Training in progress, epoch 45, checkpoint
Browse files- checkpoint-23805/README.md +9 -0
- checkpoint-23805/adapter_config.json +25 -0
- checkpoint-23805/adapter_model.safetensors +3 -0
- checkpoint-23805/added_tokens.json +5 -0
- checkpoint-23805/merges.txt +0 -0
- checkpoint-23805/optimizer.pt +3 -0
- checkpoint-23805/rng_state_0.pth +3 -0
- checkpoint-23805/rng_state_1.pth +3 -0
- checkpoint-23805/rng_state_2.pth +3 -0
- checkpoint-23805/rng_state_3.pth +3 -0
- checkpoint-23805/scheduler.pt +3 -0
- checkpoint-23805/special_tokens_map.json +14 -0
- checkpoint-23805/tokenizer.json +0 -0
- checkpoint-23805/tokenizer_config.json +43 -0
- checkpoint-23805/trainer_state.json +2374 -0
- checkpoint-23805/training_args.bin +3 -0
- checkpoint-23805/vocab.json +0 -0
checkpoint-23805/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-23805/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-23805/adapter_model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:4a66895c43a60a7f6ea94723b5284ae989d5065ccf1b4f32134868542f1487d3
|
| 3 |
+
size 112130416
|
checkpoint-23805/added_tokens.json
ADDED
|
@@ -0,0 +1,5 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"<|endoftext|>": 151643,
|
| 3 |
+
"<|im_end|>": 151645,
|
| 4 |
+
"<|im_start|>": 151644
|
| 5 |
+
}
|
checkpoint-23805/merges.txt
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
checkpoint-23805/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:21fabf5352762013a97f5e5f14147d91fc31acdf611e37eb3130ba553a4d4080
|
| 3 |
+
size 224537202
|
checkpoint-23805/rng_state_0.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:c5b5857125e1007e5e73b9e7ccf0a664f092d7a15042ed0afa81e80826bfc5a2
|
| 3 |
+
size 15024
|
checkpoint-23805/rng_state_1.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:9942976a73efbaea6855545ceee16e6d5d70eaec0414831813d950adeb30b89b
|
| 3 |
+
size 15024
|
checkpoint-23805/rng_state_2.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:2b3f505e34138e5dff6367c8e8dab6ad535882b27bb8aefa3cbfd4919266c42f
|
| 3 |
+
size 15024
|
checkpoint-23805/rng_state_3.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:70b6b9a0f0b84c824f94e2f635654514c443da00ddefd1f7b90c6f0b9e4e1361
|
| 3 |
+
size 15024
|
checkpoint-23805/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:df8c0a8f399c804404ba8bfd0eba875765ac3aeb74840bee980f74a6d1302703
|
| 3 |
+
size 1064
|
checkpoint-23805/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-23805/tokenizer.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
checkpoint-23805/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-23805/trainer_state.json
ADDED
|
@@ -0,0 +1,2374 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 45.0,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 23805,
|
| 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": 0.3393945097923279,
|
| 14 |
+
"learning_rate": 0.0003,
|
| 15 |
+
"loss": 1.951,
|
| 16 |
+
"step": 100
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.3780718336483932,
|
| 20 |
+
"grad_norm": 8017.20361328125,
|
| 21 |
+
"learning_rate": 0.0003,
|
| 22 |
+
"loss": 2.0285,
|
| 23 |
+
"step": 200
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.5671077504725898,
|
| 27 |
+
"grad_norm": 9146.7236328125,
|
| 28 |
+
"learning_rate": 0.0003,
|
| 29 |
+
"loss": 6.5404,
|
| 30 |
+
"step": 300
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.7561436672967864,
|
| 34 |
+
"grad_norm": 574.934326171875,
|
| 35 |
+
"learning_rate": 0.0003,
|
| 36 |
+
"loss": 7.7357,
|
| 37 |
+
"step": 400
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.945179584120983,
|
| 41 |
+
"grad_norm": 25128.55859375,
|
| 42 |
+
"learning_rate": 0.0003,
|
| 43 |
+
"loss": 6.3771,
|
| 44 |
+
"step": 500
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 1.0,
|
| 48 |
+
"eval_accuracy": 0.24974358974358973,
|
| 49 |
+
"eval_loss": 10.19681453704834,
|
| 50 |
+
"eval_runtime": 5.1408,
|
| 51 |
+
"eval_samples_per_second": 97.26,
|
| 52 |
+
"eval_steps_per_second": 12.255,
|
| 53 |
+
"step": 529
|
| 54 |
+
},
|
| 55 |
+
{
|
| 56 |
+
"epoch": 1.0,
|
| 57 |
+
"eval_exact_match": 0.0,
|
| 58 |
+
"eval_f1": 0.0,
|
| 59 |
+
"step": 529
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 1.1342155009451795,
|
| 63 |
+
"grad_norm": 1332.8885498046875,
|
| 64 |
+
"learning_rate": 0.0003,
|
| 65 |
+
"loss": 11.0858,
|
| 66 |
+
"step": 600
|
| 67 |
+
},
|
| 68 |
+
{
|
| 69 |
+
"epoch": 1.3232514177693762,
|
| 70 |
+
"grad_norm": 3666.881591796875,
|
| 71 |
+
"learning_rate": 0.0003,
|
| 72 |
+
"loss": 9.483,
|
| 73 |
+
"step": 700
|
| 74 |
+
},
|
| 75 |
+
{
|
| 76 |
+
"epoch": 1.5122873345935728,
|
| 77 |
+
"grad_norm": 356847.0,
|
| 78 |
+
"learning_rate": 0.0003,
|
| 79 |
+
"loss": 8.069,
|
| 80 |
+
"step": 800
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 1.7013232514177694,
|
| 84 |
+
"grad_norm": 1061.877685546875,
|
| 85 |
+
"learning_rate": 0.0003,
|
| 86 |
+
"loss": 8.7493,
|
| 87 |
+
"step": 900
|
| 88 |
+
},
|
| 89 |
+
{
|
| 90 |
+
"epoch": 1.8903591682419658,
|
| 91 |
+
"grad_norm": 46670.3046875,
|
| 92 |
+
"learning_rate": 0.0003,
|
| 93 |
+
"loss": 8.3415,
|
| 94 |
+
"step": 1000
|
| 95 |
+
},
|
| 96 |
+
{
|
| 97 |
+
"epoch": 2.0,
|
| 98 |
+
"eval_accuracy": 0.20507692307692307,
|
| 99 |
+
"eval_loss": 7.504091262817383,
|
| 100 |
+
"eval_runtime": 5.2437,
|
| 101 |
+
"eval_samples_per_second": 95.352,
|
| 102 |
+
"eval_steps_per_second": 12.014,
|
| 103 |
+
"step": 1058
|
| 104 |
+
},
|
| 105 |
+
{
|
| 106 |
+
"epoch": 2.0,
|
| 107 |
+
"eval_exact_match": 0.0,
|
| 108 |
+
"eval_f1": 0.0,
|
| 109 |
+
"step": 1058
|
| 110 |
+
},
|
| 111 |
+
{
|
| 112 |
+
"epoch": 2.0793950850661624,
|
| 113 |
+
"grad_norm": 2.0727381706237793,
|
| 114 |
+
"learning_rate": 0.0003,
|
| 115 |
+
"loss": 8.3511,
|
| 116 |
+
"step": 1100
|
| 117 |
+
},
|
| 118 |
+
{
|
| 119 |
+
"epoch": 2.268431001890359,
|
| 120 |
+
"grad_norm": 1.8185893297195435,
|
| 121 |
+
"learning_rate": 0.0003,
|
| 122 |
+
"loss": 7.3065,
|
| 123 |
+
"step": 1200
|
| 124 |
+
},
|
| 125 |
+
{
|
| 126 |
+
"epoch": 2.4574669187145557,
|
| 127 |
+
"grad_norm": 1620.93701171875,
|
| 128 |
+
"learning_rate": 0.0003,
|
| 129 |
+
"loss": 7.1302,
|
| 130 |
+
"step": 1300
|
| 131 |
+
},
|
| 132 |
+
{
|
| 133 |
+
"epoch": 2.6465028355387523,
|
| 134 |
+
"grad_norm": 1.0691380500793457,
|
| 135 |
+
"learning_rate": 0.0003,
|
| 136 |
+
"loss": 7.4423,
|
| 137 |
+
"step": 1400
|
| 138 |
+
},
|
| 139 |
+
{
|
| 140 |
+
"epoch": 2.835538752362949,
|
| 141 |
+
"grad_norm": 1.0605100393295288,
|
| 142 |
+
"learning_rate": 0.0003,
|
| 143 |
+
"loss": 7.6715,
|
| 144 |
+
"step": 1500
|
| 145 |
+
},
|
| 146 |
+
{
|
| 147 |
+
"epoch": 3.0,
|
| 148 |
+
"eval_accuracy": 0.28507692307692306,
|
| 149 |
+
"eval_loss": 7.084368705749512,
|
| 150 |
+
"eval_runtime": 5.493,
|
| 151 |
+
"eval_samples_per_second": 91.026,
|
| 152 |
+
"eval_steps_per_second": 11.469,
|
| 153 |
+
"step": 1587
|
| 154 |
+
},
|
| 155 |
+
{
|
| 156 |
+
"epoch": 3.0,
|
| 157 |
+
"eval_exact_match": 0.0,
|
| 158 |
+
"eval_f1": 0.08,
|
| 159 |
+
"step": 1587
|
| 160 |
+
},
|
| 161 |
+
{
|
| 162 |
+
"epoch": 3.0245746691871456,
|
| 163 |
+
"grad_norm": 1967676800.0,
|
| 164 |
+
"learning_rate": 0.0003,
|
| 165 |
+
"loss": 7.8617,
|
| 166 |
+
"step": 1600
|
| 167 |
+
},
|
| 168 |
+
{
|
| 169 |
+
"epoch": 3.213610586011342,
|
| 170 |
+
"grad_norm": 83918778793984.0,
|
| 171 |
+
"learning_rate": 0.0003,
|
| 172 |
+
"loss": 8.083,
|
| 173 |
+
"step": 1700
|
| 174 |
+
},
|
| 175 |
+
{
|
| 176 |
+
"epoch": 3.402646502835539,
|
| 177 |
+
"grad_norm": 12349449.0,
|
| 178 |
+
"learning_rate": 0.0003,
|
| 179 |
+
"loss": 8.0382,
|
| 180 |
+
"step": 1800
|
| 181 |
+
},
|
| 182 |
+
{
|
| 183 |
+
"epoch": 3.5916824196597354,
|
| 184 |
+
"grad_norm": 52939337728.0,
|
| 185 |
+
"learning_rate": 0.0003,
|
| 186 |
+
"loss": 7.8761,
|
| 187 |
+
"step": 1900
|
| 188 |
+
},
|
| 189 |
+
{
|
| 190 |
+
"epoch": 3.780718336483932,
|
| 191 |
+
"grad_norm": 1381182.25,
|
| 192 |
+
"learning_rate": 0.0003,
|
| 193 |
+
"loss": 7.4665,
|
| 194 |
+
"step": 2000
|
| 195 |
+
},
|
| 196 |
+
{
|
| 197 |
+
"epoch": 3.9697542533081287,
|
| 198 |
+
"grad_norm": 105.80550384521484,
|
| 199 |
+
"learning_rate": 0.0003,
|
| 200 |
+
"loss": 7.8182,
|
| 201 |
+
"step": 2100
|
| 202 |
+
},
|
| 203 |
+
{
|
| 204 |
+
"epoch": 4.0,
|
| 205 |
+
"eval_accuracy": 0.23933333333333334,
|
| 206 |
+
"eval_loss": 7.05845308303833,
|
| 207 |
+
"eval_runtime": 5.0891,
|
| 208 |
+
"eval_samples_per_second": 98.248,
|
| 209 |
+
"eval_steps_per_second": 12.379,
|
| 210 |
+
"step": 2116
|
| 211 |
+
},
|
| 212 |
+
{
|
| 213 |
+
"epoch": 4.0,
|
| 214 |
+
"eval_exact_match": 0.0,
|
| 215 |
+
"eval_f1": 0.0,
|
| 216 |
+
"step": 2116
|
| 217 |
+
},
|
| 218 |
+
{
|
| 219 |
+
"epoch": 4.158790170132325,
|
| 220 |
+
"grad_norm": 16.110862731933594,
|
| 221 |
+
"learning_rate": 0.0003,
|
| 222 |
+
"loss": 7.9728,
|
| 223 |
+
"step": 2200
|
| 224 |
+
},
|
| 225 |
+
{
|
| 226 |
+
"epoch": 4.3478260869565215,
|
| 227 |
+
"grad_norm": 4.362427234649658,
|
| 228 |
+
"learning_rate": 0.0003,
|
| 229 |
+
"loss": 7.2875,
|
| 230 |
+
"step": 2300
|
| 231 |
+
},
|
| 232 |
+
{
|
| 233 |
+
"epoch": 4.536862003780718,
|
| 234 |
+
"grad_norm": 18039.095703125,
|
| 235 |
+
"learning_rate": 0.0003,
|
| 236 |
+
"loss": 7.3972,
|
| 237 |
+
"step": 2400
|
| 238 |
+
},
|
| 239 |
+
{
|
| 240 |
+
"epoch": 4.725897920604915,
|
| 241 |
+
"grad_norm": 2.3812594051160474e+17,
|
| 242 |
+
"learning_rate": 0.0003,
|
| 243 |
+
"loss": 7.4631,
|
| 244 |
+
"step": 2500
|
| 245 |
+
},
|
| 246 |
+
{
|
| 247 |
+
"epoch": 4.914933837429111,
|
| 248 |
+
"grad_norm": 3.9549436569213867,
|
| 249 |
+
"learning_rate": 0.0003,
|
| 250 |
+
"loss": 7.3322,
|
| 251 |
+
"step": 2600
|
| 252 |
+
},
|
| 253 |
+
{
|
| 254 |
+
"epoch": 5.0,
|
| 255 |
+
"eval_accuracy": 0.22430769230769232,
|
| 256 |
+
"eval_loss": 6.72421932220459,
|
| 257 |
+
"eval_runtime": 5.1689,
|
| 258 |
+
"eval_samples_per_second": 96.731,
|
| 259 |
+
"eval_steps_per_second": 12.188,
|
| 260 |
+
"step": 2645
|
| 261 |
+
},
|
| 262 |
+
{
|
| 263 |
+
"epoch": 5.0,
|
| 264 |
+
"eval_exact_match": 0.0,
|
| 265 |
+
"eval_f1": 0.0,
|
| 266 |
+
"step": 2645
|
| 267 |
+
},
|
| 268 |
+
{
|
| 269 |
+
"epoch": 5.103969754253308,
|
| 270 |
+
"grad_norm": 120114.2421875,
|
| 271 |
+
"learning_rate": 0.0003,
|
| 272 |
+
"loss": 7.2428,
|
| 273 |
+
"step": 2700
|
| 274 |
+
},
|
| 275 |
+
{
|
| 276 |
+
"epoch": 5.293005671077505,
|
| 277 |
+
"grad_norm": 391.66119384765625,
|
| 278 |
+
"learning_rate": 0.0003,
|
| 279 |
+
"loss": 7.2495,
|
| 280 |
+
"step": 2800
|
| 281 |
+
},
|
| 282 |
+
{
|
| 283 |
+
"epoch": 5.482041587901701,
|
| 284 |
+
"grad_norm": 599371136.0,
|
| 285 |
+
"learning_rate": 0.0003,
|
| 286 |
+
"loss": 7.5915,
|
| 287 |
+
"step": 2900
|
| 288 |
+
},
|
| 289 |
+
{
|
| 290 |
+
"epoch": 5.671077504725898,
|
| 291 |
+
"grad_norm": 5585.97314453125,
|
| 292 |
+
"learning_rate": 0.0003,
|
| 293 |
+
"loss": 7.3014,
|
| 294 |
+
"step": 3000
|
| 295 |
+
},
|
| 296 |
+
{
|
| 297 |
+
"epoch": 5.8601134215500945,
|
| 298 |
+
"grad_norm": 268239548448768.0,
|
| 299 |
+
"learning_rate": 0.0003,
|
| 300 |
+
"loss": 7.259,
|
| 301 |
+
"step": 3100
|
| 302 |
+
},
|
| 303 |
+
{
|
| 304 |
+
"epoch": 6.0,
|
| 305 |
+
"eval_accuracy": 0.24984615384615386,
|
| 306 |
+
"eval_loss": 6.531904220581055,
|
| 307 |
+
"eval_runtime": 5.0344,
|
| 308 |
+
"eval_samples_per_second": 99.317,
|
| 309 |
+
"eval_steps_per_second": 12.514,
|
| 310 |
+
"step": 3174
|
| 311 |
+
},
|
| 312 |
+
{
|
| 313 |
+
"epoch": 6.0,
|
| 314 |
+
"eval_exact_match": 0.0,
|
| 315 |
+
"eval_f1": 0.0,
|
| 316 |
+
"step": 3174
|
| 317 |
+
},
|
| 318 |
+
{
|
| 319 |
+
"epoch": 6.049149338374291,
|
| 320 |
+
"grad_norm": 21535.734375,
|
| 321 |
+
"learning_rate": 0.0003,
|
| 322 |
+
"loss": 7.5829,
|
| 323 |
+
"step": 3200
|
| 324 |
+
},
|
| 325 |
+
{
|
| 326 |
+
"epoch": 6.238185255198488,
|
| 327 |
+
"grad_norm": 40871530496.0,
|
| 328 |
+
"learning_rate": 0.0003,
|
| 329 |
+
"loss": 8.6364,
|
| 330 |
+
"step": 3300
|
| 331 |
+
},
|
| 332 |
+
{
|
| 333 |
+
"epoch": 6.427221172022684,
|
| 334 |
+
"grad_norm": 13.242510795593262,
|
| 335 |
+
"learning_rate": 0.0003,
|
| 336 |
+
"loss": 7.6599,
|
| 337 |
+
"step": 3400
|
| 338 |
+
},
|
| 339 |
+
{
|
| 340 |
+
"epoch": 6.616257088846881,
|
| 341 |
+
"grad_norm": 3.0841150283813477,
|
| 342 |
+
"learning_rate": 0.0003,
|
| 343 |
+
"loss": 7.5796,
|
| 344 |
+
"step": 3500
|
| 345 |
+
},
|
| 346 |
+
{
|
| 347 |
+
"epoch": 6.805293005671078,
|
| 348 |
+
"grad_norm": 1.6124571561813354,
|
| 349 |
+
"learning_rate": 0.0003,
|
| 350 |
+
"loss": 7.3499,
|
| 351 |
+
"step": 3600
|
| 352 |
+
},
|
| 353 |
+
{
|
| 354 |
+
"epoch": 6.994328922495274,
|
| 355 |
+
"grad_norm": 1.8358014822006226,
|
| 356 |
+
"learning_rate": 0.0003,
|
| 357 |
+
"loss": 7.2206,
|
| 358 |
+
"step": 3700
|
| 359 |
+
},
|
| 360 |
+
{
|
| 361 |
+
"epoch": 7.0,
|
| 362 |
+
"eval_accuracy": 0.22430769230769232,
|
| 363 |
+
"eval_loss": 6.724715709686279,
|
| 364 |
+
"eval_runtime": 5.0476,
|
| 365 |
+
"eval_samples_per_second": 99.056,
|
| 366 |
+
"eval_steps_per_second": 12.481,
|
| 367 |
+
"step": 3703
|
| 368 |
+
},
|
| 369 |
+
{
|
| 370 |
+
"epoch": 7.0,
|
| 371 |
+
"eval_exact_match": 0.0,
|
| 372 |
+
"eval_f1": 0.0,
|
| 373 |
+
"step": 3703
|
| 374 |
+
},
|
| 375 |
+
{
|
| 376 |
+
"epoch": 7.183364839319471,
|
| 377 |
+
"grad_norm": 5.786651134490967,
|
| 378 |
+
"learning_rate": 0.0003,
|
| 379 |
+
"loss": 7.2316,
|
| 380 |
+
"step": 3800
|
| 381 |
+
},
|
| 382 |
+
{
|
| 383 |
+
"epoch": 7.3724007561436675,
|
| 384 |
+
"grad_norm": 3.7182490825653076,
|
| 385 |
+
"learning_rate": 0.0003,
|
| 386 |
+
"loss": 7.2591,
|
| 387 |
+
"step": 3900
|
| 388 |
+
},
|
| 389 |
+
{
|
| 390 |
+
"epoch": 7.561436672967864,
|
| 391 |
+
"grad_norm": 31.241104125976562,
|
| 392 |
+
"learning_rate": 0.0003,
|
| 393 |
+
"loss": 7.2557,
|
| 394 |
+
"step": 4000
|
| 395 |
+
},
|
| 396 |
+
{
|
| 397 |
+
"epoch": 7.750472589792061,
|
| 398 |
+
"grad_norm": 896688896.0,
|
| 399 |
+
"learning_rate": 0.0003,
|
| 400 |
+
"loss": 8.4436,
|
| 401 |
+
"step": 4100
|
| 402 |
+
},
|
| 403 |
+
{
|
| 404 |
+
"epoch": 7.939508506616257,
|
| 405 |
+
"grad_norm": 1.1150538843415904e+19,
|
| 406 |
+
"learning_rate": 0.0003,
|
| 407 |
+
"loss": 8.7572,
|
| 408 |
+
"step": 4200
|
| 409 |
+
},
|
| 410 |
+
{
|
| 411 |
+
"epoch": 8.0,
|
| 412 |
+
"eval_accuracy": 0.26184615384615384,
|
| 413 |
+
"eval_loss": 6.479637622833252,
|
| 414 |
+
"eval_runtime": 5.1473,
|
| 415 |
+
"eval_samples_per_second": 97.138,
|
| 416 |
+
"eval_steps_per_second": 12.239,
|
| 417 |
+
"step": 4232
|
| 418 |
+
},
|
| 419 |
+
{
|
| 420 |
+
"epoch": 8.0,
|
| 421 |
+
"eval_exact_match": 0.0,
|
| 422 |
+
"eval_f1": 0.0,
|
| 423 |
+
"step": 4232
|
| 424 |
+
},
|
| 425 |
+
{
|
| 426 |
+
"epoch": 8.128544423440454,
|
| 427 |
+
"grad_norm": 7.2513813972473145,
|
| 428 |
+
"learning_rate": 0.0003,
|
| 429 |
+
"loss": 8.1381,
|
| 430 |
+
"step": 4300
|
| 431 |
+
},
|
| 432 |
+
{
|
| 433 |
+
"epoch": 8.31758034026465,
|
| 434 |
+
"grad_norm": 14.863628387451172,
|
| 435 |
+
"learning_rate": 0.0003,
|
| 436 |
+
"loss": 7.3335,
|
| 437 |
+
"step": 4400
|
| 438 |
+
},
|
| 439 |
+
{
|
| 440 |
+
"epoch": 8.506616257088847,
|
| 441 |
+
"grad_norm": 5.358551979064941,
|
| 442 |
+
"learning_rate": 0.0003,
|
| 443 |
+
"loss": 8.3921,
|
| 444 |
+
"step": 4500
|
| 445 |
+
},
|
| 446 |
+
{
|
| 447 |
+
"epoch": 8.695652173913043,
|
| 448 |
+
"grad_norm": 113.71702575683594,
|
| 449 |
+
"learning_rate": 0.0003,
|
| 450 |
+
"loss": 7.221,
|
| 451 |
+
"step": 4600
|
| 452 |
+
},
|
| 453 |
+
{
|
| 454 |
+
"epoch": 8.88468809073724,
|
| 455 |
+
"grad_norm": 2.3997488021850586,
|
| 456 |
+
"learning_rate": 0.0003,
|
| 457 |
+
"loss": 8.4835,
|
| 458 |
+
"step": 4700
|
| 459 |
+
},
|
| 460 |
+
{
|
| 461 |
+
"epoch": 9.0,
|
| 462 |
+
"eval_accuracy": 0.26184615384615384,
|
| 463 |
+
"eval_loss": 6.614655017852783,
|
| 464 |
+
"eval_runtime": 5.026,
|
| 465 |
+
"eval_samples_per_second": 99.483,
|
| 466 |
+
"eval_steps_per_second": 12.535,
|
| 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": 7.822358131408691,
|
| 478 |
+
"learning_rate": 0.0003,
|
| 479 |
+
"loss": 7.2836,
|
| 480 |
+
"step": 4800
|
| 481 |
+
},
|
| 482 |
+
{
|
| 483 |
+
"epoch": 9.262759924385634,
|
| 484 |
+
"grad_norm": 89279359352832.0,
|
| 485 |
+
"learning_rate": 0.0003,
|
| 486 |
+
"loss": 7.2712,
|
| 487 |
+
"step": 4900
|
| 488 |
+
},
|
| 489 |
+
{
|
| 490 |
+
"epoch": 9.45179584120983,
|
| 491 |
+
"grad_norm": 656.6845703125,
|
| 492 |
+
"learning_rate": 0.0003,
|
| 493 |
+
"loss": 7.2623,
|
| 494 |
+
"step": 5000
|
| 495 |
+
},
|
| 496 |
+
{
|
| 497 |
+
"epoch": 9.640831758034027,
|
| 498 |
+
"grad_norm": 13841700.0,
|
| 499 |
+
"learning_rate": 0.0003,
|
| 500 |
+
"loss": 7.4981,
|
| 501 |
+
"step": 5100
|
| 502 |
+
},
|
| 503 |
+
{
|
| 504 |
+
"epoch": 9.829867674858223,
|
| 505 |
+
"grad_norm": 768177.9375,
|
| 506 |
+
"learning_rate": 0.0003,
|
| 507 |
+
"loss": 7.6044,
|
| 508 |
+
"step": 5200
|
| 509 |
+
},
|
| 510 |
+
{
|
| 511 |
+
"epoch": 10.0,
|
| 512 |
+
"eval_accuracy": 0.2508717948717949,
|
| 513 |
+
"eval_loss": 8.167869567871094,
|
| 514 |
+
"eval_runtime": 5.1342,
|
| 515 |
+
"eval_samples_per_second": 97.387,
|
| 516 |
+
"eval_steps_per_second": 12.271,
|
| 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": 389293.03125,
|
| 528 |
+
"learning_rate": 0.0003,
|
| 529 |
+
"loss": 7.9142,
|
| 530 |
+
"step": 5300
|
| 531 |
+
},
|
| 532 |
+
{
|
| 533 |
+
"epoch": 10.207939508506616,
|
| 534 |
+
"grad_norm": 3129429.25,
|
| 535 |
+
"learning_rate": 0.0003,
|
| 536 |
+
"loss": 8.998,
|
| 537 |
+
"step": 5400
|
| 538 |
+
},
|
| 539 |
+
{
|
| 540 |
+
"epoch": 10.396975425330814,
|
| 541 |
+
"grad_norm": 6832450.5,
|
| 542 |
+
"learning_rate": 0.0003,
|
| 543 |
+
"loss": 8.9548,
|
| 544 |
+
"step": 5500
|
| 545 |
+
},
|
| 546 |
+
{
|
| 547 |
+
"epoch": 10.58601134215501,
|
| 548 |
+
"grad_norm": 24994486.0,
|
| 549 |
+
"learning_rate": 0.0003,
|
| 550 |
+
"loss": 9.3085,
|
| 551 |
+
"step": 5600
|
| 552 |
+
},
|
| 553 |
+
{
|
| 554 |
+
"epoch": 10.775047258979207,
|
| 555 |
+
"grad_norm": 10634.3076171875,
|
| 556 |
+
"learning_rate": 0.0003,
|
| 557 |
+
"loss": 9.3052,
|
| 558 |
+
"step": 5700
|
| 559 |
+
},
|
| 560 |
+
{
|
| 561 |
+
"epoch": 10.964083175803403,
|
| 562 |
+
"grad_norm": 37.100101470947266,
|
| 563 |
+
"learning_rate": 0.0003,
|
| 564 |
+
"loss": 10.0839,
|
| 565 |
+
"step": 5800
|
| 566 |
+
},
|
| 567 |
+
{
|
| 568 |
+
"epoch": 11.0,
|
| 569 |
+
"eval_accuracy": 0.2362051282051282,
|
| 570 |
+
"eval_loss": 6.820600509643555,
|
| 571 |
+
"eval_runtime": 5.0262,
|
| 572 |
+
"eval_samples_per_second": 99.478,
|
| 573 |
+
"eval_steps_per_second": 12.534,
|
| 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": 319.2048645019531,
|
| 585 |
+
"learning_rate": 0.0003,
|
| 586 |
+
"loss": 7.4258,
|
| 587 |
+
"step": 5900
|
| 588 |
+
},
|
| 589 |
+
{
|
| 590 |
+
"epoch": 11.342155009451796,
|
| 591 |
+
"grad_norm": 72.28472900390625,
|
| 592 |
+
"learning_rate": 0.0003,
|
| 593 |
+
"loss": 7.2609,
|
| 594 |
+
"step": 6000
|
| 595 |
+
},
|
| 596 |
+
{
|
| 597 |
+
"epoch": 11.531190926275993,
|
| 598 |
+
"grad_norm": 11.851283073425293,
|
| 599 |
+
"learning_rate": 0.0003,
|
| 600 |
+
"loss": 7.5323,
|
| 601 |
+
"step": 6100
|
| 602 |
+
},
|
| 603 |
+
{
|
| 604 |
+
"epoch": 11.720226843100189,
|
| 605 |
+
"grad_norm": 6.574771404266357,
|
| 606 |
+
"learning_rate": 0.0003,
|
| 607 |
+
"loss": 7.3028,
|
| 608 |
+
"step": 6200
|
| 609 |
+
},
|
| 610 |
+
{
|
| 611 |
+
"epoch": 11.909262759924385,
|
| 612 |
+
"grad_norm": 4.578986644744873,
|
| 613 |
+
"learning_rate": 0.0003,
|
| 614 |
+
"loss": 7.2957,
|
| 615 |
+
"step": 6300
|
| 616 |
+
},
|
| 617 |
+
{
|
| 618 |
+
"epoch": 12.0,
|
| 619 |
+
"eval_accuracy": 0.26184615384615384,
|
| 620 |
+
"eval_loss": 6.52316427230835,
|
| 621 |
+
"eval_runtime": 5.0329,
|
| 622 |
+
"eval_samples_per_second": 99.346,
|
| 623 |
+
"eval_steps_per_second": 12.518,
|
| 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": 21.886821746826172,
|
| 635 |
+
"learning_rate": 0.0003,
|
| 636 |
+
"loss": 7.189,
|
| 637 |
+
"step": 6400
|
| 638 |
+
},
|
| 639 |
+
{
|
| 640 |
+
"epoch": 12.287334593572778,
|
| 641 |
+
"grad_norm": 102.8978500366211,
|
| 642 |
+
"learning_rate": 0.0003,
|
| 643 |
+
"loss": 7.2076,
|
| 644 |
+
"step": 6500
|
| 645 |
+
},
|
| 646 |
+
{
|
| 647 |
+
"epoch": 12.476370510396976,
|
| 648 |
+
"grad_norm": 4.47935676574707,
|
| 649 |
+
"learning_rate": 0.0003,
|
| 650 |
+
"loss": 7.2369,
|
| 651 |
+
"step": 6600
|
| 652 |
+
},
|
| 653 |
+
{
|
| 654 |
+
"epoch": 12.665406427221171,
|
| 655 |
+
"grad_norm": 259.2627868652344,
|
| 656 |
+
"learning_rate": 0.0003,
|
| 657 |
+
"loss": 7.3119,
|
| 658 |
+
"step": 6700
|
| 659 |
+
},
|
| 660 |
+
{
|
| 661 |
+
"epoch": 12.854442344045369,
|
| 662 |
+
"grad_norm": 473.829345703125,
|
| 663 |
+
"learning_rate": 0.0003,
|
| 664 |
+
"loss": 7.3484,
|
| 665 |
+
"step": 6800
|
| 666 |
+
},
|
| 667 |
+
{
|
| 668 |
+
"epoch": 13.0,
|
| 669 |
+
"eval_accuracy": 0.26184615384615384,
|
| 670 |
+
"eval_loss": 6.655651569366455,
|
| 671 |
+
"eval_runtime": 5.7482,
|
| 672 |
+
"eval_samples_per_second": 86.984,
|
| 673 |
+
"eval_steps_per_second": 10.96,
|
| 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": 150.06381225585938,
|
| 685 |
+
"learning_rate": 0.0003,
|
| 686 |
+
"loss": 7.2438,
|
| 687 |
+
"step": 6900
|
| 688 |
+
},
|
| 689 |
+
{
|
| 690 |
+
"epoch": 13.232514177693762,
|
| 691 |
+
"grad_norm": 1546.5465087890625,
|
| 692 |
+
"learning_rate": 0.0003,
|
| 693 |
+
"loss": 7.2503,
|
| 694 |
+
"step": 7000
|
| 695 |
+
},
|
| 696 |
+
{
|
| 697 |
+
"epoch": 13.421550094517958,
|
| 698 |
+
"grad_norm": 18108432.0,
|
| 699 |
+
"learning_rate": 0.0003,
|
| 700 |
+
"loss": 7.5797,
|
| 701 |
+
"step": 7100
|
| 702 |
+
},
|
| 703 |
+
{
|
| 704 |
+
"epoch": 13.610586011342155,
|
| 705 |
+
"grad_norm": 379725545472.0,
|
| 706 |
+
"learning_rate": 0.0003,
|
| 707 |
+
"loss": 8.6953,
|
| 708 |
+
"step": 7200
|
| 709 |
+
},
|
| 710 |
+
{
|
| 711 |
+
"epoch": 13.799621928166351,
|
| 712 |
+
"grad_norm": 14.351493835449219,
|
| 713 |
+
"learning_rate": 0.0003,
|
| 714 |
+
"loss": 7.5947,
|
| 715 |
+
"step": 7300
|
| 716 |
+
},
|
| 717 |
+
{
|
| 718 |
+
"epoch": 13.988657844990549,
|
| 719 |
+
"grad_norm": 3.4325509071350098,
|
| 720 |
+
"learning_rate": 0.0003,
|
| 721 |
+
"loss": 8.2657,
|
| 722 |
+
"step": 7400
|
| 723 |
+
},
|
| 724 |
+
{
|
| 725 |
+
"epoch": 14.0,
|
| 726 |
+
"eval_accuracy": 0.2362051282051282,
|
| 727 |
+
"eval_loss": 6.780774116516113,
|
| 728 |
+
"eval_runtime": 5.7058,
|
| 729 |
+
"eval_samples_per_second": 87.629,
|
| 730 |
+
"eval_steps_per_second": 11.041,
|
| 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": 3408474406912.0,
|
| 742 |
+
"learning_rate": 0.0003,
|
| 743 |
+
"loss": 7.5487,
|
| 744 |
+
"step": 7500
|
| 745 |
+
},
|
| 746 |
+
{
|
| 747 |
+
"epoch": 14.366729678638942,
|
| 748 |
+
"grad_norm": 766890.1875,
|
| 749 |
+
"learning_rate": 0.0003,
|
| 750 |
+
"loss": 7.4765,
|
| 751 |
+
"step": 7600
|
| 752 |
+
},
|
| 753 |
+
{
|
| 754 |
+
"epoch": 14.555765595463138,
|
| 755 |
+
"grad_norm": 4749101.0,
|
| 756 |
+
"learning_rate": 0.0003,
|
| 757 |
+
"loss": 7.4927,
|
| 758 |
+
"step": 7700
|
| 759 |
+
},
|
| 760 |
+
{
|
| 761 |
+
"epoch": 14.744801512287335,
|
| 762 |
+
"grad_norm": 21509612.0,
|
| 763 |
+
"learning_rate": 0.0003,
|
| 764 |
+
"loss": 7.4864,
|
| 765 |
+
"step": 7800
|
| 766 |
+
},
|
| 767 |
+
{
|
| 768 |
+
"epoch": 14.93383742911153,
|
| 769 |
+
"grad_norm": 1.0536716136244838e+18,
|
| 770 |
+
"learning_rate": 0.0003,
|
| 771 |
+
"loss": 7.4391,
|
| 772 |
+
"step": 7900
|
| 773 |
+
},
|
| 774 |
+
{
|
| 775 |
+
"epoch": 15.0,
|
| 776 |
+
"eval_accuracy": 0.1995897435897436,
|
| 777 |
+
"eval_loss": NaN,
|
| 778 |
+
"eval_runtime": 5.788,
|
| 779 |
+
"eval_samples_per_second": 86.386,
|
| 780 |
+
"eval_steps_per_second": 10.885,
|
| 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": NaN,
|
| 792 |
+
"learning_rate": 0.0003,
|
| 793 |
+
"loss": 602.3478,
|
| 794 |
+
"step": 8000
|
| 795 |
+
},
|
| 796 |
+
{
|
| 797 |
+
"epoch": 15.311909262759924,
|
| 798 |
+
"grad_norm": NaN,
|
| 799 |
+
"learning_rate": 0.0003,
|
| 800 |
+
"loss": 0.0,
|
| 801 |
+
"step": 8100
|
| 802 |
+
},
|
| 803 |
+
{
|
| 804 |
+
"epoch": 15.500945179584122,
|
| 805 |
+
"grad_norm": NaN,
|
| 806 |
+
"learning_rate": 0.0003,
|
| 807 |
+
"loss": 0.0,
|
| 808 |
+
"step": 8200
|
| 809 |
+
},
|
| 810 |
+
{
|
| 811 |
+
"epoch": 15.689981096408317,
|
| 812 |
+
"grad_norm": NaN,
|
| 813 |
+
"learning_rate": 0.0003,
|
| 814 |
+
"loss": 0.0,
|
| 815 |
+
"step": 8300
|
| 816 |
+
},
|
| 817 |
+
{
|
| 818 |
+
"epoch": 15.879017013232515,
|
| 819 |
+
"grad_norm": NaN,
|
| 820 |
+
"learning_rate": 0.0003,
|
| 821 |
+
"loss": 0.0,
|
| 822 |
+
"step": 8400
|
| 823 |
+
},
|
| 824 |
+
{
|
| 825 |
+
"epoch": 16.0,
|
| 826 |
+
"eval_accuracy": 0.1995897435897436,
|
| 827 |
+
"eval_loss": NaN,
|
| 828 |
+
"eval_runtime": 5.1552,
|
| 829 |
+
"eval_samples_per_second": 96.99,
|
| 830 |
+
"eval_steps_per_second": 12.221,
|
| 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": NaN,
|
| 842 |
+
"learning_rate": 0.0003,
|
| 843 |
+
"loss": 0.0,
|
| 844 |
+
"step": 8500
|
| 845 |
+
},
|
| 846 |
+
{
|
| 847 |
+
"epoch": 16.257088846880908,
|
| 848 |
+
"grad_norm": NaN,
|
| 849 |
+
"learning_rate": 0.0003,
|
| 850 |
+
"loss": 0.0,
|
| 851 |
+
"step": 8600
|
| 852 |
+
},
|
| 853 |
+
{
|
| 854 |
+
"epoch": 16.446124763705104,
|
| 855 |
+
"grad_norm": NaN,
|
| 856 |
+
"learning_rate": 0.0003,
|
| 857 |
+
"loss": 0.0,
|
| 858 |
+
"step": 8700
|
| 859 |
+
},
|
| 860 |
+
{
|
| 861 |
+
"epoch": 16.6351606805293,
|
| 862 |
+
"grad_norm": NaN,
|
| 863 |
+
"learning_rate": 0.0003,
|
| 864 |
+
"loss": 0.0,
|
| 865 |
+
"step": 8800
|
| 866 |
+
},
|
| 867 |
+
{
|
| 868 |
+
"epoch": 16.8241965973535,
|
| 869 |
+
"grad_norm": NaN,
|
| 870 |
+
"learning_rate": 0.0003,
|
| 871 |
+
"loss": 0.0,
|
| 872 |
+
"step": 8900
|
| 873 |
+
},
|
| 874 |
+
{
|
| 875 |
+
"epoch": 17.0,
|
| 876 |
+
"eval_accuracy": 0.1995897435897436,
|
| 877 |
+
"eval_loss": NaN,
|
| 878 |
+
"eval_runtime": 5.2516,
|
| 879 |
+
"eval_samples_per_second": 95.209,
|
| 880 |
+
"eval_steps_per_second": 11.996,
|
| 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": NaN,
|
| 892 |
+
"learning_rate": 0.0003,
|
| 893 |
+
"loss": 0.0,
|
| 894 |
+
"step": 9000
|
| 895 |
+
},
|
| 896 |
+
{
|
| 897 |
+
"epoch": 17.20226843100189,
|
| 898 |
+
"grad_norm": NaN,
|
| 899 |
+
"learning_rate": 0.0003,
|
| 900 |
+
"loss": 0.0,
|
| 901 |
+
"step": 9100
|
| 902 |
+
},
|
| 903 |
+
{
|
| 904 |
+
"epoch": 17.391304347826086,
|
| 905 |
+
"grad_norm": NaN,
|
| 906 |
+
"learning_rate": 0.0003,
|
| 907 |
+
"loss": 0.0,
|
| 908 |
+
"step": 9200
|
| 909 |
+
},
|
| 910 |
+
{
|
| 911 |
+
"epoch": 17.58034026465028,
|
| 912 |
+
"grad_norm": NaN,
|
| 913 |
+
"learning_rate": 0.0003,
|
| 914 |
+
"loss": 0.0,
|
| 915 |
+
"step": 9300
|
| 916 |
+
},
|
| 917 |
+
{
|
| 918 |
+
"epoch": 17.76937618147448,
|
| 919 |
+
"grad_norm": NaN,
|
| 920 |
+
"learning_rate": 0.0003,
|
| 921 |
+
"loss": 0.0,
|
| 922 |
+
"step": 9400
|
| 923 |
+
},
|
| 924 |
+
{
|
| 925 |
+
"epoch": 17.958412098298677,
|
| 926 |
+
"grad_norm": NaN,
|
| 927 |
+
"learning_rate": 0.0003,
|
| 928 |
+
"loss": 0.0,
|
| 929 |
+
"step": 9500
|
| 930 |
+
},
|
| 931 |
+
{
|
| 932 |
+
"epoch": 18.0,
|
| 933 |
+
"eval_accuracy": 0.1995897435897436,
|
| 934 |
+
"eval_loss": NaN,
|
| 935 |
+
"eval_runtime": 5.7505,
|
| 936 |
+
"eval_samples_per_second": 86.949,
|
| 937 |
+
"eval_steps_per_second": 10.956,
|
| 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": NaN,
|
| 949 |
+
"learning_rate": 0.0003,
|
| 950 |
+
"loss": 0.0,
|
| 951 |
+
"step": 9600
|
| 952 |
+
},
|
| 953 |
+
{
|
| 954 |
+
"epoch": 18.33648393194707,
|
| 955 |
+
"grad_norm": NaN,
|
| 956 |
+
"learning_rate": 0.0003,
|
| 957 |
+
"loss": 0.0,
|
| 958 |
+
"step": 9700
|
| 959 |
+
},
|
| 960 |
+
{
|
| 961 |
+
"epoch": 18.525519848771268,
|
| 962 |
+
"grad_norm": NaN,
|
| 963 |
+
"learning_rate": 0.0003,
|
| 964 |
+
"loss": 0.0,
|
| 965 |
+
"step": 9800
|
| 966 |
+
},
|
| 967 |
+
{
|
| 968 |
+
"epoch": 18.714555765595463,
|
| 969 |
+
"grad_norm": NaN,
|
| 970 |
+
"learning_rate": 0.0003,
|
| 971 |
+
"loss": 0.0,
|
| 972 |
+
"step": 9900
|
| 973 |
+
},
|
| 974 |
+
{
|
| 975 |
+
"epoch": 18.90359168241966,
|
| 976 |
+
"grad_norm": NaN,
|
| 977 |
+
"learning_rate": 0.0003,
|
| 978 |
+
"loss": 0.0,
|
| 979 |
+
"step": 10000
|
| 980 |
+
},
|
| 981 |
+
{
|
| 982 |
+
"epoch": 19.0,
|
| 983 |
+
"eval_accuracy": 0.1995897435897436,
|
| 984 |
+
"eval_loss": NaN,
|
| 985 |
+
"eval_runtime": 5.3461,
|
| 986 |
+
"eval_samples_per_second": 93.526,
|
| 987 |
+
"eval_steps_per_second": 11.784,
|
| 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": NaN,
|
| 999 |
+
"learning_rate": 0.0003,
|
| 1000 |
+
"loss": 0.0,
|
| 1001 |
+
"step": 10100
|
| 1002 |
+
},
|
| 1003 |
+
{
|
| 1004 |
+
"epoch": 19.281663516068054,
|
| 1005 |
+
"grad_norm": NaN,
|
| 1006 |
+
"learning_rate": 0.0003,
|
| 1007 |
+
"loss": 0.0,
|
| 1008 |
+
"step": 10200
|
| 1009 |
+
},
|
| 1010 |
+
{
|
| 1011 |
+
"epoch": 19.47069943289225,
|
| 1012 |
+
"grad_norm": NaN,
|
| 1013 |
+
"learning_rate": 0.0003,
|
| 1014 |
+
"loss": 0.0,
|
| 1015 |
+
"step": 10300
|
| 1016 |
+
},
|
| 1017 |
+
{
|
| 1018 |
+
"epoch": 19.659735349716446,
|
| 1019 |
+
"grad_norm": NaN,
|
| 1020 |
+
"learning_rate": 0.0003,
|
| 1021 |
+
"loss": 0.0,
|
| 1022 |
+
"step": 10400
|
| 1023 |
+
},
|
| 1024 |
+
{
|
| 1025 |
+
"epoch": 19.84877126654064,
|
| 1026 |
+
"grad_norm": NaN,
|
| 1027 |
+
"learning_rate": 0.0003,
|
| 1028 |
+
"loss": 0.0,
|
| 1029 |
+
"step": 10500
|
| 1030 |
+
},
|
| 1031 |
+
{
|
| 1032 |
+
"epoch": 20.0,
|
| 1033 |
+
"eval_accuracy": 0.1995897435897436,
|
| 1034 |
+
"eval_loss": NaN,
|
| 1035 |
+
"eval_runtime": 5.0526,
|
| 1036 |
+
"eval_samples_per_second": 98.959,
|
| 1037 |
+
"eval_steps_per_second": 12.469,
|
| 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": NaN,
|
| 1049 |
+
"learning_rate": 0.0003,
|
| 1050 |
+
"loss": 0.0,
|
| 1051 |
+
"step": 10600
|
| 1052 |
+
},
|
| 1053 |
+
{
|
| 1054 |
+
"epoch": 20.226843100189036,
|
| 1055 |
+
"grad_norm": NaN,
|
| 1056 |
+
"learning_rate": 0.0003,
|
| 1057 |
+
"loss": 0.0,
|
| 1058 |
+
"step": 10700
|
| 1059 |
+
},
|
| 1060 |
+
{
|
| 1061 |
+
"epoch": 20.415879017013232,
|
| 1062 |
+
"grad_norm": NaN,
|
| 1063 |
+
"learning_rate": 0.0003,
|
| 1064 |
+
"loss": 0.0,
|
| 1065 |
+
"step": 10800
|
| 1066 |
+
},
|
| 1067 |
+
{
|
| 1068 |
+
"epoch": 20.604914933837428,
|
| 1069 |
+
"grad_norm": NaN,
|
| 1070 |
+
"learning_rate": 0.0003,
|
| 1071 |
+
"loss": 0.0,
|
| 1072 |
+
"step": 10900
|
| 1073 |
+
},
|
| 1074 |
+
{
|
| 1075 |
+
"epoch": 20.793950850661627,
|
| 1076 |
+
"grad_norm": NaN,
|
| 1077 |
+
"learning_rate": 0.0003,
|
| 1078 |
+
"loss": 0.0,
|
| 1079 |
+
"step": 11000
|
| 1080 |
+
},
|
| 1081 |
+
{
|
| 1082 |
+
"epoch": 20.982986767485823,
|
| 1083 |
+
"grad_norm": NaN,
|
| 1084 |
+
"learning_rate": 0.0003,
|
| 1085 |
+
"loss": 0.0,
|
| 1086 |
+
"step": 11100
|
| 1087 |
+
},
|
| 1088 |
+
{
|
| 1089 |
+
"epoch": 21.0,
|
| 1090 |
+
"eval_accuracy": 0.1995897435897436,
|
| 1091 |
+
"eval_loss": NaN,
|
| 1092 |
+
"eval_runtime": 5.7608,
|
| 1093 |
+
"eval_samples_per_second": 86.793,
|
| 1094 |
+
"eval_steps_per_second": 10.936,
|
| 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": NaN,
|
| 1106 |
+
"learning_rate": 0.0003,
|
| 1107 |
+
"loss": 0.0,
|
| 1108 |
+
"step": 11200
|
| 1109 |
+
},
|
| 1110 |
+
{
|
| 1111 |
+
"epoch": 21.361058601134214,
|
| 1112 |
+
"grad_norm": NaN,
|
| 1113 |
+
"learning_rate": 0.0003,
|
| 1114 |
+
"loss": 0.0,
|
| 1115 |
+
"step": 11300
|
| 1116 |
+
},
|
| 1117 |
+
{
|
| 1118 |
+
"epoch": 21.550094517958414,
|
| 1119 |
+
"grad_norm": NaN,
|
| 1120 |
+
"learning_rate": 0.0003,
|
| 1121 |
+
"loss": 0.0,
|
| 1122 |
+
"step": 11400
|
| 1123 |
+
},
|
| 1124 |
+
{
|
| 1125 |
+
"epoch": 21.73913043478261,
|
| 1126 |
+
"grad_norm": NaN,
|
| 1127 |
+
"learning_rate": 0.0003,
|
| 1128 |
+
"loss": 0.0,
|
| 1129 |
+
"step": 11500
|
| 1130 |
+
},
|
| 1131 |
+
{
|
| 1132 |
+
"epoch": 21.928166351606805,
|
| 1133 |
+
"grad_norm": NaN,
|
| 1134 |
+
"learning_rate": 0.0003,
|
| 1135 |
+
"loss": 0.0,
|
| 1136 |
+
"step": 11600
|
| 1137 |
+
},
|
| 1138 |
+
{
|
| 1139 |
+
"epoch": 22.0,
|
| 1140 |
+
"eval_accuracy": 0.1995897435897436,
|
| 1141 |
+
"eval_loss": NaN,
|
| 1142 |
+
"eval_runtime": 5.0399,
|
| 1143 |
+
"eval_samples_per_second": 99.209,
|
| 1144 |
+
"eval_steps_per_second": 12.5,
|
| 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": NaN,
|
| 1156 |
+
"learning_rate": 0.0003,
|
| 1157 |
+
"loss": 0.0,
|
| 1158 |
+
"step": 11700
|
| 1159 |
+
},
|
| 1160 |
+
{
|
| 1161 |
+
"epoch": 22.3062381852552,
|
| 1162 |
+
"grad_norm": NaN,
|
| 1163 |
+
"learning_rate": 0.0003,
|
| 1164 |
+
"loss": 0.0,
|
| 1165 |
+
"step": 11800
|
| 1166 |
+
},
|
| 1167 |
+
{
|
| 1168 |
+
"epoch": 22.495274102079396,
|
| 1169 |
+
"grad_norm": NaN,
|
| 1170 |
+
"learning_rate": 0.0003,
|
| 1171 |
+
"loss": 0.0,
|
| 1172 |
+
"step": 11900
|
| 1173 |
+
},
|
| 1174 |
+
{
|
| 1175 |
+
"epoch": 22.68431001890359,
|
| 1176 |
+
"grad_norm": NaN,
|
| 1177 |
+
"learning_rate": 0.0003,
|
| 1178 |
+
"loss": 0.0,
|
| 1179 |
+
"step": 12000
|
| 1180 |
+
},
|
| 1181 |
+
{
|
| 1182 |
+
"epoch": 22.873345935727787,
|
| 1183 |
+
"grad_norm": NaN,
|
| 1184 |
+
"learning_rate": 0.0003,
|
| 1185 |
+
"loss": 0.0,
|
| 1186 |
+
"step": 12100
|
| 1187 |
+
},
|
| 1188 |
+
{
|
| 1189 |
+
"epoch": 23.0,
|
| 1190 |
+
"eval_accuracy": 0.1995897435897436,
|
| 1191 |
+
"eval_loss": NaN,
|
| 1192 |
+
"eval_runtime": 5.2777,
|
| 1193 |
+
"eval_samples_per_second": 94.738,
|
| 1194 |
+
"eval_steps_per_second": 11.937,
|
| 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": NaN,
|
| 1206 |
+
"learning_rate": 0.0003,
|
| 1207 |
+
"loss": 0.0,
|
| 1208 |
+
"step": 12200
|
| 1209 |
+
},
|
| 1210 |
+
{
|
| 1211 |
+
"epoch": 23.251417769376182,
|
| 1212 |
+
"grad_norm": NaN,
|
| 1213 |
+
"learning_rate": 0.0003,
|
| 1214 |
+
"loss": 0.0,
|
| 1215 |
+
"step": 12300
|
| 1216 |
+
},
|
| 1217 |
+
{
|
| 1218 |
+
"epoch": 23.440453686200378,
|
| 1219 |
+
"grad_norm": NaN,
|
| 1220 |
+
"learning_rate": 0.0003,
|
| 1221 |
+
"loss": 0.0,
|
| 1222 |
+
"step": 12400
|
| 1223 |
+
},
|
| 1224 |
+
{
|
| 1225 |
+
"epoch": 23.629489603024574,
|
| 1226 |
+
"grad_norm": NaN,
|
| 1227 |
+
"learning_rate": 0.0003,
|
| 1228 |
+
"loss": 0.0,
|
| 1229 |
+
"step": 12500
|
| 1230 |
+
},
|
| 1231 |
+
{
|
| 1232 |
+
"epoch": 23.81852551984877,
|
| 1233 |
+
"grad_norm": NaN,
|
| 1234 |
+
"learning_rate": 0.0003,
|
| 1235 |
+
"loss": 0.0,
|
| 1236 |
+
"step": 12600
|
| 1237 |
+
},
|
| 1238 |
+
{
|
| 1239 |
+
"epoch": 24.0,
|
| 1240 |
+
"eval_accuracy": 0.1995897435897436,
|
| 1241 |
+
"eval_loss": NaN,
|
| 1242 |
+
"eval_runtime": 5.0268,
|
| 1243 |
+
"eval_samples_per_second": 99.467,
|
| 1244 |
+
"eval_steps_per_second": 12.533,
|
| 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": NaN,
|
| 1256 |
+
"learning_rate": 0.0003,
|
| 1257 |
+
"loss": 0.0,
|
| 1258 |
+
"step": 12700
|
| 1259 |
+
},
|
| 1260 |
+
{
|
| 1261 |
+
"epoch": 24.196597353497165,
|
| 1262 |
+
"grad_norm": NaN,
|
| 1263 |
+
"learning_rate": 0.0003,
|
| 1264 |
+
"loss": 0.0,
|
| 1265 |
+
"step": 12800
|
| 1266 |
+
},
|
| 1267 |
+
{
|
| 1268 |
+
"epoch": 24.38563327032136,
|
| 1269 |
+
"grad_norm": NaN,
|
| 1270 |
+
"learning_rate": 0.0003,
|
| 1271 |
+
"loss": 0.0,
|
| 1272 |
+
"step": 12900
|
| 1273 |
+
},
|
| 1274 |
+
{
|
| 1275 |
+
"epoch": 24.574669187145556,
|
| 1276 |
+
"grad_norm": NaN,
|
| 1277 |
+
"learning_rate": 0.0003,
|
| 1278 |
+
"loss": 0.0,
|
| 1279 |
+
"step": 13000
|
| 1280 |
+
},
|
| 1281 |
+
{
|
| 1282 |
+
"epoch": 24.763705103969755,
|
| 1283 |
+
"grad_norm": NaN,
|
| 1284 |
+
"learning_rate": 0.0003,
|
| 1285 |
+
"loss": 0.0,
|
| 1286 |
+
"step": 13100
|
| 1287 |
+
},
|
| 1288 |
+
{
|
| 1289 |
+
"epoch": 24.95274102079395,
|
| 1290 |
+
"grad_norm": NaN,
|
| 1291 |
+
"learning_rate": 0.0003,
|
| 1292 |
+
"loss": 0.0,
|
| 1293 |
+
"step": 13200
|
| 1294 |
+
},
|
| 1295 |
+
{
|
| 1296 |
+
"epoch": 25.0,
|
| 1297 |
+
"eval_accuracy": 0.1995897435897436,
|
| 1298 |
+
"eval_loss": NaN,
|
| 1299 |
+
"eval_runtime": 5.3305,
|
| 1300 |
+
"eval_samples_per_second": 93.8,
|
| 1301 |
+
"eval_steps_per_second": 11.819,
|
| 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": NaN,
|
| 1313 |
+
"learning_rate": 0.0003,
|
| 1314 |
+
"loss": 0.0,
|
| 1315 |
+
"step": 13300
|
| 1316 |
+
},
|
| 1317 |
+
{
|
| 1318 |
+
"epoch": 25.330812854442343,
|
| 1319 |
+
"grad_norm": NaN,
|
| 1320 |
+
"learning_rate": 0.0003,
|
| 1321 |
+
"loss": 0.0,
|
| 1322 |
+
"step": 13400
|
| 1323 |
+
},
|
| 1324 |
+
{
|
| 1325 |
+
"epoch": 25.519848771266542,
|
| 1326 |
+
"grad_norm": NaN,
|
| 1327 |
+
"learning_rate": 0.0003,
|
| 1328 |
+
"loss": 0.0,
|
| 1329 |
+
"step": 13500
|
| 1330 |
+
},
|
| 1331 |
+
{
|
| 1332 |
+
"epoch": 25.708884688090738,
|
| 1333 |
+
"grad_norm": NaN,
|
| 1334 |
+
"learning_rate": 0.0003,
|
| 1335 |
+
"loss": 0.0,
|
| 1336 |
+
"step": 13600
|
| 1337 |
+
},
|
| 1338 |
+
{
|
| 1339 |
+
"epoch": 25.897920604914933,
|
| 1340 |
+
"grad_norm": NaN,
|
| 1341 |
+
"learning_rate": 0.0003,
|
| 1342 |
+
"loss": 0.0,
|
| 1343 |
+
"step": 13700
|
| 1344 |
+
},
|
| 1345 |
+
{
|
| 1346 |
+
"epoch": 26.0,
|
| 1347 |
+
"eval_accuracy": 0.1995897435897436,
|
| 1348 |
+
"eval_loss": NaN,
|
| 1349 |
+
"eval_runtime": 5.173,
|
| 1350 |
+
"eval_samples_per_second": 96.656,
|
| 1351 |
+
"eval_steps_per_second": 12.179,
|
| 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": NaN,
|
| 1363 |
+
"learning_rate": 0.0003,
|
| 1364 |
+
"loss": 0.0,
|
| 1365 |
+
"step": 13800
|
| 1366 |
+
},
|
| 1367 |
+
{
|
| 1368 |
+
"epoch": 26.27599243856333,
|
| 1369 |
+
"grad_norm": NaN,
|
| 1370 |
+
"learning_rate": 0.0003,
|
| 1371 |
+
"loss": 0.0,
|
| 1372 |
+
"step": 13900
|
| 1373 |
+
},
|
| 1374 |
+
{
|
| 1375 |
+
"epoch": 26.465028355387524,
|
| 1376 |
+
"grad_norm": NaN,
|
| 1377 |
+
"learning_rate": 0.0003,
|
| 1378 |
+
"loss": 0.0,
|
| 1379 |
+
"step": 14000
|
| 1380 |
+
},
|
| 1381 |
+
{
|
| 1382 |
+
"epoch": 26.65406427221172,
|
| 1383 |
+
"grad_norm": NaN,
|
| 1384 |
+
"learning_rate": 0.0003,
|
| 1385 |
+
"loss": 0.0,
|
| 1386 |
+
"step": 14100
|
| 1387 |
+
},
|
| 1388 |
+
{
|
| 1389 |
+
"epoch": 26.843100189035916,
|
| 1390 |
+
"grad_norm": NaN,
|
| 1391 |
+
"learning_rate": 0.0003,
|
| 1392 |
+
"loss": 0.0,
|
| 1393 |
+
"step": 14200
|
| 1394 |
+
},
|
| 1395 |
+
{
|
| 1396 |
+
"epoch": 27.0,
|
| 1397 |
+
"eval_accuracy": 0.1995897435897436,
|
| 1398 |
+
"eval_loss": NaN,
|
| 1399 |
+
"eval_runtime": 5.2291,
|
| 1400 |
+
"eval_samples_per_second": 95.618,
|
| 1401 |
+
"eval_steps_per_second": 12.048,
|
| 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": NaN,
|
| 1413 |
+
"learning_rate": 0.0003,
|
| 1414 |
+
"loss": 0.0,
|
| 1415 |
+
"step": 14300
|
| 1416 |
+
},
|
| 1417 |
+
{
|
| 1418 |
+
"epoch": 27.22117202268431,
|
| 1419 |
+
"grad_norm": NaN,
|
| 1420 |
+
"learning_rate": 0.0003,
|
| 1421 |
+
"loss": 0.0,
|
| 1422 |
+
"step": 14400
|
| 1423 |
+
},
|
| 1424 |
+
{
|
| 1425 |
+
"epoch": 27.410207939508506,
|
| 1426 |
+
"grad_norm": NaN,
|
| 1427 |
+
"learning_rate": 0.0003,
|
| 1428 |
+
"loss": 0.0,
|
| 1429 |
+
"step": 14500
|
| 1430 |
+
},
|
| 1431 |
+
{
|
| 1432 |
+
"epoch": 27.599243856332702,
|
| 1433 |
+
"grad_norm": NaN,
|
| 1434 |
+
"learning_rate": 0.0003,
|
| 1435 |
+
"loss": 0.0,
|
| 1436 |
+
"step": 14600
|
| 1437 |
+
},
|
| 1438 |
+
{
|
| 1439 |
+
"epoch": 27.7882797731569,
|
| 1440 |
+
"grad_norm": NaN,
|
| 1441 |
+
"learning_rate": 0.0003,
|
| 1442 |
+
"loss": 0.0,
|
| 1443 |
+
"step": 14700
|
| 1444 |
+
},
|
| 1445 |
+
{
|
| 1446 |
+
"epoch": 27.977315689981097,
|
| 1447 |
+
"grad_norm": NaN,
|
| 1448 |
+
"learning_rate": 0.0003,
|
| 1449 |
+
"loss": 0.0,
|
| 1450 |
+
"step": 14800
|
| 1451 |
+
},
|
| 1452 |
+
{
|
| 1453 |
+
"epoch": 28.0,
|
| 1454 |
+
"eval_accuracy": 0.1995897435897436,
|
| 1455 |
+
"eval_loss": NaN,
|
| 1456 |
+
"eval_runtime": 5.7263,
|
| 1457 |
+
"eval_samples_per_second": 87.316,
|
| 1458 |
+
"eval_steps_per_second": 11.002,
|
| 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": NaN,
|
| 1470 |
+
"learning_rate": 0.0003,
|
| 1471 |
+
"loss": 0.0,
|
| 1472 |
+
"step": 14900
|
| 1473 |
+
},
|
| 1474 |
+
{
|
| 1475 |
+
"epoch": 28.35538752362949,
|
| 1476 |
+
"grad_norm": NaN,
|
| 1477 |
+
"learning_rate": 0.0003,
|
| 1478 |
+
"loss": 0.0,
|
| 1479 |
+
"step": 15000
|
| 1480 |
+
},
|
| 1481 |
+
{
|
| 1482 |
+
"epoch": 28.544423440453688,
|
| 1483 |
+
"grad_norm": NaN,
|
| 1484 |
+
"learning_rate": 0.0003,
|
| 1485 |
+
"loss": 0.0,
|
| 1486 |
+
"step": 15100
|
| 1487 |
+
},
|
| 1488 |
+
{
|
| 1489 |
+
"epoch": 28.733459357277884,
|
| 1490 |
+
"grad_norm": NaN,
|
| 1491 |
+
"learning_rate": 0.0003,
|
| 1492 |
+
"loss": 0.0,
|
| 1493 |
+
"step": 15200
|
| 1494 |
+
},
|
| 1495 |
+
{
|
| 1496 |
+
"epoch": 28.92249527410208,
|
| 1497 |
+
"grad_norm": NaN,
|
| 1498 |
+
"learning_rate": 0.0003,
|
| 1499 |
+
"loss": 0.0,
|
| 1500 |
+
"step": 15300
|
| 1501 |
+
},
|
| 1502 |
+
{
|
| 1503 |
+
"epoch": 29.0,
|
| 1504 |
+
"eval_accuracy": 0.1995897435897436,
|
| 1505 |
+
"eval_loss": NaN,
|
| 1506 |
+
"eval_runtime": 5.7859,
|
| 1507 |
+
"eval_samples_per_second": 86.417,
|
| 1508 |
+
"eval_steps_per_second": 10.889,
|
| 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": NaN,
|
| 1520 |
+
"learning_rate": 0.0003,
|
| 1521 |
+
"loss": 0.0,
|
| 1522 |
+
"step": 15400
|
| 1523 |
+
},
|
| 1524 |
+
{
|
| 1525 |
+
"epoch": 29.300567107750474,
|
| 1526 |
+
"grad_norm": NaN,
|
| 1527 |
+
"learning_rate": 0.0003,
|
| 1528 |
+
"loss": 0.0,
|
| 1529 |
+
"step": 15500
|
| 1530 |
+
},
|
| 1531 |
+
{
|
| 1532 |
+
"epoch": 29.48960302457467,
|
| 1533 |
+
"grad_norm": NaN,
|
| 1534 |
+
"learning_rate": 0.0003,
|
| 1535 |
+
"loss": 0.0,
|
| 1536 |
+
"step": 15600
|
| 1537 |
+
},
|
| 1538 |
+
{
|
| 1539 |
+
"epoch": 29.678638941398866,
|
| 1540 |
+
"grad_norm": NaN,
|
| 1541 |
+
"learning_rate": 0.0003,
|
| 1542 |
+
"loss": 0.0,
|
| 1543 |
+
"step": 15700
|
| 1544 |
+
},
|
| 1545 |
+
{
|
| 1546 |
+
"epoch": 29.86767485822306,
|
| 1547 |
+
"grad_norm": NaN,
|
| 1548 |
+
"learning_rate": 0.0003,
|
| 1549 |
+
"loss": 0.0,
|
| 1550 |
+
"step": 15800
|
| 1551 |
+
},
|
| 1552 |
+
{
|
| 1553 |
+
"epoch": 30.0,
|
| 1554 |
+
"eval_accuracy": 0.1995897435897436,
|
| 1555 |
+
"eval_loss": NaN,
|
| 1556 |
+
"eval_runtime": 5.0784,
|
| 1557 |
+
"eval_samples_per_second": 98.456,
|
| 1558 |
+
"eval_steps_per_second": 12.406,
|
| 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": NaN,
|
| 1570 |
+
"learning_rate": 0.0003,
|
| 1571 |
+
"loss": 0.0,
|
| 1572 |
+
"step": 15900
|
| 1573 |
+
},
|
| 1574 |
+
{
|
| 1575 |
+
"epoch": 30.245746691871457,
|
| 1576 |
+
"grad_norm": NaN,
|
| 1577 |
+
"learning_rate": 0.0003,
|
| 1578 |
+
"loss": 0.0,
|
| 1579 |
+
"step": 16000
|
| 1580 |
+
},
|
| 1581 |
+
{
|
| 1582 |
+
"epoch": 30.434782608695652,
|
| 1583 |
+
"grad_norm": NaN,
|
| 1584 |
+
"learning_rate": 0.0003,
|
| 1585 |
+
"loss": 0.0,
|
| 1586 |
+
"step": 16100
|
| 1587 |
+
},
|
| 1588 |
+
{
|
| 1589 |
+
"epoch": 30.623818525519848,
|
| 1590 |
+
"grad_norm": NaN,
|
| 1591 |
+
"learning_rate": 0.0003,
|
| 1592 |
+
"loss": 0.0,
|
| 1593 |
+
"step": 16200
|
| 1594 |
+
},
|
| 1595 |
+
{
|
| 1596 |
+
"epoch": 30.812854442344044,
|
| 1597 |
+
"grad_norm": NaN,
|
| 1598 |
+
"learning_rate": 0.0003,
|
| 1599 |
+
"loss": 0.0,
|
| 1600 |
+
"step": 16300
|
| 1601 |
+
},
|
| 1602 |
+
{
|
| 1603 |
+
"epoch": 31.0,
|
| 1604 |
+
"eval_accuracy": 0.1995897435897436,
|
| 1605 |
+
"eval_loss": NaN,
|
| 1606 |
+
"eval_runtime": 5.2879,
|
| 1607 |
+
"eval_samples_per_second": 94.556,
|
| 1608 |
+
"eval_steps_per_second": 11.914,
|
| 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": NaN,
|
| 1620 |
+
"learning_rate": 0.0003,
|
| 1621 |
+
"loss": 0.0,
|
| 1622 |
+
"step": 16400
|
| 1623 |
+
},
|
| 1624 |
+
{
|
| 1625 |
+
"epoch": 31.19092627599244,
|
| 1626 |
+
"grad_norm": NaN,
|
| 1627 |
+
"learning_rate": 0.0003,
|
| 1628 |
+
"loss": 0.0,
|
| 1629 |
+
"step": 16500
|
| 1630 |
+
},
|
| 1631 |
+
{
|
| 1632 |
+
"epoch": 31.379962192816635,
|
| 1633 |
+
"grad_norm": NaN,
|
| 1634 |
+
"learning_rate": 0.0003,
|
| 1635 |
+
"loss": 0.0,
|
| 1636 |
+
"step": 16600
|
| 1637 |
+
},
|
| 1638 |
+
{
|
| 1639 |
+
"epoch": 31.56899810964083,
|
| 1640 |
+
"grad_norm": NaN,
|
| 1641 |
+
"learning_rate": 0.0003,
|
| 1642 |
+
"loss": 0.0,
|
| 1643 |
+
"step": 16700
|
| 1644 |
+
},
|
| 1645 |
+
{
|
| 1646 |
+
"epoch": 31.75803402646503,
|
| 1647 |
+
"grad_norm": NaN,
|
| 1648 |
+
"learning_rate": 0.0003,
|
| 1649 |
+
"loss": 0.0,
|
| 1650 |
+
"step": 16800
|
| 1651 |
+
},
|
| 1652 |
+
{
|
| 1653 |
+
"epoch": 31.947069943289225,
|
| 1654 |
+
"grad_norm": NaN,
|
| 1655 |
+
"learning_rate": 0.0003,
|
| 1656 |
+
"loss": 0.0,
|
| 1657 |
+
"step": 16900
|
| 1658 |
+
},
|
| 1659 |
+
{
|
| 1660 |
+
"epoch": 32.0,
|
| 1661 |
+
"eval_accuracy": 0.1995897435897436,
|
| 1662 |
+
"eval_loss": NaN,
|
| 1663 |
+
"eval_runtime": 5.324,
|
| 1664 |
+
"eval_samples_per_second": 93.915,
|
| 1665 |
+
"eval_steps_per_second": 11.833,
|
| 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": NaN,
|
| 1677 |
+
"learning_rate": 0.0003,
|
| 1678 |
+
"loss": 0.0,
|
| 1679 |
+
"step": 17000
|
| 1680 |
+
},
|
| 1681 |
+
{
|
| 1682 |
+
"epoch": 32.32514177693762,
|
| 1683 |
+
"grad_norm": NaN,
|
| 1684 |
+
"learning_rate": 0.0003,
|
| 1685 |
+
"loss": 0.0,
|
| 1686 |
+
"step": 17100
|
| 1687 |
+
},
|
| 1688 |
+
{
|
| 1689 |
+
"epoch": 32.514177693761816,
|
| 1690 |
+
"grad_norm": NaN,
|
| 1691 |
+
"learning_rate": 0.0003,
|
| 1692 |
+
"loss": 0.0,
|
| 1693 |
+
"step": 17200
|
| 1694 |
+
},
|
| 1695 |
+
{
|
| 1696 |
+
"epoch": 32.70321361058601,
|
| 1697 |
+
"grad_norm": NaN,
|
| 1698 |
+
"learning_rate": 0.0003,
|
| 1699 |
+
"loss": 0.0,
|
| 1700 |
+
"step": 17300
|
| 1701 |
+
},
|
| 1702 |
+
{
|
| 1703 |
+
"epoch": 32.89224952741021,
|
| 1704 |
+
"grad_norm": NaN,
|
| 1705 |
+
"learning_rate": 0.0003,
|
| 1706 |
+
"loss": 0.0,
|
| 1707 |
+
"step": 17400
|
| 1708 |
+
},
|
| 1709 |
+
{
|
| 1710 |
+
"epoch": 33.0,
|
| 1711 |
+
"eval_accuracy": 0.1995897435897436,
|
| 1712 |
+
"eval_loss": NaN,
|
| 1713 |
+
"eval_runtime": 5.5399,
|
| 1714 |
+
"eval_samples_per_second": 90.254,
|
| 1715 |
+
"eval_steps_per_second": 11.372,
|
| 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": NaN,
|
| 1727 |
+
"learning_rate": 0.0003,
|
| 1728 |
+
"loss": 0.0,
|
| 1729 |
+
"step": 17500
|
| 1730 |
+
},
|
| 1731 |
+
{
|
| 1732 |
+
"epoch": 33.2703213610586,
|
| 1733 |
+
"grad_norm": NaN,
|
| 1734 |
+
"learning_rate": 0.0003,
|
| 1735 |
+
"loss": 0.0,
|
| 1736 |
+
"step": 17600
|
| 1737 |
+
},
|
| 1738 |
+
{
|
| 1739 |
+
"epoch": 33.459357277882795,
|
| 1740 |
+
"grad_norm": NaN,
|
| 1741 |
+
"learning_rate": 0.0003,
|
| 1742 |
+
"loss": 0.0,
|
| 1743 |
+
"step": 17700
|
| 1744 |
+
},
|
| 1745 |
+
{
|
| 1746 |
+
"epoch": 33.648393194707,
|
| 1747 |
+
"grad_norm": NaN,
|
| 1748 |
+
"learning_rate": 0.0003,
|
| 1749 |
+
"loss": 0.0,
|
| 1750 |
+
"step": 17800
|
| 1751 |
+
},
|
| 1752 |
+
{
|
| 1753 |
+
"epoch": 33.83742911153119,
|
| 1754 |
+
"grad_norm": NaN,
|
| 1755 |
+
"learning_rate": 0.0003,
|
| 1756 |
+
"loss": 0.0,
|
| 1757 |
+
"step": 17900
|
| 1758 |
+
},
|
| 1759 |
+
{
|
| 1760 |
+
"epoch": 34.0,
|
| 1761 |
+
"eval_accuracy": 0.1995897435897436,
|
| 1762 |
+
"eval_loss": NaN,
|
| 1763 |
+
"eval_runtime": 5.2928,
|
| 1764 |
+
"eval_samples_per_second": 94.468,
|
| 1765 |
+
"eval_steps_per_second": 11.903,
|
| 1766 |
+
"step": 17986
|
| 1767 |
+
},
|
| 1768 |
+
{
|
| 1769 |
+
"epoch": 34.0,
|
| 1770 |
+
"eval_exact_match": 0.0,
|
| 1771 |
+
"eval_f1": 0.0,
|
| 1772 |
+
"step": 17986
|
| 1773 |
+
},
|
| 1774 |
+
{
|
| 1775 |
+
"epoch": 34.02646502835539,
|
| 1776 |
+
"grad_norm": NaN,
|
| 1777 |
+
"learning_rate": 0.0003,
|
| 1778 |
+
"loss": 0.0,
|
| 1779 |
+
"step": 18000
|
| 1780 |
+
},
|
| 1781 |
+
{
|
| 1782 |
+
"epoch": 34.215500945179585,
|
| 1783 |
+
"grad_norm": NaN,
|
| 1784 |
+
"learning_rate": 0.0003,
|
| 1785 |
+
"loss": 0.0,
|
| 1786 |
+
"step": 18100
|
| 1787 |
+
},
|
| 1788 |
+
{
|
| 1789 |
+
"epoch": 34.40453686200378,
|
| 1790 |
+
"grad_norm": NaN,
|
| 1791 |
+
"learning_rate": 0.0003,
|
| 1792 |
+
"loss": 0.0,
|
| 1793 |
+
"step": 18200
|
| 1794 |
+
},
|
| 1795 |
+
{
|
| 1796 |
+
"epoch": 34.593572778827976,
|
| 1797 |
+
"grad_norm": NaN,
|
| 1798 |
+
"learning_rate": 0.0003,
|
| 1799 |
+
"loss": 0.0,
|
| 1800 |
+
"step": 18300
|
| 1801 |
+
},
|
| 1802 |
+
{
|
| 1803 |
+
"epoch": 34.78260869565217,
|
| 1804 |
+
"grad_norm": NaN,
|
| 1805 |
+
"learning_rate": 0.0003,
|
| 1806 |
+
"loss": 0.0,
|
| 1807 |
+
"step": 18400
|
| 1808 |
+
},
|
| 1809 |
+
{
|
| 1810 |
+
"epoch": 34.97164461247637,
|
| 1811 |
+
"grad_norm": NaN,
|
| 1812 |
+
"learning_rate": 0.0003,
|
| 1813 |
+
"loss": 0.0,
|
| 1814 |
+
"step": 18500
|
| 1815 |
+
},
|
| 1816 |
+
{
|
| 1817 |
+
"epoch": 35.0,
|
| 1818 |
+
"eval_accuracy": 0.1995897435897436,
|
| 1819 |
+
"eval_loss": NaN,
|
| 1820 |
+
"eval_runtime": 5.7658,
|
| 1821 |
+
"eval_samples_per_second": 86.718,
|
| 1822 |
+
"eval_steps_per_second": 10.926,
|
| 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": NaN,
|
| 1834 |
+
"learning_rate": 0.0003,
|
| 1835 |
+
"loss": 0.0,
|
| 1836 |
+
"step": 18600
|
| 1837 |
+
},
|
| 1838 |
+
{
|
| 1839 |
+
"epoch": 35.349716446124766,
|
| 1840 |
+
"grad_norm": NaN,
|
| 1841 |
+
"learning_rate": 0.0003,
|
| 1842 |
+
"loss": 0.0,
|
| 1843 |
+
"step": 18700
|
| 1844 |
+
},
|
| 1845 |
+
{
|
| 1846 |
+
"epoch": 35.53875236294896,
|
| 1847 |
+
"grad_norm": NaN,
|
| 1848 |
+
"learning_rate": 0.0003,
|
| 1849 |
+
"loss": 0.0,
|
| 1850 |
+
"step": 18800
|
| 1851 |
+
},
|
| 1852 |
+
{
|
| 1853 |
+
"epoch": 35.72778827977316,
|
| 1854 |
+
"grad_norm": NaN,
|
| 1855 |
+
"learning_rate": 0.0003,
|
| 1856 |
+
"loss": 0.0,
|
| 1857 |
+
"step": 18900
|
| 1858 |
+
},
|
| 1859 |
+
{
|
| 1860 |
+
"epoch": 35.916824196597354,
|
| 1861 |
+
"grad_norm": NaN,
|
| 1862 |
+
"learning_rate": 0.0003,
|
| 1863 |
+
"loss": 0.0,
|
| 1864 |
+
"step": 19000
|
| 1865 |
+
},
|
| 1866 |
+
{
|
| 1867 |
+
"epoch": 36.0,
|
| 1868 |
+
"eval_accuracy": 0.1995897435897436,
|
| 1869 |
+
"eval_loss": NaN,
|
| 1870 |
+
"eval_runtime": 5.7508,
|
| 1871 |
+
"eval_samples_per_second": 86.945,
|
| 1872 |
+
"eval_steps_per_second": 10.955,
|
| 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": NaN,
|
| 1884 |
+
"learning_rate": 0.0003,
|
| 1885 |
+
"loss": 0.0,
|
| 1886 |
+
"step": 19100
|
| 1887 |
+
},
|
| 1888 |
+
{
|
| 1889 |
+
"epoch": 36.294896030245745,
|
| 1890 |
+
"grad_norm": NaN,
|
| 1891 |
+
"learning_rate": 0.0003,
|
| 1892 |
+
"loss": 0.0,
|
| 1893 |
+
"step": 19200
|
| 1894 |
+
},
|
| 1895 |
+
{
|
| 1896 |
+
"epoch": 36.48393194706994,
|
| 1897 |
+
"grad_norm": NaN,
|
| 1898 |
+
"learning_rate": 0.0003,
|
| 1899 |
+
"loss": 0.0,
|
| 1900 |
+
"step": 19300
|
| 1901 |
+
},
|
| 1902 |
+
{
|
| 1903 |
+
"epoch": 36.67296786389414,
|
| 1904 |
+
"grad_norm": NaN,
|
| 1905 |
+
"learning_rate": 0.0003,
|
| 1906 |
+
"loss": 0.0,
|
| 1907 |
+
"step": 19400
|
| 1908 |
+
},
|
| 1909 |
+
{
|
| 1910 |
+
"epoch": 36.86200378071834,
|
| 1911 |
+
"grad_norm": NaN,
|
| 1912 |
+
"learning_rate": 0.0003,
|
| 1913 |
+
"loss": 0.0,
|
| 1914 |
+
"step": 19500
|
| 1915 |
+
},
|
| 1916 |
+
{
|
| 1917 |
+
"epoch": 37.0,
|
| 1918 |
+
"eval_accuracy": 0.1995897435897436,
|
| 1919 |
+
"eval_loss": NaN,
|
| 1920 |
+
"eval_runtime": 6.1328,
|
| 1921 |
+
"eval_samples_per_second": 81.529,
|
| 1922 |
+
"eval_steps_per_second": 10.273,
|
| 1923 |
+
"step": 19573
|
| 1924 |
+
},
|
| 1925 |
+
{
|
| 1926 |
+
"epoch": 37.0,
|
| 1927 |
+
"eval_exact_match": 0.0,
|
| 1928 |
+
"eval_f1": 0.0,
|
| 1929 |
+
"step": 19573
|
| 1930 |
+
},
|
| 1931 |
+
{
|
| 1932 |
+
"epoch": 37.051039697542535,
|
| 1933 |
+
"grad_norm": NaN,
|
| 1934 |
+
"learning_rate": 0.0003,
|
| 1935 |
+
"loss": 0.0,
|
| 1936 |
+
"step": 19600
|
| 1937 |
+
},
|
| 1938 |
+
{
|
| 1939 |
+
"epoch": 37.24007561436673,
|
| 1940 |
+
"grad_norm": NaN,
|
| 1941 |
+
"learning_rate": 0.0003,
|
| 1942 |
+
"loss": 0.0,
|
| 1943 |
+
"step": 19700
|
| 1944 |
+
},
|
| 1945 |
+
{
|
| 1946 |
+
"epoch": 37.42911153119093,
|
| 1947 |
+
"grad_norm": NaN,
|
| 1948 |
+
"learning_rate": 0.0003,
|
| 1949 |
+
"loss": 0.0,
|
| 1950 |
+
"step": 19800
|
| 1951 |
+
},
|
| 1952 |
+
{
|
| 1953 |
+
"epoch": 37.61814744801512,
|
| 1954 |
+
"grad_norm": NaN,
|
| 1955 |
+
"learning_rate": 0.0003,
|
| 1956 |
+
"loss": 0.0,
|
| 1957 |
+
"step": 19900
|
| 1958 |
+
},
|
| 1959 |
+
{
|
| 1960 |
+
"epoch": 37.80718336483932,
|
| 1961 |
+
"grad_norm": NaN,
|
| 1962 |
+
"learning_rate": 0.0003,
|
| 1963 |
+
"loss": 0.0,
|
| 1964 |
+
"step": 20000
|
| 1965 |
+
},
|
| 1966 |
+
{
|
| 1967 |
+
"epoch": 37.996219281663514,
|
| 1968 |
+
"grad_norm": NaN,
|
| 1969 |
+
"learning_rate": 0.0003,
|
| 1970 |
+
"loss": 0.0,
|
| 1971 |
+
"step": 20100
|
| 1972 |
+
},
|
| 1973 |
+
{
|
| 1974 |
+
"epoch": 38.0,
|
| 1975 |
+
"eval_accuracy": 0.1995897435897436,
|
| 1976 |
+
"eval_loss": NaN,
|
| 1977 |
+
"eval_runtime": 4.9755,
|
| 1978 |
+
"eval_samples_per_second": 100.493,
|
| 1979 |
+
"eval_steps_per_second": 12.662,
|
| 1980 |
+
"step": 20102
|
| 1981 |
+
},
|
| 1982 |
+
{
|
| 1983 |
+
"epoch": 38.0,
|
| 1984 |
+
"eval_exact_match": 0.0,
|
| 1985 |
+
"eval_f1": 0.0,
|
| 1986 |
+
"step": 20102
|
| 1987 |
+
},
|
| 1988 |
+
{
|
| 1989 |
+
"epoch": 38.18525519848771,
|
| 1990 |
+
"grad_norm": NaN,
|
| 1991 |
+
"learning_rate": 0.0003,
|
| 1992 |
+
"loss": 0.0,
|
| 1993 |
+
"step": 20200
|
| 1994 |
+
},
|
| 1995 |
+
{
|
| 1996 |
+
"epoch": 38.37429111531191,
|
| 1997 |
+
"grad_norm": NaN,
|
| 1998 |
+
"learning_rate": 0.0003,
|
| 1999 |
+
"loss": 0.0,
|
| 2000 |
+
"step": 20300
|
| 2001 |
+
},
|
| 2002 |
+
{
|
| 2003 |
+
"epoch": 38.56332703213611,
|
| 2004 |
+
"grad_norm": NaN,
|
| 2005 |
+
"learning_rate": 0.0003,
|
| 2006 |
+
"loss": 0.0,
|
| 2007 |
+
"step": 20400
|
| 2008 |
+
},
|
| 2009 |
+
{
|
| 2010 |
+
"epoch": 38.752362948960304,
|
| 2011 |
+
"grad_norm": NaN,
|
| 2012 |
+
"learning_rate": 0.0003,
|
| 2013 |
+
"loss": 0.0,
|
| 2014 |
+
"step": 20500
|
| 2015 |
+
},
|
| 2016 |
+
{
|
| 2017 |
+
"epoch": 38.9413988657845,
|
| 2018 |
+
"grad_norm": NaN,
|
| 2019 |
+
"learning_rate": 0.0003,
|
| 2020 |
+
"loss": 0.0,
|
| 2021 |
+
"step": 20600
|
| 2022 |
+
},
|
| 2023 |
+
{
|
| 2024 |
+
"epoch": 39.0,
|
| 2025 |
+
"eval_accuracy": 0.1995897435897436,
|
| 2026 |
+
"eval_loss": NaN,
|
| 2027 |
+
"eval_runtime": 4.9998,
|
| 2028 |
+
"eval_samples_per_second": 100.005,
|
| 2029 |
+
"eval_steps_per_second": 12.601,
|
| 2030 |
+
"step": 20631
|
| 2031 |
+
},
|
| 2032 |
+
{
|
| 2033 |
+
"epoch": 39.0,
|
| 2034 |
+
"eval_exact_match": 0.0,
|
| 2035 |
+
"eval_f1": 0.0,
|
| 2036 |
+
"step": 20631
|
| 2037 |
+
},
|
| 2038 |
+
{
|
| 2039 |
+
"epoch": 39.130434782608695,
|
| 2040 |
+
"grad_norm": NaN,
|
| 2041 |
+
"learning_rate": 0.0003,
|
| 2042 |
+
"loss": 0.0,
|
| 2043 |
+
"step": 20700
|
| 2044 |
+
},
|
| 2045 |
+
{
|
| 2046 |
+
"epoch": 39.31947069943289,
|
| 2047 |
+
"grad_norm": NaN,
|
| 2048 |
+
"learning_rate": 0.0003,
|
| 2049 |
+
"loss": 0.0,
|
| 2050 |
+
"step": 20800
|
| 2051 |
+
},
|
| 2052 |
+
{
|
| 2053 |
+
"epoch": 39.50850661625709,
|
| 2054 |
+
"grad_norm": NaN,
|
| 2055 |
+
"learning_rate": 0.0003,
|
| 2056 |
+
"loss": 0.0,
|
| 2057 |
+
"step": 20900
|
| 2058 |
+
},
|
| 2059 |
+
{
|
| 2060 |
+
"epoch": 39.69754253308128,
|
| 2061 |
+
"grad_norm": NaN,
|
| 2062 |
+
"learning_rate": 0.0003,
|
| 2063 |
+
"loss": 0.0,
|
| 2064 |
+
"step": 21000
|
| 2065 |
+
},
|
| 2066 |
+
{
|
| 2067 |
+
"epoch": 39.886578449905485,
|
| 2068 |
+
"grad_norm": NaN,
|
| 2069 |
+
"learning_rate": 0.0003,
|
| 2070 |
+
"loss": 0.0,
|
| 2071 |
+
"step": 21100
|
| 2072 |
+
},
|
| 2073 |
+
{
|
| 2074 |
+
"epoch": 40.0,
|
| 2075 |
+
"eval_accuracy": 0.1995897435897436,
|
| 2076 |
+
"eval_loss": NaN,
|
| 2077 |
+
"eval_runtime": 5.4234,
|
| 2078 |
+
"eval_samples_per_second": 92.192,
|
| 2079 |
+
"eval_steps_per_second": 11.616,
|
| 2080 |
+
"step": 21160
|
| 2081 |
+
},
|
| 2082 |
+
{
|
| 2083 |
+
"epoch": 40.0,
|
| 2084 |
+
"eval_exact_match": 0.0,
|
| 2085 |
+
"eval_f1": 0.0,
|
| 2086 |
+
"step": 21160
|
| 2087 |
+
},
|
| 2088 |
+
{
|
| 2089 |
+
"epoch": 40.07561436672968,
|
| 2090 |
+
"grad_norm": NaN,
|
| 2091 |
+
"learning_rate": 0.0003,
|
| 2092 |
+
"loss": 0.0,
|
| 2093 |
+
"step": 21200
|
| 2094 |
+
},
|
| 2095 |
+
{
|
| 2096 |
+
"epoch": 40.26465028355388,
|
| 2097 |
+
"grad_norm": NaN,
|
| 2098 |
+
"learning_rate": 0.0003,
|
| 2099 |
+
"loss": 0.0,
|
| 2100 |
+
"step": 21300
|
| 2101 |
+
},
|
| 2102 |
+
{
|
| 2103 |
+
"epoch": 40.45368620037807,
|
| 2104 |
+
"grad_norm": NaN,
|
| 2105 |
+
"learning_rate": 0.0003,
|
| 2106 |
+
"loss": 0.0,
|
| 2107 |
+
"step": 21400
|
| 2108 |
+
},
|
| 2109 |
+
{
|
| 2110 |
+
"epoch": 40.64272211720227,
|
| 2111 |
+
"grad_norm": NaN,
|
| 2112 |
+
"learning_rate": 0.0003,
|
| 2113 |
+
"loss": 0.0,
|
| 2114 |
+
"step": 21500
|
| 2115 |
+
},
|
| 2116 |
+
{
|
| 2117 |
+
"epoch": 40.831758034026464,
|
| 2118 |
+
"grad_norm": NaN,
|
| 2119 |
+
"learning_rate": 0.0003,
|
| 2120 |
+
"loss": 0.0,
|
| 2121 |
+
"step": 21600
|
| 2122 |
+
},
|
| 2123 |
+
{
|
| 2124 |
+
"epoch": 41.0,
|
| 2125 |
+
"eval_accuracy": 0.1995897435897436,
|
| 2126 |
+
"eval_loss": NaN,
|
| 2127 |
+
"eval_runtime": 4.9807,
|
| 2128 |
+
"eval_samples_per_second": 100.388,
|
| 2129 |
+
"eval_steps_per_second": 12.649,
|
| 2130 |
+
"step": 21689
|
| 2131 |
+
},
|
| 2132 |
+
{
|
| 2133 |
+
"epoch": 41.0,
|
| 2134 |
+
"eval_exact_match": 0.0,
|
| 2135 |
+
"eval_f1": 0.0,
|
| 2136 |
+
"step": 21689
|
| 2137 |
+
},
|
| 2138 |
+
{
|
| 2139 |
+
"epoch": 41.02079395085066,
|
| 2140 |
+
"grad_norm": NaN,
|
| 2141 |
+
"learning_rate": 0.0003,
|
| 2142 |
+
"loss": 0.0,
|
| 2143 |
+
"step": 21700
|
| 2144 |
+
},
|
| 2145 |
+
{
|
| 2146 |
+
"epoch": 41.209829867674856,
|
| 2147 |
+
"grad_norm": NaN,
|
| 2148 |
+
"learning_rate": 0.0003,
|
| 2149 |
+
"loss": 0.0,
|
| 2150 |
+
"step": 21800
|
| 2151 |
+
},
|
| 2152 |
+
{
|
| 2153 |
+
"epoch": 41.39886578449905,
|
| 2154 |
+
"grad_norm": NaN,
|
| 2155 |
+
"learning_rate": 0.0003,
|
| 2156 |
+
"loss": 0.0,
|
| 2157 |
+
"step": 21900
|
| 2158 |
+
},
|
| 2159 |
+
{
|
| 2160 |
+
"epoch": 41.587901701323254,
|
| 2161 |
+
"grad_norm": NaN,
|
| 2162 |
+
"learning_rate": 0.0003,
|
| 2163 |
+
"loss": 0.0,
|
| 2164 |
+
"step": 22000
|
| 2165 |
+
},
|
| 2166 |
+
{
|
| 2167 |
+
"epoch": 41.77693761814745,
|
| 2168 |
+
"grad_norm": NaN,
|
| 2169 |
+
"learning_rate": 0.0003,
|
| 2170 |
+
"loss": 0.0,
|
| 2171 |
+
"step": 22100
|
| 2172 |
+
},
|
| 2173 |
+
{
|
| 2174 |
+
"epoch": 41.965973534971646,
|
| 2175 |
+
"grad_norm": NaN,
|
| 2176 |
+
"learning_rate": 0.0003,
|
| 2177 |
+
"loss": 0.0,
|
| 2178 |
+
"step": 22200
|
| 2179 |
+
},
|
| 2180 |
+
{
|
| 2181 |
+
"epoch": 42.0,
|
| 2182 |
+
"eval_accuracy": 0.1995897435897436,
|
| 2183 |
+
"eval_loss": NaN,
|
| 2184 |
+
"eval_runtime": 5.4872,
|
| 2185 |
+
"eval_samples_per_second": 91.12,
|
| 2186 |
+
"eval_steps_per_second": 11.481,
|
| 2187 |
+
"step": 22218
|
| 2188 |
+
},
|
| 2189 |
+
{
|
| 2190 |
+
"epoch": 42.0,
|
| 2191 |
+
"eval_exact_match": 0.0,
|
| 2192 |
+
"eval_f1": 0.0,
|
| 2193 |
+
"step": 22218
|
| 2194 |
+
},
|
| 2195 |
+
{
|
| 2196 |
+
"epoch": 42.15500945179584,
|
| 2197 |
+
"grad_norm": NaN,
|
| 2198 |
+
"learning_rate": 0.0003,
|
| 2199 |
+
"loss": 0.0,
|
| 2200 |
+
"step": 22300
|
| 2201 |
+
},
|
| 2202 |
+
{
|
| 2203 |
+
"epoch": 42.34404536862004,
|
| 2204 |
+
"grad_norm": NaN,
|
| 2205 |
+
"learning_rate": 0.0003,
|
| 2206 |
+
"loss": 0.0,
|
| 2207 |
+
"step": 22400
|
| 2208 |
+
},
|
| 2209 |
+
{
|
| 2210 |
+
"epoch": 42.53308128544423,
|
| 2211 |
+
"grad_norm": NaN,
|
| 2212 |
+
"learning_rate": 0.0003,
|
| 2213 |
+
"loss": 0.0,
|
| 2214 |
+
"step": 22500
|
| 2215 |
+
},
|
| 2216 |
+
{
|
| 2217 |
+
"epoch": 42.72211720226843,
|
| 2218 |
+
"grad_norm": NaN,
|
| 2219 |
+
"learning_rate": 0.0003,
|
| 2220 |
+
"loss": 0.0,
|
| 2221 |
+
"step": 22600
|
| 2222 |
+
},
|
| 2223 |
+
{
|
| 2224 |
+
"epoch": 42.911153119092624,
|
| 2225 |
+
"grad_norm": NaN,
|
| 2226 |
+
"learning_rate": 0.0003,
|
| 2227 |
+
"loss": 0.0,
|
| 2228 |
+
"step": 22700
|
| 2229 |
+
},
|
| 2230 |
+
{
|
| 2231 |
+
"epoch": 43.0,
|
| 2232 |
+
"eval_accuracy": 0.1995897435897436,
|
| 2233 |
+
"eval_loss": NaN,
|
| 2234 |
+
"eval_runtime": 5.7071,
|
| 2235 |
+
"eval_samples_per_second": 87.61,
|
| 2236 |
+
"eval_steps_per_second": 11.039,
|
| 2237 |
+
"step": 22747
|
| 2238 |
+
},
|
| 2239 |
+
{
|
| 2240 |
+
"epoch": 43.0,
|
| 2241 |
+
"eval_exact_match": 0.0,
|
| 2242 |
+
"eval_f1": 0.0,
|
| 2243 |
+
"step": 22747
|
| 2244 |
+
},
|
| 2245 |
+
{
|
| 2246 |
+
"epoch": 43.10018903591683,
|
| 2247 |
+
"grad_norm": NaN,
|
| 2248 |
+
"learning_rate": 0.0003,
|
| 2249 |
+
"loss": 0.0,
|
| 2250 |
+
"step": 22800
|
| 2251 |
+
},
|
| 2252 |
+
{
|
| 2253 |
+
"epoch": 43.28922495274102,
|
| 2254 |
+
"grad_norm": NaN,
|
| 2255 |
+
"learning_rate": 0.0003,
|
| 2256 |
+
"loss": 0.0,
|
| 2257 |
+
"step": 22900
|
| 2258 |
+
},
|
| 2259 |
+
{
|
| 2260 |
+
"epoch": 43.47826086956522,
|
| 2261 |
+
"grad_norm": NaN,
|
| 2262 |
+
"learning_rate": 0.0003,
|
| 2263 |
+
"loss": 0.0,
|
| 2264 |
+
"step": 23000
|
| 2265 |
+
},
|
| 2266 |
+
{
|
| 2267 |
+
"epoch": 43.667296786389414,
|
| 2268 |
+
"grad_norm": NaN,
|
| 2269 |
+
"learning_rate": 0.0003,
|
| 2270 |
+
"loss": 0.0,
|
| 2271 |
+
"step": 23100
|
| 2272 |
+
},
|
| 2273 |
+
{
|
| 2274 |
+
"epoch": 43.85633270321361,
|
| 2275 |
+
"grad_norm": NaN,
|
| 2276 |
+
"learning_rate": 0.0003,
|
| 2277 |
+
"loss": 0.0,
|
| 2278 |
+
"step": 23200
|
| 2279 |
+
},
|
| 2280 |
+
{
|
| 2281 |
+
"epoch": 44.0,
|
| 2282 |
+
"eval_accuracy": 0.1995897435897436,
|
| 2283 |
+
"eval_loss": NaN,
|
| 2284 |
+
"eval_runtime": 5.3411,
|
| 2285 |
+
"eval_samples_per_second": 93.613,
|
| 2286 |
+
"eval_steps_per_second": 11.795,
|
| 2287 |
+
"step": 23276
|
| 2288 |
+
},
|
| 2289 |
+
{
|
| 2290 |
+
"epoch": 44.0,
|
| 2291 |
+
"eval_exact_match": 0.0,
|
| 2292 |
+
"eval_f1": 0.0,
|
| 2293 |
+
"step": 23276
|
| 2294 |
+
},
|
| 2295 |
+
{
|
| 2296 |
+
"epoch": 44.045368620037806,
|
| 2297 |
+
"grad_norm": NaN,
|
| 2298 |
+
"learning_rate": 0.0003,
|
| 2299 |
+
"loss": 0.0,
|
| 2300 |
+
"step": 23300
|
| 2301 |
+
},
|
| 2302 |
+
{
|
| 2303 |
+
"epoch": 44.234404536862,
|
| 2304 |
+
"grad_norm": NaN,
|
| 2305 |
+
"learning_rate": 0.0003,
|
| 2306 |
+
"loss": 0.0,
|
| 2307 |
+
"step": 23400
|
| 2308 |
+
},
|
| 2309 |
+
{
|
| 2310 |
+
"epoch": 44.4234404536862,
|
| 2311 |
+
"grad_norm": NaN,
|
| 2312 |
+
"learning_rate": 0.0003,
|
| 2313 |
+
"loss": 0.0,
|
| 2314 |
+
"step": 23500
|
| 2315 |
+
},
|
| 2316 |
+
{
|
| 2317 |
+
"epoch": 44.6124763705104,
|
| 2318 |
+
"grad_norm": NaN,
|
| 2319 |
+
"learning_rate": 0.0003,
|
| 2320 |
+
"loss": 0.0,
|
| 2321 |
+
"step": 23600
|
| 2322 |
+
},
|
| 2323 |
+
{
|
| 2324 |
+
"epoch": 44.801512287334596,
|
| 2325 |
+
"grad_norm": NaN,
|
| 2326 |
+
"learning_rate": 0.0003,
|
| 2327 |
+
"loss": 0.0,
|
| 2328 |
+
"step": 23700
|
| 2329 |
+
},
|
| 2330 |
+
{
|
| 2331 |
+
"epoch": 44.99054820415879,
|
| 2332 |
+
"grad_norm": NaN,
|
| 2333 |
+
"learning_rate": 0.0003,
|
| 2334 |
+
"loss": 0.0,
|
| 2335 |
+
"step": 23800
|
| 2336 |
+
},
|
| 2337 |
+
{
|
| 2338 |
+
"epoch": 45.0,
|
| 2339 |
+
"eval_accuracy": 0.1995897435897436,
|
| 2340 |
+
"eval_loss": NaN,
|
| 2341 |
+
"eval_runtime": 5.0286,
|
| 2342 |
+
"eval_samples_per_second": 99.431,
|
| 2343 |
+
"eval_steps_per_second": 12.528,
|
| 2344 |
+
"step": 23805
|
| 2345 |
+
},
|
| 2346 |
+
{
|
| 2347 |
+
"epoch": 45.0,
|
| 2348 |
+
"eval_exact_match": 0.0,
|
| 2349 |
+
"eval_f1": 0.0,
|
| 2350 |
+
"step": 23805
|
| 2351 |
+
}
|
| 2352 |
+
],
|
| 2353 |
+
"logging_steps": 100,
|
| 2354 |
+
"max_steps": 26450,
|
| 2355 |
+
"num_input_tokens_seen": 0,
|
| 2356 |
+
"num_train_epochs": 50,
|
| 2357 |
+
"save_steps": 500,
|
| 2358 |
+
"stateful_callbacks": {
|
| 2359 |
+
"TrainerControl": {
|
| 2360 |
+
"args": {
|
| 2361 |
+
"should_epoch_stop": false,
|
| 2362 |
+
"should_evaluate": false,
|
| 2363 |
+
"should_log": false,
|
| 2364 |
+
"should_save": true,
|
| 2365 |
+
"should_training_stop": false
|
| 2366 |
+
},
|
| 2367 |
+
"attributes": {}
|
| 2368 |
+
}
|
| 2369 |
+
},
|
| 2370 |
+
"total_flos": 1.936908220228436e+18,
|
| 2371 |
+
"train_batch_size": 2,
|
| 2372 |
+
"trial_name": null,
|
| 2373 |
+
"trial_params": null
|
| 2374 |
+
}
|
checkpoint-23805/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:186f3ae462881ca0bb3a6b5295a97a245e0a78172be9c14948a2d07a12aa4241
|
| 3 |
+
size 5304
|
checkpoint-23805/vocab.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|