Training in progress, epoch 49, checkpoint
Browse files- checkpoint-12250/README.md +9 -0
- checkpoint-12250/adapter_config.json +25 -0
- checkpoint-12250/adapter_model.safetensors +3 -0
- checkpoint-12250/added_tokens.json +5 -0
- checkpoint-12250/merges.txt +0 -0
- checkpoint-12250/optimizer.pt +3 -0
- checkpoint-12250/rng_state_0.pth +3 -0
- checkpoint-12250/rng_state_1.pth +3 -0
- checkpoint-12250/rng_state_2.pth +3 -0
- checkpoint-12250/rng_state_3.pth +3 -0
- checkpoint-12250/scheduler.pt +3 -0
- checkpoint-12250/special_tokens_map.json +14 -0
- checkpoint-12250/tokenizer.json +0 -0
- checkpoint-12250/tokenizer_config.json +43 -0
- checkpoint-12250/trainer_state.json +1622 -0
- checkpoint-12250/training_args.bin +3 -0
- checkpoint-12250/vocab.json +0 -0
checkpoint-12250/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-12250/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-12250/adapter_model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:c5d3bf79b7ce354f5b21be03419d1c232aebde47b4ce5101ec1c87e7065d003d
|
| 3 |
+
size 112130416
|
checkpoint-12250/added_tokens.json
ADDED
|
@@ -0,0 +1,5 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"<|endoftext|>": 151643,
|
| 3 |
+
"<|im_end|>": 151645,
|
| 4 |
+
"<|im_start|>": 151644
|
| 5 |
+
}
|
checkpoint-12250/merges.txt
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
checkpoint-12250/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:234a08c0d97bf9d511f156cdb31f0916304350555c67ea5d0e7cf35416e2e06e
|
| 3 |
+
size 224537202
|
checkpoint-12250/rng_state_0.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:47ed3999d8423a72d96d9c1449233357bc568df81857e4dfc5a9e0797e678979
|
| 3 |
+
size 15024
|
checkpoint-12250/rng_state_1.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:1d9a20c32bd1292bb72f172554c238d58bf26ce8a1d82c9a46a52d31a5ee9544
|
| 3 |
+
size 15024
|
checkpoint-12250/rng_state_2.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:9bc40f8852a1b08848493043dc470e735f761df43fc4853eee6544428ce36704
|
| 3 |
+
size 15024
|
checkpoint-12250/rng_state_3.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:9113d010fc7d15b2849d37a349768515aa8767542aa1930873c51277855e0f81
|
| 3 |
+
size 15024
|
checkpoint-12250/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:1d31e4721197cf5f86f5d999991b438ca2b09d4b5792aab0abff6ae3b0eba634
|
| 3 |
+
size 1064
|
checkpoint-12250/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-12250/tokenizer.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
checkpoint-12250/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-12250/trainer_state.json
ADDED
|
@@ -0,0 +1,1622 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 49.0,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 12250,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.4,
|
| 13 |
+
"grad_norm": 0.48472875356674194,
|
| 14 |
+
"learning_rate": 0.0003,
|
| 15 |
+
"loss": 2.3219,
|
| 16 |
+
"step": 100
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.8,
|
| 20 |
+
"grad_norm": 0.4583953619003296,
|
| 21 |
+
"learning_rate": 0.0003,
|
| 22 |
+
"loss": 2.2487,
|
| 23 |
+
"step": 200
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 1.0,
|
| 27 |
+
"eval_accuracy": 0.5168253968253969,
|
| 28 |
+
"eval_loss": 2.311781167984009,
|
| 29 |
+
"eval_runtime": 5.4425,
|
| 30 |
+
"eval_samples_per_second": 91.87,
|
| 31 |
+
"eval_steps_per_second": 11.576,
|
| 32 |
+
"step": 250
|
| 33 |
+
},
|
| 34 |
+
{
|
| 35 |
+
"epoch": 1.0,
|
| 36 |
+
"eval_exact_match": 18.6,
|
| 37 |
+
"eval_f1": 26.999121042356347,
|
| 38 |
+
"step": 250
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 1.2,
|
| 42 |
+
"grad_norm": 0.6610094904899597,
|
| 43 |
+
"learning_rate": 0.0003,
|
| 44 |
+
"loss": 2.083,
|
| 45 |
+
"step": 300
|
| 46 |
+
},
|
| 47 |
+
{
|
| 48 |
+
"epoch": 1.6,
|
| 49 |
+
"grad_norm": 0.7361629605293274,
|
| 50 |
+
"learning_rate": 0.0003,
|
| 51 |
+
"loss": 1.9274,
|
| 52 |
+
"step": 400
|
| 53 |
+
},
|
| 54 |
+
{
|
| 55 |
+
"epoch": 2.0,
|
| 56 |
+
"grad_norm": 0.6793610453605652,
|
| 57 |
+
"learning_rate": 0.0003,
|
| 58 |
+
"loss": 1.9113,
|
| 59 |
+
"step": 500
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 2.0,
|
| 63 |
+
"eval_accuracy": 0.5153650793650794,
|
| 64 |
+
"eval_loss": 2.3749232292175293,
|
| 65 |
+
"eval_runtime": 5.2167,
|
| 66 |
+
"eval_samples_per_second": 95.846,
|
| 67 |
+
"eval_steps_per_second": 12.077,
|
| 68 |
+
"step": 500
|
| 69 |
+
},
|
| 70 |
+
{
|
| 71 |
+
"epoch": 2.0,
|
| 72 |
+
"eval_exact_match": 18.2,
|
| 73 |
+
"eval_f1": 26.209704184704197,
|
| 74 |
+
"step": 500
|
| 75 |
+
},
|
| 76 |
+
{
|
| 77 |
+
"epoch": 2.4,
|
| 78 |
+
"grad_norm": 0.8843489289283752,
|
| 79 |
+
"learning_rate": 0.0003,
|
| 80 |
+
"loss": 1.4602,
|
| 81 |
+
"step": 600
|
| 82 |
+
},
|
| 83 |
+
{
|
| 84 |
+
"epoch": 2.8,
|
| 85 |
+
"grad_norm": 0.8380613923072815,
|
| 86 |
+
"learning_rate": 0.0003,
|
| 87 |
+
"loss": 1.4924,
|
| 88 |
+
"step": 700
|
| 89 |
+
},
|
| 90 |
+
{
|
| 91 |
+
"epoch": 3.0,
|
| 92 |
+
"eval_accuracy": 0.5098730158730159,
|
| 93 |
+
"eval_loss": 2.5289740562438965,
|
| 94 |
+
"eval_runtime": 5.3631,
|
| 95 |
+
"eval_samples_per_second": 93.23,
|
| 96 |
+
"eval_steps_per_second": 11.747,
|
| 97 |
+
"step": 750
|
| 98 |
+
},
|
| 99 |
+
{
|
| 100 |
+
"epoch": 3.0,
|
| 101 |
+
"eval_exact_match": 16.6,
|
| 102 |
+
"eval_f1": 24.526153291153296,
|
| 103 |
+
"step": 750
|
| 104 |
+
},
|
| 105 |
+
{
|
| 106 |
+
"epoch": 3.2,
|
| 107 |
+
"grad_norm": 1.068247675895691,
|
| 108 |
+
"learning_rate": 0.0003,
|
| 109 |
+
"loss": 1.2869,
|
| 110 |
+
"step": 800
|
| 111 |
+
},
|
| 112 |
+
{
|
| 113 |
+
"epoch": 3.6,
|
| 114 |
+
"grad_norm": 1.041741967201233,
|
| 115 |
+
"learning_rate": 0.0003,
|
| 116 |
+
"loss": 1.0749,
|
| 117 |
+
"step": 900
|
| 118 |
+
},
|
| 119 |
+
{
|
| 120 |
+
"epoch": 4.0,
|
| 121 |
+
"grad_norm": 1.267968773841858,
|
| 122 |
+
"learning_rate": 0.0003,
|
| 123 |
+
"loss": 1.1157,
|
| 124 |
+
"step": 1000
|
| 125 |
+
},
|
| 126 |
+
{
|
| 127 |
+
"epoch": 4.0,
|
| 128 |
+
"eval_accuracy": 0.5038412698412699,
|
| 129 |
+
"eval_loss": 2.794942855834961,
|
| 130 |
+
"eval_runtime": 5.0088,
|
| 131 |
+
"eval_samples_per_second": 99.825,
|
| 132 |
+
"eval_steps_per_second": 12.578,
|
| 133 |
+
"step": 1000
|
| 134 |
+
},
|
| 135 |
+
{
|
| 136 |
+
"epoch": 4.0,
|
| 137 |
+
"eval_exact_match": 16.4,
|
| 138 |
+
"eval_f1": 23.978561993561993,
|
| 139 |
+
"step": 1000
|
| 140 |
+
},
|
| 141 |
+
{
|
| 142 |
+
"epoch": 4.4,
|
| 143 |
+
"grad_norm": 1.1971005201339722,
|
| 144 |
+
"learning_rate": 0.0003,
|
| 145 |
+
"loss": 0.7367,
|
| 146 |
+
"step": 1100
|
| 147 |
+
},
|
| 148 |
+
{
|
| 149 |
+
"epoch": 4.8,
|
| 150 |
+
"grad_norm": 1.2383127212524414,
|
| 151 |
+
"learning_rate": 0.0003,
|
| 152 |
+
"loss": 0.7814,
|
| 153 |
+
"step": 1200
|
| 154 |
+
},
|
| 155 |
+
{
|
| 156 |
+
"epoch": 5.0,
|
| 157 |
+
"eval_accuracy": 0.49853968253968256,
|
| 158 |
+
"eval_loss": 3.0137181282043457,
|
| 159 |
+
"eval_runtime": 5.1368,
|
| 160 |
+
"eval_samples_per_second": 97.337,
|
| 161 |
+
"eval_steps_per_second": 12.264,
|
| 162 |
+
"step": 1250
|
| 163 |
+
},
|
| 164 |
+
{
|
| 165 |
+
"epoch": 5.0,
|
| 166 |
+
"eval_exact_match": 14.8,
|
| 167 |
+
"eval_f1": 21.880294705294705,
|
| 168 |
+
"step": 1250
|
| 169 |
+
},
|
| 170 |
+
{
|
| 171 |
+
"epoch": 5.2,
|
| 172 |
+
"grad_norm": 0.9558005928993225,
|
| 173 |
+
"learning_rate": 0.0003,
|
| 174 |
+
"loss": 0.6786,
|
| 175 |
+
"step": 1300
|
| 176 |
+
},
|
| 177 |
+
{
|
| 178 |
+
"epoch": 5.6,
|
| 179 |
+
"grad_norm": 1.1341361999511719,
|
| 180 |
+
"learning_rate": 0.0003,
|
| 181 |
+
"loss": 0.5741,
|
| 182 |
+
"step": 1400
|
| 183 |
+
},
|
| 184 |
+
{
|
| 185 |
+
"epoch": 6.0,
|
| 186 |
+
"grad_norm": 1.1375033855438232,
|
| 187 |
+
"learning_rate": 0.0003,
|
| 188 |
+
"loss": 0.605,
|
| 189 |
+
"step": 1500
|
| 190 |
+
},
|
| 191 |
+
{
|
| 192 |
+
"epoch": 6.0,
|
| 193 |
+
"eval_accuracy": 0.49657142857142855,
|
| 194 |
+
"eval_loss": 3.220507860183716,
|
| 195 |
+
"eval_runtime": 5.2708,
|
| 196 |
+
"eval_samples_per_second": 94.863,
|
| 197 |
+
"eval_steps_per_second": 11.953,
|
| 198 |
+
"step": 1500
|
| 199 |
+
},
|
| 200 |
+
{
|
| 201 |
+
"epoch": 6.0,
|
| 202 |
+
"eval_exact_match": 15.4,
|
| 203 |
+
"eval_f1": 22.281807987334307,
|
| 204 |
+
"step": 1500
|
| 205 |
+
},
|
| 206 |
+
{
|
| 207 |
+
"epoch": 6.4,
|
| 208 |
+
"grad_norm": 1.118528962135315,
|
| 209 |
+
"learning_rate": 0.0003,
|
| 210 |
+
"loss": 0.4684,
|
| 211 |
+
"step": 1600
|
| 212 |
+
},
|
| 213 |
+
{
|
| 214 |
+
"epoch": 6.8,
|
| 215 |
+
"grad_norm": 0.9311291575431824,
|
| 216 |
+
"learning_rate": 0.0003,
|
| 217 |
+
"loss": 0.4968,
|
| 218 |
+
"step": 1700
|
| 219 |
+
},
|
| 220 |
+
{
|
| 221 |
+
"epoch": 7.0,
|
| 222 |
+
"eval_accuracy": 0.4946984126984127,
|
| 223 |
+
"eval_loss": 3.3760199546813965,
|
| 224 |
+
"eval_runtime": 5.0734,
|
| 225 |
+
"eval_samples_per_second": 98.554,
|
| 226 |
+
"eval_steps_per_second": 12.418,
|
| 227 |
+
"step": 1750
|
| 228 |
+
},
|
| 229 |
+
{
|
| 230 |
+
"epoch": 7.0,
|
| 231 |
+
"eval_exact_match": 14.8,
|
| 232 |
+
"eval_f1": 21.79480811586075,
|
| 233 |
+
"step": 1750
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 7.2,
|
| 237 |
+
"grad_norm": 1.0299904346466064,
|
| 238 |
+
"learning_rate": 0.0003,
|
| 239 |
+
"loss": 0.4631,
|
| 240 |
+
"step": 1800
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 7.6,
|
| 244 |
+
"grad_norm": 0.7518107295036316,
|
| 245 |
+
"learning_rate": 0.0003,
|
| 246 |
+
"loss": 0.4411,
|
| 247 |
+
"step": 1900
|
| 248 |
+
},
|
| 249 |
+
{
|
| 250 |
+
"epoch": 8.0,
|
| 251 |
+
"grad_norm": 1.1140515804290771,
|
| 252 |
+
"learning_rate": 0.0003,
|
| 253 |
+
"loss": 0.4597,
|
| 254 |
+
"step": 2000
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 8.0,
|
| 258 |
+
"eval_accuracy": 0.49234920634920637,
|
| 259 |
+
"eval_loss": 3.4634218215942383,
|
| 260 |
+
"eval_runtime": 5.8652,
|
| 261 |
+
"eval_samples_per_second": 85.249,
|
| 262 |
+
"eval_steps_per_second": 10.741,
|
| 263 |
+
"step": 2000
|
| 264 |
+
},
|
| 265 |
+
{
|
| 266 |
+
"epoch": 8.0,
|
| 267 |
+
"eval_exact_match": 15.0,
|
| 268 |
+
"eval_f1": 21.617115196631318,
|
| 269 |
+
"step": 2000
|
| 270 |
+
},
|
| 271 |
+
{
|
| 272 |
+
"epoch": 8.4,
|
| 273 |
+
"grad_norm": 1.2363754510879517,
|
| 274 |
+
"learning_rate": 0.0003,
|
| 275 |
+
"loss": 0.4026,
|
| 276 |
+
"step": 2100
|
| 277 |
+
},
|
| 278 |
+
{
|
| 279 |
+
"epoch": 8.8,
|
| 280 |
+
"grad_norm": 0.8174892067909241,
|
| 281 |
+
"learning_rate": 0.0003,
|
| 282 |
+
"loss": 0.4242,
|
| 283 |
+
"step": 2200
|
| 284 |
+
},
|
| 285 |
+
{
|
| 286 |
+
"epoch": 9.0,
|
| 287 |
+
"eval_accuracy": 0.4955873015873016,
|
| 288 |
+
"eval_loss": 3.4379212856292725,
|
| 289 |
+
"eval_runtime": 5.214,
|
| 290 |
+
"eval_samples_per_second": 95.896,
|
| 291 |
+
"eval_steps_per_second": 12.083,
|
| 292 |
+
"step": 2250
|
| 293 |
+
},
|
| 294 |
+
{
|
| 295 |
+
"epoch": 9.0,
|
| 296 |
+
"eval_exact_match": 15.0,
|
| 297 |
+
"eval_f1": 22.36426184926185,
|
| 298 |
+
"step": 2250
|
| 299 |
+
},
|
| 300 |
+
{
|
| 301 |
+
"epoch": 9.2,
|
| 302 |
+
"grad_norm": 0.551261842250824,
|
| 303 |
+
"learning_rate": 0.0003,
|
| 304 |
+
"loss": 0.4135,
|
| 305 |
+
"step": 2300
|
| 306 |
+
},
|
| 307 |
+
{
|
| 308 |
+
"epoch": 9.6,
|
| 309 |
+
"grad_norm": 0.6361530423164368,
|
| 310 |
+
"learning_rate": 0.0003,
|
| 311 |
+
"loss": 0.3999,
|
| 312 |
+
"step": 2400
|
| 313 |
+
},
|
| 314 |
+
{
|
| 315 |
+
"epoch": 10.0,
|
| 316 |
+
"grad_norm": 0.9359388947486877,
|
| 317 |
+
"learning_rate": 0.0003,
|
| 318 |
+
"loss": 0.4223,
|
| 319 |
+
"step": 2500
|
| 320 |
+
},
|
| 321 |
+
{
|
| 322 |
+
"epoch": 10.0,
|
| 323 |
+
"eval_accuracy": 0.494952380952381,
|
| 324 |
+
"eval_loss": 3.600064277648926,
|
| 325 |
+
"eval_runtime": 5.8832,
|
| 326 |
+
"eval_samples_per_second": 84.988,
|
| 327 |
+
"eval_steps_per_second": 10.709,
|
| 328 |
+
"step": 2500
|
| 329 |
+
},
|
| 330 |
+
{
|
| 331 |
+
"epoch": 10.0,
|
| 332 |
+
"eval_exact_match": 14.8,
|
| 333 |
+
"eval_f1": 21.784345099345103,
|
| 334 |
+
"step": 2500
|
| 335 |
+
},
|
| 336 |
+
{
|
| 337 |
+
"epoch": 10.4,
|
| 338 |
+
"grad_norm": 0.7834428548812866,
|
| 339 |
+
"learning_rate": 0.0003,
|
| 340 |
+
"loss": 0.3855,
|
| 341 |
+
"step": 2600
|
| 342 |
+
},
|
| 343 |
+
{
|
| 344 |
+
"epoch": 10.8,
|
| 345 |
+
"grad_norm": 1.0201836824417114,
|
| 346 |
+
"learning_rate": 0.0003,
|
| 347 |
+
"loss": 0.4,
|
| 348 |
+
"step": 2700
|
| 349 |
+
},
|
| 350 |
+
{
|
| 351 |
+
"epoch": 11.0,
|
| 352 |
+
"eval_accuracy": 0.4955555555555556,
|
| 353 |
+
"eval_loss": 3.571756362915039,
|
| 354 |
+
"eval_runtime": 5.6646,
|
| 355 |
+
"eval_samples_per_second": 88.267,
|
| 356 |
+
"eval_steps_per_second": 11.122,
|
| 357 |
+
"step": 2750
|
| 358 |
+
},
|
| 359 |
+
{
|
| 360 |
+
"epoch": 11.0,
|
| 361 |
+
"eval_exact_match": 15.0,
|
| 362 |
+
"eval_f1": 22.02530358530358,
|
| 363 |
+
"step": 2750
|
| 364 |
+
},
|
| 365 |
+
{
|
| 366 |
+
"epoch": 11.2,
|
| 367 |
+
"grad_norm": 0.738467812538147,
|
| 368 |
+
"learning_rate": 0.0003,
|
| 369 |
+
"loss": 0.3977,
|
| 370 |
+
"step": 2800
|
| 371 |
+
},
|
| 372 |
+
{
|
| 373 |
+
"epoch": 11.6,
|
| 374 |
+
"grad_norm": 0.535826563835144,
|
| 375 |
+
"learning_rate": 0.0003,
|
| 376 |
+
"loss": 0.3904,
|
| 377 |
+
"step": 2900
|
| 378 |
+
},
|
| 379 |
+
{
|
| 380 |
+
"epoch": 12.0,
|
| 381 |
+
"grad_norm": 0.8586340546607971,
|
| 382 |
+
"learning_rate": 0.0003,
|
| 383 |
+
"loss": 0.4007,
|
| 384 |
+
"step": 3000
|
| 385 |
+
},
|
| 386 |
+
{
|
| 387 |
+
"epoch": 12.0,
|
| 388 |
+
"eval_accuracy": 0.49323809523809525,
|
| 389 |
+
"eval_loss": 3.568419933319092,
|
| 390 |
+
"eval_runtime": 5.5173,
|
| 391 |
+
"eval_samples_per_second": 90.624,
|
| 392 |
+
"eval_steps_per_second": 11.419,
|
| 393 |
+
"step": 3000
|
| 394 |
+
},
|
| 395 |
+
{
|
| 396 |
+
"epoch": 12.0,
|
| 397 |
+
"eval_exact_match": 15.2,
|
| 398 |
+
"eval_f1": 21.94431457431457,
|
| 399 |
+
"step": 3000
|
| 400 |
+
},
|
| 401 |
+
{
|
| 402 |
+
"epoch": 12.4,
|
| 403 |
+
"grad_norm": 0.6024760007858276,
|
| 404 |
+
"learning_rate": 0.0003,
|
| 405 |
+
"loss": 0.3721,
|
| 406 |
+
"step": 3100
|
| 407 |
+
},
|
| 408 |
+
{
|
| 409 |
+
"epoch": 12.8,
|
| 410 |
+
"grad_norm": 0.7764168381690979,
|
| 411 |
+
"learning_rate": 0.0003,
|
| 412 |
+
"loss": 0.3929,
|
| 413 |
+
"step": 3200
|
| 414 |
+
},
|
| 415 |
+
{
|
| 416 |
+
"epoch": 13.0,
|
| 417 |
+
"eval_accuracy": 0.49314285714285716,
|
| 418 |
+
"eval_loss": 3.602919816970825,
|
| 419 |
+
"eval_runtime": 5.1504,
|
| 420 |
+
"eval_samples_per_second": 97.08,
|
| 421 |
+
"eval_steps_per_second": 12.232,
|
| 422 |
+
"step": 3250
|
| 423 |
+
},
|
| 424 |
+
{
|
| 425 |
+
"epoch": 13.0,
|
| 426 |
+
"eval_exact_match": 13.8,
|
| 427 |
+
"eval_f1": 20.855486735486743,
|
| 428 |
+
"step": 3250
|
| 429 |
+
},
|
| 430 |
+
{
|
| 431 |
+
"epoch": 13.2,
|
| 432 |
+
"grad_norm": 0.668953537940979,
|
| 433 |
+
"learning_rate": 0.0003,
|
| 434 |
+
"loss": 0.3886,
|
| 435 |
+
"step": 3300
|
| 436 |
+
},
|
| 437 |
+
{
|
| 438 |
+
"epoch": 13.6,
|
| 439 |
+
"grad_norm": 0.7968106865882874,
|
| 440 |
+
"learning_rate": 0.0003,
|
| 441 |
+
"loss": 0.3845,
|
| 442 |
+
"step": 3400
|
| 443 |
+
},
|
| 444 |
+
{
|
| 445 |
+
"epoch": 14.0,
|
| 446 |
+
"grad_norm": 0.7598584294319153,
|
| 447 |
+
"learning_rate": 0.0003,
|
| 448 |
+
"loss": 0.4003,
|
| 449 |
+
"step": 3500
|
| 450 |
+
},
|
| 451 |
+
{
|
| 452 |
+
"epoch": 14.0,
|
| 453 |
+
"eval_accuracy": 0.49206349206349204,
|
| 454 |
+
"eval_loss": 3.584076404571533,
|
| 455 |
+
"eval_runtime": 5.5138,
|
| 456 |
+
"eval_samples_per_second": 90.682,
|
| 457 |
+
"eval_steps_per_second": 11.426,
|
| 458 |
+
"step": 3500
|
| 459 |
+
},
|
| 460 |
+
{
|
| 461 |
+
"epoch": 14.0,
|
| 462 |
+
"eval_exact_match": 15.6,
|
| 463 |
+
"eval_f1": 22.729776889776886,
|
| 464 |
+
"step": 3500
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 14.4,
|
| 468 |
+
"grad_norm": 0.7232977151870728,
|
| 469 |
+
"learning_rate": 0.0003,
|
| 470 |
+
"loss": 0.3713,
|
| 471 |
+
"step": 3600
|
| 472 |
+
},
|
| 473 |
+
{
|
| 474 |
+
"epoch": 14.8,
|
| 475 |
+
"grad_norm": 0.6823480725288391,
|
| 476 |
+
"learning_rate": 0.0003,
|
| 477 |
+
"loss": 0.3834,
|
| 478 |
+
"step": 3700
|
| 479 |
+
},
|
| 480 |
+
{
|
| 481 |
+
"epoch": 15.0,
|
| 482 |
+
"eval_accuracy": 0.4925079365079365,
|
| 483 |
+
"eval_loss": 3.6552798748016357,
|
| 484 |
+
"eval_runtime": 5.0332,
|
| 485 |
+
"eval_samples_per_second": 99.341,
|
| 486 |
+
"eval_steps_per_second": 12.517,
|
| 487 |
+
"step": 3750
|
| 488 |
+
},
|
| 489 |
+
{
|
| 490 |
+
"epoch": 15.0,
|
| 491 |
+
"eval_exact_match": 13.8,
|
| 492 |
+
"eval_f1": 20.994360084360082,
|
| 493 |
+
"step": 3750
|
| 494 |
+
},
|
| 495 |
+
{
|
| 496 |
+
"epoch": 15.2,
|
| 497 |
+
"grad_norm": 0.5730023384094238,
|
| 498 |
+
"learning_rate": 0.0003,
|
| 499 |
+
"loss": 0.3795,
|
| 500 |
+
"step": 3800
|
| 501 |
+
},
|
| 502 |
+
{
|
| 503 |
+
"epoch": 15.6,
|
| 504 |
+
"grad_norm": 0.943156898021698,
|
| 505 |
+
"learning_rate": 0.0003,
|
| 506 |
+
"loss": 0.3786,
|
| 507 |
+
"step": 3900
|
| 508 |
+
},
|
| 509 |
+
{
|
| 510 |
+
"epoch": 16.0,
|
| 511 |
+
"grad_norm": 0.8560875654220581,
|
| 512 |
+
"learning_rate": 0.0003,
|
| 513 |
+
"loss": 0.3955,
|
| 514 |
+
"step": 4000
|
| 515 |
+
},
|
| 516 |
+
{
|
| 517 |
+
"epoch": 16.0,
|
| 518 |
+
"eval_accuracy": 0.49196825396825394,
|
| 519 |
+
"eval_loss": 3.6384711265563965,
|
| 520 |
+
"eval_runtime": 5.73,
|
| 521 |
+
"eval_samples_per_second": 87.26,
|
| 522 |
+
"eval_steps_per_second": 10.995,
|
| 523 |
+
"step": 4000
|
| 524 |
+
},
|
| 525 |
+
{
|
| 526 |
+
"epoch": 16.0,
|
| 527 |
+
"eval_exact_match": 14.4,
|
| 528 |
+
"eval_f1": 21.078827926459503,
|
| 529 |
+
"step": 4000
|
| 530 |
+
},
|
| 531 |
+
{
|
| 532 |
+
"epoch": 16.4,
|
| 533 |
+
"grad_norm": 0.5993083715438843,
|
| 534 |
+
"learning_rate": 0.0003,
|
| 535 |
+
"loss": 0.3651,
|
| 536 |
+
"step": 4100
|
| 537 |
+
},
|
| 538 |
+
{
|
| 539 |
+
"epoch": 16.8,
|
| 540 |
+
"grad_norm": 0.8744499683380127,
|
| 541 |
+
"learning_rate": 0.0003,
|
| 542 |
+
"loss": 0.3843,
|
| 543 |
+
"step": 4200
|
| 544 |
+
},
|
| 545 |
+
{
|
| 546 |
+
"epoch": 17.0,
|
| 547 |
+
"eval_accuracy": 0.49082539682539683,
|
| 548 |
+
"eval_loss": 3.658356189727783,
|
| 549 |
+
"eval_runtime": 5.2853,
|
| 550 |
+
"eval_samples_per_second": 94.602,
|
| 551 |
+
"eval_steps_per_second": 11.92,
|
| 552 |
+
"step": 4250
|
| 553 |
+
},
|
| 554 |
+
{
|
| 555 |
+
"epoch": 17.0,
|
| 556 |
+
"eval_exact_match": 15.0,
|
| 557 |
+
"eval_f1": 20.76140470640471,
|
| 558 |
+
"step": 4250
|
| 559 |
+
},
|
| 560 |
+
{
|
| 561 |
+
"epoch": 17.2,
|
| 562 |
+
"grad_norm": 0.7331182360649109,
|
| 563 |
+
"learning_rate": 0.0003,
|
| 564 |
+
"loss": 0.3791,
|
| 565 |
+
"step": 4300
|
| 566 |
+
},
|
| 567 |
+
{
|
| 568 |
+
"epoch": 17.6,
|
| 569 |
+
"grad_norm": 0.9493301510810852,
|
| 570 |
+
"learning_rate": 0.0003,
|
| 571 |
+
"loss": 0.3737,
|
| 572 |
+
"step": 4400
|
| 573 |
+
},
|
| 574 |
+
{
|
| 575 |
+
"epoch": 18.0,
|
| 576 |
+
"grad_norm": 0.7808287143707275,
|
| 577 |
+
"learning_rate": 0.0003,
|
| 578 |
+
"loss": 0.3916,
|
| 579 |
+
"step": 4500
|
| 580 |
+
},
|
| 581 |
+
{
|
| 582 |
+
"epoch": 18.0,
|
| 583 |
+
"eval_accuracy": 0.49247619047619046,
|
| 584 |
+
"eval_loss": 3.659234046936035,
|
| 585 |
+
"eval_runtime": 5.7356,
|
| 586 |
+
"eval_samples_per_second": 87.176,
|
| 587 |
+
"eval_steps_per_second": 10.984,
|
| 588 |
+
"step": 4500
|
| 589 |
+
},
|
| 590 |
+
{
|
| 591 |
+
"epoch": 18.0,
|
| 592 |
+
"eval_exact_match": 14.8,
|
| 593 |
+
"eval_f1": 21.322143412143415,
|
| 594 |
+
"step": 4500
|
| 595 |
+
},
|
| 596 |
+
{
|
| 597 |
+
"epoch": 18.4,
|
| 598 |
+
"grad_norm": 0.7472721934318542,
|
| 599 |
+
"learning_rate": 0.0003,
|
| 600 |
+
"loss": 0.3692,
|
| 601 |
+
"step": 4600
|
| 602 |
+
},
|
| 603 |
+
{
|
| 604 |
+
"epoch": 18.8,
|
| 605 |
+
"grad_norm": 0.5398291349411011,
|
| 606 |
+
"learning_rate": 0.0003,
|
| 607 |
+
"loss": 0.3825,
|
| 608 |
+
"step": 4700
|
| 609 |
+
},
|
| 610 |
+
{
|
| 611 |
+
"epoch": 19.0,
|
| 612 |
+
"eval_accuracy": 0.4913015873015873,
|
| 613 |
+
"eval_loss": 3.660428762435913,
|
| 614 |
+
"eval_runtime": 5.6417,
|
| 615 |
+
"eval_samples_per_second": 88.626,
|
| 616 |
+
"eval_steps_per_second": 11.167,
|
| 617 |
+
"step": 4750
|
| 618 |
+
},
|
| 619 |
+
{
|
| 620 |
+
"epoch": 19.0,
|
| 621 |
+
"eval_exact_match": 15.8,
|
| 622 |
+
"eval_f1": 23.026702741702746,
|
| 623 |
+
"step": 4750
|
| 624 |
+
},
|
| 625 |
+
{
|
| 626 |
+
"epoch": 19.2,
|
| 627 |
+
"grad_norm": 0.5717331171035767,
|
| 628 |
+
"learning_rate": 0.0003,
|
| 629 |
+
"loss": 0.3746,
|
| 630 |
+
"step": 4800
|
| 631 |
+
},
|
| 632 |
+
{
|
| 633 |
+
"epoch": 19.6,
|
| 634 |
+
"grad_norm": 0.8405028581619263,
|
| 635 |
+
"learning_rate": 0.0003,
|
| 636 |
+
"loss": 0.3756,
|
| 637 |
+
"step": 4900
|
| 638 |
+
},
|
| 639 |
+
{
|
| 640 |
+
"epoch": 20.0,
|
| 641 |
+
"grad_norm": 0.7853202223777771,
|
| 642 |
+
"learning_rate": 0.0003,
|
| 643 |
+
"loss": 0.387,
|
| 644 |
+
"step": 5000
|
| 645 |
+
},
|
| 646 |
+
{
|
| 647 |
+
"epoch": 20.0,
|
| 648 |
+
"eval_accuracy": 0.49044444444444446,
|
| 649 |
+
"eval_loss": 3.7399282455444336,
|
| 650 |
+
"eval_runtime": 5.2464,
|
| 651 |
+
"eval_samples_per_second": 95.303,
|
| 652 |
+
"eval_steps_per_second": 12.008,
|
| 653 |
+
"step": 5000
|
| 654 |
+
},
|
| 655 |
+
{
|
| 656 |
+
"epoch": 20.0,
|
| 657 |
+
"eval_exact_match": 13.8,
|
| 658 |
+
"eval_f1": 21.514708624708625,
|
| 659 |
+
"step": 5000
|
| 660 |
+
},
|
| 661 |
+
{
|
| 662 |
+
"epoch": 20.4,
|
| 663 |
+
"grad_norm": 0.8046431541442871,
|
| 664 |
+
"learning_rate": 0.0003,
|
| 665 |
+
"loss": 0.3643,
|
| 666 |
+
"step": 5100
|
| 667 |
+
},
|
| 668 |
+
{
|
| 669 |
+
"epoch": 20.8,
|
| 670 |
+
"grad_norm": 0.6186045408248901,
|
| 671 |
+
"learning_rate": 0.0003,
|
| 672 |
+
"loss": 0.3832,
|
| 673 |
+
"step": 5200
|
| 674 |
+
},
|
| 675 |
+
{
|
| 676 |
+
"epoch": 21.0,
|
| 677 |
+
"eval_accuracy": 0.4913968253968254,
|
| 678 |
+
"eval_loss": 3.6563804149627686,
|
| 679 |
+
"eval_runtime": 5.7336,
|
| 680 |
+
"eval_samples_per_second": 87.206,
|
| 681 |
+
"eval_steps_per_second": 10.988,
|
| 682 |
+
"step": 5250
|
| 683 |
+
},
|
| 684 |
+
{
|
| 685 |
+
"epoch": 21.0,
|
| 686 |
+
"eval_exact_match": 13.8,
|
| 687 |
+
"eval_f1": 20.836753246753247,
|
| 688 |
+
"step": 5250
|
| 689 |
+
},
|
| 690 |
+
{
|
| 691 |
+
"epoch": 21.2,
|
| 692 |
+
"grad_norm": 0.4458799660205841,
|
| 693 |
+
"learning_rate": 0.0003,
|
| 694 |
+
"loss": 0.3733,
|
| 695 |
+
"step": 5300
|
| 696 |
+
},
|
| 697 |
+
{
|
| 698 |
+
"epoch": 21.6,
|
| 699 |
+
"grad_norm": 0.6755269765853882,
|
| 700 |
+
"learning_rate": 0.0003,
|
| 701 |
+
"loss": 0.3675,
|
| 702 |
+
"step": 5400
|
| 703 |
+
},
|
| 704 |
+
{
|
| 705 |
+
"epoch": 22.0,
|
| 706 |
+
"grad_norm": 0.8081554174423218,
|
| 707 |
+
"learning_rate": 0.0003,
|
| 708 |
+
"loss": 0.384,
|
| 709 |
+
"step": 5500
|
| 710 |
+
},
|
| 711 |
+
{
|
| 712 |
+
"epoch": 22.0,
|
| 713 |
+
"eval_accuracy": 0.4899047619047619,
|
| 714 |
+
"eval_loss": 3.686178207397461,
|
| 715 |
+
"eval_runtime": 5.0623,
|
| 716 |
+
"eval_samples_per_second": 98.77,
|
| 717 |
+
"eval_steps_per_second": 12.445,
|
| 718 |
+
"step": 5500
|
| 719 |
+
},
|
| 720 |
+
{
|
| 721 |
+
"epoch": 22.0,
|
| 722 |
+
"eval_exact_match": 15.6,
|
| 723 |
+
"eval_f1": 22.18729215229216,
|
| 724 |
+
"step": 5500
|
| 725 |
+
},
|
| 726 |
+
{
|
| 727 |
+
"epoch": 22.4,
|
| 728 |
+
"grad_norm": 0.5928174257278442,
|
| 729 |
+
"learning_rate": 0.0003,
|
| 730 |
+
"loss": 0.3561,
|
| 731 |
+
"step": 5600
|
| 732 |
+
},
|
| 733 |
+
{
|
| 734 |
+
"epoch": 22.8,
|
| 735 |
+
"grad_norm": 0.5147334933280945,
|
| 736 |
+
"learning_rate": 0.0003,
|
| 737 |
+
"loss": 0.3753,
|
| 738 |
+
"step": 5700
|
| 739 |
+
},
|
| 740 |
+
{
|
| 741 |
+
"epoch": 23.0,
|
| 742 |
+
"eval_accuracy": 0.49063492063492065,
|
| 743 |
+
"eval_loss": 3.6690633296966553,
|
| 744 |
+
"eval_runtime": 5.6639,
|
| 745 |
+
"eval_samples_per_second": 88.278,
|
| 746 |
+
"eval_steps_per_second": 11.123,
|
| 747 |
+
"step": 5750
|
| 748 |
+
},
|
| 749 |
+
{
|
| 750 |
+
"epoch": 23.0,
|
| 751 |
+
"eval_exact_match": 14.2,
|
| 752 |
+
"eval_f1": 21.33359729642339,
|
| 753 |
+
"step": 5750
|
| 754 |
+
},
|
| 755 |
+
{
|
| 756 |
+
"epoch": 23.2,
|
| 757 |
+
"grad_norm": 0.8577277660369873,
|
| 758 |
+
"learning_rate": 0.0003,
|
| 759 |
+
"loss": 0.3655,
|
| 760 |
+
"step": 5800
|
| 761 |
+
},
|
| 762 |
+
{
|
| 763 |
+
"epoch": 23.6,
|
| 764 |
+
"grad_norm": 0.6065458655357361,
|
| 765 |
+
"learning_rate": 0.0003,
|
| 766 |
+
"loss": 0.3609,
|
| 767 |
+
"step": 5900
|
| 768 |
+
},
|
| 769 |
+
{
|
| 770 |
+
"epoch": 24.0,
|
| 771 |
+
"grad_norm": 1.1300039291381836,
|
| 772 |
+
"learning_rate": 0.0003,
|
| 773 |
+
"loss": 0.3816,
|
| 774 |
+
"step": 6000
|
| 775 |
+
},
|
| 776 |
+
{
|
| 777 |
+
"epoch": 24.0,
|
| 778 |
+
"eval_accuracy": 0.4908571428571429,
|
| 779 |
+
"eval_loss": 3.7181262969970703,
|
| 780 |
+
"eval_runtime": 5.7672,
|
| 781 |
+
"eval_samples_per_second": 86.698,
|
| 782 |
+
"eval_steps_per_second": 10.924,
|
| 783 |
+
"step": 6000
|
| 784 |
+
},
|
| 785 |
+
{
|
| 786 |
+
"epoch": 24.0,
|
| 787 |
+
"eval_exact_match": 13.6,
|
| 788 |
+
"eval_f1": 20.043032849503437,
|
| 789 |
+
"step": 6000
|
| 790 |
+
},
|
| 791 |
+
{
|
| 792 |
+
"epoch": 24.4,
|
| 793 |
+
"grad_norm": 0.722024142742157,
|
| 794 |
+
"learning_rate": 0.0003,
|
| 795 |
+
"loss": 0.3555,
|
| 796 |
+
"step": 6100
|
| 797 |
+
},
|
| 798 |
+
{
|
| 799 |
+
"epoch": 24.8,
|
| 800 |
+
"grad_norm": 0.773781955242157,
|
| 801 |
+
"learning_rate": 0.0003,
|
| 802 |
+
"loss": 0.3711,
|
| 803 |
+
"step": 6200
|
| 804 |
+
},
|
| 805 |
+
{
|
| 806 |
+
"epoch": 25.0,
|
| 807 |
+
"eval_accuracy": 0.4895555555555556,
|
| 808 |
+
"eval_loss": 3.715923309326172,
|
| 809 |
+
"eval_runtime": 5.35,
|
| 810 |
+
"eval_samples_per_second": 93.457,
|
| 811 |
+
"eval_steps_per_second": 11.776,
|
| 812 |
+
"step": 6250
|
| 813 |
+
},
|
| 814 |
+
{
|
| 815 |
+
"epoch": 25.0,
|
| 816 |
+
"eval_exact_match": 14.2,
|
| 817 |
+
"eval_f1": 21.817198315959928,
|
| 818 |
+
"step": 6250
|
| 819 |
+
},
|
| 820 |
+
{
|
| 821 |
+
"epoch": 25.2,
|
| 822 |
+
"grad_norm": 0.5301027894020081,
|
| 823 |
+
"learning_rate": 0.0003,
|
| 824 |
+
"loss": 0.365,
|
| 825 |
+
"step": 6300
|
| 826 |
+
},
|
| 827 |
+
{
|
| 828 |
+
"epoch": 25.6,
|
| 829 |
+
"grad_norm": 0.6550167202949524,
|
| 830 |
+
"learning_rate": 0.0003,
|
| 831 |
+
"loss": 0.3632,
|
| 832 |
+
"step": 6400
|
| 833 |
+
},
|
| 834 |
+
{
|
| 835 |
+
"epoch": 26.0,
|
| 836 |
+
"grad_norm": 0.9597300291061401,
|
| 837 |
+
"learning_rate": 0.0003,
|
| 838 |
+
"loss": 0.3758,
|
| 839 |
+
"step": 6500
|
| 840 |
+
},
|
| 841 |
+
{
|
| 842 |
+
"epoch": 26.0,
|
| 843 |
+
"eval_accuracy": 0.49073015873015874,
|
| 844 |
+
"eval_loss": 3.7265737056732178,
|
| 845 |
+
"eval_runtime": 5.7349,
|
| 846 |
+
"eval_samples_per_second": 87.185,
|
| 847 |
+
"eval_steps_per_second": 10.985,
|
| 848 |
+
"step": 6500
|
| 849 |
+
},
|
| 850 |
+
{
|
| 851 |
+
"epoch": 26.0,
|
| 852 |
+
"eval_exact_match": 13.0,
|
| 853 |
+
"eval_f1": 19.8548159030512,
|
| 854 |
+
"step": 6500
|
| 855 |
+
},
|
| 856 |
+
{
|
| 857 |
+
"epoch": 26.4,
|
| 858 |
+
"grad_norm": 0.5059230327606201,
|
| 859 |
+
"learning_rate": 0.0003,
|
| 860 |
+
"loss": 0.3536,
|
| 861 |
+
"step": 6600
|
| 862 |
+
},
|
| 863 |
+
{
|
| 864 |
+
"epoch": 26.8,
|
| 865 |
+
"grad_norm": 0.5931850671768188,
|
| 866 |
+
"learning_rate": 0.0003,
|
| 867 |
+
"loss": 0.372,
|
| 868 |
+
"step": 6700
|
| 869 |
+
},
|
| 870 |
+
{
|
| 871 |
+
"epoch": 27.0,
|
| 872 |
+
"eval_accuracy": 0.4915873015873016,
|
| 873 |
+
"eval_loss": 3.7632603645324707,
|
| 874 |
+
"eval_runtime": 4.9892,
|
| 875 |
+
"eval_samples_per_second": 100.216,
|
| 876 |
+
"eval_steps_per_second": 12.627,
|
| 877 |
+
"step": 6750
|
| 878 |
+
},
|
| 879 |
+
{
|
| 880 |
+
"epoch": 27.0,
|
| 881 |
+
"eval_exact_match": 14.6,
|
| 882 |
+
"eval_f1": 21.288471691706988,
|
| 883 |
+
"step": 6750
|
| 884 |
+
},
|
| 885 |
+
{
|
| 886 |
+
"epoch": 27.2,
|
| 887 |
+
"grad_norm": 0.8211339712142944,
|
| 888 |
+
"learning_rate": 0.0003,
|
| 889 |
+
"loss": 0.3667,
|
| 890 |
+
"step": 6800
|
| 891 |
+
},
|
| 892 |
+
{
|
| 893 |
+
"epoch": 27.6,
|
| 894 |
+
"grad_norm": 0.7241945862770081,
|
| 895 |
+
"learning_rate": 0.0003,
|
| 896 |
+
"loss": 0.3655,
|
| 897 |
+
"step": 6900
|
| 898 |
+
},
|
| 899 |
+
{
|
| 900 |
+
"epoch": 28.0,
|
| 901 |
+
"grad_norm": 0.8376013040542603,
|
| 902 |
+
"learning_rate": 0.0003,
|
| 903 |
+
"loss": 0.3813,
|
| 904 |
+
"step": 7000
|
| 905 |
+
},
|
| 906 |
+
{
|
| 907 |
+
"epoch": 28.0,
|
| 908 |
+
"eval_accuracy": 0.48926984126984124,
|
| 909 |
+
"eval_loss": 3.751114845275879,
|
| 910 |
+
"eval_runtime": 5.7876,
|
| 911 |
+
"eval_samples_per_second": 86.391,
|
| 912 |
+
"eval_steps_per_second": 10.885,
|
| 913 |
+
"step": 7000
|
| 914 |
+
},
|
| 915 |
+
{
|
| 916 |
+
"epoch": 28.0,
|
| 917 |
+
"eval_exact_match": 14.4,
|
| 918 |
+
"eval_f1": 21.447649735885033,
|
| 919 |
+
"step": 7000
|
| 920 |
+
},
|
| 921 |
+
{
|
| 922 |
+
"epoch": 28.4,
|
| 923 |
+
"grad_norm": 1.0523216724395752,
|
| 924 |
+
"learning_rate": 0.0003,
|
| 925 |
+
"loss": 0.3588,
|
| 926 |
+
"step": 7100
|
| 927 |
+
},
|
| 928 |
+
{
|
| 929 |
+
"epoch": 28.8,
|
| 930 |
+
"grad_norm": 1.0963407754898071,
|
| 931 |
+
"learning_rate": 0.0003,
|
| 932 |
+
"loss": 0.3699,
|
| 933 |
+
"step": 7200
|
| 934 |
+
},
|
| 935 |
+
{
|
| 936 |
+
"epoch": 29.0,
|
| 937 |
+
"eval_accuracy": 0.4903174603174603,
|
| 938 |
+
"eval_loss": 3.7290866374969482,
|
| 939 |
+
"eval_runtime": 5.3378,
|
| 940 |
+
"eval_samples_per_second": 93.671,
|
| 941 |
+
"eval_steps_per_second": 11.803,
|
| 942 |
+
"step": 7250
|
| 943 |
+
},
|
| 944 |
+
{
|
| 945 |
+
"epoch": 29.0,
|
| 946 |
+
"eval_exact_match": 14.4,
|
| 947 |
+
"eval_f1": 21.026401539636833,
|
| 948 |
+
"step": 7250
|
| 949 |
+
},
|
| 950 |
+
{
|
| 951 |
+
"epoch": 29.2,
|
| 952 |
+
"grad_norm": 0.6481629610061646,
|
| 953 |
+
"learning_rate": 0.0003,
|
| 954 |
+
"loss": 0.3655,
|
| 955 |
+
"step": 7300
|
| 956 |
+
},
|
| 957 |
+
{
|
| 958 |
+
"epoch": 29.6,
|
| 959 |
+
"grad_norm": 0.6700736284255981,
|
| 960 |
+
"learning_rate": 0.0003,
|
| 961 |
+
"loss": 0.3657,
|
| 962 |
+
"step": 7400
|
| 963 |
+
},
|
| 964 |
+
{
|
| 965 |
+
"epoch": 30.0,
|
| 966 |
+
"grad_norm": 0.6046570539474487,
|
| 967 |
+
"learning_rate": 0.0003,
|
| 968 |
+
"loss": 0.3772,
|
| 969 |
+
"step": 7500
|
| 970 |
+
},
|
| 971 |
+
{
|
| 972 |
+
"epoch": 30.0,
|
| 973 |
+
"eval_accuracy": 0.48857142857142855,
|
| 974 |
+
"eval_loss": 3.7827305793762207,
|
| 975 |
+
"eval_runtime": 4.9957,
|
| 976 |
+
"eval_samples_per_second": 100.086,
|
| 977 |
+
"eval_steps_per_second": 12.611,
|
| 978 |
+
"step": 7500
|
| 979 |
+
},
|
| 980 |
+
{
|
| 981 |
+
"epoch": 30.0,
|
| 982 |
+
"eval_exact_match": 14.2,
|
| 983 |
+
"eval_f1": 20.93110366757426,
|
| 984 |
+
"step": 7500
|
| 985 |
+
},
|
| 986 |
+
{
|
| 987 |
+
"epoch": 30.4,
|
| 988 |
+
"grad_norm": 0.5793018937110901,
|
| 989 |
+
"learning_rate": 0.0003,
|
| 990 |
+
"loss": 0.3515,
|
| 991 |
+
"step": 7600
|
| 992 |
+
},
|
| 993 |
+
{
|
| 994 |
+
"epoch": 30.8,
|
| 995 |
+
"grad_norm": 0.733734667301178,
|
| 996 |
+
"learning_rate": 0.0003,
|
| 997 |
+
"loss": 0.3673,
|
| 998 |
+
"step": 7700
|
| 999 |
+
},
|
| 1000 |
+
{
|
| 1001 |
+
"epoch": 31.0,
|
| 1002 |
+
"eval_accuracy": 0.48917460317460315,
|
| 1003 |
+
"eval_loss": 3.8032257556915283,
|
| 1004 |
+
"eval_runtime": 5.7717,
|
| 1005 |
+
"eval_samples_per_second": 86.629,
|
| 1006 |
+
"eval_steps_per_second": 10.915,
|
| 1007 |
+
"step": 7750
|
| 1008 |
+
},
|
| 1009 |
+
{
|
| 1010 |
+
"epoch": 31.0,
|
| 1011 |
+
"eval_exact_match": 13.2,
|
| 1012 |
+
"eval_f1": 19.896597454832754,
|
| 1013 |
+
"step": 7750
|
| 1014 |
+
},
|
| 1015 |
+
{
|
| 1016 |
+
"epoch": 31.2,
|
| 1017 |
+
"grad_norm": 0.8759809732437134,
|
| 1018 |
+
"learning_rate": 0.0003,
|
| 1019 |
+
"loss": 0.3642,
|
| 1020 |
+
"step": 7800
|
| 1021 |
+
},
|
| 1022 |
+
{
|
| 1023 |
+
"epoch": 31.6,
|
| 1024 |
+
"grad_norm": 0.7398084998130798,
|
| 1025 |
+
"learning_rate": 0.0003,
|
| 1026 |
+
"loss": 0.3647,
|
| 1027 |
+
"step": 7900
|
| 1028 |
+
},
|
| 1029 |
+
{
|
| 1030 |
+
"epoch": 32.0,
|
| 1031 |
+
"grad_norm": 0.9378136992454529,
|
| 1032 |
+
"learning_rate": 0.0003,
|
| 1033 |
+
"loss": 0.3708,
|
| 1034 |
+
"step": 8000
|
| 1035 |
+
},
|
| 1036 |
+
{
|
| 1037 |
+
"epoch": 32.0,
|
| 1038 |
+
"eval_accuracy": 0.4894603174603175,
|
| 1039 |
+
"eval_loss": 3.830268144607544,
|
| 1040 |
+
"eval_runtime": 5.6155,
|
| 1041 |
+
"eval_samples_per_second": 89.039,
|
| 1042 |
+
"eval_steps_per_second": 11.219,
|
| 1043 |
+
"step": 8000
|
| 1044 |
+
},
|
| 1045 |
+
{
|
| 1046 |
+
"epoch": 32.0,
|
| 1047 |
+
"eval_exact_match": 14.0,
|
| 1048 |
+
"eval_f1": 20.862043120278422,
|
| 1049 |
+
"step": 8000
|
| 1050 |
+
},
|
| 1051 |
+
{
|
| 1052 |
+
"epoch": 32.4,
|
| 1053 |
+
"grad_norm": 0.4455881118774414,
|
| 1054 |
+
"learning_rate": 0.0003,
|
| 1055 |
+
"loss": 0.346,
|
| 1056 |
+
"step": 8100
|
| 1057 |
+
},
|
| 1058 |
+
{
|
| 1059 |
+
"epoch": 32.8,
|
| 1060 |
+
"grad_norm": 0.6031106114387512,
|
| 1061 |
+
"learning_rate": 0.0003,
|
| 1062 |
+
"loss": 0.3632,
|
| 1063 |
+
"step": 8200
|
| 1064 |
+
},
|
| 1065 |
+
{
|
| 1066 |
+
"epoch": 33.0,
|
| 1067 |
+
"eval_accuracy": 0.4886984126984127,
|
| 1068 |
+
"eval_loss": 3.821800947189331,
|
| 1069 |
+
"eval_runtime": 5.21,
|
| 1070 |
+
"eval_samples_per_second": 95.968,
|
| 1071 |
+
"eval_steps_per_second": 12.092,
|
| 1072 |
+
"step": 8250
|
| 1073 |
+
},
|
| 1074 |
+
{
|
| 1075 |
+
"epoch": 33.0,
|
| 1076 |
+
"eval_exact_match": 14.0,
|
| 1077 |
+
"eval_f1": 20.771498109733407,
|
| 1078 |
+
"step": 8250
|
| 1079 |
+
},
|
| 1080 |
+
{
|
| 1081 |
+
"epoch": 33.2,
|
| 1082 |
+
"grad_norm": 0.6439555287361145,
|
| 1083 |
+
"learning_rate": 0.0003,
|
| 1084 |
+
"loss": 0.3553,
|
| 1085 |
+
"step": 8300
|
| 1086 |
+
},
|
| 1087 |
+
{
|
| 1088 |
+
"epoch": 33.6,
|
| 1089 |
+
"grad_norm": 0.6980018019676208,
|
| 1090 |
+
"learning_rate": 0.0003,
|
| 1091 |
+
"loss": 0.3533,
|
| 1092 |
+
"step": 8400
|
| 1093 |
+
},
|
| 1094 |
+
{
|
| 1095 |
+
"epoch": 34.0,
|
| 1096 |
+
"grad_norm": 0.6426189541816711,
|
| 1097 |
+
"learning_rate": 0.0003,
|
| 1098 |
+
"loss": 0.3692,
|
| 1099 |
+
"step": 8500
|
| 1100 |
+
},
|
| 1101 |
+
{
|
| 1102 |
+
"epoch": 34.0,
|
| 1103 |
+
"eval_accuracy": 0.49,
|
| 1104 |
+
"eval_loss": 3.7488105297088623,
|
| 1105 |
+
"eval_runtime": 4.9468,
|
| 1106 |
+
"eval_samples_per_second": 101.075,
|
| 1107 |
+
"eval_steps_per_second": 12.735,
|
| 1108 |
+
"step": 8500
|
| 1109 |
+
},
|
| 1110 |
+
{
|
| 1111 |
+
"epoch": 34.0,
|
| 1112 |
+
"eval_exact_match": 14.8,
|
| 1113 |
+
"eval_f1": 21.277384282384293,
|
| 1114 |
+
"step": 8500
|
| 1115 |
+
},
|
| 1116 |
+
{
|
| 1117 |
+
"epoch": 34.4,
|
| 1118 |
+
"grad_norm": 0.7607336044311523,
|
| 1119 |
+
"learning_rate": 0.0003,
|
| 1120 |
+
"loss": 0.3475,
|
| 1121 |
+
"step": 8600
|
| 1122 |
+
},
|
| 1123 |
+
{
|
| 1124 |
+
"epoch": 34.8,
|
| 1125 |
+
"grad_norm": 0.590248167514801,
|
| 1126 |
+
"learning_rate": 0.0003,
|
| 1127 |
+
"loss": 0.3678,
|
| 1128 |
+
"step": 8700
|
| 1129 |
+
},
|
| 1130 |
+
{
|
| 1131 |
+
"epoch": 35.0,
|
| 1132 |
+
"eval_accuracy": 0.4868571428571429,
|
| 1133 |
+
"eval_loss": 3.852421760559082,
|
| 1134 |
+
"eval_runtime": 5.7678,
|
| 1135 |
+
"eval_samples_per_second": 86.688,
|
| 1136 |
+
"eval_steps_per_second": 10.923,
|
| 1137 |
+
"step": 8750
|
| 1138 |
+
},
|
| 1139 |
+
{
|
| 1140 |
+
"epoch": 35.0,
|
| 1141 |
+
"eval_exact_match": 13.4,
|
| 1142 |
+
"eval_f1": 20.523333496568796,
|
| 1143 |
+
"step": 8750
|
| 1144 |
+
},
|
| 1145 |
+
{
|
| 1146 |
+
"epoch": 35.2,
|
| 1147 |
+
"grad_norm": 0.4356330931186676,
|
| 1148 |
+
"learning_rate": 0.0003,
|
| 1149 |
+
"loss": 0.3592,
|
| 1150 |
+
"step": 8800
|
| 1151 |
+
},
|
| 1152 |
+
{
|
| 1153 |
+
"epoch": 35.6,
|
| 1154 |
+
"grad_norm": 0.8419196009635925,
|
| 1155 |
+
"learning_rate": 0.0003,
|
| 1156 |
+
"loss": 0.3609,
|
| 1157 |
+
"step": 8900
|
| 1158 |
+
},
|
| 1159 |
+
{
|
| 1160 |
+
"epoch": 36.0,
|
| 1161 |
+
"grad_norm": 0.7648630142211914,
|
| 1162 |
+
"learning_rate": 0.0003,
|
| 1163 |
+
"loss": 0.3762,
|
| 1164 |
+
"step": 9000
|
| 1165 |
+
},
|
| 1166 |
+
{
|
| 1167 |
+
"epoch": 36.0,
|
| 1168 |
+
"eval_accuracy": 0.4875238095238095,
|
| 1169 |
+
"eval_loss": 3.8220598697662354,
|
| 1170 |
+
"eval_runtime": 5.6351,
|
| 1171 |
+
"eval_samples_per_second": 88.729,
|
| 1172 |
+
"eval_steps_per_second": 11.18,
|
| 1173 |
+
"step": 9000
|
| 1174 |
+
},
|
| 1175 |
+
{
|
| 1176 |
+
"epoch": 36.0,
|
| 1177 |
+
"eval_exact_match": 13.4,
|
| 1178 |
+
"eval_f1": 20.17646758470288,
|
| 1179 |
+
"step": 9000
|
| 1180 |
+
},
|
| 1181 |
+
{
|
| 1182 |
+
"epoch": 36.4,
|
| 1183 |
+
"grad_norm": 0.7656174898147583,
|
| 1184 |
+
"learning_rate": 0.0003,
|
| 1185 |
+
"loss": 0.3553,
|
| 1186 |
+
"step": 9100
|
| 1187 |
+
},
|
| 1188 |
+
{
|
| 1189 |
+
"epoch": 36.8,
|
| 1190 |
+
"grad_norm": 0.7704755067825317,
|
| 1191 |
+
"learning_rate": 0.0003,
|
| 1192 |
+
"loss": 0.3702,
|
| 1193 |
+
"step": 9200
|
| 1194 |
+
},
|
| 1195 |
+
{
|
| 1196 |
+
"epoch": 37.0,
|
| 1197 |
+
"eval_accuracy": 0.4862222222222222,
|
| 1198 |
+
"eval_loss": 3.8083200454711914,
|
| 1199 |
+
"eval_runtime": 5.2407,
|
| 1200 |
+
"eval_samples_per_second": 95.407,
|
| 1201 |
+
"eval_steps_per_second": 12.021,
|
| 1202 |
+
"step": 9250
|
| 1203 |
+
},
|
| 1204 |
+
{
|
| 1205 |
+
"epoch": 37.0,
|
| 1206 |
+
"eval_exact_match": 13.4,
|
| 1207 |
+
"eval_f1": 20.7023073005426,
|
| 1208 |
+
"step": 9250
|
| 1209 |
+
},
|
| 1210 |
+
{
|
| 1211 |
+
"epoch": 37.2,
|
| 1212 |
+
"grad_norm": 1.0103473663330078,
|
| 1213 |
+
"learning_rate": 0.0003,
|
| 1214 |
+
"loss": 0.361,
|
| 1215 |
+
"step": 9300
|
| 1216 |
+
},
|
| 1217 |
+
{
|
| 1218 |
+
"epoch": 37.6,
|
| 1219 |
+
"grad_norm": 0.7305041551589966,
|
| 1220 |
+
"learning_rate": 0.0003,
|
| 1221 |
+
"loss": 0.3583,
|
| 1222 |
+
"step": 9400
|
| 1223 |
+
},
|
| 1224 |
+
{
|
| 1225 |
+
"epoch": 38.0,
|
| 1226 |
+
"grad_norm": 0.8439377546310425,
|
| 1227 |
+
"learning_rate": 0.0003,
|
| 1228 |
+
"loss": 0.3745,
|
| 1229 |
+
"step": 9500
|
| 1230 |
+
},
|
| 1231 |
+
{
|
| 1232 |
+
"epoch": 38.0,
|
| 1233 |
+
"eval_accuracy": 0.48603174603174604,
|
| 1234 |
+
"eval_loss": 3.83286452293396,
|
| 1235 |
+
"eval_runtime": 5.1879,
|
| 1236 |
+
"eval_samples_per_second": 96.377,
|
| 1237 |
+
"eval_steps_per_second": 12.144,
|
| 1238 |
+
"step": 9500
|
| 1239 |
+
},
|
| 1240 |
+
{
|
| 1241 |
+
"epoch": 38.0,
|
| 1242 |
+
"eval_exact_match": 15.4,
|
| 1243 |
+
"eval_f1": 21.705345210345218,
|
| 1244 |
+
"step": 9500
|
| 1245 |
+
},
|
| 1246 |
+
{
|
| 1247 |
+
"epoch": 38.4,
|
| 1248 |
+
"grad_norm": 0.44817036390304565,
|
| 1249 |
+
"learning_rate": 0.0003,
|
| 1250 |
+
"loss": 0.3493,
|
| 1251 |
+
"step": 9600
|
| 1252 |
+
},
|
| 1253 |
+
{
|
| 1254 |
+
"epoch": 38.8,
|
| 1255 |
+
"grad_norm": 0.8559367060661316,
|
| 1256 |
+
"learning_rate": 0.0003,
|
| 1257 |
+
"loss": 0.3611,
|
| 1258 |
+
"step": 9700
|
| 1259 |
+
},
|
| 1260 |
+
{
|
| 1261 |
+
"epoch": 39.0,
|
| 1262 |
+
"eval_accuracy": 0.48784126984126985,
|
| 1263 |
+
"eval_loss": 3.8968539237976074,
|
| 1264 |
+
"eval_runtime": 4.9847,
|
| 1265 |
+
"eval_samples_per_second": 100.306,
|
| 1266 |
+
"eval_steps_per_second": 12.639,
|
| 1267 |
+
"step": 9750
|
| 1268 |
+
},
|
| 1269 |
+
{
|
| 1270 |
+
"epoch": 39.0,
|
| 1271 |
+
"eval_exact_match": 14.4,
|
| 1272 |
+
"eval_f1": 20.261010819246113,
|
| 1273 |
+
"step": 9750
|
| 1274 |
+
},
|
| 1275 |
+
{
|
| 1276 |
+
"epoch": 39.2,
|
| 1277 |
+
"grad_norm": 0.6502194404602051,
|
| 1278 |
+
"learning_rate": 0.0003,
|
| 1279 |
+
"loss": 0.358,
|
| 1280 |
+
"step": 9800
|
| 1281 |
+
},
|
| 1282 |
+
{
|
| 1283 |
+
"epoch": 39.6,
|
| 1284 |
+
"grad_norm": 1.0540180206298828,
|
| 1285 |
+
"learning_rate": 0.0003,
|
| 1286 |
+
"loss": 0.3528,
|
| 1287 |
+
"step": 9900
|
| 1288 |
+
},
|
| 1289 |
+
{
|
| 1290 |
+
"epoch": 40.0,
|
| 1291 |
+
"grad_norm": 0.6658892631530762,
|
| 1292 |
+
"learning_rate": 0.0003,
|
| 1293 |
+
"loss": 0.3648,
|
| 1294 |
+
"step": 10000
|
| 1295 |
+
},
|
| 1296 |
+
{
|
| 1297 |
+
"epoch": 40.0,
|
| 1298 |
+
"eval_accuracy": 0.4868571428571429,
|
| 1299 |
+
"eval_loss": 3.84969425201416,
|
| 1300 |
+
"eval_runtime": 5.7485,
|
| 1301 |
+
"eval_samples_per_second": 86.98,
|
| 1302 |
+
"eval_steps_per_second": 10.959,
|
| 1303 |
+
"step": 10000
|
| 1304 |
+
},
|
| 1305 |
+
{
|
| 1306 |
+
"epoch": 40.0,
|
| 1307 |
+
"eval_exact_match": 13.0,
|
| 1308 |
+
"eval_f1": 20.333614881850174,
|
| 1309 |
+
"step": 10000
|
| 1310 |
+
},
|
| 1311 |
+
{
|
| 1312 |
+
"epoch": 40.4,
|
| 1313 |
+
"grad_norm": 0.5723844766616821,
|
| 1314 |
+
"learning_rate": 0.0003,
|
| 1315 |
+
"loss": 0.347,
|
| 1316 |
+
"step": 10100
|
| 1317 |
+
},
|
| 1318 |
+
{
|
| 1319 |
+
"epoch": 40.8,
|
| 1320 |
+
"grad_norm": 0.5007913708686829,
|
| 1321 |
+
"learning_rate": 0.0003,
|
| 1322 |
+
"loss": 0.3616,
|
| 1323 |
+
"step": 10200
|
| 1324 |
+
},
|
| 1325 |
+
{
|
| 1326 |
+
"epoch": 41.0,
|
| 1327 |
+
"eval_accuracy": 0.4865079365079365,
|
| 1328 |
+
"eval_loss": 3.8460936546325684,
|
| 1329 |
+
"eval_runtime": 5.3089,
|
| 1330 |
+
"eval_samples_per_second": 94.181,
|
| 1331 |
+
"eval_steps_per_second": 11.867,
|
| 1332 |
+
"step": 10250
|
| 1333 |
+
},
|
| 1334 |
+
{
|
| 1335 |
+
"epoch": 41.0,
|
| 1336 |
+
"eval_exact_match": 12.8,
|
| 1337 |
+
"eval_f1": 19.745857084092385,
|
| 1338 |
+
"step": 10250
|
| 1339 |
+
},
|
| 1340 |
+
{
|
| 1341 |
+
"epoch": 41.2,
|
| 1342 |
+
"grad_norm": 0.597031831741333,
|
| 1343 |
+
"learning_rate": 0.0003,
|
| 1344 |
+
"loss": 0.3535,
|
| 1345 |
+
"step": 10300
|
| 1346 |
+
},
|
| 1347 |
+
{
|
| 1348 |
+
"epoch": 41.6,
|
| 1349 |
+
"grad_norm": 0.4940139651298523,
|
| 1350 |
+
"learning_rate": 0.0003,
|
| 1351 |
+
"loss": 0.3511,
|
| 1352 |
+
"step": 10400
|
| 1353 |
+
},
|
| 1354 |
+
{
|
| 1355 |
+
"epoch": 42.0,
|
| 1356 |
+
"grad_norm": 0.9585108757019043,
|
| 1357 |
+
"learning_rate": 0.0003,
|
| 1358 |
+
"loss": 0.3659,
|
| 1359 |
+
"step": 10500
|
| 1360 |
+
},
|
| 1361 |
+
{
|
| 1362 |
+
"epoch": 42.0,
|
| 1363 |
+
"eval_accuracy": 0.48768253968253966,
|
| 1364 |
+
"eval_loss": 3.8721938133239746,
|
| 1365 |
+
"eval_runtime": 4.8871,
|
| 1366 |
+
"eval_samples_per_second": 102.31,
|
| 1367 |
+
"eval_steps_per_second": 12.891,
|
| 1368 |
+
"step": 10500
|
| 1369 |
+
},
|
| 1370 |
+
{
|
| 1371 |
+
"epoch": 42.0,
|
| 1372 |
+
"eval_exact_match": 13.4,
|
| 1373 |
+
"eval_f1": 20.243903482138784,
|
| 1374 |
+
"step": 10500
|
| 1375 |
+
},
|
| 1376 |
+
{
|
| 1377 |
+
"epoch": 42.4,
|
| 1378 |
+
"grad_norm": 0.5968766212463379,
|
| 1379 |
+
"learning_rate": 0.0003,
|
| 1380 |
+
"loss": 0.345,
|
| 1381 |
+
"step": 10600
|
| 1382 |
+
},
|
| 1383 |
+
{
|
| 1384 |
+
"epoch": 42.8,
|
| 1385 |
+
"grad_norm": 0.5495741367340088,
|
| 1386 |
+
"learning_rate": 0.0003,
|
| 1387 |
+
"loss": 0.3585,
|
| 1388 |
+
"step": 10700
|
| 1389 |
+
},
|
| 1390 |
+
{
|
| 1391 |
+
"epoch": 43.0,
|
| 1392 |
+
"eval_accuracy": 0.48742857142857143,
|
| 1393 |
+
"eval_loss": 3.876279830932617,
|
| 1394 |
+
"eval_runtime": 5.7458,
|
| 1395 |
+
"eval_samples_per_second": 87.021,
|
| 1396 |
+
"eval_steps_per_second": 10.965,
|
| 1397 |
+
"step": 10750
|
| 1398 |
+
},
|
| 1399 |
+
{
|
| 1400 |
+
"epoch": 43.0,
|
| 1401 |
+
"eval_exact_match": 13.2,
|
| 1402 |
+
"eval_f1": 20.778424516659825,
|
| 1403 |
+
"step": 10750
|
| 1404 |
+
},
|
| 1405 |
+
{
|
| 1406 |
+
"epoch": 43.2,
|
| 1407 |
+
"grad_norm": 0.6515761613845825,
|
| 1408 |
+
"learning_rate": 0.0003,
|
| 1409 |
+
"loss": 0.3534,
|
| 1410 |
+
"step": 10800
|
| 1411 |
+
},
|
| 1412 |
+
{
|
| 1413 |
+
"epoch": 43.6,
|
| 1414 |
+
"grad_norm": 0.6179555654525757,
|
| 1415 |
+
"learning_rate": 0.0003,
|
| 1416 |
+
"loss": 0.3506,
|
| 1417 |
+
"step": 10900
|
| 1418 |
+
},
|
| 1419 |
+
{
|
| 1420 |
+
"epoch": 44.0,
|
| 1421 |
+
"grad_norm": 0.5593932271003723,
|
| 1422 |
+
"learning_rate": 0.0003,
|
| 1423 |
+
"loss": 0.3628,
|
| 1424 |
+
"step": 11000
|
| 1425 |
+
},
|
| 1426 |
+
{
|
| 1427 |
+
"epoch": 44.0,
|
| 1428 |
+
"eval_accuracy": 0.48768253968253966,
|
| 1429 |
+
"eval_loss": 3.850708484649658,
|
| 1430 |
+
"eval_runtime": 5.1171,
|
| 1431 |
+
"eval_samples_per_second": 97.711,
|
| 1432 |
+
"eval_steps_per_second": 12.312,
|
| 1433 |
+
"step": 11000
|
| 1434 |
+
},
|
| 1435 |
+
{
|
| 1436 |
+
"epoch": 44.0,
|
| 1437 |
+
"eval_exact_match": 14.2,
|
| 1438 |
+
"eval_f1": 22.1020778808395,
|
| 1439 |
+
"step": 11000
|
| 1440 |
+
},
|
| 1441 |
+
{
|
| 1442 |
+
"epoch": 44.4,
|
| 1443 |
+
"grad_norm": 0.6734414100646973,
|
| 1444 |
+
"learning_rate": 0.0003,
|
| 1445 |
+
"loss": 0.3459,
|
| 1446 |
+
"step": 11100
|
| 1447 |
+
},
|
| 1448 |
+
{
|
| 1449 |
+
"epoch": 44.8,
|
| 1450 |
+
"grad_norm": 0.840034008026123,
|
| 1451 |
+
"learning_rate": 0.0003,
|
| 1452 |
+
"loss": 0.3616,
|
| 1453 |
+
"step": 11200
|
| 1454 |
+
},
|
| 1455 |
+
{
|
| 1456 |
+
"epoch": 45.0,
|
| 1457 |
+
"eval_accuracy": 0.4875555555555556,
|
| 1458 |
+
"eval_loss": 3.878772020339966,
|
| 1459 |
+
"eval_runtime": 4.9819,
|
| 1460 |
+
"eval_samples_per_second": 100.364,
|
| 1461 |
+
"eval_steps_per_second": 12.646,
|
| 1462 |
+
"step": 11250
|
| 1463 |
+
},
|
| 1464 |
+
{
|
| 1465 |
+
"epoch": 45.0,
|
| 1466 |
+
"eval_exact_match": 13.6,
|
| 1467 |
+
"eval_f1": 20.89890237213767,
|
| 1468 |
+
"step": 11250
|
| 1469 |
+
},
|
| 1470 |
+
{
|
| 1471 |
+
"epoch": 45.2,
|
| 1472 |
+
"grad_norm": 0.6616114377975464,
|
| 1473 |
+
"learning_rate": 0.0003,
|
| 1474 |
+
"loss": 0.3521,
|
| 1475 |
+
"step": 11300
|
| 1476 |
+
},
|
| 1477 |
+
{
|
| 1478 |
+
"epoch": 45.6,
|
| 1479 |
+
"grad_norm": 0.5298745632171631,
|
| 1480 |
+
"learning_rate": 0.0003,
|
| 1481 |
+
"loss": 0.3577,
|
| 1482 |
+
"step": 11400
|
| 1483 |
+
},
|
| 1484 |
+
{
|
| 1485 |
+
"epoch": 46.0,
|
| 1486 |
+
"grad_norm": 0.7476080060005188,
|
| 1487 |
+
"learning_rate": 0.0003,
|
| 1488 |
+
"loss": 0.367,
|
| 1489 |
+
"step": 11500
|
| 1490 |
+
},
|
| 1491 |
+
{
|
| 1492 |
+
"epoch": 46.0,
|
| 1493 |
+
"eval_accuracy": 0.4874920634920635,
|
| 1494 |
+
"eval_loss": 3.868834972381592,
|
| 1495 |
+
"eval_runtime": 5.6354,
|
| 1496 |
+
"eval_samples_per_second": 88.724,
|
| 1497 |
+
"eval_steps_per_second": 11.179,
|
| 1498 |
+
"step": 11500
|
| 1499 |
+
},
|
| 1500 |
+
{
|
| 1501 |
+
"epoch": 46.0,
|
| 1502 |
+
"eval_exact_match": 12.8,
|
| 1503 |
+
"eval_f1": 20.124567099567106,
|
| 1504 |
+
"step": 11500
|
| 1505 |
+
},
|
| 1506 |
+
{
|
| 1507 |
+
"epoch": 46.4,
|
| 1508 |
+
"grad_norm": 0.672135591506958,
|
| 1509 |
+
"learning_rate": 0.0003,
|
| 1510 |
+
"loss": 0.3523,
|
| 1511 |
+
"step": 11600
|
| 1512 |
+
},
|
| 1513 |
+
{
|
| 1514 |
+
"epoch": 46.8,
|
| 1515 |
+
"grad_norm": 0.4491010308265686,
|
| 1516 |
+
"learning_rate": 0.0003,
|
| 1517 |
+
"loss": 0.3629,
|
| 1518 |
+
"step": 11700
|
| 1519 |
+
},
|
| 1520 |
+
{
|
| 1521 |
+
"epoch": 47.0,
|
| 1522 |
+
"eval_accuracy": 0.4867619047619048,
|
| 1523 |
+
"eval_loss": 3.9209814071655273,
|
| 1524 |
+
"eval_runtime": 5.4435,
|
| 1525 |
+
"eval_samples_per_second": 91.853,
|
| 1526 |
+
"eval_steps_per_second": 11.574,
|
| 1527 |
+
"step": 11750
|
| 1528 |
+
},
|
| 1529 |
+
{
|
| 1530 |
+
"epoch": 47.0,
|
| 1531 |
+
"eval_exact_match": 12.6,
|
| 1532 |
+
"eval_f1": 19.623137581372877,
|
| 1533 |
+
"step": 11750
|
| 1534 |
+
},
|
| 1535 |
+
{
|
| 1536 |
+
"epoch": 47.2,
|
| 1537 |
+
"grad_norm": 0.6844795942306519,
|
| 1538 |
+
"learning_rate": 0.0003,
|
| 1539 |
+
"loss": 0.3558,
|
| 1540 |
+
"step": 11800
|
| 1541 |
+
},
|
| 1542 |
+
{
|
| 1543 |
+
"epoch": 47.6,
|
| 1544 |
+
"grad_norm": 1.0474852323532104,
|
| 1545 |
+
"learning_rate": 0.0003,
|
| 1546 |
+
"loss": 0.3563,
|
| 1547 |
+
"step": 11900
|
| 1548 |
+
},
|
| 1549 |
+
{
|
| 1550 |
+
"epoch": 48.0,
|
| 1551 |
+
"grad_norm": 0.5210233330726624,
|
| 1552 |
+
"learning_rate": 0.0003,
|
| 1553 |
+
"loss": 0.366,
|
| 1554 |
+
"step": 12000
|
| 1555 |
+
},
|
| 1556 |
+
{
|
| 1557 |
+
"epoch": 48.0,
|
| 1558 |
+
"eval_accuracy": 0.4860634920634921,
|
| 1559 |
+
"eval_loss": 3.930508852005005,
|
| 1560 |
+
"eval_runtime": 5.7117,
|
| 1561 |
+
"eval_samples_per_second": 87.54,
|
| 1562 |
+
"eval_steps_per_second": 11.03,
|
| 1563 |
+
"step": 12000
|
| 1564 |
+
},
|
| 1565 |
+
{
|
| 1566 |
+
"epoch": 48.0,
|
| 1567 |
+
"eval_exact_match": 14.0,
|
| 1568 |
+
"eval_f1": 21.42872199695729,
|
| 1569 |
+
"step": 12000
|
| 1570 |
+
},
|
| 1571 |
+
{
|
| 1572 |
+
"epoch": 48.4,
|
| 1573 |
+
"grad_norm": 0.590672492980957,
|
| 1574 |
+
"learning_rate": 0.0003,
|
| 1575 |
+
"loss": 0.3453,
|
| 1576 |
+
"step": 12100
|
| 1577 |
+
},
|
| 1578 |
+
{
|
| 1579 |
+
"epoch": 48.8,
|
| 1580 |
+
"grad_norm": 0.6198665499687195,
|
| 1581 |
+
"learning_rate": 0.0003,
|
| 1582 |
+
"loss": 0.3608,
|
| 1583 |
+
"step": 12200
|
| 1584 |
+
},
|
| 1585 |
+
{
|
| 1586 |
+
"epoch": 49.0,
|
| 1587 |
+
"eval_accuracy": 0.4874603174603175,
|
| 1588 |
+
"eval_loss": 3.926318407058716,
|
| 1589 |
+
"eval_runtime": 5.7747,
|
| 1590 |
+
"eval_samples_per_second": 86.585,
|
| 1591 |
+
"eval_steps_per_second": 10.91,
|
| 1592 |
+
"step": 12250
|
| 1593 |
+
},
|
| 1594 |
+
{
|
| 1595 |
+
"epoch": 49.0,
|
| 1596 |
+
"eval_exact_match": 14.4,
|
| 1597 |
+
"eval_f1": 21.45760644584174,
|
| 1598 |
+
"step": 12250
|
| 1599 |
+
}
|
| 1600 |
+
],
|
| 1601 |
+
"logging_steps": 100,
|
| 1602 |
+
"max_steps": 12500,
|
| 1603 |
+
"num_input_tokens_seen": 0,
|
| 1604 |
+
"num_train_epochs": 50,
|
| 1605 |
+
"save_steps": 500,
|
| 1606 |
+
"stateful_callbacks": {
|
| 1607 |
+
"TrainerControl": {
|
| 1608 |
+
"args": {
|
| 1609 |
+
"should_epoch_stop": false,
|
| 1610 |
+
"should_evaluate": false,
|
| 1611 |
+
"should_log": false,
|
| 1612 |
+
"should_save": true,
|
| 1613 |
+
"should_training_stop": false
|
| 1614 |
+
},
|
| 1615 |
+
"attributes": {}
|
| 1616 |
+
}
|
| 1617 |
+
},
|
| 1618 |
+
"total_flos": 3.169169561060639e+17,
|
| 1619 |
+
"train_batch_size": 1,
|
| 1620 |
+
"trial_name": null,
|
| 1621 |
+
"trial_params": null
|
| 1622 |
+
}
|
checkpoint-12250/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:dae73bb50794744b259c1eaaeb67f6ae0b1d0f39b0a32b5f1162fc82e3a5d6a6
|
| 3 |
+
size 5304
|
checkpoint-12250/vocab.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|