Training in progress, epoch 37, checkpoint
Browse files- checkpoint-19573/README.md +9 -0
- checkpoint-19573/adapter_config.json +25 -0
- checkpoint-19573/adapter_model.safetensors +3 -0
- checkpoint-19573/added_tokens.json +5 -0
- checkpoint-19573/merges.txt +0 -0
- checkpoint-19573/optimizer.pt +3 -0
- checkpoint-19573/rng_state_0.pth +3 -0
- checkpoint-19573/rng_state_1.pth +3 -0
- checkpoint-19573/rng_state_2.pth +3 -0
- checkpoint-19573/rng_state_3.pth +3 -0
- checkpoint-19573/scheduler.pt +3 -0
- checkpoint-19573/special_tokens_map.json +14 -0
- checkpoint-19573/tokenizer.json +0 -0
- checkpoint-19573/tokenizer_config.json +43 -0
- checkpoint-19573/trainer_state.json +1953 -0
- checkpoint-19573/training_args.bin +3 -0
- checkpoint-19573/vocab.json +0 -0
checkpoint-19573/README.md
ADDED
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
---
|
2 |
+
library_name: peft
|
3 |
+
---
|
4 |
+
## Training procedure
|
5 |
+
|
6 |
+
### Framework versions
|
7 |
+
|
8 |
+
|
9 |
+
- PEFT 0.5.0
|
checkpoint-19573/adapter_config.json
ADDED
@@ -0,0 +1,25 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"auto_mapping": null,
|
3 |
+
"base_model_name_or_path": "Qwen/Qwen1.5-4B",
|
4 |
+
"bias": "none",
|
5 |
+
"fan_in_fan_out": false,
|
6 |
+
"inference_mode": true,
|
7 |
+
"init_lora_weights": true,
|
8 |
+
"layers_pattern": null,
|
9 |
+
"layers_to_transform": null,
|
10 |
+
"lora_alpha": 16,
|
11 |
+
"lora_dropout": 0.05,
|
12 |
+
"modules_to_save": null,
|
13 |
+
"peft_type": "LORA",
|
14 |
+
"r": 16,
|
15 |
+
"revision": null,
|
16 |
+
"target_modules": [
|
17 |
+
"q_proj",
|
18 |
+
"k_proj",
|
19 |
+
"v_proj",
|
20 |
+
"down_proj",
|
21 |
+
"gate_proj",
|
22 |
+
"up_proj"
|
23 |
+
],
|
24 |
+
"task_type": "CAUSAL_LM"
|
25 |
+
}
|
checkpoint-19573/adapter_model.safetensors
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:7c0ba356c9742ccba9203383bed339a04d89792a1a9c54659df949a90979ed07
|
3 |
+
size 112130416
|
checkpoint-19573/added_tokens.json
ADDED
@@ -0,0 +1,5 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"<|endoftext|>": 151643,
|
3 |
+
"<|im_end|>": 151645,
|
4 |
+
"<|im_start|>": 151644
|
5 |
+
}
|
checkpoint-19573/merges.txt
ADDED
The diff for this file is too large to render.
See raw diff
|
|
checkpoint-19573/optimizer.pt
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:cfd959743fc4f16b61b2b7d242521eae5c7246a977dbc970e61c1f503ec5114b
|
3 |
+
size 224537202
|
checkpoint-19573/rng_state_0.pth
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:0258a2f8708fb6b80e3655ac83585f0102689fd86a07a25207691112df50ba01
|
3 |
+
size 15024
|
checkpoint-19573/rng_state_1.pth
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:0c6ada00d5b172e98821ef26d8ff51fdf8712349212f9e989e381bddaef99c16
|
3 |
+
size 15024
|
checkpoint-19573/rng_state_2.pth
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:a4f2905bd912f5946c902ea81a185ba7639ffaccf8bd29cf0f1485ddb3c6d273
|
3 |
+
size 15024
|
checkpoint-19573/rng_state_3.pth
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:496c65188546c88cda352f1d9d5533b07c60057f36a8d6adbfcb9b33a9fa701b
|
3 |
+
size 15024
|
checkpoint-19573/scheduler.pt
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:0d2ea2497fb0fbcf074c0d32f5ecc3d35d5b74f3f4dc83b4a7abdfbe97a6dc80
|
3 |
+
size 1064
|
checkpoint-19573/special_tokens_map.json
ADDED
@@ -0,0 +1,14 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
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": "<|endoftext|>"
|
14 |
+
}
|
checkpoint-19573/tokenizer.json
ADDED
The diff for this file is too large to render.
See raw diff
|
|
checkpoint-19573/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 loop.first and messages[0]['role'] != 'system' %}{{ '<|im_start|>system\nYou are a helpful assistant<|im_end|>\n' }}{% endif %}{{'<|im_start|>' + message['role'] + '\n' + message['content'] + '<|im_end|>' + '\n'}}{% endfor %}{% if add_generation_prompt %}{{ '<|im_start|>assistant\n' }}{% endif %}",
|
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 |
+
}
|
checkpoint-19573/trainer_state.json
ADDED
@@ -0,0 +1,1953 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"best_metric": null,
|
3 |
+
"best_model_checkpoint": null,
|
4 |
+
"epoch": 37.0,
|
5 |
+
"eval_steps": 500,
|
6 |
+
"global_step": 19573,
|
7 |
+
"is_hyper_param_search": false,
|
8 |
+
"is_local_process_zero": true,
|
9 |
+
"is_world_process_zero": true,
|
10 |
+
"log_history": [
|
11 |
+
{
|
12 |
+
"epoch": 0.1890359168241966,
|
13 |
+
"grad_norm": 0.5558584332466125,
|
14 |
+
"learning_rate": 3e-05,
|
15 |
+
"loss": 2.1103,
|
16 |
+
"step": 100
|
17 |
+
},
|
18 |
+
{
|
19 |
+
"epoch": 0.3780718336483932,
|
20 |
+
"grad_norm": 0.4733140468597412,
|
21 |
+
"learning_rate": 3e-05,
|
22 |
+
"loss": 1.931,
|
23 |
+
"step": 200
|
24 |
+
},
|
25 |
+
{
|
26 |
+
"epoch": 0.5671077504725898,
|
27 |
+
"grad_norm": 0.48580050468444824,
|
28 |
+
"learning_rate": 3e-05,
|
29 |
+
"loss": 1.8948,
|
30 |
+
"step": 300
|
31 |
+
},
|
32 |
+
{
|
33 |
+
"epoch": 0.7561436672967864,
|
34 |
+
"grad_norm": 0.40655234456062317,
|
35 |
+
"learning_rate": 3e-05,
|
36 |
+
"loss": 1.8988,
|
37 |
+
"step": 400
|
38 |
+
},
|
39 |
+
{
|
40 |
+
"epoch": 0.945179584120983,
|
41 |
+
"grad_norm": 0.6303955316543579,
|
42 |
+
"learning_rate": 3e-05,
|
43 |
+
"loss": 1.8745,
|
44 |
+
"step": 500
|
45 |
+
},
|
46 |
+
{
|
47 |
+
"epoch": 1.0,
|
48 |
+
"eval_accuracy": 0.5701538461538461,
|
49 |
+
"eval_loss": 1.6455246210098267,
|
50 |
+
"eval_runtime": 5.0805,
|
51 |
+
"eval_samples_per_second": 98.415,
|
52 |
+
"eval_steps_per_second": 12.4,
|
53 |
+
"step": 529
|
54 |
+
},
|
55 |
+
{
|
56 |
+
"epoch": 1.0,
|
57 |
+
"eval_exact_match": 15.0,
|
58 |
+
"eval_f1": 23.927619047619068,
|
59 |
+
"step": 529
|
60 |
+
},
|
61 |
+
{
|
62 |
+
"epoch": 1.1342155009451795,
|
63 |
+
"grad_norm": 0.4278818666934967,
|
64 |
+
"learning_rate": 3e-05,
|
65 |
+
"loss": 1.8798,
|
66 |
+
"step": 600
|
67 |
+
},
|
68 |
+
{
|
69 |
+
"epoch": 1.3232514177693762,
|
70 |
+
"grad_norm": 0.449830025434494,
|
71 |
+
"learning_rate": 3e-05,
|
72 |
+
"loss": 1.8613,
|
73 |
+
"step": 700
|
74 |
+
},
|
75 |
+
{
|
76 |
+
"epoch": 1.5122873345935728,
|
77 |
+
"grad_norm": 0.5130725502967834,
|
78 |
+
"learning_rate": 3e-05,
|
79 |
+
"loss": 1.8709,
|
80 |
+
"step": 800
|
81 |
+
},
|
82 |
+
{
|
83 |
+
"epoch": 1.7013232514177694,
|
84 |
+
"grad_norm": 0.4889715611934662,
|
85 |
+
"learning_rate": 3e-05,
|
86 |
+
"loss": 1.8448,
|
87 |
+
"step": 900
|
88 |
+
},
|
89 |
+
{
|
90 |
+
"epoch": 1.8903591682419658,
|
91 |
+
"grad_norm": 11.186380386352539,
|
92 |
+
"learning_rate": 3e-05,
|
93 |
+
"loss": 1.8764,
|
94 |
+
"step": 1000
|
95 |
+
},
|
96 |
+
{
|
97 |
+
"epoch": 2.0,
|
98 |
+
"eval_accuracy": 0.5723589743589743,
|
99 |
+
"eval_loss": 1.6143561601638794,
|
100 |
+
"eval_runtime": 5.0466,
|
101 |
+
"eval_samples_per_second": 99.077,
|
102 |
+
"eval_steps_per_second": 12.484,
|
103 |
+
"step": 1058
|
104 |
+
},
|
105 |
+
{
|
106 |
+
"epoch": 2.0,
|
107 |
+
"eval_exact_match": 15.6,
|
108 |
+
"eval_f1": 25.825238095238113,
|
109 |
+
"step": 1058
|
110 |
+
},
|
111 |
+
{
|
112 |
+
"epoch": 2.0793950850661624,
|
113 |
+
"grad_norm": 0.5414931774139404,
|
114 |
+
"learning_rate": 3e-05,
|
115 |
+
"loss": 1.8398,
|
116 |
+
"step": 1100
|
117 |
+
},
|
118 |
+
{
|
119 |
+
"epoch": 2.268431001890359,
|
120 |
+
"grad_norm": 7.441005706787109,
|
121 |
+
"learning_rate": 3e-05,
|
122 |
+
"loss": 1.8348,
|
123 |
+
"step": 1200
|
124 |
+
},
|
125 |
+
{
|
126 |
+
"epoch": 2.4574669187145557,
|
127 |
+
"grad_norm": 24.312145233154297,
|
128 |
+
"learning_rate": 3e-05,
|
129 |
+
"loss": 1.8421,
|
130 |
+
"step": 1300
|
131 |
+
},
|
132 |
+
{
|
133 |
+
"epoch": 2.6465028355387523,
|
134 |
+
"grad_norm": 0.7407662868499756,
|
135 |
+
"learning_rate": 3e-05,
|
136 |
+
"loss": 1.8292,
|
137 |
+
"step": 1400
|
138 |
+
},
|
139 |
+
{
|
140 |
+
"epoch": 2.835538752362949,
|
141 |
+
"grad_norm": 0.9653642177581787,
|
142 |
+
"learning_rate": 3e-05,
|
143 |
+
"loss": 1.828,
|
144 |
+
"step": 1500
|
145 |
+
},
|
146 |
+
{
|
147 |
+
"epoch": 3.0,
|
148 |
+
"eval_accuracy": 0.5734358974358974,
|
149 |
+
"eval_loss": 1.617864727973938,
|
150 |
+
"eval_runtime": 5.2153,
|
151 |
+
"eval_samples_per_second": 95.872,
|
152 |
+
"eval_steps_per_second": 12.08,
|
153 |
+
"step": 1587
|
154 |
+
},
|
155 |
+
{
|
156 |
+
"epoch": 3.0,
|
157 |
+
"eval_exact_match": 16.2,
|
158 |
+
"eval_f1": 25.811428571428586,
|
159 |
+
"step": 1587
|
160 |
+
},
|
161 |
+
{
|
162 |
+
"epoch": 3.0245746691871456,
|
163 |
+
"grad_norm": 450.6055603027344,
|
164 |
+
"learning_rate": 3e-05,
|
165 |
+
"loss": 1.8204,
|
166 |
+
"step": 1600
|
167 |
+
},
|
168 |
+
{
|
169 |
+
"epoch": 3.213610586011342,
|
170 |
+
"grad_norm": 2.1127028465270996,
|
171 |
+
"learning_rate": 3e-05,
|
172 |
+
"loss": 1.8361,
|
173 |
+
"step": 1700
|
174 |
+
},
|
175 |
+
{
|
176 |
+
"epoch": 3.402646502835539,
|
177 |
+
"grad_norm": 564.9320068359375,
|
178 |
+
"learning_rate": 3e-05,
|
179 |
+
"loss": 1.8338,
|
180 |
+
"step": 1800
|
181 |
+
},
|
182 |
+
{
|
183 |
+
"epoch": 3.5916824196597354,
|
184 |
+
"grad_norm": 52.5078010559082,
|
185 |
+
"learning_rate": 3e-05,
|
186 |
+
"loss": 1.8075,
|
187 |
+
"step": 1900
|
188 |
+
},
|
189 |
+
{
|
190 |
+
"epoch": 3.780718336483932,
|
191 |
+
"grad_norm": 27.87557601928711,
|
192 |
+
"learning_rate": 3e-05,
|
193 |
+
"loss": 1.8175,
|
194 |
+
"step": 2000
|
195 |
+
},
|
196 |
+
{
|
197 |
+
"epoch": 3.9697542533081287,
|
198 |
+
"grad_norm": 4.810656547546387,
|
199 |
+
"learning_rate": 3e-05,
|
200 |
+
"loss": 1.8218,
|
201 |
+
"step": 2100
|
202 |
+
},
|
203 |
+
{
|
204 |
+
"epoch": 4.0,
|
205 |
+
"eval_accuracy": 0.5728205128205128,
|
206 |
+
"eval_loss": 1.6224406957626343,
|
207 |
+
"eval_runtime": 5.0588,
|
208 |
+
"eval_samples_per_second": 98.838,
|
209 |
+
"eval_steps_per_second": 12.454,
|
210 |
+
"step": 2116
|
211 |
+
},
|
212 |
+
{
|
213 |
+
"epoch": 4.0,
|
214 |
+
"eval_exact_match": 17.4,
|
215 |
+
"eval_f1": 27.194285714285726,
|
216 |
+
"step": 2116
|
217 |
+
},
|
218 |
+
{
|
219 |
+
"epoch": 4.158790170132325,
|
220 |
+
"grad_norm": 13.004076957702637,
|
221 |
+
"learning_rate": 3e-05,
|
222 |
+
"loss": 1.8089,
|
223 |
+
"step": 2200
|
224 |
+
},
|
225 |
+
{
|
226 |
+
"epoch": 4.3478260869565215,
|
227 |
+
"grad_norm": 2.23368501663208,
|
228 |
+
"learning_rate": 3e-05,
|
229 |
+
"loss": 1.8292,
|
230 |
+
"step": 2300
|
231 |
+
},
|
232 |
+
{
|
233 |
+
"epoch": 4.536862003780718,
|
234 |
+
"grad_norm": 0.8635586500167847,
|
235 |
+
"learning_rate": 3e-05,
|
236 |
+
"loss": 1.7992,
|
237 |
+
"step": 2400
|
238 |
+
},
|
239 |
+
{
|
240 |
+
"epoch": 4.725897920604915,
|
241 |
+
"grad_norm": 6.5326738357543945,
|
242 |
+
"learning_rate": 3e-05,
|
243 |
+
"loss": 1.7955,
|
244 |
+
"step": 2500
|
245 |
+
},
|
246 |
+
{
|
247 |
+
"epoch": 4.914933837429111,
|
248 |
+
"grad_norm": 463.4647521972656,
|
249 |
+
"learning_rate": 3e-05,
|
250 |
+
"loss": 1.7953,
|
251 |
+
"step": 2600
|
252 |
+
},
|
253 |
+
{
|
254 |
+
"epoch": 5.0,
|
255 |
+
"eval_accuracy": 0.5740512820512821,
|
256 |
+
"eval_loss": 1.6217337846755981,
|
257 |
+
"eval_runtime": 5.1898,
|
258 |
+
"eval_samples_per_second": 96.342,
|
259 |
+
"eval_steps_per_second": 12.139,
|
260 |
+
"step": 2645
|
261 |
+
},
|
262 |
+
{
|
263 |
+
"epoch": 5.0,
|
264 |
+
"eval_exact_match": 17.6,
|
265 |
+
"eval_f1": 26.161428571428587,
|
266 |
+
"step": 2645
|
267 |
+
},
|
268 |
+
{
|
269 |
+
"epoch": 5.103969754253308,
|
270 |
+
"grad_norm": 203.82815551757812,
|
271 |
+
"learning_rate": 3e-05,
|
272 |
+
"loss": 1.7845,
|
273 |
+
"step": 2700
|
274 |
+
},
|
275 |
+
{
|
276 |
+
"epoch": 5.293005671077505,
|
277 |
+
"grad_norm": 660.8076782226562,
|
278 |
+
"learning_rate": 3e-05,
|
279 |
+
"loss": 1.7909,
|
280 |
+
"step": 2800
|
281 |
+
},
|
282 |
+
{
|
283 |
+
"epoch": 5.482041587901701,
|
284 |
+
"grad_norm": 5727.51806640625,
|
285 |
+
"learning_rate": 3e-05,
|
286 |
+
"loss": 1.7909,
|
287 |
+
"step": 2900
|
288 |
+
},
|
289 |
+
{
|
290 |
+
"epoch": 5.671077504725898,
|
291 |
+
"grad_norm": 4880698.0,
|
292 |
+
"learning_rate": 3e-05,
|
293 |
+
"loss": 3.5781,
|
294 |
+
"step": 3000
|
295 |
+
},
|
296 |
+
{
|
297 |
+
"epoch": 5.8601134215500945,
|
298 |
+
"grad_norm": 1192.6517333984375,
|
299 |
+
"learning_rate": 3e-05,
|
300 |
+
"loss": 3.8703,
|
301 |
+
"step": 3100
|
302 |
+
},
|
303 |
+
{
|
304 |
+
"epoch": 6.0,
|
305 |
+
"eval_accuracy": 0.5433846153846154,
|
306 |
+
"eval_loss": 1.886108636856079,
|
307 |
+
"eval_runtime": 5.5162,
|
308 |
+
"eval_samples_per_second": 90.643,
|
309 |
+
"eval_steps_per_second": 11.421,
|
310 |
+
"step": 3174
|
311 |
+
},
|
312 |
+
{
|
313 |
+
"epoch": 6.0,
|
314 |
+
"eval_exact_match": 11.4,
|
315 |
+
"eval_f1": 20.358571428571437,
|
316 |
+
"step": 3174
|
317 |
+
},
|
318 |
+
{
|
319 |
+
"epoch": 6.049149338374291,
|
320 |
+
"grad_norm": 532.3778686523438,
|
321 |
+
"learning_rate": 3e-05,
|
322 |
+
"loss": 2.1701,
|
323 |
+
"step": 3200
|
324 |
+
},
|
325 |
+
{
|
326 |
+
"epoch": 6.238185255198488,
|
327 |
+
"grad_norm": 565.6697998046875,
|
328 |
+
"learning_rate": 3e-05,
|
329 |
+
"loss": 1.8533,
|
330 |
+
"step": 3300
|
331 |
+
},
|
332 |
+
{
|
333 |
+
"epoch": 6.427221172022684,
|
334 |
+
"grad_norm": 25206.482421875,
|
335 |
+
"learning_rate": 3e-05,
|
336 |
+
"loss": 2.623,
|
337 |
+
"step": 3400
|
338 |
+
},
|
339 |
+
{
|
340 |
+
"epoch": 6.616257088846881,
|
341 |
+
"grad_norm": 71.88556671142578,
|
342 |
+
"learning_rate": 3e-05,
|
343 |
+
"loss": 3.1573,
|
344 |
+
"step": 3500
|
345 |
+
},
|
346 |
+
{
|
347 |
+
"epoch": 6.805293005671078,
|
348 |
+
"grad_norm": 269563.15625,
|
349 |
+
"learning_rate": 3e-05,
|
350 |
+
"loss": 2.2696,
|
351 |
+
"step": 3600
|
352 |
+
},
|
353 |
+
{
|
354 |
+
"epoch": 6.994328922495274,
|
355 |
+
"grad_norm": 11768.7880859375,
|
356 |
+
"learning_rate": 3e-05,
|
357 |
+
"loss": 3.1506,
|
358 |
+
"step": 3700
|
359 |
+
},
|
360 |
+
{
|
361 |
+
"epoch": 7.0,
|
362 |
+
"eval_accuracy": 0.2834871794871795,
|
363 |
+
"eval_loss": 8.311681747436523,
|
364 |
+
"eval_runtime": 5.3223,
|
365 |
+
"eval_samples_per_second": 93.944,
|
366 |
+
"eval_steps_per_second": 11.837,
|
367 |
+
"step": 3703
|
368 |
+
},
|
369 |
+
{
|
370 |
+
"epoch": 7.0,
|
371 |
+
"eval_exact_match": 0.0,
|
372 |
+
"eval_f1": 0.027458851580988985,
|
373 |
+
"step": 3703
|
374 |
+
},
|
375 |
+
{
|
376 |
+
"epoch": 7.183364839319471,
|
377 |
+
"grad_norm": 6220.75244140625,
|
378 |
+
"learning_rate": 3e-05,
|
379 |
+
"loss": 4.273,
|
380 |
+
"step": 3800
|
381 |
+
},
|
382 |
+
{
|
383 |
+
"epoch": 7.3724007561436675,
|
384 |
+
"grad_norm": 3387.953857421875,
|
385 |
+
"learning_rate": 3e-05,
|
386 |
+
"loss": 7.7199,
|
387 |
+
"step": 3900
|
388 |
+
},
|
389 |
+
{
|
390 |
+
"epoch": 7.561436672967864,
|
391 |
+
"grad_norm": 372.8681640625,
|
392 |
+
"learning_rate": 3e-05,
|
393 |
+
"loss": 7.7045,
|
394 |
+
"step": 4000
|
395 |
+
},
|
396 |
+
{
|
397 |
+
"epoch": 7.750472589792061,
|
398 |
+
"grad_norm": 159.64283752441406,
|
399 |
+
"learning_rate": 3e-05,
|
400 |
+
"loss": 6.8901,
|
401 |
+
"step": 4100
|
402 |
+
},
|
403 |
+
{
|
404 |
+
"epoch": 7.939508506616257,
|
405 |
+
"grad_norm": 751876.125,
|
406 |
+
"learning_rate": 3e-05,
|
407 |
+
"loss": 6.5238,
|
408 |
+
"step": 4200
|
409 |
+
},
|
410 |
+
{
|
411 |
+
"epoch": 8.0,
|
412 |
+
"eval_accuracy": 0.22764102564102565,
|
413 |
+
"eval_loss": 7.12347936630249,
|
414 |
+
"eval_runtime": 5.3818,
|
415 |
+
"eval_samples_per_second": 92.905,
|
416 |
+
"eval_steps_per_second": 11.706,
|
417 |
+
"step": 4232
|
418 |
+
},
|
419 |
+
{
|
420 |
+
"epoch": 8.0,
|
421 |
+
"eval_exact_match": 0.0,
|
422 |
+
"eval_f1": 0.013333333333333336,
|
423 |
+
"step": 4232
|
424 |
+
},
|
425 |
+
{
|
426 |
+
"epoch": 8.128544423440454,
|
427 |
+
"grad_norm": 1035664128.0,
|
428 |
+
"learning_rate": 3e-05,
|
429 |
+
"loss": 5.4157,
|
430 |
+
"step": 4300
|
431 |
+
},
|
432 |
+
{
|
433 |
+
"epoch": 8.31758034026465,
|
434 |
+
"grad_norm": 11571242.0,
|
435 |
+
"learning_rate": 3e-05,
|
436 |
+
"loss": 5.8257,
|
437 |
+
"step": 4400
|
438 |
+
},
|
439 |
+
{
|
440 |
+
"epoch": 8.506616257088847,
|
441 |
+
"grad_norm": 88236256.0,
|
442 |
+
"learning_rate": 3e-05,
|
443 |
+
"loss": 6.3609,
|
444 |
+
"step": 4500
|
445 |
+
},
|
446 |
+
{
|
447 |
+
"epoch": 8.695652173913043,
|
448 |
+
"grad_norm": 105426080.0,
|
449 |
+
"learning_rate": 3e-05,
|
450 |
+
"loss": 6.1824,
|
451 |
+
"step": 4600
|
452 |
+
},
|
453 |
+
{
|
454 |
+
"epoch": 8.88468809073724,
|
455 |
+
"grad_norm": 51438844.0,
|
456 |
+
"learning_rate": 3e-05,
|
457 |
+
"loss": 6.1818,
|
458 |
+
"step": 4700
|
459 |
+
},
|
460 |
+
{
|
461 |
+
"epoch": 9.0,
|
462 |
+
"eval_accuracy": 0.1995897435897436,
|
463 |
+
"eval_loss": 11.187646865844727,
|
464 |
+
"eval_runtime": 5.154,
|
465 |
+
"eval_samples_per_second": 97.013,
|
466 |
+
"eval_steps_per_second": 12.224,
|
467 |
+
"step": 4761
|
468 |
+
},
|
469 |
+
{
|
470 |
+
"epoch": 9.0,
|
471 |
+
"eval_exact_match": 0.0,
|
472 |
+
"eval_f1": 0.0,
|
473 |
+
"step": 4761
|
474 |
+
},
|
475 |
+
{
|
476 |
+
"epoch": 9.073724007561436,
|
477 |
+
"grad_norm": 8742571.0,
|
478 |
+
"learning_rate": 3e-05,
|
479 |
+
"loss": 7.5029,
|
480 |
+
"step": 4800
|
481 |
+
},
|
482 |
+
{
|
483 |
+
"epoch": 9.262759924385634,
|
484 |
+
"grad_norm": 4885384192.0,
|
485 |
+
"learning_rate": 3e-05,
|
486 |
+
"loss": 7.805,
|
487 |
+
"step": 4900
|
488 |
+
},
|
489 |
+
{
|
490 |
+
"epoch": 9.45179584120983,
|
491 |
+
"grad_norm": 919074304.0,
|
492 |
+
"learning_rate": 3e-05,
|
493 |
+
"loss": 8.0712,
|
494 |
+
"step": 5000
|
495 |
+
},
|
496 |
+
{
|
497 |
+
"epoch": 9.640831758034027,
|
498 |
+
"grad_norm": 9706883072.0,
|
499 |
+
"learning_rate": 3e-05,
|
500 |
+
"loss": 8.4915,
|
501 |
+
"step": 5100
|
502 |
+
},
|
503 |
+
{
|
504 |
+
"epoch": 9.829867674858223,
|
505 |
+
"grad_norm": 35723780096.0,
|
506 |
+
"learning_rate": 3e-05,
|
507 |
+
"loss": 8.3286,
|
508 |
+
"step": 5200
|
509 |
+
},
|
510 |
+
{
|
511 |
+
"epoch": 10.0,
|
512 |
+
"eval_accuracy": 0.1995897435897436,
|
513 |
+
"eval_loss": 13.374640464782715,
|
514 |
+
"eval_runtime": 5.2089,
|
515 |
+
"eval_samples_per_second": 95.99,
|
516 |
+
"eval_steps_per_second": 12.095,
|
517 |
+
"step": 5290
|
518 |
+
},
|
519 |
+
{
|
520 |
+
"epoch": 10.0,
|
521 |
+
"eval_exact_match": 0.0,
|
522 |
+
"eval_f1": 0.0,
|
523 |
+
"step": 5290
|
524 |
+
},
|
525 |
+
{
|
526 |
+
"epoch": 10.01890359168242,
|
527 |
+
"grad_norm": 172018032.0,
|
528 |
+
"learning_rate": 3e-05,
|
529 |
+
"loss": 8.4403,
|
530 |
+
"step": 5300
|
531 |
+
},
|
532 |
+
{
|
533 |
+
"epoch": 10.207939508506616,
|
534 |
+
"grad_norm": 329336992.0,
|
535 |
+
"learning_rate": 3e-05,
|
536 |
+
"loss": 8.3233,
|
537 |
+
"step": 5400
|
538 |
+
},
|
539 |
+
{
|
540 |
+
"epoch": 10.396975425330814,
|
541 |
+
"grad_norm": 74210560.0,
|
542 |
+
"learning_rate": 3e-05,
|
543 |
+
"loss": 8.3673,
|
544 |
+
"step": 5500
|
545 |
+
},
|
546 |
+
{
|
547 |
+
"epoch": 10.58601134215501,
|
548 |
+
"grad_norm": 497968480.0,
|
549 |
+
"learning_rate": 3e-05,
|
550 |
+
"loss": 9.8311,
|
551 |
+
"step": 5600
|
552 |
+
},
|
553 |
+
{
|
554 |
+
"epoch": 10.775047258979207,
|
555 |
+
"grad_norm": 34138104.0,
|
556 |
+
"learning_rate": 3e-05,
|
557 |
+
"loss": 10.8161,
|
558 |
+
"step": 5700
|
559 |
+
},
|
560 |
+
{
|
561 |
+
"epoch": 10.964083175803403,
|
562 |
+
"grad_norm": 308232992.0,
|
563 |
+
"learning_rate": 3e-05,
|
564 |
+
"loss": 10.2827,
|
565 |
+
"step": 5800
|
566 |
+
},
|
567 |
+
{
|
568 |
+
"epoch": 11.0,
|
569 |
+
"eval_accuracy": 0.1995897435897436,
|
570 |
+
"eval_loss": 12.878591537475586,
|
571 |
+
"eval_runtime": 5.1038,
|
572 |
+
"eval_samples_per_second": 97.966,
|
573 |
+
"eval_steps_per_second": 12.344,
|
574 |
+
"step": 5819
|
575 |
+
},
|
576 |
+
{
|
577 |
+
"epoch": 11.0,
|
578 |
+
"eval_exact_match": 0.0,
|
579 |
+
"eval_f1": 0.0,
|
580 |
+
"step": 5819
|
581 |
+
},
|
582 |
+
{
|
583 |
+
"epoch": 11.1531190926276,
|
584 |
+
"grad_norm": 255807824.0,
|
585 |
+
"learning_rate": 3e-05,
|
586 |
+
"loss": 9.8782,
|
587 |
+
"step": 5900
|
588 |
+
},
|
589 |
+
{
|
590 |
+
"epoch": 11.342155009451796,
|
591 |
+
"grad_norm": 27976615936.0,
|
592 |
+
"learning_rate": 3e-05,
|
593 |
+
"loss": 9.879,
|
594 |
+
"step": 6000
|
595 |
+
},
|
596 |
+
{
|
597 |
+
"epoch": 11.531190926275993,
|
598 |
+
"grad_norm": 13783190732800.0,
|
599 |
+
"learning_rate": 3e-05,
|
600 |
+
"loss": 9.8617,
|
601 |
+
"step": 6100
|
602 |
+
},
|
603 |
+
{
|
604 |
+
"epoch": 11.720226843100189,
|
605 |
+
"grad_norm": 3467661017088.0,
|
606 |
+
"learning_rate": 3e-05,
|
607 |
+
"loss": 10.1028,
|
608 |
+
"step": 6200
|
609 |
+
},
|
610 |
+
{
|
611 |
+
"epoch": 11.909262759924385,
|
612 |
+
"grad_norm": 333752.3125,
|
613 |
+
"learning_rate": 3e-05,
|
614 |
+
"loss": 9.4972,
|
615 |
+
"step": 6300
|
616 |
+
},
|
617 |
+
{
|
618 |
+
"epoch": 12.0,
|
619 |
+
"eval_accuracy": 0.1995897435897436,
|
620 |
+
"eval_loss": 13.161654472351074,
|
621 |
+
"eval_runtime": 5.0927,
|
622 |
+
"eval_samples_per_second": 98.18,
|
623 |
+
"eval_steps_per_second": 12.371,
|
624 |
+
"step": 6348
|
625 |
+
},
|
626 |
+
{
|
627 |
+
"epoch": 12.0,
|
628 |
+
"eval_exact_match": 0.0,
|
629 |
+
"eval_f1": 0.0,
|
630 |
+
"step": 6348
|
631 |
+
},
|
632 |
+
{
|
633 |
+
"epoch": 12.098298676748582,
|
634 |
+
"grad_norm": 6690696704.0,
|
635 |
+
"learning_rate": 3e-05,
|
636 |
+
"loss": 9.2939,
|
637 |
+
"step": 6400
|
638 |
+
},
|
639 |
+
{
|
640 |
+
"epoch": 12.287334593572778,
|
641 |
+
"grad_norm": 148906835968.0,
|
642 |
+
"learning_rate": 3e-05,
|
643 |
+
"loss": 9.3499,
|
644 |
+
"step": 6500
|
645 |
+
},
|
646 |
+
{
|
647 |
+
"epoch": 12.476370510396976,
|
648 |
+
"grad_norm": 15515338752.0,
|
649 |
+
"learning_rate": 3e-05,
|
650 |
+
"loss": 9.3496,
|
651 |
+
"step": 6600
|
652 |
+
},
|
653 |
+
{
|
654 |
+
"epoch": 12.665406427221171,
|
655 |
+
"grad_norm": 76318760960.0,
|
656 |
+
"learning_rate": 3e-05,
|
657 |
+
"loss": 9.2728,
|
658 |
+
"step": 6700
|
659 |
+
},
|
660 |
+
{
|
661 |
+
"epoch": 12.854442344045369,
|
662 |
+
"grad_norm": 85637283840.0,
|
663 |
+
"learning_rate": 3e-05,
|
664 |
+
"loss": 9.4453,
|
665 |
+
"step": 6800
|
666 |
+
},
|
667 |
+
{
|
668 |
+
"epoch": 13.0,
|
669 |
+
"eval_accuracy": 0.1995897435897436,
|
670 |
+
"eval_loss": 12.748088836669922,
|
671 |
+
"eval_runtime": 5.1332,
|
672 |
+
"eval_samples_per_second": 97.405,
|
673 |
+
"eval_steps_per_second": 12.273,
|
674 |
+
"step": 6877
|
675 |
+
},
|
676 |
+
{
|
677 |
+
"epoch": 13.0,
|
678 |
+
"eval_exact_match": 0.0,
|
679 |
+
"eval_f1": 0.0,
|
680 |
+
"step": 6877
|
681 |
+
},
|
682 |
+
{
|
683 |
+
"epoch": 13.043478260869565,
|
684 |
+
"grad_norm": 370487552.0,
|
685 |
+
"learning_rate": 3e-05,
|
686 |
+
"loss": 9.8639,
|
687 |
+
"step": 6900
|
688 |
+
},
|
689 |
+
{
|
690 |
+
"epoch": 13.232514177693762,
|
691 |
+
"grad_norm": 243782.78125,
|
692 |
+
"learning_rate": 3e-05,
|
693 |
+
"loss": 9.2996,
|
694 |
+
"step": 7000
|
695 |
+
},
|
696 |
+
{
|
697 |
+
"epoch": 13.421550094517958,
|
698 |
+
"grad_norm": 572714688.0,
|
699 |
+
"learning_rate": 3e-05,
|
700 |
+
"loss": 9.0366,
|
701 |
+
"step": 7100
|
702 |
+
},
|
703 |
+
{
|
704 |
+
"epoch": 13.610586011342155,
|
705 |
+
"grad_norm": 807026112.0,
|
706 |
+
"learning_rate": 3e-05,
|
707 |
+
"loss": 8.9777,
|
708 |
+
"step": 7200
|
709 |
+
},
|
710 |
+
{
|
711 |
+
"epoch": 13.799621928166351,
|
712 |
+
"grad_norm": 401406112.0,
|
713 |
+
"learning_rate": 3e-05,
|
714 |
+
"loss": 10.3094,
|
715 |
+
"step": 7300
|
716 |
+
},
|
717 |
+
{
|
718 |
+
"epoch": 13.988657844990549,
|
719 |
+
"grad_norm": 20012445696.0,
|
720 |
+
"learning_rate": 3e-05,
|
721 |
+
"loss": 11.1388,
|
722 |
+
"step": 7400
|
723 |
+
},
|
724 |
+
{
|
725 |
+
"epoch": 14.0,
|
726 |
+
"eval_accuracy": 0.1995897435897436,
|
727 |
+
"eval_loss": 11.845151901245117,
|
728 |
+
"eval_runtime": 5.7887,
|
729 |
+
"eval_samples_per_second": 86.376,
|
730 |
+
"eval_steps_per_second": 10.883,
|
731 |
+
"step": 7406
|
732 |
+
},
|
733 |
+
{
|
734 |
+
"epoch": 14.0,
|
735 |
+
"eval_exact_match": 0.0,
|
736 |
+
"eval_f1": 0.0,
|
737 |
+
"step": 7406
|
738 |
+
},
|
739 |
+
{
|
740 |
+
"epoch": 14.177693761814744,
|
741 |
+
"grad_norm": 61066740.0,
|
742 |
+
"learning_rate": 3e-05,
|
743 |
+
"loss": 9.4144,
|
744 |
+
"step": 7500
|
745 |
+
},
|
746 |
+
{
|
747 |
+
"epoch": 14.366729678638942,
|
748 |
+
"grad_norm": 58445844.0,
|
749 |
+
"learning_rate": 3e-05,
|
750 |
+
"loss": 10.1294,
|
751 |
+
"step": 7600
|
752 |
+
},
|
753 |
+
{
|
754 |
+
"epoch": 14.555765595463138,
|
755 |
+
"grad_norm": 40712032.0,
|
756 |
+
"learning_rate": 3e-05,
|
757 |
+
"loss": 12.1512,
|
758 |
+
"step": 7700
|
759 |
+
},
|
760 |
+
{
|
761 |
+
"epoch": 14.744801512287335,
|
762 |
+
"grad_norm": 118321208.0,
|
763 |
+
"learning_rate": 3e-05,
|
764 |
+
"loss": 10.8949,
|
765 |
+
"step": 7800
|
766 |
+
},
|
767 |
+
{
|
768 |
+
"epoch": 14.93383742911153,
|
769 |
+
"grad_norm": 61603804.0,
|
770 |
+
"learning_rate": 3e-05,
|
771 |
+
"loss": 10.849,
|
772 |
+
"step": 7900
|
773 |
+
},
|
774 |
+
{
|
775 |
+
"epoch": 15.0,
|
776 |
+
"eval_accuracy": 0.1995897435897436,
|
777 |
+
"eval_loss": 14.529706954956055,
|
778 |
+
"eval_runtime": 5.7662,
|
779 |
+
"eval_samples_per_second": 86.712,
|
780 |
+
"eval_steps_per_second": 10.926,
|
781 |
+
"step": 7935
|
782 |
+
},
|
783 |
+
{
|
784 |
+
"epoch": 15.0,
|
785 |
+
"eval_exact_match": 0.0,
|
786 |
+
"eval_f1": 0.0,
|
787 |
+
"step": 7935
|
788 |
+
},
|
789 |
+
{
|
790 |
+
"epoch": 15.122873345935728,
|
791 |
+
"grad_norm": 1755613568.0,
|
792 |
+
"learning_rate": 3e-05,
|
793 |
+
"loss": 10.7302,
|
794 |
+
"step": 8000
|
795 |
+
},
|
796 |
+
{
|
797 |
+
"epoch": 15.311909262759924,
|
798 |
+
"grad_norm": 1215193088.0,
|
799 |
+
"learning_rate": 3e-05,
|
800 |
+
"loss": 10.1428,
|
801 |
+
"step": 8100
|
802 |
+
},
|
803 |
+
{
|
804 |
+
"epoch": 15.500945179584122,
|
805 |
+
"grad_norm": 40452497408.0,
|
806 |
+
"learning_rate": 3e-05,
|
807 |
+
"loss": 10.0763,
|
808 |
+
"step": 8200
|
809 |
+
},
|
810 |
+
{
|
811 |
+
"epoch": 15.689981096408317,
|
812 |
+
"grad_norm": 10178840576.0,
|
813 |
+
"learning_rate": 3e-05,
|
814 |
+
"loss": 10.2204,
|
815 |
+
"step": 8300
|
816 |
+
},
|
817 |
+
{
|
818 |
+
"epoch": 15.879017013232515,
|
819 |
+
"grad_norm": 10069305344.0,
|
820 |
+
"learning_rate": 3e-05,
|
821 |
+
"loss": 9.9088,
|
822 |
+
"step": 8400
|
823 |
+
},
|
824 |
+
{
|
825 |
+
"epoch": 16.0,
|
826 |
+
"eval_accuracy": 0.22523076923076923,
|
827 |
+
"eval_loss": 14.802911758422852,
|
828 |
+
"eval_runtime": 5.0764,
|
829 |
+
"eval_samples_per_second": 98.494,
|
830 |
+
"eval_steps_per_second": 12.41,
|
831 |
+
"step": 8464
|
832 |
+
},
|
833 |
+
{
|
834 |
+
"epoch": 16.0,
|
835 |
+
"eval_exact_match": 0.0,
|
836 |
+
"eval_f1": 0.0,
|
837 |
+
"step": 8464
|
838 |
+
},
|
839 |
+
{
|
840 |
+
"epoch": 16.068052930056712,
|
841 |
+
"grad_norm": 9244783616.0,
|
842 |
+
"learning_rate": 3e-05,
|
843 |
+
"loss": 10.2574,
|
844 |
+
"step": 8500
|
845 |
+
},
|
846 |
+
{
|
847 |
+
"epoch": 16.257088846880908,
|
848 |
+
"grad_norm": 556177664.0,
|
849 |
+
"learning_rate": 3e-05,
|
850 |
+
"loss": 10.6479,
|
851 |
+
"step": 8600
|
852 |
+
},
|
853 |
+
{
|
854 |
+
"epoch": 16.446124763705104,
|
855 |
+
"grad_norm": 44021864.0,
|
856 |
+
"learning_rate": 3e-05,
|
857 |
+
"loss": 11.8164,
|
858 |
+
"step": 8700
|
859 |
+
},
|
860 |
+
{
|
861 |
+
"epoch": 16.6351606805293,
|
862 |
+
"grad_norm": 17835092.0,
|
863 |
+
"learning_rate": 3e-05,
|
864 |
+
"loss": 12.391,
|
865 |
+
"step": 8800
|
866 |
+
},
|
867 |
+
{
|
868 |
+
"epoch": 16.8241965973535,
|
869 |
+
"grad_norm": 39907684.0,
|
870 |
+
"learning_rate": 3e-05,
|
871 |
+
"loss": 12.3715,
|
872 |
+
"step": 8900
|
873 |
+
},
|
874 |
+
{
|
875 |
+
"epoch": 17.0,
|
876 |
+
"eval_accuracy": 0.1995897435897436,
|
877 |
+
"eval_loss": 15.876049995422363,
|
878 |
+
"eval_runtime": 5.1121,
|
879 |
+
"eval_samples_per_second": 97.807,
|
880 |
+
"eval_steps_per_second": 12.324,
|
881 |
+
"step": 8993
|
882 |
+
},
|
883 |
+
{
|
884 |
+
"epoch": 17.0,
|
885 |
+
"eval_exact_match": 0.0,
|
886 |
+
"eval_f1": 0.0,
|
887 |
+
"step": 8993
|
888 |
+
},
|
889 |
+
{
|
890 |
+
"epoch": 17.013232514177695,
|
891 |
+
"grad_norm": 5489768.0,
|
892 |
+
"learning_rate": 3e-05,
|
893 |
+
"loss": 12.9394,
|
894 |
+
"step": 9000
|
895 |
+
},
|
896 |
+
{
|
897 |
+
"epoch": 17.20226843100189,
|
898 |
+
"grad_norm": 552121280.0,
|
899 |
+
"learning_rate": 3e-05,
|
900 |
+
"loss": 15.0591,
|
901 |
+
"step": 9100
|
902 |
+
},
|
903 |
+
{
|
904 |
+
"epoch": 17.391304347826086,
|
905 |
+
"grad_norm": 4284707.5,
|
906 |
+
"learning_rate": 3e-05,
|
907 |
+
"loss": 14.7709,
|
908 |
+
"step": 9200
|
909 |
+
},
|
910 |
+
{
|
911 |
+
"epoch": 17.58034026465028,
|
912 |
+
"grad_norm": 32860224.0,
|
913 |
+
"learning_rate": 3e-05,
|
914 |
+
"loss": 15.33,
|
915 |
+
"step": 9300
|
916 |
+
},
|
917 |
+
{
|
918 |
+
"epoch": 17.76937618147448,
|
919 |
+
"grad_norm": 500232672.0,
|
920 |
+
"learning_rate": 3e-05,
|
921 |
+
"loss": 15.2314,
|
922 |
+
"step": 9400
|
923 |
+
},
|
924 |
+
{
|
925 |
+
"epoch": 17.958412098298677,
|
926 |
+
"grad_norm": 138954320.0,
|
927 |
+
"learning_rate": 3e-05,
|
928 |
+
"loss": 14.0399,
|
929 |
+
"step": 9500
|
930 |
+
},
|
931 |
+
{
|
932 |
+
"epoch": 18.0,
|
933 |
+
"eval_accuracy": 0.1995897435897436,
|
934 |
+
"eval_loss": 15.959660530090332,
|
935 |
+
"eval_runtime": 5.4552,
|
936 |
+
"eval_samples_per_second": 91.656,
|
937 |
+
"eval_steps_per_second": 11.549,
|
938 |
+
"step": 9522
|
939 |
+
},
|
940 |
+
{
|
941 |
+
"epoch": 18.0,
|
942 |
+
"eval_exact_match": 0.0,
|
943 |
+
"eval_f1": 0.0,
|
944 |
+
"step": 9522
|
945 |
+
},
|
946 |
+
{
|
947 |
+
"epoch": 18.147448015122873,
|
948 |
+
"grad_norm": 17529152.0,
|
949 |
+
"learning_rate": 3e-05,
|
950 |
+
"loss": 15.3582,
|
951 |
+
"step": 9600
|
952 |
+
},
|
953 |
+
{
|
954 |
+
"epoch": 18.33648393194707,
|
955 |
+
"grad_norm": 14452664.0,
|
956 |
+
"learning_rate": 3e-05,
|
957 |
+
"loss": 16.0595,
|
958 |
+
"step": 9700
|
959 |
+
},
|
960 |
+
{
|
961 |
+
"epoch": 18.525519848771268,
|
962 |
+
"grad_norm": 9175341.0,
|
963 |
+
"learning_rate": 3e-05,
|
964 |
+
"loss": 15.2152,
|
965 |
+
"step": 9800
|
966 |
+
},
|
967 |
+
{
|
968 |
+
"epoch": 18.714555765595463,
|
969 |
+
"grad_norm": 3109339.75,
|
970 |
+
"learning_rate": 3e-05,
|
971 |
+
"loss": 15.6083,
|
972 |
+
"step": 9900
|
973 |
+
},
|
974 |
+
{
|
975 |
+
"epoch": 18.90359168241966,
|
976 |
+
"grad_norm": 1141542.125,
|
977 |
+
"learning_rate": 3e-05,
|
978 |
+
"loss": 15.6927,
|
979 |
+
"step": 10000
|
980 |
+
},
|
981 |
+
{
|
982 |
+
"epoch": 19.0,
|
983 |
+
"eval_accuracy": 0.1995897435897436,
|
984 |
+
"eval_loss": 16.572154998779297,
|
985 |
+
"eval_runtime": 5.784,
|
986 |
+
"eval_samples_per_second": 86.445,
|
987 |
+
"eval_steps_per_second": 10.892,
|
988 |
+
"step": 10051
|
989 |
+
},
|
990 |
+
{
|
991 |
+
"epoch": 19.0,
|
992 |
+
"eval_exact_match": 0.0,
|
993 |
+
"eval_f1": 0.0,
|
994 |
+
"step": 10051
|
995 |
+
},
|
996 |
+
{
|
997 |
+
"epoch": 19.092627599243855,
|
998 |
+
"grad_norm": 3051544.0,
|
999 |
+
"learning_rate": 3e-05,
|
1000 |
+
"loss": 15.4752,
|
1001 |
+
"step": 10100
|
1002 |
+
},
|
1003 |
+
{
|
1004 |
+
"epoch": 19.281663516068054,
|
1005 |
+
"grad_norm": 32653528.0,
|
1006 |
+
"learning_rate": 3e-05,
|
1007 |
+
"loss": 15.3329,
|
1008 |
+
"step": 10200
|
1009 |
+
},
|
1010 |
+
{
|
1011 |
+
"epoch": 19.47069943289225,
|
1012 |
+
"grad_norm": 373167.34375,
|
1013 |
+
"learning_rate": 3e-05,
|
1014 |
+
"loss": 15.1992,
|
1015 |
+
"step": 10300
|
1016 |
+
},
|
1017 |
+
{
|
1018 |
+
"epoch": 19.659735349716446,
|
1019 |
+
"grad_norm": 761316507648.0,
|
1020 |
+
"learning_rate": 3e-05,
|
1021 |
+
"loss": 14.5326,
|
1022 |
+
"step": 10400
|
1023 |
+
},
|
1024 |
+
{
|
1025 |
+
"epoch": 19.84877126654064,
|
1026 |
+
"grad_norm": 15194965.0,
|
1027 |
+
"learning_rate": 3e-05,
|
1028 |
+
"loss": 14.6703,
|
1029 |
+
"step": 10500
|
1030 |
+
},
|
1031 |
+
{
|
1032 |
+
"epoch": 20.0,
|
1033 |
+
"eval_accuracy": 0.2026153846153846,
|
1034 |
+
"eval_loss": 12.01396656036377,
|
1035 |
+
"eval_runtime": 5.164,
|
1036 |
+
"eval_samples_per_second": 96.824,
|
1037 |
+
"eval_steps_per_second": 12.2,
|
1038 |
+
"step": 10580
|
1039 |
+
},
|
1040 |
+
{
|
1041 |
+
"epoch": 20.0,
|
1042 |
+
"eval_exact_match": 0.0,
|
1043 |
+
"eval_f1": 0.0,
|
1044 |
+
"step": 10580
|
1045 |
+
},
|
1046 |
+
{
|
1047 |
+
"epoch": 20.03780718336484,
|
1048 |
+
"grad_norm": 1443913.625,
|
1049 |
+
"learning_rate": 3e-05,
|
1050 |
+
"loss": 13.4546,
|
1051 |
+
"step": 10600
|
1052 |
+
},
|
1053 |
+
{
|
1054 |
+
"epoch": 20.226843100189036,
|
1055 |
+
"grad_norm": 38678328.0,
|
1056 |
+
"learning_rate": 3e-05,
|
1057 |
+
"loss": 15.1074,
|
1058 |
+
"step": 10700
|
1059 |
+
},
|
1060 |
+
{
|
1061 |
+
"epoch": 20.415879017013232,
|
1062 |
+
"grad_norm": 235378832.0,
|
1063 |
+
"learning_rate": 3e-05,
|
1064 |
+
"loss": 15.8794,
|
1065 |
+
"step": 10800
|
1066 |
+
},
|
1067 |
+
{
|
1068 |
+
"epoch": 20.604914933837428,
|
1069 |
+
"grad_norm": 7457460.5,
|
1070 |
+
"learning_rate": 3e-05,
|
1071 |
+
"loss": 15.9728,
|
1072 |
+
"step": 10900
|
1073 |
+
},
|
1074 |
+
{
|
1075 |
+
"epoch": 20.793950850661627,
|
1076 |
+
"grad_norm": 360401824.0,
|
1077 |
+
"learning_rate": 3e-05,
|
1078 |
+
"loss": 16.3397,
|
1079 |
+
"step": 11000
|
1080 |
+
},
|
1081 |
+
{
|
1082 |
+
"epoch": 20.982986767485823,
|
1083 |
+
"grad_norm": 5285015.0,
|
1084 |
+
"learning_rate": 3e-05,
|
1085 |
+
"loss": 16.1349,
|
1086 |
+
"step": 11100
|
1087 |
+
},
|
1088 |
+
{
|
1089 |
+
"epoch": 21.0,
|
1090 |
+
"eval_accuracy": 0.1995897435897436,
|
1091 |
+
"eval_loss": 16.624380111694336,
|
1092 |
+
"eval_runtime": 5.0669,
|
1093 |
+
"eval_samples_per_second": 98.68,
|
1094 |
+
"eval_steps_per_second": 12.434,
|
1095 |
+
"step": 11109
|
1096 |
+
},
|
1097 |
+
{
|
1098 |
+
"epoch": 21.0,
|
1099 |
+
"eval_exact_match": 0.0,
|
1100 |
+
"eval_f1": 0.0,
|
1101 |
+
"step": 11109
|
1102 |
+
},
|
1103 |
+
{
|
1104 |
+
"epoch": 21.17202268431002,
|
1105 |
+
"grad_norm": 3120185999360.0,
|
1106 |
+
"learning_rate": 3e-05,
|
1107 |
+
"loss": 15.2325,
|
1108 |
+
"step": 11200
|
1109 |
+
},
|
1110 |
+
{
|
1111 |
+
"epoch": 21.361058601134214,
|
1112 |
+
"grad_norm": 459053662208.0,
|
1113 |
+
"learning_rate": 3e-05,
|
1114 |
+
"loss": 14.9804,
|
1115 |
+
"step": 11300
|
1116 |
+
},
|
1117 |
+
{
|
1118 |
+
"epoch": 21.550094517958414,
|
1119 |
+
"grad_norm": 33530312704.0,
|
1120 |
+
"learning_rate": 3e-05,
|
1121 |
+
"loss": 13.4482,
|
1122 |
+
"step": 11400
|
1123 |
+
},
|
1124 |
+
{
|
1125 |
+
"epoch": 21.73913043478261,
|
1126 |
+
"grad_norm": 524117879029760.0,
|
1127 |
+
"learning_rate": 3e-05,
|
1128 |
+
"loss": 12.8065,
|
1129 |
+
"step": 11500
|
1130 |
+
},
|
1131 |
+
{
|
1132 |
+
"epoch": 21.928166351606805,
|
1133 |
+
"grad_norm": 73415352385536.0,
|
1134 |
+
"learning_rate": 3e-05,
|
1135 |
+
"loss": 13.1619,
|
1136 |
+
"step": 11600
|
1137 |
+
},
|
1138 |
+
{
|
1139 |
+
"epoch": 22.0,
|
1140 |
+
"eval_accuracy": 0.1995897435897436,
|
1141 |
+
"eval_loss": 15.610301971435547,
|
1142 |
+
"eval_runtime": 5.8348,
|
1143 |
+
"eval_samples_per_second": 85.693,
|
1144 |
+
"eval_steps_per_second": 10.797,
|
1145 |
+
"step": 11638
|
1146 |
+
},
|
1147 |
+
{
|
1148 |
+
"epoch": 22.0,
|
1149 |
+
"eval_exact_match": 0.0,
|
1150 |
+
"eval_f1": 0.0,
|
1151 |
+
"step": 11638
|
1152 |
+
},
|
1153 |
+
{
|
1154 |
+
"epoch": 22.117202268431,
|
1155 |
+
"grad_norm": 24663130374144.0,
|
1156 |
+
"learning_rate": 3e-05,
|
1157 |
+
"loss": 13.4567,
|
1158 |
+
"step": 11700
|
1159 |
+
},
|
1160 |
+
{
|
1161 |
+
"epoch": 22.3062381852552,
|
1162 |
+
"grad_norm": 1152798.75,
|
1163 |
+
"learning_rate": 3e-05,
|
1164 |
+
"loss": 13.3323,
|
1165 |
+
"step": 11800
|
1166 |
+
},
|
1167 |
+
{
|
1168 |
+
"epoch": 22.495274102079396,
|
1169 |
+
"grad_norm": 6407195.0,
|
1170 |
+
"learning_rate": 3e-05,
|
1171 |
+
"loss": 12.224,
|
1172 |
+
"step": 11900
|
1173 |
+
},
|
1174 |
+
{
|
1175 |
+
"epoch": 22.68431001890359,
|
1176 |
+
"grad_norm": 3026044.75,
|
1177 |
+
"learning_rate": 3e-05,
|
1178 |
+
"loss": 11.6534,
|
1179 |
+
"step": 12000
|
1180 |
+
},
|
1181 |
+
{
|
1182 |
+
"epoch": 22.873345935727787,
|
1183 |
+
"grad_norm": 433589.25,
|
1184 |
+
"learning_rate": 3e-05,
|
1185 |
+
"loss": 11.176,
|
1186 |
+
"step": 12100
|
1187 |
+
},
|
1188 |
+
{
|
1189 |
+
"epoch": 23.0,
|
1190 |
+
"eval_accuracy": 0.20005128205128206,
|
1191 |
+
"eval_loss": 9.37710952758789,
|
1192 |
+
"eval_runtime": 5.4109,
|
1193 |
+
"eval_samples_per_second": 92.407,
|
1194 |
+
"eval_steps_per_second": 11.643,
|
1195 |
+
"step": 12167
|
1196 |
+
},
|
1197 |
+
{
|
1198 |
+
"epoch": 23.0,
|
1199 |
+
"eval_exact_match": 0.0,
|
1200 |
+
"eval_f1": 0.0,
|
1201 |
+
"step": 12167
|
1202 |
+
},
|
1203 |
+
{
|
1204 |
+
"epoch": 23.062381852551987,
|
1205 |
+
"grad_norm": 221080977408.0,
|
1206 |
+
"learning_rate": 3e-05,
|
1207 |
+
"loss": 10.2085,
|
1208 |
+
"step": 12200
|
1209 |
+
},
|
1210 |
+
{
|
1211 |
+
"epoch": 23.251417769376182,
|
1212 |
+
"grad_norm": 819255836672.0,
|
1213 |
+
"learning_rate": 3e-05,
|
1214 |
+
"loss": 10.3606,
|
1215 |
+
"step": 12300
|
1216 |
+
},
|
1217 |
+
{
|
1218 |
+
"epoch": 23.440453686200378,
|
1219 |
+
"grad_norm": 1077065216.0,
|
1220 |
+
"learning_rate": 3e-05,
|
1221 |
+
"loss": 10.3544,
|
1222 |
+
"step": 12400
|
1223 |
+
},
|
1224 |
+
{
|
1225 |
+
"epoch": 23.629489603024574,
|
1226 |
+
"grad_norm": 147283214336.0,
|
1227 |
+
"learning_rate": 3e-05,
|
1228 |
+
"loss": 10.6161,
|
1229 |
+
"step": 12500
|
1230 |
+
},
|
1231 |
+
{
|
1232 |
+
"epoch": 23.81852551984877,
|
1233 |
+
"grad_norm": 479103936.0,
|
1234 |
+
"learning_rate": 3e-05,
|
1235 |
+
"loss": 11.0412,
|
1236 |
+
"step": 12600
|
1237 |
+
},
|
1238 |
+
{
|
1239 |
+
"epoch": 24.0,
|
1240 |
+
"eval_accuracy": 0.2362051282051282,
|
1241 |
+
"eval_loss": 8.926570892333984,
|
1242 |
+
"eval_runtime": 5.7624,
|
1243 |
+
"eval_samples_per_second": 86.769,
|
1244 |
+
"eval_steps_per_second": 10.933,
|
1245 |
+
"step": 12696
|
1246 |
+
},
|
1247 |
+
{
|
1248 |
+
"epoch": 24.0,
|
1249 |
+
"eval_exact_match": 0.0,
|
1250 |
+
"eval_f1": 0.0,
|
1251 |
+
"step": 12696
|
1252 |
+
},
|
1253 |
+
{
|
1254 |
+
"epoch": 24.00756143667297,
|
1255 |
+
"grad_norm": 18736.490234375,
|
1256 |
+
"learning_rate": 3e-05,
|
1257 |
+
"loss": 10.5723,
|
1258 |
+
"step": 12700
|
1259 |
+
},
|
1260 |
+
{
|
1261 |
+
"epoch": 24.196597353497165,
|
1262 |
+
"grad_norm": 10122.208984375,
|
1263 |
+
"learning_rate": 3e-05,
|
1264 |
+
"loss": 9.1505,
|
1265 |
+
"step": 12800
|
1266 |
+
},
|
1267 |
+
{
|
1268 |
+
"epoch": 24.38563327032136,
|
1269 |
+
"grad_norm": 884941.875,
|
1270 |
+
"learning_rate": 3e-05,
|
1271 |
+
"loss": 8.6122,
|
1272 |
+
"step": 12900
|
1273 |
+
},
|
1274 |
+
{
|
1275 |
+
"epoch": 24.574669187145556,
|
1276 |
+
"grad_norm": 51996.421875,
|
1277 |
+
"learning_rate": 3e-05,
|
1278 |
+
"loss": 8.5405,
|
1279 |
+
"step": 13000
|
1280 |
+
},
|
1281 |
+
{
|
1282 |
+
"epoch": 24.763705103969755,
|
1283 |
+
"grad_norm": 11833.41796875,
|
1284 |
+
"learning_rate": 3e-05,
|
1285 |
+
"loss": 8.4026,
|
1286 |
+
"step": 13100
|
1287 |
+
},
|
1288 |
+
{
|
1289 |
+
"epoch": 24.95274102079395,
|
1290 |
+
"grad_norm": 645689536.0,
|
1291 |
+
"learning_rate": 3e-05,
|
1292 |
+
"loss": 8.8444,
|
1293 |
+
"step": 13200
|
1294 |
+
},
|
1295 |
+
{
|
1296 |
+
"epoch": 25.0,
|
1297 |
+
"eval_accuracy": 0.20235897435897435,
|
1298 |
+
"eval_loss": 10.400785446166992,
|
1299 |
+
"eval_runtime": 5.8159,
|
1300 |
+
"eval_samples_per_second": 85.971,
|
1301 |
+
"eval_steps_per_second": 10.832,
|
1302 |
+
"step": 13225
|
1303 |
+
},
|
1304 |
+
{
|
1305 |
+
"epoch": 25.0,
|
1306 |
+
"eval_exact_match": 0.0,
|
1307 |
+
"eval_f1": 0.0,
|
1308 |
+
"step": 13225
|
1309 |
+
},
|
1310 |
+
{
|
1311 |
+
"epoch": 25.141776937618147,
|
1312 |
+
"grad_norm": 33455.7734375,
|
1313 |
+
"learning_rate": 3e-05,
|
1314 |
+
"loss": 8.9314,
|
1315 |
+
"step": 13300
|
1316 |
+
},
|
1317 |
+
{
|
1318 |
+
"epoch": 25.330812854442343,
|
1319 |
+
"grad_norm": 665939.625,
|
1320 |
+
"learning_rate": 3e-05,
|
1321 |
+
"loss": 8.9419,
|
1322 |
+
"step": 13400
|
1323 |
+
},
|
1324 |
+
{
|
1325 |
+
"epoch": 25.519848771266542,
|
1326 |
+
"grad_norm": 366503.25,
|
1327 |
+
"learning_rate": 3e-05,
|
1328 |
+
"loss": 9.1085,
|
1329 |
+
"step": 13500
|
1330 |
+
},
|
1331 |
+
{
|
1332 |
+
"epoch": 25.708884688090738,
|
1333 |
+
"grad_norm": 1146194755584.0,
|
1334 |
+
"learning_rate": 3e-05,
|
1335 |
+
"loss": 9.1301,
|
1336 |
+
"step": 13600
|
1337 |
+
},
|
1338 |
+
{
|
1339 |
+
"epoch": 25.897920604914933,
|
1340 |
+
"grad_norm": 5115191230464.0,
|
1341 |
+
"learning_rate": 3e-05,
|
1342 |
+
"loss": 8.9435,
|
1343 |
+
"step": 13700
|
1344 |
+
},
|
1345 |
+
{
|
1346 |
+
"epoch": 26.0,
|
1347 |
+
"eval_accuracy": 0.1995897435897436,
|
1348 |
+
"eval_loss": 12.846183776855469,
|
1349 |
+
"eval_runtime": 5.1674,
|
1350 |
+
"eval_samples_per_second": 96.76,
|
1351 |
+
"eval_steps_per_second": 12.192,
|
1352 |
+
"step": 13754
|
1353 |
+
},
|
1354 |
+
{
|
1355 |
+
"epoch": 26.0,
|
1356 |
+
"eval_exact_match": 0.0,
|
1357 |
+
"eval_f1": 0.0,
|
1358 |
+
"step": 13754
|
1359 |
+
},
|
1360 |
+
{
|
1361 |
+
"epoch": 26.08695652173913,
|
1362 |
+
"grad_norm": 119.87647247314453,
|
1363 |
+
"learning_rate": 3e-05,
|
1364 |
+
"loss": 10.1216,
|
1365 |
+
"step": 13800
|
1366 |
+
},
|
1367 |
+
{
|
1368 |
+
"epoch": 26.27599243856333,
|
1369 |
+
"grad_norm": 5832.49462890625,
|
1370 |
+
"learning_rate": 3e-05,
|
1371 |
+
"loss": 8.5386,
|
1372 |
+
"step": 13900
|
1373 |
+
},
|
1374 |
+
{
|
1375 |
+
"epoch": 26.465028355387524,
|
1376 |
+
"grad_norm": 7308.30908203125,
|
1377 |
+
"learning_rate": 3e-05,
|
1378 |
+
"loss": 8.1093,
|
1379 |
+
"step": 14000
|
1380 |
+
},
|
1381 |
+
{
|
1382 |
+
"epoch": 26.65406427221172,
|
1383 |
+
"grad_norm": 21071.32421875,
|
1384 |
+
"learning_rate": 3e-05,
|
1385 |
+
"loss": 8.0342,
|
1386 |
+
"step": 14100
|
1387 |
+
},
|
1388 |
+
{
|
1389 |
+
"epoch": 26.843100189035916,
|
1390 |
+
"grad_norm": 69589.0625,
|
1391 |
+
"learning_rate": 3e-05,
|
1392 |
+
"loss": 8.0313,
|
1393 |
+
"step": 14200
|
1394 |
+
},
|
1395 |
+
{
|
1396 |
+
"epoch": 27.0,
|
1397 |
+
"eval_accuracy": 0.23364102564102565,
|
1398 |
+
"eval_loss": 7.521541595458984,
|
1399 |
+
"eval_runtime": 5.0339,
|
1400 |
+
"eval_samples_per_second": 99.326,
|
1401 |
+
"eval_steps_per_second": 12.515,
|
1402 |
+
"step": 14283
|
1403 |
+
},
|
1404 |
+
{
|
1405 |
+
"epoch": 27.0,
|
1406 |
+
"eval_exact_match": 0.0,
|
1407 |
+
"eval_f1": 0.0,
|
1408 |
+
"step": 14283
|
1409 |
+
},
|
1410 |
+
{
|
1411 |
+
"epoch": 27.032136105860115,
|
1412 |
+
"grad_norm": 158736.625,
|
1413 |
+
"learning_rate": 3e-05,
|
1414 |
+
"loss": 8.1019,
|
1415 |
+
"step": 14300
|
1416 |
+
},
|
1417 |
+
{
|
1418 |
+
"epoch": 27.22117202268431,
|
1419 |
+
"grad_norm": 12491.8095703125,
|
1420 |
+
"learning_rate": 3e-05,
|
1421 |
+
"loss": 8.2181,
|
1422 |
+
"step": 14400
|
1423 |
+
},
|
1424 |
+
{
|
1425 |
+
"epoch": 27.410207939508506,
|
1426 |
+
"grad_norm": 123181.3671875,
|
1427 |
+
"learning_rate": 3e-05,
|
1428 |
+
"loss": 8.3091,
|
1429 |
+
"step": 14500
|
1430 |
+
},
|
1431 |
+
{
|
1432 |
+
"epoch": 27.599243856332702,
|
1433 |
+
"grad_norm": 620153.5,
|
1434 |
+
"learning_rate": 3e-05,
|
1435 |
+
"loss": 8.4058,
|
1436 |
+
"step": 14600
|
1437 |
+
},
|
1438 |
+
{
|
1439 |
+
"epoch": 27.7882797731569,
|
1440 |
+
"grad_norm": 337683.1875,
|
1441 |
+
"learning_rate": 3e-05,
|
1442 |
+
"loss": 8.619,
|
1443 |
+
"step": 14700
|
1444 |
+
},
|
1445 |
+
{
|
1446 |
+
"epoch": 27.977315689981097,
|
1447 |
+
"grad_norm": 5629.42431640625,
|
1448 |
+
"learning_rate": 3e-05,
|
1449 |
+
"loss": 8.3923,
|
1450 |
+
"step": 14800
|
1451 |
+
},
|
1452 |
+
{
|
1453 |
+
"epoch": 28.0,
|
1454 |
+
"eval_accuracy": 0.21205128205128204,
|
1455 |
+
"eval_loss": 7.625494003295898,
|
1456 |
+
"eval_runtime": 5.1068,
|
1457 |
+
"eval_samples_per_second": 97.909,
|
1458 |
+
"eval_steps_per_second": 12.336,
|
1459 |
+
"step": 14812
|
1460 |
+
},
|
1461 |
+
{
|
1462 |
+
"epoch": 28.0,
|
1463 |
+
"eval_exact_match": 0.0,
|
1464 |
+
"eval_f1": 0.0,
|
1465 |
+
"step": 14812
|
1466 |
+
},
|
1467 |
+
{
|
1468 |
+
"epoch": 28.166351606805293,
|
1469 |
+
"grad_norm": 779905.0625,
|
1470 |
+
"learning_rate": 3e-05,
|
1471 |
+
"loss": 8.2129,
|
1472 |
+
"step": 14900
|
1473 |
+
},
|
1474 |
+
{
|
1475 |
+
"epoch": 28.35538752362949,
|
1476 |
+
"grad_norm": 92020.96875,
|
1477 |
+
"learning_rate": 3e-05,
|
1478 |
+
"loss": 8.3803,
|
1479 |
+
"step": 15000
|
1480 |
+
},
|
1481 |
+
{
|
1482 |
+
"epoch": 28.544423440453688,
|
1483 |
+
"grad_norm": 1754453.125,
|
1484 |
+
"learning_rate": 3e-05,
|
1485 |
+
"loss": 10.3285,
|
1486 |
+
"step": 15100
|
1487 |
+
},
|
1488 |
+
{
|
1489 |
+
"epoch": 28.733459357277884,
|
1490 |
+
"grad_norm": 9245.0390625,
|
1491 |
+
"learning_rate": 3e-05,
|
1492 |
+
"loss": 10.3239,
|
1493 |
+
"step": 15200
|
1494 |
+
},
|
1495 |
+
{
|
1496 |
+
"epoch": 28.92249527410208,
|
1497 |
+
"grad_norm": 57860676.0,
|
1498 |
+
"learning_rate": 3e-05,
|
1499 |
+
"loss": 8.5368,
|
1500 |
+
"step": 15300
|
1501 |
+
},
|
1502 |
+
{
|
1503 |
+
"epoch": 29.0,
|
1504 |
+
"eval_accuracy": 0.1995897435897436,
|
1505 |
+
"eval_loss": 9.528727531433105,
|
1506 |
+
"eval_runtime": 5.1078,
|
1507 |
+
"eval_samples_per_second": 97.89,
|
1508 |
+
"eval_steps_per_second": 12.334,
|
1509 |
+
"step": 15341
|
1510 |
+
},
|
1511 |
+
{
|
1512 |
+
"epoch": 29.0,
|
1513 |
+
"eval_exact_match": 0.0,
|
1514 |
+
"eval_f1": 0.0,
|
1515 |
+
"step": 15341
|
1516 |
+
},
|
1517 |
+
{
|
1518 |
+
"epoch": 29.111531190926275,
|
1519 |
+
"grad_norm": 95621760.0,
|
1520 |
+
"learning_rate": 3e-05,
|
1521 |
+
"loss": 9.2102,
|
1522 |
+
"step": 15400
|
1523 |
+
},
|
1524 |
+
{
|
1525 |
+
"epoch": 29.300567107750474,
|
1526 |
+
"grad_norm": 68808096.0,
|
1527 |
+
"learning_rate": 3e-05,
|
1528 |
+
"loss": 9.1584,
|
1529 |
+
"step": 15500
|
1530 |
+
},
|
1531 |
+
{
|
1532 |
+
"epoch": 29.48960302457467,
|
1533 |
+
"grad_norm": 394869760.0,
|
1534 |
+
"learning_rate": 3e-05,
|
1535 |
+
"loss": 9.0096,
|
1536 |
+
"step": 15600
|
1537 |
+
},
|
1538 |
+
{
|
1539 |
+
"epoch": 29.678638941398866,
|
1540 |
+
"grad_norm": 295963264.0,
|
1541 |
+
"learning_rate": 3e-05,
|
1542 |
+
"loss": 8.1208,
|
1543 |
+
"step": 15700
|
1544 |
+
},
|
1545 |
+
{
|
1546 |
+
"epoch": 29.86767485822306,
|
1547 |
+
"grad_norm": 471883200.0,
|
1548 |
+
"learning_rate": 3e-05,
|
1549 |
+
"loss": 7.964,
|
1550 |
+
"step": 15800
|
1551 |
+
},
|
1552 |
+
{
|
1553 |
+
"epoch": 30.0,
|
1554 |
+
"eval_accuracy": 0.19994871794871794,
|
1555 |
+
"eval_loss": 7.713045120239258,
|
1556 |
+
"eval_runtime": 5.3921,
|
1557 |
+
"eval_samples_per_second": 92.728,
|
1558 |
+
"eval_steps_per_second": 11.684,
|
1559 |
+
"step": 15870
|
1560 |
+
},
|
1561 |
+
{
|
1562 |
+
"epoch": 30.0,
|
1563 |
+
"eval_exact_match": 0.0,
|
1564 |
+
"eval_f1": 0.0,
|
1565 |
+
"step": 15870
|
1566 |
+
},
|
1567 |
+
{
|
1568 |
+
"epoch": 30.056710775047257,
|
1569 |
+
"grad_norm": 1943886.25,
|
1570 |
+
"learning_rate": 3e-05,
|
1571 |
+
"loss": 8.051,
|
1572 |
+
"step": 15900
|
1573 |
+
},
|
1574 |
+
{
|
1575 |
+
"epoch": 30.245746691871457,
|
1576 |
+
"grad_norm": 252297.234375,
|
1577 |
+
"learning_rate": 3e-05,
|
1578 |
+
"loss": 8.2152,
|
1579 |
+
"step": 16000
|
1580 |
+
},
|
1581 |
+
{
|
1582 |
+
"epoch": 30.434782608695652,
|
1583 |
+
"grad_norm": 20118.697265625,
|
1584 |
+
"learning_rate": 3e-05,
|
1585 |
+
"loss": 8.586,
|
1586 |
+
"step": 16100
|
1587 |
+
},
|
1588 |
+
{
|
1589 |
+
"epoch": 30.623818525519848,
|
1590 |
+
"grad_norm": 30278450.0,
|
1591 |
+
"learning_rate": 3e-05,
|
1592 |
+
"loss": 8.594,
|
1593 |
+
"step": 16200
|
1594 |
+
},
|
1595 |
+
{
|
1596 |
+
"epoch": 30.812854442344044,
|
1597 |
+
"grad_norm": 491771.5,
|
1598 |
+
"learning_rate": 3e-05,
|
1599 |
+
"loss": 8.861,
|
1600 |
+
"step": 16300
|
1601 |
+
},
|
1602 |
+
{
|
1603 |
+
"epoch": 31.0,
|
1604 |
+
"eval_accuracy": 0.2362051282051282,
|
1605 |
+
"eval_loss": 7.903472423553467,
|
1606 |
+
"eval_runtime": 5.622,
|
1607 |
+
"eval_samples_per_second": 88.937,
|
1608 |
+
"eval_steps_per_second": 11.206,
|
1609 |
+
"step": 16399
|
1610 |
+
},
|
1611 |
+
{
|
1612 |
+
"epoch": 31.0,
|
1613 |
+
"eval_exact_match": 0.0,
|
1614 |
+
"eval_f1": 0.0,
|
1615 |
+
"step": 16399
|
1616 |
+
},
|
1617 |
+
{
|
1618 |
+
"epoch": 31.001890359168243,
|
1619 |
+
"grad_norm": 67330.3125,
|
1620 |
+
"learning_rate": 3e-05,
|
1621 |
+
"loss": 8.3547,
|
1622 |
+
"step": 16400
|
1623 |
+
},
|
1624 |
+
{
|
1625 |
+
"epoch": 31.19092627599244,
|
1626 |
+
"grad_norm": 437164.03125,
|
1627 |
+
"learning_rate": 3e-05,
|
1628 |
+
"loss": 9.0483,
|
1629 |
+
"step": 16500
|
1630 |
+
},
|
1631 |
+
{
|
1632 |
+
"epoch": 31.379962192816635,
|
1633 |
+
"grad_norm": 383445.78125,
|
1634 |
+
"learning_rate": 3e-05,
|
1635 |
+
"loss": 8.8258,
|
1636 |
+
"step": 16600
|
1637 |
+
},
|
1638 |
+
{
|
1639 |
+
"epoch": 31.56899810964083,
|
1640 |
+
"grad_norm": 26452162510848.0,
|
1641 |
+
"learning_rate": 3e-05,
|
1642 |
+
"loss": 9.2178,
|
1643 |
+
"step": 16700
|
1644 |
+
},
|
1645 |
+
{
|
1646 |
+
"epoch": 31.75803402646503,
|
1647 |
+
"grad_norm": 1477470060544.0,
|
1648 |
+
"learning_rate": 3e-05,
|
1649 |
+
"loss": 9.0897,
|
1650 |
+
"step": 16800
|
1651 |
+
},
|
1652 |
+
{
|
1653 |
+
"epoch": 31.947069943289225,
|
1654 |
+
"grad_norm": 3547858688.0,
|
1655 |
+
"learning_rate": 3e-05,
|
1656 |
+
"loss": 9.1201,
|
1657 |
+
"step": 16900
|
1658 |
+
},
|
1659 |
+
{
|
1660 |
+
"epoch": 32.0,
|
1661 |
+
"eval_accuracy": 0.2362051282051282,
|
1662 |
+
"eval_loss": 8.878704071044922,
|
1663 |
+
"eval_runtime": 5.7948,
|
1664 |
+
"eval_samples_per_second": 86.284,
|
1665 |
+
"eval_steps_per_second": 10.872,
|
1666 |
+
"step": 16928
|
1667 |
+
},
|
1668 |
+
{
|
1669 |
+
"epoch": 32.0,
|
1670 |
+
"eval_exact_match": 0.0,
|
1671 |
+
"eval_f1": 0.0,
|
1672 |
+
"step": 16928
|
1673 |
+
},
|
1674 |
+
{
|
1675 |
+
"epoch": 32.136105860113425,
|
1676 |
+
"grad_norm": 25736851226624.0,
|
1677 |
+
"learning_rate": 3e-05,
|
1678 |
+
"loss": 9.0721,
|
1679 |
+
"step": 17000
|
1680 |
+
},
|
1681 |
+
{
|
1682 |
+
"epoch": 32.32514177693762,
|
1683 |
+
"grad_norm": 83071770624.0,
|
1684 |
+
"learning_rate": 3e-05,
|
1685 |
+
"loss": 9.1184,
|
1686 |
+
"step": 17100
|
1687 |
+
},
|
1688 |
+
{
|
1689 |
+
"epoch": 32.514177693761816,
|
1690 |
+
"grad_norm": 362845667328.0,
|
1691 |
+
"learning_rate": 3e-05,
|
1692 |
+
"loss": 9.1277,
|
1693 |
+
"step": 17200
|
1694 |
+
},
|
1695 |
+
{
|
1696 |
+
"epoch": 32.70321361058601,
|
1697 |
+
"grad_norm": 6446509981696.0,
|
1698 |
+
"learning_rate": 3e-05,
|
1699 |
+
"loss": 9.1827,
|
1700 |
+
"step": 17300
|
1701 |
+
},
|
1702 |
+
{
|
1703 |
+
"epoch": 32.89224952741021,
|
1704 |
+
"grad_norm": 8676001710080.0,
|
1705 |
+
"learning_rate": 3e-05,
|
1706 |
+
"loss": 9.1903,
|
1707 |
+
"step": 17400
|
1708 |
+
},
|
1709 |
+
{
|
1710 |
+
"epoch": 33.0,
|
1711 |
+
"eval_accuracy": 0.2003076923076923,
|
1712 |
+
"eval_loss": 12.143142700195312,
|
1713 |
+
"eval_runtime": 5.0907,
|
1714 |
+
"eval_samples_per_second": 98.218,
|
1715 |
+
"eval_steps_per_second": 12.375,
|
1716 |
+
"step": 17457
|
1717 |
+
},
|
1718 |
+
{
|
1719 |
+
"epoch": 33.0,
|
1720 |
+
"eval_exact_match": 0.0,
|
1721 |
+
"eval_f1": 0.0,
|
1722 |
+
"step": 17457
|
1723 |
+
},
|
1724 |
+
{
|
1725 |
+
"epoch": 33.0812854442344,
|
1726 |
+
"grad_norm": 2477152272384.0,
|
1727 |
+
"learning_rate": 3e-05,
|
1728 |
+
"loss": 10.3355,
|
1729 |
+
"step": 17500
|
1730 |
+
},
|
1731 |
+
{
|
1732 |
+
"epoch": 33.2703213610586,
|
1733 |
+
"grad_norm": 1226347118592.0,
|
1734 |
+
"learning_rate": 3e-05,
|
1735 |
+
"loss": 9.7714,
|
1736 |
+
"step": 17600
|
1737 |
+
},
|
1738 |
+
{
|
1739 |
+
"epoch": 33.459357277882795,
|
1740 |
+
"grad_norm": 6838296248320.0,
|
1741 |
+
"learning_rate": 3e-05,
|
1742 |
+
"loss": 9.9791,
|
1743 |
+
"step": 17700
|
1744 |
+
},
|
1745 |
+
{
|
1746 |
+
"epoch": 33.648393194707,
|
1747 |
+
"grad_norm": 1059247161344.0,
|
1748 |
+
"learning_rate": 3e-05,
|
1749 |
+
"loss": 8.9139,
|
1750 |
+
"step": 17800
|
1751 |
+
},
|
1752 |
+
{
|
1753 |
+
"epoch": 33.83742911153119,
|
1754 |
+
"grad_norm": 783651008.0,
|
1755 |
+
"learning_rate": 3e-05,
|
1756 |
+
"loss": 8.6886,
|
1757 |
+
"step": 17900
|
1758 |
+
},
|
1759 |
+
{
|
1760 |
+
"epoch": 34.0,
|
1761 |
+
"eval_accuracy": 0.2073846153846154,
|
1762 |
+
"eval_loss": 8.974021911621094,
|
1763 |
+
"eval_runtime": 5.0381,
|
1764 |
+
"eval_samples_per_second": 99.245,
|
1765 |
+
"eval_steps_per_second": 12.505,
|
1766 |
+
"step": 17986
|
1767 |
+
},
|
1768 |
+
{
|
1769 |
+
"epoch": 34.0,
|
1770 |
+
"eval_exact_match": 0.0,
|
1771 |
+
"eval_f1": 0.048600856756742646,
|
1772 |
+
"step": 17986
|
1773 |
+
},
|
1774 |
+
{
|
1775 |
+
"epoch": 34.02646502835539,
|
1776 |
+
"grad_norm": 950630948536320.0,
|
1777 |
+
"learning_rate": 3e-05,
|
1778 |
+
"loss": 9.4673,
|
1779 |
+
"step": 18000
|
1780 |
+
},
|
1781 |
+
{
|
1782 |
+
"epoch": 34.215500945179585,
|
1783 |
+
"grad_norm": 450093280.0,
|
1784 |
+
"learning_rate": 3e-05,
|
1785 |
+
"loss": 8.886,
|
1786 |
+
"step": 18100
|
1787 |
+
},
|
1788 |
+
{
|
1789 |
+
"epoch": 34.40453686200378,
|
1790 |
+
"grad_norm": 87242632.0,
|
1791 |
+
"learning_rate": 3e-05,
|
1792 |
+
"loss": 10.0368,
|
1793 |
+
"step": 18200
|
1794 |
+
},
|
1795 |
+
{
|
1796 |
+
"epoch": 34.593572778827976,
|
1797 |
+
"grad_norm": 57701945344.0,
|
1798 |
+
"learning_rate": 3e-05,
|
1799 |
+
"loss": 9.5221,
|
1800 |
+
"step": 18300
|
1801 |
+
},
|
1802 |
+
{
|
1803 |
+
"epoch": 34.78260869565217,
|
1804 |
+
"grad_norm": 1806314112.0,
|
1805 |
+
"learning_rate": 3e-05,
|
1806 |
+
"loss": 9.5085,
|
1807 |
+
"step": 18400
|
1808 |
+
},
|
1809 |
+
{
|
1810 |
+
"epoch": 34.97164461247637,
|
1811 |
+
"grad_norm": 2088303.0,
|
1812 |
+
"learning_rate": 3e-05,
|
1813 |
+
"loss": 9.166,
|
1814 |
+
"step": 18500
|
1815 |
+
},
|
1816 |
+
{
|
1817 |
+
"epoch": 35.0,
|
1818 |
+
"eval_accuracy": 0.22430769230769232,
|
1819 |
+
"eval_loss": 8.694220542907715,
|
1820 |
+
"eval_runtime": 6.1359,
|
1821 |
+
"eval_samples_per_second": 81.487,
|
1822 |
+
"eval_steps_per_second": 10.267,
|
1823 |
+
"step": 18515
|
1824 |
+
},
|
1825 |
+
{
|
1826 |
+
"epoch": 35.0,
|
1827 |
+
"eval_exact_match": 0.0,
|
1828 |
+
"eval_f1": 0.0,
|
1829 |
+
"step": 18515
|
1830 |
+
},
|
1831 |
+
{
|
1832 |
+
"epoch": 35.16068052930057,
|
1833 |
+
"grad_norm": 993076051968.0,
|
1834 |
+
"learning_rate": 3e-05,
|
1835 |
+
"loss": 9.1192,
|
1836 |
+
"step": 18600
|
1837 |
+
},
|
1838 |
+
{
|
1839 |
+
"epoch": 35.349716446124766,
|
1840 |
+
"grad_norm": 126950965248.0,
|
1841 |
+
"learning_rate": 3e-05,
|
1842 |
+
"loss": 8.607,
|
1843 |
+
"step": 18700
|
1844 |
+
},
|
1845 |
+
{
|
1846 |
+
"epoch": 35.53875236294896,
|
1847 |
+
"grad_norm": 816930160640.0,
|
1848 |
+
"learning_rate": 3e-05,
|
1849 |
+
"loss": 8.1998,
|
1850 |
+
"step": 18800
|
1851 |
+
},
|
1852 |
+
{
|
1853 |
+
"epoch": 35.72778827977316,
|
1854 |
+
"grad_norm": 80379322368.0,
|
1855 |
+
"learning_rate": 3e-05,
|
1856 |
+
"loss": 9.6685,
|
1857 |
+
"step": 18900
|
1858 |
+
},
|
1859 |
+
{
|
1860 |
+
"epoch": 35.916824196597354,
|
1861 |
+
"grad_norm": 34520543232.0,
|
1862 |
+
"learning_rate": 3e-05,
|
1863 |
+
"loss": 10.7328,
|
1864 |
+
"step": 19000
|
1865 |
+
},
|
1866 |
+
{
|
1867 |
+
"epoch": 36.0,
|
1868 |
+
"eval_accuracy": 0.23302564102564102,
|
1869 |
+
"eval_loss": 11.274002075195312,
|
1870 |
+
"eval_runtime": 5.2319,
|
1871 |
+
"eval_samples_per_second": 95.567,
|
1872 |
+
"eval_steps_per_second": 12.041,
|
1873 |
+
"step": 19044
|
1874 |
+
},
|
1875 |
+
{
|
1876 |
+
"epoch": 36.0,
|
1877 |
+
"eval_exact_match": 0.0,
|
1878 |
+
"eval_f1": 0.0,
|
1879 |
+
"step": 19044
|
1880 |
+
},
|
1881 |
+
{
|
1882 |
+
"epoch": 36.10586011342155,
|
1883 |
+
"grad_norm": 1875878400.0,
|
1884 |
+
"learning_rate": 3e-05,
|
1885 |
+
"loss": 10.3668,
|
1886 |
+
"step": 19100
|
1887 |
+
},
|
1888 |
+
{
|
1889 |
+
"epoch": 36.294896030245745,
|
1890 |
+
"grad_norm": 2460117368832.0,
|
1891 |
+
"learning_rate": 3e-05,
|
1892 |
+
"loss": 10.0013,
|
1893 |
+
"step": 19200
|
1894 |
+
},
|
1895 |
+
{
|
1896 |
+
"epoch": 36.48393194706994,
|
1897 |
+
"grad_norm": 1223807336448.0,
|
1898 |
+
"learning_rate": 3e-05,
|
1899 |
+
"loss": 9.937,
|
1900 |
+
"step": 19300
|
1901 |
+
},
|
1902 |
+
{
|
1903 |
+
"epoch": 36.67296786389414,
|
1904 |
+
"grad_norm": 116063576064.0,
|
1905 |
+
"learning_rate": 3e-05,
|
1906 |
+
"loss": 10.6002,
|
1907 |
+
"step": 19400
|
1908 |
+
},
|
1909 |
+
{
|
1910 |
+
"epoch": 36.86200378071834,
|
1911 |
+
"grad_norm": 19499100160.0,
|
1912 |
+
"learning_rate": 3e-05,
|
1913 |
+
"loss": 10.7473,
|
1914 |
+
"step": 19500
|
1915 |
+
},
|
1916 |
+
{
|
1917 |
+
"epoch": 37.0,
|
1918 |
+
"eval_accuracy": 0.24461538461538462,
|
1919 |
+
"eval_loss": 9.314560890197754,
|
1920 |
+
"eval_runtime": 5.0345,
|
1921 |
+
"eval_samples_per_second": 99.315,
|
1922 |
+
"eval_steps_per_second": 12.514,
|
1923 |
+
"step": 19573
|
1924 |
+
},
|
1925 |
+
{
|
1926 |
+
"epoch": 37.0,
|
1927 |
+
"eval_exact_match": 0.0,
|
1928 |
+
"eval_f1": 0.01212121212121212,
|
1929 |
+
"step": 19573
|
1930 |
+
}
|
1931 |
+
],
|
1932 |
+
"logging_steps": 100,
|
1933 |
+
"max_steps": 26450,
|
1934 |
+
"num_input_tokens_seen": 0,
|
1935 |
+
"num_train_epochs": 50,
|
1936 |
+
"save_steps": 500,
|
1937 |
+
"stateful_callbacks": {
|
1938 |
+
"TrainerControl": {
|
1939 |
+
"args": {
|
1940 |
+
"should_epoch_stop": false,
|
1941 |
+
"should_evaluate": false,
|
1942 |
+
"should_log": false,
|
1943 |
+
"should_save": true,
|
1944 |
+
"should_training_stop": false
|
1945 |
+
},
|
1946 |
+
"attributes": {}
|
1947 |
+
}
|
1948 |
+
},
|
1949 |
+
"total_flos": 1.5927922886380093e+18,
|
1950 |
+
"train_batch_size": 2,
|
1951 |
+
"trial_name": null,
|
1952 |
+
"trial_params": null
|
1953 |
+
}
|
checkpoint-19573/training_args.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:3895667ea94410ecd283430024ce6f8557816c9d86efd03783de342ef988d79c
|
3 |
+
size 5304
|
checkpoint-19573/vocab.json
ADDED
The diff for this file is too large to render.
See raw diff
|
|