Training in progress, epoch 48, checkpoint
Browse files- checkpoint-25392/README.md +9 -0
- checkpoint-25392/adapter_config.json +25 -0
- checkpoint-25392/adapter_model.safetensors +3 -0
- checkpoint-25392/added_tokens.json +5 -0
- checkpoint-25392/merges.txt +0 -0
- checkpoint-25392/optimizer.pt +3 -0
- checkpoint-25392/rng_state_0.pth +3 -0
- checkpoint-25392/rng_state_1.pth +3 -0
- checkpoint-25392/rng_state_2.pth +3 -0
- checkpoint-25392/rng_state_3.pth +3 -0
- checkpoint-25392/scheduler.pt +3 -0
- checkpoint-25392/special_tokens_map.json +14 -0
- checkpoint-25392/tokenizer.json +0 -0
- checkpoint-25392/tokenizer_config.json +43 -0
- checkpoint-25392/trainer_state.json +2524 -0
- checkpoint-25392/training_args.bin +3 -0
- checkpoint-25392/vocab.json +0 -0
checkpoint-25392/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-25392/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-25392/adapter_model.safetensors
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:6b8eb86c25a3bc17ea62bcde61656e8aa8ba1d83586950960eda5673b754a435
|
3 |
+
size 112130416
|
checkpoint-25392/added_tokens.json
ADDED
@@ -0,0 +1,5 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"<|endoftext|>": 151643,
|
3 |
+
"<|im_end|>": 151645,
|
4 |
+
"<|im_start|>": 151644
|
5 |
+
}
|
checkpoint-25392/merges.txt
ADDED
The diff for this file is too large to render.
See raw diff
|
|
checkpoint-25392/optimizer.pt
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:ea60b90ae1bebd9df1e9937171e946e70a646fdc63ca0a28765272db024f9442
|
3 |
+
size 224537202
|
checkpoint-25392/rng_state_0.pth
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:d5b9c8a5189c5696b681d17dc9237b2de9f2f078e958c3667d02c2bdbd04ca2c
|
3 |
+
size 15024
|
checkpoint-25392/rng_state_1.pth
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:af02a069a9b6a1d4746f40dffecb808e6c368b8edc0234632e964ff6d3eab0ac
|
3 |
+
size 15024
|
checkpoint-25392/rng_state_2.pth
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:4aaea61f18ea337d362f70700bb1cb1fe4c0909d07d37d98b0bce676ed5319e2
|
3 |
+
size 15024
|
checkpoint-25392/rng_state_3.pth
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:d35e7cdaf1227e2aa82021618a2cc4e23451c03d40d58798cc1bdda0a1b7c350
|
3 |
+
size 15024
|
checkpoint-25392/scheduler.pt
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:f86c80039322cf3f5506cd85e08b2a8b2698fbd0c15ae91e833f3bb76d030c45
|
3 |
+
size 1064
|
checkpoint-25392/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-25392/tokenizer.json
ADDED
The diff for this file is too large to render.
See raw diff
|
|
checkpoint-25392/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-25392/trainer_state.json
ADDED
@@ -0,0 +1,2524 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"best_metric": null,
|
3 |
+
"best_model_checkpoint": null,
|
4 |
+
"epoch": 48.0,
|
5 |
+
"eval_steps": 500,
|
6 |
+
"global_step": 25392,
|
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": 0.5107442140579224,
|
14 |
+
"learning_rate": 5e-05,
|
15 |
+
"loss": 2.0482,
|
16 |
+
"step": 100
|
17 |
+
},
|
18 |
+
{
|
19 |
+
"epoch": 0.3780718336483932,
|
20 |
+
"grad_norm": 0.5563424229621887,
|
21 |
+
"learning_rate": 5e-05,
|
22 |
+
"loss": 1.8835,
|
23 |
+
"step": 200
|
24 |
+
},
|
25 |
+
{
|
26 |
+
"epoch": 0.5671077504725898,
|
27 |
+
"grad_norm": 0.4735371172428131,
|
28 |
+
"learning_rate": 5e-05,
|
29 |
+
"loss": 1.8512,
|
30 |
+
"step": 300
|
31 |
+
},
|
32 |
+
{
|
33 |
+
"epoch": 0.7561436672967864,
|
34 |
+
"grad_norm": 0.4543495178222656,
|
35 |
+
"learning_rate": 5e-05,
|
36 |
+
"loss": 1.8489,
|
37 |
+
"step": 400
|
38 |
+
},
|
39 |
+
{
|
40 |
+
"epoch": 0.945179584120983,
|
41 |
+
"grad_norm": 0.5007112622261047,
|
42 |
+
"learning_rate": 5e-05,
|
43 |
+
"loss": 1.8439,
|
44 |
+
"step": 500
|
45 |
+
},
|
46 |
+
{
|
47 |
+
"epoch": 1.0,
|
48 |
+
"eval_accuracy": 0.5733333333333334,
|
49 |
+
"eval_loss": 1.6194567680358887,
|
50 |
+
"eval_runtime": 5.4154,
|
51 |
+
"eval_samples_per_second": 92.33,
|
52 |
+
"eval_steps_per_second": 11.634,
|
53 |
+
"step": 529
|
54 |
+
},
|
55 |
+
{
|
56 |
+
"epoch": 1.0,
|
57 |
+
"eval_exact_match": 17.0,
|
58 |
+
"eval_f1": 26.880952380952387,
|
59 |
+
"step": 529
|
60 |
+
},
|
61 |
+
{
|
62 |
+
"epoch": 1.1342155009451795,
|
63 |
+
"grad_norm": 0.5181360244750977,
|
64 |
+
"learning_rate": 5e-05,
|
65 |
+
"loss": 1.8217,
|
66 |
+
"step": 600
|
67 |
+
},
|
68 |
+
{
|
69 |
+
"epoch": 1.3232514177693762,
|
70 |
+
"grad_norm": 1.1348918676376343,
|
71 |
+
"learning_rate": 5e-05,
|
72 |
+
"loss": 1.7866,
|
73 |
+
"step": 700
|
74 |
+
},
|
75 |
+
{
|
76 |
+
"epoch": 1.5122873345935728,
|
77 |
+
"grad_norm": 0.661758303642273,
|
78 |
+
"learning_rate": 5e-05,
|
79 |
+
"loss": 1.7934,
|
80 |
+
"step": 800
|
81 |
+
},
|
82 |
+
{
|
83 |
+
"epoch": 1.7013232514177694,
|
84 |
+
"grad_norm": 0.5634772777557373,
|
85 |
+
"learning_rate": 5e-05,
|
86 |
+
"loss": 1.7731,
|
87 |
+
"step": 900
|
88 |
+
},
|
89 |
+
{
|
90 |
+
"epoch": 1.8903591682419658,
|
91 |
+
"grad_norm": 0.6698930263519287,
|
92 |
+
"learning_rate": 5e-05,
|
93 |
+
"loss": 1.7969,
|
94 |
+
"step": 1000
|
95 |
+
},
|
96 |
+
{
|
97 |
+
"epoch": 2.0,
|
98 |
+
"eval_accuracy": 0.5751794871794872,
|
99 |
+
"eval_loss": 1.603736162185669,
|
100 |
+
"eval_runtime": 5.0986,
|
101 |
+
"eval_samples_per_second": 98.066,
|
102 |
+
"eval_steps_per_second": 12.356,
|
103 |
+
"step": 1058
|
104 |
+
},
|
105 |
+
{
|
106 |
+
"epoch": 2.0,
|
107 |
+
"eval_exact_match": 17.6,
|
108 |
+
"eval_f1": 27.218095238095238,
|
109 |
+
"step": 1058
|
110 |
+
},
|
111 |
+
{
|
112 |
+
"epoch": 2.0793950850661624,
|
113 |
+
"grad_norm": 0.6265193223953247,
|
114 |
+
"learning_rate": 5e-05,
|
115 |
+
"loss": 1.7416,
|
116 |
+
"step": 1100
|
117 |
+
},
|
118 |
+
{
|
119 |
+
"epoch": 2.268431001890359,
|
120 |
+
"grad_norm": 0.812364935874939,
|
121 |
+
"learning_rate": 5e-05,
|
122 |
+
"loss": 1.6957,
|
123 |
+
"step": 1200
|
124 |
+
},
|
125 |
+
{
|
126 |
+
"epoch": 2.4574669187145557,
|
127 |
+
"grad_norm": 1.0972926616668701,
|
128 |
+
"learning_rate": 5e-05,
|
129 |
+
"loss": 1.7033,
|
130 |
+
"step": 1300
|
131 |
+
},
|
132 |
+
{
|
133 |
+
"epoch": 2.6465028355387523,
|
134 |
+
"grad_norm": 1.0831658840179443,
|
135 |
+
"learning_rate": 5e-05,
|
136 |
+
"loss": 1.6935,
|
137 |
+
"step": 1400
|
138 |
+
},
|
139 |
+
{
|
140 |
+
"epoch": 2.835538752362949,
|
141 |
+
"grad_norm": 1.0022494792938232,
|
142 |
+
"learning_rate": 5e-05,
|
143 |
+
"loss": 1.6799,
|
144 |
+
"step": 1500
|
145 |
+
},
|
146 |
+
{
|
147 |
+
"epoch": 3.0,
|
148 |
+
"eval_accuracy": 0.573025641025641,
|
149 |
+
"eval_loss": 1.6428505182266235,
|
150 |
+
"eval_runtime": 5.1488,
|
151 |
+
"eval_samples_per_second": 97.111,
|
152 |
+
"eval_steps_per_second": 12.236,
|
153 |
+
"step": 1587
|
154 |
+
},
|
155 |
+
{
|
156 |
+
"epoch": 3.0,
|
157 |
+
"eval_exact_match": 20.2,
|
158 |
+
"eval_f1": 27.93809523809523,
|
159 |
+
"step": 1587
|
160 |
+
},
|
161 |
+
{
|
162 |
+
"epoch": 3.0245746691871456,
|
163 |
+
"grad_norm": 0.8240587711334229,
|
164 |
+
"learning_rate": 5e-05,
|
165 |
+
"loss": 1.6618,
|
166 |
+
"step": 1600
|
167 |
+
},
|
168 |
+
{
|
169 |
+
"epoch": 3.213610586011342,
|
170 |
+
"grad_norm": 1.2994645833969116,
|
171 |
+
"learning_rate": 5e-05,
|
172 |
+
"loss": 1.5736,
|
173 |
+
"step": 1700
|
174 |
+
},
|
175 |
+
{
|
176 |
+
"epoch": 3.402646502835539,
|
177 |
+
"grad_norm": 1.0891159772872925,
|
178 |
+
"learning_rate": 5e-05,
|
179 |
+
"loss": 1.5798,
|
180 |
+
"step": 1800
|
181 |
+
},
|
182 |
+
{
|
183 |
+
"epoch": 3.5916824196597354,
|
184 |
+
"grad_norm": 1.3826475143432617,
|
185 |
+
"learning_rate": 5e-05,
|
186 |
+
"loss": 1.5957,
|
187 |
+
"step": 1900
|
188 |
+
},
|
189 |
+
{
|
190 |
+
"epoch": 3.780718336483932,
|
191 |
+
"grad_norm": 0.9901586174964905,
|
192 |
+
"learning_rate": 5e-05,
|
193 |
+
"loss": 1.6026,
|
194 |
+
"step": 2000
|
195 |
+
},
|
196 |
+
{
|
197 |
+
"epoch": 3.9697542533081287,
|
198 |
+
"grad_norm": 1.0607813596725464,
|
199 |
+
"learning_rate": 5e-05,
|
200 |
+
"loss": 1.6106,
|
201 |
+
"step": 2100
|
202 |
+
},
|
203 |
+
{
|
204 |
+
"epoch": 4.0,
|
205 |
+
"eval_accuracy": 0.5708717948717948,
|
206 |
+
"eval_loss": 1.7143596410751343,
|
207 |
+
"eval_runtime": 5.3942,
|
208 |
+
"eval_samples_per_second": 92.693,
|
209 |
+
"eval_steps_per_second": 11.679,
|
210 |
+
"step": 2116
|
211 |
+
},
|
212 |
+
{
|
213 |
+
"epoch": 4.0,
|
214 |
+
"eval_exact_match": 19.0,
|
215 |
+
"eval_f1": 27.71428571428571,
|
216 |
+
"step": 2116
|
217 |
+
},
|
218 |
+
{
|
219 |
+
"epoch": 4.158790170132325,
|
220 |
+
"grad_norm": 1.3189178705215454,
|
221 |
+
"learning_rate": 5e-05,
|
222 |
+
"loss": 1.5029,
|
223 |
+
"step": 2200
|
224 |
+
},
|
225 |
+
{
|
226 |
+
"epoch": 4.3478260869565215,
|
227 |
+
"grad_norm": 1.062267541885376,
|
228 |
+
"learning_rate": 5e-05,
|
229 |
+
"loss": 1.51,
|
230 |
+
"step": 2300
|
231 |
+
},
|
232 |
+
{
|
233 |
+
"epoch": 4.536862003780718,
|
234 |
+
"grad_norm": 1.3810843229293823,
|
235 |
+
"learning_rate": 5e-05,
|
236 |
+
"loss": 1.4867,
|
237 |
+
"step": 2400
|
238 |
+
},
|
239 |
+
{
|
240 |
+
"epoch": 4.725897920604915,
|
241 |
+
"grad_norm": 1.1991653442382812,
|
242 |
+
"learning_rate": 5e-05,
|
243 |
+
"loss": 1.501,
|
244 |
+
"step": 2500
|
245 |
+
},
|
246 |
+
{
|
247 |
+
"epoch": 4.914933837429111,
|
248 |
+
"grad_norm": 1.310654878616333,
|
249 |
+
"learning_rate": 5e-05,
|
250 |
+
"loss": 1.5034,
|
251 |
+
"step": 2600
|
252 |
+
},
|
253 |
+
{
|
254 |
+
"epoch": 5.0,
|
255 |
+
"eval_accuracy": 0.5681025641025641,
|
256 |
+
"eval_loss": 1.8328346014022827,
|
257 |
+
"eval_runtime": 5.7763,
|
258 |
+
"eval_samples_per_second": 86.561,
|
259 |
+
"eval_steps_per_second": 10.907,
|
260 |
+
"step": 2645
|
261 |
+
},
|
262 |
+
{
|
263 |
+
"epoch": 5.0,
|
264 |
+
"eval_exact_match": 18.8,
|
265 |
+
"eval_f1": 26.6015873015873,
|
266 |
+
"step": 2645
|
267 |
+
},
|
268 |
+
{
|
269 |
+
"epoch": 5.103969754253308,
|
270 |
+
"grad_norm": 1.1406631469726562,
|
271 |
+
"learning_rate": 5e-05,
|
272 |
+
"loss": 1.4605,
|
273 |
+
"step": 2700
|
274 |
+
},
|
275 |
+
{
|
276 |
+
"epoch": 5.293005671077505,
|
277 |
+
"grad_norm": 1.447221279144287,
|
278 |
+
"learning_rate": 5e-05,
|
279 |
+
"loss": 1.4204,
|
280 |
+
"step": 2800
|
281 |
+
},
|
282 |
+
{
|
283 |
+
"epoch": 5.482041587901701,
|
284 |
+
"grad_norm": 1.461799144744873,
|
285 |
+
"learning_rate": 5e-05,
|
286 |
+
"loss": 1.4233,
|
287 |
+
"step": 2900
|
288 |
+
},
|
289 |
+
{
|
290 |
+
"epoch": 5.671077504725898,
|
291 |
+
"grad_norm": 1.4184092283248901,
|
292 |
+
"learning_rate": 5e-05,
|
293 |
+
"loss": 1.4265,
|
294 |
+
"step": 3000
|
295 |
+
},
|
296 |
+
{
|
297 |
+
"epoch": 5.8601134215500945,
|
298 |
+
"grad_norm": 0.9850455522537231,
|
299 |
+
"learning_rate": 5e-05,
|
300 |
+
"loss": 1.4139,
|
301 |
+
"step": 3100
|
302 |
+
},
|
303 |
+
{
|
304 |
+
"epoch": 6.0,
|
305 |
+
"eval_accuracy": 0.564974358974359,
|
306 |
+
"eval_loss": 1.931396484375,
|
307 |
+
"eval_runtime": 5.3358,
|
308 |
+
"eval_samples_per_second": 93.706,
|
309 |
+
"eval_steps_per_second": 11.807,
|
310 |
+
"step": 3174
|
311 |
+
},
|
312 |
+
{
|
313 |
+
"epoch": 6.0,
|
314 |
+
"eval_exact_match": 17.6,
|
315 |
+
"eval_f1": 25.464285714285715,
|
316 |
+
"step": 3174
|
317 |
+
},
|
318 |
+
{
|
319 |
+
"epoch": 6.049149338374291,
|
320 |
+
"grad_norm": 1.2274116277694702,
|
321 |
+
"learning_rate": 5e-05,
|
322 |
+
"loss": 1.4159,
|
323 |
+
"step": 3200
|
324 |
+
},
|
325 |
+
{
|
326 |
+
"epoch": 6.238185255198488,
|
327 |
+
"grad_norm": 1.2107988595962524,
|
328 |
+
"learning_rate": 5e-05,
|
329 |
+
"loss": 1.3579,
|
330 |
+
"step": 3300
|
331 |
+
},
|
332 |
+
{
|
333 |
+
"epoch": 6.427221172022684,
|
334 |
+
"grad_norm": 1.2965952157974243,
|
335 |
+
"learning_rate": 5e-05,
|
336 |
+
"loss": 1.3485,
|
337 |
+
"step": 3400
|
338 |
+
},
|
339 |
+
{
|
340 |
+
"epoch": 6.616257088846881,
|
341 |
+
"grad_norm": 1.3035151958465576,
|
342 |
+
"learning_rate": 5e-05,
|
343 |
+
"loss": 1.36,
|
344 |
+
"step": 3500
|
345 |
+
},
|
346 |
+
{
|
347 |
+
"epoch": 6.805293005671078,
|
348 |
+
"grad_norm": 1.1840296983718872,
|
349 |
+
"learning_rate": 5e-05,
|
350 |
+
"loss": 1.3599,
|
351 |
+
"step": 3600
|
352 |
+
},
|
353 |
+
{
|
354 |
+
"epoch": 6.994328922495274,
|
355 |
+
"grad_norm": 1.780287265777588,
|
356 |
+
"learning_rate": 5e-05,
|
357 |
+
"loss": 1.355,
|
358 |
+
"step": 3700
|
359 |
+
},
|
360 |
+
{
|
361 |
+
"epoch": 7.0,
|
362 |
+
"eval_accuracy": 0.565076923076923,
|
363 |
+
"eval_loss": 1.9659509658813477,
|
364 |
+
"eval_runtime": 5.3127,
|
365 |
+
"eval_samples_per_second": 94.114,
|
366 |
+
"eval_steps_per_second": 11.858,
|
367 |
+
"step": 3703
|
368 |
+
},
|
369 |
+
{
|
370 |
+
"epoch": 7.0,
|
371 |
+
"eval_exact_match": 18.0,
|
372 |
+
"eval_f1": 26.546666666666663,
|
373 |
+
"step": 3703
|
374 |
+
},
|
375 |
+
{
|
376 |
+
"epoch": 7.183364839319471,
|
377 |
+
"grad_norm": 1.1608421802520752,
|
378 |
+
"learning_rate": 5e-05,
|
379 |
+
"loss": 1.298,
|
380 |
+
"step": 3800
|
381 |
+
},
|
382 |
+
{
|
383 |
+
"epoch": 7.3724007561436675,
|
384 |
+
"grad_norm": 1.083270788192749,
|
385 |
+
"learning_rate": 5e-05,
|
386 |
+
"loss": 1.3088,
|
387 |
+
"step": 3900
|
388 |
+
},
|
389 |
+
{
|
390 |
+
"epoch": 7.561436672967864,
|
391 |
+
"grad_norm": 1.3489899635314941,
|
392 |
+
"learning_rate": 5e-05,
|
393 |
+
"loss": 1.299,
|
394 |
+
"step": 4000
|
395 |
+
},
|
396 |
+
{
|
397 |
+
"epoch": 7.750472589792061,
|
398 |
+
"grad_norm": 1.1663447618484497,
|
399 |
+
"learning_rate": 5e-05,
|
400 |
+
"loss": 1.2841,
|
401 |
+
"step": 4100
|
402 |
+
},
|
403 |
+
{
|
404 |
+
"epoch": 7.939508506616257,
|
405 |
+
"grad_norm": 1.9308583736419678,
|
406 |
+
"learning_rate": 5e-05,
|
407 |
+
"loss": 1.314,
|
408 |
+
"step": 4200
|
409 |
+
},
|
410 |
+
{
|
411 |
+
"epoch": 8.0,
|
412 |
+
"eval_accuracy": 0.5616923076923077,
|
413 |
+
"eval_loss": 2.1164443492889404,
|
414 |
+
"eval_runtime": 5.2165,
|
415 |
+
"eval_samples_per_second": 95.85,
|
416 |
+
"eval_steps_per_second": 12.077,
|
417 |
+
"step": 4232
|
418 |
+
},
|
419 |
+
{
|
420 |
+
"epoch": 8.0,
|
421 |
+
"eval_exact_match": 17.0,
|
422 |
+
"eval_f1": 24.659841269841277,
|
423 |
+
"step": 4232
|
424 |
+
},
|
425 |
+
{
|
426 |
+
"epoch": 8.128544423440454,
|
427 |
+
"grad_norm": 1.109632134437561,
|
428 |
+
"learning_rate": 5e-05,
|
429 |
+
"loss": 1.2558,
|
430 |
+
"step": 4300
|
431 |
+
},
|
432 |
+
{
|
433 |
+
"epoch": 8.31758034026465,
|
434 |
+
"grad_norm": 1.1132491827011108,
|
435 |
+
"learning_rate": 5e-05,
|
436 |
+
"loss": 1.2264,
|
437 |
+
"step": 4400
|
438 |
+
},
|
439 |
+
{
|
440 |
+
"epoch": 8.506616257088847,
|
441 |
+
"grad_norm": 1.1017369031906128,
|
442 |
+
"learning_rate": 5e-05,
|
443 |
+
"loss": 1.2649,
|
444 |
+
"step": 4500
|
445 |
+
},
|
446 |
+
{
|
447 |
+
"epoch": 8.695652173913043,
|
448 |
+
"grad_norm": 1.0874963998794556,
|
449 |
+
"learning_rate": 5e-05,
|
450 |
+
"loss": 1.2512,
|
451 |
+
"step": 4600
|
452 |
+
},
|
453 |
+
{
|
454 |
+
"epoch": 8.88468809073724,
|
455 |
+
"grad_norm": 1.2617064714431763,
|
456 |
+
"learning_rate": 5e-05,
|
457 |
+
"loss": 1.261,
|
458 |
+
"step": 4700
|
459 |
+
},
|
460 |
+
{
|
461 |
+
"epoch": 9.0,
|
462 |
+
"eval_accuracy": 0.5622051282051282,
|
463 |
+
"eval_loss": 2.1635327339172363,
|
464 |
+
"eval_runtime": 5.0832,
|
465 |
+
"eval_samples_per_second": 98.363,
|
466 |
+
"eval_steps_per_second": 12.394,
|
467 |
+
"step": 4761
|
468 |
+
},
|
469 |
+
{
|
470 |
+
"epoch": 9.0,
|
471 |
+
"eval_exact_match": 17.8,
|
472 |
+
"eval_f1": 25.451428571428576,
|
473 |
+
"step": 4761
|
474 |
+
},
|
475 |
+
{
|
476 |
+
"epoch": 9.073724007561436,
|
477 |
+
"grad_norm": 1.2258027791976929,
|
478 |
+
"learning_rate": 5e-05,
|
479 |
+
"loss": 1.2298,
|
480 |
+
"step": 4800
|
481 |
+
},
|
482 |
+
{
|
483 |
+
"epoch": 9.262759924385634,
|
484 |
+
"grad_norm": 1.8261148929595947,
|
485 |
+
"learning_rate": 5e-05,
|
486 |
+
"loss": 1.1748,
|
487 |
+
"step": 4900
|
488 |
+
},
|
489 |
+
{
|
490 |
+
"epoch": 9.45179584120983,
|
491 |
+
"grad_norm": 1.1221072673797607,
|
492 |
+
"learning_rate": 5e-05,
|
493 |
+
"loss": 1.19,
|
494 |
+
"step": 5000
|
495 |
+
},
|
496 |
+
{
|
497 |
+
"epoch": 9.640831758034027,
|
498 |
+
"grad_norm": 1.3320170640945435,
|
499 |
+
"learning_rate": 5e-05,
|
500 |
+
"loss": 1.1932,
|
501 |
+
"step": 5100
|
502 |
+
},
|
503 |
+
{
|
504 |
+
"epoch": 9.829867674858223,
|
505 |
+
"grad_norm": 1.194077730178833,
|
506 |
+
"learning_rate": 5e-05,
|
507 |
+
"loss": 1.217,
|
508 |
+
"step": 5200
|
509 |
+
},
|
510 |
+
{
|
511 |
+
"epoch": 10.0,
|
512 |
+
"eval_accuracy": 0.5612820512820513,
|
513 |
+
"eval_loss": 2.244041919708252,
|
514 |
+
"eval_runtime": 5.1606,
|
515 |
+
"eval_samples_per_second": 96.889,
|
516 |
+
"eval_steps_per_second": 12.208,
|
517 |
+
"step": 5290
|
518 |
+
},
|
519 |
+
{
|
520 |
+
"epoch": 10.0,
|
521 |
+
"eval_exact_match": 18.4,
|
522 |
+
"eval_f1": 26.067662337662348,
|
523 |
+
"step": 5290
|
524 |
+
},
|
525 |
+
{
|
526 |
+
"epoch": 10.01890359168242,
|
527 |
+
"grad_norm": 1.1352814435958862,
|
528 |
+
"learning_rate": 5e-05,
|
529 |
+
"loss": 1.2093,
|
530 |
+
"step": 5300
|
531 |
+
},
|
532 |
+
{
|
533 |
+
"epoch": 10.207939508506616,
|
534 |
+
"grad_norm": 1.2741838693618774,
|
535 |
+
"learning_rate": 5e-05,
|
536 |
+
"loss": 1.1467,
|
537 |
+
"step": 5400
|
538 |
+
},
|
539 |
+
{
|
540 |
+
"epoch": 10.396975425330814,
|
541 |
+
"grad_norm": 1.6901249885559082,
|
542 |
+
"learning_rate": 5e-05,
|
543 |
+
"loss": 1.1288,
|
544 |
+
"step": 5500
|
545 |
+
},
|
546 |
+
{
|
547 |
+
"epoch": 10.58601134215501,
|
548 |
+
"grad_norm": 1.1944465637207031,
|
549 |
+
"learning_rate": 5e-05,
|
550 |
+
"loss": 1.1399,
|
551 |
+
"step": 5600
|
552 |
+
},
|
553 |
+
{
|
554 |
+
"epoch": 10.775047258979207,
|
555 |
+
"grad_norm": 1.2234572172164917,
|
556 |
+
"learning_rate": 5e-05,
|
557 |
+
"loss": 1.1678,
|
558 |
+
"step": 5700
|
559 |
+
},
|
560 |
+
{
|
561 |
+
"epoch": 10.964083175803403,
|
562 |
+
"grad_norm": 1.2781612873077393,
|
563 |
+
"learning_rate": 5e-05,
|
564 |
+
"loss": 1.146,
|
565 |
+
"step": 5800
|
566 |
+
},
|
567 |
+
{
|
568 |
+
"epoch": 11.0,
|
569 |
+
"eval_accuracy": 0.5618461538461539,
|
570 |
+
"eval_loss": 2.296630859375,
|
571 |
+
"eval_runtime": 5.1078,
|
572 |
+
"eval_samples_per_second": 97.889,
|
573 |
+
"eval_steps_per_second": 12.334,
|
574 |
+
"step": 5819
|
575 |
+
},
|
576 |
+
{
|
577 |
+
"epoch": 11.0,
|
578 |
+
"eval_exact_match": 16.8,
|
579 |
+
"eval_f1": 23.63305916305917,
|
580 |
+
"step": 5819
|
581 |
+
},
|
582 |
+
{
|
583 |
+
"epoch": 11.1531190926276,
|
584 |
+
"grad_norm": 1.138975739479065,
|
585 |
+
"learning_rate": 5e-05,
|
586 |
+
"loss": 1.0842,
|
587 |
+
"step": 5900
|
588 |
+
},
|
589 |
+
{
|
590 |
+
"epoch": 11.342155009451796,
|
591 |
+
"grad_norm": 1.727989912033081,
|
592 |
+
"learning_rate": 5e-05,
|
593 |
+
"loss": 1.0806,
|
594 |
+
"step": 6000
|
595 |
+
},
|
596 |
+
{
|
597 |
+
"epoch": 11.531190926275993,
|
598 |
+
"grad_norm": 1.3104358911514282,
|
599 |
+
"learning_rate": 5e-05,
|
600 |
+
"loss": 1.0953,
|
601 |
+
"step": 6100
|
602 |
+
},
|
603 |
+
{
|
604 |
+
"epoch": 11.720226843100189,
|
605 |
+
"grad_norm": 1.3629850149154663,
|
606 |
+
"learning_rate": 5e-05,
|
607 |
+
"loss": 1.0945,
|
608 |
+
"step": 6200
|
609 |
+
},
|
610 |
+
{
|
611 |
+
"epoch": 11.909262759924385,
|
612 |
+
"grad_norm": 1.3276056051254272,
|
613 |
+
"learning_rate": 5e-05,
|
614 |
+
"loss": 1.1069,
|
615 |
+
"step": 6300
|
616 |
+
},
|
617 |
+
{
|
618 |
+
"epoch": 12.0,
|
619 |
+
"eval_accuracy": 0.562,
|
620 |
+
"eval_loss": 2.3118155002593994,
|
621 |
+
"eval_runtime": 5.3837,
|
622 |
+
"eval_samples_per_second": 92.873,
|
623 |
+
"eval_steps_per_second": 11.702,
|
624 |
+
"step": 6348
|
625 |
+
},
|
626 |
+
{
|
627 |
+
"epoch": 12.0,
|
628 |
+
"eval_exact_match": 17.4,
|
629 |
+
"eval_f1": 24.43924963924964,
|
630 |
+
"step": 6348
|
631 |
+
},
|
632 |
+
{
|
633 |
+
"epoch": 12.098298676748582,
|
634 |
+
"grad_norm": 1.2608078718185425,
|
635 |
+
"learning_rate": 5e-05,
|
636 |
+
"loss": 1.0418,
|
637 |
+
"step": 6400
|
638 |
+
},
|
639 |
+
{
|
640 |
+
"epoch": 12.287334593572778,
|
641 |
+
"grad_norm": 1.4456945657730103,
|
642 |
+
"learning_rate": 5e-05,
|
643 |
+
"loss": 1.0258,
|
644 |
+
"step": 6500
|
645 |
+
},
|
646 |
+
{
|
647 |
+
"epoch": 12.476370510396976,
|
648 |
+
"grad_norm": 1.4195287227630615,
|
649 |
+
"learning_rate": 5e-05,
|
650 |
+
"loss": 1.0255,
|
651 |
+
"step": 6600
|
652 |
+
},
|
653 |
+
{
|
654 |
+
"epoch": 12.665406427221171,
|
655 |
+
"grad_norm": 1.2595598697662354,
|
656 |
+
"learning_rate": 5e-05,
|
657 |
+
"loss": 1.0511,
|
658 |
+
"step": 6700
|
659 |
+
},
|
660 |
+
{
|
661 |
+
"epoch": 12.854442344045369,
|
662 |
+
"grad_norm": 1.5901292562484741,
|
663 |
+
"learning_rate": 5e-05,
|
664 |
+
"loss": 1.045,
|
665 |
+
"step": 6800
|
666 |
+
},
|
667 |
+
{
|
668 |
+
"epoch": 13.0,
|
669 |
+
"eval_accuracy": 0.5614871794871795,
|
670 |
+
"eval_loss": 2.3580987453460693,
|
671 |
+
"eval_runtime": 5.3496,
|
672 |
+
"eval_samples_per_second": 93.466,
|
673 |
+
"eval_steps_per_second": 11.777,
|
674 |
+
"step": 6877
|
675 |
+
},
|
676 |
+
{
|
677 |
+
"epoch": 13.0,
|
678 |
+
"eval_exact_match": 17.2,
|
679 |
+
"eval_f1": 24.350995670995676,
|
680 |
+
"step": 6877
|
681 |
+
},
|
682 |
+
{
|
683 |
+
"epoch": 13.043478260869565,
|
684 |
+
"grad_norm": 1.471511721611023,
|
685 |
+
"learning_rate": 5e-05,
|
686 |
+
"loss": 1.0404,
|
687 |
+
"step": 6900
|
688 |
+
},
|
689 |
+
{
|
690 |
+
"epoch": 13.232514177693762,
|
691 |
+
"grad_norm": 1.2928757667541504,
|
692 |
+
"learning_rate": 5e-05,
|
693 |
+
"loss": 0.9573,
|
694 |
+
"step": 7000
|
695 |
+
},
|
696 |
+
{
|
697 |
+
"epoch": 13.421550094517958,
|
698 |
+
"grad_norm": 1.5545622110366821,
|
699 |
+
"learning_rate": 5e-05,
|
700 |
+
"loss": 0.9583,
|
701 |
+
"step": 7100
|
702 |
+
},
|
703 |
+
{
|
704 |
+
"epoch": 13.610586011342155,
|
705 |
+
"grad_norm": 1.6132512092590332,
|
706 |
+
"learning_rate": 5e-05,
|
707 |
+
"loss": 0.981,
|
708 |
+
"step": 7200
|
709 |
+
},
|
710 |
+
{
|
711 |
+
"epoch": 13.799621928166351,
|
712 |
+
"grad_norm": 1.7548742294311523,
|
713 |
+
"learning_rate": 5e-05,
|
714 |
+
"loss": 0.9999,
|
715 |
+
"step": 7300
|
716 |
+
},
|
717 |
+
{
|
718 |
+
"epoch": 13.988657844990549,
|
719 |
+
"grad_norm": 1.4219907522201538,
|
720 |
+
"learning_rate": 5e-05,
|
721 |
+
"loss": 1.0028,
|
722 |
+
"step": 7400
|
723 |
+
},
|
724 |
+
{
|
725 |
+
"epoch": 14.0,
|
726 |
+
"eval_accuracy": 0.5618974358974359,
|
727 |
+
"eval_loss": 2.413728952407837,
|
728 |
+
"eval_runtime": 5.0464,
|
729 |
+
"eval_samples_per_second": 99.08,
|
730 |
+
"eval_steps_per_second": 12.484,
|
731 |
+
"step": 7406
|
732 |
+
},
|
733 |
+
{
|
734 |
+
"epoch": 14.0,
|
735 |
+
"eval_exact_match": 16.6,
|
736 |
+
"eval_f1": 24.201789321789327,
|
737 |
+
"step": 7406
|
738 |
+
},
|
739 |
+
{
|
740 |
+
"epoch": 14.177693761814744,
|
741 |
+
"grad_norm": 1.5799965858459473,
|
742 |
+
"learning_rate": 5e-05,
|
743 |
+
"loss": 0.9096,
|
744 |
+
"step": 7500
|
745 |
+
},
|
746 |
+
{
|
747 |
+
"epoch": 14.366729678638942,
|
748 |
+
"grad_norm": 1.5961594581604004,
|
749 |
+
"learning_rate": 5e-05,
|
750 |
+
"loss": 0.9075,
|
751 |
+
"step": 7600
|
752 |
+
},
|
753 |
+
{
|
754 |
+
"epoch": 14.555765595463138,
|
755 |
+
"grad_norm": 1.5756512880325317,
|
756 |
+
"learning_rate": 5e-05,
|
757 |
+
"loss": 0.9222,
|
758 |
+
"step": 7700
|
759 |
+
},
|
760 |
+
{
|
761 |
+
"epoch": 14.744801512287335,
|
762 |
+
"grad_norm": 1.515251636505127,
|
763 |
+
"learning_rate": 5e-05,
|
764 |
+
"loss": 0.9445,
|
765 |
+
"step": 7800
|
766 |
+
},
|
767 |
+
{
|
768 |
+
"epoch": 14.93383742911153,
|
769 |
+
"grad_norm": 2.6804299354553223,
|
770 |
+
"learning_rate": 5e-05,
|
771 |
+
"loss": 0.9153,
|
772 |
+
"step": 7900
|
773 |
+
},
|
774 |
+
{
|
775 |
+
"epoch": 15.0,
|
776 |
+
"eval_accuracy": 0.5596923076923077,
|
777 |
+
"eval_loss": 2.4756176471710205,
|
778 |
+
"eval_runtime": 5.1956,
|
779 |
+
"eval_samples_per_second": 96.236,
|
780 |
+
"eval_steps_per_second": 12.126,
|
781 |
+
"step": 7935
|
782 |
+
},
|
783 |
+
{
|
784 |
+
"epoch": 15.0,
|
785 |
+
"eval_exact_match": 16.8,
|
786 |
+
"eval_f1": 24.146709956709966,
|
787 |
+
"step": 7935
|
788 |
+
},
|
789 |
+
{
|
790 |
+
"epoch": 15.122873345935728,
|
791 |
+
"grad_norm": 1.7786214351654053,
|
792 |
+
"learning_rate": 5e-05,
|
793 |
+
"loss": 0.8874,
|
794 |
+
"step": 8000
|
795 |
+
},
|
796 |
+
{
|
797 |
+
"epoch": 15.311909262759924,
|
798 |
+
"grad_norm": 1.709060788154602,
|
799 |
+
"learning_rate": 5e-05,
|
800 |
+
"loss": 0.8538,
|
801 |
+
"step": 8100
|
802 |
+
},
|
803 |
+
{
|
804 |
+
"epoch": 15.500945179584122,
|
805 |
+
"grad_norm": 2.3245434761047363,
|
806 |
+
"learning_rate": 5e-05,
|
807 |
+
"loss": 0.8591,
|
808 |
+
"step": 8200
|
809 |
+
},
|
810 |
+
{
|
811 |
+
"epoch": 15.689981096408317,
|
812 |
+
"grad_norm": 1.6101887226104736,
|
813 |
+
"learning_rate": 5e-05,
|
814 |
+
"loss": 0.8593,
|
815 |
+
"step": 8300
|
816 |
+
},
|
817 |
+
{
|
818 |
+
"epoch": 15.879017013232515,
|
819 |
+
"grad_norm": 1.5917792320251465,
|
820 |
+
"learning_rate": 5e-05,
|
821 |
+
"loss": 0.8748,
|
822 |
+
"step": 8400
|
823 |
+
},
|
824 |
+
{
|
825 |
+
"epoch": 16.0,
|
826 |
+
"eval_accuracy": 0.5604615384615385,
|
827 |
+
"eval_loss": 2.4434447288513184,
|
828 |
+
"eval_runtime": 5.1072,
|
829 |
+
"eval_samples_per_second": 97.901,
|
830 |
+
"eval_steps_per_second": 12.336,
|
831 |
+
"step": 8464
|
832 |
+
},
|
833 |
+
{
|
834 |
+
"epoch": 16.0,
|
835 |
+
"eval_exact_match": 15.6,
|
836 |
+
"eval_f1": 22.917027417027416,
|
837 |
+
"step": 8464
|
838 |
+
},
|
839 |
+
{
|
840 |
+
"epoch": 16.068052930056712,
|
841 |
+
"grad_norm": 1.6529980897903442,
|
842 |
+
"learning_rate": 5e-05,
|
843 |
+
"loss": 0.8413,
|
844 |
+
"step": 8500
|
845 |
+
},
|
846 |
+
{
|
847 |
+
"epoch": 16.257088846880908,
|
848 |
+
"grad_norm": 1.931520938873291,
|
849 |
+
"learning_rate": 5e-05,
|
850 |
+
"loss": 0.7863,
|
851 |
+
"step": 8600
|
852 |
+
},
|
853 |
+
{
|
854 |
+
"epoch": 16.446124763705104,
|
855 |
+
"grad_norm": 1.8401894569396973,
|
856 |
+
"learning_rate": 5e-05,
|
857 |
+
"loss": 0.799,
|
858 |
+
"step": 8700
|
859 |
+
},
|
860 |
+
{
|
861 |
+
"epoch": 16.6351606805293,
|
862 |
+
"grad_norm": 1.7952015399932861,
|
863 |
+
"learning_rate": 5e-05,
|
864 |
+
"loss": 0.8238,
|
865 |
+
"step": 8800
|
866 |
+
},
|
867 |
+
{
|
868 |
+
"epoch": 16.8241965973535,
|
869 |
+
"grad_norm": 1.8028634786605835,
|
870 |
+
"learning_rate": 5e-05,
|
871 |
+
"loss": 0.807,
|
872 |
+
"step": 8900
|
873 |
+
},
|
874 |
+
{
|
875 |
+
"epoch": 17.0,
|
876 |
+
"eval_accuracy": 0.5606153846153846,
|
877 |
+
"eval_loss": 2.4520959854125977,
|
878 |
+
"eval_runtime": 5.1769,
|
879 |
+
"eval_samples_per_second": 96.584,
|
880 |
+
"eval_steps_per_second": 12.17,
|
881 |
+
"step": 8993
|
882 |
+
},
|
883 |
+
{
|
884 |
+
"epoch": 17.0,
|
885 |
+
"eval_exact_match": 17.4,
|
886 |
+
"eval_f1": 25.104011544011545,
|
887 |
+
"step": 8993
|
888 |
+
},
|
889 |
+
{
|
890 |
+
"epoch": 17.013232514177695,
|
891 |
+
"grad_norm": 1.7197688817977905,
|
892 |
+
"learning_rate": 5e-05,
|
893 |
+
"loss": 0.8249,
|
894 |
+
"step": 9000
|
895 |
+
},
|
896 |
+
{
|
897 |
+
"epoch": 17.20226843100189,
|
898 |
+
"grad_norm": 1.768463134765625,
|
899 |
+
"learning_rate": 5e-05,
|
900 |
+
"loss": 0.7304,
|
901 |
+
"step": 9100
|
902 |
+
},
|
903 |
+
{
|
904 |
+
"epoch": 17.391304347826086,
|
905 |
+
"grad_norm": 1.9696024656295776,
|
906 |
+
"learning_rate": 5e-05,
|
907 |
+
"loss": 0.7445,
|
908 |
+
"step": 9200
|
909 |
+
},
|
910 |
+
{
|
911 |
+
"epoch": 17.58034026465028,
|
912 |
+
"grad_norm": 2.093703269958496,
|
913 |
+
"learning_rate": 5e-05,
|
914 |
+
"loss": 0.755,
|
915 |
+
"step": 9300
|
916 |
+
},
|
917 |
+
{
|
918 |
+
"epoch": 17.76937618147448,
|
919 |
+
"grad_norm": 1.909550666809082,
|
920 |
+
"learning_rate": 5e-05,
|
921 |
+
"loss": 0.7698,
|
922 |
+
"step": 9400
|
923 |
+
},
|
924 |
+
{
|
925 |
+
"epoch": 17.958412098298677,
|
926 |
+
"grad_norm": 1.6835806369781494,
|
927 |
+
"learning_rate": 5e-05,
|
928 |
+
"loss": 0.7711,
|
929 |
+
"step": 9500
|
930 |
+
},
|
931 |
+
{
|
932 |
+
"epoch": 18.0,
|
933 |
+
"eval_accuracy": 0.5603589743589743,
|
934 |
+
"eval_loss": 2.473491668701172,
|
935 |
+
"eval_runtime": 5.4257,
|
936 |
+
"eval_samples_per_second": 92.155,
|
937 |
+
"eval_steps_per_second": 11.611,
|
938 |
+
"step": 9522
|
939 |
+
},
|
940 |
+
{
|
941 |
+
"epoch": 18.0,
|
942 |
+
"eval_exact_match": 17.0,
|
943 |
+
"eval_f1": 24.074487734487732,
|
944 |
+
"step": 9522
|
945 |
+
},
|
946 |
+
{
|
947 |
+
"epoch": 18.147448015122873,
|
948 |
+
"grad_norm": 2.0483736991882324,
|
949 |
+
"learning_rate": 5e-05,
|
950 |
+
"loss": 0.7048,
|
951 |
+
"step": 9600
|
952 |
+
},
|
953 |
+
{
|
954 |
+
"epoch": 18.33648393194707,
|
955 |
+
"grad_norm": 1.987457275390625,
|
956 |
+
"learning_rate": 5e-05,
|
957 |
+
"loss": 0.6863,
|
958 |
+
"step": 9700
|
959 |
+
},
|
960 |
+
{
|
961 |
+
"epoch": 18.525519848771268,
|
962 |
+
"grad_norm": 2.1744563579559326,
|
963 |
+
"learning_rate": 5e-05,
|
964 |
+
"loss": 0.7045,
|
965 |
+
"step": 9800
|
966 |
+
},
|
967 |
+
{
|
968 |
+
"epoch": 18.714555765595463,
|
969 |
+
"grad_norm": 1.9871633052825928,
|
970 |
+
"learning_rate": 5e-05,
|
971 |
+
"loss": 0.7076,
|
972 |
+
"step": 9900
|
973 |
+
},
|
974 |
+
{
|
975 |
+
"epoch": 18.90359168241966,
|
976 |
+
"grad_norm": 1.8895001411437988,
|
977 |
+
"learning_rate": 5e-05,
|
978 |
+
"loss": 0.7202,
|
979 |
+
"step": 10000
|
980 |
+
},
|
981 |
+
{
|
982 |
+
"epoch": 19.0,
|
983 |
+
"eval_accuracy": 0.5601538461538461,
|
984 |
+
"eval_loss": 2.5168700218200684,
|
985 |
+
"eval_runtime": 5.0434,
|
986 |
+
"eval_samples_per_second": 99.14,
|
987 |
+
"eval_steps_per_second": 12.492,
|
988 |
+
"step": 10051
|
989 |
+
},
|
990 |
+
{
|
991 |
+
"epoch": 19.0,
|
992 |
+
"eval_exact_match": 17.2,
|
993 |
+
"eval_f1": 24.260360750360757,
|
994 |
+
"step": 10051
|
995 |
+
},
|
996 |
+
{
|
997 |
+
"epoch": 19.092627599243855,
|
998 |
+
"grad_norm": 1.7914209365844727,
|
999 |
+
"learning_rate": 5e-05,
|
1000 |
+
"loss": 0.6728,
|
1001 |
+
"step": 10100
|
1002 |
+
},
|
1003 |
+
{
|
1004 |
+
"epoch": 19.281663516068054,
|
1005 |
+
"grad_norm": 2.198495388031006,
|
1006 |
+
"learning_rate": 5e-05,
|
1007 |
+
"loss": 0.6241,
|
1008 |
+
"step": 10200
|
1009 |
+
},
|
1010 |
+
{
|
1011 |
+
"epoch": 19.47069943289225,
|
1012 |
+
"grad_norm": 1.81365966796875,
|
1013 |
+
"learning_rate": 5e-05,
|
1014 |
+
"loss": 0.6558,
|
1015 |
+
"step": 10300
|
1016 |
+
},
|
1017 |
+
{
|
1018 |
+
"epoch": 19.659735349716446,
|
1019 |
+
"grad_norm": 1.6571800708770752,
|
1020 |
+
"learning_rate": 5e-05,
|
1021 |
+
"loss": 0.6652,
|
1022 |
+
"step": 10400
|
1023 |
+
},
|
1024 |
+
{
|
1025 |
+
"epoch": 19.84877126654064,
|
1026 |
+
"grad_norm": 1.6469954252243042,
|
1027 |
+
"learning_rate": 5e-05,
|
1028 |
+
"loss": 0.6637,
|
1029 |
+
"step": 10500
|
1030 |
+
},
|
1031 |
+
{
|
1032 |
+
"epoch": 20.0,
|
1033 |
+
"eval_accuracy": 0.5596410256410257,
|
1034 |
+
"eval_loss": 2.516073703765869,
|
1035 |
+
"eval_runtime": 5.8234,
|
1036 |
+
"eval_samples_per_second": 85.86,
|
1037 |
+
"eval_steps_per_second": 10.818,
|
1038 |
+
"step": 10580
|
1039 |
+
},
|
1040 |
+
{
|
1041 |
+
"epoch": 20.0,
|
1042 |
+
"eval_exact_match": 17.6,
|
1043 |
+
"eval_f1": 25.225079365079367,
|
1044 |
+
"step": 10580
|
1045 |
+
},
|
1046 |
+
{
|
1047 |
+
"epoch": 20.03780718336484,
|
1048 |
+
"grad_norm": 1.7114077806472778,
|
1049 |
+
"learning_rate": 5e-05,
|
1050 |
+
"loss": 0.66,
|
1051 |
+
"step": 10600
|
1052 |
+
},
|
1053 |
+
{
|
1054 |
+
"epoch": 20.226843100189036,
|
1055 |
+
"grad_norm": 1.9962202310562134,
|
1056 |
+
"learning_rate": 5e-05,
|
1057 |
+
"loss": 0.5938,
|
1058 |
+
"step": 10700
|
1059 |
+
},
|
1060 |
+
{
|
1061 |
+
"epoch": 20.415879017013232,
|
1062 |
+
"grad_norm": 2.397592067718506,
|
1063 |
+
"learning_rate": 5e-05,
|
1064 |
+
"loss": 0.6042,
|
1065 |
+
"step": 10800
|
1066 |
+
},
|
1067 |
+
{
|
1068 |
+
"epoch": 20.604914933837428,
|
1069 |
+
"grad_norm": 2.0724213123321533,
|
1070 |
+
"learning_rate": 5e-05,
|
1071 |
+
"loss": 0.6124,
|
1072 |
+
"step": 10900
|
1073 |
+
},
|
1074 |
+
{
|
1075 |
+
"epoch": 20.793950850661627,
|
1076 |
+
"grad_norm": 1.9385457038879395,
|
1077 |
+
"learning_rate": 5e-05,
|
1078 |
+
"loss": 0.6078,
|
1079 |
+
"step": 11000
|
1080 |
+
},
|
1081 |
+
{
|
1082 |
+
"epoch": 20.982986767485823,
|
1083 |
+
"grad_norm": 2.034963607788086,
|
1084 |
+
"learning_rate": 5e-05,
|
1085 |
+
"loss": 0.6257,
|
1086 |
+
"step": 11100
|
1087 |
+
},
|
1088 |
+
{
|
1089 |
+
"epoch": 21.0,
|
1090 |
+
"eval_accuracy": 0.5597948717948718,
|
1091 |
+
"eval_loss": 2.5497183799743652,
|
1092 |
+
"eval_runtime": 5.0442,
|
1093 |
+
"eval_samples_per_second": 99.123,
|
1094 |
+
"eval_steps_per_second": 12.489,
|
1095 |
+
"step": 11109
|
1096 |
+
},
|
1097 |
+
{
|
1098 |
+
"epoch": 21.0,
|
1099 |
+
"eval_exact_match": 17.4,
|
1100 |
+
"eval_f1": 24.755238095238095,
|
1101 |
+
"step": 11109
|
1102 |
+
},
|
1103 |
+
{
|
1104 |
+
"epoch": 21.17202268431002,
|
1105 |
+
"grad_norm": 1.9768177270889282,
|
1106 |
+
"learning_rate": 5e-05,
|
1107 |
+
"loss": 0.5458,
|
1108 |
+
"step": 11200
|
1109 |
+
},
|
1110 |
+
{
|
1111 |
+
"epoch": 21.361058601134214,
|
1112 |
+
"grad_norm": 2.14420747756958,
|
1113 |
+
"learning_rate": 5e-05,
|
1114 |
+
"loss": 0.558,
|
1115 |
+
"step": 11300
|
1116 |
+
},
|
1117 |
+
{
|
1118 |
+
"epoch": 21.550094517958414,
|
1119 |
+
"grad_norm": 2.5267951488494873,
|
1120 |
+
"learning_rate": 5e-05,
|
1121 |
+
"loss": 0.5646,
|
1122 |
+
"step": 11400
|
1123 |
+
},
|
1124 |
+
{
|
1125 |
+
"epoch": 21.73913043478261,
|
1126 |
+
"grad_norm": 2.114386558532715,
|
1127 |
+
"learning_rate": 5e-05,
|
1128 |
+
"loss": 0.572,
|
1129 |
+
"step": 11500
|
1130 |
+
},
|
1131 |
+
{
|
1132 |
+
"epoch": 21.928166351606805,
|
1133 |
+
"grad_norm": 2.424217700958252,
|
1134 |
+
"learning_rate": 5e-05,
|
1135 |
+
"loss": 0.5928,
|
1136 |
+
"step": 11600
|
1137 |
+
},
|
1138 |
+
{
|
1139 |
+
"epoch": 22.0,
|
1140 |
+
"eval_accuracy": 0.5593846153846154,
|
1141 |
+
"eval_loss": 2.579535722732544,
|
1142 |
+
"eval_runtime": 5.7849,
|
1143 |
+
"eval_samples_per_second": 86.433,
|
1144 |
+
"eval_steps_per_second": 10.89,
|
1145 |
+
"step": 11638
|
1146 |
+
},
|
1147 |
+
{
|
1148 |
+
"epoch": 22.0,
|
1149 |
+
"eval_exact_match": 16.8,
|
1150 |
+
"eval_f1": 23.666507936507937,
|
1151 |
+
"step": 11638
|
1152 |
+
},
|
1153 |
+
{
|
1154 |
+
"epoch": 22.117202268431,
|
1155 |
+
"grad_norm": 1.7813427448272705,
|
1156 |
+
"learning_rate": 5e-05,
|
1157 |
+
"loss": 0.5342,
|
1158 |
+
"step": 11700
|
1159 |
+
},
|
1160 |
+
{
|
1161 |
+
"epoch": 22.3062381852552,
|
1162 |
+
"grad_norm": 1.989479422569275,
|
1163 |
+
"learning_rate": 5e-05,
|
1164 |
+
"loss": 0.5157,
|
1165 |
+
"step": 11800
|
1166 |
+
},
|
1167 |
+
{
|
1168 |
+
"epoch": 22.495274102079396,
|
1169 |
+
"grad_norm": 1.9694567918777466,
|
1170 |
+
"learning_rate": 5e-05,
|
1171 |
+
"loss": 0.5236,
|
1172 |
+
"step": 11900
|
1173 |
+
},
|
1174 |
+
{
|
1175 |
+
"epoch": 22.68431001890359,
|
1176 |
+
"grad_norm": 2.0226142406463623,
|
1177 |
+
"learning_rate": 5e-05,
|
1178 |
+
"loss": 0.5315,
|
1179 |
+
"step": 12000
|
1180 |
+
},
|
1181 |
+
{
|
1182 |
+
"epoch": 22.873345935727787,
|
1183 |
+
"grad_norm": 2.0204625129699707,
|
1184 |
+
"learning_rate": 5e-05,
|
1185 |
+
"loss": 0.5378,
|
1186 |
+
"step": 12100
|
1187 |
+
},
|
1188 |
+
{
|
1189 |
+
"epoch": 23.0,
|
1190 |
+
"eval_accuracy": 0.558974358974359,
|
1191 |
+
"eval_loss": 2.5709009170532227,
|
1192 |
+
"eval_runtime": 5.0788,
|
1193 |
+
"eval_samples_per_second": 98.449,
|
1194 |
+
"eval_steps_per_second": 12.405,
|
1195 |
+
"step": 12167
|
1196 |
+
},
|
1197 |
+
{
|
1198 |
+
"epoch": 23.0,
|
1199 |
+
"eval_exact_match": 16.4,
|
1200 |
+
"eval_f1": 23.69809523809524,
|
1201 |
+
"step": 12167
|
1202 |
+
},
|
1203 |
+
{
|
1204 |
+
"epoch": 23.062381852551987,
|
1205 |
+
"grad_norm": 2.0633811950683594,
|
1206 |
+
"learning_rate": 5e-05,
|
1207 |
+
"loss": 0.5183,
|
1208 |
+
"step": 12200
|
1209 |
+
},
|
1210 |
+
{
|
1211 |
+
"epoch": 23.251417769376182,
|
1212 |
+
"grad_norm": 2.3566079139709473,
|
1213 |
+
"learning_rate": 5e-05,
|
1214 |
+
"loss": 0.4761,
|
1215 |
+
"step": 12300
|
1216 |
+
},
|
1217 |
+
{
|
1218 |
+
"epoch": 23.440453686200378,
|
1219 |
+
"grad_norm": 2.368450880050659,
|
1220 |
+
"learning_rate": 5e-05,
|
1221 |
+
"loss": 0.487,
|
1222 |
+
"step": 12400
|
1223 |
+
},
|
1224 |
+
{
|
1225 |
+
"epoch": 23.629489603024574,
|
1226 |
+
"grad_norm": 2.1096372604370117,
|
1227 |
+
"learning_rate": 5e-05,
|
1228 |
+
"loss": 0.4945,
|
1229 |
+
"step": 12500
|
1230 |
+
},
|
1231 |
+
{
|
1232 |
+
"epoch": 23.81852551984877,
|
1233 |
+
"grad_norm": 2.2136454582214355,
|
1234 |
+
"learning_rate": 5e-05,
|
1235 |
+
"loss": 0.5035,
|
1236 |
+
"step": 12600
|
1237 |
+
},
|
1238 |
+
{
|
1239 |
+
"epoch": 24.0,
|
1240 |
+
"eval_accuracy": 0.5586153846153846,
|
1241 |
+
"eval_loss": 2.5898056030273438,
|
1242 |
+
"eval_runtime": 5.7458,
|
1243 |
+
"eval_samples_per_second": 87.021,
|
1244 |
+
"eval_steps_per_second": 10.965,
|
1245 |
+
"step": 12696
|
1246 |
+
},
|
1247 |
+
{
|
1248 |
+
"epoch": 24.0,
|
1249 |
+
"eval_exact_match": 17.8,
|
1250 |
+
"eval_f1": 24.89603174603175,
|
1251 |
+
"step": 12696
|
1252 |
+
},
|
1253 |
+
{
|
1254 |
+
"epoch": 24.00756143667297,
|
1255 |
+
"grad_norm": 2.317958116531372,
|
1256 |
+
"learning_rate": 5e-05,
|
1257 |
+
"loss": 0.5106,
|
1258 |
+
"step": 12700
|
1259 |
+
},
|
1260 |
+
{
|
1261 |
+
"epoch": 24.196597353497165,
|
1262 |
+
"grad_norm": 2.0783536434173584,
|
1263 |
+
"learning_rate": 5e-05,
|
1264 |
+
"loss": 0.4404,
|
1265 |
+
"step": 12800
|
1266 |
+
},
|
1267 |
+
{
|
1268 |
+
"epoch": 24.38563327032136,
|
1269 |
+
"grad_norm": 2.322791337966919,
|
1270 |
+
"learning_rate": 5e-05,
|
1271 |
+
"loss": 0.452,
|
1272 |
+
"step": 12900
|
1273 |
+
},
|
1274 |
+
{
|
1275 |
+
"epoch": 24.574669187145556,
|
1276 |
+
"grad_norm": 1.9763044118881226,
|
1277 |
+
"learning_rate": 5e-05,
|
1278 |
+
"loss": 0.4617,
|
1279 |
+
"step": 13000
|
1280 |
+
},
|
1281 |
+
{
|
1282 |
+
"epoch": 24.763705103969755,
|
1283 |
+
"grad_norm": 2.462392568588257,
|
1284 |
+
"learning_rate": 5e-05,
|
1285 |
+
"loss": 0.4731,
|
1286 |
+
"step": 13100
|
1287 |
+
},
|
1288 |
+
{
|
1289 |
+
"epoch": 24.95274102079395,
|
1290 |
+
"grad_norm": 2.2400307655334473,
|
1291 |
+
"learning_rate": 5e-05,
|
1292 |
+
"loss": 0.4751,
|
1293 |
+
"step": 13200
|
1294 |
+
},
|
1295 |
+
{
|
1296 |
+
"epoch": 25.0,
|
1297 |
+
"eval_accuracy": 0.5594871794871795,
|
1298 |
+
"eval_loss": 2.611313581466675,
|
1299 |
+
"eval_runtime": 5.3564,
|
1300 |
+
"eval_samples_per_second": 93.346,
|
1301 |
+
"eval_steps_per_second": 11.762,
|
1302 |
+
"step": 13225
|
1303 |
+
},
|
1304 |
+
{
|
1305 |
+
"epoch": 25.0,
|
1306 |
+
"eval_exact_match": 15.8,
|
1307 |
+
"eval_f1": 22.969047619047625,
|
1308 |
+
"step": 13225
|
1309 |
+
},
|
1310 |
+
{
|
1311 |
+
"epoch": 25.141776937618147,
|
1312 |
+
"grad_norm": 2.0931668281555176,
|
1313 |
+
"learning_rate": 5e-05,
|
1314 |
+
"loss": 0.4299,
|
1315 |
+
"step": 13300
|
1316 |
+
},
|
1317 |
+
{
|
1318 |
+
"epoch": 25.330812854442343,
|
1319 |
+
"grad_norm": 2.7056140899658203,
|
1320 |
+
"learning_rate": 5e-05,
|
1321 |
+
"loss": 0.422,
|
1322 |
+
"step": 13400
|
1323 |
+
},
|
1324 |
+
{
|
1325 |
+
"epoch": 25.519848771266542,
|
1326 |
+
"grad_norm": 1.9907532930374146,
|
1327 |
+
"learning_rate": 5e-05,
|
1328 |
+
"loss": 0.4374,
|
1329 |
+
"step": 13500
|
1330 |
+
},
|
1331 |
+
{
|
1332 |
+
"epoch": 25.708884688090738,
|
1333 |
+
"grad_norm": 2.3209221363067627,
|
1334 |
+
"learning_rate": 5e-05,
|
1335 |
+
"loss": 0.4376,
|
1336 |
+
"step": 13600
|
1337 |
+
},
|
1338 |
+
{
|
1339 |
+
"epoch": 25.897920604914933,
|
1340 |
+
"grad_norm": 3.2339212894439697,
|
1341 |
+
"learning_rate": 5e-05,
|
1342 |
+
"loss": 0.4432,
|
1343 |
+
"step": 13700
|
1344 |
+
},
|
1345 |
+
{
|
1346 |
+
"epoch": 26.0,
|
1347 |
+
"eval_accuracy": 0.5602051282051282,
|
1348 |
+
"eval_loss": 2.6182327270507812,
|
1349 |
+
"eval_runtime": 5.0321,
|
1350 |
+
"eval_samples_per_second": 99.362,
|
1351 |
+
"eval_steps_per_second": 12.52,
|
1352 |
+
"step": 13754
|
1353 |
+
},
|
1354 |
+
{
|
1355 |
+
"epoch": 26.0,
|
1356 |
+
"eval_exact_match": 15.4,
|
1357 |
+
"eval_f1": 23.059047619047618,
|
1358 |
+
"step": 13754
|
1359 |
+
},
|
1360 |
+
{
|
1361 |
+
"epoch": 26.08695652173913,
|
1362 |
+
"grad_norm": 1.9935250282287598,
|
1363 |
+
"learning_rate": 5e-05,
|
1364 |
+
"loss": 0.4265,
|
1365 |
+
"step": 13800
|
1366 |
+
},
|
1367 |
+
{
|
1368 |
+
"epoch": 26.27599243856333,
|
1369 |
+
"grad_norm": 2.5698649883270264,
|
1370 |
+
"learning_rate": 5e-05,
|
1371 |
+
"loss": 0.3949,
|
1372 |
+
"step": 13900
|
1373 |
+
},
|
1374 |
+
{
|
1375 |
+
"epoch": 26.465028355387524,
|
1376 |
+
"grad_norm": 2.568392038345337,
|
1377 |
+
"learning_rate": 5e-05,
|
1378 |
+
"loss": 0.4041,
|
1379 |
+
"step": 14000
|
1380 |
+
},
|
1381 |
+
{
|
1382 |
+
"epoch": 26.65406427221172,
|
1383 |
+
"grad_norm": 2.094651699066162,
|
1384 |
+
"learning_rate": 5e-05,
|
1385 |
+
"loss": 0.4128,
|
1386 |
+
"step": 14100
|
1387 |
+
},
|
1388 |
+
{
|
1389 |
+
"epoch": 26.843100189035916,
|
1390 |
+
"grad_norm": 2.240649700164795,
|
1391 |
+
"learning_rate": 5e-05,
|
1392 |
+
"loss": 0.4191,
|
1393 |
+
"step": 14200
|
1394 |
+
},
|
1395 |
+
{
|
1396 |
+
"epoch": 27.0,
|
1397 |
+
"eval_accuracy": 0.56,
|
1398 |
+
"eval_loss": 2.614851951599121,
|
1399 |
+
"eval_runtime": 5.636,
|
1400 |
+
"eval_samples_per_second": 88.715,
|
1401 |
+
"eval_steps_per_second": 11.178,
|
1402 |
+
"step": 14283
|
1403 |
+
},
|
1404 |
+
{
|
1405 |
+
"epoch": 27.0,
|
1406 |
+
"eval_exact_match": 17.2,
|
1407 |
+
"eval_f1": 24.941948051948046,
|
1408 |
+
"step": 14283
|
1409 |
+
},
|
1410 |
+
{
|
1411 |
+
"epoch": 27.032136105860115,
|
1412 |
+
"grad_norm": 1.8800567388534546,
|
1413 |
+
"learning_rate": 5e-05,
|
1414 |
+
"loss": 0.4135,
|
1415 |
+
"step": 14300
|
1416 |
+
},
|
1417 |
+
{
|
1418 |
+
"epoch": 27.22117202268431,
|
1419 |
+
"grad_norm": 2.525048017501831,
|
1420 |
+
"learning_rate": 5e-05,
|
1421 |
+
"loss": 0.3717,
|
1422 |
+
"step": 14400
|
1423 |
+
},
|
1424 |
+
{
|
1425 |
+
"epoch": 27.410207939508506,
|
1426 |
+
"grad_norm": 2.3803513050079346,
|
1427 |
+
"learning_rate": 5e-05,
|
1428 |
+
"loss": 0.3817,
|
1429 |
+
"step": 14500
|
1430 |
+
},
|
1431 |
+
{
|
1432 |
+
"epoch": 27.599243856332702,
|
1433 |
+
"grad_norm": 2.369356155395508,
|
1434 |
+
"learning_rate": 5e-05,
|
1435 |
+
"loss": 0.388,
|
1436 |
+
"step": 14600
|
1437 |
+
},
|
1438 |
+
{
|
1439 |
+
"epoch": 27.7882797731569,
|
1440 |
+
"grad_norm": 2.701702356338501,
|
1441 |
+
"learning_rate": 5e-05,
|
1442 |
+
"loss": 0.3943,
|
1443 |
+
"step": 14700
|
1444 |
+
},
|
1445 |
+
{
|
1446 |
+
"epoch": 27.977315689981097,
|
1447 |
+
"grad_norm": 2.2755825519561768,
|
1448 |
+
"learning_rate": 5e-05,
|
1449 |
+
"loss": 0.4015,
|
1450 |
+
"step": 14800
|
1451 |
+
},
|
1452 |
+
{
|
1453 |
+
"epoch": 28.0,
|
1454 |
+
"eval_accuracy": 0.5592820512820513,
|
1455 |
+
"eval_loss": 2.653672218322754,
|
1456 |
+
"eval_runtime": 5.3543,
|
1457 |
+
"eval_samples_per_second": 93.383,
|
1458 |
+
"eval_steps_per_second": 11.766,
|
1459 |
+
"step": 14812
|
1460 |
+
},
|
1461 |
+
{
|
1462 |
+
"epoch": 28.0,
|
1463 |
+
"eval_exact_match": 17.4,
|
1464 |
+
"eval_f1": 24.478888888888893,
|
1465 |
+
"step": 14812
|
1466 |
+
},
|
1467 |
+
{
|
1468 |
+
"epoch": 28.166351606805293,
|
1469 |
+
"grad_norm": 2.0458974838256836,
|
1470 |
+
"learning_rate": 5e-05,
|
1471 |
+
"loss": 0.3566,
|
1472 |
+
"step": 14900
|
1473 |
+
},
|
1474 |
+
{
|
1475 |
+
"epoch": 28.35538752362949,
|
1476 |
+
"grad_norm": 2.040117025375366,
|
1477 |
+
"learning_rate": 5e-05,
|
1478 |
+
"loss": 0.3602,
|
1479 |
+
"step": 15000
|
1480 |
+
},
|
1481 |
+
{
|
1482 |
+
"epoch": 28.544423440453688,
|
1483 |
+
"grad_norm": 1.947757363319397,
|
1484 |
+
"learning_rate": 5e-05,
|
1485 |
+
"loss": 0.3666,
|
1486 |
+
"step": 15100
|
1487 |
+
},
|
1488 |
+
{
|
1489 |
+
"epoch": 28.733459357277884,
|
1490 |
+
"grad_norm": 2.0905520915985107,
|
1491 |
+
"learning_rate": 5e-05,
|
1492 |
+
"loss": 0.3747,
|
1493 |
+
"step": 15200
|
1494 |
+
},
|
1495 |
+
{
|
1496 |
+
"epoch": 28.92249527410208,
|
1497 |
+
"grad_norm": 2.215851306915283,
|
1498 |
+
"learning_rate": 5e-05,
|
1499 |
+
"loss": 0.3798,
|
1500 |
+
"step": 15300
|
1501 |
+
},
|
1502 |
+
{
|
1503 |
+
"epoch": 29.0,
|
1504 |
+
"eval_accuracy": 0.5600512820512821,
|
1505 |
+
"eval_loss": 2.6304192543029785,
|
1506 |
+
"eval_runtime": 5.0359,
|
1507 |
+
"eval_samples_per_second": 99.288,
|
1508 |
+
"eval_steps_per_second": 12.51,
|
1509 |
+
"step": 15341
|
1510 |
+
},
|
1511 |
+
{
|
1512 |
+
"epoch": 29.0,
|
1513 |
+
"eval_exact_match": 16.2,
|
1514 |
+
"eval_f1": 23.909062049062054,
|
1515 |
+
"step": 15341
|
1516 |
+
},
|
1517 |
+
{
|
1518 |
+
"epoch": 29.111531190926275,
|
1519 |
+
"grad_norm": 2.5050926208496094,
|
1520 |
+
"learning_rate": 5e-05,
|
1521 |
+
"loss": 0.3538,
|
1522 |
+
"step": 15400
|
1523 |
+
},
|
1524 |
+
{
|
1525 |
+
"epoch": 29.300567107750474,
|
1526 |
+
"grad_norm": 2.0059566497802734,
|
1527 |
+
"learning_rate": 5e-05,
|
1528 |
+
"loss": 0.3365,
|
1529 |
+
"step": 15500
|
1530 |
+
},
|
1531 |
+
{
|
1532 |
+
"epoch": 29.48960302457467,
|
1533 |
+
"grad_norm": 2.7678000926971436,
|
1534 |
+
"learning_rate": 5e-05,
|
1535 |
+
"loss": 0.3523,
|
1536 |
+
"step": 15600
|
1537 |
+
},
|
1538 |
+
{
|
1539 |
+
"epoch": 29.678638941398866,
|
1540 |
+
"grad_norm": 2.6234450340270996,
|
1541 |
+
"learning_rate": 5e-05,
|
1542 |
+
"loss": 0.3552,
|
1543 |
+
"step": 15700
|
1544 |
+
},
|
1545 |
+
{
|
1546 |
+
"epoch": 29.86767485822306,
|
1547 |
+
"grad_norm": 1.9958972930908203,
|
1548 |
+
"learning_rate": 5e-05,
|
1549 |
+
"loss": 0.3629,
|
1550 |
+
"step": 15800
|
1551 |
+
},
|
1552 |
+
{
|
1553 |
+
"epoch": 30.0,
|
1554 |
+
"eval_accuracy": 0.56,
|
1555 |
+
"eval_loss": 2.6483516693115234,
|
1556 |
+
"eval_runtime": 5.0621,
|
1557 |
+
"eval_samples_per_second": 98.774,
|
1558 |
+
"eval_steps_per_second": 12.446,
|
1559 |
+
"step": 15870
|
1560 |
+
},
|
1561 |
+
{
|
1562 |
+
"epoch": 30.0,
|
1563 |
+
"eval_exact_match": 16.4,
|
1564 |
+
"eval_f1": 24.021746031746037,
|
1565 |
+
"step": 15870
|
1566 |
+
},
|
1567 |
+
{
|
1568 |
+
"epoch": 30.056710775047257,
|
1569 |
+
"grad_norm": 2.5568833351135254,
|
1570 |
+
"learning_rate": 5e-05,
|
1571 |
+
"loss": 0.3505,
|
1572 |
+
"step": 15900
|
1573 |
+
},
|
1574 |
+
{
|
1575 |
+
"epoch": 30.245746691871457,
|
1576 |
+
"grad_norm": 2.0313165187835693,
|
1577 |
+
"learning_rate": 5e-05,
|
1578 |
+
"loss": 0.3252,
|
1579 |
+
"step": 16000
|
1580 |
+
},
|
1581 |
+
{
|
1582 |
+
"epoch": 30.434782608695652,
|
1583 |
+
"grad_norm": 1.8152469396591187,
|
1584 |
+
"learning_rate": 5e-05,
|
1585 |
+
"loss": 0.335,
|
1586 |
+
"step": 16100
|
1587 |
+
},
|
1588 |
+
{
|
1589 |
+
"epoch": 30.623818525519848,
|
1590 |
+
"grad_norm": 2.0004210472106934,
|
1591 |
+
"learning_rate": 5e-05,
|
1592 |
+
"loss": 0.34,
|
1593 |
+
"step": 16200
|
1594 |
+
},
|
1595 |
+
{
|
1596 |
+
"epoch": 30.812854442344044,
|
1597 |
+
"grad_norm": 2.3558757305145264,
|
1598 |
+
"learning_rate": 5e-05,
|
1599 |
+
"loss": 0.3474,
|
1600 |
+
"step": 16300
|
1601 |
+
},
|
1602 |
+
{
|
1603 |
+
"epoch": 31.0,
|
1604 |
+
"eval_accuracy": 0.5605641025641026,
|
1605 |
+
"eval_loss": 2.6599507331848145,
|
1606 |
+
"eval_runtime": 5.3592,
|
1607 |
+
"eval_samples_per_second": 93.297,
|
1608 |
+
"eval_steps_per_second": 11.755,
|
1609 |
+
"step": 16399
|
1610 |
+
},
|
1611 |
+
{
|
1612 |
+
"epoch": 31.0,
|
1613 |
+
"eval_exact_match": 16.8,
|
1614 |
+
"eval_f1": 23.902857142857147,
|
1615 |
+
"step": 16399
|
1616 |
+
},
|
1617 |
+
{
|
1618 |
+
"epoch": 31.001890359168243,
|
1619 |
+
"grad_norm": 1.641440749168396,
|
1620 |
+
"learning_rate": 5e-05,
|
1621 |
+
"loss": 0.3478,
|
1622 |
+
"step": 16400
|
1623 |
+
},
|
1624 |
+
{
|
1625 |
+
"epoch": 31.19092627599244,
|
1626 |
+
"grad_norm": 2.425560712814331,
|
1627 |
+
"learning_rate": 5e-05,
|
1628 |
+
"loss": 0.3067,
|
1629 |
+
"step": 16500
|
1630 |
+
},
|
1631 |
+
{
|
1632 |
+
"epoch": 31.379962192816635,
|
1633 |
+
"grad_norm": 1.850221872329712,
|
1634 |
+
"learning_rate": 5e-05,
|
1635 |
+
"loss": 0.3202,
|
1636 |
+
"step": 16600
|
1637 |
+
},
|
1638 |
+
{
|
1639 |
+
"epoch": 31.56899810964083,
|
1640 |
+
"grad_norm": 1.4244951009750366,
|
1641 |
+
"learning_rate": 5e-05,
|
1642 |
+
"loss": 0.3301,
|
1643 |
+
"step": 16700
|
1644 |
+
},
|
1645 |
+
{
|
1646 |
+
"epoch": 31.75803402646503,
|
1647 |
+
"grad_norm": 2.093834638595581,
|
1648 |
+
"learning_rate": 5e-05,
|
1649 |
+
"loss": 0.3293,
|
1650 |
+
"step": 16800
|
1651 |
+
},
|
1652 |
+
{
|
1653 |
+
"epoch": 31.947069943289225,
|
1654 |
+
"grad_norm": 2.0703086853027344,
|
1655 |
+
"learning_rate": 5e-05,
|
1656 |
+
"loss": 0.3369,
|
1657 |
+
"step": 16900
|
1658 |
+
},
|
1659 |
+
{
|
1660 |
+
"epoch": 32.0,
|
1661 |
+
"eval_accuracy": 0.5597435897435897,
|
1662 |
+
"eval_loss": 2.6675195693969727,
|
1663 |
+
"eval_runtime": 5.1201,
|
1664 |
+
"eval_samples_per_second": 97.655,
|
1665 |
+
"eval_steps_per_second": 12.305,
|
1666 |
+
"step": 16928
|
1667 |
+
},
|
1668 |
+
{
|
1669 |
+
"epoch": 32.0,
|
1670 |
+
"eval_exact_match": 15.6,
|
1671 |
+
"eval_f1": 22.95126984126984,
|
1672 |
+
"step": 16928
|
1673 |
+
},
|
1674 |
+
{
|
1675 |
+
"epoch": 32.136105860113425,
|
1676 |
+
"grad_norm": 2.0121419429779053,
|
1677 |
+
"learning_rate": 5e-05,
|
1678 |
+
"loss": 0.3134,
|
1679 |
+
"step": 17000
|
1680 |
+
},
|
1681 |
+
{
|
1682 |
+
"epoch": 32.32514177693762,
|
1683 |
+
"grad_norm": 1.8783822059631348,
|
1684 |
+
"learning_rate": 5e-05,
|
1685 |
+
"loss": 0.3102,
|
1686 |
+
"step": 17100
|
1687 |
+
},
|
1688 |
+
{
|
1689 |
+
"epoch": 32.514177693761816,
|
1690 |
+
"grad_norm": 1.5319064855575562,
|
1691 |
+
"learning_rate": 5e-05,
|
1692 |
+
"loss": 0.3089,
|
1693 |
+
"step": 17200
|
1694 |
+
},
|
1695 |
+
{
|
1696 |
+
"epoch": 32.70321361058601,
|
1697 |
+
"grad_norm": 2.6580235958099365,
|
1698 |
+
"learning_rate": 5e-05,
|
1699 |
+
"loss": 0.3175,
|
1700 |
+
"step": 17300
|
1701 |
+
},
|
1702 |
+
{
|
1703 |
+
"epoch": 32.89224952741021,
|
1704 |
+
"grad_norm": 2.121335744857788,
|
1705 |
+
"learning_rate": 5e-05,
|
1706 |
+
"loss": 0.3275,
|
1707 |
+
"step": 17400
|
1708 |
+
},
|
1709 |
+
{
|
1710 |
+
"epoch": 33.0,
|
1711 |
+
"eval_accuracy": 0.5603589743589743,
|
1712 |
+
"eval_loss": 2.658357858657837,
|
1713 |
+
"eval_runtime": 5.4806,
|
1714 |
+
"eval_samples_per_second": 91.231,
|
1715 |
+
"eval_steps_per_second": 11.495,
|
1716 |
+
"step": 17457
|
1717 |
+
},
|
1718 |
+
{
|
1719 |
+
"epoch": 33.0,
|
1720 |
+
"eval_exact_match": 16.4,
|
1721 |
+
"eval_f1": 24.0452380952381,
|
1722 |
+
"step": 17457
|
1723 |
+
},
|
1724 |
+
{
|
1725 |
+
"epoch": 33.0812854442344,
|
1726 |
+
"grad_norm": 2.1282331943511963,
|
1727 |
+
"learning_rate": 5e-05,
|
1728 |
+
"loss": 0.311,
|
1729 |
+
"step": 17500
|
1730 |
+
},
|
1731 |
+
{
|
1732 |
+
"epoch": 33.2703213610586,
|
1733 |
+
"grad_norm": 2.212125062942505,
|
1734 |
+
"learning_rate": 5e-05,
|
1735 |
+
"loss": 0.2951,
|
1736 |
+
"step": 17600
|
1737 |
+
},
|
1738 |
+
{
|
1739 |
+
"epoch": 33.459357277882795,
|
1740 |
+
"grad_norm": 3.389835834503174,
|
1741 |
+
"learning_rate": 5e-05,
|
1742 |
+
"loss": 0.3068,
|
1743 |
+
"step": 17700
|
1744 |
+
},
|
1745 |
+
{
|
1746 |
+
"epoch": 33.648393194707,
|
1747 |
+
"grad_norm": 1.812578558921814,
|
1748 |
+
"learning_rate": 5e-05,
|
1749 |
+
"loss": 0.3091,
|
1750 |
+
"step": 17800
|
1751 |
+
},
|
1752 |
+
{
|
1753 |
+
"epoch": 33.83742911153119,
|
1754 |
+
"grad_norm": 2.434966802597046,
|
1755 |
+
"learning_rate": 5e-05,
|
1756 |
+
"loss": 0.311,
|
1757 |
+
"step": 17900
|
1758 |
+
},
|
1759 |
+
{
|
1760 |
+
"epoch": 34.0,
|
1761 |
+
"eval_accuracy": 0.5600512820512821,
|
1762 |
+
"eval_loss": 2.676753520965576,
|
1763 |
+
"eval_runtime": 5.3579,
|
1764 |
+
"eval_samples_per_second": 93.32,
|
1765 |
+
"eval_steps_per_second": 11.758,
|
1766 |
+
"step": 17986
|
1767 |
+
},
|
1768 |
+
{
|
1769 |
+
"epoch": 34.0,
|
1770 |
+
"eval_exact_match": 16.6,
|
1771 |
+
"eval_f1": 24.913650793650795,
|
1772 |
+
"step": 17986
|
1773 |
+
},
|
1774 |
+
{
|
1775 |
+
"epoch": 34.02646502835539,
|
1776 |
+
"grad_norm": 1.5352602005004883,
|
1777 |
+
"learning_rate": 5e-05,
|
1778 |
+
"loss": 0.3142,
|
1779 |
+
"step": 18000
|
1780 |
+
},
|
1781 |
+
{
|
1782 |
+
"epoch": 34.215500945179585,
|
1783 |
+
"grad_norm": 1.4355374574661255,
|
1784 |
+
"learning_rate": 5e-05,
|
1785 |
+
"loss": 0.2897,
|
1786 |
+
"step": 18100
|
1787 |
+
},
|
1788 |
+
{
|
1789 |
+
"epoch": 34.40453686200378,
|
1790 |
+
"grad_norm": 1.7993170022964478,
|
1791 |
+
"learning_rate": 5e-05,
|
1792 |
+
"loss": 0.2923,
|
1793 |
+
"step": 18200
|
1794 |
+
},
|
1795 |
+
{
|
1796 |
+
"epoch": 34.593572778827976,
|
1797 |
+
"grad_norm": 1.9478981494903564,
|
1798 |
+
"learning_rate": 5e-05,
|
1799 |
+
"loss": 0.3001,
|
1800 |
+
"step": 18300
|
1801 |
+
},
|
1802 |
+
{
|
1803 |
+
"epoch": 34.78260869565217,
|
1804 |
+
"grad_norm": 3.045933246612549,
|
1805 |
+
"learning_rate": 5e-05,
|
1806 |
+
"loss": 0.3034,
|
1807 |
+
"step": 18400
|
1808 |
+
},
|
1809 |
+
{
|
1810 |
+
"epoch": 34.97164461247637,
|
1811 |
+
"grad_norm": 1.8796017169952393,
|
1812 |
+
"learning_rate": 5e-05,
|
1813 |
+
"loss": 0.31,
|
1814 |
+
"step": 18500
|
1815 |
+
},
|
1816 |
+
{
|
1817 |
+
"epoch": 35.0,
|
1818 |
+
"eval_accuracy": 0.5601538461538461,
|
1819 |
+
"eval_loss": 2.6845180988311768,
|
1820 |
+
"eval_runtime": 5.7492,
|
1821 |
+
"eval_samples_per_second": 86.969,
|
1822 |
+
"eval_steps_per_second": 10.958,
|
1823 |
+
"step": 18515
|
1824 |
+
},
|
1825 |
+
{
|
1826 |
+
"epoch": 35.0,
|
1827 |
+
"eval_exact_match": 16.2,
|
1828 |
+
"eval_f1": 23.91369408369409,
|
1829 |
+
"step": 18515
|
1830 |
+
},
|
1831 |
+
{
|
1832 |
+
"epoch": 35.16068052930057,
|
1833 |
+
"grad_norm": 1.7619277238845825,
|
1834 |
+
"learning_rate": 5e-05,
|
1835 |
+
"loss": 0.2853,
|
1836 |
+
"step": 18600
|
1837 |
+
},
|
1838 |
+
{
|
1839 |
+
"epoch": 35.349716446124766,
|
1840 |
+
"grad_norm": 1.7933720350265503,
|
1841 |
+
"learning_rate": 5e-05,
|
1842 |
+
"loss": 0.2853,
|
1843 |
+
"step": 18700
|
1844 |
+
},
|
1845 |
+
{
|
1846 |
+
"epoch": 35.53875236294896,
|
1847 |
+
"grad_norm": 1.476181149482727,
|
1848 |
+
"learning_rate": 5e-05,
|
1849 |
+
"loss": 0.2955,
|
1850 |
+
"step": 18800
|
1851 |
+
},
|
1852 |
+
{
|
1853 |
+
"epoch": 35.72778827977316,
|
1854 |
+
"grad_norm": 1.6427425146102905,
|
1855 |
+
"learning_rate": 5e-05,
|
1856 |
+
"loss": 0.2934,
|
1857 |
+
"step": 18900
|
1858 |
+
},
|
1859 |
+
{
|
1860 |
+
"epoch": 35.916824196597354,
|
1861 |
+
"grad_norm": 1.7125171422958374,
|
1862 |
+
"learning_rate": 5e-05,
|
1863 |
+
"loss": 0.3009,
|
1864 |
+
"step": 19000
|
1865 |
+
},
|
1866 |
+
{
|
1867 |
+
"epoch": 36.0,
|
1868 |
+
"eval_accuracy": 0.5604615384615385,
|
1869 |
+
"eval_loss": 2.6663565635681152,
|
1870 |
+
"eval_runtime": 5.0382,
|
1871 |
+
"eval_samples_per_second": 99.241,
|
1872 |
+
"eval_steps_per_second": 12.504,
|
1873 |
+
"step": 19044
|
1874 |
+
},
|
1875 |
+
{
|
1876 |
+
"epoch": 36.0,
|
1877 |
+
"eval_exact_match": 16.0,
|
1878 |
+
"eval_f1": 23.221269841269848,
|
1879 |
+
"step": 19044
|
1880 |
+
},
|
1881 |
+
{
|
1882 |
+
"epoch": 36.10586011342155,
|
1883 |
+
"grad_norm": 1.757238507270813,
|
1884 |
+
"learning_rate": 5e-05,
|
1885 |
+
"loss": 0.286,
|
1886 |
+
"step": 19100
|
1887 |
+
},
|
1888 |
+
{
|
1889 |
+
"epoch": 36.294896030245745,
|
1890 |
+
"grad_norm": 1.9411671161651611,
|
1891 |
+
"learning_rate": 5e-05,
|
1892 |
+
"loss": 0.2761,
|
1893 |
+
"step": 19200
|
1894 |
+
},
|
1895 |
+
{
|
1896 |
+
"epoch": 36.48393194706994,
|
1897 |
+
"grad_norm": 2.305509328842163,
|
1898 |
+
"learning_rate": 5e-05,
|
1899 |
+
"loss": 0.2873,
|
1900 |
+
"step": 19300
|
1901 |
+
},
|
1902 |
+
{
|
1903 |
+
"epoch": 36.67296786389414,
|
1904 |
+
"grad_norm": 1.6525070667266846,
|
1905 |
+
"learning_rate": 5e-05,
|
1906 |
+
"loss": 0.2881,
|
1907 |
+
"step": 19400
|
1908 |
+
},
|
1909 |
+
{
|
1910 |
+
"epoch": 36.86200378071834,
|
1911 |
+
"grad_norm": 1.562827706336975,
|
1912 |
+
"learning_rate": 5e-05,
|
1913 |
+
"loss": 0.2959,
|
1914 |
+
"step": 19500
|
1915 |
+
},
|
1916 |
+
{
|
1917 |
+
"epoch": 37.0,
|
1918 |
+
"eval_accuracy": 0.5594871794871795,
|
1919 |
+
"eval_loss": 2.738389492034912,
|
1920 |
+
"eval_runtime": 5.0296,
|
1921 |
+
"eval_samples_per_second": 99.412,
|
1922 |
+
"eval_steps_per_second": 12.526,
|
1923 |
+
"step": 19573
|
1924 |
+
},
|
1925 |
+
{
|
1926 |
+
"epoch": 37.0,
|
1927 |
+
"eval_exact_match": 16.0,
|
1928 |
+
"eval_f1": 23.01206349206349,
|
1929 |
+
"step": 19573
|
1930 |
+
},
|
1931 |
+
{
|
1932 |
+
"epoch": 37.051039697542535,
|
1933 |
+
"grad_norm": 2.588029146194458,
|
1934 |
+
"learning_rate": 5e-05,
|
1935 |
+
"loss": 0.2885,
|
1936 |
+
"step": 19600
|
1937 |
+
},
|
1938 |
+
{
|
1939 |
+
"epoch": 37.24007561436673,
|
1940 |
+
"grad_norm": 2.2393579483032227,
|
1941 |
+
"learning_rate": 5e-05,
|
1942 |
+
"loss": 0.267,
|
1943 |
+
"step": 19700
|
1944 |
+
},
|
1945 |
+
{
|
1946 |
+
"epoch": 37.42911153119093,
|
1947 |
+
"grad_norm": 1.5349841117858887,
|
1948 |
+
"learning_rate": 5e-05,
|
1949 |
+
"loss": 0.2787,
|
1950 |
+
"step": 19800
|
1951 |
+
},
|
1952 |
+
{
|
1953 |
+
"epoch": 37.61814744801512,
|
1954 |
+
"grad_norm": 1.8498040437698364,
|
1955 |
+
"learning_rate": 5e-05,
|
1956 |
+
"loss": 0.2873,
|
1957 |
+
"step": 19900
|
1958 |
+
},
|
1959 |
+
{
|
1960 |
+
"epoch": 37.80718336483932,
|
1961 |
+
"grad_norm": 1.9959454536437988,
|
1962 |
+
"learning_rate": 5e-05,
|
1963 |
+
"loss": 0.2879,
|
1964 |
+
"step": 20000
|
1965 |
+
},
|
1966 |
+
{
|
1967 |
+
"epoch": 37.996219281663514,
|
1968 |
+
"grad_norm": 1.814253330230713,
|
1969 |
+
"learning_rate": 5e-05,
|
1970 |
+
"loss": 0.2927,
|
1971 |
+
"step": 20100
|
1972 |
+
},
|
1973 |
+
{
|
1974 |
+
"epoch": 38.0,
|
1975 |
+
"eval_accuracy": 0.5592820512820513,
|
1976 |
+
"eval_loss": 2.731750965118408,
|
1977 |
+
"eval_runtime": 5.032,
|
1978 |
+
"eval_samples_per_second": 99.364,
|
1979 |
+
"eval_steps_per_second": 12.52,
|
1980 |
+
"step": 20102
|
1981 |
+
},
|
1982 |
+
{
|
1983 |
+
"epoch": 38.0,
|
1984 |
+
"eval_exact_match": 16.4,
|
1985 |
+
"eval_f1": 23.767460317460323,
|
1986 |
+
"step": 20102
|
1987 |
+
},
|
1988 |
+
{
|
1989 |
+
"epoch": 38.18525519848771,
|
1990 |
+
"grad_norm": 1.635722041130066,
|
1991 |
+
"learning_rate": 5e-05,
|
1992 |
+
"loss": 0.2633,
|
1993 |
+
"step": 20200
|
1994 |
+
},
|
1995 |
+
{
|
1996 |
+
"epoch": 38.37429111531191,
|
1997 |
+
"grad_norm": 1.6726531982421875,
|
1998 |
+
"learning_rate": 5e-05,
|
1999 |
+
"loss": 0.2747,
|
2000 |
+
"step": 20300
|
2001 |
+
},
|
2002 |
+
{
|
2003 |
+
"epoch": 38.56332703213611,
|
2004 |
+
"grad_norm": 1.9620853662490845,
|
2005 |
+
"learning_rate": 5e-05,
|
2006 |
+
"loss": 0.2751,
|
2007 |
+
"step": 20400
|
2008 |
+
},
|
2009 |
+
{
|
2010 |
+
"epoch": 38.752362948960304,
|
2011 |
+
"grad_norm": 1.823510766029358,
|
2012 |
+
"learning_rate": 5e-05,
|
2013 |
+
"loss": 0.2868,
|
2014 |
+
"step": 20500
|
2015 |
+
},
|
2016 |
+
{
|
2017 |
+
"epoch": 38.9413988657845,
|
2018 |
+
"grad_norm": 2.833876132965088,
|
2019 |
+
"learning_rate": 5e-05,
|
2020 |
+
"loss": 0.2859,
|
2021 |
+
"step": 20600
|
2022 |
+
},
|
2023 |
+
{
|
2024 |
+
"epoch": 39.0,
|
2025 |
+
"eval_accuracy": 0.5603589743589743,
|
2026 |
+
"eval_loss": 2.7081727981567383,
|
2027 |
+
"eval_runtime": 5.0504,
|
2028 |
+
"eval_samples_per_second": 99.002,
|
2029 |
+
"eval_steps_per_second": 12.474,
|
2030 |
+
"step": 20631
|
2031 |
+
},
|
2032 |
+
{
|
2033 |
+
"epoch": 39.0,
|
2034 |
+
"eval_exact_match": 16.2,
|
2035 |
+
"eval_f1": 23.982424242424248,
|
2036 |
+
"step": 20631
|
2037 |
+
},
|
2038 |
+
{
|
2039 |
+
"epoch": 39.130434782608695,
|
2040 |
+
"grad_norm": 1.8951789140701294,
|
2041 |
+
"learning_rate": 5e-05,
|
2042 |
+
"loss": 0.2682,
|
2043 |
+
"step": 20700
|
2044 |
+
},
|
2045 |
+
{
|
2046 |
+
"epoch": 39.31947069943289,
|
2047 |
+
"grad_norm": 1.8282488584518433,
|
2048 |
+
"learning_rate": 5e-05,
|
2049 |
+
"loss": 0.2679,
|
2050 |
+
"step": 20800
|
2051 |
+
},
|
2052 |
+
{
|
2053 |
+
"epoch": 39.50850661625709,
|
2054 |
+
"grad_norm": 1.6153210401535034,
|
2055 |
+
"learning_rate": 5e-05,
|
2056 |
+
"loss": 0.2734,
|
2057 |
+
"step": 20900
|
2058 |
+
},
|
2059 |
+
{
|
2060 |
+
"epoch": 39.69754253308128,
|
2061 |
+
"grad_norm": 1.6458935737609863,
|
2062 |
+
"learning_rate": 5e-05,
|
2063 |
+
"loss": 0.278,
|
2064 |
+
"step": 21000
|
2065 |
+
},
|
2066 |
+
{
|
2067 |
+
"epoch": 39.886578449905485,
|
2068 |
+
"grad_norm": 1.6835973262786865,
|
2069 |
+
"learning_rate": 5e-05,
|
2070 |
+
"loss": 0.2826,
|
2071 |
+
"step": 21100
|
2072 |
+
},
|
2073 |
+
{
|
2074 |
+
"epoch": 40.0,
|
2075 |
+
"eval_accuracy": 0.5604102564102564,
|
2076 |
+
"eval_loss": 2.6441874504089355,
|
2077 |
+
"eval_runtime": 5.0452,
|
2078 |
+
"eval_samples_per_second": 99.104,
|
2079 |
+
"eval_steps_per_second": 12.487,
|
2080 |
+
"step": 21160
|
2081 |
+
},
|
2082 |
+
{
|
2083 |
+
"epoch": 40.0,
|
2084 |
+
"eval_exact_match": 16.0,
|
2085 |
+
"eval_f1": 23.70051948051949,
|
2086 |
+
"step": 21160
|
2087 |
+
},
|
2088 |
+
{
|
2089 |
+
"epoch": 40.07561436672968,
|
2090 |
+
"grad_norm": 1.868228554725647,
|
2091 |
+
"learning_rate": 5e-05,
|
2092 |
+
"loss": 0.2739,
|
2093 |
+
"step": 21200
|
2094 |
+
},
|
2095 |
+
{
|
2096 |
+
"epoch": 40.26465028355388,
|
2097 |
+
"grad_norm": 1.665651798248291,
|
2098 |
+
"learning_rate": 5e-05,
|
2099 |
+
"loss": 0.2619,
|
2100 |
+
"step": 21300
|
2101 |
+
},
|
2102 |
+
{
|
2103 |
+
"epoch": 40.45368620037807,
|
2104 |
+
"grad_norm": 1.808684229850769,
|
2105 |
+
"learning_rate": 5e-05,
|
2106 |
+
"loss": 0.2666,
|
2107 |
+
"step": 21400
|
2108 |
+
},
|
2109 |
+
{
|
2110 |
+
"epoch": 40.64272211720227,
|
2111 |
+
"grad_norm": 1.3911728858947754,
|
2112 |
+
"learning_rate": 5e-05,
|
2113 |
+
"loss": 0.2758,
|
2114 |
+
"step": 21500
|
2115 |
+
},
|
2116 |
+
{
|
2117 |
+
"epoch": 40.831758034026464,
|
2118 |
+
"grad_norm": 1.9357768297195435,
|
2119 |
+
"learning_rate": 5e-05,
|
2120 |
+
"loss": 0.2766,
|
2121 |
+
"step": 21600
|
2122 |
+
},
|
2123 |
+
{
|
2124 |
+
"epoch": 41.0,
|
2125 |
+
"eval_accuracy": 0.5605128205128205,
|
2126 |
+
"eval_loss": 2.668642520904541,
|
2127 |
+
"eval_runtime": 5.349,
|
2128 |
+
"eval_samples_per_second": 93.476,
|
2129 |
+
"eval_steps_per_second": 11.778,
|
2130 |
+
"step": 21689
|
2131 |
+
},
|
2132 |
+
{
|
2133 |
+
"epoch": 41.0,
|
2134 |
+
"eval_exact_match": 15.4,
|
2135 |
+
"eval_f1": 22.827027417027416,
|
2136 |
+
"step": 21689
|
2137 |
+
},
|
2138 |
+
{
|
2139 |
+
"epoch": 41.02079395085066,
|
2140 |
+
"grad_norm": 1.5084640979766846,
|
2141 |
+
"learning_rate": 5e-05,
|
2142 |
+
"loss": 0.276,
|
2143 |
+
"step": 21700
|
2144 |
+
},
|
2145 |
+
{
|
2146 |
+
"epoch": 41.209829867674856,
|
2147 |
+
"grad_norm": 2.262157440185547,
|
2148 |
+
"learning_rate": 5e-05,
|
2149 |
+
"loss": 0.2591,
|
2150 |
+
"step": 21800
|
2151 |
+
},
|
2152 |
+
{
|
2153 |
+
"epoch": 41.39886578449905,
|
2154 |
+
"grad_norm": 1.668727993965149,
|
2155 |
+
"learning_rate": 5e-05,
|
2156 |
+
"loss": 0.2637,
|
2157 |
+
"step": 21900
|
2158 |
+
},
|
2159 |
+
{
|
2160 |
+
"epoch": 41.587901701323254,
|
2161 |
+
"grad_norm": 2.1757819652557373,
|
2162 |
+
"learning_rate": 5e-05,
|
2163 |
+
"loss": 0.2681,
|
2164 |
+
"step": 22000
|
2165 |
+
},
|
2166 |
+
{
|
2167 |
+
"epoch": 41.77693761814745,
|
2168 |
+
"grad_norm": 1.980452060699463,
|
2169 |
+
"learning_rate": 5e-05,
|
2170 |
+
"loss": 0.275,
|
2171 |
+
"step": 22100
|
2172 |
+
},
|
2173 |
+
{
|
2174 |
+
"epoch": 41.965973534971646,
|
2175 |
+
"grad_norm": 1.2051122188568115,
|
2176 |
+
"learning_rate": 5e-05,
|
2177 |
+
"loss": 0.2755,
|
2178 |
+
"step": 22200
|
2179 |
+
},
|
2180 |
+
{
|
2181 |
+
"epoch": 42.0,
|
2182 |
+
"eval_accuracy": 0.56,
|
2183 |
+
"eval_loss": 2.7107934951782227,
|
2184 |
+
"eval_runtime": 5.1459,
|
2185 |
+
"eval_samples_per_second": 97.165,
|
2186 |
+
"eval_steps_per_second": 12.243,
|
2187 |
+
"step": 22218
|
2188 |
+
},
|
2189 |
+
{
|
2190 |
+
"epoch": 42.0,
|
2191 |
+
"eval_exact_match": 16.8,
|
2192 |
+
"eval_f1": 23.684805194805197,
|
2193 |
+
"step": 22218
|
2194 |
+
},
|
2195 |
+
{
|
2196 |
+
"epoch": 42.15500945179584,
|
2197 |
+
"grad_norm": 1.3517260551452637,
|
2198 |
+
"learning_rate": 5e-05,
|
2199 |
+
"loss": 0.2595,
|
2200 |
+
"step": 22300
|
2201 |
+
},
|
2202 |
+
{
|
2203 |
+
"epoch": 42.34404536862004,
|
2204 |
+
"grad_norm": 1.849206805229187,
|
2205 |
+
"learning_rate": 5e-05,
|
2206 |
+
"loss": 0.2538,
|
2207 |
+
"step": 22400
|
2208 |
+
},
|
2209 |
+
{
|
2210 |
+
"epoch": 42.53308128544423,
|
2211 |
+
"grad_norm": 1.4720873832702637,
|
2212 |
+
"learning_rate": 5e-05,
|
2213 |
+
"loss": 0.2672,
|
2214 |
+
"step": 22500
|
2215 |
+
},
|
2216 |
+
{
|
2217 |
+
"epoch": 42.72211720226843,
|
2218 |
+
"grad_norm": 1.668751835823059,
|
2219 |
+
"learning_rate": 5e-05,
|
2220 |
+
"loss": 0.2707,
|
2221 |
+
"step": 22600
|
2222 |
+
},
|
2223 |
+
{
|
2224 |
+
"epoch": 42.911153119092624,
|
2225 |
+
"grad_norm": 2.0918915271759033,
|
2226 |
+
"learning_rate": 5e-05,
|
2227 |
+
"loss": 0.2751,
|
2228 |
+
"step": 22700
|
2229 |
+
},
|
2230 |
+
{
|
2231 |
+
"epoch": 43.0,
|
2232 |
+
"eval_accuracy": 0.5601538461538461,
|
2233 |
+
"eval_loss": 2.7315948009490967,
|
2234 |
+
"eval_runtime": 5.0195,
|
2235 |
+
"eval_samples_per_second": 99.612,
|
2236 |
+
"eval_steps_per_second": 12.551,
|
2237 |
+
"step": 22747
|
2238 |
+
},
|
2239 |
+
{
|
2240 |
+
"epoch": 43.0,
|
2241 |
+
"eval_exact_match": 18.2,
|
2242 |
+
"eval_f1": 24.878571428571433,
|
2243 |
+
"step": 22747
|
2244 |
+
},
|
2245 |
+
{
|
2246 |
+
"epoch": 43.10018903591683,
|
2247 |
+
"grad_norm": 1.3155404329299927,
|
2248 |
+
"learning_rate": 5e-05,
|
2249 |
+
"loss": 0.2601,
|
2250 |
+
"step": 22800
|
2251 |
+
},
|
2252 |
+
{
|
2253 |
+
"epoch": 43.28922495274102,
|
2254 |
+
"grad_norm": 1.6259891986846924,
|
2255 |
+
"learning_rate": 5e-05,
|
2256 |
+
"loss": 0.2545,
|
2257 |
+
"step": 22900
|
2258 |
+
},
|
2259 |
+
{
|
2260 |
+
"epoch": 43.47826086956522,
|
2261 |
+
"grad_norm": 2.0335798263549805,
|
2262 |
+
"learning_rate": 5e-05,
|
2263 |
+
"loss": 0.2599,
|
2264 |
+
"step": 23000
|
2265 |
+
},
|
2266 |
+
{
|
2267 |
+
"epoch": 43.667296786389414,
|
2268 |
+
"grad_norm": 1.4336028099060059,
|
2269 |
+
"learning_rate": 5e-05,
|
2270 |
+
"loss": 0.2641,
|
2271 |
+
"step": 23100
|
2272 |
+
},
|
2273 |
+
{
|
2274 |
+
"epoch": 43.85633270321361,
|
2275 |
+
"grad_norm": 1.5379899740219116,
|
2276 |
+
"learning_rate": 5e-05,
|
2277 |
+
"loss": 0.2725,
|
2278 |
+
"step": 23200
|
2279 |
+
},
|
2280 |
+
{
|
2281 |
+
"epoch": 44.0,
|
2282 |
+
"eval_accuracy": 0.5597948717948718,
|
2283 |
+
"eval_loss": 2.7413489818573,
|
2284 |
+
"eval_runtime": 5.1078,
|
2285 |
+
"eval_samples_per_second": 97.889,
|
2286 |
+
"eval_steps_per_second": 12.334,
|
2287 |
+
"step": 23276
|
2288 |
+
},
|
2289 |
+
{
|
2290 |
+
"epoch": 44.0,
|
2291 |
+
"eval_exact_match": 16.4,
|
2292 |
+
"eval_f1": 22.825714285714295,
|
2293 |
+
"step": 23276
|
2294 |
+
},
|
2295 |
+
{
|
2296 |
+
"epoch": 44.045368620037806,
|
2297 |
+
"grad_norm": 1.7726408243179321,
|
2298 |
+
"learning_rate": 5e-05,
|
2299 |
+
"loss": 0.2678,
|
2300 |
+
"step": 23300
|
2301 |
+
},
|
2302 |
+
{
|
2303 |
+
"epoch": 44.234404536862,
|
2304 |
+
"grad_norm": 1.8952834606170654,
|
2305 |
+
"learning_rate": 5e-05,
|
2306 |
+
"loss": 0.2497,
|
2307 |
+
"step": 23400
|
2308 |
+
},
|
2309 |
+
{
|
2310 |
+
"epoch": 44.4234404536862,
|
2311 |
+
"grad_norm": 1.2246063947677612,
|
2312 |
+
"learning_rate": 5e-05,
|
2313 |
+
"loss": 0.2522,
|
2314 |
+
"step": 23500
|
2315 |
+
},
|
2316 |
+
{
|
2317 |
+
"epoch": 44.6124763705104,
|
2318 |
+
"grad_norm": 1.498618721961975,
|
2319 |
+
"learning_rate": 5e-05,
|
2320 |
+
"loss": 0.2668,
|
2321 |
+
"step": 23600
|
2322 |
+
},
|
2323 |
+
{
|
2324 |
+
"epoch": 44.801512287334596,
|
2325 |
+
"grad_norm": 1.5657014846801758,
|
2326 |
+
"learning_rate": 5e-05,
|
2327 |
+
"loss": 0.2697,
|
2328 |
+
"step": 23700
|
2329 |
+
},
|
2330 |
+
{
|
2331 |
+
"epoch": 44.99054820415879,
|
2332 |
+
"grad_norm": 1.5949833393096924,
|
2333 |
+
"learning_rate": 5e-05,
|
2334 |
+
"loss": 0.2711,
|
2335 |
+
"step": 23800
|
2336 |
+
},
|
2337 |
+
{
|
2338 |
+
"epoch": 45.0,
|
2339 |
+
"eval_accuracy": 0.5592307692307692,
|
2340 |
+
"eval_loss": 2.7097885608673096,
|
2341 |
+
"eval_runtime": 5.0937,
|
2342 |
+
"eval_samples_per_second": 98.161,
|
2343 |
+
"eval_steps_per_second": 12.368,
|
2344 |
+
"step": 23805
|
2345 |
+
},
|
2346 |
+
{
|
2347 |
+
"epoch": 45.0,
|
2348 |
+
"eval_exact_match": 16.2,
|
2349 |
+
"eval_f1": 23.22766233766234,
|
2350 |
+
"step": 23805
|
2351 |
+
},
|
2352 |
+
{
|
2353 |
+
"epoch": 45.17958412098299,
|
2354 |
+
"grad_norm": 3.5289559364318848,
|
2355 |
+
"learning_rate": 5e-05,
|
2356 |
+
"loss": 0.2513,
|
2357 |
+
"step": 23900
|
2358 |
+
},
|
2359 |
+
{
|
2360 |
+
"epoch": 45.36862003780718,
|
2361 |
+
"grad_norm": 1.690945029258728,
|
2362 |
+
"learning_rate": 5e-05,
|
2363 |
+
"loss": 0.2628,
|
2364 |
+
"step": 24000
|
2365 |
+
},
|
2366 |
+
{
|
2367 |
+
"epoch": 45.55765595463138,
|
2368 |
+
"grad_norm": 2.572908878326416,
|
2369 |
+
"learning_rate": 5e-05,
|
2370 |
+
"loss": 0.2546,
|
2371 |
+
"step": 24100
|
2372 |
+
},
|
2373 |
+
{
|
2374 |
+
"epoch": 45.746691871455575,
|
2375 |
+
"grad_norm": 2.0459492206573486,
|
2376 |
+
"learning_rate": 5e-05,
|
2377 |
+
"loss": 0.2608,
|
2378 |
+
"step": 24200
|
2379 |
+
},
|
2380 |
+
{
|
2381 |
+
"epoch": 45.93572778827977,
|
2382 |
+
"grad_norm": 1.8246817588806152,
|
2383 |
+
"learning_rate": 5e-05,
|
2384 |
+
"loss": 0.2662,
|
2385 |
+
"step": 24300
|
2386 |
+
},
|
2387 |
+
{
|
2388 |
+
"epoch": 46.0,
|
2389 |
+
"eval_accuracy": 0.5595384615384615,
|
2390 |
+
"eval_loss": 2.7349929809570312,
|
2391 |
+
"eval_runtime": 5.9124,
|
2392 |
+
"eval_samples_per_second": 84.569,
|
2393 |
+
"eval_steps_per_second": 10.656,
|
2394 |
+
"step": 24334
|
2395 |
+
},
|
2396 |
+
{
|
2397 |
+
"epoch": 46.0,
|
2398 |
+
"eval_exact_match": 16.6,
|
2399 |
+
"eval_f1": 24.38528138528139,
|
2400 |
+
"step": 24334
|
2401 |
+
},
|
2402 |
+
{
|
2403 |
+
"epoch": 46.12476370510397,
|
2404 |
+
"grad_norm": 1.241733431816101,
|
2405 |
+
"learning_rate": 5e-05,
|
2406 |
+
"loss": 0.2548,
|
2407 |
+
"step": 24400
|
2408 |
+
},
|
2409 |
+
{
|
2410 |
+
"epoch": 46.31379962192817,
|
2411 |
+
"grad_norm": 1.538244605064392,
|
2412 |
+
"learning_rate": 5e-05,
|
2413 |
+
"loss": 0.2485,
|
2414 |
+
"step": 24500
|
2415 |
+
},
|
2416 |
+
{
|
2417 |
+
"epoch": 46.502835538752365,
|
2418 |
+
"grad_norm": 1.4510152339935303,
|
2419 |
+
"learning_rate": 5e-05,
|
2420 |
+
"loss": 0.2548,
|
2421 |
+
"step": 24600
|
2422 |
+
},
|
2423 |
+
{
|
2424 |
+
"epoch": 46.69187145557656,
|
2425 |
+
"grad_norm": 1.2538402080535889,
|
2426 |
+
"learning_rate": 5e-05,
|
2427 |
+
"loss": 0.262,
|
2428 |
+
"step": 24700
|
2429 |
+
},
|
2430 |
+
{
|
2431 |
+
"epoch": 46.880907372400756,
|
2432 |
+
"grad_norm": 1.49154794216156,
|
2433 |
+
"learning_rate": 5e-05,
|
2434 |
+
"loss": 0.2636,
|
2435 |
+
"step": 24800
|
2436 |
+
},
|
2437 |
+
{
|
2438 |
+
"epoch": 47.0,
|
2439 |
+
"eval_accuracy": 0.56,
|
2440 |
+
"eval_loss": 2.7406322956085205,
|
2441 |
+
"eval_runtime": 5.8521,
|
2442 |
+
"eval_samples_per_second": 85.439,
|
2443 |
+
"eval_steps_per_second": 10.765,
|
2444 |
+
"step": 24863
|
2445 |
+
},
|
2446 |
+
{
|
2447 |
+
"epoch": 47.0,
|
2448 |
+
"eval_exact_match": 16.0,
|
2449 |
+
"eval_f1": 23.067503607503607,
|
2450 |
+
"step": 24863
|
2451 |
+
},
|
2452 |
+
{
|
2453 |
+
"epoch": 47.06994328922495,
|
2454 |
+
"grad_norm": 1.242047667503357,
|
2455 |
+
"learning_rate": 5e-05,
|
2456 |
+
"loss": 0.2582,
|
2457 |
+
"step": 24900
|
2458 |
+
},
|
2459 |
+
{
|
2460 |
+
"epoch": 47.25897920604915,
|
2461 |
+
"grad_norm": 1.1271893978118896,
|
2462 |
+
"learning_rate": 5e-05,
|
2463 |
+
"loss": 0.2502,
|
2464 |
+
"step": 25000
|
2465 |
+
},
|
2466 |
+
{
|
2467 |
+
"epoch": 47.44801512287334,
|
2468 |
+
"grad_norm": 1.5100948810577393,
|
2469 |
+
"learning_rate": 5e-05,
|
2470 |
+
"loss": 0.2499,
|
2471 |
+
"step": 25100
|
2472 |
+
},
|
2473 |
+
{
|
2474 |
+
"epoch": 47.63705103969754,
|
2475 |
+
"grad_norm": 1.4169464111328125,
|
2476 |
+
"learning_rate": 5e-05,
|
2477 |
+
"loss": 0.2538,
|
2478 |
+
"step": 25200
|
2479 |
+
},
|
2480 |
+
{
|
2481 |
+
"epoch": 47.82608695652174,
|
2482 |
+
"grad_norm": 1.4983057975769043,
|
2483 |
+
"learning_rate": 5e-05,
|
2484 |
+
"loss": 0.2628,
|
2485 |
+
"step": 25300
|
2486 |
+
},
|
2487 |
+
{
|
2488 |
+
"epoch": 48.0,
|
2489 |
+
"eval_accuracy": 0.5598974358974359,
|
2490 |
+
"eval_loss": 2.764431953430176,
|
2491 |
+
"eval_runtime": 5.1934,
|
2492 |
+
"eval_samples_per_second": 96.276,
|
2493 |
+
"eval_steps_per_second": 12.131,
|
2494 |
+
"step": 25392
|
2495 |
+
},
|
2496 |
+
{
|
2497 |
+
"epoch": 48.0,
|
2498 |
+
"eval_exact_match": 16.4,
|
2499 |
+
"eval_f1": 23.456031746031748,
|
2500 |
+
"step": 25392
|
2501 |
+
}
|
2502 |
+
],
|
2503 |
+
"logging_steps": 100,
|
2504 |
+
"max_steps": 26450,
|
2505 |
+
"num_input_tokens_seen": 0,
|
2506 |
+
"num_train_epochs": 50,
|
2507 |
+
"save_steps": 500,
|
2508 |
+
"stateful_callbacks": {
|
2509 |
+
"TrainerControl": {
|
2510 |
+
"args": {
|
2511 |
+
"should_epoch_stop": false,
|
2512 |
+
"should_evaluate": false,
|
2513 |
+
"should_log": false,
|
2514 |
+
"should_save": true,
|
2515 |
+
"should_training_stop": false
|
2516 |
+
},
|
2517 |
+
"attributes": {}
|
2518 |
+
}
|
2519 |
+
},
|
2520 |
+
"total_flos": 1.5642442537338143e+18,
|
2521 |
+
"train_batch_size": 1,
|
2522 |
+
"trial_name": null,
|
2523 |
+
"trial_params": null
|
2524 |
+
}
|
checkpoint-25392/training_args.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:f01d18f717251d8791982fc93ed5352f2d247add98e1102e2a52cc0a32128bb7
|
3 |
+
size 5304
|
checkpoint-25392/vocab.json
ADDED
The diff for this file is too large to render.
See raw diff
|
|