Upload folder using huggingface_hub
Browse files- added_tokens.json +5 -0
- checkpoint-4218/config.json +32 -0
- checkpoint-4218/generation_config.json +7 -0
- checkpoint-4218/model.safetensors +3 -0
- checkpoint-4218/optimizer.pt +3 -0
- checkpoint-4218/rng_state.pth +3 -0
- checkpoint-4218/scheduler.pt +3 -0
- checkpoint-4218/trainer_state.json +2980 -0
- checkpoint-4218/training_args.bin +3 -0
- config.json +32 -0
- generation_config.json +7 -0
- model.safetensors +3 -0
- special_tokens_map.json +44 -0
- tokenizer.json +0 -0
- tokenizer.model +3 -0
- tokenizer_config.json +69 -0
- training_args.bin +3 -0
added_tokens.json
ADDED
|
@@ -0,0 +1,5 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"[BOS]": 32001,
|
| 3 |
+
"[EOS]": 32002,
|
| 4 |
+
"[PAD]": 32000
|
| 5 |
+
}
|
checkpoint-4218/config.json
ADDED
|
@@ -0,0 +1,32 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "appvoid/arco-mini-run-5",
|
| 3 |
+
"architectures": [
|
| 4 |
+
"LlamaForCausalLM"
|
| 5 |
+
],
|
| 6 |
+
"attention_bias": false,
|
| 7 |
+
"attention_dropout": 0.0,
|
| 8 |
+
"bos_token_id": 1,
|
| 9 |
+
"eos_token_id": 2,
|
| 10 |
+
"head_dim": 96,
|
| 11 |
+
"hidden_act": "silu",
|
| 12 |
+
"hidden_size": 1536,
|
| 13 |
+
"initializer_range": 0.02,
|
| 14 |
+
"intermediate_size": 3072,
|
| 15 |
+
"max_position_embeddings": 2048,
|
| 16 |
+
"mlp_bias": false,
|
| 17 |
+
"model_type": "llama",
|
| 18 |
+
"num_attention_heads": 8,
|
| 19 |
+
"num_hidden_layers": 8,
|
| 20 |
+
"num_key_value_heads": 8,
|
| 21 |
+
"pad_token_id": 0,
|
| 22 |
+
"pretraining_tp": 1,
|
| 23 |
+
"rms_norm_eps": 1e-05,
|
| 24 |
+
"rope_scaling": null,
|
| 25 |
+
"rope_theta": 100000,
|
| 26 |
+
"sliding_window": null,
|
| 27 |
+
"tie_word_embeddings": false,
|
| 28 |
+
"torch_dtype": "float32",
|
| 29 |
+
"transformers_version": "4.45.1",
|
| 30 |
+
"use_cache": true,
|
| 31 |
+
"vocab_size": 32003
|
| 32 |
+
}
|
checkpoint-4218/generation_config.json
ADDED
|
@@ -0,0 +1,7 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_from_model_config": true,
|
| 3 |
+
"bos_token_id": 1,
|
| 4 |
+
"eos_token_id": 2,
|
| 5 |
+
"pad_token_id": 0,
|
| 6 |
+
"transformers_version": "4.45.1"
|
| 7 |
+
}
|
checkpoint-4218/model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:9ed892552e4757b7d11b8ff1d74dc2449cf88d885f52872c693b3159f6d36b57
|
| 3 |
+
size 997345472
|
checkpoint-4218/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:ce50b63e3e3c18683454ee37a11561440eca8e7b734c5c88a05d11025d626acf
|
| 3 |
+
size 1994737722
|
checkpoint-4218/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:386fcc8cc1089aade9450d86fb239ea3483f455fd2d78d8378645feecfec9d69
|
| 3 |
+
size 14244
|
checkpoint-4218/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:9e1f9b966746dac160df707e9211e9209bb9e54cf18c6d8ce33924a240310f96
|
| 3 |
+
size 1064
|
checkpoint-4218/trainer_state.json
ADDED
|
@@ -0,0 +1,2980 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 0.9998222222222222,
|
| 5 |
+
"eval_steps": 5000,
|
| 6 |
+
"global_step": 4218,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.0023703703703703703,
|
| 13 |
+
"grad_norm": 0.542885959148407,
|
| 14 |
+
"learning_rate": 1.0000000000000002e-06,
|
| 15 |
+
"loss": 1.4997,
|
| 16 |
+
"step": 10
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.004740740740740741,
|
| 20 |
+
"grad_norm": 0.44538265466690063,
|
| 21 |
+
"learning_rate": 2.0000000000000003e-06,
|
| 22 |
+
"loss": 1.4925,
|
| 23 |
+
"step": 20
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.0071111111111111115,
|
| 27 |
+
"grad_norm": 0.4945567548274994,
|
| 28 |
+
"learning_rate": 3e-06,
|
| 29 |
+
"loss": 1.4985,
|
| 30 |
+
"step": 30
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.009481481481481481,
|
| 34 |
+
"grad_norm": 0.4560663402080536,
|
| 35 |
+
"learning_rate": 4.000000000000001e-06,
|
| 36 |
+
"loss": 1.4031,
|
| 37 |
+
"step": 40
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.011851851851851851,
|
| 41 |
+
"grad_norm": 0.4691298007965088,
|
| 42 |
+
"learning_rate": 5e-06,
|
| 43 |
+
"loss": 1.4175,
|
| 44 |
+
"step": 50
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.014222222222222223,
|
| 48 |
+
"grad_norm": 0.44202300906181335,
|
| 49 |
+
"learning_rate": 6e-06,
|
| 50 |
+
"loss": 1.4337,
|
| 51 |
+
"step": 60
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.016592592592592593,
|
| 55 |
+
"grad_norm": 0.5069476366043091,
|
| 56 |
+
"learning_rate": 7.000000000000001e-06,
|
| 57 |
+
"loss": 1.4629,
|
| 58 |
+
"step": 70
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.018962962962962963,
|
| 62 |
+
"grad_norm": 0.4806945025920868,
|
| 63 |
+
"learning_rate": 8.000000000000001e-06,
|
| 64 |
+
"loss": 1.5706,
|
| 65 |
+
"step": 80
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.021333333333333333,
|
| 69 |
+
"grad_norm": 0.5269841551780701,
|
| 70 |
+
"learning_rate": 9e-06,
|
| 71 |
+
"loss": 1.5625,
|
| 72 |
+
"step": 90
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 0.023703703703703703,
|
| 76 |
+
"grad_norm": 0.37831586599349976,
|
| 77 |
+
"learning_rate": 1e-05,
|
| 78 |
+
"loss": 1.4083,
|
| 79 |
+
"step": 100
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.026074074074074072,
|
| 83 |
+
"grad_norm": 0.442981094121933,
|
| 84 |
+
"learning_rate": 1.1000000000000001e-05,
|
| 85 |
+
"loss": 1.3799,
|
| 86 |
+
"step": 110
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.028444444444444446,
|
| 90 |
+
"grad_norm": 0.47675761580467224,
|
| 91 |
+
"learning_rate": 1.2e-05,
|
| 92 |
+
"loss": 1.5356,
|
| 93 |
+
"step": 120
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 0.030814814814814816,
|
| 97 |
+
"grad_norm": 0.5033993721008301,
|
| 98 |
+
"learning_rate": 1.3000000000000001e-05,
|
| 99 |
+
"loss": 1.504,
|
| 100 |
+
"step": 130
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"epoch": 0.033185185185185186,
|
| 104 |
+
"grad_norm": 0.4628155827522278,
|
| 105 |
+
"learning_rate": 1.4000000000000001e-05,
|
| 106 |
+
"loss": 1.3474,
|
| 107 |
+
"step": 140
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 0.035555555555555556,
|
| 111 |
+
"grad_norm": 0.41637757420539856,
|
| 112 |
+
"learning_rate": 1.5e-05,
|
| 113 |
+
"loss": 1.4352,
|
| 114 |
+
"step": 150
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 0.037925925925925925,
|
| 118 |
+
"grad_norm": 0.5029244422912598,
|
| 119 |
+
"learning_rate": 1.6000000000000003e-05,
|
| 120 |
+
"loss": 1.3224,
|
| 121 |
+
"step": 160
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.040296296296296295,
|
| 125 |
+
"grad_norm": 0.6434731483459473,
|
| 126 |
+
"learning_rate": 1.7000000000000003e-05,
|
| 127 |
+
"loss": 1.5611,
|
| 128 |
+
"step": 170
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 0.042666666666666665,
|
| 132 |
+
"grad_norm": 0.42424359917640686,
|
| 133 |
+
"learning_rate": 1.8e-05,
|
| 134 |
+
"loss": 1.4191,
|
| 135 |
+
"step": 180
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 0.045037037037037035,
|
| 139 |
+
"grad_norm": 0.4729703962802887,
|
| 140 |
+
"learning_rate": 1.9e-05,
|
| 141 |
+
"loss": 1.3284,
|
| 142 |
+
"step": 190
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"epoch": 0.047407407407407405,
|
| 146 |
+
"grad_norm": 0.48806190490722656,
|
| 147 |
+
"learning_rate": 2e-05,
|
| 148 |
+
"loss": 1.48,
|
| 149 |
+
"step": 200
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 0.049777777777777775,
|
| 153 |
+
"grad_norm": 0.4987320303916931,
|
| 154 |
+
"learning_rate": 2.1e-05,
|
| 155 |
+
"loss": 1.4535,
|
| 156 |
+
"step": 210
|
| 157 |
+
},
|
| 158 |
+
{
|
| 159 |
+
"epoch": 0.052148148148148145,
|
| 160 |
+
"grad_norm": 0.46912866830825806,
|
| 161 |
+
"learning_rate": 2.2000000000000003e-05,
|
| 162 |
+
"loss": 1.4846,
|
| 163 |
+
"step": 220
|
| 164 |
+
},
|
| 165 |
+
{
|
| 166 |
+
"epoch": 0.05451851851851852,
|
| 167 |
+
"grad_norm": 0.4369196593761444,
|
| 168 |
+
"learning_rate": 2.3000000000000003e-05,
|
| 169 |
+
"loss": 1.5564,
|
| 170 |
+
"step": 230
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 0.05688888888888889,
|
| 174 |
+
"grad_norm": 0.48074963688850403,
|
| 175 |
+
"learning_rate": 2.4e-05,
|
| 176 |
+
"loss": 1.3142,
|
| 177 |
+
"step": 240
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 0.05925925925925926,
|
| 181 |
+
"grad_norm": 0.450253427028656,
|
| 182 |
+
"learning_rate": 2.5e-05,
|
| 183 |
+
"loss": 1.3877,
|
| 184 |
+
"step": 250
|
| 185 |
+
},
|
| 186 |
+
{
|
| 187 |
+
"epoch": 0.06162962962962963,
|
| 188 |
+
"grad_norm": 0.4517356753349304,
|
| 189 |
+
"learning_rate": 2.6000000000000002e-05,
|
| 190 |
+
"loss": 1.3969,
|
| 191 |
+
"step": 260
|
| 192 |
+
},
|
| 193 |
+
{
|
| 194 |
+
"epoch": 0.064,
|
| 195 |
+
"grad_norm": 0.47781577706336975,
|
| 196 |
+
"learning_rate": 2.7000000000000002e-05,
|
| 197 |
+
"loss": 1.5352,
|
| 198 |
+
"step": 270
|
| 199 |
+
},
|
| 200 |
+
{
|
| 201 |
+
"epoch": 0.06637037037037037,
|
| 202 |
+
"grad_norm": 0.5579633712768555,
|
| 203 |
+
"learning_rate": 2.8000000000000003e-05,
|
| 204 |
+
"loss": 1.5436,
|
| 205 |
+
"step": 280
|
| 206 |
+
},
|
| 207 |
+
{
|
| 208 |
+
"epoch": 0.06874074074074074,
|
| 209 |
+
"grad_norm": 0.4838034510612488,
|
| 210 |
+
"learning_rate": 2.9e-05,
|
| 211 |
+
"loss": 1.3564,
|
| 212 |
+
"step": 290
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 0.07111111111111111,
|
| 216 |
+
"grad_norm": 0.5685828328132629,
|
| 217 |
+
"learning_rate": 3e-05,
|
| 218 |
+
"loss": 1.4429,
|
| 219 |
+
"step": 300
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 0.07348148148148148,
|
| 223 |
+
"grad_norm": 0.5230541229248047,
|
| 224 |
+
"learning_rate": 3.1e-05,
|
| 225 |
+
"loss": 1.3933,
|
| 226 |
+
"step": 310
|
| 227 |
+
},
|
| 228 |
+
{
|
| 229 |
+
"epoch": 0.07585185185185185,
|
| 230 |
+
"grad_norm": 0.45525529980659485,
|
| 231 |
+
"learning_rate": 3.2000000000000005e-05,
|
| 232 |
+
"loss": 1.4224,
|
| 233 |
+
"step": 320
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 0.07822222222222222,
|
| 237 |
+
"grad_norm": 0.47926583886146545,
|
| 238 |
+
"learning_rate": 3.3e-05,
|
| 239 |
+
"loss": 1.4466,
|
| 240 |
+
"step": 330
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 0.08059259259259259,
|
| 244 |
+
"grad_norm": 0.38689500093460083,
|
| 245 |
+
"learning_rate": 3.4000000000000007e-05,
|
| 246 |
+
"loss": 1.5105,
|
| 247 |
+
"step": 340
|
| 248 |
+
},
|
| 249 |
+
{
|
| 250 |
+
"epoch": 0.08296296296296296,
|
| 251 |
+
"grad_norm": 0.4488411843776703,
|
| 252 |
+
"learning_rate": 3.5e-05,
|
| 253 |
+
"loss": 1.3392,
|
| 254 |
+
"step": 350
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 0.08533333333333333,
|
| 258 |
+
"grad_norm": 0.5381152033805847,
|
| 259 |
+
"learning_rate": 3.6e-05,
|
| 260 |
+
"loss": 1.375,
|
| 261 |
+
"step": 360
|
| 262 |
+
},
|
| 263 |
+
{
|
| 264 |
+
"epoch": 0.0877037037037037,
|
| 265 |
+
"grad_norm": 0.5788478255271912,
|
| 266 |
+
"learning_rate": 3.7e-05,
|
| 267 |
+
"loss": 1.3522,
|
| 268 |
+
"step": 370
|
| 269 |
+
},
|
| 270 |
+
{
|
| 271 |
+
"epoch": 0.09007407407407407,
|
| 272 |
+
"grad_norm": 0.501133143901825,
|
| 273 |
+
"learning_rate": 3.8e-05,
|
| 274 |
+
"loss": 1.3294,
|
| 275 |
+
"step": 380
|
| 276 |
+
},
|
| 277 |
+
{
|
| 278 |
+
"epoch": 0.09244444444444444,
|
| 279 |
+
"grad_norm": 0.5300689935684204,
|
| 280 |
+
"learning_rate": 3.9000000000000006e-05,
|
| 281 |
+
"loss": 1.5623,
|
| 282 |
+
"step": 390
|
| 283 |
+
},
|
| 284 |
+
{
|
| 285 |
+
"epoch": 0.09481481481481481,
|
| 286 |
+
"grad_norm": 0.5409078001976013,
|
| 287 |
+
"learning_rate": 4e-05,
|
| 288 |
+
"loss": 1.4098,
|
| 289 |
+
"step": 400
|
| 290 |
+
},
|
| 291 |
+
{
|
| 292 |
+
"epoch": 0.09718518518518518,
|
| 293 |
+
"grad_norm": 0.5598166584968567,
|
| 294 |
+
"learning_rate": 4.1e-05,
|
| 295 |
+
"loss": 1.4104,
|
| 296 |
+
"step": 410
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 0.09955555555555555,
|
| 300 |
+
"grad_norm": 0.5656659603118896,
|
| 301 |
+
"learning_rate": 4.2e-05,
|
| 302 |
+
"loss": 1.3782,
|
| 303 |
+
"step": 420
|
| 304 |
+
},
|
| 305 |
+
{
|
| 306 |
+
"epoch": 0.10192592592592592,
|
| 307 |
+
"grad_norm": 0.5094364881515503,
|
| 308 |
+
"learning_rate": 4.3e-05,
|
| 309 |
+
"loss": 1.6233,
|
| 310 |
+
"step": 430
|
| 311 |
+
},
|
| 312 |
+
{
|
| 313 |
+
"epoch": 0.10429629629629629,
|
| 314 |
+
"grad_norm": 0.5540050268173218,
|
| 315 |
+
"learning_rate": 4.4000000000000006e-05,
|
| 316 |
+
"loss": 1.512,
|
| 317 |
+
"step": 440
|
| 318 |
+
},
|
| 319 |
+
{
|
| 320 |
+
"epoch": 0.10666666666666667,
|
| 321 |
+
"grad_norm": 0.5693063139915466,
|
| 322 |
+
"learning_rate": 4.5e-05,
|
| 323 |
+
"loss": 1.641,
|
| 324 |
+
"step": 450
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 0.10903703703703704,
|
| 328 |
+
"grad_norm": 0.5556958913803101,
|
| 329 |
+
"learning_rate": 4.600000000000001e-05,
|
| 330 |
+
"loss": 1.4532,
|
| 331 |
+
"step": 460
|
| 332 |
+
},
|
| 333 |
+
{
|
| 334 |
+
"epoch": 0.11140740740740741,
|
| 335 |
+
"grad_norm": 0.5052928924560547,
|
| 336 |
+
"learning_rate": 4.7e-05,
|
| 337 |
+
"loss": 1.4863,
|
| 338 |
+
"step": 470
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 0.11377777777777778,
|
| 342 |
+
"grad_norm": 0.5321051478385925,
|
| 343 |
+
"learning_rate": 4.8e-05,
|
| 344 |
+
"loss": 1.353,
|
| 345 |
+
"step": 480
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 0.11614814814814815,
|
| 349 |
+
"grad_norm": 0.658074140548706,
|
| 350 |
+
"learning_rate": 4.9e-05,
|
| 351 |
+
"loss": 1.563,
|
| 352 |
+
"step": 490
|
| 353 |
+
},
|
| 354 |
+
{
|
| 355 |
+
"epoch": 0.11851851851851852,
|
| 356 |
+
"grad_norm": 0.4817732870578766,
|
| 357 |
+
"learning_rate": 5e-05,
|
| 358 |
+
"loss": 1.5347,
|
| 359 |
+
"step": 500
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"epoch": 0.12088888888888889,
|
| 363 |
+
"grad_norm": 0.6876205801963806,
|
| 364 |
+
"learning_rate": 4.9865519096288324e-05,
|
| 365 |
+
"loss": 1.6524,
|
| 366 |
+
"step": 510
|
| 367 |
+
},
|
| 368 |
+
{
|
| 369 |
+
"epoch": 0.12325925925925926,
|
| 370 |
+
"grad_norm": 0.5238626003265381,
|
| 371 |
+
"learning_rate": 4.973103819257665e-05,
|
| 372 |
+
"loss": 1.5766,
|
| 373 |
+
"step": 520
|
| 374 |
+
},
|
| 375 |
+
{
|
| 376 |
+
"epoch": 0.12562962962962962,
|
| 377 |
+
"grad_norm": 0.4588116705417633,
|
| 378 |
+
"learning_rate": 4.959655728886498e-05,
|
| 379 |
+
"loss": 1.4258,
|
| 380 |
+
"step": 530
|
| 381 |
+
},
|
| 382 |
+
{
|
| 383 |
+
"epoch": 0.128,
|
| 384 |
+
"grad_norm": 0.529692530632019,
|
| 385 |
+
"learning_rate": 4.946207638515331e-05,
|
| 386 |
+
"loss": 1.5574,
|
| 387 |
+
"step": 540
|
| 388 |
+
},
|
| 389 |
+
{
|
| 390 |
+
"epoch": 0.13037037037037036,
|
| 391 |
+
"grad_norm": 0.475524365901947,
|
| 392 |
+
"learning_rate": 4.932759548144163e-05,
|
| 393 |
+
"loss": 1.377,
|
| 394 |
+
"step": 550
|
| 395 |
+
},
|
| 396 |
+
{
|
| 397 |
+
"epoch": 0.13274074074074074,
|
| 398 |
+
"grad_norm": 0.48722413182258606,
|
| 399 |
+
"learning_rate": 4.919311457772996e-05,
|
| 400 |
+
"loss": 1.3156,
|
| 401 |
+
"step": 560
|
| 402 |
+
},
|
| 403 |
+
{
|
| 404 |
+
"epoch": 0.1351111111111111,
|
| 405 |
+
"grad_norm": 0.6309683322906494,
|
| 406 |
+
"learning_rate": 4.905863367401829e-05,
|
| 407 |
+
"loss": 1.5581,
|
| 408 |
+
"step": 570
|
| 409 |
+
},
|
| 410 |
+
{
|
| 411 |
+
"epoch": 0.13748148148148148,
|
| 412 |
+
"grad_norm": 0.5029247999191284,
|
| 413 |
+
"learning_rate": 4.892415277030662e-05,
|
| 414 |
+
"loss": 1.5968,
|
| 415 |
+
"step": 580
|
| 416 |
+
},
|
| 417 |
+
{
|
| 418 |
+
"epoch": 0.13985185185185184,
|
| 419 |
+
"grad_norm": 0.421310156583786,
|
| 420 |
+
"learning_rate": 4.878967186659494e-05,
|
| 421 |
+
"loss": 1.4881,
|
| 422 |
+
"step": 590
|
| 423 |
+
},
|
| 424 |
+
{
|
| 425 |
+
"epoch": 0.14222222222222222,
|
| 426 |
+
"grad_norm": 0.5082572102546692,
|
| 427 |
+
"learning_rate": 4.865519096288327e-05,
|
| 428 |
+
"loss": 1.6574,
|
| 429 |
+
"step": 600
|
| 430 |
+
},
|
| 431 |
+
{
|
| 432 |
+
"epoch": 0.1445925925925926,
|
| 433 |
+
"grad_norm": 0.5082793235778809,
|
| 434 |
+
"learning_rate": 4.85207100591716e-05,
|
| 435 |
+
"loss": 1.3606,
|
| 436 |
+
"step": 610
|
| 437 |
+
},
|
| 438 |
+
{
|
| 439 |
+
"epoch": 0.14696296296296296,
|
| 440 |
+
"grad_norm": 0.5169036388397217,
|
| 441 |
+
"learning_rate": 4.838622915545993e-05,
|
| 442 |
+
"loss": 1.4457,
|
| 443 |
+
"step": 620
|
| 444 |
+
},
|
| 445 |
+
{
|
| 446 |
+
"epoch": 0.14933333333333335,
|
| 447 |
+
"grad_norm": 0.5509771704673767,
|
| 448 |
+
"learning_rate": 4.825174825174825e-05,
|
| 449 |
+
"loss": 1.4754,
|
| 450 |
+
"step": 630
|
| 451 |
+
},
|
| 452 |
+
{
|
| 453 |
+
"epoch": 0.1517037037037037,
|
| 454 |
+
"grad_norm": 0.630851149559021,
|
| 455 |
+
"learning_rate": 4.811726734803658e-05,
|
| 456 |
+
"loss": 1.5453,
|
| 457 |
+
"step": 640
|
| 458 |
+
},
|
| 459 |
+
{
|
| 460 |
+
"epoch": 0.15407407407407409,
|
| 461 |
+
"grad_norm": 0.5343595147132874,
|
| 462 |
+
"learning_rate": 4.798278644432491e-05,
|
| 463 |
+
"loss": 1.4822,
|
| 464 |
+
"step": 650
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 0.15644444444444444,
|
| 468 |
+
"grad_norm": 0.5070016980171204,
|
| 469 |
+
"learning_rate": 4.7848305540613237e-05,
|
| 470 |
+
"loss": 1.3835,
|
| 471 |
+
"step": 660
|
| 472 |
+
},
|
| 473 |
+
{
|
| 474 |
+
"epoch": 0.15881481481481483,
|
| 475 |
+
"grad_norm": 0.6097332835197449,
|
| 476 |
+
"learning_rate": 4.771382463690156e-05,
|
| 477 |
+
"loss": 1.546,
|
| 478 |
+
"step": 670
|
| 479 |
+
},
|
| 480 |
+
{
|
| 481 |
+
"epoch": 0.16118518518518518,
|
| 482 |
+
"grad_norm": 0.5894319415092468,
|
| 483 |
+
"learning_rate": 4.757934373318989e-05,
|
| 484 |
+
"loss": 1.3605,
|
| 485 |
+
"step": 680
|
| 486 |
+
},
|
| 487 |
+
{
|
| 488 |
+
"epoch": 0.16355555555555557,
|
| 489 |
+
"grad_norm": 0.4879942238330841,
|
| 490 |
+
"learning_rate": 4.7444862829478216e-05,
|
| 491 |
+
"loss": 1.4391,
|
| 492 |
+
"step": 690
|
| 493 |
+
},
|
| 494 |
+
{
|
| 495 |
+
"epoch": 0.16592592592592592,
|
| 496 |
+
"grad_norm": 0.49390801787376404,
|
| 497 |
+
"learning_rate": 4.7310381925766545e-05,
|
| 498 |
+
"loss": 1.2228,
|
| 499 |
+
"step": 700
|
| 500 |
+
},
|
| 501 |
+
{
|
| 502 |
+
"epoch": 0.1682962962962963,
|
| 503 |
+
"grad_norm": 0.6193021535873413,
|
| 504 |
+
"learning_rate": 4.717590102205487e-05,
|
| 505 |
+
"loss": 1.7474,
|
| 506 |
+
"step": 710
|
| 507 |
+
},
|
| 508 |
+
{
|
| 509 |
+
"epoch": 0.17066666666666666,
|
| 510 |
+
"grad_norm": 0.4410654604434967,
|
| 511 |
+
"learning_rate": 4.7041420118343196e-05,
|
| 512 |
+
"loss": 1.4255,
|
| 513 |
+
"step": 720
|
| 514 |
+
},
|
| 515 |
+
{
|
| 516 |
+
"epoch": 0.17303703703703704,
|
| 517 |
+
"grad_norm": 0.5690642595291138,
|
| 518 |
+
"learning_rate": 4.6906939214631525e-05,
|
| 519 |
+
"loss": 1.4252,
|
| 520 |
+
"step": 730
|
| 521 |
+
},
|
| 522 |
+
{
|
| 523 |
+
"epoch": 0.1754074074074074,
|
| 524 |
+
"grad_norm": 0.4561966359615326,
|
| 525 |
+
"learning_rate": 4.6772458310919854e-05,
|
| 526 |
+
"loss": 1.4214,
|
| 527 |
+
"step": 740
|
| 528 |
+
},
|
| 529 |
+
{
|
| 530 |
+
"epoch": 0.17777777777777778,
|
| 531 |
+
"grad_norm": 0.5871363282203674,
|
| 532 |
+
"learning_rate": 4.6637977407208176e-05,
|
| 533 |
+
"loss": 1.5272,
|
| 534 |
+
"step": 750
|
| 535 |
+
},
|
| 536 |
+
{
|
| 537 |
+
"epoch": 0.18014814814814814,
|
| 538 |
+
"grad_norm": 0.5293774604797363,
|
| 539 |
+
"learning_rate": 4.6503496503496505e-05,
|
| 540 |
+
"loss": 1.5507,
|
| 541 |
+
"step": 760
|
| 542 |
+
},
|
| 543 |
+
{
|
| 544 |
+
"epoch": 0.18251851851851852,
|
| 545 |
+
"grad_norm": 0.483826220035553,
|
| 546 |
+
"learning_rate": 4.636901559978483e-05,
|
| 547 |
+
"loss": 1.4058,
|
| 548 |
+
"step": 770
|
| 549 |
+
},
|
| 550 |
+
{
|
| 551 |
+
"epoch": 0.18488888888888888,
|
| 552 |
+
"grad_norm": 0.551902174949646,
|
| 553 |
+
"learning_rate": 4.623453469607316e-05,
|
| 554 |
+
"loss": 1.3707,
|
| 555 |
+
"step": 780
|
| 556 |
+
},
|
| 557 |
+
{
|
| 558 |
+
"epoch": 0.18725925925925926,
|
| 559 |
+
"grad_norm": 0.5492023825645447,
|
| 560 |
+
"learning_rate": 4.6100053792361484e-05,
|
| 561 |
+
"loss": 1.4901,
|
| 562 |
+
"step": 790
|
| 563 |
+
},
|
| 564 |
+
{
|
| 565 |
+
"epoch": 0.18962962962962962,
|
| 566 |
+
"grad_norm": 0.5409772396087646,
|
| 567 |
+
"learning_rate": 4.596557288864981e-05,
|
| 568 |
+
"loss": 1.6356,
|
| 569 |
+
"step": 800
|
| 570 |
+
},
|
| 571 |
+
{
|
| 572 |
+
"epoch": 0.192,
|
| 573 |
+
"grad_norm": 0.5469648838043213,
|
| 574 |
+
"learning_rate": 4.583109198493814e-05,
|
| 575 |
+
"loss": 1.4574,
|
| 576 |
+
"step": 810
|
| 577 |
+
},
|
| 578 |
+
{
|
| 579 |
+
"epoch": 0.19437037037037036,
|
| 580 |
+
"grad_norm": 0.5523713827133179,
|
| 581 |
+
"learning_rate": 4.569661108122647e-05,
|
| 582 |
+
"loss": 1.5304,
|
| 583 |
+
"step": 820
|
| 584 |
+
},
|
| 585 |
+
{
|
| 586 |
+
"epoch": 0.19674074074074074,
|
| 587 |
+
"grad_norm": 0.4884456992149353,
|
| 588 |
+
"learning_rate": 4.556213017751479e-05,
|
| 589 |
+
"loss": 1.4173,
|
| 590 |
+
"step": 830
|
| 591 |
+
},
|
| 592 |
+
{
|
| 593 |
+
"epoch": 0.1991111111111111,
|
| 594 |
+
"grad_norm": 0.5865374803543091,
|
| 595 |
+
"learning_rate": 4.542764927380312e-05,
|
| 596 |
+
"loss": 1.4178,
|
| 597 |
+
"step": 840
|
| 598 |
+
},
|
| 599 |
+
{
|
| 600 |
+
"epoch": 0.20148148148148148,
|
| 601 |
+
"grad_norm": 0.5571750402450562,
|
| 602 |
+
"learning_rate": 4.529316837009145e-05,
|
| 603 |
+
"loss": 1.5356,
|
| 604 |
+
"step": 850
|
| 605 |
+
},
|
| 606 |
+
{
|
| 607 |
+
"epoch": 0.20385185185185184,
|
| 608 |
+
"grad_norm": 0.567616879940033,
|
| 609 |
+
"learning_rate": 4.515868746637978e-05,
|
| 610 |
+
"loss": 1.4496,
|
| 611 |
+
"step": 860
|
| 612 |
+
},
|
| 613 |
+
{
|
| 614 |
+
"epoch": 0.20622222222222222,
|
| 615 |
+
"grad_norm": 0.5077497959136963,
|
| 616 |
+
"learning_rate": 4.50242065626681e-05,
|
| 617 |
+
"loss": 1.4757,
|
| 618 |
+
"step": 870
|
| 619 |
+
},
|
| 620 |
+
{
|
| 621 |
+
"epoch": 0.20859259259259258,
|
| 622 |
+
"grad_norm": 0.5118802189826965,
|
| 623 |
+
"learning_rate": 4.488972565895643e-05,
|
| 624 |
+
"loss": 1.3845,
|
| 625 |
+
"step": 880
|
| 626 |
+
},
|
| 627 |
+
{
|
| 628 |
+
"epoch": 0.21096296296296296,
|
| 629 |
+
"grad_norm": 0.43292248249053955,
|
| 630 |
+
"learning_rate": 4.475524475524476e-05,
|
| 631 |
+
"loss": 1.421,
|
| 632 |
+
"step": 890
|
| 633 |
+
},
|
| 634 |
+
{
|
| 635 |
+
"epoch": 0.21333333333333335,
|
| 636 |
+
"grad_norm": 0.5365243554115295,
|
| 637 |
+
"learning_rate": 4.462076385153308e-05,
|
| 638 |
+
"loss": 1.4586,
|
| 639 |
+
"step": 900
|
| 640 |
+
},
|
| 641 |
+
{
|
| 642 |
+
"epoch": 0.2157037037037037,
|
| 643 |
+
"grad_norm": 0.4912022054195404,
|
| 644 |
+
"learning_rate": 4.448628294782141e-05,
|
| 645 |
+
"loss": 1.5385,
|
| 646 |
+
"step": 910
|
| 647 |
+
},
|
| 648 |
+
{
|
| 649 |
+
"epoch": 0.2180740740740741,
|
| 650 |
+
"grad_norm": 0.5855193734169006,
|
| 651 |
+
"learning_rate": 4.435180204410974e-05,
|
| 652 |
+
"loss": 1.5718,
|
| 653 |
+
"step": 920
|
| 654 |
+
},
|
| 655 |
+
{
|
| 656 |
+
"epoch": 0.22044444444444444,
|
| 657 |
+
"grad_norm": 0.5224360227584839,
|
| 658 |
+
"learning_rate": 4.421732114039807e-05,
|
| 659 |
+
"loss": 1.3853,
|
| 660 |
+
"step": 930
|
| 661 |
+
},
|
| 662 |
+
{
|
| 663 |
+
"epoch": 0.22281481481481483,
|
| 664 |
+
"grad_norm": 0.4283509850502014,
|
| 665 |
+
"learning_rate": 4.408284023668639e-05,
|
| 666 |
+
"loss": 1.3758,
|
| 667 |
+
"step": 940
|
| 668 |
+
},
|
| 669 |
+
{
|
| 670 |
+
"epoch": 0.22518518518518518,
|
| 671 |
+
"grad_norm": 0.44806018471717834,
|
| 672 |
+
"learning_rate": 4.394835933297472e-05,
|
| 673 |
+
"loss": 1.4089,
|
| 674 |
+
"step": 950
|
| 675 |
+
},
|
| 676 |
+
{
|
| 677 |
+
"epoch": 0.22755555555555557,
|
| 678 |
+
"grad_norm": 0.4234403967857361,
|
| 679 |
+
"learning_rate": 4.381387842926305e-05,
|
| 680 |
+
"loss": 1.321,
|
| 681 |
+
"step": 960
|
| 682 |
+
},
|
| 683 |
+
{
|
| 684 |
+
"epoch": 0.22992592592592592,
|
| 685 |
+
"grad_norm": 0.45570847392082214,
|
| 686 |
+
"learning_rate": 4.3679397525551376e-05,
|
| 687 |
+
"loss": 1.384,
|
| 688 |
+
"step": 970
|
| 689 |
+
},
|
| 690 |
+
{
|
| 691 |
+
"epoch": 0.2322962962962963,
|
| 692 |
+
"grad_norm": 0.6098482608795166,
|
| 693 |
+
"learning_rate": 4.35449166218397e-05,
|
| 694 |
+
"loss": 1.5165,
|
| 695 |
+
"step": 980
|
| 696 |
+
},
|
| 697 |
+
{
|
| 698 |
+
"epoch": 0.23466666666666666,
|
| 699 |
+
"grad_norm": 0.47981974482536316,
|
| 700 |
+
"learning_rate": 4.341043571812803e-05,
|
| 701 |
+
"loss": 1.4827,
|
| 702 |
+
"step": 990
|
| 703 |
+
},
|
| 704 |
+
{
|
| 705 |
+
"epoch": 0.23703703703703705,
|
| 706 |
+
"grad_norm": 0.567845344543457,
|
| 707 |
+
"learning_rate": 4.3275954814416356e-05,
|
| 708 |
+
"loss": 1.4494,
|
| 709 |
+
"step": 1000
|
| 710 |
+
},
|
| 711 |
+
{
|
| 712 |
+
"epoch": 0.2394074074074074,
|
| 713 |
+
"grad_norm": 0.5508958697319031,
|
| 714 |
+
"learning_rate": 4.3141473910704685e-05,
|
| 715 |
+
"loss": 1.5681,
|
| 716 |
+
"step": 1010
|
| 717 |
+
},
|
| 718 |
+
{
|
| 719 |
+
"epoch": 0.24177777777777779,
|
| 720 |
+
"grad_norm": 0.6119508743286133,
|
| 721 |
+
"learning_rate": 4.300699300699301e-05,
|
| 722 |
+
"loss": 1.4522,
|
| 723 |
+
"step": 1020
|
| 724 |
+
},
|
| 725 |
+
{
|
| 726 |
+
"epoch": 0.24414814814814814,
|
| 727 |
+
"grad_norm": 0.654909074306488,
|
| 728 |
+
"learning_rate": 4.2872512103281336e-05,
|
| 729 |
+
"loss": 1.6317,
|
| 730 |
+
"step": 1030
|
| 731 |
+
},
|
| 732 |
+
{
|
| 733 |
+
"epoch": 0.24651851851851853,
|
| 734 |
+
"grad_norm": 0.5818801522254944,
|
| 735 |
+
"learning_rate": 4.2738031199569664e-05,
|
| 736 |
+
"loss": 1.4917,
|
| 737 |
+
"step": 1040
|
| 738 |
+
},
|
| 739 |
+
{
|
| 740 |
+
"epoch": 0.24888888888888888,
|
| 741 |
+
"grad_norm": 0.5295186638832092,
|
| 742 |
+
"learning_rate": 4.260355029585799e-05,
|
| 743 |
+
"loss": 1.6393,
|
| 744 |
+
"step": 1050
|
| 745 |
+
},
|
| 746 |
+
{
|
| 747 |
+
"epoch": 0.25125925925925924,
|
| 748 |
+
"grad_norm": 0.5558478832244873,
|
| 749 |
+
"learning_rate": 4.2469069392146315e-05,
|
| 750 |
+
"loss": 1.5607,
|
| 751 |
+
"step": 1060
|
| 752 |
+
},
|
| 753 |
+
{
|
| 754 |
+
"epoch": 0.25362962962962965,
|
| 755 |
+
"grad_norm": 0.5266067385673523,
|
| 756 |
+
"learning_rate": 4.2334588488434644e-05,
|
| 757 |
+
"loss": 1.4366,
|
| 758 |
+
"step": 1070
|
| 759 |
+
},
|
| 760 |
+
{
|
| 761 |
+
"epoch": 0.256,
|
| 762 |
+
"grad_norm": 0.4949641823768616,
|
| 763 |
+
"learning_rate": 4.220010758472297e-05,
|
| 764 |
+
"loss": 1.4392,
|
| 765 |
+
"step": 1080
|
| 766 |
+
},
|
| 767 |
+
{
|
| 768 |
+
"epoch": 0.25837037037037036,
|
| 769 |
+
"grad_norm": 0.48148399591445923,
|
| 770 |
+
"learning_rate": 4.20656266810113e-05,
|
| 771 |
+
"loss": 1.5301,
|
| 772 |
+
"step": 1090
|
| 773 |
+
},
|
| 774 |
+
{
|
| 775 |
+
"epoch": 0.2607407407407407,
|
| 776 |
+
"grad_norm": 0.5564059615135193,
|
| 777 |
+
"learning_rate": 4.1931145777299624e-05,
|
| 778 |
+
"loss": 1.3645,
|
| 779 |
+
"step": 1100
|
| 780 |
+
},
|
| 781 |
+
{
|
| 782 |
+
"epoch": 0.26311111111111113,
|
| 783 |
+
"grad_norm": 0.6419994235038757,
|
| 784 |
+
"learning_rate": 4.179666487358795e-05,
|
| 785 |
+
"loss": 1.4543,
|
| 786 |
+
"step": 1110
|
| 787 |
+
},
|
| 788 |
+
{
|
| 789 |
+
"epoch": 0.2654814814814815,
|
| 790 |
+
"grad_norm": 0.5205827355384827,
|
| 791 |
+
"learning_rate": 4.166218396987628e-05,
|
| 792 |
+
"loss": 1.5358,
|
| 793 |
+
"step": 1120
|
| 794 |
+
},
|
| 795 |
+
{
|
| 796 |
+
"epoch": 0.26785185185185184,
|
| 797 |
+
"grad_norm": 0.45430988073349,
|
| 798 |
+
"learning_rate": 4.152770306616461e-05,
|
| 799 |
+
"loss": 1.5483,
|
| 800 |
+
"step": 1130
|
| 801 |
+
},
|
| 802 |
+
{
|
| 803 |
+
"epoch": 0.2702222222222222,
|
| 804 |
+
"grad_norm": 0.5467645525932312,
|
| 805 |
+
"learning_rate": 4.139322216245293e-05,
|
| 806 |
+
"loss": 1.4702,
|
| 807 |
+
"step": 1140
|
| 808 |
+
},
|
| 809 |
+
{
|
| 810 |
+
"epoch": 0.2725925925925926,
|
| 811 |
+
"grad_norm": 0.47259363532066345,
|
| 812 |
+
"learning_rate": 4.125874125874126e-05,
|
| 813 |
+
"loss": 1.451,
|
| 814 |
+
"step": 1150
|
| 815 |
+
},
|
| 816 |
+
{
|
| 817 |
+
"epoch": 0.27496296296296296,
|
| 818 |
+
"grad_norm": 0.48951438069343567,
|
| 819 |
+
"learning_rate": 4.112426035502959e-05,
|
| 820 |
+
"loss": 1.4095,
|
| 821 |
+
"step": 1160
|
| 822 |
+
},
|
| 823 |
+
{
|
| 824 |
+
"epoch": 0.2773333333333333,
|
| 825 |
+
"grad_norm": 0.601701021194458,
|
| 826 |
+
"learning_rate": 4.098977945131792e-05,
|
| 827 |
+
"loss": 1.6848,
|
| 828 |
+
"step": 1170
|
| 829 |
+
},
|
| 830 |
+
{
|
| 831 |
+
"epoch": 0.2797037037037037,
|
| 832 |
+
"grad_norm": 0.5379857420921326,
|
| 833 |
+
"learning_rate": 4.085529854760624e-05,
|
| 834 |
+
"loss": 1.3598,
|
| 835 |
+
"step": 1180
|
| 836 |
+
},
|
| 837 |
+
{
|
| 838 |
+
"epoch": 0.2820740740740741,
|
| 839 |
+
"grad_norm": 0.6498066186904907,
|
| 840 |
+
"learning_rate": 4.072081764389457e-05,
|
| 841 |
+
"loss": 1.4655,
|
| 842 |
+
"step": 1190
|
| 843 |
+
},
|
| 844 |
+
{
|
| 845 |
+
"epoch": 0.28444444444444444,
|
| 846 |
+
"grad_norm": 0.5294344425201416,
|
| 847 |
+
"learning_rate": 4.05863367401829e-05,
|
| 848 |
+
"loss": 1.3851,
|
| 849 |
+
"step": 1200
|
| 850 |
+
},
|
| 851 |
+
{
|
| 852 |
+
"epoch": 0.2868148148148148,
|
| 853 |
+
"grad_norm": 0.5410310626029968,
|
| 854 |
+
"learning_rate": 4.045185583647123e-05,
|
| 855 |
+
"loss": 1.5091,
|
| 856 |
+
"step": 1210
|
| 857 |
+
},
|
| 858 |
+
{
|
| 859 |
+
"epoch": 0.2891851851851852,
|
| 860 |
+
"grad_norm": 0.5395278334617615,
|
| 861 |
+
"learning_rate": 4.031737493275955e-05,
|
| 862 |
+
"loss": 1.3487,
|
| 863 |
+
"step": 1220
|
| 864 |
+
},
|
| 865 |
+
{
|
| 866 |
+
"epoch": 0.29155555555555557,
|
| 867 |
+
"grad_norm": 0.637909471988678,
|
| 868 |
+
"learning_rate": 4.018289402904788e-05,
|
| 869 |
+
"loss": 1.5848,
|
| 870 |
+
"step": 1230
|
| 871 |
+
},
|
| 872 |
+
{
|
| 873 |
+
"epoch": 0.2939259259259259,
|
| 874 |
+
"grad_norm": 0.4254130721092224,
|
| 875 |
+
"learning_rate": 4.004841312533621e-05,
|
| 876 |
+
"loss": 1.4186,
|
| 877 |
+
"step": 1240
|
| 878 |
+
},
|
| 879 |
+
{
|
| 880 |
+
"epoch": 0.2962962962962963,
|
| 881 |
+
"grad_norm": 0.5799821019172668,
|
| 882 |
+
"learning_rate": 3.9913932221624536e-05,
|
| 883 |
+
"loss": 1.4289,
|
| 884 |
+
"step": 1250
|
| 885 |
+
},
|
| 886 |
+
{
|
| 887 |
+
"epoch": 0.2986666666666667,
|
| 888 |
+
"grad_norm": 0.43250229954719543,
|
| 889 |
+
"learning_rate": 3.977945131791286e-05,
|
| 890 |
+
"loss": 1.5458,
|
| 891 |
+
"step": 1260
|
| 892 |
+
},
|
| 893 |
+
{
|
| 894 |
+
"epoch": 0.30103703703703705,
|
| 895 |
+
"grad_norm": 0.4722803831100464,
|
| 896 |
+
"learning_rate": 3.964497041420119e-05,
|
| 897 |
+
"loss": 1.4716,
|
| 898 |
+
"step": 1270
|
| 899 |
+
},
|
| 900 |
+
{
|
| 901 |
+
"epoch": 0.3034074074074074,
|
| 902 |
+
"grad_norm": 0.5686700940132141,
|
| 903 |
+
"learning_rate": 3.9510489510489516e-05,
|
| 904 |
+
"loss": 1.6452,
|
| 905 |
+
"step": 1280
|
| 906 |
+
},
|
| 907 |
+
{
|
| 908 |
+
"epoch": 0.30577777777777776,
|
| 909 |
+
"grad_norm": 0.493028461933136,
|
| 910 |
+
"learning_rate": 3.9376008606777844e-05,
|
| 911 |
+
"loss": 1.4516,
|
| 912 |
+
"step": 1290
|
| 913 |
+
},
|
| 914 |
+
{
|
| 915 |
+
"epoch": 0.30814814814814817,
|
| 916 |
+
"grad_norm": 0.5100602507591248,
|
| 917 |
+
"learning_rate": 3.9241527703066166e-05,
|
| 918 |
+
"loss": 1.3366,
|
| 919 |
+
"step": 1300
|
| 920 |
+
},
|
| 921 |
+
{
|
| 922 |
+
"epoch": 0.3105185185185185,
|
| 923 |
+
"grad_norm": 0.6535771489143372,
|
| 924 |
+
"learning_rate": 3.910704679935449e-05,
|
| 925 |
+
"loss": 1.4312,
|
| 926 |
+
"step": 1310
|
| 927 |
+
},
|
| 928 |
+
{
|
| 929 |
+
"epoch": 0.3128888888888889,
|
| 930 |
+
"grad_norm": 0.48823079466819763,
|
| 931 |
+
"learning_rate": 3.8972565895642824e-05,
|
| 932 |
+
"loss": 1.4888,
|
| 933 |
+
"step": 1320
|
| 934 |
+
},
|
| 935 |
+
{
|
| 936 |
+
"epoch": 0.31525925925925924,
|
| 937 |
+
"grad_norm": 0.4459994435310364,
|
| 938 |
+
"learning_rate": 3.8838084991931146e-05,
|
| 939 |
+
"loss": 1.4637,
|
| 940 |
+
"step": 1330
|
| 941 |
+
},
|
| 942 |
+
{
|
| 943 |
+
"epoch": 0.31762962962962965,
|
| 944 |
+
"grad_norm": 0.5344628691673279,
|
| 945 |
+
"learning_rate": 3.8703604088219475e-05,
|
| 946 |
+
"loss": 1.6118,
|
| 947 |
+
"step": 1340
|
| 948 |
+
},
|
| 949 |
+
{
|
| 950 |
+
"epoch": 0.32,
|
| 951 |
+
"grad_norm": 0.44893643260002136,
|
| 952 |
+
"learning_rate": 3.85691231845078e-05,
|
| 953 |
+
"loss": 1.4521,
|
| 954 |
+
"step": 1350
|
| 955 |
+
},
|
| 956 |
+
{
|
| 957 |
+
"epoch": 0.32237037037037036,
|
| 958 |
+
"grad_norm": 0.4381811022758484,
|
| 959 |
+
"learning_rate": 3.8434642280796126e-05,
|
| 960 |
+
"loss": 1.5751,
|
| 961 |
+
"step": 1360
|
| 962 |
+
},
|
| 963 |
+
{
|
| 964 |
+
"epoch": 0.3247407407407407,
|
| 965 |
+
"grad_norm": 0.5791207551956177,
|
| 966 |
+
"learning_rate": 3.8300161377084455e-05,
|
| 967 |
+
"loss": 1.5563,
|
| 968 |
+
"step": 1370
|
| 969 |
+
},
|
| 970 |
+
{
|
| 971 |
+
"epoch": 0.32711111111111113,
|
| 972 |
+
"grad_norm": 0.5584151148796082,
|
| 973 |
+
"learning_rate": 3.8165680473372784e-05,
|
| 974 |
+
"loss": 1.3673,
|
| 975 |
+
"step": 1380
|
| 976 |
+
},
|
| 977 |
+
{
|
| 978 |
+
"epoch": 0.3294814814814815,
|
| 979 |
+
"grad_norm": 0.6110686659812927,
|
| 980 |
+
"learning_rate": 3.8031199569661106e-05,
|
| 981 |
+
"loss": 1.5949,
|
| 982 |
+
"step": 1390
|
| 983 |
+
},
|
| 984 |
+
{
|
| 985 |
+
"epoch": 0.33185185185185184,
|
| 986 |
+
"grad_norm": 0.5519852638244629,
|
| 987 |
+
"learning_rate": 3.7896718665949434e-05,
|
| 988 |
+
"loss": 1.5133,
|
| 989 |
+
"step": 1400
|
| 990 |
+
},
|
| 991 |
+
{
|
| 992 |
+
"epoch": 0.3342222222222222,
|
| 993 |
+
"grad_norm": 0.5732788443565369,
|
| 994 |
+
"learning_rate": 3.776223776223776e-05,
|
| 995 |
+
"loss": 1.4601,
|
| 996 |
+
"step": 1410
|
| 997 |
+
},
|
| 998 |
+
{
|
| 999 |
+
"epoch": 0.3365925925925926,
|
| 1000 |
+
"grad_norm": 0.5789920091629028,
|
| 1001 |
+
"learning_rate": 3.762775685852609e-05,
|
| 1002 |
+
"loss": 1.6461,
|
| 1003 |
+
"step": 1420
|
| 1004 |
+
},
|
| 1005 |
+
{
|
| 1006 |
+
"epoch": 0.33896296296296297,
|
| 1007 |
+
"grad_norm": 0.592776358127594,
|
| 1008 |
+
"learning_rate": 3.7493275954814414e-05,
|
| 1009 |
+
"loss": 1.5558,
|
| 1010 |
+
"step": 1430
|
| 1011 |
+
},
|
| 1012 |
+
{
|
| 1013 |
+
"epoch": 0.3413333333333333,
|
| 1014 |
+
"grad_norm": 0.5435842871665955,
|
| 1015 |
+
"learning_rate": 3.735879505110274e-05,
|
| 1016 |
+
"loss": 1.5095,
|
| 1017 |
+
"step": 1440
|
| 1018 |
+
},
|
| 1019 |
+
{
|
| 1020 |
+
"epoch": 0.3437037037037037,
|
| 1021 |
+
"grad_norm": 0.6474444270133972,
|
| 1022 |
+
"learning_rate": 3.722431414739107e-05,
|
| 1023 |
+
"loss": 1.429,
|
| 1024 |
+
"step": 1450
|
| 1025 |
+
},
|
| 1026 |
+
{
|
| 1027 |
+
"epoch": 0.3460740740740741,
|
| 1028 |
+
"grad_norm": 0.4926964342594147,
|
| 1029 |
+
"learning_rate": 3.70898332436794e-05,
|
| 1030 |
+
"loss": 1.4543,
|
| 1031 |
+
"step": 1460
|
| 1032 |
+
},
|
| 1033 |
+
{
|
| 1034 |
+
"epoch": 0.34844444444444445,
|
| 1035 |
+
"grad_norm": 0.5748719573020935,
|
| 1036 |
+
"learning_rate": 3.695535233996772e-05,
|
| 1037 |
+
"loss": 1.5844,
|
| 1038 |
+
"step": 1470
|
| 1039 |
+
},
|
| 1040 |
+
{
|
| 1041 |
+
"epoch": 0.3508148148148148,
|
| 1042 |
+
"grad_norm": 0.5535377264022827,
|
| 1043 |
+
"learning_rate": 3.682087143625605e-05,
|
| 1044 |
+
"loss": 1.5348,
|
| 1045 |
+
"step": 1480
|
| 1046 |
+
},
|
| 1047 |
+
{
|
| 1048 |
+
"epoch": 0.35318518518518516,
|
| 1049 |
+
"grad_norm": 0.4644632339477539,
|
| 1050 |
+
"learning_rate": 3.668639053254438e-05,
|
| 1051 |
+
"loss": 1.3157,
|
| 1052 |
+
"step": 1490
|
| 1053 |
+
},
|
| 1054 |
+
{
|
| 1055 |
+
"epoch": 0.35555555555555557,
|
| 1056 |
+
"grad_norm": 0.5858569145202637,
|
| 1057 |
+
"learning_rate": 3.655190962883271e-05,
|
| 1058 |
+
"loss": 1.4678,
|
| 1059 |
+
"step": 1500
|
| 1060 |
+
},
|
| 1061 |
+
{
|
| 1062 |
+
"epoch": 0.3579259259259259,
|
| 1063 |
+
"grad_norm": 0.5428529381752014,
|
| 1064 |
+
"learning_rate": 3.641742872512103e-05,
|
| 1065 |
+
"loss": 1.5561,
|
| 1066 |
+
"step": 1510
|
| 1067 |
+
},
|
| 1068 |
+
{
|
| 1069 |
+
"epoch": 0.3602962962962963,
|
| 1070 |
+
"grad_norm": 0.5255948901176453,
|
| 1071 |
+
"learning_rate": 3.628294782140936e-05,
|
| 1072 |
+
"loss": 1.5802,
|
| 1073 |
+
"step": 1520
|
| 1074 |
+
},
|
| 1075 |
+
{
|
| 1076 |
+
"epoch": 0.3626666666666667,
|
| 1077 |
+
"grad_norm": 0.534137487411499,
|
| 1078 |
+
"learning_rate": 3.614846691769769e-05,
|
| 1079 |
+
"loss": 1.5142,
|
| 1080 |
+
"step": 1530
|
| 1081 |
+
},
|
| 1082 |
+
{
|
| 1083 |
+
"epoch": 0.36503703703703705,
|
| 1084 |
+
"grad_norm": 0.5558648705482483,
|
| 1085 |
+
"learning_rate": 3.601398601398602e-05,
|
| 1086 |
+
"loss": 1.5613,
|
| 1087 |
+
"step": 1540
|
| 1088 |
+
},
|
| 1089 |
+
{
|
| 1090 |
+
"epoch": 0.3674074074074074,
|
| 1091 |
+
"grad_norm": 0.5890410542488098,
|
| 1092 |
+
"learning_rate": 3.587950511027434e-05,
|
| 1093 |
+
"loss": 1.4679,
|
| 1094 |
+
"step": 1550
|
| 1095 |
+
},
|
| 1096 |
+
{
|
| 1097 |
+
"epoch": 0.36977777777777776,
|
| 1098 |
+
"grad_norm": 0.4830753803253174,
|
| 1099 |
+
"learning_rate": 3.574502420656267e-05,
|
| 1100 |
+
"loss": 1.4698,
|
| 1101 |
+
"step": 1560
|
| 1102 |
+
},
|
| 1103 |
+
{
|
| 1104 |
+
"epoch": 0.3721481481481482,
|
| 1105 |
+
"grad_norm": 0.5191047787666321,
|
| 1106 |
+
"learning_rate": 3.5610543302851e-05,
|
| 1107 |
+
"loss": 1.528,
|
| 1108 |
+
"step": 1570
|
| 1109 |
+
},
|
| 1110 |
+
{
|
| 1111 |
+
"epoch": 0.37451851851851853,
|
| 1112 |
+
"grad_norm": 0.5316727161407471,
|
| 1113 |
+
"learning_rate": 3.5476062399139326e-05,
|
| 1114 |
+
"loss": 1.4427,
|
| 1115 |
+
"step": 1580
|
| 1116 |
+
},
|
| 1117 |
+
{
|
| 1118 |
+
"epoch": 0.3768888888888889,
|
| 1119 |
+
"grad_norm": 0.553815484046936,
|
| 1120 |
+
"learning_rate": 3.534158149542765e-05,
|
| 1121 |
+
"loss": 1.548,
|
| 1122 |
+
"step": 1590
|
| 1123 |
+
},
|
| 1124 |
+
{
|
| 1125 |
+
"epoch": 0.37925925925925924,
|
| 1126 |
+
"grad_norm": 0.47779569029808044,
|
| 1127 |
+
"learning_rate": 3.520710059171598e-05,
|
| 1128 |
+
"loss": 1.4533,
|
| 1129 |
+
"step": 1600
|
| 1130 |
+
},
|
| 1131 |
+
{
|
| 1132 |
+
"epoch": 0.38162962962962965,
|
| 1133 |
+
"grad_norm": 0.5595371127128601,
|
| 1134 |
+
"learning_rate": 3.5072619688004306e-05,
|
| 1135 |
+
"loss": 1.4503,
|
| 1136 |
+
"step": 1610
|
| 1137 |
+
},
|
| 1138 |
+
{
|
| 1139 |
+
"epoch": 0.384,
|
| 1140 |
+
"grad_norm": 0.5166143774986267,
|
| 1141 |
+
"learning_rate": 3.4938138784292635e-05,
|
| 1142 |
+
"loss": 1.3783,
|
| 1143 |
+
"step": 1620
|
| 1144 |
+
},
|
| 1145 |
+
{
|
| 1146 |
+
"epoch": 0.38637037037037036,
|
| 1147 |
+
"grad_norm": 0.6249716877937317,
|
| 1148 |
+
"learning_rate": 3.480365788058096e-05,
|
| 1149 |
+
"loss": 1.4494,
|
| 1150 |
+
"step": 1630
|
| 1151 |
+
},
|
| 1152 |
+
{
|
| 1153 |
+
"epoch": 0.3887407407407407,
|
| 1154 |
+
"grad_norm": 0.484937846660614,
|
| 1155 |
+
"learning_rate": 3.4669176976869286e-05,
|
| 1156 |
+
"loss": 1.421,
|
| 1157 |
+
"step": 1640
|
| 1158 |
+
},
|
| 1159 |
+
{
|
| 1160 |
+
"epoch": 0.39111111111111113,
|
| 1161 |
+
"grad_norm": 0.5464750528335571,
|
| 1162 |
+
"learning_rate": 3.4534696073157615e-05,
|
| 1163 |
+
"loss": 1.266,
|
| 1164 |
+
"step": 1650
|
| 1165 |
+
},
|
| 1166 |
+
{
|
| 1167 |
+
"epoch": 0.3934814814814815,
|
| 1168 |
+
"grad_norm": 0.48874956369400024,
|
| 1169 |
+
"learning_rate": 3.440021516944594e-05,
|
| 1170 |
+
"loss": 1.5355,
|
| 1171 |
+
"step": 1660
|
| 1172 |
+
},
|
| 1173 |
+
{
|
| 1174 |
+
"epoch": 0.39585185185185184,
|
| 1175 |
+
"grad_norm": 0.47555652260780334,
|
| 1176 |
+
"learning_rate": 3.4265734265734265e-05,
|
| 1177 |
+
"loss": 1.5799,
|
| 1178 |
+
"step": 1670
|
| 1179 |
+
},
|
| 1180 |
+
{
|
| 1181 |
+
"epoch": 0.3982222222222222,
|
| 1182 |
+
"grad_norm": 0.49769505858421326,
|
| 1183 |
+
"learning_rate": 3.4131253362022594e-05,
|
| 1184 |
+
"loss": 1.3748,
|
| 1185 |
+
"step": 1680
|
| 1186 |
+
},
|
| 1187 |
+
{
|
| 1188 |
+
"epoch": 0.4005925925925926,
|
| 1189 |
+
"grad_norm": 0.4664982259273529,
|
| 1190 |
+
"learning_rate": 3.399677245831092e-05,
|
| 1191 |
+
"loss": 1.4894,
|
| 1192 |
+
"step": 1690
|
| 1193 |
+
},
|
| 1194 |
+
{
|
| 1195 |
+
"epoch": 0.40296296296296297,
|
| 1196 |
+
"grad_norm": 0.5216518044471741,
|
| 1197 |
+
"learning_rate": 3.3862291554599245e-05,
|
| 1198 |
+
"loss": 1.4645,
|
| 1199 |
+
"step": 1700
|
| 1200 |
+
},
|
| 1201 |
+
{
|
| 1202 |
+
"epoch": 0.4053333333333333,
|
| 1203 |
+
"grad_norm": 0.6157680749893188,
|
| 1204 |
+
"learning_rate": 3.3727810650887574e-05,
|
| 1205 |
+
"loss": 1.4002,
|
| 1206 |
+
"step": 1710
|
| 1207 |
+
},
|
| 1208 |
+
{
|
| 1209 |
+
"epoch": 0.4077037037037037,
|
| 1210 |
+
"grad_norm": 0.5828937888145447,
|
| 1211 |
+
"learning_rate": 3.35933297471759e-05,
|
| 1212 |
+
"loss": 1.4816,
|
| 1213 |
+
"step": 1720
|
| 1214 |
+
},
|
| 1215 |
+
{
|
| 1216 |
+
"epoch": 0.4100740740740741,
|
| 1217 |
+
"grad_norm": 0.5792407989501953,
|
| 1218 |
+
"learning_rate": 3.345884884346423e-05,
|
| 1219 |
+
"loss": 1.3557,
|
| 1220 |
+
"step": 1730
|
| 1221 |
+
},
|
| 1222 |
+
{
|
| 1223 |
+
"epoch": 0.41244444444444445,
|
| 1224 |
+
"grad_norm": 0.4985092580318451,
|
| 1225 |
+
"learning_rate": 3.3324367939752554e-05,
|
| 1226 |
+
"loss": 1.4572,
|
| 1227 |
+
"step": 1740
|
| 1228 |
+
},
|
| 1229 |
+
{
|
| 1230 |
+
"epoch": 0.4148148148148148,
|
| 1231 |
+
"grad_norm": 0.5901199579238892,
|
| 1232 |
+
"learning_rate": 3.318988703604088e-05,
|
| 1233 |
+
"loss": 1.5292,
|
| 1234 |
+
"step": 1750
|
| 1235 |
+
},
|
| 1236 |
+
{
|
| 1237 |
+
"epoch": 0.41718518518518516,
|
| 1238 |
+
"grad_norm": 0.5087295174598694,
|
| 1239 |
+
"learning_rate": 3.305540613232921e-05,
|
| 1240 |
+
"loss": 1.3405,
|
| 1241 |
+
"step": 1760
|
| 1242 |
+
},
|
| 1243 |
+
{
|
| 1244 |
+
"epoch": 0.41955555555555557,
|
| 1245 |
+
"grad_norm": 0.5455463528633118,
|
| 1246 |
+
"learning_rate": 3.292092522861754e-05,
|
| 1247 |
+
"loss": 1.4262,
|
| 1248 |
+
"step": 1770
|
| 1249 |
+
},
|
| 1250 |
+
{
|
| 1251 |
+
"epoch": 0.4219259259259259,
|
| 1252 |
+
"grad_norm": 0.46563345193862915,
|
| 1253 |
+
"learning_rate": 3.278644432490586e-05,
|
| 1254 |
+
"loss": 1.4328,
|
| 1255 |
+
"step": 1780
|
| 1256 |
+
},
|
| 1257 |
+
{
|
| 1258 |
+
"epoch": 0.4242962962962963,
|
| 1259 |
+
"grad_norm": 0.545524537563324,
|
| 1260 |
+
"learning_rate": 3.265196342119419e-05,
|
| 1261 |
+
"loss": 1.4826,
|
| 1262 |
+
"step": 1790
|
| 1263 |
+
},
|
| 1264 |
+
{
|
| 1265 |
+
"epoch": 0.4266666666666667,
|
| 1266 |
+
"grad_norm": 0.4182009994983673,
|
| 1267 |
+
"learning_rate": 3.251748251748252e-05,
|
| 1268 |
+
"loss": 1.4909,
|
| 1269 |
+
"step": 1800
|
| 1270 |
+
},
|
| 1271 |
+
{
|
| 1272 |
+
"epoch": 0.42903703703703705,
|
| 1273 |
+
"grad_norm": 0.39127054810523987,
|
| 1274 |
+
"learning_rate": 3.238300161377085e-05,
|
| 1275 |
+
"loss": 1.4389,
|
| 1276 |
+
"step": 1810
|
| 1277 |
+
},
|
| 1278 |
+
{
|
| 1279 |
+
"epoch": 0.4314074074074074,
|
| 1280 |
+
"grad_norm": 0.46866652369499207,
|
| 1281 |
+
"learning_rate": 3.224852071005917e-05,
|
| 1282 |
+
"loss": 1.3992,
|
| 1283 |
+
"step": 1820
|
| 1284 |
+
},
|
| 1285 |
+
{
|
| 1286 |
+
"epoch": 0.43377777777777776,
|
| 1287 |
+
"grad_norm": 0.5216823816299438,
|
| 1288 |
+
"learning_rate": 3.21140398063475e-05,
|
| 1289 |
+
"loss": 1.3525,
|
| 1290 |
+
"step": 1830
|
| 1291 |
+
},
|
| 1292 |
+
{
|
| 1293 |
+
"epoch": 0.4361481481481482,
|
| 1294 |
+
"grad_norm": 0.49909713864326477,
|
| 1295 |
+
"learning_rate": 3.197955890263583e-05,
|
| 1296 |
+
"loss": 1.5491,
|
| 1297 |
+
"step": 1840
|
| 1298 |
+
},
|
| 1299 |
+
{
|
| 1300 |
+
"epoch": 0.43851851851851853,
|
| 1301 |
+
"grad_norm": 0.4957892596721649,
|
| 1302 |
+
"learning_rate": 3.184507799892416e-05,
|
| 1303 |
+
"loss": 1.4723,
|
| 1304 |
+
"step": 1850
|
| 1305 |
+
},
|
| 1306 |
+
{
|
| 1307 |
+
"epoch": 0.4408888888888889,
|
| 1308 |
+
"grad_norm": 0.518822431564331,
|
| 1309 |
+
"learning_rate": 3.171059709521248e-05,
|
| 1310 |
+
"loss": 1.535,
|
| 1311 |
+
"step": 1860
|
| 1312 |
+
},
|
| 1313 |
+
{
|
| 1314 |
+
"epoch": 0.44325925925925924,
|
| 1315 |
+
"grad_norm": 0.6380564570426941,
|
| 1316 |
+
"learning_rate": 3.157611619150081e-05,
|
| 1317 |
+
"loss": 1.5652,
|
| 1318 |
+
"step": 1870
|
| 1319 |
+
},
|
| 1320 |
+
{
|
| 1321 |
+
"epoch": 0.44562962962962965,
|
| 1322 |
+
"grad_norm": 0.49906617403030396,
|
| 1323 |
+
"learning_rate": 3.144163528778914e-05,
|
| 1324 |
+
"loss": 1.3624,
|
| 1325 |
+
"step": 1880
|
| 1326 |
+
},
|
| 1327 |
+
{
|
| 1328 |
+
"epoch": 0.448,
|
| 1329 |
+
"grad_norm": 0.5234742760658264,
|
| 1330 |
+
"learning_rate": 3.1307154384077466e-05,
|
| 1331 |
+
"loss": 1.4192,
|
| 1332 |
+
"step": 1890
|
| 1333 |
+
},
|
| 1334 |
+
{
|
| 1335 |
+
"epoch": 0.45037037037037037,
|
| 1336 |
+
"grad_norm": 0.5430870056152344,
|
| 1337 |
+
"learning_rate": 3.117267348036579e-05,
|
| 1338 |
+
"loss": 1.4674,
|
| 1339 |
+
"step": 1900
|
| 1340 |
+
},
|
| 1341 |
+
{
|
| 1342 |
+
"epoch": 0.4527407407407407,
|
| 1343 |
+
"grad_norm": 0.5488291382789612,
|
| 1344 |
+
"learning_rate": 3.103819257665412e-05,
|
| 1345 |
+
"loss": 1.4759,
|
| 1346 |
+
"step": 1910
|
| 1347 |
+
},
|
| 1348 |
+
{
|
| 1349 |
+
"epoch": 0.45511111111111113,
|
| 1350 |
+
"grad_norm": 0.4655541181564331,
|
| 1351 |
+
"learning_rate": 3.0903711672942446e-05,
|
| 1352 |
+
"loss": 1.4559,
|
| 1353 |
+
"step": 1920
|
| 1354 |
+
},
|
| 1355 |
+
{
|
| 1356 |
+
"epoch": 0.4574814814814815,
|
| 1357 |
+
"grad_norm": 0.442128449678421,
|
| 1358 |
+
"learning_rate": 3.0769230769230774e-05,
|
| 1359 |
+
"loss": 1.3129,
|
| 1360 |
+
"step": 1930
|
| 1361 |
+
},
|
| 1362 |
+
{
|
| 1363 |
+
"epoch": 0.45985185185185184,
|
| 1364 |
+
"grad_norm": 0.5909174084663391,
|
| 1365 |
+
"learning_rate": 3.0634749865519096e-05,
|
| 1366 |
+
"loss": 1.5915,
|
| 1367 |
+
"step": 1940
|
| 1368 |
+
},
|
| 1369 |
+
{
|
| 1370 |
+
"epoch": 0.4622222222222222,
|
| 1371 |
+
"grad_norm": 0.41102078557014465,
|
| 1372 |
+
"learning_rate": 3.0500268961807425e-05,
|
| 1373 |
+
"loss": 1.3717,
|
| 1374 |
+
"step": 1950
|
| 1375 |
+
},
|
| 1376 |
+
{
|
| 1377 |
+
"epoch": 0.4645925925925926,
|
| 1378 |
+
"grad_norm": 0.5467662811279297,
|
| 1379 |
+
"learning_rate": 3.036578805809575e-05,
|
| 1380 |
+
"loss": 1.4838,
|
| 1381 |
+
"step": 1960
|
| 1382 |
+
},
|
| 1383 |
+
{
|
| 1384 |
+
"epoch": 0.46696296296296297,
|
| 1385 |
+
"grad_norm": 0.6555057764053345,
|
| 1386 |
+
"learning_rate": 3.023130715438408e-05,
|
| 1387 |
+
"loss": 1.4289,
|
| 1388 |
+
"step": 1970
|
| 1389 |
+
},
|
| 1390 |
+
{
|
| 1391 |
+
"epoch": 0.4693333333333333,
|
| 1392 |
+
"grad_norm": 0.4430755078792572,
|
| 1393 |
+
"learning_rate": 3.0096826250672405e-05,
|
| 1394 |
+
"loss": 1.5556,
|
| 1395 |
+
"step": 1980
|
| 1396 |
+
},
|
| 1397 |
+
{
|
| 1398 |
+
"epoch": 0.4717037037037037,
|
| 1399 |
+
"grad_norm": 0.48016276955604553,
|
| 1400 |
+
"learning_rate": 2.9962345346960734e-05,
|
| 1401 |
+
"loss": 1.5461,
|
| 1402 |
+
"step": 1990
|
| 1403 |
+
},
|
| 1404 |
+
{
|
| 1405 |
+
"epoch": 0.4740740740740741,
|
| 1406 |
+
"grad_norm": 0.5283887982368469,
|
| 1407 |
+
"learning_rate": 2.982786444324906e-05,
|
| 1408 |
+
"loss": 1.4598,
|
| 1409 |
+
"step": 2000
|
| 1410 |
+
},
|
| 1411 |
+
{
|
| 1412 |
+
"epoch": 0.47644444444444445,
|
| 1413 |
+
"grad_norm": 0.5336430668830872,
|
| 1414 |
+
"learning_rate": 2.9693383539537388e-05,
|
| 1415 |
+
"loss": 1.5837,
|
| 1416 |
+
"step": 2010
|
| 1417 |
+
},
|
| 1418 |
+
{
|
| 1419 |
+
"epoch": 0.4788148148148148,
|
| 1420 |
+
"grad_norm": 0.39814135432243347,
|
| 1421 |
+
"learning_rate": 2.9558902635825713e-05,
|
| 1422 |
+
"loss": 1.3496,
|
| 1423 |
+
"step": 2020
|
| 1424 |
+
},
|
| 1425 |
+
{
|
| 1426 |
+
"epoch": 0.48118518518518516,
|
| 1427 |
+
"grad_norm": 0.6095125079154968,
|
| 1428 |
+
"learning_rate": 2.9424421732114042e-05,
|
| 1429 |
+
"loss": 1.5925,
|
| 1430 |
+
"step": 2030
|
| 1431 |
+
},
|
| 1432 |
+
{
|
| 1433 |
+
"epoch": 0.48355555555555557,
|
| 1434 |
+
"grad_norm": 0.5880560874938965,
|
| 1435 |
+
"learning_rate": 2.9289940828402368e-05,
|
| 1436 |
+
"loss": 1.3603,
|
| 1437 |
+
"step": 2040
|
| 1438 |
+
},
|
| 1439 |
+
{
|
| 1440 |
+
"epoch": 0.48592592592592593,
|
| 1441 |
+
"grad_norm": 0.5470516085624695,
|
| 1442 |
+
"learning_rate": 2.9155459924690697e-05,
|
| 1443 |
+
"loss": 1.5022,
|
| 1444 |
+
"step": 2050
|
| 1445 |
+
},
|
| 1446 |
+
{
|
| 1447 |
+
"epoch": 0.4882962962962963,
|
| 1448 |
+
"grad_norm": 0.45742228627204895,
|
| 1449 |
+
"learning_rate": 2.9020979020979022e-05,
|
| 1450 |
+
"loss": 1.4699,
|
| 1451 |
+
"step": 2060
|
| 1452 |
+
},
|
| 1453 |
+
{
|
| 1454 |
+
"epoch": 0.49066666666666664,
|
| 1455 |
+
"grad_norm": 0.5314275622367859,
|
| 1456 |
+
"learning_rate": 2.888649811726735e-05,
|
| 1457 |
+
"loss": 1.5043,
|
| 1458 |
+
"step": 2070
|
| 1459 |
+
},
|
| 1460 |
+
{
|
| 1461 |
+
"epoch": 0.49303703703703705,
|
| 1462 |
+
"grad_norm": 0.5969755053520203,
|
| 1463 |
+
"learning_rate": 2.8752017213555676e-05,
|
| 1464 |
+
"loss": 1.4709,
|
| 1465 |
+
"step": 2080
|
| 1466 |
+
},
|
| 1467 |
+
{
|
| 1468 |
+
"epoch": 0.4954074074074074,
|
| 1469 |
+
"grad_norm": 0.5115885138511658,
|
| 1470 |
+
"learning_rate": 2.8617536309844002e-05,
|
| 1471 |
+
"loss": 1.4031,
|
| 1472 |
+
"step": 2090
|
| 1473 |
+
},
|
| 1474 |
+
{
|
| 1475 |
+
"epoch": 0.49777777777777776,
|
| 1476 |
+
"grad_norm": 0.5907914042472839,
|
| 1477 |
+
"learning_rate": 2.848305540613233e-05,
|
| 1478 |
+
"loss": 1.3509,
|
| 1479 |
+
"step": 2100
|
| 1480 |
+
},
|
| 1481 |
+
{
|
| 1482 |
+
"epoch": 0.5001481481481481,
|
| 1483 |
+
"grad_norm": 0.48430949449539185,
|
| 1484 |
+
"learning_rate": 2.8348574502420656e-05,
|
| 1485 |
+
"loss": 1.4393,
|
| 1486 |
+
"step": 2110
|
| 1487 |
+
},
|
| 1488 |
+
{
|
| 1489 |
+
"epoch": 0.5025185185185185,
|
| 1490 |
+
"grad_norm": 0.5502893328666687,
|
| 1491 |
+
"learning_rate": 2.8214093598708985e-05,
|
| 1492 |
+
"loss": 1.5571,
|
| 1493 |
+
"step": 2120
|
| 1494 |
+
},
|
| 1495 |
+
{
|
| 1496 |
+
"epoch": 0.5048888888888889,
|
| 1497 |
+
"grad_norm": 0.48268720507621765,
|
| 1498 |
+
"learning_rate": 2.807961269499731e-05,
|
| 1499 |
+
"loss": 1.3895,
|
| 1500 |
+
"step": 2130
|
| 1501 |
+
},
|
| 1502 |
+
{
|
| 1503 |
+
"epoch": 0.5072592592592593,
|
| 1504 |
+
"grad_norm": 0.6141895651817322,
|
| 1505 |
+
"learning_rate": 2.794513179128564e-05,
|
| 1506 |
+
"loss": 1.5278,
|
| 1507 |
+
"step": 2140
|
| 1508 |
+
},
|
| 1509 |
+
{
|
| 1510 |
+
"epoch": 0.5096296296296297,
|
| 1511 |
+
"grad_norm": 0.48447638750076294,
|
| 1512 |
+
"learning_rate": 2.7810650887573965e-05,
|
| 1513 |
+
"loss": 1.456,
|
| 1514 |
+
"step": 2150
|
| 1515 |
+
},
|
| 1516 |
+
{
|
| 1517 |
+
"epoch": 0.512,
|
| 1518 |
+
"grad_norm": 0.4536721408367157,
|
| 1519 |
+
"learning_rate": 2.7676169983862293e-05,
|
| 1520 |
+
"loss": 1.4259,
|
| 1521 |
+
"step": 2160
|
| 1522 |
+
},
|
| 1523 |
+
{
|
| 1524 |
+
"epoch": 0.5143703703703704,
|
| 1525 |
+
"grad_norm": 0.5519189238548279,
|
| 1526 |
+
"learning_rate": 2.754168908015062e-05,
|
| 1527 |
+
"loss": 1.508,
|
| 1528 |
+
"step": 2170
|
| 1529 |
+
},
|
| 1530 |
+
{
|
| 1531 |
+
"epoch": 0.5167407407407407,
|
| 1532 |
+
"grad_norm": 0.4641801416873932,
|
| 1533 |
+
"learning_rate": 2.7407208176438948e-05,
|
| 1534 |
+
"loss": 1.5087,
|
| 1535 |
+
"step": 2180
|
| 1536 |
+
},
|
| 1537 |
+
{
|
| 1538 |
+
"epoch": 0.5191111111111111,
|
| 1539 |
+
"grad_norm": 0.5566359162330627,
|
| 1540 |
+
"learning_rate": 2.7272727272727273e-05,
|
| 1541 |
+
"loss": 1.4994,
|
| 1542 |
+
"step": 2190
|
| 1543 |
+
},
|
| 1544 |
+
{
|
| 1545 |
+
"epoch": 0.5214814814814814,
|
| 1546 |
+
"grad_norm": 0.5316601991653442,
|
| 1547 |
+
"learning_rate": 2.7138246369015602e-05,
|
| 1548 |
+
"loss": 1.375,
|
| 1549 |
+
"step": 2200
|
| 1550 |
+
},
|
| 1551 |
+
{
|
| 1552 |
+
"epoch": 0.5238518518518519,
|
| 1553 |
+
"grad_norm": 0.545514702796936,
|
| 1554 |
+
"learning_rate": 2.7003765465303927e-05,
|
| 1555 |
+
"loss": 1.4449,
|
| 1556 |
+
"step": 2210
|
| 1557 |
+
},
|
| 1558 |
+
{
|
| 1559 |
+
"epoch": 0.5262222222222223,
|
| 1560 |
+
"grad_norm": 0.5452851057052612,
|
| 1561 |
+
"learning_rate": 2.6869284561592256e-05,
|
| 1562 |
+
"loss": 1.639,
|
| 1563 |
+
"step": 2220
|
| 1564 |
+
},
|
| 1565 |
+
{
|
| 1566 |
+
"epoch": 0.5285925925925926,
|
| 1567 |
+
"grad_norm": 0.5291896462440491,
|
| 1568 |
+
"learning_rate": 2.673480365788058e-05,
|
| 1569 |
+
"loss": 1.3638,
|
| 1570 |
+
"step": 2230
|
| 1571 |
+
},
|
| 1572 |
+
{
|
| 1573 |
+
"epoch": 0.530962962962963,
|
| 1574 |
+
"grad_norm": 0.4708302319049835,
|
| 1575 |
+
"learning_rate": 2.660032275416891e-05,
|
| 1576 |
+
"loss": 1.1973,
|
| 1577 |
+
"step": 2240
|
| 1578 |
+
},
|
| 1579 |
+
{
|
| 1580 |
+
"epoch": 0.5333333333333333,
|
| 1581 |
+
"grad_norm": 0.4936722218990326,
|
| 1582 |
+
"learning_rate": 2.6465841850457236e-05,
|
| 1583 |
+
"loss": 1.543,
|
| 1584 |
+
"step": 2250
|
| 1585 |
+
},
|
| 1586 |
+
{
|
| 1587 |
+
"epoch": 0.5357037037037037,
|
| 1588 |
+
"grad_norm": 0.5722488760948181,
|
| 1589 |
+
"learning_rate": 2.6331360946745565e-05,
|
| 1590 |
+
"loss": 1.531,
|
| 1591 |
+
"step": 2260
|
| 1592 |
+
},
|
| 1593 |
+
{
|
| 1594 |
+
"epoch": 0.538074074074074,
|
| 1595 |
+
"grad_norm": 0.5386027097702026,
|
| 1596 |
+
"learning_rate": 2.619688004303389e-05,
|
| 1597 |
+
"loss": 1.4335,
|
| 1598 |
+
"step": 2270
|
| 1599 |
+
},
|
| 1600 |
+
{
|
| 1601 |
+
"epoch": 0.5404444444444444,
|
| 1602 |
+
"grad_norm": 0.5803340673446655,
|
| 1603 |
+
"learning_rate": 2.606239913932222e-05,
|
| 1604 |
+
"loss": 1.4049,
|
| 1605 |
+
"step": 2280
|
| 1606 |
+
},
|
| 1607 |
+
{
|
| 1608 |
+
"epoch": 0.5428148148148149,
|
| 1609 |
+
"grad_norm": 0.3970150053501129,
|
| 1610 |
+
"learning_rate": 2.5927918235610544e-05,
|
| 1611 |
+
"loss": 1.3923,
|
| 1612 |
+
"step": 2290
|
| 1613 |
+
},
|
| 1614 |
+
{
|
| 1615 |
+
"epoch": 0.5451851851851852,
|
| 1616 |
+
"grad_norm": 0.45682525634765625,
|
| 1617 |
+
"learning_rate": 2.5793437331898873e-05,
|
| 1618 |
+
"loss": 1.4838,
|
| 1619 |
+
"step": 2300
|
| 1620 |
+
},
|
| 1621 |
+
{
|
| 1622 |
+
"epoch": 0.5475555555555556,
|
| 1623 |
+
"grad_norm": 0.5088069438934326,
|
| 1624 |
+
"learning_rate": 2.56589564281872e-05,
|
| 1625 |
+
"loss": 1.4416,
|
| 1626 |
+
"step": 2310
|
| 1627 |
+
},
|
| 1628 |
+
{
|
| 1629 |
+
"epoch": 0.5499259259259259,
|
| 1630 |
+
"grad_norm": 0.5557109713554382,
|
| 1631 |
+
"learning_rate": 2.5524475524475528e-05,
|
| 1632 |
+
"loss": 1.4099,
|
| 1633 |
+
"step": 2320
|
| 1634 |
+
},
|
| 1635 |
+
{
|
| 1636 |
+
"epoch": 0.5522962962962963,
|
| 1637 |
+
"grad_norm": 0.4954288601875305,
|
| 1638 |
+
"learning_rate": 2.5389994620763853e-05,
|
| 1639 |
+
"loss": 1.3828,
|
| 1640 |
+
"step": 2330
|
| 1641 |
+
},
|
| 1642 |
+
{
|
| 1643 |
+
"epoch": 0.5546666666666666,
|
| 1644 |
+
"grad_norm": 0.5320334434509277,
|
| 1645 |
+
"learning_rate": 2.5255513717052182e-05,
|
| 1646 |
+
"loss": 1.3686,
|
| 1647 |
+
"step": 2340
|
| 1648 |
+
},
|
| 1649 |
+
{
|
| 1650 |
+
"epoch": 0.557037037037037,
|
| 1651 |
+
"grad_norm": 0.511646032333374,
|
| 1652 |
+
"learning_rate": 2.5121032813340507e-05,
|
| 1653 |
+
"loss": 1.4752,
|
| 1654 |
+
"step": 2350
|
| 1655 |
+
},
|
| 1656 |
+
{
|
| 1657 |
+
"epoch": 0.5594074074074074,
|
| 1658 |
+
"grad_norm": 0.4852311909198761,
|
| 1659 |
+
"learning_rate": 2.4986551909628833e-05,
|
| 1660 |
+
"loss": 1.5352,
|
| 1661 |
+
"step": 2360
|
| 1662 |
+
},
|
| 1663 |
+
{
|
| 1664 |
+
"epoch": 0.5617777777777778,
|
| 1665 |
+
"grad_norm": 0.5558280944824219,
|
| 1666 |
+
"learning_rate": 2.485207100591716e-05,
|
| 1667 |
+
"loss": 1.3955,
|
| 1668 |
+
"step": 2370
|
| 1669 |
+
},
|
| 1670 |
+
{
|
| 1671 |
+
"epoch": 0.5641481481481482,
|
| 1672 |
+
"grad_norm": 0.5369210243225098,
|
| 1673 |
+
"learning_rate": 2.4717590102205487e-05,
|
| 1674 |
+
"loss": 1.4265,
|
| 1675 |
+
"step": 2380
|
| 1676 |
+
},
|
| 1677 |
+
{
|
| 1678 |
+
"epoch": 0.5665185185185185,
|
| 1679 |
+
"grad_norm": 0.5134137868881226,
|
| 1680 |
+
"learning_rate": 2.4583109198493816e-05,
|
| 1681 |
+
"loss": 1.3688,
|
| 1682 |
+
"step": 2390
|
| 1683 |
+
},
|
| 1684 |
+
{
|
| 1685 |
+
"epoch": 0.5688888888888889,
|
| 1686 |
+
"grad_norm": 0.47109952569007874,
|
| 1687 |
+
"learning_rate": 2.444862829478214e-05,
|
| 1688 |
+
"loss": 1.4207,
|
| 1689 |
+
"step": 2400
|
| 1690 |
+
},
|
| 1691 |
+
{
|
| 1692 |
+
"epoch": 0.5712592592592592,
|
| 1693 |
+
"grad_norm": 0.4982026517391205,
|
| 1694 |
+
"learning_rate": 2.431414739107047e-05,
|
| 1695 |
+
"loss": 1.4249,
|
| 1696 |
+
"step": 2410
|
| 1697 |
+
},
|
| 1698 |
+
{
|
| 1699 |
+
"epoch": 0.5736296296296296,
|
| 1700 |
+
"grad_norm": 0.5209967494010925,
|
| 1701 |
+
"learning_rate": 2.4179666487358796e-05,
|
| 1702 |
+
"loss": 1.4708,
|
| 1703 |
+
"step": 2420
|
| 1704 |
+
},
|
| 1705 |
+
{
|
| 1706 |
+
"epoch": 0.576,
|
| 1707 |
+
"grad_norm": 0.5762905478477478,
|
| 1708 |
+
"learning_rate": 2.4045185583647124e-05,
|
| 1709 |
+
"loss": 1.4194,
|
| 1710 |
+
"step": 2430
|
| 1711 |
+
},
|
| 1712 |
+
{
|
| 1713 |
+
"epoch": 0.5783703703703704,
|
| 1714 |
+
"grad_norm": 0.4918428659439087,
|
| 1715 |
+
"learning_rate": 2.391070467993545e-05,
|
| 1716 |
+
"loss": 1.6279,
|
| 1717 |
+
"step": 2440
|
| 1718 |
+
},
|
| 1719 |
+
{
|
| 1720 |
+
"epoch": 0.5807407407407408,
|
| 1721 |
+
"grad_norm": 0.5050658583641052,
|
| 1722 |
+
"learning_rate": 2.377622377622378e-05,
|
| 1723 |
+
"loss": 1.5029,
|
| 1724 |
+
"step": 2450
|
| 1725 |
+
},
|
| 1726 |
+
{
|
| 1727 |
+
"epoch": 0.5831111111111111,
|
| 1728 |
+
"grad_norm": 0.49715667963027954,
|
| 1729 |
+
"learning_rate": 2.3641742872512104e-05,
|
| 1730 |
+
"loss": 1.4228,
|
| 1731 |
+
"step": 2460
|
| 1732 |
+
},
|
| 1733 |
+
{
|
| 1734 |
+
"epoch": 0.5854814814814815,
|
| 1735 |
+
"grad_norm": 0.4800516664981842,
|
| 1736 |
+
"learning_rate": 2.3507261968800433e-05,
|
| 1737 |
+
"loss": 1.3595,
|
| 1738 |
+
"step": 2470
|
| 1739 |
+
},
|
| 1740 |
+
{
|
| 1741 |
+
"epoch": 0.5878518518518518,
|
| 1742 |
+
"grad_norm": 0.5617285966873169,
|
| 1743 |
+
"learning_rate": 2.337278106508876e-05,
|
| 1744 |
+
"loss": 1.5974,
|
| 1745 |
+
"step": 2480
|
| 1746 |
+
},
|
| 1747 |
+
{
|
| 1748 |
+
"epoch": 0.5902222222222222,
|
| 1749 |
+
"grad_norm": 0.5133258700370789,
|
| 1750 |
+
"learning_rate": 2.3238300161377087e-05,
|
| 1751 |
+
"loss": 1.4437,
|
| 1752 |
+
"step": 2490
|
| 1753 |
+
},
|
| 1754 |
+
{
|
| 1755 |
+
"epoch": 0.5925925925925926,
|
| 1756 |
+
"grad_norm": 0.5644205212593079,
|
| 1757 |
+
"learning_rate": 2.3103819257665413e-05,
|
| 1758 |
+
"loss": 1.4633,
|
| 1759 |
+
"step": 2500
|
| 1760 |
+
},
|
| 1761 |
+
{
|
| 1762 |
+
"epoch": 0.5949629629629629,
|
| 1763 |
+
"grad_norm": 0.5411229133605957,
|
| 1764 |
+
"learning_rate": 2.296933835395374e-05,
|
| 1765 |
+
"loss": 1.3376,
|
| 1766 |
+
"step": 2510
|
| 1767 |
+
},
|
| 1768 |
+
{
|
| 1769 |
+
"epoch": 0.5973333333333334,
|
| 1770 |
+
"grad_norm": 0.5555963516235352,
|
| 1771 |
+
"learning_rate": 2.2834857450242067e-05,
|
| 1772 |
+
"loss": 1.5361,
|
| 1773 |
+
"step": 2520
|
| 1774 |
+
},
|
| 1775 |
+
{
|
| 1776 |
+
"epoch": 0.5997037037037037,
|
| 1777 |
+
"grad_norm": 0.48606014251708984,
|
| 1778 |
+
"learning_rate": 2.2700376546530396e-05,
|
| 1779 |
+
"loss": 1.4386,
|
| 1780 |
+
"step": 2530
|
| 1781 |
+
},
|
| 1782 |
+
{
|
| 1783 |
+
"epoch": 0.6020740740740741,
|
| 1784 |
+
"grad_norm": 0.4902474582195282,
|
| 1785 |
+
"learning_rate": 2.256589564281872e-05,
|
| 1786 |
+
"loss": 1.4536,
|
| 1787 |
+
"step": 2540
|
| 1788 |
+
},
|
| 1789 |
+
{
|
| 1790 |
+
"epoch": 0.6044444444444445,
|
| 1791 |
+
"grad_norm": 0.5565341114997864,
|
| 1792 |
+
"learning_rate": 2.243141473910705e-05,
|
| 1793 |
+
"loss": 1.4801,
|
| 1794 |
+
"step": 2550
|
| 1795 |
+
},
|
| 1796 |
+
{
|
| 1797 |
+
"epoch": 0.6068148148148148,
|
| 1798 |
+
"grad_norm": 0.5383167862892151,
|
| 1799 |
+
"learning_rate": 2.2296933835395375e-05,
|
| 1800 |
+
"loss": 1.3528,
|
| 1801 |
+
"step": 2560
|
| 1802 |
+
},
|
| 1803 |
+
{
|
| 1804 |
+
"epoch": 0.6091851851851852,
|
| 1805 |
+
"grad_norm": 0.4610041379928589,
|
| 1806 |
+
"learning_rate": 2.2162452931683704e-05,
|
| 1807 |
+
"loss": 1.5063,
|
| 1808 |
+
"step": 2570
|
| 1809 |
+
},
|
| 1810 |
+
{
|
| 1811 |
+
"epoch": 0.6115555555555555,
|
| 1812 |
+
"grad_norm": 0.5119171142578125,
|
| 1813 |
+
"learning_rate": 2.202797202797203e-05,
|
| 1814 |
+
"loss": 1.4559,
|
| 1815 |
+
"step": 2580
|
| 1816 |
+
},
|
| 1817 |
+
{
|
| 1818 |
+
"epoch": 0.6139259259259259,
|
| 1819 |
+
"grad_norm": 0.46622559428215027,
|
| 1820 |
+
"learning_rate": 2.189349112426036e-05,
|
| 1821 |
+
"loss": 1.485,
|
| 1822 |
+
"step": 2590
|
| 1823 |
+
},
|
| 1824 |
+
{
|
| 1825 |
+
"epoch": 0.6162962962962963,
|
| 1826 |
+
"grad_norm": 0.5610603094100952,
|
| 1827 |
+
"learning_rate": 2.1759010220548684e-05,
|
| 1828 |
+
"loss": 1.3863,
|
| 1829 |
+
"step": 2600
|
| 1830 |
+
},
|
| 1831 |
+
{
|
| 1832 |
+
"epoch": 0.6186666666666667,
|
| 1833 |
+
"grad_norm": 0.5185586214065552,
|
| 1834 |
+
"learning_rate": 2.162452931683701e-05,
|
| 1835 |
+
"loss": 1.4574,
|
| 1836 |
+
"step": 2610
|
| 1837 |
+
},
|
| 1838 |
+
{
|
| 1839 |
+
"epoch": 0.621037037037037,
|
| 1840 |
+
"grad_norm": 0.5091121196746826,
|
| 1841 |
+
"learning_rate": 2.1490048413125338e-05,
|
| 1842 |
+
"loss": 1.3745,
|
| 1843 |
+
"step": 2620
|
| 1844 |
+
},
|
| 1845 |
+
{
|
| 1846 |
+
"epoch": 0.6234074074074074,
|
| 1847 |
+
"grad_norm": 0.39684295654296875,
|
| 1848 |
+
"learning_rate": 2.1355567509413664e-05,
|
| 1849 |
+
"loss": 1.5785,
|
| 1850 |
+
"step": 2630
|
| 1851 |
+
},
|
| 1852 |
+
{
|
| 1853 |
+
"epoch": 0.6257777777777778,
|
| 1854 |
+
"grad_norm": 0.5499323606491089,
|
| 1855 |
+
"learning_rate": 2.1221086605701993e-05,
|
| 1856 |
+
"loss": 1.4926,
|
| 1857 |
+
"step": 2640
|
| 1858 |
+
},
|
| 1859 |
+
{
|
| 1860 |
+
"epoch": 0.6281481481481481,
|
| 1861 |
+
"grad_norm": 0.448824942111969,
|
| 1862 |
+
"learning_rate": 2.1086605701990318e-05,
|
| 1863 |
+
"loss": 1.5199,
|
| 1864 |
+
"step": 2650
|
| 1865 |
+
},
|
| 1866 |
+
{
|
| 1867 |
+
"epoch": 0.6305185185185185,
|
| 1868 |
+
"grad_norm": 0.5647756457328796,
|
| 1869 |
+
"learning_rate": 2.0952124798278647e-05,
|
| 1870 |
+
"loss": 1.493,
|
| 1871 |
+
"step": 2660
|
| 1872 |
+
},
|
| 1873 |
+
{
|
| 1874 |
+
"epoch": 0.6328888888888888,
|
| 1875 |
+
"grad_norm": 0.5426878929138184,
|
| 1876 |
+
"learning_rate": 2.0817643894566972e-05,
|
| 1877 |
+
"loss": 1.4232,
|
| 1878 |
+
"step": 2670
|
| 1879 |
+
},
|
| 1880 |
+
{
|
| 1881 |
+
"epoch": 0.6352592592592593,
|
| 1882 |
+
"grad_norm": 0.5105384588241577,
|
| 1883 |
+
"learning_rate": 2.0683162990855298e-05,
|
| 1884 |
+
"loss": 1.5136,
|
| 1885 |
+
"step": 2680
|
| 1886 |
+
},
|
| 1887 |
+
{
|
| 1888 |
+
"epoch": 0.6376296296296297,
|
| 1889 |
+
"grad_norm": 0.5488259792327881,
|
| 1890 |
+
"learning_rate": 2.0548682087143627e-05,
|
| 1891 |
+
"loss": 1.4877,
|
| 1892 |
+
"step": 2690
|
| 1893 |
+
},
|
| 1894 |
+
{
|
| 1895 |
+
"epoch": 0.64,
|
| 1896 |
+
"grad_norm": 0.47102248668670654,
|
| 1897 |
+
"learning_rate": 2.0414201183431952e-05,
|
| 1898 |
+
"loss": 1.2976,
|
| 1899 |
+
"step": 2700
|
| 1900 |
+
},
|
| 1901 |
+
{
|
| 1902 |
+
"epoch": 0.6423703703703704,
|
| 1903 |
+
"grad_norm": 0.4708435535430908,
|
| 1904 |
+
"learning_rate": 2.027972027972028e-05,
|
| 1905 |
+
"loss": 1.4773,
|
| 1906 |
+
"step": 2710
|
| 1907 |
+
},
|
| 1908 |
+
{
|
| 1909 |
+
"epoch": 0.6447407407407407,
|
| 1910 |
+
"grad_norm": 0.5073569416999817,
|
| 1911 |
+
"learning_rate": 2.0145239376008606e-05,
|
| 1912 |
+
"loss": 1.3551,
|
| 1913 |
+
"step": 2720
|
| 1914 |
+
},
|
| 1915 |
+
{
|
| 1916 |
+
"epoch": 0.6471111111111111,
|
| 1917 |
+
"grad_norm": 0.4934346377849579,
|
| 1918 |
+
"learning_rate": 2.0010758472296935e-05,
|
| 1919 |
+
"loss": 1.2423,
|
| 1920 |
+
"step": 2730
|
| 1921 |
+
},
|
| 1922 |
+
{
|
| 1923 |
+
"epoch": 0.6494814814814814,
|
| 1924 |
+
"grad_norm": 0.5030198097229004,
|
| 1925 |
+
"learning_rate": 1.987627756858526e-05,
|
| 1926 |
+
"loss": 1.4272,
|
| 1927 |
+
"step": 2740
|
| 1928 |
+
},
|
| 1929 |
+
{
|
| 1930 |
+
"epoch": 0.6518518518518519,
|
| 1931 |
+
"grad_norm": 0.4713825583457947,
|
| 1932 |
+
"learning_rate": 1.974179666487359e-05,
|
| 1933 |
+
"loss": 1.4898,
|
| 1934 |
+
"step": 2750
|
| 1935 |
+
},
|
| 1936 |
+
{
|
| 1937 |
+
"epoch": 0.6542222222222223,
|
| 1938 |
+
"grad_norm": 0.430649995803833,
|
| 1939 |
+
"learning_rate": 1.9607315761161915e-05,
|
| 1940 |
+
"loss": 1.3365,
|
| 1941 |
+
"step": 2760
|
| 1942 |
+
},
|
| 1943 |
+
{
|
| 1944 |
+
"epoch": 0.6565925925925926,
|
| 1945 |
+
"grad_norm": 0.58051598072052,
|
| 1946 |
+
"learning_rate": 1.9472834857450244e-05,
|
| 1947 |
+
"loss": 1.4567,
|
| 1948 |
+
"step": 2770
|
| 1949 |
+
},
|
| 1950 |
+
{
|
| 1951 |
+
"epoch": 0.658962962962963,
|
| 1952 |
+
"grad_norm": 0.46255138516426086,
|
| 1953 |
+
"learning_rate": 1.933835395373857e-05,
|
| 1954 |
+
"loss": 1.5205,
|
| 1955 |
+
"step": 2780
|
| 1956 |
+
},
|
| 1957 |
+
{
|
| 1958 |
+
"epoch": 0.6613333333333333,
|
| 1959 |
+
"grad_norm": 0.5674681663513184,
|
| 1960 |
+
"learning_rate": 1.9203873050026898e-05,
|
| 1961 |
+
"loss": 1.3594,
|
| 1962 |
+
"step": 2790
|
| 1963 |
+
},
|
| 1964 |
+
{
|
| 1965 |
+
"epoch": 0.6637037037037037,
|
| 1966 |
+
"grad_norm": 0.5063351392745972,
|
| 1967 |
+
"learning_rate": 1.9069392146315223e-05,
|
| 1968 |
+
"loss": 1.2901,
|
| 1969 |
+
"step": 2800
|
| 1970 |
+
},
|
| 1971 |
+
{
|
| 1972 |
+
"epoch": 0.666074074074074,
|
| 1973 |
+
"grad_norm": 0.4963226914405823,
|
| 1974 |
+
"learning_rate": 1.8934911242603552e-05,
|
| 1975 |
+
"loss": 1.4437,
|
| 1976 |
+
"step": 2810
|
| 1977 |
+
},
|
| 1978 |
+
{
|
| 1979 |
+
"epoch": 0.6684444444444444,
|
| 1980 |
+
"grad_norm": 0.5070900917053223,
|
| 1981 |
+
"learning_rate": 1.8800430338891878e-05,
|
| 1982 |
+
"loss": 1.376,
|
| 1983 |
+
"step": 2820
|
| 1984 |
+
},
|
| 1985 |
+
{
|
| 1986 |
+
"epoch": 0.6708148148148149,
|
| 1987 |
+
"grad_norm": 0.5724377036094666,
|
| 1988 |
+
"learning_rate": 1.8665949435180206e-05,
|
| 1989 |
+
"loss": 1.5226,
|
| 1990 |
+
"step": 2830
|
| 1991 |
+
},
|
| 1992 |
+
{
|
| 1993 |
+
"epoch": 0.6731851851851852,
|
| 1994 |
+
"grad_norm": 0.5261855125427246,
|
| 1995 |
+
"learning_rate": 1.8531468531468532e-05,
|
| 1996 |
+
"loss": 1.438,
|
| 1997 |
+
"step": 2840
|
| 1998 |
+
},
|
| 1999 |
+
{
|
| 2000 |
+
"epoch": 0.6755555555555556,
|
| 2001 |
+
"grad_norm": 0.5292350053787231,
|
| 2002 |
+
"learning_rate": 1.839698762775686e-05,
|
| 2003 |
+
"loss": 1.4071,
|
| 2004 |
+
"step": 2850
|
| 2005 |
+
},
|
| 2006 |
+
{
|
| 2007 |
+
"epoch": 0.6779259259259259,
|
| 2008 |
+
"grad_norm": 0.4596816301345825,
|
| 2009 |
+
"learning_rate": 1.8262506724045186e-05,
|
| 2010 |
+
"loss": 1.337,
|
| 2011 |
+
"step": 2860
|
| 2012 |
+
},
|
| 2013 |
+
{
|
| 2014 |
+
"epoch": 0.6802962962962963,
|
| 2015 |
+
"grad_norm": 0.5225928425788879,
|
| 2016 |
+
"learning_rate": 1.8128025820333515e-05,
|
| 2017 |
+
"loss": 1.4363,
|
| 2018 |
+
"step": 2870
|
| 2019 |
+
},
|
| 2020 |
+
{
|
| 2021 |
+
"epoch": 0.6826666666666666,
|
| 2022 |
+
"grad_norm": 0.49359938502311707,
|
| 2023 |
+
"learning_rate": 1.799354491662184e-05,
|
| 2024 |
+
"loss": 1.3388,
|
| 2025 |
+
"step": 2880
|
| 2026 |
+
},
|
| 2027 |
+
{
|
| 2028 |
+
"epoch": 0.685037037037037,
|
| 2029 |
+
"grad_norm": 0.5156022906303406,
|
| 2030 |
+
"learning_rate": 1.785906401291017e-05,
|
| 2031 |
+
"loss": 1.4515,
|
| 2032 |
+
"step": 2890
|
| 2033 |
+
},
|
| 2034 |
+
{
|
| 2035 |
+
"epoch": 0.6874074074074074,
|
| 2036 |
+
"grad_norm": 0.5047289133071899,
|
| 2037 |
+
"learning_rate": 1.7724583109198495e-05,
|
| 2038 |
+
"loss": 1.4584,
|
| 2039 |
+
"step": 2900
|
| 2040 |
+
},
|
| 2041 |
+
{
|
| 2042 |
+
"epoch": 0.6897777777777778,
|
| 2043 |
+
"grad_norm": 0.4975475072860718,
|
| 2044 |
+
"learning_rate": 1.7590102205486824e-05,
|
| 2045 |
+
"loss": 1.5042,
|
| 2046 |
+
"step": 2910
|
| 2047 |
+
},
|
| 2048 |
+
{
|
| 2049 |
+
"epoch": 0.6921481481481482,
|
| 2050 |
+
"grad_norm": 0.5997641086578369,
|
| 2051 |
+
"learning_rate": 1.745562130177515e-05,
|
| 2052 |
+
"loss": 1.5658,
|
| 2053 |
+
"step": 2920
|
| 2054 |
+
},
|
| 2055 |
+
{
|
| 2056 |
+
"epoch": 0.6945185185185185,
|
| 2057 |
+
"grad_norm": 0.5376483201980591,
|
| 2058 |
+
"learning_rate": 1.7321140398063478e-05,
|
| 2059 |
+
"loss": 1.3429,
|
| 2060 |
+
"step": 2930
|
| 2061 |
+
},
|
| 2062 |
+
{
|
| 2063 |
+
"epoch": 0.6968888888888889,
|
| 2064 |
+
"grad_norm": 0.4973870813846588,
|
| 2065 |
+
"learning_rate": 1.7186659494351803e-05,
|
| 2066 |
+
"loss": 1.4287,
|
| 2067 |
+
"step": 2940
|
| 2068 |
+
},
|
| 2069 |
+
{
|
| 2070 |
+
"epoch": 0.6992592592592592,
|
| 2071 |
+
"grad_norm": 0.5504077076911926,
|
| 2072 |
+
"learning_rate": 1.7052178590640132e-05,
|
| 2073 |
+
"loss": 1.5161,
|
| 2074 |
+
"step": 2950
|
| 2075 |
+
},
|
| 2076 |
+
{
|
| 2077 |
+
"epoch": 0.7016296296296296,
|
| 2078 |
+
"grad_norm": 0.4603710174560547,
|
| 2079 |
+
"learning_rate": 1.6917697686928457e-05,
|
| 2080 |
+
"loss": 1.463,
|
| 2081 |
+
"step": 2960
|
| 2082 |
+
},
|
| 2083 |
+
{
|
| 2084 |
+
"epoch": 0.704,
|
| 2085 |
+
"grad_norm": 0.5116856694221497,
|
| 2086 |
+
"learning_rate": 1.6783216783216786e-05,
|
| 2087 |
+
"loss": 1.3862,
|
| 2088 |
+
"step": 2970
|
| 2089 |
+
},
|
| 2090 |
+
{
|
| 2091 |
+
"epoch": 0.7063703703703703,
|
| 2092 |
+
"grad_norm": 0.49981990456581116,
|
| 2093 |
+
"learning_rate": 1.6648735879505112e-05,
|
| 2094 |
+
"loss": 1.4209,
|
| 2095 |
+
"step": 2980
|
| 2096 |
+
},
|
| 2097 |
+
{
|
| 2098 |
+
"epoch": 0.7087407407407408,
|
| 2099 |
+
"grad_norm": 0.5085658431053162,
|
| 2100 |
+
"learning_rate": 1.651425497579344e-05,
|
| 2101 |
+
"loss": 1.4031,
|
| 2102 |
+
"step": 2990
|
| 2103 |
+
},
|
| 2104 |
+
{
|
| 2105 |
+
"epoch": 0.7111111111111111,
|
| 2106 |
+
"grad_norm": 0.5499709844589233,
|
| 2107 |
+
"learning_rate": 1.6379774072081766e-05,
|
| 2108 |
+
"loss": 1.5093,
|
| 2109 |
+
"step": 3000
|
| 2110 |
+
},
|
| 2111 |
+
{
|
| 2112 |
+
"epoch": 0.7134814814814815,
|
| 2113 |
+
"grad_norm": 0.5997831225395203,
|
| 2114 |
+
"learning_rate": 1.6245293168370095e-05,
|
| 2115 |
+
"loss": 1.517,
|
| 2116 |
+
"step": 3010
|
| 2117 |
+
},
|
| 2118 |
+
{
|
| 2119 |
+
"epoch": 0.7158518518518519,
|
| 2120 |
+
"grad_norm": 0.5795171856880188,
|
| 2121 |
+
"learning_rate": 1.611081226465842e-05,
|
| 2122 |
+
"loss": 1.4279,
|
| 2123 |
+
"step": 3020
|
| 2124 |
+
},
|
| 2125 |
+
{
|
| 2126 |
+
"epoch": 0.7182222222222222,
|
| 2127 |
+
"grad_norm": 0.5227158069610596,
|
| 2128 |
+
"learning_rate": 1.5976331360946746e-05,
|
| 2129 |
+
"loss": 1.5101,
|
| 2130 |
+
"step": 3030
|
| 2131 |
+
},
|
| 2132 |
+
{
|
| 2133 |
+
"epoch": 0.7205925925925926,
|
| 2134 |
+
"grad_norm": 0.6352266669273376,
|
| 2135 |
+
"learning_rate": 1.5841850457235075e-05,
|
| 2136 |
+
"loss": 1.4849,
|
| 2137 |
+
"step": 3040
|
| 2138 |
+
},
|
| 2139 |
+
{
|
| 2140 |
+
"epoch": 0.7229629629629629,
|
| 2141 |
+
"grad_norm": 0.6044921278953552,
|
| 2142 |
+
"learning_rate": 1.57073695535234e-05,
|
| 2143 |
+
"loss": 1.5239,
|
| 2144 |
+
"step": 3050
|
| 2145 |
+
},
|
| 2146 |
+
{
|
| 2147 |
+
"epoch": 0.7253333333333334,
|
| 2148 |
+
"grad_norm": 0.4992562532424927,
|
| 2149 |
+
"learning_rate": 1.557288864981173e-05,
|
| 2150 |
+
"loss": 1.37,
|
| 2151 |
+
"step": 3060
|
| 2152 |
+
},
|
| 2153 |
+
{
|
| 2154 |
+
"epoch": 0.7277037037037037,
|
| 2155 |
+
"grad_norm": 0.5439409017562866,
|
| 2156 |
+
"learning_rate": 1.5438407746100054e-05,
|
| 2157 |
+
"loss": 1.3724,
|
| 2158 |
+
"step": 3070
|
| 2159 |
+
},
|
| 2160 |
+
{
|
| 2161 |
+
"epoch": 0.7300740740740741,
|
| 2162 |
+
"grad_norm": 0.5415698289871216,
|
| 2163 |
+
"learning_rate": 1.530392684238838e-05,
|
| 2164 |
+
"loss": 1.4562,
|
| 2165 |
+
"step": 3080
|
| 2166 |
+
},
|
| 2167 |
+
{
|
| 2168 |
+
"epoch": 0.7324444444444445,
|
| 2169 |
+
"grad_norm": 0.4034167230129242,
|
| 2170 |
+
"learning_rate": 1.5169445938676707e-05,
|
| 2171 |
+
"loss": 1.3212,
|
| 2172 |
+
"step": 3090
|
| 2173 |
+
},
|
| 2174 |
+
{
|
| 2175 |
+
"epoch": 0.7348148148148148,
|
| 2176 |
+
"grad_norm": 0.5307872891426086,
|
| 2177 |
+
"learning_rate": 1.5034965034965034e-05,
|
| 2178 |
+
"loss": 1.4165,
|
| 2179 |
+
"step": 3100
|
| 2180 |
+
},
|
| 2181 |
+
{
|
| 2182 |
+
"epoch": 0.7371851851851852,
|
| 2183 |
+
"grad_norm": 0.5271874666213989,
|
| 2184 |
+
"learning_rate": 1.4900484131253361e-05,
|
| 2185 |
+
"loss": 1.4077,
|
| 2186 |
+
"step": 3110
|
| 2187 |
+
},
|
| 2188 |
+
{
|
| 2189 |
+
"epoch": 0.7395555555555555,
|
| 2190 |
+
"grad_norm": 0.5995745658874512,
|
| 2191 |
+
"learning_rate": 1.4766003227541688e-05,
|
| 2192 |
+
"loss": 1.4732,
|
| 2193 |
+
"step": 3120
|
| 2194 |
+
},
|
| 2195 |
+
{
|
| 2196 |
+
"epoch": 0.7419259259259259,
|
| 2197 |
+
"grad_norm": 0.45358097553253174,
|
| 2198 |
+
"learning_rate": 1.4631522323830015e-05,
|
| 2199 |
+
"loss": 1.4083,
|
| 2200 |
+
"step": 3130
|
| 2201 |
+
},
|
| 2202 |
+
{
|
| 2203 |
+
"epoch": 0.7442962962962963,
|
| 2204 |
+
"grad_norm": 0.47864630818367004,
|
| 2205 |
+
"learning_rate": 1.4497041420118343e-05,
|
| 2206 |
+
"loss": 1.4363,
|
| 2207 |
+
"step": 3140
|
| 2208 |
+
},
|
| 2209 |
+
{
|
| 2210 |
+
"epoch": 0.7466666666666667,
|
| 2211 |
+
"grad_norm": 0.46107572317123413,
|
| 2212 |
+
"learning_rate": 1.436256051640667e-05,
|
| 2213 |
+
"loss": 1.5163,
|
| 2214 |
+
"step": 3150
|
| 2215 |
+
},
|
| 2216 |
+
{
|
| 2217 |
+
"epoch": 0.7490370370370371,
|
| 2218 |
+
"grad_norm": 0.5413241386413574,
|
| 2219 |
+
"learning_rate": 1.4228079612694997e-05,
|
| 2220 |
+
"loss": 1.3225,
|
| 2221 |
+
"step": 3160
|
| 2222 |
+
},
|
| 2223 |
+
{
|
| 2224 |
+
"epoch": 0.7514074074074074,
|
| 2225 |
+
"grad_norm": 0.4649742841720581,
|
| 2226 |
+
"learning_rate": 1.4093598708983324e-05,
|
| 2227 |
+
"loss": 1.4101,
|
| 2228 |
+
"step": 3170
|
| 2229 |
+
},
|
| 2230 |
+
{
|
| 2231 |
+
"epoch": 0.7537777777777778,
|
| 2232 |
+
"grad_norm": 0.5219136476516724,
|
| 2233 |
+
"learning_rate": 1.3959117805271651e-05,
|
| 2234 |
+
"loss": 1.4523,
|
| 2235 |
+
"step": 3180
|
| 2236 |
+
},
|
| 2237 |
+
{
|
| 2238 |
+
"epoch": 0.7561481481481481,
|
| 2239 |
+
"grad_norm": 0.5591155886650085,
|
| 2240 |
+
"learning_rate": 1.3824636901559978e-05,
|
| 2241 |
+
"loss": 1.5966,
|
| 2242 |
+
"step": 3190
|
| 2243 |
+
},
|
| 2244 |
+
{
|
| 2245 |
+
"epoch": 0.7585185185185185,
|
| 2246 |
+
"grad_norm": 0.5293004512786865,
|
| 2247 |
+
"learning_rate": 1.3690155997848305e-05,
|
| 2248 |
+
"loss": 1.4613,
|
| 2249 |
+
"step": 3200
|
| 2250 |
+
},
|
| 2251 |
+
{
|
| 2252 |
+
"epoch": 0.7608888888888888,
|
| 2253 |
+
"grad_norm": 0.46828821301460266,
|
| 2254 |
+
"learning_rate": 1.3555675094136632e-05,
|
| 2255 |
+
"loss": 1.5187,
|
| 2256 |
+
"step": 3210
|
| 2257 |
+
},
|
| 2258 |
+
{
|
| 2259 |
+
"epoch": 0.7632592592592593,
|
| 2260 |
+
"grad_norm": 0.4590572416782379,
|
| 2261 |
+
"learning_rate": 1.342119419042496e-05,
|
| 2262 |
+
"loss": 1.4418,
|
| 2263 |
+
"step": 3220
|
| 2264 |
+
},
|
| 2265 |
+
{
|
| 2266 |
+
"epoch": 0.7656296296296297,
|
| 2267 |
+
"grad_norm": 0.6020212769508362,
|
| 2268 |
+
"learning_rate": 1.3286713286713287e-05,
|
| 2269 |
+
"loss": 1.5557,
|
| 2270 |
+
"step": 3230
|
| 2271 |
+
},
|
| 2272 |
+
{
|
| 2273 |
+
"epoch": 0.768,
|
| 2274 |
+
"grad_norm": 0.542536199092865,
|
| 2275 |
+
"learning_rate": 1.3152232383001614e-05,
|
| 2276 |
+
"loss": 1.3689,
|
| 2277 |
+
"step": 3240
|
| 2278 |
+
},
|
| 2279 |
+
{
|
| 2280 |
+
"epoch": 0.7703703703703704,
|
| 2281 |
+
"grad_norm": 0.5394562482833862,
|
| 2282 |
+
"learning_rate": 1.3017751479289941e-05,
|
| 2283 |
+
"loss": 1.4272,
|
| 2284 |
+
"step": 3250
|
| 2285 |
+
},
|
| 2286 |
+
{
|
| 2287 |
+
"epoch": 0.7727407407407407,
|
| 2288 |
+
"grad_norm": 0.38109496235847473,
|
| 2289 |
+
"learning_rate": 1.2883270575578268e-05,
|
| 2290 |
+
"loss": 1.419,
|
| 2291 |
+
"step": 3260
|
| 2292 |
+
},
|
| 2293 |
+
{
|
| 2294 |
+
"epoch": 0.7751111111111111,
|
| 2295 |
+
"grad_norm": 0.4617583155632019,
|
| 2296 |
+
"learning_rate": 1.2748789671866595e-05,
|
| 2297 |
+
"loss": 1.3547,
|
| 2298 |
+
"step": 3270
|
| 2299 |
+
},
|
| 2300 |
+
{
|
| 2301 |
+
"epoch": 0.7774814814814814,
|
| 2302 |
+
"grad_norm": 0.5739762783050537,
|
| 2303 |
+
"learning_rate": 1.2614308768154922e-05,
|
| 2304 |
+
"loss": 1.5191,
|
| 2305 |
+
"step": 3280
|
| 2306 |
+
},
|
| 2307 |
+
{
|
| 2308 |
+
"epoch": 0.7798518518518519,
|
| 2309 |
+
"grad_norm": 0.42539921402931213,
|
| 2310 |
+
"learning_rate": 1.247982786444325e-05,
|
| 2311 |
+
"loss": 1.3347,
|
| 2312 |
+
"step": 3290
|
| 2313 |
+
},
|
| 2314 |
+
{
|
| 2315 |
+
"epoch": 0.7822222222222223,
|
| 2316 |
+
"grad_norm": 0.5273600816726685,
|
| 2317 |
+
"learning_rate": 1.2345346960731577e-05,
|
| 2318 |
+
"loss": 1.3758,
|
| 2319 |
+
"step": 3300
|
| 2320 |
+
},
|
| 2321 |
+
{
|
| 2322 |
+
"epoch": 0.7845925925925926,
|
| 2323 |
+
"grad_norm": 0.4796091616153717,
|
| 2324 |
+
"learning_rate": 1.2210866057019904e-05,
|
| 2325 |
+
"loss": 1.3903,
|
| 2326 |
+
"step": 3310
|
| 2327 |
+
},
|
| 2328 |
+
{
|
| 2329 |
+
"epoch": 0.786962962962963,
|
| 2330 |
+
"grad_norm": 0.46542009711265564,
|
| 2331 |
+
"learning_rate": 1.2076385153308231e-05,
|
| 2332 |
+
"loss": 1.406,
|
| 2333 |
+
"step": 3320
|
| 2334 |
+
},
|
| 2335 |
+
{
|
| 2336 |
+
"epoch": 0.7893333333333333,
|
| 2337 |
+
"grad_norm": 0.47980108857154846,
|
| 2338 |
+
"learning_rate": 1.1941904249596558e-05,
|
| 2339 |
+
"loss": 1.3991,
|
| 2340 |
+
"step": 3330
|
| 2341 |
+
},
|
| 2342 |
+
{
|
| 2343 |
+
"epoch": 0.7917037037037037,
|
| 2344 |
+
"grad_norm": 0.507736086845398,
|
| 2345 |
+
"learning_rate": 1.1807423345884885e-05,
|
| 2346 |
+
"loss": 1.4583,
|
| 2347 |
+
"step": 3340
|
| 2348 |
+
},
|
| 2349 |
+
{
|
| 2350 |
+
"epoch": 0.794074074074074,
|
| 2351 |
+
"grad_norm": 0.5380430817604065,
|
| 2352 |
+
"learning_rate": 1.1672942442173212e-05,
|
| 2353 |
+
"loss": 1.2621,
|
| 2354 |
+
"step": 3350
|
| 2355 |
+
},
|
| 2356 |
+
{
|
| 2357 |
+
"epoch": 0.7964444444444444,
|
| 2358 |
+
"grad_norm": 0.6689913272857666,
|
| 2359 |
+
"learning_rate": 1.153846153846154e-05,
|
| 2360 |
+
"loss": 1.5355,
|
| 2361 |
+
"step": 3360
|
| 2362 |
+
},
|
| 2363 |
+
{
|
| 2364 |
+
"epoch": 0.7988148148148149,
|
| 2365 |
+
"grad_norm": 0.5129537582397461,
|
| 2366 |
+
"learning_rate": 1.1403980634749865e-05,
|
| 2367 |
+
"loss": 1.5895,
|
| 2368 |
+
"step": 3370
|
| 2369 |
+
},
|
| 2370 |
+
{
|
| 2371 |
+
"epoch": 0.8011851851851852,
|
| 2372 |
+
"grad_norm": 0.5240408182144165,
|
| 2373 |
+
"learning_rate": 1.1269499731038192e-05,
|
| 2374 |
+
"loss": 1.4896,
|
| 2375 |
+
"step": 3380
|
| 2376 |
+
},
|
| 2377 |
+
{
|
| 2378 |
+
"epoch": 0.8035555555555556,
|
| 2379 |
+
"grad_norm": 0.5004174709320068,
|
| 2380 |
+
"learning_rate": 1.113501882732652e-05,
|
| 2381 |
+
"loss": 1.4303,
|
| 2382 |
+
"step": 3390
|
| 2383 |
+
},
|
| 2384 |
+
{
|
| 2385 |
+
"epoch": 0.8059259259259259,
|
| 2386 |
+
"grad_norm": 0.46890896558761597,
|
| 2387 |
+
"learning_rate": 1.1000537923614846e-05,
|
| 2388 |
+
"loss": 1.3536,
|
| 2389 |
+
"step": 3400
|
| 2390 |
+
},
|
| 2391 |
+
{
|
| 2392 |
+
"epoch": 0.8082962962962963,
|
| 2393 |
+
"grad_norm": 0.5150523781776428,
|
| 2394 |
+
"learning_rate": 1.0866057019903174e-05,
|
| 2395 |
+
"loss": 1.3952,
|
| 2396 |
+
"step": 3410
|
| 2397 |
+
},
|
| 2398 |
+
{
|
| 2399 |
+
"epoch": 0.8106666666666666,
|
| 2400 |
+
"grad_norm": 0.4322206676006317,
|
| 2401 |
+
"learning_rate": 1.07315761161915e-05,
|
| 2402 |
+
"loss": 1.2252,
|
| 2403 |
+
"step": 3420
|
| 2404 |
+
},
|
| 2405 |
+
{
|
| 2406 |
+
"epoch": 0.813037037037037,
|
| 2407 |
+
"grad_norm": 0.5782944560050964,
|
| 2408 |
+
"learning_rate": 1.0597095212479828e-05,
|
| 2409 |
+
"loss": 1.4982,
|
| 2410 |
+
"step": 3430
|
| 2411 |
+
},
|
| 2412 |
+
{
|
| 2413 |
+
"epoch": 0.8154074074074074,
|
| 2414 |
+
"grad_norm": 0.6032952666282654,
|
| 2415 |
+
"learning_rate": 1.0462614308768155e-05,
|
| 2416 |
+
"loss": 1.4614,
|
| 2417 |
+
"step": 3440
|
| 2418 |
+
},
|
| 2419 |
+
{
|
| 2420 |
+
"epoch": 0.8177777777777778,
|
| 2421 |
+
"grad_norm": 0.453756183385849,
|
| 2422 |
+
"learning_rate": 1.0328133405056482e-05,
|
| 2423 |
+
"loss": 1.5017,
|
| 2424 |
+
"step": 3450
|
| 2425 |
+
},
|
| 2426 |
+
{
|
| 2427 |
+
"epoch": 0.8201481481481482,
|
| 2428 |
+
"grad_norm": 0.5315883159637451,
|
| 2429 |
+
"learning_rate": 1.019365250134481e-05,
|
| 2430 |
+
"loss": 1.5354,
|
| 2431 |
+
"step": 3460
|
| 2432 |
+
},
|
| 2433 |
+
{
|
| 2434 |
+
"epoch": 0.8225185185185185,
|
| 2435 |
+
"grad_norm": 0.5065041184425354,
|
| 2436 |
+
"learning_rate": 1.0059171597633136e-05,
|
| 2437 |
+
"loss": 1.4245,
|
| 2438 |
+
"step": 3470
|
| 2439 |
+
},
|
| 2440 |
+
{
|
| 2441 |
+
"epoch": 0.8248888888888889,
|
| 2442 |
+
"grad_norm": 0.542103111743927,
|
| 2443 |
+
"learning_rate": 9.924690693921463e-06,
|
| 2444 |
+
"loss": 1.3599,
|
| 2445 |
+
"step": 3480
|
| 2446 |
+
},
|
| 2447 |
+
{
|
| 2448 |
+
"epoch": 0.8272592592592592,
|
| 2449 |
+
"grad_norm": 0.5330160856246948,
|
| 2450 |
+
"learning_rate": 9.79020979020979e-06,
|
| 2451 |
+
"loss": 1.3502,
|
| 2452 |
+
"step": 3490
|
| 2453 |
+
},
|
| 2454 |
+
{
|
| 2455 |
+
"epoch": 0.8296296296296296,
|
| 2456 |
+
"grad_norm": 0.4731038212776184,
|
| 2457 |
+
"learning_rate": 9.655728886498118e-06,
|
| 2458 |
+
"loss": 1.2931,
|
| 2459 |
+
"step": 3500
|
| 2460 |
+
},
|
| 2461 |
+
{
|
| 2462 |
+
"epoch": 0.832,
|
| 2463 |
+
"grad_norm": 0.4719734787940979,
|
| 2464 |
+
"learning_rate": 9.521247982786445e-06,
|
| 2465 |
+
"loss": 1.3185,
|
| 2466 |
+
"step": 3510
|
| 2467 |
+
},
|
| 2468 |
+
{
|
| 2469 |
+
"epoch": 0.8343703703703703,
|
| 2470 |
+
"grad_norm": 0.5551607012748718,
|
| 2471 |
+
"learning_rate": 9.386767079074772e-06,
|
| 2472 |
+
"loss": 1.4045,
|
| 2473 |
+
"step": 3520
|
| 2474 |
+
},
|
| 2475 |
+
{
|
| 2476 |
+
"epoch": 0.8367407407407408,
|
| 2477 |
+
"grad_norm": 0.5661736130714417,
|
| 2478 |
+
"learning_rate": 9.252286175363099e-06,
|
| 2479 |
+
"loss": 1.6005,
|
| 2480 |
+
"step": 3530
|
| 2481 |
+
},
|
| 2482 |
+
{
|
| 2483 |
+
"epoch": 0.8391111111111111,
|
| 2484 |
+
"grad_norm": 0.5772873759269714,
|
| 2485 |
+
"learning_rate": 9.117805271651426e-06,
|
| 2486 |
+
"loss": 1.3975,
|
| 2487 |
+
"step": 3540
|
| 2488 |
+
},
|
| 2489 |
+
{
|
| 2490 |
+
"epoch": 0.8414814814814815,
|
| 2491 |
+
"grad_norm": 0.5180752873420715,
|
| 2492 |
+
"learning_rate": 8.983324367939753e-06,
|
| 2493 |
+
"loss": 1.5147,
|
| 2494 |
+
"step": 3550
|
| 2495 |
+
},
|
| 2496 |
+
{
|
| 2497 |
+
"epoch": 0.8438518518518519,
|
| 2498 |
+
"grad_norm": 0.5256723165512085,
|
| 2499 |
+
"learning_rate": 8.84884346422808e-06,
|
| 2500 |
+
"loss": 1.4673,
|
| 2501 |
+
"step": 3560
|
| 2502 |
+
},
|
| 2503 |
+
{
|
| 2504 |
+
"epoch": 0.8462222222222222,
|
| 2505 |
+
"grad_norm": 0.4829583168029785,
|
| 2506 |
+
"learning_rate": 8.714362560516406e-06,
|
| 2507 |
+
"loss": 1.5102,
|
| 2508 |
+
"step": 3570
|
| 2509 |
+
},
|
| 2510 |
+
{
|
| 2511 |
+
"epoch": 0.8485925925925926,
|
| 2512 |
+
"grad_norm": 0.5027347207069397,
|
| 2513 |
+
"learning_rate": 8.579881656804733e-06,
|
| 2514 |
+
"loss": 1.6514,
|
| 2515 |
+
"step": 3580
|
| 2516 |
+
},
|
| 2517 |
+
{
|
| 2518 |
+
"epoch": 0.8509629629629629,
|
| 2519 |
+
"grad_norm": 0.5117186903953552,
|
| 2520 |
+
"learning_rate": 8.44540075309306e-06,
|
| 2521 |
+
"loss": 1.4451,
|
| 2522 |
+
"step": 3590
|
| 2523 |
+
},
|
| 2524 |
+
{
|
| 2525 |
+
"epoch": 0.8533333333333334,
|
| 2526 |
+
"grad_norm": 0.4994155466556549,
|
| 2527 |
+
"learning_rate": 8.310919849381387e-06,
|
| 2528 |
+
"loss": 1.3408,
|
| 2529 |
+
"step": 3600
|
| 2530 |
+
},
|
| 2531 |
+
{
|
| 2532 |
+
"epoch": 0.8557037037037037,
|
| 2533 |
+
"grad_norm": 0.4399481415748596,
|
| 2534 |
+
"learning_rate": 8.176438945669715e-06,
|
| 2535 |
+
"loss": 1.4044,
|
| 2536 |
+
"step": 3610
|
| 2537 |
+
},
|
| 2538 |
+
{
|
| 2539 |
+
"epoch": 0.8580740740740741,
|
| 2540 |
+
"grad_norm": 0.49395203590393066,
|
| 2541 |
+
"learning_rate": 8.041958041958042e-06,
|
| 2542 |
+
"loss": 1.3895,
|
| 2543 |
+
"step": 3620
|
| 2544 |
+
},
|
| 2545 |
+
{
|
| 2546 |
+
"epoch": 0.8604444444444445,
|
| 2547 |
+
"grad_norm": 0.4922611713409424,
|
| 2548 |
+
"learning_rate": 7.907477138246369e-06,
|
| 2549 |
+
"loss": 1.5503,
|
| 2550 |
+
"step": 3630
|
| 2551 |
+
},
|
| 2552 |
+
{
|
| 2553 |
+
"epoch": 0.8628148148148148,
|
| 2554 |
+
"grad_norm": 0.5255241990089417,
|
| 2555 |
+
"learning_rate": 7.772996234534696e-06,
|
| 2556 |
+
"loss": 1.4736,
|
| 2557 |
+
"step": 3640
|
| 2558 |
+
},
|
| 2559 |
+
{
|
| 2560 |
+
"epoch": 0.8651851851851852,
|
| 2561 |
+
"grad_norm": 0.4713379442691803,
|
| 2562 |
+
"learning_rate": 7.638515330823023e-06,
|
| 2563 |
+
"loss": 1.4395,
|
| 2564 |
+
"step": 3650
|
| 2565 |
+
},
|
| 2566 |
+
{
|
| 2567 |
+
"epoch": 0.8675555555555555,
|
| 2568 |
+
"grad_norm": 0.4290190637111664,
|
| 2569 |
+
"learning_rate": 7.50403442711135e-06,
|
| 2570 |
+
"loss": 1.4265,
|
| 2571 |
+
"step": 3660
|
| 2572 |
+
},
|
| 2573 |
+
{
|
| 2574 |
+
"epoch": 0.8699259259259259,
|
| 2575 |
+
"grad_norm": 0.5157113075256348,
|
| 2576 |
+
"learning_rate": 7.369553523399677e-06,
|
| 2577 |
+
"loss": 1.4363,
|
| 2578 |
+
"step": 3670
|
| 2579 |
+
},
|
| 2580 |
+
{
|
| 2581 |
+
"epoch": 0.8722962962962963,
|
| 2582 |
+
"grad_norm": 0.5694654583930969,
|
| 2583 |
+
"learning_rate": 7.2350726196880045e-06,
|
| 2584 |
+
"loss": 1.4602,
|
| 2585 |
+
"step": 3680
|
| 2586 |
+
},
|
| 2587 |
+
{
|
| 2588 |
+
"epoch": 0.8746666666666667,
|
| 2589 |
+
"grad_norm": 0.47969937324523926,
|
| 2590 |
+
"learning_rate": 7.100591715976332e-06,
|
| 2591 |
+
"loss": 1.2716,
|
| 2592 |
+
"step": 3690
|
| 2593 |
+
},
|
| 2594 |
+
{
|
| 2595 |
+
"epoch": 0.8770370370370371,
|
| 2596 |
+
"grad_norm": 0.4939590096473694,
|
| 2597 |
+
"learning_rate": 6.966110812264659e-06,
|
| 2598 |
+
"loss": 1.4356,
|
| 2599 |
+
"step": 3700
|
| 2600 |
+
},
|
| 2601 |
+
{
|
| 2602 |
+
"epoch": 0.8794074074074074,
|
| 2603 |
+
"grad_norm": 0.4829910695552826,
|
| 2604 |
+
"learning_rate": 6.831629908552986e-06,
|
| 2605 |
+
"loss": 1.4482,
|
| 2606 |
+
"step": 3710
|
| 2607 |
+
},
|
| 2608 |
+
{
|
| 2609 |
+
"epoch": 0.8817777777777778,
|
| 2610 |
+
"grad_norm": 0.473178505897522,
|
| 2611 |
+
"learning_rate": 6.697149004841313e-06,
|
| 2612 |
+
"loss": 1.3772,
|
| 2613 |
+
"step": 3720
|
| 2614 |
+
},
|
| 2615 |
+
{
|
| 2616 |
+
"epoch": 0.8841481481481481,
|
| 2617 |
+
"grad_norm": 0.5327422618865967,
|
| 2618 |
+
"learning_rate": 6.56266810112964e-06,
|
| 2619 |
+
"loss": 1.5707,
|
| 2620 |
+
"step": 3730
|
| 2621 |
+
},
|
| 2622 |
+
{
|
| 2623 |
+
"epoch": 0.8865185185185185,
|
| 2624 |
+
"grad_norm": 0.4069652259349823,
|
| 2625 |
+
"learning_rate": 6.428187197417967e-06,
|
| 2626 |
+
"loss": 1.3501,
|
| 2627 |
+
"step": 3740
|
| 2628 |
+
},
|
| 2629 |
+
{
|
| 2630 |
+
"epoch": 0.8888888888888888,
|
| 2631 |
+
"grad_norm": 0.5368005037307739,
|
| 2632 |
+
"learning_rate": 6.2937062937062944e-06,
|
| 2633 |
+
"loss": 1.6791,
|
| 2634 |
+
"step": 3750
|
| 2635 |
+
},
|
| 2636 |
+
{
|
| 2637 |
+
"epoch": 0.8912592592592593,
|
| 2638 |
+
"grad_norm": 0.5622044801712036,
|
| 2639 |
+
"learning_rate": 6.159225389994621e-06,
|
| 2640 |
+
"loss": 1.2727,
|
| 2641 |
+
"step": 3760
|
| 2642 |
+
},
|
| 2643 |
+
{
|
| 2644 |
+
"epoch": 0.8936296296296297,
|
| 2645 |
+
"grad_norm": 0.5526837110519409,
|
| 2646 |
+
"learning_rate": 6.024744486282948e-06,
|
| 2647 |
+
"loss": 1.478,
|
| 2648 |
+
"step": 3770
|
| 2649 |
+
},
|
| 2650 |
+
{
|
| 2651 |
+
"epoch": 0.896,
|
| 2652 |
+
"grad_norm": 0.5442282557487488,
|
| 2653 |
+
"learning_rate": 5.890263582571275e-06,
|
| 2654 |
+
"loss": 1.3917,
|
| 2655 |
+
"step": 3780
|
| 2656 |
+
},
|
| 2657 |
+
{
|
| 2658 |
+
"epoch": 0.8983703703703704,
|
| 2659 |
+
"grad_norm": 0.5711065530776978,
|
| 2660 |
+
"learning_rate": 5.755782678859602e-06,
|
| 2661 |
+
"loss": 1.4748,
|
| 2662 |
+
"step": 3790
|
| 2663 |
+
},
|
| 2664 |
+
{
|
| 2665 |
+
"epoch": 0.9007407407407407,
|
| 2666 |
+
"grad_norm": 0.5068963766098022,
|
| 2667 |
+
"learning_rate": 5.621301775147929e-06,
|
| 2668 |
+
"loss": 1.3375,
|
| 2669 |
+
"step": 3800
|
| 2670 |
+
},
|
| 2671 |
+
{
|
| 2672 |
+
"epoch": 0.9031111111111111,
|
| 2673 |
+
"grad_norm": 0.6032804846763611,
|
| 2674 |
+
"learning_rate": 5.486820871436256e-06,
|
| 2675 |
+
"loss": 1.553,
|
| 2676 |
+
"step": 3810
|
| 2677 |
+
},
|
| 2678 |
+
{
|
| 2679 |
+
"epoch": 0.9054814814814814,
|
| 2680 |
+
"grad_norm": 0.4718996286392212,
|
| 2681 |
+
"learning_rate": 5.352339967724583e-06,
|
| 2682 |
+
"loss": 1.4586,
|
| 2683 |
+
"step": 3820
|
| 2684 |
+
},
|
| 2685 |
+
{
|
| 2686 |
+
"epoch": 0.9078518518518518,
|
| 2687 |
+
"grad_norm": 0.583185076713562,
|
| 2688 |
+
"learning_rate": 5.21785906401291e-06,
|
| 2689 |
+
"loss": 1.4294,
|
| 2690 |
+
"step": 3830
|
| 2691 |
+
},
|
| 2692 |
+
{
|
| 2693 |
+
"epoch": 0.9102222222222223,
|
| 2694 |
+
"grad_norm": 0.5688157677650452,
|
| 2695 |
+
"learning_rate": 5.083378160301237e-06,
|
| 2696 |
+
"loss": 1.3434,
|
| 2697 |
+
"step": 3840
|
| 2698 |
+
},
|
| 2699 |
+
{
|
| 2700 |
+
"epoch": 0.9125925925925926,
|
| 2701 |
+
"grad_norm": 0.43667495250701904,
|
| 2702 |
+
"learning_rate": 4.948897256589564e-06,
|
| 2703 |
+
"loss": 1.5107,
|
| 2704 |
+
"step": 3850
|
| 2705 |
+
},
|
| 2706 |
+
{
|
| 2707 |
+
"epoch": 0.914962962962963,
|
| 2708 |
+
"grad_norm": 0.5924187302589417,
|
| 2709 |
+
"learning_rate": 4.814416352877891e-06,
|
| 2710 |
+
"loss": 1.3558,
|
| 2711 |
+
"step": 3860
|
| 2712 |
+
},
|
| 2713 |
+
{
|
| 2714 |
+
"epoch": 0.9173333333333333,
|
| 2715 |
+
"grad_norm": 0.538318932056427,
|
| 2716 |
+
"learning_rate": 4.679935449166218e-06,
|
| 2717 |
+
"loss": 1.379,
|
| 2718 |
+
"step": 3870
|
| 2719 |
+
},
|
| 2720 |
+
{
|
| 2721 |
+
"epoch": 0.9197037037037037,
|
| 2722 |
+
"grad_norm": 0.4840611219406128,
|
| 2723 |
+
"learning_rate": 4.5454545454545455e-06,
|
| 2724 |
+
"loss": 1.2776,
|
| 2725 |
+
"step": 3880
|
| 2726 |
+
},
|
| 2727 |
+
{
|
| 2728 |
+
"epoch": 0.922074074074074,
|
| 2729 |
+
"grad_norm": 0.5927128195762634,
|
| 2730 |
+
"learning_rate": 4.410973641742873e-06,
|
| 2731 |
+
"loss": 1.4598,
|
| 2732 |
+
"step": 3890
|
| 2733 |
+
},
|
| 2734 |
+
{
|
| 2735 |
+
"epoch": 0.9244444444444444,
|
| 2736 |
+
"grad_norm": 0.5672274827957153,
|
| 2737 |
+
"learning_rate": 4.2764927380312e-06,
|
| 2738 |
+
"loss": 1.467,
|
| 2739 |
+
"step": 3900
|
| 2740 |
+
},
|
| 2741 |
+
{
|
| 2742 |
+
"epoch": 0.9268148148148149,
|
| 2743 |
+
"grad_norm": 0.5424569249153137,
|
| 2744 |
+
"learning_rate": 4.142011834319527e-06,
|
| 2745 |
+
"loss": 1.532,
|
| 2746 |
+
"step": 3910
|
| 2747 |
+
},
|
| 2748 |
+
{
|
| 2749 |
+
"epoch": 0.9291851851851852,
|
| 2750 |
+
"grad_norm": 0.5921524167060852,
|
| 2751 |
+
"learning_rate": 4.007530930607853e-06,
|
| 2752 |
+
"loss": 1.5244,
|
| 2753 |
+
"step": 3920
|
| 2754 |
+
},
|
| 2755 |
+
{
|
| 2756 |
+
"epoch": 0.9315555555555556,
|
| 2757 |
+
"grad_norm": 0.4295691251754761,
|
| 2758 |
+
"learning_rate": 3.87305002689618e-06,
|
| 2759 |
+
"loss": 1.4358,
|
| 2760 |
+
"step": 3930
|
| 2761 |
+
},
|
| 2762 |
+
{
|
| 2763 |
+
"epoch": 0.9339259259259259,
|
| 2764 |
+
"grad_norm": 0.5117975473403931,
|
| 2765 |
+
"learning_rate": 3.738569123184508e-06,
|
| 2766 |
+
"loss": 1.4241,
|
| 2767 |
+
"step": 3940
|
| 2768 |
+
},
|
| 2769 |
+
{
|
| 2770 |
+
"epoch": 0.9362962962962963,
|
| 2771 |
+
"grad_norm": 0.5273323655128479,
|
| 2772 |
+
"learning_rate": 3.604088219472835e-06,
|
| 2773 |
+
"loss": 1.3188,
|
| 2774 |
+
"step": 3950
|
| 2775 |
+
},
|
| 2776 |
+
{
|
| 2777 |
+
"epoch": 0.9386666666666666,
|
| 2778 |
+
"grad_norm": 0.4947817325592041,
|
| 2779 |
+
"learning_rate": 3.469607315761162e-06,
|
| 2780 |
+
"loss": 1.444,
|
| 2781 |
+
"step": 3960
|
| 2782 |
+
},
|
| 2783 |
+
{
|
| 2784 |
+
"epoch": 0.941037037037037,
|
| 2785 |
+
"grad_norm": 0.5852746367454529,
|
| 2786 |
+
"learning_rate": 3.3351264120494893e-06,
|
| 2787 |
+
"loss": 1.4334,
|
| 2788 |
+
"step": 3970
|
| 2789 |
+
},
|
| 2790 |
+
{
|
| 2791 |
+
"epoch": 0.9434074074074074,
|
| 2792 |
+
"grad_norm": 0.5183681845664978,
|
| 2793 |
+
"learning_rate": 3.2006455083378165e-06,
|
| 2794 |
+
"loss": 1.4422,
|
| 2795 |
+
"step": 3980
|
| 2796 |
+
},
|
| 2797 |
+
{
|
| 2798 |
+
"epoch": 0.9457777777777778,
|
| 2799 |
+
"grad_norm": 0.5193647146224976,
|
| 2800 |
+
"learning_rate": 3.066164604626143e-06,
|
| 2801 |
+
"loss": 1.3769,
|
| 2802 |
+
"step": 3990
|
| 2803 |
+
},
|
| 2804 |
+
{
|
| 2805 |
+
"epoch": 0.9481481481481482,
|
| 2806 |
+
"grad_norm": 0.5597278475761414,
|
| 2807 |
+
"learning_rate": 2.9316837009144703e-06,
|
| 2808 |
+
"loss": 1.5406,
|
| 2809 |
+
"step": 4000
|
| 2810 |
+
},
|
| 2811 |
+
{
|
| 2812 |
+
"epoch": 0.9505185185185185,
|
| 2813 |
+
"grad_norm": 0.5173184275627136,
|
| 2814 |
+
"learning_rate": 2.7972027972027974e-06,
|
| 2815 |
+
"loss": 1.5082,
|
| 2816 |
+
"step": 4010
|
| 2817 |
+
},
|
| 2818 |
+
{
|
| 2819 |
+
"epoch": 0.9528888888888889,
|
| 2820 |
+
"grad_norm": 0.5205141305923462,
|
| 2821 |
+
"learning_rate": 2.6627218934911246e-06,
|
| 2822 |
+
"loss": 1.4971,
|
| 2823 |
+
"step": 4020
|
| 2824 |
+
},
|
| 2825 |
+
{
|
| 2826 |
+
"epoch": 0.9552592592592593,
|
| 2827 |
+
"grad_norm": 0.48843199014663696,
|
| 2828 |
+
"learning_rate": 2.5282409897794517e-06,
|
| 2829 |
+
"loss": 1.4113,
|
| 2830 |
+
"step": 4030
|
| 2831 |
+
},
|
| 2832 |
+
{
|
| 2833 |
+
"epoch": 0.9576296296296296,
|
| 2834 |
+
"grad_norm": 0.5854997634887695,
|
| 2835 |
+
"learning_rate": 2.3937600860677784e-06,
|
| 2836 |
+
"loss": 1.5666,
|
| 2837 |
+
"step": 4040
|
| 2838 |
+
},
|
| 2839 |
+
{
|
| 2840 |
+
"epoch": 0.96,
|
| 2841 |
+
"grad_norm": 0.5264609456062317,
|
| 2842 |
+
"learning_rate": 2.2592791823561056e-06,
|
| 2843 |
+
"loss": 1.3334,
|
| 2844 |
+
"step": 4050
|
| 2845 |
+
},
|
| 2846 |
+
{
|
| 2847 |
+
"epoch": 0.9623703703703703,
|
| 2848 |
+
"grad_norm": 0.5044777989387512,
|
| 2849 |
+
"learning_rate": 2.1247982786444327e-06,
|
| 2850 |
+
"loss": 1.4251,
|
| 2851 |
+
"step": 4060
|
| 2852 |
+
},
|
| 2853 |
+
{
|
| 2854 |
+
"epoch": 0.9647407407407408,
|
| 2855 |
+
"grad_norm": 0.5678290128707886,
|
| 2856 |
+
"learning_rate": 1.99031737493276e-06,
|
| 2857 |
+
"loss": 1.4359,
|
| 2858 |
+
"step": 4070
|
| 2859 |
+
},
|
| 2860 |
+
{
|
| 2861 |
+
"epoch": 0.9671111111111111,
|
| 2862 |
+
"grad_norm": 0.5562720894813538,
|
| 2863 |
+
"learning_rate": 1.8558364712210868e-06,
|
| 2864 |
+
"loss": 1.4333,
|
| 2865 |
+
"step": 4080
|
| 2866 |
+
},
|
| 2867 |
+
{
|
| 2868 |
+
"epoch": 0.9694814814814815,
|
| 2869 |
+
"grad_norm": 0.5118197798728943,
|
| 2870 |
+
"learning_rate": 1.7213555675094137e-06,
|
| 2871 |
+
"loss": 1.4257,
|
| 2872 |
+
"step": 4090
|
| 2873 |
+
},
|
| 2874 |
+
{
|
| 2875 |
+
"epoch": 0.9718518518518519,
|
| 2876 |
+
"grad_norm": 0.4556100070476532,
|
| 2877 |
+
"learning_rate": 1.5868746637977408e-06,
|
| 2878 |
+
"loss": 1.4414,
|
| 2879 |
+
"step": 4100
|
| 2880 |
+
},
|
| 2881 |
+
{
|
| 2882 |
+
"epoch": 0.9742222222222222,
|
| 2883 |
+
"grad_norm": 0.5370482802391052,
|
| 2884 |
+
"learning_rate": 1.452393760086068e-06,
|
| 2885 |
+
"loss": 1.6166,
|
| 2886 |
+
"step": 4110
|
| 2887 |
+
},
|
| 2888 |
+
{
|
| 2889 |
+
"epoch": 0.9765925925925926,
|
| 2890 |
+
"grad_norm": 0.5851370096206665,
|
| 2891 |
+
"learning_rate": 1.3179128563743949e-06,
|
| 2892 |
+
"loss": 1.5649,
|
| 2893 |
+
"step": 4120
|
| 2894 |
+
},
|
| 2895 |
+
{
|
| 2896 |
+
"epoch": 0.9789629629629629,
|
| 2897 |
+
"grad_norm": 0.5804405212402344,
|
| 2898 |
+
"learning_rate": 1.183431952662722e-06,
|
| 2899 |
+
"loss": 1.3898,
|
| 2900 |
+
"step": 4130
|
| 2901 |
+
},
|
| 2902 |
+
{
|
| 2903 |
+
"epoch": 0.9813333333333333,
|
| 2904 |
+
"grad_norm": 0.5411353707313538,
|
| 2905 |
+
"learning_rate": 1.0489510489510491e-06,
|
| 2906 |
+
"loss": 1.4885,
|
| 2907 |
+
"step": 4140
|
| 2908 |
+
},
|
| 2909 |
+
{
|
| 2910 |
+
"epoch": 0.9837037037037037,
|
| 2911 |
+
"grad_norm": 0.5271933078765869,
|
| 2912 |
+
"learning_rate": 9.14470145239376e-07,
|
| 2913 |
+
"loss": 1.3816,
|
| 2914 |
+
"step": 4150
|
| 2915 |
+
},
|
| 2916 |
+
{
|
| 2917 |
+
"epoch": 0.9860740740740741,
|
| 2918 |
+
"grad_norm": 0.5039179921150208,
|
| 2919 |
+
"learning_rate": 7.799892415277031e-07,
|
| 2920 |
+
"loss": 1.2985,
|
| 2921 |
+
"step": 4160
|
| 2922 |
+
},
|
| 2923 |
+
{
|
| 2924 |
+
"epoch": 0.9884444444444445,
|
| 2925 |
+
"grad_norm": 0.47758767008781433,
|
| 2926 |
+
"learning_rate": 6.455083378160301e-07,
|
| 2927 |
+
"loss": 1.4848,
|
| 2928 |
+
"step": 4170
|
| 2929 |
+
},
|
| 2930 |
+
{
|
| 2931 |
+
"epoch": 0.9908148148148148,
|
| 2932 |
+
"grad_norm": 0.5355851054191589,
|
| 2933 |
+
"learning_rate": 5.110274341043572e-07,
|
| 2934 |
+
"loss": 1.5264,
|
| 2935 |
+
"step": 4180
|
| 2936 |
+
},
|
| 2937 |
+
{
|
| 2938 |
+
"epoch": 0.9931851851851852,
|
| 2939 |
+
"grad_norm": 0.5918956398963928,
|
| 2940 |
+
"learning_rate": 3.7654653039268424e-07,
|
| 2941 |
+
"loss": 1.4337,
|
| 2942 |
+
"step": 4190
|
| 2943 |
+
},
|
| 2944 |
+
{
|
| 2945 |
+
"epoch": 0.9955555555555555,
|
| 2946 |
+
"grad_norm": 0.4990881085395813,
|
| 2947 |
+
"learning_rate": 2.4206562668101127e-07,
|
| 2948 |
+
"loss": 1.3851,
|
| 2949 |
+
"step": 4200
|
| 2950 |
+
},
|
| 2951 |
+
{
|
| 2952 |
+
"epoch": 0.9979259259259259,
|
| 2953 |
+
"grad_norm": 0.49343588948249817,
|
| 2954 |
+
"learning_rate": 1.0758472296933835e-07,
|
| 2955 |
+
"loss": 1.5421,
|
| 2956 |
+
"step": 4210
|
| 2957 |
+
}
|
| 2958 |
+
],
|
| 2959 |
+
"logging_steps": 10,
|
| 2960 |
+
"max_steps": 4218,
|
| 2961 |
+
"num_input_tokens_seen": 0,
|
| 2962 |
+
"num_train_epochs": 1,
|
| 2963 |
+
"save_steps": 5000,
|
| 2964 |
+
"stateful_callbacks": {
|
| 2965 |
+
"TrainerControl": {
|
| 2966 |
+
"args": {
|
| 2967 |
+
"should_epoch_stop": false,
|
| 2968 |
+
"should_evaluate": false,
|
| 2969 |
+
"should_log": false,
|
| 2970 |
+
"should_save": true,
|
| 2971 |
+
"should_training_stop": true
|
| 2972 |
+
},
|
| 2973 |
+
"attributes": {}
|
| 2974 |
+
}
|
| 2975 |
+
},
|
| 2976 |
+
"total_flos": 1.660058442305372e+17,
|
| 2977 |
+
"train_batch_size": 4,
|
| 2978 |
+
"trial_name": null,
|
| 2979 |
+
"trial_params": null
|
| 2980 |
+
}
|
checkpoint-4218/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:f6edf39eb5dcce229997ed5659e24fe29f23ba447c7d41fe76b0f4e10d66ddf6
|
| 3 |
+
size 5176
|
config.json
ADDED
|
@@ -0,0 +1,32 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "appvoid/arco-mini-run-5",
|
| 3 |
+
"architectures": [
|
| 4 |
+
"LlamaForCausalLM"
|
| 5 |
+
],
|
| 6 |
+
"attention_bias": false,
|
| 7 |
+
"attention_dropout": 0.0,
|
| 8 |
+
"bos_token_id": 1,
|
| 9 |
+
"eos_token_id": 2,
|
| 10 |
+
"head_dim": 96,
|
| 11 |
+
"hidden_act": "silu",
|
| 12 |
+
"hidden_size": 1536,
|
| 13 |
+
"initializer_range": 0.02,
|
| 14 |
+
"intermediate_size": 3072,
|
| 15 |
+
"max_position_embeddings": 2048,
|
| 16 |
+
"mlp_bias": false,
|
| 17 |
+
"model_type": "llama",
|
| 18 |
+
"num_attention_heads": 8,
|
| 19 |
+
"num_hidden_layers": 8,
|
| 20 |
+
"num_key_value_heads": 8,
|
| 21 |
+
"pad_token_id": 0,
|
| 22 |
+
"pretraining_tp": 1,
|
| 23 |
+
"rms_norm_eps": 1e-05,
|
| 24 |
+
"rope_scaling": null,
|
| 25 |
+
"rope_theta": 100000,
|
| 26 |
+
"sliding_window": null,
|
| 27 |
+
"tie_word_embeddings": false,
|
| 28 |
+
"torch_dtype": "float32",
|
| 29 |
+
"transformers_version": "4.45.1",
|
| 30 |
+
"use_cache": true,
|
| 31 |
+
"vocab_size": 32003
|
| 32 |
+
}
|
generation_config.json
ADDED
|
@@ -0,0 +1,7 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_from_model_config": true,
|
| 3 |
+
"bos_token_id": 1,
|
| 4 |
+
"eos_token_id": 2,
|
| 5 |
+
"pad_token_id": 0,
|
| 6 |
+
"transformers_version": "4.45.1"
|
| 7 |
+
}
|
model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:9ed892552e4757b7d11b8ff1d74dc2449cf88d885f52872c693b3159f6d36b57
|
| 3 |
+
size 997345472
|
special_tokens_map.json
ADDED
|
@@ -0,0 +1,44 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"bos_token": {
|
| 3 |
+
"content": "[BOS]",
|
| 4 |
+
"lstrip": false,
|
| 5 |
+
"normalized": false,
|
| 6 |
+
"rstrip": false,
|
| 7 |
+
"single_word": false
|
| 8 |
+
},
|
| 9 |
+
"cls_token": {
|
| 10 |
+
"content": "</s>",
|
| 11 |
+
"lstrip": false,
|
| 12 |
+
"normalized": false,
|
| 13 |
+
"rstrip": false,
|
| 14 |
+
"single_word": false
|
| 15 |
+
},
|
| 16 |
+
"eos_token": {
|
| 17 |
+
"content": "[EOS]",
|
| 18 |
+
"lstrip": false,
|
| 19 |
+
"normalized": false,
|
| 20 |
+
"rstrip": false,
|
| 21 |
+
"single_word": false
|
| 22 |
+
},
|
| 23 |
+
"pad_token": {
|
| 24 |
+
"content": "[PAD]",
|
| 25 |
+
"lstrip": false,
|
| 26 |
+
"normalized": false,
|
| 27 |
+
"rstrip": false,
|
| 28 |
+
"single_word": false
|
| 29 |
+
},
|
| 30 |
+
"sep_token": {
|
| 31 |
+
"content": "</s>",
|
| 32 |
+
"lstrip": false,
|
| 33 |
+
"normalized": false,
|
| 34 |
+
"rstrip": false,
|
| 35 |
+
"single_word": false
|
| 36 |
+
},
|
| 37 |
+
"unk_token": {
|
| 38 |
+
"content": "<unk>",
|
| 39 |
+
"lstrip": false,
|
| 40 |
+
"normalized": false,
|
| 41 |
+
"rstrip": false,
|
| 42 |
+
"single_word": false
|
| 43 |
+
}
|
| 44 |
+
}
|
tokenizer.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
tokenizer.model
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:dadfd56d766715c61d2ef780a525ab43b8e6da4de6865bda3d95fdef5e134055
|
| 3 |
+
size 493443
|
tokenizer_config.json
ADDED
|
@@ -0,0 +1,69 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"add_bos_token": false,
|
| 3 |
+
"add_eos_token": false,
|
| 4 |
+
"add_prefix_space": true,
|
| 5 |
+
"added_tokens_decoder": {
|
| 6 |
+
"0": {
|
| 7 |
+
"content": "<unk>",
|
| 8 |
+
"lstrip": false,
|
| 9 |
+
"normalized": false,
|
| 10 |
+
"rstrip": false,
|
| 11 |
+
"single_word": false,
|
| 12 |
+
"special": true
|
| 13 |
+
},
|
| 14 |
+
"1": {
|
| 15 |
+
"content": "<s>",
|
| 16 |
+
"lstrip": false,
|
| 17 |
+
"normalized": false,
|
| 18 |
+
"rstrip": false,
|
| 19 |
+
"single_word": false,
|
| 20 |
+
"special": true
|
| 21 |
+
},
|
| 22 |
+
"2": {
|
| 23 |
+
"content": "</s>",
|
| 24 |
+
"lstrip": false,
|
| 25 |
+
"normalized": false,
|
| 26 |
+
"rstrip": false,
|
| 27 |
+
"single_word": false,
|
| 28 |
+
"special": true
|
| 29 |
+
},
|
| 30 |
+
"32000": {
|
| 31 |
+
"content": "[PAD]",
|
| 32 |
+
"lstrip": false,
|
| 33 |
+
"normalized": false,
|
| 34 |
+
"rstrip": false,
|
| 35 |
+
"single_word": false,
|
| 36 |
+
"special": true
|
| 37 |
+
},
|
| 38 |
+
"32001": {
|
| 39 |
+
"content": "[BOS]",
|
| 40 |
+
"lstrip": false,
|
| 41 |
+
"normalized": false,
|
| 42 |
+
"rstrip": false,
|
| 43 |
+
"single_word": false,
|
| 44 |
+
"special": true
|
| 45 |
+
},
|
| 46 |
+
"32002": {
|
| 47 |
+
"content": "[EOS]",
|
| 48 |
+
"lstrip": false,
|
| 49 |
+
"normalized": false,
|
| 50 |
+
"rstrip": false,
|
| 51 |
+
"single_word": false,
|
| 52 |
+
"special": true
|
| 53 |
+
}
|
| 54 |
+
},
|
| 55 |
+
"additional_special_tokens": [],
|
| 56 |
+
"bos_token": "[BOS]",
|
| 57 |
+
"clean_up_tokenization_spaces": false,
|
| 58 |
+
"cls_token": "</s>",
|
| 59 |
+
"eos_token": "[EOS]",
|
| 60 |
+
"legacy": true,
|
| 61 |
+
"model_max_length": 1000000000000000019884624838656,
|
| 62 |
+
"pad_token": "[PAD]",
|
| 63 |
+
"sep_token": "</s>",
|
| 64 |
+
"sp_model_kwargs": {},
|
| 65 |
+
"spaces_between_special_tokens": false,
|
| 66 |
+
"tokenizer_class": "LlamaTokenizer",
|
| 67 |
+
"unk_token": "<unk>",
|
| 68 |
+
"use_default_system_prompt": false
|
| 69 |
+
}
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:f6edf39eb5dcce229997ed5659e24fe29f23ba447c7d41fe76b0f4e10d66ddf6
|
| 3 |
+
size 5176
|