Mia Fournier
commited on
Commit
•
64ddfa1
1
Parent(s):
f723a5e
Upload folder using huggingface_hub
Browse files- added_tokens.json +5 -0
- config.json +27 -0
- generation_config.json +7 -0
- merges.txt +0 -0
- model.safetensors +3 -0
- optimizer.pt +3 -0
- rng_state.pth +3 -0
- scheduler.pt +3 -0
- special_tokens_map.json +20 -0
- tokenizer.json +0 -0
- tokenizer_config.json +43 -0
- trainer_state.json +2715 -0
- training_args.bin +3 -0
- vocab.json +0 -0
added_tokens.json
ADDED
|
@@ -0,0 +1,5 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"<|endoftext|>": 151643,
|
| 3 |
+
"<|im_end|>": 151645,
|
| 4 |
+
"<|im_start|>": 151644
|
| 5 |
+
}
|
config.json
ADDED
|
@@ -0,0 +1,27 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "MangyMango/testing1",
|
| 3 |
+
"architectures": [
|
| 4 |
+
"Qwen2ForCausalLM"
|
| 5 |
+
],
|
| 6 |
+
"attention_dropout": 0.0,
|
| 7 |
+
"eos_token_id": 151643,
|
| 8 |
+
"hidden_act": "silu",
|
| 9 |
+
"hidden_size": 1536,
|
| 10 |
+
"initializer_range": 0.02,
|
| 11 |
+
"intermediate_size": 8960,
|
| 12 |
+
"max_position_embeddings": 131072,
|
| 13 |
+
"max_window_layers": 28,
|
| 14 |
+
"model_type": "qwen2",
|
| 15 |
+
"num_attention_heads": 12,
|
| 16 |
+
"num_hidden_layers": 28,
|
| 17 |
+
"num_key_value_heads": 2,
|
| 18 |
+
"rms_norm_eps": 1e-06,
|
| 19 |
+
"rope_theta": 1000000.0,
|
| 20 |
+
"sliding_window": null,
|
| 21 |
+
"tie_word_embeddings": true,
|
| 22 |
+
"torch_dtype": "bfloat16",
|
| 23 |
+
"transformers_version": "4.44.0.dev0",
|
| 24 |
+
"use_cache": false,
|
| 25 |
+
"use_sliding_window": false,
|
| 26 |
+
"vocab_size": 151936
|
| 27 |
+
}
|
generation_config.json
ADDED
|
@@ -0,0 +1,7 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"bos_token_id": 151643,
|
| 3 |
+
"do_sample": true,
|
| 4 |
+
"eos_token_id": 151643,
|
| 5 |
+
"max_new_tokens": 2048,
|
| 6 |
+
"transformers_version": "4.44.0.dev0"
|
| 7 |
+
}
|
merges.txt
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:7cc8093d8748bd1a89ef3c69a3aa8b8c0dd2ea4819c7ad0149deac76342efe61
|
| 3 |
+
size 3087467144
|
optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:1f275a7f70afd86d3fb0dfb6c51c63f1488f40d5cd4b6b4a0526024ecf6208dc
|
| 3 |
+
size 6175147196
|
rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:d245e05e72192c132e0f2edb6fdcae0c578c890f0fe912f17ec7b0bba2d38cc3
|
| 3 |
+
size 14244
|
scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:af3b6ec6e263e8506305011321cc19af20c5e04b33437bc261ef469e8b957f07
|
| 3 |
+
size 1064
|
special_tokens_map.json
ADDED
|
@@ -0,0 +1,20 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"additional_special_tokens": [
|
| 3 |
+
"<|im_start|>",
|
| 4 |
+
"<|im_end|>"
|
| 5 |
+
],
|
| 6 |
+
"eos_token": {
|
| 7 |
+
"content": "<|endoftext|>",
|
| 8 |
+
"lstrip": false,
|
| 9 |
+
"normalized": false,
|
| 10 |
+
"rstrip": false,
|
| 11 |
+
"single_word": false
|
| 12 |
+
},
|
| 13 |
+
"pad_token": {
|
| 14 |
+
"content": "<|endoftext|>",
|
| 15 |
+
"lstrip": false,
|
| 16 |
+
"normalized": false,
|
| 17 |
+
"rstrip": false,
|
| 18 |
+
"single_word": false
|
| 19 |
+
}
|
| 20 |
+
}
|
tokenizer.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
tokenizer_config.json
ADDED
|
@@ -0,0 +1,43 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"add_prefix_space": false,
|
| 3 |
+
"added_tokens_decoder": {
|
| 4 |
+
"151643": {
|
| 5 |
+
"content": "<|endoftext|>",
|
| 6 |
+
"lstrip": false,
|
| 7 |
+
"normalized": false,
|
| 8 |
+
"rstrip": false,
|
| 9 |
+
"single_word": false,
|
| 10 |
+
"special": true
|
| 11 |
+
},
|
| 12 |
+
"151644": {
|
| 13 |
+
"content": "<|im_start|>",
|
| 14 |
+
"lstrip": false,
|
| 15 |
+
"normalized": false,
|
| 16 |
+
"rstrip": false,
|
| 17 |
+
"single_word": false,
|
| 18 |
+
"special": true
|
| 19 |
+
},
|
| 20 |
+
"151645": {
|
| 21 |
+
"content": "<|im_end|>",
|
| 22 |
+
"lstrip": false,
|
| 23 |
+
"normalized": false,
|
| 24 |
+
"rstrip": false,
|
| 25 |
+
"single_word": false,
|
| 26 |
+
"special": true
|
| 27 |
+
}
|
| 28 |
+
},
|
| 29 |
+
"additional_special_tokens": [
|
| 30 |
+
"<|im_start|>",
|
| 31 |
+
"<|im_end|>"
|
| 32 |
+
],
|
| 33 |
+
"bos_token": null,
|
| 34 |
+
"chat_template": "{% for message in messages %}{% if message['role'] == 'user' %}{{ '### Instruction: ' + message['content'] + '\n\n' }}{% elif message['role'] == 'assistant' %}{{ '### Response: ' + message['content'] + eos_token}}{% endif %}{% endfor %}",
|
| 35 |
+
"clean_up_tokenization_spaces": false,
|
| 36 |
+
"eos_token": "<|endoftext|>",
|
| 37 |
+
"errors": "replace",
|
| 38 |
+
"model_max_length": 32768,
|
| 39 |
+
"pad_token": "<|endoftext|>",
|
| 40 |
+
"split_special_tokens": false,
|
| 41 |
+
"tokenizer_class": "Qwen2Tokenizer",
|
| 42 |
+
"unk_token": null
|
| 43 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,2715 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 1.9814460437640404,
|
| 5 |
+
"eval_steps": 47,
|
| 6 |
+
"global_step": 374,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.005324902238122972,
|
| 13 |
+
"grad_norm": 1.5,
|
| 14 |
+
"learning_rate": 1.4285714285714286e-06,
|
| 15 |
+
"loss": 1.7903,
|
| 16 |
+
"step": 1
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.005324902238122972,
|
| 20 |
+
"eval_loss": 1.775637149810791,
|
| 21 |
+
"eval_runtime": 77.1748,
|
| 22 |
+
"eval_samples_per_second": 15.29,
|
| 23 |
+
"eval_steps_per_second": 15.29,
|
| 24 |
+
"step": 1
|
| 25 |
+
},
|
| 26 |
+
{
|
| 27 |
+
"epoch": 0.010649804476245944,
|
| 28 |
+
"grad_norm": 1.484375,
|
| 29 |
+
"learning_rate": 2.8571428571428573e-06,
|
| 30 |
+
"loss": 1.761,
|
| 31 |
+
"step": 2
|
| 32 |
+
},
|
| 33 |
+
{
|
| 34 |
+
"epoch": 0.015974706714368916,
|
| 35 |
+
"grad_norm": 1.5546875,
|
| 36 |
+
"learning_rate": 4.2857142857142855e-06,
|
| 37 |
+
"loss": 1.8033,
|
| 38 |
+
"step": 3
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 0.02129960895249189,
|
| 42 |
+
"grad_norm": 1.3671875,
|
| 43 |
+
"learning_rate": 5.7142857142857145e-06,
|
| 44 |
+
"loss": 1.7335,
|
| 45 |
+
"step": 4
|
| 46 |
+
},
|
| 47 |
+
{
|
| 48 |
+
"epoch": 0.02662451119061486,
|
| 49 |
+
"grad_norm": 1.484375,
|
| 50 |
+
"learning_rate": 7.1428571428571436e-06,
|
| 51 |
+
"loss": 1.7532,
|
| 52 |
+
"step": 5
|
| 53 |
+
},
|
| 54 |
+
{
|
| 55 |
+
"epoch": 0.03194941342873783,
|
| 56 |
+
"grad_norm": 1.3203125,
|
| 57 |
+
"learning_rate": 8.571428571428571e-06,
|
| 58 |
+
"loss": 1.7546,
|
| 59 |
+
"step": 6
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 0.037274315666860805,
|
| 63 |
+
"grad_norm": 1.21875,
|
| 64 |
+
"learning_rate": 1e-05,
|
| 65 |
+
"loss": 1.8068,
|
| 66 |
+
"step": 7
|
| 67 |
+
},
|
| 68 |
+
{
|
| 69 |
+
"epoch": 0.04259921790498378,
|
| 70 |
+
"grad_norm": 1.1796875,
|
| 71 |
+
"learning_rate": 1.1428571428571429e-05,
|
| 72 |
+
"loss": 1.8295,
|
| 73 |
+
"step": 8
|
| 74 |
+
},
|
| 75 |
+
{
|
| 76 |
+
"epoch": 0.04792412014310675,
|
| 77 |
+
"grad_norm": 1.078125,
|
| 78 |
+
"learning_rate": 1.2857142857142859e-05,
|
| 79 |
+
"loss": 1.7343,
|
| 80 |
+
"step": 9
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 0.05324902238122972,
|
| 84 |
+
"grad_norm": 1.25,
|
| 85 |
+
"learning_rate": 1.4285714285714287e-05,
|
| 86 |
+
"loss": 1.8318,
|
| 87 |
+
"step": 10
|
| 88 |
+
},
|
| 89 |
+
{
|
| 90 |
+
"epoch": 0.058573924619352694,
|
| 91 |
+
"grad_norm": 1.1796875,
|
| 92 |
+
"learning_rate": 1.5714285714285715e-05,
|
| 93 |
+
"loss": 1.8064,
|
| 94 |
+
"step": 11
|
| 95 |
+
},
|
| 96 |
+
{
|
| 97 |
+
"epoch": 0.06389882685747567,
|
| 98 |
+
"grad_norm": 1.1484375,
|
| 99 |
+
"learning_rate": 1.7142857142857142e-05,
|
| 100 |
+
"loss": 1.7317,
|
| 101 |
+
"step": 12
|
| 102 |
+
},
|
| 103 |
+
{
|
| 104 |
+
"epoch": 0.06922372909559864,
|
| 105 |
+
"grad_norm": 1.2578125,
|
| 106 |
+
"learning_rate": 1.8571428571428575e-05,
|
| 107 |
+
"loss": 1.8281,
|
| 108 |
+
"step": 13
|
| 109 |
+
},
|
| 110 |
+
{
|
| 111 |
+
"epoch": 0.07454863133372161,
|
| 112 |
+
"grad_norm": 1.1328125,
|
| 113 |
+
"learning_rate": 2e-05,
|
| 114 |
+
"loss": 1.6915,
|
| 115 |
+
"step": 14
|
| 116 |
+
},
|
| 117 |
+
{
|
| 118 |
+
"epoch": 0.07987353357184458,
|
| 119 |
+
"grad_norm": 1.1484375,
|
| 120 |
+
"learning_rate": 1.9999619230641714e-05,
|
| 121 |
+
"loss": 1.7509,
|
| 122 |
+
"step": 15
|
| 123 |
+
},
|
| 124 |
+
{
|
| 125 |
+
"epoch": 0.08519843580996755,
|
| 126 |
+
"grad_norm": 1.0859375,
|
| 127 |
+
"learning_rate": 1.9998476951563914e-05,
|
| 128 |
+
"loss": 1.7416,
|
| 129 |
+
"step": 16
|
| 130 |
+
},
|
| 131 |
+
{
|
| 132 |
+
"epoch": 0.09052333804809053,
|
| 133 |
+
"grad_norm": 1.0859375,
|
| 134 |
+
"learning_rate": 1.9996573249755573e-05,
|
| 135 |
+
"loss": 1.7,
|
| 136 |
+
"step": 17
|
| 137 |
+
},
|
| 138 |
+
{
|
| 139 |
+
"epoch": 0.0958482402862135,
|
| 140 |
+
"grad_norm": 1.1640625,
|
| 141 |
+
"learning_rate": 1.999390827019096e-05,
|
| 142 |
+
"loss": 1.771,
|
| 143 |
+
"step": 18
|
| 144 |
+
},
|
| 145 |
+
{
|
| 146 |
+
"epoch": 0.10117314252433647,
|
| 147 |
+
"grad_norm": 1.1171875,
|
| 148 |
+
"learning_rate": 1.999048221581858e-05,
|
| 149 |
+
"loss": 1.7097,
|
| 150 |
+
"step": 19
|
| 151 |
+
},
|
| 152 |
+
{
|
| 153 |
+
"epoch": 0.10649804476245944,
|
| 154 |
+
"grad_norm": 1.1015625,
|
| 155 |
+
"learning_rate": 1.9986295347545738e-05,
|
| 156 |
+
"loss": 1.7411,
|
| 157 |
+
"step": 20
|
| 158 |
+
},
|
| 159 |
+
{
|
| 160 |
+
"epoch": 0.11182294700058241,
|
| 161 |
+
"grad_norm": 1.046875,
|
| 162 |
+
"learning_rate": 1.998134798421867e-05,
|
| 163 |
+
"loss": 1.7664,
|
| 164 |
+
"step": 21
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 0.11714784923870539,
|
| 168 |
+
"grad_norm": 1.1171875,
|
| 169 |
+
"learning_rate": 1.9975640502598243e-05,
|
| 170 |
+
"loss": 1.7283,
|
| 171 |
+
"step": 22
|
| 172 |
+
},
|
| 173 |
+
{
|
| 174 |
+
"epoch": 0.12247275147682836,
|
| 175 |
+
"grad_norm": 1.1015625,
|
| 176 |
+
"learning_rate": 1.9969173337331283e-05,
|
| 177 |
+
"loss": 1.8261,
|
| 178 |
+
"step": 23
|
| 179 |
+
},
|
| 180 |
+
{
|
| 181 |
+
"epoch": 0.12779765371495133,
|
| 182 |
+
"grad_norm": 1.078125,
|
| 183 |
+
"learning_rate": 1.9961946980917457e-05,
|
| 184 |
+
"loss": 1.7096,
|
| 185 |
+
"step": 24
|
| 186 |
+
},
|
| 187 |
+
{
|
| 188 |
+
"epoch": 0.1331225559530743,
|
| 189 |
+
"grad_norm": 1.078125,
|
| 190 |
+
"learning_rate": 1.9953961983671792e-05,
|
| 191 |
+
"loss": 1.8025,
|
| 192 |
+
"step": 25
|
| 193 |
+
},
|
| 194 |
+
{
|
| 195 |
+
"epoch": 0.13844745819119728,
|
| 196 |
+
"grad_norm": 1.0625,
|
| 197 |
+
"learning_rate": 1.9945218953682736e-05,
|
| 198 |
+
"loss": 1.7506,
|
| 199 |
+
"step": 26
|
| 200 |
+
},
|
| 201 |
+
{
|
| 202 |
+
"epoch": 0.14377236042932023,
|
| 203 |
+
"grad_norm": 1.0546875,
|
| 204 |
+
"learning_rate": 1.9935718556765878e-05,
|
| 205 |
+
"loss": 1.7492,
|
| 206 |
+
"step": 27
|
| 207 |
+
},
|
| 208 |
+
{
|
| 209 |
+
"epoch": 0.14909726266744322,
|
| 210 |
+
"grad_norm": 1.078125,
|
| 211 |
+
"learning_rate": 1.9925461516413224e-05,
|
| 212 |
+
"loss": 1.7108,
|
| 213 |
+
"step": 28
|
| 214 |
+
},
|
| 215 |
+
{
|
| 216 |
+
"epoch": 0.15442216490556618,
|
| 217 |
+
"grad_norm": 1.0546875,
|
| 218 |
+
"learning_rate": 1.9914448613738107e-05,
|
| 219 |
+
"loss": 1.6542,
|
| 220 |
+
"step": 29
|
| 221 |
+
},
|
| 222 |
+
{
|
| 223 |
+
"epoch": 0.15974706714368916,
|
| 224 |
+
"grad_norm": 1.0234375,
|
| 225 |
+
"learning_rate": 1.9902680687415704e-05,
|
| 226 |
+
"loss": 1.7548,
|
| 227 |
+
"step": 30
|
| 228 |
+
},
|
| 229 |
+
{
|
| 230 |
+
"epoch": 0.16507196938181212,
|
| 231 |
+
"grad_norm": 1.1171875,
|
| 232 |
+
"learning_rate": 1.989015863361917e-05,
|
| 233 |
+
"loss": 1.7417,
|
| 234 |
+
"step": 31
|
| 235 |
+
},
|
| 236 |
+
{
|
| 237 |
+
"epoch": 0.1703968716199351,
|
| 238 |
+
"grad_norm": 1.0546875,
|
| 239 |
+
"learning_rate": 1.9876883405951378e-05,
|
| 240 |
+
"loss": 1.6839,
|
| 241 |
+
"step": 32
|
| 242 |
+
},
|
| 243 |
+
{
|
| 244 |
+
"epoch": 0.17572177385805807,
|
| 245 |
+
"grad_norm": 1.0,
|
| 246 |
+
"learning_rate": 1.9862856015372315e-05,
|
| 247 |
+
"loss": 1.6547,
|
| 248 |
+
"step": 33
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 0.18104667609618105,
|
| 252 |
+
"grad_norm": 1.0625,
|
| 253 |
+
"learning_rate": 1.9848077530122083e-05,
|
| 254 |
+
"loss": 1.7696,
|
| 255 |
+
"step": 34
|
| 256 |
+
},
|
| 257 |
+
{
|
| 258 |
+
"epoch": 0.186371578334304,
|
| 259 |
+
"grad_norm": 1.1484375,
|
| 260 |
+
"learning_rate": 1.983254907563955e-05,
|
| 261 |
+
"loss": 1.6861,
|
| 262 |
+
"step": 35
|
| 263 |
+
},
|
| 264 |
+
{
|
| 265 |
+
"epoch": 0.191696480572427,
|
| 266 |
+
"grad_norm": 1.0703125,
|
| 267 |
+
"learning_rate": 1.9816271834476642e-05,
|
| 268 |
+
"loss": 1.7726,
|
| 269 |
+
"step": 36
|
| 270 |
+
},
|
| 271 |
+
{
|
| 272 |
+
"epoch": 0.19702138281054996,
|
| 273 |
+
"grad_norm": 1.0078125,
|
| 274 |
+
"learning_rate": 1.9799247046208297e-05,
|
| 275 |
+
"loss": 1.7046,
|
| 276 |
+
"step": 37
|
| 277 |
+
},
|
| 278 |
+
{
|
| 279 |
+
"epoch": 0.20234628504867294,
|
| 280 |
+
"grad_norm": 1.15625,
|
| 281 |
+
"learning_rate": 1.9781476007338058e-05,
|
| 282 |
+
"loss": 1.7071,
|
| 283 |
+
"step": 38
|
| 284 |
+
},
|
| 285 |
+
{
|
| 286 |
+
"epoch": 0.2076711872867959,
|
| 287 |
+
"grad_norm": 1.0859375,
|
| 288 |
+
"learning_rate": 1.9762960071199334e-05,
|
| 289 |
+
"loss": 1.8033,
|
| 290 |
+
"step": 39
|
| 291 |
+
},
|
| 292 |
+
{
|
| 293 |
+
"epoch": 0.21299608952491889,
|
| 294 |
+
"grad_norm": 1.0859375,
|
| 295 |
+
"learning_rate": 1.9743700647852356e-05,
|
| 296 |
+
"loss": 1.8173,
|
| 297 |
+
"step": 40
|
| 298 |
+
},
|
| 299 |
+
{
|
| 300 |
+
"epoch": 0.21832099176304184,
|
| 301 |
+
"grad_norm": 1.1171875,
|
| 302 |
+
"learning_rate": 1.9723699203976768e-05,
|
| 303 |
+
"loss": 1.7546,
|
| 304 |
+
"step": 41
|
| 305 |
+
},
|
| 306 |
+
{
|
| 307 |
+
"epoch": 0.22364589400116483,
|
| 308 |
+
"grad_norm": 1.09375,
|
| 309 |
+
"learning_rate": 1.9702957262759964e-05,
|
| 310 |
+
"loss": 1.7599,
|
| 311 |
+
"step": 42
|
| 312 |
+
},
|
| 313 |
+
{
|
| 314 |
+
"epoch": 0.2289707962392878,
|
| 315 |
+
"grad_norm": 1.03125,
|
| 316 |
+
"learning_rate": 1.968147640378108e-05,
|
| 317 |
+
"loss": 1.6765,
|
| 318 |
+
"step": 43
|
| 319 |
+
},
|
| 320 |
+
{
|
| 321 |
+
"epoch": 0.23429569847741077,
|
| 322 |
+
"grad_norm": 1.046875,
|
| 323 |
+
"learning_rate": 1.9659258262890683e-05,
|
| 324 |
+
"loss": 1.7651,
|
| 325 |
+
"step": 44
|
| 326 |
+
},
|
| 327 |
+
{
|
| 328 |
+
"epoch": 0.23962060071553373,
|
| 329 |
+
"grad_norm": 1.0546875,
|
| 330 |
+
"learning_rate": 1.963630453208623e-05,
|
| 331 |
+
"loss": 1.6026,
|
| 332 |
+
"step": 45
|
| 333 |
+
},
|
| 334 |
+
{
|
| 335 |
+
"epoch": 0.24494550295365672,
|
| 336 |
+
"grad_norm": 1.1015625,
|
| 337 |
+
"learning_rate": 1.961261695938319e-05,
|
| 338 |
+
"loss": 1.7446,
|
| 339 |
+
"step": 46
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 0.2502704051917797,
|
| 343 |
+
"grad_norm": 1.015625,
|
| 344 |
+
"learning_rate": 1.958819734868193e-05,
|
| 345 |
+
"loss": 1.7374,
|
| 346 |
+
"step": 47
|
| 347 |
+
},
|
| 348 |
+
{
|
| 349 |
+
"epoch": 0.2502704051917797,
|
| 350 |
+
"eval_loss": 1.7246959209442139,
|
| 351 |
+
"eval_runtime": 84.4189,
|
| 352 |
+
"eval_samples_per_second": 13.978,
|
| 353 |
+
"eval_steps_per_second": 13.978,
|
| 354 |
+
"step": 47
|
| 355 |
+
},
|
| 356 |
+
{
|
| 357 |
+
"epoch": 0.25559530742990266,
|
| 358 |
+
"grad_norm": 1.0390625,
|
| 359 |
+
"learning_rate": 1.9563047559630356e-05,
|
| 360 |
+
"loss": 1.7224,
|
| 361 |
+
"step": 48
|
| 362 |
+
},
|
| 363 |
+
{
|
| 364 |
+
"epoch": 0.26092020966802565,
|
| 365 |
+
"grad_norm": 1.0703125,
|
| 366 |
+
"learning_rate": 1.953716950748227e-05,
|
| 367 |
+
"loss": 1.7919,
|
| 368 |
+
"step": 49
|
| 369 |
+
},
|
| 370 |
+
{
|
| 371 |
+
"epoch": 0.2662451119061486,
|
| 372 |
+
"grad_norm": 1.0625,
|
| 373 |
+
"learning_rate": 1.9510565162951538e-05,
|
| 374 |
+
"loss": 1.728,
|
| 375 |
+
"step": 50
|
| 376 |
+
},
|
| 377 |
+
{
|
| 378 |
+
"epoch": 0.27157001414427157,
|
| 379 |
+
"grad_norm": 1.109375,
|
| 380 |
+
"learning_rate": 1.9483236552061996e-05,
|
| 381 |
+
"loss": 1.758,
|
| 382 |
+
"step": 51
|
| 383 |
+
},
|
| 384 |
+
{
|
| 385 |
+
"epoch": 0.27689491638239455,
|
| 386 |
+
"grad_norm": 1.0078125,
|
| 387 |
+
"learning_rate": 1.945518575599317e-05,
|
| 388 |
+
"loss": 1.6651,
|
| 389 |
+
"step": 52
|
| 390 |
+
},
|
| 391 |
+
{
|
| 392 |
+
"epoch": 0.28221981862051754,
|
| 393 |
+
"grad_norm": 1.1015625,
|
| 394 |
+
"learning_rate": 1.9426414910921785e-05,
|
| 395 |
+
"loss": 1.7097,
|
| 396 |
+
"step": 53
|
| 397 |
+
},
|
| 398 |
+
{
|
| 399 |
+
"epoch": 0.28754472085864047,
|
| 400 |
+
"grad_norm": 1.0703125,
|
| 401 |
+
"learning_rate": 1.9396926207859085e-05,
|
| 402 |
+
"loss": 1.7315,
|
| 403 |
+
"step": 54
|
| 404 |
+
},
|
| 405 |
+
{
|
| 406 |
+
"epoch": 0.29286962309676345,
|
| 407 |
+
"grad_norm": 1.140625,
|
| 408 |
+
"learning_rate": 1.9366721892483976e-05,
|
| 409 |
+
"loss": 1.7535,
|
| 410 |
+
"step": 55
|
| 411 |
+
},
|
| 412 |
+
{
|
| 413 |
+
"epoch": 0.29819452533488644,
|
| 414 |
+
"grad_norm": 1.0859375,
|
| 415 |
+
"learning_rate": 1.9335804264972018e-05,
|
| 416 |
+
"loss": 1.7806,
|
| 417 |
+
"step": 56
|
| 418 |
+
},
|
| 419 |
+
{
|
| 420 |
+
"epoch": 0.3035194275730094,
|
| 421 |
+
"grad_norm": 1.046875,
|
| 422 |
+
"learning_rate": 1.9304175679820247e-05,
|
| 423 |
+
"loss": 1.7565,
|
| 424 |
+
"step": 57
|
| 425 |
+
},
|
| 426 |
+
{
|
| 427 |
+
"epoch": 0.30884432981113236,
|
| 428 |
+
"grad_norm": 1.0234375,
|
| 429 |
+
"learning_rate": 1.9271838545667876e-05,
|
| 430 |
+
"loss": 1.7056,
|
| 431 |
+
"step": 58
|
| 432 |
+
},
|
| 433 |
+
{
|
| 434 |
+
"epoch": 0.31416923204925534,
|
| 435 |
+
"grad_norm": 0.9921875,
|
| 436 |
+
"learning_rate": 1.9238795325112867e-05,
|
| 437 |
+
"loss": 1.7266,
|
| 438 |
+
"step": 59
|
| 439 |
+
},
|
| 440 |
+
{
|
| 441 |
+
"epoch": 0.31949413428737833,
|
| 442 |
+
"grad_norm": 1.046875,
|
| 443 |
+
"learning_rate": 1.9205048534524405e-05,
|
| 444 |
+
"loss": 1.6798,
|
| 445 |
+
"step": 60
|
| 446 |
+
},
|
| 447 |
+
{
|
| 448 |
+
"epoch": 0.3248190365255013,
|
| 449 |
+
"grad_norm": 1.1328125,
|
| 450 |
+
"learning_rate": 1.917060074385124e-05,
|
| 451 |
+
"loss": 1.6847,
|
| 452 |
+
"step": 61
|
| 453 |
+
},
|
| 454 |
+
{
|
| 455 |
+
"epoch": 0.33014393876362425,
|
| 456 |
+
"grad_norm": 1.0390625,
|
| 457 |
+
"learning_rate": 1.913545457642601e-05,
|
| 458 |
+
"loss": 1.6441,
|
| 459 |
+
"step": 62
|
| 460 |
+
},
|
| 461 |
+
{
|
| 462 |
+
"epoch": 0.33546884100174723,
|
| 463 |
+
"grad_norm": 1.0234375,
|
| 464 |
+
"learning_rate": 1.9099612708765432e-05,
|
| 465 |
+
"loss": 1.6839,
|
| 466 |
+
"step": 63
|
| 467 |
+
},
|
| 468 |
+
{
|
| 469 |
+
"epoch": 0.3407937432398702,
|
| 470 |
+
"grad_norm": 1.0625,
|
| 471 |
+
"learning_rate": 1.9063077870366504e-05,
|
| 472 |
+
"loss": 1.6653,
|
| 473 |
+
"step": 64
|
| 474 |
+
},
|
| 475 |
+
{
|
| 476 |
+
"epoch": 0.3461186454779932,
|
| 477 |
+
"grad_norm": 1.0,
|
| 478 |
+
"learning_rate": 1.902585284349861e-05,
|
| 479 |
+
"loss": 1.6688,
|
| 480 |
+
"step": 65
|
| 481 |
+
},
|
| 482 |
+
{
|
| 483 |
+
"epoch": 0.35144354771611613,
|
| 484 |
+
"grad_norm": 1.046875,
|
| 485 |
+
"learning_rate": 1.8987940462991673e-05,
|
| 486 |
+
"loss": 1.6317,
|
| 487 |
+
"step": 66
|
| 488 |
+
},
|
| 489 |
+
{
|
| 490 |
+
"epoch": 0.3567684499542391,
|
| 491 |
+
"grad_norm": 0.94921875,
|
| 492 |
+
"learning_rate": 1.894934361602025e-05,
|
| 493 |
+
"loss": 1.6889,
|
| 494 |
+
"step": 67
|
| 495 |
+
},
|
| 496 |
+
{
|
| 497 |
+
"epoch": 0.3620933521923621,
|
| 498 |
+
"grad_norm": 1.1015625,
|
| 499 |
+
"learning_rate": 1.891006524188368e-05,
|
| 500 |
+
"loss": 1.7744,
|
| 501 |
+
"step": 68
|
| 502 |
+
},
|
| 503 |
+
{
|
| 504 |
+
"epoch": 0.3674182544304851,
|
| 505 |
+
"grad_norm": 1.109375,
|
| 506 |
+
"learning_rate": 1.887010833178222e-05,
|
| 507 |
+
"loss": 1.7525,
|
| 508 |
+
"step": 69
|
| 509 |
+
},
|
| 510 |
+
{
|
| 511 |
+
"epoch": 0.372743156668608,
|
| 512 |
+
"grad_norm": 1.03125,
|
| 513 |
+
"learning_rate": 1.8829475928589272e-05,
|
| 514 |
+
"loss": 1.5786,
|
| 515 |
+
"step": 70
|
| 516 |
+
},
|
| 517 |
+
{
|
| 518 |
+
"epoch": 0.378068058906731,
|
| 519 |
+
"grad_norm": 0.984375,
|
| 520 |
+
"learning_rate": 1.8788171126619653e-05,
|
| 521 |
+
"loss": 1.7715,
|
| 522 |
+
"step": 71
|
| 523 |
+
},
|
| 524 |
+
{
|
| 525 |
+
"epoch": 0.383392961144854,
|
| 526 |
+
"grad_norm": 1.0078125,
|
| 527 |
+
"learning_rate": 1.874619707139396e-05,
|
| 528 |
+
"loss": 1.6526,
|
| 529 |
+
"step": 72
|
| 530 |
+
},
|
| 531 |
+
{
|
| 532 |
+
"epoch": 0.388717863382977,
|
| 533 |
+
"grad_norm": 1.0390625,
|
| 534 |
+
"learning_rate": 1.8703556959398998e-05,
|
| 535 |
+
"loss": 1.7511,
|
| 536 |
+
"step": 73
|
| 537 |
+
},
|
| 538 |
+
{
|
| 539 |
+
"epoch": 0.3940427656210999,
|
| 540 |
+
"grad_norm": 1.046875,
|
| 541 |
+
"learning_rate": 1.866025403784439e-05,
|
| 542 |
+
"loss": 1.789,
|
| 543 |
+
"step": 74
|
| 544 |
+
},
|
| 545 |
+
{
|
| 546 |
+
"epoch": 0.3993676678592229,
|
| 547 |
+
"grad_norm": 1.078125,
|
| 548 |
+
"learning_rate": 1.861629160441526e-05,
|
| 549 |
+
"loss": 1.7705,
|
| 550 |
+
"step": 75
|
| 551 |
+
},
|
| 552 |
+
{
|
| 553 |
+
"epoch": 0.4046925700973459,
|
| 554 |
+
"grad_norm": 1.0625,
|
| 555 |
+
"learning_rate": 1.8571673007021124e-05,
|
| 556 |
+
"loss": 1.7042,
|
| 557 |
+
"step": 76
|
| 558 |
+
},
|
| 559 |
+
{
|
| 560 |
+
"epoch": 0.4100174723354688,
|
| 561 |
+
"grad_norm": 0.98828125,
|
| 562 |
+
"learning_rate": 1.8526401643540924e-05,
|
| 563 |
+
"loss": 1.6971,
|
| 564 |
+
"step": 77
|
| 565 |
+
},
|
| 566 |
+
{
|
| 567 |
+
"epoch": 0.4153423745735918,
|
| 568 |
+
"grad_norm": 0.9765625,
|
| 569 |
+
"learning_rate": 1.848048096156426e-05,
|
| 570 |
+
"loss": 1.6873,
|
| 571 |
+
"step": 78
|
| 572 |
+
},
|
| 573 |
+
{
|
| 574 |
+
"epoch": 0.4206672768117148,
|
| 575 |
+
"grad_norm": 0.953125,
|
| 576 |
+
"learning_rate": 1.843391445812886e-05,
|
| 577 |
+
"loss": 1.6937,
|
| 578 |
+
"step": 79
|
| 579 |
+
},
|
| 580 |
+
{
|
| 581 |
+
"epoch": 0.42599217904983777,
|
| 582 |
+
"grad_norm": 1.0234375,
|
| 583 |
+
"learning_rate": 1.8386705679454243e-05,
|
| 584 |
+
"loss": 1.6798,
|
| 585 |
+
"step": 80
|
| 586 |
+
},
|
| 587 |
+
{
|
| 588 |
+
"epoch": 0.4313170812879607,
|
| 589 |
+
"grad_norm": 1.0703125,
|
| 590 |
+
"learning_rate": 1.8338858220671683e-05,
|
| 591 |
+
"loss": 1.7316,
|
| 592 |
+
"step": 81
|
| 593 |
+
},
|
| 594 |
+
{
|
| 595 |
+
"epoch": 0.4366419835260837,
|
| 596 |
+
"grad_norm": 1.0390625,
|
| 597 |
+
"learning_rate": 1.8290375725550417e-05,
|
| 598 |
+
"loss": 1.6617,
|
| 599 |
+
"step": 82
|
| 600 |
+
},
|
| 601 |
+
{
|
| 602 |
+
"epoch": 0.4419668857642067,
|
| 603 |
+
"grad_norm": 1.078125,
|
| 604 |
+
"learning_rate": 1.8241261886220155e-05,
|
| 605 |
+
"loss": 1.7817,
|
| 606 |
+
"step": 83
|
| 607 |
+
},
|
| 608 |
+
{
|
| 609 |
+
"epoch": 0.44729178800232966,
|
| 610 |
+
"grad_norm": 1.046875,
|
| 611 |
+
"learning_rate": 1.819152044288992e-05,
|
| 612 |
+
"loss": 1.7649,
|
| 613 |
+
"step": 84
|
| 614 |
+
},
|
| 615 |
+
{
|
| 616 |
+
"epoch": 0.4526166902404526,
|
| 617 |
+
"grad_norm": 1.0546875,
|
| 618 |
+
"learning_rate": 1.8141155183563195e-05,
|
| 619 |
+
"loss": 1.7158,
|
| 620 |
+
"step": 85
|
| 621 |
+
},
|
| 622 |
+
{
|
| 623 |
+
"epoch": 0.4579415924785756,
|
| 624 |
+
"grad_norm": 1.1171875,
|
| 625 |
+
"learning_rate": 1.8090169943749477e-05,
|
| 626 |
+
"loss": 1.7345,
|
| 627 |
+
"step": 86
|
| 628 |
+
},
|
| 629 |
+
{
|
| 630 |
+
"epoch": 0.46326649471669856,
|
| 631 |
+
"grad_norm": 1.0234375,
|
| 632 |
+
"learning_rate": 1.8038568606172172e-05,
|
| 633 |
+
"loss": 1.6563,
|
| 634 |
+
"step": 87
|
| 635 |
+
},
|
| 636 |
+
{
|
| 637 |
+
"epoch": 0.46859139695482155,
|
| 638 |
+
"grad_norm": 1.0234375,
|
| 639 |
+
"learning_rate": 1.798635510047293e-05,
|
| 640 |
+
"loss": 1.726,
|
| 641 |
+
"step": 88
|
| 642 |
+
},
|
| 643 |
+
{
|
| 644 |
+
"epoch": 0.4739162991929445,
|
| 645 |
+
"grad_norm": 0.9921875,
|
| 646 |
+
"learning_rate": 1.7933533402912354e-05,
|
| 647 |
+
"loss": 1.6767,
|
| 648 |
+
"step": 89
|
| 649 |
+
},
|
| 650 |
+
{
|
| 651 |
+
"epoch": 0.47924120143106747,
|
| 652 |
+
"grad_norm": 1.0234375,
|
| 653 |
+
"learning_rate": 1.788010753606722e-05,
|
| 654 |
+
"loss": 1.6585,
|
| 655 |
+
"step": 90
|
| 656 |
+
},
|
| 657 |
+
{
|
| 658 |
+
"epoch": 0.48456610366919045,
|
| 659 |
+
"grad_norm": 0.9609375,
|
| 660 |
+
"learning_rate": 1.782608156852414e-05,
|
| 661 |
+
"loss": 1.6202,
|
| 662 |
+
"step": 91
|
| 663 |
+
},
|
| 664 |
+
{
|
| 665 |
+
"epoch": 0.48989100590731344,
|
| 666 |
+
"grad_norm": 1.0234375,
|
| 667 |
+
"learning_rate": 1.777145961456971e-05,
|
| 668 |
+
"loss": 1.7857,
|
| 669 |
+
"step": 92
|
| 670 |
+
},
|
| 671 |
+
{
|
| 672 |
+
"epoch": 0.49521590814543637,
|
| 673 |
+
"grad_norm": 1.0546875,
|
| 674 |
+
"learning_rate": 1.7716245833877202e-05,
|
| 675 |
+
"loss": 1.7479,
|
| 676 |
+
"step": 93
|
| 677 |
+
},
|
| 678 |
+
{
|
| 679 |
+
"epoch": 0.5005408103835594,
|
| 680 |
+
"grad_norm": 1.0625,
|
| 681 |
+
"learning_rate": 1.766044443118978e-05,
|
| 682 |
+
"loss": 1.7516,
|
| 683 |
+
"step": 94
|
| 684 |
+
},
|
| 685 |
+
{
|
| 686 |
+
"epoch": 0.5005408103835594,
|
| 687 |
+
"eval_loss": 1.7072687149047852,
|
| 688 |
+
"eval_runtime": 77.7379,
|
| 689 |
+
"eval_samples_per_second": 15.179,
|
| 690 |
+
"eval_steps_per_second": 15.179,
|
| 691 |
+
"step": 94
|
| 692 |
+
},
|
| 693 |
+
{
|
| 694 |
+
"epoch": 0.5058657126216823,
|
| 695 |
+
"grad_norm": 1.0703125,
|
| 696 |
+
"learning_rate": 1.7604059656000313e-05,
|
| 697 |
+
"loss": 1.6983,
|
| 698 |
+
"step": 95
|
| 699 |
+
},
|
| 700 |
+
{
|
| 701 |
+
"epoch": 0.5111906148598053,
|
| 702 |
+
"grad_norm": 1.078125,
|
| 703 |
+
"learning_rate": 1.7547095802227723e-05,
|
| 704 |
+
"loss": 1.7144,
|
| 705 |
+
"step": 96
|
| 706 |
+
},
|
| 707 |
+
{
|
| 708 |
+
"epoch": 0.5165155170979283,
|
| 709 |
+
"grad_norm": 1.0546875,
|
| 710 |
+
"learning_rate": 1.7489557207890025e-05,
|
| 711 |
+
"loss": 1.754,
|
| 712 |
+
"step": 97
|
| 713 |
+
},
|
| 714 |
+
{
|
| 715 |
+
"epoch": 0.5218404193360513,
|
| 716 |
+
"grad_norm": 1.03125,
|
| 717 |
+
"learning_rate": 1.7431448254773943e-05,
|
| 718 |
+
"loss": 1.7862,
|
| 719 |
+
"step": 98
|
| 720 |
+
},
|
| 721 |
+
{
|
| 722 |
+
"epoch": 0.5271653215741742,
|
| 723 |
+
"grad_norm": 1.046875,
|
| 724 |
+
"learning_rate": 1.737277336810124e-05,
|
| 725 |
+
"loss": 1.7244,
|
| 726 |
+
"step": 99
|
| 727 |
+
},
|
| 728 |
+
{
|
| 729 |
+
"epoch": 0.5324902238122972,
|
| 730 |
+
"grad_norm": 1.0546875,
|
| 731 |
+
"learning_rate": 1.7313537016191706e-05,
|
| 732 |
+
"loss": 1.7129,
|
| 733 |
+
"step": 100
|
| 734 |
+
},
|
| 735 |
+
{
|
| 736 |
+
"epoch": 0.5378151260504201,
|
| 737 |
+
"grad_norm": 1.0390625,
|
| 738 |
+
"learning_rate": 1.7253743710122877e-05,
|
| 739 |
+
"loss": 1.6302,
|
| 740 |
+
"step": 101
|
| 741 |
+
},
|
| 742 |
+
{
|
| 743 |
+
"epoch": 0.5431400282885431,
|
| 744 |
+
"grad_norm": 0.99609375,
|
| 745 |
+
"learning_rate": 1.7193398003386514e-05,
|
| 746 |
+
"loss": 1.6831,
|
| 747 |
+
"step": 102
|
| 748 |
+
},
|
| 749 |
+
{
|
| 750 |
+
"epoch": 0.5484649305266661,
|
| 751 |
+
"grad_norm": 1.0,
|
| 752 |
+
"learning_rate": 1.713250449154182e-05,
|
| 753 |
+
"loss": 1.74,
|
| 754 |
+
"step": 103
|
| 755 |
+
},
|
| 756 |
+
{
|
| 757 |
+
"epoch": 0.5537898327647891,
|
| 758 |
+
"grad_norm": 1.0,
|
| 759 |
+
"learning_rate": 1.7071067811865477e-05,
|
| 760 |
+
"loss": 1.7333,
|
| 761 |
+
"step": 104
|
| 762 |
+
},
|
| 763 |
+
{
|
| 764 |
+
"epoch": 0.5591147350029121,
|
| 765 |
+
"grad_norm": 1.0546875,
|
| 766 |
+
"learning_rate": 1.700909264299851e-05,
|
| 767 |
+
"loss": 1.7359,
|
| 768 |
+
"step": 105
|
| 769 |
+
},
|
| 770 |
+
{
|
| 771 |
+
"epoch": 0.5644396372410351,
|
| 772 |
+
"grad_norm": 0.98828125,
|
| 773 |
+
"learning_rate": 1.6946583704589973e-05,
|
| 774 |
+
"loss": 1.7432,
|
| 775 |
+
"step": 106
|
| 776 |
+
},
|
| 777 |
+
{
|
| 778 |
+
"epoch": 0.569764539479158,
|
| 779 |
+
"grad_norm": 1.0078125,
|
| 780 |
+
"learning_rate": 1.688354575693754e-05,
|
| 781 |
+
"loss": 1.6974,
|
| 782 |
+
"step": 107
|
| 783 |
+
},
|
| 784 |
+
{
|
| 785 |
+
"epoch": 0.5750894417172809,
|
| 786 |
+
"grad_norm": 1.03125,
|
| 787 |
+
"learning_rate": 1.6819983600624986e-05,
|
| 788 |
+
"loss": 1.7011,
|
| 789 |
+
"step": 108
|
| 790 |
+
},
|
| 791 |
+
{
|
| 792 |
+
"epoch": 0.5804143439554039,
|
| 793 |
+
"grad_norm": 1.0703125,
|
| 794 |
+
"learning_rate": 1.6755902076156606e-05,
|
| 795 |
+
"loss": 1.665,
|
| 796 |
+
"step": 109
|
| 797 |
+
},
|
| 798 |
+
{
|
| 799 |
+
"epoch": 0.5857392461935269,
|
| 800 |
+
"grad_norm": 1.125,
|
| 801 |
+
"learning_rate": 1.6691306063588583e-05,
|
| 802 |
+
"loss": 1.8047,
|
| 803 |
+
"step": 110
|
| 804 |
+
},
|
| 805 |
+
{
|
| 806 |
+
"epoch": 0.5910641484316499,
|
| 807 |
+
"grad_norm": 0.9921875,
|
| 808 |
+
"learning_rate": 1.6626200482157378e-05,
|
| 809 |
+
"loss": 1.6664,
|
| 810 |
+
"step": 111
|
| 811 |
+
},
|
| 812 |
+
{
|
| 813 |
+
"epoch": 0.5963890506697729,
|
| 814 |
+
"grad_norm": 1.0234375,
|
| 815 |
+
"learning_rate": 1.6560590289905074e-05,
|
| 816 |
+
"loss": 1.7291,
|
| 817 |
+
"step": 112
|
| 818 |
+
},
|
| 819 |
+
{
|
| 820 |
+
"epoch": 0.6017139529078959,
|
| 821 |
+
"grad_norm": 1.0859375,
|
| 822 |
+
"learning_rate": 1.6494480483301836e-05,
|
| 823 |
+
"loss": 1.709,
|
| 824 |
+
"step": 113
|
| 825 |
+
},
|
| 826 |
+
{
|
| 827 |
+
"epoch": 0.6070388551460189,
|
| 828 |
+
"grad_norm": 1.0234375,
|
| 829 |
+
"learning_rate": 1.6427876096865394e-05,
|
| 830 |
+
"loss": 1.7286,
|
| 831 |
+
"step": 114
|
| 832 |
+
},
|
| 833 |
+
{
|
| 834 |
+
"epoch": 0.6123637573841417,
|
| 835 |
+
"grad_norm": 1.0078125,
|
| 836 |
+
"learning_rate": 1.636078220277764e-05,
|
| 837 |
+
"loss": 1.6269,
|
| 838 |
+
"step": 115
|
| 839 |
+
},
|
| 840 |
+
{
|
| 841 |
+
"epoch": 0.6176886596222647,
|
| 842 |
+
"grad_norm": 1.0234375,
|
| 843 |
+
"learning_rate": 1.6293203910498375e-05,
|
| 844 |
+
"loss": 1.6286,
|
| 845 |
+
"step": 116
|
| 846 |
+
},
|
| 847 |
+
{
|
| 848 |
+
"epoch": 0.6230135618603877,
|
| 849 |
+
"grad_norm": 1.09375,
|
| 850 |
+
"learning_rate": 1.6225146366376198e-05,
|
| 851 |
+
"loss": 1.78,
|
| 852 |
+
"step": 117
|
| 853 |
+
},
|
| 854 |
+
{
|
| 855 |
+
"epoch": 0.6283384640985107,
|
| 856 |
+
"grad_norm": 1.078125,
|
| 857 |
+
"learning_rate": 1.6156614753256583e-05,
|
| 858 |
+
"loss": 1.7336,
|
| 859 |
+
"step": 118
|
| 860 |
+
},
|
| 861 |
+
{
|
| 862 |
+
"epoch": 0.6336633663366337,
|
| 863 |
+
"grad_norm": 1.0625,
|
| 864 |
+
"learning_rate": 1.608761429008721e-05,
|
| 865 |
+
"loss": 1.6578,
|
| 866 |
+
"step": 119
|
| 867 |
+
},
|
| 868 |
+
{
|
| 869 |
+
"epoch": 0.6389882685747567,
|
| 870 |
+
"grad_norm": 1.015625,
|
| 871 |
+
"learning_rate": 1.6018150231520486e-05,
|
| 872 |
+
"loss": 1.6736,
|
| 873 |
+
"step": 120
|
| 874 |
+
},
|
| 875 |
+
{
|
| 876 |
+
"epoch": 0.6443131708128796,
|
| 877 |
+
"grad_norm": 1.0546875,
|
| 878 |
+
"learning_rate": 1.5948227867513416e-05,
|
| 879 |
+
"loss": 1.5976,
|
| 880 |
+
"step": 121
|
| 881 |
+
},
|
| 882 |
+
{
|
| 883 |
+
"epoch": 0.6496380730510026,
|
| 884 |
+
"grad_norm": 1.0078125,
|
| 885 |
+
"learning_rate": 1.5877852522924733e-05,
|
| 886 |
+
"loss": 1.7155,
|
| 887 |
+
"step": 122
|
| 888 |
+
},
|
| 889 |
+
{
|
| 890 |
+
"epoch": 0.6549629752891255,
|
| 891 |
+
"grad_norm": 1.0078125,
|
| 892 |
+
"learning_rate": 1.5807029557109398e-05,
|
| 893 |
+
"loss": 1.6595,
|
| 894 |
+
"step": 123
|
| 895 |
+
},
|
| 896 |
+
{
|
| 897 |
+
"epoch": 0.6602878775272485,
|
| 898 |
+
"grad_norm": 1.0234375,
|
| 899 |
+
"learning_rate": 1.573576436351046e-05,
|
| 900 |
+
"loss": 1.7354,
|
| 901 |
+
"step": 124
|
| 902 |
+
},
|
| 903 |
+
{
|
| 904 |
+
"epoch": 0.6656127797653715,
|
| 905 |
+
"grad_norm": 1.0703125,
|
| 906 |
+
"learning_rate": 1.566406236924833e-05,
|
| 907 |
+
"loss": 1.7401,
|
| 908 |
+
"step": 125
|
| 909 |
+
},
|
| 910 |
+
{
|
| 911 |
+
"epoch": 0.6709376820034945,
|
| 912 |
+
"grad_norm": 1.0625,
|
| 913 |
+
"learning_rate": 1.5591929034707468e-05,
|
| 914 |
+
"loss": 1.6774,
|
| 915 |
+
"step": 126
|
| 916 |
+
},
|
| 917 |
+
{
|
| 918 |
+
"epoch": 0.6762625842416174,
|
| 919 |
+
"grad_norm": 1.0078125,
|
| 920 |
+
"learning_rate": 1.5519369853120584e-05,
|
| 921 |
+
"loss": 1.6818,
|
| 922 |
+
"step": 127
|
| 923 |
+
},
|
| 924 |
+
{
|
| 925 |
+
"epoch": 0.6815874864797404,
|
| 926 |
+
"grad_norm": 1.0390625,
|
| 927 |
+
"learning_rate": 1.5446390350150272e-05,
|
| 928 |
+
"loss": 1.598,
|
| 929 |
+
"step": 128
|
| 930 |
+
},
|
| 931 |
+
{
|
| 932 |
+
"epoch": 0.6869123887178634,
|
| 933 |
+
"grad_norm": 1.015625,
|
| 934 |
+
"learning_rate": 1.5372996083468242e-05,
|
| 935 |
+
"loss": 1.7103,
|
| 936 |
+
"step": 129
|
| 937 |
+
},
|
| 938 |
+
{
|
| 939 |
+
"epoch": 0.6922372909559864,
|
| 940 |
+
"grad_norm": 1.015625,
|
| 941 |
+
"learning_rate": 1.529919264233205e-05,
|
| 942 |
+
"loss": 1.6196,
|
| 943 |
+
"step": 130
|
| 944 |
+
},
|
| 945 |
+
{
|
| 946 |
+
"epoch": 0.6975621931941093,
|
| 947 |
+
"grad_norm": 1.046875,
|
| 948 |
+
"learning_rate": 1.5224985647159489e-05,
|
| 949 |
+
"loss": 1.618,
|
| 950 |
+
"step": 131
|
| 951 |
+
},
|
| 952 |
+
{
|
| 953 |
+
"epoch": 0.7028870954322323,
|
| 954 |
+
"grad_norm": 1.03125,
|
| 955 |
+
"learning_rate": 1.5150380749100545e-05,
|
| 956 |
+
"loss": 1.7159,
|
| 957 |
+
"step": 132
|
| 958 |
+
},
|
| 959 |
+
{
|
| 960 |
+
"epoch": 0.7082119976703553,
|
| 961 |
+
"grad_norm": 0.9921875,
|
| 962 |
+
"learning_rate": 1.5075383629607043e-05,
|
| 963 |
+
"loss": 1.6372,
|
| 964 |
+
"step": 133
|
| 965 |
+
},
|
| 966 |
+
{
|
| 967 |
+
"epoch": 0.7135368999084782,
|
| 968 |
+
"grad_norm": 1.0625,
|
| 969 |
+
"learning_rate": 1.5000000000000002e-05,
|
| 970 |
+
"loss": 1.7368,
|
| 971 |
+
"step": 134
|
| 972 |
+
},
|
| 973 |
+
{
|
| 974 |
+
"epoch": 0.7188618021466012,
|
| 975 |
+
"grad_norm": 0.97265625,
|
| 976 |
+
"learning_rate": 1.4924235601034673e-05,
|
| 977 |
+
"loss": 1.6675,
|
| 978 |
+
"step": 135
|
| 979 |
+
},
|
| 980 |
+
{
|
| 981 |
+
"epoch": 0.7241867043847242,
|
| 982 |
+
"grad_norm": 1.078125,
|
| 983 |
+
"learning_rate": 1.4848096202463373e-05,
|
| 984 |
+
"loss": 1.6651,
|
| 985 |
+
"step": 136
|
| 986 |
+
},
|
| 987 |
+
{
|
| 988 |
+
"epoch": 0.7295116066228472,
|
| 989 |
+
"grad_norm": 1.0078125,
|
| 990 |
+
"learning_rate": 1.4771587602596085e-05,
|
| 991 |
+
"loss": 1.6842,
|
| 992 |
+
"step": 137
|
| 993 |
+
},
|
| 994 |
+
{
|
| 995 |
+
"epoch": 0.7348365088609702,
|
| 996 |
+
"grad_norm": 1.015625,
|
| 997 |
+
"learning_rate": 1.469471562785891e-05,
|
| 998 |
+
"loss": 1.682,
|
| 999 |
+
"step": 138
|
| 1000 |
+
},
|
| 1001 |
+
{
|
| 1002 |
+
"epoch": 0.7401614110990931,
|
| 1003 |
+
"grad_norm": 1.03125,
|
| 1004 |
+
"learning_rate": 1.4617486132350343e-05,
|
| 1005 |
+
"loss": 1.697,
|
| 1006 |
+
"step": 139
|
| 1007 |
+
},
|
| 1008 |
+
{
|
| 1009 |
+
"epoch": 0.745486313337216,
|
| 1010 |
+
"grad_norm": 1.03125,
|
| 1011 |
+
"learning_rate": 1.4539904997395468e-05,
|
| 1012 |
+
"loss": 1.6511,
|
| 1013 |
+
"step": 140
|
| 1014 |
+
},
|
| 1015 |
+
{
|
| 1016 |
+
"epoch": 0.750811215575339,
|
| 1017 |
+
"grad_norm": 1.0625,
|
| 1018 |
+
"learning_rate": 1.4461978131098089e-05,
|
| 1019 |
+
"loss": 1.6586,
|
| 1020 |
+
"step": 141
|
| 1021 |
+
},
|
| 1022 |
+
{
|
| 1023 |
+
"epoch": 0.750811215575339,
|
| 1024 |
+
"eval_loss": 1.6974835395812988,
|
| 1025 |
+
"eval_runtime": 77.2891,
|
| 1026 |
+
"eval_samples_per_second": 15.267,
|
| 1027 |
+
"eval_steps_per_second": 15.267,
|
| 1028 |
+
"step": 141
|
| 1029 |
+
},
|
| 1030 |
+
{
|
| 1031 |
+
"epoch": 0.756136117813462,
|
| 1032 |
+
"grad_norm": 1.0625,
|
| 1033 |
+
"learning_rate": 1.4383711467890776e-05,
|
| 1034 |
+
"loss": 1.7338,
|
| 1035 |
+
"step": 142
|
| 1036 |
+
},
|
| 1037 |
+
{
|
| 1038 |
+
"epoch": 0.761461020051585,
|
| 1039 |
+
"grad_norm": 1.0234375,
|
| 1040 |
+
"learning_rate": 1.4305110968082953e-05,
|
| 1041 |
+
"loss": 1.6623,
|
| 1042 |
+
"step": 143
|
| 1043 |
+
},
|
| 1044 |
+
{
|
| 1045 |
+
"epoch": 0.766785922289708,
|
| 1046 |
+
"grad_norm": 1.03125,
|
| 1047 |
+
"learning_rate": 1.4226182617406996e-05,
|
| 1048 |
+
"loss": 1.6748,
|
| 1049 |
+
"step": 144
|
| 1050 |
+
},
|
| 1051 |
+
{
|
| 1052 |
+
"epoch": 0.772110824527831,
|
| 1053 |
+
"grad_norm": 1.0,
|
| 1054 |
+
"learning_rate": 1.4146932426562391e-05,
|
| 1055 |
+
"loss": 1.7057,
|
| 1056 |
+
"step": 145
|
| 1057 |
+
},
|
| 1058 |
+
{
|
| 1059 |
+
"epoch": 0.777435726765954,
|
| 1060 |
+
"grad_norm": 1.0234375,
|
| 1061 |
+
"learning_rate": 1.4067366430758004e-05,
|
| 1062 |
+
"loss": 1.6403,
|
| 1063 |
+
"step": 146
|
| 1064 |
+
},
|
| 1065 |
+
{
|
| 1066 |
+
"epoch": 0.7827606290040768,
|
| 1067 |
+
"grad_norm": 1.09375,
|
| 1068 |
+
"learning_rate": 1.3987490689252463e-05,
|
| 1069 |
+
"loss": 1.7262,
|
| 1070 |
+
"step": 147
|
| 1071 |
+
},
|
| 1072 |
+
{
|
| 1073 |
+
"epoch": 0.7880855312421998,
|
| 1074 |
+
"grad_norm": 1.0703125,
|
| 1075 |
+
"learning_rate": 1.3907311284892737e-05,
|
| 1076 |
+
"loss": 1.7561,
|
| 1077 |
+
"step": 148
|
| 1078 |
+
},
|
| 1079 |
+
{
|
| 1080 |
+
"epoch": 0.7934104334803228,
|
| 1081 |
+
"grad_norm": 1.109375,
|
| 1082 |
+
"learning_rate": 1.3826834323650899e-05,
|
| 1083 |
+
"loss": 1.6847,
|
| 1084 |
+
"step": 149
|
| 1085 |
+
},
|
| 1086 |
+
{
|
| 1087 |
+
"epoch": 0.7987353357184458,
|
| 1088 |
+
"grad_norm": 1.015625,
|
| 1089 |
+
"learning_rate": 1.3746065934159123e-05,
|
| 1090 |
+
"loss": 1.5979,
|
| 1091 |
+
"step": 150
|
| 1092 |
+
},
|
| 1093 |
+
{
|
| 1094 |
+
"epoch": 0.8040602379565688,
|
| 1095 |
+
"grad_norm": 1.0859375,
|
| 1096 |
+
"learning_rate": 1.3665012267242974e-05,
|
| 1097 |
+
"loss": 1.6537,
|
| 1098 |
+
"step": 151
|
| 1099 |
+
},
|
| 1100 |
+
{
|
| 1101 |
+
"epoch": 0.8093851401946918,
|
| 1102 |
+
"grad_norm": 1.03125,
|
| 1103 |
+
"learning_rate": 1.3583679495453e-05,
|
| 1104 |
+
"loss": 1.7491,
|
| 1105 |
+
"step": 152
|
| 1106 |
+
},
|
| 1107 |
+
{
|
| 1108 |
+
"epoch": 0.8147100424328148,
|
| 1109 |
+
"grad_norm": 1.0,
|
| 1110 |
+
"learning_rate": 1.3502073812594677e-05,
|
| 1111 |
+
"loss": 1.6909,
|
| 1112 |
+
"step": 153
|
| 1113 |
+
},
|
| 1114 |
+
{
|
| 1115 |
+
"epoch": 0.8200349446709376,
|
| 1116 |
+
"grad_norm": 1.0390625,
|
| 1117 |
+
"learning_rate": 1.342020143325669e-05,
|
| 1118 |
+
"loss": 1.6917,
|
| 1119 |
+
"step": 154
|
| 1120 |
+
},
|
| 1121 |
+
{
|
| 1122 |
+
"epoch": 0.8253598469090606,
|
| 1123 |
+
"grad_norm": 0.99609375,
|
| 1124 |
+
"learning_rate": 1.333806859233771e-05,
|
| 1125 |
+
"loss": 1.692,
|
| 1126 |
+
"step": 155
|
| 1127 |
+
},
|
| 1128 |
+
{
|
| 1129 |
+
"epoch": 0.8306847491471836,
|
| 1130 |
+
"grad_norm": 1.0390625,
|
| 1131 |
+
"learning_rate": 1.3255681544571568e-05,
|
| 1132 |
+
"loss": 1.6995,
|
| 1133 |
+
"step": 156
|
| 1134 |
+
},
|
| 1135 |
+
{
|
| 1136 |
+
"epoch": 0.8360096513853066,
|
| 1137 |
+
"grad_norm": 1.03125,
|
| 1138 |
+
"learning_rate": 1.3173046564050923e-05,
|
| 1139 |
+
"loss": 1.6612,
|
| 1140 |
+
"step": 157
|
| 1141 |
+
},
|
| 1142 |
+
{
|
| 1143 |
+
"epoch": 0.8413345536234296,
|
| 1144 |
+
"grad_norm": 1.09375,
|
| 1145 |
+
"learning_rate": 1.3090169943749475e-05,
|
| 1146 |
+
"loss": 1.7376,
|
| 1147 |
+
"step": 158
|
| 1148 |
+
},
|
| 1149 |
+
{
|
| 1150 |
+
"epoch": 0.8466594558615526,
|
| 1151 |
+
"grad_norm": 1.0,
|
| 1152 |
+
"learning_rate": 1.300705799504273e-05,
|
| 1153 |
+
"loss": 1.6703,
|
| 1154 |
+
"step": 159
|
| 1155 |
+
},
|
| 1156 |
+
{
|
| 1157 |
+
"epoch": 0.8519843580996755,
|
| 1158 |
+
"grad_norm": 1.0546875,
|
| 1159 |
+
"learning_rate": 1.2923717047227368e-05,
|
| 1160 |
+
"loss": 1.6611,
|
| 1161 |
+
"step": 160
|
| 1162 |
+
},
|
| 1163 |
+
{
|
| 1164 |
+
"epoch": 0.8573092603377985,
|
| 1165 |
+
"grad_norm": 0.95703125,
|
| 1166 |
+
"learning_rate": 1.284015344703923e-05,
|
| 1167 |
+
"loss": 1.6215,
|
| 1168 |
+
"step": 161
|
| 1169 |
+
},
|
| 1170 |
+
{
|
| 1171 |
+
"epoch": 0.8626341625759214,
|
| 1172 |
+
"grad_norm": 1.046875,
|
| 1173 |
+
"learning_rate": 1.2756373558169992e-05,
|
| 1174 |
+
"loss": 1.7199,
|
| 1175 |
+
"step": 162
|
| 1176 |
+
},
|
| 1177 |
+
{
|
| 1178 |
+
"epoch": 0.8679590648140444,
|
| 1179 |
+
"grad_norm": 1.0078125,
|
| 1180 |
+
"learning_rate": 1.267238376078257e-05,
|
| 1181 |
+
"loss": 1.6418,
|
| 1182 |
+
"step": 163
|
| 1183 |
+
},
|
| 1184 |
+
{
|
| 1185 |
+
"epoch": 0.8732839670521674,
|
| 1186 |
+
"grad_norm": 0.9140625,
|
| 1187 |
+
"learning_rate": 1.2588190451025209e-05,
|
| 1188 |
+
"loss": 1.6476,
|
| 1189 |
+
"step": 164
|
| 1190 |
+
},
|
| 1191 |
+
{
|
| 1192 |
+
"epoch": 0.8786088692902904,
|
| 1193 |
+
"grad_norm": 0.96484375,
|
| 1194 |
+
"learning_rate": 1.2503800040544417e-05,
|
| 1195 |
+
"loss": 1.6088,
|
| 1196 |
+
"step": 165
|
| 1197 |
+
},
|
| 1198 |
+
{
|
| 1199 |
+
"epoch": 0.8839337715284133,
|
| 1200 |
+
"grad_norm": 0.9921875,
|
| 1201 |
+
"learning_rate": 1.2419218955996677e-05,
|
| 1202 |
+
"loss": 1.6919,
|
| 1203 |
+
"step": 166
|
| 1204 |
+
},
|
| 1205 |
+
{
|
| 1206 |
+
"epoch": 0.8892586737665363,
|
| 1207 |
+
"grad_norm": 1.0625,
|
| 1208 |
+
"learning_rate": 1.2334453638559057e-05,
|
| 1209 |
+
"loss": 1.6118,
|
| 1210 |
+
"step": 167
|
| 1211 |
+
},
|
| 1212 |
+
{
|
| 1213 |
+
"epoch": 0.8945835760046593,
|
| 1214 |
+
"grad_norm": 1.0546875,
|
| 1215 |
+
"learning_rate": 1.2249510543438652e-05,
|
| 1216 |
+
"loss": 1.7246,
|
| 1217 |
+
"step": 168
|
| 1218 |
+
},
|
| 1219 |
+
{
|
| 1220 |
+
"epoch": 0.8999084782427823,
|
| 1221 |
+
"grad_norm": 1.0390625,
|
| 1222 |
+
"learning_rate": 1.2164396139381029e-05,
|
| 1223 |
+
"loss": 1.6821,
|
| 1224 |
+
"step": 169
|
| 1225 |
+
},
|
| 1226 |
+
{
|
| 1227 |
+
"epoch": 0.9052333804809052,
|
| 1228 |
+
"grad_norm": 1.0625,
|
| 1229 |
+
"learning_rate": 1.2079116908177592e-05,
|
| 1230 |
+
"loss": 1.6474,
|
| 1231 |
+
"step": 170
|
| 1232 |
+
},
|
| 1233 |
+
{
|
| 1234 |
+
"epoch": 0.9105582827190282,
|
| 1235 |
+
"grad_norm": 1.1015625,
|
| 1236 |
+
"learning_rate": 1.1993679344171973e-05,
|
| 1237 |
+
"loss": 1.7538,
|
| 1238 |
+
"step": 171
|
| 1239 |
+
},
|
| 1240 |
+
{
|
| 1241 |
+
"epoch": 0.9158831849571512,
|
| 1242 |
+
"grad_norm": 1.0078125,
|
| 1243 |
+
"learning_rate": 1.190808995376545e-05,
|
| 1244 |
+
"loss": 1.6299,
|
| 1245 |
+
"step": 172
|
| 1246 |
+
},
|
| 1247 |
+
{
|
| 1248 |
+
"epoch": 0.9212080871952741,
|
| 1249 |
+
"grad_norm": 1.0078125,
|
| 1250 |
+
"learning_rate": 1.1822355254921478e-05,
|
| 1251 |
+
"loss": 1.6671,
|
| 1252 |
+
"step": 173
|
| 1253 |
+
},
|
| 1254 |
+
{
|
| 1255 |
+
"epoch": 0.9265329894333971,
|
| 1256 |
+
"grad_norm": 1.0078125,
|
| 1257 |
+
"learning_rate": 1.1736481776669307e-05,
|
| 1258 |
+
"loss": 1.6996,
|
| 1259 |
+
"step": 174
|
| 1260 |
+
},
|
| 1261 |
+
{
|
| 1262 |
+
"epoch": 0.9318578916715201,
|
| 1263 |
+
"grad_norm": 1.0546875,
|
| 1264 |
+
"learning_rate": 1.1650476058606776e-05,
|
| 1265 |
+
"loss": 1.6863,
|
| 1266 |
+
"step": 175
|
| 1267 |
+
},
|
| 1268 |
+
{
|
| 1269 |
+
"epoch": 0.9371827939096431,
|
| 1270 |
+
"grad_norm": 1.1171875,
|
| 1271 |
+
"learning_rate": 1.156434465040231e-05,
|
| 1272 |
+
"loss": 1.7013,
|
| 1273 |
+
"step": 176
|
| 1274 |
+
},
|
| 1275 |
+
{
|
| 1276 |
+
"epoch": 0.9425076961477661,
|
| 1277 |
+
"grad_norm": 0.9921875,
|
| 1278 |
+
"learning_rate": 1.1478094111296109e-05,
|
| 1279 |
+
"loss": 1.7416,
|
| 1280 |
+
"step": 177
|
| 1281 |
+
},
|
| 1282 |
+
{
|
| 1283 |
+
"epoch": 0.947832598385889,
|
| 1284 |
+
"grad_norm": 1.03125,
|
| 1285 |
+
"learning_rate": 1.1391731009600655e-05,
|
| 1286 |
+
"loss": 1.7964,
|
| 1287 |
+
"step": 178
|
| 1288 |
+
},
|
| 1289 |
+
{
|
| 1290 |
+
"epoch": 0.9531575006240119,
|
| 1291 |
+
"grad_norm": 1.046875,
|
| 1292 |
+
"learning_rate": 1.130526192220052e-05,
|
| 1293 |
+
"loss": 1.761,
|
| 1294 |
+
"step": 179
|
| 1295 |
+
},
|
| 1296 |
+
{
|
| 1297 |
+
"epoch": 0.9584824028621349,
|
| 1298 |
+
"grad_norm": 1.0234375,
|
| 1299 |
+
"learning_rate": 1.1218693434051475e-05,
|
| 1300 |
+
"loss": 1.6999,
|
| 1301 |
+
"step": 180
|
| 1302 |
+
},
|
| 1303 |
+
{
|
| 1304 |
+
"epoch": 0.9638073051002579,
|
| 1305 |
+
"grad_norm": 1.0546875,
|
| 1306 |
+
"learning_rate": 1.113203213767907e-05,
|
| 1307 |
+
"loss": 1.69,
|
| 1308 |
+
"step": 181
|
| 1309 |
+
},
|
| 1310 |
+
{
|
| 1311 |
+
"epoch": 0.9691322073383809,
|
| 1312 |
+
"grad_norm": 0.953125,
|
| 1313 |
+
"learning_rate": 1.1045284632676535e-05,
|
| 1314 |
+
"loss": 1.6777,
|
| 1315 |
+
"step": 182
|
| 1316 |
+
},
|
| 1317 |
+
{
|
| 1318 |
+
"epoch": 0.9744571095765039,
|
| 1319 |
+
"grad_norm": 1.0234375,
|
| 1320 |
+
"learning_rate": 1.0958457525202241e-05,
|
| 1321 |
+
"loss": 1.6792,
|
| 1322 |
+
"step": 183
|
| 1323 |
+
},
|
| 1324 |
+
{
|
| 1325 |
+
"epoch": 0.9797820118146269,
|
| 1326 |
+
"grad_norm": 1.03125,
|
| 1327 |
+
"learning_rate": 1.0871557427476585e-05,
|
| 1328 |
+
"loss": 1.6724,
|
| 1329 |
+
"step": 184
|
| 1330 |
+
},
|
| 1331 |
+
{
|
| 1332 |
+
"epoch": 0.9851069140527499,
|
| 1333 |
+
"grad_norm": 1.0625,
|
| 1334 |
+
"learning_rate": 1.0784590957278452e-05,
|
| 1335 |
+
"loss": 1.7415,
|
| 1336 |
+
"step": 185
|
| 1337 |
+
},
|
| 1338 |
+
{
|
| 1339 |
+
"epoch": 0.9904318162908727,
|
| 1340 |
+
"grad_norm": 1.0,
|
| 1341 |
+
"learning_rate": 1.0697564737441254e-05,
|
| 1342 |
+
"loss": 1.6796,
|
| 1343 |
+
"step": 186
|
| 1344 |
+
},
|
| 1345 |
+
{
|
| 1346 |
+
"epoch": 0.9957567185289957,
|
| 1347 |
+
"grad_norm": 1.015625,
|
| 1348 |
+
"learning_rate": 1.0610485395348571e-05,
|
| 1349 |
+
"loss": 1.6186,
|
| 1350 |
+
"step": 187
|
| 1351 |
+
},
|
| 1352 |
+
{
|
| 1353 |
+
"epoch": 1.0010816207671187,
|
| 1354 |
+
"grad_norm": 0.984375,
|
| 1355 |
+
"learning_rate": 1.0523359562429441e-05,
|
| 1356 |
+
"loss": 1.6741,
|
| 1357 |
+
"step": 188
|
| 1358 |
+
},
|
| 1359 |
+
{
|
| 1360 |
+
"epoch": 1.0010816207671187,
|
| 1361 |
+
"eval_loss": 1.6922072172164917,
|
| 1362 |
+
"eval_runtime": 77.3424,
|
| 1363 |
+
"eval_samples_per_second": 15.257,
|
| 1364 |
+
"eval_steps_per_second": 15.257,
|
| 1365 |
+
"step": 188
|
| 1366 |
+
},
|
| 1367 |
+
{
|
| 1368 |
+
"epoch": 1.0064065230052417,
|
| 1369 |
+
"grad_norm": 0.9921875,
|
| 1370 |
+
"learning_rate": 1.0436193873653362e-05,
|
| 1371 |
+
"loss": 1.7202,
|
| 1372 |
+
"step": 189
|
| 1373 |
+
},
|
| 1374 |
+
{
|
| 1375 |
+
"epoch": 1.0016640319494134,
|
| 1376 |
+
"grad_norm": 0.9921875,
|
| 1377 |
+
"learning_rate": 1.0348994967025012e-05,
|
| 1378 |
+
"loss": 1.639,
|
| 1379 |
+
"step": 190
|
| 1380 |
+
},
|
| 1381 |
+
{
|
| 1382 |
+
"epoch": 1.0069889341875364,
|
| 1383 |
+
"grad_norm": 1.0546875,
|
| 1384 |
+
"learning_rate": 1.0261769483078734e-05,
|
| 1385 |
+
"loss": 1.6223,
|
| 1386 |
+
"step": 191
|
| 1387 |
+
},
|
| 1388 |
+
{
|
| 1389 |
+
"epoch": 1.0123138364256594,
|
| 1390 |
+
"grad_norm": 1.0078125,
|
| 1391 |
+
"learning_rate": 1.0174524064372837e-05,
|
| 1392 |
+
"loss": 1.7193,
|
| 1393 |
+
"step": 192
|
| 1394 |
+
},
|
| 1395 |
+
{
|
| 1396 |
+
"epoch": 1.0176387386637824,
|
| 1397 |
+
"grad_norm": 1.0078125,
|
| 1398 |
+
"learning_rate": 1.008726535498374e-05,
|
| 1399 |
+
"loss": 1.6904,
|
| 1400 |
+
"step": 193
|
| 1401 |
+
},
|
| 1402 |
+
{
|
| 1403 |
+
"epoch": 1.0229636409019054,
|
| 1404 |
+
"grad_norm": 1.03125,
|
| 1405 |
+
"learning_rate": 1e-05,
|
| 1406 |
+
"loss": 1.6921,
|
| 1407 |
+
"step": 194
|
| 1408 |
+
},
|
| 1409 |
+
{
|
| 1410 |
+
"epoch": 1.0282885431400284,
|
| 1411 |
+
"grad_norm": 0.9921875,
|
| 1412 |
+
"learning_rate": 9.912734645016262e-06,
|
| 1413 |
+
"loss": 1.6593,
|
| 1414 |
+
"step": 195
|
| 1415 |
+
},
|
| 1416 |
+
{
|
| 1417 |
+
"epoch": 1.0336134453781514,
|
| 1418 |
+
"grad_norm": 0.97265625,
|
| 1419 |
+
"learning_rate": 9.825475935627165e-06,
|
| 1420 |
+
"loss": 1.6469,
|
| 1421 |
+
"step": 196
|
| 1422 |
+
},
|
| 1423 |
+
{
|
| 1424 |
+
"epoch": 1.0389383476162741,
|
| 1425 |
+
"grad_norm": 0.9921875,
|
| 1426 |
+
"learning_rate": 9.738230516921272e-06,
|
| 1427 |
+
"loss": 1.5877,
|
| 1428 |
+
"step": 197
|
| 1429 |
+
},
|
| 1430 |
+
{
|
| 1431 |
+
"epoch": 1.0442632498543971,
|
| 1432 |
+
"grad_norm": 1.03125,
|
| 1433 |
+
"learning_rate": 9.651005032974994e-06,
|
| 1434 |
+
"loss": 1.6726,
|
| 1435 |
+
"step": 198
|
| 1436 |
+
},
|
| 1437 |
+
{
|
| 1438 |
+
"epoch": 1.04958815209252,
|
| 1439 |
+
"grad_norm": 1.0,
|
| 1440 |
+
"learning_rate": 9.563806126346643e-06,
|
| 1441 |
+
"loss": 1.6607,
|
| 1442 |
+
"step": 199
|
| 1443 |
+
},
|
| 1444 |
+
{
|
| 1445 |
+
"epoch": 1.054913054330643,
|
| 1446 |
+
"grad_norm": 1.015625,
|
| 1447 |
+
"learning_rate": 9.476640437570562e-06,
|
| 1448 |
+
"loss": 1.6926,
|
| 1449 |
+
"step": 200
|
| 1450 |
+
},
|
| 1451 |
+
{
|
| 1452 |
+
"epoch": 1.060237956568766,
|
| 1453 |
+
"grad_norm": 1.015625,
|
| 1454 |
+
"learning_rate": 9.38951460465143e-06,
|
| 1455 |
+
"loss": 1.6241,
|
| 1456 |
+
"step": 201
|
| 1457 |
+
},
|
| 1458 |
+
{
|
| 1459 |
+
"epoch": 1.065562858806889,
|
| 1460 |
+
"grad_norm": 0.9921875,
|
| 1461 |
+
"learning_rate": 9.302435262558748e-06,
|
| 1462 |
+
"loss": 1.6431,
|
| 1463 |
+
"step": 202
|
| 1464 |
+
},
|
| 1465 |
+
{
|
| 1466 |
+
"epoch": 1.070887761045012,
|
| 1467 |
+
"grad_norm": 1.0234375,
|
| 1468 |
+
"learning_rate": 9.215409042721553e-06,
|
| 1469 |
+
"loss": 1.6339,
|
| 1470 |
+
"step": 203
|
| 1471 |
+
},
|
| 1472 |
+
{
|
| 1473 |
+
"epoch": 1.076212663283135,
|
| 1474 |
+
"grad_norm": 1.0,
|
| 1475 |
+
"learning_rate": 9.128442572523418e-06,
|
| 1476 |
+
"loss": 1.6701,
|
| 1477 |
+
"step": 204
|
| 1478 |
+
},
|
| 1479 |
+
{
|
| 1480 |
+
"epoch": 1.081537565521258,
|
| 1481 |
+
"grad_norm": 0.98828125,
|
| 1482 |
+
"learning_rate": 9.04154247479776e-06,
|
| 1483 |
+
"loss": 1.6059,
|
| 1484 |
+
"step": 205
|
| 1485 |
+
},
|
| 1486 |
+
{
|
| 1487 |
+
"epoch": 1.086862467759381,
|
| 1488 |
+
"grad_norm": 1.0234375,
|
| 1489 |
+
"learning_rate": 8.954715367323468e-06,
|
| 1490 |
+
"loss": 1.6638,
|
| 1491 |
+
"step": 206
|
| 1492 |
+
},
|
| 1493 |
+
{
|
| 1494 |
+
"epoch": 1.092187369997504,
|
| 1495 |
+
"grad_norm": 1.09375,
|
| 1496 |
+
"learning_rate": 8.867967862320935e-06,
|
| 1497 |
+
"loss": 1.6428,
|
| 1498 |
+
"step": 207
|
| 1499 |
+
},
|
| 1500 |
+
{
|
| 1501 |
+
"epoch": 1.097512272235627,
|
| 1502 |
+
"grad_norm": 0.9765625,
|
| 1503 |
+
"learning_rate": 8.781306565948528e-06,
|
| 1504 |
+
"loss": 1.6424,
|
| 1505 |
+
"step": 208
|
| 1506 |
+
},
|
| 1507 |
+
{
|
| 1508 |
+
"epoch": 1.10283717447375,
|
| 1509 |
+
"grad_norm": 0.984375,
|
| 1510 |
+
"learning_rate": 8.694738077799487e-06,
|
| 1511 |
+
"loss": 1.6105,
|
| 1512 |
+
"step": 209
|
| 1513 |
+
},
|
| 1514 |
+
{
|
| 1515 |
+
"epoch": 1.108162076711873,
|
| 1516 |
+
"grad_norm": 1.0,
|
| 1517 |
+
"learning_rate": 8.60826899039935e-06,
|
| 1518 |
+
"loss": 1.7068,
|
| 1519 |
+
"step": 210
|
| 1520 |
+
},
|
| 1521 |
+
{
|
| 1522 |
+
"epoch": 1.113486978949996,
|
| 1523 |
+
"grad_norm": 1.0859375,
|
| 1524 |
+
"learning_rate": 8.521905888703894e-06,
|
| 1525 |
+
"loss": 1.7118,
|
| 1526 |
+
"step": 211
|
| 1527 |
+
},
|
| 1528 |
+
{
|
| 1529 |
+
"epoch": 1.118811881188119,
|
| 1530 |
+
"grad_norm": 1.0625,
|
| 1531 |
+
"learning_rate": 8.43565534959769e-06,
|
| 1532 |
+
"loss": 1.6553,
|
| 1533 |
+
"step": 212
|
| 1534 |
+
},
|
| 1535 |
+
{
|
| 1536 |
+
"epoch": 1.1241367834262417,
|
| 1537 |
+
"grad_norm": 1.0703125,
|
| 1538 |
+
"learning_rate": 8.349523941393224e-06,
|
| 1539 |
+
"loss": 1.6909,
|
| 1540 |
+
"step": 213
|
| 1541 |
+
},
|
| 1542 |
+
{
|
| 1543 |
+
"epoch": 1.1294616856643647,
|
| 1544 |
+
"grad_norm": 1.0546875,
|
| 1545 |
+
"learning_rate": 8.263518223330698e-06,
|
| 1546 |
+
"loss": 1.7384,
|
| 1547 |
+
"step": 214
|
| 1548 |
+
},
|
| 1549 |
+
{
|
| 1550 |
+
"epoch": 1.1347865879024877,
|
| 1551 |
+
"grad_norm": 1.03125,
|
| 1552 |
+
"learning_rate": 8.177644745078525e-06,
|
| 1553 |
+
"loss": 1.6182,
|
| 1554 |
+
"step": 215
|
| 1555 |
+
},
|
| 1556 |
+
{
|
| 1557 |
+
"epoch": 1.1401114901406106,
|
| 1558 |
+
"grad_norm": 1.125,
|
| 1559 |
+
"learning_rate": 8.091910046234552e-06,
|
| 1560 |
+
"loss": 1.7063,
|
| 1561 |
+
"step": 216
|
| 1562 |
+
},
|
| 1563 |
+
{
|
| 1564 |
+
"epoch": 1.1454363923787336,
|
| 1565 |
+
"grad_norm": 0.96484375,
|
| 1566 |
+
"learning_rate": 8.00632065582803e-06,
|
| 1567 |
+
"loss": 1.6621,
|
| 1568 |
+
"step": 217
|
| 1569 |
+
},
|
| 1570 |
+
{
|
| 1571 |
+
"epoch": 1.1507612946168566,
|
| 1572 |
+
"grad_norm": 0.98828125,
|
| 1573 |
+
"learning_rate": 7.92088309182241e-06,
|
| 1574 |
+
"loss": 1.7231,
|
| 1575 |
+
"step": 218
|
| 1576 |
+
},
|
| 1577 |
+
{
|
| 1578 |
+
"epoch": 1.1560861968549796,
|
| 1579 |
+
"grad_norm": 1.015625,
|
| 1580 |
+
"learning_rate": 7.835603860618973e-06,
|
| 1581 |
+
"loss": 1.6707,
|
| 1582 |
+
"step": 219
|
| 1583 |
+
},
|
| 1584 |
+
{
|
| 1585 |
+
"epoch": 1.1614110990931026,
|
| 1586 |
+
"grad_norm": 1.015625,
|
| 1587 |
+
"learning_rate": 7.750489456561351e-06,
|
| 1588 |
+
"loss": 1.6226,
|
| 1589 |
+
"step": 220
|
| 1590 |
+
},
|
| 1591 |
+
{
|
| 1592 |
+
"epoch": 1.1667360013312256,
|
| 1593 |
+
"grad_norm": 1.0703125,
|
| 1594 |
+
"learning_rate": 7.66554636144095e-06,
|
| 1595 |
+
"loss": 1.6862,
|
| 1596 |
+
"step": 221
|
| 1597 |
+
},
|
| 1598 |
+
{
|
| 1599 |
+
"epoch": 1.1720609035693486,
|
| 1600 |
+
"grad_norm": 1.0625,
|
| 1601 |
+
"learning_rate": 7.580781044003324e-06,
|
| 1602 |
+
"loss": 1.7325,
|
| 1603 |
+
"step": 222
|
| 1604 |
+
},
|
| 1605 |
+
{
|
| 1606 |
+
"epoch": 1.1773858058074715,
|
| 1607 |
+
"grad_norm": 1.03125,
|
| 1608 |
+
"learning_rate": 7.496199959455584e-06,
|
| 1609 |
+
"loss": 1.6573,
|
| 1610 |
+
"step": 223
|
| 1611 |
+
},
|
| 1612 |
+
{
|
| 1613 |
+
"epoch": 1.1827107080455945,
|
| 1614 |
+
"grad_norm": 1.03125,
|
| 1615 |
+
"learning_rate": 7.411809548974792e-06,
|
| 1616 |
+
"loss": 1.6594,
|
| 1617 |
+
"step": 224
|
| 1618 |
+
},
|
| 1619 |
+
{
|
| 1620 |
+
"epoch": 1.1880356102837175,
|
| 1621 |
+
"grad_norm": 1.015625,
|
| 1622 |
+
"learning_rate": 7.327616239217432e-06,
|
| 1623 |
+
"loss": 1.6523,
|
| 1624 |
+
"step": 225
|
| 1625 |
+
},
|
| 1626 |
+
{
|
| 1627 |
+
"epoch": 1.1933605125218405,
|
| 1628 |
+
"grad_norm": 1.0546875,
|
| 1629 |
+
"learning_rate": 7.243626441830009e-06,
|
| 1630 |
+
"loss": 1.6895,
|
| 1631 |
+
"step": 226
|
| 1632 |
+
},
|
| 1633 |
+
{
|
| 1634 |
+
"epoch": 1.1986854147599635,
|
| 1635 |
+
"grad_norm": 0.99609375,
|
| 1636 |
+
"learning_rate": 7.159846552960774e-06,
|
| 1637 |
+
"loss": 1.6883,
|
| 1638 |
+
"step": 227
|
| 1639 |
+
},
|
| 1640 |
+
{
|
| 1641 |
+
"epoch": 1.2040103169980862,
|
| 1642 |
+
"grad_norm": 0.9921875,
|
| 1643 |
+
"learning_rate": 7.076282952772634e-06,
|
| 1644 |
+
"loss": 1.5706,
|
| 1645 |
+
"step": 228
|
| 1646 |
+
},
|
| 1647 |
+
{
|
| 1648 |
+
"epoch": 1.2093352192362095,
|
| 1649 |
+
"grad_norm": 0.95703125,
|
| 1650 |
+
"learning_rate": 6.992942004957271e-06,
|
| 1651 |
+
"loss": 1.5938,
|
| 1652 |
+
"step": 229
|
| 1653 |
+
},
|
| 1654 |
+
{
|
| 1655 |
+
"epoch": 1.2146601214743322,
|
| 1656 |
+
"grad_norm": 1.046875,
|
| 1657 |
+
"learning_rate": 6.909830056250527e-06,
|
| 1658 |
+
"loss": 1.6963,
|
| 1659 |
+
"step": 230
|
| 1660 |
+
},
|
| 1661 |
+
{
|
| 1662 |
+
"epoch": 1.2199850237124552,
|
| 1663 |
+
"grad_norm": 1.0703125,
|
| 1664 |
+
"learning_rate": 6.826953435949081e-06,
|
| 1665 |
+
"loss": 1.6565,
|
| 1666 |
+
"step": 231
|
| 1667 |
+
},
|
| 1668 |
+
{
|
| 1669 |
+
"epoch": 1.2253099259505782,
|
| 1670 |
+
"grad_norm": 1.0078125,
|
| 1671 |
+
"learning_rate": 6.744318455428436e-06,
|
| 1672 |
+
"loss": 1.657,
|
| 1673 |
+
"step": 232
|
| 1674 |
+
},
|
| 1675 |
+
{
|
| 1676 |
+
"epoch": 1.2306348281887012,
|
| 1677 |
+
"grad_norm": 1.09375,
|
| 1678 |
+
"learning_rate": 6.661931407662292e-06,
|
| 1679 |
+
"loss": 1.6942,
|
| 1680 |
+
"step": 233
|
| 1681 |
+
},
|
| 1682 |
+
{
|
| 1683 |
+
"epoch": 1.2359597304268242,
|
| 1684 |
+
"grad_norm": 0.94921875,
|
| 1685 |
+
"learning_rate": 6.579798566743314e-06,
|
| 1686 |
+
"loss": 1.6247,
|
| 1687 |
+
"step": 234
|
| 1688 |
+
},
|
| 1689 |
+
{
|
| 1690 |
+
"epoch": 1.2412846326649472,
|
| 1691 |
+
"grad_norm": 1.0078125,
|
| 1692 |
+
"learning_rate": 6.497926187405326e-06,
|
| 1693 |
+
"loss": 1.6845,
|
| 1694 |
+
"step": 235
|
| 1695 |
+
},
|
| 1696 |
+
{
|
| 1697 |
+
"epoch": 1.2412846326649472,
|
| 1698 |
+
"eval_loss": 1.6904007196426392,
|
| 1699 |
+
"eval_runtime": 77.2338,
|
| 1700 |
+
"eval_samples_per_second": 15.278,
|
| 1701 |
+
"eval_steps_per_second": 15.278,
|
| 1702 |
+
"step": 235
|
| 1703 |
+
},
|
| 1704 |
+
{
|
| 1705 |
+
"epoch": 1.2466095349030701,
|
| 1706 |
+
"grad_norm": 1.046875,
|
| 1707 |
+
"learning_rate": 6.4163205045469975e-06,
|
| 1708 |
+
"loss": 1.6958,
|
| 1709 |
+
"step": 236
|
| 1710 |
+
},
|
| 1711 |
+
{
|
| 1712 |
+
"epoch": 1.2519344371411931,
|
| 1713 |
+
"grad_norm": 1.109375,
|
| 1714 |
+
"learning_rate": 6.334987732757028e-06,
|
| 1715 |
+
"loss": 1.589,
|
| 1716 |
+
"step": 237
|
| 1717 |
+
},
|
| 1718 |
+
{
|
| 1719 |
+
"epoch": 1.257259339379316,
|
| 1720 |
+
"grad_norm": 1.0390625,
|
| 1721 |
+
"learning_rate": 6.25393406584088e-06,
|
| 1722 |
+
"loss": 1.6417,
|
| 1723 |
+
"step": 238
|
| 1724 |
+
},
|
| 1725 |
+
{
|
| 1726 |
+
"epoch": 1.262584241617439,
|
| 1727 |
+
"grad_norm": 0.9609375,
|
| 1728 |
+
"learning_rate": 6.173165676349103e-06,
|
| 1729 |
+
"loss": 1.6412,
|
| 1730 |
+
"step": 239
|
| 1731 |
+
},
|
| 1732 |
+
{
|
| 1733 |
+
"epoch": 1.267909143855562,
|
| 1734 |
+
"grad_norm": 1.046875,
|
| 1735 |
+
"learning_rate": 6.092688715107265e-06,
|
| 1736 |
+
"loss": 1.6539,
|
| 1737 |
+
"step": 240
|
| 1738 |
+
},
|
| 1739 |
+
{
|
| 1740 |
+
"epoch": 1.273234046093685,
|
| 1741 |
+
"grad_norm": 1.03125,
|
| 1742 |
+
"learning_rate": 6.0125093107475385e-06,
|
| 1743 |
+
"loss": 1.6194,
|
| 1744 |
+
"step": 241
|
| 1745 |
+
},
|
| 1746 |
+
{
|
| 1747 |
+
"epoch": 1.278558948331808,
|
| 1748 |
+
"grad_norm": 0.9921875,
|
| 1749 |
+
"learning_rate": 5.932633569242e-06,
|
| 1750 |
+
"loss": 1.6566,
|
| 1751 |
+
"step": 242
|
| 1752 |
+
},
|
| 1753 |
+
{
|
| 1754 |
+
"epoch": 1.2838838505699308,
|
| 1755 |
+
"grad_norm": 1.0234375,
|
| 1756 |
+
"learning_rate": 5.853067573437612e-06,
|
| 1757 |
+
"loss": 1.6734,
|
| 1758 |
+
"step": 243
|
| 1759 |
+
},
|
| 1760 |
+
{
|
| 1761 |
+
"epoch": 1.289208752808054,
|
| 1762 |
+
"grad_norm": 1.046875,
|
| 1763 |
+
"learning_rate": 5.773817382593008e-06,
|
| 1764 |
+
"loss": 1.7109,
|
| 1765 |
+
"step": 244
|
| 1766 |
+
},
|
| 1767 |
+
{
|
| 1768 |
+
"epoch": 1.2945336550461768,
|
| 1769 |
+
"grad_norm": 1.015625,
|
| 1770 |
+
"learning_rate": 5.694889031917047e-06,
|
| 1771 |
+
"loss": 1.5984,
|
| 1772 |
+
"step": 245
|
| 1773 |
+
},
|
| 1774 |
+
{
|
| 1775 |
+
"epoch": 1.2998585572842998,
|
| 1776 |
+
"grad_norm": 1.046875,
|
| 1777 |
+
"learning_rate": 5.616288532109225e-06,
|
| 1778 |
+
"loss": 1.7059,
|
| 1779 |
+
"step": 246
|
| 1780 |
+
},
|
| 1781 |
+
{
|
| 1782 |
+
"epoch": 1.3051834595224228,
|
| 1783 |
+
"grad_norm": 1.0390625,
|
| 1784 |
+
"learning_rate": 5.5380218689019125e-06,
|
| 1785 |
+
"loss": 1.6666,
|
| 1786 |
+
"step": 247
|
| 1787 |
+
},
|
| 1788 |
+
{
|
| 1789 |
+
"epoch": 1.3105083617605457,
|
| 1790 |
+
"grad_norm": 1.0234375,
|
| 1791 |
+
"learning_rate": 5.460095002604533e-06,
|
| 1792 |
+
"loss": 1.6752,
|
| 1793 |
+
"step": 248
|
| 1794 |
+
},
|
| 1795 |
+
{
|
| 1796 |
+
"epoch": 1.3158332639986687,
|
| 1797 |
+
"grad_norm": 1.0625,
|
| 1798 |
+
"learning_rate": 5.382513867649663e-06,
|
| 1799 |
+
"loss": 1.684,
|
| 1800 |
+
"step": 249
|
| 1801 |
+
},
|
| 1802 |
+
{
|
| 1803 |
+
"epoch": 1.3211581662367917,
|
| 1804 |
+
"grad_norm": 0.98046875,
|
| 1805 |
+
"learning_rate": 5.305284372141095e-06,
|
| 1806 |
+
"loss": 1.6072,
|
| 1807 |
+
"step": 250
|
| 1808 |
+
},
|
| 1809 |
+
{
|
| 1810 |
+
"epoch": 1.3264830684749147,
|
| 1811 |
+
"grad_norm": 0.9296875,
|
| 1812 |
+
"learning_rate": 5.228412397403916e-06,
|
| 1813 |
+
"loss": 1.4978,
|
| 1814 |
+
"step": 251
|
| 1815 |
+
},
|
| 1816 |
+
{
|
| 1817 |
+
"epoch": 1.3318079707130377,
|
| 1818 |
+
"grad_norm": 1.0390625,
|
| 1819 |
+
"learning_rate": 5.151903797536631e-06,
|
| 1820 |
+
"loss": 1.6164,
|
| 1821 |
+
"step": 252
|
| 1822 |
+
},
|
| 1823 |
+
{
|
| 1824 |
+
"epoch": 1.3371328729511607,
|
| 1825 |
+
"grad_norm": 1.0703125,
|
| 1826 |
+
"learning_rate": 5.075764398965331e-06,
|
| 1827 |
+
"loss": 1.5904,
|
| 1828 |
+
"step": 253
|
| 1829 |
+
},
|
| 1830 |
+
{
|
| 1831 |
+
"epoch": 1.3424577751892837,
|
| 1832 |
+
"grad_norm": 1.0,
|
| 1833 |
+
"learning_rate": 5.000000000000003e-06,
|
| 1834 |
+
"loss": 1.5924,
|
| 1835 |
+
"step": 254
|
| 1836 |
+
},
|
| 1837 |
+
{
|
| 1838 |
+
"epoch": 1.3477826774274067,
|
| 1839 |
+
"grad_norm": 1.0390625,
|
| 1840 |
+
"learning_rate": 4.924616370392962e-06,
|
| 1841 |
+
"loss": 1.6521,
|
| 1842 |
+
"step": 255
|
| 1843 |
+
},
|
| 1844 |
+
{
|
| 1845 |
+
"epoch": 1.3531075796655296,
|
| 1846 |
+
"grad_norm": 1.0234375,
|
| 1847 |
+
"learning_rate": 4.849619250899458e-06,
|
| 1848 |
+
"loss": 1.6124,
|
| 1849 |
+
"step": 256
|
| 1850 |
+
},
|
| 1851 |
+
{
|
| 1852 |
+
"epoch": 1.3584324819036526,
|
| 1853 |
+
"grad_norm": 1.0234375,
|
| 1854 |
+
"learning_rate": 4.775014352840512e-06,
|
| 1855 |
+
"loss": 1.6634,
|
| 1856 |
+
"step": 257
|
| 1857 |
+
},
|
| 1858 |
+
{
|
| 1859 |
+
"epoch": 1.3637573841417754,
|
| 1860 |
+
"grad_norm": 0.984375,
|
| 1861 |
+
"learning_rate": 4.700807357667953e-06,
|
| 1862 |
+
"loss": 1.628,
|
| 1863 |
+
"step": 258
|
| 1864 |
+
},
|
| 1865 |
+
{
|
| 1866 |
+
"epoch": 1.3690822863798986,
|
| 1867 |
+
"grad_norm": 1.046875,
|
| 1868 |
+
"learning_rate": 4.627003916531761e-06,
|
| 1869 |
+
"loss": 1.7282,
|
| 1870 |
+
"step": 259
|
| 1871 |
+
},
|
| 1872 |
+
{
|
| 1873 |
+
"epoch": 1.3744071886180214,
|
| 1874 |
+
"grad_norm": 1.0625,
|
| 1875 |
+
"learning_rate": 4.5536096498497295e-06,
|
| 1876 |
+
"loss": 1.7081,
|
| 1877 |
+
"step": 260
|
| 1878 |
+
},
|
| 1879 |
+
{
|
| 1880 |
+
"epoch": 1.3797320908561446,
|
| 1881 |
+
"grad_norm": 1.03125,
|
| 1882 |
+
"learning_rate": 4.480630146879419e-06,
|
| 1883 |
+
"loss": 1.6772,
|
| 1884 |
+
"step": 261
|
| 1885 |
+
},
|
| 1886 |
+
{
|
| 1887 |
+
"epoch": 1.3850569930942673,
|
| 1888 |
+
"grad_norm": 1.0234375,
|
| 1889 |
+
"learning_rate": 4.408070965292534e-06,
|
| 1890 |
+
"loss": 1.695,
|
| 1891 |
+
"step": 262
|
| 1892 |
+
},
|
| 1893 |
+
{
|
| 1894 |
+
"epoch": 1.3903818953323903,
|
| 1895 |
+
"grad_norm": 0.97265625,
|
| 1896 |
+
"learning_rate": 4.335937630751675e-06,
|
| 1897 |
+
"loss": 1.6414,
|
| 1898 |
+
"step": 263
|
| 1899 |
+
},
|
| 1900 |
+
{
|
| 1901 |
+
"epoch": 1.3957067975705133,
|
| 1902 |
+
"grad_norm": 1.015625,
|
| 1903 |
+
"learning_rate": 4.264235636489542e-06,
|
| 1904 |
+
"loss": 1.6425,
|
| 1905 |
+
"step": 264
|
| 1906 |
+
},
|
| 1907 |
+
{
|
| 1908 |
+
"epoch": 1.4010316998086363,
|
| 1909 |
+
"grad_norm": 1.0390625,
|
| 1910 |
+
"learning_rate": 4.192970442890602e-06,
|
| 1911 |
+
"loss": 1.6466,
|
| 1912 |
+
"step": 265
|
| 1913 |
+
},
|
| 1914 |
+
{
|
| 1915 |
+
"epoch": 1.4063566020467593,
|
| 1916 |
+
"grad_norm": 1.0,
|
| 1917 |
+
"learning_rate": 4.12214747707527e-06,
|
| 1918 |
+
"loss": 1.6378,
|
| 1919 |
+
"step": 266
|
| 1920 |
+
},
|
| 1921 |
+
{
|
| 1922 |
+
"epoch": 1.4116815042848823,
|
| 1923 |
+
"grad_norm": 1.015625,
|
| 1924 |
+
"learning_rate": 4.051772132486589e-06,
|
| 1925 |
+
"loss": 1.718,
|
| 1926 |
+
"step": 267
|
| 1927 |
+
},
|
| 1928 |
+
{
|
| 1929 |
+
"epoch": 1.4170064065230052,
|
| 1930 |
+
"grad_norm": 1.1015625,
|
| 1931 |
+
"learning_rate": 3.981849768479516e-06,
|
| 1932 |
+
"loss": 1.6534,
|
| 1933 |
+
"step": 268
|
| 1934 |
+
},
|
| 1935 |
+
{
|
| 1936 |
+
"epoch": 1.4223313087611282,
|
| 1937 |
+
"grad_norm": 0.98046875,
|
| 1938 |
+
"learning_rate": 3.912385709912794e-06,
|
| 1939 |
+
"loss": 1.6769,
|
| 1940 |
+
"step": 269
|
| 1941 |
+
},
|
| 1942 |
+
{
|
| 1943 |
+
"epoch": 1.4276562109992512,
|
| 1944 |
+
"grad_norm": 1.046875,
|
| 1945 |
+
"learning_rate": 3.8433852467434175e-06,
|
| 1946 |
+
"loss": 1.6337,
|
| 1947 |
+
"step": 270
|
| 1948 |
+
},
|
| 1949 |
+
{
|
| 1950 |
+
"epoch": 1.4329811132373742,
|
| 1951 |
+
"grad_norm": 1.0625,
|
| 1952 |
+
"learning_rate": 3.774853633623806e-06,
|
| 1953 |
+
"loss": 1.7151,
|
| 1954 |
+
"step": 271
|
| 1955 |
+
},
|
| 1956 |
+
{
|
| 1957 |
+
"epoch": 1.4383060154754972,
|
| 1958 |
+
"grad_norm": 1.0234375,
|
| 1959 |
+
"learning_rate": 3.7067960895016277e-06,
|
| 1960 |
+
"loss": 1.6018,
|
| 1961 |
+
"step": 272
|
| 1962 |
+
},
|
| 1963 |
+
{
|
| 1964 |
+
"epoch": 1.4436309177136202,
|
| 1965 |
+
"grad_norm": 1.03125,
|
| 1966 |
+
"learning_rate": 3.6392177972223596e-06,
|
| 1967 |
+
"loss": 1.597,
|
| 1968 |
+
"step": 273
|
| 1969 |
+
},
|
| 1970 |
+
{
|
| 1971 |
+
"epoch": 1.4489558199517432,
|
| 1972 |
+
"grad_norm": 1.0390625,
|
| 1973 |
+
"learning_rate": 3.5721239031346067e-06,
|
| 1974 |
+
"loss": 1.6465,
|
| 1975 |
+
"step": 274
|
| 1976 |
+
},
|
| 1977 |
+
{
|
| 1978 |
+
"epoch": 1.454280722189866,
|
| 1979 |
+
"grad_norm": 1.03125,
|
| 1980 |
+
"learning_rate": 3.505519516698165e-06,
|
| 1981 |
+
"loss": 1.7173,
|
| 1982 |
+
"step": 275
|
| 1983 |
+
},
|
| 1984 |
+
{
|
| 1985 |
+
"epoch": 1.4596056244279891,
|
| 1986 |
+
"grad_norm": 1.015625,
|
| 1987 |
+
"learning_rate": 3.4394097100949286e-06,
|
| 1988 |
+
"loss": 1.5711,
|
| 1989 |
+
"step": 276
|
| 1990 |
+
},
|
| 1991 |
+
{
|
| 1992 |
+
"epoch": 1.464930526666112,
|
| 1993 |
+
"grad_norm": 1.03125,
|
| 1994 |
+
"learning_rate": 3.3737995178426276e-06,
|
| 1995 |
+
"loss": 1.7184,
|
| 1996 |
+
"step": 277
|
| 1997 |
+
},
|
| 1998 |
+
{
|
| 1999 |
+
"epoch": 1.4702554289042349,
|
| 2000 |
+
"grad_norm": 1.0234375,
|
| 2001 |
+
"learning_rate": 3.308693936411421e-06,
|
| 2002 |
+
"loss": 1.6222,
|
| 2003 |
+
"step": 278
|
| 2004 |
+
},
|
| 2005 |
+
{
|
| 2006 |
+
"epoch": 1.4755803311423579,
|
| 2007 |
+
"grad_norm": 0.99609375,
|
| 2008 |
+
"learning_rate": 3.2440979238433977e-06,
|
| 2009 |
+
"loss": 1.7164,
|
| 2010 |
+
"step": 279
|
| 2011 |
+
},
|
| 2012 |
+
{
|
| 2013 |
+
"epoch": 1.4809052333804809,
|
| 2014 |
+
"grad_norm": 1.03125,
|
| 2015 |
+
"learning_rate": 3.1800163993750166e-06,
|
| 2016 |
+
"loss": 1.8001,
|
| 2017 |
+
"step": 280
|
| 2018 |
+
},
|
| 2019 |
+
{
|
| 2020 |
+
"epoch": 1.4862301356186038,
|
| 2021 |
+
"grad_norm": 1.125,
|
| 2022 |
+
"learning_rate": 3.116454243062459e-06,
|
| 2023 |
+
"loss": 1.6933,
|
| 2024 |
+
"step": 281
|
| 2025 |
+
},
|
| 2026 |
+
{
|
| 2027 |
+
"epoch": 1.4915550378567268,
|
| 2028 |
+
"grad_norm": 1.0234375,
|
| 2029 |
+
"learning_rate": 3.0534162954100264e-06,
|
| 2030 |
+
"loss": 1.6367,
|
| 2031 |
+
"step": 282
|
| 2032 |
+
},
|
| 2033 |
+
{
|
| 2034 |
+
"epoch": 1.4915550378567268,
|
| 2035 |
+
"eval_loss": 1.6898616552352905,
|
| 2036 |
+
"eval_runtime": 78.3804,
|
| 2037 |
+
"eval_samples_per_second": 15.055,
|
| 2038 |
+
"eval_steps_per_second": 15.055,
|
| 2039 |
+
"step": 282
|
| 2040 |
+
},
|
| 2041 |
+
{
|
| 2042 |
+
"epoch": 1.4968799400948498,
|
| 2043 |
+
"grad_norm": 1.015625,
|
| 2044 |
+
"learning_rate": 2.990907357001491e-06,
|
| 2045 |
+
"loss": 1.6462,
|
| 2046 |
+
"step": 283
|
| 2047 |
+
},
|
| 2048 |
+
{
|
| 2049 |
+
"epoch": 1.5022048423329728,
|
| 2050 |
+
"grad_norm": 0.9921875,
|
| 2051 |
+
"learning_rate": 2.9289321881345257e-06,
|
| 2052 |
+
"loss": 1.6232,
|
| 2053 |
+
"step": 284
|
| 2054 |
+
},
|
| 2055 |
+
{
|
| 2056 |
+
"epoch": 1.5075297445710958,
|
| 2057 |
+
"grad_norm": 1.015625,
|
| 2058 |
+
"learning_rate": 2.867495508458186e-06,
|
| 2059 |
+
"loss": 1.7229,
|
| 2060 |
+
"step": 285
|
| 2061 |
+
},
|
| 2062 |
+
{
|
| 2063 |
+
"epoch": 1.5128546468092188,
|
| 2064 |
+
"grad_norm": 1.03125,
|
| 2065 |
+
"learning_rate": 2.8066019966134907e-06,
|
| 2066 |
+
"loss": 1.6581,
|
| 2067 |
+
"step": 286
|
| 2068 |
+
},
|
| 2069 |
+
{
|
| 2070 |
+
"epoch": 1.5181795490473418,
|
| 2071 |
+
"grad_norm": 0.98046875,
|
| 2072 |
+
"learning_rate": 2.746256289877126e-06,
|
| 2073 |
+
"loss": 1.6565,
|
| 2074 |
+
"step": 287
|
| 2075 |
+
},
|
| 2076 |
+
{
|
| 2077 |
+
"epoch": 1.5235044512854645,
|
| 2078 |
+
"grad_norm": 1.015625,
|
| 2079 |
+
"learning_rate": 2.6864629838082957e-06,
|
| 2080 |
+
"loss": 1.646,
|
| 2081 |
+
"step": 288
|
| 2082 |
+
},
|
| 2083 |
+
{
|
| 2084 |
+
"epoch": 1.5288293535235877,
|
| 2085 |
+
"grad_norm": 0.97265625,
|
| 2086 |
+
"learning_rate": 2.6272266318987606e-06,
|
| 2087 |
+
"loss": 1.6165,
|
| 2088 |
+
"step": 289
|
| 2089 |
+
},
|
| 2090 |
+
{
|
| 2091 |
+
"epoch": 1.5341542557617105,
|
| 2092 |
+
"grad_norm": 0.953125,
|
| 2093 |
+
"learning_rate": 2.5685517452260566e-06,
|
| 2094 |
+
"loss": 1.6342,
|
| 2095 |
+
"step": 290
|
| 2096 |
+
},
|
| 2097 |
+
{
|
| 2098 |
+
"epoch": 1.5394791579998337,
|
| 2099 |
+
"grad_norm": 1.0390625,
|
| 2100 |
+
"learning_rate": 2.5104427921099783e-06,
|
| 2101 |
+
"loss": 1.6765,
|
| 2102 |
+
"step": 291
|
| 2103 |
+
},
|
| 2104 |
+
{
|
| 2105 |
+
"epoch": 1.5448040602379565,
|
| 2106 |
+
"grad_norm": 1.0078125,
|
| 2107 |
+
"learning_rate": 2.45290419777228e-06,
|
| 2108 |
+
"loss": 1.6634,
|
| 2109 |
+
"step": 292
|
| 2110 |
+
},
|
| 2111 |
+
{
|
| 2112 |
+
"epoch": 1.5501289624760797,
|
| 2113 |
+
"grad_norm": 0.98828125,
|
| 2114 |
+
"learning_rate": 2.395940343999691e-06,
|
| 2115 |
+
"loss": 1.6478,
|
| 2116 |
+
"step": 293
|
| 2117 |
+
},
|
| 2118 |
+
{
|
| 2119 |
+
"epoch": 1.5554538647142024,
|
| 2120 |
+
"grad_norm": 1.0546875,
|
| 2121 |
+
"learning_rate": 2.339555568810221e-06,
|
| 2122 |
+
"loss": 1.6321,
|
| 2123 |
+
"step": 294
|
| 2124 |
+
},
|
| 2125 |
+
{
|
| 2126 |
+
"epoch": 1.5607787669523256,
|
| 2127 |
+
"grad_norm": 1.03125,
|
| 2128 |
+
"learning_rate": 2.2837541661228024e-06,
|
| 2129 |
+
"loss": 1.7033,
|
| 2130 |
+
"step": 295
|
| 2131 |
+
},
|
| 2132 |
+
{
|
| 2133 |
+
"epoch": 1.5661036691904484,
|
| 2134 |
+
"grad_norm": 0.95703125,
|
| 2135 |
+
"learning_rate": 2.2285403854302912e-06,
|
| 2136 |
+
"loss": 1.562,
|
| 2137 |
+
"step": 296
|
| 2138 |
+
},
|
| 2139 |
+
{
|
| 2140 |
+
"epoch": 1.5714285714285714,
|
| 2141 |
+
"grad_norm": 1.0078125,
|
| 2142 |
+
"learning_rate": 2.173918431475861e-06,
|
| 2143 |
+
"loss": 1.6686,
|
| 2144 |
+
"step": 297
|
| 2145 |
+
},
|
| 2146 |
+
{
|
| 2147 |
+
"epoch": 1.5767534736666944,
|
| 2148 |
+
"grad_norm": 1.015625,
|
| 2149 |
+
"learning_rate": 2.119892463932781e-06,
|
| 2150 |
+
"loss": 1.5676,
|
| 2151 |
+
"step": 298
|
| 2152 |
+
},
|
| 2153 |
+
{
|
| 2154 |
+
"epoch": 1.5820783759048174,
|
| 2155 |
+
"grad_norm": 0.9921875,
|
| 2156 |
+
"learning_rate": 2.0664665970876496e-06,
|
| 2157 |
+
"loss": 1.6761,
|
| 2158 |
+
"step": 299
|
| 2159 |
+
},
|
| 2160 |
+
{
|
| 2161 |
+
"epoch": 1.5874032781429404,
|
| 2162 |
+
"grad_norm": 1.015625,
|
| 2163 |
+
"learning_rate": 2.013644899527074e-06,
|
| 2164 |
+
"loss": 1.6786,
|
| 2165 |
+
"step": 300
|
| 2166 |
+
},
|
| 2167 |
+
{
|
| 2168 |
+
"epoch": 1.5927281803810633,
|
| 2169 |
+
"grad_norm": 1.0625,
|
| 2170 |
+
"learning_rate": 1.961431393827827e-06,
|
| 2171 |
+
"loss": 1.6917,
|
| 2172 |
+
"step": 301
|
| 2173 |
+
},
|
| 2174 |
+
{
|
| 2175 |
+
"epoch": 1.5980530826191863,
|
| 2176 |
+
"grad_norm": 1.0,
|
| 2177 |
+
"learning_rate": 1.9098300562505266e-06,
|
| 2178 |
+
"loss": 1.6204,
|
| 2179 |
+
"step": 302
|
| 2180 |
+
},
|
| 2181 |
+
{
|
| 2182 |
+
"epoch": 1.6033779848573093,
|
| 2183 |
+
"grad_norm": 1.046875,
|
| 2184 |
+
"learning_rate": 1.858844816436809e-06,
|
| 2185 |
+
"loss": 1.6689,
|
| 2186 |
+
"step": 303
|
| 2187 |
+
},
|
| 2188 |
+
{
|
| 2189 |
+
"epoch": 1.6087028870954323,
|
| 2190 |
+
"grad_norm": 0.98046875,
|
| 2191 |
+
"learning_rate": 1.808479557110081e-06,
|
| 2192 |
+
"loss": 1.7041,
|
| 2193 |
+
"step": 304
|
| 2194 |
+
},
|
| 2195 |
+
{
|
| 2196 |
+
"epoch": 1.614027789333555,
|
| 2197 |
+
"grad_norm": 0.96875,
|
| 2198 |
+
"learning_rate": 1.7587381137798432e-06,
|
| 2199 |
+
"loss": 1.6656,
|
| 2200 |
+
"step": 305
|
| 2201 |
+
},
|
| 2202 |
+
{
|
| 2203 |
+
"epoch": 1.6193526915716783,
|
| 2204 |
+
"grad_norm": 1.0,
|
| 2205 |
+
"learning_rate": 1.709624274449584e-06,
|
| 2206 |
+
"loss": 1.6269,
|
| 2207 |
+
"step": 306
|
| 2208 |
+
},
|
| 2209 |
+
{
|
| 2210 |
+
"epoch": 1.624677593809801,
|
| 2211 |
+
"grad_norm": 1.03125,
|
| 2212 |
+
"learning_rate": 1.6611417793283192e-06,
|
| 2213 |
+
"loss": 1.5737,
|
| 2214 |
+
"step": 307
|
| 2215 |
+
},
|
| 2216 |
+
{
|
| 2217 |
+
"epoch": 1.6300024960479242,
|
| 2218 |
+
"grad_norm": 1.0078125,
|
| 2219 |
+
"learning_rate": 1.6132943205457607e-06,
|
| 2220 |
+
"loss": 1.722,
|
| 2221 |
+
"step": 308
|
| 2222 |
+
},
|
| 2223 |
+
{
|
| 2224 |
+
"epoch": 1.635327398286047,
|
| 2225 |
+
"grad_norm": 0.99609375,
|
| 2226 |
+
"learning_rate": 1.566085541871145e-06,
|
| 2227 |
+
"loss": 1.683,
|
| 2228 |
+
"step": 309
|
| 2229 |
+
},
|
| 2230 |
+
{
|
| 2231 |
+
"epoch": 1.6406523005241702,
|
| 2232 |
+
"grad_norm": 1.0546875,
|
| 2233 |
+
"learning_rate": 1.5195190384357405e-06,
|
| 2234 |
+
"loss": 1.6843,
|
| 2235 |
+
"step": 310
|
| 2236 |
+
},
|
| 2237 |
+
{
|
| 2238 |
+
"epoch": 1.645977202762293,
|
| 2239 |
+
"grad_norm": 1.0390625,
|
| 2240 |
+
"learning_rate": 1.4735983564590784e-06,
|
| 2241 |
+
"loss": 1.6979,
|
| 2242 |
+
"step": 311
|
| 2243 |
+
},
|
| 2244 |
+
{
|
| 2245 |
+
"epoch": 1.651302105000416,
|
| 2246 |
+
"grad_norm": 1.015625,
|
| 2247 |
+
"learning_rate": 1.4283269929788779e-06,
|
| 2248 |
+
"loss": 1.6396,
|
| 2249 |
+
"step": 312
|
| 2250 |
+
},
|
| 2251 |
+
{
|
| 2252 |
+
"epoch": 1.656627007238539,
|
| 2253 |
+
"grad_norm": 0.99609375,
|
| 2254 |
+
"learning_rate": 1.3837083955847418e-06,
|
| 2255 |
+
"loss": 1.6738,
|
| 2256 |
+
"step": 313
|
| 2257 |
+
},
|
| 2258 |
+
{
|
| 2259 |
+
"epoch": 1.661951909476662,
|
| 2260 |
+
"grad_norm": 0.94921875,
|
| 2261 |
+
"learning_rate": 1.339745962155613e-06,
|
| 2262 |
+
"loss": 1.6566,
|
| 2263 |
+
"step": 314
|
| 2264 |
+
},
|
| 2265 |
+
{
|
| 2266 |
+
"epoch": 1.667276811714785,
|
| 2267 |
+
"grad_norm": 1.0078125,
|
| 2268 |
+
"learning_rate": 1.2964430406010032e-06,
|
| 2269 |
+
"loss": 1.6679,
|
| 2270 |
+
"step": 315
|
| 2271 |
+
},
|
| 2272 |
+
{
|
| 2273 |
+
"epoch": 1.672601713952908,
|
| 2274 |
+
"grad_norm": 1.0,
|
| 2275 |
+
"learning_rate": 1.2538029286060428e-06,
|
| 2276 |
+
"loss": 1.6892,
|
| 2277 |
+
"step": 316
|
| 2278 |
+
},
|
| 2279 |
+
{
|
| 2280 |
+
"epoch": 1.677926616191031,
|
| 2281 |
+
"grad_norm": 1.0390625,
|
| 2282 |
+
"learning_rate": 1.2118288733803474e-06,
|
| 2283 |
+
"loss": 1.6914,
|
| 2284 |
+
"step": 317
|
| 2285 |
+
},
|
| 2286 |
+
{
|
| 2287 |
+
"epoch": 1.6832515184291539,
|
| 2288 |
+
"grad_norm": 1.015625,
|
| 2289 |
+
"learning_rate": 1.1705240714107301e-06,
|
| 2290 |
+
"loss": 1.5954,
|
| 2291 |
+
"step": 318
|
| 2292 |
+
},
|
| 2293 |
+
{
|
| 2294 |
+
"epoch": 1.6885764206672769,
|
| 2295 |
+
"grad_norm": 0.9765625,
|
| 2296 |
+
"learning_rate": 1.129891668217783e-06,
|
| 2297 |
+
"loss": 1.633,
|
| 2298 |
+
"step": 319
|
| 2299 |
+
},
|
| 2300 |
+
{
|
| 2301 |
+
"epoch": 1.6939013229053996,
|
| 2302 |
+
"grad_norm": 1.0546875,
|
| 2303 |
+
"learning_rate": 1.0899347581163222e-06,
|
| 2304 |
+
"loss": 1.6976,
|
| 2305 |
+
"step": 320
|
| 2306 |
+
},
|
| 2307 |
+
{
|
| 2308 |
+
"epoch": 1.6992262251435228,
|
| 2309 |
+
"grad_norm": 0.9921875,
|
| 2310 |
+
"learning_rate": 1.0506563839797501e-06,
|
| 2311 |
+
"loss": 1.6283,
|
| 2312 |
+
"step": 321
|
| 2313 |
+
},
|
| 2314 |
+
{
|
| 2315 |
+
"epoch": 1.7045511273816456,
|
| 2316 |
+
"grad_norm": 0.9765625,
|
| 2317 |
+
"learning_rate": 1.012059537008332e-06,
|
| 2318 |
+
"loss": 1.6094,
|
| 2319 |
+
"step": 322
|
| 2320 |
+
},
|
| 2321 |
+
{
|
| 2322 |
+
"epoch": 1.7098760296197688,
|
| 2323 |
+
"grad_norm": 1.0390625,
|
| 2324 |
+
"learning_rate": 9.74147156501396e-07,
|
| 2325 |
+
"loss": 1.6049,
|
| 2326 |
+
"step": 323
|
| 2327 |
+
},
|
| 2328 |
+
{
|
| 2329 |
+
"epoch": 1.7152009318578916,
|
| 2330 |
+
"grad_norm": 1.0546875,
|
| 2331 |
+
"learning_rate": 9.369221296335007e-07,
|
| 2332 |
+
"loss": 1.6529,
|
| 2333 |
+
"step": 324
|
| 2334 |
+
},
|
| 2335 |
+
{
|
| 2336 |
+
"epoch": 1.7205258340960148,
|
| 2337 |
+
"grad_norm": 1.09375,
|
| 2338 |
+
"learning_rate": 9.00387291234569e-07,
|
| 2339 |
+
"loss": 1.7539,
|
| 2340 |
+
"step": 325
|
| 2341 |
+
},
|
| 2342 |
+
{
|
| 2343 |
+
"epoch": 1.7258507363341375,
|
| 2344 |
+
"grad_norm": 1.03125,
|
| 2345 |
+
"learning_rate": 8.645454235739903e-07,
|
| 2346 |
+
"loss": 1.6844,
|
| 2347 |
+
"step": 326
|
| 2348 |
+
},
|
| 2349 |
+
{
|
| 2350 |
+
"epoch": 1.7311756385722605,
|
| 2351 |
+
"grad_norm": 0.98046875,
|
| 2352 |
+
"learning_rate": 8.293992561487596e-07,
|
| 2353 |
+
"loss": 1.6409,
|
| 2354 |
+
"step": 327
|
| 2355 |
+
},
|
| 2356 |
+
{
|
| 2357 |
+
"epoch": 1.7365005408103835,
|
| 2358 |
+
"grad_norm": 1.03125,
|
| 2359 |
+
"learning_rate": 7.949514654755963e-07,
|
| 2360 |
+
"loss": 1.7356,
|
| 2361 |
+
"step": 328
|
| 2362 |
+
},
|
| 2363 |
+
{
|
| 2364 |
+
"epoch": 1.7418254430485065,
|
| 2365 |
+
"grad_norm": 1.0625,
|
| 2366 |
+
"learning_rate": 7.612046748871327e-07,
|
| 2367 |
+
"loss": 1.6681,
|
| 2368 |
+
"step": 329
|
| 2369 |
+
},
|
| 2370 |
+
{
|
| 2371 |
+
"epoch": 1.7418254430485065,
|
| 2372 |
+
"eval_loss": 1.6896613836288452,
|
| 2373 |
+
"eval_runtime": 77.1918,
|
| 2374 |
+
"eval_samples_per_second": 15.287,
|
| 2375 |
+
"eval_steps_per_second": 15.287,
|
| 2376 |
+
"step": 329
|
| 2377 |
+
},
|
| 2378 |
+
{
|
| 2379 |
+
"epoch": 1.7471503452866295,
|
| 2380 |
+
"grad_norm": 1.046875,
|
| 2381 |
+
"learning_rate": 7.281614543321269e-07,
|
| 2382 |
+
"loss": 1.6493,
|
| 2383 |
+
"step": 330
|
| 2384 |
+
},
|
| 2385 |
+
{
|
| 2386 |
+
"epoch": 1.7524752475247525,
|
| 2387 |
+
"grad_norm": 1.0234375,
|
| 2388 |
+
"learning_rate": 6.958243201797554e-07,
|
| 2389 |
+
"loss": 1.6299,
|
| 2390 |
+
"step": 331
|
| 2391 |
+
},
|
| 2392 |
+
{
|
| 2393 |
+
"epoch": 1.7578001497628755,
|
| 2394 |
+
"grad_norm": 1.0234375,
|
| 2395 |
+
"learning_rate": 6.641957350279838e-07,
|
| 2396 |
+
"loss": 1.6882,
|
| 2397 |
+
"step": 332
|
| 2398 |
+
},
|
| 2399 |
+
{
|
| 2400 |
+
"epoch": 1.7631250520009984,
|
| 2401 |
+
"grad_norm": 1.0390625,
|
| 2402 |
+
"learning_rate": 6.332781075160244e-07,
|
| 2403 |
+
"loss": 1.6256,
|
| 2404 |
+
"step": 333
|
| 2405 |
+
},
|
| 2406 |
+
{
|
| 2407 |
+
"epoch": 1.7684499542391214,
|
| 2408 |
+
"grad_norm": 1.0234375,
|
| 2409 |
+
"learning_rate": 6.030737921409169e-07,
|
| 2410 |
+
"loss": 1.5845,
|
| 2411 |
+
"step": 334
|
| 2412 |
+
},
|
| 2413 |
+
{
|
| 2414 |
+
"epoch": 1.7737748564772442,
|
| 2415 |
+
"grad_norm": 0.984375,
|
| 2416 |
+
"learning_rate": 5.735850890782158e-07,
|
| 2417 |
+
"loss": 1.6066,
|
| 2418 |
+
"step": 335
|
| 2419 |
+
},
|
| 2420 |
+
{
|
| 2421 |
+
"epoch": 1.7790997587153674,
|
| 2422 |
+
"grad_norm": 1.03125,
|
| 2423 |
+
"learning_rate": 5.448142440068316e-07,
|
| 2424 |
+
"loss": 1.6751,
|
| 2425 |
+
"step": 336
|
| 2426 |
+
},
|
| 2427 |
+
{
|
| 2428 |
+
"epoch": 1.7844246609534902,
|
| 2429 |
+
"grad_norm": 0.97265625,
|
| 2430 |
+
"learning_rate": 5.167634479380068e-07,
|
| 2431 |
+
"loss": 1.6368,
|
| 2432 |
+
"step": 337
|
| 2433 |
+
},
|
| 2434 |
+
{
|
| 2435 |
+
"epoch": 1.7897495631916134,
|
| 2436 |
+
"grad_norm": 1.0234375,
|
| 2437 |
+
"learning_rate": 4.894348370484648e-07,
|
| 2438 |
+
"loss": 1.7731,
|
| 2439 |
+
"step": 338
|
| 2440 |
+
},
|
| 2441 |
+
{
|
| 2442 |
+
"epoch": 1.7950744654297361,
|
| 2443 |
+
"grad_norm": 1.03125,
|
| 2444 |
+
"learning_rate": 4.628304925177318e-07,
|
| 2445 |
+
"loss": 1.6714,
|
| 2446 |
+
"step": 339
|
| 2447 |
+
},
|
| 2448 |
+
{
|
| 2449 |
+
"epoch": 1.8003993676678594,
|
| 2450 |
+
"grad_norm": 1.1015625,
|
| 2451 |
+
"learning_rate": 4.3695244036964567e-07,
|
| 2452 |
+
"loss": 1.7154,
|
| 2453 |
+
"step": 340
|
| 2454 |
+
},
|
| 2455 |
+
{
|
| 2456 |
+
"epoch": 1.8057242699059821,
|
| 2457 |
+
"grad_norm": 0.94921875,
|
| 2458 |
+
"learning_rate": 4.118026513180695e-07,
|
| 2459 |
+
"loss": 1.5225,
|
| 2460 |
+
"step": 341
|
| 2461 |
+
},
|
| 2462 |
+
{
|
| 2463 |
+
"epoch": 1.8110491721441053,
|
| 2464 |
+
"grad_norm": 1.0,
|
| 2465 |
+
"learning_rate": 3.8738304061681107e-07,
|
| 2466 |
+
"loss": 1.6143,
|
| 2467 |
+
"step": 342
|
| 2468 |
+
},
|
| 2469 |
+
{
|
| 2470 |
+
"epoch": 1.816374074382228,
|
| 2471 |
+
"grad_norm": 1.0,
|
| 2472 |
+
"learning_rate": 3.6369546791377054e-07,
|
| 2473 |
+
"loss": 1.6301,
|
| 2474 |
+
"step": 343
|
| 2475 |
+
},
|
| 2476 |
+
{
|
| 2477 |
+
"epoch": 1.821698976620351,
|
| 2478 |
+
"grad_norm": 0.95703125,
|
| 2479 |
+
"learning_rate": 3.4074173710931804e-07,
|
| 2480 |
+
"loss": 1.5861,
|
| 2481 |
+
"step": 344
|
| 2482 |
+
},
|
| 2483 |
+
{
|
| 2484 |
+
"epoch": 1.827023878858474,
|
| 2485 |
+
"grad_norm": 0.9921875,
|
| 2486 |
+
"learning_rate": 3.185235962189237e-07,
|
| 2487 |
+
"loss": 1.6469,
|
| 2488 |
+
"step": 345
|
| 2489 |
+
},
|
| 2490 |
+
{
|
| 2491 |
+
"epoch": 1.832348781096597,
|
| 2492 |
+
"grad_norm": 0.95703125,
|
| 2493 |
+
"learning_rate": 2.970427372400353e-07,
|
| 2494 |
+
"loss": 1.5941,
|
| 2495 |
+
"step": 346
|
| 2496 |
+
},
|
| 2497 |
+
{
|
| 2498 |
+
"epoch": 1.83767368333472,
|
| 2499 |
+
"grad_norm": 0.9609375,
|
| 2500 |
+
"learning_rate": 2.7630079602323447e-07,
|
| 2501 |
+
"loss": 1.6405,
|
| 2502 |
+
"step": 347
|
| 2503 |
+
},
|
| 2504 |
+
{
|
| 2505 |
+
"epoch": 1.842998585572843,
|
| 2506 |
+
"grad_norm": 1.03125,
|
| 2507 |
+
"learning_rate": 2.5629935214764866e-07,
|
| 2508 |
+
"loss": 1.6329,
|
| 2509 |
+
"step": 348
|
| 2510 |
+
},
|
| 2511 |
+
{
|
| 2512 |
+
"epoch": 1.848323487810966,
|
| 2513 |
+
"grad_norm": 1.03125,
|
| 2514 |
+
"learning_rate": 2.370399288006664e-07,
|
| 2515 |
+
"loss": 1.62,
|
| 2516 |
+
"step": 349
|
| 2517 |
+
},
|
| 2518 |
+
{
|
| 2519 |
+
"epoch": 1.853648390049089,
|
| 2520 |
+
"grad_norm": 1.0078125,
|
| 2521 |
+
"learning_rate": 2.1852399266194312e-07,
|
| 2522 |
+
"loss": 1.5775,
|
| 2523 |
+
"step": 350
|
| 2524 |
+
},
|
| 2525 |
+
{
|
| 2526 |
+
"epoch": 1.858973292287212,
|
| 2527 |
+
"grad_norm": 0.984375,
|
| 2528 |
+
"learning_rate": 2.0075295379170413e-07,
|
| 2529 |
+
"loss": 1.6226,
|
| 2530 |
+
"step": 351
|
| 2531 |
+
},
|
| 2532 |
+
{
|
| 2533 |
+
"epoch": 1.8642981945253347,
|
| 2534 |
+
"grad_norm": 1.0234375,
|
| 2535 |
+
"learning_rate": 1.8372816552336025e-07,
|
| 2536 |
+
"loss": 1.5957,
|
| 2537 |
+
"step": 352
|
| 2538 |
+
},
|
| 2539 |
+
{
|
| 2540 |
+
"epoch": 1.869623096763458,
|
| 2541 |
+
"grad_norm": 1.015625,
|
| 2542 |
+
"learning_rate": 1.6745092436045495e-07,
|
| 2543 |
+
"loss": 1.6389,
|
| 2544 |
+
"step": 353
|
| 2545 |
+
},
|
| 2546 |
+
{
|
| 2547 |
+
"epoch": 1.8749479990015807,
|
| 2548 |
+
"grad_norm": 1.0546875,
|
| 2549 |
+
"learning_rate": 1.519224698779198e-07,
|
| 2550 |
+
"loss": 1.6588,
|
| 2551 |
+
"step": 354
|
| 2552 |
+
},
|
| 2553 |
+
{
|
| 2554 |
+
"epoch": 1.880272901239704,
|
| 2555 |
+
"grad_norm": 0.9765625,
|
| 2556 |
+
"learning_rate": 1.3714398462768563e-07,
|
| 2557 |
+
"loss": 1.5915,
|
| 2558 |
+
"step": 355
|
| 2559 |
+
},
|
| 2560 |
+
{
|
| 2561 |
+
"epoch": 1.8855978034778267,
|
| 2562 |
+
"grad_norm": 1.03125,
|
| 2563 |
+
"learning_rate": 1.231165940486234e-07,
|
| 2564 |
+
"loss": 1.6454,
|
| 2565 |
+
"step": 356
|
| 2566 |
+
},
|
| 2567 |
+
{
|
| 2568 |
+
"epoch": 1.89092270571595,
|
| 2569 |
+
"grad_norm": 1.03125,
|
| 2570 |
+
"learning_rate": 1.0984136638083176e-07,
|
| 2571 |
+
"loss": 1.6706,
|
| 2572 |
+
"step": 357
|
| 2573 |
+
},
|
| 2574 |
+
{
|
| 2575 |
+
"epoch": 1.8962476079540727,
|
| 2576 |
+
"grad_norm": 1.046875,
|
| 2577 |
+
"learning_rate": 9.731931258429638e-08,
|
| 2578 |
+
"loss": 1.6479,
|
| 2579 |
+
"step": 358
|
| 2580 |
+
},
|
| 2581 |
+
{
|
| 2582 |
+
"epoch": 1.9015725101921956,
|
| 2583 |
+
"grad_norm": 0.93359375,
|
| 2584 |
+
"learning_rate": 8.555138626189619e-08,
|
| 2585 |
+
"loss": 1.6405,
|
| 2586 |
+
"step": 359
|
| 2587 |
+
},
|
| 2588 |
+
{
|
| 2589 |
+
"epoch": 1.9068974124303186,
|
| 2590 |
+
"grad_norm": 1.0234375,
|
| 2591 |
+
"learning_rate": 7.453848358678018e-08,
|
| 2592 |
+
"loss": 1.6638,
|
| 2593 |
+
"step": 360
|
| 2594 |
+
},
|
| 2595 |
+
{
|
| 2596 |
+
"epoch": 1.9122223146684416,
|
| 2597 |
+
"grad_norm": 1.046875,
|
| 2598 |
+
"learning_rate": 6.428144323412544e-08,
|
| 2599 |
+
"loss": 1.6709,
|
| 2600 |
+
"step": 361
|
| 2601 |
+
},
|
| 2602 |
+
{
|
| 2603 |
+
"epoch": 1.9175472169065646,
|
| 2604 |
+
"grad_norm": 1.0546875,
|
| 2605 |
+
"learning_rate": 5.4781046317267103e-08,
|
| 2606 |
+
"loss": 1.6191,
|
| 2607 |
+
"step": 362
|
| 2608 |
+
},
|
| 2609 |
+
{
|
| 2610 |
+
"epoch": 1.9228721191446876,
|
| 2611 |
+
"grad_norm": 1.0390625,
|
| 2612 |
+
"learning_rate": 4.603801632821148e-08,
|
| 2613 |
+
"loss": 1.7144,
|
| 2614 |
+
"step": 363
|
| 2615 |
+
},
|
| 2616 |
+
{
|
| 2617 |
+
"epoch": 1.9281970213828106,
|
| 2618 |
+
"grad_norm": 1.0078125,
|
| 2619 |
+
"learning_rate": 3.805301908254455e-08,
|
| 2620 |
+
"loss": 1.6358,
|
| 2621 |
+
"step": 364
|
| 2622 |
+
},
|
| 2623 |
+
{
|
| 2624 |
+
"epoch": 1.9335219236209336,
|
| 2625 |
+
"grad_norm": 1.0078125,
|
| 2626 |
+
"learning_rate": 3.082666266872036e-08,
|
| 2627 |
+
"loss": 1.6961,
|
| 2628 |
+
"step": 365
|
| 2629 |
+
},
|
| 2630 |
+
{
|
| 2631 |
+
"epoch": 1.9388468258590565,
|
| 2632 |
+
"grad_norm": 1.046875,
|
| 2633 |
+
"learning_rate": 2.4359497401758026e-08,
|
| 2634 |
+
"loss": 1.6074,
|
| 2635 |
+
"step": 366
|
| 2636 |
+
},
|
| 2637 |
+
{
|
| 2638 |
+
"epoch": 1.9441717280971793,
|
| 2639 |
+
"grad_norm": 1.03125,
|
| 2640 |
+
"learning_rate": 1.86520157813308e-08,
|
| 2641 |
+
"loss": 1.5905,
|
| 2642 |
+
"step": 367
|
| 2643 |
+
},
|
| 2644 |
+
{
|
| 2645 |
+
"epoch": 1.9494966303353025,
|
| 2646 |
+
"grad_norm": 0.97265625,
|
| 2647 |
+
"learning_rate": 1.370465245426167e-08,
|
| 2648 |
+
"loss": 1.6317,
|
| 2649 |
+
"step": 368
|
| 2650 |
+
},
|
| 2651 |
+
{
|
| 2652 |
+
"epoch": 1.9548215325734253,
|
| 2653 |
+
"grad_norm": 0.98828125,
|
| 2654 |
+
"learning_rate": 9.517784181422018e-09,
|
| 2655 |
+
"loss": 1.5783,
|
| 2656 |
+
"step": 369
|
| 2657 |
+
},
|
| 2658 |
+
{
|
| 2659 |
+
"epoch": 1.9601464348115485,
|
| 2660 |
+
"grad_norm": 0.98828125,
|
| 2661 |
+
"learning_rate": 6.091729809042379e-09,
|
| 2662 |
+
"loss": 1.7104,
|
| 2663 |
+
"step": 370
|
| 2664 |
+
},
|
| 2665 |
+
{
|
| 2666 |
+
"epoch": 1.9654713370496713,
|
| 2667 |
+
"grad_norm": 1.0546875,
|
| 2668 |
+
"learning_rate": 3.4267502444274013e-09,
|
| 2669 |
+
"loss": 1.7328,
|
| 2670 |
+
"step": 371
|
| 2671 |
+
},
|
| 2672 |
+
{
|
| 2673 |
+
"epoch": 1.9707962392877945,
|
| 2674 |
+
"grad_norm": 1.03125,
|
| 2675 |
+
"learning_rate": 1.5230484360873043e-09,
|
| 2676 |
+
"loss": 1.6749,
|
| 2677 |
+
"step": 372
|
| 2678 |
+
},
|
| 2679 |
+
{
|
| 2680 |
+
"epoch": 1.9761211415259172,
|
| 2681 |
+
"grad_norm": 0.98046875,
|
| 2682 |
+
"learning_rate": 3.807693582869032e-10,
|
| 2683 |
+
"loss": 1.5706,
|
| 2684 |
+
"step": 373
|
| 2685 |
+
},
|
| 2686 |
+
{
|
| 2687 |
+
"epoch": 1.9814460437640404,
|
| 2688 |
+
"grad_norm": 1.0703125,
|
| 2689 |
+
"learning_rate": 0.0,
|
| 2690 |
+
"loss": 1.662,
|
| 2691 |
+
"step": 374
|
| 2692 |
+
}
|
| 2693 |
+
],
|
| 2694 |
+
"logging_steps": 1,
|
| 2695 |
+
"max_steps": 374,
|
| 2696 |
+
"num_input_tokens_seen": 0,
|
| 2697 |
+
"num_train_epochs": 2,
|
| 2698 |
+
"save_steps": 187,
|
| 2699 |
+
"stateful_callbacks": {
|
| 2700 |
+
"TrainerControl": {
|
| 2701 |
+
"args": {
|
| 2702 |
+
"should_epoch_stop": false,
|
| 2703 |
+
"should_evaluate": false,
|
| 2704 |
+
"should_log": false,
|
| 2705 |
+
"should_save": true,
|
| 2706 |
+
"should_training_stop": true
|
| 2707 |
+
},
|
| 2708 |
+
"attributes": {}
|
| 2709 |
+
}
|
| 2710 |
+
},
|
| 2711 |
+
"total_flos": 3.8540467560146534e+17,
|
| 2712 |
+
"train_batch_size": 1,
|
| 2713 |
+
"trial_name": null,
|
| 2714 |
+
"trial_params": null
|
| 2715 |
+
}
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:35b4b979a5cff4130ad7b93748ff9ed06a157ef9e649ffec51ce3ecfa2d21603
|
| 3 |
+
size 6136
|
vocab.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|