1.125, + "step": 1840 + }, + { + "epoch": 2.0175342465753423, + "grad_norm": 2.4791345596313477, + "learning_rate": 1.8501218521527215e-06, + "loss": 1.6937, + "step": 1841 + }, + { + "epoch": 2.0186301369863013, + "grad_norm": 2.7243971824645996, + "learning_rate": 1.8480909829406987e-06, + "loss": 1.6674, + "step": 1842 + }, + { + "epoch": 2.0197260273972604, + "grad_norm": 2.618680477142334, + "learning_rate": 1.846060113728676e-06, + "loss": 1.6994, + "step": 1843 + }, + { + "epoch": 2.020821917808219, + "grad_norm": 2.597929000854492, + "learning_rate": 1.8440292445166533e-06, + "loss": 1.7375, + "step": 1844 + }, + { + "epoch": 2.021917808219178, + "grad_norm": 2.860177993774414, + "learning_rate": 1.8419983753046306e-06, + "loss": 1.7128, + "step": 1845 + }, + { + "epoch": 2.023013698630137, + "grad_norm": 2.606898307800293, + "learning_rate": 1.8399675060926078e-06, + "loss": 1.7735, + "step": 1846 + }, + { + "epoch": 2.024109589041096, + "grad_norm": 2.5152604579925537, + "learning_rate": 1.8379366368805851e-06, + "loss": 1.75, + "step": 1847 + }, + { + "epoch": 2.0252054794520546, + "grad_norm": 2.678071975708008, + "learning_rate": 1.8359057676685624e-06, + "loss": 1.7114, + "step": 1848 + }, + { + "epoch": 2.0263013698630137, + "grad_norm": 2.673515558242798, + "learning_rate": 1.8338748984565397e-06, + "loss": 1.6992, + "step": 1849 + }, + { + "epoch": 2.0273972602739727, + "grad_norm": 2.5902516841888428, + "learning_rate": 1.831844029244517e-06, + "loss": 1.7195, + "step": 1850 + }, + { + "epoch": 2.0273972602739727, + "eval_loss": 1.953579068183899, + "eval_runtime": 37.304, + "eval_samples_per_second": 26.807, + "eval_steps_per_second": 1.126, + "step": 1850 + }, + { + "epoch": 2.0284931506849313, + "grad_norm": 2.4912915229797363, + "learning_rate": 1.829813160032494e-06, + "loss": 1.6955, + "step": 1851 + }, + { + "epoch": 2.0295890410958903, + "grad_norm": 2.683821678161621, + "learning_rate": 1.8277822908204713e-06, + "loss": 1.7057, + "step": 1852 + }, + { + "epoch": 2.0306849315068494, + "grad_norm": 2.9541101455688477, + "learning_rate": 1.8257514216084486e-06, + "loss": 1.7377, + "step": 1853 + }, + { + "epoch": 2.0317808219178084, + "grad_norm": 2.783851146697998, + "learning_rate": 1.8237205523964258e-06, + "loss": 1.7197, + "step": 1854 + }, + { + "epoch": 2.032876712328767, + "grad_norm": 2.5224950313568115, + "learning_rate": 1.8216896831844031e-06, + "loss": 1.7396, + "step": 1855 + }, + { + "epoch": 2.033972602739726, + "grad_norm": 2.627767562866211, + "learning_rate": 1.8196588139723804e-06, + "loss": 1.7773, + "step": 1856 + }, + { + "epoch": 2.035068493150685, + "grad_norm": 2.443873405456543, + "learning_rate": 1.8176279447603577e-06, + "loss": 1.6656, + "step": 1857 + }, + { + "epoch": 2.0361643835616436, + "grad_norm": 2.648056745529175, + "learning_rate": 1.815597075548335e-06, + "loss": 1.6954, + "step": 1858 + }, + { + "epoch": 2.0372602739726027, + "grad_norm": 2.536935329437256, + "learning_rate": 1.8135662063363122e-06, + "loss": 1.7427, + "step": 1859 + }, + { + "epoch": 2.0383561643835617, + "grad_norm": 2.42183256149292, + "learning_rate": 1.8115353371242895e-06, + "loss": 1.7634, + "step": 1860 + }, + { + "epoch": 2.0383561643835617, + "eval_loss": 1.953397512435913, + "eval_runtime": 37.2886, + "eval_samples_per_second": 26.818, + "eval_steps_per_second": 1.126, + "step": 1860 + }, + { + "epoch": 2.0394520547945207, + "grad_norm": 2.538804769515991, + "learning_rate": 1.8095044679122668e-06, + "loss": 1.6712, + "step": 1861 + }, + { + "epoch": 2.0405479452054793, + "grad_norm": 2.5820095539093018, + "learning_rate": 1.807473598700244e-06, + "loss": 1.7215, + "step": 1862 + }, + { + "epoch": 2.0416438356164384, + "grad_norm": 2.6997382640838623, + "learning_rate": 1.805442729488221e-06, + "loss": 1.7117, + "step": 1863 + }, + { + "epoch": 2.0427397260273974, + "grad_norm": 2.775996685028076, + "learning_rate": 1.8034118602761982e-06, + "loss": 1.7355, + "step": 1864 + }, + { + "epoch": 2.043835616438356, + "grad_norm": 2.6333167552948, + "learning_rate": 1.8013809910641754e-06, + "loss": 1.7343, + "step": 1865 + }, + { + "epoch": 2.044931506849315, + "grad_norm": 2.5830278396606445, + "learning_rate": 1.7993501218521527e-06, + "loss": 1.7339, + "step": 1866 + }, + { + "epoch": 2.046027397260274, + "grad_norm": 2.6145055294036865, + "learning_rate": 1.79731925264013e-06, + "loss": 1.6835, + "step": 1867 + }, + { + "epoch": 2.047123287671233, + "grad_norm": 2.568159580230713, + "learning_rate": 1.7952883834281073e-06, + "loss": 1.7574, + "step": 1868 + }, + { + "epoch": 2.0482191780821917, + "grad_norm": 2.5235254764556885, + "learning_rate": 1.7932575142160846e-06, + "loss": 1.7067, + "step": 1869 + }, + { + "epoch": 2.0493150684931507, + "grad_norm": 2.464035987854004, + "learning_rate": 1.7912266450040618e-06, + "loss": 1.7451, + "step": 1870 + }, + { + "epoch": 2.0493150684931507, + "eval_loss": 1.9485031366348267, + "eval_runtime": 37.2648, + "eval_samples_per_second": 26.835, + "eval_steps_per_second": 1.127, + "step": 1870 + }, + { + "epoch": 2.0504109589041097, + "grad_norm": 2.4328689575195312, + "learning_rate": 1.789195775792039e-06, + "loss": 1.7508, + "step": 1871 + }, + { + "epoch": 2.0515068493150683, + "grad_norm": 2.5483932495117188, + "learning_rate": 1.7871649065800164e-06, + "loss": 1.6488, + "step": 1872 + }, + { + "epoch": 2.0526027397260274, + "grad_norm": 2.5290846824645996, + "learning_rate": 1.7851340373679937e-06, + "loss": 1.7939, + "step": 1873 + }, + { + "epoch": 2.0536986301369864, + "grad_norm": 2.506943702697754, + "learning_rate": 1.783103168155971e-06, + "loss": 1.6813, + "step": 1874 + }, + { + "epoch": 2.0547945205479454, + "grad_norm": 2.5863115787506104, + "learning_rate": 1.7810722989439482e-06, + "loss": 1.7104, + "step": 1875 + }, + { + "epoch": 2.055890410958904, + "grad_norm": 2.6402666568756104, + "learning_rate": 1.7790414297319253e-06, + "loss": 1.7311, + "step": 1876 + }, + { + "epoch": 2.056986301369863, + "grad_norm": 2.6700708866119385, + "learning_rate": 1.7770105605199025e-06, + "loss": 1.695, + "step": 1877 + }, + { + "epoch": 2.058082191780822, + "grad_norm": 2.580693244934082, + "learning_rate": 1.7749796913078798e-06, + "loss": 1.6763, + "step": 1878 + }, + { + "epoch": 2.0591780821917807, + "grad_norm": 2.6035709381103516, + "learning_rate": 1.772948822095857e-06, + "loss": 1.7389, + "step": 1879 + }, + { + "epoch": 2.0602739726027397, + "grad_norm": 2.508784532546997, + "learning_rate": 1.7709179528838344e-06, + "loss": 1.7762, + "step": 1880 + }, + { + "epoch": 2.0602739726027397, + "eval_loss": 1.9516879320144653, + "eval_runtime": 37.2284, + "eval_samples_per_second": 26.861, + "eval_steps_per_second": 1.128, + "step": 1880 + }, + { + "epoch": 2.0613698630136987, + "grad_norm": 2.47988224029541, + "learning_rate": 1.7688870836718116e-06, + "loss": 1.7659, + "step": 1881 + }, + { + "epoch": 2.0624657534246573, + "grad_norm": 2.5782723426818848, + "learning_rate": 1.766856214459789e-06, + "loss": 1.7179, + "step": 1882 + }, + { + "epoch": 2.0635616438356164, + "grad_norm": 2.5616347789764404, + "learning_rate": 1.7648253452477662e-06, + "loss": 1.7163, + "step": 1883 + }, + { + "epoch": 2.0646575342465754, + "grad_norm": 2.4464340209960938, + "learning_rate": 1.7627944760357435e-06, + "loss": 1.7033, + "step": 1884 + }, + { + "epoch": 2.0657534246575344, + "grad_norm": 2.6022887229919434, + "learning_rate": 1.7607636068237208e-06, + "loss": 1.7823, + "step": 1885 + }, + { + "epoch": 2.066849315068493, + "grad_norm": 2.5734951496124268, + "learning_rate": 1.758732737611698e-06, + "loss": 1.7486, + "step": 1886 + }, + { + "epoch": 2.067945205479452, + "grad_norm": 2.6162502765655518, + "learning_rate": 1.7567018683996753e-06, + "loss": 1.7285, + "step": 1887 + }, + { + "epoch": 2.069041095890411, + "grad_norm": 2.654008388519287, + "learning_rate": 1.7546709991876524e-06, + "loss": 1.7195, + "step": 1888 + }, + { + "epoch": 2.0701369863013697, + "grad_norm": 2.64469575881958, + "learning_rate": 1.7526401299756296e-06, + "loss": 1.7247, + "step": 1889 + }, + { + "epoch": 2.0712328767123287, + "grad_norm": 2.5627946853637695, + "learning_rate": 1.750609260763607e-06, + "loss": 1.7254, + "step": 1890 + }, + { + "epoch": 2.0712328767123287, + "eval_loss": 1.951324224472046, + "eval_runtime": 37.2566, + "eval_samples_per_second": 26.841, + "eval_steps_per_second": 1.127, + "step": 1890 + }, + { + "epoch": 2.0723287671232877, + "grad_norm": 2.656520128250122, + "learning_rate": 1.7485783915515842e-06, + "loss": 1.7236, + "step": 1891 + }, + { + "epoch": 2.0734246575342468, + "grad_norm": 2.6355814933776855, + "learning_rate": 1.7465475223395615e-06, + "loss": 1.7118, + "step": 1892 + }, + { + "epoch": 2.0745205479452054, + "grad_norm": 2.5709362030029297, + "learning_rate": 1.7445166531275387e-06, + "loss": 1.685, + "step": 1893 + }, + { + "epoch": 2.0756164383561644, + "grad_norm": 2.555422782897949, + "learning_rate": 1.742485783915516e-06, + "loss": 1.8356, + "step": 1894 + }, + { + "epoch": 2.0767123287671234, + "grad_norm": 2.6279077529907227, + "learning_rate": 1.7404549147034933e-06, + "loss": 1.8079, + "step": 1895 + }, + { + "epoch": 2.077808219178082, + "grad_norm": 2.5886247158050537, + "learning_rate": 1.7384240454914706e-06, + "loss": 1.639, + "step": 1896 + }, + { + "epoch": 2.078904109589041, + "grad_norm": 2.562755584716797, + "learning_rate": 1.7363931762794478e-06, + "loss": 1.7345, + "step": 1897 + }, + { + "epoch": 2.08, + "grad_norm": 2.670344591140747, + "learning_rate": 1.7343623070674251e-06, + "loss": 1.7594, + "step": 1898 + }, + { + "epoch": 2.081095890410959, + "grad_norm": 2.567506790161133, + "learning_rate": 1.7323314378554024e-06, + "loss": 1.7161, + "step": 1899 + }, + { + "epoch": 2.0821917808219177, + "grad_norm": 2.562826156616211, + "learning_rate": 1.7303005686433797e-06, + "loss": 1.7275, + "step": 1900 + }, + { + "epoch": 2.0821917808219177, + "eval_loss": 1.9498006105422974, + "eval_runtime": 37.3069, + "eval_samples_per_second": 26.805, + "eval_steps_per_second": 1.126, + "step": 1900 + }, + { + "epoch": 2.0832876712328767, + "grad_norm": 2.538393974304199, + "learning_rate": 1.7282696994313567e-06, + "loss": 1.7187, + "step": 1901 + }, + { + "epoch": 2.0843835616438358, + "grad_norm": 2.687603235244751, + "learning_rate": 1.726238830219334e-06, + "loss": 1.7349, + "step": 1902 + }, + { + "epoch": 2.0854794520547943, + "grad_norm": 2.5806703567504883, + "learning_rate": 1.7242079610073113e-06, + "loss": 1.7135, + "step": 1903 + }, + { + "epoch": 2.0865753424657534, + "grad_norm": 2.468123435974121, + "learning_rate": 1.7221770917952886e-06, + "loss": 1.7946, + "step": 1904 + }, + { + "epoch": 2.0876712328767124, + "grad_norm": 2.514617443084717, + "learning_rate": 1.7201462225832658e-06, + "loss": 1.7403, + "step": 1905 + }, + { + "epoch": 2.0887671232876714, + "grad_norm": 2.7283825874328613, + "learning_rate": 1.7181153533712431e-06, + "loss": 1.7369, + "step": 1906 + }, + { + "epoch": 2.08986301369863, + "grad_norm": 2.5634899139404297, + "learning_rate": 1.7160844841592204e-06, + "loss": 1.6337, + "step": 1907 + }, + { + "epoch": 2.090958904109589, + "grad_norm": 2.635281801223755, + "learning_rate": 1.7140536149471977e-06, + "loss": 1.6836, + "step": 1908 + }, + { + "epoch": 2.092054794520548, + "grad_norm": 2.550048351287842, + "learning_rate": 1.712022745735175e-06, + "loss": 1.7171, + "step": 1909 + }, + { + "epoch": 2.0931506849315067, + "grad_norm": 2.62712025642395, + "learning_rate": 1.7099918765231522e-06, + "loss": 1.7027, + "step": 1910 + }, + { + "epoch": 2.0931506849315067, + "eval_loss": 1.9503322839736938, + "eval_runtime": 37.3435, + "eval_samples_per_second": 26.778, + "eval_steps_per_second": 1.125, + "step": 1910 + }, + { + "epoch": 2.0942465753424657, + "grad_norm": 2.540708541870117, + "learning_rate": 1.7079610073111295e-06, + "loss": 1.7358, + "step": 1911 + }, + { + "epoch": 2.0953424657534248, + "grad_norm": 2.6046950817108154, + "learning_rate": 1.7059301380991068e-06, + "loss": 1.6538, + "step": 1912 + }, + { + "epoch": 2.0964383561643833, + "grad_norm": 2.4760568141937256, + "learning_rate": 1.7038992688870836e-06, + "loss": 1.7207, + "step": 1913 + }, + { + "epoch": 2.0975342465753424, + "grad_norm": 2.428621292114258, + "learning_rate": 1.701868399675061e-06, + "loss": 1.7092, + "step": 1914 + }, + { + "epoch": 2.0986301369863014, + "grad_norm": 2.6177377700805664, + "learning_rate": 1.6998375304630382e-06, + "loss": 1.7429, + "step": 1915 + }, + { + "epoch": 2.0997260273972604, + "grad_norm": 2.4981932640075684, + "learning_rate": 1.6978066612510155e-06, + "loss": 1.7996, + "step": 1916 + }, + { + "epoch": 2.100821917808219, + "grad_norm": 2.617943286895752, + "learning_rate": 1.6957757920389927e-06, + "loss": 1.7716, + "step": 1917 + }, + { + "epoch": 2.101917808219178, + "grad_norm": 2.7581987380981445, + "learning_rate": 1.69374492282697e-06, + "loss": 1.7194, + "step": 1918 + }, + { + "epoch": 2.103013698630137, + "grad_norm": 2.622145652770996, + "learning_rate": 1.6917140536149473e-06, + "loss": 1.6887, + "step": 1919 + }, + { + "epoch": 2.1041095890410957, + "grad_norm": 2.742100477218628, + "learning_rate": 1.6896831844029246e-06, + "loss": 1.7787, + "step": 1920 + }, + { + "epoch": 2.1041095890410957, + "eval_loss": 1.9521653652191162, + "eval_runtime": 37.5529, + "eval_samples_per_second": 26.629, + "eval_steps_per_second": 1.118, + "step": 1920 + }, + { + "epoch": 2.1052054794520547, + "grad_norm": 2.5809876918792725, + "learning_rate": 1.6876523151909018e-06, + "loss": 1.7392, + "step": 1921 + }, + { + "epoch": 2.1063013698630138, + "grad_norm": 2.7006125450134277, + "learning_rate": 1.6856214459788791e-06, + "loss": 1.7442, + "step": 1922 + }, + { + "epoch": 2.107397260273973, + "grad_norm": 2.566088914871216, + "learning_rate": 1.6835905767668564e-06, + "loss": 1.6862, + "step": 1923 + }, + { + "epoch": 2.1084931506849314, + "grad_norm": 2.508967638015747, + "learning_rate": 1.6815597075548337e-06, + "loss": 1.695, + "step": 1924 + }, + { + "epoch": 2.1095890410958904, + "grad_norm": 2.6175615787506104, + "learning_rate": 1.679528838342811e-06, + "loss": 1.785, + "step": 1925 + }, + { + "epoch": 2.1106849315068494, + "grad_norm": 2.6141958236694336, + "learning_rate": 1.677497969130788e-06, + "loss": 1.6952, + "step": 1926 + }, + { + "epoch": 2.111780821917808, + "grad_norm": 2.4894959926605225, + "learning_rate": 1.6754670999187653e-06, + "loss": 1.6947, + "step": 1927 + }, + { + "epoch": 2.112876712328767, + "grad_norm": 2.5664544105529785, + "learning_rate": 1.6734362307067426e-06, + "loss": 1.731, + "step": 1928 + }, + { + "epoch": 2.113972602739726, + "grad_norm": 2.5422523021698, + "learning_rate": 1.6714053614947198e-06, + "loss": 1.6765, + "step": 1929 + }, + { + "epoch": 2.115068493150685, + "grad_norm": 2.6428608894348145, + "learning_rate": 1.669374492282697e-06, + "loss": 1.6834, + "step": 1930 + }, + { + "epoch": 2.115068493150685, + "eval_loss": 1.952999472618103, + "eval_runtime": 37.6765, + "eval_samples_per_second": 26.542, + "eval_steps_per_second": 1.115, + "step": 1930 + }, + { + "epoch": 2.1161643835616437, + "grad_norm": 2.6088430881500244, + "learning_rate": 1.6673436230706744e-06, + "loss": 1.7713, + "step": 1931 + }, + { + "epoch": 2.1172602739726027, + "grad_norm": 2.49733304977417, + "learning_rate": 1.6653127538586517e-06, + "loss": 1.7461, + "step": 1932 + }, + { + "epoch": 2.118356164383562, + "grad_norm": 2.5760915279388428, + "learning_rate": 1.663281884646629e-06, + "loss": 1.6837, + "step": 1933 + }, + { + "epoch": 2.1194520547945204, + "grad_norm": 2.4771194458007812, + "learning_rate": 1.6612510154346062e-06, + "loss": 1.7191, + "step": 1934 + }, + { + "epoch": 2.1205479452054794, + "grad_norm": 2.6267881393432617, + "learning_rate": 1.6592201462225835e-06, + "loss": 1.7819, + "step": 1935 + }, + { + "epoch": 2.1216438356164384, + "grad_norm": 2.6205623149871826, + "learning_rate": 1.6571892770105608e-06, + "loss": 1.7443, + "step": 1936 + }, + { + "epoch": 2.1227397260273975, + "grad_norm": 2.660942792892456, + "learning_rate": 1.655158407798538e-06, + "loss": 1.783, + "step": 1937 + }, + { + "epoch": 2.123835616438356, + "grad_norm": 2.5452628135681152, + "learning_rate": 1.653127538586515e-06, + "loss": 1.7316, + "step": 1938 + }, + { + "epoch": 2.124931506849315, + "grad_norm": 2.542837142944336, + "learning_rate": 1.6510966693744924e-06, + "loss": 1.7624, + "step": 1939 + }, + { + "epoch": 2.126027397260274, + "grad_norm": 2.535205841064453, + "learning_rate": 1.6490658001624696e-06, + "loss": 1.7664, + "step": 1940 + }, + { + "epoch": 2.126027397260274, + "eval_loss": 1.9492406845092773, + "eval_runtime": 37.3497, + "eval_samples_per_second": 26.774, + "eval_steps_per_second": 1.125, + "step": 1940 + }, + { + "epoch": 2.1271232876712327, + "grad_norm": 2.7622485160827637, + "learning_rate": 1.647034930950447e-06, + "loss": 1.7603, + "step": 1941 + }, + { + "epoch": 2.1282191780821917, + "grad_norm": 2.629058599472046, + "learning_rate": 1.6450040617384242e-06, + "loss": 1.7036, + "step": 1942 + }, + { + "epoch": 2.1293150684931508, + "grad_norm": 2.543135643005371, + "learning_rate": 1.6429731925264015e-06, + "loss": 1.7172, + "step": 1943 + }, + { + "epoch": 2.1304109589041094, + "grad_norm": 2.792086601257324, + "learning_rate": 1.6409423233143788e-06, + "loss": 1.7486, + "step": 1944 + }, + { + "epoch": 2.1315068493150684, + "grad_norm": 2.7788681983947754, + "learning_rate": 1.638911454102356e-06, + "loss": 1.7136, + "step": 1945 + }, + { + "epoch": 2.1326027397260274, + "grad_norm": 2.5364747047424316, + "learning_rate": 1.6368805848903333e-06, + "loss": 1.6766, + "step": 1946 + }, + { + "epoch": 2.1336986301369865, + "grad_norm": 2.5749800205230713, + "learning_rate": 1.6348497156783106e-06, + "loss": 1.7627, + "step": 1947 + }, + { + "epoch": 2.134794520547945, + "grad_norm": 2.4931676387786865, + "learning_rate": 1.6328188464662879e-06, + "loss": 1.7049, + "step": 1948 + }, + { + "epoch": 2.135890410958904, + "grad_norm": 2.547396183013916, + "learning_rate": 1.6307879772542651e-06, + "loss": 1.7165, + "step": 1949 + }, + { + "epoch": 2.136986301369863, + "grad_norm": 2.5881359577178955, + "learning_rate": 1.628757108042242e-06, + "loss": 1.7508, + "step": 1950 + }, + { + "epoch": 2.136986301369863, + "eval_loss": 1.9464337825775146, + "eval_runtime": 37.284, + "eval_samples_per_second": 26.821, + "eval_steps_per_second": 1.126, + "step": 1950 + }, + { + "epoch": 2.138082191780822, + "grad_norm": 2.6131961345672607, + "learning_rate": 1.6267262388302193e-06, + "loss": 1.7717, + "step": 1951 + }, + { + "epoch": 2.1391780821917807, + "grad_norm": 2.5644335746765137, + "learning_rate": 1.6246953696181965e-06, + "loss": 1.7139, + "step": 1952 + }, + { + "epoch": 2.1402739726027398, + "grad_norm": 2.5851385593414307, + "learning_rate": 1.6226645004061738e-06, + "loss": 1.7473, + "step": 1953 + }, + { + "epoch": 2.141369863013699, + "grad_norm": 2.493952989578247, + "learning_rate": 1.620633631194151e-06, + "loss": 1.8209, + "step": 1954 + }, + { + "epoch": 2.1424657534246574, + "grad_norm": 2.5686917304992676, + "learning_rate": 1.6186027619821284e-06, + "loss": 1.7527, + "step": 1955 + }, + { + "epoch": 2.1435616438356164, + "grad_norm": 2.6349050998687744, + "learning_rate": 1.6165718927701056e-06, + "loss": 1.7033, + "step": 1956 + }, + { + "epoch": 2.1446575342465755, + "grad_norm": 2.7004778385162354, + "learning_rate": 1.614541023558083e-06, + "loss": 1.7498, + "step": 1957 + }, + { + "epoch": 2.145753424657534, + "grad_norm": 2.5406992435455322, + "learning_rate": 1.6125101543460602e-06, + "loss": 1.7605, + "step": 1958 + }, + { + "epoch": 2.146849315068493, + "grad_norm": 2.796125650405884, + "learning_rate": 1.6104792851340375e-06, + "loss": 1.7087, + "step": 1959 + }, + { + "epoch": 2.147945205479452, + "grad_norm": 2.703362226486206, + "learning_rate": 1.6084484159220147e-06, + "loss": 1.751, + "step": 1960 + }, + { + "epoch": 2.147945205479452, + "eval_loss": 1.9470759630203247, + "eval_runtime": 37.2721, + "eval_samples_per_second": 26.83, + "eval_steps_per_second": 1.127, + "step": 1960 + }, + { + "epoch": 2.149041095890411, + "grad_norm": 2.6464223861694336, + "learning_rate": 1.606417546709992e-06, + "loss": 1.7147, + "step": 1961 + }, + { + "epoch": 2.1501369863013697, + "grad_norm": 2.6499273777008057, + "learning_rate": 1.6043866774979693e-06, + "loss": 1.7015, + "step": 1962 + }, + { + "epoch": 2.1512328767123288, + "grad_norm": 2.583768844604492, + "learning_rate": 1.6023558082859464e-06, + "loss": 1.7158, + "step": 1963 + }, + { + "epoch": 2.152328767123288, + "grad_norm": 2.6245381832122803, + "learning_rate": 1.6003249390739236e-06, + "loss": 1.6824, + "step": 1964 + }, + { + "epoch": 2.1534246575342464, + "grad_norm": 2.497676134109497, + "learning_rate": 1.598294069861901e-06, + "loss": 1.7345, + "step": 1965 + }, + { + "epoch": 2.1545205479452054, + "grad_norm": 2.7218856811523438, + "learning_rate": 1.5962632006498782e-06, + "loss": 1.6988, + "step": 1966 + }, + { + "epoch": 2.1556164383561645, + "grad_norm": 2.7654666900634766, + "learning_rate": 1.5942323314378555e-06, + "loss": 1.7635, + "step": 1967 + }, + { + "epoch": 2.1567123287671235, + "grad_norm": 2.598526954650879, + "learning_rate": 1.5922014622258327e-06, + "loss": 1.6769, + "step": 1968 + }, + { + "epoch": 2.157808219178082, + "grad_norm": 2.5995986461639404, + "learning_rate": 1.59017059301381e-06, + "loss": 1.6928, + "step": 1969 + }, + { + "epoch": 2.158904109589041, + "grad_norm": 2.5692837238311768, + "learning_rate": 1.5881397238017873e-06, + "loss": 1.7415, + "step": 1970 + }, + { + "epoch": 2.158904109589041, + "eval_loss": 1.9472142457962036, + "eval_runtime": 37.258, + "eval_samples_per_second": 26.84, + "eval_steps_per_second": 1.127, + "step": 1970 + }, + { + "epoch": 2.16, + "grad_norm": 2.622732162475586, + "learning_rate": 1.5861088545897646e-06, + "loss": 1.7383, + "step": 1971 + }, + { + "epoch": 2.1610958904109587, + "grad_norm": 2.573005437850952, + "learning_rate": 1.5840779853777418e-06, + "loss": 1.6696, + "step": 1972 + }, + { + "epoch": 2.1621917808219178, + "grad_norm": 2.7022411823272705, + "learning_rate": 1.5820471161657191e-06, + "loss": 1.7231, + "step": 1973 + }, + { + "epoch": 2.163287671232877, + "grad_norm": 2.739145278930664, + "learning_rate": 1.5800162469536964e-06, + "loss": 1.7012, + "step": 1974 + }, + { + "epoch": 2.1643835616438354, + "grad_norm": 2.7034459114074707, + "learning_rate": 1.5779853777416735e-06, + "loss": 1.769, + "step": 1975 + }, + { + "epoch": 2.1654794520547944, + "grad_norm": 2.745359182357788, + "learning_rate": 1.5759545085296507e-06, + "loss": 1.7424, + "step": 1976 + }, + { + "epoch": 2.1665753424657535, + "grad_norm": 2.722721576690674, + "learning_rate": 1.573923639317628e-06, + "loss": 1.733, + "step": 1977 + }, + { + "epoch": 2.1676712328767125, + "grad_norm": 2.689074754714966, + "learning_rate": 1.5718927701056053e-06, + "loss": 1.7468, + "step": 1978 + }, + { + "epoch": 2.168767123287671, + "grad_norm": 2.6735317707061768, + "learning_rate": 1.5698619008935826e-06, + "loss": 1.7312, + "step": 1979 + }, + { + "epoch": 2.16986301369863, + "grad_norm": 2.6908175945281982, + "learning_rate": 1.5678310316815598e-06, + "loss": 1.7502, + "step": 1980 + }, + { + "epoch": 2.16986301369863, + "eval_loss": 1.947867512702942, + "eval_runtime": 37.1858, + "eval_samples_per_second": 26.892, + "eval_steps_per_second": 1.129, + "step": 1980 + }, + { + "epoch": 2.170958904109589, + "grad_norm": 2.524951934814453, + "learning_rate": 1.5658001624695371e-06, + "loss": 1.6863, + "step": 1981 + }, + { + "epoch": 2.172054794520548, + "grad_norm": 2.746180295944214, + "learning_rate": 1.5637692932575144e-06, + "loss": 1.7621, + "step": 1982 + }, + { + "epoch": 2.1731506849315068, + "grad_norm": 2.4799301624298096, + "learning_rate": 1.5617384240454917e-06, + "loss": 1.7121, + "step": 1983 + }, + { + "epoch": 2.174246575342466, + "grad_norm": 2.6821515560150146, + "learning_rate": 1.559707554833469e-06, + "loss": 1.7271, + "step": 1984 + }, + { + "epoch": 2.175342465753425, + "grad_norm": 2.572122812271118, + "learning_rate": 1.5576766856214462e-06, + "loss": 1.7671, + "step": 1985 + }, + { + "epoch": 2.1764383561643834, + "grad_norm": 2.614806652069092, + "learning_rate": 1.5556458164094235e-06, + "loss": 1.6411, + "step": 1986 + }, + { + "epoch": 2.1775342465753424, + "grad_norm": 2.602883815765381, + "learning_rate": 1.5536149471974008e-06, + "loss": 1.7143, + "step": 1987 + }, + { + "epoch": 2.1786301369863015, + "grad_norm": 2.5893354415893555, + "learning_rate": 1.5515840779853778e-06, + "loss": 1.6733, + "step": 1988 + }, + { + "epoch": 2.17972602739726, + "grad_norm": 2.652308702468872, + "learning_rate": 1.549553208773355e-06, + "loss": 1.7311, + "step": 1989 + }, + { + "epoch": 2.180821917808219, + "grad_norm": 2.7129695415496826, + "learning_rate": 1.5475223395613324e-06, + "loss": 1.7001, + "step": 1990 + }, + { + "epoch": 2.180821917808219, + "eval_loss": 1.9512702226638794, + "eval_runtime": 37.225, + "eval_samples_per_second": 26.864, + "eval_steps_per_second": 1.128, + "step": 1990 + }, + { + "epoch": 2.181917808219178, + "grad_norm": 2.563638210296631, + "learning_rate": 1.5454914703493097e-06, + "loss": 1.7118, + "step": 1991 + }, + { + "epoch": 2.183013698630137, + "grad_norm": 2.629056453704834, + "learning_rate": 1.543460601137287e-06, + "loss": 1.8041, + "step": 1992 + }, + { + "epoch": 2.1841095890410958, + "grad_norm": 2.6072700023651123, + "learning_rate": 1.5414297319252642e-06, + "loss": 1.7568, + "step": 1993 + }, + { + "epoch": 2.185205479452055, + "grad_norm": 2.525784730911255, + "learning_rate": 1.5393988627132415e-06, + "loss": 1.748, + "step": 1994 + }, + { + "epoch": 2.186301369863014, + "grad_norm": 2.564699649810791, + "learning_rate": 1.5373679935012188e-06, + "loss": 1.7148, + "step": 1995 + }, + { + "epoch": 2.1873972602739724, + "grad_norm": 2.7365031242370605, + "learning_rate": 1.535337124289196e-06, + "loss": 1.6714, + "step": 1996 + }, + { + "epoch": 2.1884931506849314, + "grad_norm": 2.719954252243042, + "learning_rate": 1.5333062550771733e-06, + "loss": 1.7301, + "step": 1997 + }, + { + "epoch": 2.1895890410958905, + "grad_norm": 2.7979655265808105, + "learning_rate": 1.5312753858651506e-06, + "loss": 1.8163, + "step": 1998 + }, + { + "epoch": 2.1906849315068495, + "grad_norm": 2.6182446479797363, + "learning_rate": 1.5292445166531279e-06, + "loss": 1.713, + "step": 1999 + }, + { + "epoch": 2.191780821917808, + "grad_norm": 2.7218992710113525, + "learning_rate": 1.5272136474411047e-06, + "loss": 1.7355, + "step": 2000 + }, + { + "epoch": 2.191780821917808, + "eval_loss": 1.950900673866272, + "eval_runtime": 37.2927, + "eval_samples_per_second": 26.815, + "eval_steps_per_second": 1.126, + "step": 2000 + }, + { + "epoch": 2.192876712328767, + "grad_norm": 2.7384393215179443, + "learning_rate": 1.525182778229082e-06, + "loss": 1.7777, + "step": 2001 + }, + { + "epoch": 2.193972602739726, + "grad_norm": 2.7223575115203857, + "learning_rate": 1.5231519090170593e-06, + "loss": 1.7136, + "step": 2002 + }, + { + "epoch": 2.1950684931506848, + "grad_norm": 2.5952577590942383, + "learning_rate": 1.5211210398050365e-06, + "loss": 1.6964, + "step": 2003 + }, + { + "epoch": 2.196164383561644, + "grad_norm": 2.626406669616699, + "learning_rate": 1.5190901705930138e-06, + "loss": 1.6996, + "step": 2004 + }, + { + "epoch": 2.197260273972603, + "grad_norm": 2.6563360691070557, + "learning_rate": 1.517059301380991e-06, + "loss": 1.7428, + "step": 2005 + }, + { + "epoch": 2.198356164383562, + "grad_norm": 2.647664785385132, + "learning_rate": 1.5150284321689684e-06, + "loss": 1.6953, + "step": 2006 + }, + { + "epoch": 2.1994520547945204, + "grad_norm": 2.5670430660247803, + "learning_rate": 1.5129975629569456e-06, + "loss": 1.7215, + "step": 2007 + }, + { + "epoch": 2.2005479452054795, + "grad_norm": 2.6148288249969482, + "learning_rate": 1.510966693744923e-06, + "loss": 1.724, + "step": 2008 + }, + { + "epoch": 2.2016438356164385, + "grad_norm": 2.8430588245391846, + "learning_rate": 1.5089358245329002e-06, + "loss": 1.7255, + "step": 2009 + }, + { + "epoch": 2.202739726027397, + "grad_norm": 2.5937509536743164, + "learning_rate": 1.5069049553208775e-06, + "loss": 1.7055, + "step": 2010 + }, + { + "epoch": 2.202739726027397, + "eval_loss": 1.9469048976898193, + "eval_runtime": 37.2173, + "eval_samples_per_second": 26.869, + "eval_steps_per_second": 1.129, + "step": 2010 + }, + { + "epoch": 2.203835616438356, + "grad_norm": 2.7167937755584717, + "learning_rate": 1.5048740861088547e-06, + "loss": 1.7474, + "step": 2011 + }, + { + "epoch": 2.204931506849315, + "grad_norm": 2.53837251663208, + "learning_rate": 1.502843216896832e-06, + "loss": 1.74, + "step": 2012 + }, + { + "epoch": 2.206027397260274, + "grad_norm": 2.604931592941284, + "learning_rate": 1.500812347684809e-06, + "loss": 1.6573, + "step": 2013 + }, + { + "epoch": 2.207123287671233, + "grad_norm": 2.6670846939086914, + "learning_rate": 1.4987814784727864e-06, + "loss": 1.7208, + "step": 2014 + }, + { + "epoch": 2.208219178082192, + "grad_norm": 2.677746534347534, + "learning_rate": 1.4967506092607636e-06, + "loss": 1.728, + "step": 2015 + }, + { + "epoch": 2.209315068493151, + "grad_norm": 2.6490612030029297, + "learning_rate": 1.494719740048741e-06, + "loss": 1.6728, + "step": 2016 + }, + { + "epoch": 2.2104109589041094, + "grad_norm": 2.6571807861328125, + "learning_rate": 1.4926888708367182e-06, + "loss": 1.7528, + "step": 2017 + }, + { + "epoch": 2.2115068493150685, + "grad_norm": 2.5933258533477783, + "learning_rate": 1.4906580016246955e-06, + "loss": 1.757, + "step": 2018 + }, + { + "epoch": 2.2126027397260275, + "grad_norm": 2.6451070308685303, + "learning_rate": 1.4886271324126727e-06, + "loss": 1.763, + "step": 2019 + }, + { + "epoch": 2.213698630136986, + "grad_norm": 2.5259275436401367, + "learning_rate": 1.48659626320065e-06, + "loss": 1.7979, + "step": 2020 + }, + { + "epoch": 2.213698630136986, + "eval_loss": 1.9480665922164917, + "eval_runtime": 37.5325, + "eval_samples_per_second": 26.644, + "eval_steps_per_second": 1.119, + "step": 2020 + }, + { + "epoch": 2.214794520547945, + "grad_norm": 2.6959381103515625, + "learning_rate": 1.4845653939886273e-06, + "loss": 1.6376, + "step": 2021 + }, + { + "epoch": 2.215890410958904, + "grad_norm": 2.6179497241973877, + "learning_rate": 1.4825345247766046e-06, + "loss": 1.6575, + "step": 2022 + }, + { + "epoch": 2.216986301369863, + "grad_norm": 2.5618393421173096, + "learning_rate": 1.4805036555645818e-06, + "loss": 1.6486, + "step": 2023 + }, + { + "epoch": 2.2180821917808218, + "grad_norm": 2.545051336288452, + "learning_rate": 1.4784727863525591e-06, + "loss": 1.7218, + "step": 2024 + }, + { + "epoch": 2.219178082191781, + "grad_norm": 2.6248512268066406, + "learning_rate": 1.4764419171405362e-06, + "loss": 1.691, + "step": 2025 + }, + { + "epoch": 2.22027397260274, + "grad_norm": 2.65645170211792, + "learning_rate": 1.4744110479285135e-06, + "loss": 1.7602, + "step": 2026 + }, + { + "epoch": 2.2213698630136984, + "grad_norm": 2.6771113872528076, + "learning_rate": 1.4723801787164907e-06, + "loss": 1.7573, + "step": 2027 + }, + { + "epoch": 2.2224657534246575, + "grad_norm": 2.5466039180755615, + "learning_rate": 1.470349309504468e-06, + "loss": 1.7604, + "step": 2028 + }, + { + "epoch": 2.2235616438356165, + "grad_norm": 2.685079574584961, + "learning_rate": 1.4683184402924453e-06, + "loss": 1.7341, + "step": 2029 + }, + { + "epoch": 2.2246575342465755, + "grad_norm": 2.6854352951049805, + "learning_rate": 1.4662875710804226e-06, + "loss": 1.7427, + "step": 2030 + }, + { + "epoch": 2.2246575342465755, + "eval_loss": 1.9459377527236938, + "eval_runtime": 37.5255, + "eval_samples_per_second": 26.649, + "eval_steps_per_second": 1.119, + "step": 2030 + }, + { + "epoch": 2.225753424657534, + "grad_norm": 2.667675733566284, + "learning_rate": 1.4642567018683998e-06, + "loss": 1.8328, + "step": 2031 + }, + { + "epoch": 2.226849315068493, + "grad_norm": 2.603386640548706, + "learning_rate": 1.4622258326563771e-06, + "loss": 1.7193, + "step": 2032 + }, + { + "epoch": 2.227945205479452, + "grad_norm": 2.4960451126098633, + "learning_rate": 1.4601949634443544e-06, + "loss": 1.695, + "step": 2033 + }, + { + "epoch": 2.2290410958904108, + "grad_norm": 2.755707025527954, + "learning_rate": 1.4581640942323317e-06, + "loss": 1.7475, + "step": 2034 + }, + { + "epoch": 2.23013698630137, + "grad_norm": 2.5849006175994873, + "learning_rate": 1.456133225020309e-06, + "loss": 1.8106, + "step": 2035 + }, + { + "epoch": 2.231232876712329, + "grad_norm": 2.690680980682373, + "learning_rate": 1.4541023558082862e-06, + "loss": 1.7368, + "step": 2036 + }, + { + "epoch": 2.232328767123288, + "grad_norm": 2.6760425567626953, + "learning_rate": 1.4520714865962635e-06, + "loss": 1.722, + "step": 2037 + }, + { + "epoch": 2.2334246575342465, + "grad_norm": 2.656379461288452, + "learning_rate": 1.4500406173842406e-06, + "loss": 1.7623, + "step": 2038 + }, + { + "epoch": 2.2345205479452055, + "grad_norm": 2.8213517665863037, + "learning_rate": 1.4480097481722178e-06, + "loss": 1.7344, + "step": 2039 + }, + { + "epoch": 2.2356164383561645, + "grad_norm": 2.877163887023926, + "learning_rate": 1.4459788789601951e-06, + "loss": 1.7184, + "step": 2040 + }, + { + "epoch": 2.2356164383561645, + "eval_loss": 1.9430633783340454, + "eval_runtime": 37.3102, + "eval_samples_per_second": 26.802, + "eval_steps_per_second": 1.126, + "step": 2040 + }, + { + "epoch": 2.236712328767123, + "grad_norm": 2.7285237312316895, + "learning_rate": 1.4439480097481724e-06, + "loss": 1.6826, + "step": 2041 + }, + { + "epoch": 2.237808219178082, + "grad_norm": 2.5208237171173096, + "learning_rate": 1.4419171405361497e-06, + "loss": 1.6788, + "step": 2042 + }, + { + "epoch": 2.238904109589041, + "grad_norm": 2.7415900230407715, + "learning_rate": 1.439886271324127e-06, + "loss": 1.7423, + "step": 2043 + }, + { + "epoch": 2.24, + "grad_norm": 2.760897397994995, + "learning_rate": 1.4378554021121042e-06, + "loss": 1.7341, + "step": 2044 + }, + { + "epoch": 2.241095890410959, + "grad_norm": 2.5650556087493896, + "learning_rate": 1.4358245329000815e-06, + "loss": 1.7036, + "step": 2045 + }, + { + "epoch": 2.242191780821918, + "grad_norm": 2.60780930519104, + "learning_rate": 1.4337936636880588e-06, + "loss": 1.7719, + "step": 2046 + }, + { + "epoch": 2.243287671232877, + "grad_norm": 2.592473268508911, + "learning_rate": 1.431762794476036e-06, + "loss": 1.7421, + "step": 2047 + }, + { + "epoch": 2.2443835616438355, + "grad_norm": 2.6672041416168213, + "learning_rate": 1.4297319252640133e-06, + "loss": 1.6865, + "step": 2048 + }, + { + "epoch": 2.2454794520547945, + "grad_norm": 2.6182847023010254, + "learning_rate": 1.4277010560519906e-06, + "loss": 1.731, + "step": 2049 + }, + { + "epoch": 2.2465753424657535, + "grad_norm": 2.8213913440704346, + "learning_rate": 1.4256701868399674e-06, + "loss": 1.6845, + "step": 2050 + }, + { + "epoch": 2.2465753424657535, + "eval_loss": 1.9434115886688232, + "eval_runtime": 37.3048, + "eval_samples_per_second": 26.806, + "eval_steps_per_second": 1.126, + "step": 2050 + }, + { + "epoch": 2.247671232876712, + "grad_norm": 2.7278084754943848, + "learning_rate": 1.4236393176279447e-06, + "loss": 1.8077, + "step": 2051 + }, + { + "epoch": 2.248767123287671, + "grad_norm": 2.5665392875671387, + "learning_rate": 1.421608448415922e-06, + "loss": 1.7235, + "step": 2052 + }, + { + "epoch": 2.24986301369863, + "grad_norm": 2.680154323577881, + "learning_rate": 1.4195775792038993e-06, + "loss": 1.8164, + "step": 2053 + }, + { + "epoch": 2.250958904109589, + "grad_norm": 2.770051956176758, + "learning_rate": 1.4175467099918765e-06, + "loss": 1.6948, + "step": 2054 + }, + { + "epoch": 2.252054794520548, + "grad_norm": 2.6189041137695312, + "learning_rate": 1.4155158407798538e-06, + "loss": 1.7369, + "step": 2055 + }, + { + "epoch": 2.253150684931507, + "grad_norm": 2.6655054092407227, + "learning_rate": 1.413484971567831e-06, + "loss": 1.7456, + "step": 2056 + }, + { + "epoch": 2.254246575342466, + "grad_norm": 2.6520164012908936, + "learning_rate": 1.4114541023558084e-06, + "loss": 1.6719, + "step": 2057 + }, + { + "epoch": 2.255342465753425, + "grad_norm": 2.594533681869507, + "learning_rate": 1.4094232331437856e-06, + "loss": 1.7078, + "step": 2058 + }, + { + "epoch": 2.2564383561643835, + "grad_norm": 2.592308759689331, + "learning_rate": 1.407392363931763e-06, + "loss": 1.7517, + "step": 2059 + }, + { + "epoch": 2.2575342465753425, + "grad_norm": 2.5815367698669434, + "learning_rate": 1.4053614947197402e-06, + "loss": 1.8262, + "step": 2060 + }, + { + "epoch": 2.2575342465753425, + "eval_loss": 1.945846438407898, + "eval_runtime": 37.186, + "eval_samples_per_second": 26.892, + "eval_steps_per_second": 1.129, + "step": 2060 + }, + { + "epoch": 2.2586301369863016, + "grad_norm": 2.643691301345825, + "learning_rate": 1.4033306255077175e-06, + "loss": 1.7559, + "step": 2061 + }, + { + "epoch": 2.25972602739726, + "grad_norm": 2.6091558933258057, + "learning_rate": 1.4012997562956945e-06, + "loss": 1.6433, + "step": 2062 + }, + { + "epoch": 2.260821917808219, + "grad_norm": 2.7172136306762695, + "learning_rate": 1.3992688870836718e-06, + "loss": 1.7735, + "step": 2063 + }, + { + "epoch": 2.261917808219178, + "grad_norm": 2.5673139095306396, + "learning_rate": 1.397238017871649e-06, + "loss": 1.7402, + "step": 2064 + }, + { + "epoch": 2.263013698630137, + "grad_norm": 2.711242198944092, + "learning_rate": 1.3952071486596264e-06, + "loss": 1.7183, + "step": 2065 + }, + { + "epoch": 2.264109589041096, + "grad_norm": 2.6999948024749756, + "learning_rate": 1.3931762794476036e-06, + "loss": 1.6856, + "step": 2066 + }, + { + "epoch": 2.265205479452055, + "grad_norm": 2.670964002609253, + "learning_rate": 1.391145410235581e-06, + "loss": 1.78, + "step": 2067 + }, + { + "epoch": 2.266301369863014, + "grad_norm": 2.4599525928497314, + "learning_rate": 1.3891145410235582e-06, + "loss": 1.7143, + "step": 2068 + }, + { + "epoch": 2.2673972602739725, + "grad_norm": 2.552266836166382, + "learning_rate": 1.3870836718115355e-06, + "loss": 1.6503, + "step": 2069 + }, + { + "epoch": 2.2684931506849315, + "grad_norm": 2.5764405727386475, + "learning_rate": 1.3850528025995127e-06, + "loss": 1.7046, + "step": 2070 + }, + { + "epoch": 2.2684931506849315, + "eval_loss": 1.9457523822784424, + "eval_runtime": 37.217, + "eval_samples_per_second": 26.869, + "eval_steps_per_second": 1.129, + "step": 2070 + }, + { + "epoch": 2.2695890410958905, + "grad_norm": 2.524228572845459, + "learning_rate": 1.38302193338749e-06, + "loss": 1.7078, + "step": 2071 + }, + { + "epoch": 2.270684931506849, + "grad_norm": 2.5531418323516846, + "learning_rate": 1.3809910641754673e-06, + "loss": 1.646, + "step": 2072 + }, + { + "epoch": 2.271780821917808, + "grad_norm": 2.705824613571167, + "learning_rate": 1.3789601949634446e-06, + "loss": 1.7407, + "step": 2073 + }, + { + "epoch": 2.272876712328767, + "grad_norm": 2.5266356468200684, + "learning_rate": 1.3769293257514218e-06, + "loss": 1.6472, + "step": 2074 + }, + { + "epoch": 2.2739726027397262, + "grad_norm": 2.5785422325134277, + "learning_rate": 1.374898456539399e-06, + "loss": 1.7561, + "step": 2075 + }, + { + "epoch": 2.275068493150685, + "grad_norm": 2.663120985031128, + "learning_rate": 1.3728675873273762e-06, + "loss": 1.6901, + "step": 2076 + }, + { + "epoch": 2.276164383561644, + "grad_norm": 2.6085517406463623, + "learning_rate": 1.3708367181153535e-06, + "loss": 1.7829, + "step": 2077 + }, + { + "epoch": 2.277260273972603, + "grad_norm": 2.6439929008483887, + "learning_rate": 1.3688058489033307e-06, + "loss": 1.7684, + "step": 2078 + }, + { + "epoch": 2.2783561643835615, + "grad_norm": 2.6614694595336914, + "learning_rate": 1.366774979691308e-06, + "loss": 1.726, + "step": 2079 + }, + { + "epoch": 2.2794520547945205, + "grad_norm": 2.7339539527893066, + "learning_rate": 1.3647441104792853e-06, + "loss": 1.7577, + "step": 2080 + }, + { + "epoch": 2.2794520547945205, + "eval_loss": 1.9456545114517212, + "eval_runtime": 37.174, + "eval_samples_per_second": 26.901, + "eval_steps_per_second": 1.13, + "step": 2080 + }, + { + "epoch": 2.2805479452054795, + "grad_norm": 2.603545665740967, + "learning_rate": 1.3627132412672626e-06, + "loss": 1.7297, + "step": 2081 + }, + { + "epoch": 2.281643835616438, + "grad_norm": 2.548466682434082, + "learning_rate": 1.3606823720552398e-06, + "loss": 1.6701, + "step": 2082 + }, + { + "epoch": 2.282739726027397, + "grad_norm": 2.6043875217437744, + "learning_rate": 1.3586515028432171e-06, + "loss": 1.7369, + "step": 2083 + }, + { + "epoch": 2.283835616438356, + "grad_norm": 2.7412586212158203, + "learning_rate": 1.3566206336311944e-06, + "loss": 1.654, + "step": 2084 + }, + { + "epoch": 2.2849315068493152, + "grad_norm": 2.669410228729248, + "learning_rate": 1.3545897644191717e-06, + "loss": 1.7186, + "step": 2085 + }, + { + "epoch": 2.286027397260274, + "grad_norm": 2.7052674293518066, + "learning_rate": 1.352558895207149e-06, + "loss": 1.7005, + "step": 2086 + }, + { + "epoch": 2.287123287671233, + "grad_norm": 2.5840418338775635, + "learning_rate": 1.3505280259951258e-06, + "loss": 1.7459, + "step": 2087 + }, + { + "epoch": 2.288219178082192, + "grad_norm": 2.6355416774749756, + "learning_rate": 1.348497156783103e-06, + "loss": 1.7243, + "step": 2088 + }, + { + "epoch": 2.289315068493151, + "grad_norm": 2.6697628498077393, + "learning_rate": 1.3464662875710804e-06, + "loss": 1.7597, + "step": 2089 + }, + { + "epoch": 2.2904109589041095, + "grad_norm": 2.679842233657837, + "learning_rate": 1.3444354183590576e-06, + "loss": 1.6557, + "step": 2090 + }, + { + "epoch": 2.2904109589041095, + "eval_loss": 1.9469218254089355, + "eval_runtime": 37.1615, + "eval_samples_per_second": 26.91, + "eval_steps_per_second": 1.13, + "step": 2090 + }, + { + "epoch": 2.2915068493150685, + "grad_norm": 2.8354430198669434, + "learning_rate": 1.342404549147035e-06, + "loss": 1.7496, + "step": 2091 + }, + { + "epoch": 2.2926027397260276, + "grad_norm": 2.6957387924194336, + "learning_rate": 1.3403736799350122e-06, + "loss": 1.6872, + "step": 2092 + }, + { + "epoch": 2.293698630136986, + "grad_norm": 2.5815868377685547, + "learning_rate": 1.3383428107229895e-06, + "loss": 1.7497, + "step": 2093 + }, + { + "epoch": 2.294794520547945, + "grad_norm": 2.727085828781128, + "learning_rate": 1.3363119415109667e-06, + "loss": 1.7723, + "step": 2094 + }, + { + "epoch": 2.2958904109589042, + "grad_norm": 2.6119000911712646, + "learning_rate": 1.334281072298944e-06, + "loss": 1.7414, + "step": 2095 + }, + { + "epoch": 2.296986301369863, + "grad_norm": 2.6403088569641113, + "learning_rate": 1.3322502030869213e-06, + "loss": 1.7398, + "step": 2096 + }, + { + "epoch": 2.298082191780822, + "grad_norm": 2.5066816806793213, + "learning_rate": 1.3302193338748986e-06, + "loss": 1.6974, + "step": 2097 + }, + { + "epoch": 2.299178082191781, + "grad_norm": 2.6601762771606445, + "learning_rate": 1.3281884646628758e-06, + "loss": 1.7578, + "step": 2098 + }, + { + "epoch": 2.30027397260274, + "grad_norm": 2.571499824523926, + "learning_rate": 1.3261575954508531e-06, + "loss": 1.7573, + "step": 2099 + }, + { + "epoch": 2.3013698630136985, + "grad_norm": 2.533543825149536, + "learning_rate": 1.3241267262388302e-06, + "loss": 1.7566, + "step": 2100 + }, + { + "epoch": 2.3013698630136985, + "eval_loss": 1.9441688060760498, + "eval_runtime": 37.2158, + "eval_samples_per_second": 26.87, + "eval_steps_per_second": 1.129, + "step": 2100 + }, + { + "epoch": 2.3024657534246575, + "grad_norm": 2.782416582107544, + "learning_rate": 1.3220958570268074e-06, + "loss": 1.714, + "step": 2101 + }, + { + "epoch": 2.3035616438356166, + "grad_norm": 2.742504358291626, + "learning_rate": 1.3200649878147847e-06, + "loss": 1.6677, + "step": 2102 + }, + { + "epoch": 2.304657534246575, + "grad_norm": 2.7778940200805664, + "learning_rate": 1.318034118602762e-06, + "loss": 1.7178, + "step": 2103 + }, + { + "epoch": 2.305753424657534, + "grad_norm": 2.7003014087677, + "learning_rate": 1.3160032493907393e-06, + "loss": 1.7691, + "step": 2104 + }, + { + "epoch": 2.3068493150684932, + "grad_norm": 2.6644296646118164, + "learning_rate": 1.3139723801787166e-06, + "loss": 1.7285, + "step": 2105 + }, + { + "epoch": 2.3079452054794523, + "grad_norm": 2.7859749794006348, + "learning_rate": 1.3119415109666938e-06, + "loss": 1.8026, + "step": 2106 + }, + { + "epoch": 2.309041095890411, + "grad_norm": 2.6862375736236572, + "learning_rate": 1.309910641754671e-06, + "loss": 1.7238, + "step": 2107 + }, + { + "epoch": 2.31013698630137, + "grad_norm": 2.749908685684204, + "learning_rate": 1.3078797725426484e-06, + "loss": 1.676, + "step": 2108 + }, + { + "epoch": 2.311232876712329, + "grad_norm": 2.679384469985962, + "learning_rate": 1.3058489033306257e-06, + "loss": 1.6899, + "step": 2109 + }, + { + "epoch": 2.3123287671232875, + "grad_norm": 2.5621979236602783, + "learning_rate": 1.303818034118603e-06, + "loss": 1.6963, + "step": 2110 + }, + { + "epoch": 2.3123287671232875, + "eval_loss": 1.9453113079071045, + "eval_runtime": 37.2803, + "eval_samples_per_second": 26.824, + "eval_steps_per_second": 1.127, + "step": 2110 + }, + { + "epoch": 2.3134246575342465, + "grad_norm": 2.6183035373687744, + "learning_rate": 1.3017871649065802e-06, + "loss": 1.7272, + "step": 2111 + }, + { + "epoch": 2.3145205479452056, + "grad_norm": 2.5777511596679688, + "learning_rate": 1.2997562956945573e-06, + "loss": 1.7041, + "step": 2112 + }, + { + "epoch": 2.315616438356164, + "grad_norm": 2.784181594848633, + "learning_rate": 1.2977254264825345e-06, + "loss": 1.6835, + "step": 2113 + }, + { + "epoch": 2.316712328767123, + "grad_norm": 2.679591417312622, + "learning_rate": 1.2956945572705118e-06, + "loss": 1.7242, + "step": 2114 + }, + { + "epoch": 2.317808219178082, + "grad_norm": 2.561269521713257, + "learning_rate": 1.293663688058489e-06, + "loss": 1.7455, + "step": 2115 + }, + { + "epoch": 2.3189041095890413, + "grad_norm": 2.5087854862213135, + "learning_rate": 1.2916328188464664e-06, + "loss": 1.6824, + "step": 2116 + }, + { + "epoch": 2.32, + "grad_norm": 2.7247722148895264, + "learning_rate": 1.2896019496344436e-06, + "loss": 1.6268, + "step": 2117 + }, + { + "epoch": 2.321095890410959, + "grad_norm": 2.603994846343994, + "learning_rate": 1.287571080422421e-06, + "loss": 1.6925, + "step": 2118 + }, + { + "epoch": 2.322191780821918, + "grad_norm": 2.5525155067443848, + "learning_rate": 1.2855402112103982e-06, + "loss": 1.7643, + "step": 2119 + }, + { + "epoch": 2.323287671232877, + "grad_norm": 2.5990493297576904, + "learning_rate": 1.2835093419983755e-06, + "loss": 1.6712, + "step": 2120 + }, + { + "epoch": 2.323287671232877, + "eval_loss": 1.9467085599899292, + "eval_runtime": 37.5353, + "eval_samples_per_second": 26.642, + "eval_steps_per_second": 1.119, + "step": 2120 + }, + { + "epoch": 2.3243835616438355, + "grad_norm": 2.764711618423462, + "learning_rate": 1.2814784727863528e-06, + "loss": 1.71, + "step": 2121 + }, + { + "epoch": 2.3254794520547946, + "grad_norm": 2.6009247303009033, + "learning_rate": 1.27944760357433e-06, + "loss": 1.6956, + "step": 2122 + }, + { + "epoch": 2.3265753424657536, + "grad_norm": 2.6900346279144287, + "learning_rate": 1.2774167343623073e-06, + "loss": 1.7529, + "step": 2123 + }, + { + "epoch": 2.327671232876712, + "grad_norm": 2.6602327823638916, + "learning_rate": 1.2753858651502846e-06, + "loss": 1.7588, + "step": 2124 + }, + { + "epoch": 2.328767123287671, + "grad_norm": 2.7727952003479004, + "learning_rate": 1.2733549959382616e-06, + "loss": 1.7499, + "step": 2125 + }, + { + "epoch": 2.3298630136986302, + "grad_norm": 2.703307867050171, + "learning_rate": 1.271324126726239e-06, + "loss": 1.7169, + "step": 2126 + }, + { + "epoch": 2.330958904109589, + "grad_norm": 2.7290568351745605, + "learning_rate": 1.2692932575142162e-06, + "loss": 1.7076, + "step": 2127 + }, + { + "epoch": 2.332054794520548, + "grad_norm": 2.664865255355835, + "learning_rate": 1.2672623883021935e-06, + "loss": 1.6655, + "step": 2128 + }, + { + "epoch": 2.333150684931507, + "grad_norm": 2.6096653938293457, + "learning_rate": 1.2652315190901707e-06, + "loss": 1.6644, + "step": 2129 + }, + { + "epoch": 2.334246575342466, + "grad_norm": 2.7201335430145264, + "learning_rate": 1.263200649878148e-06, + "loss": 1.7924, + "step": 2130 + }, + { + "epoch": 2.334246575342466, + "eval_loss": 1.9422904253005981, + "eval_runtime": 37.5421, + "eval_samples_per_second": 26.637, + "eval_steps_per_second": 1.119, + "step": 2130 + }, + { + "epoch": 2.3353424657534245, + "grad_norm": 2.519702434539795, + "learning_rate": 1.2611697806661253e-06, + "loss": 1.7108, + "step": 2131 + }, + { + "epoch": 2.3364383561643836, + "grad_norm": 2.6535212993621826, + "learning_rate": 1.2591389114541026e-06, + "loss": 1.7243, + "step": 2132 + }, + { + "epoch": 2.3375342465753426, + "grad_norm": 2.651458501815796, + "learning_rate": 1.2571080422420798e-06, + "loss": 1.7158, + "step": 2133 + }, + { + "epoch": 2.338630136986301, + "grad_norm": 2.6745946407318115, + "learning_rate": 1.2550771730300571e-06, + "loss": 1.7014, + "step": 2134 + }, + { + "epoch": 2.33972602739726, + "grad_norm": 2.6088216304779053, + "learning_rate": 1.2530463038180344e-06, + "loss": 1.7724, + "step": 2135 + }, + { + "epoch": 2.3408219178082192, + "grad_norm": 2.703014373779297, + "learning_rate": 1.2510154346060117e-06, + "loss": 1.7104, + "step": 2136 + }, + { + "epoch": 2.3419178082191783, + "grad_norm": 2.57769513130188, + "learning_rate": 1.2489845653939887e-06, + "loss": 1.7259, + "step": 2137 + }, + { + "epoch": 2.343013698630137, + "grad_norm": 2.596695899963379, + "learning_rate": 1.246953696181966e-06, + "loss": 1.7571, + "step": 2138 + }, + { + "epoch": 2.344109589041096, + "grad_norm": 2.6988871097564697, + "learning_rate": 1.2449228269699433e-06, + "loss": 1.8002, + "step": 2139 + }, + { + "epoch": 2.345205479452055, + "grad_norm": 2.6352357864379883, + "learning_rate": 1.2428919577579204e-06, + "loss": 1.7213, + "step": 2140 + }, + { + "epoch": 2.345205479452055, + "eval_loss": 1.9401096105575562, + "eval_runtime": 37.4404, + "eval_samples_per_second": 26.709, + "eval_steps_per_second": 1.122, + "step": 2140 + }, + { + "epoch": 2.3463013698630135, + "grad_norm": 2.7089645862579346, + "learning_rate": 1.2408610885458976e-06, + "loss": 1.6737, + "step": 2141 + }, + { + "epoch": 2.3473972602739726, + "grad_norm": 2.6364030838012695, + "learning_rate": 1.238830219333875e-06, + "loss": 1.7186, + "step": 2142 + }, + { + "epoch": 2.3484931506849316, + "grad_norm": 2.5563573837280273, + "learning_rate": 1.2367993501218522e-06, + "loss": 1.6956, + "step": 2143 + }, + { + "epoch": 2.34958904109589, + "grad_norm": 2.638512372970581, + "learning_rate": 1.2347684809098295e-06, + "loss": 1.7693, + "step": 2144 + }, + { + "epoch": 2.350684931506849, + "grad_norm": 2.733036518096924, + "learning_rate": 1.2327376116978067e-06, + "loss": 1.7327, + "step": 2145 + }, + { + "epoch": 2.3517808219178082, + "grad_norm": 2.6013307571411133, + "learning_rate": 1.230706742485784e-06, + "loss": 1.6539, + "step": 2146 + }, + { + "epoch": 2.3528767123287673, + "grad_norm": 2.7629592418670654, + "learning_rate": 1.2286758732737613e-06, + "loss": 1.666, + "step": 2147 + }, + { + "epoch": 2.353972602739726, + "grad_norm": 2.6855108737945557, + "learning_rate": 1.2266450040617386e-06, + "loss": 1.8044, + "step": 2148 + }, + { + "epoch": 2.355068493150685, + "grad_norm": 2.607609748840332, + "learning_rate": 1.2246141348497158e-06, + "loss": 1.8022, + "step": 2149 + }, + { + "epoch": 2.356164383561644, + "grad_norm": 2.5609383583068848, + "learning_rate": 1.2225832656376931e-06, + "loss": 1.73, + "step": 2150 + }, + { + "epoch": 2.356164383561644, + "eval_loss": 1.9387279748916626, + "eval_runtime": 37.2896, + "eval_samples_per_second": 26.817, + "eval_steps_per_second": 1.126, + "step": 2150 + }, + { + "epoch": 2.357260273972603, + "grad_norm": 2.523988723754883, + "learning_rate": 1.2205523964256704e-06, + "loss": 1.7289, + "step": 2151 + }, + { + "epoch": 2.3583561643835615, + "grad_norm": 2.675995111465454, + "learning_rate": 1.2185215272136477e-06, + "loss": 1.7036, + "step": 2152 + }, + { + "epoch": 2.3594520547945206, + "grad_norm": 2.535367012023926, + "learning_rate": 1.2164906580016247e-06, + "loss": 1.7309, + "step": 2153 + }, + { + "epoch": 2.3605479452054796, + "grad_norm": 2.6444637775421143, + "learning_rate": 1.214459788789602e-06, + "loss": 1.7192, + "step": 2154 + }, + { + "epoch": 2.361643835616438, + "grad_norm": 2.5746796131134033, + "learning_rate": 1.2124289195775793e-06, + "loss": 1.6343, + "step": 2155 + }, + { + "epoch": 2.3627397260273972, + "grad_norm": 2.612849712371826, + "learning_rate": 1.2103980503655566e-06, + "loss": 1.6686, + "step": 2156 + }, + { + "epoch": 2.3638356164383563, + "grad_norm": 2.6836912631988525, + "learning_rate": 1.2083671811535338e-06, + "loss": 1.7173, + "step": 2157 + }, + { + "epoch": 2.364931506849315, + "grad_norm": 2.716177225112915, + "learning_rate": 1.2063363119415111e-06, + "loss": 1.695, + "step": 2158 + }, + { + "epoch": 2.366027397260274, + "grad_norm": 2.6918227672576904, + "learning_rate": 1.2043054427294882e-06, + "loss": 1.684, + "step": 2159 + }, + { + "epoch": 2.367123287671233, + "grad_norm": 2.6265008449554443, + "learning_rate": 1.2022745735174654e-06, + "loss": 1.6959, + "step": 2160 + }, + { + "epoch": 2.367123287671233, + "eval_loss": 1.939719319343567, + "eval_runtime": 37.2557, + "eval_samples_per_second": 26.842, + "eval_steps_per_second": 1.127, + "step": 2160 + }, + { + "epoch": 2.368219178082192, + "grad_norm": 2.638829469680786, + "learning_rate": 1.2002437043054427e-06, + "loss": 1.7362, + "step": 2161 + }, + { + "epoch": 2.3693150684931505, + "grad_norm": 2.6369569301605225, + "learning_rate": 1.19821283509342e-06, + "loss": 1.7765, + "step": 2162 + }, + { + "epoch": 2.3704109589041096, + "grad_norm": 2.6272799968719482, + "learning_rate": 1.1961819658813973e-06, + "loss": 1.7695, + "step": 2163 + }, + { + "epoch": 2.3715068493150686, + "grad_norm": 2.601388931274414, + "learning_rate": 1.1941510966693746e-06, + "loss": 1.7259, + "step": 2164 + }, + { + "epoch": 2.3726027397260276, + "grad_norm": 2.5515146255493164, + "learning_rate": 1.1921202274573518e-06, + "loss": 1.7048, + "step": 2165 + }, + { + "epoch": 2.3736986301369862, + "grad_norm": 2.6139824390411377, + "learning_rate": 1.190089358245329e-06, + "loss": 1.6951, + "step": 2166 + }, + { + "epoch": 2.3747945205479453, + "grad_norm": 2.7428653240203857, + "learning_rate": 1.1880584890333064e-06, + "loss": 1.7771, + "step": 2167 + }, + { + "epoch": 2.3758904109589043, + "grad_norm": 2.690804958343506, + "learning_rate": 1.1860276198212837e-06, + "loss": 1.7073, + "step": 2168 + }, + { + "epoch": 2.376986301369863, + "grad_norm": 2.64790415763855, + "learning_rate": 1.183996750609261e-06, + "loss": 1.7344, + "step": 2169 + }, + { + "epoch": 2.378082191780822, + "grad_norm": 2.555832624435425, + "learning_rate": 1.1819658813972382e-06, + "loss": 1.7452, + "step": 2170 + }, + { + "epoch": 2.378082191780822, + "eval_loss": 1.9401137828826904, + "eval_runtime": 37.2362, + "eval_samples_per_second": 26.856, + "eval_steps_per_second": 1.128, + "step": 2170 + }, + { + "epoch": 2.379178082191781, + "grad_norm": 2.604097366333008, + "learning_rate": 1.1799350121852153e-06, + "loss": 1.7176, + "step": 2171 + }, + { + "epoch": 2.3802739726027395, + "grad_norm": 2.579312324523926, + "learning_rate": 1.1779041429731925e-06, + "loss": 1.7588, + "step": 2172 + }, + { + "epoch": 2.3813698630136986, + "grad_norm": 2.555727481842041, + "learning_rate": 1.1758732737611698e-06, + "loss": 1.7612, + "step": 2173 + }, + { + "epoch": 2.3824657534246576, + "grad_norm": 2.6929352283477783, + "learning_rate": 1.173842404549147e-06, + "loss": 1.6707, + "step": 2174 + }, + { + "epoch": 2.383561643835616, + "grad_norm": 2.6732795238494873, + "learning_rate": 1.1718115353371244e-06, + "loss": 1.7743, + "step": 2175 + }, + { + "epoch": 2.3846575342465752, + "grad_norm": 2.673490047454834, + "learning_rate": 1.1697806661251016e-06, + "loss": 1.6968, + "step": 2176 + }, + { + "epoch": 2.3857534246575343, + "grad_norm": 2.657837152481079, + "learning_rate": 1.167749796913079e-06, + "loss": 1.7633, + "step": 2177 + }, + { + "epoch": 2.3868493150684933, + "grad_norm": 2.6677606105804443, + "learning_rate": 1.1657189277010562e-06, + "loss": 1.7646, + "step": 2178 + }, + { + "epoch": 2.387945205479452, + "grad_norm": 2.63075852394104, + "learning_rate": 1.1636880584890335e-06, + "loss": 1.7019, + "step": 2179 + }, + { + "epoch": 2.389041095890411, + "grad_norm": 2.874946117401123, + "learning_rate": 1.1616571892770108e-06, + "loss": 1.691, + "step": 2180 + }, + { + "epoch": 2.389041095890411, + "eval_loss": 1.9399209022521973, + "eval_runtime": 37.1991, + "eval_samples_per_second": 26.882, + "eval_steps_per_second": 1.129, + "step": 2180 + }, + { + "epoch": 2.39013698630137, + "grad_norm": 2.734825611114502, + "learning_rate": 1.159626320064988e-06, + "loss": 1.7551, + "step": 2181 + }, + { + "epoch": 2.391232876712329, + "grad_norm": 2.5722298622131348, + "learning_rate": 1.1575954508529653e-06, + "loss": 1.7279, + "step": 2182 + }, + { + "epoch": 2.3923287671232876, + "grad_norm": 2.6449334621429443, + "learning_rate": 1.1555645816409426e-06, + "loss": 1.7591, + "step": 2183 + }, + { + "epoch": 2.3934246575342466, + "grad_norm": 2.774374485015869, + "learning_rate": 1.1535337124289196e-06, + "loss": 1.7243, + "step": 2184 + }, + { + "epoch": 2.3945205479452056, + "grad_norm": 2.554065227508545, + "learning_rate": 1.151502843216897e-06, + "loss": 1.6767, + "step": 2185 + }, + { + "epoch": 2.3956164383561642, + "grad_norm": 2.5883984565734863, + "learning_rate": 1.1494719740048742e-06, + "loss": 1.6416, + "step": 2186 + }, + { + "epoch": 2.3967123287671233, + "grad_norm": 2.5799498558044434, + "learning_rate": 1.1474411047928515e-06, + "loss": 1.7162, + "step": 2187 + }, + { + "epoch": 2.3978082191780823, + "grad_norm": 2.64442777633667, + "learning_rate": 1.1454102355808287e-06, + "loss": 1.709, + "step": 2188 + }, + { + "epoch": 2.398904109589041, + "grad_norm": 2.6813411712646484, + "learning_rate": 1.143379366368806e-06, + "loss": 1.6903, + "step": 2189 + }, + { + "epoch": 2.4, + "grad_norm": 2.6276469230651855, + "learning_rate": 1.141348497156783e-06, + "loss": 1.6734, + "step": 2190 + }, + { + "epoch": 2.4, + "eval_loss": 1.9423893690109253, + "eval_runtime": 37.1963, + "eval_samples_per_second": 26.884, + "eval_steps_per_second": 1.129, + "step": 2190 + }, + { + "epoch": 2.401095890410959, + "grad_norm": 2.6805641651153564, + "learning_rate": 1.1393176279447604e-06, + "loss": 1.8038, + "step": 2191 + }, + { + "epoch": 2.402191780821918, + "grad_norm": 2.813957691192627, + "learning_rate": 1.1372867587327376e-06, + "loss": 1.7895, + "step": 2192 + }, + { + "epoch": 2.4032876712328766, + "grad_norm": 2.505695343017578, + "learning_rate": 1.135255889520715e-06, + "loss": 1.6398, + "step": 2193 + }, + { + "epoch": 2.4043835616438356, + "grad_norm": 2.7426211833953857, + "learning_rate": 1.1332250203086922e-06, + "loss": 1.7472, + "step": 2194 + }, + { + "epoch": 2.4054794520547946, + "grad_norm": 2.627737522125244, + "learning_rate": 1.1311941510966695e-06, + "loss": 1.7272, + "step": 2195 + }, + { + "epoch": 2.4065753424657537, + "grad_norm": 2.615032911300659, + "learning_rate": 1.1291632818846467e-06, + "loss": 1.7004, + "step": 2196 + }, + { + "epoch": 2.4076712328767123, + "grad_norm": 2.5147581100463867, + "learning_rate": 1.127132412672624e-06, + "loss": 1.6894, + "step": 2197 + }, + { + "epoch": 2.4087671232876713, + "grad_norm": 2.7482388019561768, + "learning_rate": 1.1251015434606013e-06, + "loss": 1.6738, + "step": 2198 + }, + { + "epoch": 2.4098630136986303, + "grad_norm": 2.7309365272521973, + "learning_rate": 1.1230706742485786e-06, + "loss": 1.7222, + "step": 2199 + }, + { + "epoch": 2.410958904109589, + "grad_norm": 2.8322935104370117, + "learning_rate": 1.1210398050365558e-06, + "loss": 1.7762, + "step": 2200 + }, + { + "epoch": 2.410958904109589, + "eval_loss": 1.9395875930786133, + "eval_runtime": 37.2409, + "eval_samples_per_second": 26.852, + "eval_steps_per_second": 1.128, + "step": 2200 + }, + { + "epoch": 2.412054794520548, + "grad_norm": 2.7829575538635254, + "learning_rate": 1.1190089358245331e-06, + "loss": 1.7092, + "step": 2201 + }, + { + "epoch": 2.413150684931507, + "grad_norm": 2.518369197845459, + "learning_rate": 1.1169780666125102e-06, + "loss": 1.655, + "step": 2202 + }, + { + "epoch": 2.4142465753424656, + "grad_norm": 2.715089797973633, + "learning_rate": 1.1149471974004875e-06, + "loss": 1.7461, + "step": 2203 + }, + { + "epoch": 2.4153424657534246, + "grad_norm": 2.7772066593170166, + "learning_rate": 1.1129163281884647e-06, + "loss": 1.6931, + "step": 2204 + }, + { + "epoch": 2.4164383561643836, + "grad_norm": 2.813166379928589, + "learning_rate": 1.110885458976442e-06, + "loss": 1.7514, + "step": 2205 + }, + { + "epoch": 2.417534246575342, + "grad_norm": 2.6345536708831787, + "learning_rate": 1.1088545897644193e-06, + "loss": 1.6865, + "step": 2206 + }, + { + "epoch": 2.4186301369863012, + "grad_norm": 2.6846344470977783, + "learning_rate": 1.1068237205523966e-06, + "loss": 1.7564, + "step": 2207 + }, + { + "epoch": 2.4197260273972603, + "grad_norm": 2.6230814456939697, + "learning_rate": 1.1047928513403736e-06, + "loss": 1.7499, + "step": 2208 + }, + { + "epoch": 2.4208219178082193, + "grad_norm": 2.6820080280303955, + "learning_rate": 1.102761982128351e-06, + "loss": 1.7494, + "step": 2209 + }, + { + "epoch": 2.421917808219178, + "grad_norm": 2.572378396987915, + "learning_rate": 1.1007311129163282e-06, + "loss": 1.6585, + "step": 2210 + }, + { + "epoch": 2.421917808219178, + "eval_loss": 1.9389609098434448, + "eval_runtime": 37.2665, + "eval_samples_per_second": 26.834, + "eval_steps_per_second": 1.127, + "step": 2210 + }, + { + "epoch": 2.423013698630137, + "grad_norm": 2.7218871116638184, + "learning_rate": 1.0987002437043055e-06, + "loss": 1.7187, + "step": 2211 + }, + { + "epoch": 2.424109589041096, + "grad_norm": 2.7090065479278564, + "learning_rate": 1.0966693744922827e-06, + "loss": 1.6983, + "step": 2212 + }, + { + "epoch": 2.425205479452055, + "grad_norm": 2.6419243812561035, + "learning_rate": 1.09463850528026e-06, + "loss": 1.6485, + "step": 2213 + }, + { + "epoch": 2.4263013698630136, + "grad_norm": 2.635533571243286, + "learning_rate": 1.0926076360682373e-06, + "loss": 1.7023, + "step": 2214 + }, + { + "epoch": 2.4273972602739726, + "grad_norm": 2.644904375076294, + "learning_rate": 1.0905767668562146e-06, + "loss": 1.6774, + "step": 2215 + }, + { + "epoch": 2.4284931506849317, + "grad_norm": 2.6821508407592773, + "learning_rate": 1.0885458976441918e-06, + "loss": 1.6995, + "step": 2216 + }, + { + "epoch": 2.4295890410958902, + "grad_norm": 2.674916982650757, + "learning_rate": 1.0865150284321691e-06, + "loss": 1.7888, + "step": 2217 + }, + { + "epoch": 2.4306849315068493, + "grad_norm": 2.6966474056243896, + "learning_rate": 1.0844841592201464e-06, + "loss": 1.7303, + "step": 2218 + }, + { + "epoch": 2.4317808219178083, + "grad_norm": 2.6421585083007812, + "learning_rate": 1.0824532900081237e-06, + "loss": 1.7497, + "step": 2219 + }, + { + "epoch": 2.432876712328767, + "grad_norm": 2.8147826194763184, + "learning_rate": 1.080422420796101e-06, + "loss": 1.7088, + "step": 2220 + }, + { + "epoch": 2.432876712328767, + "eval_loss": 1.9437190294265747, + "eval_runtime": 37.5268, + "eval_samples_per_second": 26.648, + "eval_steps_per_second": 1.119, + "step": 2220 + }, + { + "epoch": 2.433972602739726, + "grad_norm": 2.6919329166412354, + "learning_rate": 1.078391551584078e-06, + "loss": 1.7168, + "step": 2221 + }, + { + "epoch": 2.435068493150685, + "grad_norm": 2.6179897785186768, + "learning_rate": 1.0763606823720553e-06, + "loss": 1.7477, + "step": 2222 + }, + { + "epoch": 2.436164383561644, + "grad_norm": 2.7324624061584473, + "learning_rate": 1.0743298131600326e-06, + "loss": 1.7921, + "step": 2223 + }, + { + "epoch": 2.4372602739726026, + "grad_norm": 2.6431500911712646, + "learning_rate": 1.0722989439480098e-06, + "loss": 1.7432, + "step": 2224 + }, + { + "epoch": 2.4383561643835616, + "grad_norm": 2.593528985977173, + "learning_rate": 1.070268074735987e-06, + "loss": 1.7097, + "step": 2225 + }, + { + "epoch": 2.4394520547945207, + "grad_norm": 2.6277310848236084, + "learning_rate": 1.0682372055239644e-06, + "loss": 1.6911, + "step": 2226 + }, + { + "epoch": 2.4405479452054797, + "grad_norm": 2.5935580730438232, + "learning_rate": 1.0662063363119414e-06, + "loss": 1.6933, + "step": 2227 + }, + { + "epoch": 2.4416438356164383, + "grad_norm": 2.672347068786621, + "learning_rate": 1.0641754670999187e-06, + "loss": 1.6826, + "step": 2228 + }, + { + "epoch": 2.4427397260273973, + "grad_norm": 2.7214746475219727, + "learning_rate": 1.062144597887896e-06, + "loss": 1.7152, + "step": 2229 + }, + { + "epoch": 2.4438356164383563, + "grad_norm": 2.7052741050720215, + "learning_rate": 1.0601137286758733e-06, + "loss": 1.7119, + "step": 2230 + }, + { + "epoch": 2.4438356164383563, + "eval_loss": 1.9452862739562988, + "eval_runtime": 37.6622, + "eval_samples_per_second": 26.552, + "eval_steps_per_second": 1.115, + "step": 2230 + }, + { + "epoch": 2.444931506849315, + "grad_norm": 2.7374205589294434, + "learning_rate": 1.0580828594638505e-06, + "loss": 1.7904, + "step": 2231 + }, + { + "epoch": 2.446027397260274, + "grad_norm": 2.6822078227996826, + "learning_rate": 1.0560519902518278e-06, + "loss": 1.6238, + "step": 2232 + }, + { + "epoch": 2.447123287671233, + "grad_norm": 2.6262941360473633, + "learning_rate": 1.054021121039805e-06, + "loss": 1.7501, + "step": 2233 + }, + { + "epoch": 2.4482191780821916, + "grad_norm": 2.689777374267578, + "learning_rate": 1.0519902518277824e-06, + "loss": 1.6604, + "step": 2234 + }, + { + "epoch": 2.4493150684931506, + "grad_norm": 2.6919682025909424, + "learning_rate": 1.0499593826157596e-06, + "loss": 1.6493, + "step": 2235 + }, + { + "epoch": 2.4504109589041096, + "grad_norm": 2.732959270477295, + "learning_rate": 1.047928513403737e-06, + "loss": 1.7429, + "step": 2236 + }, + { + "epoch": 2.4515068493150687, + "grad_norm": 2.6201255321502686, + "learning_rate": 1.0458976441917142e-06, + "loss": 1.7131, + "step": 2237 + }, + { + "epoch": 2.4526027397260273, + "grad_norm": 2.655271291732788, + "learning_rate": 1.0438667749796915e-06, + "loss": 1.634, + "step": 2238 + }, + { + "epoch": 2.4536986301369863, + "grad_norm": 2.6041507720947266, + "learning_rate": 1.0418359057676688e-06, + "loss": 1.6384, + "step": 2239 + }, + { + "epoch": 2.4547945205479453, + "grad_norm": 2.582871198654175, + "learning_rate": 1.0398050365556458e-06, + "loss": 1.6903, + "step": 2240 + }, + { + "epoch": 2.4547945205479453, + "eval_loss": 1.9449145793914795, + "eval_runtime": 37.4116, + "eval_samples_per_second": 26.73, + "eval_steps_per_second": 1.123, + "step": 2240 + }, + { + "epoch": 2.455890410958904, + "grad_norm": 2.647946357727051, + "learning_rate": 1.037774167343623e-06, + "loss": 1.752, + "step": 2241 + }, + { + "epoch": 2.456986301369863, + "grad_norm": 2.6174609661102295, + "learning_rate": 1.0357432981316004e-06, + "loss": 1.6745, + "step": 2242 + }, + { + "epoch": 2.458082191780822, + "grad_norm": 2.7191693782806396, + "learning_rate": 1.0337124289195776e-06, + "loss": 1.7291, + "step": 2243 + }, + { + "epoch": 2.459178082191781, + "grad_norm": 2.6612565517425537, + "learning_rate": 1.031681559707555e-06, + "loss": 1.7523, + "step": 2244 + }, + { + "epoch": 2.4602739726027396, + "grad_norm": 2.617875337600708, + "learning_rate": 1.0296506904955322e-06, + "loss": 1.7597, + "step": 2245 + }, + { + "epoch": 2.4613698630136986, + "grad_norm": 2.730677843093872, + "learning_rate": 1.0276198212835095e-06, + "loss": 1.7146, + "step": 2246 + }, + { + "epoch": 2.4624657534246577, + "grad_norm": 2.6297593116760254, + "learning_rate": 1.0255889520714867e-06, + "loss": 1.7183, + "step": 2247 + }, + { + "epoch": 2.4635616438356163, + "grad_norm": 2.600847005844116, + "learning_rate": 1.023558082859464e-06, + "loss": 1.7086, + "step": 2248 + }, + { + "epoch": 2.4646575342465753, + "grad_norm": 2.773501396179199, + "learning_rate": 1.0215272136474413e-06, + "loss": 1.6708, + "step": 2249 + }, + { + "epoch": 2.4657534246575343, + "grad_norm": 2.6344449520111084, + "learning_rate": 1.0194963444354186e-06, + "loss": 1.7818, + "step": 2250 + }, + { + "epoch": 2.4657534246575343, + "eval_loss": 1.940114974975586, + "eval_runtime": 37.3629, + "eval_samples_per_second": 26.764, + "eval_steps_per_second": 1.124, + "step": 2250 + }, + { + "epoch": 2.466849315068493, + "grad_norm": 2.6345012187957764, + "learning_rate": 1.0174654752233958e-06, + "loss": 1.6525, + "step": 2251 + }, + { + "epoch": 2.467945205479452, + "grad_norm": 2.6919827461242676, + "learning_rate": 1.015434606011373e-06, + "loss": 1.7457, + "step": 2252 + }, + { + "epoch": 2.469041095890411, + "grad_norm": 2.6676719188690186, + "learning_rate": 1.0134037367993502e-06, + "loss": 1.6934, + "step": 2253 + }, + { + "epoch": 2.47013698630137, + "grad_norm": 2.623911142349243, + "learning_rate": 1.0113728675873275e-06, + "loss": 1.7089, + "step": 2254 + }, + { + "epoch": 2.4712328767123286, + "grad_norm": 2.6342320442199707, + "learning_rate": 1.0093419983753047e-06, + "loss": 1.7123, + "step": 2255 + }, + { + "epoch": 2.4723287671232876, + "grad_norm": 2.8066930770874023, + "learning_rate": 1.007311129163282e-06, + "loss": 1.7095, + "step": 2256 + }, + { + "epoch": 2.4734246575342467, + "grad_norm": 2.6137008666992188, + "learning_rate": 1.0052802599512593e-06, + "loss": 1.6602, + "step": 2257 + }, + { + "epoch": 2.4745205479452057, + "grad_norm": 2.6158881187438965, + "learning_rate": 1.0032493907392364e-06, + "loss": 1.6659, + "step": 2258 + }, + { + "epoch": 2.4756164383561643, + "grad_norm": 2.7409005165100098, + "learning_rate": 1.0012185215272136e-06, + "loss": 1.7194, + "step": 2259 + }, + { + "epoch": 2.4767123287671233, + "grad_norm": 2.783179759979248, + "learning_rate": 9.99187652315191e-07, + "loss": 1.658, + "step": 2260 + }, + { + "epoch": 2.4767123287671233, + "eval_loss": 1.9401936531066895, + "eval_runtime": 37.2693, + "eval_samples_per_second": 26.832, + "eval_steps_per_second": 1.127, + "step": 2260 + }, + { + "epoch": 2.4778082191780824, + "grad_norm": 2.7862110137939453, + "learning_rate": 9.971567831031682e-07, + "loss": 1.7535, + "step": 2261 + }, + { + "epoch": 2.478904109589041, + "grad_norm": 2.623033046722412, + "learning_rate": 9.951259138911455e-07, + "loss": 1.7285, + "step": 2262 + }, + { + "epoch": 2.48, + "grad_norm": 2.5325446128845215, + "learning_rate": 9.930950446791227e-07, + "loss": 1.6486, + "step": 2263 + }, + { + "epoch": 2.481095890410959, + "grad_norm": 2.6013123989105225, + "learning_rate": 9.910641754671e-07, + "loss": 1.7677, + "step": 2264 + }, + { + "epoch": 2.4821917808219176, + "grad_norm": 2.706526279449463, + "learning_rate": 9.890333062550773e-07, + "loss": 1.81, + "step": 2265 + }, + { + "epoch": 2.4832876712328766, + "grad_norm": 2.6389000415802, + "learning_rate": 9.870024370430546e-07, + "loss": 1.7104, + "step": 2266 + }, + { + "epoch": 2.4843835616438357, + "grad_norm": 2.8673291206359863, + "learning_rate": 9.849715678310318e-07, + "loss": 1.7439, + "step": 2267 + }, + { + "epoch": 2.4854794520547947, + "grad_norm": 2.574463367462158, + "learning_rate": 9.829406986190091e-07, + "loss": 1.7134, + "step": 2268 + }, + { + "epoch": 2.4865753424657533, + "grad_norm": 2.6605708599090576, + "learning_rate": 9.809098294069864e-07, + "loss": 1.7266, + "step": 2269 + }, + { + "epoch": 2.4876712328767123, + "grad_norm": 2.6304521560668945, + "learning_rate": 9.788789601949637e-07, + "loss": 1.6388, + "step": 2270 + }, + { + "epoch": 2.4876712328767123, + "eval_loss": 1.9415092468261719, + "eval_runtime": 37.2274, + "eval_samples_per_second": 26.862, + "eval_steps_per_second": 1.128, + "step": 2270 + }, + { + "epoch": 2.4887671232876714, + "grad_norm": 2.6577560901641846, + "learning_rate": 9.768480909829407e-07, + "loss": 1.7252, + "step": 2271 + }, + { + "epoch": 2.48986301369863, + "grad_norm": 2.801530122756958, + "learning_rate": 9.74817221770918e-07, + "loss": 1.7774, + "step": 2272 + }, + { + "epoch": 2.490958904109589, + "grad_norm": 2.761235237121582, + "learning_rate": 9.727863525588953e-07, + "loss": 1.6889, + "step": 2273 + }, + { + "epoch": 2.492054794520548, + "grad_norm": 2.719083547592163, + "learning_rate": 9.707554833468726e-07, + "loss": 1.7231, + "step": 2274 + }, + { + "epoch": 2.493150684931507, + "grad_norm": 2.5667307376861572, + "learning_rate": 9.687246141348498e-07, + "loss": 1.7301, + "step": 2275 + }, + { + "epoch": 2.4942465753424656, + "grad_norm": 2.6383678913116455, + "learning_rate": 9.666937449228271e-07, + "loss": 1.6999, + "step": 2276 + }, + { + "epoch": 2.4953424657534247, + "grad_norm": 2.7380318641662598, + "learning_rate": 9.646628757108042e-07, + "loss": 1.7202, + "step": 2277 + }, + { + "epoch": 2.4964383561643837, + "grad_norm": 2.715460777282715, + "learning_rate": 9.626320064987814e-07, + "loss": 1.6626, + "step": 2278 + }, + { + "epoch": 2.4975342465753423, + "grad_norm": 2.5860559940338135, + "learning_rate": 9.606011372867587e-07, + "loss": 1.7345, + "step": 2279 + }, + { + "epoch": 2.4986301369863013, + "grad_norm": 2.614272356033325, + "learning_rate": 9.58570268074736e-07, + "loss": 1.6728, + "step": 2280 + }, + { + "epoch": 2.4986301369863013, + "eval_loss": 1.9393996000289917, + "eval_runtime": 37.2508, + "eval_samples_per_second": 26.845, + "eval_steps_per_second": 1.127, + "step": 2280 + }, + { + "epoch": 2.4997260273972604, + "grad_norm": 2.6709139347076416, + "learning_rate": 9.565393988627133e-07, + "loss": 1.7094, + "step": 2281 + }, + { + "epoch": 2.500821917808219, + "grad_norm": 2.468263626098633, + "learning_rate": 9.545085296506906e-07, + "loss": 1.7044, + "step": 2282 + }, + { + "epoch": 2.501917808219178, + "grad_norm": 2.646242141723633, + "learning_rate": 9.524776604386677e-07, + "loss": 1.7144, + "step": 2283 + }, + { + "epoch": 2.503013698630137, + "grad_norm": 2.6524763107299805, + "learning_rate": 9.50446791226645e-07, + "loss": 1.7221, + "step": 2284 + }, + { + "epoch": 2.504109589041096, + "grad_norm": 2.681304693222046, + "learning_rate": 9.484159220146223e-07, + "loss": 1.6446, + "step": 2285 + }, + { + "epoch": 2.5052054794520546, + "grad_norm": 2.5855886936187744, + "learning_rate": 9.463850528025995e-07, + "loss": 1.757, + "step": 2286 + }, + { + "epoch": 2.5063013698630137, + "grad_norm": 2.5410566329956055, + "learning_rate": 9.443541835905768e-07, + "loss": 1.7487, + "step": 2287 + }, + { + "epoch": 2.5073972602739727, + "grad_norm": 2.726349115371704, + "learning_rate": 9.423233143785541e-07, + "loss": 1.7841, + "step": 2288 + }, + { + "epoch": 2.5084931506849317, + "grad_norm": 2.7225544452667236, + "learning_rate": 9.402924451665313e-07, + "loss": 1.7921, + "step": 2289 + }, + { + "epoch": 2.5095890410958903, + "grad_norm": 2.7081549167633057, + "learning_rate": 9.382615759545085e-07, + "loss": 1.6726, + "step": 2290 + }, + { + "epoch": 2.5095890410958903, + "eval_loss": 1.9398577213287354, + "eval_runtime": 37.2122, + "eval_samples_per_second": 26.873, + "eval_steps_per_second": 1.129, + "step": 2290 + }, + { + "epoch": 2.5106849315068493, + "grad_norm": 2.6577272415161133, + "learning_rate": 9.362307067424858e-07, + "loss": 1.7571, + "step": 2291 + }, + { + "epoch": 2.5117808219178084, + "grad_norm": 2.659438133239746, + "learning_rate": 9.341998375304631e-07, + "loss": 1.7113, + "step": 2292 + }, + { + "epoch": 2.512876712328767, + "grad_norm": 2.6679470539093018, + "learning_rate": 9.321689683184404e-07, + "loss": 1.7718, + "step": 2293 + }, + { + "epoch": 2.513972602739726, + "grad_norm": 2.703716993331909, + "learning_rate": 9.301380991064176e-07, + "loss": 1.6617, + "step": 2294 + }, + { + "epoch": 2.515068493150685, + "grad_norm": 2.66267728805542, + "learning_rate": 9.281072298943949e-07, + "loss": 1.6479, + "step": 2295 + }, + { + "epoch": 2.5161643835616436, + "grad_norm": 2.7160699367523193, + "learning_rate": 9.260763606823721e-07, + "loss": 1.6931, + "step": 2296 + }, + { + "epoch": 2.5172602739726027, + "grad_norm": 2.7481086254119873, + "learning_rate": 9.240454914703494e-07, + "loss": 1.73, + "step": 2297 + }, + { + "epoch": 2.5183561643835617, + "grad_norm": 2.702737331390381, + "learning_rate": 9.220146222583266e-07, + "loss": 1.7414, + "step": 2298 + }, + { + "epoch": 2.5194520547945203, + "grad_norm": 2.78155517578125, + "learning_rate": 9.199837530463039e-07, + "loss": 1.684, + "step": 2299 + }, + { + "epoch": 2.5205479452054793, + "grad_norm": 2.5439021587371826, + "learning_rate": 9.179528838342812e-07, + "loss": 1.7116, + "step": 2300 + }, + { + "epoch": 2.5205479452054793, + "eval_loss": 1.9369380474090576, + "eval_runtime": 37.2139, + "eval_samples_per_second": 26.872, + "eval_steps_per_second": 1.129, + "step": 2300 + }, + { + "epoch": 2.5216438356164383, + "grad_norm": 2.5379269123077393, + "learning_rate": 9.159220146222585e-07, + "loss": 1.6888, + "step": 2301 + }, + { + "epoch": 2.5227397260273974, + "grad_norm": 2.6183183193206787, + "learning_rate": 9.138911454102356e-07, + "loss": 1.7297, + "step": 2302 + }, + { + "epoch": 2.5238356164383564, + "grad_norm": 2.6776952743530273, + "learning_rate": 9.118602761982129e-07, + "loss": 1.5998, + "step": 2303 + }, + { + "epoch": 2.524931506849315, + "grad_norm": 2.71006441116333, + "learning_rate": 9.098294069861902e-07, + "loss": 1.7272, + "step": 2304 + }, + { + "epoch": 2.526027397260274, + "grad_norm": 2.6362080574035645, + "learning_rate": 9.077985377741675e-07, + "loss": 1.7147, + "step": 2305 + }, + { + "epoch": 2.527123287671233, + "grad_norm": 2.629568576812744, + "learning_rate": 9.057676685621447e-07, + "loss": 1.6762, + "step": 2306 + }, + { + "epoch": 2.5282191780821917, + "grad_norm": 2.4955074787139893, + "learning_rate": 9.03736799350122e-07, + "loss": 1.642, + "step": 2307 + }, + { + "epoch": 2.5293150684931507, + "grad_norm": 2.610558271408081, + "learning_rate": 9.017059301380991e-07, + "loss": 1.7773, + "step": 2308 + }, + { + "epoch": 2.5304109589041097, + "grad_norm": 2.5644302368164062, + "learning_rate": 8.996750609260764e-07, + "loss": 1.7524, + "step": 2309 + }, + { + "epoch": 2.5315068493150683, + "grad_norm": 2.6172549724578857, + "learning_rate": 8.976441917140536e-07, + "loss": 1.7865, + "step": 2310 + }, + { + "epoch": 2.5315068493150683, + "eval_loss": 1.9368717670440674, + "eval_runtime": 37.2323, + "eval_samples_per_second": 26.858, + "eval_steps_per_second": 1.128, + "step": 2310 + }, + { + "epoch": 2.5326027397260273, + "grad_norm": 2.6722280979156494, + "learning_rate": 8.956133225020309e-07, + "loss": 1.7449, + "step": 2311 + }, + { + "epoch": 2.5336986301369864, + "grad_norm": 2.6534717082977295, + "learning_rate": 8.935824532900082e-07, + "loss": 1.7844, + "step": 2312 + }, + { + "epoch": 2.534794520547945, + "grad_norm": 2.566446542739868, + "learning_rate": 8.915515840779855e-07, + "loss": 1.6826, + "step": 2313 + }, + { + "epoch": 2.535890410958904, + "grad_norm": 2.6202712059020996, + "learning_rate": 8.895207148659626e-07, + "loss": 1.6535, + "step": 2314 + }, + { + "epoch": 2.536986301369863, + "grad_norm": 2.6147568225860596, + "learning_rate": 8.874898456539399e-07, + "loss": 1.7338, + "step": 2315 + }, + { + "epoch": 2.538082191780822, + "grad_norm": 2.6190552711486816, + "learning_rate": 8.854589764419172e-07, + "loss": 1.7297, + "step": 2316 + }, + { + "epoch": 2.539178082191781, + "grad_norm": 2.7838358879089355, + "learning_rate": 8.834281072298945e-07, + "loss": 1.7372, + "step": 2317 + }, + { + "epoch": 2.5402739726027397, + "grad_norm": 2.6310222148895264, + "learning_rate": 8.813972380178717e-07, + "loss": 1.6706, + "step": 2318 + }, + { + "epoch": 2.5413698630136987, + "grad_norm": 2.7847073078155518, + "learning_rate": 8.79366368805849e-07, + "loss": 1.6967, + "step": 2319 + }, + { + "epoch": 2.5424657534246577, + "grad_norm": 2.6233162879943848, + "learning_rate": 8.773354995938262e-07, + "loss": 1.7215, + "step": 2320 + }, + { + "epoch": 2.5424657534246577, + "eval_loss": 1.9374815225601196, + "eval_runtime": 37.5549, + "eval_samples_per_second": 26.628, + "eval_steps_per_second": 1.118, + "step": 2320 + }, + { + "epoch": 2.5435616438356163, + "grad_norm": 2.6680665016174316, + "learning_rate": 8.753046303818035e-07, + "loss": 1.6705, + "step": 2321 + }, + { + "epoch": 2.5446575342465754, + "grad_norm": 2.570737600326538, + "learning_rate": 8.732737611697807e-07, + "loss": 1.6617, + "step": 2322 + }, + { + "epoch": 2.5457534246575344, + "grad_norm": 2.6836602687835693, + "learning_rate": 8.71242891957758e-07, + "loss": 1.6989, + "step": 2323 + }, + { + "epoch": 2.546849315068493, + "grad_norm": 2.637993812561035, + "learning_rate": 8.692120227457353e-07, + "loss": 1.7281, + "step": 2324 + }, + { + "epoch": 2.547945205479452, + "grad_norm": 2.61149001121521, + "learning_rate": 8.671811535337126e-07, + "loss": 1.7459, + "step": 2325 + }, + { + "epoch": 2.549041095890411, + "grad_norm": 2.6190171241760254, + "learning_rate": 8.651502843216898e-07, + "loss": 1.6925, + "step": 2326 + }, + { + "epoch": 2.5501369863013696, + "grad_norm": 2.6211154460906982, + "learning_rate": 8.63119415109667e-07, + "loss": 1.7999, + "step": 2327 + }, + { + "epoch": 2.5512328767123287, + "grad_norm": 2.6447153091430664, + "learning_rate": 8.610885458976443e-07, + "loss": 1.7356, + "step": 2328 + }, + { + "epoch": 2.5523287671232877, + "grad_norm": 2.6078338623046875, + "learning_rate": 8.590576766856216e-07, + "loss": 1.7281, + "step": 2329 + }, + { + "epoch": 2.5534246575342463, + "grad_norm": 2.5882375240325928, + "learning_rate": 8.570268074735988e-07, + "loss": 1.6858, + "step": 2330 + }, + { + "epoch": 2.5534246575342463, + "eval_loss": 1.9356836080551147, + "eval_runtime": 37.5283, + "eval_samples_per_second": 26.647, + "eval_steps_per_second": 1.119, + "step": 2330 + }, + { + "epoch": 2.5545205479452053, + "grad_norm": 2.6218528747558594, + "learning_rate": 8.549959382615761e-07, + "loss": 1.6712, + "step": 2331 + }, + { + "epoch": 2.5556164383561644, + "grad_norm": 2.60943341255188, + "learning_rate": 8.529650690495534e-07, + "loss": 1.6924, + "step": 2332 + }, + { + "epoch": 2.5567123287671234, + "grad_norm": 2.8169174194335938, + "learning_rate": 8.509341998375305e-07, + "loss": 1.715, + "step": 2333 + }, + { + "epoch": 2.5578082191780824, + "grad_norm": 2.601630210876465, + "learning_rate": 8.489033306255077e-07, + "loss": 1.6473, + "step": 2334 + }, + { + "epoch": 2.558904109589041, + "grad_norm": 2.622537612915039, + "learning_rate": 8.46872461413485e-07, + "loss": 1.6131, + "step": 2335 + }, + { + "epoch": 2.56, + "grad_norm": 2.57839035987854, + "learning_rate": 8.448415922014623e-07, + "loss": 1.6713, + "step": 2336 + }, + { + "epoch": 2.561095890410959, + "grad_norm": 2.7201271057128906, + "learning_rate": 8.428107229894396e-07, + "loss": 1.6825, + "step": 2337 + }, + { + "epoch": 2.5621917808219177, + "grad_norm": 2.57549786567688, + "learning_rate": 8.407798537774168e-07, + "loss": 1.7242, + "step": 2338 + }, + { + "epoch": 2.5632876712328767, + "grad_norm": 2.8006980419158936, + "learning_rate": 8.38748984565394e-07, + "loss": 1.7053, + "step": 2339 + }, + { + "epoch": 2.5643835616438357, + "grad_norm": 2.7859253883361816, + "learning_rate": 8.367181153533713e-07, + "loss": 1.6751, + "step": 2340 + }, + { + "epoch": 2.5643835616438357, + "eval_loss": 1.9371570348739624, + "eval_runtime": 37.3729, + "eval_samples_per_second": 26.757, + "eval_steps_per_second": 1.124, + "step": 2340 + }, + { + "epoch": 2.5654794520547943, + "grad_norm": 2.664045572280884, + "learning_rate": 8.346872461413486e-07, + "loss": 1.6977, + "step": 2341 + }, + { + "epoch": 2.5665753424657534, + "grad_norm": 2.7048933506011963, + "learning_rate": 8.326563769293258e-07, + "loss": 1.691, + "step": 2342 + }, + { + "epoch": 2.5676712328767124, + "grad_norm": 2.6367626190185547, + "learning_rate": 8.306255077173031e-07, + "loss": 1.6999, + "step": 2343 + }, + { + "epoch": 2.568767123287671, + "grad_norm": 2.588322162628174, + "learning_rate": 8.285946385052804e-07, + "loss": 1.7915, + "step": 2344 + }, + { + "epoch": 2.56986301369863, + "grad_norm": 2.6249992847442627, + "learning_rate": 8.265637692932575e-07, + "loss": 1.6534, + "step": 2345 + }, + { + "epoch": 2.570958904109589, + "grad_norm": 2.6095449924468994, + "learning_rate": 8.245329000812348e-07, + "loss": 1.7339, + "step": 2346 + }, + { + "epoch": 2.572054794520548, + "grad_norm": 2.6527903079986572, + "learning_rate": 8.225020308692121e-07, + "loss": 1.7199, + "step": 2347 + }, + { + "epoch": 2.573150684931507, + "grad_norm": 2.6879372596740723, + "learning_rate": 8.204711616571894e-07, + "loss": 1.6669, + "step": 2348 + }, + { + "epoch": 2.5742465753424657, + "grad_norm": 2.6295018196105957, + "learning_rate": 8.184402924451667e-07, + "loss": 1.7001, + "step": 2349 + }, + { + "epoch": 2.5753424657534247, + "grad_norm": 2.669123888015747, + "learning_rate": 8.164094232331439e-07, + "loss": 1.6573, + "step": 2350 + }, + { + "epoch": 2.5753424657534247, + "eval_loss": 1.9378976821899414, + "eval_runtime": 37.3297, + "eval_samples_per_second": 26.788, + "eval_steps_per_second": 1.125, + "step": 2350 + }, + { + "epoch": 2.5764383561643838, + "grad_norm": 2.5638668537139893, + "learning_rate": 8.14378554021121e-07, + "loss": 1.7396, + "step": 2351 + }, + { + "epoch": 2.5775342465753424, + "grad_norm": 2.613858461380005, + "learning_rate": 8.123476848090983e-07, + "loss": 1.7081, + "step": 2352 + }, + { + "epoch": 2.5786301369863014, + "grad_norm": 2.633657693862915, + "learning_rate": 8.103168155970755e-07, + "loss": 1.6905, + "step": 2353 + }, + { + "epoch": 2.5797260273972604, + "grad_norm": 2.600965738296509, + "learning_rate": 8.082859463850528e-07, + "loss": 1.6979, + "step": 2354 + }, + { + "epoch": 2.580821917808219, + "grad_norm": 2.7222094535827637, + "learning_rate": 8.062550771730301e-07, + "loss": 1.7267, + "step": 2355 + }, + { + "epoch": 2.581917808219178, + "grad_norm": 2.783076047897339, + "learning_rate": 8.042242079610074e-07, + "loss": 1.6906, + "step": 2356 + }, + { + "epoch": 2.583013698630137, + "grad_norm": 2.671109199523926, + "learning_rate": 8.021933387489846e-07, + "loss": 1.6822, + "step": 2357 + }, + { + "epoch": 2.5841095890410957, + "grad_norm": 2.602358818054199, + "learning_rate": 8.001624695369618e-07, + "loss": 1.6834, + "step": 2358 + }, + { + "epoch": 2.5852054794520547, + "grad_norm": 2.665668249130249, + "learning_rate": 7.981316003249391e-07, + "loss": 1.7003, + "step": 2359 + }, + { + "epoch": 2.5863013698630137, + "grad_norm": 2.827331781387329, + "learning_rate": 7.961007311129164e-07, + "loss": 1.7663, + "step": 2360 + }, + { + "epoch": 2.5863013698630137, + "eval_loss": 1.9384022951126099, + "eval_runtime": 37.2505, + "eval_samples_per_second": 26.845, + "eval_steps_per_second": 1.128, + "step": 2360 + }, + { + "epoch": 2.5873972602739728, + "grad_norm": 2.603614568710327, + "learning_rate": 7.940698619008936e-07, + "loss": 1.6496, + "step": 2361 + }, + { + "epoch": 2.5884931506849314, + "grad_norm": 2.695448875427246, + "learning_rate": 7.920389926888709e-07, + "loss": 1.7325, + "step": 2362 + }, + { + "epoch": 2.5895890410958904, + "grad_norm": 2.6337499618530273, + "learning_rate": 7.900081234768482e-07, + "loss": 1.8151, + "step": 2363 + }, + { + "epoch": 2.5906849315068494, + "grad_norm": 2.6526639461517334, + "learning_rate": 7.879772542648254e-07, + "loss": 1.6684, + "step": 2364 + }, + { + "epoch": 2.5917808219178085, + "grad_norm": 2.5932109355926514, + "learning_rate": 7.859463850528026e-07, + "loss": 1.7238, + "step": 2365 + }, + { + "epoch": 2.592876712328767, + "grad_norm": 2.6547865867614746, + "learning_rate": 7.839155158407799e-07, + "loss": 1.7204, + "step": 2366 + }, + { + "epoch": 2.593972602739726, + "grad_norm": 2.6299901008605957, + "learning_rate": 7.818846466287572e-07, + "loss": 1.72, + "step": 2367 + }, + { + "epoch": 2.595068493150685, + "grad_norm": 2.69770884513855, + "learning_rate": 7.798537774167345e-07, + "loss": 1.7157, + "step": 2368 + }, + { + "epoch": 2.5961643835616437, + "grad_norm": 2.721971273422241, + "learning_rate": 7.778229082047117e-07, + "loss": 1.7384, + "step": 2369 + }, + { + "epoch": 2.5972602739726027, + "grad_norm": 2.6927170753479004, + "learning_rate": 7.757920389926889e-07, + "loss": 1.6823, + "step": 2370 + }, + { + "epoch": 2.5972602739726027, + "eval_loss": 1.935421109199524, + "eval_runtime": 37.2369, + "eval_samples_per_second": 26.855, + "eval_steps_per_second": 1.128, + "step": 2370 + }, + { + "epoch": 2.5983561643835618, + "grad_norm": 2.6370513439178467, + "learning_rate": 7.737611697806662e-07, + "loss": 1.7193, + "step": 2371 + }, + { + "epoch": 2.5994520547945204, + "grad_norm": 2.6238462924957275, + "learning_rate": 7.717303005686435e-07, + "loss": 1.685, + "step": 2372 + }, + { + "epoch": 2.6005479452054794, + "grad_norm": 2.643711566925049, + "learning_rate": 7.696994313566207e-07, + "loss": 1.7318, + "step": 2373 + }, + { + "epoch": 2.6016438356164384, + "grad_norm": 2.6582396030426025, + "learning_rate": 7.67668562144598e-07, + "loss": 1.7782, + "step": 2374 + }, + { + "epoch": 2.602739726027397, + "grad_norm": 2.6774542331695557, + "learning_rate": 7.656376929325753e-07, + "loss": 1.6587, + "step": 2375 + }, + { + "epoch": 2.603835616438356, + "grad_norm": 2.611940622329712, + "learning_rate": 7.636068237205524e-07, + "loss": 1.6779, + "step": 2376 + }, + { + "epoch": 2.604931506849315, + "grad_norm": 2.6520285606384277, + "learning_rate": 7.615759545085296e-07, + "loss": 1.7437, + "step": 2377 + }, + { + "epoch": 2.606027397260274, + "grad_norm": 2.666217565536499, + "learning_rate": 7.595450852965069e-07, + "loss": 1.6655, + "step": 2378 + }, + { + "epoch": 2.607123287671233, + "grad_norm": 2.695321559906006, + "learning_rate": 7.575142160844842e-07, + "loss": 1.6835, + "step": 2379 + }, + { + "epoch": 2.6082191780821917, + "grad_norm": 2.6860222816467285, + "learning_rate": 7.554833468724615e-07, + "loss": 1.7316, + "step": 2380 + }, + { + "epoch": 2.6082191780821917, + "eval_loss": 1.9367016553878784, + "eval_runtime": 37.1903, + "eval_samples_per_second": 26.889, + "eval_steps_per_second": 1.129, + "step": 2380 + }, + { + "epoch": 2.6093150684931508, + "grad_norm": 2.6422245502471924, + "learning_rate": 7.534524776604387e-07, + "loss": 1.6533, + "step": 2381 + }, + { + "epoch": 2.61041095890411, + "grad_norm": 2.6544158458709717, + "learning_rate": 7.51421608448416e-07, + "loss": 1.6734, + "step": 2382 + }, + { + "epoch": 2.6115068493150684, + "grad_norm": 2.648153305053711, + "learning_rate": 7.493907392363932e-07, + "loss": 1.7374, + "step": 2383 + }, + { + "epoch": 2.6126027397260274, + "grad_norm": 2.646836042404175, + "learning_rate": 7.473598700243705e-07, + "loss": 1.6794, + "step": 2384 + }, + { + "epoch": 2.6136986301369864, + "grad_norm": 2.633972644805908, + "learning_rate": 7.453290008123477e-07, + "loss": 1.633, + "step": 2385 + }, + { + "epoch": 2.614794520547945, + "grad_norm": 2.580915927886963, + "learning_rate": 7.43298131600325e-07, + "loss": 1.6981, + "step": 2386 + }, + { + "epoch": 2.615890410958904, + "grad_norm": 2.605191946029663, + "learning_rate": 7.412672623883023e-07, + "loss": 1.711, + "step": 2387 + }, + { + "epoch": 2.616986301369863, + "grad_norm": 2.7410099506378174, + "learning_rate": 7.392363931762796e-07, + "loss": 1.7175, + "step": 2388 + }, + { + "epoch": 2.6180821917808217, + "grad_norm": 2.688540458679199, + "learning_rate": 7.372055239642567e-07, + "loss": 1.6447, + "step": 2389 + }, + { + "epoch": 2.6191780821917807, + "grad_norm": 2.719954252243042, + "learning_rate": 7.35174654752234e-07, + "loss": 1.7034, + "step": 2390 + }, + { + "epoch": 2.6191780821917807, + "eval_loss": 1.934234380722046, + "eval_runtime": 37.1917, + "eval_samples_per_second": 26.888, + "eval_steps_per_second": 1.129, + "step": 2390 + }, + { + "epoch": 2.6202739726027398, + "grad_norm": 2.7232038974761963, + "learning_rate": 7.331437855402113e-07, + "loss": 1.7318, + "step": 2391 + }, + { + "epoch": 2.621369863013699, + "grad_norm": 2.586639404296875, + "learning_rate": 7.311129163281886e-07, + "loss": 1.7529, + "step": 2392 + }, + { + "epoch": 2.6224657534246574, + "grad_norm": 2.643853187561035, + "learning_rate": 7.290820471161658e-07, + "loss": 1.6809, + "step": 2393 + }, + { + "epoch": 2.6235616438356164, + "grad_norm": 2.6359288692474365, + "learning_rate": 7.270511779041431e-07, + "loss": 1.7261, + "step": 2394 + }, + { + "epoch": 2.6246575342465754, + "grad_norm": 2.6692726612091064, + "learning_rate": 7.250203086921203e-07, + "loss": 1.7311, + "step": 2395 + }, + { + "epoch": 2.6257534246575345, + "grad_norm": 2.6712090969085693, + "learning_rate": 7.229894394800976e-07, + "loss": 1.6367, + "step": 2396 + }, + { + "epoch": 2.626849315068493, + "grad_norm": 2.618422031402588, + "learning_rate": 7.209585702680748e-07, + "loss": 1.7017, + "step": 2397 + }, + { + "epoch": 2.627945205479452, + "grad_norm": 2.6770448684692383, + "learning_rate": 7.189277010560521e-07, + "loss": 1.7193, + "step": 2398 + }, + { + "epoch": 2.629041095890411, + "grad_norm": 2.605620861053467, + "learning_rate": 7.168968318440294e-07, + "loss": 1.6515, + "step": 2399 + }, + { + "epoch": 2.6301369863013697, + "grad_norm": 2.616734266281128, + "learning_rate": 7.148659626320067e-07, + "loss": 1.685, + "step": 2400 + }, + { + "epoch": 2.6301369863013697, + "eval_loss": 1.9330971240997314, + "eval_runtime": 37.1928, + "eval_samples_per_second": 26.887, + "eval_steps_per_second": 1.129, + "step": 2400 + }, + { + "epoch": 2.6312328767123287, + "grad_norm": 2.622939348220825, + "learning_rate": 7.128350934199837e-07, + "loss": 1.655, + "step": 2401 + }, + { + "epoch": 2.632328767123288, + "grad_norm": 2.67498517036438, + "learning_rate": 7.10804224207961e-07, + "loss": 1.7659, + "step": 2402 + }, + { + "epoch": 2.6334246575342464, + "grad_norm": 2.6155238151550293, + "learning_rate": 7.087733549959383e-07, + "loss": 1.7605, + "step": 2403 + }, + { + "epoch": 2.6345205479452054, + "grad_norm": 2.527864933013916, + "learning_rate": 7.067424857839155e-07, + "loss": 1.6955, + "step": 2404 + }, + { + "epoch": 2.6356164383561644, + "grad_norm": 2.6684014797210693, + "learning_rate": 7.047116165718928e-07, + "loss": 1.7006, + "step": 2405 + }, + { + "epoch": 2.636712328767123, + "grad_norm": 2.736567258834839, + "learning_rate": 7.026807473598701e-07, + "loss": 1.6942, + "step": 2406 + }, + { + "epoch": 2.637808219178082, + "grad_norm": 2.7265021800994873, + "learning_rate": 7.006498781478473e-07, + "loss": 1.7116, + "step": 2407 + }, + { + "epoch": 2.638904109589041, + "grad_norm": 2.747013568878174, + "learning_rate": 6.986190089358245e-07, + "loss": 1.7766, + "step": 2408 + }, + { + "epoch": 2.64, + "grad_norm": 2.719531536102295, + "learning_rate": 6.965881397238018e-07, + "loss": 1.6573, + "step": 2409 + }, + { + "epoch": 2.641095890410959, + "grad_norm": 2.570411205291748, + "learning_rate": 6.945572705117791e-07, + "loss": 1.7131, + "step": 2410 + }, + { + "epoch": 2.641095890410959, + "eval_loss": 1.9303901195526123, + "eval_runtime": 37.2544, + "eval_samples_per_second": 26.842, + "eval_steps_per_second": 1.127, + "step": 2410 + }, + { + "epoch": 2.6421917808219177, + "grad_norm": 2.6145083904266357, + "learning_rate": 6.925264012997564e-07, + "loss": 1.7183, + "step": 2411 + }, + { + "epoch": 2.643287671232877, + "grad_norm": 2.6236915588378906, + "learning_rate": 6.904955320877336e-07, + "loss": 1.7132, + "step": 2412 + }, + { + "epoch": 2.644383561643836, + "grad_norm": 2.7300617694854736, + "learning_rate": 6.884646628757109e-07, + "loss": 1.7181, + "step": 2413 + }, + { + "epoch": 2.6454794520547944, + "grad_norm": 2.6430609226226807, + "learning_rate": 6.864337936636881e-07, + "loss": 1.7242, + "step": 2414 + }, + { + "epoch": 2.6465753424657534, + "grad_norm": 2.6691088676452637, + "learning_rate": 6.844029244516654e-07, + "loss": 1.7957, + "step": 2415 + }, + { + "epoch": 2.6476712328767125, + "grad_norm": 2.6629152297973633, + "learning_rate": 6.823720552396426e-07, + "loss": 1.7569, + "step": 2416 + }, + { + "epoch": 2.648767123287671, + "grad_norm": 2.6918444633483887, + "learning_rate": 6.803411860276199e-07, + "loss": 1.6753, + "step": 2417 + }, + { + "epoch": 2.64986301369863, + "grad_norm": 2.6504695415496826, + "learning_rate": 6.783103168155972e-07, + "loss": 1.7234, + "step": 2418 + }, + { + "epoch": 2.650958904109589, + "grad_norm": 2.5938878059387207, + "learning_rate": 6.762794476035745e-07, + "loss": 1.7098, + "step": 2419 + }, + { + "epoch": 2.6520547945205477, + "grad_norm": 2.6518945693969727, + "learning_rate": 6.742485783915515e-07, + "loss": 1.6973, + "step": 2420 + }, + { + "epoch": 2.6520547945205477, + "eval_loss": 1.9311248064041138, + "eval_runtime": 37.5332, + "eval_samples_per_second": 26.643, + "eval_steps_per_second": 1.119, + "step": 2420 + }, + { + "epoch": 2.6531506849315067, + "grad_norm": 2.7135426998138428, + "learning_rate": 6.722177091795288e-07, + "loss": 1.7364, + "step": 2421 + }, + { + "epoch": 2.6542465753424658, + "grad_norm": 2.7694764137268066, + "learning_rate": 6.701868399675061e-07, + "loss": 1.7292, + "step": 2422 + }, + { + "epoch": 2.655342465753425, + "grad_norm": 2.7132647037506104, + "learning_rate": 6.681559707554834e-07, + "loss": 1.7074, + "step": 2423 + }, + { + "epoch": 2.6564383561643834, + "grad_norm": 2.643613576889038, + "learning_rate": 6.661251015434606e-07, + "loss": 1.6761, + "step": 2424 + }, + { + "epoch": 2.6575342465753424, + "grad_norm": 2.7570016384124756, + "learning_rate": 6.640942323314379e-07, + "loss": 1.6834, + "step": 2425 + }, + { + "epoch": 2.6586301369863015, + "grad_norm": 2.6339755058288574, + "learning_rate": 6.620633631194151e-07, + "loss": 1.6736, + "step": 2426 + }, + { + "epoch": 2.6597260273972605, + "grad_norm": 2.6660594940185547, + "learning_rate": 6.600324939073924e-07, + "loss": 1.6849, + "step": 2427 + }, + { + "epoch": 2.660821917808219, + "grad_norm": 2.6688084602355957, + "learning_rate": 6.580016246953696e-07, + "loss": 1.7178, + "step": 2428 + }, + { + "epoch": 2.661917808219178, + "grad_norm": 2.6950714588165283, + "learning_rate": 6.559707554833469e-07, + "loss": 1.6978, + "step": 2429 + }, + { + "epoch": 2.663013698630137, + "grad_norm": 2.711935043334961, + "learning_rate": 6.539398862713242e-07, + "loss": 1.7468, + "step": 2430 + }, + { + "epoch": 2.663013698630137, + "eval_loss": 1.9298992156982422, + "eval_runtime": 37.4036, + "eval_samples_per_second": 26.735, + "eval_steps_per_second": 1.123, + "step": 2430 + }, + { + "epoch": 2.6641095890410957, + "grad_norm": 2.593010663986206, + "learning_rate": 6.519090170593015e-07, + "loss": 1.6863, + "step": 2431 + }, + { + "epoch": 2.6652054794520548, + "grad_norm": 2.6305296421051025, + "learning_rate": 6.498781478472786e-07, + "loss": 1.7474, + "step": 2432 + }, + { + "epoch": 2.666301369863014, + "grad_norm": 2.6826376914978027, + "learning_rate": 6.478472786352559e-07, + "loss": 1.7136, + "step": 2433 + }, + { + "epoch": 2.6673972602739724, + "grad_norm": 2.6297945976257324, + "learning_rate": 6.458164094232332e-07, + "loss": 1.7667, + "step": 2434 + }, + { + "epoch": 2.6684931506849314, + "grad_norm": 2.7016513347625732, + "learning_rate": 6.437855402112105e-07, + "loss": 1.7143, + "step": 2435 + }, + { + "epoch": 2.6695890410958905, + "grad_norm": 2.7064952850341797, + "learning_rate": 6.417546709991877e-07, + "loss": 1.6453, + "step": 2436 + }, + { + "epoch": 2.670684931506849, + "grad_norm": 2.7441818714141846, + "learning_rate": 6.39723801787165e-07, + "loss": 1.676, + "step": 2437 + }, + { + "epoch": 2.671780821917808, + "grad_norm": 2.7477304935455322, + "learning_rate": 6.376929325751423e-07, + "loss": 1.754, + "step": 2438 + }, + { + "epoch": 2.672876712328767, + "grad_norm": 2.6238012313842773, + "learning_rate": 6.356620633631195e-07, + "loss": 1.6926, + "step": 2439 + }, + { + "epoch": 2.673972602739726, + "grad_norm": 2.625750780105591, + "learning_rate": 6.336311941510967e-07, + "loss": 1.6856, + "step": 2440 + }, + { + "epoch": 2.673972602739726, + "eval_loss": 1.9298005104064941, + "eval_runtime": 37.2569, + "eval_samples_per_second": 26.841, + "eval_steps_per_second": 1.127, + "step": 2440 + }, + { + "epoch": 2.675068493150685, + "grad_norm": 2.650430202484131, + "learning_rate": 6.31600324939074e-07, + "loss": 1.694, + "step": 2441 + }, + { + "epoch": 2.6761643835616438, + "grad_norm": 2.681729316711426, + "learning_rate": 6.295694557270513e-07, + "loss": 1.7124, + "step": 2442 + }, + { + "epoch": 2.677260273972603, + "grad_norm": 2.6914284229278564, + "learning_rate": 6.275385865150286e-07, + "loss": 1.7497, + "step": 2443 + }, + { + "epoch": 2.678356164383562, + "grad_norm": 2.6426215171813965, + "learning_rate": 6.255077173030058e-07, + "loss": 1.7147, + "step": 2444 + }, + { + "epoch": 2.6794520547945204, + "grad_norm": 2.7941248416900635, + "learning_rate": 6.23476848090983e-07, + "loss": 1.704, + "step": 2445 + }, + { + "epoch": 2.6805479452054795, + "grad_norm": 2.599816083908081, + "learning_rate": 6.214459788789602e-07, + "loss": 1.7141, + "step": 2446 + }, + { + "epoch": 2.6816438356164385, + "grad_norm": 2.6136841773986816, + "learning_rate": 6.194151096669375e-07, + "loss": 1.6938, + "step": 2447 + }, + { + "epoch": 2.682739726027397, + "grad_norm": 2.63150691986084, + "learning_rate": 6.173842404549147e-07, + "loss": 1.7304, + "step": 2448 + }, + { + "epoch": 2.683835616438356, + "grad_norm": 2.6653857231140137, + "learning_rate": 6.15353371242892e-07, + "loss": 1.6979, + "step": 2449 + }, + { + "epoch": 2.684931506849315, + "grad_norm": 2.632521867752075, + "learning_rate": 6.133225020308693e-07, + "loss": 1.712, + "step": 2450 + }, + { + "epoch": 2.684931506849315, + "eval_loss": 1.9305981397628784, + "eval_runtime": 37.2989, + "eval_samples_per_second": 26.81, + "eval_steps_per_second": 1.126, + "step": 2450 + }, + { + "epoch": 2.6860273972602737, + "grad_norm": 2.6429569721221924, + "learning_rate": 6.112916328188466e-07, + "loss": 1.6156, + "step": 2451 + }, + { + "epoch": 2.6871232876712328, + "grad_norm": 2.576719045639038, + "learning_rate": 6.092607636068238e-07, + "loss": 1.7034, + "step": 2452 + }, + { + "epoch": 2.688219178082192, + "grad_norm": 2.7339651584625244, + "learning_rate": 6.07229894394801e-07, + "loss": 1.7756, + "step": 2453 + }, + { + "epoch": 2.689315068493151, + "grad_norm": 2.6168463230133057, + "learning_rate": 6.051990251827783e-07, + "loss": 1.6746, + "step": 2454 + }, + { + "epoch": 2.69041095890411, + "grad_norm": 2.554588556289673, + "learning_rate": 6.031681559707556e-07, + "loss": 1.6721, + "step": 2455 + }, + { + "epoch": 2.6915068493150685, + "grad_norm": 2.6572208404541016, + "learning_rate": 6.011372867587327e-07, + "loss": 1.7241, + "step": 2456 + }, + { + "epoch": 2.6926027397260275, + "grad_norm": 2.746213436126709, + "learning_rate": 5.9910641754671e-07, + "loss": 1.7228, + "step": 2457 + }, + { + "epoch": 2.6936986301369865, + "grad_norm": 2.6031382083892822, + "learning_rate": 5.970755483346873e-07, + "loss": 1.6683, + "step": 2458 + }, + { + "epoch": 2.694794520547945, + "grad_norm": 2.672415018081665, + "learning_rate": 5.950446791226646e-07, + "loss": 1.6981, + "step": 2459 + }, + { + "epoch": 2.695890410958904, + "grad_norm": 2.5940663814544678, + "learning_rate": 5.930138099106418e-07, + "loss": 1.684, + "step": 2460 + }, + { + "epoch": 2.695890410958904, + "eval_loss": 1.9297748804092407, + "eval_runtime": 37.2936, + "eval_samples_per_second": 26.814, + "eval_steps_per_second": 1.126, + "step": 2460 + }, + { + "epoch": 2.696986301369863, + "grad_norm": 2.6395111083984375, + "learning_rate": 5.909829406986191e-07, + "loss": 1.7055, + "step": 2461 + }, + { + "epoch": 2.6980821917808218, + "grad_norm": 2.6723620891571045, + "learning_rate": 5.889520714865963e-07, + "loss": 1.6866, + "step": 2462 + }, + { + "epoch": 2.699178082191781, + "grad_norm": 2.717759132385254, + "learning_rate": 5.869212022745735e-07, + "loss": 1.7176, + "step": 2463 + }, + { + "epoch": 2.70027397260274, + "grad_norm": 2.563380479812622, + "learning_rate": 5.848903330625508e-07, + "loss": 1.6588, + "step": 2464 + }, + { + "epoch": 2.7013698630136984, + "grad_norm": 2.6206467151641846, + "learning_rate": 5.828594638505281e-07, + "loss": 1.7233, + "step": 2465 + }, + { + "epoch": 2.7024657534246574, + "grad_norm": 2.7347686290740967, + "learning_rate": 5.808285946385054e-07, + "loss": 1.7118, + "step": 2466 + }, + { + "epoch": 2.7035616438356165, + "grad_norm": 2.6237869262695312, + "learning_rate": 5.787977254264827e-07, + "loss": 1.6843, + "step": 2467 + }, + { + "epoch": 2.704657534246575, + "grad_norm": 2.6551127433776855, + "learning_rate": 5.767668562144598e-07, + "loss": 1.7234, + "step": 2468 + }, + { + "epoch": 2.705753424657534, + "grad_norm": 2.7673237323760986, + "learning_rate": 5.747359870024371e-07, + "loss": 1.7436, + "step": 2469 + }, + { + "epoch": 2.706849315068493, + "grad_norm": 2.626408338546753, + "learning_rate": 5.727051177904144e-07, + "loss": 1.6945, + "step": 2470 + }, + { + "epoch": 2.706849315068493, + "eval_loss": 1.9326976537704468, + "eval_runtime": 37.1855, + "eval_samples_per_second": 26.892, + "eval_steps_per_second": 1.129, + "step": 2470 + }, + { + "epoch": 2.707945205479452, + "grad_norm": 2.65018630027771, + "learning_rate": 5.706742485783915e-07, + "loss": 1.7005, + "step": 2471 + }, + { + "epoch": 2.709041095890411, + "grad_norm": 2.7302255630493164, + "learning_rate": 5.686433793663688e-07, + "loss": 1.7411, + "step": 2472 + }, + { + "epoch": 2.71013698630137, + "grad_norm": 2.6798901557922363, + "learning_rate": 5.666125101543461e-07, + "loss": 1.6764, + "step": 2473 + }, + { + "epoch": 2.711232876712329, + "grad_norm": 2.7441818714141846, + "learning_rate": 5.645816409423234e-07, + "loss": 1.6428, + "step": 2474 + }, + { + "epoch": 2.712328767123288, + "grad_norm": 2.7589964866638184, + "learning_rate": 5.625507717303006e-07, + "loss": 1.7632, + "step": 2475 + }, + { + "epoch": 2.7134246575342464, + "grad_norm": 2.712848424911499, + "learning_rate": 5.605199025182779e-07, + "loss": 1.7217, + "step": 2476 + }, + { + "epoch": 2.7145205479452055, + "grad_norm": 2.6010446548461914, + "learning_rate": 5.584890333062551e-07, + "loss": 1.686, + "step": 2477 + }, + { + "epoch": 2.7156164383561645, + "grad_norm": 2.6970198154449463, + "learning_rate": 5.564581640942324e-07, + "loss": 1.6769, + "step": 2478 + }, + { + "epoch": 2.716712328767123, + "grad_norm": 2.7168943881988525, + "learning_rate": 5.544272948822096e-07, + "loss": 1.7434, + "step": 2479 + }, + { + "epoch": 2.717808219178082, + "grad_norm": 2.6607320308685303, + "learning_rate": 5.523964256701868e-07, + "loss": 1.6781, + "step": 2480 + }, + { + "epoch": 2.717808219178082, + "eval_loss": 1.9310208559036255, + "eval_runtime": 37.1878, + "eval_samples_per_second": 26.891, + "eval_steps_per_second": 1.129, + "step": 2480 + }, + { + "epoch": 2.718904109589041, + "grad_norm": 2.6367454528808594, + "learning_rate": 5.503655564581641e-07, + "loss": 1.6893, + "step": 2481 + }, + { + "epoch": 2.7199999999999998, + "grad_norm": 2.6094961166381836, + "learning_rate": 5.483346872461414e-07, + "loss": 1.7248, + "step": 2482 + }, + { + "epoch": 2.721095890410959, + "grad_norm": 2.5507280826568604, + "learning_rate": 5.463038180341186e-07, + "loss": 1.6823, + "step": 2483 + }, + { + "epoch": 2.722191780821918, + "grad_norm": 2.695561170578003, + "learning_rate": 5.442729488220959e-07, + "loss": 1.6919, + "step": 2484 + }, + { + "epoch": 2.723287671232877, + "grad_norm": 2.7629032135009766, + "learning_rate": 5.422420796100732e-07, + "loss": 1.6787, + "step": 2485 + }, + { + "epoch": 2.724383561643836, + "grad_norm": 2.6458840370178223, + "learning_rate": 5.402112103980505e-07, + "loss": 1.7204, + "step": 2486 + }, + { + "epoch": 2.7254794520547945, + "grad_norm": 2.748023271560669, + "learning_rate": 5.381803411860276e-07, + "loss": 1.6915, + "step": 2487 + }, + { + "epoch": 2.7265753424657535, + "grad_norm": 2.712794542312622, + "learning_rate": 5.361494719740049e-07, + "loss": 1.7314, + "step": 2488 + }, + { + "epoch": 2.7276712328767125, + "grad_norm": 2.6901073455810547, + "learning_rate": 5.341186027619822e-07, + "loss": 1.7719, + "step": 2489 + }, + { + "epoch": 2.728767123287671, + "grad_norm": 2.674168825149536, + "learning_rate": 5.320877335499594e-07, + "loss": 1.6601, + "step": 2490 + }, + { + "epoch": 2.728767123287671, + "eval_loss": 1.9327023029327393, + "eval_runtime": 37.1502, + "eval_samples_per_second": 26.918, + "eval_steps_per_second": 1.131, + "step": 2490 + }, + { + "epoch": 2.72986301369863, + "grad_norm": 2.6993298530578613, + "learning_rate": 5.300568643379366e-07, + "loss": 1.679, + "step": 2491 + }, + { + "epoch": 2.730958904109589, + "grad_norm": 2.6257898807525635, + "learning_rate": 5.280259951259139e-07, + "loss": 1.7667, + "step": 2492 + }, + { + "epoch": 2.732054794520548, + "grad_norm": 2.6909453868865967, + "learning_rate": 5.259951259138912e-07, + "loss": 1.7052, + "step": 2493 + }, + { + "epoch": 2.733150684931507, + "grad_norm": 2.609600782394409, + "learning_rate": 5.239642567018685e-07, + "loss": 1.6757, + "step": 2494 + }, + { + "epoch": 2.734246575342466, + "grad_norm": 2.911816358566284, + "learning_rate": 5.219333874898457e-07, + "loss": 1.6777, + "step": 2495 + }, + { + "epoch": 2.7353424657534244, + "grad_norm": 2.6345603466033936, + "learning_rate": 5.199025182778229e-07, + "loss": 1.6931, + "step": 2496 + }, + { + "epoch": 2.7364383561643835, + "grad_norm": 2.684359312057495, + "learning_rate": 5.178716490658002e-07, + "loss": 1.7098, + "step": 2497 + }, + { + "epoch": 2.7375342465753425, + "grad_norm": 2.694227695465088, + "learning_rate": 5.158407798537775e-07, + "loss": 1.6653, + "step": 2498 + }, + { + "epoch": 2.7386301369863015, + "grad_norm": 2.6043031215667725, + "learning_rate": 5.138099106417547e-07, + "loss": 1.6284, + "step": 2499 + }, + { + "epoch": 2.73972602739726, + "grad_norm": 2.6505203247070312, + "learning_rate": 5.11779041429732e-07, + "loss": 1.6002, + "step": 2500 + }, + { + "epoch": 2.73972602739726, + "eval_loss": 1.9312472343444824, + "eval_runtime": 37.1619, + "eval_samples_per_second": 26.909, + "eval_steps_per_second": 1.13, + "step": 2500 + }, + { + "epoch": 2.740821917808219, + "grad_norm": 2.673102855682373, + "learning_rate": 5.097481722177093e-07, + "loss": 1.6781, + "step": 2501 + }, + { + "epoch": 2.741917808219178, + "grad_norm": 2.694951057434082, + "learning_rate": 5.077173030056865e-07, + "loss": 1.6758, + "step": 2502 + }, + { + "epoch": 2.743013698630137, + "grad_norm": 2.7397866249084473, + "learning_rate": 5.056864337936637e-07, + "loss": 1.7493, + "step": 2503 + }, + { + "epoch": 2.744109589041096, + "grad_norm": 2.6714928150177, + "learning_rate": 5.03655564581641e-07, + "loss": 1.7482, + "step": 2504 + }, + { + "epoch": 2.745205479452055, + "grad_norm": 2.659916400909424, + "learning_rate": 5.016246953696182e-07, + "loss": 1.7065, + "step": 2505 + }, + { + "epoch": 2.746301369863014, + "grad_norm": 2.625319242477417, + "learning_rate": 4.995938261575955e-07, + "loss": 1.8302, + "step": 2506 + }, + { + "epoch": 2.7473972602739725, + "grad_norm": 2.693692684173584, + "learning_rate": 4.975629569455727e-07, + "loss": 1.6995, + "step": 2507 + }, + { + "epoch": 2.7484931506849315, + "grad_norm": 2.730882406234741, + "learning_rate": 4.9553208773355e-07, + "loss": 1.7171, + "step": 2508 + }, + { + "epoch": 2.7495890410958905, + "grad_norm": 2.5712368488311768, + "learning_rate": 4.935012185215273e-07, + "loss": 1.7869, + "step": 2509 + }, + { + "epoch": 2.750684931506849, + "grad_norm": 2.52966046333313, + "learning_rate": 4.914703493095046e-07, + "loss": 1.687, + "step": 2510 + }, + { + "epoch": 2.750684931506849, + "eval_loss": 1.9297986030578613, + "eval_runtime": 37.2591, + "eval_samples_per_second": 26.839, + "eval_steps_per_second": 1.127, + "step": 2510 + }, + { + "epoch": 2.751780821917808, + "grad_norm": 2.6121788024902344, + "learning_rate": 4.894394800974818e-07, + "loss": 1.6527, + "step": 2511 + }, + { + "epoch": 2.752876712328767, + "grad_norm": 2.5516674518585205, + "learning_rate": 4.87408610885459e-07, + "loss": 1.6697, + "step": 2512 + }, + { + "epoch": 2.7539726027397258, + "grad_norm": 2.679359197616577, + "learning_rate": 4.853777416734363e-07, + "loss": 1.7923, + "step": 2513 + }, + { + "epoch": 2.755068493150685, + "grad_norm": 2.5589449405670166, + "learning_rate": 4.833468724614136e-07, + "loss": 1.7436, + "step": 2514 + }, + { + "epoch": 2.756164383561644, + "grad_norm": 2.5719308853149414, + "learning_rate": 4.813160032493907e-07, + "loss": 1.6924, + "step": 2515 + }, + { + "epoch": 2.757260273972603, + "grad_norm": 2.6262130737304688, + "learning_rate": 4.79285134037368e-07, + "loss": 1.6781, + "step": 2516 + }, + { + "epoch": 2.758356164383562, + "grad_norm": 2.661180019378662, + "learning_rate": 4.772542648253453e-07, + "loss": 1.666, + "step": 2517 + }, + { + "epoch": 2.7594520547945205, + "grad_norm": 2.7254981994628906, + "learning_rate": 4.752233956133225e-07, + "loss": 1.6764, + "step": 2518 + }, + { + "epoch": 2.7605479452054795, + "grad_norm": 2.5897748470306396, + "learning_rate": 4.731925264012998e-07, + "loss": 1.6853, + "step": 2519 + }, + { + "epoch": 2.7616438356164386, + "grad_norm": 2.7142059803009033, + "learning_rate": 4.7116165718927705e-07, + "loss": 1.7576, + "step": 2520 + }, + { + "epoch": 2.7616438356164386, + "eval_loss": 1.9287093877792358, + "eval_runtime": 37.499, + "eval_samples_per_second": 26.667, + "eval_steps_per_second": 1.12, + "step": 2520 + }, + { + "epoch": 2.762739726027397, + "grad_norm": 2.723893165588379, + "learning_rate": 4.6913078797725427e-07, + "loss": 1.7089, + "step": 2521 + }, + { + "epoch": 2.763835616438356, + "grad_norm": 2.6650211811065674, + "learning_rate": 4.6709991876523155e-07, + "loss": 1.6936, + "step": 2522 + }, + { + "epoch": 2.764931506849315, + "grad_norm": 2.674407720565796, + "learning_rate": 4.650690495532088e-07, + "loss": 1.6771, + "step": 2523 + }, + { + "epoch": 2.766027397260274, + "grad_norm": 2.674057722091675, + "learning_rate": 4.6303818034118605e-07, + "loss": 1.7758, + "step": 2524 + }, + { + "epoch": 2.767123287671233, + "grad_norm": 2.7330195903778076, + "learning_rate": 4.610073111291633e-07, + "loss": 1.7414, + "step": 2525 + }, + { + "epoch": 2.768219178082192, + "grad_norm": 2.6220991611480713, + "learning_rate": 4.589764419171406e-07, + "loss": 1.6475, + "step": 2526 + }, + { + "epoch": 2.7693150684931505, + "grad_norm": 2.663470506668091, + "learning_rate": 4.569455727051178e-07, + "loss": 1.7021, + "step": 2527 + }, + { + "epoch": 2.7704109589041095, + "grad_norm": 2.6512386798858643, + "learning_rate": 4.549147034930951e-07, + "loss": 1.662, + "step": 2528 + }, + { + "epoch": 2.7715068493150685, + "grad_norm": 2.6922953128814697, + "learning_rate": 4.5288383428107237e-07, + "loss": 1.7606, + "step": 2529 + }, + { + "epoch": 2.7726027397260276, + "grad_norm": 2.5963351726531982, + "learning_rate": 4.5085296506904954e-07, + "loss": 1.7219, + "step": 2530 + }, + { + "epoch": 2.7726027397260276, + "eval_loss": 1.9284989833831787, + "eval_runtime": 37.3552, + "eval_samples_per_second": 26.77, + "eval_steps_per_second": 1.124, + "step": 2530 + }, + { + "epoch": 2.773698630136986, + "grad_norm": 2.7230756282806396, + "learning_rate": 4.488220958570268e-07, + "loss": 1.6907, + "step": 2531 + }, + { + "epoch": 2.774794520547945, + "grad_norm": 2.6310737133026123, + "learning_rate": 4.467912266450041e-07, + "loss": 1.6675, + "step": 2532 + }, + { + "epoch": 2.775890410958904, + "grad_norm": 2.652928590774536, + "learning_rate": 4.447603574329813e-07, + "loss": 1.6579, + "step": 2533 + }, + { + "epoch": 2.7769863013698632, + "grad_norm": 2.635474681854248, + "learning_rate": 4.427294882209586e-07, + "loss": 1.7347, + "step": 2534 + }, + { + "epoch": 2.778082191780822, + "grad_norm": 2.6199090480804443, + "learning_rate": 4.4069861900893587e-07, + "loss": 1.6644, + "step": 2535 + }, + { + "epoch": 2.779178082191781, + "grad_norm": 2.7695133686065674, + "learning_rate": 4.386677497969131e-07, + "loss": 1.7151, + "step": 2536 + }, + { + "epoch": 2.78027397260274, + "grad_norm": 2.6705634593963623, + "learning_rate": 4.3663688058489037e-07, + "loss": 1.8027, + "step": 2537 + }, + { + "epoch": 2.7813698630136985, + "grad_norm": 2.7524545192718506, + "learning_rate": 4.3460601137286764e-07, + "loss": 1.7093, + "step": 2538 + }, + { + "epoch": 2.7824657534246575, + "grad_norm": 2.6585731506347656, + "learning_rate": 4.325751421608449e-07, + "loss": 1.7429, + "step": 2539 + }, + { + "epoch": 2.7835616438356166, + "grad_norm": 2.6227736473083496, + "learning_rate": 4.3054427294882214e-07, + "loss": 1.7089, + "step": 2540 + }, + { + "epoch": 2.7835616438356166, + "eval_loss": 1.9279123544692993, + "eval_runtime": 37.2954, + "eval_samples_per_second": 26.813, + "eval_steps_per_second": 1.126, + "step": 2540 + }, + { + "epoch": 2.784657534246575, + "grad_norm": 2.5640761852264404, + "learning_rate": 4.285134037367994e-07, + "loss": 1.7332, + "step": 2541 + }, + { + "epoch": 2.785753424657534, + "grad_norm": 2.7220466136932373, + "learning_rate": 4.264825345247767e-07, + "loss": 1.6978, + "step": 2542 + }, + { + "epoch": 2.786849315068493, + "grad_norm": 2.659855365753174, + "learning_rate": 4.2445166531275386e-07, + "loss": 1.6211, + "step": 2543 + }, + { + "epoch": 2.787945205479452, + "grad_norm": 2.6177449226379395, + "learning_rate": 4.2242079610073114e-07, + "loss": 1.6771, + "step": 2544 + }, + { + "epoch": 2.789041095890411, + "grad_norm": 2.7482223510742188, + "learning_rate": 4.203899268887084e-07, + "loss": 1.6339, + "step": 2545 + }, + { + "epoch": 2.79013698630137, + "grad_norm": 2.66460919380188, + "learning_rate": 4.1835905767668564e-07, + "loss": 1.7428, + "step": 2546 + }, + { + "epoch": 2.791232876712329, + "grad_norm": 2.6975536346435547, + "learning_rate": 4.163281884646629e-07, + "loss": 1.7647, + "step": 2547 + }, + { + "epoch": 2.792328767123288, + "grad_norm": 2.5990328788757324, + "learning_rate": 4.142973192526402e-07, + "loss": 1.7001, + "step": 2548 + }, + { + "epoch": 2.7934246575342465, + "grad_norm": 2.886885643005371, + "learning_rate": 4.122664500406174e-07, + "loss": 1.7358, + "step": 2549 + }, + { + "epoch": 2.7945205479452055, + "grad_norm": 2.685453176498413, + "learning_rate": 4.102355808285947e-07, + "loss": 1.7167, + "step": 2550 + }, + { + "epoch": 2.7945205479452055, + "eval_loss": 1.926743984222412, + "eval_runtime": 37.3249, + "eval_samples_per_second": 26.792, + "eval_steps_per_second": 1.125, + "step": 2550 + }, + { + "epoch": 2.7956164383561646, + "grad_norm": 3.710097551345825, + "learning_rate": 4.0820471161657196e-07, + "loss": 1.6883, + "step": 2551 + }, + { + "epoch": 2.796712328767123, + "grad_norm": 2.782146692276001, + "learning_rate": 4.0617384240454913e-07, + "loss": 1.7271, + "step": 2552 + }, + { + "epoch": 2.797808219178082, + "grad_norm": 2.653635263442993, + "learning_rate": 4.041429731925264e-07, + "loss": 1.7343, + "step": 2553 + }, + { + "epoch": 2.7989041095890412, + "grad_norm": 2.6325294971466064, + "learning_rate": 4.021121039805037e-07, + "loss": 1.6424, + "step": 2554 + }, + { + "epoch": 2.8, + "grad_norm": 2.666888952255249, + "learning_rate": 4.000812347684809e-07, + "loss": 1.6722, + "step": 2555 + }, + { + "epoch": 2.801095890410959, + "grad_norm": 2.586986780166626, + "learning_rate": 3.980503655564582e-07, + "loss": 1.6779, + "step": 2556 + }, + { + "epoch": 2.802191780821918, + "grad_norm": 2.744933605194092, + "learning_rate": 3.9601949634443546e-07, + "loss": 1.7067, + "step": 2557 + }, + { + "epoch": 2.8032876712328765, + "grad_norm": 2.675821542739868, + "learning_rate": 3.939886271324127e-07, + "loss": 1.6975, + "step": 2558 + }, + { + "epoch": 2.8043835616438355, + "grad_norm": 2.746312379837036, + "learning_rate": 3.9195775792038996e-07, + "loss": 1.6871, + "step": 2559 + }, + { + "epoch": 2.8054794520547945, + "grad_norm": 2.7854371070861816, + "learning_rate": 3.8992688870836723e-07, + "loss": 1.7325, + "step": 2560 + }, + { + "epoch": 2.8054794520547945, + "eval_loss": 1.9262990951538086, + "eval_runtime": 37.3194, + "eval_samples_per_second": 26.796, + "eval_steps_per_second": 1.125, + "step": 2560 + }, + { + "epoch": 2.8065753424657536, + "grad_norm": 2.6016507148742676, + "learning_rate": 3.8789601949634446e-07, + "loss": 1.7339, + "step": 2561 + }, + { + "epoch": 2.807671232876712, + "grad_norm": 2.6750402450561523, + "learning_rate": 3.8586515028432173e-07, + "loss": 1.6845, + "step": 2562 + }, + { + "epoch": 2.808767123287671, + "grad_norm": 2.6950154304504395, + "learning_rate": 3.83834281072299e-07, + "loss": 1.6109, + "step": 2563 + }, + { + "epoch": 2.8098630136986302, + "grad_norm": 2.6395761966705322, + "learning_rate": 3.818034118602762e-07, + "loss": 1.702, + "step": 2564 + }, + { + "epoch": 2.8109589041095893, + "grad_norm": 2.621403217315674, + "learning_rate": 3.7977254264825345e-07, + "loss": 1.6529, + "step": 2565 + }, + { + "epoch": 2.812054794520548, + "grad_norm": 2.663494348526001, + "learning_rate": 3.7774167343623073e-07, + "loss": 1.7134, + "step": 2566 + }, + { + "epoch": 2.813150684931507, + "grad_norm": 2.6931262016296387, + "learning_rate": 3.75710804224208e-07, + "loss": 1.7277, + "step": 2567 + }, + { + "epoch": 2.814246575342466, + "grad_norm": 2.70450758934021, + "learning_rate": 3.7367993501218523e-07, + "loss": 1.7418, + "step": 2568 + }, + { + "epoch": 2.8153424657534245, + "grad_norm": 2.7160189151763916, + "learning_rate": 3.716490658001625e-07, + "loss": 1.6762, + "step": 2569 + }, + { + "epoch": 2.8164383561643835, + "grad_norm": 2.765512466430664, + "learning_rate": 3.696181965881398e-07, + "loss": 1.693, + "step": 2570 + }, + { + "epoch": 2.8164383561643835, + "eval_loss": 1.9258683919906616, + "eval_runtime": 37.2514, + "eval_samples_per_second": 26.845, + "eval_steps_per_second": 1.127, + "step": 2570 + }, + { + "epoch": 2.8175342465753426, + "grad_norm": 2.706547737121582, + "learning_rate": 3.67587327376117e-07, + "loss": 1.6424, + "step": 2571 + }, + { + "epoch": 2.818630136986301, + "grad_norm": 2.6400201320648193, + "learning_rate": 3.655564581640943e-07, + "loss": 1.6969, + "step": 2572 + }, + { + "epoch": 2.81972602739726, + "grad_norm": 2.6580376625061035, + "learning_rate": 3.6352558895207155e-07, + "loss": 1.7062, + "step": 2573 + }, + { + "epoch": 2.8208219178082192, + "grad_norm": 2.7226128578186035, + "learning_rate": 3.614947197400488e-07, + "loss": 1.6472, + "step": 2574 + }, + { + "epoch": 2.821917808219178, + "grad_norm": 2.5525615215301514, + "learning_rate": 3.5946385052802605e-07, + "loss": 1.7036, + "step": 2575 + }, + { + "epoch": 2.823013698630137, + "grad_norm": 2.747749090194702, + "learning_rate": 3.5743298131600333e-07, + "loss": 1.6979, + "step": 2576 + }, + { + "epoch": 2.824109589041096, + "grad_norm": 2.6271979808807373, + "learning_rate": 3.554021121039805e-07, + "loss": 1.6729, + "step": 2577 + }, + { + "epoch": 2.825205479452055, + "grad_norm": 2.8085596561431885, + "learning_rate": 3.533712428919578e-07, + "loss": 1.7494, + "step": 2578 + }, + { + "epoch": 2.826301369863014, + "grad_norm": 2.587289571762085, + "learning_rate": 3.5134037367993505e-07, + "loss": 1.6996, + "step": 2579 + }, + { + "epoch": 2.8273972602739725, + "grad_norm": 2.7216591835021973, + "learning_rate": 3.4930950446791227e-07, + "loss": 1.691, + "step": 2580 + }, + { + "epoch": 2.8273972602739725, + "eval_loss": 1.9263840913772583, + "eval_runtime": 37.169, + "eval_samples_per_second": 26.904, + "eval_steps_per_second": 1.13, + "step": 2580 + }, + { + "epoch": 2.8284931506849316, + "grad_norm": 2.6477060317993164, + "learning_rate": 3.4727863525588955e-07, + "loss": 1.6906, + "step": 2581 + }, + { + "epoch": 2.8295890410958906, + "grad_norm": 2.629814624786377, + "learning_rate": 3.452477660438668e-07, + "loss": 1.699, + "step": 2582 + }, + { + "epoch": 2.830684931506849, + "grad_norm": 2.6733274459838867, + "learning_rate": 3.4321689683184405e-07, + "loss": 1.736, + "step": 2583 + }, + { + "epoch": 2.831780821917808, + "grad_norm": 2.664885997772217, + "learning_rate": 3.411860276198213e-07, + "loss": 1.6923, + "step": 2584 + }, + { + "epoch": 2.8328767123287673, + "grad_norm": 2.544177532196045, + "learning_rate": 3.391551584077986e-07, + "loss": 1.665, + "step": 2585 + }, + { + "epoch": 2.833972602739726, + "grad_norm": 2.691823720932007, + "learning_rate": 3.3712428919577577e-07, + "loss": 1.7273, + "step": 2586 + }, + { + "epoch": 2.835068493150685, + "grad_norm": 2.6187548637390137, + "learning_rate": 3.3509341998375304e-07, + "loss": 1.7064, + "step": 2587 + }, + { + "epoch": 2.836164383561644, + "grad_norm": 2.701310873031616, + "learning_rate": 3.330625507717303e-07, + "loss": 1.6525, + "step": 2588 + }, + { + "epoch": 2.8372602739726025, + "grad_norm": 2.7446582317352295, + "learning_rate": 3.3103168155970754e-07, + "loss": 1.6849, + "step": 2589 + }, + { + "epoch": 2.8383561643835615, + "grad_norm": 2.686530113220215, + "learning_rate": 3.290008123476848e-07, + "loss": 1.7715, + "step": 2590 + }, + { + "epoch": 2.8383561643835615, + "eval_loss": 1.9255309104919434, + "eval_runtime": 37.1892, + "eval_samples_per_second": 26.89, + "eval_steps_per_second": 1.129, + "step": 2590 + }, + { + "epoch": 2.8394520547945206, + "grad_norm": 2.6728549003601074, + "learning_rate": 3.269699431356621e-07, + "loss": 1.7696, + "step": 2591 + }, + { + "epoch": 2.8405479452054796, + "grad_norm": 2.677741050720215, + "learning_rate": 3.249390739236393e-07, + "loss": 1.6749, + "step": 2592 + }, + { + "epoch": 2.8416438356164386, + "grad_norm": 2.633995532989502, + "learning_rate": 3.229082047116166e-07, + "loss": 1.6446, + "step": 2593 + }, + { + "epoch": 2.842739726027397, + "grad_norm": 2.6043264865875244, + "learning_rate": 3.2087733549959387e-07, + "loss": 1.7563, + "step": 2594 + }, + { + "epoch": 2.8438356164383563, + "grad_norm": 2.757563591003418, + "learning_rate": 3.1884646628757114e-07, + "loss": 1.7432, + "step": 2595 + }, + { + "epoch": 2.8449315068493153, + "grad_norm": 2.588893413543701, + "learning_rate": 3.1681559707554837e-07, + "loss": 1.7265, + "step": 2596 + }, + { + "epoch": 2.846027397260274, + "grad_norm": 2.6549947261810303, + "learning_rate": 3.1478472786352564e-07, + "loss": 1.7149, + "step": 2597 + }, + { + "epoch": 2.847123287671233, + "grad_norm": 2.570131540298462, + "learning_rate": 3.127538586515029e-07, + "loss": 1.655, + "step": 2598 + }, + { + "epoch": 2.848219178082192, + "grad_norm": 2.7245287895202637, + "learning_rate": 3.107229894394801e-07, + "loss": 1.6985, + "step": 2599 + }, + { + "epoch": 2.8493150684931505, + "grad_norm": 2.735067844390869, + "learning_rate": 3.0869212022745736e-07, + "loss": 1.6963, + "step": 2600 + }, + { + "epoch": 2.8493150684931505, + "eval_loss": 1.9269450902938843, + "eval_runtime": 37.1831, + "eval_samples_per_second": 26.894, + "eval_steps_per_second": 1.13, + "step": 2600 + } + ], + "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": 8.105004464340992e+17, + "train_batch_size": 3, + "trial_name": null, + "trial_params": null +} diff --git a/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-2600/zero_to_fp32.py b/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-2600/zero_to_fp32.py new file mode 100644 index 0000000000000000000000000000000000000000..24cc342e78d1a006c782b3a4cd68d9ce786d8fd8 --- /dev/null +++ b/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-2600/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/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-2700/rng_state_0.pth b/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-2700/rng_state_0.pth new file mode 100644 index 0000000000000000000000000000000000000000..12d8f134d0ace768aa9a4fc9a1c7787b2ce68b40 --- /dev/null +++ b/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-2700/rng_state_0.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:1fd745605ae8f150c3645104631b2d3c3b452eecda3ab1f66d0a3f474af022ff +size 16567 diff --git a/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-2700/rng_state_1.pth b/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-2700/rng_state_1.pth new file mode 100644 index 0000000000000000000000000000000000000000..7255d4b96382a76a01d62b0f3fb8ed7ba214a8cf --- /dev/null +++ b/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-2700/rng_state_1.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:cab5d049f3d3fc102f2df0c1651bc9d13107a3c0c772fd671f399b9d7d1de545 +size 16567 diff --git a/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-2700/rng_state_2.pth b/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-2700/rng_state_2.pth new file mode 100644 index 0000000000000000000000000000000000000000..cf2513c6be393ffd7f18b9966efe1ea94961ce76 --- /dev/null +++ b/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-2700/rng_state_2.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:1e03f5804d1958af28b457d1c95eadeccb6d55a564daf236124a0bd65854c268 +size 16567 diff --git a/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-2700/scheduler.pt b/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-2700/scheduler.pt new file mode 100644 index 0000000000000000000000000000000000000000..8caf38bf27eafc2d6fbde62c34932a03a613e4d7 --- /dev/null +++ b/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-2700/scheduler.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:5c9cc8f24cbcc65e106abf6b0ed79c412a415f21119804893567224bda4dd535 +size 627 diff --git a/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-2700/tokenizer.json b/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-2700/tokenizer.json new file mode 100644 index 0000000000000000000000000000000000000000..21e1edfa2183a57886f456b02efd11cce1a7df83 --- /dev/null +++ b/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-2700/tokenizer.json @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:e8485b8900f2ef3a891091dbe5ef0edc56f6922c0a23c7dfbb4baf64854f7de4 +size 17210102 diff --git a/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-300/config.json b/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-300/config.json new file mode 100644 index 0000000000000000000000000000000000000000..fb0f9a53a507ffcdb60410deb71d60ef801bf350 --- /dev/null +++ b/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-300/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/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-300/generation_config.json b/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-300/generation_config.json new file mode 100644 index 0000000000000000000000000000000000000000..9c389b841a9a29ddd904e02d1eb0e08dcce82ad9 --- /dev/null +++ b/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-300/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/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-300/latest b/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-300/latest new file mode 100644 index 0000000000000000000000000000000000000000..6761b575fffac7f1984044dcb6446b3a51da04c8 --- /dev/null +++ b/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-300/latest @@ -0,0 +1 @@ +global_step300 \ No newline at end of file diff --git a/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-300/model.safetensors.index.json b/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-300/model.safetensors.index.json new file mode 100644 index 0000000000000000000000000000000000000000..ed64de846d720b9a7859dc20575fea8e8ca51940 --- /dev/null +++ b/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-300/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/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-300/special_tokens_map.json b/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-300/special_tokens_map.json new file mode 100644 index 0000000000000000000000000000000000000000..6949c5975ee0e961ef61cf31010dce04df0a03f8 --- /dev/null +++ b/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-300/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/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-300/tokenizer_config.json b/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-300/tokenizer_config.json new file mode 100644 index 0000000000000000000000000000000000000000..5684b14e2756b56aa397c1324e239b442162b34a --- /dev/null +++ b/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-300/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/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-300/trainer_state.json b/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-300/trainer_state.json new file mode 100644 index 0000000000000000000000000000000000000000..d99afbdd23d53f64fd4c9320607263151722197e --- /dev/null +++ b/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-300/trainer_state.json @@ -0,0 +1,2373 @@ +{ + "best_metric": null, + "best_model_checkpoint": null, + "epoch": 0.3287671232876712, + "eval_steps": 10, + "global_step": 300, + "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": 3.0476, + "step": 1 + }, + { + "epoch": 0.002191780821917808, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.0289, + "step": 2 + }, + { + "epoch": 0.0032876712328767125, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 2.9935, + "step": 3 + }, + { + "epoch": 0.004383561643835616, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.0139, + "step": 4 + }, + { + "epoch": 0.005479452054794521, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 2.9892, + "step": 5 + }, + { + "epoch": 0.006575342465753425, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.1252, + "step": 6 + }, + { + "epoch": 0.007671232876712329, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.0291, + "step": 7 + }, + { + "epoch": 0.008767123287671232, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.0607, + "step": 8 + }, + { + "epoch": 0.009863013698630137, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.0306, + "step": 9 + }, + { + "epoch": 0.010958904109589041, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.0811, + "step": 10 + }, + { + "epoch": 0.010958904109589041, + "eval_loss": 3.0768930912017822, + "eval_runtime": 38.0377, + "eval_samples_per_second": 26.29, + "eval_steps_per_second": 1.104, + "step": 10 + }, + { + "epoch": 0.012054794520547946, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.0693, + "step": 11 + }, + { + "epoch": 0.01315068493150685, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 2.9668, + "step": 12 + }, + { + "epoch": 0.014246575342465753, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.052, + "step": 13 + }, + { + "epoch": 0.015342465753424657, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.0363, + "step": 14 + }, + { + "epoch": 0.01643835616438356, + "grad_norm": 5.58175802230835, + "learning_rate": 1.8248175182481753e-08, + "loss": 3.0778, + "step": 15 + }, + { + "epoch": 0.017534246575342465, + "grad_norm": 6.360531806945801, + "learning_rate": 3.6496350364963505e-08, + "loss": 3.0257, + "step": 16 + }, + { + "epoch": 0.01863013698630137, + "grad_norm": 6.453113555908203, + "learning_rate": 5.4744525547445264e-08, + "loss": 3.0262, + "step": 17 + }, + { + "epoch": 0.019726027397260273, + "grad_norm": 6.789405345916748, + "learning_rate": 7.299270072992701e-08, + "loss": 3.0671, + "step": 18 + }, + { + "epoch": 0.020821917808219178, + "grad_norm": 6.185514450073242, + "learning_rate": 9.124087591240876e-08, + "loss": 3.0077, + "step": 19 + }, + { + "epoch": 0.021917808219178082, + "grad_norm": 5.719798564910889, + "learning_rate": 1.0948905109489053e-07, + "loss": 3.089, + "step": 20 + }, + { + "epoch": 0.021917808219178082, + "eval_loss": 3.07546329498291, + "eval_runtime": 37.5318, + "eval_samples_per_second": 26.644, + "eval_steps_per_second": 1.119, + "step": 20 + }, + { + "epoch": 0.023013698630136987, + "grad_norm": 6.307768821716309, + "learning_rate": 1.2773722627737227e-07, + "loss": 3.0641, + "step": 21 + }, + { + "epoch": 0.02410958904109589, + "grad_norm": 6.255112171173096, + "learning_rate": 1.4598540145985402e-07, + "loss": 2.9849, + "step": 22 + }, + { + "epoch": 0.025205479452054796, + "grad_norm": 5.812101364135742, + "learning_rate": 1.642335766423358e-07, + "loss": 3.0211, + "step": 23 + }, + { + "epoch": 0.0263013698630137, + "grad_norm": 6.262075901031494, + "learning_rate": 1.8248175182481753e-07, + "loss": 3.0654, + "step": 24 + }, + { + "epoch": 0.0273972602739726, + "grad_norm": 5.419229507446289, + "learning_rate": 2.0072992700729928e-07, + "loss": 3.0661, + "step": 25 + }, + { + "epoch": 0.028493150684931506, + "grad_norm": 4.750737190246582, + "learning_rate": 2.1897810218978106e-07, + "loss": 3.0494, + "step": 26 + }, + { + "epoch": 0.02958904109589041, + "grad_norm": 4.732361316680908, + "learning_rate": 2.3722627737226278e-07, + "loss": 3.0471, + "step": 27 + }, + { + "epoch": 0.030684931506849315, + "grad_norm": 4.749786853790283, + "learning_rate": 2.5547445255474454e-07, + "loss": 3.0107, + "step": 28 + }, + { + "epoch": 0.03178082191780822, + "grad_norm": 4.451244831085205, + "learning_rate": 2.737226277372263e-07, + "loss": 2.9987, + "step": 29 + }, + { + "epoch": 0.03287671232876712, + "grad_norm": 4.5158257484436035, + "learning_rate": 2.9197080291970804e-07, + "loss": 2.9686, + "step": 30 + }, + { + "epoch": 0.03287671232876712, + "eval_loss": 3.031872034072876, + "eval_runtime": 37.2608, + "eval_samples_per_second": 26.838, + "eval_steps_per_second": 1.127, + "step": 30 + }, + { + "epoch": 0.03397260273972603, + "grad_norm": 3.8239498138427734, + "learning_rate": 3.102189781021898e-07, + "loss": 3.0057, + "step": 31 + }, + { + "epoch": 0.03506849315068493, + "grad_norm": 4.037254810333252, + "learning_rate": 3.284671532846716e-07, + "loss": 2.9733, + "step": 32 + }, + { + "epoch": 0.03616438356164384, + "grad_norm": 4.283931732177734, + "learning_rate": 3.467153284671533e-07, + "loss": 2.9515, + "step": 33 + }, + { + "epoch": 0.03726027397260274, + "grad_norm": 4.1784987449646, + "learning_rate": 3.6496350364963505e-07, + "loss": 2.9483, + "step": 34 + }, + { + "epoch": 0.038356164383561646, + "grad_norm": 3.9210753440856934, + "learning_rate": 3.8321167883211686e-07, + "loss": 2.9679, + "step": 35 + }, + { + "epoch": 0.03945205479452055, + "grad_norm": 3.72884202003479, + "learning_rate": 4.0145985401459856e-07, + "loss": 2.9989, + "step": 36 + }, + { + "epoch": 0.040547945205479455, + "grad_norm": 3.2882330417633057, + "learning_rate": 4.197080291970803e-07, + "loss": 2.9384, + "step": 37 + }, + { + "epoch": 0.041643835616438356, + "grad_norm": 3.515293836593628, + "learning_rate": 4.379562043795621e-07, + "loss": 2.8986, + "step": 38 + }, + { + "epoch": 0.042739726027397264, + "grad_norm": 3.568453073501587, + "learning_rate": 4.562043795620438e-07, + "loss": 2.9854, + "step": 39 + }, + { + "epoch": 0.043835616438356165, + "grad_norm": 3.389789342880249, + "learning_rate": 4.7445255474452557e-07, + "loss": 2.8891, + "step": 40 + }, + { + "epoch": 0.043835616438356165, + "eval_loss": 2.964235544204712, + "eval_runtime": 37.3406, + "eval_samples_per_second": 26.781, + "eval_steps_per_second": 1.125, + "step": 40 + }, + { + "epoch": 0.044931506849315066, + "grad_norm": 3.1826283931732178, + "learning_rate": 4.927007299270073e-07, + "loss": 2.9271, + "step": 41 + }, + { + "epoch": 0.046027397260273974, + "grad_norm": 2.941539764404297, + "learning_rate": 5.109489051094891e-07, + "loss": 2.9486, + "step": 42 + }, + { + "epoch": 0.047123287671232875, + "grad_norm": 3.0076773166656494, + "learning_rate": 5.291970802919708e-07, + "loss": 2.9252, + "step": 43 + }, + { + "epoch": 0.04821917808219178, + "grad_norm": 2.79028582572937, + "learning_rate": 5.474452554744526e-07, + "loss": 2.9277, + "step": 44 + }, + { + "epoch": 0.049315068493150684, + "grad_norm": 2.603456735610962, + "learning_rate": 5.656934306569343e-07, + "loss": 2.9144, + "step": 45 + }, + { + "epoch": 0.05041095890410959, + "grad_norm": 2.674888849258423, + "learning_rate": 5.839416058394161e-07, + "loss": 2.8732, + "step": 46 + }, + { + "epoch": 0.05150684931506849, + "grad_norm": 2.972093343734741, + "learning_rate": 6.021897810218979e-07, + "loss": 2.8623, + "step": 47 + }, + { + "epoch": 0.0526027397260274, + "grad_norm": 2.757766008377075, + "learning_rate": 6.204379562043796e-07, + "loss": 2.8941, + "step": 48 + }, + { + "epoch": 0.0536986301369863, + "grad_norm": 2.910634994506836, + "learning_rate": 6.386861313868614e-07, + "loss": 2.8101, + "step": 49 + }, + { + "epoch": 0.0547945205479452, + "grad_norm": 2.5542638301849365, + "learning_rate": 6.569343065693432e-07, + "loss": 2.9262, + "step": 50 + }, + { + "epoch": 0.0547945205479452, + "eval_loss": 2.883579969406128, + "eval_runtime": 37.326, + "eval_samples_per_second": 26.791, + "eval_steps_per_second": 1.125, + "step": 50 + }, + { + "epoch": 0.05589041095890411, + "grad_norm": 2.333974599838257, + "learning_rate": 6.751824817518248e-07, + "loss": 2.8217, + "step": 51 + }, + { + "epoch": 0.05698630136986301, + "grad_norm": 2.181022882461548, + "learning_rate": 6.934306569343066e-07, + "loss": 2.8483, + "step": 52 + }, + { + "epoch": 0.05808219178082192, + "grad_norm": 2.12200927734375, + "learning_rate": 7.116788321167883e-07, + "loss": 2.7788, + "step": 53 + }, + { + "epoch": 0.05917808219178082, + "grad_norm": 2.052626371383667, + "learning_rate": 7.299270072992701e-07, + "loss": 2.7774, + "step": 54 + }, + { + "epoch": 0.06027397260273973, + "grad_norm": 2.11487078666687, + "learning_rate": 7.48175182481752e-07, + "loss": 2.8045, + "step": 55 + }, + { + "epoch": 0.06136986301369863, + "grad_norm": 2.083588123321533, + "learning_rate": 7.664233576642337e-07, + "loss": 2.8119, + "step": 56 + }, + { + "epoch": 0.06246575342465754, + "grad_norm": 2.2079808712005615, + "learning_rate": 7.846715328467154e-07, + "loss": 2.82, + "step": 57 + }, + { + "epoch": 0.06356164383561644, + "grad_norm": 2.100802421569824, + "learning_rate": 8.029197080291971e-07, + "loss": 2.794, + "step": 58 + }, + { + "epoch": 0.06465753424657535, + "grad_norm": 1.8322066068649292, + "learning_rate": 8.211678832116789e-07, + "loss": 2.7444, + "step": 59 + }, + { + "epoch": 0.06575342465753424, + "grad_norm": 2.077005386352539, + "learning_rate": 8.394160583941606e-07, + "loss": 2.7093, + "step": 60 + }, + { + "epoch": 0.06575342465753424, + "eval_loss": 2.8177874088287354, + "eval_runtime": 37.2903, + "eval_samples_per_second": 26.817, + "eval_steps_per_second": 1.126, + "step": 60 + }, + { + "epoch": 0.06684931506849315, + "grad_norm": 2.1688592433929443, + "learning_rate": 8.576642335766425e-07, + "loss": 2.7446, + "step": 61 + }, + { + "epoch": 0.06794520547945206, + "grad_norm": 2.1709237098693848, + "learning_rate": 8.759124087591242e-07, + "loss": 2.736, + "step": 62 + }, + { + "epoch": 0.06904109589041096, + "grad_norm": 2.3578364849090576, + "learning_rate": 8.94160583941606e-07, + "loss": 2.7583, + "step": 63 + }, + { + "epoch": 0.07013698630136986, + "grad_norm": 1.9025483131408691, + "learning_rate": 9.124087591240876e-07, + "loss": 2.7651, + "step": 64 + }, + { + "epoch": 0.07123287671232877, + "grad_norm": 1.9450963735580444, + "learning_rate": 9.306569343065694e-07, + "loss": 2.7186, + "step": 65 + }, + { + "epoch": 0.07232876712328767, + "grad_norm": 1.957723617553711, + "learning_rate": 9.489051094890511e-07, + "loss": 2.7421, + "step": 66 + }, + { + "epoch": 0.07342465753424658, + "grad_norm": 1.9823631048202515, + "learning_rate": 9.67153284671533e-07, + "loss": 2.7258, + "step": 67 + }, + { + "epoch": 0.07452054794520548, + "grad_norm": 1.945495843887329, + "learning_rate": 9.854014598540146e-07, + "loss": 2.7733, + "step": 68 + }, + { + "epoch": 0.07561643835616438, + "grad_norm": 1.9684759378433228, + "learning_rate": 1.0036496350364965e-06, + "loss": 2.7041, + "step": 69 + }, + { + "epoch": 0.07671232876712329, + "grad_norm": 2.087336540222168, + "learning_rate": 1.0218978102189781e-06, + "loss": 2.6662, + "step": 70 + }, + { + "epoch": 0.07671232876712329, + "eval_loss": 2.7529091835021973, + "eval_runtime": 37.3122, + "eval_samples_per_second": 26.801, + "eval_steps_per_second": 1.126, + "step": 70 + }, + { + "epoch": 0.07780821917808219, + "grad_norm": 1.7790117263793945, + "learning_rate": 1.04014598540146e-06, + "loss": 2.7379, + "step": 71 + }, + { + "epoch": 0.0789041095890411, + "grad_norm": 2.2780861854553223, + "learning_rate": 1.0583941605839416e-06, + "loss": 2.6234, + "step": 72 + }, + { + "epoch": 0.08, + "grad_norm": 2.1144700050354004, + "learning_rate": 1.0766423357664233e-06, + "loss": 2.6248, + "step": 73 + }, + { + "epoch": 0.08109589041095891, + "grad_norm": 2.0247490406036377, + "learning_rate": 1.0948905109489052e-06, + "loss": 2.6456, + "step": 74 + }, + { + "epoch": 0.0821917808219178, + "grad_norm": 2.702547311782837, + "learning_rate": 1.113138686131387e-06, + "loss": 2.6759, + "step": 75 + }, + { + "epoch": 0.08328767123287671, + "grad_norm": 1.8208750486373901, + "learning_rate": 1.1313868613138687e-06, + "loss": 2.6845, + "step": 76 + }, + { + "epoch": 0.08438356164383562, + "grad_norm": 2.3829221725463867, + "learning_rate": 1.1496350364963505e-06, + "loss": 2.61, + "step": 77 + }, + { + "epoch": 0.08547945205479453, + "grad_norm": 2.580885887145996, + "learning_rate": 1.1678832116788322e-06, + "loss": 2.5896, + "step": 78 + }, + { + "epoch": 0.08657534246575342, + "grad_norm": 2.2477641105651855, + "learning_rate": 1.1861313868613138e-06, + "loss": 2.596, + "step": 79 + }, + { + "epoch": 0.08767123287671233, + "grad_norm": 2.1895623207092285, + "learning_rate": 1.2043795620437959e-06, + "loss": 2.5906, + "step": 80 + }, + { + "epoch": 0.08767123287671233, + "eval_loss": 2.6815176010131836, + "eval_runtime": 37.3342, + "eval_samples_per_second": 26.785, + "eval_steps_per_second": 1.125, + "step": 80 + }, + { + "epoch": 0.08876712328767124, + "grad_norm": 2.217090606689453, + "learning_rate": 1.2226277372262775e-06, + "loss": 2.6005, + "step": 81 + }, + { + "epoch": 0.08986301369863013, + "grad_norm": 2.4679362773895264, + "learning_rate": 1.2408759124087592e-06, + "loss": 2.6333, + "step": 82 + }, + { + "epoch": 0.09095890410958904, + "grad_norm": 2.2590575218200684, + "learning_rate": 1.259124087591241e-06, + "loss": 2.5814, + "step": 83 + }, + { + "epoch": 0.09205479452054795, + "grad_norm": 1.877977728843689, + "learning_rate": 1.2773722627737229e-06, + "loss": 2.5948, + "step": 84 + }, + { + "epoch": 0.09315068493150686, + "grad_norm": 2.1535534858703613, + "learning_rate": 1.2956204379562045e-06, + "loss": 2.5781, + "step": 85 + }, + { + "epoch": 0.09424657534246575, + "grad_norm": 2.5001144409179688, + "learning_rate": 1.3138686131386864e-06, + "loss": 2.5339, + "step": 86 + }, + { + "epoch": 0.09534246575342466, + "grad_norm": 2.3279049396514893, + "learning_rate": 1.332116788321168e-06, + "loss": 2.545, + "step": 87 + }, + { + "epoch": 0.09643835616438357, + "grad_norm": 2.2230734825134277, + "learning_rate": 1.3503649635036497e-06, + "loss": 2.4745, + "step": 88 + }, + { + "epoch": 0.09753424657534246, + "grad_norm": 2.8762636184692383, + "learning_rate": 1.3686131386861315e-06, + "loss": 2.516, + "step": 89 + }, + { + "epoch": 0.09863013698630137, + "grad_norm": 2.153506278991699, + "learning_rate": 1.3868613138686132e-06, + "loss": 2.4547, + "step": 90 + }, + { + "epoch": 0.09863013698630137, + "eval_loss": 2.603238344192505, + "eval_runtime": 37.4162, + "eval_samples_per_second": 26.726, + "eval_steps_per_second": 1.123, + "step": 90 + }, + { + "epoch": 0.09972602739726028, + "grad_norm": 2.4143614768981934, + "learning_rate": 1.405109489051095e-06, + "loss": 2.547, + "step": 91 + }, + { + "epoch": 0.10082191780821918, + "grad_norm": 2.354632616043091, + "learning_rate": 1.4233576642335767e-06, + "loss": 2.5009, + "step": 92 + }, + { + "epoch": 0.10191780821917808, + "grad_norm": 2.6605563163757324, + "learning_rate": 1.4416058394160583e-06, + "loss": 2.5304, + "step": 93 + }, + { + "epoch": 0.10301369863013699, + "grad_norm": 2.3724122047424316, + "learning_rate": 1.4598540145985402e-06, + "loss": 2.4925, + "step": 94 + }, + { + "epoch": 0.10410958904109589, + "grad_norm": 3.6323022842407227, + "learning_rate": 1.478102189781022e-06, + "loss": 2.4592, + "step": 95 + }, + { + "epoch": 0.1052054794520548, + "grad_norm": 2.3407089710235596, + "learning_rate": 1.496350364963504e-06, + "loss": 2.5223, + "step": 96 + }, + { + "epoch": 0.1063013698630137, + "grad_norm": 2.8238413333892822, + "learning_rate": 1.5145985401459856e-06, + "loss": 2.489, + "step": 97 + }, + { + "epoch": 0.1073972602739726, + "grad_norm": 2.6433658599853516, + "learning_rate": 1.5328467153284674e-06, + "loss": 2.5101, + "step": 98 + }, + { + "epoch": 0.10849315068493151, + "grad_norm": 2.562761068344116, + "learning_rate": 1.551094890510949e-06, + "loss": 2.4642, + "step": 99 + }, + { + "epoch": 0.1095890410958904, + "grad_norm": 2.4501149654388428, + "learning_rate": 1.5693430656934307e-06, + "loss": 2.4102, + "step": 100 + }, + { + "epoch": 0.1095890410958904, + "eval_loss": 2.5452804565429688, + "eval_runtime": 37.4224, + "eval_samples_per_second": 26.722, + "eval_steps_per_second": 1.122, + "step": 100 + }, + { + "epoch": 0.11068493150684931, + "grad_norm": 2.475864887237549, + "learning_rate": 1.5875912408759126e-06, + "loss": 2.4147, + "step": 101 + }, + { + "epoch": 0.11178082191780822, + "grad_norm": 2.558875322341919, + "learning_rate": 1.6058394160583942e-06, + "loss": 2.4694, + "step": 102 + }, + { + "epoch": 0.11287671232876713, + "grad_norm": 2.1085734367370605, + "learning_rate": 1.624087591240876e-06, + "loss": 2.5099, + "step": 103 + }, + { + "epoch": 0.11397260273972602, + "grad_norm": 2.410731792449951, + "learning_rate": 1.6423357664233577e-06, + "loss": 2.432, + "step": 104 + }, + { + "epoch": 0.11506849315068493, + "grad_norm": 2.2800745964050293, + "learning_rate": 1.6605839416058394e-06, + "loss": 2.3669, + "step": 105 + }, + { + "epoch": 0.11616438356164384, + "grad_norm": 2.2234933376312256, + "learning_rate": 1.6788321167883212e-06, + "loss": 2.4145, + "step": 106 + }, + { + "epoch": 0.11726027397260275, + "grad_norm": 2.3367040157318115, + "learning_rate": 1.6970802919708029e-06, + "loss": 2.4272, + "step": 107 + }, + { + "epoch": 0.11835616438356164, + "grad_norm": 2.7002811431884766, + "learning_rate": 1.715328467153285e-06, + "loss": 2.3478, + "step": 108 + }, + { + "epoch": 0.11945205479452055, + "grad_norm": 2.6628332138061523, + "learning_rate": 1.7335766423357666e-06, + "loss": 2.4096, + "step": 109 + }, + { + "epoch": 0.12054794520547946, + "grad_norm": 2.4174230098724365, + "learning_rate": 1.7518248175182485e-06, + "loss": 2.407, + "step": 110 + }, + { + "epoch": 0.12054794520547946, + "eval_loss": 2.491746664047241, + "eval_runtime": 37.2555, + "eval_samples_per_second": 26.842, + "eval_steps_per_second": 1.127, + "step": 110 + }, + { + "epoch": 0.12164383561643835, + "grad_norm": 2.24796986579895, + "learning_rate": 1.77007299270073e-06, + "loss": 2.4369, + "step": 111 + }, + { + "epoch": 0.12273972602739726, + "grad_norm": 2.720939874649048, + "learning_rate": 1.788321167883212e-06, + "loss": 2.4116, + "step": 112 + }, + { + "epoch": 0.12383561643835617, + "grad_norm": 2.6907670497894287, + "learning_rate": 1.8065693430656936e-06, + "loss": 2.4577, + "step": 113 + }, + { + "epoch": 0.12493150684931507, + "grad_norm": 2.559636354446411, + "learning_rate": 1.8248175182481753e-06, + "loss": 2.3999, + "step": 114 + }, + { + "epoch": 0.12602739726027398, + "grad_norm": 2.6730551719665527, + "learning_rate": 1.8430656934306571e-06, + "loss": 2.3428, + "step": 115 + }, + { + "epoch": 0.12712328767123288, + "grad_norm": 3.172179937362671, + "learning_rate": 1.8613138686131388e-06, + "loss": 2.3722, + "step": 116 + }, + { + "epoch": 0.12821917808219177, + "grad_norm": 2.6757171154022217, + "learning_rate": 1.8795620437956206e-06, + "loss": 2.3572, + "step": 117 + }, + { + "epoch": 0.1293150684931507, + "grad_norm": 2.5840766429901123, + "learning_rate": 1.8978102189781023e-06, + "loss": 2.353, + "step": 118 + }, + { + "epoch": 0.1304109589041096, + "grad_norm": 2.6248581409454346, + "learning_rate": 1.916058394160584e-06, + "loss": 2.36, + "step": 119 + }, + { + "epoch": 0.13150684931506848, + "grad_norm": 2.428516387939453, + "learning_rate": 1.934306569343066e-06, + "loss": 2.3821, + "step": 120 + }, + { + "epoch": 0.13150684931506848, + "eval_loss": 2.451565980911255, + "eval_runtime": 37.3669, + "eval_samples_per_second": 26.762, + "eval_steps_per_second": 1.124, + "step": 120 + }, + { + "epoch": 0.1326027397260274, + "grad_norm": 2.9956233501434326, + "learning_rate": 1.952554744525548e-06, + "loss": 2.2778, + "step": 121 + }, + { + "epoch": 0.1336986301369863, + "grad_norm": 2.4120893478393555, + "learning_rate": 1.9708029197080293e-06, + "loss": 2.325, + "step": 122 + }, + { + "epoch": 0.13479452054794522, + "grad_norm": 2.1410024166107178, + "learning_rate": 1.989051094890511e-06, + "loss": 2.3248, + "step": 123 + }, + { + "epoch": 0.1358904109589041, + "grad_norm": 2.923799514770508, + "learning_rate": 2.007299270072993e-06, + "loss": 2.327, + "step": 124 + }, + { + "epoch": 0.136986301369863, + "grad_norm": 3.0770339965820312, + "learning_rate": 2.025547445255475e-06, + "loss": 2.3847, + "step": 125 + }, + { + "epoch": 0.13808219178082193, + "grad_norm": 2.6408181190490723, + "learning_rate": 2.0437956204379563e-06, + "loss": 2.3196, + "step": 126 + }, + { + "epoch": 0.13917808219178082, + "grad_norm": 2.9127118587493896, + "learning_rate": 2.062043795620438e-06, + "loss": 2.3287, + "step": 127 + }, + { + "epoch": 0.14027397260273972, + "grad_norm": 2.9805052280426025, + "learning_rate": 2.08029197080292e-06, + "loss": 2.3243, + "step": 128 + }, + { + "epoch": 0.14136986301369864, + "grad_norm": 3.1342010498046875, + "learning_rate": 2.0985401459854014e-06, + "loss": 2.38, + "step": 129 + }, + { + "epoch": 0.14246575342465753, + "grad_norm": 2.678422451019287, + "learning_rate": 2.1167883211678833e-06, + "loss": 2.2727, + "step": 130 + }, + { + "epoch": 0.14246575342465753, + "eval_loss": 2.4187021255493164, + "eval_runtime": 37.3117, + "eval_samples_per_second": 26.801, + "eval_steps_per_second": 1.126, + "step": 130 + }, + { + "epoch": 0.14356164383561643, + "grad_norm": 2.9281492233276367, + "learning_rate": 2.135036496350365e-06, + "loss": 2.3327, + "step": 131 + }, + { + "epoch": 0.14465753424657535, + "grad_norm": 2.9592933654785156, + "learning_rate": 2.1532846715328466e-06, + "loss": 2.3674, + "step": 132 + }, + { + "epoch": 0.14575342465753424, + "grad_norm": 3.404761552810669, + "learning_rate": 2.171532846715329e-06, + "loss": 2.2679, + "step": 133 + }, + { + "epoch": 0.14684931506849316, + "grad_norm": 3.635571241378784, + "learning_rate": 2.1897810218978103e-06, + "loss": 2.2799, + "step": 134 + }, + { + "epoch": 0.14794520547945206, + "grad_norm": 3.9130170345306396, + "learning_rate": 2.208029197080292e-06, + "loss": 2.3263, + "step": 135 + }, + { + "epoch": 0.14904109589041095, + "grad_norm": 4.300438404083252, + "learning_rate": 2.226277372262774e-06, + "loss": 2.2969, + "step": 136 + }, + { + "epoch": 0.15013698630136987, + "grad_norm": 2.863511085510254, + "learning_rate": 2.244525547445256e-06, + "loss": 2.297, + "step": 137 + }, + { + "epoch": 0.15123287671232877, + "grad_norm": 4.910321235656738, + "learning_rate": 2.2627737226277373e-06, + "loss": 2.3269, + "step": 138 + }, + { + "epoch": 0.15232876712328766, + "grad_norm": 3.3323862552642822, + "learning_rate": 2.281021897810219e-06, + "loss": 2.3195, + "step": 139 + }, + { + "epoch": 0.15342465753424658, + "grad_norm": 3.2672693729400635, + "learning_rate": 2.299270072992701e-06, + "loss": 2.2797, + "step": 140 + }, + { + "epoch": 0.15342465753424658, + "eval_loss": 2.3943886756896973, + "eval_runtime": 37.3137, + "eval_samples_per_second": 26.8, + "eval_steps_per_second": 1.126, + "step": 140 + }, + { + "epoch": 0.15452054794520548, + "grad_norm": 3.9229774475097656, + "learning_rate": 2.3175182481751825e-06, + "loss": 2.2718, + "step": 141 + }, + { + "epoch": 0.15561643835616437, + "grad_norm": 2.6409823894500732, + "learning_rate": 2.3357664233576643e-06, + "loss": 2.2828, + "step": 142 + }, + { + "epoch": 0.1567123287671233, + "grad_norm": 4.6122612953186035, + "learning_rate": 2.354014598540146e-06, + "loss": 2.247, + "step": 143 + }, + { + "epoch": 0.1578082191780822, + "grad_norm": 3.391357898712158, + "learning_rate": 2.3722627737226276e-06, + "loss": 2.2863, + "step": 144 + }, + { + "epoch": 0.1589041095890411, + "grad_norm": 3.3550853729248047, + "learning_rate": 2.39051094890511e-06, + "loss": 2.2789, + "step": 145 + }, + { + "epoch": 0.16, + "grad_norm": 3.7376883029937744, + "learning_rate": 2.4087591240875918e-06, + "loss": 2.2297, + "step": 146 + }, + { + "epoch": 0.1610958904109589, + "grad_norm": 3.005157709121704, + "learning_rate": 2.427007299270073e-06, + "loss": 2.2467, + "step": 147 + }, + { + "epoch": 0.16219178082191782, + "grad_norm": 3.4616103172302246, + "learning_rate": 2.445255474452555e-06, + "loss": 2.2612, + "step": 148 + }, + { + "epoch": 0.1632876712328767, + "grad_norm": 3.3595101833343506, + "learning_rate": 2.463503649635037e-06, + "loss": 2.3021, + "step": 149 + }, + { + "epoch": 0.1643835616438356, + "grad_norm": 4.831241607666016, + "learning_rate": 2.4817518248175183e-06, + "loss": 2.2174, + "step": 150 + }, + { + "epoch": 0.1643835616438356, + "eval_loss": 2.3677749633789062, + "eval_runtime": 37.3092, + "eval_samples_per_second": 26.803, + "eval_steps_per_second": 1.126, + "step": 150 + }, + { + "epoch": 0.16547945205479453, + "grad_norm": 2.778820037841797, + "learning_rate": 2.5e-06, + "loss": 2.1856, + "step": 151 + }, + { + "epoch": 0.16657534246575342, + "grad_norm": 4.566643714904785, + "learning_rate": 2.518248175182482e-06, + "loss": 2.3186, + "step": 152 + }, + { + "epoch": 0.16767123287671232, + "grad_norm": 2.9908089637756348, + "learning_rate": 2.5364963503649635e-06, + "loss": 2.2127, + "step": 153 + }, + { + "epoch": 0.16876712328767124, + "grad_norm": 2.9702792167663574, + "learning_rate": 2.5547445255474458e-06, + "loss": 2.2071, + "step": 154 + }, + { + "epoch": 0.16986301369863013, + "grad_norm": 2.850044012069702, + "learning_rate": 2.5729927007299272e-06, + "loss": 2.2585, + "step": 155 + }, + { + "epoch": 0.17095890410958905, + "grad_norm": 2.926091432571411, + "learning_rate": 2.591240875912409e-06, + "loss": 2.2397, + "step": 156 + }, + { + "epoch": 0.17205479452054795, + "grad_norm": 2.8720169067382812, + "learning_rate": 2.6094890510948905e-06, + "loss": 2.2306, + "step": 157 + }, + { + "epoch": 0.17315068493150684, + "grad_norm": 2.8837177753448486, + "learning_rate": 2.627737226277373e-06, + "loss": 2.2666, + "step": 158 + }, + { + "epoch": 0.17424657534246576, + "grad_norm": 3.119621753692627, + "learning_rate": 2.6459854014598542e-06, + "loss": 2.2722, + "step": 159 + }, + { + "epoch": 0.17534246575342466, + "grad_norm": 2.7085700035095215, + "learning_rate": 2.664233576642336e-06, + "loss": 2.2346, + "step": 160 + }, + { + "epoch": 0.17534246575342466, + "eval_loss": 2.352083921432495, + "eval_runtime": 37.245, + "eval_samples_per_second": 26.849, + "eval_steps_per_second": 1.128, + "step": 160 + }, + { + "epoch": 0.17643835616438355, + "grad_norm": 2.9964346885681152, + "learning_rate": 2.6824817518248175e-06, + "loss": 2.2155, + "step": 161 + }, + { + "epoch": 0.17753424657534247, + "grad_norm": 3.1535208225250244, + "learning_rate": 2.7007299270072994e-06, + "loss": 2.235, + "step": 162 + }, + { + "epoch": 0.17863013698630137, + "grad_norm": 2.7240943908691406, + "learning_rate": 2.718978102189781e-06, + "loss": 2.2095, + "step": 163 + }, + { + "epoch": 0.17972602739726026, + "grad_norm": 3.122654676437378, + "learning_rate": 2.737226277372263e-06, + "loss": 2.2667, + "step": 164 + }, + { + "epoch": 0.18082191780821918, + "grad_norm": 2.9718410968780518, + "learning_rate": 2.755474452554745e-06, + "loss": 2.2499, + "step": 165 + }, + { + "epoch": 0.18191780821917808, + "grad_norm": 3.0081896781921387, + "learning_rate": 2.7737226277372264e-06, + "loss": 2.2089, + "step": 166 + }, + { + "epoch": 0.18301369863013697, + "grad_norm": 2.829509973526001, + "learning_rate": 2.7919708029197082e-06, + "loss": 2.3139, + "step": 167 + }, + { + "epoch": 0.1841095890410959, + "grad_norm": 3.4480397701263428, + "learning_rate": 2.81021897810219e-06, + "loss": 2.2705, + "step": 168 + }, + { + "epoch": 0.1852054794520548, + "grad_norm": 3.1107025146484375, + "learning_rate": 2.828467153284672e-06, + "loss": 2.1895, + "step": 169 + }, + { + "epoch": 0.1863013698630137, + "grad_norm": 2.916790008544922, + "learning_rate": 2.8467153284671534e-06, + "loss": 2.2317, + "step": 170 + }, + { + "epoch": 0.1863013698630137, + "eval_loss": 2.336153030395508, + "eval_runtime": 37.2401, + "eval_samples_per_second": 26.853, + "eval_steps_per_second": 1.128, + "step": 170 + }, + { + "epoch": 0.1873972602739726, + "grad_norm": 3.2582287788391113, + "learning_rate": 2.8649635036496353e-06, + "loss": 2.143, + "step": 171 + }, + { + "epoch": 0.1884931506849315, + "grad_norm": 2.430973768234253, + "learning_rate": 2.8832116788321167e-06, + "loss": 2.1977, + "step": 172 + }, + { + "epoch": 0.18958904109589042, + "grad_norm": 2.8763599395751953, + "learning_rate": 2.901459854014599e-06, + "loss": 2.1551, + "step": 173 + }, + { + "epoch": 0.19068493150684931, + "grad_norm": 2.9242653846740723, + "learning_rate": 2.9197080291970804e-06, + "loss": 2.2569, + "step": 174 + }, + { + "epoch": 0.1917808219178082, + "grad_norm": 2.623551607131958, + "learning_rate": 2.9379562043795623e-06, + "loss": 2.2329, + "step": 175 + }, + { + "epoch": 0.19287671232876713, + "grad_norm": 2.9511337280273438, + "learning_rate": 2.956204379562044e-06, + "loss": 2.2154, + "step": 176 + }, + { + "epoch": 0.19397260273972602, + "grad_norm": 2.415118932723999, + "learning_rate": 2.9744525547445256e-06, + "loss": 2.2637, + "step": 177 + }, + { + "epoch": 0.19506849315068492, + "grad_norm": 2.9664924144744873, + "learning_rate": 2.992700729927008e-06, + "loss": 2.2263, + "step": 178 + }, + { + "epoch": 0.19616438356164384, + "grad_norm": 2.7214572429656982, + "learning_rate": 3.0109489051094893e-06, + "loss": 2.1667, + "step": 179 + }, + { + "epoch": 0.19726027397260273, + "grad_norm": 2.8712823390960693, + "learning_rate": 3.029197080291971e-06, + "loss": 2.1661, + "step": 180 + }, + { + "epoch": 0.19726027397260273, + "eval_loss": 2.3191800117492676, + "eval_runtime": 37.24, + "eval_samples_per_second": 26.853, + "eval_steps_per_second": 1.128, + "step": 180 + }, + { + "epoch": 0.19835616438356166, + "grad_norm": 2.959709405899048, + "learning_rate": 3.0474452554744526e-06, + "loss": 2.2483, + "step": 181 + }, + { + "epoch": 0.19945205479452055, + "grad_norm": 2.567089319229126, + "learning_rate": 3.065693430656935e-06, + "loss": 2.2382, + "step": 182 + }, + { + "epoch": 0.20054794520547944, + "grad_norm": 2.47188401222229, + "learning_rate": 3.0839416058394163e-06, + "loss": 2.2148, + "step": 183 + }, + { + "epoch": 0.20164383561643837, + "grad_norm": 2.3435261249542236, + "learning_rate": 3.102189781021898e-06, + "loss": 2.2334, + "step": 184 + }, + { + "epoch": 0.20273972602739726, + "grad_norm": 2.5280468463897705, + "learning_rate": 3.1204379562043796e-06, + "loss": 2.2214, + "step": 185 + }, + { + "epoch": 0.20383561643835615, + "grad_norm": 2.4808971881866455, + "learning_rate": 3.1386861313868614e-06, + "loss": 2.1554, + "step": 186 + }, + { + "epoch": 0.20493150684931508, + "grad_norm": 2.9937026500701904, + "learning_rate": 3.156934306569343e-06, + "loss": 2.1389, + "step": 187 + }, + { + "epoch": 0.20602739726027397, + "grad_norm": 2.9220476150512695, + "learning_rate": 3.175182481751825e-06, + "loss": 2.2515, + "step": 188 + }, + { + "epoch": 0.20712328767123286, + "grad_norm": 3.0449001789093018, + "learning_rate": 3.193430656934307e-06, + "loss": 2.2016, + "step": 189 + }, + { + "epoch": 0.20821917808219179, + "grad_norm": 2.8027074337005615, + "learning_rate": 3.2116788321167884e-06, + "loss": 2.1823, + "step": 190 + }, + { + "epoch": 0.20821917808219179, + "eval_loss": 2.309677839279175, + "eval_runtime": 37.2442, + "eval_samples_per_second": 26.85, + "eval_steps_per_second": 1.128, + "step": 190 + }, + { + "epoch": 0.20931506849315068, + "grad_norm": 2.9158318042755127, + "learning_rate": 3.2299270072992707e-06, + "loss": 2.1928, + "step": 191 + }, + { + "epoch": 0.2104109589041096, + "grad_norm": 2.5328080654144287, + "learning_rate": 3.248175182481752e-06, + "loss": 2.1828, + "step": 192 + }, + { + "epoch": 0.2115068493150685, + "grad_norm": 3.1709704399108887, + "learning_rate": 3.266423357664234e-06, + "loss": 2.1052, + "step": 193 + }, + { + "epoch": 0.2126027397260274, + "grad_norm": 2.9156367778778076, + "learning_rate": 3.2846715328467155e-06, + "loss": 2.1943, + "step": 194 + }, + { + "epoch": 0.2136986301369863, + "grad_norm": 2.879408836364746, + "learning_rate": 3.3029197080291973e-06, + "loss": 2.1782, + "step": 195 + }, + { + "epoch": 0.2147945205479452, + "grad_norm": 3.748483180999756, + "learning_rate": 3.3211678832116788e-06, + "loss": 2.2107, + "step": 196 + }, + { + "epoch": 0.2158904109589041, + "grad_norm": 3.197510004043579, + "learning_rate": 3.339416058394161e-06, + "loss": 2.1845, + "step": 197 + }, + { + "epoch": 0.21698630136986302, + "grad_norm": 3.2596049308776855, + "learning_rate": 3.3576642335766425e-06, + "loss": 2.1816, + "step": 198 + }, + { + "epoch": 0.21808219178082192, + "grad_norm": 3.3329780101776123, + "learning_rate": 3.3759124087591243e-06, + "loss": 2.181, + "step": 199 + }, + { + "epoch": 0.2191780821917808, + "grad_norm": 2.7302327156066895, + "learning_rate": 3.3941605839416058e-06, + "loss": 2.1761, + "step": 200 + }, + { + "epoch": 0.2191780821917808, + "eval_loss": 2.2909653186798096, + "eval_runtime": 37.284, + "eval_samples_per_second": 26.821, + "eval_steps_per_second": 1.126, + "step": 200 + }, + { + "epoch": 0.22027397260273973, + "grad_norm": 3.002255916595459, + "learning_rate": 3.412408759124088e-06, + "loss": 2.1868, + "step": 201 + }, + { + "epoch": 0.22136986301369863, + "grad_norm": 2.6694891452789307, + "learning_rate": 3.43065693430657e-06, + "loss": 2.192, + "step": 202 + }, + { + "epoch": 0.22246575342465755, + "grad_norm": 2.9225986003875732, + "learning_rate": 3.4489051094890513e-06, + "loss": 2.1303, + "step": 203 + }, + { + "epoch": 0.22356164383561644, + "grad_norm": 2.7844080924987793, + "learning_rate": 3.467153284671533e-06, + "loss": 2.1788, + "step": 204 + }, + { + "epoch": 0.22465753424657534, + "grad_norm": 2.726123332977295, + "learning_rate": 3.4854014598540146e-06, + "loss": 2.1193, + "step": 205 + }, + { + "epoch": 0.22575342465753426, + "grad_norm": 2.838040590286255, + "learning_rate": 3.503649635036497e-06, + "loss": 2.1674, + "step": 206 + }, + { + "epoch": 0.22684931506849315, + "grad_norm": 2.6863937377929688, + "learning_rate": 3.5218978102189783e-06, + "loss": 2.161, + "step": 207 + }, + { + "epoch": 0.22794520547945205, + "grad_norm": 2.5116159915924072, + "learning_rate": 3.54014598540146e-06, + "loss": 2.179, + "step": 208 + }, + { + "epoch": 0.22904109589041097, + "grad_norm": 3.381863594055176, + "learning_rate": 3.5583941605839416e-06, + "loss": 2.2035, + "step": 209 + }, + { + "epoch": 0.23013698630136986, + "grad_norm": 3.2741382122039795, + "learning_rate": 3.576642335766424e-06, + "loss": 2.2518, + "step": 210 + }, + { + "epoch": 0.23013698630136986, + "eval_loss": 2.2773635387420654, + "eval_runtime": 37.2658, + "eval_samples_per_second": 26.834, + "eval_steps_per_second": 1.127, + "step": 210 + }, + { + "epoch": 0.23123287671232876, + "grad_norm": 2.917370319366455, + "learning_rate": 3.5948905109489054e-06, + "loss": 2.2265, + "step": 211 + }, + { + "epoch": 0.23232876712328768, + "grad_norm": 2.8392834663391113, + "learning_rate": 3.6131386861313872e-06, + "loss": 2.0709, + "step": 212 + }, + { + "epoch": 0.23342465753424657, + "grad_norm": 2.753612518310547, + "learning_rate": 3.6313868613138687e-06, + "loss": 2.0752, + "step": 213 + }, + { + "epoch": 0.2345205479452055, + "grad_norm": 3.1477746963500977, + "learning_rate": 3.6496350364963505e-06, + "loss": 2.1169, + "step": 214 + }, + { + "epoch": 0.2356164383561644, + "grad_norm": 2.5448100566864014, + "learning_rate": 3.667883211678833e-06, + "loss": 2.1338, + "step": 215 + }, + { + "epoch": 0.23671232876712328, + "grad_norm": 3.1622560024261475, + "learning_rate": 3.6861313868613142e-06, + "loss": 2.1065, + "step": 216 + }, + { + "epoch": 0.2378082191780822, + "grad_norm": 2.7054336071014404, + "learning_rate": 3.704379562043796e-06, + "loss": 2.1104, + "step": 217 + }, + { + "epoch": 0.2389041095890411, + "grad_norm": 2.964141845703125, + "learning_rate": 3.7226277372262775e-06, + "loss": 2.1949, + "step": 218 + }, + { + "epoch": 0.24, + "grad_norm": 2.7586112022399902, + "learning_rate": 3.7408759124087594e-06, + "loss": 2.165, + "step": 219 + }, + { + "epoch": 0.2410958904109589, + "grad_norm": 2.9035964012145996, + "learning_rate": 3.7591240875912412e-06, + "loss": 2.1192, + "step": 220 + }, + { + "epoch": 0.2410958904109589, + "eval_loss": 2.2649471759796143, + "eval_runtime": 37.4311, + "eval_samples_per_second": 26.716, + "eval_steps_per_second": 1.122, + "step": 220 + }, + { + "epoch": 0.2421917808219178, + "grad_norm": 2.4212076663970947, + "learning_rate": 3.777372262773723e-06, + "loss": 2.1389, + "step": 221 + }, + { + "epoch": 0.2432876712328767, + "grad_norm": 2.973227024078369, + "learning_rate": 3.7956204379562045e-06, + "loss": 2.132, + "step": 222 + }, + { + "epoch": 0.24438356164383562, + "grad_norm": 2.9831430912017822, + "learning_rate": 3.8138686131386864e-06, + "loss": 2.1142, + "step": 223 + }, + { + "epoch": 0.24547945205479452, + "grad_norm": 3.2880353927612305, + "learning_rate": 3.832116788321168e-06, + "loss": 2.168, + "step": 224 + }, + { + "epoch": 0.2465753424657534, + "grad_norm": 3.3920998573303223, + "learning_rate": 3.85036496350365e-06, + "loss": 2.1277, + "step": 225 + }, + { + "epoch": 0.24767123287671233, + "grad_norm": 2.7170026302337646, + "learning_rate": 3.868613138686132e-06, + "loss": 2.0936, + "step": 226 + }, + { + "epoch": 0.24876712328767123, + "grad_norm": 3.6220011711120605, + "learning_rate": 3.886861313868613e-06, + "loss": 2.1259, + "step": 227 + }, + { + "epoch": 0.24986301369863015, + "grad_norm": 3.4229273796081543, + "learning_rate": 3.905109489051096e-06, + "loss": 2.1421, + "step": 228 + }, + { + "epoch": 0.25095890410958904, + "grad_norm": 3.0358338356018066, + "learning_rate": 3.923357664233577e-06, + "loss": 2.1747, + "step": 229 + }, + { + "epoch": 0.25205479452054796, + "grad_norm": 2.611771821975708, + "learning_rate": 3.9416058394160585e-06, + "loss": 2.1282, + "step": 230 + }, + { + "epoch": 0.25205479452054796, + "eval_loss": 2.254887580871582, + "eval_runtime": 37.5943, + "eval_samples_per_second": 26.6, + "eval_steps_per_second": 1.117, + "step": 230 + }, + { + "epoch": 0.25315068493150683, + "grad_norm": 3.846412181854248, + "learning_rate": 3.95985401459854e-06, + "loss": 2.1892, + "step": 231 + }, + { + "epoch": 0.25424657534246575, + "grad_norm": 2.488525390625, + "learning_rate": 3.978102189781022e-06, + "loss": 2.2064, + "step": 232 + }, + { + "epoch": 0.2553424657534247, + "grad_norm": 3.114964485168457, + "learning_rate": 3.996350364963504e-06, + "loss": 2.0653, + "step": 233 + }, + { + "epoch": 0.25643835616438354, + "grad_norm": 2.5559768676757812, + "learning_rate": 4.014598540145986e-06, + "loss": 2.1503, + "step": 234 + }, + { + "epoch": 0.25753424657534246, + "grad_norm": 3.2347540855407715, + "learning_rate": 4.032846715328467e-06, + "loss": 2.0523, + "step": 235 + }, + { + "epoch": 0.2586301369863014, + "grad_norm": 2.8086111545562744, + "learning_rate": 4.05109489051095e-06, + "loss": 2.1206, + "step": 236 + }, + { + "epoch": 0.25972602739726025, + "grad_norm": 2.909231185913086, + "learning_rate": 4.069343065693431e-06, + "loss": 2.139, + "step": 237 + }, + { + "epoch": 0.2608219178082192, + "grad_norm": 3.2171082496643066, + "learning_rate": 4.0875912408759126e-06, + "loss": 2.0908, + "step": 238 + }, + { + "epoch": 0.2619178082191781, + "grad_norm": 2.3851265907287598, + "learning_rate": 4.105839416058395e-06, + "loss": 2.1133, + "step": 239 + }, + { + "epoch": 0.26301369863013696, + "grad_norm": 3.3912737369537354, + "learning_rate": 4.124087591240876e-06, + "loss": 2.1911, + "step": 240 + }, + { + "epoch": 0.26301369863013696, + "eval_loss": 2.2436957359313965, + "eval_runtime": 37.2947, + "eval_samples_per_second": 26.813, + "eval_steps_per_second": 1.126, + "step": 240 + }, + { + "epoch": 0.2641095890410959, + "grad_norm": 2.82906174659729, + "learning_rate": 4.1423357664233586e-06, + "loss": 2.11, + "step": 241 + }, + { + "epoch": 0.2652054794520548, + "grad_norm": 3.0317952632904053, + "learning_rate": 4.16058394160584e-06, + "loss": 2.0953, + "step": 242 + }, + { + "epoch": 0.2663013698630137, + "grad_norm": 3.6438329219818115, + "learning_rate": 4.1788321167883214e-06, + "loss": 2.1746, + "step": 243 + }, + { + "epoch": 0.2673972602739726, + "grad_norm": 2.6690592765808105, + "learning_rate": 4.197080291970803e-06, + "loss": 2.0555, + "step": 244 + }, + { + "epoch": 0.2684931506849315, + "grad_norm": 3.135376453399658, + "learning_rate": 4.215328467153285e-06, + "loss": 2.1394, + "step": 245 + }, + { + "epoch": 0.26958904109589044, + "grad_norm": 2.5474987030029297, + "learning_rate": 4.233576642335767e-06, + "loss": 2.0822, + "step": 246 + }, + { + "epoch": 0.2706849315068493, + "grad_norm": 2.883108377456665, + "learning_rate": 4.251824817518249e-06, + "loss": 2.0869, + "step": 247 + }, + { + "epoch": 0.2717808219178082, + "grad_norm": 2.9801037311553955, + "learning_rate": 4.27007299270073e-06, + "loss": 2.1194, + "step": 248 + }, + { + "epoch": 0.27287671232876715, + "grad_norm": 2.7688217163085938, + "learning_rate": 4.288321167883212e-06, + "loss": 2.0908, + "step": 249 + }, + { + "epoch": 0.273972602739726, + "grad_norm": 3.205148458480835, + "learning_rate": 4.306569343065693e-06, + "loss": 2.0762, + "step": 250 + }, + { + "epoch": 0.273972602739726, + "eval_loss": 2.238699197769165, + "eval_runtime": 37.2428, + "eval_samples_per_second": 26.851, + "eval_steps_per_second": 1.128, + "step": 250 + }, + { + "epoch": 0.27506849315068493, + "grad_norm": 2.8262946605682373, + "learning_rate": 4.3248175182481755e-06, + "loss": 2.0509, + "step": 251 + }, + { + "epoch": 0.27616438356164386, + "grad_norm": 3.156261682510376, + "learning_rate": 4.343065693430658e-06, + "loss": 2.1213, + "step": 252 + }, + { + "epoch": 0.2772602739726027, + "grad_norm": 2.6367998123168945, + "learning_rate": 4.361313868613139e-06, + "loss": 2.1071, + "step": 253 + }, + { + "epoch": 0.27835616438356164, + "grad_norm": 2.871534824371338, + "learning_rate": 4.379562043795621e-06, + "loss": 2.1008, + "step": 254 + }, + { + "epoch": 0.27945205479452057, + "grad_norm": 2.839432716369629, + "learning_rate": 4.397810218978102e-06, + "loss": 2.1293, + "step": 255 + }, + { + "epoch": 0.28054794520547943, + "grad_norm": 2.627143383026123, + "learning_rate": 4.416058394160584e-06, + "loss": 2.1141, + "step": 256 + }, + { + "epoch": 0.28164383561643835, + "grad_norm": 2.8292996883392334, + "learning_rate": 4.434306569343066e-06, + "loss": 2.0981, + "step": 257 + }, + { + "epoch": 0.2827397260273973, + "grad_norm": 2.5711069107055664, + "learning_rate": 4.452554744525548e-06, + "loss": 2.0707, + "step": 258 + }, + { + "epoch": 0.28383561643835614, + "grad_norm": 2.488954544067383, + "learning_rate": 4.4708029197080295e-06, + "loss": 2.1172, + "step": 259 + }, + { + "epoch": 0.28493150684931506, + "grad_norm": 2.442209005355835, + "learning_rate": 4.489051094890512e-06, + "loss": 2.1119, + "step": 260 + }, + { + "epoch": 0.28493150684931506, + "eval_loss": 2.2306911945343018, + "eval_runtime": 37.2412, + "eval_samples_per_second": 26.852, + "eval_steps_per_second": 1.128, + "step": 260 + }, + { + "epoch": 0.286027397260274, + "grad_norm": 2.9147934913635254, + "learning_rate": 4.507299270072993e-06, + "loss": 2.1517, + "step": 261 + }, + { + "epoch": 0.28712328767123285, + "grad_norm": 2.549833297729492, + "learning_rate": 4.525547445255475e-06, + "loss": 2.1047, + "step": 262 + }, + { + "epoch": 0.2882191780821918, + "grad_norm": 3.3405206203460693, + "learning_rate": 4.543795620437956e-06, + "loss": 2.0975, + "step": 263 + }, + { + "epoch": 0.2893150684931507, + "grad_norm": 2.554431676864624, + "learning_rate": 4.562043795620438e-06, + "loss": 2.0622, + "step": 264 + }, + { + "epoch": 0.29041095890410956, + "grad_norm": 3.5529873371124268, + "learning_rate": 4.580291970802921e-06, + "loss": 2.1567, + "step": 265 + }, + { + "epoch": 0.2915068493150685, + "grad_norm": 2.421682119369507, + "learning_rate": 4.598540145985402e-06, + "loss": 2.1619, + "step": 266 + }, + { + "epoch": 0.2926027397260274, + "grad_norm": 2.918288230895996, + "learning_rate": 4.6167883211678835e-06, + "loss": 2.1238, + "step": 267 + }, + { + "epoch": 0.2936986301369863, + "grad_norm": 3.0737621784210205, + "learning_rate": 4.635036496350365e-06, + "loss": 2.0728, + "step": 268 + }, + { + "epoch": 0.2947945205479452, + "grad_norm": 2.5862410068511963, + "learning_rate": 4.653284671532847e-06, + "loss": 2.0675, + "step": 269 + }, + { + "epoch": 0.2958904109589041, + "grad_norm": 2.9004979133605957, + "learning_rate": 4.671532846715329e-06, + "loss": 2.1248, + "step": 270 + }, + { + "epoch": 0.2958904109589041, + "eval_loss": 2.2179365158081055, + "eval_runtime": 37.2352, + "eval_samples_per_second": 26.856, + "eval_steps_per_second": 1.128, + "step": 270 + }, + { + "epoch": 0.29698630136986304, + "grad_norm": 2.4784982204437256, + "learning_rate": 4.689781021897811e-06, + "loss": 2.1276, + "step": 271 + }, + { + "epoch": 0.2980821917808219, + "grad_norm": 2.7971739768981934, + "learning_rate": 4.708029197080292e-06, + "loss": 2.0931, + "step": 272 + }, + { + "epoch": 0.2991780821917808, + "grad_norm": 2.5839481353759766, + "learning_rate": 4.726277372262774e-06, + "loss": 2.1371, + "step": 273 + }, + { + "epoch": 0.30027397260273975, + "grad_norm": 2.7168564796447754, + "learning_rate": 4.744525547445255e-06, + "loss": 2.1086, + "step": 274 + }, + { + "epoch": 0.3013698630136986, + "grad_norm": 2.6566879749298096, + "learning_rate": 4.7627737226277375e-06, + "loss": 2.0368, + "step": 275 + }, + { + "epoch": 0.30246575342465754, + "grad_norm": 2.5068323612213135, + "learning_rate": 4.78102189781022e-06, + "loss": 2.1077, + "step": 276 + }, + { + "epoch": 0.30356164383561646, + "grad_norm": 2.631408214569092, + "learning_rate": 4.799270072992701e-06, + "loss": 2.1013, + "step": 277 + }, + { + "epoch": 0.3046575342465753, + "grad_norm": 2.5642812252044678, + "learning_rate": 4.8175182481751835e-06, + "loss": 2.0739, + "step": 278 + }, + { + "epoch": 0.30575342465753425, + "grad_norm": 2.3326497077941895, + "learning_rate": 4.835766423357665e-06, + "loss": 2.0971, + "step": 279 + }, + { + "epoch": 0.30684931506849317, + "grad_norm": 2.70701265335083, + "learning_rate": 4.854014598540146e-06, + "loss": 2.0904, + "step": 280 + }, + { + "epoch": 0.30684931506849317, + "eval_loss": 2.215972900390625, + "eval_runtime": 37.2196, + "eval_samples_per_second": 26.868, + "eval_steps_per_second": 1.128, + "step": 280 + }, + { + "epoch": 0.30794520547945203, + "grad_norm": 2.615201234817505, + "learning_rate": 4.872262773722628e-06, + "loss": 2.039, + "step": 281 + }, + { + "epoch": 0.30904109589041096, + "grad_norm": 2.2821998596191406, + "learning_rate": 4.89051094890511e-06, + "loss": 2.0529, + "step": 282 + }, + { + "epoch": 0.3101369863013699, + "grad_norm": 2.767153024673462, + "learning_rate": 4.9087591240875915e-06, + "loss": 2.0351, + "step": 283 + }, + { + "epoch": 0.31123287671232874, + "grad_norm": 2.537121295928955, + "learning_rate": 4.927007299270074e-06, + "loss": 2.0752, + "step": 284 + }, + { + "epoch": 0.31232876712328766, + "grad_norm": 2.835052013397217, + "learning_rate": 4.945255474452555e-06, + "loss": 2.0892, + "step": 285 + }, + { + "epoch": 0.3134246575342466, + "grad_norm": 2.5154924392700195, + "learning_rate": 4.963503649635037e-06, + "loss": 2.0569, + "step": 286 + }, + { + "epoch": 0.31452054794520545, + "grad_norm": 2.8212568759918213, + "learning_rate": 4.981751824817518e-06, + "loss": 2.1588, + "step": 287 + }, + { + "epoch": 0.3156164383561644, + "grad_norm": 2.3840994834899902, + "learning_rate": 5e-06, + "loss": 2.1409, + "step": 288 + }, + { + "epoch": 0.3167123287671233, + "grad_norm": 2.9641315937042236, + "learning_rate": 4.997969130787978e-06, + "loss": 2.0892, + "step": 289 + }, + { + "epoch": 0.3178082191780822, + "grad_norm": 2.4516749382019043, + "learning_rate": 4.995938261575955e-06, + "loss": 2.1105, + "step": 290 + }, + { + "epoch": 0.3178082191780822, + "eval_loss": 2.207075834274292, + "eval_runtime": 37.1304, + "eval_samples_per_second": 26.932, + "eval_steps_per_second": 1.131, + "step": 290 + }, + { + "epoch": 0.3189041095890411, + "grad_norm": 2.768869400024414, + "learning_rate": 4.993907392363932e-06, + "loss": 2.0231, + "step": 291 + }, + { + "epoch": 0.32, + "grad_norm": 2.771475315093994, + "learning_rate": 4.9918765231519095e-06, + "loss": 2.1018, + "step": 292 + }, + { + "epoch": 0.32109589041095893, + "grad_norm": 2.773983955383301, + "learning_rate": 4.989845653939887e-06, + "loss": 2.0317, + "step": 293 + }, + { + "epoch": 0.3221917808219178, + "grad_norm": 2.4447343349456787, + "learning_rate": 4.987814784727864e-06, + "loss": 2.0739, + "step": 294 + }, + { + "epoch": 0.3232876712328767, + "grad_norm": 2.444261312484741, + "learning_rate": 4.985783915515841e-06, + "loss": 2.073, + "step": 295 + }, + { + "epoch": 0.32438356164383564, + "grad_norm": 2.871792793273926, + "learning_rate": 4.983753046303819e-06, + "loss": 2.058, + "step": 296 + }, + { + "epoch": 0.3254794520547945, + "grad_norm": 2.3651537895202637, + "learning_rate": 4.981722177091796e-06, + "loss": 2.0749, + "step": 297 + }, + { + "epoch": 0.3265753424657534, + "grad_norm": 2.3395845890045166, + "learning_rate": 4.979691307879773e-06, + "loss": 2.065, + "step": 298 + }, + { + "epoch": 0.32767123287671235, + "grad_norm": 2.6275217533111572, + "learning_rate": 4.9776604386677504e-06, + "loss": 2.0945, + "step": 299 + }, + { + "epoch": 0.3287671232876712, + "grad_norm": 2.866285562515259, + "learning_rate": 4.975629569455728e-06, + "loss": 2.0702, + "step": 300 + }, + { + "epoch": 0.3287671232876712, + "eval_loss": 2.200012683868408, + "eval_runtime": 37.2082, + "eval_samples_per_second": 26.876, + "eval_steps_per_second": 1.129, + "step": 300 + } + ], + "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": 9.35192822808576e+16, + "train_batch_size": 3, + "trial_name": null, + "trial_params": null +} diff --git a/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-300/zero_to_fp32.py b/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-300/zero_to_fp32.py new file mode 100644 index 0000000000000000000000000000000000000000..24cc342e78d1a006c782b3a4cd68d9ce786d8fd8 --- /dev/null +++ b/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-300/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/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-600/rng_state_0.pth b/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-600/rng_state_0.pth new file mode 100644 index 0000000000000000000000000000000000000000..d5f37008b65b2b6b774ba8ca3e08ea7f2886beeb --- /dev/null +++ b/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-600/rng_state_0.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:898e9b2b9cf458e184c0b3debd303a4fd7c10dad3b82242d58682e89cace8b1e +size 16567 diff --git a/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-600/rng_state_1.pth b/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-600/rng_state_1.pth new file mode 100644 index 0000000000000000000000000000000000000000..43698f33ac8bdb91266f12645eda8b36787da828 --- /dev/null +++ b/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-600/rng_state_1.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:050132a910188198ac7405dcc27a0ec731bdadfd7fd9c07e7468ac838c90fae0 +size 16567 diff --git a/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-600/rng_state_2.pth b/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-600/rng_state_2.pth new file mode 100644 index 0000000000000000000000000000000000000000..90d0b3f139f4cb1368a680fad6f12df4c66f078f --- /dev/null +++ b/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-600/rng_state_2.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:21ab24991874894fe74f6d6882102cbe2b931a423ab076862a8f74e33fa92f52 +size 16567 diff --git a/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-600/scheduler.pt b/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-600/scheduler.pt new file mode 100644 index 0000000000000000000000000000000000000000..4a88517ed88bc72e58abab86c72bf63fbafacd85 --- /dev/null +++ b/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-600/scheduler.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:d5ec84c03afb654562bc87ec6e7ade731cf26fc3b08c72d0a97b6b921259c65e +size 627 diff --git a/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-600/tokenizer.json b/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-600/tokenizer.json new file mode 100644 index 0000000000000000000000000000000000000000..21e1edfa2183a57886f456b02efd11cce1a7df83 --- /dev/null +++ b/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-600/tokenizer.json @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:e8485b8900f2ef3a891091dbe5ef0edc56f6922c0a23c7dfbb4baf64854f7de4 +size 17210102 diff --git a/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-600/training_args.bin b/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-600/training_args.bin new file mode 100644 index 0000000000000000000000000000000000000000..0628b4aecba57cabec5230705ebfea1d004cfc77 --- /dev/null +++ b/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-600/training_args.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:f1b22d390bcc8f3a3c3fa0ce7ebfd361692a465dd5cb0c24a819e263b73e1c2e +size 6011 diff --git a/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-900/rng_state_0.pth b/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-900/rng_state_0.pth new file mode 100644 index 0000000000000000000000000000000000000000..5a018a68bd5303fcda57dcb6093948dc236b697a --- /dev/null +++ b/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-900/rng_state_0.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:34bfd29981738a04d7cb9b0b31f032749b5bd5fa1b120104d856a6f5b060af88 +size 16567 diff --git a/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-900/rng_state_1.pth b/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-900/rng_state_1.pth new file mode 100644 index 0000000000000000000000000000000000000000..85d20c3dd29d7f71371d2ce18d57829a2ebd22bc --- /dev/null +++ b/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-900/rng_state_1.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:e9e7ca9eed1d152be37852a8850dde9294edfaf37097b42c75c54b278336ef61 +size 16567 diff --git a/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-900/rng_state_2.pth b/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-900/rng_state_2.pth new file mode 100644 index 0000000000000000000000000000000000000000..137d87c143da2b66811890c8cf1e8b013ef6eacf --- /dev/null +++ b/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-900/rng_state_2.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:62c45c6121ed4cfe9395ee5153d26d376e4f02f8dfa1bbceafb8a5416efcdd14 +size 16567 diff --git a/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-900/scheduler.pt b/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-900/scheduler.pt new file mode 100644 index 0000000000000000000000000000000000000000..86063610918de03fe10ff5915f245b21a80f8845 --- /dev/null +++ b/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-900/scheduler.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:977be84d836c6e1ba1cd1d60ad9c4b8ba9f4c0a63264746f477be2406b5629d7 +size 627 diff --git a/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-900/tokenizer.json b/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-900/tokenizer.json new file mode 100644 index 0000000000000000000000000000000000000000..21e1edfa2183a57886f456b02efd11cce1a7df83 --- /dev/null +++ b/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-900/tokenizer.json @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:e8485b8900f2ef3a891091dbe5ef0edc56f6922c0a23c7dfbb4baf64854f7de4 +size 17210102 diff --git a/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-900/training_args.bin b/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-900/training_args.bin new file mode 100644 index 0000000000000000000000000000000000000000..0628b4aecba57cabec5230705ebfea1d004cfc77 --- /dev/null +++ b/checkpoints/Llama-3.2-3B/babylm_shuffle_deterministic21_10M_seed0/runs/checkpoint-900/training_args.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:f1b22d390bcc8f3a3c3fa0ce7ebfd361692a465dd5cb0c24a819e263b73e1c2e +size 6011 diff --git a/checkpoints/Llama-3.2-3B/babylm_shuffle_even_odd_10M_seed0/runs/checkpoint-2100/model-00001-of-00002.safetensors b/checkpoints/Llama-3.2-3B/babylm_shuffle_even_odd_10M_seed0/runs/checkpoint-2100/model-00001-of-00002.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..4a56e145176183f4005d09bac11a4032610a3236 --- /dev/null +++ b/checkpoints/Llama-3.2-3B/babylm_shuffle_even_odd_10M_seed0/runs/checkpoint-2100/model-00001-of-00002.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:910d7b9c9dc00394bd66251a7e801cb2ffc322109a5a4b8b365d4a4fed0aeceb +size 4965798912 diff --git a/checkpoints/Llama-3.2-3B/babylm_shuffle_even_odd_10M_seed0/runs/checkpoint-2200/model-00001-of-00002.safetensors b/checkpoints/Llama-3.2-3B/babylm_shuffle_even_odd_10M_seed0/runs/checkpoint-2200/model-00001-of-00002.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..aee1ceac91a4d9377bfff76195e988c25f5a5420 --- /dev/null +++ b/checkpoints/Llama-3.2-3B/babylm_shuffle_even_odd_10M_seed0/runs/checkpoint-2200/model-00001-of-00002.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:31469791bea2815267e7644da17d4f2c4930a7ce001dbb62fae01b597a213973 +size 4965798912 diff --git a/checkpoints/Llama-3.2-3B/babylm_shuffle_even_odd_10M_seed0/runs/checkpoint-2300/model-00001-of-00002.safetensors b/checkpoints/Llama-3.2-3B/babylm_shuffle_even_odd_10M_seed0/runs/checkpoint-2300/model-00001-of-00002.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..9befcf11c94a03b1b17cf78275f5129a2f81ff5a --- /dev/null +++ b/checkpoints/Llama-3.2-3B/babylm_shuffle_even_odd_10M_seed0/runs/checkpoint-2300/model-00001-of-00002.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:6629e947cb1560d610b1ccb4bae543c040ffe43786114878afe962ad8969e13c +size 4965798912 diff --git a/checkpoints/Llama-3.2-3B/babylm_shuffle_even_odd_10M_seed0/runs/checkpoint-2400/model-00001-of-00002.safetensors b/checkpoints/Llama-3.2-3B/babylm_shuffle_even_odd_10M_seed0/runs/checkpoint-2400/model-00001-of-00002.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..8122ffbec40ec20ef924a97e43c48d8f656bfbb1 --- /dev/null +++ b/checkpoints/Llama-3.2-3B/babylm_shuffle_even_odd_10M_seed0/runs/checkpoint-2400/model-00001-of-00002.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:6a4db62df4df78a38acf1b83eb7a0e76b02ee6ba70330081c70035173c27dd99 +size 4965798912 diff --git a/checkpoints/Llama-3.2-3B/babylm_shuffle_even_odd_10M_seed0/runs/checkpoint-2500/model-00001-of-00002.safetensors b/checkpoints/Llama-3.2-3B/babylm_shuffle_even_odd_10M_seed0/runs/checkpoint-2500/model-00001-of-00002.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..fcce80babcf4fcf98bc755782691d91b0f630673 --- /dev/null +++ b/checkpoints/Llama-3.2-3B/babylm_shuffle_even_odd_10M_seed0/runs/checkpoint-2500/model-00001-of-00002.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:3c14930712290f165691c5f8ad35b87796c0ea951f1d0c0576dc6d015bec8677 +size 4965798912 diff --git a/checkpoints/Llama-3.2-3B/babylm_shuffle_even_odd_10M_seed0/runs/checkpoint-2600/model-00001-of-00002.safetensors b/checkpoints/Llama-3.2-3B/babylm_shuffle_even_odd_10M_seed0/runs/checkpoint-2600/model-00001-of-00002.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..2caf9998beba0adfcd43db7592bf874ad78de226 --- /dev/null +++ b/checkpoints/Llama-3.2-3B/babylm_shuffle_even_odd_10M_seed0/runs/checkpoint-2600/model-00001-of-00002.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:66e751abfbf18756bb61a8a58611e4e4f341c2f33137b88f270eb7ee66edb0ab +size 4965798912 diff --git a/checkpoints/Llama-3.2-3B/babylm_shuffle_even_odd_10M_seed0/runs/checkpoint-300/model-00001-of-00002.safetensors b/checkpoints/Llama-3.2-3B/babylm_shuffle_even_odd_10M_seed0/runs/checkpoint-300/model-00001-of-00002.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..d639ba0da48cb78334c0f706fe680b780a30e14f --- /dev/null +++ b/checkpoints/Llama-3.2-3B/babylm_shuffle_even_odd_10M_seed0/runs/checkpoint-300/model-00001-of-00002.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:54c29fc5e9d7182dcbbbc86ea6d34614b195c5fb57964e00f9699b96ecf8d9e0 +size 4965798912