Auto upload zain 2026-08-12T21:54:25.586111
Browse files- zain/Activation/out/glu-silu-100L_run/checkpoint-1400/config.json +36 -0
- zain/Activation/out/glu-silu-100L_run/checkpoint-1400/model.safetensors +3 -0
- zain/Activation/out/glu-silu-100L_run/checkpoint-1400/optimizer.pt +3 -0
- zain/Activation/out/glu-silu-100L_run/checkpoint-1400/rng_state.pth +3 -0
- zain/Activation/out/glu-silu-100L_run/checkpoint-1400/scheduler.pt +3 -0
- zain/Activation/out/glu-silu-100L_run/checkpoint-1400/tokenizer.json +0 -0
- zain/Activation/out/glu-silu-100L_run/checkpoint-1400/tokenizer_config.json +13 -0
- zain/Activation/out/glu-silu-100L_run/checkpoint-1400/trainer_state.json +748 -0
- zain/Activation/out/glu-silu-100L_run/checkpoint-1400/training_args.bin +3 -0
- zain/Activation/out/glu-silu-100L_run/checkpoint-1500/config.json +36 -0
- zain/Activation/out/glu-silu-100L_run/checkpoint-1500/model.safetensors +3 -0
- zain/Activation/out/glu-silu-100L_run/checkpoint-1500/optimizer.pt +3 -0
- zain/Activation/out/glu-silu-100L_run/checkpoint-1500/rng_state.pth +3 -0
- zain/Activation/out/glu-silu-100L_run/checkpoint-1500/scheduler.pt +3 -0
- zain/Activation/out/glu-silu-100L_run/checkpoint-1500/tokenizer.json +0 -0
- zain/Activation/out/glu-silu-100L_run/checkpoint-1500/tokenizer_config.json +13 -0
- zain/Activation/out/glu-silu-100L_run/checkpoint-1500/trainer_state.json +799 -0
- zain/Activation/out/glu-silu-100L_run/checkpoint-1500/training_args.bin +3 -0
- zain/Activation/out/glu-silu-100L_run/checkpoint-1600/config.json +36 -0
- zain/Activation/out/glu-silu-100L_run/checkpoint-1600/model.safetensors +3 -0
- zain/Activation/out/glu-silu-100L_run/checkpoint-1600/optimizer.pt +3 -0
- zain/Activation/out/glu-silu-100L_run/checkpoint-1600/rng_state.pth +3 -0
- zain/Activation/out/glu-silu-100L_run/checkpoint-1600/scheduler.pt +3 -0
- zain/Activation/out/glu-silu-100L_run/checkpoint-1600/tokenizer.json +0 -0
- zain/Activation/out/glu-silu-100L_run/checkpoint-1600/tokenizer_config.json +13 -0
- zain/Activation/out/glu-silu-100L_run/checkpoint-1600/trainer_state.json +850 -0
- zain/Activation/out/glu-silu-100L_run/checkpoint-1600/training_args.bin +3 -0
- zain/Activation/out/glu-silu-100L_run/checkpoint-1700/config.json +36 -0
- zain/Activation/out/glu-silu-100L_run/checkpoint-1700/model.safetensors +3 -0
- zain/Activation/out/glu-silu-100L_run/checkpoint-1700/optimizer.pt +3 -0
- zain/Activation/out/glu-silu-100L_run/checkpoint-1700/rng_state.pth +3 -0
- zain/Activation/out/glu-silu-100L_run/checkpoint-1700/scheduler.pt +3 -0
- zain/Activation/out/glu-silu-100L_run/checkpoint-1700/tokenizer.json +0 -0
- zain/Activation/out/glu-silu-100L_run/checkpoint-1700/tokenizer_config.json +13 -0
- zain/Activation/out/glu-silu-100L_run/checkpoint-1700/trainer_state.json +901 -0
- zain/Activation/out/glu-silu-100L_run/checkpoint-1700/training_args.bin +3 -0
- zain/Activation/out/glu-silu-100L_run/checkpoint-1800/config.json +36 -0
- zain/Activation/out/glu-silu-100L_run/checkpoint-1800/model.safetensors +3 -0
- zain/Activation/out/glu-silu-100L_run/checkpoint-1800/optimizer.pt +3 -0
- zain/Activation/out/glu-silu-100L_run/checkpoint-1800/rng_state.pth +3 -0
- zain/Activation/out/glu-silu-100L_run/checkpoint-1800/scheduler.pt +3 -0
- zain/Activation/out/glu-silu-100L_run/checkpoint-1800/tokenizer.json +0 -0
- zain/Activation/out/glu-silu-100L_run/checkpoint-1800/tokenizer_config.json +13 -0
- zain/Activation/out/glu-silu-100L_run/checkpoint-1800/trainer_state.json +952 -0
- zain/Activation/out/glu-silu-100L_run/checkpoint-1800/training_args.bin +3 -0
- zain/Activation/out/glu-silu-100L_run/training_log.jsonl +36 -0
- zain/Activation/wandb/debug-internal.log +42 -0
- zain/Activation/wandb/run-20260812_213624-1sryr5l2/files/output.log +62 -1
- zain/Activation/wandb/run-20260812_213624-1sryr5l2/logs/debug-internal.log +42 -0
- zain/Activation/wandb/run-20260812_213624-1sryr5l2/run-1sryr5l2.wandb +2 -2
zain/Activation/out/glu-silu-100L_run/checkpoint-1400/config.json
ADDED
|
@@ -0,0 +1,36 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"activation": "silu",
|
| 3 |
+
"architectures": [
|
| 4 |
+
"TinyLlamaForCausalLM"
|
| 5 |
+
],
|
| 6 |
+
"attention_bias": false,
|
| 7 |
+
"attention_dropout": 0.0,
|
| 8 |
+
"bos_token_id": 1,
|
| 9 |
+
"dtype": "bfloat16",
|
| 10 |
+
"eos_token_id": 2,
|
| 11 |
+
"head_dim": 32,
|
| 12 |
+
"hidden_act": "silu",
|
| 13 |
+
"hidden_size": 128,
|
| 14 |
+
"initializer_range": 0.02,
|
| 15 |
+
"intermediate_size": 256,
|
| 16 |
+
"max_position_embeddings": 512,
|
| 17 |
+
"mlp_bias": false,
|
| 18 |
+
"mlp_type": "glu",
|
| 19 |
+
"model_type": "tiny_llama",
|
| 20 |
+
"num_attention_heads": 4,
|
| 21 |
+
"num_hidden_layers": 100,
|
| 22 |
+
"num_key_value_heads": 4,
|
| 23 |
+
"pad_token_id": 0,
|
| 24 |
+
"pretraining_tp": 1,
|
| 25 |
+
"rms_norm_eps": 1e-06,
|
| 26 |
+
"rope_parameters": {
|
| 27 |
+
"rope_theta": 10000.0,
|
| 28 |
+
"rope_type": "default"
|
| 29 |
+
},
|
| 30 |
+
"tie_word_embeddings": true,
|
| 31 |
+
"tokenizer_name": "w-ahmad/tiny-stories-tokenizer",
|
| 32 |
+
"transformers_version": "5.16.0.dev0",
|
| 33 |
+
"use_cache": false,
|
| 34 |
+
"vocab_size": 4096,
|
| 35 |
+
"waleed_beta": 10.0
|
| 36 |
+
}
|
zain/Activation/out/glu-silu-100L_run/checkpoint-1400/model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:822477b49eef76ece4fd80c222a9e20632ca5b9b3baa15271ff09e5dc29b13d1
|
| 3 |
+
size 33967272
|
zain/Activation/out/glu-silu-100L_run/checkpoint-1400/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:cbd0e5a4c60f7555d68a984cbffc3d92aa3095f37ce3b633343acbeb3562da0d
|
| 3 |
+
size 68504996
|
zain/Activation/out/glu-silu-100L_run/checkpoint-1400/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:52d4940820055e4d79b82c43038fd1599197353607649f88dcebf26376772128
|
| 3 |
+
size 14244
|
zain/Activation/out/glu-silu-100L_run/checkpoint-1400/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:c7b5f8184ea93ad1b1d686071b34fd1fe8500efdf9d9cf11c4738b7a0da36fbb
|
| 3 |
+
size 1064
|
zain/Activation/out/glu-silu-100L_run/checkpoint-1400/tokenizer.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
zain/Activation/out/glu-silu-100L_run/checkpoint-1400/tokenizer_config.json
ADDED
|
@@ -0,0 +1,13 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"add_prefix_space": false,
|
| 3 |
+
"backend": "tokenizers",
|
| 4 |
+
"bos_token": "<|endoftext|>",
|
| 5 |
+
"eos_token": "<|endoftext|>",
|
| 6 |
+
"errors": "replace",
|
| 7 |
+
"is_local": false,
|
| 8 |
+
"local_files_only": false,
|
| 9 |
+
"model_max_length": 1024,
|
| 10 |
+
"pad_token": "<|endoftext|>",
|
| 11 |
+
"tokenizer_class": "GPT2Tokenizer",
|
| 12 |
+
"unk_token": "<|endoftext|>"
|
| 13 |
+
}
|
zain/Activation/out/glu-silu-100L_run/checkpoint-1400/trainer_state.json
ADDED
|
@@ -0,0 +1,748 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_global_step": null,
|
| 3 |
+
"best_metric": null,
|
| 4 |
+
"best_model_checkpoint": null,
|
| 5 |
+
"epoch": 0.09437141894169195,
|
| 6 |
+
"eval_steps": 50,
|
| 7 |
+
"global_step": 1400,
|
| 8 |
+
"is_hyper_param_search": false,
|
| 9 |
+
"is_local_process_zero": true,
|
| 10 |
+
"is_world_process_zero": true,
|
| 11 |
+
"log_history": [
|
| 12 |
+
{
|
| 13 |
+
"epoch": 0.0013481631277384564,
|
| 14 |
+
"grad_norm": 2.078125,
|
| 15 |
+
"learning_rate": 2.66e-05,
|
| 16 |
+
"loss": 8.29962387084961,
|
| 17 |
+
"step": 20
|
| 18 |
+
},
|
| 19 |
+
{
|
| 20 |
+
"epoch": 0.002696326255476913,
|
| 21 |
+
"grad_norm": 1.3515625,
|
| 22 |
+
"learning_rate": 5.46e-05,
|
| 23 |
+
"loss": 8.047650909423828,
|
| 24 |
+
"step": 40
|
| 25 |
+
},
|
| 26 |
+
{
|
| 27 |
+
"epoch": 0.003370407819346141,
|
| 28 |
+
"eval_loss": 7.835535049438477,
|
| 29 |
+
"eval_runtime": 15.2165,
|
| 30 |
+
"eval_samples_per_second": 626.095,
|
| 31 |
+
"eval_steps_per_second": 0.657,
|
| 32 |
+
"step": 50
|
| 33 |
+
},
|
| 34 |
+
{
|
| 35 |
+
"epoch": 0.004044489383215369,
|
| 36 |
+
"grad_norm": 1.2734375,
|
| 37 |
+
"learning_rate": 8.259999999999999e-05,
|
| 38 |
+
"loss": 7.8378746032714846,
|
| 39 |
+
"step": 60
|
| 40 |
+
},
|
| 41 |
+
{
|
| 42 |
+
"epoch": 0.005392652510953826,
|
| 43 |
+
"grad_norm": 1.265625,
|
| 44 |
+
"learning_rate": 0.0001106,
|
| 45 |
+
"loss": 7.5728271484375,
|
| 46 |
+
"step": 80
|
| 47 |
+
},
|
| 48 |
+
{
|
| 49 |
+
"epoch": 0.006740815638692282,
|
| 50 |
+
"grad_norm": 1.1953125,
|
| 51 |
+
"learning_rate": 0.0001386,
|
| 52 |
+
"loss": 7.2332916259765625,
|
| 53 |
+
"step": 100
|
| 54 |
+
},
|
| 55 |
+
{
|
| 56 |
+
"epoch": 0.006740815638692282,
|
| 57 |
+
"eval_loss": 7.039053440093994,
|
| 58 |
+
"eval_runtime": 15.2074,
|
| 59 |
+
"eval_samples_per_second": 626.471,
|
| 60 |
+
"eval_steps_per_second": 0.658,
|
| 61 |
+
"step": 100
|
| 62 |
+
},
|
| 63 |
+
{
|
| 64 |
+
"epoch": 0.008088978766430738,
|
| 65 |
+
"grad_norm": 1.1328125,
|
| 66 |
+
"learning_rate": 0.00016659999999999998,
|
| 67 |
+
"loss": 6.8608039855957035,
|
| 68 |
+
"step": 120
|
| 69 |
+
},
|
| 70 |
+
{
|
| 71 |
+
"epoch": 0.009437141894169195,
|
| 72 |
+
"grad_norm": 0.94921875,
|
| 73 |
+
"learning_rate": 0.00019460000000000001,
|
| 74 |
+
"loss": 6.517842864990234,
|
| 75 |
+
"step": 140
|
| 76 |
+
},
|
| 77 |
+
{
|
| 78 |
+
"epoch": 0.010111223458038422,
|
| 79 |
+
"eval_loss": 6.229617595672607,
|
| 80 |
+
"eval_runtime": 15.1978,
|
| 81 |
+
"eval_samples_per_second": 626.867,
|
| 82 |
+
"eval_steps_per_second": 0.658,
|
| 83 |
+
"step": 150
|
| 84 |
+
},
|
| 85 |
+
{
|
| 86 |
+
"epoch": 0.010785305021907651,
|
| 87 |
+
"grad_norm": 0.80078125,
|
| 88 |
+
"learning_rate": 0.0002226,
|
| 89 |
+
"loss": 6.2399169921875,
|
| 90 |
+
"step": 160
|
| 91 |
+
},
|
| 92 |
+
{
|
| 93 |
+
"epoch": 0.012133468149646108,
|
| 94 |
+
"grad_norm": 0.6640625,
|
| 95 |
+
"learning_rate": 0.00025059999999999997,
|
| 96 |
+
"loss": 6.036548614501953,
|
| 97 |
+
"step": 180
|
| 98 |
+
},
|
| 99 |
+
{
|
| 100 |
+
"epoch": 0.013481631277384564,
|
| 101 |
+
"grad_norm": 0.61328125,
|
| 102 |
+
"learning_rate": 0.0002786,
|
| 103 |
+
"loss": 5.857078552246094,
|
| 104 |
+
"step": 200
|
| 105 |
+
},
|
| 106 |
+
{
|
| 107 |
+
"epoch": 0.013481631277384564,
|
| 108 |
+
"eval_loss": 5.765108585357666,
|
| 109 |
+
"eval_runtime": 15.5271,
|
| 110 |
+
"eval_samples_per_second": 613.574,
|
| 111 |
+
"eval_steps_per_second": 0.644,
|
| 112 |
+
"step": 200
|
| 113 |
+
},
|
| 114 |
+
{
|
| 115 |
+
"epoch": 0.01482979440512302,
|
| 116 |
+
"grad_norm": 0.61328125,
|
| 117 |
+
"learning_rate": 0.00030659999999999997,
|
| 118 |
+
"loss": 5.680564117431641,
|
| 119 |
+
"step": 220
|
| 120 |
+
},
|
| 121 |
+
{
|
| 122 |
+
"epoch": 0.016177957532861477,
|
| 123 |
+
"grad_norm": 2.0625,
|
| 124 |
+
"learning_rate": 0.0003346,
|
| 125 |
+
"loss": 5.472259902954102,
|
| 126 |
+
"step": 240
|
| 127 |
+
},
|
| 128 |
+
{
|
| 129 |
+
"epoch": 0.016852039096730706,
|
| 130 |
+
"eval_loss": 5.240335464477539,
|
| 131 |
+
"eval_runtime": 15.2596,
|
| 132 |
+
"eval_samples_per_second": 624.33,
|
| 133 |
+
"eval_steps_per_second": 0.655,
|
| 134 |
+
"step": 250
|
| 135 |
+
},
|
| 136 |
+
{
|
| 137 |
+
"epoch": 0.01752612066059993,
|
| 138 |
+
"grad_norm": 1.7421875,
|
| 139 |
+
"learning_rate": 0.00036260000000000003,
|
| 140 |
+
"loss": 5.24955940246582,
|
| 141 |
+
"step": 260
|
| 142 |
+
},
|
| 143 |
+
{
|
| 144 |
+
"epoch": 0.01887428378833839,
|
| 145 |
+
"grad_norm": 0.62109375,
|
| 146 |
+
"learning_rate": 0.0003906,
|
| 147 |
+
"loss": 4.995759963989258,
|
| 148 |
+
"step": 280
|
| 149 |
+
},
|
| 150 |
+
{
|
| 151 |
+
"epoch": 0.020222446916076844,
|
| 152 |
+
"grad_norm": 1.390625,
|
| 153 |
+
"learning_rate": 0.0004186,
|
| 154 |
+
"loss": 4.769353866577148,
|
| 155 |
+
"step": 300
|
| 156 |
+
},
|
| 157 |
+
{
|
| 158 |
+
"epoch": 0.020222446916076844,
|
| 159 |
+
"eval_loss": 4.669629096984863,
|
| 160 |
+
"eval_runtime": 15.2595,
|
| 161 |
+
"eval_samples_per_second": 624.332,
|
| 162 |
+
"eval_steps_per_second": 0.655,
|
| 163 |
+
"step": 300
|
| 164 |
+
},
|
| 165 |
+
{
|
| 166 |
+
"epoch": 0.021570610043815303,
|
| 167 |
+
"grad_norm": 0.73828125,
|
| 168 |
+
"learning_rate": 0.0004466,
|
| 169 |
+
"loss": 4.569867324829102,
|
| 170 |
+
"step": 320
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 0.022918773171553757,
|
| 174 |
+
"grad_norm": 0.765625,
|
| 175 |
+
"learning_rate": 0.00047460000000000004,
|
| 176 |
+
"loss": 4.389446258544922,
|
| 177 |
+
"step": 340
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 0.023592854735422986,
|
| 181 |
+
"eval_loss": 4.231207847595215,
|
| 182 |
+
"eval_runtime": 15.2143,
|
| 183 |
+
"eval_samples_per_second": 626.188,
|
| 184 |
+
"eval_steps_per_second": 0.657,
|
| 185 |
+
"step": 350
|
| 186 |
+
},
|
| 187 |
+
{
|
| 188 |
+
"epoch": 0.024266936299292215,
|
| 189 |
+
"grad_norm": 1.2421875,
|
| 190 |
+
"learning_rate": 0.0005026,
|
| 191 |
+
"loss": 4.243848419189453,
|
| 192 |
+
"step": 360
|
| 193 |
+
},
|
| 194 |
+
{
|
| 195 |
+
"epoch": 0.02561509942703067,
|
| 196 |
+
"grad_norm": 0.7109375,
|
| 197 |
+
"learning_rate": 0.0005306,
|
| 198 |
+
"loss": 4.1090648651123045,
|
| 199 |
+
"step": 380
|
| 200 |
+
},
|
| 201 |
+
{
|
| 202 |
+
"epoch": 0.026963262554769128,
|
| 203 |
+
"grad_norm": 0.484375,
|
| 204 |
+
"learning_rate": 0.0005586,
|
| 205 |
+
"loss": 3.966172790527344,
|
| 206 |
+
"step": 400
|
| 207 |
+
},
|
| 208 |
+
{
|
| 209 |
+
"epoch": 0.026963262554769128,
|
| 210 |
+
"eval_loss": 3.923773765563965,
|
| 211 |
+
"eval_runtime": 15.5308,
|
| 212 |
+
"eval_samples_per_second": 613.426,
|
| 213 |
+
"eval_steps_per_second": 0.644,
|
| 214 |
+
"step": 400
|
| 215 |
+
},
|
| 216 |
+
{
|
| 217 |
+
"epoch": 0.028311425682507583,
|
| 218 |
+
"grad_norm": 0.6640625,
|
| 219 |
+
"learning_rate": 0.0005866,
|
| 220 |
+
"loss": 3.8559597015380858,
|
| 221 |
+
"step": 420
|
| 222 |
+
},
|
| 223 |
+
{
|
| 224 |
+
"epoch": 0.02965958881024604,
|
| 225 |
+
"grad_norm": 0.515625,
|
| 226 |
+
"learning_rate": 0.0006146,
|
| 227 |
+
"loss": 3.7943866729736326,
|
| 228 |
+
"step": 440
|
| 229 |
+
},
|
| 230 |
+
{
|
| 231 |
+
"epoch": 0.030333670374115267,
|
| 232 |
+
"eval_loss": 3.6977553367614746,
|
| 233 |
+
"eval_runtime": 15.2001,
|
| 234 |
+
"eval_samples_per_second": 626.77,
|
| 235 |
+
"eval_steps_per_second": 0.658,
|
| 236 |
+
"step": 450
|
| 237 |
+
},
|
| 238 |
+
{
|
| 239 |
+
"epoch": 0.031007751937984496,
|
| 240 |
+
"grad_norm": 0.51953125,
|
| 241 |
+
"learning_rate": 0.0006426,
|
| 242 |
+
"loss": 3.686443328857422,
|
| 243 |
+
"step": 460
|
| 244 |
+
},
|
| 245 |
+
{
|
| 246 |
+
"epoch": 0.032355915065722954,
|
| 247 |
+
"grad_norm": 0.62109375,
|
| 248 |
+
"learning_rate": 0.0006705999999999999,
|
| 249 |
+
"loss": 3.6123157501220704,
|
| 250 |
+
"step": 480
|
| 251 |
+
},
|
| 252 |
+
{
|
| 253 |
+
"epoch": 0.03370407819346141,
|
| 254 |
+
"grad_norm": 0.70703125,
|
| 255 |
+
"learning_rate": 0.0006986,
|
| 256 |
+
"loss": 3.542654800415039,
|
| 257 |
+
"step": 500
|
| 258 |
+
},
|
| 259 |
+
{
|
| 260 |
+
"epoch": 0.03370407819346141,
|
| 261 |
+
"eval_loss": 3.513820171356201,
|
| 262 |
+
"eval_runtime": 15.2193,
|
| 263 |
+
"eval_samples_per_second": 625.983,
|
| 264 |
+
"eval_steps_per_second": 0.657,
|
| 265 |
+
"step": 500
|
| 266 |
+
},
|
| 267 |
+
{
|
| 268 |
+
"epoch": 0.03505224132119986,
|
| 269 |
+
"grad_norm": 0.48046875,
|
| 270 |
+
"learning_rate": 0.0007,
|
| 271 |
+
"loss": 3.472199249267578,
|
| 272 |
+
"step": 520
|
| 273 |
+
},
|
| 274 |
+
{
|
| 275 |
+
"epoch": 0.03640040444893832,
|
| 276 |
+
"grad_norm": 0.53515625,
|
| 277 |
+
"learning_rate": 0.0007,
|
| 278 |
+
"loss": 3.396291732788086,
|
| 279 |
+
"step": 540
|
| 280 |
+
},
|
| 281 |
+
{
|
| 282 |
+
"epoch": 0.03707448601280755,
|
| 283 |
+
"eval_loss": 3.3581483364105225,
|
| 284 |
+
"eval_runtime": 15.3044,
|
| 285 |
+
"eval_samples_per_second": 622.502,
|
| 286 |
+
"eval_steps_per_second": 0.653,
|
| 287 |
+
"step": 550
|
| 288 |
+
},
|
| 289 |
+
{
|
| 290 |
+
"epoch": 0.03774856757667678,
|
| 291 |
+
"grad_norm": 0.47265625,
|
| 292 |
+
"learning_rate": 0.0007,
|
| 293 |
+
"loss": 3.347300720214844,
|
| 294 |
+
"step": 560
|
| 295 |
+
},
|
| 296 |
+
{
|
| 297 |
+
"epoch": 0.03909673070441524,
|
| 298 |
+
"grad_norm": 0.458984375,
|
| 299 |
+
"learning_rate": 0.0007,
|
| 300 |
+
"loss": 3.2885902404785154,
|
| 301 |
+
"step": 580
|
| 302 |
+
},
|
| 303 |
+
{
|
| 304 |
+
"epoch": 0.04044489383215369,
|
| 305 |
+
"grad_norm": 0.4375,
|
| 306 |
+
"learning_rate": 0.0007,
|
| 307 |
+
"loss": 3.25723876953125,
|
| 308 |
+
"step": 600
|
| 309 |
+
},
|
| 310 |
+
{
|
| 311 |
+
"epoch": 0.04044489383215369,
|
| 312 |
+
"eval_loss": 3.2229573726654053,
|
| 313 |
+
"eval_runtime": 15.2967,
|
| 314 |
+
"eval_samples_per_second": 622.812,
|
| 315 |
+
"eval_steps_per_second": 0.654,
|
| 316 |
+
"step": 600
|
| 317 |
+
},
|
| 318 |
+
{
|
| 319 |
+
"epoch": 0.04179305695989215,
|
| 320 |
+
"grad_norm": 0.458984375,
|
| 321 |
+
"learning_rate": 0.0007,
|
| 322 |
+
"loss": 3.1902334213256838,
|
| 323 |
+
"step": 620
|
| 324 |
+
},
|
| 325 |
+
{
|
| 326 |
+
"epoch": 0.043141220087630605,
|
| 327 |
+
"grad_norm": 0.40625,
|
| 328 |
+
"learning_rate": 0.0007,
|
| 329 |
+
"loss": 3.167861557006836,
|
| 330 |
+
"step": 640
|
| 331 |
+
},
|
| 332 |
+
{
|
| 333 |
+
"epoch": 0.043815301651499834,
|
| 334 |
+
"eval_loss": 3.120636463165283,
|
| 335 |
+
"eval_runtime": 15.3751,
|
| 336 |
+
"eval_samples_per_second": 619.639,
|
| 337 |
+
"eval_steps_per_second": 0.65,
|
| 338 |
+
"step": 650
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 0.044489383215369056,
|
| 342 |
+
"grad_norm": 0.494140625,
|
| 343 |
+
"learning_rate": 0.0007,
|
| 344 |
+
"loss": 3.1089324951171875,
|
| 345 |
+
"step": 660
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 0.045837546343107514,
|
| 349 |
+
"grad_norm": 0.435546875,
|
| 350 |
+
"learning_rate": 0.0007,
|
| 351 |
+
"loss": 3.0828268051147463,
|
| 352 |
+
"step": 680
|
| 353 |
+
},
|
| 354 |
+
{
|
| 355 |
+
"epoch": 0.04718570947084597,
|
| 356 |
+
"grad_norm": 0.427734375,
|
| 357 |
+
"learning_rate": 0.0007,
|
| 358 |
+
"loss": 3.0396488189697264,
|
| 359 |
+
"step": 700
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"epoch": 0.04718570947084597,
|
| 363 |
+
"eval_loss": 3.0276408195495605,
|
| 364 |
+
"eval_runtime": 15.2735,
|
| 365 |
+
"eval_samples_per_second": 623.761,
|
| 366 |
+
"eval_steps_per_second": 0.655,
|
| 367 |
+
"step": 700
|
| 368 |
+
},
|
| 369 |
+
{
|
| 370 |
+
"epoch": 0.04853387259858443,
|
| 371 |
+
"grad_norm": 0.4375,
|
| 372 |
+
"learning_rate": 0.0007,
|
| 373 |
+
"loss": 3.0034805297851563,
|
| 374 |
+
"step": 720
|
| 375 |
+
},
|
| 376 |
+
{
|
| 377 |
+
"epoch": 0.04988203572632288,
|
| 378 |
+
"grad_norm": 0.52734375,
|
| 379 |
+
"learning_rate": 0.0007,
|
| 380 |
+
"loss": 2.96556396484375,
|
| 381 |
+
"step": 740
|
| 382 |
+
},
|
| 383 |
+
{
|
| 384 |
+
"epoch": 0.05055611729019211,
|
| 385 |
+
"eval_loss": 2.950009346008301,
|
| 386 |
+
"eval_runtime": 15.243,
|
| 387 |
+
"eval_samples_per_second": 625.009,
|
| 388 |
+
"eval_steps_per_second": 0.656,
|
| 389 |
+
"step": 750
|
| 390 |
+
},
|
| 391 |
+
{
|
| 392 |
+
"epoch": 0.05123019885406134,
|
| 393 |
+
"grad_norm": 0.451171875,
|
| 394 |
+
"learning_rate": 0.0007,
|
| 395 |
+
"loss": 2.950779151916504,
|
| 396 |
+
"step": 760
|
| 397 |
+
},
|
| 398 |
+
{
|
| 399 |
+
"epoch": 0.0525783619817998,
|
| 400 |
+
"grad_norm": 0.400390625,
|
| 401 |
+
"learning_rate": 0.0007,
|
| 402 |
+
"loss": 2.956203079223633,
|
| 403 |
+
"step": 780
|
| 404 |
+
},
|
| 405 |
+
{
|
| 406 |
+
"epoch": 0.053926525109538256,
|
| 407 |
+
"grad_norm": 0.40625,
|
| 408 |
+
"learning_rate": 0.0007,
|
| 409 |
+
"loss": 2.9017578125,
|
| 410 |
+
"step": 800
|
| 411 |
+
},
|
| 412 |
+
{
|
| 413 |
+
"epoch": 0.053926525109538256,
|
| 414 |
+
"eval_loss": 2.8844006061553955,
|
| 415 |
+
"eval_runtime": 15.2903,
|
| 416 |
+
"eval_samples_per_second": 623.073,
|
| 417 |
+
"eval_steps_per_second": 0.654,
|
| 418 |
+
"step": 800
|
| 419 |
+
},
|
| 420 |
+
{
|
| 421 |
+
"epoch": 0.05527468823727671,
|
| 422 |
+
"grad_norm": 0.41015625,
|
| 423 |
+
"learning_rate": 0.0007,
|
| 424 |
+
"loss": 2.887515640258789,
|
| 425 |
+
"step": 820
|
| 426 |
+
},
|
| 427 |
+
{
|
| 428 |
+
"epoch": 0.056622851365015166,
|
| 429 |
+
"grad_norm": 0.41796875,
|
| 430 |
+
"learning_rate": 0.0007,
|
| 431 |
+
"loss": 2.8556928634643555,
|
| 432 |
+
"step": 840
|
| 433 |
+
},
|
| 434 |
+
{
|
| 435 |
+
"epoch": 0.057296932928884395,
|
| 436 |
+
"eval_loss": 2.821610450744629,
|
| 437 |
+
"eval_runtime": 15.2253,
|
| 438 |
+
"eval_samples_per_second": 625.733,
|
| 439 |
+
"eval_steps_per_second": 0.657,
|
| 440 |
+
"step": 850
|
| 441 |
+
},
|
| 442 |
+
{
|
| 443 |
+
"epoch": 0.057971014492753624,
|
| 444 |
+
"grad_norm": 0.423828125,
|
| 445 |
+
"learning_rate": 0.0007,
|
| 446 |
+
"loss": 2.8130855560302734,
|
| 447 |
+
"step": 860
|
| 448 |
+
},
|
| 449 |
+
{
|
| 450 |
+
"epoch": 0.05931917762049208,
|
| 451 |
+
"grad_norm": 0.38671875,
|
| 452 |
+
"learning_rate": 0.0007,
|
| 453 |
+
"loss": 2.790329170227051,
|
| 454 |
+
"step": 880
|
| 455 |
+
},
|
| 456 |
+
{
|
| 457 |
+
"epoch": 0.06066734074823053,
|
| 458 |
+
"grad_norm": 0.427734375,
|
| 459 |
+
"learning_rate": 0.0007,
|
| 460 |
+
"loss": 2.7760162353515625,
|
| 461 |
+
"step": 900
|
| 462 |
+
},
|
| 463 |
+
{
|
| 464 |
+
"epoch": 0.06066734074823053,
|
| 465 |
+
"eval_loss": 2.7661373615264893,
|
| 466 |
+
"eval_runtime": 15.2736,
|
| 467 |
+
"eval_samples_per_second": 623.755,
|
| 468 |
+
"eval_steps_per_second": 0.655,
|
| 469 |
+
"step": 900
|
| 470 |
+
},
|
| 471 |
+
{
|
| 472 |
+
"epoch": 0.06201550387596899,
|
| 473 |
+
"grad_norm": 0.435546875,
|
| 474 |
+
"learning_rate": 0.0007,
|
| 475 |
+
"loss": 2.7462047576904296,
|
| 476 |
+
"step": 920
|
| 477 |
+
},
|
| 478 |
+
{
|
| 479 |
+
"epoch": 0.06336366700370745,
|
| 480 |
+
"grad_norm": 0.4140625,
|
| 481 |
+
"learning_rate": 0.0007,
|
| 482 |
+
"loss": 2.732274627685547,
|
| 483 |
+
"step": 940
|
| 484 |
+
},
|
| 485 |
+
{
|
| 486 |
+
"epoch": 0.06403774856757667,
|
| 487 |
+
"eval_loss": 2.713942289352417,
|
| 488 |
+
"eval_runtime": 15.6062,
|
| 489 |
+
"eval_samples_per_second": 610.463,
|
| 490 |
+
"eval_steps_per_second": 0.641,
|
| 491 |
+
"step": 950
|
| 492 |
+
},
|
| 493 |
+
{
|
| 494 |
+
"epoch": 0.06471183013144591,
|
| 495 |
+
"grad_norm": 0.419921875,
|
| 496 |
+
"learning_rate": 0.0007,
|
| 497 |
+
"loss": 2.705318069458008,
|
| 498 |
+
"step": 960
|
| 499 |
+
},
|
| 500 |
+
{
|
| 501 |
+
"epoch": 0.06605999325918437,
|
| 502 |
+
"grad_norm": 0.4296875,
|
| 503 |
+
"learning_rate": 0.0007,
|
| 504 |
+
"loss": 2.6892818450927733,
|
| 505 |
+
"step": 980
|
| 506 |
+
},
|
| 507 |
+
{
|
| 508 |
+
"epoch": 0.06740815638692282,
|
| 509 |
+
"grad_norm": 0.4453125,
|
| 510 |
+
"learning_rate": 0.0007,
|
| 511 |
+
"loss": 2.687115287780762,
|
| 512 |
+
"step": 1000
|
| 513 |
+
},
|
| 514 |
+
{
|
| 515 |
+
"epoch": 0.06740815638692282,
|
| 516 |
+
"eval_loss": 2.668647289276123,
|
| 517 |
+
"eval_runtime": 15.2597,
|
| 518 |
+
"eval_samples_per_second": 624.326,
|
| 519 |
+
"eval_steps_per_second": 0.655,
|
| 520 |
+
"step": 1000
|
| 521 |
+
},
|
| 522 |
+
{
|
| 523 |
+
"epoch": 0.06875631951466127,
|
| 524 |
+
"grad_norm": 0.388671875,
|
| 525 |
+
"learning_rate": 0.0007,
|
| 526 |
+
"loss": 2.6533071517944338,
|
| 527 |
+
"step": 1020
|
| 528 |
+
},
|
| 529 |
+
{
|
| 530 |
+
"epoch": 0.07010448264239973,
|
| 531 |
+
"grad_norm": 0.392578125,
|
| 532 |
+
"learning_rate": 0.0007,
|
| 533 |
+
"loss": 2.6390922546386717,
|
| 534 |
+
"step": 1040
|
| 535 |
+
},
|
| 536 |
+
{
|
| 537 |
+
"epoch": 0.07077856420626896,
|
| 538 |
+
"eval_loss": 2.6291816234588623,
|
| 539 |
+
"eval_runtime": 15.2457,
|
| 540 |
+
"eval_samples_per_second": 624.896,
|
| 541 |
+
"eval_steps_per_second": 0.656,
|
| 542 |
+
"step": 1050
|
| 543 |
+
},
|
| 544 |
+
{
|
| 545 |
+
"epoch": 0.07145264577013818,
|
| 546 |
+
"grad_norm": 0.4375,
|
| 547 |
+
"learning_rate": 0.0007,
|
| 548 |
+
"loss": 2.624461555480957,
|
| 549 |
+
"step": 1060
|
| 550 |
+
},
|
| 551 |
+
{
|
| 552 |
+
"epoch": 0.07280080889787664,
|
| 553 |
+
"grad_norm": 0.396484375,
|
| 554 |
+
"learning_rate": 0.0007,
|
| 555 |
+
"loss": 2.6044427871704103,
|
| 556 |
+
"step": 1080
|
| 557 |
+
},
|
| 558 |
+
{
|
| 559 |
+
"epoch": 0.0741489720256151,
|
| 560 |
+
"grad_norm": 0.423828125,
|
| 561 |
+
"learning_rate": 0.0007,
|
| 562 |
+
"loss": 2.597646713256836,
|
| 563 |
+
"step": 1100
|
| 564 |
+
},
|
| 565 |
+
{
|
| 566 |
+
"epoch": 0.0741489720256151,
|
| 567 |
+
"eval_loss": 2.589696168899536,
|
| 568 |
+
"eval_runtime": 15.3967,
|
| 569 |
+
"eval_samples_per_second": 618.768,
|
| 570 |
+
"eval_steps_per_second": 0.649,
|
| 571 |
+
"step": 1100
|
| 572 |
+
},
|
| 573 |
+
{
|
| 574 |
+
"epoch": 0.07549713515335356,
|
| 575 |
+
"grad_norm": 0.369140625,
|
| 576 |
+
"learning_rate": 0.0007,
|
| 577 |
+
"loss": 2.561947250366211,
|
| 578 |
+
"step": 1120
|
| 579 |
+
},
|
| 580 |
+
{
|
| 581 |
+
"epoch": 0.07684529828109202,
|
| 582 |
+
"grad_norm": 0.390625,
|
| 583 |
+
"learning_rate": 0.0007,
|
| 584 |
+
"loss": 2.54862117767334,
|
| 585 |
+
"step": 1140
|
| 586 |
+
},
|
| 587 |
+
{
|
| 588 |
+
"epoch": 0.07751937984496124,
|
| 589 |
+
"eval_loss": 2.559185743331909,
|
| 590 |
+
"eval_runtime": 15.2877,
|
| 591 |
+
"eval_samples_per_second": 623.181,
|
| 592 |
+
"eval_steps_per_second": 0.654,
|
| 593 |
+
"step": 1150
|
| 594 |
+
},
|
| 595 |
+
{
|
| 596 |
+
"epoch": 0.07819346140883048,
|
| 597 |
+
"grad_norm": 0.37890625,
|
| 598 |
+
"learning_rate": 0.0007,
|
| 599 |
+
"loss": 2.5469717025756835,
|
| 600 |
+
"step": 1160
|
| 601 |
+
},
|
| 602 |
+
{
|
| 603 |
+
"epoch": 0.07954162453656892,
|
| 604 |
+
"grad_norm": 0.400390625,
|
| 605 |
+
"learning_rate": 0.0007,
|
| 606 |
+
"loss": 2.540451240539551,
|
| 607 |
+
"step": 1180
|
| 608 |
+
},
|
| 609 |
+
{
|
| 610 |
+
"epoch": 0.08088978766430738,
|
| 611 |
+
"grad_norm": 0.40625,
|
| 612 |
+
"learning_rate": 0.0007,
|
| 613 |
+
"loss": 2.53134708404541,
|
| 614 |
+
"step": 1200
|
| 615 |
+
},
|
| 616 |
+
{
|
| 617 |
+
"epoch": 0.08088978766430738,
|
| 618 |
+
"eval_loss": 2.521665096282959,
|
| 619 |
+
"eval_runtime": 15.2822,
|
| 620 |
+
"eval_samples_per_second": 623.407,
|
| 621 |
+
"eval_steps_per_second": 0.654,
|
| 622 |
+
"step": 1200
|
| 623 |
+
},
|
| 624 |
+
{
|
| 625 |
+
"epoch": 0.08223795079204584,
|
| 626 |
+
"grad_norm": 0.37109375,
|
| 627 |
+
"learning_rate": 0.0007,
|
| 628 |
+
"loss": 2.513539123535156,
|
| 629 |
+
"step": 1220
|
| 630 |
+
},
|
| 631 |
+
{
|
| 632 |
+
"epoch": 0.0835861139197843,
|
| 633 |
+
"grad_norm": 0.36328125,
|
| 634 |
+
"learning_rate": 0.0007,
|
| 635 |
+
"loss": 2.5009738922119142,
|
| 636 |
+
"step": 1240
|
| 637 |
+
},
|
| 638 |
+
{
|
| 639 |
+
"epoch": 0.08426019548365352,
|
| 640 |
+
"eval_loss": 2.489459753036499,
|
| 641 |
+
"eval_runtime": 15.2493,
|
| 642 |
+
"eval_samples_per_second": 624.749,
|
| 643 |
+
"eval_steps_per_second": 0.656,
|
| 644 |
+
"step": 1250
|
| 645 |
+
},
|
| 646 |
+
{
|
| 647 |
+
"epoch": 0.08493427704752275,
|
| 648 |
+
"grad_norm": 0.388671875,
|
| 649 |
+
"learning_rate": 0.0007,
|
| 650 |
+
"loss": 2.4894077301025392,
|
| 651 |
+
"step": 1260
|
| 652 |
+
},
|
| 653 |
+
{
|
| 654 |
+
"epoch": 0.08628244017526121,
|
| 655 |
+
"grad_norm": 0.369140625,
|
| 656 |
+
"learning_rate": 0.0007,
|
| 657 |
+
"loss": 2.4697439193725588,
|
| 658 |
+
"step": 1280
|
| 659 |
+
},
|
| 660 |
+
{
|
| 661 |
+
"epoch": 0.08763060330299967,
|
| 662 |
+
"grad_norm": 0.423828125,
|
| 663 |
+
"learning_rate": 0.0007,
|
| 664 |
+
"loss": 2.461294174194336,
|
| 665 |
+
"step": 1300
|
| 666 |
+
},
|
| 667 |
+
{
|
| 668 |
+
"epoch": 0.08763060330299967,
|
| 669 |
+
"eval_loss": 2.4627816677093506,
|
| 670 |
+
"eval_runtime": 15.421,
|
| 671 |
+
"eval_samples_per_second": 617.795,
|
| 672 |
+
"eval_steps_per_second": 0.648,
|
| 673 |
+
"step": 1300
|
| 674 |
+
},
|
| 675 |
+
{
|
| 676 |
+
"epoch": 0.08897876643073811,
|
| 677 |
+
"grad_norm": 0.36328125,
|
| 678 |
+
"learning_rate": 0.0007,
|
| 679 |
+
"loss": 2.452680206298828,
|
| 680 |
+
"step": 1320
|
| 681 |
+
},
|
| 682 |
+
{
|
| 683 |
+
"epoch": 0.09032692955847657,
|
| 684 |
+
"grad_norm": 0.404296875,
|
| 685 |
+
"learning_rate": 0.0007,
|
| 686 |
+
"loss": 2.4396684646606444,
|
| 687 |
+
"step": 1340
|
| 688 |
+
},
|
| 689 |
+
{
|
| 690 |
+
"epoch": 0.0910010111223458,
|
| 691 |
+
"eval_loss": 2.437220335006714,
|
| 692 |
+
"eval_runtime": 15.3026,
|
| 693 |
+
"eval_samples_per_second": 622.576,
|
| 694 |
+
"eval_steps_per_second": 0.653,
|
| 695 |
+
"step": 1350
|
| 696 |
+
},
|
| 697 |
+
{
|
| 698 |
+
"epoch": 0.09167509268621503,
|
| 699 |
+
"grad_norm": 0.412109375,
|
| 700 |
+
"learning_rate": 0.0007,
|
| 701 |
+
"loss": 2.4397760391235352,
|
| 702 |
+
"step": 1360
|
| 703 |
+
},
|
| 704 |
+
{
|
| 705 |
+
"epoch": 0.09302325581395349,
|
| 706 |
+
"grad_norm": 0.353515625,
|
| 707 |
+
"learning_rate": 0.0007,
|
| 708 |
+
"loss": 2.4158536911010744,
|
| 709 |
+
"step": 1380
|
| 710 |
+
},
|
| 711 |
+
{
|
| 712 |
+
"epoch": 0.09437141894169195,
|
| 713 |
+
"grad_norm": 0.345703125,
|
| 714 |
+
"learning_rate": 0.0007,
|
| 715 |
+
"loss": 2.406942367553711,
|
| 716 |
+
"step": 1400
|
| 717 |
+
},
|
| 718 |
+
{
|
| 719 |
+
"epoch": 0.09437141894169195,
|
| 720 |
+
"eval_loss": 2.4096601009368896,
|
| 721 |
+
"eval_runtime": 15.357,
|
| 722 |
+
"eval_samples_per_second": 620.368,
|
| 723 |
+
"eval_steps_per_second": 0.651,
|
| 724 |
+
"step": 1400
|
| 725 |
+
}
|
| 726 |
+
],
|
| 727 |
+
"logging_steps": 20,
|
| 728 |
+
"max_steps": 2500,
|
| 729 |
+
"num_input_tokens_seen": 0,
|
| 730 |
+
"num_train_epochs": 1,
|
| 731 |
+
"save_steps": 100,
|
| 732 |
+
"stateful_callbacks": {
|
| 733 |
+
"TrainerControl": {
|
| 734 |
+
"args": {
|
| 735 |
+
"should_epoch_stop": false,
|
| 736 |
+
"should_evaluate": false,
|
| 737 |
+
"should_log": false,
|
| 738 |
+
"should_save": true,
|
| 739 |
+
"should_training_stop": false
|
| 740 |
+
},
|
| 741 |
+
"attributes": {}
|
| 742 |
+
}
|
| 743 |
+
},
|
| 744 |
+
"total_flos": 4516797323673600.0,
|
| 745 |
+
"train_batch_size": 64,
|
| 746 |
+
"trial_name": null,
|
| 747 |
+
"trial_params": null
|
| 748 |
+
}
|
zain/Activation/out/glu-silu-100L_run/checkpoint-1400/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:fd90512e56f813612dc668676d6f8cab72da110a42f23c61cafebf91064625a1
|
| 3 |
+
size 4920
|
zain/Activation/out/glu-silu-100L_run/checkpoint-1500/config.json
ADDED
|
@@ -0,0 +1,36 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"activation": "silu",
|
| 3 |
+
"architectures": [
|
| 4 |
+
"TinyLlamaForCausalLM"
|
| 5 |
+
],
|
| 6 |
+
"attention_bias": false,
|
| 7 |
+
"attention_dropout": 0.0,
|
| 8 |
+
"bos_token_id": 1,
|
| 9 |
+
"dtype": "bfloat16",
|
| 10 |
+
"eos_token_id": 2,
|
| 11 |
+
"head_dim": 32,
|
| 12 |
+
"hidden_act": "silu",
|
| 13 |
+
"hidden_size": 128,
|
| 14 |
+
"initializer_range": 0.02,
|
| 15 |
+
"intermediate_size": 256,
|
| 16 |
+
"max_position_embeddings": 512,
|
| 17 |
+
"mlp_bias": false,
|
| 18 |
+
"mlp_type": "glu",
|
| 19 |
+
"model_type": "tiny_llama",
|
| 20 |
+
"num_attention_heads": 4,
|
| 21 |
+
"num_hidden_layers": 100,
|
| 22 |
+
"num_key_value_heads": 4,
|
| 23 |
+
"pad_token_id": 0,
|
| 24 |
+
"pretraining_tp": 1,
|
| 25 |
+
"rms_norm_eps": 1e-06,
|
| 26 |
+
"rope_parameters": {
|
| 27 |
+
"rope_theta": 10000.0,
|
| 28 |
+
"rope_type": "default"
|
| 29 |
+
},
|
| 30 |
+
"tie_word_embeddings": true,
|
| 31 |
+
"tokenizer_name": "w-ahmad/tiny-stories-tokenizer",
|
| 32 |
+
"transformers_version": "5.16.0.dev0",
|
| 33 |
+
"use_cache": false,
|
| 34 |
+
"vocab_size": 4096,
|
| 35 |
+
"waleed_beta": 10.0
|
| 36 |
+
}
|
zain/Activation/out/glu-silu-100L_run/checkpoint-1500/model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:40a6bbdaacbaf59fb1a527a6ed139e10182978da2410657f0fb219c8630045ba
|
| 3 |
+
size 33967272
|
zain/Activation/out/glu-silu-100L_run/checkpoint-1500/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:973fff485fab8b7a1264522f88b528a6f6b5fe1d6c2a211092fb3b1f713d73e6
|
| 3 |
+
size 68504996
|
zain/Activation/out/glu-silu-100L_run/checkpoint-1500/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:c6dc93c57336e63d1a7c6701812162f34ece4833af3a7a404c317dbe67b08e9c
|
| 3 |
+
size 14244
|
zain/Activation/out/glu-silu-100L_run/checkpoint-1500/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:6a37538486c6fa3324bc6a6a3b2ad67db8646e004089a61bb15e3c6cba879c55
|
| 3 |
+
size 1064
|
zain/Activation/out/glu-silu-100L_run/checkpoint-1500/tokenizer.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
zain/Activation/out/glu-silu-100L_run/checkpoint-1500/tokenizer_config.json
ADDED
|
@@ -0,0 +1,13 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"add_prefix_space": false,
|
| 3 |
+
"backend": "tokenizers",
|
| 4 |
+
"bos_token": "<|endoftext|>",
|
| 5 |
+
"eos_token": "<|endoftext|>",
|
| 6 |
+
"errors": "replace",
|
| 7 |
+
"is_local": false,
|
| 8 |
+
"local_files_only": false,
|
| 9 |
+
"model_max_length": 1024,
|
| 10 |
+
"pad_token": "<|endoftext|>",
|
| 11 |
+
"tokenizer_class": "GPT2Tokenizer",
|
| 12 |
+
"unk_token": "<|endoftext|>"
|
| 13 |
+
}
|
zain/Activation/out/glu-silu-100L_run/checkpoint-1500/trainer_state.json
ADDED
|
@@ -0,0 +1,799 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_global_step": null,
|
| 3 |
+
"best_metric": null,
|
| 4 |
+
"best_model_checkpoint": null,
|
| 5 |
+
"epoch": 0.10111223458038422,
|
| 6 |
+
"eval_steps": 50,
|
| 7 |
+
"global_step": 1500,
|
| 8 |
+
"is_hyper_param_search": false,
|
| 9 |
+
"is_local_process_zero": true,
|
| 10 |
+
"is_world_process_zero": true,
|
| 11 |
+
"log_history": [
|
| 12 |
+
{
|
| 13 |
+
"epoch": 0.0013481631277384564,
|
| 14 |
+
"grad_norm": 2.078125,
|
| 15 |
+
"learning_rate": 2.66e-05,
|
| 16 |
+
"loss": 8.29962387084961,
|
| 17 |
+
"step": 20
|
| 18 |
+
},
|
| 19 |
+
{
|
| 20 |
+
"epoch": 0.002696326255476913,
|
| 21 |
+
"grad_norm": 1.3515625,
|
| 22 |
+
"learning_rate": 5.46e-05,
|
| 23 |
+
"loss": 8.047650909423828,
|
| 24 |
+
"step": 40
|
| 25 |
+
},
|
| 26 |
+
{
|
| 27 |
+
"epoch": 0.003370407819346141,
|
| 28 |
+
"eval_loss": 7.835535049438477,
|
| 29 |
+
"eval_runtime": 15.2165,
|
| 30 |
+
"eval_samples_per_second": 626.095,
|
| 31 |
+
"eval_steps_per_second": 0.657,
|
| 32 |
+
"step": 50
|
| 33 |
+
},
|
| 34 |
+
{
|
| 35 |
+
"epoch": 0.004044489383215369,
|
| 36 |
+
"grad_norm": 1.2734375,
|
| 37 |
+
"learning_rate": 8.259999999999999e-05,
|
| 38 |
+
"loss": 7.8378746032714846,
|
| 39 |
+
"step": 60
|
| 40 |
+
},
|
| 41 |
+
{
|
| 42 |
+
"epoch": 0.005392652510953826,
|
| 43 |
+
"grad_norm": 1.265625,
|
| 44 |
+
"learning_rate": 0.0001106,
|
| 45 |
+
"loss": 7.5728271484375,
|
| 46 |
+
"step": 80
|
| 47 |
+
},
|
| 48 |
+
{
|
| 49 |
+
"epoch": 0.006740815638692282,
|
| 50 |
+
"grad_norm": 1.1953125,
|
| 51 |
+
"learning_rate": 0.0001386,
|
| 52 |
+
"loss": 7.2332916259765625,
|
| 53 |
+
"step": 100
|
| 54 |
+
},
|
| 55 |
+
{
|
| 56 |
+
"epoch": 0.006740815638692282,
|
| 57 |
+
"eval_loss": 7.039053440093994,
|
| 58 |
+
"eval_runtime": 15.2074,
|
| 59 |
+
"eval_samples_per_second": 626.471,
|
| 60 |
+
"eval_steps_per_second": 0.658,
|
| 61 |
+
"step": 100
|
| 62 |
+
},
|
| 63 |
+
{
|
| 64 |
+
"epoch": 0.008088978766430738,
|
| 65 |
+
"grad_norm": 1.1328125,
|
| 66 |
+
"learning_rate": 0.00016659999999999998,
|
| 67 |
+
"loss": 6.8608039855957035,
|
| 68 |
+
"step": 120
|
| 69 |
+
},
|
| 70 |
+
{
|
| 71 |
+
"epoch": 0.009437141894169195,
|
| 72 |
+
"grad_norm": 0.94921875,
|
| 73 |
+
"learning_rate": 0.00019460000000000001,
|
| 74 |
+
"loss": 6.517842864990234,
|
| 75 |
+
"step": 140
|
| 76 |
+
},
|
| 77 |
+
{
|
| 78 |
+
"epoch": 0.010111223458038422,
|
| 79 |
+
"eval_loss": 6.229617595672607,
|
| 80 |
+
"eval_runtime": 15.1978,
|
| 81 |
+
"eval_samples_per_second": 626.867,
|
| 82 |
+
"eval_steps_per_second": 0.658,
|
| 83 |
+
"step": 150
|
| 84 |
+
},
|
| 85 |
+
{
|
| 86 |
+
"epoch": 0.010785305021907651,
|
| 87 |
+
"grad_norm": 0.80078125,
|
| 88 |
+
"learning_rate": 0.0002226,
|
| 89 |
+
"loss": 6.2399169921875,
|
| 90 |
+
"step": 160
|
| 91 |
+
},
|
| 92 |
+
{
|
| 93 |
+
"epoch": 0.012133468149646108,
|
| 94 |
+
"grad_norm": 0.6640625,
|
| 95 |
+
"learning_rate": 0.00025059999999999997,
|
| 96 |
+
"loss": 6.036548614501953,
|
| 97 |
+
"step": 180
|
| 98 |
+
},
|
| 99 |
+
{
|
| 100 |
+
"epoch": 0.013481631277384564,
|
| 101 |
+
"grad_norm": 0.61328125,
|
| 102 |
+
"learning_rate": 0.0002786,
|
| 103 |
+
"loss": 5.857078552246094,
|
| 104 |
+
"step": 200
|
| 105 |
+
},
|
| 106 |
+
{
|
| 107 |
+
"epoch": 0.013481631277384564,
|
| 108 |
+
"eval_loss": 5.765108585357666,
|
| 109 |
+
"eval_runtime": 15.5271,
|
| 110 |
+
"eval_samples_per_second": 613.574,
|
| 111 |
+
"eval_steps_per_second": 0.644,
|
| 112 |
+
"step": 200
|
| 113 |
+
},
|
| 114 |
+
{
|
| 115 |
+
"epoch": 0.01482979440512302,
|
| 116 |
+
"grad_norm": 0.61328125,
|
| 117 |
+
"learning_rate": 0.00030659999999999997,
|
| 118 |
+
"loss": 5.680564117431641,
|
| 119 |
+
"step": 220
|
| 120 |
+
},
|
| 121 |
+
{
|
| 122 |
+
"epoch": 0.016177957532861477,
|
| 123 |
+
"grad_norm": 2.0625,
|
| 124 |
+
"learning_rate": 0.0003346,
|
| 125 |
+
"loss": 5.472259902954102,
|
| 126 |
+
"step": 240
|
| 127 |
+
},
|
| 128 |
+
{
|
| 129 |
+
"epoch": 0.016852039096730706,
|
| 130 |
+
"eval_loss": 5.240335464477539,
|
| 131 |
+
"eval_runtime": 15.2596,
|
| 132 |
+
"eval_samples_per_second": 624.33,
|
| 133 |
+
"eval_steps_per_second": 0.655,
|
| 134 |
+
"step": 250
|
| 135 |
+
},
|
| 136 |
+
{
|
| 137 |
+
"epoch": 0.01752612066059993,
|
| 138 |
+
"grad_norm": 1.7421875,
|
| 139 |
+
"learning_rate": 0.00036260000000000003,
|
| 140 |
+
"loss": 5.24955940246582,
|
| 141 |
+
"step": 260
|
| 142 |
+
},
|
| 143 |
+
{
|
| 144 |
+
"epoch": 0.01887428378833839,
|
| 145 |
+
"grad_norm": 0.62109375,
|
| 146 |
+
"learning_rate": 0.0003906,
|
| 147 |
+
"loss": 4.995759963989258,
|
| 148 |
+
"step": 280
|
| 149 |
+
},
|
| 150 |
+
{
|
| 151 |
+
"epoch": 0.020222446916076844,
|
| 152 |
+
"grad_norm": 1.390625,
|
| 153 |
+
"learning_rate": 0.0004186,
|
| 154 |
+
"loss": 4.769353866577148,
|
| 155 |
+
"step": 300
|
| 156 |
+
},
|
| 157 |
+
{
|
| 158 |
+
"epoch": 0.020222446916076844,
|
| 159 |
+
"eval_loss": 4.669629096984863,
|
| 160 |
+
"eval_runtime": 15.2595,
|
| 161 |
+
"eval_samples_per_second": 624.332,
|
| 162 |
+
"eval_steps_per_second": 0.655,
|
| 163 |
+
"step": 300
|
| 164 |
+
},
|
| 165 |
+
{
|
| 166 |
+
"epoch": 0.021570610043815303,
|
| 167 |
+
"grad_norm": 0.73828125,
|
| 168 |
+
"learning_rate": 0.0004466,
|
| 169 |
+
"loss": 4.569867324829102,
|
| 170 |
+
"step": 320
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 0.022918773171553757,
|
| 174 |
+
"grad_norm": 0.765625,
|
| 175 |
+
"learning_rate": 0.00047460000000000004,
|
| 176 |
+
"loss": 4.389446258544922,
|
| 177 |
+
"step": 340
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 0.023592854735422986,
|
| 181 |
+
"eval_loss": 4.231207847595215,
|
| 182 |
+
"eval_runtime": 15.2143,
|
| 183 |
+
"eval_samples_per_second": 626.188,
|
| 184 |
+
"eval_steps_per_second": 0.657,
|
| 185 |
+
"step": 350
|
| 186 |
+
},
|
| 187 |
+
{
|
| 188 |
+
"epoch": 0.024266936299292215,
|
| 189 |
+
"grad_norm": 1.2421875,
|
| 190 |
+
"learning_rate": 0.0005026,
|
| 191 |
+
"loss": 4.243848419189453,
|
| 192 |
+
"step": 360
|
| 193 |
+
},
|
| 194 |
+
{
|
| 195 |
+
"epoch": 0.02561509942703067,
|
| 196 |
+
"grad_norm": 0.7109375,
|
| 197 |
+
"learning_rate": 0.0005306,
|
| 198 |
+
"loss": 4.1090648651123045,
|
| 199 |
+
"step": 380
|
| 200 |
+
},
|
| 201 |
+
{
|
| 202 |
+
"epoch": 0.026963262554769128,
|
| 203 |
+
"grad_norm": 0.484375,
|
| 204 |
+
"learning_rate": 0.0005586,
|
| 205 |
+
"loss": 3.966172790527344,
|
| 206 |
+
"step": 400
|
| 207 |
+
},
|
| 208 |
+
{
|
| 209 |
+
"epoch": 0.026963262554769128,
|
| 210 |
+
"eval_loss": 3.923773765563965,
|
| 211 |
+
"eval_runtime": 15.5308,
|
| 212 |
+
"eval_samples_per_second": 613.426,
|
| 213 |
+
"eval_steps_per_second": 0.644,
|
| 214 |
+
"step": 400
|
| 215 |
+
},
|
| 216 |
+
{
|
| 217 |
+
"epoch": 0.028311425682507583,
|
| 218 |
+
"grad_norm": 0.6640625,
|
| 219 |
+
"learning_rate": 0.0005866,
|
| 220 |
+
"loss": 3.8559597015380858,
|
| 221 |
+
"step": 420
|
| 222 |
+
},
|
| 223 |
+
{
|
| 224 |
+
"epoch": 0.02965958881024604,
|
| 225 |
+
"grad_norm": 0.515625,
|
| 226 |
+
"learning_rate": 0.0006146,
|
| 227 |
+
"loss": 3.7943866729736326,
|
| 228 |
+
"step": 440
|
| 229 |
+
},
|
| 230 |
+
{
|
| 231 |
+
"epoch": 0.030333670374115267,
|
| 232 |
+
"eval_loss": 3.6977553367614746,
|
| 233 |
+
"eval_runtime": 15.2001,
|
| 234 |
+
"eval_samples_per_second": 626.77,
|
| 235 |
+
"eval_steps_per_second": 0.658,
|
| 236 |
+
"step": 450
|
| 237 |
+
},
|
| 238 |
+
{
|
| 239 |
+
"epoch": 0.031007751937984496,
|
| 240 |
+
"grad_norm": 0.51953125,
|
| 241 |
+
"learning_rate": 0.0006426,
|
| 242 |
+
"loss": 3.686443328857422,
|
| 243 |
+
"step": 460
|
| 244 |
+
},
|
| 245 |
+
{
|
| 246 |
+
"epoch": 0.032355915065722954,
|
| 247 |
+
"grad_norm": 0.62109375,
|
| 248 |
+
"learning_rate": 0.0006705999999999999,
|
| 249 |
+
"loss": 3.6123157501220704,
|
| 250 |
+
"step": 480
|
| 251 |
+
},
|
| 252 |
+
{
|
| 253 |
+
"epoch": 0.03370407819346141,
|
| 254 |
+
"grad_norm": 0.70703125,
|
| 255 |
+
"learning_rate": 0.0006986,
|
| 256 |
+
"loss": 3.542654800415039,
|
| 257 |
+
"step": 500
|
| 258 |
+
},
|
| 259 |
+
{
|
| 260 |
+
"epoch": 0.03370407819346141,
|
| 261 |
+
"eval_loss": 3.513820171356201,
|
| 262 |
+
"eval_runtime": 15.2193,
|
| 263 |
+
"eval_samples_per_second": 625.983,
|
| 264 |
+
"eval_steps_per_second": 0.657,
|
| 265 |
+
"step": 500
|
| 266 |
+
},
|
| 267 |
+
{
|
| 268 |
+
"epoch": 0.03505224132119986,
|
| 269 |
+
"grad_norm": 0.48046875,
|
| 270 |
+
"learning_rate": 0.0007,
|
| 271 |
+
"loss": 3.472199249267578,
|
| 272 |
+
"step": 520
|
| 273 |
+
},
|
| 274 |
+
{
|
| 275 |
+
"epoch": 0.03640040444893832,
|
| 276 |
+
"grad_norm": 0.53515625,
|
| 277 |
+
"learning_rate": 0.0007,
|
| 278 |
+
"loss": 3.396291732788086,
|
| 279 |
+
"step": 540
|
| 280 |
+
},
|
| 281 |
+
{
|
| 282 |
+
"epoch": 0.03707448601280755,
|
| 283 |
+
"eval_loss": 3.3581483364105225,
|
| 284 |
+
"eval_runtime": 15.3044,
|
| 285 |
+
"eval_samples_per_second": 622.502,
|
| 286 |
+
"eval_steps_per_second": 0.653,
|
| 287 |
+
"step": 550
|
| 288 |
+
},
|
| 289 |
+
{
|
| 290 |
+
"epoch": 0.03774856757667678,
|
| 291 |
+
"grad_norm": 0.47265625,
|
| 292 |
+
"learning_rate": 0.0007,
|
| 293 |
+
"loss": 3.347300720214844,
|
| 294 |
+
"step": 560
|
| 295 |
+
},
|
| 296 |
+
{
|
| 297 |
+
"epoch": 0.03909673070441524,
|
| 298 |
+
"grad_norm": 0.458984375,
|
| 299 |
+
"learning_rate": 0.0007,
|
| 300 |
+
"loss": 3.2885902404785154,
|
| 301 |
+
"step": 580
|
| 302 |
+
},
|
| 303 |
+
{
|
| 304 |
+
"epoch": 0.04044489383215369,
|
| 305 |
+
"grad_norm": 0.4375,
|
| 306 |
+
"learning_rate": 0.0007,
|
| 307 |
+
"loss": 3.25723876953125,
|
| 308 |
+
"step": 600
|
| 309 |
+
},
|
| 310 |
+
{
|
| 311 |
+
"epoch": 0.04044489383215369,
|
| 312 |
+
"eval_loss": 3.2229573726654053,
|
| 313 |
+
"eval_runtime": 15.2967,
|
| 314 |
+
"eval_samples_per_second": 622.812,
|
| 315 |
+
"eval_steps_per_second": 0.654,
|
| 316 |
+
"step": 600
|
| 317 |
+
},
|
| 318 |
+
{
|
| 319 |
+
"epoch": 0.04179305695989215,
|
| 320 |
+
"grad_norm": 0.458984375,
|
| 321 |
+
"learning_rate": 0.0007,
|
| 322 |
+
"loss": 3.1902334213256838,
|
| 323 |
+
"step": 620
|
| 324 |
+
},
|
| 325 |
+
{
|
| 326 |
+
"epoch": 0.043141220087630605,
|
| 327 |
+
"grad_norm": 0.40625,
|
| 328 |
+
"learning_rate": 0.0007,
|
| 329 |
+
"loss": 3.167861557006836,
|
| 330 |
+
"step": 640
|
| 331 |
+
},
|
| 332 |
+
{
|
| 333 |
+
"epoch": 0.043815301651499834,
|
| 334 |
+
"eval_loss": 3.120636463165283,
|
| 335 |
+
"eval_runtime": 15.3751,
|
| 336 |
+
"eval_samples_per_second": 619.639,
|
| 337 |
+
"eval_steps_per_second": 0.65,
|
| 338 |
+
"step": 650
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 0.044489383215369056,
|
| 342 |
+
"grad_norm": 0.494140625,
|
| 343 |
+
"learning_rate": 0.0007,
|
| 344 |
+
"loss": 3.1089324951171875,
|
| 345 |
+
"step": 660
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 0.045837546343107514,
|
| 349 |
+
"grad_norm": 0.435546875,
|
| 350 |
+
"learning_rate": 0.0007,
|
| 351 |
+
"loss": 3.0828268051147463,
|
| 352 |
+
"step": 680
|
| 353 |
+
},
|
| 354 |
+
{
|
| 355 |
+
"epoch": 0.04718570947084597,
|
| 356 |
+
"grad_norm": 0.427734375,
|
| 357 |
+
"learning_rate": 0.0007,
|
| 358 |
+
"loss": 3.0396488189697264,
|
| 359 |
+
"step": 700
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"epoch": 0.04718570947084597,
|
| 363 |
+
"eval_loss": 3.0276408195495605,
|
| 364 |
+
"eval_runtime": 15.2735,
|
| 365 |
+
"eval_samples_per_second": 623.761,
|
| 366 |
+
"eval_steps_per_second": 0.655,
|
| 367 |
+
"step": 700
|
| 368 |
+
},
|
| 369 |
+
{
|
| 370 |
+
"epoch": 0.04853387259858443,
|
| 371 |
+
"grad_norm": 0.4375,
|
| 372 |
+
"learning_rate": 0.0007,
|
| 373 |
+
"loss": 3.0034805297851563,
|
| 374 |
+
"step": 720
|
| 375 |
+
},
|
| 376 |
+
{
|
| 377 |
+
"epoch": 0.04988203572632288,
|
| 378 |
+
"grad_norm": 0.52734375,
|
| 379 |
+
"learning_rate": 0.0007,
|
| 380 |
+
"loss": 2.96556396484375,
|
| 381 |
+
"step": 740
|
| 382 |
+
},
|
| 383 |
+
{
|
| 384 |
+
"epoch": 0.05055611729019211,
|
| 385 |
+
"eval_loss": 2.950009346008301,
|
| 386 |
+
"eval_runtime": 15.243,
|
| 387 |
+
"eval_samples_per_second": 625.009,
|
| 388 |
+
"eval_steps_per_second": 0.656,
|
| 389 |
+
"step": 750
|
| 390 |
+
},
|
| 391 |
+
{
|
| 392 |
+
"epoch": 0.05123019885406134,
|
| 393 |
+
"grad_norm": 0.451171875,
|
| 394 |
+
"learning_rate": 0.0007,
|
| 395 |
+
"loss": 2.950779151916504,
|
| 396 |
+
"step": 760
|
| 397 |
+
},
|
| 398 |
+
{
|
| 399 |
+
"epoch": 0.0525783619817998,
|
| 400 |
+
"grad_norm": 0.400390625,
|
| 401 |
+
"learning_rate": 0.0007,
|
| 402 |
+
"loss": 2.956203079223633,
|
| 403 |
+
"step": 780
|
| 404 |
+
},
|
| 405 |
+
{
|
| 406 |
+
"epoch": 0.053926525109538256,
|
| 407 |
+
"grad_norm": 0.40625,
|
| 408 |
+
"learning_rate": 0.0007,
|
| 409 |
+
"loss": 2.9017578125,
|
| 410 |
+
"step": 800
|
| 411 |
+
},
|
| 412 |
+
{
|
| 413 |
+
"epoch": 0.053926525109538256,
|
| 414 |
+
"eval_loss": 2.8844006061553955,
|
| 415 |
+
"eval_runtime": 15.2903,
|
| 416 |
+
"eval_samples_per_second": 623.073,
|
| 417 |
+
"eval_steps_per_second": 0.654,
|
| 418 |
+
"step": 800
|
| 419 |
+
},
|
| 420 |
+
{
|
| 421 |
+
"epoch": 0.05527468823727671,
|
| 422 |
+
"grad_norm": 0.41015625,
|
| 423 |
+
"learning_rate": 0.0007,
|
| 424 |
+
"loss": 2.887515640258789,
|
| 425 |
+
"step": 820
|
| 426 |
+
},
|
| 427 |
+
{
|
| 428 |
+
"epoch": 0.056622851365015166,
|
| 429 |
+
"grad_norm": 0.41796875,
|
| 430 |
+
"learning_rate": 0.0007,
|
| 431 |
+
"loss": 2.8556928634643555,
|
| 432 |
+
"step": 840
|
| 433 |
+
},
|
| 434 |
+
{
|
| 435 |
+
"epoch": 0.057296932928884395,
|
| 436 |
+
"eval_loss": 2.821610450744629,
|
| 437 |
+
"eval_runtime": 15.2253,
|
| 438 |
+
"eval_samples_per_second": 625.733,
|
| 439 |
+
"eval_steps_per_second": 0.657,
|
| 440 |
+
"step": 850
|
| 441 |
+
},
|
| 442 |
+
{
|
| 443 |
+
"epoch": 0.057971014492753624,
|
| 444 |
+
"grad_norm": 0.423828125,
|
| 445 |
+
"learning_rate": 0.0007,
|
| 446 |
+
"loss": 2.8130855560302734,
|
| 447 |
+
"step": 860
|
| 448 |
+
},
|
| 449 |
+
{
|
| 450 |
+
"epoch": 0.05931917762049208,
|
| 451 |
+
"grad_norm": 0.38671875,
|
| 452 |
+
"learning_rate": 0.0007,
|
| 453 |
+
"loss": 2.790329170227051,
|
| 454 |
+
"step": 880
|
| 455 |
+
},
|
| 456 |
+
{
|
| 457 |
+
"epoch": 0.06066734074823053,
|
| 458 |
+
"grad_norm": 0.427734375,
|
| 459 |
+
"learning_rate": 0.0007,
|
| 460 |
+
"loss": 2.7760162353515625,
|
| 461 |
+
"step": 900
|
| 462 |
+
},
|
| 463 |
+
{
|
| 464 |
+
"epoch": 0.06066734074823053,
|
| 465 |
+
"eval_loss": 2.7661373615264893,
|
| 466 |
+
"eval_runtime": 15.2736,
|
| 467 |
+
"eval_samples_per_second": 623.755,
|
| 468 |
+
"eval_steps_per_second": 0.655,
|
| 469 |
+
"step": 900
|
| 470 |
+
},
|
| 471 |
+
{
|
| 472 |
+
"epoch": 0.06201550387596899,
|
| 473 |
+
"grad_norm": 0.435546875,
|
| 474 |
+
"learning_rate": 0.0007,
|
| 475 |
+
"loss": 2.7462047576904296,
|
| 476 |
+
"step": 920
|
| 477 |
+
},
|
| 478 |
+
{
|
| 479 |
+
"epoch": 0.06336366700370745,
|
| 480 |
+
"grad_norm": 0.4140625,
|
| 481 |
+
"learning_rate": 0.0007,
|
| 482 |
+
"loss": 2.732274627685547,
|
| 483 |
+
"step": 940
|
| 484 |
+
},
|
| 485 |
+
{
|
| 486 |
+
"epoch": 0.06403774856757667,
|
| 487 |
+
"eval_loss": 2.713942289352417,
|
| 488 |
+
"eval_runtime": 15.6062,
|
| 489 |
+
"eval_samples_per_second": 610.463,
|
| 490 |
+
"eval_steps_per_second": 0.641,
|
| 491 |
+
"step": 950
|
| 492 |
+
},
|
| 493 |
+
{
|
| 494 |
+
"epoch": 0.06471183013144591,
|
| 495 |
+
"grad_norm": 0.419921875,
|
| 496 |
+
"learning_rate": 0.0007,
|
| 497 |
+
"loss": 2.705318069458008,
|
| 498 |
+
"step": 960
|
| 499 |
+
},
|
| 500 |
+
{
|
| 501 |
+
"epoch": 0.06605999325918437,
|
| 502 |
+
"grad_norm": 0.4296875,
|
| 503 |
+
"learning_rate": 0.0007,
|
| 504 |
+
"loss": 2.6892818450927733,
|
| 505 |
+
"step": 980
|
| 506 |
+
},
|
| 507 |
+
{
|
| 508 |
+
"epoch": 0.06740815638692282,
|
| 509 |
+
"grad_norm": 0.4453125,
|
| 510 |
+
"learning_rate": 0.0007,
|
| 511 |
+
"loss": 2.687115287780762,
|
| 512 |
+
"step": 1000
|
| 513 |
+
},
|
| 514 |
+
{
|
| 515 |
+
"epoch": 0.06740815638692282,
|
| 516 |
+
"eval_loss": 2.668647289276123,
|
| 517 |
+
"eval_runtime": 15.2597,
|
| 518 |
+
"eval_samples_per_second": 624.326,
|
| 519 |
+
"eval_steps_per_second": 0.655,
|
| 520 |
+
"step": 1000
|
| 521 |
+
},
|
| 522 |
+
{
|
| 523 |
+
"epoch": 0.06875631951466127,
|
| 524 |
+
"grad_norm": 0.388671875,
|
| 525 |
+
"learning_rate": 0.0007,
|
| 526 |
+
"loss": 2.6533071517944338,
|
| 527 |
+
"step": 1020
|
| 528 |
+
},
|
| 529 |
+
{
|
| 530 |
+
"epoch": 0.07010448264239973,
|
| 531 |
+
"grad_norm": 0.392578125,
|
| 532 |
+
"learning_rate": 0.0007,
|
| 533 |
+
"loss": 2.6390922546386717,
|
| 534 |
+
"step": 1040
|
| 535 |
+
},
|
| 536 |
+
{
|
| 537 |
+
"epoch": 0.07077856420626896,
|
| 538 |
+
"eval_loss": 2.6291816234588623,
|
| 539 |
+
"eval_runtime": 15.2457,
|
| 540 |
+
"eval_samples_per_second": 624.896,
|
| 541 |
+
"eval_steps_per_second": 0.656,
|
| 542 |
+
"step": 1050
|
| 543 |
+
},
|
| 544 |
+
{
|
| 545 |
+
"epoch": 0.07145264577013818,
|
| 546 |
+
"grad_norm": 0.4375,
|
| 547 |
+
"learning_rate": 0.0007,
|
| 548 |
+
"loss": 2.624461555480957,
|
| 549 |
+
"step": 1060
|
| 550 |
+
},
|
| 551 |
+
{
|
| 552 |
+
"epoch": 0.07280080889787664,
|
| 553 |
+
"grad_norm": 0.396484375,
|
| 554 |
+
"learning_rate": 0.0007,
|
| 555 |
+
"loss": 2.6044427871704103,
|
| 556 |
+
"step": 1080
|
| 557 |
+
},
|
| 558 |
+
{
|
| 559 |
+
"epoch": 0.0741489720256151,
|
| 560 |
+
"grad_norm": 0.423828125,
|
| 561 |
+
"learning_rate": 0.0007,
|
| 562 |
+
"loss": 2.597646713256836,
|
| 563 |
+
"step": 1100
|
| 564 |
+
},
|
| 565 |
+
{
|
| 566 |
+
"epoch": 0.0741489720256151,
|
| 567 |
+
"eval_loss": 2.589696168899536,
|
| 568 |
+
"eval_runtime": 15.3967,
|
| 569 |
+
"eval_samples_per_second": 618.768,
|
| 570 |
+
"eval_steps_per_second": 0.649,
|
| 571 |
+
"step": 1100
|
| 572 |
+
},
|
| 573 |
+
{
|
| 574 |
+
"epoch": 0.07549713515335356,
|
| 575 |
+
"grad_norm": 0.369140625,
|
| 576 |
+
"learning_rate": 0.0007,
|
| 577 |
+
"loss": 2.561947250366211,
|
| 578 |
+
"step": 1120
|
| 579 |
+
},
|
| 580 |
+
{
|
| 581 |
+
"epoch": 0.07684529828109202,
|
| 582 |
+
"grad_norm": 0.390625,
|
| 583 |
+
"learning_rate": 0.0007,
|
| 584 |
+
"loss": 2.54862117767334,
|
| 585 |
+
"step": 1140
|
| 586 |
+
},
|
| 587 |
+
{
|
| 588 |
+
"epoch": 0.07751937984496124,
|
| 589 |
+
"eval_loss": 2.559185743331909,
|
| 590 |
+
"eval_runtime": 15.2877,
|
| 591 |
+
"eval_samples_per_second": 623.181,
|
| 592 |
+
"eval_steps_per_second": 0.654,
|
| 593 |
+
"step": 1150
|
| 594 |
+
},
|
| 595 |
+
{
|
| 596 |
+
"epoch": 0.07819346140883048,
|
| 597 |
+
"grad_norm": 0.37890625,
|
| 598 |
+
"learning_rate": 0.0007,
|
| 599 |
+
"loss": 2.5469717025756835,
|
| 600 |
+
"step": 1160
|
| 601 |
+
},
|
| 602 |
+
{
|
| 603 |
+
"epoch": 0.07954162453656892,
|
| 604 |
+
"grad_norm": 0.400390625,
|
| 605 |
+
"learning_rate": 0.0007,
|
| 606 |
+
"loss": 2.540451240539551,
|
| 607 |
+
"step": 1180
|
| 608 |
+
},
|
| 609 |
+
{
|
| 610 |
+
"epoch": 0.08088978766430738,
|
| 611 |
+
"grad_norm": 0.40625,
|
| 612 |
+
"learning_rate": 0.0007,
|
| 613 |
+
"loss": 2.53134708404541,
|
| 614 |
+
"step": 1200
|
| 615 |
+
},
|
| 616 |
+
{
|
| 617 |
+
"epoch": 0.08088978766430738,
|
| 618 |
+
"eval_loss": 2.521665096282959,
|
| 619 |
+
"eval_runtime": 15.2822,
|
| 620 |
+
"eval_samples_per_second": 623.407,
|
| 621 |
+
"eval_steps_per_second": 0.654,
|
| 622 |
+
"step": 1200
|
| 623 |
+
},
|
| 624 |
+
{
|
| 625 |
+
"epoch": 0.08223795079204584,
|
| 626 |
+
"grad_norm": 0.37109375,
|
| 627 |
+
"learning_rate": 0.0007,
|
| 628 |
+
"loss": 2.513539123535156,
|
| 629 |
+
"step": 1220
|
| 630 |
+
},
|
| 631 |
+
{
|
| 632 |
+
"epoch": 0.0835861139197843,
|
| 633 |
+
"grad_norm": 0.36328125,
|
| 634 |
+
"learning_rate": 0.0007,
|
| 635 |
+
"loss": 2.5009738922119142,
|
| 636 |
+
"step": 1240
|
| 637 |
+
},
|
| 638 |
+
{
|
| 639 |
+
"epoch": 0.08426019548365352,
|
| 640 |
+
"eval_loss": 2.489459753036499,
|
| 641 |
+
"eval_runtime": 15.2493,
|
| 642 |
+
"eval_samples_per_second": 624.749,
|
| 643 |
+
"eval_steps_per_second": 0.656,
|
| 644 |
+
"step": 1250
|
| 645 |
+
},
|
| 646 |
+
{
|
| 647 |
+
"epoch": 0.08493427704752275,
|
| 648 |
+
"grad_norm": 0.388671875,
|
| 649 |
+
"learning_rate": 0.0007,
|
| 650 |
+
"loss": 2.4894077301025392,
|
| 651 |
+
"step": 1260
|
| 652 |
+
},
|
| 653 |
+
{
|
| 654 |
+
"epoch": 0.08628244017526121,
|
| 655 |
+
"grad_norm": 0.369140625,
|
| 656 |
+
"learning_rate": 0.0007,
|
| 657 |
+
"loss": 2.4697439193725588,
|
| 658 |
+
"step": 1280
|
| 659 |
+
},
|
| 660 |
+
{
|
| 661 |
+
"epoch": 0.08763060330299967,
|
| 662 |
+
"grad_norm": 0.423828125,
|
| 663 |
+
"learning_rate": 0.0007,
|
| 664 |
+
"loss": 2.461294174194336,
|
| 665 |
+
"step": 1300
|
| 666 |
+
},
|
| 667 |
+
{
|
| 668 |
+
"epoch": 0.08763060330299967,
|
| 669 |
+
"eval_loss": 2.4627816677093506,
|
| 670 |
+
"eval_runtime": 15.421,
|
| 671 |
+
"eval_samples_per_second": 617.795,
|
| 672 |
+
"eval_steps_per_second": 0.648,
|
| 673 |
+
"step": 1300
|
| 674 |
+
},
|
| 675 |
+
{
|
| 676 |
+
"epoch": 0.08897876643073811,
|
| 677 |
+
"grad_norm": 0.36328125,
|
| 678 |
+
"learning_rate": 0.0007,
|
| 679 |
+
"loss": 2.452680206298828,
|
| 680 |
+
"step": 1320
|
| 681 |
+
},
|
| 682 |
+
{
|
| 683 |
+
"epoch": 0.09032692955847657,
|
| 684 |
+
"grad_norm": 0.404296875,
|
| 685 |
+
"learning_rate": 0.0007,
|
| 686 |
+
"loss": 2.4396684646606444,
|
| 687 |
+
"step": 1340
|
| 688 |
+
},
|
| 689 |
+
{
|
| 690 |
+
"epoch": 0.0910010111223458,
|
| 691 |
+
"eval_loss": 2.437220335006714,
|
| 692 |
+
"eval_runtime": 15.3026,
|
| 693 |
+
"eval_samples_per_second": 622.576,
|
| 694 |
+
"eval_steps_per_second": 0.653,
|
| 695 |
+
"step": 1350
|
| 696 |
+
},
|
| 697 |
+
{
|
| 698 |
+
"epoch": 0.09167509268621503,
|
| 699 |
+
"grad_norm": 0.412109375,
|
| 700 |
+
"learning_rate": 0.0007,
|
| 701 |
+
"loss": 2.4397760391235352,
|
| 702 |
+
"step": 1360
|
| 703 |
+
},
|
| 704 |
+
{
|
| 705 |
+
"epoch": 0.09302325581395349,
|
| 706 |
+
"grad_norm": 0.353515625,
|
| 707 |
+
"learning_rate": 0.0007,
|
| 708 |
+
"loss": 2.4158536911010744,
|
| 709 |
+
"step": 1380
|
| 710 |
+
},
|
| 711 |
+
{
|
| 712 |
+
"epoch": 0.09437141894169195,
|
| 713 |
+
"grad_norm": 0.345703125,
|
| 714 |
+
"learning_rate": 0.0007,
|
| 715 |
+
"loss": 2.406942367553711,
|
| 716 |
+
"step": 1400
|
| 717 |
+
},
|
| 718 |
+
{
|
| 719 |
+
"epoch": 0.09437141894169195,
|
| 720 |
+
"eval_loss": 2.4096601009368896,
|
| 721 |
+
"eval_runtime": 15.357,
|
| 722 |
+
"eval_samples_per_second": 620.368,
|
| 723 |
+
"eval_steps_per_second": 0.651,
|
| 724 |
+
"step": 1400
|
| 725 |
+
},
|
| 726 |
+
{
|
| 727 |
+
"epoch": 0.0957195820694304,
|
| 728 |
+
"grad_norm": 0.36328125,
|
| 729 |
+
"learning_rate": 0.0007,
|
| 730 |
+
"loss": 2.4161144256591798,
|
| 731 |
+
"step": 1420
|
| 732 |
+
},
|
| 733 |
+
{
|
| 734 |
+
"epoch": 0.09706774519716886,
|
| 735 |
+
"grad_norm": 0.36328125,
|
| 736 |
+
"learning_rate": 0.0007,
|
| 737 |
+
"loss": 2.392984390258789,
|
| 738 |
+
"step": 1440
|
| 739 |
+
},
|
| 740 |
+
{
|
| 741 |
+
"epoch": 0.09774182676103808,
|
| 742 |
+
"eval_loss": 2.3937113285064697,
|
| 743 |
+
"eval_runtime": 15.3555,
|
| 744 |
+
"eval_samples_per_second": 620.43,
|
| 745 |
+
"eval_steps_per_second": 0.651,
|
| 746 |
+
"step": 1450
|
| 747 |
+
},
|
| 748 |
+
{
|
| 749 |
+
"epoch": 0.09841590832490732,
|
| 750 |
+
"grad_norm": 0.384765625,
|
| 751 |
+
"learning_rate": 0.0007,
|
| 752 |
+
"loss": 2.3820331573486326,
|
| 753 |
+
"step": 1460
|
| 754 |
+
},
|
| 755 |
+
{
|
| 756 |
+
"epoch": 0.09976407145264576,
|
| 757 |
+
"grad_norm": 0.33984375,
|
| 758 |
+
"learning_rate": 0.0007,
|
| 759 |
+
"loss": 2.3687986373901366,
|
| 760 |
+
"step": 1480
|
| 761 |
+
},
|
| 762 |
+
{
|
| 763 |
+
"epoch": 0.10111223458038422,
|
| 764 |
+
"grad_norm": 0.38671875,
|
| 765 |
+
"learning_rate": 0.0007,
|
| 766 |
+
"loss": 2.355032730102539,
|
| 767 |
+
"step": 1500
|
| 768 |
+
},
|
| 769 |
+
{
|
| 770 |
+
"epoch": 0.10111223458038422,
|
| 771 |
+
"eval_loss": 2.3681657314300537,
|
| 772 |
+
"eval_runtime": 15.3336,
|
| 773 |
+
"eval_samples_per_second": 621.317,
|
| 774 |
+
"eval_steps_per_second": 0.652,
|
| 775 |
+
"step": 1500
|
| 776 |
+
}
|
| 777 |
+
],
|
| 778 |
+
"logging_steps": 20,
|
| 779 |
+
"max_steps": 2500,
|
| 780 |
+
"num_input_tokens_seen": 0,
|
| 781 |
+
"num_train_epochs": 1,
|
| 782 |
+
"save_steps": 100,
|
| 783 |
+
"stateful_callbacks": {
|
| 784 |
+
"TrainerControl": {
|
| 785 |
+
"args": {
|
| 786 |
+
"should_epoch_stop": false,
|
| 787 |
+
"should_evaluate": false,
|
| 788 |
+
"should_log": false,
|
| 789 |
+
"should_save": true,
|
| 790 |
+
"should_training_stop": false
|
| 791 |
+
},
|
| 792 |
+
"attributes": {}
|
| 793 |
+
}
|
| 794 |
+
},
|
| 795 |
+
"total_flos": 4839425703936000.0,
|
| 796 |
+
"train_batch_size": 64,
|
| 797 |
+
"trial_name": null,
|
| 798 |
+
"trial_params": null
|
| 799 |
+
}
|
zain/Activation/out/glu-silu-100L_run/checkpoint-1500/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:fd90512e56f813612dc668676d6f8cab72da110a42f23c61cafebf91064625a1
|
| 3 |
+
size 4920
|
zain/Activation/out/glu-silu-100L_run/checkpoint-1600/config.json
ADDED
|
@@ -0,0 +1,36 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"activation": "silu",
|
| 3 |
+
"architectures": [
|
| 4 |
+
"TinyLlamaForCausalLM"
|
| 5 |
+
],
|
| 6 |
+
"attention_bias": false,
|
| 7 |
+
"attention_dropout": 0.0,
|
| 8 |
+
"bos_token_id": 1,
|
| 9 |
+
"dtype": "bfloat16",
|
| 10 |
+
"eos_token_id": 2,
|
| 11 |
+
"head_dim": 32,
|
| 12 |
+
"hidden_act": "silu",
|
| 13 |
+
"hidden_size": 128,
|
| 14 |
+
"initializer_range": 0.02,
|
| 15 |
+
"intermediate_size": 256,
|
| 16 |
+
"max_position_embeddings": 512,
|
| 17 |
+
"mlp_bias": false,
|
| 18 |
+
"mlp_type": "glu",
|
| 19 |
+
"model_type": "tiny_llama",
|
| 20 |
+
"num_attention_heads": 4,
|
| 21 |
+
"num_hidden_layers": 100,
|
| 22 |
+
"num_key_value_heads": 4,
|
| 23 |
+
"pad_token_id": 0,
|
| 24 |
+
"pretraining_tp": 1,
|
| 25 |
+
"rms_norm_eps": 1e-06,
|
| 26 |
+
"rope_parameters": {
|
| 27 |
+
"rope_theta": 10000.0,
|
| 28 |
+
"rope_type": "default"
|
| 29 |
+
},
|
| 30 |
+
"tie_word_embeddings": true,
|
| 31 |
+
"tokenizer_name": "w-ahmad/tiny-stories-tokenizer",
|
| 32 |
+
"transformers_version": "5.16.0.dev0",
|
| 33 |
+
"use_cache": false,
|
| 34 |
+
"vocab_size": 4096,
|
| 35 |
+
"waleed_beta": 10.0
|
| 36 |
+
}
|
zain/Activation/out/glu-silu-100L_run/checkpoint-1600/model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:64315a4039dcbd96e06cc4bbdd281305344bc2887c7aa227f5b56a91c67964ba
|
| 3 |
+
size 33967272
|
zain/Activation/out/glu-silu-100L_run/checkpoint-1600/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:ed1b1928a8c2037fb288bc2ecfa751a3e600744c0336cbb55ab428930f285b6b
|
| 3 |
+
size 68504996
|
zain/Activation/out/glu-silu-100L_run/checkpoint-1600/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:cbf1acd569be92a8afde23369728c9d2c1c586d4743ed904baa10b88431b0449
|
| 3 |
+
size 14244
|
zain/Activation/out/glu-silu-100L_run/checkpoint-1600/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:a5207460c1b01907ad3d1f40124c12e85beccaebe656a961bcd0bff18a30ffd6
|
| 3 |
+
size 1064
|
zain/Activation/out/glu-silu-100L_run/checkpoint-1600/tokenizer.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
zain/Activation/out/glu-silu-100L_run/checkpoint-1600/tokenizer_config.json
ADDED
|
@@ -0,0 +1,13 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"add_prefix_space": false,
|
| 3 |
+
"backend": "tokenizers",
|
| 4 |
+
"bos_token": "<|endoftext|>",
|
| 5 |
+
"eos_token": "<|endoftext|>",
|
| 6 |
+
"errors": "replace",
|
| 7 |
+
"is_local": false,
|
| 8 |
+
"local_files_only": false,
|
| 9 |
+
"model_max_length": 1024,
|
| 10 |
+
"pad_token": "<|endoftext|>",
|
| 11 |
+
"tokenizer_class": "GPT2Tokenizer",
|
| 12 |
+
"unk_token": "<|endoftext|>"
|
| 13 |
+
}
|
zain/Activation/out/glu-silu-100L_run/checkpoint-1600/trainer_state.json
ADDED
|
@@ -0,0 +1,850 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_global_step": null,
|
| 3 |
+
"best_metric": null,
|
| 4 |
+
"best_model_checkpoint": null,
|
| 5 |
+
"epoch": 0.10785305021907651,
|
| 6 |
+
"eval_steps": 50,
|
| 7 |
+
"global_step": 1600,
|
| 8 |
+
"is_hyper_param_search": false,
|
| 9 |
+
"is_local_process_zero": true,
|
| 10 |
+
"is_world_process_zero": true,
|
| 11 |
+
"log_history": [
|
| 12 |
+
{
|
| 13 |
+
"epoch": 0.0013481631277384564,
|
| 14 |
+
"grad_norm": 2.078125,
|
| 15 |
+
"learning_rate": 2.66e-05,
|
| 16 |
+
"loss": 8.29962387084961,
|
| 17 |
+
"step": 20
|
| 18 |
+
},
|
| 19 |
+
{
|
| 20 |
+
"epoch": 0.002696326255476913,
|
| 21 |
+
"grad_norm": 1.3515625,
|
| 22 |
+
"learning_rate": 5.46e-05,
|
| 23 |
+
"loss": 8.047650909423828,
|
| 24 |
+
"step": 40
|
| 25 |
+
},
|
| 26 |
+
{
|
| 27 |
+
"epoch": 0.003370407819346141,
|
| 28 |
+
"eval_loss": 7.835535049438477,
|
| 29 |
+
"eval_runtime": 15.2165,
|
| 30 |
+
"eval_samples_per_second": 626.095,
|
| 31 |
+
"eval_steps_per_second": 0.657,
|
| 32 |
+
"step": 50
|
| 33 |
+
},
|
| 34 |
+
{
|
| 35 |
+
"epoch": 0.004044489383215369,
|
| 36 |
+
"grad_norm": 1.2734375,
|
| 37 |
+
"learning_rate": 8.259999999999999e-05,
|
| 38 |
+
"loss": 7.8378746032714846,
|
| 39 |
+
"step": 60
|
| 40 |
+
},
|
| 41 |
+
{
|
| 42 |
+
"epoch": 0.005392652510953826,
|
| 43 |
+
"grad_norm": 1.265625,
|
| 44 |
+
"learning_rate": 0.0001106,
|
| 45 |
+
"loss": 7.5728271484375,
|
| 46 |
+
"step": 80
|
| 47 |
+
},
|
| 48 |
+
{
|
| 49 |
+
"epoch": 0.006740815638692282,
|
| 50 |
+
"grad_norm": 1.1953125,
|
| 51 |
+
"learning_rate": 0.0001386,
|
| 52 |
+
"loss": 7.2332916259765625,
|
| 53 |
+
"step": 100
|
| 54 |
+
},
|
| 55 |
+
{
|
| 56 |
+
"epoch": 0.006740815638692282,
|
| 57 |
+
"eval_loss": 7.039053440093994,
|
| 58 |
+
"eval_runtime": 15.2074,
|
| 59 |
+
"eval_samples_per_second": 626.471,
|
| 60 |
+
"eval_steps_per_second": 0.658,
|
| 61 |
+
"step": 100
|
| 62 |
+
},
|
| 63 |
+
{
|
| 64 |
+
"epoch": 0.008088978766430738,
|
| 65 |
+
"grad_norm": 1.1328125,
|
| 66 |
+
"learning_rate": 0.00016659999999999998,
|
| 67 |
+
"loss": 6.8608039855957035,
|
| 68 |
+
"step": 120
|
| 69 |
+
},
|
| 70 |
+
{
|
| 71 |
+
"epoch": 0.009437141894169195,
|
| 72 |
+
"grad_norm": 0.94921875,
|
| 73 |
+
"learning_rate": 0.00019460000000000001,
|
| 74 |
+
"loss": 6.517842864990234,
|
| 75 |
+
"step": 140
|
| 76 |
+
},
|
| 77 |
+
{
|
| 78 |
+
"epoch": 0.010111223458038422,
|
| 79 |
+
"eval_loss": 6.229617595672607,
|
| 80 |
+
"eval_runtime": 15.1978,
|
| 81 |
+
"eval_samples_per_second": 626.867,
|
| 82 |
+
"eval_steps_per_second": 0.658,
|
| 83 |
+
"step": 150
|
| 84 |
+
},
|
| 85 |
+
{
|
| 86 |
+
"epoch": 0.010785305021907651,
|
| 87 |
+
"grad_norm": 0.80078125,
|
| 88 |
+
"learning_rate": 0.0002226,
|
| 89 |
+
"loss": 6.2399169921875,
|
| 90 |
+
"step": 160
|
| 91 |
+
},
|
| 92 |
+
{
|
| 93 |
+
"epoch": 0.012133468149646108,
|
| 94 |
+
"grad_norm": 0.6640625,
|
| 95 |
+
"learning_rate": 0.00025059999999999997,
|
| 96 |
+
"loss": 6.036548614501953,
|
| 97 |
+
"step": 180
|
| 98 |
+
},
|
| 99 |
+
{
|
| 100 |
+
"epoch": 0.013481631277384564,
|
| 101 |
+
"grad_norm": 0.61328125,
|
| 102 |
+
"learning_rate": 0.0002786,
|
| 103 |
+
"loss": 5.857078552246094,
|
| 104 |
+
"step": 200
|
| 105 |
+
},
|
| 106 |
+
{
|
| 107 |
+
"epoch": 0.013481631277384564,
|
| 108 |
+
"eval_loss": 5.765108585357666,
|
| 109 |
+
"eval_runtime": 15.5271,
|
| 110 |
+
"eval_samples_per_second": 613.574,
|
| 111 |
+
"eval_steps_per_second": 0.644,
|
| 112 |
+
"step": 200
|
| 113 |
+
},
|
| 114 |
+
{
|
| 115 |
+
"epoch": 0.01482979440512302,
|
| 116 |
+
"grad_norm": 0.61328125,
|
| 117 |
+
"learning_rate": 0.00030659999999999997,
|
| 118 |
+
"loss": 5.680564117431641,
|
| 119 |
+
"step": 220
|
| 120 |
+
},
|
| 121 |
+
{
|
| 122 |
+
"epoch": 0.016177957532861477,
|
| 123 |
+
"grad_norm": 2.0625,
|
| 124 |
+
"learning_rate": 0.0003346,
|
| 125 |
+
"loss": 5.472259902954102,
|
| 126 |
+
"step": 240
|
| 127 |
+
},
|
| 128 |
+
{
|
| 129 |
+
"epoch": 0.016852039096730706,
|
| 130 |
+
"eval_loss": 5.240335464477539,
|
| 131 |
+
"eval_runtime": 15.2596,
|
| 132 |
+
"eval_samples_per_second": 624.33,
|
| 133 |
+
"eval_steps_per_second": 0.655,
|
| 134 |
+
"step": 250
|
| 135 |
+
},
|
| 136 |
+
{
|
| 137 |
+
"epoch": 0.01752612066059993,
|
| 138 |
+
"grad_norm": 1.7421875,
|
| 139 |
+
"learning_rate": 0.00036260000000000003,
|
| 140 |
+
"loss": 5.24955940246582,
|
| 141 |
+
"step": 260
|
| 142 |
+
},
|
| 143 |
+
{
|
| 144 |
+
"epoch": 0.01887428378833839,
|
| 145 |
+
"grad_norm": 0.62109375,
|
| 146 |
+
"learning_rate": 0.0003906,
|
| 147 |
+
"loss": 4.995759963989258,
|
| 148 |
+
"step": 280
|
| 149 |
+
},
|
| 150 |
+
{
|
| 151 |
+
"epoch": 0.020222446916076844,
|
| 152 |
+
"grad_norm": 1.390625,
|
| 153 |
+
"learning_rate": 0.0004186,
|
| 154 |
+
"loss": 4.769353866577148,
|
| 155 |
+
"step": 300
|
| 156 |
+
},
|
| 157 |
+
{
|
| 158 |
+
"epoch": 0.020222446916076844,
|
| 159 |
+
"eval_loss": 4.669629096984863,
|
| 160 |
+
"eval_runtime": 15.2595,
|
| 161 |
+
"eval_samples_per_second": 624.332,
|
| 162 |
+
"eval_steps_per_second": 0.655,
|
| 163 |
+
"step": 300
|
| 164 |
+
},
|
| 165 |
+
{
|
| 166 |
+
"epoch": 0.021570610043815303,
|
| 167 |
+
"grad_norm": 0.73828125,
|
| 168 |
+
"learning_rate": 0.0004466,
|
| 169 |
+
"loss": 4.569867324829102,
|
| 170 |
+
"step": 320
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 0.022918773171553757,
|
| 174 |
+
"grad_norm": 0.765625,
|
| 175 |
+
"learning_rate": 0.00047460000000000004,
|
| 176 |
+
"loss": 4.389446258544922,
|
| 177 |
+
"step": 340
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 0.023592854735422986,
|
| 181 |
+
"eval_loss": 4.231207847595215,
|
| 182 |
+
"eval_runtime": 15.2143,
|
| 183 |
+
"eval_samples_per_second": 626.188,
|
| 184 |
+
"eval_steps_per_second": 0.657,
|
| 185 |
+
"step": 350
|
| 186 |
+
},
|
| 187 |
+
{
|
| 188 |
+
"epoch": 0.024266936299292215,
|
| 189 |
+
"grad_norm": 1.2421875,
|
| 190 |
+
"learning_rate": 0.0005026,
|
| 191 |
+
"loss": 4.243848419189453,
|
| 192 |
+
"step": 360
|
| 193 |
+
},
|
| 194 |
+
{
|
| 195 |
+
"epoch": 0.02561509942703067,
|
| 196 |
+
"grad_norm": 0.7109375,
|
| 197 |
+
"learning_rate": 0.0005306,
|
| 198 |
+
"loss": 4.1090648651123045,
|
| 199 |
+
"step": 380
|
| 200 |
+
},
|
| 201 |
+
{
|
| 202 |
+
"epoch": 0.026963262554769128,
|
| 203 |
+
"grad_norm": 0.484375,
|
| 204 |
+
"learning_rate": 0.0005586,
|
| 205 |
+
"loss": 3.966172790527344,
|
| 206 |
+
"step": 400
|
| 207 |
+
},
|
| 208 |
+
{
|
| 209 |
+
"epoch": 0.026963262554769128,
|
| 210 |
+
"eval_loss": 3.923773765563965,
|
| 211 |
+
"eval_runtime": 15.5308,
|
| 212 |
+
"eval_samples_per_second": 613.426,
|
| 213 |
+
"eval_steps_per_second": 0.644,
|
| 214 |
+
"step": 400
|
| 215 |
+
},
|
| 216 |
+
{
|
| 217 |
+
"epoch": 0.028311425682507583,
|
| 218 |
+
"grad_norm": 0.6640625,
|
| 219 |
+
"learning_rate": 0.0005866,
|
| 220 |
+
"loss": 3.8559597015380858,
|
| 221 |
+
"step": 420
|
| 222 |
+
},
|
| 223 |
+
{
|
| 224 |
+
"epoch": 0.02965958881024604,
|
| 225 |
+
"grad_norm": 0.515625,
|
| 226 |
+
"learning_rate": 0.0006146,
|
| 227 |
+
"loss": 3.7943866729736326,
|
| 228 |
+
"step": 440
|
| 229 |
+
},
|
| 230 |
+
{
|
| 231 |
+
"epoch": 0.030333670374115267,
|
| 232 |
+
"eval_loss": 3.6977553367614746,
|
| 233 |
+
"eval_runtime": 15.2001,
|
| 234 |
+
"eval_samples_per_second": 626.77,
|
| 235 |
+
"eval_steps_per_second": 0.658,
|
| 236 |
+
"step": 450
|
| 237 |
+
},
|
| 238 |
+
{
|
| 239 |
+
"epoch": 0.031007751937984496,
|
| 240 |
+
"grad_norm": 0.51953125,
|
| 241 |
+
"learning_rate": 0.0006426,
|
| 242 |
+
"loss": 3.686443328857422,
|
| 243 |
+
"step": 460
|
| 244 |
+
},
|
| 245 |
+
{
|
| 246 |
+
"epoch": 0.032355915065722954,
|
| 247 |
+
"grad_norm": 0.62109375,
|
| 248 |
+
"learning_rate": 0.0006705999999999999,
|
| 249 |
+
"loss": 3.6123157501220704,
|
| 250 |
+
"step": 480
|
| 251 |
+
},
|
| 252 |
+
{
|
| 253 |
+
"epoch": 0.03370407819346141,
|
| 254 |
+
"grad_norm": 0.70703125,
|
| 255 |
+
"learning_rate": 0.0006986,
|
| 256 |
+
"loss": 3.542654800415039,
|
| 257 |
+
"step": 500
|
| 258 |
+
},
|
| 259 |
+
{
|
| 260 |
+
"epoch": 0.03370407819346141,
|
| 261 |
+
"eval_loss": 3.513820171356201,
|
| 262 |
+
"eval_runtime": 15.2193,
|
| 263 |
+
"eval_samples_per_second": 625.983,
|
| 264 |
+
"eval_steps_per_second": 0.657,
|
| 265 |
+
"step": 500
|
| 266 |
+
},
|
| 267 |
+
{
|
| 268 |
+
"epoch": 0.03505224132119986,
|
| 269 |
+
"grad_norm": 0.48046875,
|
| 270 |
+
"learning_rate": 0.0007,
|
| 271 |
+
"loss": 3.472199249267578,
|
| 272 |
+
"step": 520
|
| 273 |
+
},
|
| 274 |
+
{
|
| 275 |
+
"epoch": 0.03640040444893832,
|
| 276 |
+
"grad_norm": 0.53515625,
|
| 277 |
+
"learning_rate": 0.0007,
|
| 278 |
+
"loss": 3.396291732788086,
|
| 279 |
+
"step": 540
|
| 280 |
+
},
|
| 281 |
+
{
|
| 282 |
+
"epoch": 0.03707448601280755,
|
| 283 |
+
"eval_loss": 3.3581483364105225,
|
| 284 |
+
"eval_runtime": 15.3044,
|
| 285 |
+
"eval_samples_per_second": 622.502,
|
| 286 |
+
"eval_steps_per_second": 0.653,
|
| 287 |
+
"step": 550
|
| 288 |
+
},
|
| 289 |
+
{
|
| 290 |
+
"epoch": 0.03774856757667678,
|
| 291 |
+
"grad_norm": 0.47265625,
|
| 292 |
+
"learning_rate": 0.0007,
|
| 293 |
+
"loss": 3.347300720214844,
|
| 294 |
+
"step": 560
|
| 295 |
+
},
|
| 296 |
+
{
|
| 297 |
+
"epoch": 0.03909673070441524,
|
| 298 |
+
"grad_norm": 0.458984375,
|
| 299 |
+
"learning_rate": 0.0007,
|
| 300 |
+
"loss": 3.2885902404785154,
|
| 301 |
+
"step": 580
|
| 302 |
+
},
|
| 303 |
+
{
|
| 304 |
+
"epoch": 0.04044489383215369,
|
| 305 |
+
"grad_norm": 0.4375,
|
| 306 |
+
"learning_rate": 0.0007,
|
| 307 |
+
"loss": 3.25723876953125,
|
| 308 |
+
"step": 600
|
| 309 |
+
},
|
| 310 |
+
{
|
| 311 |
+
"epoch": 0.04044489383215369,
|
| 312 |
+
"eval_loss": 3.2229573726654053,
|
| 313 |
+
"eval_runtime": 15.2967,
|
| 314 |
+
"eval_samples_per_second": 622.812,
|
| 315 |
+
"eval_steps_per_second": 0.654,
|
| 316 |
+
"step": 600
|
| 317 |
+
},
|
| 318 |
+
{
|
| 319 |
+
"epoch": 0.04179305695989215,
|
| 320 |
+
"grad_norm": 0.458984375,
|
| 321 |
+
"learning_rate": 0.0007,
|
| 322 |
+
"loss": 3.1902334213256838,
|
| 323 |
+
"step": 620
|
| 324 |
+
},
|
| 325 |
+
{
|
| 326 |
+
"epoch": 0.043141220087630605,
|
| 327 |
+
"grad_norm": 0.40625,
|
| 328 |
+
"learning_rate": 0.0007,
|
| 329 |
+
"loss": 3.167861557006836,
|
| 330 |
+
"step": 640
|
| 331 |
+
},
|
| 332 |
+
{
|
| 333 |
+
"epoch": 0.043815301651499834,
|
| 334 |
+
"eval_loss": 3.120636463165283,
|
| 335 |
+
"eval_runtime": 15.3751,
|
| 336 |
+
"eval_samples_per_second": 619.639,
|
| 337 |
+
"eval_steps_per_second": 0.65,
|
| 338 |
+
"step": 650
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 0.044489383215369056,
|
| 342 |
+
"grad_norm": 0.494140625,
|
| 343 |
+
"learning_rate": 0.0007,
|
| 344 |
+
"loss": 3.1089324951171875,
|
| 345 |
+
"step": 660
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 0.045837546343107514,
|
| 349 |
+
"grad_norm": 0.435546875,
|
| 350 |
+
"learning_rate": 0.0007,
|
| 351 |
+
"loss": 3.0828268051147463,
|
| 352 |
+
"step": 680
|
| 353 |
+
},
|
| 354 |
+
{
|
| 355 |
+
"epoch": 0.04718570947084597,
|
| 356 |
+
"grad_norm": 0.427734375,
|
| 357 |
+
"learning_rate": 0.0007,
|
| 358 |
+
"loss": 3.0396488189697264,
|
| 359 |
+
"step": 700
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"epoch": 0.04718570947084597,
|
| 363 |
+
"eval_loss": 3.0276408195495605,
|
| 364 |
+
"eval_runtime": 15.2735,
|
| 365 |
+
"eval_samples_per_second": 623.761,
|
| 366 |
+
"eval_steps_per_second": 0.655,
|
| 367 |
+
"step": 700
|
| 368 |
+
},
|
| 369 |
+
{
|
| 370 |
+
"epoch": 0.04853387259858443,
|
| 371 |
+
"grad_norm": 0.4375,
|
| 372 |
+
"learning_rate": 0.0007,
|
| 373 |
+
"loss": 3.0034805297851563,
|
| 374 |
+
"step": 720
|
| 375 |
+
},
|
| 376 |
+
{
|
| 377 |
+
"epoch": 0.04988203572632288,
|
| 378 |
+
"grad_norm": 0.52734375,
|
| 379 |
+
"learning_rate": 0.0007,
|
| 380 |
+
"loss": 2.96556396484375,
|
| 381 |
+
"step": 740
|
| 382 |
+
},
|
| 383 |
+
{
|
| 384 |
+
"epoch": 0.05055611729019211,
|
| 385 |
+
"eval_loss": 2.950009346008301,
|
| 386 |
+
"eval_runtime": 15.243,
|
| 387 |
+
"eval_samples_per_second": 625.009,
|
| 388 |
+
"eval_steps_per_second": 0.656,
|
| 389 |
+
"step": 750
|
| 390 |
+
},
|
| 391 |
+
{
|
| 392 |
+
"epoch": 0.05123019885406134,
|
| 393 |
+
"grad_norm": 0.451171875,
|
| 394 |
+
"learning_rate": 0.0007,
|
| 395 |
+
"loss": 2.950779151916504,
|
| 396 |
+
"step": 760
|
| 397 |
+
},
|
| 398 |
+
{
|
| 399 |
+
"epoch": 0.0525783619817998,
|
| 400 |
+
"grad_norm": 0.400390625,
|
| 401 |
+
"learning_rate": 0.0007,
|
| 402 |
+
"loss": 2.956203079223633,
|
| 403 |
+
"step": 780
|
| 404 |
+
},
|
| 405 |
+
{
|
| 406 |
+
"epoch": 0.053926525109538256,
|
| 407 |
+
"grad_norm": 0.40625,
|
| 408 |
+
"learning_rate": 0.0007,
|
| 409 |
+
"loss": 2.9017578125,
|
| 410 |
+
"step": 800
|
| 411 |
+
},
|
| 412 |
+
{
|
| 413 |
+
"epoch": 0.053926525109538256,
|
| 414 |
+
"eval_loss": 2.8844006061553955,
|
| 415 |
+
"eval_runtime": 15.2903,
|
| 416 |
+
"eval_samples_per_second": 623.073,
|
| 417 |
+
"eval_steps_per_second": 0.654,
|
| 418 |
+
"step": 800
|
| 419 |
+
},
|
| 420 |
+
{
|
| 421 |
+
"epoch": 0.05527468823727671,
|
| 422 |
+
"grad_norm": 0.41015625,
|
| 423 |
+
"learning_rate": 0.0007,
|
| 424 |
+
"loss": 2.887515640258789,
|
| 425 |
+
"step": 820
|
| 426 |
+
},
|
| 427 |
+
{
|
| 428 |
+
"epoch": 0.056622851365015166,
|
| 429 |
+
"grad_norm": 0.41796875,
|
| 430 |
+
"learning_rate": 0.0007,
|
| 431 |
+
"loss": 2.8556928634643555,
|
| 432 |
+
"step": 840
|
| 433 |
+
},
|
| 434 |
+
{
|
| 435 |
+
"epoch": 0.057296932928884395,
|
| 436 |
+
"eval_loss": 2.821610450744629,
|
| 437 |
+
"eval_runtime": 15.2253,
|
| 438 |
+
"eval_samples_per_second": 625.733,
|
| 439 |
+
"eval_steps_per_second": 0.657,
|
| 440 |
+
"step": 850
|
| 441 |
+
},
|
| 442 |
+
{
|
| 443 |
+
"epoch": 0.057971014492753624,
|
| 444 |
+
"grad_norm": 0.423828125,
|
| 445 |
+
"learning_rate": 0.0007,
|
| 446 |
+
"loss": 2.8130855560302734,
|
| 447 |
+
"step": 860
|
| 448 |
+
},
|
| 449 |
+
{
|
| 450 |
+
"epoch": 0.05931917762049208,
|
| 451 |
+
"grad_norm": 0.38671875,
|
| 452 |
+
"learning_rate": 0.0007,
|
| 453 |
+
"loss": 2.790329170227051,
|
| 454 |
+
"step": 880
|
| 455 |
+
},
|
| 456 |
+
{
|
| 457 |
+
"epoch": 0.06066734074823053,
|
| 458 |
+
"grad_norm": 0.427734375,
|
| 459 |
+
"learning_rate": 0.0007,
|
| 460 |
+
"loss": 2.7760162353515625,
|
| 461 |
+
"step": 900
|
| 462 |
+
},
|
| 463 |
+
{
|
| 464 |
+
"epoch": 0.06066734074823053,
|
| 465 |
+
"eval_loss": 2.7661373615264893,
|
| 466 |
+
"eval_runtime": 15.2736,
|
| 467 |
+
"eval_samples_per_second": 623.755,
|
| 468 |
+
"eval_steps_per_second": 0.655,
|
| 469 |
+
"step": 900
|
| 470 |
+
},
|
| 471 |
+
{
|
| 472 |
+
"epoch": 0.06201550387596899,
|
| 473 |
+
"grad_norm": 0.435546875,
|
| 474 |
+
"learning_rate": 0.0007,
|
| 475 |
+
"loss": 2.7462047576904296,
|
| 476 |
+
"step": 920
|
| 477 |
+
},
|
| 478 |
+
{
|
| 479 |
+
"epoch": 0.06336366700370745,
|
| 480 |
+
"grad_norm": 0.4140625,
|
| 481 |
+
"learning_rate": 0.0007,
|
| 482 |
+
"loss": 2.732274627685547,
|
| 483 |
+
"step": 940
|
| 484 |
+
},
|
| 485 |
+
{
|
| 486 |
+
"epoch": 0.06403774856757667,
|
| 487 |
+
"eval_loss": 2.713942289352417,
|
| 488 |
+
"eval_runtime": 15.6062,
|
| 489 |
+
"eval_samples_per_second": 610.463,
|
| 490 |
+
"eval_steps_per_second": 0.641,
|
| 491 |
+
"step": 950
|
| 492 |
+
},
|
| 493 |
+
{
|
| 494 |
+
"epoch": 0.06471183013144591,
|
| 495 |
+
"grad_norm": 0.419921875,
|
| 496 |
+
"learning_rate": 0.0007,
|
| 497 |
+
"loss": 2.705318069458008,
|
| 498 |
+
"step": 960
|
| 499 |
+
},
|
| 500 |
+
{
|
| 501 |
+
"epoch": 0.06605999325918437,
|
| 502 |
+
"grad_norm": 0.4296875,
|
| 503 |
+
"learning_rate": 0.0007,
|
| 504 |
+
"loss": 2.6892818450927733,
|
| 505 |
+
"step": 980
|
| 506 |
+
},
|
| 507 |
+
{
|
| 508 |
+
"epoch": 0.06740815638692282,
|
| 509 |
+
"grad_norm": 0.4453125,
|
| 510 |
+
"learning_rate": 0.0007,
|
| 511 |
+
"loss": 2.687115287780762,
|
| 512 |
+
"step": 1000
|
| 513 |
+
},
|
| 514 |
+
{
|
| 515 |
+
"epoch": 0.06740815638692282,
|
| 516 |
+
"eval_loss": 2.668647289276123,
|
| 517 |
+
"eval_runtime": 15.2597,
|
| 518 |
+
"eval_samples_per_second": 624.326,
|
| 519 |
+
"eval_steps_per_second": 0.655,
|
| 520 |
+
"step": 1000
|
| 521 |
+
},
|
| 522 |
+
{
|
| 523 |
+
"epoch": 0.06875631951466127,
|
| 524 |
+
"grad_norm": 0.388671875,
|
| 525 |
+
"learning_rate": 0.0007,
|
| 526 |
+
"loss": 2.6533071517944338,
|
| 527 |
+
"step": 1020
|
| 528 |
+
},
|
| 529 |
+
{
|
| 530 |
+
"epoch": 0.07010448264239973,
|
| 531 |
+
"grad_norm": 0.392578125,
|
| 532 |
+
"learning_rate": 0.0007,
|
| 533 |
+
"loss": 2.6390922546386717,
|
| 534 |
+
"step": 1040
|
| 535 |
+
},
|
| 536 |
+
{
|
| 537 |
+
"epoch": 0.07077856420626896,
|
| 538 |
+
"eval_loss": 2.6291816234588623,
|
| 539 |
+
"eval_runtime": 15.2457,
|
| 540 |
+
"eval_samples_per_second": 624.896,
|
| 541 |
+
"eval_steps_per_second": 0.656,
|
| 542 |
+
"step": 1050
|
| 543 |
+
},
|
| 544 |
+
{
|
| 545 |
+
"epoch": 0.07145264577013818,
|
| 546 |
+
"grad_norm": 0.4375,
|
| 547 |
+
"learning_rate": 0.0007,
|
| 548 |
+
"loss": 2.624461555480957,
|
| 549 |
+
"step": 1060
|
| 550 |
+
},
|
| 551 |
+
{
|
| 552 |
+
"epoch": 0.07280080889787664,
|
| 553 |
+
"grad_norm": 0.396484375,
|
| 554 |
+
"learning_rate": 0.0007,
|
| 555 |
+
"loss": 2.6044427871704103,
|
| 556 |
+
"step": 1080
|
| 557 |
+
},
|
| 558 |
+
{
|
| 559 |
+
"epoch": 0.0741489720256151,
|
| 560 |
+
"grad_norm": 0.423828125,
|
| 561 |
+
"learning_rate": 0.0007,
|
| 562 |
+
"loss": 2.597646713256836,
|
| 563 |
+
"step": 1100
|
| 564 |
+
},
|
| 565 |
+
{
|
| 566 |
+
"epoch": 0.0741489720256151,
|
| 567 |
+
"eval_loss": 2.589696168899536,
|
| 568 |
+
"eval_runtime": 15.3967,
|
| 569 |
+
"eval_samples_per_second": 618.768,
|
| 570 |
+
"eval_steps_per_second": 0.649,
|
| 571 |
+
"step": 1100
|
| 572 |
+
},
|
| 573 |
+
{
|
| 574 |
+
"epoch": 0.07549713515335356,
|
| 575 |
+
"grad_norm": 0.369140625,
|
| 576 |
+
"learning_rate": 0.0007,
|
| 577 |
+
"loss": 2.561947250366211,
|
| 578 |
+
"step": 1120
|
| 579 |
+
},
|
| 580 |
+
{
|
| 581 |
+
"epoch": 0.07684529828109202,
|
| 582 |
+
"grad_norm": 0.390625,
|
| 583 |
+
"learning_rate": 0.0007,
|
| 584 |
+
"loss": 2.54862117767334,
|
| 585 |
+
"step": 1140
|
| 586 |
+
},
|
| 587 |
+
{
|
| 588 |
+
"epoch": 0.07751937984496124,
|
| 589 |
+
"eval_loss": 2.559185743331909,
|
| 590 |
+
"eval_runtime": 15.2877,
|
| 591 |
+
"eval_samples_per_second": 623.181,
|
| 592 |
+
"eval_steps_per_second": 0.654,
|
| 593 |
+
"step": 1150
|
| 594 |
+
},
|
| 595 |
+
{
|
| 596 |
+
"epoch": 0.07819346140883048,
|
| 597 |
+
"grad_norm": 0.37890625,
|
| 598 |
+
"learning_rate": 0.0007,
|
| 599 |
+
"loss": 2.5469717025756835,
|
| 600 |
+
"step": 1160
|
| 601 |
+
},
|
| 602 |
+
{
|
| 603 |
+
"epoch": 0.07954162453656892,
|
| 604 |
+
"grad_norm": 0.400390625,
|
| 605 |
+
"learning_rate": 0.0007,
|
| 606 |
+
"loss": 2.540451240539551,
|
| 607 |
+
"step": 1180
|
| 608 |
+
},
|
| 609 |
+
{
|
| 610 |
+
"epoch": 0.08088978766430738,
|
| 611 |
+
"grad_norm": 0.40625,
|
| 612 |
+
"learning_rate": 0.0007,
|
| 613 |
+
"loss": 2.53134708404541,
|
| 614 |
+
"step": 1200
|
| 615 |
+
},
|
| 616 |
+
{
|
| 617 |
+
"epoch": 0.08088978766430738,
|
| 618 |
+
"eval_loss": 2.521665096282959,
|
| 619 |
+
"eval_runtime": 15.2822,
|
| 620 |
+
"eval_samples_per_second": 623.407,
|
| 621 |
+
"eval_steps_per_second": 0.654,
|
| 622 |
+
"step": 1200
|
| 623 |
+
},
|
| 624 |
+
{
|
| 625 |
+
"epoch": 0.08223795079204584,
|
| 626 |
+
"grad_norm": 0.37109375,
|
| 627 |
+
"learning_rate": 0.0007,
|
| 628 |
+
"loss": 2.513539123535156,
|
| 629 |
+
"step": 1220
|
| 630 |
+
},
|
| 631 |
+
{
|
| 632 |
+
"epoch": 0.0835861139197843,
|
| 633 |
+
"grad_norm": 0.36328125,
|
| 634 |
+
"learning_rate": 0.0007,
|
| 635 |
+
"loss": 2.5009738922119142,
|
| 636 |
+
"step": 1240
|
| 637 |
+
},
|
| 638 |
+
{
|
| 639 |
+
"epoch": 0.08426019548365352,
|
| 640 |
+
"eval_loss": 2.489459753036499,
|
| 641 |
+
"eval_runtime": 15.2493,
|
| 642 |
+
"eval_samples_per_second": 624.749,
|
| 643 |
+
"eval_steps_per_second": 0.656,
|
| 644 |
+
"step": 1250
|
| 645 |
+
},
|
| 646 |
+
{
|
| 647 |
+
"epoch": 0.08493427704752275,
|
| 648 |
+
"grad_norm": 0.388671875,
|
| 649 |
+
"learning_rate": 0.0007,
|
| 650 |
+
"loss": 2.4894077301025392,
|
| 651 |
+
"step": 1260
|
| 652 |
+
},
|
| 653 |
+
{
|
| 654 |
+
"epoch": 0.08628244017526121,
|
| 655 |
+
"grad_norm": 0.369140625,
|
| 656 |
+
"learning_rate": 0.0007,
|
| 657 |
+
"loss": 2.4697439193725588,
|
| 658 |
+
"step": 1280
|
| 659 |
+
},
|
| 660 |
+
{
|
| 661 |
+
"epoch": 0.08763060330299967,
|
| 662 |
+
"grad_norm": 0.423828125,
|
| 663 |
+
"learning_rate": 0.0007,
|
| 664 |
+
"loss": 2.461294174194336,
|
| 665 |
+
"step": 1300
|
| 666 |
+
},
|
| 667 |
+
{
|
| 668 |
+
"epoch": 0.08763060330299967,
|
| 669 |
+
"eval_loss": 2.4627816677093506,
|
| 670 |
+
"eval_runtime": 15.421,
|
| 671 |
+
"eval_samples_per_second": 617.795,
|
| 672 |
+
"eval_steps_per_second": 0.648,
|
| 673 |
+
"step": 1300
|
| 674 |
+
},
|
| 675 |
+
{
|
| 676 |
+
"epoch": 0.08897876643073811,
|
| 677 |
+
"grad_norm": 0.36328125,
|
| 678 |
+
"learning_rate": 0.0007,
|
| 679 |
+
"loss": 2.452680206298828,
|
| 680 |
+
"step": 1320
|
| 681 |
+
},
|
| 682 |
+
{
|
| 683 |
+
"epoch": 0.09032692955847657,
|
| 684 |
+
"grad_norm": 0.404296875,
|
| 685 |
+
"learning_rate": 0.0007,
|
| 686 |
+
"loss": 2.4396684646606444,
|
| 687 |
+
"step": 1340
|
| 688 |
+
},
|
| 689 |
+
{
|
| 690 |
+
"epoch": 0.0910010111223458,
|
| 691 |
+
"eval_loss": 2.437220335006714,
|
| 692 |
+
"eval_runtime": 15.3026,
|
| 693 |
+
"eval_samples_per_second": 622.576,
|
| 694 |
+
"eval_steps_per_second": 0.653,
|
| 695 |
+
"step": 1350
|
| 696 |
+
},
|
| 697 |
+
{
|
| 698 |
+
"epoch": 0.09167509268621503,
|
| 699 |
+
"grad_norm": 0.412109375,
|
| 700 |
+
"learning_rate": 0.0007,
|
| 701 |
+
"loss": 2.4397760391235352,
|
| 702 |
+
"step": 1360
|
| 703 |
+
},
|
| 704 |
+
{
|
| 705 |
+
"epoch": 0.09302325581395349,
|
| 706 |
+
"grad_norm": 0.353515625,
|
| 707 |
+
"learning_rate": 0.0007,
|
| 708 |
+
"loss": 2.4158536911010744,
|
| 709 |
+
"step": 1380
|
| 710 |
+
},
|
| 711 |
+
{
|
| 712 |
+
"epoch": 0.09437141894169195,
|
| 713 |
+
"grad_norm": 0.345703125,
|
| 714 |
+
"learning_rate": 0.0007,
|
| 715 |
+
"loss": 2.406942367553711,
|
| 716 |
+
"step": 1400
|
| 717 |
+
},
|
| 718 |
+
{
|
| 719 |
+
"epoch": 0.09437141894169195,
|
| 720 |
+
"eval_loss": 2.4096601009368896,
|
| 721 |
+
"eval_runtime": 15.357,
|
| 722 |
+
"eval_samples_per_second": 620.368,
|
| 723 |
+
"eval_steps_per_second": 0.651,
|
| 724 |
+
"step": 1400
|
| 725 |
+
},
|
| 726 |
+
{
|
| 727 |
+
"epoch": 0.0957195820694304,
|
| 728 |
+
"grad_norm": 0.36328125,
|
| 729 |
+
"learning_rate": 0.0007,
|
| 730 |
+
"loss": 2.4161144256591798,
|
| 731 |
+
"step": 1420
|
| 732 |
+
},
|
| 733 |
+
{
|
| 734 |
+
"epoch": 0.09706774519716886,
|
| 735 |
+
"grad_norm": 0.36328125,
|
| 736 |
+
"learning_rate": 0.0007,
|
| 737 |
+
"loss": 2.392984390258789,
|
| 738 |
+
"step": 1440
|
| 739 |
+
},
|
| 740 |
+
{
|
| 741 |
+
"epoch": 0.09774182676103808,
|
| 742 |
+
"eval_loss": 2.3937113285064697,
|
| 743 |
+
"eval_runtime": 15.3555,
|
| 744 |
+
"eval_samples_per_second": 620.43,
|
| 745 |
+
"eval_steps_per_second": 0.651,
|
| 746 |
+
"step": 1450
|
| 747 |
+
},
|
| 748 |
+
{
|
| 749 |
+
"epoch": 0.09841590832490732,
|
| 750 |
+
"grad_norm": 0.384765625,
|
| 751 |
+
"learning_rate": 0.0007,
|
| 752 |
+
"loss": 2.3820331573486326,
|
| 753 |
+
"step": 1460
|
| 754 |
+
},
|
| 755 |
+
{
|
| 756 |
+
"epoch": 0.09976407145264576,
|
| 757 |
+
"grad_norm": 0.33984375,
|
| 758 |
+
"learning_rate": 0.0007,
|
| 759 |
+
"loss": 2.3687986373901366,
|
| 760 |
+
"step": 1480
|
| 761 |
+
},
|
| 762 |
+
{
|
| 763 |
+
"epoch": 0.10111223458038422,
|
| 764 |
+
"grad_norm": 0.38671875,
|
| 765 |
+
"learning_rate": 0.0007,
|
| 766 |
+
"loss": 2.355032730102539,
|
| 767 |
+
"step": 1500
|
| 768 |
+
},
|
| 769 |
+
{
|
| 770 |
+
"epoch": 0.10111223458038422,
|
| 771 |
+
"eval_loss": 2.3681657314300537,
|
| 772 |
+
"eval_runtime": 15.3336,
|
| 773 |
+
"eval_samples_per_second": 621.317,
|
| 774 |
+
"eval_steps_per_second": 0.652,
|
| 775 |
+
"step": 1500
|
| 776 |
+
},
|
| 777 |
+
{
|
| 778 |
+
"epoch": 0.10246039770812268,
|
| 779 |
+
"grad_norm": 0.392578125,
|
| 780 |
+
"learning_rate": 0.0007,
|
| 781 |
+
"loss": 2.3572179794311525,
|
| 782 |
+
"step": 1520
|
| 783 |
+
},
|
| 784 |
+
{
|
| 785 |
+
"epoch": 0.10380856083586114,
|
| 786 |
+
"grad_norm": 0.35546875,
|
| 787 |
+
"learning_rate": 0.0007,
|
| 788 |
+
"loss": 2.345840835571289,
|
| 789 |
+
"step": 1540
|
| 790 |
+
},
|
| 791 |
+
{
|
| 792 |
+
"epoch": 0.10448264239973036,
|
| 793 |
+
"eval_loss": 2.3475193977355957,
|
| 794 |
+
"eval_runtime": 15.3453,
|
| 795 |
+
"eval_samples_per_second": 620.842,
|
| 796 |
+
"eval_steps_per_second": 0.652,
|
| 797 |
+
"step": 1550
|
| 798 |
+
},
|
| 799 |
+
{
|
| 800 |
+
"epoch": 0.1051567239635996,
|
| 801 |
+
"grad_norm": 0.388671875,
|
| 802 |
+
"learning_rate": 0.0007,
|
| 803 |
+
"loss": 2.335231971740723,
|
| 804 |
+
"step": 1560
|
| 805 |
+
},
|
| 806 |
+
{
|
| 807 |
+
"epoch": 0.10650488709133805,
|
| 808 |
+
"grad_norm": 0.359375,
|
| 809 |
+
"learning_rate": 0.0007,
|
| 810 |
+
"loss": 2.3543283462524416,
|
| 811 |
+
"step": 1580
|
| 812 |
+
},
|
| 813 |
+
{
|
| 814 |
+
"epoch": 0.10785305021907651,
|
| 815 |
+
"grad_norm": 0.39453125,
|
| 816 |
+
"learning_rate": 0.0007,
|
| 817 |
+
"loss": 2.3237049102783205,
|
| 818 |
+
"step": 1600
|
| 819 |
+
},
|
| 820 |
+
{
|
| 821 |
+
"epoch": 0.10785305021907651,
|
| 822 |
+
"eval_loss": 2.3377676010131836,
|
| 823 |
+
"eval_runtime": 15.2858,
|
| 824 |
+
"eval_samples_per_second": 623.259,
|
| 825 |
+
"eval_steps_per_second": 0.654,
|
| 826 |
+
"step": 1600
|
| 827 |
+
}
|
| 828 |
+
],
|
| 829 |
+
"logging_steps": 20,
|
| 830 |
+
"max_steps": 2500,
|
| 831 |
+
"num_input_tokens_seen": 0,
|
| 832 |
+
"num_train_epochs": 1,
|
| 833 |
+
"save_steps": 100,
|
| 834 |
+
"stateful_callbacks": {
|
| 835 |
+
"TrainerControl": {
|
| 836 |
+
"args": {
|
| 837 |
+
"should_epoch_stop": false,
|
| 838 |
+
"should_evaluate": false,
|
| 839 |
+
"should_log": false,
|
| 840 |
+
"should_save": true,
|
| 841 |
+
"should_training_stop": false
|
| 842 |
+
},
|
| 843 |
+
"attributes": {}
|
| 844 |
+
}
|
| 845 |
+
},
|
| 846 |
+
"total_flos": 5162054084198400.0,
|
| 847 |
+
"train_batch_size": 64,
|
| 848 |
+
"trial_name": null,
|
| 849 |
+
"trial_params": null
|
| 850 |
+
}
|
zain/Activation/out/glu-silu-100L_run/checkpoint-1600/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:fd90512e56f813612dc668676d6f8cab72da110a42f23c61cafebf91064625a1
|
| 3 |
+
size 4920
|
zain/Activation/out/glu-silu-100L_run/checkpoint-1700/config.json
ADDED
|
@@ -0,0 +1,36 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"activation": "silu",
|
| 3 |
+
"architectures": [
|
| 4 |
+
"TinyLlamaForCausalLM"
|
| 5 |
+
],
|
| 6 |
+
"attention_bias": false,
|
| 7 |
+
"attention_dropout": 0.0,
|
| 8 |
+
"bos_token_id": 1,
|
| 9 |
+
"dtype": "bfloat16",
|
| 10 |
+
"eos_token_id": 2,
|
| 11 |
+
"head_dim": 32,
|
| 12 |
+
"hidden_act": "silu",
|
| 13 |
+
"hidden_size": 128,
|
| 14 |
+
"initializer_range": 0.02,
|
| 15 |
+
"intermediate_size": 256,
|
| 16 |
+
"max_position_embeddings": 512,
|
| 17 |
+
"mlp_bias": false,
|
| 18 |
+
"mlp_type": "glu",
|
| 19 |
+
"model_type": "tiny_llama",
|
| 20 |
+
"num_attention_heads": 4,
|
| 21 |
+
"num_hidden_layers": 100,
|
| 22 |
+
"num_key_value_heads": 4,
|
| 23 |
+
"pad_token_id": 0,
|
| 24 |
+
"pretraining_tp": 1,
|
| 25 |
+
"rms_norm_eps": 1e-06,
|
| 26 |
+
"rope_parameters": {
|
| 27 |
+
"rope_theta": 10000.0,
|
| 28 |
+
"rope_type": "default"
|
| 29 |
+
},
|
| 30 |
+
"tie_word_embeddings": true,
|
| 31 |
+
"tokenizer_name": "w-ahmad/tiny-stories-tokenizer",
|
| 32 |
+
"transformers_version": "5.16.0.dev0",
|
| 33 |
+
"use_cache": false,
|
| 34 |
+
"vocab_size": 4096,
|
| 35 |
+
"waleed_beta": 10.0
|
| 36 |
+
}
|
zain/Activation/out/glu-silu-100L_run/checkpoint-1700/model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:55fa08f968f4be79e976594f1d93283a3f0894e4b6c721466be01a88a6a29e7f
|
| 3 |
+
size 33967272
|
zain/Activation/out/glu-silu-100L_run/checkpoint-1700/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:c8362849f4262d5eaefe1be46c05ac9ee235447745b31b08f2f367ad3e20ee0f
|
| 3 |
+
size 68504996
|
zain/Activation/out/glu-silu-100L_run/checkpoint-1700/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:ae8d8ba6af5b03ddf91f8fca0d75b831aa7a8fb44968fbf47d7d42fb9ff92ad9
|
| 3 |
+
size 14244
|
zain/Activation/out/glu-silu-100L_run/checkpoint-1700/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:da239a902a87900de7861a410c642a5ae70d481b0020a3040c363cc3ced9b7a4
|
| 3 |
+
size 1064
|
zain/Activation/out/glu-silu-100L_run/checkpoint-1700/tokenizer.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
zain/Activation/out/glu-silu-100L_run/checkpoint-1700/tokenizer_config.json
ADDED
|
@@ -0,0 +1,13 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"add_prefix_space": false,
|
| 3 |
+
"backend": "tokenizers",
|
| 4 |
+
"bos_token": "<|endoftext|>",
|
| 5 |
+
"eos_token": "<|endoftext|>",
|
| 6 |
+
"errors": "replace",
|
| 7 |
+
"is_local": false,
|
| 8 |
+
"local_files_only": false,
|
| 9 |
+
"model_max_length": 1024,
|
| 10 |
+
"pad_token": "<|endoftext|>",
|
| 11 |
+
"tokenizer_class": "GPT2Tokenizer",
|
| 12 |
+
"unk_token": "<|endoftext|>"
|
| 13 |
+
}
|
zain/Activation/out/glu-silu-100L_run/checkpoint-1700/trainer_state.json
ADDED
|
@@ -0,0 +1,901 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_global_step": null,
|
| 3 |
+
"best_metric": null,
|
| 4 |
+
"best_model_checkpoint": null,
|
| 5 |
+
"epoch": 0.11459386585776879,
|
| 6 |
+
"eval_steps": 50,
|
| 7 |
+
"global_step": 1700,
|
| 8 |
+
"is_hyper_param_search": false,
|
| 9 |
+
"is_local_process_zero": true,
|
| 10 |
+
"is_world_process_zero": true,
|
| 11 |
+
"log_history": [
|
| 12 |
+
{
|
| 13 |
+
"epoch": 0.0013481631277384564,
|
| 14 |
+
"grad_norm": 2.078125,
|
| 15 |
+
"learning_rate": 2.66e-05,
|
| 16 |
+
"loss": 8.29962387084961,
|
| 17 |
+
"step": 20
|
| 18 |
+
},
|
| 19 |
+
{
|
| 20 |
+
"epoch": 0.002696326255476913,
|
| 21 |
+
"grad_norm": 1.3515625,
|
| 22 |
+
"learning_rate": 5.46e-05,
|
| 23 |
+
"loss": 8.047650909423828,
|
| 24 |
+
"step": 40
|
| 25 |
+
},
|
| 26 |
+
{
|
| 27 |
+
"epoch": 0.003370407819346141,
|
| 28 |
+
"eval_loss": 7.835535049438477,
|
| 29 |
+
"eval_runtime": 15.2165,
|
| 30 |
+
"eval_samples_per_second": 626.095,
|
| 31 |
+
"eval_steps_per_second": 0.657,
|
| 32 |
+
"step": 50
|
| 33 |
+
},
|
| 34 |
+
{
|
| 35 |
+
"epoch": 0.004044489383215369,
|
| 36 |
+
"grad_norm": 1.2734375,
|
| 37 |
+
"learning_rate": 8.259999999999999e-05,
|
| 38 |
+
"loss": 7.8378746032714846,
|
| 39 |
+
"step": 60
|
| 40 |
+
},
|
| 41 |
+
{
|
| 42 |
+
"epoch": 0.005392652510953826,
|
| 43 |
+
"grad_norm": 1.265625,
|
| 44 |
+
"learning_rate": 0.0001106,
|
| 45 |
+
"loss": 7.5728271484375,
|
| 46 |
+
"step": 80
|
| 47 |
+
},
|
| 48 |
+
{
|
| 49 |
+
"epoch": 0.006740815638692282,
|
| 50 |
+
"grad_norm": 1.1953125,
|
| 51 |
+
"learning_rate": 0.0001386,
|
| 52 |
+
"loss": 7.2332916259765625,
|
| 53 |
+
"step": 100
|
| 54 |
+
},
|
| 55 |
+
{
|
| 56 |
+
"epoch": 0.006740815638692282,
|
| 57 |
+
"eval_loss": 7.039053440093994,
|
| 58 |
+
"eval_runtime": 15.2074,
|
| 59 |
+
"eval_samples_per_second": 626.471,
|
| 60 |
+
"eval_steps_per_second": 0.658,
|
| 61 |
+
"step": 100
|
| 62 |
+
},
|
| 63 |
+
{
|
| 64 |
+
"epoch": 0.008088978766430738,
|
| 65 |
+
"grad_norm": 1.1328125,
|
| 66 |
+
"learning_rate": 0.00016659999999999998,
|
| 67 |
+
"loss": 6.8608039855957035,
|
| 68 |
+
"step": 120
|
| 69 |
+
},
|
| 70 |
+
{
|
| 71 |
+
"epoch": 0.009437141894169195,
|
| 72 |
+
"grad_norm": 0.94921875,
|
| 73 |
+
"learning_rate": 0.00019460000000000001,
|
| 74 |
+
"loss": 6.517842864990234,
|
| 75 |
+
"step": 140
|
| 76 |
+
},
|
| 77 |
+
{
|
| 78 |
+
"epoch": 0.010111223458038422,
|
| 79 |
+
"eval_loss": 6.229617595672607,
|
| 80 |
+
"eval_runtime": 15.1978,
|
| 81 |
+
"eval_samples_per_second": 626.867,
|
| 82 |
+
"eval_steps_per_second": 0.658,
|
| 83 |
+
"step": 150
|
| 84 |
+
},
|
| 85 |
+
{
|
| 86 |
+
"epoch": 0.010785305021907651,
|
| 87 |
+
"grad_norm": 0.80078125,
|
| 88 |
+
"learning_rate": 0.0002226,
|
| 89 |
+
"loss": 6.2399169921875,
|
| 90 |
+
"step": 160
|
| 91 |
+
},
|
| 92 |
+
{
|
| 93 |
+
"epoch": 0.012133468149646108,
|
| 94 |
+
"grad_norm": 0.6640625,
|
| 95 |
+
"learning_rate": 0.00025059999999999997,
|
| 96 |
+
"loss": 6.036548614501953,
|
| 97 |
+
"step": 180
|
| 98 |
+
},
|
| 99 |
+
{
|
| 100 |
+
"epoch": 0.013481631277384564,
|
| 101 |
+
"grad_norm": 0.61328125,
|
| 102 |
+
"learning_rate": 0.0002786,
|
| 103 |
+
"loss": 5.857078552246094,
|
| 104 |
+
"step": 200
|
| 105 |
+
},
|
| 106 |
+
{
|
| 107 |
+
"epoch": 0.013481631277384564,
|
| 108 |
+
"eval_loss": 5.765108585357666,
|
| 109 |
+
"eval_runtime": 15.5271,
|
| 110 |
+
"eval_samples_per_second": 613.574,
|
| 111 |
+
"eval_steps_per_second": 0.644,
|
| 112 |
+
"step": 200
|
| 113 |
+
},
|
| 114 |
+
{
|
| 115 |
+
"epoch": 0.01482979440512302,
|
| 116 |
+
"grad_norm": 0.61328125,
|
| 117 |
+
"learning_rate": 0.00030659999999999997,
|
| 118 |
+
"loss": 5.680564117431641,
|
| 119 |
+
"step": 220
|
| 120 |
+
},
|
| 121 |
+
{
|
| 122 |
+
"epoch": 0.016177957532861477,
|
| 123 |
+
"grad_norm": 2.0625,
|
| 124 |
+
"learning_rate": 0.0003346,
|
| 125 |
+
"loss": 5.472259902954102,
|
| 126 |
+
"step": 240
|
| 127 |
+
},
|
| 128 |
+
{
|
| 129 |
+
"epoch": 0.016852039096730706,
|
| 130 |
+
"eval_loss": 5.240335464477539,
|
| 131 |
+
"eval_runtime": 15.2596,
|
| 132 |
+
"eval_samples_per_second": 624.33,
|
| 133 |
+
"eval_steps_per_second": 0.655,
|
| 134 |
+
"step": 250
|
| 135 |
+
},
|
| 136 |
+
{
|
| 137 |
+
"epoch": 0.01752612066059993,
|
| 138 |
+
"grad_norm": 1.7421875,
|
| 139 |
+
"learning_rate": 0.00036260000000000003,
|
| 140 |
+
"loss": 5.24955940246582,
|
| 141 |
+
"step": 260
|
| 142 |
+
},
|
| 143 |
+
{
|
| 144 |
+
"epoch": 0.01887428378833839,
|
| 145 |
+
"grad_norm": 0.62109375,
|
| 146 |
+
"learning_rate": 0.0003906,
|
| 147 |
+
"loss": 4.995759963989258,
|
| 148 |
+
"step": 280
|
| 149 |
+
},
|
| 150 |
+
{
|
| 151 |
+
"epoch": 0.020222446916076844,
|
| 152 |
+
"grad_norm": 1.390625,
|
| 153 |
+
"learning_rate": 0.0004186,
|
| 154 |
+
"loss": 4.769353866577148,
|
| 155 |
+
"step": 300
|
| 156 |
+
},
|
| 157 |
+
{
|
| 158 |
+
"epoch": 0.020222446916076844,
|
| 159 |
+
"eval_loss": 4.669629096984863,
|
| 160 |
+
"eval_runtime": 15.2595,
|
| 161 |
+
"eval_samples_per_second": 624.332,
|
| 162 |
+
"eval_steps_per_second": 0.655,
|
| 163 |
+
"step": 300
|
| 164 |
+
},
|
| 165 |
+
{
|
| 166 |
+
"epoch": 0.021570610043815303,
|
| 167 |
+
"grad_norm": 0.73828125,
|
| 168 |
+
"learning_rate": 0.0004466,
|
| 169 |
+
"loss": 4.569867324829102,
|
| 170 |
+
"step": 320
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 0.022918773171553757,
|
| 174 |
+
"grad_norm": 0.765625,
|
| 175 |
+
"learning_rate": 0.00047460000000000004,
|
| 176 |
+
"loss": 4.389446258544922,
|
| 177 |
+
"step": 340
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 0.023592854735422986,
|
| 181 |
+
"eval_loss": 4.231207847595215,
|
| 182 |
+
"eval_runtime": 15.2143,
|
| 183 |
+
"eval_samples_per_second": 626.188,
|
| 184 |
+
"eval_steps_per_second": 0.657,
|
| 185 |
+
"step": 350
|
| 186 |
+
},
|
| 187 |
+
{
|
| 188 |
+
"epoch": 0.024266936299292215,
|
| 189 |
+
"grad_norm": 1.2421875,
|
| 190 |
+
"learning_rate": 0.0005026,
|
| 191 |
+
"loss": 4.243848419189453,
|
| 192 |
+
"step": 360
|
| 193 |
+
},
|
| 194 |
+
{
|
| 195 |
+
"epoch": 0.02561509942703067,
|
| 196 |
+
"grad_norm": 0.7109375,
|
| 197 |
+
"learning_rate": 0.0005306,
|
| 198 |
+
"loss": 4.1090648651123045,
|
| 199 |
+
"step": 380
|
| 200 |
+
},
|
| 201 |
+
{
|
| 202 |
+
"epoch": 0.026963262554769128,
|
| 203 |
+
"grad_norm": 0.484375,
|
| 204 |
+
"learning_rate": 0.0005586,
|
| 205 |
+
"loss": 3.966172790527344,
|
| 206 |
+
"step": 400
|
| 207 |
+
},
|
| 208 |
+
{
|
| 209 |
+
"epoch": 0.026963262554769128,
|
| 210 |
+
"eval_loss": 3.923773765563965,
|
| 211 |
+
"eval_runtime": 15.5308,
|
| 212 |
+
"eval_samples_per_second": 613.426,
|
| 213 |
+
"eval_steps_per_second": 0.644,
|
| 214 |
+
"step": 400
|
| 215 |
+
},
|
| 216 |
+
{
|
| 217 |
+
"epoch": 0.028311425682507583,
|
| 218 |
+
"grad_norm": 0.6640625,
|
| 219 |
+
"learning_rate": 0.0005866,
|
| 220 |
+
"loss": 3.8559597015380858,
|
| 221 |
+
"step": 420
|
| 222 |
+
},
|
| 223 |
+
{
|
| 224 |
+
"epoch": 0.02965958881024604,
|
| 225 |
+
"grad_norm": 0.515625,
|
| 226 |
+
"learning_rate": 0.0006146,
|
| 227 |
+
"loss": 3.7943866729736326,
|
| 228 |
+
"step": 440
|
| 229 |
+
},
|
| 230 |
+
{
|
| 231 |
+
"epoch": 0.030333670374115267,
|
| 232 |
+
"eval_loss": 3.6977553367614746,
|
| 233 |
+
"eval_runtime": 15.2001,
|
| 234 |
+
"eval_samples_per_second": 626.77,
|
| 235 |
+
"eval_steps_per_second": 0.658,
|
| 236 |
+
"step": 450
|
| 237 |
+
},
|
| 238 |
+
{
|
| 239 |
+
"epoch": 0.031007751937984496,
|
| 240 |
+
"grad_norm": 0.51953125,
|
| 241 |
+
"learning_rate": 0.0006426,
|
| 242 |
+
"loss": 3.686443328857422,
|
| 243 |
+
"step": 460
|
| 244 |
+
},
|
| 245 |
+
{
|
| 246 |
+
"epoch": 0.032355915065722954,
|
| 247 |
+
"grad_norm": 0.62109375,
|
| 248 |
+
"learning_rate": 0.0006705999999999999,
|
| 249 |
+
"loss": 3.6123157501220704,
|
| 250 |
+
"step": 480
|
| 251 |
+
},
|
| 252 |
+
{
|
| 253 |
+
"epoch": 0.03370407819346141,
|
| 254 |
+
"grad_norm": 0.70703125,
|
| 255 |
+
"learning_rate": 0.0006986,
|
| 256 |
+
"loss": 3.542654800415039,
|
| 257 |
+
"step": 500
|
| 258 |
+
},
|
| 259 |
+
{
|
| 260 |
+
"epoch": 0.03370407819346141,
|
| 261 |
+
"eval_loss": 3.513820171356201,
|
| 262 |
+
"eval_runtime": 15.2193,
|
| 263 |
+
"eval_samples_per_second": 625.983,
|
| 264 |
+
"eval_steps_per_second": 0.657,
|
| 265 |
+
"step": 500
|
| 266 |
+
},
|
| 267 |
+
{
|
| 268 |
+
"epoch": 0.03505224132119986,
|
| 269 |
+
"grad_norm": 0.48046875,
|
| 270 |
+
"learning_rate": 0.0007,
|
| 271 |
+
"loss": 3.472199249267578,
|
| 272 |
+
"step": 520
|
| 273 |
+
},
|
| 274 |
+
{
|
| 275 |
+
"epoch": 0.03640040444893832,
|
| 276 |
+
"grad_norm": 0.53515625,
|
| 277 |
+
"learning_rate": 0.0007,
|
| 278 |
+
"loss": 3.396291732788086,
|
| 279 |
+
"step": 540
|
| 280 |
+
},
|
| 281 |
+
{
|
| 282 |
+
"epoch": 0.03707448601280755,
|
| 283 |
+
"eval_loss": 3.3581483364105225,
|
| 284 |
+
"eval_runtime": 15.3044,
|
| 285 |
+
"eval_samples_per_second": 622.502,
|
| 286 |
+
"eval_steps_per_second": 0.653,
|
| 287 |
+
"step": 550
|
| 288 |
+
},
|
| 289 |
+
{
|
| 290 |
+
"epoch": 0.03774856757667678,
|
| 291 |
+
"grad_norm": 0.47265625,
|
| 292 |
+
"learning_rate": 0.0007,
|
| 293 |
+
"loss": 3.347300720214844,
|
| 294 |
+
"step": 560
|
| 295 |
+
},
|
| 296 |
+
{
|
| 297 |
+
"epoch": 0.03909673070441524,
|
| 298 |
+
"grad_norm": 0.458984375,
|
| 299 |
+
"learning_rate": 0.0007,
|
| 300 |
+
"loss": 3.2885902404785154,
|
| 301 |
+
"step": 580
|
| 302 |
+
},
|
| 303 |
+
{
|
| 304 |
+
"epoch": 0.04044489383215369,
|
| 305 |
+
"grad_norm": 0.4375,
|
| 306 |
+
"learning_rate": 0.0007,
|
| 307 |
+
"loss": 3.25723876953125,
|
| 308 |
+
"step": 600
|
| 309 |
+
},
|
| 310 |
+
{
|
| 311 |
+
"epoch": 0.04044489383215369,
|
| 312 |
+
"eval_loss": 3.2229573726654053,
|
| 313 |
+
"eval_runtime": 15.2967,
|
| 314 |
+
"eval_samples_per_second": 622.812,
|
| 315 |
+
"eval_steps_per_second": 0.654,
|
| 316 |
+
"step": 600
|
| 317 |
+
},
|
| 318 |
+
{
|
| 319 |
+
"epoch": 0.04179305695989215,
|
| 320 |
+
"grad_norm": 0.458984375,
|
| 321 |
+
"learning_rate": 0.0007,
|
| 322 |
+
"loss": 3.1902334213256838,
|
| 323 |
+
"step": 620
|
| 324 |
+
},
|
| 325 |
+
{
|
| 326 |
+
"epoch": 0.043141220087630605,
|
| 327 |
+
"grad_norm": 0.40625,
|
| 328 |
+
"learning_rate": 0.0007,
|
| 329 |
+
"loss": 3.167861557006836,
|
| 330 |
+
"step": 640
|
| 331 |
+
},
|
| 332 |
+
{
|
| 333 |
+
"epoch": 0.043815301651499834,
|
| 334 |
+
"eval_loss": 3.120636463165283,
|
| 335 |
+
"eval_runtime": 15.3751,
|
| 336 |
+
"eval_samples_per_second": 619.639,
|
| 337 |
+
"eval_steps_per_second": 0.65,
|
| 338 |
+
"step": 650
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 0.044489383215369056,
|
| 342 |
+
"grad_norm": 0.494140625,
|
| 343 |
+
"learning_rate": 0.0007,
|
| 344 |
+
"loss": 3.1089324951171875,
|
| 345 |
+
"step": 660
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 0.045837546343107514,
|
| 349 |
+
"grad_norm": 0.435546875,
|
| 350 |
+
"learning_rate": 0.0007,
|
| 351 |
+
"loss": 3.0828268051147463,
|
| 352 |
+
"step": 680
|
| 353 |
+
},
|
| 354 |
+
{
|
| 355 |
+
"epoch": 0.04718570947084597,
|
| 356 |
+
"grad_norm": 0.427734375,
|
| 357 |
+
"learning_rate": 0.0007,
|
| 358 |
+
"loss": 3.0396488189697264,
|
| 359 |
+
"step": 700
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"epoch": 0.04718570947084597,
|
| 363 |
+
"eval_loss": 3.0276408195495605,
|
| 364 |
+
"eval_runtime": 15.2735,
|
| 365 |
+
"eval_samples_per_second": 623.761,
|
| 366 |
+
"eval_steps_per_second": 0.655,
|
| 367 |
+
"step": 700
|
| 368 |
+
},
|
| 369 |
+
{
|
| 370 |
+
"epoch": 0.04853387259858443,
|
| 371 |
+
"grad_norm": 0.4375,
|
| 372 |
+
"learning_rate": 0.0007,
|
| 373 |
+
"loss": 3.0034805297851563,
|
| 374 |
+
"step": 720
|
| 375 |
+
},
|
| 376 |
+
{
|
| 377 |
+
"epoch": 0.04988203572632288,
|
| 378 |
+
"grad_norm": 0.52734375,
|
| 379 |
+
"learning_rate": 0.0007,
|
| 380 |
+
"loss": 2.96556396484375,
|
| 381 |
+
"step": 740
|
| 382 |
+
},
|
| 383 |
+
{
|
| 384 |
+
"epoch": 0.05055611729019211,
|
| 385 |
+
"eval_loss": 2.950009346008301,
|
| 386 |
+
"eval_runtime": 15.243,
|
| 387 |
+
"eval_samples_per_second": 625.009,
|
| 388 |
+
"eval_steps_per_second": 0.656,
|
| 389 |
+
"step": 750
|
| 390 |
+
},
|
| 391 |
+
{
|
| 392 |
+
"epoch": 0.05123019885406134,
|
| 393 |
+
"grad_norm": 0.451171875,
|
| 394 |
+
"learning_rate": 0.0007,
|
| 395 |
+
"loss": 2.950779151916504,
|
| 396 |
+
"step": 760
|
| 397 |
+
},
|
| 398 |
+
{
|
| 399 |
+
"epoch": 0.0525783619817998,
|
| 400 |
+
"grad_norm": 0.400390625,
|
| 401 |
+
"learning_rate": 0.0007,
|
| 402 |
+
"loss": 2.956203079223633,
|
| 403 |
+
"step": 780
|
| 404 |
+
},
|
| 405 |
+
{
|
| 406 |
+
"epoch": 0.053926525109538256,
|
| 407 |
+
"grad_norm": 0.40625,
|
| 408 |
+
"learning_rate": 0.0007,
|
| 409 |
+
"loss": 2.9017578125,
|
| 410 |
+
"step": 800
|
| 411 |
+
},
|
| 412 |
+
{
|
| 413 |
+
"epoch": 0.053926525109538256,
|
| 414 |
+
"eval_loss": 2.8844006061553955,
|
| 415 |
+
"eval_runtime": 15.2903,
|
| 416 |
+
"eval_samples_per_second": 623.073,
|
| 417 |
+
"eval_steps_per_second": 0.654,
|
| 418 |
+
"step": 800
|
| 419 |
+
},
|
| 420 |
+
{
|
| 421 |
+
"epoch": 0.05527468823727671,
|
| 422 |
+
"grad_norm": 0.41015625,
|
| 423 |
+
"learning_rate": 0.0007,
|
| 424 |
+
"loss": 2.887515640258789,
|
| 425 |
+
"step": 820
|
| 426 |
+
},
|
| 427 |
+
{
|
| 428 |
+
"epoch": 0.056622851365015166,
|
| 429 |
+
"grad_norm": 0.41796875,
|
| 430 |
+
"learning_rate": 0.0007,
|
| 431 |
+
"loss": 2.8556928634643555,
|
| 432 |
+
"step": 840
|
| 433 |
+
},
|
| 434 |
+
{
|
| 435 |
+
"epoch": 0.057296932928884395,
|
| 436 |
+
"eval_loss": 2.821610450744629,
|
| 437 |
+
"eval_runtime": 15.2253,
|
| 438 |
+
"eval_samples_per_second": 625.733,
|
| 439 |
+
"eval_steps_per_second": 0.657,
|
| 440 |
+
"step": 850
|
| 441 |
+
},
|
| 442 |
+
{
|
| 443 |
+
"epoch": 0.057971014492753624,
|
| 444 |
+
"grad_norm": 0.423828125,
|
| 445 |
+
"learning_rate": 0.0007,
|
| 446 |
+
"loss": 2.8130855560302734,
|
| 447 |
+
"step": 860
|
| 448 |
+
},
|
| 449 |
+
{
|
| 450 |
+
"epoch": 0.05931917762049208,
|
| 451 |
+
"grad_norm": 0.38671875,
|
| 452 |
+
"learning_rate": 0.0007,
|
| 453 |
+
"loss": 2.790329170227051,
|
| 454 |
+
"step": 880
|
| 455 |
+
},
|
| 456 |
+
{
|
| 457 |
+
"epoch": 0.06066734074823053,
|
| 458 |
+
"grad_norm": 0.427734375,
|
| 459 |
+
"learning_rate": 0.0007,
|
| 460 |
+
"loss": 2.7760162353515625,
|
| 461 |
+
"step": 900
|
| 462 |
+
},
|
| 463 |
+
{
|
| 464 |
+
"epoch": 0.06066734074823053,
|
| 465 |
+
"eval_loss": 2.7661373615264893,
|
| 466 |
+
"eval_runtime": 15.2736,
|
| 467 |
+
"eval_samples_per_second": 623.755,
|
| 468 |
+
"eval_steps_per_second": 0.655,
|
| 469 |
+
"step": 900
|
| 470 |
+
},
|
| 471 |
+
{
|
| 472 |
+
"epoch": 0.06201550387596899,
|
| 473 |
+
"grad_norm": 0.435546875,
|
| 474 |
+
"learning_rate": 0.0007,
|
| 475 |
+
"loss": 2.7462047576904296,
|
| 476 |
+
"step": 920
|
| 477 |
+
},
|
| 478 |
+
{
|
| 479 |
+
"epoch": 0.06336366700370745,
|
| 480 |
+
"grad_norm": 0.4140625,
|
| 481 |
+
"learning_rate": 0.0007,
|
| 482 |
+
"loss": 2.732274627685547,
|
| 483 |
+
"step": 940
|
| 484 |
+
},
|
| 485 |
+
{
|
| 486 |
+
"epoch": 0.06403774856757667,
|
| 487 |
+
"eval_loss": 2.713942289352417,
|
| 488 |
+
"eval_runtime": 15.6062,
|
| 489 |
+
"eval_samples_per_second": 610.463,
|
| 490 |
+
"eval_steps_per_second": 0.641,
|
| 491 |
+
"step": 950
|
| 492 |
+
},
|
| 493 |
+
{
|
| 494 |
+
"epoch": 0.06471183013144591,
|
| 495 |
+
"grad_norm": 0.419921875,
|
| 496 |
+
"learning_rate": 0.0007,
|
| 497 |
+
"loss": 2.705318069458008,
|
| 498 |
+
"step": 960
|
| 499 |
+
},
|
| 500 |
+
{
|
| 501 |
+
"epoch": 0.06605999325918437,
|
| 502 |
+
"grad_norm": 0.4296875,
|
| 503 |
+
"learning_rate": 0.0007,
|
| 504 |
+
"loss": 2.6892818450927733,
|
| 505 |
+
"step": 980
|
| 506 |
+
},
|
| 507 |
+
{
|
| 508 |
+
"epoch": 0.06740815638692282,
|
| 509 |
+
"grad_norm": 0.4453125,
|
| 510 |
+
"learning_rate": 0.0007,
|
| 511 |
+
"loss": 2.687115287780762,
|
| 512 |
+
"step": 1000
|
| 513 |
+
},
|
| 514 |
+
{
|
| 515 |
+
"epoch": 0.06740815638692282,
|
| 516 |
+
"eval_loss": 2.668647289276123,
|
| 517 |
+
"eval_runtime": 15.2597,
|
| 518 |
+
"eval_samples_per_second": 624.326,
|
| 519 |
+
"eval_steps_per_second": 0.655,
|
| 520 |
+
"step": 1000
|
| 521 |
+
},
|
| 522 |
+
{
|
| 523 |
+
"epoch": 0.06875631951466127,
|
| 524 |
+
"grad_norm": 0.388671875,
|
| 525 |
+
"learning_rate": 0.0007,
|
| 526 |
+
"loss": 2.6533071517944338,
|
| 527 |
+
"step": 1020
|
| 528 |
+
},
|
| 529 |
+
{
|
| 530 |
+
"epoch": 0.07010448264239973,
|
| 531 |
+
"grad_norm": 0.392578125,
|
| 532 |
+
"learning_rate": 0.0007,
|
| 533 |
+
"loss": 2.6390922546386717,
|
| 534 |
+
"step": 1040
|
| 535 |
+
},
|
| 536 |
+
{
|
| 537 |
+
"epoch": 0.07077856420626896,
|
| 538 |
+
"eval_loss": 2.6291816234588623,
|
| 539 |
+
"eval_runtime": 15.2457,
|
| 540 |
+
"eval_samples_per_second": 624.896,
|
| 541 |
+
"eval_steps_per_second": 0.656,
|
| 542 |
+
"step": 1050
|
| 543 |
+
},
|
| 544 |
+
{
|
| 545 |
+
"epoch": 0.07145264577013818,
|
| 546 |
+
"grad_norm": 0.4375,
|
| 547 |
+
"learning_rate": 0.0007,
|
| 548 |
+
"loss": 2.624461555480957,
|
| 549 |
+
"step": 1060
|
| 550 |
+
},
|
| 551 |
+
{
|
| 552 |
+
"epoch": 0.07280080889787664,
|
| 553 |
+
"grad_norm": 0.396484375,
|
| 554 |
+
"learning_rate": 0.0007,
|
| 555 |
+
"loss": 2.6044427871704103,
|
| 556 |
+
"step": 1080
|
| 557 |
+
},
|
| 558 |
+
{
|
| 559 |
+
"epoch": 0.0741489720256151,
|
| 560 |
+
"grad_norm": 0.423828125,
|
| 561 |
+
"learning_rate": 0.0007,
|
| 562 |
+
"loss": 2.597646713256836,
|
| 563 |
+
"step": 1100
|
| 564 |
+
},
|
| 565 |
+
{
|
| 566 |
+
"epoch": 0.0741489720256151,
|
| 567 |
+
"eval_loss": 2.589696168899536,
|
| 568 |
+
"eval_runtime": 15.3967,
|
| 569 |
+
"eval_samples_per_second": 618.768,
|
| 570 |
+
"eval_steps_per_second": 0.649,
|
| 571 |
+
"step": 1100
|
| 572 |
+
},
|
| 573 |
+
{
|
| 574 |
+
"epoch": 0.07549713515335356,
|
| 575 |
+
"grad_norm": 0.369140625,
|
| 576 |
+
"learning_rate": 0.0007,
|
| 577 |
+
"loss": 2.561947250366211,
|
| 578 |
+
"step": 1120
|
| 579 |
+
},
|
| 580 |
+
{
|
| 581 |
+
"epoch": 0.07684529828109202,
|
| 582 |
+
"grad_norm": 0.390625,
|
| 583 |
+
"learning_rate": 0.0007,
|
| 584 |
+
"loss": 2.54862117767334,
|
| 585 |
+
"step": 1140
|
| 586 |
+
},
|
| 587 |
+
{
|
| 588 |
+
"epoch": 0.07751937984496124,
|
| 589 |
+
"eval_loss": 2.559185743331909,
|
| 590 |
+
"eval_runtime": 15.2877,
|
| 591 |
+
"eval_samples_per_second": 623.181,
|
| 592 |
+
"eval_steps_per_second": 0.654,
|
| 593 |
+
"step": 1150
|
| 594 |
+
},
|
| 595 |
+
{
|
| 596 |
+
"epoch": 0.07819346140883048,
|
| 597 |
+
"grad_norm": 0.37890625,
|
| 598 |
+
"learning_rate": 0.0007,
|
| 599 |
+
"loss": 2.5469717025756835,
|
| 600 |
+
"step": 1160
|
| 601 |
+
},
|
| 602 |
+
{
|
| 603 |
+
"epoch": 0.07954162453656892,
|
| 604 |
+
"grad_norm": 0.400390625,
|
| 605 |
+
"learning_rate": 0.0007,
|
| 606 |
+
"loss": 2.540451240539551,
|
| 607 |
+
"step": 1180
|
| 608 |
+
},
|
| 609 |
+
{
|
| 610 |
+
"epoch": 0.08088978766430738,
|
| 611 |
+
"grad_norm": 0.40625,
|
| 612 |
+
"learning_rate": 0.0007,
|
| 613 |
+
"loss": 2.53134708404541,
|
| 614 |
+
"step": 1200
|
| 615 |
+
},
|
| 616 |
+
{
|
| 617 |
+
"epoch": 0.08088978766430738,
|
| 618 |
+
"eval_loss": 2.521665096282959,
|
| 619 |
+
"eval_runtime": 15.2822,
|
| 620 |
+
"eval_samples_per_second": 623.407,
|
| 621 |
+
"eval_steps_per_second": 0.654,
|
| 622 |
+
"step": 1200
|
| 623 |
+
},
|
| 624 |
+
{
|
| 625 |
+
"epoch": 0.08223795079204584,
|
| 626 |
+
"grad_norm": 0.37109375,
|
| 627 |
+
"learning_rate": 0.0007,
|
| 628 |
+
"loss": 2.513539123535156,
|
| 629 |
+
"step": 1220
|
| 630 |
+
},
|
| 631 |
+
{
|
| 632 |
+
"epoch": 0.0835861139197843,
|
| 633 |
+
"grad_norm": 0.36328125,
|
| 634 |
+
"learning_rate": 0.0007,
|
| 635 |
+
"loss": 2.5009738922119142,
|
| 636 |
+
"step": 1240
|
| 637 |
+
},
|
| 638 |
+
{
|
| 639 |
+
"epoch": 0.08426019548365352,
|
| 640 |
+
"eval_loss": 2.489459753036499,
|
| 641 |
+
"eval_runtime": 15.2493,
|
| 642 |
+
"eval_samples_per_second": 624.749,
|
| 643 |
+
"eval_steps_per_second": 0.656,
|
| 644 |
+
"step": 1250
|
| 645 |
+
},
|
| 646 |
+
{
|
| 647 |
+
"epoch": 0.08493427704752275,
|
| 648 |
+
"grad_norm": 0.388671875,
|
| 649 |
+
"learning_rate": 0.0007,
|
| 650 |
+
"loss": 2.4894077301025392,
|
| 651 |
+
"step": 1260
|
| 652 |
+
},
|
| 653 |
+
{
|
| 654 |
+
"epoch": 0.08628244017526121,
|
| 655 |
+
"grad_norm": 0.369140625,
|
| 656 |
+
"learning_rate": 0.0007,
|
| 657 |
+
"loss": 2.4697439193725588,
|
| 658 |
+
"step": 1280
|
| 659 |
+
},
|
| 660 |
+
{
|
| 661 |
+
"epoch": 0.08763060330299967,
|
| 662 |
+
"grad_norm": 0.423828125,
|
| 663 |
+
"learning_rate": 0.0007,
|
| 664 |
+
"loss": 2.461294174194336,
|
| 665 |
+
"step": 1300
|
| 666 |
+
},
|
| 667 |
+
{
|
| 668 |
+
"epoch": 0.08763060330299967,
|
| 669 |
+
"eval_loss": 2.4627816677093506,
|
| 670 |
+
"eval_runtime": 15.421,
|
| 671 |
+
"eval_samples_per_second": 617.795,
|
| 672 |
+
"eval_steps_per_second": 0.648,
|
| 673 |
+
"step": 1300
|
| 674 |
+
},
|
| 675 |
+
{
|
| 676 |
+
"epoch": 0.08897876643073811,
|
| 677 |
+
"grad_norm": 0.36328125,
|
| 678 |
+
"learning_rate": 0.0007,
|
| 679 |
+
"loss": 2.452680206298828,
|
| 680 |
+
"step": 1320
|
| 681 |
+
},
|
| 682 |
+
{
|
| 683 |
+
"epoch": 0.09032692955847657,
|
| 684 |
+
"grad_norm": 0.404296875,
|
| 685 |
+
"learning_rate": 0.0007,
|
| 686 |
+
"loss": 2.4396684646606444,
|
| 687 |
+
"step": 1340
|
| 688 |
+
},
|
| 689 |
+
{
|
| 690 |
+
"epoch": 0.0910010111223458,
|
| 691 |
+
"eval_loss": 2.437220335006714,
|
| 692 |
+
"eval_runtime": 15.3026,
|
| 693 |
+
"eval_samples_per_second": 622.576,
|
| 694 |
+
"eval_steps_per_second": 0.653,
|
| 695 |
+
"step": 1350
|
| 696 |
+
},
|
| 697 |
+
{
|
| 698 |
+
"epoch": 0.09167509268621503,
|
| 699 |
+
"grad_norm": 0.412109375,
|
| 700 |
+
"learning_rate": 0.0007,
|
| 701 |
+
"loss": 2.4397760391235352,
|
| 702 |
+
"step": 1360
|
| 703 |
+
},
|
| 704 |
+
{
|
| 705 |
+
"epoch": 0.09302325581395349,
|
| 706 |
+
"grad_norm": 0.353515625,
|
| 707 |
+
"learning_rate": 0.0007,
|
| 708 |
+
"loss": 2.4158536911010744,
|
| 709 |
+
"step": 1380
|
| 710 |
+
},
|
| 711 |
+
{
|
| 712 |
+
"epoch": 0.09437141894169195,
|
| 713 |
+
"grad_norm": 0.345703125,
|
| 714 |
+
"learning_rate": 0.0007,
|
| 715 |
+
"loss": 2.406942367553711,
|
| 716 |
+
"step": 1400
|
| 717 |
+
},
|
| 718 |
+
{
|
| 719 |
+
"epoch": 0.09437141894169195,
|
| 720 |
+
"eval_loss": 2.4096601009368896,
|
| 721 |
+
"eval_runtime": 15.357,
|
| 722 |
+
"eval_samples_per_second": 620.368,
|
| 723 |
+
"eval_steps_per_second": 0.651,
|
| 724 |
+
"step": 1400
|
| 725 |
+
},
|
| 726 |
+
{
|
| 727 |
+
"epoch": 0.0957195820694304,
|
| 728 |
+
"grad_norm": 0.36328125,
|
| 729 |
+
"learning_rate": 0.0007,
|
| 730 |
+
"loss": 2.4161144256591798,
|
| 731 |
+
"step": 1420
|
| 732 |
+
},
|
| 733 |
+
{
|
| 734 |
+
"epoch": 0.09706774519716886,
|
| 735 |
+
"grad_norm": 0.36328125,
|
| 736 |
+
"learning_rate": 0.0007,
|
| 737 |
+
"loss": 2.392984390258789,
|
| 738 |
+
"step": 1440
|
| 739 |
+
},
|
| 740 |
+
{
|
| 741 |
+
"epoch": 0.09774182676103808,
|
| 742 |
+
"eval_loss": 2.3937113285064697,
|
| 743 |
+
"eval_runtime": 15.3555,
|
| 744 |
+
"eval_samples_per_second": 620.43,
|
| 745 |
+
"eval_steps_per_second": 0.651,
|
| 746 |
+
"step": 1450
|
| 747 |
+
},
|
| 748 |
+
{
|
| 749 |
+
"epoch": 0.09841590832490732,
|
| 750 |
+
"grad_norm": 0.384765625,
|
| 751 |
+
"learning_rate": 0.0007,
|
| 752 |
+
"loss": 2.3820331573486326,
|
| 753 |
+
"step": 1460
|
| 754 |
+
},
|
| 755 |
+
{
|
| 756 |
+
"epoch": 0.09976407145264576,
|
| 757 |
+
"grad_norm": 0.33984375,
|
| 758 |
+
"learning_rate": 0.0007,
|
| 759 |
+
"loss": 2.3687986373901366,
|
| 760 |
+
"step": 1480
|
| 761 |
+
},
|
| 762 |
+
{
|
| 763 |
+
"epoch": 0.10111223458038422,
|
| 764 |
+
"grad_norm": 0.38671875,
|
| 765 |
+
"learning_rate": 0.0007,
|
| 766 |
+
"loss": 2.355032730102539,
|
| 767 |
+
"step": 1500
|
| 768 |
+
},
|
| 769 |
+
{
|
| 770 |
+
"epoch": 0.10111223458038422,
|
| 771 |
+
"eval_loss": 2.3681657314300537,
|
| 772 |
+
"eval_runtime": 15.3336,
|
| 773 |
+
"eval_samples_per_second": 621.317,
|
| 774 |
+
"eval_steps_per_second": 0.652,
|
| 775 |
+
"step": 1500
|
| 776 |
+
},
|
| 777 |
+
{
|
| 778 |
+
"epoch": 0.10246039770812268,
|
| 779 |
+
"grad_norm": 0.392578125,
|
| 780 |
+
"learning_rate": 0.0007,
|
| 781 |
+
"loss": 2.3572179794311525,
|
| 782 |
+
"step": 1520
|
| 783 |
+
},
|
| 784 |
+
{
|
| 785 |
+
"epoch": 0.10380856083586114,
|
| 786 |
+
"grad_norm": 0.35546875,
|
| 787 |
+
"learning_rate": 0.0007,
|
| 788 |
+
"loss": 2.345840835571289,
|
| 789 |
+
"step": 1540
|
| 790 |
+
},
|
| 791 |
+
{
|
| 792 |
+
"epoch": 0.10448264239973036,
|
| 793 |
+
"eval_loss": 2.3475193977355957,
|
| 794 |
+
"eval_runtime": 15.3453,
|
| 795 |
+
"eval_samples_per_second": 620.842,
|
| 796 |
+
"eval_steps_per_second": 0.652,
|
| 797 |
+
"step": 1550
|
| 798 |
+
},
|
| 799 |
+
{
|
| 800 |
+
"epoch": 0.1051567239635996,
|
| 801 |
+
"grad_norm": 0.388671875,
|
| 802 |
+
"learning_rate": 0.0007,
|
| 803 |
+
"loss": 2.335231971740723,
|
| 804 |
+
"step": 1560
|
| 805 |
+
},
|
| 806 |
+
{
|
| 807 |
+
"epoch": 0.10650488709133805,
|
| 808 |
+
"grad_norm": 0.359375,
|
| 809 |
+
"learning_rate": 0.0007,
|
| 810 |
+
"loss": 2.3543283462524416,
|
| 811 |
+
"step": 1580
|
| 812 |
+
},
|
| 813 |
+
{
|
| 814 |
+
"epoch": 0.10785305021907651,
|
| 815 |
+
"grad_norm": 0.39453125,
|
| 816 |
+
"learning_rate": 0.0007,
|
| 817 |
+
"loss": 2.3237049102783205,
|
| 818 |
+
"step": 1600
|
| 819 |
+
},
|
| 820 |
+
{
|
| 821 |
+
"epoch": 0.10785305021907651,
|
| 822 |
+
"eval_loss": 2.3377676010131836,
|
| 823 |
+
"eval_runtime": 15.2858,
|
| 824 |
+
"eval_samples_per_second": 623.259,
|
| 825 |
+
"eval_steps_per_second": 0.654,
|
| 826 |
+
"step": 1600
|
| 827 |
+
},
|
| 828 |
+
{
|
| 829 |
+
"epoch": 0.10920121334681497,
|
| 830 |
+
"grad_norm": 0.357421875,
|
| 831 |
+
"learning_rate": 0.0007,
|
| 832 |
+
"loss": 2.3196277618408203,
|
| 833 |
+
"step": 1620
|
| 834 |
+
},
|
| 835 |
+
{
|
| 836 |
+
"epoch": 0.11054937647455342,
|
| 837 |
+
"grad_norm": 0.341796875,
|
| 838 |
+
"learning_rate": 0.0007,
|
| 839 |
+
"loss": 2.3308162689208984,
|
| 840 |
+
"step": 1640
|
| 841 |
+
},
|
| 842 |
+
{
|
| 843 |
+
"epoch": 0.11122345803842265,
|
| 844 |
+
"eval_loss": 2.313627004623413,
|
| 845 |
+
"eval_runtime": 15.276,
|
| 846 |
+
"eval_samples_per_second": 623.659,
|
| 847 |
+
"eval_steps_per_second": 0.655,
|
| 848 |
+
"step": 1650
|
| 849 |
+
},
|
| 850 |
+
{
|
| 851 |
+
"epoch": 0.11189753960229187,
|
| 852 |
+
"grad_norm": 0.353515625,
|
| 853 |
+
"learning_rate": 0.0007,
|
| 854 |
+
"loss": 2.3104904174804686,
|
| 855 |
+
"step": 1660
|
| 856 |
+
},
|
| 857 |
+
{
|
| 858 |
+
"epoch": 0.11324570273003033,
|
| 859 |
+
"grad_norm": 0.353515625,
|
| 860 |
+
"learning_rate": 0.0007,
|
| 861 |
+
"loss": 2.2955816268920897,
|
| 862 |
+
"step": 1680
|
| 863 |
+
},
|
| 864 |
+
{
|
| 865 |
+
"epoch": 0.11459386585776879,
|
| 866 |
+
"grad_norm": 0.345703125,
|
| 867 |
+
"learning_rate": 0.0007,
|
| 868 |
+
"loss": 2.3044401168823243,
|
| 869 |
+
"step": 1700
|
| 870 |
+
},
|
| 871 |
+
{
|
| 872 |
+
"epoch": 0.11459386585776879,
|
| 873 |
+
"eval_loss": 2.2976956367492676,
|
| 874 |
+
"eval_runtime": 15.3404,
|
| 875 |
+
"eval_samples_per_second": 621.039,
|
| 876 |
+
"eval_steps_per_second": 0.652,
|
| 877 |
+
"step": 1700
|
| 878 |
+
}
|
| 879 |
+
],
|
| 880 |
+
"logging_steps": 20,
|
| 881 |
+
"max_steps": 2500,
|
| 882 |
+
"num_input_tokens_seen": 0,
|
| 883 |
+
"num_train_epochs": 1,
|
| 884 |
+
"save_steps": 100,
|
| 885 |
+
"stateful_callbacks": {
|
| 886 |
+
"TrainerControl": {
|
| 887 |
+
"args": {
|
| 888 |
+
"should_epoch_stop": false,
|
| 889 |
+
"should_evaluate": false,
|
| 890 |
+
"should_log": false,
|
| 891 |
+
"should_save": true,
|
| 892 |
+
"should_training_stop": false
|
| 893 |
+
},
|
| 894 |
+
"attributes": {}
|
| 895 |
+
}
|
| 896 |
+
},
|
| 897 |
+
"total_flos": 5484682464460800.0,
|
| 898 |
+
"train_batch_size": 64,
|
| 899 |
+
"trial_name": null,
|
| 900 |
+
"trial_params": null
|
| 901 |
+
}
|
zain/Activation/out/glu-silu-100L_run/checkpoint-1700/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:fd90512e56f813612dc668676d6f8cab72da110a42f23c61cafebf91064625a1
|
| 3 |
+
size 4920
|
zain/Activation/out/glu-silu-100L_run/checkpoint-1800/config.json
ADDED
|
@@ -0,0 +1,36 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"activation": "silu",
|
| 3 |
+
"architectures": [
|
| 4 |
+
"TinyLlamaForCausalLM"
|
| 5 |
+
],
|
| 6 |
+
"attention_bias": false,
|
| 7 |
+
"attention_dropout": 0.0,
|
| 8 |
+
"bos_token_id": 1,
|
| 9 |
+
"dtype": "bfloat16",
|
| 10 |
+
"eos_token_id": 2,
|
| 11 |
+
"head_dim": 32,
|
| 12 |
+
"hidden_act": "silu",
|
| 13 |
+
"hidden_size": 128,
|
| 14 |
+
"initializer_range": 0.02,
|
| 15 |
+
"intermediate_size": 256,
|
| 16 |
+
"max_position_embeddings": 512,
|
| 17 |
+
"mlp_bias": false,
|
| 18 |
+
"mlp_type": "glu",
|
| 19 |
+
"model_type": "tiny_llama",
|
| 20 |
+
"num_attention_heads": 4,
|
| 21 |
+
"num_hidden_layers": 100,
|
| 22 |
+
"num_key_value_heads": 4,
|
| 23 |
+
"pad_token_id": 0,
|
| 24 |
+
"pretraining_tp": 1,
|
| 25 |
+
"rms_norm_eps": 1e-06,
|
| 26 |
+
"rope_parameters": {
|
| 27 |
+
"rope_theta": 10000.0,
|
| 28 |
+
"rope_type": "default"
|
| 29 |
+
},
|
| 30 |
+
"tie_word_embeddings": true,
|
| 31 |
+
"tokenizer_name": "w-ahmad/tiny-stories-tokenizer",
|
| 32 |
+
"transformers_version": "5.16.0.dev0",
|
| 33 |
+
"use_cache": false,
|
| 34 |
+
"vocab_size": 4096,
|
| 35 |
+
"waleed_beta": 10.0
|
| 36 |
+
}
|
zain/Activation/out/glu-silu-100L_run/checkpoint-1800/model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:14c31a2908d700d4d078005075309da1bde4eeac083c88ae3cce77a15ef8fa86
|
| 3 |
+
size 33967272
|
zain/Activation/out/glu-silu-100L_run/checkpoint-1800/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:42c0d1ed4da56a36dbf36e383739198fc1f277550627fb8181469ec825183faa
|
| 3 |
+
size 68504996
|
zain/Activation/out/glu-silu-100L_run/checkpoint-1800/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:a22625c60bf14250424f99a809ffcecb5f25c776dd8079c8624dc295555b32c0
|
| 3 |
+
size 14244
|
zain/Activation/out/glu-silu-100L_run/checkpoint-1800/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:8214bf4c471bc465884f1c219edb27abec42cbdf74a9a2fe9320000a4ce60c16
|
| 3 |
+
size 1064
|
zain/Activation/out/glu-silu-100L_run/checkpoint-1800/tokenizer.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
zain/Activation/out/glu-silu-100L_run/checkpoint-1800/tokenizer_config.json
ADDED
|
@@ -0,0 +1,13 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"add_prefix_space": false,
|
| 3 |
+
"backend": "tokenizers",
|
| 4 |
+
"bos_token": "<|endoftext|>",
|
| 5 |
+
"eos_token": "<|endoftext|>",
|
| 6 |
+
"errors": "replace",
|
| 7 |
+
"is_local": false,
|
| 8 |
+
"local_files_only": false,
|
| 9 |
+
"model_max_length": 1024,
|
| 10 |
+
"pad_token": "<|endoftext|>",
|
| 11 |
+
"tokenizer_class": "GPT2Tokenizer",
|
| 12 |
+
"unk_token": "<|endoftext|>"
|
| 13 |
+
}
|
zain/Activation/out/glu-silu-100L_run/checkpoint-1800/trainer_state.json
ADDED
|
@@ -0,0 +1,952 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_global_step": null,
|
| 3 |
+
"best_metric": null,
|
| 4 |
+
"best_model_checkpoint": null,
|
| 5 |
+
"epoch": 0.12133468149646107,
|
| 6 |
+
"eval_steps": 50,
|
| 7 |
+
"global_step": 1800,
|
| 8 |
+
"is_hyper_param_search": false,
|
| 9 |
+
"is_local_process_zero": true,
|
| 10 |
+
"is_world_process_zero": true,
|
| 11 |
+
"log_history": [
|
| 12 |
+
{
|
| 13 |
+
"epoch": 0.0013481631277384564,
|
| 14 |
+
"grad_norm": 2.078125,
|
| 15 |
+
"learning_rate": 2.66e-05,
|
| 16 |
+
"loss": 8.29962387084961,
|
| 17 |
+
"step": 20
|
| 18 |
+
},
|
| 19 |
+
{
|
| 20 |
+
"epoch": 0.002696326255476913,
|
| 21 |
+
"grad_norm": 1.3515625,
|
| 22 |
+
"learning_rate": 5.46e-05,
|
| 23 |
+
"loss": 8.047650909423828,
|
| 24 |
+
"step": 40
|
| 25 |
+
},
|
| 26 |
+
{
|
| 27 |
+
"epoch": 0.003370407819346141,
|
| 28 |
+
"eval_loss": 7.835535049438477,
|
| 29 |
+
"eval_runtime": 15.2165,
|
| 30 |
+
"eval_samples_per_second": 626.095,
|
| 31 |
+
"eval_steps_per_second": 0.657,
|
| 32 |
+
"step": 50
|
| 33 |
+
},
|
| 34 |
+
{
|
| 35 |
+
"epoch": 0.004044489383215369,
|
| 36 |
+
"grad_norm": 1.2734375,
|
| 37 |
+
"learning_rate": 8.259999999999999e-05,
|
| 38 |
+
"loss": 7.8378746032714846,
|
| 39 |
+
"step": 60
|
| 40 |
+
},
|
| 41 |
+
{
|
| 42 |
+
"epoch": 0.005392652510953826,
|
| 43 |
+
"grad_norm": 1.265625,
|
| 44 |
+
"learning_rate": 0.0001106,
|
| 45 |
+
"loss": 7.5728271484375,
|
| 46 |
+
"step": 80
|
| 47 |
+
},
|
| 48 |
+
{
|
| 49 |
+
"epoch": 0.006740815638692282,
|
| 50 |
+
"grad_norm": 1.1953125,
|
| 51 |
+
"learning_rate": 0.0001386,
|
| 52 |
+
"loss": 7.2332916259765625,
|
| 53 |
+
"step": 100
|
| 54 |
+
},
|
| 55 |
+
{
|
| 56 |
+
"epoch": 0.006740815638692282,
|
| 57 |
+
"eval_loss": 7.039053440093994,
|
| 58 |
+
"eval_runtime": 15.2074,
|
| 59 |
+
"eval_samples_per_second": 626.471,
|
| 60 |
+
"eval_steps_per_second": 0.658,
|
| 61 |
+
"step": 100
|
| 62 |
+
},
|
| 63 |
+
{
|
| 64 |
+
"epoch": 0.008088978766430738,
|
| 65 |
+
"grad_norm": 1.1328125,
|
| 66 |
+
"learning_rate": 0.00016659999999999998,
|
| 67 |
+
"loss": 6.8608039855957035,
|
| 68 |
+
"step": 120
|
| 69 |
+
},
|
| 70 |
+
{
|
| 71 |
+
"epoch": 0.009437141894169195,
|
| 72 |
+
"grad_norm": 0.94921875,
|
| 73 |
+
"learning_rate": 0.00019460000000000001,
|
| 74 |
+
"loss": 6.517842864990234,
|
| 75 |
+
"step": 140
|
| 76 |
+
},
|
| 77 |
+
{
|
| 78 |
+
"epoch": 0.010111223458038422,
|
| 79 |
+
"eval_loss": 6.229617595672607,
|
| 80 |
+
"eval_runtime": 15.1978,
|
| 81 |
+
"eval_samples_per_second": 626.867,
|
| 82 |
+
"eval_steps_per_second": 0.658,
|
| 83 |
+
"step": 150
|
| 84 |
+
},
|
| 85 |
+
{
|
| 86 |
+
"epoch": 0.010785305021907651,
|
| 87 |
+
"grad_norm": 0.80078125,
|
| 88 |
+
"learning_rate": 0.0002226,
|
| 89 |
+
"loss": 6.2399169921875,
|
| 90 |
+
"step": 160
|
| 91 |
+
},
|
| 92 |
+
{
|
| 93 |
+
"epoch": 0.012133468149646108,
|
| 94 |
+
"grad_norm": 0.6640625,
|
| 95 |
+
"learning_rate": 0.00025059999999999997,
|
| 96 |
+
"loss": 6.036548614501953,
|
| 97 |
+
"step": 180
|
| 98 |
+
},
|
| 99 |
+
{
|
| 100 |
+
"epoch": 0.013481631277384564,
|
| 101 |
+
"grad_norm": 0.61328125,
|
| 102 |
+
"learning_rate": 0.0002786,
|
| 103 |
+
"loss": 5.857078552246094,
|
| 104 |
+
"step": 200
|
| 105 |
+
},
|
| 106 |
+
{
|
| 107 |
+
"epoch": 0.013481631277384564,
|
| 108 |
+
"eval_loss": 5.765108585357666,
|
| 109 |
+
"eval_runtime": 15.5271,
|
| 110 |
+
"eval_samples_per_second": 613.574,
|
| 111 |
+
"eval_steps_per_second": 0.644,
|
| 112 |
+
"step": 200
|
| 113 |
+
},
|
| 114 |
+
{
|
| 115 |
+
"epoch": 0.01482979440512302,
|
| 116 |
+
"grad_norm": 0.61328125,
|
| 117 |
+
"learning_rate": 0.00030659999999999997,
|
| 118 |
+
"loss": 5.680564117431641,
|
| 119 |
+
"step": 220
|
| 120 |
+
},
|
| 121 |
+
{
|
| 122 |
+
"epoch": 0.016177957532861477,
|
| 123 |
+
"grad_norm": 2.0625,
|
| 124 |
+
"learning_rate": 0.0003346,
|
| 125 |
+
"loss": 5.472259902954102,
|
| 126 |
+
"step": 240
|
| 127 |
+
},
|
| 128 |
+
{
|
| 129 |
+
"epoch": 0.016852039096730706,
|
| 130 |
+
"eval_loss": 5.240335464477539,
|
| 131 |
+
"eval_runtime": 15.2596,
|
| 132 |
+
"eval_samples_per_second": 624.33,
|
| 133 |
+
"eval_steps_per_second": 0.655,
|
| 134 |
+
"step": 250
|
| 135 |
+
},
|
| 136 |
+
{
|
| 137 |
+
"epoch": 0.01752612066059993,
|
| 138 |
+
"grad_norm": 1.7421875,
|
| 139 |
+
"learning_rate": 0.00036260000000000003,
|
| 140 |
+
"loss": 5.24955940246582,
|
| 141 |
+
"step": 260
|
| 142 |
+
},
|
| 143 |
+
{
|
| 144 |
+
"epoch": 0.01887428378833839,
|
| 145 |
+
"grad_norm": 0.62109375,
|
| 146 |
+
"learning_rate": 0.0003906,
|
| 147 |
+
"loss": 4.995759963989258,
|
| 148 |
+
"step": 280
|
| 149 |
+
},
|
| 150 |
+
{
|
| 151 |
+
"epoch": 0.020222446916076844,
|
| 152 |
+
"grad_norm": 1.390625,
|
| 153 |
+
"learning_rate": 0.0004186,
|
| 154 |
+
"loss": 4.769353866577148,
|
| 155 |
+
"step": 300
|
| 156 |
+
},
|
| 157 |
+
{
|
| 158 |
+
"epoch": 0.020222446916076844,
|
| 159 |
+
"eval_loss": 4.669629096984863,
|
| 160 |
+
"eval_runtime": 15.2595,
|
| 161 |
+
"eval_samples_per_second": 624.332,
|
| 162 |
+
"eval_steps_per_second": 0.655,
|
| 163 |
+
"step": 300
|
| 164 |
+
},
|
| 165 |
+
{
|
| 166 |
+
"epoch": 0.021570610043815303,
|
| 167 |
+
"grad_norm": 0.73828125,
|
| 168 |
+
"learning_rate": 0.0004466,
|
| 169 |
+
"loss": 4.569867324829102,
|
| 170 |
+
"step": 320
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 0.022918773171553757,
|
| 174 |
+
"grad_norm": 0.765625,
|
| 175 |
+
"learning_rate": 0.00047460000000000004,
|
| 176 |
+
"loss": 4.389446258544922,
|
| 177 |
+
"step": 340
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 0.023592854735422986,
|
| 181 |
+
"eval_loss": 4.231207847595215,
|
| 182 |
+
"eval_runtime": 15.2143,
|
| 183 |
+
"eval_samples_per_second": 626.188,
|
| 184 |
+
"eval_steps_per_second": 0.657,
|
| 185 |
+
"step": 350
|
| 186 |
+
},
|
| 187 |
+
{
|
| 188 |
+
"epoch": 0.024266936299292215,
|
| 189 |
+
"grad_norm": 1.2421875,
|
| 190 |
+
"learning_rate": 0.0005026,
|
| 191 |
+
"loss": 4.243848419189453,
|
| 192 |
+
"step": 360
|
| 193 |
+
},
|
| 194 |
+
{
|
| 195 |
+
"epoch": 0.02561509942703067,
|
| 196 |
+
"grad_norm": 0.7109375,
|
| 197 |
+
"learning_rate": 0.0005306,
|
| 198 |
+
"loss": 4.1090648651123045,
|
| 199 |
+
"step": 380
|
| 200 |
+
},
|
| 201 |
+
{
|
| 202 |
+
"epoch": 0.026963262554769128,
|
| 203 |
+
"grad_norm": 0.484375,
|
| 204 |
+
"learning_rate": 0.0005586,
|
| 205 |
+
"loss": 3.966172790527344,
|
| 206 |
+
"step": 400
|
| 207 |
+
},
|
| 208 |
+
{
|
| 209 |
+
"epoch": 0.026963262554769128,
|
| 210 |
+
"eval_loss": 3.923773765563965,
|
| 211 |
+
"eval_runtime": 15.5308,
|
| 212 |
+
"eval_samples_per_second": 613.426,
|
| 213 |
+
"eval_steps_per_second": 0.644,
|
| 214 |
+
"step": 400
|
| 215 |
+
},
|
| 216 |
+
{
|
| 217 |
+
"epoch": 0.028311425682507583,
|
| 218 |
+
"grad_norm": 0.6640625,
|
| 219 |
+
"learning_rate": 0.0005866,
|
| 220 |
+
"loss": 3.8559597015380858,
|
| 221 |
+
"step": 420
|
| 222 |
+
},
|
| 223 |
+
{
|
| 224 |
+
"epoch": 0.02965958881024604,
|
| 225 |
+
"grad_norm": 0.515625,
|
| 226 |
+
"learning_rate": 0.0006146,
|
| 227 |
+
"loss": 3.7943866729736326,
|
| 228 |
+
"step": 440
|
| 229 |
+
},
|
| 230 |
+
{
|
| 231 |
+
"epoch": 0.030333670374115267,
|
| 232 |
+
"eval_loss": 3.6977553367614746,
|
| 233 |
+
"eval_runtime": 15.2001,
|
| 234 |
+
"eval_samples_per_second": 626.77,
|
| 235 |
+
"eval_steps_per_second": 0.658,
|
| 236 |
+
"step": 450
|
| 237 |
+
},
|
| 238 |
+
{
|
| 239 |
+
"epoch": 0.031007751937984496,
|
| 240 |
+
"grad_norm": 0.51953125,
|
| 241 |
+
"learning_rate": 0.0006426,
|
| 242 |
+
"loss": 3.686443328857422,
|
| 243 |
+
"step": 460
|
| 244 |
+
},
|
| 245 |
+
{
|
| 246 |
+
"epoch": 0.032355915065722954,
|
| 247 |
+
"grad_norm": 0.62109375,
|
| 248 |
+
"learning_rate": 0.0006705999999999999,
|
| 249 |
+
"loss": 3.6123157501220704,
|
| 250 |
+
"step": 480
|
| 251 |
+
},
|
| 252 |
+
{
|
| 253 |
+
"epoch": 0.03370407819346141,
|
| 254 |
+
"grad_norm": 0.70703125,
|
| 255 |
+
"learning_rate": 0.0006986,
|
| 256 |
+
"loss": 3.542654800415039,
|
| 257 |
+
"step": 500
|
| 258 |
+
},
|
| 259 |
+
{
|
| 260 |
+
"epoch": 0.03370407819346141,
|
| 261 |
+
"eval_loss": 3.513820171356201,
|
| 262 |
+
"eval_runtime": 15.2193,
|
| 263 |
+
"eval_samples_per_second": 625.983,
|
| 264 |
+
"eval_steps_per_second": 0.657,
|
| 265 |
+
"step": 500
|
| 266 |
+
},
|
| 267 |
+
{
|
| 268 |
+
"epoch": 0.03505224132119986,
|
| 269 |
+
"grad_norm": 0.48046875,
|
| 270 |
+
"learning_rate": 0.0007,
|
| 271 |
+
"loss": 3.472199249267578,
|
| 272 |
+
"step": 520
|
| 273 |
+
},
|
| 274 |
+
{
|
| 275 |
+
"epoch": 0.03640040444893832,
|
| 276 |
+
"grad_norm": 0.53515625,
|
| 277 |
+
"learning_rate": 0.0007,
|
| 278 |
+
"loss": 3.396291732788086,
|
| 279 |
+
"step": 540
|
| 280 |
+
},
|
| 281 |
+
{
|
| 282 |
+
"epoch": 0.03707448601280755,
|
| 283 |
+
"eval_loss": 3.3581483364105225,
|
| 284 |
+
"eval_runtime": 15.3044,
|
| 285 |
+
"eval_samples_per_second": 622.502,
|
| 286 |
+
"eval_steps_per_second": 0.653,
|
| 287 |
+
"step": 550
|
| 288 |
+
},
|
| 289 |
+
{
|
| 290 |
+
"epoch": 0.03774856757667678,
|
| 291 |
+
"grad_norm": 0.47265625,
|
| 292 |
+
"learning_rate": 0.0007,
|
| 293 |
+
"loss": 3.347300720214844,
|
| 294 |
+
"step": 560
|
| 295 |
+
},
|
| 296 |
+
{
|
| 297 |
+
"epoch": 0.03909673070441524,
|
| 298 |
+
"grad_norm": 0.458984375,
|
| 299 |
+
"learning_rate": 0.0007,
|
| 300 |
+
"loss": 3.2885902404785154,
|
| 301 |
+
"step": 580
|
| 302 |
+
},
|
| 303 |
+
{
|
| 304 |
+
"epoch": 0.04044489383215369,
|
| 305 |
+
"grad_norm": 0.4375,
|
| 306 |
+
"learning_rate": 0.0007,
|
| 307 |
+
"loss": 3.25723876953125,
|
| 308 |
+
"step": 600
|
| 309 |
+
},
|
| 310 |
+
{
|
| 311 |
+
"epoch": 0.04044489383215369,
|
| 312 |
+
"eval_loss": 3.2229573726654053,
|
| 313 |
+
"eval_runtime": 15.2967,
|
| 314 |
+
"eval_samples_per_second": 622.812,
|
| 315 |
+
"eval_steps_per_second": 0.654,
|
| 316 |
+
"step": 600
|
| 317 |
+
},
|
| 318 |
+
{
|
| 319 |
+
"epoch": 0.04179305695989215,
|
| 320 |
+
"grad_norm": 0.458984375,
|
| 321 |
+
"learning_rate": 0.0007,
|
| 322 |
+
"loss": 3.1902334213256838,
|
| 323 |
+
"step": 620
|
| 324 |
+
},
|
| 325 |
+
{
|
| 326 |
+
"epoch": 0.043141220087630605,
|
| 327 |
+
"grad_norm": 0.40625,
|
| 328 |
+
"learning_rate": 0.0007,
|
| 329 |
+
"loss": 3.167861557006836,
|
| 330 |
+
"step": 640
|
| 331 |
+
},
|
| 332 |
+
{
|
| 333 |
+
"epoch": 0.043815301651499834,
|
| 334 |
+
"eval_loss": 3.120636463165283,
|
| 335 |
+
"eval_runtime": 15.3751,
|
| 336 |
+
"eval_samples_per_second": 619.639,
|
| 337 |
+
"eval_steps_per_second": 0.65,
|
| 338 |
+
"step": 650
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 0.044489383215369056,
|
| 342 |
+
"grad_norm": 0.494140625,
|
| 343 |
+
"learning_rate": 0.0007,
|
| 344 |
+
"loss": 3.1089324951171875,
|
| 345 |
+
"step": 660
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 0.045837546343107514,
|
| 349 |
+
"grad_norm": 0.435546875,
|
| 350 |
+
"learning_rate": 0.0007,
|
| 351 |
+
"loss": 3.0828268051147463,
|
| 352 |
+
"step": 680
|
| 353 |
+
},
|
| 354 |
+
{
|
| 355 |
+
"epoch": 0.04718570947084597,
|
| 356 |
+
"grad_norm": 0.427734375,
|
| 357 |
+
"learning_rate": 0.0007,
|
| 358 |
+
"loss": 3.0396488189697264,
|
| 359 |
+
"step": 700
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"epoch": 0.04718570947084597,
|
| 363 |
+
"eval_loss": 3.0276408195495605,
|
| 364 |
+
"eval_runtime": 15.2735,
|
| 365 |
+
"eval_samples_per_second": 623.761,
|
| 366 |
+
"eval_steps_per_second": 0.655,
|
| 367 |
+
"step": 700
|
| 368 |
+
},
|
| 369 |
+
{
|
| 370 |
+
"epoch": 0.04853387259858443,
|
| 371 |
+
"grad_norm": 0.4375,
|
| 372 |
+
"learning_rate": 0.0007,
|
| 373 |
+
"loss": 3.0034805297851563,
|
| 374 |
+
"step": 720
|
| 375 |
+
},
|
| 376 |
+
{
|
| 377 |
+
"epoch": 0.04988203572632288,
|
| 378 |
+
"grad_norm": 0.52734375,
|
| 379 |
+
"learning_rate": 0.0007,
|
| 380 |
+
"loss": 2.96556396484375,
|
| 381 |
+
"step": 740
|
| 382 |
+
},
|
| 383 |
+
{
|
| 384 |
+
"epoch": 0.05055611729019211,
|
| 385 |
+
"eval_loss": 2.950009346008301,
|
| 386 |
+
"eval_runtime": 15.243,
|
| 387 |
+
"eval_samples_per_second": 625.009,
|
| 388 |
+
"eval_steps_per_second": 0.656,
|
| 389 |
+
"step": 750
|
| 390 |
+
},
|
| 391 |
+
{
|
| 392 |
+
"epoch": 0.05123019885406134,
|
| 393 |
+
"grad_norm": 0.451171875,
|
| 394 |
+
"learning_rate": 0.0007,
|
| 395 |
+
"loss": 2.950779151916504,
|
| 396 |
+
"step": 760
|
| 397 |
+
},
|
| 398 |
+
{
|
| 399 |
+
"epoch": 0.0525783619817998,
|
| 400 |
+
"grad_norm": 0.400390625,
|
| 401 |
+
"learning_rate": 0.0007,
|
| 402 |
+
"loss": 2.956203079223633,
|
| 403 |
+
"step": 780
|
| 404 |
+
},
|
| 405 |
+
{
|
| 406 |
+
"epoch": 0.053926525109538256,
|
| 407 |
+
"grad_norm": 0.40625,
|
| 408 |
+
"learning_rate": 0.0007,
|
| 409 |
+
"loss": 2.9017578125,
|
| 410 |
+
"step": 800
|
| 411 |
+
},
|
| 412 |
+
{
|
| 413 |
+
"epoch": 0.053926525109538256,
|
| 414 |
+
"eval_loss": 2.8844006061553955,
|
| 415 |
+
"eval_runtime": 15.2903,
|
| 416 |
+
"eval_samples_per_second": 623.073,
|
| 417 |
+
"eval_steps_per_second": 0.654,
|
| 418 |
+
"step": 800
|
| 419 |
+
},
|
| 420 |
+
{
|
| 421 |
+
"epoch": 0.05527468823727671,
|
| 422 |
+
"grad_norm": 0.41015625,
|
| 423 |
+
"learning_rate": 0.0007,
|
| 424 |
+
"loss": 2.887515640258789,
|
| 425 |
+
"step": 820
|
| 426 |
+
},
|
| 427 |
+
{
|
| 428 |
+
"epoch": 0.056622851365015166,
|
| 429 |
+
"grad_norm": 0.41796875,
|
| 430 |
+
"learning_rate": 0.0007,
|
| 431 |
+
"loss": 2.8556928634643555,
|
| 432 |
+
"step": 840
|
| 433 |
+
},
|
| 434 |
+
{
|
| 435 |
+
"epoch": 0.057296932928884395,
|
| 436 |
+
"eval_loss": 2.821610450744629,
|
| 437 |
+
"eval_runtime": 15.2253,
|
| 438 |
+
"eval_samples_per_second": 625.733,
|
| 439 |
+
"eval_steps_per_second": 0.657,
|
| 440 |
+
"step": 850
|
| 441 |
+
},
|
| 442 |
+
{
|
| 443 |
+
"epoch": 0.057971014492753624,
|
| 444 |
+
"grad_norm": 0.423828125,
|
| 445 |
+
"learning_rate": 0.0007,
|
| 446 |
+
"loss": 2.8130855560302734,
|
| 447 |
+
"step": 860
|
| 448 |
+
},
|
| 449 |
+
{
|
| 450 |
+
"epoch": 0.05931917762049208,
|
| 451 |
+
"grad_norm": 0.38671875,
|
| 452 |
+
"learning_rate": 0.0007,
|
| 453 |
+
"loss": 2.790329170227051,
|
| 454 |
+
"step": 880
|
| 455 |
+
},
|
| 456 |
+
{
|
| 457 |
+
"epoch": 0.06066734074823053,
|
| 458 |
+
"grad_norm": 0.427734375,
|
| 459 |
+
"learning_rate": 0.0007,
|
| 460 |
+
"loss": 2.7760162353515625,
|
| 461 |
+
"step": 900
|
| 462 |
+
},
|
| 463 |
+
{
|
| 464 |
+
"epoch": 0.06066734074823053,
|
| 465 |
+
"eval_loss": 2.7661373615264893,
|
| 466 |
+
"eval_runtime": 15.2736,
|
| 467 |
+
"eval_samples_per_second": 623.755,
|
| 468 |
+
"eval_steps_per_second": 0.655,
|
| 469 |
+
"step": 900
|
| 470 |
+
},
|
| 471 |
+
{
|
| 472 |
+
"epoch": 0.06201550387596899,
|
| 473 |
+
"grad_norm": 0.435546875,
|
| 474 |
+
"learning_rate": 0.0007,
|
| 475 |
+
"loss": 2.7462047576904296,
|
| 476 |
+
"step": 920
|
| 477 |
+
},
|
| 478 |
+
{
|
| 479 |
+
"epoch": 0.06336366700370745,
|
| 480 |
+
"grad_norm": 0.4140625,
|
| 481 |
+
"learning_rate": 0.0007,
|
| 482 |
+
"loss": 2.732274627685547,
|
| 483 |
+
"step": 940
|
| 484 |
+
},
|
| 485 |
+
{
|
| 486 |
+
"epoch": 0.06403774856757667,
|
| 487 |
+
"eval_loss": 2.713942289352417,
|
| 488 |
+
"eval_runtime": 15.6062,
|
| 489 |
+
"eval_samples_per_second": 610.463,
|
| 490 |
+
"eval_steps_per_second": 0.641,
|
| 491 |
+
"step": 950
|
| 492 |
+
},
|
| 493 |
+
{
|
| 494 |
+
"epoch": 0.06471183013144591,
|
| 495 |
+
"grad_norm": 0.419921875,
|
| 496 |
+
"learning_rate": 0.0007,
|
| 497 |
+
"loss": 2.705318069458008,
|
| 498 |
+
"step": 960
|
| 499 |
+
},
|
| 500 |
+
{
|
| 501 |
+
"epoch": 0.06605999325918437,
|
| 502 |
+
"grad_norm": 0.4296875,
|
| 503 |
+
"learning_rate": 0.0007,
|
| 504 |
+
"loss": 2.6892818450927733,
|
| 505 |
+
"step": 980
|
| 506 |
+
},
|
| 507 |
+
{
|
| 508 |
+
"epoch": 0.06740815638692282,
|
| 509 |
+
"grad_norm": 0.4453125,
|
| 510 |
+
"learning_rate": 0.0007,
|
| 511 |
+
"loss": 2.687115287780762,
|
| 512 |
+
"step": 1000
|
| 513 |
+
},
|
| 514 |
+
{
|
| 515 |
+
"epoch": 0.06740815638692282,
|
| 516 |
+
"eval_loss": 2.668647289276123,
|
| 517 |
+
"eval_runtime": 15.2597,
|
| 518 |
+
"eval_samples_per_second": 624.326,
|
| 519 |
+
"eval_steps_per_second": 0.655,
|
| 520 |
+
"step": 1000
|
| 521 |
+
},
|
| 522 |
+
{
|
| 523 |
+
"epoch": 0.06875631951466127,
|
| 524 |
+
"grad_norm": 0.388671875,
|
| 525 |
+
"learning_rate": 0.0007,
|
| 526 |
+
"loss": 2.6533071517944338,
|
| 527 |
+
"step": 1020
|
| 528 |
+
},
|
| 529 |
+
{
|
| 530 |
+
"epoch": 0.07010448264239973,
|
| 531 |
+
"grad_norm": 0.392578125,
|
| 532 |
+
"learning_rate": 0.0007,
|
| 533 |
+
"loss": 2.6390922546386717,
|
| 534 |
+
"step": 1040
|
| 535 |
+
},
|
| 536 |
+
{
|
| 537 |
+
"epoch": 0.07077856420626896,
|
| 538 |
+
"eval_loss": 2.6291816234588623,
|
| 539 |
+
"eval_runtime": 15.2457,
|
| 540 |
+
"eval_samples_per_second": 624.896,
|
| 541 |
+
"eval_steps_per_second": 0.656,
|
| 542 |
+
"step": 1050
|
| 543 |
+
},
|
| 544 |
+
{
|
| 545 |
+
"epoch": 0.07145264577013818,
|
| 546 |
+
"grad_norm": 0.4375,
|
| 547 |
+
"learning_rate": 0.0007,
|
| 548 |
+
"loss": 2.624461555480957,
|
| 549 |
+
"step": 1060
|
| 550 |
+
},
|
| 551 |
+
{
|
| 552 |
+
"epoch": 0.07280080889787664,
|
| 553 |
+
"grad_norm": 0.396484375,
|
| 554 |
+
"learning_rate": 0.0007,
|
| 555 |
+
"loss": 2.6044427871704103,
|
| 556 |
+
"step": 1080
|
| 557 |
+
},
|
| 558 |
+
{
|
| 559 |
+
"epoch": 0.0741489720256151,
|
| 560 |
+
"grad_norm": 0.423828125,
|
| 561 |
+
"learning_rate": 0.0007,
|
| 562 |
+
"loss": 2.597646713256836,
|
| 563 |
+
"step": 1100
|
| 564 |
+
},
|
| 565 |
+
{
|
| 566 |
+
"epoch": 0.0741489720256151,
|
| 567 |
+
"eval_loss": 2.589696168899536,
|
| 568 |
+
"eval_runtime": 15.3967,
|
| 569 |
+
"eval_samples_per_second": 618.768,
|
| 570 |
+
"eval_steps_per_second": 0.649,
|
| 571 |
+
"step": 1100
|
| 572 |
+
},
|
| 573 |
+
{
|
| 574 |
+
"epoch": 0.07549713515335356,
|
| 575 |
+
"grad_norm": 0.369140625,
|
| 576 |
+
"learning_rate": 0.0007,
|
| 577 |
+
"loss": 2.561947250366211,
|
| 578 |
+
"step": 1120
|
| 579 |
+
},
|
| 580 |
+
{
|
| 581 |
+
"epoch": 0.07684529828109202,
|
| 582 |
+
"grad_norm": 0.390625,
|
| 583 |
+
"learning_rate": 0.0007,
|
| 584 |
+
"loss": 2.54862117767334,
|
| 585 |
+
"step": 1140
|
| 586 |
+
},
|
| 587 |
+
{
|
| 588 |
+
"epoch": 0.07751937984496124,
|
| 589 |
+
"eval_loss": 2.559185743331909,
|
| 590 |
+
"eval_runtime": 15.2877,
|
| 591 |
+
"eval_samples_per_second": 623.181,
|
| 592 |
+
"eval_steps_per_second": 0.654,
|
| 593 |
+
"step": 1150
|
| 594 |
+
},
|
| 595 |
+
{
|
| 596 |
+
"epoch": 0.07819346140883048,
|
| 597 |
+
"grad_norm": 0.37890625,
|
| 598 |
+
"learning_rate": 0.0007,
|
| 599 |
+
"loss": 2.5469717025756835,
|
| 600 |
+
"step": 1160
|
| 601 |
+
},
|
| 602 |
+
{
|
| 603 |
+
"epoch": 0.07954162453656892,
|
| 604 |
+
"grad_norm": 0.400390625,
|
| 605 |
+
"learning_rate": 0.0007,
|
| 606 |
+
"loss": 2.540451240539551,
|
| 607 |
+
"step": 1180
|
| 608 |
+
},
|
| 609 |
+
{
|
| 610 |
+
"epoch": 0.08088978766430738,
|
| 611 |
+
"grad_norm": 0.40625,
|
| 612 |
+
"learning_rate": 0.0007,
|
| 613 |
+
"loss": 2.53134708404541,
|
| 614 |
+
"step": 1200
|
| 615 |
+
},
|
| 616 |
+
{
|
| 617 |
+
"epoch": 0.08088978766430738,
|
| 618 |
+
"eval_loss": 2.521665096282959,
|
| 619 |
+
"eval_runtime": 15.2822,
|
| 620 |
+
"eval_samples_per_second": 623.407,
|
| 621 |
+
"eval_steps_per_second": 0.654,
|
| 622 |
+
"step": 1200
|
| 623 |
+
},
|
| 624 |
+
{
|
| 625 |
+
"epoch": 0.08223795079204584,
|
| 626 |
+
"grad_norm": 0.37109375,
|
| 627 |
+
"learning_rate": 0.0007,
|
| 628 |
+
"loss": 2.513539123535156,
|
| 629 |
+
"step": 1220
|
| 630 |
+
},
|
| 631 |
+
{
|
| 632 |
+
"epoch": 0.0835861139197843,
|
| 633 |
+
"grad_norm": 0.36328125,
|
| 634 |
+
"learning_rate": 0.0007,
|
| 635 |
+
"loss": 2.5009738922119142,
|
| 636 |
+
"step": 1240
|
| 637 |
+
},
|
| 638 |
+
{
|
| 639 |
+
"epoch": 0.08426019548365352,
|
| 640 |
+
"eval_loss": 2.489459753036499,
|
| 641 |
+
"eval_runtime": 15.2493,
|
| 642 |
+
"eval_samples_per_second": 624.749,
|
| 643 |
+
"eval_steps_per_second": 0.656,
|
| 644 |
+
"step": 1250
|
| 645 |
+
},
|
| 646 |
+
{
|
| 647 |
+
"epoch": 0.08493427704752275,
|
| 648 |
+
"grad_norm": 0.388671875,
|
| 649 |
+
"learning_rate": 0.0007,
|
| 650 |
+
"loss": 2.4894077301025392,
|
| 651 |
+
"step": 1260
|
| 652 |
+
},
|
| 653 |
+
{
|
| 654 |
+
"epoch": 0.08628244017526121,
|
| 655 |
+
"grad_norm": 0.369140625,
|
| 656 |
+
"learning_rate": 0.0007,
|
| 657 |
+
"loss": 2.4697439193725588,
|
| 658 |
+
"step": 1280
|
| 659 |
+
},
|
| 660 |
+
{
|
| 661 |
+
"epoch": 0.08763060330299967,
|
| 662 |
+
"grad_norm": 0.423828125,
|
| 663 |
+
"learning_rate": 0.0007,
|
| 664 |
+
"loss": 2.461294174194336,
|
| 665 |
+
"step": 1300
|
| 666 |
+
},
|
| 667 |
+
{
|
| 668 |
+
"epoch": 0.08763060330299967,
|
| 669 |
+
"eval_loss": 2.4627816677093506,
|
| 670 |
+
"eval_runtime": 15.421,
|
| 671 |
+
"eval_samples_per_second": 617.795,
|
| 672 |
+
"eval_steps_per_second": 0.648,
|
| 673 |
+
"step": 1300
|
| 674 |
+
},
|
| 675 |
+
{
|
| 676 |
+
"epoch": 0.08897876643073811,
|
| 677 |
+
"grad_norm": 0.36328125,
|
| 678 |
+
"learning_rate": 0.0007,
|
| 679 |
+
"loss": 2.452680206298828,
|
| 680 |
+
"step": 1320
|
| 681 |
+
},
|
| 682 |
+
{
|
| 683 |
+
"epoch": 0.09032692955847657,
|
| 684 |
+
"grad_norm": 0.404296875,
|
| 685 |
+
"learning_rate": 0.0007,
|
| 686 |
+
"loss": 2.4396684646606444,
|
| 687 |
+
"step": 1340
|
| 688 |
+
},
|
| 689 |
+
{
|
| 690 |
+
"epoch": 0.0910010111223458,
|
| 691 |
+
"eval_loss": 2.437220335006714,
|
| 692 |
+
"eval_runtime": 15.3026,
|
| 693 |
+
"eval_samples_per_second": 622.576,
|
| 694 |
+
"eval_steps_per_second": 0.653,
|
| 695 |
+
"step": 1350
|
| 696 |
+
},
|
| 697 |
+
{
|
| 698 |
+
"epoch": 0.09167509268621503,
|
| 699 |
+
"grad_norm": 0.412109375,
|
| 700 |
+
"learning_rate": 0.0007,
|
| 701 |
+
"loss": 2.4397760391235352,
|
| 702 |
+
"step": 1360
|
| 703 |
+
},
|
| 704 |
+
{
|
| 705 |
+
"epoch": 0.09302325581395349,
|
| 706 |
+
"grad_norm": 0.353515625,
|
| 707 |
+
"learning_rate": 0.0007,
|
| 708 |
+
"loss": 2.4158536911010744,
|
| 709 |
+
"step": 1380
|
| 710 |
+
},
|
| 711 |
+
{
|
| 712 |
+
"epoch": 0.09437141894169195,
|
| 713 |
+
"grad_norm": 0.345703125,
|
| 714 |
+
"learning_rate": 0.0007,
|
| 715 |
+
"loss": 2.406942367553711,
|
| 716 |
+
"step": 1400
|
| 717 |
+
},
|
| 718 |
+
{
|
| 719 |
+
"epoch": 0.09437141894169195,
|
| 720 |
+
"eval_loss": 2.4096601009368896,
|
| 721 |
+
"eval_runtime": 15.357,
|
| 722 |
+
"eval_samples_per_second": 620.368,
|
| 723 |
+
"eval_steps_per_second": 0.651,
|
| 724 |
+
"step": 1400
|
| 725 |
+
},
|
| 726 |
+
{
|
| 727 |
+
"epoch": 0.0957195820694304,
|
| 728 |
+
"grad_norm": 0.36328125,
|
| 729 |
+
"learning_rate": 0.0007,
|
| 730 |
+
"loss": 2.4161144256591798,
|
| 731 |
+
"step": 1420
|
| 732 |
+
},
|
| 733 |
+
{
|
| 734 |
+
"epoch": 0.09706774519716886,
|
| 735 |
+
"grad_norm": 0.36328125,
|
| 736 |
+
"learning_rate": 0.0007,
|
| 737 |
+
"loss": 2.392984390258789,
|
| 738 |
+
"step": 1440
|
| 739 |
+
},
|
| 740 |
+
{
|
| 741 |
+
"epoch": 0.09774182676103808,
|
| 742 |
+
"eval_loss": 2.3937113285064697,
|
| 743 |
+
"eval_runtime": 15.3555,
|
| 744 |
+
"eval_samples_per_second": 620.43,
|
| 745 |
+
"eval_steps_per_second": 0.651,
|
| 746 |
+
"step": 1450
|
| 747 |
+
},
|
| 748 |
+
{
|
| 749 |
+
"epoch": 0.09841590832490732,
|
| 750 |
+
"grad_norm": 0.384765625,
|
| 751 |
+
"learning_rate": 0.0007,
|
| 752 |
+
"loss": 2.3820331573486326,
|
| 753 |
+
"step": 1460
|
| 754 |
+
},
|
| 755 |
+
{
|
| 756 |
+
"epoch": 0.09976407145264576,
|
| 757 |
+
"grad_norm": 0.33984375,
|
| 758 |
+
"learning_rate": 0.0007,
|
| 759 |
+
"loss": 2.3687986373901366,
|
| 760 |
+
"step": 1480
|
| 761 |
+
},
|
| 762 |
+
{
|
| 763 |
+
"epoch": 0.10111223458038422,
|
| 764 |
+
"grad_norm": 0.38671875,
|
| 765 |
+
"learning_rate": 0.0007,
|
| 766 |
+
"loss": 2.355032730102539,
|
| 767 |
+
"step": 1500
|
| 768 |
+
},
|
| 769 |
+
{
|
| 770 |
+
"epoch": 0.10111223458038422,
|
| 771 |
+
"eval_loss": 2.3681657314300537,
|
| 772 |
+
"eval_runtime": 15.3336,
|
| 773 |
+
"eval_samples_per_second": 621.317,
|
| 774 |
+
"eval_steps_per_second": 0.652,
|
| 775 |
+
"step": 1500
|
| 776 |
+
},
|
| 777 |
+
{
|
| 778 |
+
"epoch": 0.10246039770812268,
|
| 779 |
+
"grad_norm": 0.392578125,
|
| 780 |
+
"learning_rate": 0.0007,
|
| 781 |
+
"loss": 2.3572179794311525,
|
| 782 |
+
"step": 1520
|
| 783 |
+
},
|
| 784 |
+
{
|
| 785 |
+
"epoch": 0.10380856083586114,
|
| 786 |
+
"grad_norm": 0.35546875,
|
| 787 |
+
"learning_rate": 0.0007,
|
| 788 |
+
"loss": 2.345840835571289,
|
| 789 |
+
"step": 1540
|
| 790 |
+
},
|
| 791 |
+
{
|
| 792 |
+
"epoch": 0.10448264239973036,
|
| 793 |
+
"eval_loss": 2.3475193977355957,
|
| 794 |
+
"eval_runtime": 15.3453,
|
| 795 |
+
"eval_samples_per_second": 620.842,
|
| 796 |
+
"eval_steps_per_second": 0.652,
|
| 797 |
+
"step": 1550
|
| 798 |
+
},
|
| 799 |
+
{
|
| 800 |
+
"epoch": 0.1051567239635996,
|
| 801 |
+
"grad_norm": 0.388671875,
|
| 802 |
+
"learning_rate": 0.0007,
|
| 803 |
+
"loss": 2.335231971740723,
|
| 804 |
+
"step": 1560
|
| 805 |
+
},
|
| 806 |
+
{
|
| 807 |
+
"epoch": 0.10650488709133805,
|
| 808 |
+
"grad_norm": 0.359375,
|
| 809 |
+
"learning_rate": 0.0007,
|
| 810 |
+
"loss": 2.3543283462524416,
|
| 811 |
+
"step": 1580
|
| 812 |
+
},
|
| 813 |
+
{
|
| 814 |
+
"epoch": 0.10785305021907651,
|
| 815 |
+
"grad_norm": 0.39453125,
|
| 816 |
+
"learning_rate": 0.0007,
|
| 817 |
+
"loss": 2.3237049102783205,
|
| 818 |
+
"step": 1600
|
| 819 |
+
},
|
| 820 |
+
{
|
| 821 |
+
"epoch": 0.10785305021907651,
|
| 822 |
+
"eval_loss": 2.3377676010131836,
|
| 823 |
+
"eval_runtime": 15.2858,
|
| 824 |
+
"eval_samples_per_second": 623.259,
|
| 825 |
+
"eval_steps_per_second": 0.654,
|
| 826 |
+
"step": 1600
|
| 827 |
+
},
|
| 828 |
+
{
|
| 829 |
+
"epoch": 0.10920121334681497,
|
| 830 |
+
"grad_norm": 0.357421875,
|
| 831 |
+
"learning_rate": 0.0007,
|
| 832 |
+
"loss": 2.3196277618408203,
|
| 833 |
+
"step": 1620
|
| 834 |
+
},
|
| 835 |
+
{
|
| 836 |
+
"epoch": 0.11054937647455342,
|
| 837 |
+
"grad_norm": 0.341796875,
|
| 838 |
+
"learning_rate": 0.0007,
|
| 839 |
+
"loss": 2.3308162689208984,
|
| 840 |
+
"step": 1640
|
| 841 |
+
},
|
| 842 |
+
{
|
| 843 |
+
"epoch": 0.11122345803842265,
|
| 844 |
+
"eval_loss": 2.313627004623413,
|
| 845 |
+
"eval_runtime": 15.276,
|
| 846 |
+
"eval_samples_per_second": 623.659,
|
| 847 |
+
"eval_steps_per_second": 0.655,
|
| 848 |
+
"step": 1650
|
| 849 |
+
},
|
| 850 |
+
{
|
| 851 |
+
"epoch": 0.11189753960229187,
|
| 852 |
+
"grad_norm": 0.353515625,
|
| 853 |
+
"learning_rate": 0.0007,
|
| 854 |
+
"loss": 2.3104904174804686,
|
| 855 |
+
"step": 1660
|
| 856 |
+
},
|
| 857 |
+
{
|
| 858 |
+
"epoch": 0.11324570273003033,
|
| 859 |
+
"grad_norm": 0.353515625,
|
| 860 |
+
"learning_rate": 0.0007,
|
| 861 |
+
"loss": 2.2955816268920897,
|
| 862 |
+
"step": 1680
|
| 863 |
+
},
|
| 864 |
+
{
|
| 865 |
+
"epoch": 0.11459386585776879,
|
| 866 |
+
"grad_norm": 0.345703125,
|
| 867 |
+
"learning_rate": 0.0007,
|
| 868 |
+
"loss": 2.3044401168823243,
|
| 869 |
+
"step": 1700
|
| 870 |
+
},
|
| 871 |
+
{
|
| 872 |
+
"epoch": 0.11459386585776879,
|
| 873 |
+
"eval_loss": 2.2976956367492676,
|
| 874 |
+
"eval_runtime": 15.3404,
|
| 875 |
+
"eval_samples_per_second": 621.039,
|
| 876 |
+
"eval_steps_per_second": 0.652,
|
| 877 |
+
"step": 1700
|
| 878 |
+
},
|
| 879 |
+
{
|
| 880 |
+
"epoch": 0.11594202898550725,
|
| 881 |
+
"grad_norm": 0.373046875,
|
| 882 |
+
"learning_rate": 0.0007,
|
| 883 |
+
"loss": 2.280081939697266,
|
| 884 |
+
"step": 1720
|
| 885 |
+
},
|
| 886 |
+
{
|
| 887 |
+
"epoch": 0.1172901921132457,
|
| 888 |
+
"grad_norm": 0.373046875,
|
| 889 |
+
"learning_rate": 0.0007,
|
| 890 |
+
"loss": 2.269576644897461,
|
| 891 |
+
"step": 1740
|
| 892 |
+
},
|
| 893 |
+
{
|
| 894 |
+
"epoch": 0.11796427367711493,
|
| 895 |
+
"eval_loss": 2.28181791305542,
|
| 896 |
+
"eval_runtime": 15.2137,
|
| 897 |
+
"eval_samples_per_second": 626.211,
|
| 898 |
+
"eval_steps_per_second": 0.657,
|
| 899 |
+
"step": 1750
|
| 900 |
+
},
|
| 901 |
+
{
|
| 902 |
+
"epoch": 0.11863835524098416,
|
| 903 |
+
"grad_norm": 0.359375,
|
| 904 |
+
"learning_rate": 0.0007,
|
| 905 |
+
"loss": 2.273081398010254,
|
| 906 |
+
"step": 1760
|
| 907 |
+
},
|
| 908 |
+
{
|
| 909 |
+
"epoch": 0.11998651836872262,
|
| 910 |
+
"grad_norm": 0.345703125,
|
| 911 |
+
"learning_rate": 0.0007,
|
| 912 |
+
"loss": 2.2877025604248047,
|
| 913 |
+
"step": 1780
|
| 914 |
+
},
|
| 915 |
+
{
|
| 916 |
+
"epoch": 0.12133468149646107,
|
| 917 |
+
"grad_norm": 0.3671875,
|
| 918 |
+
"learning_rate": 0.0007,
|
| 919 |
+
"loss": 2.2828989028930664,
|
| 920 |
+
"step": 1800
|
| 921 |
+
},
|
| 922 |
+
{
|
| 923 |
+
"epoch": 0.12133468149646107,
|
| 924 |
+
"eval_loss": 2.2671334743499756,
|
| 925 |
+
"eval_runtime": 15.1615,
|
| 926 |
+
"eval_samples_per_second": 628.369,
|
| 927 |
+
"eval_steps_per_second": 0.66,
|
| 928 |
+
"step": 1800
|
| 929 |
+
}
|
| 930 |
+
],
|
| 931 |
+
"logging_steps": 20,
|
| 932 |
+
"max_steps": 2500,
|
| 933 |
+
"num_input_tokens_seen": 0,
|
| 934 |
+
"num_train_epochs": 1,
|
| 935 |
+
"save_steps": 100,
|
| 936 |
+
"stateful_callbacks": {
|
| 937 |
+
"TrainerControl": {
|
| 938 |
+
"args": {
|
| 939 |
+
"should_epoch_stop": false,
|
| 940 |
+
"should_evaluate": false,
|
| 941 |
+
"should_log": false,
|
| 942 |
+
"should_save": true,
|
| 943 |
+
"should_training_stop": false
|
| 944 |
+
},
|
| 945 |
+
"attributes": {}
|
| 946 |
+
}
|
| 947 |
+
},
|
| 948 |
+
"total_flos": 5807310844723200.0,
|
| 949 |
+
"train_batch_size": 64,
|
| 950 |
+
"trial_name": null,
|
| 951 |
+
"trial_params": null
|
| 952 |
+
}
|
zain/Activation/out/glu-silu-100L_run/checkpoint-1800/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:fd90512e56f813612dc668676d6f8cab72da110a42f23c61cafebf91064625a1
|
| 3 |
+
size 4920
|
zain/Activation/out/glu-silu-100L_run/training_log.jsonl
CHANGED
|
@@ -92,3 +92,39 @@
|
|
| 92 |
{"step": 1300, "epoch": 0.08763060330299967, "timestamp": 1786571342.4780235, "loss": 2.461294174194336, "grad_norm": 0.423828125, "learning_rate": 0.0007, "train/total_time_seconds": 307.07267947494984, "train/time_per_step_avg": 0.23488603215664625, "train/epoch_time_elapsed": 757.1364300251007, "train/estimated_remaining_minutes": 4.724195068845382}
|
| 93 |
{"step": 1300, "epoch": 0.08763060330299967, "timestamp": 1786571357.9005606, "eval_loss": 2.4627816677093506, "eval_runtime": 15.421, "eval_samples_per_second": 617.795, "eval_steps_per_second": 0.648, "train/total_time_seconds": 307.07267947494984, "train/time_per_step_avg": 0.23488603215664625, "train/epoch_time_elapsed": 772.5589650757611, "train/estimated_remaining_minutes": 4.724195068845382}
|
| 94 |
{"step": 1320, "epoch": 0.08897876643073811, "timestamp": 1786571364.13814, "loss": 2.452680206298828, "grad_norm": 0.36328125, "learning_rate": 0.0007, "train/total_time_seconds": 312.06660179793835, "train/time_per_step_avg": 0.23786656107753515, "train/epoch_time_elapsed": 778.7965461052954, "train/estimated_remaining_minutes": 4.649477147999587}
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 92 |
{"step": 1300, "epoch": 0.08763060330299967, "timestamp": 1786571342.4780235, "loss": 2.461294174194336, "grad_norm": 0.423828125, "learning_rate": 0.0007, "train/total_time_seconds": 307.07267947494984, "train/time_per_step_avg": 0.23488603215664625, "train/epoch_time_elapsed": 757.1364300251007, "train/estimated_remaining_minutes": 4.724195068845382}
|
| 93 |
{"step": 1300, "epoch": 0.08763060330299967, "timestamp": 1786571357.9005606, "eval_loss": 2.4627816677093506, "eval_runtime": 15.421, "eval_samples_per_second": 617.795, "eval_steps_per_second": 0.648, "train/total_time_seconds": 307.07267947494984, "train/time_per_step_avg": 0.23488603215664625, "train/epoch_time_elapsed": 772.5589650757611, "train/estimated_remaining_minutes": 4.724195068845382}
|
| 94 |
{"step": 1320, "epoch": 0.08897876643073811, "timestamp": 1786571364.13814, "loss": 2.452680206298828, "grad_norm": 0.36328125, "learning_rate": 0.0007, "train/total_time_seconds": 312.06660179793835, "train/time_per_step_avg": 0.23786656107753515, "train/epoch_time_elapsed": 778.7965461052954, "train/estimated_remaining_minutes": 4.649477147999587}
|
| 95 |
+
{"step": 1340, "epoch": 0.09032692955847657, "timestamp": 1786571369.85727, "loss": 2.4396684646606444, "grad_norm": 0.404296875, "learning_rate": 0.0007, "train/total_time_seconds": 316.7943187877536, "train/time_per_step_avg": 0.23815039787441492, "train/epoch_time_elapsed": 784.5156761556864, "train/estimated_remaining_minutes": 4.570664300917838}
|
| 96 |
+
{"step": 1350, "epoch": 0.0910010111223458, "timestamp": 1786571388.0114145, "eval_loss": 2.437220335006714, "eval_runtime": 15.3026, "eval_samples_per_second": 622.576, "eval_steps_per_second": 0.653, "train/total_time_seconds": 319.1534503661096, "train/time_per_step_avg": 0.23833153448998928, "train/epoch_time_elapsed": 802.6698179095984, "train/estimated_remaining_minutes": 4.531190961987976}
|
| 97 |
+
{"step": 1360, "epoch": 0.09167509268621503, "timestamp": 1786571390.8419049, "loss": 2.4397760391235352, "grad_norm": 0.412109375, "learning_rate": 0.0007, "train/total_time_seconds": 321.49719002470374, "train/time_per_step_avg": 0.2383602423593402, "train/epoch_time_elapsed": 805.5003115534782, "train/estimated_remaining_minutes": 4.491504860639243}
|
| 98 |
+
{"step": 1380, "epoch": 0.09302325581395349, "timestamp": 1786571396.5560505, "loss": 2.4158536911010744, "grad_norm": 0.353515625, "learning_rate": 0.0007, "train/total_time_seconds": 326.22012339904904, "train/time_per_step_avg": 0.23858194533735513, "train/epoch_time_elapsed": 811.2144560106099, "train/estimated_remaining_minutes": 4.412639350325302}
|
| 99 |
+
{"step": 1400, "epoch": 0.09437141894169195, "timestamp": 1786571402.2409782, "loss": 2.406942367553711, "grad_norm": 0.345703125, "learning_rate": 0.0007, "train/total_time_seconds": 330.9265225343406, "train/time_per_step_avg": 0.23853843059390784, "train/epoch_time_elapsed": 816.8993840627372, "train/estimated_remaining_minutes": 4.333561604616365}
|
| 100 |
+
{"step": 1400, "epoch": 0.09437141894169195, "timestamp": 1786571417.5995984, "eval_loss": 2.4096601009368896, "eval_runtime": 15.357, "eval_samples_per_second": 620.368, "eval_steps_per_second": 0.651, "train/total_time_seconds": 330.9265225343406, "train/time_per_step_avg": 0.23853843059390784, "train/epoch_time_elapsed": 832.2580012530088, "train/estimated_remaining_minutes": 4.333561604616365}
|
| 101 |
+
{"step": 1420, "epoch": 0.0957195820694304, "timestamp": 1786571423.498136, "loss": 2.4161144256591798, "grad_norm": 0.36328125, "learning_rate": 0.0007, "train/total_time_seconds": 335.6057075113058, "train/time_per_step_avg": 0.2353910571336746, "train/epoch_time_elapsed": 838.1565419510007, "train/estimated_remaining_minutes": 4.254156855777116}
|
| 102 |
+
{"step": 1440, "epoch": 0.09706774519716886, "timestamp": 1786571429.2081707, "loss": 2.392984390258789, "grad_norm": 0.36328125, "learning_rate": 0.0007, "train/total_time_seconds": 340.3155130036175, "train/time_per_step_avg": 0.23521194215863944, "train/epoch_time_elapsed": 843.8665771856904, "train/estimated_remaining_minutes": 4.175167173424011}
|
| 103 |
+
{"step": 1450, "epoch": 0.09774182676103808, "timestamp": 1786571447.4222121, "eval_loss": 2.3937113285064697, "eval_runtime": 15.3555, "eval_samples_per_second": 620.43, "eval_steps_per_second": 0.651, "train/total_time_seconds": 342.66722694411874, "train/time_per_step_avg": 0.2351377657800913, "train/epoch_time_elapsed": 862.0806160643697, "train/estimated_remaining_minutes": 4.135638945877295}
|
| 104 |
+
{"step": 1460, "epoch": 0.09841590832490732, "timestamp": 1786571450.2504935, "loss": 2.3820331573486326, "grad_norm": 0.384765625, "learning_rate": 0.0007, "train/total_time_seconds": 345.0103280842304, "train/time_per_step_avg": 0.2351313805952668, "train/epoch_time_elapsed": 864.9088996425271, "train/estimated_remaining_minutes": 4.096013027484014}
|
| 105 |
+
{"step": 1480, "epoch": 0.09976407145264576, "timestamp": 1786571455.9186325, "loss": 2.3687986373901366, "grad_norm": 0.33984375, "learning_rate": 0.0007, "train/total_time_seconds": 349.7033684179187, "train/time_per_step_avg": 0.2348324501886964, "train/epoch_time_elapsed": 870.5770386792719, "train/estimated_remaining_minutes": 4.016863015611228}
|
| 106 |
+
{"step": 1500, "epoch": 0.10111223458038422, "timestamp": 1786571461.6036294, "loss": 2.355032730102539, "grad_norm": 0.38671875, "learning_rate": 0.0007, "train/total_time_seconds": 354.4035275951028, "train/time_per_step_avg": 0.23477005060762168, "train/epoch_time_elapsed": 876.2620359696448, "train/estimated_remaining_minutes": 3.93781697327892}
|
| 107 |
+
{"step": 1500, "epoch": 0.10111223458038422, "timestamp": 1786571476.9387658, "eval_loss": 2.3681657314300537, "eval_runtime": 15.3336, "eval_samples_per_second": 621.317, "eval_steps_per_second": 0.652, "train/total_time_seconds": 354.4035275951028, "train/time_per_step_avg": 0.23477005060762168, "train/epoch_time_elapsed": 891.5971701219678, "train/estimated_remaining_minutes": 3.93781697327892}
|
| 108 |
+
{"step": 1520, "epoch": 0.10246039770812268, "timestamp": 1786571483.158526, "loss": 2.3572179794311525, "grad_norm": 0.392578125, "learning_rate": 0.0007, "train/total_time_seconds": 359.0897579006851, "train/time_per_step_avg": 0.23484050389379263, "train/epoch_time_elapsed": 897.8169320560992, "train/estimated_remaining_minutes": 3.858639942353853}
|
| 109 |
+
{"step": 1540, "epoch": 0.10380856083586114, "timestamp": 1786571488.8240943, "loss": 2.345840835571289, "grad_norm": 0.35546875, "learning_rate": 0.0007, "train/total_time_seconds": 363.7835675626993, "train/time_per_step_avg": 0.23468054559081794, "train/epoch_time_elapsed": 903.4824999682605, "train/estimated_remaining_minutes": 3.779569533118954}
|
| 110 |
+
{"step": 1550, "epoch": 0.10448264239973036, "timestamp": 1786571507.0314584, "eval_loss": 2.3475193977355957, "eval_runtime": 15.3453, "eval_samples_per_second": 620.842, "eval_steps_per_second": 0.652, "train/total_time_seconds": 366.1316197067499, "train/time_per_step_avg": 0.23464392762631178, "train/epoch_time_elapsed": 921.689861100167, "train/estimated_remaining_minutes": 3.7400541798001337}
|
| 111 |
+
{"step": 1560, "epoch": 0.1051567239635996, "timestamp": 1786571509.876675, "loss": 2.335231971740723, "grad_norm": 0.388671875, "learning_rate": 0.0007, "train/total_time_seconds": 368.4845058917999, "train/time_per_step_avg": 0.23474177807569505, "train/epoch_time_elapsed": 924.5350807644427, "train/estimated_remaining_minutes": 3.7005922600244863}
|
| 112 |
+
{"step": 1580, "epoch": 0.10650488709133805, "timestamp": 1786571515.5551944, "loss": 2.3543283462524416, "grad_norm": 0.359375, "learning_rate": 0.0007, "train/total_time_seconds": 373.1850628070533, "train/time_per_step_avg": 0.23481694389134644, "train/epoch_time_elapsed": 930.2136003486812, "train/estimated_remaining_minutes": 3.621627191798408}
|
| 113 |
+
{"step": 1600, "epoch": 0.10785305021907651, "timestamp": 1786571521.2161312, "loss": 2.3237049102783205, "grad_norm": 0.39453125, "learning_rate": 0.0007, "train/total_time_seconds": 377.86905916780233, "train/time_per_step_avg": 0.23465531572699547, "train/epoch_time_elapsed": 935.8745375089347, "train/estimated_remaining_minutes": 3.542522429698147}
|
| 114 |
+
{"step": 1600, "epoch": 0.10785305021907651, "timestamp": 1786571536.5034702, "eval_loss": 2.3377676010131836, "eval_runtime": 15.2858, "eval_samples_per_second": 623.259, "eval_steps_per_second": 0.654, "train/total_time_seconds": 377.86905916780233, "train/time_per_step_avg": 0.23465531572699547, "train/epoch_time_elapsed": 951.1618747338653, "train/estimated_remaining_minutes": 3.542522429698147}
|
| 115 |
+
{"step": 1620, "epoch": 0.10920121334681497, "timestamp": 1786571542.4343424, "loss": 2.3196277618408203, "grad_norm": 0.357421875, "learning_rate": 0.0007, "train/total_time_seconds": 382.560747820884, "train/time_per_step_avg": 0.23470989920198917, "train/epoch_time_elapsed": 957.0927486978471, "train/estimated_remaining_minutes": 3.4635129432343406}
|
| 116 |
+
{"step": 1640, "epoch": 0.11054937647455342, "timestamp": 1786571548.0853047, "loss": 2.3308162689208984, "grad_norm": 0.341796875, "learning_rate": 0.0007, "train/total_time_seconds": 387.236578900367, "train/time_per_step_avg": 0.23453011337667704, "train/epoch_time_elapsed": 962.7437107637525, "train/estimated_remaining_minutes": 3.384384734291826}
|
| 117 |
+
{"step": 1650, "epoch": 0.11122345803842265, "timestamp": 1786571566.1831853, "eval_loss": 2.313627004623413, "eval_runtime": 15.276, "eval_samples_per_second": 623.659, "eval_steps_per_second": 0.655, "train/total_time_seconds": 389.57430966570973, "train/time_per_step_avg": 0.23442689958959817, "train/epoch_time_elapsed": 980.8415895439684, "train/estimated_remaining_minutes": 3.3448299314732655}
|
| 118 |
+
{"step": 1660, "epoch": 0.11189753960229187, "timestamp": 1786571569.0301538, "loss": 2.3104904174804686, "grad_norm": 0.353515625, "learning_rate": 0.0007, "train/total_time_seconds": 391.929935477674, "train/time_per_step_avg": 0.2344542958587408, "train/epoch_time_elapsed": 983.6885597929358, "train/estimated_remaining_minutes": 3.305433190775564}
|
| 119 |
+
{"step": 1680, "epoch": 0.11324570273003033, "timestamp": 1786571574.7093, "loss": 2.2955816268920897, "grad_norm": 0.353515625, "learning_rate": 0.0007, "train/total_time_seconds": 396.6305368691683, "train/time_per_step_avg": 0.23445474062114954, "train/epoch_time_elapsed": 989.3677054867148, "train/estimated_remaining_minutes": 3.2265579388166468}
|
| 120 |
+
{"step": 1700, "epoch": 0.11459386585776879, "timestamp": 1786571580.3731654, "loss": 2.3044401168823243, "grad_norm": 0.345703125, "learning_rate": 0.0007, "train/total_time_seconds": 401.3185408115387, "train/time_per_step_avg": 0.2344948164373636, "train/epoch_time_elapsed": 995.0315714627504, "train/estimated_remaining_minutes": 3.147596398521872}
|
| 121 |
+
{"step": 1700, "epoch": 0.11459386585776879, "timestamp": 1786571595.7150414, "eval_loss": 2.2976956367492676, "eval_runtime": 15.3404, "eval_samples_per_second": 621.039, "eval_steps_per_second": 0.652, "train/total_time_seconds": 401.3185408115387, "train/time_per_step_avg": 0.2344948164373636, "train/epoch_time_elapsed": 1010.373444840312, "train/estimated_remaining_minutes": 3.147596398521872}
|
| 122 |
+
{"step": 1720, "epoch": 0.11594202898550725, "timestamp": 1786571601.9694974, "loss": 2.280081939697266, "grad_norm": 0.373046875, "learning_rate": 0.0007, "train/total_time_seconds": 406.30108135566115, "train/time_per_step_avg": 0.23740333534777164, "train/epoch_time_elapsed": 1016.6279033198953, "train/estimated_remaining_minutes": 3.0708802660602297}
|
| 123 |
+
{"step": 1740, "epoch": 0.1172901921132457, "timestamp": 1786571607.65943, "loss": 2.269576644897461, "grad_norm": 0.373046875, "learning_rate": 0.0007, "train/total_time_seconds": 411.01078770309687, "train/time_per_step_avg": 0.23774208802729846, "train/epoch_time_elapsed": 1022.31783612445, "train/estimated_remaining_minutes": 2.992032554160475}
|
| 124 |
+
{"step": 1750, "epoch": 0.11796427367711493, "timestamp": 1786571625.6988165, "eval_loss": 2.28181791305542, "eval_runtime": 15.2137, "eval_samples_per_second": 626.211, "eval_steps_per_second": 0.657, "train/total_time_seconds": 413.350578032434, "train/time_per_step_avg": 0.2377626836672425, "train/epoch_time_elapsed": 1040.3572210036218, "train/estimated_remaining_minutes": 2.9525041288031}
|
| 125 |
+
{"step": 1760, "epoch": 0.11863835524098416, "timestamp": 1786571628.5253522, "loss": 2.273081398010254, "grad_norm": 0.359375, "learning_rate": 0.0007, "train/total_time_seconds": 415.68846390768886, "train/time_per_step_avg": 0.2375852843001485, "train/epoch_time_elapsed": 1043.1837587207556, "train/estimated_remaining_minutes": 2.912968402383426}
|
| 126 |
+
{"step": 1780, "epoch": 0.11998651836872262, "timestamp": 1786571634.1754513, "loss": 2.2877025604248047, "grad_norm": 0.345703125, "learning_rate": 0.0007, "train/total_time_seconds": 420.3682208955288, "train/time_per_step_avg": 0.23737684026360512, "train/epoch_time_elapsed": 1048.833857037127, "train/estimated_remaining_minutes": 2.833943062217048}
|
| 127 |
+
{"step": 1800, "epoch": 0.12133468149646107, "timestamp": 1786571639.8574946, "loss": 2.2828989028930664, "grad_norm": 0.3671875, "learning_rate": 0.0007, "train/total_time_seconds": 425.08155478164554, "train/time_per_step_avg": 0.2376301397010684, "train/epoch_time_elapsed": 1054.5159012190998, "train/estimated_remaining_minutes": 2.7551582254365914}
|
| 128 |
+
{"step": 1800, "epoch": 0.12133468149646107, "timestamp": 1786571655.0203607, "eval_loss": 2.2671334743499756, "eval_runtime": 15.1615, "eval_samples_per_second": 628.369, "eval_steps_per_second": 0.66, "train/total_time_seconds": 425.08155478164554, "train/time_per_step_avg": 0.2376301397010684, "train/epoch_time_elapsed": 1069.6787635795772, "train/estimated_remaining_minutes": 2.7551582254365914}
|
| 129 |
+
{"step": 1820, "epoch": 0.12268284462419952, "timestamp": 1786571660.9795668, "loss": 2.2743486404418944, "grad_norm": 0.388671875, "learning_rate": 0.0007, "train/total_time_seconds": 429.7808753103018, "train/time_per_step_avg": 0.23479793954640626, "train/epoch_time_elapsed": 1075.637973241508, "train/estimated_remaining_minutes": 2.6762911649359453}
|
| 130 |
+
{"step": 1840, "epoch": 0.12403100775193798, "timestamp": 1786571666.6678298, "loss": 2.2429489135742187, "grad_norm": 0.3671875, "learning_rate": 0.0007, "train/total_time_seconds": 434.4591994471848, "train/time_per_step_avg": 0.23448411744087935, "train/epoch_time_elapsed": 1081.3262354955077, "train/estimated_remaining_minutes": 2.5973104314777355}
|
zain/Activation/wandb/debug-internal.log
CHANGED
|
@@ -109,3 +109,45 @@
|
|
| 109 |
{"time":"2026-08-12T21:48:55.692043729Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
|
| 110 |
{"time":"2026-08-12T21:49:10.573014113Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":88,"history_lines":2,"events_offset":100,"events_lines":2,"console_offset":146,"console_lines":1}
|
| 111 |
{"time":"2026-08-12T21:49:10.721360411Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 109 |
{"time":"2026-08-12T21:48:55.692043729Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
|
| 110 |
{"time":"2026-08-12T21:49:10.573014113Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":88,"history_lines":2,"events_offset":100,"events_lines":2,"console_offset":146,"console_lines":1}
|
| 111 |
{"time":"2026-08-12T21:49:10.721360411Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
|
| 112 |
+
{"time":"2026-08-12T21:49:25.57253842Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":90,"history_lines":2,"events_offset":102,"events_lines":2,"console_offset":146,"console_lines":1}
|
| 113 |
+
{"time":"2026-08-12T21:49:25.675579246Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
|
| 114 |
+
{"time":"2026-08-12T21:49:40.5724696Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":92,"history_lines":1,"events_offset":104,"events_lines":2,"console_offset":149,"console_lines":13}
|
| 115 |
+
{"time":"2026-08-12T21:49:40.726375879Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
|
| 116 |
+
{"time":"2026-08-12T21:49:55.572625084Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":93,"history_lines":2,"events_offset":106,"events_lines":2,"console_offset":158,"console_lines":1}
|
| 117 |
+
{"time":"2026-08-12T21:49:55.73298204Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
|
| 118 |
+
{"time":"2026-08-12T21:50:10.572523069Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":95,"history_lines":2,"events_offset":108,"events_lines":2,"console_offset":158,"console_lines":1}
|
| 119 |
+
{"time":"2026-08-12T21:50:10.733807706Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
|
| 120 |
+
{"time":"2026-08-12T21:50:25.573196838Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":97,"history_lines":2,"events_offset":110,"events_lines":2,"console_offset":158,"console_lines":1}
|
| 121 |
+
{"time":"2026-08-12T21:50:25.676239826Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
|
| 122 |
+
{"time":"2026-08-12T21:50:40.572938093Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":99,"history_lines":1,"events_offset":112,"events_lines":2,"console_offset":161,"console_lines":13}
|
| 123 |
+
{"time":"2026-08-12T21:50:40.698942691Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
|
| 124 |
+
{"time":"2026-08-12T21:50:55.572436606Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":100,"history_lines":2,"events_offset":114,"events_lines":2,"console_offset":170,"console_lines":1}
|
| 125 |
+
{"time":"2026-08-12T21:50:55.702809045Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
|
| 126 |
+
{"time":"2026-08-12T21:51:10.572756971Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":102,"history_lines":2,"events_offset":116,"events_lines":2,"console_offset":170,"console_lines":1}
|
| 127 |
+
{"time":"2026-08-12T21:51:10.75225332Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
|
| 128 |
+
{"time":"2026-08-12T21:51:25.572481917Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":104,"history_lines":2,"events_offset":118,"events_lines":2,"console_offset":170,"console_lines":1}
|
| 129 |
+
{"time":"2026-08-12T21:51:25.696050752Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
|
| 130 |
+
{"time":"2026-08-12T21:51:40.572475245Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":106,"history_lines":1,"events_offset":120,"events_lines":2,"console_offset":173,"console_lines":13}
|
| 131 |
+
{"time":"2026-08-12T21:51:40.708689871Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
|
| 132 |
+
{"time":"2026-08-12T21:51:55.572777206Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":107,"history_lines":3,"events_offset":122,"events_lines":2,"console_offset":182,"console_lines":1}
|
| 133 |
+
{"time":"2026-08-12T21:51:55.692182287Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
|
| 134 |
+
{"time":"2026-08-12T21:52:10.572760508Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":110,"history_lines":1,"events_offset":124,"events_lines":2,"console_offset":182,"console_lines":1}
|
| 135 |
+
{"time":"2026-08-12T21:52:10.71057509Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
|
| 136 |
+
{"time":"2026-08-12T21:52:25.573164599Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":111,"history_lines":2,"events_offset":126,"events_lines":2,"console_offset":182,"console_lines":1}
|
| 137 |
+
{"time":"2026-08-12T21:52:25.697136237Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
|
| 138 |
+
{"time":"2026-08-12T21:52:40.572773335Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":113,"history_lines":1,"events_offset":128,"events_lines":2,"console_offset":185,"console_lines":13}
|
| 139 |
+
{"time":"2026-08-12T21:52:40.68286574Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
|
| 140 |
+
{"time":"2026-08-12T21:52:55.573174111Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":114,"history_lines":3,"events_offset":130,"events_lines":2,"console_offset":194,"console_lines":1}
|
| 141 |
+
{"time":"2026-08-12T21:52:55.698734987Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
|
| 142 |
+
{"time":"2026-08-12T21:53:10.572927957Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":117,"history_lines":1,"events_offset":132,"events_lines":2,"console_offset":194,"console_lines":1}
|
| 143 |
+
{"time":"2026-08-12T21:53:10.695876795Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
|
| 144 |
+
{"time":"2026-08-12T21:53:25.572552777Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":118,"history_lines":2,"events_offset":134,"events_lines":2,"console_offset":194,"console_lines":1}
|
| 145 |
+
{"time":"2026-08-12T21:53:25.694093078Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
|
| 146 |
+
{"time":"2026-08-12T21:53:40.572777233Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":120,"history_lines":1,"events_offset":136,"events_lines":2,"console_offset":197,"console_lines":13}
|
| 147 |
+
{"time":"2026-08-12T21:53:40.677283746Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
|
| 148 |
+
{"time":"2026-08-12T21:53:55.572683833Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":121,"history_lines":3,"events_offset":138,"events_lines":2,"console_offset":206,"console_lines":1}
|
| 149 |
+
{"time":"2026-08-12T21:53:55.718391882Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
|
| 150 |
+
{"time":"2026-08-12T21:54:10.57330427Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":124,"history_lines":1,"events_offset":140,"events_lines":2,"console_offset":206,"console_lines":1}
|
| 151 |
+
{"time":"2026-08-12T21:54:10.755710662Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
|
| 152 |
+
{"time":"2026-08-12T21:54:25.573166176Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":125,"history_lines":2,"events_offset":142,"events_lines":2,"console_offset":206,"console_lines":1}
|
| 153 |
+
{"time":"2026-08-12T21:54:25.691084195Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
|
zain/Activation/wandb/run-20260812_213624-1sryr5l2/files/output.log
CHANGED
|
@@ -156,4 +156,65 @@ Writing model shards: 100%|██████████| 1/1 [00:00<00:00, 17.
|
|
| 156 |
- If you are the owner of the model architecture code, please modify your model class such that it inherits from `GenerationMixin` (after `PreTrainedModel`, otherwise you'll get an exception).
|
| 157 |
- If you are not the owner of the model architecture class, please contact the model code owner to update it.
|
| 158 |
Writing model shards: 100%|██████████| 1/1 [00:00<00:00, 17.42it/s]
|
| 159 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 156 |
- If you are the owner of the model architecture code, please modify your model class such that it inherits from `GenerationMixin` (after `PreTrainedModel`, otherwise you'll get an exception).
|
| 157 |
- If you are not the owner of the model architecture class, please contact the model code owner to update it.
|
| 158 |
Writing model shards: 100%|██████████| 1/1 [00:00<00:00, 17.42it/s]
|
| 159 |
+
56%|█████▌ | 1400/2500 [13:53<05:14, 3.50i[transformers] TinyLlamaForCausalLM has generative capabilities, as `prepare_inputs_for_generation` is explicitly defined. However, it doesn't directly inherit from `GenerationMixin`. From 👉v4.50👈 onwards, `PreTrainedModel` will NOT inherit from `GenerationMixin`, and this model will lose the ability to call `generate` and other related functions.
|
| 160 |
+
{'loss': '2.453', 'grad_norm': '0.3633', 'learning_rate': '0.0007', 'epoch': '0.08898', 'train/total_time_seconds': '312.1', 'train/time_per_step_avg': '0.2379', 'train/epoch_time_elapsed': '778.8', 'train/estimated_remaining_minutes': '4.649'}
|
| 161 |
+
{'loss': '2.44', 'grad_norm': '0.4043', 'learning_rate': '0.0007', 'epoch': '0.09033', 'train/total_time_seconds': '316.8', 'train/time_per_step_avg': '0.2382', 'train/epoch_time_elapsed': '784.5', 'train/estimated_remaining_minutes': '4.571'}
|
| 162 |
+
- If you're using `trust_remote_code=True`, you can get rid of this warning by loading the model with an auto class. See https://huggingface.co/docs/transformers/en/model_doc/auto#auto-classes
|
| 163 |
+
{'eval_loss': '2.437', 'eval_runtime': '15.3', 'eval_samples_per_second': '622.6', 'eval_steps_per_second': '0.653', 'epoch': '0.091', 'train/total_time_seconds': '319.2', 'train/time_per_step_avg': '0.2383', 'train/epoch_time_elapsed': '802.7', 'train/estimated_remaining_minutes': '4.531'}
|
| 164 |
+
{'loss': '2.44', 'grad_norm': '0.4121', 'learning_rate': '0.0007', 'epoch': '0.09168', 'train/total_time_seconds': '321.5', 'train/time_per_step_avg': '0.2384', 'train/epoch_time_elapsed': '805.5', 'train/estimated_remaining_minutes': '4.492'}
|
| 165 |
+
{'loss': '2.416', 'grad_norm': '0.3535', 'learning_rate': '0.0007', 'epoch': '0.09302', 'train/total_time_seconds': '326.2', 'train/time_per_step_avg': '0.2386', 'train/epoch_time_elapsed': '811.2', 'train/estimated_remaining_minutes': '4.413'}
|
| 166 |
+
{'loss': '2.407', 'grad_norm': '0.3457', 'learning_rate': '0.0007', 'epoch': '0.09437', 'train/total_time_seconds': '330.9', 'train/time_per_step_avg': '0.2385', 'train/epoch_time_elapsed': '816.9', 'train/estimated_remaining_minutes': '4.334'}
|
| 167 |
+
{'eval_loss': '2.41', 'eval_runtime': '15.36', 'eval_samples_per_second': '620.4', 'eval_steps_per_second': '0.651', 'epoch': '0.09437', 'train/total_time_seconds': '330.9', 'train/time_per_step_avg': '0.2385', 'train/epoch_time_elapsed': '832.3', 'train/estimated_remaining_minutes': '4.334'}
|
| 168 |
+
- If you are the owner of the model architecture code, please modify your model class such that it inherits from `GenerationMixin` (after `PreTrainedModel`, otherwise you'll get an exception).
|
| 169 |
+
- If you are not the owner of the model architecture class, please contact the model code owner to update it.
|
| 170 |
+
Writing model shards: 100%|██████████| 1/1 [00:00<00:00, 18.81it/s]
|
| 171 |
+
60%|██████ | 1500/2500 [14:52<04:43, 3.53i[transformers] TinyLlamaForCausalLM has generative capabilities, as `prepare_inputs_for_generation` is explicitly defined. However, it doesn't directly inherit from `GenerationMixin`. From 👉v4.50👈 onwards, `PreTrainedModel` will NOT inherit from `GenerationMixin`, and this model will lose the ability to call `generate` and other related functions.
|
| 172 |
+
{'loss': '2.416', 'grad_norm': '0.3633', 'learning_rate': '0.0007', 'epoch': '0.09572', 'train/total_time_seconds': '335.6', 'train/time_per_step_avg': '0.2354', 'train/epoch_time_elapsed': '838.2', 'train/estimated_remaining_minutes': '4.254'}
|
| 173 |
+
{'loss': '2.393', 'grad_norm': '0.3633', 'learning_rate': '0.0007', 'epoch': '0.09707', 'train/total_time_seconds': '340.3', 'train/time_per_step_avg': '0.2352', 'train/epoch_time_elapsed': '843.9', 'train/estimated_remaining_minutes': '4.175'}
|
| 174 |
+
- If you're using `trust_remote_code=True`, you can get rid of this warning by loading the model with an auto class. See https://huggingface.co/docs/transformers/en/model_doc/auto#auto-classes
|
| 175 |
+
{'eval_loss': '2.394', 'eval_runtime': '15.36', 'eval_samples_per_second': '620.4', 'eval_steps_per_second': '0.651', 'epoch': '0.09774', 'train/total_time_seconds': '342.7', 'train/time_per_step_avg': '0.2351', 'train/epoch_time_elapsed': '862.1', 'train/estimated_remaining_minutes': '4.136'}
|
| 176 |
+
{'loss': '2.382', 'grad_norm': '0.3848', 'learning_rate': '0.0007', 'epoch': '0.09842', 'train/total_time_seconds': '345', 'train/time_per_step_avg': '0.2351', 'train/epoch_time_elapsed': '864.9', 'train/estimated_remaining_minutes': '4.096'}
|
| 177 |
+
{'loss': '2.369', 'grad_norm': '0.3398', 'learning_rate': '0.0007', 'epoch': '0.09976', 'train/total_time_seconds': '349.7', 'train/time_per_step_avg': '0.2348', 'train/epoch_time_elapsed': '870.6', 'train/estimated_remaining_minutes': '4.017'}
|
| 178 |
+
{'loss': '2.355', 'grad_norm': '0.3867', 'learning_rate': '0.0007', 'epoch': '0.1011', 'train/total_time_seconds': '354.4', 'train/time_per_step_avg': '0.2348', 'train/epoch_time_elapsed': '876.3', 'train/estimated_remaining_minutes': '3.938'}
|
| 179 |
+
{'eval_loss': '2.368', 'eval_runtime': '15.33', 'eval_samples_per_second': '621.3', 'eval_steps_per_second': '0.652', 'epoch': '0.1011', 'train/total_time_seconds': '354.4', 'train/time_per_step_avg': '0.2348', 'train/epoch_time_elapsed': '891.6', 'train/estimated_remaining_minutes': '3.938'}
|
| 180 |
+
- If you are the owner of the model architecture code, please modify your model class such that it inherits from `GenerationMixin` (after `PreTrainedModel`, otherwise you'll get an exception).
|
| 181 |
+
- If you are not the owner of the model architecture class, please contact the model code owner to update it.
|
| 182 |
+
Writing model shards: 100%|██████████| 1/1 [00:00<00:00, 16.09it/s]
|
| 183 |
+
64%|██████▍ | 1600/2500 [15:52<04:14, 3.54i[transformers] TinyLlamaForCausalLM has generative capabilities, as `prepare_inputs_for_generation` is explicitly defined. However, it doesn't directly inherit from `GenerationMixin`. From 👉v4.50👈 onwards, `PreTrainedModel` will NOT inherit from `GenerationMixin`, and this model will lose the ability to call `generate` and other related functions.
|
| 184 |
+
{'loss': '2.357', 'grad_norm': '0.3926', 'learning_rate': '0.0007', 'epoch': '0.1025', 'train/total_time_seconds': '359.1', 'train/time_per_step_avg': '0.2348', 'train/epoch_time_elapsed': '897.8', 'train/estimated_remaining_minutes': '3.859'}
|
| 185 |
+
{'loss': '2.346', 'grad_norm': '0.3555', 'learning_rate': '0.0007', 'epoch': '0.1038', 'train/total_time_seconds': '363.8', 'train/time_per_step_avg': '0.2347', 'train/epoch_time_elapsed': '903.5', 'train/estimated_remaining_minutes': '3.78'}
|
| 186 |
+
- If you're using `trust_remote_code=True`, you can get rid of this warning by loading the model with an auto class. See https://huggingface.co/docs/transformers/en/model_doc/auto#auto-classes
|
| 187 |
+
{'eval_loss': '2.348', 'eval_runtime': '15.35', 'eval_samples_per_second': '620.8', 'eval_steps_per_second': '0.652', 'epoch': '0.1045', 'train/total_time_seconds': '366.1', 'train/time_per_step_avg': '0.2346', 'train/epoch_time_elapsed': '921.7', 'train/estimated_remaining_minutes': '3.74'}
|
| 188 |
+
{'loss': '2.335', 'grad_norm': '0.3887', 'learning_rate': '0.0007', 'epoch': '0.1052', 'train/total_time_seconds': '368.5', 'train/time_per_step_avg': '0.2347', 'train/epoch_time_elapsed': '924.5', 'train/estimated_remaining_minutes': '3.701'}
|
| 189 |
+
{'loss': '2.354', 'grad_norm': '0.3594', 'learning_rate': '0.0007', 'epoch': '0.1065', 'train/total_time_seconds': '373.2', 'train/time_per_step_avg': '0.2348', 'train/epoch_time_elapsed': '930.2', 'train/estimated_remaining_minutes': '3.622'}
|
| 190 |
+
{'loss': '2.324', 'grad_norm': '0.3945', 'learning_rate': '0.0007', 'epoch': '0.1079', 'train/total_time_seconds': '377.9', 'train/time_per_step_avg': '0.2347', 'train/epoch_time_elapsed': '935.9', 'train/estimated_remaining_minutes': '3.543'}
|
| 191 |
+
{'eval_loss': '2.338', 'eval_runtime': '15.29', 'eval_samples_per_second': '623.3', 'eval_steps_per_second': '0.654', 'epoch': '0.1079', 'train/total_time_seconds': '377.9', 'train/time_per_step_avg': '0.2347', 'train/epoch_time_elapsed': '951.2', 'train/estimated_remaining_minutes': '3.543'}
|
| 192 |
+
- If you are the owner of the model architecture code, please modify your model class such that it inherits from `GenerationMixin` (after `PreTrainedModel`, otherwise you'll get an exception).
|
| 193 |
+
- If you are not the owner of the model architecture class, please contact the model code owner to update it.
|
| 194 |
+
Writing model shards: 100%|██████████| 1/1 [00:00<00:00, 17.93it/s]
|
| 195 |
+
68%|██████▊ | 1700/2500 [16:51<03:45, 3.54i[transformers] TinyLlamaForCausalLM has generative capabilities, as `prepare_inputs_for_generation` is explicitly defined. However, it doesn't directly inherit from `GenerationMixin`. From 👉v4.50👈 onwards, `PreTrainedModel` will NOT inherit from `GenerationMixin`, and this model will lose the ability to call `generate` and other related functions.
|
| 196 |
+
{'loss': '2.32', 'grad_norm': '0.3574', 'learning_rate': '0.0007', 'epoch': '0.1092', 'train/total_time_seconds': '382.6', 'train/time_per_step_avg': '0.2347', 'train/epoch_time_elapsed': '957.1', 'train/estimated_remaining_minutes': '3.464'}
|
| 197 |
+
{'loss': '2.331', 'grad_norm': '0.3418', 'learning_rate': '0.0007', 'epoch': '0.1105', 'train/total_time_seconds': '387.2', 'train/time_per_step_avg': '0.2345', 'train/epoch_time_elapsed': '962.7', 'train/estimated_remaining_minutes': '3.384'}
|
| 198 |
+
- If you're using `trust_remote_code=True`, you can get rid of this warning by loading the model with an auto class. See https://huggingface.co/docs/transformers/en/model_doc/auto#auto-classes
|
| 199 |
+
{'eval_loss': '2.314', 'eval_runtime': '15.28', 'eval_samples_per_second': '623.7', 'eval_steps_per_second': '0.655', 'epoch': '0.1112', 'train/total_time_seconds': '389.6', 'train/time_per_step_avg': '0.2344', 'train/epoch_time_elapsed': '980.8', 'train/estimated_remaining_minutes': '3.345'}
|
| 200 |
+
{'loss': '2.31', 'grad_norm': '0.3535', 'learning_rate': '0.0007', 'epoch': '0.1119', 'train/total_time_seconds': '391.9', 'train/time_per_step_avg': '0.2345', 'train/epoch_time_elapsed': '983.7', 'train/estimated_remaining_minutes': '3.305'}
|
| 201 |
+
{'loss': '2.296', 'grad_norm': '0.3535', 'learning_rate': '0.0007', 'epoch': '0.1132', 'train/total_time_seconds': '396.6', 'train/time_per_step_avg': '0.2345', 'train/epoch_time_elapsed': '989.4', 'train/estimated_remaining_minutes': '3.227'}
|
| 202 |
+
{'loss': '2.304', 'grad_norm': '0.3457', 'learning_rate': '0.0007', 'epoch': '0.1146', 'train/total_time_seconds': '401.3', 'train/time_per_step_avg': '0.2345', 'train/epoch_time_elapsed': '995', 'train/estimated_remaining_minutes': '3.148'}
|
| 203 |
+
{'eval_loss': '2.298', 'eval_runtime': '15.34', 'eval_samples_per_second': '621', 'eval_steps_per_second': '0.652', 'epoch': '0.1146', 'train/total_time_seconds': '401.3', 'train/time_per_step_avg': '0.2345', 'train/epoch_time_elapsed': '1010', 'train/estimated_remaining_minutes': '3.148'}
|
| 204 |
+
- If you are the owner of the model architecture code, please modify your model class such that it inherits from `GenerationMixin` (after `PreTrainedModel`, otherwise you'll get an exception).
|
| 205 |
+
- If you are not the owner of the model architecture class, please contact the model code owner to update it.
|
| 206 |
+
Writing model shards: 100%|██████████| 1/1 [00:00<00:00, 12.85it/s]
|
| 207 |
+
72%|███████▏ | 1800/2500 [17:50<03:17, 3.54i[transformers] TinyLlamaForCausalLM has generative capabilities, as `prepare_inputs_for_generation` is explicitly defined. However, it doesn't directly inherit from `GenerationMixin`. From 👉v4.50👈 onwards, `PreTrainedModel` will NOT inherit from `GenerationMixin`, and this model will lose the ability to call `generate` and other related functions.
|
| 208 |
+
{'loss': '2.28', 'grad_norm': '0.373', 'learning_rate': '0.0007', 'epoch': '0.1159', 'train/total_time_seconds': '406.3', 'train/time_per_step_avg': '0.2374', 'train/epoch_time_elapsed': '1017', 'train/estimated_remaining_minutes': '3.071'}
|
| 209 |
+
{'loss': '2.27', 'grad_norm': '0.373', 'learning_rate': '0.0007', 'epoch': '0.1173', 'train/total_time_seconds': '411', 'train/time_per_step_avg': '0.2377', 'train/epoch_time_elapsed': '1022', 'train/estimated_remaining_minutes': '2.992'}
|
| 210 |
+
- If you're using `trust_remote_code=True`, you can get rid of this warning by loading the model with an auto class. See https://huggingface.co/docs/transformers/en/model_doc/auto#auto-classes
|
| 211 |
+
{'eval_loss': '2.282', 'eval_runtime': '15.21', 'eval_samples_per_second': '626.2', 'eval_steps_per_second': '0.657', 'epoch': '0.118', 'train/total_time_seconds': '413.4', 'train/time_per_step_avg': '0.2378', 'train/epoch_time_elapsed': '1040', 'train/estimated_remaining_minutes': '2.953'}
|
| 212 |
+
{'loss': '2.273', 'grad_norm': '0.3594', 'learning_rate': '0.0007', 'epoch': '0.1186', 'train/total_time_seconds': '415.7', 'train/time_per_step_avg': '0.2376', 'train/epoch_time_elapsed': '1043', 'train/estimated_remaining_minutes': '2.913'}
|
| 213 |
+
{'loss': '2.288', 'grad_norm': '0.3457', 'learning_rate': '0.0007', 'epoch': '0.12', 'train/total_time_seconds': '420.4', 'train/time_per_step_avg': '0.2374', 'train/epoch_time_elapsed': '1049', 'train/estimated_remaining_minutes': '2.834'}
|
| 214 |
+
{'loss': '2.283', 'grad_norm': '0.3672', 'learning_rate': '0.0007', 'epoch': '0.1213', 'train/total_time_seconds': '425.1', 'train/time_per_step_avg': '0.2376', 'train/epoch_time_elapsed': '1055', 'train/estimated_remaining_minutes': '2.755'}
|
| 215 |
+
{'eval_loss': '2.267', 'eval_runtime': '15.16', 'eval_samples_per_second': '628.4', 'eval_steps_per_second': '0.66', 'epoch': '0.1213', 'train/total_time_seconds': '425.1', 'train/time_per_step_avg': '0.2376', 'train/epoch_time_elapsed': '1070', 'train/estimated_remaining_minutes': '2.755'}
|
| 216 |
+
- If you are the owner of the model architecture code, please modify your model class such that it inherits from `GenerationMixin` (after `PreTrainedModel`, otherwise you'll get an exception).
|
| 217 |
+
- If you are not the owner of the model architecture class, please contact the model code owner to update it.
|
| 218 |
+
Writing model shards: 100%|██████████| 1/1 [00:00<00:00, 19.28it/s]
|
| 219 |
+
73%|███████▎ | 1836/2500 [18:01<03:10, 3.49it/s], ?it/s]
|
| 220 |
+
{'loss': '2.274', 'grad_norm': '0.3887', 'learning_rate': '0.0007', 'epoch': '0.1227', 'train/total_time_seconds': '429.8', 'train/time_per_step_avg': '0.2348', 'train/epoch_time_elapsed': '1076', 'train/estimated_remaining_minutes': '2.676'}
|
zain/Activation/wandb/run-20260812_213624-1sryr5l2/logs/debug-internal.log
CHANGED
|
@@ -109,3 +109,45 @@
|
|
| 109 |
{"time":"2026-08-12T21:48:55.692043729Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
|
| 110 |
{"time":"2026-08-12T21:49:10.573014113Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":88,"history_lines":2,"events_offset":100,"events_lines":2,"console_offset":146,"console_lines":1}
|
| 111 |
{"time":"2026-08-12T21:49:10.721360411Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 109 |
{"time":"2026-08-12T21:48:55.692043729Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
|
| 110 |
{"time":"2026-08-12T21:49:10.573014113Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":88,"history_lines":2,"events_offset":100,"events_lines":2,"console_offset":146,"console_lines":1}
|
| 111 |
{"time":"2026-08-12T21:49:10.721360411Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
|
| 112 |
+
{"time":"2026-08-12T21:49:25.57253842Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":90,"history_lines":2,"events_offset":102,"events_lines":2,"console_offset":146,"console_lines":1}
|
| 113 |
+
{"time":"2026-08-12T21:49:25.675579246Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
|
| 114 |
+
{"time":"2026-08-12T21:49:40.5724696Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":92,"history_lines":1,"events_offset":104,"events_lines":2,"console_offset":149,"console_lines":13}
|
| 115 |
+
{"time":"2026-08-12T21:49:40.726375879Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
|
| 116 |
+
{"time":"2026-08-12T21:49:55.572625084Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":93,"history_lines":2,"events_offset":106,"events_lines":2,"console_offset":158,"console_lines":1}
|
| 117 |
+
{"time":"2026-08-12T21:49:55.73298204Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
|
| 118 |
+
{"time":"2026-08-12T21:50:10.572523069Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":95,"history_lines":2,"events_offset":108,"events_lines":2,"console_offset":158,"console_lines":1}
|
| 119 |
+
{"time":"2026-08-12T21:50:10.733807706Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
|
| 120 |
+
{"time":"2026-08-12T21:50:25.573196838Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":97,"history_lines":2,"events_offset":110,"events_lines":2,"console_offset":158,"console_lines":1}
|
| 121 |
+
{"time":"2026-08-12T21:50:25.676239826Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
|
| 122 |
+
{"time":"2026-08-12T21:50:40.572938093Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":99,"history_lines":1,"events_offset":112,"events_lines":2,"console_offset":161,"console_lines":13}
|
| 123 |
+
{"time":"2026-08-12T21:50:40.698942691Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
|
| 124 |
+
{"time":"2026-08-12T21:50:55.572436606Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":100,"history_lines":2,"events_offset":114,"events_lines":2,"console_offset":170,"console_lines":1}
|
| 125 |
+
{"time":"2026-08-12T21:50:55.702809045Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
|
| 126 |
+
{"time":"2026-08-12T21:51:10.572756971Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":102,"history_lines":2,"events_offset":116,"events_lines":2,"console_offset":170,"console_lines":1}
|
| 127 |
+
{"time":"2026-08-12T21:51:10.75225332Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
|
| 128 |
+
{"time":"2026-08-12T21:51:25.572481917Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":104,"history_lines":2,"events_offset":118,"events_lines":2,"console_offset":170,"console_lines":1}
|
| 129 |
+
{"time":"2026-08-12T21:51:25.696050752Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
|
| 130 |
+
{"time":"2026-08-12T21:51:40.572475245Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":106,"history_lines":1,"events_offset":120,"events_lines":2,"console_offset":173,"console_lines":13}
|
| 131 |
+
{"time":"2026-08-12T21:51:40.708689871Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
|
| 132 |
+
{"time":"2026-08-12T21:51:55.572777206Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":107,"history_lines":3,"events_offset":122,"events_lines":2,"console_offset":182,"console_lines":1}
|
| 133 |
+
{"time":"2026-08-12T21:51:55.692182287Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
|
| 134 |
+
{"time":"2026-08-12T21:52:10.572760508Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":110,"history_lines":1,"events_offset":124,"events_lines":2,"console_offset":182,"console_lines":1}
|
| 135 |
+
{"time":"2026-08-12T21:52:10.71057509Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
|
| 136 |
+
{"time":"2026-08-12T21:52:25.573164599Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":111,"history_lines":2,"events_offset":126,"events_lines":2,"console_offset":182,"console_lines":1}
|
| 137 |
+
{"time":"2026-08-12T21:52:25.697136237Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
|
| 138 |
+
{"time":"2026-08-12T21:52:40.572773335Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":113,"history_lines":1,"events_offset":128,"events_lines":2,"console_offset":185,"console_lines":13}
|
| 139 |
+
{"time":"2026-08-12T21:52:40.68286574Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
|
| 140 |
+
{"time":"2026-08-12T21:52:55.573174111Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":114,"history_lines":3,"events_offset":130,"events_lines":2,"console_offset":194,"console_lines":1}
|
| 141 |
+
{"time":"2026-08-12T21:52:55.698734987Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
|
| 142 |
+
{"time":"2026-08-12T21:53:10.572927957Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":117,"history_lines":1,"events_offset":132,"events_lines":2,"console_offset":194,"console_lines":1}
|
| 143 |
+
{"time":"2026-08-12T21:53:10.695876795Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
|
| 144 |
+
{"time":"2026-08-12T21:53:25.572552777Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":118,"history_lines":2,"events_offset":134,"events_lines":2,"console_offset":194,"console_lines":1}
|
| 145 |
+
{"time":"2026-08-12T21:53:25.694093078Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
|
| 146 |
+
{"time":"2026-08-12T21:53:40.572777233Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":120,"history_lines":1,"events_offset":136,"events_lines":2,"console_offset":197,"console_lines":13}
|
| 147 |
+
{"time":"2026-08-12T21:53:40.677283746Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
|
| 148 |
+
{"time":"2026-08-12T21:53:55.572683833Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":121,"history_lines":3,"events_offset":138,"events_lines":2,"console_offset":206,"console_lines":1}
|
| 149 |
+
{"time":"2026-08-12T21:53:55.718391882Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
|
| 150 |
+
{"time":"2026-08-12T21:54:10.57330427Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":124,"history_lines":1,"events_offset":140,"events_lines":2,"console_offset":206,"console_lines":1}
|
| 151 |
+
{"time":"2026-08-12T21:54:10.755710662Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
|
| 152 |
+
{"time":"2026-08-12T21:54:25.573166176Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":125,"history_lines":2,"events_offset":142,"events_lines":2,"console_offset":206,"console_lines":1}
|
| 153 |
+
{"time":"2026-08-12T21:54:25.691084195Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
|
zain/Activation/wandb/run-20260812_213624-1sryr5l2/run-1sryr5l2.wandb
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
-
size
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:50a415efa35953c6c0710234234aeeac59334fd3c45644ec5437de3756a4bf47
|
| 3 |
+
size 983040
|