chenhao
commited on
Commit
•
4af8aa6
1
Parent(s):
a26b954
Upload folder using huggingface_hub
Browse files- README.md +46 -0
- adapter_config.json +27 -0
- adapter_model.safetensors +3 -0
- all_results.json +7 -0
- special_tokens_map.json +24 -0
- tokenizer.model +3 -0
- tokenizer_config.json +44 -0
- train_results.json +7 -0
- trainer_log.jsonl +32 -0
- trainer_state.json +216 -0
- training_args.bin +3 -0
- training_loss.png +0 -0
README.md
ADDED
@@ -0,0 +1,46 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
---
|
2 |
+
license: other
|
3 |
+
library_name: peft
|
4 |
+
tags:
|
5 |
+
- llama-factory
|
6 |
+
- lora
|
7 |
+
- generated_from_trainer
|
8 |
+
base_model: mistralai/Mistral-7B-v0.1
|
9 |
+
model-index:
|
10 |
+
- name: mistral-7b-medqa-v1
|
11 |
+
results: []
|
12 |
+
---
|
13 |
+
|
14 |
+
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
15 |
+
should probably proofread and complete it, then remove this comment. -->
|
16 |
+
|
17 |
+
# mistral-7b-medqa-v1
|
18 |
+
|
19 |
+
This model is a fine-tuned version of mistralai/Mistral-7B-v0.1 on the medical_meadow_medqa dataset.
|
20 |
+
|
21 |
+
### Training hyperparameters
|
22 |
+
|
23 |
+
The following hyperparameters were used during training:
|
24 |
+
- learning_rate: 0.0005
|
25 |
+
- train_batch_size: 4
|
26 |
+
- eval_batch_size: 8
|
27 |
+
- seed: 42
|
28 |
+
- gradient_accumulation_steps: 4
|
29 |
+
- total_train_batch_size: 16
|
30 |
+
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
31 |
+
- lr_scheduler_type: cosine
|
32 |
+
- lr_scheduler_warmup_steps: 20
|
33 |
+
- num_epochs: 1.0
|
34 |
+
- mixed_precision_training: Native AMP
|
35 |
+
|
36 |
+
### Training results
|
37 |
+
|
38 |
+
|
39 |
+
|
40 |
+
### Framework versions
|
41 |
+
|
42 |
+
- PEFT 0.8.2
|
43 |
+
- Transformers 4.37.2
|
44 |
+
- Pytorch 2.1.1+cu121
|
45 |
+
- Datasets 2.16.1
|
46 |
+
- Tokenizers 0.15.1
|
adapter_config.json
ADDED
@@ -0,0 +1,27 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"alpha_pattern": {},
|
3 |
+
"auto_mapping": null,
|
4 |
+
"base_model_name_or_path": "mistralai/Mistral-7B-v0.1",
|
5 |
+
"bias": "none",
|
6 |
+
"fan_in_fan_out": false,
|
7 |
+
"inference_mode": true,
|
8 |
+
"init_lora_weights": true,
|
9 |
+
"layers_pattern": null,
|
10 |
+
"layers_to_transform": null,
|
11 |
+
"loftq_config": {},
|
12 |
+
"lora_alpha": 16,
|
13 |
+
"lora_dropout": 0.2,
|
14 |
+
"megatron_config": null,
|
15 |
+
"megatron_core": "megatron.core",
|
16 |
+
"modules_to_save": null,
|
17 |
+
"peft_type": "LORA",
|
18 |
+
"r": 8,
|
19 |
+
"rank_pattern": {},
|
20 |
+
"revision": null,
|
21 |
+
"target_modules": [
|
22 |
+
"q_proj",
|
23 |
+
"v_proj"
|
24 |
+
],
|
25 |
+
"task_type": "CAUSAL_LM",
|
26 |
+
"use_rslora": false
|
27 |
+
}
|
adapter_model.safetensors
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:9eed2c63b22828645550f6ee74d632e57d3e84cd2decc2b1d1fd9c2868ac901d
|
3 |
+
size 13648432
|
all_results.json
ADDED
@@ -0,0 +1,7 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 1.0,
|
3 |
+
"train_loss": 0.16286772943268985,
|
4 |
+
"train_runtime": 1280.7232,
|
5 |
+
"train_samples_per_second": 3.904,
|
6 |
+
"train_steps_per_second": 0.244
|
7 |
+
}
|
special_tokens_map.json
ADDED
@@ -0,0 +1,24 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"bos_token": {
|
3 |
+
"content": "<s>",
|
4 |
+
"lstrip": false,
|
5 |
+
"normalized": false,
|
6 |
+
"rstrip": false,
|
7 |
+
"single_word": false
|
8 |
+
},
|
9 |
+
"eos_token": {
|
10 |
+
"content": "</s>",
|
11 |
+
"lstrip": false,
|
12 |
+
"normalized": false,
|
13 |
+
"rstrip": false,
|
14 |
+
"single_word": false
|
15 |
+
},
|
16 |
+
"pad_token": "</s>",
|
17 |
+
"unk_token": {
|
18 |
+
"content": "<unk>",
|
19 |
+
"lstrip": false,
|
20 |
+
"normalized": false,
|
21 |
+
"rstrip": false,
|
22 |
+
"single_word": false
|
23 |
+
}
|
24 |
+
}
|
tokenizer.model
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:dadfd56d766715c61d2ef780a525ab43b8e6da4de6865bda3d95fdef5e134055
|
3 |
+
size 493443
|
tokenizer_config.json
ADDED
@@ -0,0 +1,44 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"add_bos_token": true,
|
3 |
+
"add_eos_token": false,
|
4 |
+
"added_tokens_decoder": {
|
5 |
+
"0": {
|
6 |
+
"content": "<unk>",
|
7 |
+
"lstrip": false,
|
8 |
+
"normalized": false,
|
9 |
+
"rstrip": false,
|
10 |
+
"single_word": false,
|
11 |
+
"special": true
|
12 |
+
},
|
13 |
+
"1": {
|
14 |
+
"content": "<s>",
|
15 |
+
"lstrip": false,
|
16 |
+
"normalized": false,
|
17 |
+
"rstrip": false,
|
18 |
+
"single_word": false,
|
19 |
+
"special": true
|
20 |
+
},
|
21 |
+
"2": {
|
22 |
+
"content": "</s>",
|
23 |
+
"lstrip": false,
|
24 |
+
"normalized": false,
|
25 |
+
"rstrip": false,
|
26 |
+
"single_word": false,
|
27 |
+
"special": true
|
28 |
+
}
|
29 |
+
},
|
30 |
+
"additional_special_tokens": [],
|
31 |
+
"bos_token": "<s>",
|
32 |
+
"clean_up_tokenization_spaces": false,
|
33 |
+
"eos_token": "</s>",
|
34 |
+
"legacy": true,
|
35 |
+
"model_max_length": 1000000000000000019884624838656,
|
36 |
+
"pad_token": "</s>",
|
37 |
+
"padding_side": "right",
|
38 |
+
"sp_model_kwargs": {},
|
39 |
+
"spaces_between_special_tokens": false,
|
40 |
+
"split_special_tokens": false,
|
41 |
+
"tokenizer_class": "LlamaTokenizer",
|
42 |
+
"unk_token": "<unk>",
|
43 |
+
"use_default_system_prompt": false
|
44 |
+
}
|
train_results.json
ADDED
@@ -0,0 +1,7 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 1.0,
|
3 |
+
"train_loss": 0.16286772943268985,
|
4 |
+
"train_runtime": 1280.7232,
|
5 |
+
"train_samples_per_second": 3.904,
|
6 |
+
"train_steps_per_second": 0.244
|
7 |
+
}
|
trainer_log.jsonl
ADDED
@@ -0,0 +1,32 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{"current_steps": 10, "total_steps": 312, "loss": 0.7923, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 0.00025, "epoch": 0.03, "percentage": 3.21, "elapsed_time": "0:00:44", "remaining_time": "0:22:15"}
|
2 |
+
{"current_steps": 20, "total_steps": 312, "loss": 0.1889, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 0.0005, "epoch": 0.06, "percentage": 6.41, "elapsed_time": "0:01:24", "remaining_time": "0:20:33"}
|
3 |
+
{"current_steps": 30, "total_steps": 312, "loss": 0.1673, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 0.000498554476723488, "epoch": 0.1, "percentage": 9.62, "elapsed_time": "0:02:05", "remaining_time": "0:19:39"}
|
4 |
+
{"current_steps": 40, "total_steps": 312, "loss": 0.1507, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 0.0004942346231942955, "epoch": 0.13, "percentage": 12.82, "elapsed_time": "0:02:46", "remaining_time": "0:18:49"}
|
5 |
+
{"current_steps": 50, "total_steps": 312, "loss": 0.1385, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 0.0004870903950030429, "epoch": 0.16, "percentage": 16.03, "elapsed_time": "0:03:27", "remaining_time": "0:18:09"}
|
6 |
+
{"current_steps": 60, "total_steps": 312, "loss": 0.1593, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 0.00047720440933487574, "epoch": 0.19, "percentage": 19.23, "elapsed_time": "0:04:10", "remaining_time": "0:17:30"}
|
7 |
+
{"current_steps": 70, "total_steps": 312, "loss": 0.1549, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 0.0004646909895689508, "epoch": 0.22, "percentage": 22.44, "elapsed_time": "0:04:50", "remaining_time": "0:16:43"}
|
8 |
+
{"current_steps": 80, "total_steps": 312, "loss": 0.159, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 0.0004496948432215913, "epoch": 0.26, "percentage": 25.64, "elapsed_time": "0:05:30", "remaining_time": "0:15:59"}
|
9 |
+
{"current_steps": 90, "total_steps": 312, "loss": 0.1562, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 0.0004323893885216219, "epoch": 0.29, "percentage": 28.85, "elapsed_time": "0:06:10", "remaining_time": "0:15:13"}
|
10 |
+
{"current_steps": 100, "total_steps": 312, "loss": 0.1352, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 0.00041297474896967814, "epoch": 0.32, "percentage": 32.05, "elapsed_time": "0:06:53", "remaining_time": "0:14:36"}
|
11 |
+
{"current_steps": 110, "total_steps": 312, "loss": 0.1588, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 0.0003916754390727794, "epoch": 0.35, "percentage": 35.26, "elapsed_time": "0:07:33", "remaining_time": "0:13:52"}
|
12 |
+
{"current_steps": 120, "total_steps": 312, "loss": 0.1523, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 0.0003687377680167626, "epoch": 0.38, "percentage": 38.46, "elapsed_time": "0:08:15", "remaining_time": "0:13:12"}
|
13 |
+
{"current_steps": 130, "total_steps": 312, "loss": 0.1418, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 0.0003444269913009912, "epoch": 0.42, "percentage": 41.67, "elapsed_time": "0:08:55", "remaining_time": "0:12:29"}
|
14 |
+
{"current_steps": 140, "total_steps": 312, "loss": 0.1428, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 0.0003190242432743673, "epoch": 0.45, "percentage": 44.87, "elapsed_time": "0:09:38", "remaining_time": "0:11:50"}
|
15 |
+
{"current_steps": 150, "total_steps": 312, "loss": 0.1471, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 0.0002928232860453694, "epoch": 0.48, "percentage": 48.08, "elapsed_time": "0:10:18", "remaining_time": "0:11:08"}
|
16 |
+
{"current_steps": 160, "total_steps": 312, "loss": 0.1427, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 0.00026612711236232914, "epoch": 0.51, "percentage": 51.28, "elapsed_time": "0:10:59", "remaining_time": "0:10:26"}
|
17 |
+
{"current_steps": 170, "total_steps": 312, "loss": 0.1366, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 0.00023924444174886733, "epoch": 0.54, "percentage": 54.49, "elapsed_time": "0:11:40", "remaining_time": "0:09:44"}
|
18 |
+
{"current_steps": 180, "total_steps": 312, "loss": 0.1463, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 0.00021248615041383683, "epoch": 0.58, "percentage": 57.69, "elapsed_time": "0:12:21", "remaining_time": "0:09:03"}
|
19 |
+
{"current_steps": 190, "total_steps": 312, "loss": 0.1328, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 0.00018616167622095327, "epoch": 0.61, "percentage": 60.9, "elapsed_time": "0:13:03", "remaining_time": "0:08:22"}
|
20 |
+
{"current_steps": 200, "total_steps": 312, "loss": 0.134, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 0.0001605754402917186, "epoch": 0.64, "percentage": 64.1, "elapsed_time": "0:13:42", "remaining_time": "0:07:40"}
|
21 |
+
{"current_steps": 210, "total_steps": 312, "loss": 0.1417, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 0.00013602332662288535, "epoch": 0.67, "percentage": 67.31, "elapsed_time": "0:14:22", "remaining_time": "0:06:59"}
|
22 |
+
{"current_steps": 220, "total_steps": 312, "loss": 0.1194, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 0.00011278926042882027, "epoch": 0.7, "percentage": 70.51, "elapsed_time": "0:15:03", "remaining_time": "0:06:18"}
|
23 |
+
{"current_steps": 230, "total_steps": 312, "loss": 0.1359, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 9.114192477745568e-05, "epoch": 0.74, "percentage": 73.72, "elapsed_time": "0:15:44", "remaining_time": "0:05:36"}
|
24 |
+
{"current_steps": 240, "total_steps": 312, "loss": 0.1244, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 7.133165348925977e-05, "epoch": 0.77, "percentage": 76.92, "elapsed_time": "0:16:22", "remaining_time": "0:04:54"}
|
25 |
+
{"current_steps": 250, "total_steps": 312, "loss": 0.1367, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 5.3587536230321364e-05, "epoch": 0.8, "percentage": 80.13, "elapsed_time": "0:17:05", "remaining_time": "0:04:14"}
|
26 |
+
{"current_steps": 260, "total_steps": 312, "loss": 0.1256, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 3.811476927679228e-05, "epoch": 0.83, "percentage": 83.33, "elapsed_time": "0:17:47", "remaining_time": "0:03:33"}
|
27 |
+
{"current_steps": 270, "total_steps": 312, "loss": 0.1312, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 2.5092282586939186e-05, "epoch": 0.86, "percentage": 86.54, "elapsed_time": "0:18:27", "remaining_time": "0:02:52"}
|
28 |
+
{"current_steps": 280, "total_steps": 312, "loss": 0.1272, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 1.467067062178823e-05, "epoch": 0.9, "percentage": 89.74, "elapsed_time": "0:19:06", "remaining_time": "0:02:11"}
|
29 |
+
{"current_steps": 290, "total_steps": 312, "loss": 0.1241, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 6.970450842734649e-06, "epoch": 0.93, "percentage": 92.95, "elapsed_time": "0:19:49", "remaining_time": "0:01:30"}
|
30 |
+
{"current_steps": 300, "total_steps": 312, "loss": 0.1234, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 2.0806700251775058e-06, "epoch": 0.96, "percentage": 96.15, "elapsed_time": "0:20:32", "remaining_time": "0:00:49"}
|
31 |
+
{"current_steps": 310, "total_steps": 312, "loss": 0.1305, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 5.7874505027283307e-08, "epoch": 0.99, "percentage": 99.36, "elapsed_time": "0:21:12", "remaining_time": "0:00:08"}
|
32 |
+
{"current_steps": 312, "total_steps": 312, "loss": null, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": null, "epoch": 1.0, "percentage": 100.0, "elapsed_time": "0:21:20", "remaining_time": "0:00:00"}
|
trainer_state.json
ADDED
@@ -0,0 +1,216 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"best_metric": null,
|
3 |
+
"best_model_checkpoint": null,
|
4 |
+
"epoch": 0.9984,
|
5 |
+
"eval_steps": 500,
|
6 |
+
"global_step": 312,
|
7 |
+
"is_hyper_param_search": false,
|
8 |
+
"is_local_process_zero": true,
|
9 |
+
"is_world_process_zero": true,
|
10 |
+
"log_history": [
|
11 |
+
{
|
12 |
+
"epoch": 0.03,
|
13 |
+
"learning_rate": 0.00025,
|
14 |
+
"loss": 0.7923,
|
15 |
+
"step": 10
|
16 |
+
},
|
17 |
+
{
|
18 |
+
"epoch": 0.06,
|
19 |
+
"learning_rate": 0.0005,
|
20 |
+
"loss": 0.1889,
|
21 |
+
"step": 20
|
22 |
+
},
|
23 |
+
{
|
24 |
+
"epoch": 0.1,
|
25 |
+
"learning_rate": 0.000498554476723488,
|
26 |
+
"loss": 0.1673,
|
27 |
+
"step": 30
|
28 |
+
},
|
29 |
+
{
|
30 |
+
"epoch": 0.13,
|
31 |
+
"learning_rate": 0.0004942346231942955,
|
32 |
+
"loss": 0.1507,
|
33 |
+
"step": 40
|
34 |
+
},
|
35 |
+
{
|
36 |
+
"epoch": 0.16,
|
37 |
+
"learning_rate": 0.0004870903950030429,
|
38 |
+
"loss": 0.1385,
|
39 |
+
"step": 50
|
40 |
+
},
|
41 |
+
{
|
42 |
+
"epoch": 0.19,
|
43 |
+
"learning_rate": 0.00047720440933487574,
|
44 |
+
"loss": 0.1593,
|
45 |
+
"step": 60
|
46 |
+
},
|
47 |
+
{
|
48 |
+
"epoch": 0.22,
|
49 |
+
"learning_rate": 0.0004646909895689508,
|
50 |
+
"loss": 0.1549,
|
51 |
+
"step": 70
|
52 |
+
},
|
53 |
+
{
|
54 |
+
"epoch": 0.26,
|
55 |
+
"learning_rate": 0.0004496948432215913,
|
56 |
+
"loss": 0.159,
|
57 |
+
"step": 80
|
58 |
+
},
|
59 |
+
{
|
60 |
+
"epoch": 0.29,
|
61 |
+
"learning_rate": 0.0004323893885216219,
|
62 |
+
"loss": 0.1562,
|
63 |
+
"step": 90
|
64 |
+
},
|
65 |
+
{
|
66 |
+
"epoch": 0.32,
|
67 |
+
"learning_rate": 0.00041297474896967814,
|
68 |
+
"loss": 0.1352,
|
69 |
+
"step": 100
|
70 |
+
},
|
71 |
+
{
|
72 |
+
"epoch": 0.35,
|
73 |
+
"learning_rate": 0.0003916754390727794,
|
74 |
+
"loss": 0.1588,
|
75 |
+
"step": 110
|
76 |
+
},
|
77 |
+
{
|
78 |
+
"epoch": 0.38,
|
79 |
+
"learning_rate": 0.0003687377680167626,
|
80 |
+
"loss": 0.1523,
|
81 |
+
"step": 120
|
82 |
+
},
|
83 |
+
{
|
84 |
+
"epoch": 0.42,
|
85 |
+
"learning_rate": 0.0003444269913009912,
|
86 |
+
"loss": 0.1418,
|
87 |
+
"step": 130
|
88 |
+
},
|
89 |
+
{
|
90 |
+
"epoch": 0.45,
|
91 |
+
"learning_rate": 0.0003190242432743673,
|
92 |
+
"loss": 0.1428,
|
93 |
+
"step": 140
|
94 |
+
},
|
95 |
+
{
|
96 |
+
"epoch": 0.48,
|
97 |
+
"learning_rate": 0.0002928232860453694,
|
98 |
+
"loss": 0.1471,
|
99 |
+
"step": 150
|
100 |
+
},
|
101 |
+
{
|
102 |
+
"epoch": 0.51,
|
103 |
+
"learning_rate": 0.00026612711236232914,
|
104 |
+
"loss": 0.1427,
|
105 |
+
"step": 160
|
106 |
+
},
|
107 |
+
{
|
108 |
+
"epoch": 0.54,
|
109 |
+
"learning_rate": 0.00023924444174886733,
|
110 |
+
"loss": 0.1366,
|
111 |
+
"step": 170
|
112 |
+
},
|
113 |
+
{
|
114 |
+
"epoch": 0.58,
|
115 |
+
"learning_rate": 0.00021248615041383683,
|
116 |
+
"loss": 0.1463,
|
117 |
+
"step": 180
|
118 |
+
},
|
119 |
+
{
|
120 |
+
"epoch": 0.61,
|
121 |
+
"learning_rate": 0.00018616167622095327,
|
122 |
+
"loss": 0.1328,
|
123 |
+
"step": 190
|
124 |
+
},
|
125 |
+
{
|
126 |
+
"epoch": 0.64,
|
127 |
+
"learning_rate": 0.0001605754402917186,
|
128 |
+
"loss": 0.134,
|
129 |
+
"step": 200
|
130 |
+
},
|
131 |
+
{
|
132 |
+
"epoch": 0.67,
|
133 |
+
"learning_rate": 0.00013602332662288535,
|
134 |
+
"loss": 0.1417,
|
135 |
+
"step": 210
|
136 |
+
},
|
137 |
+
{
|
138 |
+
"epoch": 0.7,
|
139 |
+
"learning_rate": 0.00011278926042882027,
|
140 |
+
"loss": 0.1194,
|
141 |
+
"step": 220
|
142 |
+
},
|
143 |
+
{
|
144 |
+
"epoch": 0.74,
|
145 |
+
"learning_rate": 9.114192477745568e-05,
|
146 |
+
"loss": 0.1359,
|
147 |
+
"step": 230
|
148 |
+
},
|
149 |
+
{
|
150 |
+
"epoch": 0.77,
|
151 |
+
"learning_rate": 7.133165348925977e-05,
|
152 |
+
"loss": 0.1244,
|
153 |
+
"step": 240
|
154 |
+
},
|
155 |
+
{
|
156 |
+
"epoch": 0.8,
|
157 |
+
"learning_rate": 5.3587536230321364e-05,
|
158 |
+
"loss": 0.1367,
|
159 |
+
"step": 250
|
160 |
+
},
|
161 |
+
{
|
162 |
+
"epoch": 0.83,
|
163 |
+
"learning_rate": 3.811476927679228e-05,
|
164 |
+
"loss": 0.1256,
|
165 |
+
"step": 260
|
166 |
+
},
|
167 |
+
{
|
168 |
+
"epoch": 0.86,
|
169 |
+
"learning_rate": 2.5092282586939186e-05,
|
170 |
+
"loss": 0.1312,
|
171 |
+
"step": 270
|
172 |
+
},
|
173 |
+
{
|
174 |
+
"epoch": 0.9,
|
175 |
+
"learning_rate": 1.467067062178823e-05,
|
176 |
+
"loss": 0.1272,
|
177 |
+
"step": 280
|
178 |
+
},
|
179 |
+
{
|
180 |
+
"epoch": 0.93,
|
181 |
+
"learning_rate": 6.970450842734649e-06,
|
182 |
+
"loss": 0.1241,
|
183 |
+
"step": 290
|
184 |
+
},
|
185 |
+
{
|
186 |
+
"epoch": 0.96,
|
187 |
+
"learning_rate": 2.0806700251775058e-06,
|
188 |
+
"loss": 0.1234,
|
189 |
+
"step": 300
|
190 |
+
},
|
191 |
+
{
|
192 |
+
"epoch": 0.99,
|
193 |
+
"learning_rate": 5.7874505027283307e-08,
|
194 |
+
"loss": 0.1305,
|
195 |
+
"step": 310
|
196 |
+
},
|
197 |
+
{
|
198 |
+
"epoch": 1.0,
|
199 |
+
"step": 312,
|
200 |
+
"total_flos": 8.46080087312302e+16,
|
201 |
+
"train_loss": 0.16286772943268985,
|
202 |
+
"train_runtime": 1280.7232,
|
203 |
+
"train_samples_per_second": 3.904,
|
204 |
+
"train_steps_per_second": 0.244
|
205 |
+
}
|
206 |
+
],
|
207 |
+
"logging_steps": 10,
|
208 |
+
"max_steps": 312,
|
209 |
+
"num_input_tokens_seen": 0,
|
210 |
+
"num_train_epochs": 1,
|
211 |
+
"save_steps": 100,
|
212 |
+
"total_flos": 8.46080087312302e+16,
|
213 |
+
"train_batch_size": 4,
|
214 |
+
"trial_name": null,
|
215 |
+
"trial_params": null
|
216 |
+
}
|
training_args.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:7c16dec8e3498a9cca28ece05614b36ebe1dc05c1d57240fdb006563e36eac2f
|
3 |
+
size 4920
|
training_loss.png
ADDED