Training in progress, epoch 17, checkpoint
Browse files- checkpoint-4250/README.md +9 -0
- checkpoint-4250/adapter_config.json +25 -0
- checkpoint-4250/adapter_model.safetensors +3 -0
- checkpoint-4250/added_tokens.json +5 -0
- checkpoint-4250/merges.txt +0 -0
- checkpoint-4250/optimizer.pt +3 -0
- checkpoint-4250/rng_state_0.pth +3 -0
- checkpoint-4250/rng_state_1.pth +3 -0
- checkpoint-4250/rng_state_2.pth +3 -0
- checkpoint-4250/rng_state_3.pth +3 -0
- checkpoint-4250/scheduler.pt +3 -0
- checkpoint-4250/special_tokens_map.json +14 -0
- checkpoint-4250/tokenizer.json +0 -0
- checkpoint-4250/tokenizer_config.json +43 -0
- checkpoint-4250/trainer_state.json +582 -0
- checkpoint-4250/training_args.bin +3 -0
- checkpoint-4250/vocab.json +0 -0
checkpoint-4250/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-4250/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-4250/adapter_model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:914cc8664df15f132b9872c8dc65ac108bf104fb40f76d91cef1c1af0142e604
|
| 3 |
+
size 112130416
|
checkpoint-4250/added_tokens.json
ADDED
|
@@ -0,0 +1,5 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"<|endoftext|>": 151643,
|
| 3 |
+
"<|im_end|>": 151645,
|
| 4 |
+
"<|im_start|>": 151644
|
| 5 |
+
}
|
checkpoint-4250/merges.txt
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
checkpoint-4250/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:e7f0882f6a3d3c4ec9ce141d3c955a3ac7f9038717b4f9ca4388ed1706e36baf
|
| 3 |
+
size 224537202
|
checkpoint-4250/rng_state_0.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:ac29ed9322c2384af3986ff067eecc3e6f6933ccf2badd1dbdca82b0ecdbe310
|
| 3 |
+
size 15024
|
checkpoint-4250/rng_state_1.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:8c70b333669f75bb207fe350ce79e19359e3560cdabc92735fa4ec2beac716eb
|
| 3 |
+
size 15024
|
checkpoint-4250/rng_state_2.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:ab9872d90d6cb0adbad79b121bfd66a88030230fb4c9f354e025a737476e477d
|
| 3 |
+
size 15024
|
checkpoint-4250/rng_state_3.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:53316192c5e5baa22b0a498403b2ef54e56a7cf58978223b6ebf6bb834bc3749
|
| 3 |
+
size 15024
|
checkpoint-4250/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:9d83a8893e2d4e656ed3384c45a76b70e8e2c298fb5e8eeba5ddf4a6728df6ca
|
| 3 |
+
size 1064
|
checkpoint-4250/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-4250/tokenizer.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
checkpoint-4250/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-4250/trainer_state.json
ADDED
|
@@ -0,0 +1,582 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 17.0,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 4250,
|
| 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 |
+
"logging_steps": 100,
|
| 562 |
+
"max_steps": 12500,
|
| 563 |
+
"num_input_tokens_seen": 0,
|
| 564 |
+
"num_train_epochs": 50,
|
| 565 |
+
"save_steps": 500,
|
| 566 |
+
"stateful_callbacks": {
|
| 567 |
+
"TrainerControl": {
|
| 568 |
+
"args": {
|
| 569 |
+
"should_epoch_stop": false,
|
| 570 |
+
"should_evaluate": false,
|
| 571 |
+
"should_log": false,
|
| 572 |
+
"should_save": true,
|
| 573 |
+
"should_training_stop": false
|
| 574 |
+
},
|
| 575 |
+
"attributes": {}
|
| 576 |
+
}
|
| 577 |
+
},
|
| 578 |
+
"total_flos": 1.0995077958598656e+17,
|
| 579 |
+
"train_batch_size": 1,
|
| 580 |
+
"trial_name": null,
|
| 581 |
+
"trial_params": null
|
| 582 |
+
}
|
checkpoint-4250/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-4250/vocab.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|