Upload folder using huggingface_hub
Browse files- added_tokens.json +5 -0
- checkpoint-2812/config.json +32 -0
- checkpoint-2812/generation_config.json +7 -0
- checkpoint-2812/model.safetensors +3 -0
- checkpoint-2812/optimizer.pt +3 -0
- checkpoint-2812/rng_state.pth +3 -0
- checkpoint-2812/scheduler.pt +3 -0
- checkpoint-2812/trainer_state.json +2000 -0
- checkpoint-2812/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-2812/config.json
ADDED
|
@@ -0,0 +1,32 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "appvoid/arco-mini-run-4",
|
| 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-2812/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-2812/model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:2b74e6e089df1f5cabadddc8329432ca60d36596872755480e71adca08612f3b
|
| 3 |
+
size 997345472
|
checkpoint-2812/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:467d1fe1cf41c17dd85c3fafb4c6da2c0c19b880403ce40ad2355ff640eb6ec3
|
| 3 |
+
size 1994737722
|
checkpoint-2812/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-2812/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:9195b4b9a48e429e5bc1047d7a46165fb060d5f32bba5884316c5087a26a2879
|
| 3 |
+
size 1064
|
checkpoint-2812/trainer_state.json
ADDED
|
@@ -0,0 +1,2000 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 0.9998222222222222,
|
| 5 |
+
"eval_steps": 5000,
|
| 6 |
+
"global_step": 2812,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.0035555555555555557,
|
| 13 |
+
"grad_norm": 0.4803234934806824,
|
| 14 |
+
"learning_rate": 2.0000000000000003e-06,
|
| 15 |
+
"loss": 1.5376,
|
| 16 |
+
"step": 10
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.0071111111111111115,
|
| 20 |
+
"grad_norm": 0.4609735310077667,
|
| 21 |
+
"learning_rate": 4.000000000000001e-06,
|
| 22 |
+
"loss": 1.3932,
|
| 23 |
+
"step": 20
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.010666666666666666,
|
| 27 |
+
"grad_norm": 0.4800471365451813,
|
| 28 |
+
"learning_rate": 6e-06,
|
| 29 |
+
"loss": 1.6084,
|
| 30 |
+
"step": 30
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.014222222222222223,
|
| 34 |
+
"grad_norm": 0.513812780380249,
|
| 35 |
+
"learning_rate": 8.000000000000001e-06,
|
| 36 |
+
"loss": 1.5579,
|
| 37 |
+
"step": 40
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.017777777777777778,
|
| 41 |
+
"grad_norm": 0.5741678476333618,
|
| 42 |
+
"learning_rate": 1e-05,
|
| 43 |
+
"loss": 1.5945,
|
| 44 |
+
"step": 50
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.021333333333333333,
|
| 48 |
+
"grad_norm": 0.42774268984794617,
|
| 49 |
+
"learning_rate": 1.2e-05,
|
| 50 |
+
"loss": 1.6154,
|
| 51 |
+
"step": 60
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.024888888888888887,
|
| 55 |
+
"grad_norm": 0.37852978706359863,
|
| 56 |
+
"learning_rate": 1.4000000000000001e-05,
|
| 57 |
+
"loss": 1.5581,
|
| 58 |
+
"step": 70
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.028444444444444446,
|
| 62 |
+
"grad_norm": 0.48577743768692017,
|
| 63 |
+
"learning_rate": 1.6000000000000003e-05,
|
| 64 |
+
"loss": 1.57,
|
| 65 |
+
"step": 80
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.032,
|
| 69 |
+
"grad_norm": 0.5572775602340698,
|
| 70 |
+
"learning_rate": 1.8e-05,
|
| 71 |
+
"loss": 1.6154,
|
| 72 |
+
"step": 90
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 0.035555555555555556,
|
| 76 |
+
"grad_norm": 0.4657267928123474,
|
| 77 |
+
"learning_rate": 2e-05,
|
| 78 |
+
"loss": 1.5583,
|
| 79 |
+
"step": 100
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.03911111111111111,
|
| 83 |
+
"grad_norm": 0.48570355772972107,
|
| 84 |
+
"learning_rate": 2.2000000000000003e-05,
|
| 85 |
+
"loss": 1.6356,
|
| 86 |
+
"step": 110
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.042666666666666665,
|
| 90 |
+
"grad_norm": 0.5374159812927246,
|
| 91 |
+
"learning_rate": 2.4e-05,
|
| 92 |
+
"loss": 1.5911,
|
| 93 |
+
"step": 120
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 0.04622222222222222,
|
| 97 |
+
"grad_norm": 0.4544757306575775,
|
| 98 |
+
"learning_rate": 2.6000000000000002e-05,
|
| 99 |
+
"loss": 1.5462,
|
| 100 |
+
"step": 130
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"epoch": 0.049777777777777775,
|
| 104 |
+
"grad_norm": 0.49096816778182983,
|
| 105 |
+
"learning_rate": 2.8000000000000003e-05,
|
| 106 |
+
"loss": 1.4696,
|
| 107 |
+
"step": 140
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 0.05333333333333334,
|
| 111 |
+
"grad_norm": 0.45160388946533203,
|
| 112 |
+
"learning_rate": 3e-05,
|
| 113 |
+
"loss": 1.4397,
|
| 114 |
+
"step": 150
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 0.05688888888888889,
|
| 118 |
+
"grad_norm": 0.4654102325439453,
|
| 119 |
+
"learning_rate": 3.2000000000000005e-05,
|
| 120 |
+
"loss": 1.61,
|
| 121 |
+
"step": 160
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.060444444444444446,
|
| 125 |
+
"grad_norm": 0.5608298182487488,
|
| 126 |
+
"learning_rate": 3.4000000000000007e-05,
|
| 127 |
+
"loss": 1.5799,
|
| 128 |
+
"step": 170
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 0.064,
|
| 132 |
+
"grad_norm": 0.5087274312973022,
|
| 133 |
+
"learning_rate": 3.6e-05,
|
| 134 |
+
"loss": 1.6534,
|
| 135 |
+
"step": 180
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 0.06755555555555555,
|
| 139 |
+
"grad_norm": 0.5886797308921814,
|
| 140 |
+
"learning_rate": 3.8e-05,
|
| 141 |
+
"loss": 1.5561,
|
| 142 |
+
"step": 190
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"epoch": 0.07111111111111111,
|
| 146 |
+
"grad_norm": 0.5276679396629333,
|
| 147 |
+
"learning_rate": 4e-05,
|
| 148 |
+
"loss": 1.5705,
|
| 149 |
+
"step": 200
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 0.07466666666666667,
|
| 153 |
+
"grad_norm": 0.4331255257129669,
|
| 154 |
+
"learning_rate": 4.2e-05,
|
| 155 |
+
"loss": 1.554,
|
| 156 |
+
"step": 210
|
| 157 |
+
},
|
| 158 |
+
{
|
| 159 |
+
"epoch": 0.07822222222222222,
|
| 160 |
+
"grad_norm": 0.5068897604942322,
|
| 161 |
+
"learning_rate": 4.4000000000000006e-05,
|
| 162 |
+
"loss": 1.451,
|
| 163 |
+
"step": 220
|
| 164 |
+
},
|
| 165 |
+
{
|
| 166 |
+
"epoch": 0.08177777777777778,
|
| 167 |
+
"grad_norm": 0.48896610736846924,
|
| 168 |
+
"learning_rate": 4.600000000000001e-05,
|
| 169 |
+
"loss": 1.6522,
|
| 170 |
+
"step": 230
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 0.08533333333333333,
|
| 174 |
+
"grad_norm": 0.5350568890571594,
|
| 175 |
+
"learning_rate": 4.8e-05,
|
| 176 |
+
"loss": 1.5571,
|
| 177 |
+
"step": 240
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 0.08888888888888889,
|
| 181 |
+
"grad_norm": 0.6370730400085449,
|
| 182 |
+
"learning_rate": 5e-05,
|
| 183 |
+
"loss": 1.6112,
|
| 184 |
+
"step": 250
|
| 185 |
+
},
|
| 186 |
+
{
|
| 187 |
+
"epoch": 0.09244444444444444,
|
| 188 |
+
"grad_norm": 0.5409080386161804,
|
| 189 |
+
"learning_rate": 5.2000000000000004e-05,
|
| 190 |
+
"loss": 1.5878,
|
| 191 |
+
"step": 260
|
| 192 |
+
},
|
| 193 |
+
{
|
| 194 |
+
"epoch": 0.096,
|
| 195 |
+
"grad_norm": 0.5505415797233582,
|
| 196 |
+
"learning_rate": 5.4000000000000005e-05,
|
| 197 |
+
"loss": 1.6938,
|
| 198 |
+
"step": 270
|
| 199 |
+
},
|
| 200 |
+
{
|
| 201 |
+
"epoch": 0.09955555555555555,
|
| 202 |
+
"grad_norm": 0.557578980922699,
|
| 203 |
+
"learning_rate": 5.6000000000000006e-05,
|
| 204 |
+
"loss": 1.6776,
|
| 205 |
+
"step": 280
|
| 206 |
+
},
|
| 207 |
+
{
|
| 208 |
+
"epoch": 0.10311111111111111,
|
| 209 |
+
"grad_norm": 0.5505691170692444,
|
| 210 |
+
"learning_rate": 5.8e-05,
|
| 211 |
+
"loss": 1.7429,
|
| 212 |
+
"step": 290
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 0.10666666666666667,
|
| 216 |
+
"grad_norm": 0.5328088998794556,
|
| 217 |
+
"learning_rate": 6e-05,
|
| 218 |
+
"loss": 1.6633,
|
| 219 |
+
"step": 300
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 0.11022222222222222,
|
| 223 |
+
"grad_norm": 0.4911518394947052,
|
| 224 |
+
"learning_rate": 6.2e-05,
|
| 225 |
+
"loss": 1.4746,
|
| 226 |
+
"step": 310
|
| 227 |
+
},
|
| 228 |
+
{
|
| 229 |
+
"epoch": 0.11377777777777778,
|
| 230 |
+
"grad_norm": 0.5215829610824585,
|
| 231 |
+
"learning_rate": 6.400000000000001e-05,
|
| 232 |
+
"loss": 1.4723,
|
| 233 |
+
"step": 320
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 0.11733333333333333,
|
| 237 |
+
"grad_norm": 0.5551868081092834,
|
| 238 |
+
"learning_rate": 6.6e-05,
|
| 239 |
+
"loss": 1.5322,
|
| 240 |
+
"step": 330
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 0.12088888888888889,
|
| 244 |
+
"grad_norm": 0.607334554195404,
|
| 245 |
+
"learning_rate": 6.800000000000001e-05,
|
| 246 |
+
"loss": 1.7652,
|
| 247 |
+
"step": 340
|
| 248 |
+
},
|
| 249 |
+
{
|
| 250 |
+
"epoch": 0.12444444444444444,
|
| 251 |
+
"grad_norm": 0.4849174916744232,
|
| 252 |
+
"learning_rate": 7e-05,
|
| 253 |
+
"loss": 1.5915,
|
| 254 |
+
"step": 350
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 0.128,
|
| 258 |
+
"grad_norm": 0.5409530997276306,
|
| 259 |
+
"learning_rate": 7.2e-05,
|
| 260 |
+
"loss": 1.6263,
|
| 261 |
+
"step": 360
|
| 262 |
+
},
|
| 263 |
+
{
|
| 264 |
+
"epoch": 0.13155555555555556,
|
| 265 |
+
"grad_norm": 0.5955730080604553,
|
| 266 |
+
"learning_rate": 7.4e-05,
|
| 267 |
+
"loss": 1.598,
|
| 268 |
+
"step": 370
|
| 269 |
+
},
|
| 270 |
+
{
|
| 271 |
+
"epoch": 0.1351111111111111,
|
| 272 |
+
"grad_norm": 0.5792020559310913,
|
| 273 |
+
"learning_rate": 7.6e-05,
|
| 274 |
+
"loss": 1.7498,
|
| 275 |
+
"step": 380
|
| 276 |
+
},
|
| 277 |
+
{
|
| 278 |
+
"epoch": 0.13866666666666666,
|
| 279 |
+
"grad_norm": 0.6078203916549683,
|
| 280 |
+
"learning_rate": 7.800000000000001e-05,
|
| 281 |
+
"loss": 1.5885,
|
| 282 |
+
"step": 390
|
| 283 |
+
},
|
| 284 |
+
{
|
| 285 |
+
"epoch": 0.14222222222222222,
|
| 286 |
+
"grad_norm": 0.470548540353775,
|
| 287 |
+
"learning_rate": 8e-05,
|
| 288 |
+
"loss": 1.4596,
|
| 289 |
+
"step": 400
|
| 290 |
+
},
|
| 291 |
+
{
|
| 292 |
+
"epoch": 0.14577777777777778,
|
| 293 |
+
"grad_norm": 0.600089967250824,
|
| 294 |
+
"learning_rate": 8.2e-05,
|
| 295 |
+
"loss": 1.6343,
|
| 296 |
+
"step": 410
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 0.14933333333333335,
|
| 300 |
+
"grad_norm": 0.5364957451820374,
|
| 301 |
+
"learning_rate": 8.4e-05,
|
| 302 |
+
"loss": 1.4748,
|
| 303 |
+
"step": 420
|
| 304 |
+
},
|
| 305 |
+
{
|
| 306 |
+
"epoch": 0.15288888888888888,
|
| 307 |
+
"grad_norm": 0.5835921168327332,
|
| 308 |
+
"learning_rate": 8.6e-05,
|
| 309 |
+
"loss": 1.6809,
|
| 310 |
+
"step": 430
|
| 311 |
+
},
|
| 312 |
+
{
|
| 313 |
+
"epoch": 0.15644444444444444,
|
| 314 |
+
"grad_norm": 0.5760766863822937,
|
| 315 |
+
"learning_rate": 8.800000000000001e-05,
|
| 316 |
+
"loss": 1.6276,
|
| 317 |
+
"step": 440
|
| 318 |
+
},
|
| 319 |
+
{
|
| 320 |
+
"epoch": 0.16,
|
| 321 |
+
"grad_norm": 0.5958523154258728,
|
| 322 |
+
"learning_rate": 9e-05,
|
| 323 |
+
"loss": 1.4696,
|
| 324 |
+
"step": 450
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 0.16355555555555557,
|
| 328 |
+
"grad_norm": 0.5406598448753357,
|
| 329 |
+
"learning_rate": 9.200000000000001e-05,
|
| 330 |
+
"loss": 1.625,
|
| 331 |
+
"step": 460
|
| 332 |
+
},
|
| 333 |
+
{
|
| 334 |
+
"epoch": 0.1671111111111111,
|
| 335 |
+
"grad_norm": 0.5989697575569153,
|
| 336 |
+
"learning_rate": 9.4e-05,
|
| 337 |
+
"loss": 1.6854,
|
| 338 |
+
"step": 470
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 0.17066666666666666,
|
| 342 |
+
"grad_norm": 0.594978392124176,
|
| 343 |
+
"learning_rate": 9.6e-05,
|
| 344 |
+
"loss": 1.5774,
|
| 345 |
+
"step": 480
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 0.17422222222222222,
|
| 349 |
+
"grad_norm": 0.5095472931861877,
|
| 350 |
+
"learning_rate": 9.8e-05,
|
| 351 |
+
"loss": 1.7095,
|
| 352 |
+
"step": 490
|
| 353 |
+
},
|
| 354 |
+
{
|
| 355 |
+
"epoch": 0.17777777777777778,
|
| 356 |
+
"grad_norm": 0.7304088473320007,
|
| 357 |
+
"learning_rate": 0.0001,
|
| 358 |
+
"loss": 1.6842,
|
| 359 |
+
"step": 500
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"epoch": 0.18133333333333335,
|
| 363 |
+
"grad_norm": 0.6202882528305054,
|
| 364 |
+
"learning_rate": 9.95674740484429e-05,
|
| 365 |
+
"loss": 1.696,
|
| 366 |
+
"step": 510
|
| 367 |
+
},
|
| 368 |
+
{
|
| 369 |
+
"epoch": 0.18488888888888888,
|
| 370 |
+
"grad_norm": 0.49087047576904297,
|
| 371 |
+
"learning_rate": 9.913494809688582e-05,
|
| 372 |
+
"loss": 1.6507,
|
| 373 |
+
"step": 520
|
| 374 |
+
},
|
| 375 |
+
{
|
| 376 |
+
"epoch": 0.18844444444444444,
|
| 377 |
+
"grad_norm": 0.6279426217079163,
|
| 378 |
+
"learning_rate": 9.870242214532872e-05,
|
| 379 |
+
"loss": 1.6809,
|
| 380 |
+
"step": 530
|
| 381 |
+
},
|
| 382 |
+
{
|
| 383 |
+
"epoch": 0.192,
|
| 384 |
+
"grad_norm": 0.602878987789154,
|
| 385 |
+
"learning_rate": 9.826989619377162e-05,
|
| 386 |
+
"loss": 1.6173,
|
| 387 |
+
"step": 540
|
| 388 |
+
},
|
| 389 |
+
{
|
| 390 |
+
"epoch": 0.19555555555555557,
|
| 391 |
+
"grad_norm": 0.6200081706047058,
|
| 392 |
+
"learning_rate": 9.783737024221454e-05,
|
| 393 |
+
"loss": 1.6787,
|
| 394 |
+
"step": 550
|
| 395 |
+
},
|
| 396 |
+
{
|
| 397 |
+
"epoch": 0.1991111111111111,
|
| 398 |
+
"grad_norm": 0.6582353115081787,
|
| 399 |
+
"learning_rate": 9.740484429065745e-05,
|
| 400 |
+
"loss": 1.7095,
|
| 401 |
+
"step": 560
|
| 402 |
+
},
|
| 403 |
+
{
|
| 404 |
+
"epoch": 0.20266666666666666,
|
| 405 |
+
"grad_norm": 0.541404664516449,
|
| 406 |
+
"learning_rate": 9.697231833910035e-05,
|
| 407 |
+
"loss": 1.8418,
|
| 408 |
+
"step": 570
|
| 409 |
+
},
|
| 410 |
+
{
|
| 411 |
+
"epoch": 0.20622222222222222,
|
| 412 |
+
"grad_norm": 0.5303178429603577,
|
| 413 |
+
"learning_rate": 9.653979238754325e-05,
|
| 414 |
+
"loss": 1.6112,
|
| 415 |
+
"step": 580
|
| 416 |
+
},
|
| 417 |
+
{
|
| 418 |
+
"epoch": 0.20977777777777779,
|
| 419 |
+
"grad_norm": 0.527113676071167,
|
| 420 |
+
"learning_rate": 9.610726643598617e-05,
|
| 421 |
+
"loss": 1.6454,
|
| 422 |
+
"step": 590
|
| 423 |
+
},
|
| 424 |
+
{
|
| 425 |
+
"epoch": 0.21333333333333335,
|
| 426 |
+
"grad_norm": 0.5398814678192139,
|
| 427 |
+
"learning_rate": 9.567474048442907e-05,
|
| 428 |
+
"loss": 1.5353,
|
| 429 |
+
"step": 600
|
| 430 |
+
},
|
| 431 |
+
{
|
| 432 |
+
"epoch": 0.21688888888888888,
|
| 433 |
+
"grad_norm": 0.5357266068458557,
|
| 434 |
+
"learning_rate": 9.524221453287197e-05,
|
| 435 |
+
"loss": 1.6816,
|
| 436 |
+
"step": 610
|
| 437 |
+
},
|
| 438 |
+
{
|
| 439 |
+
"epoch": 0.22044444444444444,
|
| 440 |
+
"grad_norm": 0.4769757390022278,
|
| 441 |
+
"learning_rate": 9.480968858131488e-05,
|
| 442 |
+
"loss": 1.6625,
|
| 443 |
+
"step": 620
|
| 444 |
+
},
|
| 445 |
+
{
|
| 446 |
+
"epoch": 0.224,
|
| 447 |
+
"grad_norm": 0.5274131298065186,
|
| 448 |
+
"learning_rate": 9.437716262975779e-05,
|
| 449 |
+
"loss": 1.7129,
|
| 450 |
+
"step": 630
|
| 451 |
+
},
|
| 452 |
+
{
|
| 453 |
+
"epoch": 0.22755555555555557,
|
| 454 |
+
"grad_norm": 0.6454447507858276,
|
| 455 |
+
"learning_rate": 9.394463667820069e-05,
|
| 456 |
+
"loss": 1.5577,
|
| 457 |
+
"step": 640
|
| 458 |
+
},
|
| 459 |
+
{
|
| 460 |
+
"epoch": 0.2311111111111111,
|
| 461 |
+
"grad_norm": 0.5967287421226501,
|
| 462 |
+
"learning_rate": 9.35121107266436e-05,
|
| 463 |
+
"loss": 1.3905,
|
| 464 |
+
"step": 650
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 0.23466666666666666,
|
| 468 |
+
"grad_norm": 0.57136470079422,
|
| 469 |
+
"learning_rate": 9.307958477508652e-05,
|
| 470 |
+
"loss": 1.6029,
|
| 471 |
+
"step": 660
|
| 472 |
+
},
|
| 473 |
+
{
|
| 474 |
+
"epoch": 0.23822222222222222,
|
| 475 |
+
"grad_norm": 0.5300968885421753,
|
| 476 |
+
"learning_rate": 9.264705882352942e-05,
|
| 477 |
+
"loss": 1.7363,
|
| 478 |
+
"step": 670
|
| 479 |
+
},
|
| 480 |
+
{
|
| 481 |
+
"epoch": 0.24177777777777779,
|
| 482 |
+
"grad_norm": 0.44372138381004333,
|
| 483 |
+
"learning_rate": 9.221453287197233e-05,
|
| 484 |
+
"loss": 1.6467,
|
| 485 |
+
"step": 680
|
| 486 |
+
},
|
| 487 |
+
{
|
| 488 |
+
"epoch": 0.24533333333333332,
|
| 489 |
+
"grad_norm": 0.45939210057258606,
|
| 490 |
+
"learning_rate": 9.178200692041523e-05,
|
| 491 |
+
"loss": 1.5413,
|
| 492 |
+
"step": 690
|
| 493 |
+
},
|
| 494 |
+
{
|
| 495 |
+
"epoch": 0.24888888888888888,
|
| 496 |
+
"grad_norm": 0.5719866752624512,
|
| 497 |
+
"learning_rate": 9.134948096885813e-05,
|
| 498 |
+
"loss": 1.5702,
|
| 499 |
+
"step": 700
|
| 500 |
+
},
|
| 501 |
+
{
|
| 502 |
+
"epoch": 0.25244444444444447,
|
| 503 |
+
"grad_norm": 0.4535101652145386,
|
| 504 |
+
"learning_rate": 9.091695501730105e-05,
|
| 505 |
+
"loss": 1.5331,
|
| 506 |
+
"step": 710
|
| 507 |
+
},
|
| 508 |
+
{
|
| 509 |
+
"epoch": 0.256,
|
| 510 |
+
"grad_norm": 0.5633770227432251,
|
| 511 |
+
"learning_rate": 9.048442906574395e-05,
|
| 512 |
+
"loss": 1.696,
|
| 513 |
+
"step": 720
|
| 514 |
+
},
|
| 515 |
+
{
|
| 516 |
+
"epoch": 0.25955555555555554,
|
| 517 |
+
"grad_norm": 0.6633341908454895,
|
| 518 |
+
"learning_rate": 9.005190311418685e-05,
|
| 519 |
+
"loss": 1.5428,
|
| 520 |
+
"step": 730
|
| 521 |
+
},
|
| 522 |
+
{
|
| 523 |
+
"epoch": 0.26311111111111113,
|
| 524 |
+
"grad_norm": 0.6005223393440247,
|
| 525 |
+
"learning_rate": 8.961937716262977e-05,
|
| 526 |
+
"loss": 1.8278,
|
| 527 |
+
"step": 740
|
| 528 |
+
},
|
| 529 |
+
{
|
| 530 |
+
"epoch": 0.26666666666666666,
|
| 531 |
+
"grad_norm": 0.5239768624305725,
|
| 532 |
+
"learning_rate": 8.918685121107267e-05,
|
| 533 |
+
"loss": 1.6761,
|
| 534 |
+
"step": 750
|
| 535 |
+
},
|
| 536 |
+
{
|
| 537 |
+
"epoch": 0.2702222222222222,
|
| 538 |
+
"grad_norm": 0.4788583517074585,
|
| 539 |
+
"learning_rate": 8.875432525951558e-05,
|
| 540 |
+
"loss": 1.548,
|
| 541 |
+
"step": 760
|
| 542 |
+
},
|
| 543 |
+
{
|
| 544 |
+
"epoch": 0.2737777777777778,
|
| 545 |
+
"grad_norm": 0.5428124666213989,
|
| 546 |
+
"learning_rate": 8.832179930795848e-05,
|
| 547 |
+
"loss": 1.6972,
|
| 548 |
+
"step": 770
|
| 549 |
+
},
|
| 550 |
+
{
|
| 551 |
+
"epoch": 0.2773333333333333,
|
| 552 |
+
"grad_norm": 0.42001649737358093,
|
| 553 |
+
"learning_rate": 8.78892733564014e-05,
|
| 554 |
+
"loss": 1.6244,
|
| 555 |
+
"step": 780
|
| 556 |
+
},
|
| 557 |
+
{
|
| 558 |
+
"epoch": 0.2808888888888889,
|
| 559 |
+
"grad_norm": 0.5100952386856079,
|
| 560 |
+
"learning_rate": 8.74567474048443e-05,
|
| 561 |
+
"loss": 1.6704,
|
| 562 |
+
"step": 790
|
| 563 |
+
},
|
| 564 |
+
{
|
| 565 |
+
"epoch": 0.28444444444444444,
|
| 566 |
+
"grad_norm": 0.49174782633781433,
|
| 567 |
+
"learning_rate": 8.70242214532872e-05,
|
| 568 |
+
"loss": 1.6876,
|
| 569 |
+
"step": 800
|
| 570 |
+
},
|
| 571 |
+
{
|
| 572 |
+
"epoch": 0.288,
|
| 573 |
+
"grad_norm": 0.543880820274353,
|
| 574 |
+
"learning_rate": 8.659169550173011e-05,
|
| 575 |
+
"loss": 1.599,
|
| 576 |
+
"step": 810
|
| 577 |
+
},
|
| 578 |
+
{
|
| 579 |
+
"epoch": 0.29155555555555557,
|
| 580 |
+
"grad_norm": 0.4966542422771454,
|
| 581 |
+
"learning_rate": 8.615916955017301e-05,
|
| 582 |
+
"loss": 1.4723,
|
| 583 |
+
"step": 820
|
| 584 |
+
},
|
| 585 |
+
{
|
| 586 |
+
"epoch": 0.2951111111111111,
|
| 587 |
+
"grad_norm": 0.44916948676109314,
|
| 588 |
+
"learning_rate": 8.572664359861592e-05,
|
| 589 |
+
"loss": 1.452,
|
| 590 |
+
"step": 830
|
| 591 |
+
},
|
| 592 |
+
{
|
| 593 |
+
"epoch": 0.2986666666666667,
|
| 594 |
+
"grad_norm": 0.5791043639183044,
|
| 595 |
+
"learning_rate": 8.529411764705883e-05,
|
| 596 |
+
"loss": 1.7569,
|
| 597 |
+
"step": 840
|
| 598 |
+
},
|
| 599 |
+
{
|
| 600 |
+
"epoch": 0.3022222222222222,
|
| 601 |
+
"grad_norm": 0.5880780816078186,
|
| 602 |
+
"learning_rate": 8.486159169550173e-05,
|
| 603 |
+
"loss": 1.4779,
|
| 604 |
+
"step": 850
|
| 605 |
+
},
|
| 606 |
+
{
|
| 607 |
+
"epoch": 0.30577777777777776,
|
| 608 |
+
"grad_norm": 0.5578981041908264,
|
| 609 |
+
"learning_rate": 8.442906574394463e-05,
|
| 610 |
+
"loss": 1.6577,
|
| 611 |
+
"step": 860
|
| 612 |
+
},
|
| 613 |
+
{
|
| 614 |
+
"epoch": 0.30933333333333335,
|
| 615 |
+
"grad_norm": 0.6132334470748901,
|
| 616 |
+
"learning_rate": 8.399653979238755e-05,
|
| 617 |
+
"loss": 1.5862,
|
| 618 |
+
"step": 870
|
| 619 |
+
},
|
| 620 |
+
{
|
| 621 |
+
"epoch": 0.3128888888888889,
|
| 622 |
+
"grad_norm": 0.512484073638916,
|
| 623 |
+
"learning_rate": 8.356401384083046e-05,
|
| 624 |
+
"loss": 1.4891,
|
| 625 |
+
"step": 880
|
| 626 |
+
},
|
| 627 |
+
{
|
| 628 |
+
"epoch": 0.3164444444444444,
|
| 629 |
+
"grad_norm": 0.4793022871017456,
|
| 630 |
+
"learning_rate": 8.313148788927336e-05,
|
| 631 |
+
"loss": 1.601,
|
| 632 |
+
"step": 890
|
| 633 |
+
},
|
| 634 |
+
{
|
| 635 |
+
"epoch": 0.32,
|
| 636 |
+
"grad_norm": 0.5512364506721497,
|
| 637 |
+
"learning_rate": 8.269896193771626e-05,
|
| 638 |
+
"loss": 1.6191,
|
| 639 |
+
"step": 900
|
| 640 |
+
},
|
| 641 |
+
{
|
| 642 |
+
"epoch": 0.32355555555555554,
|
| 643 |
+
"grad_norm": 0.5528599619865417,
|
| 644 |
+
"learning_rate": 8.226643598615918e-05,
|
| 645 |
+
"loss": 1.6555,
|
| 646 |
+
"step": 910
|
| 647 |
+
},
|
| 648 |
+
{
|
| 649 |
+
"epoch": 0.32711111111111113,
|
| 650 |
+
"grad_norm": 0.4718135595321655,
|
| 651 |
+
"learning_rate": 8.183391003460208e-05,
|
| 652 |
+
"loss": 1.3796,
|
| 653 |
+
"step": 920
|
| 654 |
+
},
|
| 655 |
+
{
|
| 656 |
+
"epoch": 0.33066666666666666,
|
| 657 |
+
"grad_norm": 0.5744655728340149,
|
| 658 |
+
"learning_rate": 8.140138408304498e-05,
|
| 659 |
+
"loss": 1.7323,
|
| 660 |
+
"step": 930
|
| 661 |
+
},
|
| 662 |
+
{
|
| 663 |
+
"epoch": 0.3342222222222222,
|
| 664 |
+
"grad_norm": 0.5298731923103333,
|
| 665 |
+
"learning_rate": 8.09688581314879e-05,
|
| 666 |
+
"loss": 1.6613,
|
| 667 |
+
"step": 940
|
| 668 |
+
},
|
| 669 |
+
{
|
| 670 |
+
"epoch": 0.3377777777777778,
|
| 671 |
+
"grad_norm": 0.507189154624939,
|
| 672 |
+
"learning_rate": 8.05363321799308e-05,
|
| 673 |
+
"loss": 1.4932,
|
| 674 |
+
"step": 950
|
| 675 |
+
},
|
| 676 |
+
{
|
| 677 |
+
"epoch": 0.3413333333333333,
|
| 678 |
+
"grad_norm": 0.5233726501464844,
|
| 679 |
+
"learning_rate": 8.01038062283737e-05,
|
| 680 |
+
"loss": 1.5989,
|
| 681 |
+
"step": 960
|
| 682 |
+
},
|
| 683 |
+
{
|
| 684 |
+
"epoch": 0.3448888888888889,
|
| 685 |
+
"grad_norm": 0.5754904747009277,
|
| 686 |
+
"learning_rate": 7.967128027681662e-05,
|
| 687 |
+
"loss": 1.6657,
|
| 688 |
+
"step": 970
|
| 689 |
+
},
|
| 690 |
+
{
|
| 691 |
+
"epoch": 0.34844444444444445,
|
| 692 |
+
"grad_norm": 0.6196070909500122,
|
| 693 |
+
"learning_rate": 7.923875432525953e-05,
|
| 694 |
+
"loss": 1.6808,
|
| 695 |
+
"step": 980
|
| 696 |
+
},
|
| 697 |
+
{
|
| 698 |
+
"epoch": 0.352,
|
| 699 |
+
"grad_norm": 0.461910605430603,
|
| 700 |
+
"learning_rate": 7.880622837370243e-05,
|
| 701 |
+
"loss": 1.6012,
|
| 702 |
+
"step": 990
|
| 703 |
+
},
|
| 704 |
+
{
|
| 705 |
+
"epoch": 0.35555555555555557,
|
| 706 |
+
"grad_norm": 0.4632907807826996,
|
| 707 |
+
"learning_rate": 7.837370242214533e-05,
|
| 708 |
+
"loss": 1.5879,
|
| 709 |
+
"step": 1000
|
| 710 |
+
},
|
| 711 |
+
{
|
| 712 |
+
"epoch": 0.3591111111111111,
|
| 713 |
+
"grad_norm": 0.48907166719436646,
|
| 714 |
+
"learning_rate": 7.794117647058824e-05,
|
| 715 |
+
"loss": 1.6109,
|
| 716 |
+
"step": 1010
|
| 717 |
+
},
|
| 718 |
+
{
|
| 719 |
+
"epoch": 0.3626666666666667,
|
| 720 |
+
"grad_norm": 0.5417441129684448,
|
| 721 |
+
"learning_rate": 7.750865051903114e-05,
|
| 722 |
+
"loss": 1.5687,
|
| 723 |
+
"step": 1020
|
| 724 |
+
},
|
| 725 |
+
{
|
| 726 |
+
"epoch": 0.3662222222222222,
|
| 727 |
+
"grad_norm": 0.519092857837677,
|
| 728 |
+
"learning_rate": 7.707612456747404e-05,
|
| 729 |
+
"loss": 1.6197,
|
| 730 |
+
"step": 1030
|
| 731 |
+
},
|
| 732 |
+
{
|
| 733 |
+
"epoch": 0.36977777777777776,
|
| 734 |
+
"grad_norm": 0.45026785135269165,
|
| 735 |
+
"learning_rate": 7.664359861591696e-05,
|
| 736 |
+
"loss": 1.5194,
|
| 737 |
+
"step": 1040
|
| 738 |
+
},
|
| 739 |
+
{
|
| 740 |
+
"epoch": 0.37333333333333335,
|
| 741 |
+
"grad_norm": 0.5378026366233826,
|
| 742 |
+
"learning_rate": 7.621107266435986e-05,
|
| 743 |
+
"loss": 1.419,
|
| 744 |
+
"step": 1050
|
| 745 |
+
},
|
| 746 |
+
{
|
| 747 |
+
"epoch": 0.3768888888888889,
|
| 748 |
+
"grad_norm": 0.4889166057109833,
|
| 749 |
+
"learning_rate": 7.577854671280276e-05,
|
| 750 |
+
"loss": 1.6356,
|
| 751 |
+
"step": 1060
|
| 752 |
+
},
|
| 753 |
+
{
|
| 754 |
+
"epoch": 0.3804444444444444,
|
| 755 |
+
"grad_norm": 0.5200234055519104,
|
| 756 |
+
"learning_rate": 7.534602076124569e-05,
|
| 757 |
+
"loss": 1.7118,
|
| 758 |
+
"step": 1070
|
| 759 |
+
},
|
| 760 |
+
{
|
| 761 |
+
"epoch": 0.384,
|
| 762 |
+
"grad_norm": 0.4398309886455536,
|
| 763 |
+
"learning_rate": 7.491349480968859e-05,
|
| 764 |
+
"loss": 1.525,
|
| 765 |
+
"step": 1080
|
| 766 |
+
},
|
| 767 |
+
{
|
| 768 |
+
"epoch": 0.38755555555555554,
|
| 769 |
+
"grad_norm": 0.43217843770980835,
|
| 770 |
+
"learning_rate": 7.448096885813149e-05,
|
| 771 |
+
"loss": 1.5643,
|
| 772 |
+
"step": 1090
|
| 773 |
+
},
|
| 774 |
+
{
|
| 775 |
+
"epoch": 0.39111111111111113,
|
| 776 |
+
"grad_norm": 0.5189200043678284,
|
| 777 |
+
"learning_rate": 7.40484429065744e-05,
|
| 778 |
+
"loss": 1.6891,
|
| 779 |
+
"step": 1100
|
| 780 |
+
},
|
| 781 |
+
{
|
| 782 |
+
"epoch": 0.39466666666666667,
|
| 783 |
+
"grad_norm": 0.5237637758255005,
|
| 784 |
+
"learning_rate": 7.36159169550173e-05,
|
| 785 |
+
"loss": 1.4445,
|
| 786 |
+
"step": 1110
|
| 787 |
+
},
|
| 788 |
+
{
|
| 789 |
+
"epoch": 0.3982222222222222,
|
| 790 |
+
"grad_norm": 0.5614046454429626,
|
| 791 |
+
"learning_rate": 7.318339100346021e-05,
|
| 792 |
+
"loss": 1.6439,
|
| 793 |
+
"step": 1120
|
| 794 |
+
},
|
| 795 |
+
{
|
| 796 |
+
"epoch": 0.4017777777777778,
|
| 797 |
+
"grad_norm": 0.4191811680793762,
|
| 798 |
+
"learning_rate": 7.275086505190312e-05,
|
| 799 |
+
"loss": 1.4344,
|
| 800 |
+
"step": 1130
|
| 801 |
+
},
|
| 802 |
+
{
|
| 803 |
+
"epoch": 0.4053333333333333,
|
| 804 |
+
"grad_norm": 0.5043474435806274,
|
| 805 |
+
"learning_rate": 7.231833910034602e-05,
|
| 806 |
+
"loss": 1.7017,
|
| 807 |
+
"step": 1140
|
| 808 |
+
},
|
| 809 |
+
{
|
| 810 |
+
"epoch": 0.4088888888888889,
|
| 811 |
+
"grad_norm": 0.40949752926826477,
|
| 812 |
+
"learning_rate": 7.188581314878892e-05,
|
| 813 |
+
"loss": 1.5903,
|
| 814 |
+
"step": 1150
|
| 815 |
+
},
|
| 816 |
+
{
|
| 817 |
+
"epoch": 0.41244444444444445,
|
| 818 |
+
"grad_norm": 0.4599238634109497,
|
| 819 |
+
"learning_rate": 7.145328719723184e-05,
|
| 820 |
+
"loss": 1.5666,
|
| 821 |
+
"step": 1160
|
| 822 |
+
},
|
| 823 |
+
{
|
| 824 |
+
"epoch": 0.416,
|
| 825 |
+
"grad_norm": 0.5736513137817383,
|
| 826 |
+
"learning_rate": 7.102076124567474e-05,
|
| 827 |
+
"loss": 1.5525,
|
| 828 |
+
"step": 1170
|
| 829 |
+
},
|
| 830 |
+
{
|
| 831 |
+
"epoch": 0.41955555555555557,
|
| 832 |
+
"grad_norm": 0.5437818169593811,
|
| 833 |
+
"learning_rate": 7.058823529411765e-05,
|
| 834 |
+
"loss": 1.6668,
|
| 835 |
+
"step": 1180
|
| 836 |
+
},
|
| 837 |
+
{
|
| 838 |
+
"epoch": 0.4231111111111111,
|
| 839 |
+
"grad_norm": 0.4318547248840332,
|
| 840 |
+
"learning_rate": 7.015570934256056e-05,
|
| 841 |
+
"loss": 1.4075,
|
| 842 |
+
"step": 1190
|
| 843 |
+
},
|
| 844 |
+
{
|
| 845 |
+
"epoch": 0.4266666666666667,
|
| 846 |
+
"grad_norm": 0.44979867339134216,
|
| 847 |
+
"learning_rate": 6.972318339100347e-05,
|
| 848 |
+
"loss": 1.3973,
|
| 849 |
+
"step": 1200
|
| 850 |
+
},
|
| 851 |
+
{
|
| 852 |
+
"epoch": 0.43022222222222223,
|
| 853 |
+
"grad_norm": 0.4653863310813904,
|
| 854 |
+
"learning_rate": 6.929065743944637e-05,
|
| 855 |
+
"loss": 1.4578,
|
| 856 |
+
"step": 1210
|
| 857 |
+
},
|
| 858 |
+
{
|
| 859 |
+
"epoch": 0.43377777777777776,
|
| 860 |
+
"grad_norm": 0.5257763266563416,
|
| 861 |
+
"learning_rate": 6.885813148788927e-05,
|
| 862 |
+
"loss": 1.5667,
|
| 863 |
+
"step": 1220
|
| 864 |
+
},
|
| 865 |
+
{
|
| 866 |
+
"epoch": 0.43733333333333335,
|
| 867 |
+
"grad_norm": 0.5117728114128113,
|
| 868 |
+
"learning_rate": 6.842560553633219e-05,
|
| 869 |
+
"loss": 1.6843,
|
| 870 |
+
"step": 1230
|
| 871 |
+
},
|
| 872 |
+
{
|
| 873 |
+
"epoch": 0.4408888888888889,
|
| 874 |
+
"grad_norm": 0.4611428678035736,
|
| 875 |
+
"learning_rate": 6.799307958477509e-05,
|
| 876 |
+
"loss": 1.7192,
|
| 877 |
+
"step": 1240
|
| 878 |
+
},
|
| 879 |
+
{
|
| 880 |
+
"epoch": 0.4444444444444444,
|
| 881 |
+
"grad_norm": 0.586694061756134,
|
| 882 |
+
"learning_rate": 6.756055363321799e-05,
|
| 883 |
+
"loss": 1.6057,
|
| 884 |
+
"step": 1250
|
| 885 |
+
},
|
| 886 |
+
{
|
| 887 |
+
"epoch": 0.448,
|
| 888 |
+
"grad_norm": 0.43433499336242676,
|
| 889 |
+
"learning_rate": 6.71280276816609e-05,
|
| 890 |
+
"loss": 1.6829,
|
| 891 |
+
"step": 1260
|
| 892 |
+
},
|
| 893 |
+
{
|
| 894 |
+
"epoch": 0.45155555555555554,
|
| 895 |
+
"grad_norm": 0.5851064920425415,
|
| 896 |
+
"learning_rate": 6.66955017301038e-05,
|
| 897 |
+
"loss": 1.6039,
|
| 898 |
+
"step": 1270
|
| 899 |
+
},
|
| 900 |
+
{
|
| 901 |
+
"epoch": 0.45511111111111113,
|
| 902 |
+
"grad_norm": 0.5549812912940979,
|
| 903 |
+
"learning_rate": 6.626297577854672e-05,
|
| 904 |
+
"loss": 1.6076,
|
| 905 |
+
"step": 1280
|
| 906 |
+
},
|
| 907 |
+
{
|
| 908 |
+
"epoch": 0.45866666666666667,
|
| 909 |
+
"grad_norm": 0.4489051103591919,
|
| 910 |
+
"learning_rate": 6.583044982698962e-05,
|
| 911 |
+
"loss": 1.6347,
|
| 912 |
+
"step": 1290
|
| 913 |
+
},
|
| 914 |
+
{
|
| 915 |
+
"epoch": 0.4622222222222222,
|
| 916 |
+
"grad_norm": 0.5636203289031982,
|
| 917 |
+
"learning_rate": 6.539792387543253e-05,
|
| 918 |
+
"loss": 1.3819,
|
| 919 |
+
"step": 1300
|
| 920 |
+
},
|
| 921 |
+
{
|
| 922 |
+
"epoch": 0.4657777777777778,
|
| 923 |
+
"grad_norm": 0.5598219037055969,
|
| 924 |
+
"learning_rate": 6.496539792387544e-05,
|
| 925 |
+
"loss": 1.6914,
|
| 926 |
+
"step": 1310
|
| 927 |
+
},
|
| 928 |
+
{
|
| 929 |
+
"epoch": 0.4693333333333333,
|
| 930 |
+
"grad_norm": 0.44435766339302063,
|
| 931 |
+
"learning_rate": 6.453287197231834e-05,
|
| 932 |
+
"loss": 1.5487,
|
| 933 |
+
"step": 1320
|
| 934 |
+
},
|
| 935 |
+
{
|
| 936 |
+
"epoch": 0.4728888888888889,
|
| 937 |
+
"grad_norm": 0.4664863646030426,
|
| 938 |
+
"learning_rate": 6.410034602076125e-05,
|
| 939 |
+
"loss": 1.5156,
|
| 940 |
+
"step": 1330
|
| 941 |
+
},
|
| 942 |
+
{
|
| 943 |
+
"epoch": 0.47644444444444445,
|
| 944 |
+
"grad_norm": 0.5192257761955261,
|
| 945 |
+
"learning_rate": 6.366782006920415e-05,
|
| 946 |
+
"loss": 1.5507,
|
| 947 |
+
"step": 1340
|
| 948 |
+
},
|
| 949 |
+
{
|
| 950 |
+
"epoch": 0.48,
|
| 951 |
+
"grad_norm": 0.6014835238456726,
|
| 952 |
+
"learning_rate": 6.323529411764705e-05,
|
| 953 |
+
"loss": 1.4809,
|
| 954 |
+
"step": 1350
|
| 955 |
+
},
|
| 956 |
+
{
|
| 957 |
+
"epoch": 0.48355555555555557,
|
| 958 |
+
"grad_norm": 0.5586997270584106,
|
| 959 |
+
"learning_rate": 6.280276816608997e-05,
|
| 960 |
+
"loss": 1.5619,
|
| 961 |
+
"step": 1360
|
| 962 |
+
},
|
| 963 |
+
{
|
| 964 |
+
"epoch": 0.4871111111111111,
|
| 965 |
+
"grad_norm": 0.4059390723705292,
|
| 966 |
+
"learning_rate": 6.237024221453287e-05,
|
| 967 |
+
"loss": 1.4799,
|
| 968 |
+
"step": 1370
|
| 969 |
+
},
|
| 970 |
+
{
|
| 971 |
+
"epoch": 0.49066666666666664,
|
| 972 |
+
"grad_norm": 0.41877785325050354,
|
| 973 |
+
"learning_rate": 6.193771626297578e-05,
|
| 974 |
+
"loss": 1.4187,
|
| 975 |
+
"step": 1380
|
| 976 |
+
},
|
| 977 |
+
{
|
| 978 |
+
"epoch": 0.49422222222222223,
|
| 979 |
+
"grad_norm": 0.4456830620765686,
|
| 980 |
+
"learning_rate": 6.15051903114187e-05,
|
| 981 |
+
"loss": 1.5507,
|
| 982 |
+
"step": 1390
|
| 983 |
+
},
|
| 984 |
+
{
|
| 985 |
+
"epoch": 0.49777777777777776,
|
| 986 |
+
"grad_norm": 0.45245644450187683,
|
| 987 |
+
"learning_rate": 6.10726643598616e-05,
|
| 988 |
+
"loss": 1.5478,
|
| 989 |
+
"step": 1400
|
| 990 |
+
},
|
| 991 |
+
{
|
| 992 |
+
"epoch": 0.5013333333333333,
|
| 993 |
+
"grad_norm": 0.4327283501625061,
|
| 994 |
+
"learning_rate": 6.06401384083045e-05,
|
| 995 |
+
"loss": 1.5734,
|
| 996 |
+
"step": 1410
|
| 997 |
+
},
|
| 998 |
+
{
|
| 999 |
+
"epoch": 0.5048888888888889,
|
| 1000 |
+
"grad_norm": 0.6047600507736206,
|
| 1001 |
+
"learning_rate": 6.020761245674741e-05,
|
| 1002 |
+
"loss": 1.6916,
|
| 1003 |
+
"step": 1420
|
| 1004 |
+
},
|
| 1005 |
+
{
|
| 1006 |
+
"epoch": 0.5084444444444445,
|
| 1007 |
+
"grad_norm": 0.4880015254020691,
|
| 1008 |
+
"learning_rate": 5.9775086505190316e-05,
|
| 1009 |
+
"loss": 1.6346,
|
| 1010 |
+
"step": 1430
|
| 1011 |
+
},
|
| 1012 |
+
{
|
| 1013 |
+
"epoch": 0.512,
|
| 1014 |
+
"grad_norm": 0.4689754843711853,
|
| 1015 |
+
"learning_rate": 5.934256055363322e-05,
|
| 1016 |
+
"loss": 1.5714,
|
| 1017 |
+
"step": 1440
|
| 1018 |
+
},
|
| 1019 |
+
{
|
| 1020 |
+
"epoch": 0.5155555555555555,
|
| 1021 |
+
"grad_norm": 0.4588066041469574,
|
| 1022 |
+
"learning_rate": 5.8910034602076125e-05,
|
| 1023 |
+
"loss": 1.5547,
|
| 1024 |
+
"step": 1450
|
| 1025 |
+
},
|
| 1026 |
+
{
|
| 1027 |
+
"epoch": 0.5191111111111111,
|
| 1028 |
+
"grad_norm": 0.45055222511291504,
|
| 1029 |
+
"learning_rate": 5.847750865051903e-05,
|
| 1030 |
+
"loss": 1.4591,
|
| 1031 |
+
"step": 1460
|
| 1032 |
+
},
|
| 1033 |
+
{
|
| 1034 |
+
"epoch": 0.5226666666666666,
|
| 1035 |
+
"grad_norm": 0.49660903215408325,
|
| 1036 |
+
"learning_rate": 5.8044982698961933e-05,
|
| 1037 |
+
"loss": 1.7409,
|
| 1038 |
+
"step": 1470
|
| 1039 |
+
},
|
| 1040 |
+
{
|
| 1041 |
+
"epoch": 0.5262222222222223,
|
| 1042 |
+
"grad_norm": 0.48975858092308044,
|
| 1043 |
+
"learning_rate": 5.761245674740484e-05,
|
| 1044 |
+
"loss": 1.5158,
|
| 1045 |
+
"step": 1480
|
| 1046 |
+
},
|
| 1047 |
+
{
|
| 1048 |
+
"epoch": 0.5297777777777778,
|
| 1049 |
+
"grad_norm": 0.4958665668964386,
|
| 1050 |
+
"learning_rate": 5.7179930795847756e-05,
|
| 1051 |
+
"loss": 1.4612,
|
| 1052 |
+
"step": 1490
|
| 1053 |
+
},
|
| 1054 |
+
{
|
| 1055 |
+
"epoch": 0.5333333333333333,
|
| 1056 |
+
"grad_norm": 0.4389738142490387,
|
| 1057 |
+
"learning_rate": 5.6747404844290664e-05,
|
| 1058 |
+
"loss": 1.6067,
|
| 1059 |
+
"step": 1500
|
| 1060 |
+
},
|
| 1061 |
+
{
|
| 1062 |
+
"epoch": 0.5368888888888889,
|
| 1063 |
+
"grad_norm": 0.44060131907463074,
|
| 1064 |
+
"learning_rate": 5.631487889273357e-05,
|
| 1065 |
+
"loss": 1.3391,
|
| 1066 |
+
"step": 1510
|
| 1067 |
+
},
|
| 1068 |
+
{
|
| 1069 |
+
"epoch": 0.5404444444444444,
|
| 1070 |
+
"grad_norm": 0.5496039390563965,
|
| 1071 |
+
"learning_rate": 5.588235294117647e-05,
|
| 1072 |
+
"loss": 1.5108,
|
| 1073 |
+
"step": 1520
|
| 1074 |
+
},
|
| 1075 |
+
{
|
| 1076 |
+
"epoch": 0.544,
|
| 1077 |
+
"grad_norm": 0.5372530221939087,
|
| 1078 |
+
"learning_rate": 5.544982698961938e-05,
|
| 1079 |
+
"loss": 1.4777,
|
| 1080 |
+
"step": 1530
|
| 1081 |
+
},
|
| 1082 |
+
{
|
| 1083 |
+
"epoch": 0.5475555555555556,
|
| 1084 |
+
"grad_norm": 0.5017176270484924,
|
| 1085 |
+
"learning_rate": 5.501730103806229e-05,
|
| 1086 |
+
"loss": 1.5007,
|
| 1087 |
+
"step": 1540
|
| 1088 |
+
},
|
| 1089 |
+
{
|
| 1090 |
+
"epoch": 0.5511111111111111,
|
| 1091 |
+
"grad_norm": 0.5397398471832275,
|
| 1092 |
+
"learning_rate": 5.458477508650519e-05,
|
| 1093 |
+
"loss": 1.6057,
|
| 1094 |
+
"step": 1550
|
| 1095 |
+
},
|
| 1096 |
+
{
|
| 1097 |
+
"epoch": 0.5546666666666666,
|
| 1098 |
+
"grad_norm": 0.5731176733970642,
|
| 1099 |
+
"learning_rate": 5.41522491349481e-05,
|
| 1100 |
+
"loss": 1.6644,
|
| 1101 |
+
"step": 1560
|
| 1102 |
+
},
|
| 1103 |
+
{
|
| 1104 |
+
"epoch": 0.5582222222222222,
|
| 1105 |
+
"grad_norm": 0.48925384879112244,
|
| 1106 |
+
"learning_rate": 5.3719723183391005e-05,
|
| 1107 |
+
"loss": 1.5506,
|
| 1108 |
+
"step": 1570
|
| 1109 |
+
},
|
| 1110 |
+
{
|
| 1111 |
+
"epoch": 0.5617777777777778,
|
| 1112 |
+
"grad_norm": 0.47176703810691833,
|
| 1113 |
+
"learning_rate": 5.3287197231833906e-05,
|
| 1114 |
+
"loss": 1.5355,
|
| 1115 |
+
"step": 1580
|
| 1116 |
+
},
|
| 1117 |
+
{
|
| 1118 |
+
"epoch": 0.5653333333333334,
|
| 1119 |
+
"grad_norm": 0.45709192752838135,
|
| 1120 |
+
"learning_rate": 5.285467128027683e-05,
|
| 1121 |
+
"loss": 1.4926,
|
| 1122 |
+
"step": 1590
|
| 1123 |
+
},
|
| 1124 |
+
{
|
| 1125 |
+
"epoch": 0.5688888888888889,
|
| 1126 |
+
"grad_norm": 0.5231002569198608,
|
| 1127 |
+
"learning_rate": 5.242214532871973e-05,
|
| 1128 |
+
"loss": 1.5645,
|
| 1129 |
+
"step": 1600
|
| 1130 |
+
},
|
| 1131 |
+
{
|
| 1132 |
+
"epoch": 0.5724444444444444,
|
| 1133 |
+
"grad_norm": 0.4811376929283142,
|
| 1134 |
+
"learning_rate": 5.1989619377162636e-05,
|
| 1135 |
+
"loss": 1.5169,
|
| 1136 |
+
"step": 1610
|
| 1137 |
+
},
|
| 1138 |
+
{
|
| 1139 |
+
"epoch": 0.576,
|
| 1140 |
+
"grad_norm": 0.5673519968986511,
|
| 1141 |
+
"learning_rate": 5.1557093425605544e-05,
|
| 1142 |
+
"loss": 1.4974,
|
| 1143 |
+
"step": 1620
|
| 1144 |
+
},
|
| 1145 |
+
{
|
| 1146 |
+
"epoch": 0.5795555555555556,
|
| 1147 |
+
"grad_norm": 0.5168330669403076,
|
| 1148 |
+
"learning_rate": 5.1124567474048445e-05,
|
| 1149 |
+
"loss": 1.4242,
|
| 1150 |
+
"step": 1630
|
| 1151 |
+
},
|
| 1152 |
+
{
|
| 1153 |
+
"epoch": 0.5831111111111111,
|
| 1154 |
+
"grad_norm": 0.49934643507003784,
|
| 1155 |
+
"learning_rate": 5.069204152249135e-05,
|
| 1156 |
+
"loss": 1.6342,
|
| 1157 |
+
"step": 1640
|
| 1158 |
+
},
|
| 1159 |
+
{
|
| 1160 |
+
"epoch": 0.5866666666666667,
|
| 1161 |
+
"grad_norm": 0.44559621810913086,
|
| 1162 |
+
"learning_rate": 5.0259515570934254e-05,
|
| 1163 |
+
"loss": 1.5204,
|
| 1164 |
+
"step": 1650
|
| 1165 |
+
},
|
| 1166 |
+
{
|
| 1167 |
+
"epoch": 0.5902222222222222,
|
| 1168 |
+
"grad_norm": 0.5243204832077026,
|
| 1169 |
+
"learning_rate": 4.982698961937716e-05,
|
| 1170 |
+
"loss": 1.4703,
|
| 1171 |
+
"step": 1660
|
| 1172 |
+
},
|
| 1173 |
+
{
|
| 1174 |
+
"epoch": 0.5937777777777777,
|
| 1175 |
+
"grad_norm": 0.5117699503898621,
|
| 1176 |
+
"learning_rate": 4.9394463667820076e-05,
|
| 1177 |
+
"loss": 1.7011,
|
| 1178 |
+
"step": 1670
|
| 1179 |
+
},
|
| 1180 |
+
{
|
| 1181 |
+
"epoch": 0.5973333333333334,
|
| 1182 |
+
"grad_norm": 0.42527756094932556,
|
| 1183 |
+
"learning_rate": 4.896193771626298e-05,
|
| 1184 |
+
"loss": 1.4331,
|
| 1185 |
+
"step": 1680
|
| 1186 |
+
},
|
| 1187 |
+
{
|
| 1188 |
+
"epoch": 0.6008888888888889,
|
| 1189 |
+
"grad_norm": 0.5092193484306335,
|
| 1190 |
+
"learning_rate": 4.8529411764705885e-05,
|
| 1191 |
+
"loss": 1.4378,
|
| 1192 |
+
"step": 1690
|
| 1193 |
+
},
|
| 1194 |
+
{
|
| 1195 |
+
"epoch": 0.6044444444444445,
|
| 1196 |
+
"grad_norm": 0.5100553631782532,
|
| 1197 |
+
"learning_rate": 4.809688581314879e-05,
|
| 1198 |
+
"loss": 1.5999,
|
| 1199 |
+
"step": 1700
|
| 1200 |
+
},
|
| 1201 |
+
{
|
| 1202 |
+
"epoch": 0.608,
|
| 1203 |
+
"grad_norm": 0.5817019939422607,
|
| 1204 |
+
"learning_rate": 4.7664359861591694e-05,
|
| 1205 |
+
"loss": 1.6094,
|
| 1206 |
+
"step": 1710
|
| 1207 |
+
},
|
| 1208 |
+
{
|
| 1209 |
+
"epoch": 0.6115555555555555,
|
| 1210 |
+
"grad_norm": 0.41723236441612244,
|
| 1211 |
+
"learning_rate": 4.723183391003461e-05,
|
| 1212 |
+
"loss": 1.3555,
|
| 1213 |
+
"step": 1720
|
| 1214 |
+
},
|
| 1215 |
+
{
|
| 1216 |
+
"epoch": 0.6151111111111112,
|
| 1217 |
+
"grad_norm": 0.5209750533103943,
|
| 1218 |
+
"learning_rate": 4.679930795847751e-05,
|
| 1219 |
+
"loss": 1.56,
|
| 1220 |
+
"step": 1730
|
| 1221 |
+
},
|
| 1222 |
+
{
|
| 1223 |
+
"epoch": 0.6186666666666667,
|
| 1224 |
+
"grad_norm": 0.48648136854171753,
|
| 1225 |
+
"learning_rate": 4.636678200692042e-05,
|
| 1226 |
+
"loss": 1.48,
|
| 1227 |
+
"step": 1740
|
| 1228 |
+
},
|
| 1229 |
+
{
|
| 1230 |
+
"epoch": 0.6222222222222222,
|
| 1231 |
+
"grad_norm": 0.5282444357872009,
|
| 1232 |
+
"learning_rate": 4.5934256055363325e-05,
|
| 1233 |
+
"loss": 1.5003,
|
| 1234 |
+
"step": 1750
|
| 1235 |
+
},
|
| 1236 |
+
{
|
| 1237 |
+
"epoch": 0.6257777777777778,
|
| 1238 |
+
"grad_norm": 0.4915744364261627,
|
| 1239 |
+
"learning_rate": 4.5501730103806226e-05,
|
| 1240 |
+
"loss": 1.4575,
|
| 1241 |
+
"step": 1760
|
| 1242 |
+
},
|
| 1243 |
+
{
|
| 1244 |
+
"epoch": 0.6293333333333333,
|
| 1245 |
+
"grad_norm": 0.568629801273346,
|
| 1246 |
+
"learning_rate": 4.506920415224914e-05,
|
| 1247 |
+
"loss": 1.4804,
|
| 1248 |
+
"step": 1770
|
| 1249 |
+
},
|
| 1250 |
+
{
|
| 1251 |
+
"epoch": 0.6328888888888888,
|
| 1252 |
+
"grad_norm": 0.48679119348526,
|
| 1253 |
+
"learning_rate": 4.463667820069204e-05,
|
| 1254 |
+
"loss": 1.5015,
|
| 1255 |
+
"step": 1780
|
| 1256 |
+
},
|
| 1257 |
+
{
|
| 1258 |
+
"epoch": 0.6364444444444445,
|
| 1259 |
+
"grad_norm": 0.5196161866188049,
|
| 1260 |
+
"learning_rate": 4.420415224913495e-05,
|
| 1261 |
+
"loss": 1.561,
|
| 1262 |
+
"step": 1790
|
| 1263 |
+
},
|
| 1264 |
+
{
|
| 1265 |
+
"epoch": 0.64,
|
| 1266 |
+
"grad_norm": 0.4946150779724121,
|
| 1267 |
+
"learning_rate": 4.377162629757786e-05,
|
| 1268 |
+
"loss": 1.5151,
|
| 1269 |
+
"step": 1800
|
| 1270 |
+
},
|
| 1271 |
+
{
|
| 1272 |
+
"epoch": 0.6435555555555555,
|
| 1273 |
+
"grad_norm": 0.40968433022499084,
|
| 1274 |
+
"learning_rate": 4.333910034602076e-05,
|
| 1275 |
+
"loss": 1.5373,
|
| 1276 |
+
"step": 1810
|
| 1277 |
+
},
|
| 1278 |
+
{
|
| 1279 |
+
"epoch": 0.6471111111111111,
|
| 1280 |
+
"grad_norm": 0.6429114937782288,
|
| 1281 |
+
"learning_rate": 4.290657439446367e-05,
|
| 1282 |
+
"loss": 1.4867,
|
| 1283 |
+
"step": 1820
|
| 1284 |
+
},
|
| 1285 |
+
{
|
| 1286 |
+
"epoch": 0.6506666666666666,
|
| 1287 |
+
"grad_norm": 0.474758118391037,
|
| 1288 |
+
"learning_rate": 4.247404844290658e-05,
|
| 1289 |
+
"loss": 1.5035,
|
| 1290 |
+
"step": 1830
|
| 1291 |
+
},
|
| 1292 |
+
{
|
| 1293 |
+
"epoch": 0.6542222222222223,
|
| 1294 |
+
"grad_norm": 0.51218581199646,
|
| 1295 |
+
"learning_rate": 4.204152249134948e-05,
|
| 1296 |
+
"loss": 1.5725,
|
| 1297 |
+
"step": 1840
|
| 1298 |
+
},
|
| 1299 |
+
{
|
| 1300 |
+
"epoch": 0.6577777777777778,
|
| 1301 |
+
"grad_norm": 0.49607232213020325,
|
| 1302 |
+
"learning_rate": 4.160899653979239e-05,
|
| 1303 |
+
"loss": 1.5352,
|
| 1304 |
+
"step": 1850
|
| 1305 |
+
},
|
| 1306 |
+
{
|
| 1307 |
+
"epoch": 0.6613333333333333,
|
| 1308 |
+
"grad_norm": 0.4859633147716522,
|
| 1309 |
+
"learning_rate": 4.11764705882353e-05,
|
| 1310 |
+
"loss": 1.4925,
|
| 1311 |
+
"step": 1860
|
| 1312 |
+
},
|
| 1313 |
+
{
|
| 1314 |
+
"epoch": 0.6648888888888889,
|
| 1315 |
+
"grad_norm": 0.4592611491680145,
|
| 1316 |
+
"learning_rate": 4.0743944636678205e-05,
|
| 1317 |
+
"loss": 1.5156,
|
| 1318 |
+
"step": 1870
|
| 1319 |
+
},
|
| 1320 |
+
{
|
| 1321 |
+
"epoch": 0.6684444444444444,
|
| 1322 |
+
"grad_norm": 0.47916117310523987,
|
| 1323 |
+
"learning_rate": 4.031141868512111e-05,
|
| 1324 |
+
"loss": 1.545,
|
| 1325 |
+
"step": 1880
|
| 1326 |
+
},
|
| 1327 |
+
{
|
| 1328 |
+
"epoch": 0.672,
|
| 1329 |
+
"grad_norm": 0.5305891633033752,
|
| 1330 |
+
"learning_rate": 3.9878892733564014e-05,
|
| 1331 |
+
"loss": 1.5384,
|
| 1332 |
+
"step": 1890
|
| 1333 |
+
},
|
| 1334 |
+
{
|
| 1335 |
+
"epoch": 0.6755555555555556,
|
| 1336 |
+
"grad_norm": 0.4583602547645569,
|
| 1337 |
+
"learning_rate": 3.944636678200692e-05,
|
| 1338 |
+
"loss": 1.5635,
|
| 1339 |
+
"step": 1900
|
| 1340 |
+
},
|
| 1341 |
+
{
|
| 1342 |
+
"epoch": 0.6791111111111111,
|
| 1343 |
+
"grad_norm": 0.5024811625480652,
|
| 1344 |
+
"learning_rate": 3.901384083044983e-05,
|
| 1345 |
+
"loss": 1.5016,
|
| 1346 |
+
"step": 1910
|
| 1347 |
+
},
|
| 1348 |
+
{
|
| 1349 |
+
"epoch": 0.6826666666666666,
|
| 1350 |
+
"grad_norm": 0.5373083353042603,
|
| 1351 |
+
"learning_rate": 3.858131487889274e-05,
|
| 1352 |
+
"loss": 1.5399,
|
| 1353 |
+
"step": 1920
|
| 1354 |
+
},
|
| 1355 |
+
{
|
| 1356 |
+
"epoch": 0.6862222222222222,
|
| 1357 |
+
"grad_norm": 0.415873646736145,
|
| 1358 |
+
"learning_rate": 3.8148788927335645e-05,
|
| 1359 |
+
"loss": 1.6384,
|
| 1360 |
+
"step": 1930
|
| 1361 |
+
},
|
| 1362 |
+
{
|
| 1363 |
+
"epoch": 0.6897777777777778,
|
| 1364 |
+
"grad_norm": 0.46044766902923584,
|
| 1365 |
+
"learning_rate": 3.7716262975778546e-05,
|
| 1366 |
+
"loss": 1.5588,
|
| 1367 |
+
"step": 1940
|
| 1368 |
+
},
|
| 1369 |
+
{
|
| 1370 |
+
"epoch": 0.6933333333333334,
|
| 1371 |
+
"grad_norm": 0.37672215700149536,
|
| 1372 |
+
"learning_rate": 3.7283737024221454e-05,
|
| 1373 |
+
"loss": 1.49,
|
| 1374 |
+
"step": 1950
|
| 1375 |
+
},
|
| 1376 |
+
{
|
| 1377 |
+
"epoch": 0.6968888888888889,
|
| 1378 |
+
"grad_norm": 0.40345415472984314,
|
| 1379 |
+
"learning_rate": 3.685121107266436e-05,
|
| 1380 |
+
"loss": 1.3789,
|
| 1381 |
+
"step": 1960
|
| 1382 |
+
},
|
| 1383 |
+
{
|
| 1384 |
+
"epoch": 0.7004444444444444,
|
| 1385 |
+
"grad_norm": 0.49566739797592163,
|
| 1386 |
+
"learning_rate": 3.641868512110726e-05,
|
| 1387 |
+
"loss": 1.3318,
|
| 1388 |
+
"step": 1970
|
| 1389 |
+
},
|
| 1390 |
+
{
|
| 1391 |
+
"epoch": 0.704,
|
| 1392 |
+
"grad_norm": 0.48587340116500854,
|
| 1393 |
+
"learning_rate": 3.598615916955018e-05,
|
| 1394 |
+
"loss": 1.4758,
|
| 1395 |
+
"step": 1980
|
| 1396 |
+
},
|
| 1397 |
+
{
|
| 1398 |
+
"epoch": 0.7075555555555556,
|
| 1399 |
+
"grad_norm": 0.596446692943573,
|
| 1400 |
+
"learning_rate": 3.5553633217993085e-05,
|
| 1401 |
+
"loss": 1.5347,
|
| 1402 |
+
"step": 1990
|
| 1403 |
+
},
|
| 1404 |
+
{
|
| 1405 |
+
"epoch": 0.7111111111111111,
|
| 1406 |
+
"grad_norm": 0.4951972961425781,
|
| 1407 |
+
"learning_rate": 3.5121107266435986e-05,
|
| 1408 |
+
"loss": 1.2704,
|
| 1409 |
+
"step": 2000
|
| 1410 |
+
},
|
| 1411 |
+
{
|
| 1412 |
+
"epoch": 0.7146666666666667,
|
| 1413 |
+
"grad_norm": 0.4018455743789673,
|
| 1414 |
+
"learning_rate": 3.4688581314878894e-05,
|
| 1415 |
+
"loss": 1.5217,
|
| 1416 |
+
"step": 2010
|
| 1417 |
+
},
|
| 1418 |
+
{
|
| 1419 |
+
"epoch": 0.7182222222222222,
|
| 1420 |
+
"grad_norm": 0.5961456894874573,
|
| 1421 |
+
"learning_rate": 3.4256055363321795e-05,
|
| 1422 |
+
"loss": 1.7622,
|
| 1423 |
+
"step": 2020
|
| 1424 |
+
},
|
| 1425 |
+
{
|
| 1426 |
+
"epoch": 0.7217777777777777,
|
| 1427 |
+
"grad_norm": 0.4609779119491577,
|
| 1428 |
+
"learning_rate": 3.382352941176471e-05,
|
| 1429 |
+
"loss": 1.5252,
|
| 1430 |
+
"step": 2030
|
| 1431 |
+
},
|
| 1432 |
+
{
|
| 1433 |
+
"epoch": 0.7253333333333334,
|
| 1434 |
+
"grad_norm": 0.43142765760421753,
|
| 1435 |
+
"learning_rate": 3.339100346020762e-05,
|
| 1436 |
+
"loss": 1.4461,
|
| 1437 |
+
"step": 2040
|
| 1438 |
+
},
|
| 1439 |
+
{
|
| 1440 |
+
"epoch": 0.7288888888888889,
|
| 1441 |
+
"grad_norm": 0.4517713785171509,
|
| 1442 |
+
"learning_rate": 3.295847750865052e-05,
|
| 1443 |
+
"loss": 1.5962,
|
| 1444 |
+
"step": 2050
|
| 1445 |
+
},
|
| 1446 |
+
{
|
| 1447 |
+
"epoch": 0.7324444444444445,
|
| 1448 |
+
"grad_norm": 0.48978036642074585,
|
| 1449 |
+
"learning_rate": 3.2525951557093426e-05,
|
| 1450 |
+
"loss": 1.5366,
|
| 1451 |
+
"step": 2060
|
| 1452 |
+
},
|
| 1453 |
+
{
|
| 1454 |
+
"epoch": 0.736,
|
| 1455 |
+
"grad_norm": 0.40700387954711914,
|
| 1456 |
+
"learning_rate": 3.2093425605536334e-05,
|
| 1457 |
+
"loss": 1.5572,
|
| 1458 |
+
"step": 2070
|
| 1459 |
+
},
|
| 1460 |
+
{
|
| 1461 |
+
"epoch": 0.7395555555555555,
|
| 1462 |
+
"grad_norm": 0.5392187237739563,
|
| 1463 |
+
"learning_rate": 3.166089965397924e-05,
|
| 1464 |
+
"loss": 1.5552,
|
| 1465 |
+
"step": 2080
|
| 1466 |
+
},
|
| 1467 |
+
{
|
| 1468 |
+
"epoch": 0.7431111111111111,
|
| 1469 |
+
"grad_norm": 0.44562795758247375,
|
| 1470 |
+
"learning_rate": 3.122837370242215e-05,
|
| 1471 |
+
"loss": 1.5194,
|
| 1472 |
+
"step": 2090
|
| 1473 |
+
},
|
| 1474 |
+
{
|
| 1475 |
+
"epoch": 0.7466666666666667,
|
| 1476 |
+
"grad_norm": 0.44013121724128723,
|
| 1477 |
+
"learning_rate": 3.079584775086505e-05,
|
| 1478 |
+
"loss": 1.3915,
|
| 1479 |
+
"step": 2100
|
| 1480 |
+
},
|
| 1481 |
+
{
|
| 1482 |
+
"epoch": 0.7502222222222222,
|
| 1483 |
+
"grad_norm": 0.4336005747318268,
|
| 1484 |
+
"learning_rate": 3.036332179930796e-05,
|
| 1485 |
+
"loss": 1.4715,
|
| 1486 |
+
"step": 2110
|
| 1487 |
+
},
|
| 1488 |
+
{
|
| 1489 |
+
"epoch": 0.7537777777777778,
|
| 1490 |
+
"grad_norm": 0.5743895769119263,
|
| 1491 |
+
"learning_rate": 2.9930795847750863e-05,
|
| 1492 |
+
"loss": 1.5256,
|
| 1493 |
+
"step": 2120
|
| 1494 |
+
},
|
| 1495 |
+
{
|
| 1496 |
+
"epoch": 0.7573333333333333,
|
| 1497 |
+
"grad_norm": 0.45934122800827026,
|
| 1498 |
+
"learning_rate": 2.9498269896193774e-05,
|
| 1499 |
+
"loss": 1.5873,
|
| 1500 |
+
"step": 2130
|
| 1501 |
+
},
|
| 1502 |
+
{
|
| 1503 |
+
"epoch": 0.7608888888888888,
|
| 1504 |
+
"grad_norm": 0.5065307021141052,
|
| 1505 |
+
"learning_rate": 2.9065743944636682e-05,
|
| 1506 |
+
"loss": 1.5582,
|
| 1507 |
+
"step": 2140
|
| 1508 |
+
},
|
| 1509 |
+
{
|
| 1510 |
+
"epoch": 0.7644444444444445,
|
| 1511 |
+
"grad_norm": 0.511725664138794,
|
| 1512 |
+
"learning_rate": 2.8633217993079586e-05,
|
| 1513 |
+
"loss": 1.4813,
|
| 1514 |
+
"step": 2150
|
| 1515 |
+
},
|
| 1516 |
+
{
|
| 1517 |
+
"epoch": 0.768,
|
| 1518 |
+
"grad_norm": 0.5018854737281799,
|
| 1519 |
+
"learning_rate": 2.820069204152249e-05,
|
| 1520 |
+
"loss": 1.5786,
|
| 1521 |
+
"step": 2160
|
| 1522 |
+
},
|
| 1523 |
+
{
|
| 1524 |
+
"epoch": 0.7715555555555556,
|
| 1525 |
+
"grad_norm": 0.5292341113090515,
|
| 1526 |
+
"learning_rate": 2.77681660899654e-05,
|
| 1527 |
+
"loss": 1.6471,
|
| 1528 |
+
"step": 2170
|
| 1529 |
+
},
|
| 1530 |
+
{
|
| 1531 |
+
"epoch": 0.7751111111111111,
|
| 1532 |
+
"grad_norm": 0.3638977110385895,
|
| 1533 |
+
"learning_rate": 2.733564013840831e-05,
|
| 1534 |
+
"loss": 1.4863,
|
| 1535 |
+
"step": 2180
|
| 1536 |
+
},
|
| 1537 |
+
{
|
| 1538 |
+
"epoch": 0.7786666666666666,
|
| 1539 |
+
"grad_norm": 0.49934521317481995,
|
| 1540 |
+
"learning_rate": 2.6903114186851214e-05,
|
| 1541 |
+
"loss": 1.4587,
|
| 1542 |
+
"step": 2190
|
| 1543 |
+
},
|
| 1544 |
+
{
|
| 1545 |
+
"epoch": 0.7822222222222223,
|
| 1546 |
+
"grad_norm": 0.5519063472747803,
|
| 1547 |
+
"learning_rate": 2.647058823529412e-05,
|
| 1548 |
+
"loss": 1.4915,
|
| 1549 |
+
"step": 2200
|
| 1550 |
+
},
|
| 1551 |
+
{
|
| 1552 |
+
"epoch": 0.7857777777777778,
|
| 1553 |
+
"grad_norm": 0.5289978384971619,
|
| 1554 |
+
"learning_rate": 2.6038062283737023e-05,
|
| 1555 |
+
"loss": 1.5924,
|
| 1556 |
+
"step": 2210
|
| 1557 |
+
},
|
| 1558 |
+
{
|
| 1559 |
+
"epoch": 0.7893333333333333,
|
| 1560 |
+
"grad_norm": 0.4768216907978058,
|
| 1561 |
+
"learning_rate": 2.560553633217993e-05,
|
| 1562 |
+
"loss": 1.4847,
|
| 1563 |
+
"step": 2220
|
| 1564 |
+
},
|
| 1565 |
+
{
|
| 1566 |
+
"epoch": 0.7928888888888889,
|
| 1567 |
+
"grad_norm": 0.5400416254997253,
|
| 1568 |
+
"learning_rate": 2.5173010380622842e-05,
|
| 1569 |
+
"loss": 1.5664,
|
| 1570 |
+
"step": 2230
|
| 1571 |
+
},
|
| 1572 |
+
{
|
| 1573 |
+
"epoch": 0.7964444444444444,
|
| 1574 |
+
"grad_norm": 0.5170167684555054,
|
| 1575 |
+
"learning_rate": 2.4740484429065743e-05,
|
| 1576 |
+
"loss": 1.7191,
|
| 1577 |
+
"step": 2240
|
| 1578 |
+
},
|
| 1579 |
+
{
|
| 1580 |
+
"epoch": 0.8,
|
| 1581 |
+
"grad_norm": 0.4661620259284973,
|
| 1582 |
+
"learning_rate": 2.430795847750865e-05,
|
| 1583 |
+
"loss": 1.584,
|
| 1584 |
+
"step": 2250
|
| 1585 |
+
},
|
| 1586 |
+
{
|
| 1587 |
+
"epoch": 0.8035555555555556,
|
| 1588 |
+
"grad_norm": 0.48751363158226013,
|
| 1589 |
+
"learning_rate": 2.387543252595156e-05,
|
| 1590 |
+
"loss": 1.6166,
|
| 1591 |
+
"step": 2260
|
| 1592 |
+
},
|
| 1593 |
+
{
|
| 1594 |
+
"epoch": 0.8071111111111111,
|
| 1595 |
+
"grad_norm": 0.48189136385917664,
|
| 1596 |
+
"learning_rate": 2.3442906574394467e-05,
|
| 1597 |
+
"loss": 1.412,
|
| 1598 |
+
"step": 2270
|
| 1599 |
+
},
|
| 1600 |
+
{
|
| 1601 |
+
"epoch": 0.8106666666666666,
|
| 1602 |
+
"grad_norm": 0.4377134144306183,
|
| 1603 |
+
"learning_rate": 2.301038062283737e-05,
|
| 1604 |
+
"loss": 1.4633,
|
| 1605 |
+
"step": 2280
|
| 1606 |
+
},
|
| 1607 |
+
{
|
| 1608 |
+
"epoch": 0.8142222222222222,
|
| 1609 |
+
"grad_norm": 0.4900270998477936,
|
| 1610 |
+
"learning_rate": 2.2577854671280275e-05,
|
| 1611 |
+
"loss": 1.5968,
|
| 1612 |
+
"step": 2290
|
| 1613 |
+
},
|
| 1614 |
+
{
|
| 1615 |
+
"epoch": 0.8177777777777778,
|
| 1616 |
+
"grad_norm": 0.5673394203186035,
|
| 1617 |
+
"learning_rate": 2.2145328719723187e-05,
|
| 1618 |
+
"loss": 1.4443,
|
| 1619 |
+
"step": 2300
|
| 1620 |
+
},
|
| 1621 |
+
{
|
| 1622 |
+
"epoch": 0.8213333333333334,
|
| 1623 |
+
"grad_norm": 0.5573922395706177,
|
| 1624 |
+
"learning_rate": 2.171280276816609e-05,
|
| 1625 |
+
"loss": 1.5951,
|
| 1626 |
+
"step": 2310
|
| 1627 |
+
},
|
| 1628 |
+
{
|
| 1629 |
+
"epoch": 0.8248888888888889,
|
| 1630 |
+
"grad_norm": 0.5815632343292236,
|
| 1631 |
+
"learning_rate": 2.1280276816609e-05,
|
| 1632 |
+
"loss": 1.54,
|
| 1633 |
+
"step": 2320
|
| 1634 |
+
},
|
| 1635 |
+
{
|
| 1636 |
+
"epoch": 0.8284444444444444,
|
| 1637 |
+
"grad_norm": 0.5276745557785034,
|
| 1638 |
+
"learning_rate": 2.0847750865051903e-05,
|
| 1639 |
+
"loss": 1.5903,
|
| 1640 |
+
"step": 2330
|
| 1641 |
+
},
|
| 1642 |
+
{
|
| 1643 |
+
"epoch": 0.832,
|
| 1644 |
+
"grad_norm": 0.4580680727958679,
|
| 1645 |
+
"learning_rate": 2.041522491349481e-05,
|
| 1646 |
+
"loss": 1.6143,
|
| 1647 |
+
"step": 2340
|
| 1648 |
+
},
|
| 1649 |
+
{
|
| 1650 |
+
"epoch": 0.8355555555555556,
|
| 1651 |
+
"grad_norm": 0.5015590190887451,
|
| 1652 |
+
"learning_rate": 1.998269896193772e-05,
|
| 1653 |
+
"loss": 1.4156,
|
| 1654 |
+
"step": 2350
|
| 1655 |
+
},
|
| 1656 |
+
{
|
| 1657 |
+
"epoch": 0.8391111111111111,
|
| 1658 |
+
"grad_norm": 0.508334219455719,
|
| 1659 |
+
"learning_rate": 1.9550173010380623e-05,
|
| 1660 |
+
"loss": 1.5272,
|
| 1661 |
+
"step": 2360
|
| 1662 |
+
},
|
| 1663 |
+
{
|
| 1664 |
+
"epoch": 0.8426666666666667,
|
| 1665 |
+
"grad_norm": 0.44540607929229736,
|
| 1666 |
+
"learning_rate": 1.9117647058823528e-05,
|
| 1667 |
+
"loss": 1.5522,
|
| 1668 |
+
"step": 2370
|
| 1669 |
+
},
|
| 1670 |
+
{
|
| 1671 |
+
"epoch": 0.8462222222222222,
|
| 1672 |
+
"grad_norm": 0.5358651876449585,
|
| 1673 |
+
"learning_rate": 1.868512110726644e-05,
|
| 1674 |
+
"loss": 1.4527,
|
| 1675 |
+
"step": 2380
|
| 1676 |
+
},
|
| 1677 |
+
{
|
| 1678 |
+
"epoch": 0.8497777777777777,
|
| 1679 |
+
"grad_norm": 0.5191397666931152,
|
| 1680 |
+
"learning_rate": 1.8252595155709343e-05,
|
| 1681 |
+
"loss": 1.5627,
|
| 1682 |
+
"step": 2390
|
| 1683 |
+
},
|
| 1684 |
+
{
|
| 1685 |
+
"epoch": 0.8533333333333334,
|
| 1686 |
+
"grad_norm": 0.4698735773563385,
|
| 1687 |
+
"learning_rate": 1.782006920415225e-05,
|
| 1688 |
+
"loss": 1.4434,
|
| 1689 |
+
"step": 2400
|
| 1690 |
+
},
|
| 1691 |
+
{
|
| 1692 |
+
"epoch": 0.8568888888888889,
|
| 1693 |
+
"grad_norm": 0.4450030028820038,
|
| 1694 |
+
"learning_rate": 1.7387543252595156e-05,
|
| 1695 |
+
"loss": 1.383,
|
| 1696 |
+
"step": 2410
|
| 1697 |
+
},
|
| 1698 |
+
{
|
| 1699 |
+
"epoch": 0.8604444444444445,
|
| 1700 |
+
"grad_norm": 0.47164052724838257,
|
| 1701 |
+
"learning_rate": 1.6955017301038063e-05,
|
| 1702 |
+
"loss": 1.5499,
|
| 1703 |
+
"step": 2420
|
| 1704 |
+
},
|
| 1705 |
+
{
|
| 1706 |
+
"epoch": 0.864,
|
| 1707 |
+
"grad_norm": 0.4113026559352875,
|
| 1708 |
+
"learning_rate": 1.652249134948097e-05,
|
| 1709 |
+
"loss": 1.499,
|
| 1710 |
+
"step": 2430
|
| 1711 |
+
},
|
| 1712 |
+
{
|
| 1713 |
+
"epoch": 0.8675555555555555,
|
| 1714 |
+
"grad_norm": 0.48519784212112427,
|
| 1715 |
+
"learning_rate": 1.6089965397923876e-05,
|
| 1716 |
+
"loss": 1.5236,
|
| 1717 |
+
"step": 2440
|
| 1718 |
+
},
|
| 1719 |
+
{
|
| 1720 |
+
"epoch": 0.8711111111111111,
|
| 1721 |
+
"grad_norm": 0.5262069702148438,
|
| 1722 |
+
"learning_rate": 1.5657439446366783e-05,
|
| 1723 |
+
"loss": 1.5144,
|
| 1724 |
+
"step": 2450
|
| 1725 |
+
},
|
| 1726 |
+
{
|
| 1727 |
+
"epoch": 0.8746666666666667,
|
| 1728 |
+
"grad_norm": 0.42513802647590637,
|
| 1729 |
+
"learning_rate": 1.522491349480969e-05,
|
| 1730 |
+
"loss": 1.4932,
|
| 1731 |
+
"step": 2460
|
| 1732 |
+
},
|
| 1733 |
+
{
|
| 1734 |
+
"epoch": 0.8782222222222222,
|
| 1735 |
+
"grad_norm": 0.47486934065818787,
|
| 1736 |
+
"learning_rate": 1.4792387543252596e-05,
|
| 1737 |
+
"loss": 1.4398,
|
| 1738 |
+
"step": 2470
|
| 1739 |
+
},
|
| 1740 |
+
{
|
| 1741 |
+
"epoch": 0.8817777777777778,
|
| 1742 |
+
"grad_norm": 0.5634934902191162,
|
| 1743 |
+
"learning_rate": 1.4359861591695503e-05,
|
| 1744 |
+
"loss": 1.65,
|
| 1745 |
+
"step": 2480
|
| 1746 |
+
},
|
| 1747 |
+
{
|
| 1748 |
+
"epoch": 0.8853333333333333,
|
| 1749 |
+
"grad_norm": 0.43849506974220276,
|
| 1750 |
+
"learning_rate": 1.3927335640138408e-05,
|
| 1751 |
+
"loss": 1.5201,
|
| 1752 |
+
"step": 2490
|
| 1753 |
+
},
|
| 1754 |
+
{
|
| 1755 |
+
"epoch": 0.8888888888888888,
|
| 1756 |
+
"grad_norm": 0.4285069704055786,
|
| 1757 |
+
"learning_rate": 1.3494809688581317e-05,
|
| 1758 |
+
"loss": 1.492,
|
| 1759 |
+
"step": 2500
|
| 1760 |
+
},
|
| 1761 |
+
{
|
| 1762 |
+
"epoch": 0.8924444444444445,
|
| 1763 |
+
"grad_norm": 0.5707066655158997,
|
| 1764 |
+
"learning_rate": 1.3062283737024222e-05,
|
| 1765 |
+
"loss": 1.5991,
|
| 1766 |
+
"step": 2510
|
| 1767 |
+
},
|
| 1768 |
+
{
|
| 1769 |
+
"epoch": 0.896,
|
| 1770 |
+
"grad_norm": 0.4697231948375702,
|
| 1771 |
+
"learning_rate": 1.2629757785467128e-05,
|
| 1772 |
+
"loss": 1.5585,
|
| 1773 |
+
"step": 2520
|
| 1774 |
+
},
|
| 1775 |
+
{
|
| 1776 |
+
"epoch": 0.8995555555555556,
|
| 1777 |
+
"grad_norm": 0.4958859384059906,
|
| 1778 |
+
"learning_rate": 1.2197231833910034e-05,
|
| 1779 |
+
"loss": 1.5496,
|
| 1780 |
+
"step": 2530
|
| 1781 |
+
},
|
| 1782 |
+
{
|
| 1783 |
+
"epoch": 0.9031111111111111,
|
| 1784 |
+
"grad_norm": 0.4845017194747925,
|
| 1785 |
+
"learning_rate": 1.1764705882352942e-05,
|
| 1786 |
+
"loss": 1.592,
|
| 1787 |
+
"step": 2540
|
| 1788 |
+
},
|
| 1789 |
+
{
|
| 1790 |
+
"epoch": 0.9066666666666666,
|
| 1791 |
+
"grad_norm": 0.4356600344181061,
|
| 1792 |
+
"learning_rate": 1.1332179930795848e-05,
|
| 1793 |
+
"loss": 1.5873,
|
| 1794 |
+
"step": 2550
|
| 1795 |
+
},
|
| 1796 |
+
{
|
| 1797 |
+
"epoch": 0.9102222222222223,
|
| 1798 |
+
"grad_norm": 0.3917759954929352,
|
| 1799 |
+
"learning_rate": 1.0899653979238756e-05,
|
| 1800 |
+
"loss": 1.4984,
|
| 1801 |
+
"step": 2560
|
| 1802 |
+
},
|
| 1803 |
+
{
|
| 1804 |
+
"epoch": 0.9137777777777778,
|
| 1805 |
+
"grad_norm": 0.48273682594299316,
|
| 1806 |
+
"learning_rate": 1.046712802768166e-05,
|
| 1807 |
+
"loss": 1.4288,
|
| 1808 |
+
"step": 2570
|
| 1809 |
+
},
|
| 1810 |
+
{
|
| 1811 |
+
"epoch": 0.9173333333333333,
|
| 1812 |
+
"grad_norm": 0.5662245750427246,
|
| 1813 |
+
"learning_rate": 1.0034602076124568e-05,
|
| 1814 |
+
"loss": 1.5123,
|
| 1815 |
+
"step": 2580
|
| 1816 |
+
},
|
| 1817 |
+
{
|
| 1818 |
+
"epoch": 0.9208888888888889,
|
| 1819 |
+
"grad_norm": 0.510188639163971,
|
| 1820 |
+
"learning_rate": 9.602076124567474e-06,
|
| 1821 |
+
"loss": 1.4915,
|
| 1822 |
+
"step": 2590
|
| 1823 |
+
},
|
| 1824 |
+
{
|
| 1825 |
+
"epoch": 0.9244444444444444,
|
| 1826 |
+
"grad_norm": 0.5999191403388977,
|
| 1827 |
+
"learning_rate": 9.169550173010382e-06,
|
| 1828 |
+
"loss": 1.7374,
|
| 1829 |
+
"step": 2600
|
| 1830 |
+
},
|
| 1831 |
+
{
|
| 1832 |
+
"epoch": 0.928,
|
| 1833 |
+
"grad_norm": 0.5733387470245361,
|
| 1834 |
+
"learning_rate": 8.737024221453288e-06,
|
| 1835 |
+
"loss": 1.5551,
|
| 1836 |
+
"step": 2610
|
| 1837 |
+
},
|
| 1838 |
+
{
|
| 1839 |
+
"epoch": 0.9315555555555556,
|
| 1840 |
+
"grad_norm": 0.3804311156272888,
|
| 1841 |
+
"learning_rate": 8.304498269896194e-06,
|
| 1842 |
+
"loss": 1.5631,
|
| 1843 |
+
"step": 2620
|
| 1844 |
+
},
|
| 1845 |
+
{
|
| 1846 |
+
"epoch": 0.9351111111111111,
|
| 1847 |
+
"grad_norm": 0.44998374581336975,
|
| 1848 |
+
"learning_rate": 7.8719723183391e-06,
|
| 1849 |
+
"loss": 1.7566,
|
| 1850 |
+
"step": 2630
|
| 1851 |
+
},
|
| 1852 |
+
{
|
| 1853 |
+
"epoch": 0.9386666666666666,
|
| 1854 |
+
"grad_norm": 0.4219221770763397,
|
| 1855 |
+
"learning_rate": 7.439446366782007e-06,
|
| 1856 |
+
"loss": 1.454,
|
| 1857 |
+
"step": 2640
|
| 1858 |
+
},
|
| 1859 |
+
{
|
| 1860 |
+
"epoch": 0.9422222222222222,
|
| 1861 |
+
"grad_norm": 0.48252835869789124,
|
| 1862 |
+
"learning_rate": 7.006920415224914e-06,
|
| 1863 |
+
"loss": 1.5107,
|
| 1864 |
+
"step": 2650
|
| 1865 |
+
},
|
| 1866 |
+
{
|
| 1867 |
+
"epoch": 0.9457777777777778,
|
| 1868 |
+
"grad_norm": 0.5446690320968628,
|
| 1869 |
+
"learning_rate": 6.5743944636678194e-06,
|
| 1870 |
+
"loss": 1.5333,
|
| 1871 |
+
"step": 2660
|
| 1872 |
+
},
|
| 1873 |
+
{
|
| 1874 |
+
"epoch": 0.9493333333333334,
|
| 1875 |
+
"grad_norm": 0.4676448106765747,
|
| 1876 |
+
"learning_rate": 6.141868512110726e-06,
|
| 1877 |
+
"loss": 1.4822,
|
| 1878 |
+
"step": 2670
|
| 1879 |
+
},
|
| 1880 |
+
{
|
| 1881 |
+
"epoch": 0.9528888888888889,
|
| 1882 |
+
"grad_norm": 0.41428953409194946,
|
| 1883 |
+
"learning_rate": 5.709342560553633e-06,
|
| 1884 |
+
"loss": 1.398,
|
| 1885 |
+
"step": 2680
|
| 1886 |
+
},
|
| 1887 |
+
{
|
| 1888 |
+
"epoch": 0.9564444444444444,
|
| 1889 |
+
"grad_norm": 0.39139533042907715,
|
| 1890 |
+
"learning_rate": 5.2768166089965395e-06,
|
| 1891 |
+
"loss": 1.539,
|
| 1892 |
+
"step": 2690
|
| 1893 |
+
},
|
| 1894 |
+
{
|
| 1895 |
+
"epoch": 0.96,
|
| 1896 |
+
"grad_norm": 0.5213208794593811,
|
| 1897 |
+
"learning_rate": 4.8442906574394464e-06,
|
| 1898 |
+
"loss": 1.535,
|
| 1899 |
+
"step": 2700
|
| 1900 |
+
},
|
| 1901 |
+
{
|
| 1902 |
+
"epoch": 0.9635555555555556,
|
| 1903 |
+
"grad_norm": 0.6053948402404785,
|
| 1904 |
+
"learning_rate": 4.411764705882353e-06,
|
| 1905 |
+
"loss": 1.484,
|
| 1906 |
+
"step": 2710
|
| 1907 |
+
},
|
| 1908 |
+
{
|
| 1909 |
+
"epoch": 0.9671111111111111,
|
| 1910 |
+
"grad_norm": 0.44833922386169434,
|
| 1911 |
+
"learning_rate": 3.9792387543252595e-06,
|
| 1912 |
+
"loss": 1.3107,
|
| 1913 |
+
"step": 2720
|
| 1914 |
+
},
|
| 1915 |
+
{
|
| 1916 |
+
"epoch": 0.9706666666666667,
|
| 1917 |
+
"grad_norm": 0.4769100844860077,
|
| 1918 |
+
"learning_rate": 3.5467128027681665e-06,
|
| 1919 |
+
"loss": 1.5114,
|
| 1920 |
+
"step": 2730
|
| 1921 |
+
},
|
| 1922 |
+
{
|
| 1923 |
+
"epoch": 0.9742222222222222,
|
| 1924 |
+
"grad_norm": 0.5112429857254028,
|
| 1925 |
+
"learning_rate": 3.1141868512110726e-06,
|
| 1926 |
+
"loss": 1.5357,
|
| 1927 |
+
"step": 2740
|
| 1928 |
+
},
|
| 1929 |
+
{
|
| 1930 |
+
"epoch": 0.9777777777777777,
|
| 1931 |
+
"grad_norm": 0.4884449243545532,
|
| 1932 |
+
"learning_rate": 2.6816608996539796e-06,
|
| 1933 |
+
"loss": 1.4739,
|
| 1934 |
+
"step": 2750
|
| 1935 |
+
},
|
| 1936 |
+
{
|
| 1937 |
+
"epoch": 0.9813333333333333,
|
| 1938 |
+
"grad_norm": 0.5470872521400452,
|
| 1939 |
+
"learning_rate": 2.249134948096886e-06,
|
| 1940 |
+
"loss": 1.5557,
|
| 1941 |
+
"step": 2760
|
| 1942 |
+
},
|
| 1943 |
+
{
|
| 1944 |
+
"epoch": 0.9848888888888889,
|
| 1945 |
+
"grad_norm": 0.517669141292572,
|
| 1946 |
+
"learning_rate": 1.8166089965397926e-06,
|
| 1947 |
+
"loss": 1.5388,
|
| 1948 |
+
"step": 2770
|
| 1949 |
+
},
|
| 1950 |
+
{
|
| 1951 |
+
"epoch": 0.9884444444444445,
|
| 1952 |
+
"grad_norm": 0.46402767300605774,
|
| 1953 |
+
"learning_rate": 1.3840830449826992e-06,
|
| 1954 |
+
"loss": 1.5509,
|
| 1955 |
+
"step": 2780
|
| 1956 |
+
},
|
| 1957 |
+
{
|
| 1958 |
+
"epoch": 0.992,
|
| 1959 |
+
"grad_norm": 0.5298945307731628,
|
| 1960 |
+
"learning_rate": 9.515570934256056e-07,
|
| 1961 |
+
"loss": 1.533,
|
| 1962 |
+
"step": 2790
|
| 1963 |
+
},
|
| 1964 |
+
{
|
| 1965 |
+
"epoch": 0.9955555555555555,
|
| 1966 |
+
"grad_norm": 0.5506584644317627,
|
| 1967 |
+
"learning_rate": 5.190311418685121e-07,
|
| 1968 |
+
"loss": 1.404,
|
| 1969 |
+
"step": 2800
|
| 1970 |
+
},
|
| 1971 |
+
{
|
| 1972 |
+
"epoch": 0.9991111111111111,
|
| 1973 |
+
"grad_norm": 0.5148972868919373,
|
| 1974 |
+
"learning_rate": 8.65051903114187e-08,
|
| 1975 |
+
"loss": 1.558,
|
| 1976 |
+
"step": 2810
|
| 1977 |
+
}
|
| 1978 |
+
],
|
| 1979 |
+
"logging_steps": 10,
|
| 1980 |
+
"max_steps": 2812,
|
| 1981 |
+
"num_input_tokens_seen": 0,
|
| 1982 |
+
"num_train_epochs": 1,
|
| 1983 |
+
"save_steps": 5000,
|
| 1984 |
+
"stateful_callbacks": {
|
| 1985 |
+
"TrainerControl": {
|
| 1986 |
+
"args": {
|
| 1987 |
+
"should_epoch_stop": false,
|
| 1988 |
+
"should_evaluate": false,
|
| 1989 |
+
"should_log": false,
|
| 1990 |
+
"should_save": true,
|
| 1991 |
+
"should_training_stop": true
|
| 1992 |
+
},
|
| 1993 |
+
"attributes": {}
|
| 1994 |
+
}
|
| 1995 |
+
},
|
| 1996 |
+
"total_flos": 1.1067056282035814e+17,
|
| 1997 |
+
"train_batch_size": 4,
|
| 1998 |
+
"trial_name": null,
|
| 1999 |
+
"trial_params": null
|
| 2000 |
+
}
|
checkpoint-2812/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:1c4745953a4d16c5fbbe32e5f471e8c6aa13389decd1c644a551850315969586
|
| 3 |
+
size 5176
|
config.json
ADDED
|
@@ -0,0 +1,32 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "appvoid/arco-mini-run-4",
|
| 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 sha256:2b74e6e089df1f5cabadddc8329432ca60d36596872755480e71adca08612f3b
|
| 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 sha256:1c4745953a4d16c5fbbe32e5f471e8c6aa13389decd1c644a551850315969586
|
| 3 |
+
size 5176
|