Training in progress, step 369, checkpoint
Browse files- last-checkpoint/chat_template.jinja +1 -0
- last-checkpoint/config.json +62 -0
- last-checkpoint/generation_config.json +14 -0
- last-checkpoint/model.safetensors +3 -0
- last-checkpoint/optimizer.pt +3 -0
- last-checkpoint/rng_state.pth +3 -0
- last-checkpoint/scheduler.pt +3 -0
- last-checkpoint/tokenizer.json +0 -0
- last-checkpoint/tokenizer_config.json +17 -0
- last-checkpoint/trainer_state.json +2625 -0
- last-checkpoint/training_args.bin +3 -0
last-checkpoint/chat_template.jinja
ADDED
|
@@ -0,0 +1 @@
|
|
|
|
|
|
|
| 1 |
+
{% if messages[0]['role'] == 'system' %}{% set loop_messages = messages[1:] %}{% set system_message = messages[0]['content'] %}<|hy_begin▁of▁sentence|>{{ system_message }}<|hy_place▁holder▁no▁3|>{% else %}{% set loop_messages = messages %}<|hy_begin▁of▁sentence|>{% endif %}{% for message in loop_messages %}{% if message['role'] == 'user' %}<|hy_User|>{{ message['content'] }}{% elif message['role'] == 'assistant' %}<|hy_Assistant|>{{ message['content'] }}<|hy_place▁holder▁no▁2|>{% endif %}{% endfor %}{% if add_generation_prompt %}<|hy_Assistant|>{% else %}<|hy_place▁holder▁no▁8|>{% endif %}
|
last-checkpoint/config.json
ADDED
|
@@ -0,0 +1,62 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"add_classification_head": false,
|
| 3 |
+
"architectures": [
|
| 4 |
+
"HunYuanDenseV1ForCausalLM"
|
| 5 |
+
],
|
| 6 |
+
"attention_bias": false,
|
| 7 |
+
"attention_dropout": 0.0,
|
| 8 |
+
"attention_head_dim": 128,
|
| 9 |
+
"bos_token_id": 120000,
|
| 10 |
+
"cla_share_factor": 2,
|
| 11 |
+
"class_num": 0,
|
| 12 |
+
"dense_list": [
|
| 13 |
+
2048,
|
| 14 |
+
0
|
| 15 |
+
],
|
| 16 |
+
"dtype": "bfloat16",
|
| 17 |
+
"eos_token_id": 120020,
|
| 18 |
+
"head_dim": 128,
|
| 19 |
+
"hidden_act": "silu",
|
| 20 |
+
"hidden_size": 2048,
|
| 21 |
+
"im_end_id": 5,
|
| 22 |
+
"im_newline_id": 11,
|
| 23 |
+
"im_start_id": 4,
|
| 24 |
+
"initializer_range": 0.02,
|
| 25 |
+
"intermediate_size": 6144,
|
| 26 |
+
"mask_init_id": 12,
|
| 27 |
+
"max_position_embeddings": 262144,
|
| 28 |
+
"mlp_bias": false,
|
| 29 |
+
"model_type": "hunyuan_v1_dense",
|
| 30 |
+
"norm_type": "rms",
|
| 31 |
+
"num_attention_heads": 16,
|
| 32 |
+
"num_hidden_layers": 32,
|
| 33 |
+
"num_key_value_heads": 4,
|
| 34 |
+
"org_vocab_size": 120818,
|
| 35 |
+
"pad_id": 120002,
|
| 36 |
+
"pad_token_id": 120002,
|
| 37 |
+
"pool_type": "last",
|
| 38 |
+
"pretraining_tp": 1,
|
| 39 |
+
"rms_norm_eps": 1e-05,
|
| 40 |
+
"rope_parameters": {
|
| 41 |
+
"alpha": 1000.0,
|
| 42 |
+
"beta_fast": 32,
|
| 43 |
+
"beta_slow": 1,
|
| 44 |
+
"factor": 1.0,
|
| 45 |
+
"mscale": 1.0,
|
| 46 |
+
"mscale_all_dim": 1.0,
|
| 47 |
+
"rope_theta": 10000.0,
|
| 48 |
+
"rope_type": "dynamic",
|
| 49 |
+
"type": "dynamic"
|
| 50 |
+
},
|
| 51 |
+
"sep_token_id": 120007,
|
| 52 |
+
"text_end_id": 7,
|
| 53 |
+
"text_start_id": 6,
|
| 54 |
+
"tie_word_embeddings": true,
|
| 55 |
+
"transformers_version": "5.0.0rc3",
|
| 56 |
+
"unsloth_version": "2026.1.3",
|
| 57 |
+
"use_cache": false,
|
| 58 |
+
"use_cla": false,
|
| 59 |
+
"use_qk_norm": true,
|
| 60 |
+
"use_rotary_pos_emb": true,
|
| 61 |
+
"vocab_size": 120821
|
| 62 |
+
}
|
last-checkpoint/generation_config.json
ADDED
|
@@ -0,0 +1,14 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"bos_token_id": 120000,
|
| 3 |
+
"do_sample": true,
|
| 4 |
+
"eos_token_id": [
|
| 5 |
+
120020
|
| 6 |
+
],
|
| 7 |
+
"max_length": 262144,
|
| 8 |
+
"pad_token_id": 120002,
|
| 9 |
+
"repetition_penalty": 1.05,
|
| 10 |
+
"temperature": 0.7,
|
| 11 |
+
"top_k": 20,
|
| 12 |
+
"top_p": 0.8,
|
| 13 |
+
"transformers_version": "5.0.0rc3"
|
| 14 |
+
}
|
last-checkpoint/model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:5bf9e49f60acf6e18aab9ff3178ffc2854814793a6f9204251649ba254acca40
|
| 3 |
+
size 3582214344
|
last-checkpoint/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:7c0c480d009aa6fa9f6916fb3da7ffa3c254dd12dcaff015fad1efa5852cb2e2
|
| 3 |
+
size 5116251049
|
last-checkpoint/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:7c800b778fa7e115e4c34de8529902de8b61c9a1b4bab3eb8295d06dafff030e
|
| 3 |
+
size 14645
|
last-checkpoint/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:1e26e9662bc241074c48d43e930c7804d454d5841a5f05b5e507a576850ed1a8
|
| 3 |
+
size 1465
|
last-checkpoint/tokenizer.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
last-checkpoint/tokenizer_config.json
ADDED
|
@@ -0,0 +1,17 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"backend": "tokenizers",
|
| 3 |
+
"bos_token": "<|hy_begin▁of▁sentence|>",
|
| 4 |
+
"clean_up_tokenization_spaces": true,
|
| 5 |
+
"eos_token": "<|hy_place▁holder▁no▁2|>",
|
| 6 |
+
"extra_special_tokens": [
|
| 7 |
+
"<|source|>",
|
| 8 |
+
"<|translation|>",
|
| 9 |
+
"<|im_end|>"
|
| 10 |
+
],
|
| 11 |
+
"is_local": false,
|
| 12 |
+
"model_max_length": 1000000000000000019884624838656,
|
| 13 |
+
"model_specific_special_tokens": {},
|
| 14 |
+
"pad_token": "<|hy_▁pad▁|>",
|
| 15 |
+
"padding_side": "right",
|
| 16 |
+
"tokenizer_class": "TokenizersBackend"
|
| 17 |
+
}
|
last-checkpoint/trainer_state.json
ADDED
|
@@ -0,0 +1,2625 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_global_step": null,
|
| 3 |
+
"best_metric": null,
|
| 4 |
+
"best_model_checkpoint": null,
|
| 5 |
+
"epoch": 0.21052631578947367,
|
| 6 |
+
"eval_steps": 369,
|
| 7 |
+
"global_step": 369,
|
| 8 |
+
"is_hyper_param_search": false,
|
| 9 |
+
"is_local_process_zero": true,
|
| 10 |
+
"is_world_process_zero": true,
|
| 11 |
+
"log_history": [
|
| 12 |
+
{
|
| 13 |
+
"epoch": 0.0005705320211096848,
|
| 14 |
+
"grad_norm": 83.0,
|
| 15 |
+
"learning_rate": 0.0,
|
| 16 |
+
"loss": 3.648493528366089,
|
| 17 |
+
"step": 1
|
| 18 |
+
},
|
| 19 |
+
{
|
| 20 |
+
"epoch": 0.0011410640422193695,
|
| 21 |
+
"grad_norm": 84.5,
|
| 22 |
+
"learning_rate": 1.3513513513513515e-06,
|
| 23 |
+
"loss": 3.7405500411987305,
|
| 24 |
+
"step": 2
|
| 25 |
+
},
|
| 26 |
+
{
|
| 27 |
+
"epoch": 0.0017115960633290544,
|
| 28 |
+
"grad_norm": 74.0,
|
| 29 |
+
"learning_rate": 2.702702702702703e-06,
|
| 30 |
+
"loss": 3.510922431945801,
|
| 31 |
+
"step": 3
|
| 32 |
+
},
|
| 33 |
+
{
|
| 34 |
+
"epoch": 0.002282128084438739,
|
| 35 |
+
"grad_norm": 74.0,
|
| 36 |
+
"learning_rate": 4.0540540540540545e-06,
|
| 37 |
+
"loss": 3.477842330932617,
|
| 38 |
+
"step": 4
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 0.002852660105548424,
|
| 42 |
+
"grad_norm": 48.5,
|
| 43 |
+
"learning_rate": 5.405405405405406e-06,
|
| 44 |
+
"loss": 3.2050325870513916,
|
| 45 |
+
"step": 5
|
| 46 |
+
},
|
| 47 |
+
{
|
| 48 |
+
"epoch": 0.0034231921266581087,
|
| 49 |
+
"grad_norm": 35.0,
|
| 50 |
+
"learning_rate": 6.7567567567567575e-06,
|
| 51 |
+
"loss": 2.9610347747802734,
|
| 52 |
+
"step": 6
|
| 53 |
+
},
|
| 54 |
+
{
|
| 55 |
+
"epoch": 0.003993724147767793,
|
| 56 |
+
"grad_norm": 25.75,
|
| 57 |
+
"learning_rate": 8.108108108108109e-06,
|
| 58 |
+
"loss": 2.8089160919189453,
|
| 59 |
+
"step": 7
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 0.004564256168877478,
|
| 63 |
+
"grad_norm": 15.25,
|
| 64 |
+
"learning_rate": 9.45945945945946e-06,
|
| 65 |
+
"loss": 2.672607183456421,
|
| 66 |
+
"step": 8
|
| 67 |
+
},
|
| 68 |
+
{
|
| 69 |
+
"epoch": 0.005134788189987163,
|
| 70 |
+
"grad_norm": 10.125,
|
| 71 |
+
"learning_rate": 1.0810810810810812e-05,
|
| 72 |
+
"loss": 2.4392411708831787,
|
| 73 |
+
"step": 9
|
| 74 |
+
},
|
| 75 |
+
{
|
| 76 |
+
"epoch": 0.005705320211096848,
|
| 77 |
+
"grad_norm": 8.875,
|
| 78 |
+
"learning_rate": 1.2162162162162164e-05,
|
| 79 |
+
"loss": 2.4409432411193848,
|
| 80 |
+
"step": 10
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 0.006275852232206533,
|
| 84 |
+
"grad_norm": 7.40625,
|
| 85 |
+
"learning_rate": 1.3513513513513515e-05,
|
| 86 |
+
"loss": 2.3299427032470703,
|
| 87 |
+
"step": 11
|
| 88 |
+
},
|
| 89 |
+
{
|
| 90 |
+
"epoch": 0.0068463842533162175,
|
| 91 |
+
"grad_norm": 6.6875,
|
| 92 |
+
"learning_rate": 1.4864864864864867e-05,
|
| 93 |
+
"loss": 2.2852554321289062,
|
| 94 |
+
"step": 12
|
| 95 |
+
},
|
| 96 |
+
{
|
| 97 |
+
"epoch": 0.007416916274425902,
|
| 98 |
+
"grad_norm": 6.5625,
|
| 99 |
+
"learning_rate": 1.6216216216216218e-05,
|
| 100 |
+
"loss": 2.2712786197662354,
|
| 101 |
+
"step": 13
|
| 102 |
+
},
|
| 103 |
+
{
|
| 104 |
+
"epoch": 0.007987448295535587,
|
| 105 |
+
"grad_norm": 7.1875,
|
| 106 |
+
"learning_rate": 1.756756756756757e-05,
|
| 107 |
+
"loss": 2.2143714427948,
|
| 108 |
+
"step": 14
|
| 109 |
+
},
|
| 110 |
+
{
|
| 111 |
+
"epoch": 0.008557980316645272,
|
| 112 |
+
"grad_norm": 7.0,
|
| 113 |
+
"learning_rate": 1.891891891891892e-05,
|
| 114 |
+
"loss": 2.0812437534332275,
|
| 115 |
+
"step": 15
|
| 116 |
+
},
|
| 117 |
+
{
|
| 118 |
+
"epoch": 0.009128512337754956,
|
| 119 |
+
"grad_norm": 6.28125,
|
| 120 |
+
"learning_rate": 2.0270270270270273e-05,
|
| 121 |
+
"loss": 2.068169355392456,
|
| 122 |
+
"step": 16
|
| 123 |
+
},
|
| 124 |
+
{
|
| 125 |
+
"epoch": 0.009699044358864642,
|
| 126 |
+
"grad_norm": 5.75,
|
| 127 |
+
"learning_rate": 2.1621621621621624e-05,
|
| 128 |
+
"loss": 1.8387004137039185,
|
| 129 |
+
"step": 17
|
| 130 |
+
},
|
| 131 |
+
{
|
| 132 |
+
"epoch": 0.010269576379974325,
|
| 133 |
+
"grad_norm": 4.3125,
|
| 134 |
+
"learning_rate": 2.2972972972972976e-05,
|
| 135 |
+
"loss": 1.7710001468658447,
|
| 136 |
+
"step": 18
|
| 137 |
+
},
|
| 138 |
+
{
|
| 139 |
+
"epoch": 0.01084010840108401,
|
| 140 |
+
"grad_norm": 4.25,
|
| 141 |
+
"learning_rate": 2.4324324324324327e-05,
|
| 142 |
+
"loss": 1.7796661853790283,
|
| 143 |
+
"step": 19
|
| 144 |
+
},
|
| 145 |
+
{
|
| 146 |
+
"epoch": 0.011410640422193696,
|
| 147 |
+
"grad_norm": 3.5,
|
| 148 |
+
"learning_rate": 2.5675675675675675e-05,
|
| 149 |
+
"loss": 1.6957234144210815,
|
| 150 |
+
"step": 20
|
| 151 |
+
},
|
| 152 |
+
{
|
| 153 |
+
"epoch": 0.01198117244330338,
|
| 154 |
+
"grad_norm": 3.21875,
|
| 155 |
+
"learning_rate": 2.702702702702703e-05,
|
| 156 |
+
"loss": 1.7516167163848877,
|
| 157 |
+
"step": 21
|
| 158 |
+
},
|
| 159 |
+
{
|
| 160 |
+
"epoch": 0.012551704464413066,
|
| 161 |
+
"grad_norm": 2.78125,
|
| 162 |
+
"learning_rate": 2.8378378378378378e-05,
|
| 163 |
+
"loss": 1.6087043285369873,
|
| 164 |
+
"step": 22
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 0.01312223648552275,
|
| 168 |
+
"grad_norm": 2.34375,
|
| 169 |
+
"learning_rate": 2.9729729729729733e-05,
|
| 170 |
+
"loss": 1.5943574905395508,
|
| 171 |
+
"step": 23
|
| 172 |
+
},
|
| 173 |
+
{
|
| 174 |
+
"epoch": 0.013692768506632435,
|
| 175 |
+
"grad_norm": 2.140625,
|
| 176 |
+
"learning_rate": 3.108108108108108e-05,
|
| 177 |
+
"loss": 1.599621295928955,
|
| 178 |
+
"step": 24
|
| 179 |
+
},
|
| 180 |
+
{
|
| 181 |
+
"epoch": 0.014263300527742119,
|
| 182 |
+
"grad_norm": 2.234375,
|
| 183 |
+
"learning_rate": 3.2432432432432436e-05,
|
| 184 |
+
"loss": 1.6016688346862793,
|
| 185 |
+
"step": 25
|
| 186 |
+
},
|
| 187 |
+
{
|
| 188 |
+
"epoch": 0.014833832548851804,
|
| 189 |
+
"grad_norm": 1.9609375,
|
| 190 |
+
"learning_rate": 3.3783783783783784e-05,
|
| 191 |
+
"loss": 1.5124552249908447,
|
| 192 |
+
"step": 26
|
| 193 |
+
},
|
| 194 |
+
{
|
| 195 |
+
"epoch": 0.01540436456996149,
|
| 196 |
+
"grad_norm": 1.9765625,
|
| 197 |
+
"learning_rate": 3.513513513513514e-05,
|
| 198 |
+
"loss": 1.5520291328430176,
|
| 199 |
+
"step": 27
|
| 200 |
+
},
|
| 201 |
+
{
|
| 202 |
+
"epoch": 0.015974896591071173,
|
| 203 |
+
"grad_norm": 1.90625,
|
| 204 |
+
"learning_rate": 3.648648648648649e-05,
|
| 205 |
+
"loss": 1.4819629192352295,
|
| 206 |
+
"step": 28
|
| 207 |
+
},
|
| 208 |
+
{
|
| 209 |
+
"epoch": 0.01654542861218086,
|
| 210 |
+
"grad_norm": 2.0,
|
| 211 |
+
"learning_rate": 3.783783783783784e-05,
|
| 212 |
+
"loss": 1.5304462909698486,
|
| 213 |
+
"step": 29
|
| 214 |
+
},
|
| 215 |
+
{
|
| 216 |
+
"epoch": 0.017115960633290545,
|
| 217 |
+
"grad_norm": 1.7734375,
|
| 218 |
+
"learning_rate": 3.918918918918919e-05,
|
| 219 |
+
"loss": 1.4461307525634766,
|
| 220 |
+
"step": 30
|
| 221 |
+
},
|
| 222 |
+
{
|
| 223 |
+
"epoch": 0.017686492654400227,
|
| 224 |
+
"grad_norm": 1.8125,
|
| 225 |
+
"learning_rate": 4.0540540540540545e-05,
|
| 226 |
+
"loss": 1.4548516273498535,
|
| 227 |
+
"step": 31
|
| 228 |
+
},
|
| 229 |
+
{
|
| 230 |
+
"epoch": 0.018257024675509912,
|
| 231 |
+
"grad_norm": 1.6875,
|
| 232 |
+
"learning_rate": 4.189189189189189e-05,
|
| 233 |
+
"loss": 1.435849905014038,
|
| 234 |
+
"step": 32
|
| 235 |
+
},
|
| 236 |
+
{
|
| 237 |
+
"epoch": 0.018827556696619598,
|
| 238 |
+
"grad_norm": 1.578125,
|
| 239 |
+
"learning_rate": 4.324324324324325e-05,
|
| 240 |
+
"loss": 1.4789021015167236,
|
| 241 |
+
"step": 33
|
| 242 |
+
},
|
| 243 |
+
{
|
| 244 |
+
"epoch": 0.019398088717729283,
|
| 245 |
+
"grad_norm": 1.578125,
|
| 246 |
+
"learning_rate": 4.4594594594594596e-05,
|
| 247 |
+
"loss": 1.3856297731399536,
|
| 248 |
+
"step": 34
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 0.01996862073883897,
|
| 252 |
+
"grad_norm": 1.65625,
|
| 253 |
+
"learning_rate": 4.594594594594595e-05,
|
| 254 |
+
"loss": 1.5028152465820312,
|
| 255 |
+
"step": 35
|
| 256 |
+
},
|
| 257 |
+
{
|
| 258 |
+
"epoch": 0.02053915275994865,
|
| 259 |
+
"grad_norm": 1.46875,
|
| 260 |
+
"learning_rate": 4.72972972972973e-05,
|
| 261 |
+
"loss": 1.4294812679290771,
|
| 262 |
+
"step": 36
|
| 263 |
+
},
|
| 264 |
+
{
|
| 265 |
+
"epoch": 0.021109684781058336,
|
| 266 |
+
"grad_norm": 1.65625,
|
| 267 |
+
"learning_rate": 4.8648648648648654e-05,
|
| 268 |
+
"loss": 1.3971917629241943,
|
| 269 |
+
"step": 37
|
| 270 |
+
},
|
| 271 |
+
{
|
| 272 |
+
"epoch": 0.02168021680216802,
|
| 273 |
+
"grad_norm": 1.515625,
|
| 274 |
+
"learning_rate": 5e-05,
|
| 275 |
+
"loss": 1.3995487689971924,
|
| 276 |
+
"step": 38
|
| 277 |
+
},
|
| 278 |
+
{
|
| 279 |
+
"epoch": 0.022250748823277707,
|
| 280 |
+
"grad_norm": 1.703125,
|
| 281 |
+
"learning_rate": 4.9972283813747225e-05,
|
| 282 |
+
"loss": 1.4693856239318848,
|
| 283 |
+
"step": 39
|
| 284 |
+
},
|
| 285 |
+
{
|
| 286 |
+
"epoch": 0.022821280844387393,
|
| 287 |
+
"grad_norm": 1.5703125,
|
| 288 |
+
"learning_rate": 4.994456762749446e-05,
|
| 289 |
+
"loss": 1.4715073108673096,
|
| 290 |
+
"step": 40
|
| 291 |
+
},
|
| 292 |
+
{
|
| 293 |
+
"epoch": 0.023391812865497075,
|
| 294 |
+
"grad_norm": 1.703125,
|
| 295 |
+
"learning_rate": 4.9916851441241684e-05,
|
| 296 |
+
"loss": 1.490320086479187,
|
| 297 |
+
"step": 41
|
| 298 |
+
},
|
| 299 |
+
{
|
| 300 |
+
"epoch": 0.02396234488660676,
|
| 301 |
+
"grad_norm": 1.5,
|
| 302 |
+
"learning_rate": 4.9889135254988913e-05,
|
| 303 |
+
"loss": 1.3657546043395996,
|
| 304 |
+
"step": 42
|
| 305 |
+
},
|
| 306 |
+
{
|
| 307 |
+
"epoch": 0.024532876907716446,
|
| 308 |
+
"grad_norm": 1.7109375,
|
| 309 |
+
"learning_rate": 4.986141906873614e-05,
|
| 310 |
+
"loss": 1.4324053525924683,
|
| 311 |
+
"step": 43
|
| 312 |
+
},
|
| 313 |
+
{
|
| 314 |
+
"epoch": 0.02510340892882613,
|
| 315 |
+
"grad_norm": 1.7890625,
|
| 316 |
+
"learning_rate": 4.983370288248337e-05,
|
| 317 |
+
"loss": 1.3849389553070068,
|
| 318 |
+
"step": 44
|
| 319 |
+
},
|
| 320 |
+
{
|
| 321 |
+
"epoch": 0.025673940949935817,
|
| 322 |
+
"grad_norm": 1.453125,
|
| 323 |
+
"learning_rate": 4.98059866962306e-05,
|
| 324 |
+
"loss": 1.425079345703125,
|
| 325 |
+
"step": 45
|
| 326 |
+
},
|
| 327 |
+
{
|
| 328 |
+
"epoch": 0.0262444729710455,
|
| 329 |
+
"grad_norm": 1.484375,
|
| 330 |
+
"learning_rate": 4.977827050997783e-05,
|
| 331 |
+
"loss": 1.4127968549728394,
|
| 332 |
+
"step": 46
|
| 333 |
+
},
|
| 334 |
+
{
|
| 335 |
+
"epoch": 0.026815004992155184,
|
| 336 |
+
"grad_norm": 1.5390625,
|
| 337 |
+
"learning_rate": 4.9750554323725054e-05,
|
| 338 |
+
"loss": 1.429938793182373,
|
| 339 |
+
"step": 47
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 0.02738553701326487,
|
| 343 |
+
"grad_norm": 1.4140625,
|
| 344 |
+
"learning_rate": 4.972283813747229e-05,
|
| 345 |
+
"loss": 1.4178887605667114,
|
| 346 |
+
"step": 48
|
| 347 |
+
},
|
| 348 |
+
{
|
| 349 |
+
"epoch": 0.027956069034374555,
|
| 350 |
+
"grad_norm": 1.46875,
|
| 351 |
+
"learning_rate": 4.969512195121951e-05,
|
| 352 |
+
"loss": 1.4397588968276978,
|
| 353 |
+
"step": 49
|
| 354 |
+
},
|
| 355 |
+
{
|
| 356 |
+
"epoch": 0.028526601055484237,
|
| 357 |
+
"grad_norm": 1.453125,
|
| 358 |
+
"learning_rate": 4.966740576496674e-05,
|
| 359 |
+
"loss": 1.3697854280471802,
|
| 360 |
+
"step": 50
|
| 361 |
+
},
|
| 362 |
+
{
|
| 363 |
+
"epoch": 0.029097133076593923,
|
| 364 |
+
"grad_norm": 1.3984375,
|
| 365 |
+
"learning_rate": 4.963968957871397e-05,
|
| 366 |
+
"loss": 1.3517988920211792,
|
| 367 |
+
"step": 51
|
| 368 |
+
},
|
| 369 |
+
{
|
| 370 |
+
"epoch": 0.02966766509770361,
|
| 371 |
+
"grad_norm": 1.3984375,
|
| 372 |
+
"learning_rate": 4.9611973392461195e-05,
|
| 373 |
+
"loss": 1.4193122386932373,
|
| 374 |
+
"step": 52
|
| 375 |
+
},
|
| 376 |
+
{
|
| 377 |
+
"epoch": 0.030238197118813294,
|
| 378 |
+
"grad_norm": 1.34375,
|
| 379 |
+
"learning_rate": 4.958425720620843e-05,
|
| 380 |
+
"loss": 1.37640380859375,
|
| 381 |
+
"step": 53
|
| 382 |
+
},
|
| 383 |
+
{
|
| 384 |
+
"epoch": 0.03080872913992298,
|
| 385 |
+
"grad_norm": 1.40625,
|
| 386 |
+
"learning_rate": 4.9556541019955654e-05,
|
| 387 |
+
"loss": 1.336474895477295,
|
| 388 |
+
"step": 54
|
| 389 |
+
},
|
| 390 |
+
{
|
| 391 |
+
"epoch": 0.031379261161032665,
|
| 392 |
+
"grad_norm": 1.515625,
|
| 393 |
+
"learning_rate": 4.952882483370288e-05,
|
| 394 |
+
"loss": 1.4701391458511353,
|
| 395 |
+
"step": 55
|
| 396 |
+
},
|
| 397 |
+
{
|
| 398 |
+
"epoch": 0.03194979318214235,
|
| 399 |
+
"grad_norm": 1.4765625,
|
| 400 |
+
"learning_rate": 4.950110864745011e-05,
|
| 401 |
+
"loss": 1.3760974407196045,
|
| 402 |
+
"step": 56
|
| 403 |
+
},
|
| 404 |
+
{
|
| 405 |
+
"epoch": 0.032520325203252036,
|
| 406 |
+
"grad_norm": 1.4609375,
|
| 407 |
+
"learning_rate": 4.947339246119734e-05,
|
| 408 |
+
"loss": 1.3897124528884888,
|
| 409 |
+
"step": 57
|
| 410 |
+
},
|
| 411 |
+
{
|
| 412 |
+
"epoch": 0.03309085722436172,
|
| 413 |
+
"grad_norm": 1.578125,
|
| 414 |
+
"learning_rate": 4.944567627494457e-05,
|
| 415 |
+
"loss": 1.4239261150360107,
|
| 416 |
+
"step": 58
|
| 417 |
+
},
|
| 418 |
+
{
|
| 419 |
+
"epoch": 0.0336613892454714,
|
| 420 |
+
"grad_norm": 1.53125,
|
| 421 |
+
"learning_rate": 4.94179600886918e-05,
|
| 422 |
+
"loss": 1.3669216632843018,
|
| 423 |
+
"step": 59
|
| 424 |
+
},
|
| 425 |
+
{
|
| 426 |
+
"epoch": 0.03423192126658109,
|
| 427 |
+
"grad_norm": 1.3515625,
|
| 428 |
+
"learning_rate": 4.9390243902439024e-05,
|
| 429 |
+
"loss": 1.346958041191101,
|
| 430 |
+
"step": 60
|
| 431 |
+
},
|
| 432 |
+
{
|
| 433 |
+
"epoch": 0.03480245328769077,
|
| 434 |
+
"grad_norm": 1.5234375,
|
| 435 |
+
"learning_rate": 4.936252771618626e-05,
|
| 436 |
+
"loss": 1.4235575199127197,
|
| 437 |
+
"step": 61
|
| 438 |
+
},
|
| 439 |
+
{
|
| 440 |
+
"epoch": 0.03537298530880045,
|
| 441 |
+
"grad_norm": 1.3359375,
|
| 442 |
+
"learning_rate": 4.933481152993348e-05,
|
| 443 |
+
"loss": 1.3075377941131592,
|
| 444 |
+
"step": 62
|
| 445 |
+
},
|
| 446 |
+
{
|
| 447 |
+
"epoch": 0.03594351732991014,
|
| 448 |
+
"grad_norm": 1.3203125,
|
| 449 |
+
"learning_rate": 4.930709534368071e-05,
|
| 450 |
+
"loss": 1.3214820623397827,
|
| 451 |
+
"step": 63
|
| 452 |
+
},
|
| 453 |
+
{
|
| 454 |
+
"epoch": 0.036514049351019824,
|
| 455 |
+
"grad_norm": 1.3515625,
|
| 456 |
+
"learning_rate": 4.927937915742794e-05,
|
| 457 |
+
"loss": 1.39829421043396,
|
| 458 |
+
"step": 64
|
| 459 |
+
},
|
| 460 |
+
{
|
| 461 |
+
"epoch": 0.03708458137212951,
|
| 462 |
+
"grad_norm": 1.3671875,
|
| 463 |
+
"learning_rate": 4.9251662971175164e-05,
|
| 464 |
+
"loss": 1.3523836135864258,
|
| 465 |
+
"step": 65
|
| 466 |
+
},
|
| 467 |
+
{
|
| 468 |
+
"epoch": 0.037655113393239195,
|
| 469 |
+
"grad_norm": 1.3125,
|
| 470 |
+
"learning_rate": 4.92239467849224e-05,
|
| 471 |
+
"loss": 1.3268153667449951,
|
| 472 |
+
"step": 66
|
| 473 |
+
},
|
| 474 |
+
{
|
| 475 |
+
"epoch": 0.03822564541434888,
|
| 476 |
+
"grad_norm": 1.28125,
|
| 477 |
+
"learning_rate": 4.919623059866962e-05,
|
| 478 |
+
"loss": 1.3205022811889648,
|
| 479 |
+
"step": 67
|
| 480 |
+
},
|
| 481 |
+
{
|
| 482 |
+
"epoch": 0.038796177435458566,
|
| 483 |
+
"grad_norm": 1.2734375,
|
| 484 |
+
"learning_rate": 4.916851441241685e-05,
|
| 485 |
+
"loss": 1.2956037521362305,
|
| 486 |
+
"step": 68
|
| 487 |
+
},
|
| 488 |
+
{
|
| 489 |
+
"epoch": 0.03936670945656825,
|
| 490 |
+
"grad_norm": 1.375,
|
| 491 |
+
"learning_rate": 4.914079822616408e-05,
|
| 492 |
+
"loss": 1.3702654838562012,
|
| 493 |
+
"step": 69
|
| 494 |
+
},
|
| 495 |
+
{
|
| 496 |
+
"epoch": 0.03993724147767794,
|
| 497 |
+
"grad_norm": 1.296875,
|
| 498 |
+
"learning_rate": 4.911308203991131e-05,
|
| 499 |
+
"loss": 1.388296127319336,
|
| 500 |
+
"step": 70
|
| 501 |
+
},
|
| 502 |
+
{
|
| 503 |
+
"epoch": 0.04050777349878762,
|
| 504 |
+
"grad_norm": 1.5078125,
|
| 505 |
+
"learning_rate": 4.908536585365854e-05,
|
| 506 |
+
"loss": 1.4403045177459717,
|
| 507 |
+
"step": 71
|
| 508 |
+
},
|
| 509 |
+
{
|
| 510 |
+
"epoch": 0.0410783055198973,
|
| 511 |
+
"grad_norm": 1.25,
|
| 512 |
+
"learning_rate": 4.905764966740577e-05,
|
| 513 |
+
"loss": 1.3626902103424072,
|
| 514 |
+
"step": 72
|
| 515 |
+
},
|
| 516 |
+
{
|
| 517 |
+
"epoch": 0.04164883754100699,
|
| 518 |
+
"grad_norm": 1.34375,
|
| 519 |
+
"learning_rate": 4.902993348115299e-05,
|
| 520 |
+
"loss": 1.382088303565979,
|
| 521 |
+
"step": 73
|
| 522 |
+
},
|
| 523 |
+
{
|
| 524 |
+
"epoch": 0.04221936956211667,
|
| 525 |
+
"grad_norm": 1.2578125,
|
| 526 |
+
"learning_rate": 4.900221729490023e-05,
|
| 527 |
+
"loss": 1.3237360715866089,
|
| 528 |
+
"step": 74
|
| 529 |
+
},
|
| 530 |
+
{
|
| 531 |
+
"epoch": 0.04278990158322636,
|
| 532 |
+
"grad_norm": 1.296875,
|
| 533 |
+
"learning_rate": 4.897450110864745e-05,
|
| 534 |
+
"loss": 1.319187879562378,
|
| 535 |
+
"step": 75
|
| 536 |
+
},
|
| 537 |
+
{
|
| 538 |
+
"epoch": 0.04336043360433604,
|
| 539 |
+
"grad_norm": 1.3828125,
|
| 540 |
+
"learning_rate": 4.894678492239468e-05,
|
| 541 |
+
"loss": 1.3707743883132935,
|
| 542 |
+
"step": 76
|
| 543 |
+
},
|
| 544 |
+
{
|
| 545 |
+
"epoch": 0.043930965625445725,
|
| 546 |
+
"grad_norm": 1.2578125,
|
| 547 |
+
"learning_rate": 4.891906873614191e-05,
|
| 548 |
+
"loss": 1.3658738136291504,
|
| 549 |
+
"step": 77
|
| 550 |
+
},
|
| 551 |
+
{
|
| 552 |
+
"epoch": 0.044501497646555414,
|
| 553 |
+
"grad_norm": 1.265625,
|
| 554 |
+
"learning_rate": 4.8891352549889134e-05,
|
| 555 |
+
"loss": 1.3247051239013672,
|
| 556 |
+
"step": 78
|
| 557 |
+
},
|
| 558 |
+
{
|
| 559 |
+
"epoch": 0.045072029667665096,
|
| 560 |
+
"grad_norm": 1.421875,
|
| 561 |
+
"learning_rate": 4.886363636363637e-05,
|
| 562 |
+
"loss": 1.3614035844802856,
|
| 563 |
+
"step": 79
|
| 564 |
+
},
|
| 565 |
+
{
|
| 566 |
+
"epoch": 0.045642561688774785,
|
| 567 |
+
"grad_norm": 1.2734375,
|
| 568 |
+
"learning_rate": 4.883592017738359e-05,
|
| 569 |
+
"loss": 1.2589421272277832,
|
| 570 |
+
"step": 80
|
| 571 |
+
},
|
| 572 |
+
{
|
| 573 |
+
"epoch": 0.04621309370988447,
|
| 574 |
+
"grad_norm": 1.28125,
|
| 575 |
+
"learning_rate": 4.880820399113082e-05,
|
| 576 |
+
"loss": 1.3525424003601074,
|
| 577 |
+
"step": 81
|
| 578 |
+
},
|
| 579 |
+
{
|
| 580 |
+
"epoch": 0.04678362573099415,
|
| 581 |
+
"grad_norm": 1.2578125,
|
| 582 |
+
"learning_rate": 4.878048780487805e-05,
|
| 583 |
+
"loss": 1.2903777360916138,
|
| 584 |
+
"step": 82
|
| 585 |
+
},
|
| 586 |
+
{
|
| 587 |
+
"epoch": 0.04735415775210384,
|
| 588 |
+
"grad_norm": 1.328125,
|
| 589 |
+
"learning_rate": 4.875277161862528e-05,
|
| 590 |
+
"loss": 1.3538789749145508,
|
| 591 |
+
"step": 83
|
| 592 |
+
},
|
| 593 |
+
{
|
| 594 |
+
"epoch": 0.04792468977321352,
|
| 595 |
+
"grad_norm": 1.3046875,
|
| 596 |
+
"learning_rate": 4.872505543237251e-05,
|
| 597 |
+
"loss": 1.3419591188430786,
|
| 598 |
+
"step": 84
|
| 599 |
+
},
|
| 600 |
+
{
|
| 601 |
+
"epoch": 0.04849522179432321,
|
| 602 |
+
"grad_norm": 1.28125,
|
| 603 |
+
"learning_rate": 4.869733924611974e-05,
|
| 604 |
+
"loss": 1.3367938995361328,
|
| 605 |
+
"step": 85
|
| 606 |
+
},
|
| 607 |
+
{
|
| 608 |
+
"epoch": 0.04906575381543289,
|
| 609 |
+
"grad_norm": 1.3046875,
|
| 610 |
+
"learning_rate": 4.866962305986696e-05,
|
| 611 |
+
"loss": 1.2979538440704346,
|
| 612 |
+
"step": 86
|
| 613 |
+
},
|
| 614 |
+
{
|
| 615 |
+
"epoch": 0.049636285836542574,
|
| 616 |
+
"grad_norm": 1.25,
|
| 617 |
+
"learning_rate": 4.864190687361419e-05,
|
| 618 |
+
"loss": 1.348291039466858,
|
| 619 |
+
"step": 87
|
| 620 |
+
},
|
| 621 |
+
{
|
| 622 |
+
"epoch": 0.05020681785765226,
|
| 623 |
+
"grad_norm": 1.3125,
|
| 624 |
+
"learning_rate": 4.861419068736142e-05,
|
| 625 |
+
"loss": 1.3377124071121216,
|
| 626 |
+
"step": 88
|
| 627 |
+
},
|
| 628 |
+
{
|
| 629 |
+
"epoch": 0.050777349878761945,
|
| 630 |
+
"grad_norm": 1.328125,
|
| 631 |
+
"learning_rate": 4.8586474501108644e-05,
|
| 632 |
+
"loss": 1.3180426359176636,
|
| 633 |
+
"step": 89
|
| 634 |
+
},
|
| 635 |
+
{
|
| 636 |
+
"epoch": 0.051347881899871634,
|
| 637 |
+
"grad_norm": 1.2109375,
|
| 638 |
+
"learning_rate": 4.855875831485588e-05,
|
| 639 |
+
"loss": 1.3215968608856201,
|
| 640 |
+
"step": 90
|
| 641 |
+
},
|
| 642 |
+
{
|
| 643 |
+
"epoch": 0.051918413920981316,
|
| 644 |
+
"grad_norm": 1.2421875,
|
| 645 |
+
"learning_rate": 4.85310421286031e-05,
|
| 646 |
+
"loss": 1.3354041576385498,
|
| 647 |
+
"step": 91
|
| 648 |
+
},
|
| 649 |
+
{
|
| 650 |
+
"epoch": 0.052488945942091,
|
| 651 |
+
"grad_norm": 1.28125,
|
| 652 |
+
"learning_rate": 4.850332594235034e-05,
|
| 653 |
+
"loss": 1.3552148342132568,
|
| 654 |
+
"step": 92
|
| 655 |
+
},
|
| 656 |
+
{
|
| 657 |
+
"epoch": 0.05305947796320069,
|
| 658 |
+
"grad_norm": 1.2421875,
|
| 659 |
+
"learning_rate": 4.847560975609756e-05,
|
| 660 |
+
"loss": 1.2916048765182495,
|
| 661 |
+
"step": 93
|
| 662 |
+
},
|
| 663 |
+
{
|
| 664 |
+
"epoch": 0.05363000998431037,
|
| 665 |
+
"grad_norm": 1.3125,
|
| 666 |
+
"learning_rate": 4.844789356984479e-05,
|
| 667 |
+
"loss": 1.3131537437438965,
|
| 668 |
+
"step": 94
|
| 669 |
+
},
|
| 670 |
+
{
|
| 671 |
+
"epoch": 0.05420054200542006,
|
| 672 |
+
"grad_norm": 1.296875,
|
| 673 |
+
"learning_rate": 4.842017738359202e-05,
|
| 674 |
+
"loss": 1.2902660369873047,
|
| 675 |
+
"step": 95
|
| 676 |
+
},
|
| 677 |
+
{
|
| 678 |
+
"epoch": 0.05477107402652974,
|
| 679 |
+
"grad_norm": 1.28125,
|
| 680 |
+
"learning_rate": 4.839246119733925e-05,
|
| 681 |
+
"loss": 1.3799315690994263,
|
| 682 |
+
"step": 96
|
| 683 |
+
},
|
| 684 |
+
{
|
| 685 |
+
"epoch": 0.05534160604763942,
|
| 686 |
+
"grad_norm": 1.3515625,
|
| 687 |
+
"learning_rate": 4.836474501108647e-05,
|
| 688 |
+
"loss": 1.3607311248779297,
|
| 689 |
+
"step": 97
|
| 690 |
+
},
|
| 691 |
+
{
|
| 692 |
+
"epoch": 0.05591213806874911,
|
| 693 |
+
"grad_norm": 1.2421875,
|
| 694 |
+
"learning_rate": 4.833702882483371e-05,
|
| 695 |
+
"loss": 1.3038060665130615,
|
| 696 |
+
"step": 98
|
| 697 |
+
},
|
| 698 |
+
{
|
| 699 |
+
"epoch": 0.05648267008985879,
|
| 700 |
+
"grad_norm": 1.234375,
|
| 701 |
+
"learning_rate": 4.830931263858093e-05,
|
| 702 |
+
"loss": 1.318457841873169,
|
| 703 |
+
"step": 99
|
| 704 |
+
},
|
| 705 |
+
{
|
| 706 |
+
"epoch": 0.057053202110968475,
|
| 707 |
+
"grad_norm": 1.2890625,
|
| 708 |
+
"learning_rate": 4.828159645232816e-05,
|
| 709 |
+
"loss": 1.3159422874450684,
|
| 710 |
+
"step": 100
|
| 711 |
+
},
|
| 712 |
+
{
|
| 713 |
+
"epoch": 0.057623734132078164,
|
| 714 |
+
"grad_norm": 1.25,
|
| 715 |
+
"learning_rate": 4.825388026607539e-05,
|
| 716 |
+
"loss": 1.3275076150894165,
|
| 717 |
+
"step": 101
|
| 718 |
+
},
|
| 719 |
+
{
|
| 720 |
+
"epoch": 0.058194266153187846,
|
| 721 |
+
"grad_norm": 1.1796875,
|
| 722 |
+
"learning_rate": 4.8226164079822614e-05,
|
| 723 |
+
"loss": 1.2983460426330566,
|
| 724 |
+
"step": 102
|
| 725 |
+
},
|
| 726 |
+
{
|
| 727 |
+
"epoch": 0.058764798174297535,
|
| 728 |
+
"grad_norm": 1.171875,
|
| 729 |
+
"learning_rate": 4.819844789356985e-05,
|
| 730 |
+
"loss": 1.3114261627197266,
|
| 731 |
+
"step": 103
|
| 732 |
+
},
|
| 733 |
+
{
|
| 734 |
+
"epoch": 0.05933533019540722,
|
| 735 |
+
"grad_norm": 1.1796875,
|
| 736 |
+
"learning_rate": 4.817073170731707e-05,
|
| 737 |
+
"loss": 1.266122817993164,
|
| 738 |
+
"step": 104
|
| 739 |
+
},
|
| 740 |
+
{
|
| 741 |
+
"epoch": 0.0599058622165169,
|
| 742 |
+
"grad_norm": 1.234375,
|
| 743 |
+
"learning_rate": 4.81430155210643e-05,
|
| 744 |
+
"loss": 1.3662368059158325,
|
| 745 |
+
"step": 105
|
| 746 |
+
},
|
| 747 |
+
{
|
| 748 |
+
"epoch": 0.06047639423762659,
|
| 749 |
+
"grad_norm": 1.296875,
|
| 750 |
+
"learning_rate": 4.811529933481153e-05,
|
| 751 |
+
"loss": 1.3158059120178223,
|
| 752 |
+
"step": 106
|
| 753 |
+
},
|
| 754 |
+
{
|
| 755 |
+
"epoch": 0.06104692625873627,
|
| 756 |
+
"grad_norm": 1.234375,
|
| 757 |
+
"learning_rate": 4.808758314855876e-05,
|
| 758 |
+
"loss": 1.3571752309799194,
|
| 759 |
+
"step": 107
|
| 760 |
+
},
|
| 761 |
+
{
|
| 762 |
+
"epoch": 0.06161745827984596,
|
| 763 |
+
"grad_norm": 1.34375,
|
| 764 |
+
"learning_rate": 4.805986696230599e-05,
|
| 765 |
+
"loss": 1.3249101638793945,
|
| 766 |
+
"step": 108
|
| 767 |
+
},
|
| 768 |
+
{
|
| 769 |
+
"epoch": 0.06218799030095564,
|
| 770 |
+
"grad_norm": 1.21875,
|
| 771 |
+
"learning_rate": 4.803215077605322e-05,
|
| 772 |
+
"loss": 1.3386337757110596,
|
| 773 |
+
"step": 109
|
| 774 |
+
},
|
| 775 |
+
{
|
| 776 |
+
"epoch": 0.06275852232206533,
|
| 777 |
+
"grad_norm": 1.2421875,
|
| 778 |
+
"learning_rate": 4.800443458980044e-05,
|
| 779 |
+
"loss": 1.2874070405960083,
|
| 780 |
+
"step": 110
|
| 781 |
+
},
|
| 782 |
+
{
|
| 783 |
+
"epoch": 0.06332905434317501,
|
| 784 |
+
"grad_norm": 1.28125,
|
| 785 |
+
"learning_rate": 4.797671840354768e-05,
|
| 786 |
+
"loss": 1.3232687711715698,
|
| 787 |
+
"step": 111
|
| 788 |
+
},
|
| 789 |
+
{
|
| 790 |
+
"epoch": 0.0638995863642847,
|
| 791 |
+
"grad_norm": 1.203125,
|
| 792 |
+
"learning_rate": 4.79490022172949e-05,
|
| 793 |
+
"loss": 1.3370904922485352,
|
| 794 |
+
"step": 112
|
| 795 |
+
},
|
| 796 |
+
{
|
| 797 |
+
"epoch": 0.06447011838539438,
|
| 798 |
+
"grad_norm": 1.3203125,
|
| 799 |
+
"learning_rate": 4.792128603104213e-05,
|
| 800 |
+
"loss": 1.3211901187896729,
|
| 801 |
+
"step": 113
|
| 802 |
+
},
|
| 803 |
+
{
|
| 804 |
+
"epoch": 0.06504065040650407,
|
| 805 |
+
"grad_norm": 1.2578125,
|
| 806 |
+
"learning_rate": 4.789356984478936e-05,
|
| 807 |
+
"loss": 1.3841608762741089,
|
| 808 |
+
"step": 114
|
| 809 |
+
},
|
| 810 |
+
{
|
| 811 |
+
"epoch": 0.06561118242761375,
|
| 812 |
+
"grad_norm": 1.2890625,
|
| 813 |
+
"learning_rate": 4.786585365853658e-05,
|
| 814 |
+
"loss": 1.4017915725708008,
|
| 815 |
+
"step": 115
|
| 816 |
+
},
|
| 817 |
+
{
|
| 818 |
+
"epoch": 0.06618171444872344,
|
| 819 |
+
"grad_norm": 1.4140625,
|
| 820 |
+
"learning_rate": 4.783813747228382e-05,
|
| 821 |
+
"loss": 1.4110525846481323,
|
| 822 |
+
"step": 116
|
| 823 |
+
},
|
| 824 |
+
{
|
| 825 |
+
"epoch": 0.06675224646983312,
|
| 826 |
+
"grad_norm": 1.2734375,
|
| 827 |
+
"learning_rate": 4.781042128603104e-05,
|
| 828 |
+
"loss": 1.2671241760253906,
|
| 829 |
+
"step": 117
|
| 830 |
+
},
|
| 831 |
+
{
|
| 832 |
+
"epoch": 0.0673227784909428,
|
| 833 |
+
"grad_norm": 1.21875,
|
| 834 |
+
"learning_rate": 4.778270509977827e-05,
|
| 835 |
+
"loss": 1.2970881462097168,
|
| 836 |
+
"step": 118
|
| 837 |
+
},
|
| 838 |
+
{
|
| 839 |
+
"epoch": 0.06789331051205248,
|
| 840 |
+
"grad_norm": 1.3515625,
|
| 841 |
+
"learning_rate": 4.77549889135255e-05,
|
| 842 |
+
"loss": 1.2626357078552246,
|
| 843 |
+
"step": 119
|
| 844 |
+
},
|
| 845 |
+
{
|
| 846 |
+
"epoch": 0.06846384253316218,
|
| 847 |
+
"grad_norm": 1.3203125,
|
| 848 |
+
"learning_rate": 4.772727272727273e-05,
|
| 849 |
+
"loss": 1.2779147624969482,
|
| 850 |
+
"step": 120
|
| 851 |
+
},
|
| 852 |
+
{
|
| 853 |
+
"epoch": 0.06903437455427186,
|
| 854 |
+
"grad_norm": 1.1796875,
|
| 855 |
+
"learning_rate": 4.769955654101996e-05,
|
| 856 |
+
"loss": 1.308679461479187,
|
| 857 |
+
"step": 121
|
| 858 |
+
},
|
| 859 |
+
{
|
| 860 |
+
"epoch": 0.06960490657538154,
|
| 861 |
+
"grad_norm": 1.234375,
|
| 862 |
+
"learning_rate": 4.767184035476719e-05,
|
| 863 |
+
"loss": 1.299755573272705,
|
| 864 |
+
"step": 122
|
| 865 |
+
},
|
| 866 |
+
{
|
| 867 |
+
"epoch": 0.07017543859649122,
|
| 868 |
+
"grad_norm": 1.2578125,
|
| 869 |
+
"learning_rate": 4.764412416851441e-05,
|
| 870 |
+
"loss": 1.3637490272521973,
|
| 871 |
+
"step": 123
|
| 872 |
+
},
|
| 873 |
+
{
|
| 874 |
+
"epoch": 0.0707459706176009,
|
| 875 |
+
"grad_norm": 1.21875,
|
| 876 |
+
"learning_rate": 4.761640798226164e-05,
|
| 877 |
+
"loss": 1.3058216571807861,
|
| 878 |
+
"step": 124
|
| 879 |
+
},
|
| 880 |
+
{
|
| 881 |
+
"epoch": 0.0713165026387106,
|
| 882 |
+
"grad_norm": 1.2734375,
|
| 883 |
+
"learning_rate": 4.758869179600887e-05,
|
| 884 |
+
"loss": 1.3146748542785645,
|
| 885 |
+
"step": 125
|
| 886 |
+
},
|
| 887 |
+
{
|
| 888 |
+
"epoch": 0.07188703465982028,
|
| 889 |
+
"grad_norm": 1.2109375,
|
| 890 |
+
"learning_rate": 4.75609756097561e-05,
|
| 891 |
+
"loss": 1.2844371795654297,
|
| 892 |
+
"step": 126
|
| 893 |
+
},
|
| 894 |
+
{
|
| 895 |
+
"epoch": 0.07245756668092997,
|
| 896 |
+
"grad_norm": 1.25,
|
| 897 |
+
"learning_rate": 4.753325942350333e-05,
|
| 898 |
+
"loss": 1.3195525407791138,
|
| 899 |
+
"step": 127
|
| 900 |
+
},
|
| 901 |
+
{
|
| 902 |
+
"epoch": 0.07302809870203965,
|
| 903 |
+
"grad_norm": 1.1953125,
|
| 904 |
+
"learning_rate": 4.750554323725055e-05,
|
| 905 |
+
"loss": 1.3399118185043335,
|
| 906 |
+
"step": 128
|
| 907 |
+
},
|
| 908 |
+
{
|
| 909 |
+
"epoch": 0.07359863072314933,
|
| 910 |
+
"grad_norm": 1.15625,
|
| 911 |
+
"learning_rate": 4.747782705099779e-05,
|
| 912 |
+
"loss": 1.2919648885726929,
|
| 913 |
+
"step": 129
|
| 914 |
+
},
|
| 915 |
+
{
|
| 916 |
+
"epoch": 0.07416916274425903,
|
| 917 |
+
"grad_norm": 1.2421875,
|
| 918 |
+
"learning_rate": 4.745011086474501e-05,
|
| 919 |
+
"loss": 1.277235507965088,
|
| 920 |
+
"step": 130
|
| 921 |
+
},
|
| 922 |
+
{
|
| 923 |
+
"epoch": 0.07473969476536871,
|
| 924 |
+
"grad_norm": 1.25,
|
| 925 |
+
"learning_rate": 4.742239467849224e-05,
|
| 926 |
+
"loss": 1.3034231662750244,
|
| 927 |
+
"step": 131
|
| 928 |
+
},
|
| 929 |
+
{
|
| 930 |
+
"epoch": 0.07531022678647839,
|
| 931 |
+
"grad_norm": 1.109375,
|
| 932 |
+
"learning_rate": 4.739467849223947e-05,
|
| 933 |
+
"loss": 1.2368437051773071,
|
| 934 |
+
"step": 132
|
| 935 |
+
},
|
| 936 |
+
{
|
| 937 |
+
"epoch": 0.07588075880758807,
|
| 938 |
+
"grad_norm": 1.2265625,
|
| 939 |
+
"learning_rate": 4.73669623059867e-05,
|
| 940 |
+
"loss": 1.3728649616241455,
|
| 941 |
+
"step": 133
|
| 942 |
+
},
|
| 943 |
+
{
|
| 944 |
+
"epoch": 0.07645129082869775,
|
| 945 |
+
"grad_norm": 1.109375,
|
| 946 |
+
"learning_rate": 4.733924611973393e-05,
|
| 947 |
+
"loss": 1.2506084442138672,
|
| 948 |
+
"step": 134
|
| 949 |
+
},
|
| 950 |
+
{
|
| 951 |
+
"epoch": 0.07702182284980745,
|
| 952 |
+
"grad_norm": 1.2109375,
|
| 953 |
+
"learning_rate": 4.731152993348116e-05,
|
| 954 |
+
"loss": 1.2813055515289307,
|
| 955 |
+
"step": 135
|
| 956 |
+
},
|
| 957 |
+
{
|
| 958 |
+
"epoch": 0.07759235487091713,
|
| 959 |
+
"grad_norm": 1.203125,
|
| 960 |
+
"learning_rate": 4.728381374722838e-05,
|
| 961 |
+
"loss": 1.2894189357757568,
|
| 962 |
+
"step": 136
|
| 963 |
+
},
|
| 964 |
+
{
|
| 965 |
+
"epoch": 0.07816288689202681,
|
| 966 |
+
"grad_norm": 1.203125,
|
| 967 |
+
"learning_rate": 4.725609756097561e-05,
|
| 968 |
+
"loss": 1.3396642208099365,
|
| 969 |
+
"step": 137
|
| 970 |
+
},
|
| 971 |
+
{
|
| 972 |
+
"epoch": 0.0787334189131365,
|
| 973 |
+
"grad_norm": 1.359375,
|
| 974 |
+
"learning_rate": 4.722838137472284e-05,
|
| 975 |
+
"loss": 1.3043787479400635,
|
| 976 |
+
"step": 138
|
| 977 |
+
},
|
| 978 |
+
{
|
| 979 |
+
"epoch": 0.07930395093424618,
|
| 980 |
+
"grad_norm": 1.203125,
|
| 981 |
+
"learning_rate": 4.720066518847007e-05,
|
| 982 |
+
"loss": 1.308459997177124,
|
| 983 |
+
"step": 139
|
| 984 |
+
},
|
| 985 |
+
{
|
| 986 |
+
"epoch": 0.07987448295535587,
|
| 987 |
+
"grad_norm": 1.21875,
|
| 988 |
+
"learning_rate": 4.71729490022173e-05,
|
| 989 |
+
"loss": 1.3281002044677734,
|
| 990 |
+
"step": 140
|
| 991 |
+
},
|
| 992 |
+
{
|
| 993 |
+
"epoch": 0.08044501497646556,
|
| 994 |
+
"grad_norm": 1.171875,
|
| 995 |
+
"learning_rate": 4.714523281596452e-05,
|
| 996 |
+
"loss": 1.3146984577178955,
|
| 997 |
+
"step": 141
|
| 998 |
+
},
|
| 999 |
+
{
|
| 1000 |
+
"epoch": 0.08101554699757524,
|
| 1001 |
+
"grad_norm": 1.203125,
|
| 1002 |
+
"learning_rate": 4.711751662971176e-05,
|
| 1003 |
+
"loss": 1.3078755140304565,
|
| 1004 |
+
"step": 142
|
| 1005 |
+
},
|
| 1006 |
+
{
|
| 1007 |
+
"epoch": 0.08158607901868492,
|
| 1008 |
+
"grad_norm": 1.21875,
|
| 1009 |
+
"learning_rate": 4.708980044345898e-05,
|
| 1010 |
+
"loss": 1.3129773139953613,
|
| 1011 |
+
"step": 143
|
| 1012 |
+
},
|
| 1013 |
+
{
|
| 1014 |
+
"epoch": 0.0821566110397946,
|
| 1015 |
+
"grad_norm": 1.171875,
|
| 1016 |
+
"learning_rate": 4.706208425720621e-05,
|
| 1017 |
+
"loss": 1.2827129364013672,
|
| 1018 |
+
"step": 144
|
| 1019 |
+
},
|
| 1020 |
+
{
|
| 1021 |
+
"epoch": 0.0827271430609043,
|
| 1022 |
+
"grad_norm": 1.171875,
|
| 1023 |
+
"learning_rate": 4.703436807095344e-05,
|
| 1024 |
+
"loss": 1.3232603073120117,
|
| 1025 |
+
"step": 145
|
| 1026 |
+
},
|
| 1027 |
+
{
|
| 1028 |
+
"epoch": 0.08329767508201398,
|
| 1029 |
+
"grad_norm": 1.1796875,
|
| 1030 |
+
"learning_rate": 4.700665188470067e-05,
|
| 1031 |
+
"loss": 1.220211386680603,
|
| 1032 |
+
"step": 146
|
| 1033 |
+
},
|
| 1034 |
+
{
|
| 1035 |
+
"epoch": 0.08386820710312366,
|
| 1036 |
+
"grad_norm": 1.2421875,
|
| 1037 |
+
"learning_rate": 4.69789356984479e-05,
|
| 1038 |
+
"loss": 1.3406665325164795,
|
| 1039 |
+
"step": 147
|
| 1040 |
+
},
|
| 1041 |
+
{
|
| 1042 |
+
"epoch": 0.08443873912423334,
|
| 1043 |
+
"grad_norm": 1.1875,
|
| 1044 |
+
"learning_rate": 4.695121951219512e-05,
|
| 1045 |
+
"loss": 1.2698848247528076,
|
| 1046 |
+
"step": 148
|
| 1047 |
+
},
|
| 1048 |
+
{
|
| 1049 |
+
"epoch": 0.08500927114534303,
|
| 1050 |
+
"grad_norm": 1.1171875,
|
| 1051 |
+
"learning_rate": 4.692350332594235e-05,
|
| 1052 |
+
"loss": 1.3016014099121094,
|
| 1053 |
+
"step": 149
|
| 1054 |
+
},
|
| 1055 |
+
{
|
| 1056 |
+
"epoch": 0.08557980316645272,
|
| 1057 |
+
"grad_norm": 1.1484375,
|
| 1058 |
+
"learning_rate": 4.689578713968958e-05,
|
| 1059 |
+
"loss": 1.2674150466918945,
|
| 1060 |
+
"step": 150
|
| 1061 |
+
},
|
| 1062 |
+
{
|
| 1063 |
+
"epoch": 0.0861503351875624,
|
| 1064 |
+
"grad_norm": 1.234375,
|
| 1065 |
+
"learning_rate": 4.686807095343681e-05,
|
| 1066 |
+
"loss": 1.316935420036316,
|
| 1067 |
+
"step": 151
|
| 1068 |
+
},
|
| 1069 |
+
{
|
| 1070 |
+
"epoch": 0.08672086720867209,
|
| 1071 |
+
"grad_norm": 1.125,
|
| 1072 |
+
"learning_rate": 4.684035476718403e-05,
|
| 1073 |
+
"loss": 1.263155221939087,
|
| 1074 |
+
"step": 152
|
| 1075 |
+
},
|
| 1076 |
+
{
|
| 1077 |
+
"epoch": 0.08729139922978177,
|
| 1078 |
+
"grad_norm": 1.1875,
|
| 1079 |
+
"learning_rate": 4.681263858093127e-05,
|
| 1080 |
+
"loss": 1.30006742477417,
|
| 1081 |
+
"step": 153
|
| 1082 |
+
},
|
| 1083 |
+
{
|
| 1084 |
+
"epoch": 0.08786193125089145,
|
| 1085 |
+
"grad_norm": 1.296875,
|
| 1086 |
+
"learning_rate": 4.678492239467849e-05,
|
| 1087 |
+
"loss": 1.3325148820877075,
|
| 1088 |
+
"step": 154
|
| 1089 |
+
},
|
| 1090 |
+
{
|
| 1091 |
+
"epoch": 0.08843246327200115,
|
| 1092 |
+
"grad_norm": 1.3125,
|
| 1093 |
+
"learning_rate": 4.675720620842573e-05,
|
| 1094 |
+
"loss": 1.2306278944015503,
|
| 1095 |
+
"step": 155
|
| 1096 |
+
},
|
| 1097 |
+
{
|
| 1098 |
+
"epoch": 0.08900299529311083,
|
| 1099 |
+
"grad_norm": 1.234375,
|
| 1100 |
+
"learning_rate": 4.672949002217295e-05,
|
| 1101 |
+
"loss": 1.3476486206054688,
|
| 1102 |
+
"step": 156
|
| 1103 |
+
},
|
| 1104 |
+
{
|
| 1105 |
+
"epoch": 0.08957352731422051,
|
| 1106 |
+
"grad_norm": 1.1953125,
|
| 1107 |
+
"learning_rate": 4.670177383592018e-05,
|
| 1108 |
+
"loss": 1.2401833534240723,
|
| 1109 |
+
"step": 157
|
| 1110 |
+
},
|
| 1111 |
+
{
|
| 1112 |
+
"epoch": 0.09014405933533019,
|
| 1113 |
+
"grad_norm": 1.296875,
|
| 1114 |
+
"learning_rate": 4.667405764966741e-05,
|
| 1115 |
+
"loss": 1.3140380382537842,
|
| 1116 |
+
"step": 158
|
| 1117 |
+
},
|
| 1118 |
+
{
|
| 1119 |
+
"epoch": 0.09071459135643987,
|
| 1120 |
+
"grad_norm": 1.1875,
|
| 1121 |
+
"learning_rate": 4.664634146341464e-05,
|
| 1122 |
+
"loss": 1.29231595993042,
|
| 1123 |
+
"step": 159
|
| 1124 |
+
},
|
| 1125 |
+
{
|
| 1126 |
+
"epoch": 0.09128512337754957,
|
| 1127 |
+
"grad_norm": 1.15625,
|
| 1128 |
+
"learning_rate": 4.661862527716186e-05,
|
| 1129 |
+
"loss": 1.2908031940460205,
|
| 1130 |
+
"step": 160
|
| 1131 |
+
},
|
| 1132 |
+
{
|
| 1133 |
+
"epoch": 0.09185565539865925,
|
| 1134 |
+
"grad_norm": 1.140625,
|
| 1135 |
+
"learning_rate": 4.659090909090909e-05,
|
| 1136 |
+
"loss": 1.259028434753418,
|
| 1137 |
+
"step": 161
|
| 1138 |
+
},
|
| 1139 |
+
{
|
| 1140 |
+
"epoch": 0.09242618741976893,
|
| 1141 |
+
"grad_norm": 1.203125,
|
| 1142 |
+
"learning_rate": 4.656319290465632e-05,
|
| 1143 |
+
"loss": 1.2758322954177856,
|
| 1144 |
+
"step": 162
|
| 1145 |
+
},
|
| 1146 |
+
{
|
| 1147 |
+
"epoch": 0.09299671944087862,
|
| 1148 |
+
"grad_norm": 1.1640625,
|
| 1149 |
+
"learning_rate": 4.653547671840355e-05,
|
| 1150 |
+
"loss": 1.2392590045928955,
|
| 1151 |
+
"step": 163
|
| 1152 |
+
},
|
| 1153 |
+
{
|
| 1154 |
+
"epoch": 0.0935672514619883,
|
| 1155 |
+
"grad_norm": 1.2421875,
|
| 1156 |
+
"learning_rate": 4.650776053215078e-05,
|
| 1157 |
+
"loss": 1.3232059478759766,
|
| 1158 |
+
"step": 164
|
| 1159 |
+
},
|
| 1160 |
+
{
|
| 1161 |
+
"epoch": 0.094137783483098,
|
| 1162 |
+
"grad_norm": 1.1328125,
|
| 1163 |
+
"learning_rate": 4.6480044345898e-05,
|
| 1164 |
+
"loss": 1.3052716255187988,
|
| 1165 |
+
"step": 165
|
| 1166 |
+
},
|
| 1167 |
+
{
|
| 1168 |
+
"epoch": 0.09470831550420768,
|
| 1169 |
+
"grad_norm": 1.15625,
|
| 1170 |
+
"learning_rate": 4.645232815964524e-05,
|
| 1171 |
+
"loss": 1.2643868923187256,
|
| 1172 |
+
"step": 166
|
| 1173 |
+
},
|
| 1174 |
+
{
|
| 1175 |
+
"epoch": 0.09527884752531736,
|
| 1176 |
+
"grad_norm": 1.1796875,
|
| 1177 |
+
"learning_rate": 4.642461197339246e-05,
|
| 1178 |
+
"loss": 1.3158135414123535,
|
| 1179 |
+
"step": 167
|
| 1180 |
+
},
|
| 1181 |
+
{
|
| 1182 |
+
"epoch": 0.09584937954642704,
|
| 1183 |
+
"grad_norm": 1.1484375,
|
| 1184 |
+
"learning_rate": 4.639689578713969e-05,
|
| 1185 |
+
"loss": 1.2975637912750244,
|
| 1186 |
+
"step": 168
|
| 1187 |
+
},
|
| 1188 |
+
{
|
| 1189 |
+
"epoch": 0.09641991156753672,
|
| 1190 |
+
"grad_norm": 1.09375,
|
| 1191 |
+
"learning_rate": 4.636917960088692e-05,
|
| 1192 |
+
"loss": 1.202270269393921,
|
| 1193 |
+
"step": 169
|
| 1194 |
+
},
|
| 1195 |
+
{
|
| 1196 |
+
"epoch": 0.09699044358864642,
|
| 1197 |
+
"grad_norm": 1.1015625,
|
| 1198 |
+
"learning_rate": 4.634146341463415e-05,
|
| 1199 |
+
"loss": 1.1989184617996216,
|
| 1200 |
+
"step": 170
|
| 1201 |
+
},
|
| 1202 |
+
{
|
| 1203 |
+
"epoch": 0.0975609756097561,
|
| 1204 |
+
"grad_norm": 1.15625,
|
| 1205 |
+
"learning_rate": 4.631374722838138e-05,
|
| 1206 |
+
"loss": 1.325451374053955,
|
| 1207 |
+
"step": 171
|
| 1208 |
+
},
|
| 1209 |
+
{
|
| 1210 |
+
"epoch": 0.09813150763086578,
|
| 1211 |
+
"grad_norm": 1.1484375,
|
| 1212 |
+
"learning_rate": 4.628603104212861e-05,
|
| 1213 |
+
"loss": 1.3150224685668945,
|
| 1214 |
+
"step": 172
|
| 1215 |
+
},
|
| 1216 |
+
{
|
| 1217 |
+
"epoch": 0.09870203965197547,
|
| 1218 |
+
"grad_norm": 1.1484375,
|
| 1219 |
+
"learning_rate": 4.625831485587583e-05,
|
| 1220 |
+
"loss": 1.2864487171173096,
|
| 1221 |
+
"step": 173
|
| 1222 |
+
},
|
| 1223 |
+
{
|
| 1224 |
+
"epoch": 0.09927257167308515,
|
| 1225 |
+
"grad_norm": 1.140625,
|
| 1226 |
+
"learning_rate": 4.623059866962306e-05,
|
| 1227 |
+
"loss": 1.3033939599990845,
|
| 1228 |
+
"step": 174
|
| 1229 |
+
},
|
| 1230 |
+
{
|
| 1231 |
+
"epoch": 0.09984310369419484,
|
| 1232 |
+
"grad_norm": 1.109375,
|
| 1233 |
+
"learning_rate": 4.620288248337029e-05,
|
| 1234 |
+
"loss": 1.2654147148132324,
|
| 1235 |
+
"step": 175
|
| 1236 |
+
},
|
| 1237 |
+
{
|
| 1238 |
+
"epoch": 0.10041363571530453,
|
| 1239 |
+
"grad_norm": 1.125,
|
| 1240 |
+
"learning_rate": 4.617516629711752e-05,
|
| 1241 |
+
"loss": 1.2905241250991821,
|
| 1242 |
+
"step": 176
|
| 1243 |
+
},
|
| 1244 |
+
{
|
| 1245 |
+
"epoch": 0.10098416773641421,
|
| 1246 |
+
"grad_norm": 1.1484375,
|
| 1247 |
+
"learning_rate": 4.614745011086475e-05,
|
| 1248 |
+
"loss": 1.2881019115447998,
|
| 1249 |
+
"step": 177
|
| 1250 |
+
},
|
| 1251 |
+
{
|
| 1252 |
+
"epoch": 0.10155469975752389,
|
| 1253 |
+
"grad_norm": 1.15625,
|
| 1254 |
+
"learning_rate": 4.611973392461197e-05,
|
| 1255 |
+
"loss": 1.3300973176956177,
|
| 1256 |
+
"step": 178
|
| 1257 |
+
},
|
| 1258 |
+
{
|
| 1259 |
+
"epoch": 0.10212523177863357,
|
| 1260 |
+
"grad_norm": 1.1484375,
|
| 1261 |
+
"learning_rate": 4.609201773835921e-05,
|
| 1262 |
+
"loss": 1.3166918754577637,
|
| 1263 |
+
"step": 179
|
| 1264 |
+
},
|
| 1265 |
+
{
|
| 1266 |
+
"epoch": 0.10269576379974327,
|
| 1267 |
+
"grad_norm": 1.078125,
|
| 1268 |
+
"learning_rate": 4.606430155210643e-05,
|
| 1269 |
+
"loss": 1.2149487733840942,
|
| 1270 |
+
"step": 180
|
| 1271 |
+
},
|
| 1272 |
+
{
|
| 1273 |
+
"epoch": 0.10326629582085295,
|
| 1274 |
+
"grad_norm": 1.1171875,
|
| 1275 |
+
"learning_rate": 4.603658536585366e-05,
|
| 1276 |
+
"loss": 1.284995198249817,
|
| 1277 |
+
"step": 181
|
| 1278 |
+
},
|
| 1279 |
+
{
|
| 1280 |
+
"epoch": 0.10383682784196263,
|
| 1281 |
+
"grad_norm": 1.171875,
|
| 1282 |
+
"learning_rate": 4.600886917960089e-05,
|
| 1283 |
+
"loss": 1.3197823762893677,
|
| 1284 |
+
"step": 182
|
| 1285 |
+
},
|
| 1286 |
+
{
|
| 1287 |
+
"epoch": 0.10440735986307231,
|
| 1288 |
+
"grad_norm": 1.09375,
|
| 1289 |
+
"learning_rate": 4.598115299334812e-05,
|
| 1290 |
+
"loss": 1.2414249181747437,
|
| 1291 |
+
"step": 183
|
| 1292 |
+
},
|
| 1293 |
+
{
|
| 1294 |
+
"epoch": 0.104977891884182,
|
| 1295 |
+
"grad_norm": 1.1953125,
|
| 1296 |
+
"learning_rate": 4.595343680709535e-05,
|
| 1297 |
+
"loss": 1.2936391830444336,
|
| 1298 |
+
"step": 184
|
| 1299 |
+
},
|
| 1300 |
+
{
|
| 1301 |
+
"epoch": 0.10554842390529169,
|
| 1302 |
+
"grad_norm": 1.203125,
|
| 1303 |
+
"learning_rate": 4.592572062084257e-05,
|
| 1304 |
+
"loss": 1.2889211177825928,
|
| 1305 |
+
"step": 185
|
| 1306 |
+
},
|
| 1307 |
+
{
|
| 1308 |
+
"epoch": 0.10611895592640137,
|
| 1309 |
+
"grad_norm": 1.1640625,
|
| 1310 |
+
"learning_rate": 4.58980044345898e-05,
|
| 1311 |
+
"loss": 1.2958948612213135,
|
| 1312 |
+
"step": 186
|
| 1313 |
+
},
|
| 1314 |
+
{
|
| 1315 |
+
"epoch": 0.10668948794751106,
|
| 1316 |
+
"grad_norm": 1.15625,
|
| 1317 |
+
"learning_rate": 4.587028824833703e-05,
|
| 1318 |
+
"loss": 1.3174210786819458,
|
| 1319 |
+
"step": 187
|
| 1320 |
+
},
|
| 1321 |
+
{
|
| 1322 |
+
"epoch": 0.10726001996862074,
|
| 1323 |
+
"grad_norm": 1.09375,
|
| 1324 |
+
"learning_rate": 4.584257206208426e-05,
|
| 1325 |
+
"loss": 1.3083107471466064,
|
| 1326 |
+
"step": 188
|
| 1327 |
+
},
|
| 1328 |
+
{
|
| 1329 |
+
"epoch": 0.10783055198973042,
|
| 1330 |
+
"grad_norm": 1.1015625,
|
| 1331 |
+
"learning_rate": 4.581485587583149e-05,
|
| 1332 |
+
"loss": 1.2460663318634033,
|
| 1333 |
+
"step": 189
|
| 1334 |
+
},
|
| 1335 |
+
{
|
| 1336 |
+
"epoch": 0.10840108401084012,
|
| 1337 |
+
"grad_norm": 1.109375,
|
| 1338 |
+
"learning_rate": 4.578713968957872e-05,
|
| 1339 |
+
"loss": 1.262696623802185,
|
| 1340 |
+
"step": 190
|
| 1341 |
+
},
|
| 1342 |
+
{
|
| 1343 |
+
"epoch": 0.1089716160319498,
|
| 1344 |
+
"grad_norm": 1.1796875,
|
| 1345 |
+
"learning_rate": 4.575942350332594e-05,
|
| 1346 |
+
"loss": 1.290346384048462,
|
| 1347 |
+
"step": 191
|
| 1348 |
+
},
|
| 1349 |
+
{
|
| 1350 |
+
"epoch": 0.10954214805305948,
|
| 1351 |
+
"grad_norm": 1.15625,
|
| 1352 |
+
"learning_rate": 4.573170731707318e-05,
|
| 1353 |
+
"loss": 1.2630096673965454,
|
| 1354 |
+
"step": 192
|
| 1355 |
+
},
|
| 1356 |
+
{
|
| 1357 |
+
"epoch": 0.11011268007416916,
|
| 1358 |
+
"grad_norm": 1.1015625,
|
| 1359 |
+
"learning_rate": 4.57039911308204e-05,
|
| 1360 |
+
"loss": 1.2521231174468994,
|
| 1361 |
+
"step": 193
|
| 1362 |
+
},
|
| 1363 |
+
{
|
| 1364 |
+
"epoch": 0.11068321209527884,
|
| 1365 |
+
"grad_norm": 1.1484375,
|
| 1366 |
+
"learning_rate": 4.567627494456763e-05,
|
| 1367 |
+
"loss": 1.2671630382537842,
|
| 1368 |
+
"step": 194
|
| 1369 |
+
},
|
| 1370 |
+
{
|
| 1371 |
+
"epoch": 0.11125374411638853,
|
| 1372 |
+
"grad_norm": 1.171875,
|
| 1373 |
+
"learning_rate": 4.564855875831486e-05,
|
| 1374 |
+
"loss": 1.3561689853668213,
|
| 1375 |
+
"step": 195
|
| 1376 |
+
},
|
| 1377 |
+
{
|
| 1378 |
+
"epoch": 0.11182427613749822,
|
| 1379 |
+
"grad_norm": 1.1484375,
|
| 1380 |
+
"learning_rate": 4.562084257206209e-05,
|
| 1381 |
+
"loss": 1.2499645948410034,
|
| 1382 |
+
"step": 196
|
| 1383 |
+
},
|
| 1384 |
+
{
|
| 1385 |
+
"epoch": 0.1123948081586079,
|
| 1386 |
+
"grad_norm": 1.15625,
|
| 1387 |
+
"learning_rate": 4.559312638580932e-05,
|
| 1388 |
+
"loss": 1.2348875999450684,
|
| 1389 |
+
"step": 197
|
| 1390 |
+
},
|
| 1391 |
+
{
|
| 1392 |
+
"epoch": 0.11296534017971759,
|
| 1393 |
+
"grad_norm": 1.1640625,
|
| 1394 |
+
"learning_rate": 4.556541019955654e-05,
|
| 1395 |
+
"loss": 1.322629690170288,
|
| 1396 |
+
"step": 198
|
| 1397 |
+
},
|
| 1398 |
+
{
|
| 1399 |
+
"epoch": 0.11353587220082727,
|
| 1400 |
+
"grad_norm": 1.171875,
|
| 1401 |
+
"learning_rate": 4.553769401330377e-05,
|
| 1402 |
+
"loss": 1.2846410274505615,
|
| 1403 |
+
"step": 199
|
| 1404 |
+
},
|
| 1405 |
+
{
|
| 1406 |
+
"epoch": 0.11410640422193695,
|
| 1407 |
+
"grad_norm": 1.1640625,
|
| 1408 |
+
"learning_rate": 4.5509977827051e-05,
|
| 1409 |
+
"loss": 1.311292052268982,
|
| 1410 |
+
"step": 200
|
| 1411 |
+
},
|
| 1412 |
+
{
|
| 1413 |
+
"epoch": 0.11467693624304665,
|
| 1414 |
+
"grad_norm": 1.140625,
|
| 1415 |
+
"learning_rate": 4.548226164079823e-05,
|
| 1416 |
+
"loss": 1.2933259010314941,
|
| 1417 |
+
"step": 201
|
| 1418 |
+
},
|
| 1419 |
+
{
|
| 1420 |
+
"epoch": 0.11524746826415633,
|
| 1421 |
+
"grad_norm": 1.21875,
|
| 1422 |
+
"learning_rate": 4.545454545454546e-05,
|
| 1423 |
+
"loss": 1.3615764379501343,
|
| 1424 |
+
"step": 202
|
| 1425 |
+
},
|
| 1426 |
+
{
|
| 1427 |
+
"epoch": 0.11581800028526601,
|
| 1428 |
+
"grad_norm": 1.1328125,
|
| 1429 |
+
"learning_rate": 4.542682926829269e-05,
|
| 1430 |
+
"loss": 1.187692403793335,
|
| 1431 |
+
"step": 203
|
| 1432 |
+
},
|
| 1433 |
+
{
|
| 1434 |
+
"epoch": 0.11638853230637569,
|
| 1435 |
+
"grad_norm": 1.1796875,
|
| 1436 |
+
"learning_rate": 4.539911308203991e-05,
|
| 1437 |
+
"loss": 1.2587438821792603,
|
| 1438 |
+
"step": 204
|
| 1439 |
+
},
|
| 1440 |
+
{
|
| 1441 |
+
"epoch": 0.11695906432748537,
|
| 1442 |
+
"grad_norm": 1.0546875,
|
| 1443 |
+
"learning_rate": 4.537139689578715e-05,
|
| 1444 |
+
"loss": 1.2154557704925537,
|
| 1445 |
+
"step": 205
|
| 1446 |
+
},
|
| 1447 |
+
{
|
| 1448 |
+
"epoch": 0.11752959634859507,
|
| 1449 |
+
"grad_norm": 1.09375,
|
| 1450 |
+
"learning_rate": 4.534368070953437e-05,
|
| 1451 |
+
"loss": 1.2670985460281372,
|
| 1452 |
+
"step": 206
|
| 1453 |
+
},
|
| 1454 |
+
{
|
| 1455 |
+
"epoch": 0.11810012836970475,
|
| 1456 |
+
"grad_norm": 1.171875,
|
| 1457 |
+
"learning_rate": 4.53159645232816e-05,
|
| 1458 |
+
"loss": 1.292269229888916,
|
| 1459 |
+
"step": 207
|
| 1460 |
+
},
|
| 1461 |
+
{
|
| 1462 |
+
"epoch": 0.11867066039081443,
|
| 1463 |
+
"grad_norm": 1.21875,
|
| 1464 |
+
"learning_rate": 4.528824833702883e-05,
|
| 1465 |
+
"loss": 1.2353066205978394,
|
| 1466 |
+
"step": 208
|
| 1467 |
+
},
|
| 1468 |
+
{
|
| 1469 |
+
"epoch": 0.11924119241192412,
|
| 1470 |
+
"grad_norm": 1.125,
|
| 1471 |
+
"learning_rate": 4.526053215077605e-05,
|
| 1472 |
+
"loss": 1.2745922803878784,
|
| 1473 |
+
"step": 209
|
| 1474 |
+
},
|
| 1475 |
+
{
|
| 1476 |
+
"epoch": 0.1198117244330338,
|
| 1477 |
+
"grad_norm": 1.109375,
|
| 1478 |
+
"learning_rate": 4.523281596452328e-05,
|
| 1479 |
+
"loss": 1.2637782096862793,
|
| 1480 |
+
"step": 210
|
| 1481 |
+
},
|
| 1482 |
+
{
|
| 1483 |
+
"epoch": 0.1203822564541435,
|
| 1484 |
+
"grad_norm": 1.125,
|
| 1485 |
+
"learning_rate": 4.520509977827051e-05,
|
| 1486 |
+
"loss": 1.2595422267913818,
|
| 1487 |
+
"step": 211
|
| 1488 |
+
},
|
| 1489 |
+
{
|
| 1490 |
+
"epoch": 0.12095278847525318,
|
| 1491 |
+
"grad_norm": 1.0859375,
|
| 1492 |
+
"learning_rate": 4.517738359201774e-05,
|
| 1493 |
+
"loss": 1.2515778541564941,
|
| 1494 |
+
"step": 212
|
| 1495 |
+
},
|
| 1496 |
+
{
|
| 1497 |
+
"epoch": 0.12152332049636286,
|
| 1498 |
+
"grad_norm": 1.1953125,
|
| 1499 |
+
"learning_rate": 4.514966740576497e-05,
|
| 1500 |
+
"loss": 1.2258851528167725,
|
| 1501 |
+
"step": 213
|
| 1502 |
+
},
|
| 1503 |
+
{
|
| 1504 |
+
"epoch": 0.12209385251747254,
|
| 1505 |
+
"grad_norm": 1.140625,
|
| 1506 |
+
"learning_rate": 4.51219512195122e-05,
|
| 1507 |
+
"loss": 1.2595672607421875,
|
| 1508 |
+
"step": 214
|
| 1509 |
+
},
|
| 1510 |
+
{
|
| 1511 |
+
"epoch": 0.12266438453858222,
|
| 1512 |
+
"grad_norm": 1.125,
|
| 1513 |
+
"learning_rate": 4.509423503325942e-05,
|
| 1514 |
+
"loss": 1.2574856281280518,
|
| 1515 |
+
"step": 215
|
| 1516 |
+
},
|
| 1517 |
+
{
|
| 1518 |
+
"epoch": 0.12323491655969192,
|
| 1519 |
+
"grad_norm": 1.21875,
|
| 1520 |
+
"learning_rate": 4.506651884700666e-05,
|
| 1521 |
+
"loss": 1.2860839366912842,
|
| 1522 |
+
"step": 216
|
| 1523 |
+
},
|
| 1524 |
+
{
|
| 1525 |
+
"epoch": 0.1238054485808016,
|
| 1526 |
+
"grad_norm": 1.1875,
|
| 1527 |
+
"learning_rate": 4.503880266075388e-05,
|
| 1528 |
+
"loss": 1.2748535871505737,
|
| 1529 |
+
"step": 217
|
| 1530 |
+
},
|
| 1531 |
+
{
|
| 1532 |
+
"epoch": 0.12437598060191128,
|
| 1533 |
+
"grad_norm": 1.1640625,
|
| 1534 |
+
"learning_rate": 4.501108647450111e-05,
|
| 1535 |
+
"loss": 1.2630361318588257,
|
| 1536 |
+
"step": 218
|
| 1537 |
+
},
|
| 1538 |
+
{
|
| 1539 |
+
"epoch": 0.12494651262302096,
|
| 1540 |
+
"grad_norm": 1.1328125,
|
| 1541 |
+
"learning_rate": 4.498337028824834e-05,
|
| 1542 |
+
"loss": 1.2100318670272827,
|
| 1543 |
+
"step": 219
|
| 1544 |
+
},
|
| 1545 |
+
{
|
| 1546 |
+
"epoch": 0.12551704464413066,
|
| 1547 |
+
"grad_norm": 1.1328125,
|
| 1548 |
+
"learning_rate": 4.495565410199557e-05,
|
| 1549 |
+
"loss": 1.279637098312378,
|
| 1550 |
+
"step": 220
|
| 1551 |
+
},
|
| 1552 |
+
{
|
| 1553 |
+
"epoch": 0.12608757666524034,
|
| 1554 |
+
"grad_norm": 1.09375,
|
| 1555 |
+
"learning_rate": 4.49279379157428e-05,
|
| 1556 |
+
"loss": 1.241306185722351,
|
| 1557 |
+
"step": 221
|
| 1558 |
+
},
|
| 1559 |
+
{
|
| 1560 |
+
"epoch": 0.12665810868635002,
|
| 1561 |
+
"grad_norm": 1.125,
|
| 1562 |
+
"learning_rate": 4.490022172949002e-05,
|
| 1563 |
+
"loss": 1.2467423677444458,
|
| 1564 |
+
"step": 222
|
| 1565 |
+
},
|
| 1566 |
+
{
|
| 1567 |
+
"epoch": 0.1272286407074597,
|
| 1568 |
+
"grad_norm": 1.1640625,
|
| 1569 |
+
"learning_rate": 4.487250554323725e-05,
|
| 1570 |
+
"loss": 1.2398571968078613,
|
| 1571 |
+
"step": 223
|
| 1572 |
+
},
|
| 1573 |
+
{
|
| 1574 |
+
"epoch": 0.1277991727285694,
|
| 1575 |
+
"grad_norm": 1.1171875,
|
| 1576 |
+
"learning_rate": 4.484478935698448e-05,
|
| 1577 |
+
"loss": 1.298073410987854,
|
| 1578 |
+
"step": 224
|
| 1579 |
+
},
|
| 1580 |
+
{
|
| 1581 |
+
"epoch": 0.12836970474967907,
|
| 1582 |
+
"grad_norm": 1.1640625,
|
| 1583 |
+
"learning_rate": 4.481707317073171e-05,
|
| 1584 |
+
"loss": 1.3275305032730103,
|
| 1585 |
+
"step": 225
|
| 1586 |
+
},
|
| 1587 |
+
{
|
| 1588 |
+
"epoch": 0.12894023677078875,
|
| 1589 |
+
"grad_norm": 1.0859375,
|
| 1590 |
+
"learning_rate": 4.478935698447894e-05,
|
| 1591 |
+
"loss": 1.2483649253845215,
|
| 1592 |
+
"step": 226
|
| 1593 |
+
},
|
| 1594 |
+
{
|
| 1595 |
+
"epoch": 0.12951076879189843,
|
| 1596 |
+
"grad_norm": 1.1484375,
|
| 1597 |
+
"learning_rate": 4.476164079822617e-05,
|
| 1598 |
+
"loss": 1.322462797164917,
|
| 1599 |
+
"step": 227
|
| 1600 |
+
},
|
| 1601 |
+
{
|
| 1602 |
+
"epoch": 0.13008130081300814,
|
| 1603 |
+
"grad_norm": 1.140625,
|
| 1604 |
+
"learning_rate": 4.473392461197339e-05,
|
| 1605 |
+
"loss": 1.2100863456726074,
|
| 1606 |
+
"step": 228
|
| 1607 |
+
},
|
| 1608 |
+
{
|
| 1609 |
+
"epoch": 0.13065183283411783,
|
| 1610 |
+
"grad_norm": 1.1640625,
|
| 1611 |
+
"learning_rate": 4.470620842572063e-05,
|
| 1612 |
+
"loss": 1.249301552772522,
|
| 1613 |
+
"step": 229
|
| 1614 |
+
},
|
| 1615 |
+
{
|
| 1616 |
+
"epoch": 0.1312223648552275,
|
| 1617 |
+
"grad_norm": 1.125,
|
| 1618 |
+
"learning_rate": 4.467849223946785e-05,
|
| 1619 |
+
"loss": 1.2208349704742432,
|
| 1620 |
+
"step": 230
|
| 1621 |
+
},
|
| 1622 |
+
{
|
| 1623 |
+
"epoch": 0.1317928968763372,
|
| 1624 |
+
"grad_norm": 1.171875,
|
| 1625 |
+
"learning_rate": 4.465077605321508e-05,
|
| 1626 |
+
"loss": 1.2686306238174438,
|
| 1627 |
+
"step": 231
|
| 1628 |
+
},
|
| 1629 |
+
{
|
| 1630 |
+
"epoch": 0.13236342889744687,
|
| 1631 |
+
"grad_norm": 1.1796875,
|
| 1632 |
+
"learning_rate": 4.462305986696231e-05,
|
| 1633 |
+
"loss": 1.2922316789627075,
|
| 1634 |
+
"step": 232
|
| 1635 |
+
},
|
| 1636 |
+
{
|
| 1637 |
+
"epoch": 0.13293396091855655,
|
| 1638 |
+
"grad_norm": 1.171875,
|
| 1639 |
+
"learning_rate": 4.459534368070954e-05,
|
| 1640 |
+
"loss": 1.2734718322753906,
|
| 1641 |
+
"step": 233
|
| 1642 |
+
},
|
| 1643 |
+
{
|
| 1644 |
+
"epoch": 0.13350449293966624,
|
| 1645 |
+
"grad_norm": 1.125,
|
| 1646 |
+
"learning_rate": 4.456762749445677e-05,
|
| 1647 |
+
"loss": 1.2748900651931763,
|
| 1648 |
+
"step": 234
|
| 1649 |
+
},
|
| 1650 |
+
{
|
| 1651 |
+
"epoch": 0.13407502496077592,
|
| 1652 |
+
"grad_norm": 1.125,
|
| 1653 |
+
"learning_rate": 4.453991130820399e-05,
|
| 1654 |
+
"loss": 1.2857415676116943,
|
| 1655 |
+
"step": 235
|
| 1656 |
+
},
|
| 1657 |
+
{
|
| 1658 |
+
"epoch": 0.1346455569818856,
|
| 1659 |
+
"grad_norm": 1.1640625,
|
| 1660 |
+
"learning_rate": 4.451219512195122e-05,
|
| 1661 |
+
"loss": 1.2689714431762695,
|
| 1662 |
+
"step": 236
|
| 1663 |
+
},
|
| 1664 |
+
{
|
| 1665 |
+
"epoch": 0.13521608900299528,
|
| 1666 |
+
"grad_norm": 1.1171875,
|
| 1667 |
+
"learning_rate": 4.448447893569845e-05,
|
| 1668 |
+
"loss": 1.248453140258789,
|
| 1669 |
+
"step": 237
|
| 1670 |
+
},
|
| 1671 |
+
{
|
| 1672 |
+
"epoch": 0.13578662102410496,
|
| 1673 |
+
"grad_norm": 1.1484375,
|
| 1674 |
+
"learning_rate": 4.445676274944568e-05,
|
| 1675 |
+
"loss": 1.2693870067596436,
|
| 1676 |
+
"step": 238
|
| 1677 |
+
},
|
| 1678 |
+
{
|
| 1679 |
+
"epoch": 0.13635715304521467,
|
| 1680 |
+
"grad_norm": 1.109375,
|
| 1681 |
+
"learning_rate": 4.442904656319291e-05,
|
| 1682 |
+
"loss": 1.2767329216003418,
|
| 1683 |
+
"step": 239
|
| 1684 |
+
},
|
| 1685 |
+
{
|
| 1686 |
+
"epoch": 0.13692768506632436,
|
| 1687 |
+
"grad_norm": 1.109375,
|
| 1688 |
+
"learning_rate": 4.440133037694014e-05,
|
| 1689 |
+
"loss": 1.2598170042037964,
|
| 1690 |
+
"step": 240
|
| 1691 |
+
},
|
| 1692 |
+
{
|
| 1693 |
+
"epoch": 0.13749821708743404,
|
| 1694 |
+
"grad_norm": 1.140625,
|
| 1695 |
+
"learning_rate": 4.437361419068736e-05,
|
| 1696 |
+
"loss": 1.2850111722946167,
|
| 1697 |
+
"step": 241
|
| 1698 |
+
},
|
| 1699 |
+
{
|
| 1700 |
+
"epoch": 0.13806874910854372,
|
| 1701 |
+
"grad_norm": 1.0625,
|
| 1702 |
+
"learning_rate": 4.4345898004434597e-05,
|
| 1703 |
+
"loss": 1.2005095481872559,
|
| 1704 |
+
"step": 242
|
| 1705 |
+
},
|
| 1706 |
+
{
|
| 1707 |
+
"epoch": 0.1386392811296534,
|
| 1708 |
+
"grad_norm": 1.09375,
|
| 1709 |
+
"learning_rate": 4.431818181818182e-05,
|
| 1710 |
+
"loss": 1.2896265983581543,
|
| 1711 |
+
"step": 243
|
| 1712 |
+
},
|
| 1713 |
+
{
|
| 1714 |
+
"epoch": 0.13920981315076308,
|
| 1715 |
+
"grad_norm": 1.1328125,
|
| 1716 |
+
"learning_rate": 4.429046563192905e-05,
|
| 1717 |
+
"loss": 1.3427916765213013,
|
| 1718 |
+
"step": 244
|
| 1719 |
+
},
|
| 1720 |
+
{
|
| 1721 |
+
"epoch": 0.13978034517187277,
|
| 1722 |
+
"grad_norm": 1.1484375,
|
| 1723 |
+
"learning_rate": 4.426274944567628e-05,
|
| 1724 |
+
"loss": 1.2719500064849854,
|
| 1725 |
+
"step": 245
|
| 1726 |
+
},
|
| 1727 |
+
{
|
| 1728 |
+
"epoch": 0.14035087719298245,
|
| 1729 |
+
"grad_norm": 1.1640625,
|
| 1730 |
+
"learning_rate": 4.42350332594235e-05,
|
| 1731 |
+
"loss": 1.2944797277450562,
|
| 1732 |
+
"step": 246
|
| 1733 |
+
},
|
| 1734 |
+
{
|
| 1735 |
+
"epoch": 0.14092140921409213,
|
| 1736 |
+
"grad_norm": 1.1328125,
|
| 1737 |
+
"learning_rate": 4.420731707317074e-05,
|
| 1738 |
+
"loss": 1.3022198677062988,
|
| 1739 |
+
"step": 247
|
| 1740 |
+
},
|
| 1741 |
+
{
|
| 1742 |
+
"epoch": 0.1414919412352018,
|
| 1743 |
+
"grad_norm": 1.2109375,
|
| 1744 |
+
"learning_rate": 4.417960088691796e-05,
|
| 1745 |
+
"loss": 1.286307454109192,
|
| 1746 |
+
"step": 248
|
| 1747 |
+
},
|
| 1748 |
+
{
|
| 1749 |
+
"epoch": 0.14206247325631152,
|
| 1750 |
+
"grad_norm": 1.328125,
|
| 1751 |
+
"learning_rate": 4.415188470066519e-05,
|
| 1752 |
+
"loss": 1.3540141582489014,
|
| 1753 |
+
"step": 249
|
| 1754 |
+
},
|
| 1755 |
+
{
|
| 1756 |
+
"epoch": 0.1426330052774212,
|
| 1757 |
+
"grad_norm": 1.0859375,
|
| 1758 |
+
"learning_rate": 4.412416851441242e-05,
|
| 1759 |
+
"loss": 1.2702994346618652,
|
| 1760 |
+
"step": 250
|
| 1761 |
+
},
|
| 1762 |
+
{
|
| 1763 |
+
"epoch": 0.1432035372985309,
|
| 1764 |
+
"grad_norm": 1.140625,
|
| 1765 |
+
"learning_rate": 4.409645232815965e-05,
|
| 1766 |
+
"loss": 1.2684781551361084,
|
| 1767 |
+
"step": 251
|
| 1768 |
+
},
|
| 1769 |
+
{
|
| 1770 |
+
"epoch": 0.14377406931964057,
|
| 1771 |
+
"grad_norm": 1.1484375,
|
| 1772 |
+
"learning_rate": 4.406873614190688e-05,
|
| 1773 |
+
"loss": 1.1907923221588135,
|
| 1774 |
+
"step": 252
|
| 1775 |
+
},
|
| 1776 |
+
{
|
| 1777 |
+
"epoch": 0.14434460134075025,
|
| 1778 |
+
"grad_norm": 1.1328125,
|
| 1779 |
+
"learning_rate": 4.404101995565411e-05,
|
| 1780 |
+
"loss": 1.2790608406066895,
|
| 1781 |
+
"step": 253
|
| 1782 |
+
},
|
| 1783 |
+
{
|
| 1784 |
+
"epoch": 0.14491513336185993,
|
| 1785 |
+
"grad_norm": 1.1015625,
|
| 1786 |
+
"learning_rate": 4.401330376940133e-05,
|
| 1787 |
+
"loss": 1.2878901958465576,
|
| 1788 |
+
"step": 254
|
| 1789 |
+
},
|
| 1790 |
+
{
|
| 1791 |
+
"epoch": 0.14548566538296961,
|
| 1792 |
+
"grad_norm": 1.09375,
|
| 1793 |
+
"learning_rate": 4.3985587583148566e-05,
|
| 1794 |
+
"loss": 1.2305991649627686,
|
| 1795 |
+
"step": 255
|
| 1796 |
+
},
|
| 1797 |
+
{
|
| 1798 |
+
"epoch": 0.1460561974040793,
|
| 1799 |
+
"grad_norm": 1.1796875,
|
| 1800 |
+
"learning_rate": 4.395787139689579e-05,
|
| 1801 |
+
"loss": 1.3150757551193237,
|
| 1802 |
+
"step": 256
|
| 1803 |
+
},
|
| 1804 |
+
{
|
| 1805 |
+
"epoch": 0.14662672942518898,
|
| 1806 |
+
"grad_norm": 1.0390625,
|
| 1807 |
+
"learning_rate": 4.393015521064302e-05,
|
| 1808 |
+
"loss": 1.213336706161499,
|
| 1809 |
+
"step": 257
|
| 1810 |
+
},
|
| 1811 |
+
{
|
| 1812 |
+
"epoch": 0.14719726144629866,
|
| 1813 |
+
"grad_norm": 1.0625,
|
| 1814 |
+
"learning_rate": 4.390243902439025e-05,
|
| 1815 |
+
"loss": 1.2233829498291016,
|
| 1816 |
+
"step": 258
|
| 1817 |
+
},
|
| 1818 |
+
{
|
| 1819 |
+
"epoch": 0.14776779346740837,
|
| 1820 |
+
"grad_norm": 1.078125,
|
| 1821 |
+
"learning_rate": 4.387472283813747e-05,
|
| 1822 |
+
"loss": 1.1772549152374268,
|
| 1823 |
+
"step": 259
|
| 1824 |
+
},
|
| 1825 |
+
{
|
| 1826 |
+
"epoch": 0.14833832548851805,
|
| 1827 |
+
"grad_norm": 1.171875,
|
| 1828 |
+
"learning_rate": 4.3847006651884707e-05,
|
| 1829 |
+
"loss": 1.3097314834594727,
|
| 1830 |
+
"step": 260
|
| 1831 |
+
},
|
| 1832 |
+
{
|
| 1833 |
+
"epoch": 0.14890885750962773,
|
| 1834 |
+
"grad_norm": 1.0703125,
|
| 1835 |
+
"learning_rate": 4.381929046563193e-05,
|
| 1836 |
+
"loss": 1.3049172163009644,
|
| 1837 |
+
"step": 261
|
| 1838 |
+
},
|
| 1839 |
+
{
|
| 1840 |
+
"epoch": 0.14947938953073742,
|
| 1841 |
+
"grad_norm": 1.125,
|
| 1842 |
+
"learning_rate": 4.379157427937916e-05,
|
| 1843 |
+
"loss": 1.3094444274902344,
|
| 1844 |
+
"step": 262
|
| 1845 |
+
},
|
| 1846 |
+
{
|
| 1847 |
+
"epoch": 0.1500499215518471,
|
| 1848 |
+
"grad_norm": 1.125,
|
| 1849 |
+
"learning_rate": 4.376385809312639e-05,
|
| 1850 |
+
"loss": 1.3298535346984863,
|
| 1851 |
+
"step": 263
|
| 1852 |
+
},
|
| 1853 |
+
{
|
| 1854 |
+
"epoch": 0.15062045357295678,
|
| 1855 |
+
"grad_norm": 1.0234375,
|
| 1856 |
+
"learning_rate": 4.373614190687362e-05,
|
| 1857 |
+
"loss": 1.2394543886184692,
|
| 1858 |
+
"step": 264
|
| 1859 |
+
},
|
| 1860 |
+
{
|
| 1861 |
+
"epoch": 0.15119098559406646,
|
| 1862 |
+
"grad_norm": 1.1640625,
|
| 1863 |
+
"learning_rate": 4.370842572062084e-05,
|
| 1864 |
+
"loss": 1.2180919647216797,
|
| 1865 |
+
"step": 265
|
| 1866 |
+
},
|
| 1867 |
+
{
|
| 1868 |
+
"epoch": 0.15176151761517614,
|
| 1869 |
+
"grad_norm": 1.140625,
|
| 1870 |
+
"learning_rate": 4.3680709534368077e-05,
|
| 1871 |
+
"loss": 1.2652344703674316,
|
| 1872 |
+
"step": 266
|
| 1873 |
+
},
|
| 1874 |
+
{
|
| 1875 |
+
"epoch": 0.15233204963628583,
|
| 1876 |
+
"grad_norm": 1.09375,
|
| 1877 |
+
"learning_rate": 4.36529933481153e-05,
|
| 1878 |
+
"loss": 1.2816247940063477,
|
| 1879 |
+
"step": 267
|
| 1880 |
+
},
|
| 1881 |
+
{
|
| 1882 |
+
"epoch": 0.1529025816573955,
|
| 1883 |
+
"grad_norm": 1.046875,
|
| 1884 |
+
"learning_rate": 4.3625277161862536e-05,
|
| 1885 |
+
"loss": 1.2074222564697266,
|
| 1886 |
+
"step": 268
|
| 1887 |
+
},
|
| 1888 |
+
{
|
| 1889 |
+
"epoch": 0.15347311367850522,
|
| 1890 |
+
"grad_norm": 1.0859375,
|
| 1891 |
+
"learning_rate": 4.359756097560976e-05,
|
| 1892 |
+
"loss": 1.2124351263046265,
|
| 1893 |
+
"step": 269
|
| 1894 |
+
},
|
| 1895 |
+
{
|
| 1896 |
+
"epoch": 0.1540436456996149,
|
| 1897 |
+
"grad_norm": 1.0859375,
|
| 1898 |
+
"learning_rate": 4.356984478935698e-05,
|
| 1899 |
+
"loss": 1.187751293182373,
|
| 1900 |
+
"step": 270
|
| 1901 |
+
},
|
| 1902 |
+
{
|
| 1903 |
+
"epoch": 0.15461417772072458,
|
| 1904 |
+
"grad_norm": 1.046875,
|
| 1905 |
+
"learning_rate": 4.354212860310422e-05,
|
| 1906 |
+
"loss": 1.1458532810211182,
|
| 1907 |
+
"step": 271
|
| 1908 |
+
},
|
| 1909 |
+
{
|
| 1910 |
+
"epoch": 0.15518470974183426,
|
| 1911 |
+
"grad_norm": 1.1171875,
|
| 1912 |
+
"learning_rate": 4.351441241685144e-05,
|
| 1913 |
+
"loss": 1.229477882385254,
|
| 1914 |
+
"step": 272
|
| 1915 |
+
},
|
| 1916 |
+
{
|
| 1917 |
+
"epoch": 0.15575524176294395,
|
| 1918 |
+
"grad_norm": 1.2265625,
|
| 1919 |
+
"learning_rate": 4.348669623059867e-05,
|
| 1920 |
+
"loss": 1.2863445281982422,
|
| 1921 |
+
"step": 273
|
| 1922 |
+
},
|
| 1923 |
+
{
|
| 1924 |
+
"epoch": 0.15632577378405363,
|
| 1925 |
+
"grad_norm": 1.0703125,
|
| 1926 |
+
"learning_rate": 4.34589800443459e-05,
|
| 1927 |
+
"loss": 1.226841688156128,
|
| 1928 |
+
"step": 274
|
| 1929 |
+
},
|
| 1930 |
+
{
|
| 1931 |
+
"epoch": 0.1568963058051633,
|
| 1932 |
+
"grad_norm": 1.1640625,
|
| 1933 |
+
"learning_rate": 4.343126385809313e-05,
|
| 1934 |
+
"loss": 1.2147347927093506,
|
| 1935 |
+
"step": 275
|
| 1936 |
+
},
|
| 1937 |
+
{
|
| 1938 |
+
"epoch": 0.157466837826273,
|
| 1939 |
+
"grad_norm": 1.0546875,
|
| 1940 |
+
"learning_rate": 4.340354767184036e-05,
|
| 1941 |
+
"loss": 1.2533400058746338,
|
| 1942 |
+
"step": 276
|
| 1943 |
+
},
|
| 1944 |
+
{
|
| 1945 |
+
"epoch": 0.15803736984738267,
|
| 1946 |
+
"grad_norm": 1.0859375,
|
| 1947 |
+
"learning_rate": 4.337583148558759e-05,
|
| 1948 |
+
"loss": 1.2199838161468506,
|
| 1949 |
+
"step": 277
|
| 1950 |
+
},
|
| 1951 |
+
{
|
| 1952 |
+
"epoch": 0.15860790186849236,
|
| 1953 |
+
"grad_norm": 1.0546875,
|
| 1954 |
+
"learning_rate": 4.334811529933481e-05,
|
| 1955 |
+
"loss": 1.196079969406128,
|
| 1956 |
+
"step": 278
|
| 1957 |
+
},
|
| 1958 |
+
{
|
| 1959 |
+
"epoch": 0.15917843388960207,
|
| 1960 |
+
"grad_norm": 1.140625,
|
| 1961 |
+
"learning_rate": 4.3320399113082046e-05,
|
| 1962 |
+
"loss": 1.2512052059173584,
|
| 1963 |
+
"step": 279
|
| 1964 |
+
},
|
| 1965 |
+
{
|
| 1966 |
+
"epoch": 0.15974896591071175,
|
| 1967 |
+
"grad_norm": 1.1171875,
|
| 1968 |
+
"learning_rate": 4.329268292682927e-05,
|
| 1969 |
+
"loss": 1.2729978561401367,
|
| 1970 |
+
"step": 280
|
| 1971 |
+
},
|
| 1972 |
+
{
|
| 1973 |
+
"epoch": 0.16031949793182143,
|
| 1974 |
+
"grad_norm": 1.171875,
|
| 1975 |
+
"learning_rate": 4.32649667405765e-05,
|
| 1976 |
+
"loss": 1.2414803504943848,
|
| 1977 |
+
"step": 281
|
| 1978 |
+
},
|
| 1979 |
+
{
|
| 1980 |
+
"epoch": 0.1608900299529311,
|
| 1981 |
+
"grad_norm": 1.125,
|
| 1982 |
+
"learning_rate": 4.323725055432373e-05,
|
| 1983 |
+
"loss": 1.2329685688018799,
|
| 1984 |
+
"step": 282
|
| 1985 |
+
},
|
| 1986 |
+
{
|
| 1987 |
+
"epoch": 0.1614605619740408,
|
| 1988 |
+
"grad_norm": 1.0703125,
|
| 1989 |
+
"learning_rate": 4.320953436807095e-05,
|
| 1990 |
+
"loss": 1.2458125352859497,
|
| 1991 |
+
"step": 283
|
| 1992 |
+
},
|
| 1993 |
+
{
|
| 1994 |
+
"epoch": 0.16203109399515048,
|
| 1995 |
+
"grad_norm": 1.09375,
|
| 1996 |
+
"learning_rate": 4.318181818181819e-05,
|
| 1997 |
+
"loss": 1.2762466669082642,
|
| 1998 |
+
"step": 284
|
| 1999 |
+
},
|
| 2000 |
+
{
|
| 2001 |
+
"epoch": 0.16260162601626016,
|
| 2002 |
+
"grad_norm": 1.1484375,
|
| 2003 |
+
"learning_rate": 4.315410199556541e-05,
|
| 2004 |
+
"loss": 1.2883433103561401,
|
| 2005 |
+
"step": 285
|
| 2006 |
+
},
|
| 2007 |
+
{
|
| 2008 |
+
"epoch": 0.16317215803736984,
|
| 2009 |
+
"grad_norm": 1.109375,
|
| 2010 |
+
"learning_rate": 4.312638580931264e-05,
|
| 2011 |
+
"loss": 1.261974811553955,
|
| 2012 |
+
"step": 286
|
| 2013 |
+
},
|
| 2014 |
+
{
|
| 2015 |
+
"epoch": 0.16374269005847952,
|
| 2016 |
+
"grad_norm": 1.0625,
|
| 2017 |
+
"learning_rate": 4.309866962305987e-05,
|
| 2018 |
+
"loss": 1.2657639980316162,
|
| 2019 |
+
"step": 287
|
| 2020 |
+
},
|
| 2021 |
+
{
|
| 2022 |
+
"epoch": 0.1643132220795892,
|
| 2023 |
+
"grad_norm": 1.1328125,
|
| 2024 |
+
"learning_rate": 4.30709534368071e-05,
|
| 2025 |
+
"loss": 1.295043706893921,
|
| 2026 |
+
"step": 288
|
| 2027 |
+
},
|
| 2028 |
+
{
|
| 2029 |
+
"epoch": 0.16488375410069891,
|
| 2030 |
+
"grad_norm": 1.09375,
|
| 2031 |
+
"learning_rate": 4.304323725055433e-05,
|
| 2032 |
+
"loss": 1.2336839437484741,
|
| 2033 |
+
"step": 289
|
| 2034 |
+
},
|
| 2035 |
+
{
|
| 2036 |
+
"epoch": 0.1654542861218086,
|
| 2037 |
+
"grad_norm": 1.09375,
|
| 2038 |
+
"learning_rate": 4.301552106430156e-05,
|
| 2039 |
+
"loss": 1.264127492904663,
|
| 2040 |
+
"step": 290
|
| 2041 |
+
},
|
| 2042 |
+
{
|
| 2043 |
+
"epoch": 0.16602481814291828,
|
| 2044 |
+
"grad_norm": 1.078125,
|
| 2045 |
+
"learning_rate": 4.298780487804878e-05,
|
| 2046 |
+
"loss": 1.2246544361114502,
|
| 2047 |
+
"step": 291
|
| 2048 |
+
},
|
| 2049 |
+
{
|
| 2050 |
+
"epoch": 0.16659535016402796,
|
| 2051 |
+
"grad_norm": 1.0546875,
|
| 2052 |
+
"learning_rate": 4.2960088691796016e-05,
|
| 2053 |
+
"loss": 1.2040233612060547,
|
| 2054 |
+
"step": 292
|
| 2055 |
+
},
|
| 2056 |
+
{
|
| 2057 |
+
"epoch": 0.16716588218513764,
|
| 2058 |
+
"grad_norm": 1.09375,
|
| 2059 |
+
"learning_rate": 4.293237250554324e-05,
|
| 2060 |
+
"loss": 1.2784225940704346,
|
| 2061 |
+
"step": 293
|
| 2062 |
+
},
|
| 2063 |
+
{
|
| 2064 |
+
"epoch": 0.16773641420624733,
|
| 2065 |
+
"grad_norm": 1.109375,
|
| 2066 |
+
"learning_rate": 4.290465631929047e-05,
|
| 2067 |
+
"loss": 1.3152185678482056,
|
| 2068 |
+
"step": 294
|
| 2069 |
+
},
|
| 2070 |
+
{
|
| 2071 |
+
"epoch": 0.168306946227357,
|
| 2072 |
+
"grad_norm": 1.1015625,
|
| 2073 |
+
"learning_rate": 4.28769401330377e-05,
|
| 2074 |
+
"loss": 1.2193617820739746,
|
| 2075 |
+
"step": 295
|
| 2076 |
+
},
|
| 2077 |
+
{
|
| 2078 |
+
"epoch": 0.1688774782484667,
|
| 2079 |
+
"grad_norm": 1.1015625,
|
| 2080 |
+
"learning_rate": 4.284922394678492e-05,
|
| 2081 |
+
"loss": 1.2813901901245117,
|
| 2082 |
+
"step": 296
|
| 2083 |
+
},
|
| 2084 |
+
{
|
| 2085 |
+
"epoch": 0.16944801026957637,
|
| 2086 |
+
"grad_norm": 1.09375,
|
| 2087 |
+
"learning_rate": 4.2821507760532156e-05,
|
| 2088 |
+
"loss": 1.205044150352478,
|
| 2089 |
+
"step": 297
|
| 2090 |
+
},
|
| 2091 |
+
{
|
| 2092 |
+
"epoch": 0.17001854229068605,
|
| 2093 |
+
"grad_norm": 1.1015625,
|
| 2094 |
+
"learning_rate": 4.279379157427938e-05,
|
| 2095 |
+
"loss": 1.2626889944076538,
|
| 2096 |
+
"step": 298
|
| 2097 |
+
},
|
| 2098 |
+
{
|
| 2099 |
+
"epoch": 0.17058907431179576,
|
| 2100 |
+
"grad_norm": 1.1484375,
|
| 2101 |
+
"learning_rate": 4.276607538802661e-05,
|
| 2102 |
+
"loss": 1.2680320739746094,
|
| 2103 |
+
"step": 299
|
| 2104 |
+
},
|
| 2105 |
+
{
|
| 2106 |
+
"epoch": 0.17115960633290545,
|
| 2107 |
+
"grad_norm": 1.0859375,
|
| 2108 |
+
"learning_rate": 4.273835920177384e-05,
|
| 2109 |
+
"loss": 1.2155548334121704,
|
| 2110 |
+
"step": 300
|
| 2111 |
+
},
|
| 2112 |
+
{
|
| 2113 |
+
"epoch": 0.17173013835401513,
|
| 2114 |
+
"grad_norm": 1.1875,
|
| 2115 |
+
"learning_rate": 4.271064301552107e-05,
|
| 2116 |
+
"loss": 1.2199232578277588,
|
| 2117 |
+
"step": 301
|
| 2118 |
+
},
|
| 2119 |
+
{
|
| 2120 |
+
"epoch": 0.1723006703751248,
|
| 2121 |
+
"grad_norm": 1.1171875,
|
| 2122 |
+
"learning_rate": 4.26829268292683e-05,
|
| 2123 |
+
"loss": 1.2747461795806885,
|
| 2124 |
+
"step": 302
|
| 2125 |
+
},
|
| 2126 |
+
{
|
| 2127 |
+
"epoch": 0.1728712023962345,
|
| 2128 |
+
"grad_norm": 1.0546875,
|
| 2129 |
+
"learning_rate": 4.2655210643015526e-05,
|
| 2130 |
+
"loss": 1.235656976699829,
|
| 2131 |
+
"step": 303
|
| 2132 |
+
},
|
| 2133 |
+
{
|
| 2134 |
+
"epoch": 0.17344173441734417,
|
| 2135 |
+
"grad_norm": 1.1875,
|
| 2136 |
+
"learning_rate": 4.262749445676275e-05,
|
| 2137 |
+
"loss": 1.3054769039154053,
|
| 2138 |
+
"step": 304
|
| 2139 |
+
},
|
| 2140 |
+
{
|
| 2141 |
+
"epoch": 0.17401226643845386,
|
| 2142 |
+
"grad_norm": 1.125,
|
| 2143 |
+
"learning_rate": 4.2599778270509985e-05,
|
| 2144 |
+
"loss": 1.2325561046600342,
|
| 2145 |
+
"step": 305
|
| 2146 |
+
},
|
| 2147 |
+
{
|
| 2148 |
+
"epoch": 0.17458279845956354,
|
| 2149 |
+
"grad_norm": 1.0546875,
|
| 2150 |
+
"learning_rate": 4.257206208425721e-05,
|
| 2151 |
+
"loss": 1.1963461637496948,
|
| 2152 |
+
"step": 306
|
| 2153 |
+
},
|
| 2154 |
+
{
|
| 2155 |
+
"epoch": 0.17515333048067322,
|
| 2156 |
+
"grad_norm": 1.09375,
|
| 2157 |
+
"learning_rate": 4.254434589800444e-05,
|
| 2158 |
+
"loss": 1.2029732465744019,
|
| 2159 |
+
"step": 307
|
| 2160 |
+
},
|
| 2161 |
+
{
|
| 2162 |
+
"epoch": 0.1757238625017829,
|
| 2163 |
+
"grad_norm": 1.0703125,
|
| 2164 |
+
"learning_rate": 4.251662971175167e-05,
|
| 2165 |
+
"loss": 1.289282202720642,
|
| 2166 |
+
"step": 308
|
| 2167 |
+
},
|
| 2168 |
+
{
|
| 2169 |
+
"epoch": 0.17629439452289258,
|
| 2170 |
+
"grad_norm": 1.1640625,
|
| 2171 |
+
"learning_rate": 4.248891352549889e-05,
|
| 2172 |
+
"loss": 1.2570784091949463,
|
| 2173 |
+
"step": 309
|
| 2174 |
+
},
|
| 2175 |
+
{
|
| 2176 |
+
"epoch": 0.1768649265440023,
|
| 2177 |
+
"grad_norm": 1.0703125,
|
| 2178 |
+
"learning_rate": 4.2461197339246126e-05,
|
| 2179 |
+
"loss": 1.1787132024765015,
|
| 2180 |
+
"step": 310
|
| 2181 |
+
},
|
| 2182 |
+
{
|
| 2183 |
+
"epoch": 0.17743545856511198,
|
| 2184 |
+
"grad_norm": 1.1328125,
|
| 2185 |
+
"learning_rate": 4.243348115299335e-05,
|
| 2186 |
+
"loss": 1.2079870700836182,
|
| 2187 |
+
"step": 311
|
| 2188 |
+
},
|
| 2189 |
+
{
|
| 2190 |
+
"epoch": 0.17800599058622166,
|
| 2191 |
+
"grad_norm": 1.1640625,
|
| 2192 |
+
"learning_rate": 4.240576496674058e-05,
|
| 2193 |
+
"loss": 1.2776343822479248,
|
| 2194 |
+
"step": 312
|
| 2195 |
+
},
|
| 2196 |
+
{
|
| 2197 |
+
"epoch": 0.17857652260733134,
|
| 2198 |
+
"grad_norm": 1.0625,
|
| 2199 |
+
"learning_rate": 4.237804878048781e-05,
|
| 2200 |
+
"loss": 1.1856639385223389,
|
| 2201 |
+
"step": 313
|
| 2202 |
+
},
|
| 2203 |
+
{
|
| 2204 |
+
"epoch": 0.17914705462844102,
|
| 2205 |
+
"grad_norm": 1.1328125,
|
| 2206 |
+
"learning_rate": 4.235033259423504e-05,
|
| 2207 |
+
"loss": 1.268944501876831,
|
| 2208 |
+
"step": 314
|
| 2209 |
+
},
|
| 2210 |
+
{
|
| 2211 |
+
"epoch": 0.1797175866495507,
|
| 2212 |
+
"grad_norm": 1.140625,
|
| 2213 |
+
"learning_rate": 4.2322616407982266e-05,
|
| 2214 |
+
"loss": 1.2755537033081055,
|
| 2215 |
+
"step": 315
|
| 2216 |
+
},
|
| 2217 |
+
{
|
| 2218 |
+
"epoch": 0.18028811867066039,
|
| 2219 |
+
"grad_norm": 1.0703125,
|
| 2220 |
+
"learning_rate": 4.2294900221729496e-05,
|
| 2221 |
+
"loss": 1.274179458618164,
|
| 2222 |
+
"step": 316
|
| 2223 |
+
},
|
| 2224 |
+
{
|
| 2225 |
+
"epoch": 0.18085865069177007,
|
| 2226 |
+
"grad_norm": 1.1015625,
|
| 2227 |
+
"learning_rate": 4.226718403547672e-05,
|
| 2228 |
+
"loss": 1.2530457973480225,
|
| 2229 |
+
"step": 317
|
| 2230 |
+
},
|
| 2231 |
+
{
|
| 2232 |
+
"epoch": 0.18142918271287975,
|
| 2233 |
+
"grad_norm": 1.1015625,
|
| 2234 |
+
"learning_rate": 4.2239467849223955e-05,
|
| 2235 |
+
"loss": 1.1844085454940796,
|
| 2236 |
+
"step": 318
|
| 2237 |
+
},
|
| 2238 |
+
{
|
| 2239 |
+
"epoch": 0.18199971473398943,
|
| 2240 |
+
"grad_norm": 1.1328125,
|
| 2241 |
+
"learning_rate": 4.221175166297118e-05,
|
| 2242 |
+
"loss": 1.3111554384231567,
|
| 2243 |
+
"step": 319
|
| 2244 |
+
},
|
| 2245 |
+
{
|
| 2246 |
+
"epoch": 0.18257024675509914,
|
| 2247 |
+
"grad_norm": 1.1328125,
|
| 2248 |
+
"learning_rate": 4.21840354767184e-05,
|
| 2249 |
+
"loss": 1.2178188562393188,
|
| 2250 |
+
"step": 320
|
| 2251 |
+
},
|
| 2252 |
+
{
|
| 2253 |
+
"epoch": 0.18314077877620882,
|
| 2254 |
+
"grad_norm": 1.1015625,
|
| 2255 |
+
"learning_rate": 4.2156319290465636e-05,
|
| 2256 |
+
"loss": 1.2369928359985352,
|
| 2257 |
+
"step": 321
|
| 2258 |
+
},
|
| 2259 |
+
{
|
| 2260 |
+
"epoch": 0.1837113107973185,
|
| 2261 |
+
"grad_norm": 1.0625,
|
| 2262 |
+
"learning_rate": 4.212860310421286e-05,
|
| 2263 |
+
"loss": 1.1851946115493774,
|
| 2264 |
+
"step": 322
|
| 2265 |
+
},
|
| 2266 |
+
{
|
| 2267 |
+
"epoch": 0.1842818428184282,
|
| 2268 |
+
"grad_norm": 1.09375,
|
| 2269 |
+
"learning_rate": 4.210088691796009e-05,
|
| 2270 |
+
"loss": 1.2697205543518066,
|
| 2271 |
+
"step": 323
|
| 2272 |
+
},
|
| 2273 |
+
{
|
| 2274 |
+
"epoch": 0.18485237483953787,
|
| 2275 |
+
"grad_norm": 1.078125,
|
| 2276 |
+
"learning_rate": 4.207317073170732e-05,
|
| 2277 |
+
"loss": 1.2498860359191895,
|
| 2278 |
+
"step": 324
|
| 2279 |
+
},
|
| 2280 |
+
{
|
| 2281 |
+
"epoch": 0.18542290686064755,
|
| 2282 |
+
"grad_norm": 1.109375,
|
| 2283 |
+
"learning_rate": 4.204545454545455e-05,
|
| 2284 |
+
"loss": 1.2507086992263794,
|
| 2285 |
+
"step": 325
|
| 2286 |
+
},
|
| 2287 |
+
{
|
| 2288 |
+
"epoch": 0.18599343888175723,
|
| 2289 |
+
"grad_norm": 1.03125,
|
| 2290 |
+
"learning_rate": 4.201773835920178e-05,
|
| 2291 |
+
"loss": 1.2160149812698364,
|
| 2292 |
+
"step": 326
|
| 2293 |
+
},
|
| 2294 |
+
{
|
| 2295 |
+
"epoch": 0.18656397090286692,
|
| 2296 |
+
"grad_norm": 1.078125,
|
| 2297 |
+
"learning_rate": 4.1990022172949006e-05,
|
| 2298 |
+
"loss": 1.238983392715454,
|
| 2299 |
+
"step": 327
|
| 2300 |
+
},
|
| 2301 |
+
{
|
| 2302 |
+
"epoch": 0.1871345029239766,
|
| 2303 |
+
"grad_norm": 1.03125,
|
| 2304 |
+
"learning_rate": 4.196230598669623e-05,
|
| 2305 |
+
"loss": 1.2306344509124756,
|
| 2306 |
+
"step": 328
|
| 2307 |
+
},
|
| 2308 |
+
{
|
| 2309 |
+
"epoch": 0.18770503494508628,
|
| 2310 |
+
"grad_norm": 1.1015625,
|
| 2311 |
+
"learning_rate": 4.1934589800443465e-05,
|
| 2312 |
+
"loss": 1.27529776096344,
|
| 2313 |
+
"step": 329
|
| 2314 |
+
},
|
| 2315 |
+
{
|
| 2316 |
+
"epoch": 0.188275566966196,
|
| 2317 |
+
"grad_norm": 1.0859375,
|
| 2318 |
+
"learning_rate": 4.190687361419069e-05,
|
| 2319 |
+
"loss": 1.2787272930145264,
|
| 2320 |
+
"step": 330
|
| 2321 |
+
},
|
| 2322 |
+
{
|
| 2323 |
+
"epoch": 0.18884609898730567,
|
| 2324 |
+
"grad_norm": 1.0546875,
|
| 2325 |
+
"learning_rate": 4.187915742793792e-05,
|
| 2326 |
+
"loss": 1.2454849481582642,
|
| 2327 |
+
"step": 331
|
| 2328 |
+
},
|
| 2329 |
+
{
|
| 2330 |
+
"epoch": 0.18941663100841535,
|
| 2331 |
+
"grad_norm": 1.0078125,
|
| 2332 |
+
"learning_rate": 4.185144124168515e-05,
|
| 2333 |
+
"loss": 1.2060352563858032,
|
| 2334 |
+
"step": 332
|
| 2335 |
+
},
|
| 2336 |
+
{
|
| 2337 |
+
"epoch": 0.18998716302952504,
|
| 2338 |
+
"grad_norm": 1.078125,
|
| 2339 |
+
"learning_rate": 4.182372505543237e-05,
|
| 2340 |
+
"loss": 1.2341554164886475,
|
| 2341 |
+
"step": 333
|
| 2342 |
+
},
|
| 2343 |
+
{
|
| 2344 |
+
"epoch": 0.19055769505063472,
|
| 2345 |
+
"grad_norm": 1.1015625,
|
| 2346 |
+
"learning_rate": 4.1796008869179606e-05,
|
| 2347 |
+
"loss": 1.2774791717529297,
|
| 2348 |
+
"step": 334
|
| 2349 |
+
},
|
| 2350 |
+
{
|
| 2351 |
+
"epoch": 0.1911282270717444,
|
| 2352 |
+
"grad_norm": 1.0546875,
|
| 2353 |
+
"learning_rate": 4.176829268292683e-05,
|
| 2354 |
+
"loss": 1.2547677755355835,
|
| 2355 |
+
"step": 335
|
| 2356 |
+
},
|
| 2357 |
+
{
|
| 2358 |
+
"epoch": 0.19169875909285408,
|
| 2359 |
+
"grad_norm": 1.1015625,
|
| 2360 |
+
"learning_rate": 4.174057649667406e-05,
|
| 2361 |
+
"loss": 1.286057472229004,
|
| 2362 |
+
"step": 336
|
| 2363 |
+
},
|
| 2364 |
+
{
|
| 2365 |
+
"epoch": 0.19226929111396376,
|
| 2366 |
+
"grad_norm": 1.1015625,
|
| 2367 |
+
"learning_rate": 4.171286031042129e-05,
|
| 2368 |
+
"loss": 1.2891746759414673,
|
| 2369 |
+
"step": 337
|
| 2370 |
+
},
|
| 2371 |
+
{
|
| 2372 |
+
"epoch": 0.19283982313507345,
|
| 2373 |
+
"grad_norm": 1.046875,
|
| 2374 |
+
"learning_rate": 4.168514412416852e-05,
|
| 2375 |
+
"loss": 1.2376006841659546,
|
| 2376 |
+
"step": 338
|
| 2377 |
+
},
|
| 2378 |
+
{
|
| 2379 |
+
"epoch": 0.19341035515618313,
|
| 2380 |
+
"grad_norm": 1.0703125,
|
| 2381 |
+
"learning_rate": 4.1657427937915746e-05,
|
| 2382 |
+
"loss": 1.2672202587127686,
|
| 2383 |
+
"step": 339
|
| 2384 |
+
},
|
| 2385 |
+
{
|
| 2386 |
+
"epoch": 0.19398088717729284,
|
| 2387 |
+
"grad_norm": 1.1015625,
|
| 2388 |
+
"learning_rate": 4.1629711751662976e-05,
|
| 2389 |
+
"loss": 1.2037293910980225,
|
| 2390 |
+
"step": 340
|
| 2391 |
+
},
|
| 2392 |
+
{
|
| 2393 |
+
"epoch": 0.19455141919840252,
|
| 2394 |
+
"grad_norm": 1.0859375,
|
| 2395 |
+
"learning_rate": 4.16019955654102e-05,
|
| 2396 |
+
"loss": 1.218858003616333,
|
| 2397 |
+
"step": 341
|
| 2398 |
+
},
|
| 2399 |
+
{
|
| 2400 |
+
"epoch": 0.1951219512195122,
|
| 2401 |
+
"grad_norm": 1.0859375,
|
| 2402 |
+
"learning_rate": 4.1574279379157435e-05,
|
| 2403 |
+
"loss": 1.2183986902236938,
|
| 2404 |
+
"step": 342
|
| 2405 |
+
},
|
| 2406 |
+
{
|
| 2407 |
+
"epoch": 0.19569248324062188,
|
| 2408 |
+
"grad_norm": 1.1171875,
|
| 2409 |
+
"learning_rate": 4.154656319290466e-05,
|
| 2410 |
+
"loss": 1.2573124170303345,
|
| 2411 |
+
"step": 343
|
| 2412 |
+
},
|
| 2413 |
+
{
|
| 2414 |
+
"epoch": 0.19626301526173157,
|
| 2415 |
+
"grad_norm": 1.125,
|
| 2416 |
+
"learning_rate": 4.151884700665189e-05,
|
| 2417 |
+
"loss": 1.21070396900177,
|
| 2418 |
+
"step": 344
|
| 2419 |
+
},
|
| 2420 |
+
{
|
| 2421 |
+
"epoch": 0.19683354728284125,
|
| 2422 |
+
"grad_norm": 1.09375,
|
| 2423 |
+
"learning_rate": 4.1491130820399116e-05,
|
| 2424 |
+
"loss": 1.286003589630127,
|
| 2425 |
+
"step": 345
|
| 2426 |
+
},
|
| 2427 |
+
{
|
| 2428 |
+
"epoch": 0.19740407930395093,
|
| 2429 |
+
"grad_norm": 1.1015625,
|
| 2430 |
+
"learning_rate": 4.146341463414634e-05,
|
| 2431 |
+
"loss": 1.2600152492523193,
|
| 2432 |
+
"step": 346
|
| 2433 |
+
},
|
| 2434 |
+
{
|
| 2435 |
+
"epoch": 0.1979746113250606,
|
| 2436 |
+
"grad_norm": 1.203125,
|
| 2437 |
+
"learning_rate": 4.1435698447893575e-05,
|
| 2438 |
+
"loss": 1.2338290214538574,
|
| 2439 |
+
"step": 347
|
| 2440 |
+
},
|
| 2441 |
+
{
|
| 2442 |
+
"epoch": 0.1985451433461703,
|
| 2443 |
+
"grad_norm": 1.15625,
|
| 2444 |
+
"learning_rate": 4.14079822616408e-05,
|
| 2445 |
+
"loss": 1.2722115516662598,
|
| 2446 |
+
"step": 348
|
| 2447 |
+
},
|
| 2448 |
+
{
|
| 2449 |
+
"epoch": 0.19911567536727998,
|
| 2450 |
+
"grad_norm": 1.0859375,
|
| 2451 |
+
"learning_rate": 4.138026607538803e-05,
|
| 2452 |
+
"loss": 1.1988334655761719,
|
| 2453 |
+
"step": 349
|
| 2454 |
+
},
|
| 2455 |
+
{
|
| 2456 |
+
"epoch": 0.19968620738838969,
|
| 2457 |
+
"grad_norm": 1.0859375,
|
| 2458 |
+
"learning_rate": 4.135254988913526e-05,
|
| 2459 |
+
"loss": 1.2339057922363281,
|
| 2460 |
+
"step": 350
|
| 2461 |
+
},
|
| 2462 |
+
{
|
| 2463 |
+
"epoch": 0.20025673940949937,
|
| 2464 |
+
"grad_norm": 1.125,
|
| 2465 |
+
"learning_rate": 4.1324833702882486e-05,
|
| 2466 |
+
"loss": 1.2363622188568115,
|
| 2467 |
+
"step": 351
|
| 2468 |
+
},
|
| 2469 |
+
{
|
| 2470 |
+
"epoch": 0.20082727143060905,
|
| 2471 |
+
"grad_norm": 1.0859375,
|
| 2472 |
+
"learning_rate": 4.1297117516629716e-05,
|
| 2473 |
+
"loss": 1.2658472061157227,
|
| 2474 |
+
"step": 352
|
| 2475 |
+
},
|
| 2476 |
+
{
|
| 2477 |
+
"epoch": 0.20139780345171873,
|
| 2478 |
+
"grad_norm": 1.0546875,
|
| 2479 |
+
"learning_rate": 4.1269401330376945e-05,
|
| 2480 |
+
"loss": 1.2181835174560547,
|
| 2481 |
+
"step": 353
|
| 2482 |
+
},
|
| 2483 |
+
{
|
| 2484 |
+
"epoch": 0.20196833547282841,
|
| 2485 |
+
"grad_norm": 1.1171875,
|
| 2486 |
+
"learning_rate": 4.124168514412417e-05,
|
| 2487 |
+
"loss": 1.2710312604904175,
|
| 2488 |
+
"step": 354
|
| 2489 |
+
},
|
| 2490 |
+
{
|
| 2491 |
+
"epoch": 0.2025388674939381,
|
| 2492 |
+
"grad_norm": 1.0625,
|
| 2493 |
+
"learning_rate": 4.12139689578714e-05,
|
| 2494 |
+
"loss": 1.176246166229248,
|
| 2495 |
+
"step": 355
|
| 2496 |
+
},
|
| 2497 |
+
{
|
| 2498 |
+
"epoch": 0.20310939951504778,
|
| 2499 |
+
"grad_norm": 1.0625,
|
| 2500 |
+
"learning_rate": 4.118625277161863e-05,
|
| 2501 |
+
"loss": 1.24937903881073,
|
| 2502 |
+
"step": 356
|
| 2503 |
+
},
|
| 2504 |
+
{
|
| 2505 |
+
"epoch": 0.20367993153615746,
|
| 2506 |
+
"grad_norm": 1.0703125,
|
| 2507 |
+
"learning_rate": 4.1158536585365856e-05,
|
| 2508 |
+
"loss": 1.2401498556137085,
|
| 2509 |
+
"step": 357
|
| 2510 |
+
},
|
| 2511 |
+
{
|
| 2512 |
+
"epoch": 0.20425046355726714,
|
| 2513 |
+
"grad_norm": 1.046875,
|
| 2514 |
+
"learning_rate": 4.1130820399113086e-05,
|
| 2515 |
+
"loss": 1.2015979290008545,
|
| 2516 |
+
"step": 358
|
| 2517 |
+
},
|
| 2518 |
+
{
|
| 2519 |
+
"epoch": 0.20482099557837682,
|
| 2520 |
+
"grad_norm": 1.1484375,
|
| 2521 |
+
"learning_rate": 4.110310421286031e-05,
|
| 2522 |
+
"loss": 1.2495380640029907,
|
| 2523 |
+
"step": 359
|
| 2524 |
+
},
|
| 2525 |
+
{
|
| 2526 |
+
"epoch": 0.20539152759948653,
|
| 2527 |
+
"grad_norm": 1.0703125,
|
| 2528 |
+
"learning_rate": 4.1075388026607545e-05,
|
| 2529 |
+
"loss": 1.2646973133087158,
|
| 2530 |
+
"step": 360
|
| 2531 |
+
},
|
| 2532 |
+
{
|
| 2533 |
+
"epoch": 0.20596205962059622,
|
| 2534 |
+
"grad_norm": 1.046875,
|
| 2535 |
+
"learning_rate": 4.104767184035477e-05,
|
| 2536 |
+
"loss": 1.2007383108139038,
|
| 2537 |
+
"step": 361
|
| 2538 |
+
},
|
| 2539 |
+
{
|
| 2540 |
+
"epoch": 0.2065325916417059,
|
| 2541 |
+
"grad_norm": 1.046875,
|
| 2542 |
+
"learning_rate": 4.1019955654102e-05,
|
| 2543 |
+
"loss": 1.226219892501831,
|
| 2544 |
+
"step": 362
|
| 2545 |
+
},
|
| 2546 |
+
{
|
| 2547 |
+
"epoch": 0.20710312366281558,
|
| 2548 |
+
"grad_norm": 1.0703125,
|
| 2549 |
+
"learning_rate": 4.0992239467849226e-05,
|
| 2550 |
+
"loss": 1.306444525718689,
|
| 2551 |
+
"step": 363
|
| 2552 |
+
},
|
| 2553 |
+
{
|
| 2554 |
+
"epoch": 0.20767365568392526,
|
| 2555 |
+
"grad_norm": 1.0625,
|
| 2556 |
+
"learning_rate": 4.0964523281596456e-05,
|
| 2557 |
+
"loss": 1.2141070365905762,
|
| 2558 |
+
"step": 364
|
| 2559 |
+
},
|
| 2560 |
+
{
|
| 2561 |
+
"epoch": 0.20824418770503494,
|
| 2562 |
+
"grad_norm": 1.0390625,
|
| 2563 |
+
"learning_rate": 4.0936807095343685e-05,
|
| 2564 |
+
"loss": 1.2149772644042969,
|
| 2565 |
+
"step": 365
|
| 2566 |
+
},
|
| 2567 |
+
{
|
| 2568 |
+
"epoch": 0.20881471972614463,
|
| 2569 |
+
"grad_norm": 1.0703125,
|
| 2570 |
+
"learning_rate": 4.0909090909090915e-05,
|
| 2571 |
+
"loss": 1.2671623229980469,
|
| 2572 |
+
"step": 366
|
| 2573 |
+
},
|
| 2574 |
+
{
|
| 2575 |
+
"epoch": 0.2093852517472543,
|
| 2576 |
+
"grad_norm": 1.0546875,
|
| 2577 |
+
"learning_rate": 4.088137472283814e-05,
|
| 2578 |
+
"loss": 1.2434954643249512,
|
| 2579 |
+
"step": 367
|
| 2580 |
+
},
|
| 2581 |
+
{
|
| 2582 |
+
"epoch": 0.209955783768364,
|
| 2583 |
+
"grad_norm": 1.03125,
|
| 2584 |
+
"learning_rate": 4.085365853658537e-05,
|
| 2585 |
+
"loss": 1.2326661348342896,
|
| 2586 |
+
"step": 368
|
| 2587 |
+
},
|
| 2588 |
+
{
|
| 2589 |
+
"epoch": 0.21052631578947367,
|
| 2590 |
+
"grad_norm": 1.0390625,
|
| 2591 |
+
"learning_rate": 4.0825942350332596e-05,
|
| 2592 |
+
"loss": 1.2969672679901123,
|
| 2593 |
+
"step": 369
|
| 2594 |
+
},
|
| 2595 |
+
{
|
| 2596 |
+
"epoch": 0.21052631578947367,
|
| 2597 |
+
"eval_loss": 1.238897681236267,
|
| 2598 |
+
"eval_runtime": 80.0789,
|
| 2599 |
+
"eval_samples_per_second": 11.938,
|
| 2600 |
+
"eval_steps_per_second": 2.985,
|
| 2601 |
+
"step": 369
|
| 2602 |
+
}
|
| 2603 |
+
],
|
| 2604 |
+
"logging_steps": 1,
|
| 2605 |
+
"max_steps": 1841,
|
| 2606 |
+
"num_input_tokens_seen": 0,
|
| 2607 |
+
"num_train_epochs": 2,
|
| 2608 |
+
"save_steps": 369,
|
| 2609 |
+
"stateful_callbacks": {
|
| 2610 |
+
"TrainerControl": {
|
| 2611 |
+
"args": {
|
| 2612 |
+
"should_epoch_stop": false,
|
| 2613 |
+
"should_evaluate": false,
|
| 2614 |
+
"should_log": false,
|
| 2615 |
+
"should_save": true,
|
| 2616 |
+
"should_training_stop": false
|
| 2617 |
+
},
|
| 2618 |
+
"attributes": {}
|
| 2619 |
+
}
|
| 2620 |
+
},
|
| 2621 |
+
"total_flos": 3.248585174072279e+17,
|
| 2622 |
+
"train_batch_size": 8,
|
| 2623 |
+
"trial_name": null,
|
| 2624 |
+
"trial_params": null
|
| 2625 |
+
}
|
last-checkpoint/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:35188d4fdad34d0de38cdcb4d62b3bed000a8a2f5a1b904b9e90f5f504891166
|
| 3 |
+
size 5713
|