End of training
Browse files- README.md +110 -0
- adapter_config.json +31 -0
- adapter_model.safetensors +3 -0
- added_tokens.json +3 -0
- all_results.json +18 -0
- eval_results.json +7 -0
- metrics.json +1 -0
- special_tokens_map.json +30 -0
- tokenizer.json +0 -0
- tokenizer.model +3 -0
- tokenizer_config.json +51 -0
- train_results.json +7 -0
- trainer_state.json +690 -0
- training_args.bin +3 -0
README.md
ADDED
@@ -0,0 +1,110 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
---
|
2 |
+
license: apache-2.0
|
3 |
+
library_name: peft
|
4 |
+
tags:
|
5 |
+
- generated_from_trainer
|
6 |
+
base_model: mistralai/Mistral-7B-v0.1
|
7 |
+
datasets:
|
8 |
+
- mbe
|
9 |
+
metrics:
|
10 |
+
- accuracy
|
11 |
+
model-index:
|
12 |
+
- name: Mistral-7B-v0.1_mbe_no
|
13 |
+
results: []
|
14 |
+
---
|
15 |
+
|
16 |
+
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
17 |
+
should probably proofread and complete it, then remove this comment. -->
|
18 |
+
|
19 |
+
# Mistral-7B-v0.1_mbe_no
|
20 |
+
|
21 |
+
This model is a fine-tuned version of [mistralai/Mistral-7B-v0.1](https://huggingface.co/mistralai/Mistral-7B-v0.1) on the mbe dataset.
|
22 |
+
It achieves the following results on the evaluation set:
|
23 |
+
- Loss: 0.7827
|
24 |
+
- Accuracy: 0.5493
|
25 |
+
|
26 |
+
## Model description
|
27 |
+
|
28 |
+
More information needed
|
29 |
+
|
30 |
+
## Intended uses & limitations
|
31 |
+
|
32 |
+
More information needed
|
33 |
+
|
34 |
+
## Training and evaluation data
|
35 |
+
|
36 |
+
More information needed
|
37 |
+
|
38 |
+
## Training procedure
|
39 |
+
|
40 |
+
### Training hyperparameters
|
41 |
+
|
42 |
+
The following hyperparameters were used during training:
|
43 |
+
- learning_rate: 3e-05
|
44 |
+
- train_batch_size: 4
|
45 |
+
- eval_batch_size: 4
|
46 |
+
- seed: 42
|
47 |
+
- gradient_accumulation_steps: 4
|
48 |
+
- total_train_batch_size: 16
|
49 |
+
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
50 |
+
- lr_scheduler_type: constant
|
51 |
+
- lr_scheduler_warmup_ratio: 0.03
|
52 |
+
- num_epochs: 3.0
|
53 |
+
|
54 |
+
### Training results
|
55 |
+
|
56 |
+
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|
57 |
+
|:-------------:|:-----:|:----:|:---------------:|:--------:|
|
58 |
+
| 1.5293 | 0.07 | 10 | 0.6504 | 0.3520 |
|
59 |
+
| 0.6652 | 0.13 | 20 | 0.6469 | 0.3783 |
|
60 |
+
| 0.6523 | 0.2 | 30 | 0.6430 | 0.3651 |
|
61 |
+
| 0.613 | 0.27 | 40 | 0.6341 | 0.4079 |
|
62 |
+
| 0.6586 | 0.33 | 50 | 0.6206 | 0.3882 |
|
63 |
+
| 0.586 | 0.4 | 60 | 0.6269 | 0.4178 |
|
64 |
+
| 0.594 | 0.47 | 70 | 0.6046 | 0.4276 |
|
65 |
+
| 0.6063 | 0.53 | 80 | 0.6135 | 0.4178 |
|
66 |
+
| 0.5988 | 0.6 | 90 | 0.6097 | 0.4276 |
|
67 |
+
| 0.6217 | 0.67 | 100 | 0.6098 | 0.4539 |
|
68 |
+
| 0.5817 | 0.73 | 110 | 0.6022 | 0.4539 |
|
69 |
+
| 0.6219 | 0.8 | 120 | 0.5926 | 0.4572 |
|
70 |
+
| 0.559 | 0.87 | 130 | 0.5816 | 0.4605 |
|
71 |
+
| 0.5514 | 0.93 | 140 | 0.5783 | 0.4737 |
|
72 |
+
| 0.59 | 1.0 | 150 | 0.5622 | 0.4868 |
|
73 |
+
| 0.46 | 1.07 | 160 | 0.5868 | 0.4803 |
|
74 |
+
| 0.4484 | 1.14 | 170 | 0.5667 | 0.4868 |
|
75 |
+
| 0.4162 | 1.2 | 180 | 0.5820 | 0.4803 |
|
76 |
+
| 0.4716 | 1.27 | 190 | 0.5904 | 0.4638 |
|
77 |
+
| 0.4486 | 1.34 | 200 | 0.5777 | 0.5099 |
|
78 |
+
| 0.4264 | 1.4 | 210 | 0.6482 | 0.4967 |
|
79 |
+
| 0.4236 | 1.47 | 220 | 0.5741 | 0.5033 |
|
80 |
+
| 0.4141 | 1.54 | 230 | 0.5608 | 0.5164 |
|
81 |
+
| 0.4308 | 1.6 | 240 | 0.5539 | 0.5099 |
|
82 |
+
| 0.4505 | 1.67 | 250 | 0.5495 | 0.5033 |
|
83 |
+
| 0.3958 | 1.74 | 260 | 0.5594 | 0.5099 |
|
84 |
+
| 0.4432 | 1.8 | 270 | 0.5492 | 0.5164 |
|
85 |
+
| 0.4067 | 1.87 | 280 | 0.6024 | 0.5066 |
|
86 |
+
| 0.3988 | 1.94 | 290 | 0.5607 | 0.5099 |
|
87 |
+
| 0.3992 | 2.0 | 300 | 0.5670 | 0.5164 |
|
88 |
+
| 0.2304 | 2.07 | 310 | 0.8200 | 0.5362 |
|
89 |
+
| 0.1696 | 2.14 | 320 | 0.9087 | 0.5296 |
|
90 |
+
| 0.2255 | 2.2 | 330 | 0.7566 | 0.5362 |
|
91 |
+
| 0.1923 | 2.27 | 340 | 0.7020 | 0.5197 |
|
92 |
+
| 0.281 | 2.34 | 350 | 0.6653 | 0.5033 |
|
93 |
+
| 0.2311 | 2.4 | 360 | 0.6412 | 0.5132 |
|
94 |
+
| 0.1523 | 2.47 | 370 | 0.8846 | 0.5230 |
|
95 |
+
| 0.2451 | 2.54 | 380 | 0.9252 | 0.5164 |
|
96 |
+
| 0.2022 | 2.6 | 390 | 0.7422 | 0.5197 |
|
97 |
+
| 0.217 | 2.67 | 400 | 0.7558 | 0.5329 |
|
98 |
+
| 0.165 | 2.74 | 410 | 0.7846 | 0.5428 |
|
99 |
+
| 0.2025 | 2.8 | 420 | 0.7254 | 0.5230 |
|
100 |
+
| 0.2201 | 2.87 | 430 | 0.6531 | 0.5296 |
|
101 |
+
| 0.2037 | 2.94 | 440 | 0.7827 | 0.5493 |
|
102 |
+
|
103 |
+
|
104 |
+
### Framework versions
|
105 |
+
|
106 |
+
- PEFT 0.7.1
|
107 |
+
- Transformers 4.37.2
|
108 |
+
- Pytorch 2.1.2+cu121
|
109 |
+
- Datasets 2.17.1
|
110 |
+
- Tokenizers 0.15.1
|
adapter_config.json
ADDED
@@ -0,0 +1,31 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
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": 0.1,
|
14 |
+
"megatron_config": null,
|
15 |
+
"megatron_core": "megatron.core",
|
16 |
+
"modules_to_save": null,
|
17 |
+
"peft_type": "LORA",
|
18 |
+
"r": 64,
|
19 |
+
"rank_pattern": {},
|
20 |
+
"revision": null,
|
21 |
+
"target_modules": [
|
22 |
+
"gate_proj",
|
23 |
+
"k_proj",
|
24 |
+
"o_proj",
|
25 |
+
"q_proj",
|
26 |
+
"up_proj",
|
27 |
+
"v_proj",
|
28 |
+
"down_proj"
|
29 |
+
],
|
30 |
+
"task_type": "CAUSAL_LM"
|
31 |
+
}
|
adapter_model.safetensors
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:87e2c9555bace0e632ac55bd4001d73139fb9ed91755fda1410bed2f02808894
|
3 |
+
size 335605144
|
added_tokens.json
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"[PAD]": 32000
|
3 |
+
}
|
all_results.json
ADDED
@@ -0,0 +1,18 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 2.98,
|
3 |
+
"eval_accuracy": 0.8141509433962264,
|
4 |
+
"eval_f1_macro": 0.7097996478763092,
|
5 |
+
"eval_f1_micro": 0.8141509433962264,
|
6 |
+
"eval_loss": 1.1632015705108643,
|
7 |
+
"eval_precision_macro": 0.7222302630120379,
|
8 |
+
"eval_precision_micro": 0.8141509433962264,
|
9 |
+
"eval_recall_macro": 0.7125706602249756,
|
10 |
+
"eval_recall_micro": 0.8141509433962264,
|
11 |
+
"eval_runtime": 19.7133,
|
12 |
+
"eval_samples_per_second": 25.009,
|
13 |
+
"eval_steps_per_second": 1.573,
|
14 |
+
"train_loss": 0.43748926956381573,
|
15 |
+
"train_runtime": 1886.9121,
|
16 |
+
"train_samples_per_second": 3.809,
|
17 |
+
"train_steps_per_second": 0.237
|
18 |
+
}
|
eval_results.json
ADDED
@@ -0,0 +1,7 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 2.0,
|
3 |
+
"eval_loss": 1.1632015705108643,
|
4 |
+
"eval_runtime": 19.7133,
|
5 |
+
"eval_samples_per_second": 25.009,
|
6 |
+
"eval_steps_per_second": 1.573
|
7 |
+
}
|
metrics.json
ADDED
@@ -0,0 +1 @@
|
|
|
|
|
1 |
+
{"run_name": "./output", "train_runtime": 10306.9039, "train_samples_per_second": 9.463, "train_steps_per_second": 0.148, "train_loss": 1.1488912840840697, "epoch": 2.0, "eval_loss": 1.1632015705108643, "eval_runtime": 19.7133, "eval_samples_per_second": 25.009, "eval_steps_per_second": 1.573}
|
special_tokens_map.json
ADDED
@@ -0,0 +1,30 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
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": {
|
17 |
+
"content": "[PAD]",
|
18 |
+
"lstrip": false,
|
19 |
+
"normalized": false,
|
20 |
+
"rstrip": false,
|
21 |
+
"single_word": false
|
22 |
+
},
|
23 |
+
"unk_token": {
|
24 |
+
"content": "<unk>",
|
25 |
+
"lstrip": false,
|
26 |
+
"normalized": false,
|
27 |
+
"rstrip": false,
|
28 |
+
"single_word": false
|
29 |
+
}
|
30 |
+
}
|
tokenizer.json
ADDED
The diff for this file is too large to render.
See raw diff
|
|
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 +1,51 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
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 |
+
"32000": {
|
30 |
+
"content": "[PAD]",
|
31 |
+
"lstrip": false,
|
32 |
+
"normalized": false,
|
33 |
+
"rstrip": false,
|
34 |
+
"single_word": false,
|
35 |
+
"special": true
|
36 |
+
}
|
37 |
+
},
|
38 |
+
"additional_special_tokens": [],
|
39 |
+
"bos_token": "<s>",
|
40 |
+
"clean_up_tokenization_spaces": false,
|
41 |
+
"eos_token": "</s>",
|
42 |
+
"legacy": true,
|
43 |
+
"model_max_length": 1000000000000000019884624838656,
|
44 |
+
"pad_token": "[PAD]",
|
45 |
+
"padding_side": "left",
|
46 |
+
"sp_model_kwargs": {},
|
47 |
+
"spaces_between_special_tokens": false,
|
48 |
+
"tokenizer_class": "LlamaTokenizer",
|
49 |
+
"unk_token": "<unk>",
|
50 |
+
"use_default_system_prompt": false
|
51 |
+
}
|
train_results.json
ADDED
@@ -0,0 +1,7 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 2.98,
|
3 |
+
"train_loss": 0.43748926956381573,
|
4 |
+
"train_runtime": 1886.9121,
|
5 |
+
"train_samples_per_second": 3.809,
|
6 |
+
"train_steps_per_second": 0.237
|
7 |
+
}
|
trainer_state.json
ADDED
@@ -0,0 +1,690 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"best_metric": null,
|
3 |
+
"best_model_checkpoint": null,
|
4 |
+
"epoch": 2.984974958263773,
|
5 |
+
"eval_steps": 10,
|
6 |
+
"global_step": 447,
|
7 |
+
"is_hyper_param_search": false,
|
8 |
+
"is_local_process_zero": true,
|
9 |
+
"is_world_process_zero": true,
|
10 |
+
"log_history": [
|
11 |
+
{
|
12 |
+
"epoch": 0.07,
|
13 |
+
"learning_rate": 3e-05,
|
14 |
+
"loss": 1.5293,
|
15 |
+
"step": 10
|
16 |
+
},
|
17 |
+
{
|
18 |
+
"epoch": 0.07,
|
19 |
+
"eval_accuracy": 0.3519736842105263,
|
20 |
+
"eval_loss": 0.6503910422325134,
|
21 |
+
"eval_runtime": 16.3619,
|
22 |
+
"eval_samples_per_second": 18.58,
|
23 |
+
"eval_steps_per_second": 4.645,
|
24 |
+
"step": 10
|
25 |
+
},
|
26 |
+
{
|
27 |
+
"epoch": 0.13,
|
28 |
+
"learning_rate": 3e-05,
|
29 |
+
"loss": 0.6652,
|
30 |
+
"step": 20
|
31 |
+
},
|
32 |
+
{
|
33 |
+
"epoch": 0.13,
|
34 |
+
"eval_accuracy": 0.3782894736842105,
|
35 |
+
"eval_loss": 0.6469289660453796,
|
36 |
+
"eval_runtime": 16.3596,
|
37 |
+
"eval_samples_per_second": 18.582,
|
38 |
+
"eval_steps_per_second": 4.646,
|
39 |
+
"step": 20
|
40 |
+
},
|
41 |
+
{
|
42 |
+
"epoch": 0.2,
|
43 |
+
"learning_rate": 3e-05,
|
44 |
+
"loss": 0.6523,
|
45 |
+
"step": 30
|
46 |
+
},
|
47 |
+
{
|
48 |
+
"epoch": 0.2,
|
49 |
+
"eval_accuracy": 0.3651315789473684,
|
50 |
+
"eval_loss": 0.6429744362831116,
|
51 |
+
"eval_runtime": 16.3534,
|
52 |
+
"eval_samples_per_second": 18.589,
|
53 |
+
"eval_steps_per_second": 4.647,
|
54 |
+
"step": 30
|
55 |
+
},
|
56 |
+
{
|
57 |
+
"epoch": 0.27,
|
58 |
+
"learning_rate": 3e-05,
|
59 |
+
"loss": 0.613,
|
60 |
+
"step": 40
|
61 |
+
},
|
62 |
+
{
|
63 |
+
"epoch": 0.27,
|
64 |
+
"eval_accuracy": 0.40789473684210525,
|
65 |
+
"eval_loss": 0.6341487765312195,
|
66 |
+
"eval_runtime": 16.4446,
|
67 |
+
"eval_samples_per_second": 18.486,
|
68 |
+
"eval_steps_per_second": 4.622,
|
69 |
+
"step": 40
|
70 |
+
},
|
71 |
+
{
|
72 |
+
"epoch": 0.33,
|
73 |
+
"learning_rate": 3e-05,
|
74 |
+
"loss": 0.6586,
|
75 |
+
"step": 50
|
76 |
+
},
|
77 |
+
{
|
78 |
+
"epoch": 0.33,
|
79 |
+
"eval_accuracy": 0.3881578947368421,
|
80 |
+
"eval_loss": 0.6206462979316711,
|
81 |
+
"eval_runtime": 16.366,
|
82 |
+
"eval_samples_per_second": 18.575,
|
83 |
+
"eval_steps_per_second": 4.644,
|
84 |
+
"step": 50
|
85 |
+
},
|
86 |
+
{
|
87 |
+
"epoch": 0.4,
|
88 |
+
"learning_rate": 3e-05,
|
89 |
+
"loss": 0.586,
|
90 |
+
"step": 60
|
91 |
+
},
|
92 |
+
{
|
93 |
+
"epoch": 0.4,
|
94 |
+
"eval_accuracy": 0.41776315789473684,
|
95 |
+
"eval_loss": 0.6268599033355713,
|
96 |
+
"eval_runtime": 16.3556,
|
97 |
+
"eval_samples_per_second": 18.587,
|
98 |
+
"eval_steps_per_second": 4.647,
|
99 |
+
"step": 60
|
100 |
+
},
|
101 |
+
{
|
102 |
+
"epoch": 0.47,
|
103 |
+
"learning_rate": 3e-05,
|
104 |
+
"loss": 0.594,
|
105 |
+
"step": 70
|
106 |
+
},
|
107 |
+
{
|
108 |
+
"epoch": 0.47,
|
109 |
+
"eval_accuracy": 0.4276315789473684,
|
110 |
+
"eval_loss": 0.6045505404472351,
|
111 |
+
"eval_runtime": 16.3658,
|
112 |
+
"eval_samples_per_second": 18.575,
|
113 |
+
"eval_steps_per_second": 4.644,
|
114 |
+
"step": 70
|
115 |
+
},
|
116 |
+
{
|
117 |
+
"epoch": 0.53,
|
118 |
+
"learning_rate": 3e-05,
|
119 |
+
"loss": 0.6063,
|
120 |
+
"step": 80
|
121 |
+
},
|
122 |
+
{
|
123 |
+
"epoch": 0.53,
|
124 |
+
"eval_accuracy": 0.41776315789473684,
|
125 |
+
"eval_loss": 0.6135305762290955,
|
126 |
+
"eval_runtime": 16.3648,
|
127 |
+
"eval_samples_per_second": 18.576,
|
128 |
+
"eval_steps_per_second": 4.644,
|
129 |
+
"step": 80
|
130 |
+
},
|
131 |
+
{
|
132 |
+
"epoch": 0.6,
|
133 |
+
"learning_rate": 3e-05,
|
134 |
+
"loss": 0.5988,
|
135 |
+
"step": 90
|
136 |
+
},
|
137 |
+
{
|
138 |
+
"epoch": 0.6,
|
139 |
+
"eval_accuracy": 0.4276315789473684,
|
140 |
+
"eval_loss": 0.6097424626350403,
|
141 |
+
"eval_runtime": 16.3755,
|
142 |
+
"eval_samples_per_second": 18.564,
|
143 |
+
"eval_steps_per_second": 4.641,
|
144 |
+
"step": 90
|
145 |
+
},
|
146 |
+
{
|
147 |
+
"epoch": 0.67,
|
148 |
+
"learning_rate": 3e-05,
|
149 |
+
"loss": 0.6217,
|
150 |
+
"step": 100
|
151 |
+
},
|
152 |
+
{
|
153 |
+
"epoch": 0.67,
|
154 |
+
"eval_accuracy": 0.45394736842105265,
|
155 |
+
"eval_loss": 0.6098220348358154,
|
156 |
+
"eval_runtime": 16.3674,
|
157 |
+
"eval_samples_per_second": 18.574,
|
158 |
+
"eval_steps_per_second": 4.643,
|
159 |
+
"step": 100
|
160 |
+
},
|
161 |
+
{
|
162 |
+
"epoch": 0.73,
|
163 |
+
"learning_rate": 3e-05,
|
164 |
+
"loss": 0.5817,
|
165 |
+
"step": 110
|
166 |
+
},
|
167 |
+
{
|
168 |
+
"epoch": 0.73,
|
169 |
+
"eval_accuracy": 0.45394736842105265,
|
170 |
+
"eval_loss": 0.6021943092346191,
|
171 |
+
"eval_runtime": 16.3678,
|
172 |
+
"eval_samples_per_second": 18.573,
|
173 |
+
"eval_steps_per_second": 4.643,
|
174 |
+
"step": 110
|
175 |
+
},
|
176 |
+
{
|
177 |
+
"epoch": 0.8,
|
178 |
+
"learning_rate": 3e-05,
|
179 |
+
"loss": 0.6219,
|
180 |
+
"step": 120
|
181 |
+
},
|
182 |
+
{
|
183 |
+
"epoch": 0.8,
|
184 |
+
"eval_accuracy": 0.45723684210526316,
|
185 |
+
"eval_loss": 0.5926041007041931,
|
186 |
+
"eval_runtime": 16.3627,
|
187 |
+
"eval_samples_per_second": 18.579,
|
188 |
+
"eval_steps_per_second": 4.645,
|
189 |
+
"step": 120
|
190 |
+
},
|
191 |
+
{
|
192 |
+
"epoch": 0.87,
|
193 |
+
"learning_rate": 3e-05,
|
194 |
+
"loss": 0.559,
|
195 |
+
"step": 130
|
196 |
+
},
|
197 |
+
{
|
198 |
+
"epoch": 0.87,
|
199 |
+
"eval_accuracy": 0.4605263157894737,
|
200 |
+
"eval_loss": 0.5816267728805542,
|
201 |
+
"eval_runtime": 16.3673,
|
202 |
+
"eval_samples_per_second": 18.574,
|
203 |
+
"eval_steps_per_second": 4.643,
|
204 |
+
"step": 130
|
205 |
+
},
|
206 |
+
{
|
207 |
+
"epoch": 0.93,
|
208 |
+
"learning_rate": 3e-05,
|
209 |
+
"loss": 0.5514,
|
210 |
+
"step": 140
|
211 |
+
},
|
212 |
+
{
|
213 |
+
"epoch": 0.93,
|
214 |
+
"eval_accuracy": 0.47368421052631576,
|
215 |
+
"eval_loss": 0.5783373713493347,
|
216 |
+
"eval_runtime": 16.368,
|
217 |
+
"eval_samples_per_second": 18.573,
|
218 |
+
"eval_steps_per_second": 4.643,
|
219 |
+
"step": 140
|
220 |
+
},
|
221 |
+
{
|
222 |
+
"epoch": 1.0,
|
223 |
+
"learning_rate": 3e-05,
|
224 |
+
"loss": 0.59,
|
225 |
+
"step": 150
|
226 |
+
},
|
227 |
+
{
|
228 |
+
"epoch": 1.0,
|
229 |
+
"eval_accuracy": 0.4868421052631579,
|
230 |
+
"eval_loss": 0.5621668100357056,
|
231 |
+
"eval_runtime": 16.3648,
|
232 |
+
"eval_samples_per_second": 18.576,
|
233 |
+
"eval_steps_per_second": 4.644,
|
234 |
+
"step": 150
|
235 |
+
},
|
236 |
+
{
|
237 |
+
"epoch": 1.07,
|
238 |
+
"learning_rate": 3e-05,
|
239 |
+
"loss": 0.46,
|
240 |
+
"step": 160
|
241 |
+
},
|
242 |
+
{
|
243 |
+
"epoch": 1.07,
|
244 |
+
"eval_accuracy": 0.48026315789473684,
|
245 |
+
"eval_loss": 0.5867581367492676,
|
246 |
+
"eval_runtime": 16.3579,
|
247 |
+
"eval_samples_per_second": 18.584,
|
248 |
+
"eval_steps_per_second": 4.646,
|
249 |
+
"step": 160
|
250 |
+
},
|
251 |
+
{
|
252 |
+
"epoch": 1.14,
|
253 |
+
"learning_rate": 3e-05,
|
254 |
+
"loss": 0.4484,
|
255 |
+
"step": 170
|
256 |
+
},
|
257 |
+
{
|
258 |
+
"epoch": 1.14,
|
259 |
+
"eval_accuracy": 0.4868421052631579,
|
260 |
+
"eval_loss": 0.5666611194610596,
|
261 |
+
"eval_runtime": 16.3613,
|
262 |
+
"eval_samples_per_second": 18.58,
|
263 |
+
"eval_steps_per_second": 4.645,
|
264 |
+
"step": 170
|
265 |
+
},
|
266 |
+
{
|
267 |
+
"epoch": 1.2,
|
268 |
+
"learning_rate": 3e-05,
|
269 |
+
"loss": 0.4162,
|
270 |
+
"step": 180
|
271 |
+
},
|
272 |
+
{
|
273 |
+
"epoch": 1.2,
|
274 |
+
"eval_accuracy": 0.48026315789473684,
|
275 |
+
"eval_loss": 0.5819750428199768,
|
276 |
+
"eval_runtime": 16.3725,
|
277 |
+
"eval_samples_per_second": 18.568,
|
278 |
+
"eval_steps_per_second": 4.642,
|
279 |
+
"step": 180
|
280 |
+
},
|
281 |
+
{
|
282 |
+
"epoch": 1.27,
|
283 |
+
"learning_rate": 3e-05,
|
284 |
+
"loss": 0.4716,
|
285 |
+
"step": 190
|
286 |
+
},
|
287 |
+
{
|
288 |
+
"epoch": 1.27,
|
289 |
+
"eval_accuracy": 0.46381578947368424,
|
290 |
+
"eval_loss": 0.590432345867157,
|
291 |
+
"eval_runtime": 16.361,
|
292 |
+
"eval_samples_per_second": 18.581,
|
293 |
+
"eval_steps_per_second": 4.645,
|
294 |
+
"step": 190
|
295 |
+
},
|
296 |
+
{
|
297 |
+
"epoch": 1.34,
|
298 |
+
"learning_rate": 3e-05,
|
299 |
+
"loss": 0.4486,
|
300 |
+
"step": 200
|
301 |
+
},
|
302 |
+
{
|
303 |
+
"epoch": 1.34,
|
304 |
+
"eval_accuracy": 0.5098684210526315,
|
305 |
+
"eval_loss": 0.5777420997619629,
|
306 |
+
"eval_runtime": 16.354,
|
307 |
+
"eval_samples_per_second": 18.589,
|
308 |
+
"eval_steps_per_second": 4.647,
|
309 |
+
"step": 200
|
310 |
+
},
|
311 |
+
{
|
312 |
+
"epoch": 1.4,
|
313 |
+
"learning_rate": 3e-05,
|
314 |
+
"loss": 0.4264,
|
315 |
+
"step": 210
|
316 |
+
},
|
317 |
+
{
|
318 |
+
"epoch": 1.4,
|
319 |
+
"eval_accuracy": 0.4967105263157895,
|
320 |
+
"eval_loss": 0.6482496857643127,
|
321 |
+
"eval_runtime": 16.3696,
|
322 |
+
"eval_samples_per_second": 18.571,
|
323 |
+
"eval_steps_per_second": 4.643,
|
324 |
+
"step": 210
|
325 |
+
},
|
326 |
+
{
|
327 |
+
"epoch": 1.47,
|
328 |
+
"learning_rate": 3e-05,
|
329 |
+
"loss": 0.4236,
|
330 |
+
"step": 220
|
331 |
+
},
|
332 |
+
{
|
333 |
+
"epoch": 1.47,
|
334 |
+
"eval_accuracy": 0.5032894736842105,
|
335 |
+
"eval_loss": 0.5741015076637268,
|
336 |
+
"eval_runtime": 16.3648,
|
337 |
+
"eval_samples_per_second": 18.576,
|
338 |
+
"eval_steps_per_second": 4.644,
|
339 |
+
"step": 220
|
340 |
+
},
|
341 |
+
{
|
342 |
+
"epoch": 1.54,
|
343 |
+
"learning_rate": 3e-05,
|
344 |
+
"loss": 0.4141,
|
345 |
+
"step": 230
|
346 |
+
},
|
347 |
+
{
|
348 |
+
"epoch": 1.54,
|
349 |
+
"eval_accuracy": 0.5164473684210527,
|
350 |
+
"eval_loss": 0.5607666373252869,
|
351 |
+
"eval_runtime": 16.3604,
|
352 |
+
"eval_samples_per_second": 18.581,
|
353 |
+
"eval_steps_per_second": 4.645,
|
354 |
+
"step": 230
|
355 |
+
},
|
356 |
+
{
|
357 |
+
"epoch": 1.6,
|
358 |
+
"learning_rate": 3e-05,
|
359 |
+
"loss": 0.4308,
|
360 |
+
"step": 240
|
361 |
+
},
|
362 |
+
{
|
363 |
+
"epoch": 1.6,
|
364 |
+
"eval_accuracy": 0.5098684210526315,
|
365 |
+
"eval_loss": 0.5539161562919617,
|
366 |
+
"eval_runtime": 16.3597,
|
367 |
+
"eval_samples_per_second": 18.582,
|
368 |
+
"eval_steps_per_second": 4.646,
|
369 |
+
"step": 240
|
370 |
+
},
|
371 |
+
{
|
372 |
+
"epoch": 1.67,
|
373 |
+
"learning_rate": 3e-05,
|
374 |
+
"loss": 0.4505,
|
375 |
+
"step": 250
|
376 |
+
},
|
377 |
+
{
|
378 |
+
"epoch": 1.67,
|
379 |
+
"eval_accuracy": 0.5032894736842105,
|
380 |
+
"eval_loss": 0.5494562387466431,
|
381 |
+
"eval_runtime": 16.3671,
|
382 |
+
"eval_samples_per_second": 18.574,
|
383 |
+
"eval_steps_per_second": 4.643,
|
384 |
+
"step": 250
|
385 |
+
},
|
386 |
+
{
|
387 |
+
"epoch": 1.74,
|
388 |
+
"learning_rate": 3e-05,
|
389 |
+
"loss": 0.3958,
|
390 |
+
"step": 260
|
391 |
+
},
|
392 |
+
{
|
393 |
+
"epoch": 1.74,
|
394 |
+
"eval_accuracy": 0.5098684210526315,
|
395 |
+
"eval_loss": 0.5593812465667725,
|
396 |
+
"eval_runtime": 16.3598,
|
397 |
+
"eval_samples_per_second": 18.582,
|
398 |
+
"eval_steps_per_second": 4.646,
|
399 |
+
"step": 260
|
400 |
+
},
|
401 |
+
{
|
402 |
+
"epoch": 1.8,
|
403 |
+
"learning_rate": 3e-05,
|
404 |
+
"loss": 0.4432,
|
405 |
+
"step": 270
|
406 |
+
},
|
407 |
+
{
|
408 |
+
"epoch": 1.8,
|
409 |
+
"eval_accuracy": 0.5164473684210527,
|
410 |
+
"eval_loss": 0.5492013096809387,
|
411 |
+
"eval_runtime": 16.3635,
|
412 |
+
"eval_samples_per_second": 18.578,
|
413 |
+
"eval_steps_per_second": 4.644,
|
414 |
+
"step": 270
|
415 |
+
},
|
416 |
+
{
|
417 |
+
"epoch": 1.87,
|
418 |
+
"learning_rate": 3e-05,
|
419 |
+
"loss": 0.4067,
|
420 |
+
"step": 280
|
421 |
+
},
|
422 |
+
{
|
423 |
+
"epoch": 1.87,
|
424 |
+
"eval_accuracy": 0.506578947368421,
|
425 |
+
"eval_loss": 0.6023809313774109,
|
426 |
+
"eval_runtime": 16.3658,
|
427 |
+
"eval_samples_per_second": 18.575,
|
428 |
+
"eval_steps_per_second": 4.644,
|
429 |
+
"step": 280
|
430 |
+
},
|
431 |
+
{
|
432 |
+
"epoch": 1.94,
|
433 |
+
"learning_rate": 3e-05,
|
434 |
+
"loss": 0.3988,
|
435 |
+
"step": 290
|
436 |
+
},
|
437 |
+
{
|
438 |
+
"epoch": 1.94,
|
439 |
+
"eval_accuracy": 0.5098684210526315,
|
440 |
+
"eval_loss": 0.5606762766838074,
|
441 |
+
"eval_runtime": 16.3642,
|
442 |
+
"eval_samples_per_second": 18.577,
|
443 |
+
"eval_steps_per_second": 4.644,
|
444 |
+
"step": 290
|
445 |
+
},
|
446 |
+
{
|
447 |
+
"epoch": 2.0,
|
448 |
+
"learning_rate": 3e-05,
|
449 |
+
"loss": 0.3992,
|
450 |
+
"step": 300
|
451 |
+
},
|
452 |
+
{
|
453 |
+
"epoch": 2.0,
|
454 |
+
"eval_accuracy": 0.5164473684210527,
|
455 |
+
"eval_loss": 0.5669550895690918,
|
456 |
+
"eval_runtime": 16.3688,
|
457 |
+
"eval_samples_per_second": 18.572,
|
458 |
+
"eval_steps_per_second": 4.643,
|
459 |
+
"step": 300
|
460 |
+
},
|
461 |
+
{
|
462 |
+
"epoch": 2.07,
|
463 |
+
"learning_rate": 3e-05,
|
464 |
+
"loss": 0.2304,
|
465 |
+
"step": 310
|
466 |
+
},
|
467 |
+
{
|
468 |
+
"epoch": 2.07,
|
469 |
+
"eval_accuracy": 0.5361842105263158,
|
470 |
+
"eval_loss": 0.8199814558029175,
|
471 |
+
"eval_runtime": 16.3644,
|
472 |
+
"eval_samples_per_second": 18.577,
|
473 |
+
"eval_steps_per_second": 4.644,
|
474 |
+
"step": 310
|
475 |
+
},
|
476 |
+
{
|
477 |
+
"epoch": 2.14,
|
478 |
+
"learning_rate": 3e-05,
|
479 |
+
"loss": 0.1696,
|
480 |
+
"step": 320
|
481 |
+
},
|
482 |
+
{
|
483 |
+
"epoch": 2.14,
|
484 |
+
"eval_accuracy": 0.5296052631578947,
|
485 |
+
"eval_loss": 0.9087279438972473,
|
486 |
+
"eval_runtime": 16.362,
|
487 |
+
"eval_samples_per_second": 18.58,
|
488 |
+
"eval_steps_per_second": 4.645,
|
489 |
+
"step": 320
|
490 |
+
},
|
491 |
+
{
|
492 |
+
"epoch": 2.2,
|
493 |
+
"learning_rate": 3e-05,
|
494 |
+
"loss": 0.2255,
|
495 |
+
"step": 330
|
496 |
+
},
|
497 |
+
{
|
498 |
+
"epoch": 2.2,
|
499 |
+
"eval_accuracy": 0.5361842105263158,
|
500 |
+
"eval_loss": 0.7565640211105347,
|
501 |
+
"eval_runtime": 16.3681,
|
502 |
+
"eval_samples_per_second": 18.573,
|
503 |
+
"eval_steps_per_second": 4.643,
|
504 |
+
"step": 330
|
505 |
+
},
|
506 |
+
{
|
507 |
+
"epoch": 2.27,
|
508 |
+
"learning_rate": 3e-05,
|
509 |
+
"loss": 0.1923,
|
510 |
+
"step": 340
|
511 |
+
},
|
512 |
+
{
|
513 |
+
"epoch": 2.27,
|
514 |
+
"eval_accuracy": 0.5197368421052632,
|
515 |
+
"eval_loss": 0.701989471912384,
|
516 |
+
"eval_runtime": 16.3692,
|
517 |
+
"eval_samples_per_second": 18.571,
|
518 |
+
"eval_steps_per_second": 4.643,
|
519 |
+
"step": 340
|
520 |
+
},
|
521 |
+
{
|
522 |
+
"epoch": 2.34,
|
523 |
+
"learning_rate": 3e-05,
|
524 |
+
"loss": 0.281,
|
525 |
+
"step": 350
|
526 |
+
},
|
527 |
+
{
|
528 |
+
"epoch": 2.34,
|
529 |
+
"eval_accuracy": 0.5032894736842105,
|
530 |
+
"eval_loss": 0.6653422117233276,
|
531 |
+
"eval_runtime": 16.3703,
|
532 |
+
"eval_samples_per_second": 18.57,
|
533 |
+
"eval_steps_per_second": 4.643,
|
534 |
+
"step": 350
|
535 |
+
},
|
536 |
+
{
|
537 |
+
"epoch": 2.4,
|
538 |
+
"learning_rate": 3e-05,
|
539 |
+
"loss": 0.2311,
|
540 |
+
"step": 360
|
541 |
+
},
|
542 |
+
{
|
543 |
+
"epoch": 2.4,
|
544 |
+
"eval_accuracy": 0.5131578947368421,
|
545 |
+
"eval_loss": 0.6411683559417725,
|
546 |
+
"eval_runtime": 16.3702,
|
547 |
+
"eval_samples_per_second": 18.57,
|
548 |
+
"eval_steps_per_second": 4.643,
|
549 |
+
"step": 360
|
550 |
+
},
|
551 |
+
{
|
552 |
+
"epoch": 2.47,
|
553 |
+
"learning_rate": 3e-05,
|
554 |
+
"loss": 0.1523,
|
555 |
+
"step": 370
|
556 |
+
},
|
557 |
+
{
|
558 |
+
"epoch": 2.47,
|
559 |
+
"eval_accuracy": 0.5230263157894737,
|
560 |
+
"eval_loss": 0.8846220970153809,
|
561 |
+
"eval_runtime": 16.3728,
|
562 |
+
"eval_samples_per_second": 18.567,
|
563 |
+
"eval_steps_per_second": 4.642,
|
564 |
+
"step": 370
|
565 |
+
},
|
566 |
+
{
|
567 |
+
"epoch": 2.54,
|
568 |
+
"learning_rate": 3e-05,
|
569 |
+
"loss": 0.2451,
|
570 |
+
"step": 380
|
571 |
+
},
|
572 |
+
{
|
573 |
+
"epoch": 2.54,
|
574 |
+
"eval_accuracy": 0.5164473684210527,
|
575 |
+
"eval_loss": 0.9251819252967834,
|
576 |
+
"eval_runtime": 16.379,
|
577 |
+
"eval_samples_per_second": 18.56,
|
578 |
+
"eval_steps_per_second": 4.64,
|
579 |
+
"step": 380
|
580 |
+
},
|
581 |
+
{
|
582 |
+
"epoch": 2.6,
|
583 |
+
"learning_rate": 3e-05,
|
584 |
+
"loss": 0.2022,
|
585 |
+
"step": 390
|
586 |
+
},
|
587 |
+
{
|
588 |
+
"epoch": 2.6,
|
589 |
+
"eval_accuracy": 0.5197368421052632,
|
590 |
+
"eval_loss": 0.7422206401824951,
|
591 |
+
"eval_runtime": 16.3727,
|
592 |
+
"eval_samples_per_second": 18.567,
|
593 |
+
"eval_steps_per_second": 4.642,
|
594 |
+
"step": 390
|
595 |
+
},
|
596 |
+
{
|
597 |
+
"epoch": 2.67,
|
598 |
+
"learning_rate": 3e-05,
|
599 |
+
"loss": 0.217,
|
600 |
+
"step": 400
|
601 |
+
},
|
602 |
+
{
|
603 |
+
"epoch": 2.67,
|
604 |
+
"eval_accuracy": 0.5328947368421053,
|
605 |
+
"eval_loss": 0.7557851076126099,
|
606 |
+
"eval_runtime": 16.3638,
|
607 |
+
"eval_samples_per_second": 18.578,
|
608 |
+
"eval_steps_per_second": 4.644,
|
609 |
+
"step": 400
|
610 |
+
},
|
611 |
+
{
|
612 |
+
"epoch": 2.74,
|
613 |
+
"learning_rate": 3e-05,
|
614 |
+
"loss": 0.165,
|
615 |
+
"step": 410
|
616 |
+
},
|
617 |
+
{
|
618 |
+
"epoch": 2.74,
|
619 |
+
"eval_accuracy": 0.5427631578947368,
|
620 |
+
"eval_loss": 0.7846018075942993,
|
621 |
+
"eval_runtime": 16.3674,
|
622 |
+
"eval_samples_per_second": 18.574,
|
623 |
+
"eval_steps_per_second": 4.643,
|
624 |
+
"step": 410
|
625 |
+
},
|
626 |
+
{
|
627 |
+
"epoch": 2.8,
|
628 |
+
"learning_rate": 3e-05,
|
629 |
+
"loss": 0.2025,
|
630 |
+
"step": 420
|
631 |
+
},
|
632 |
+
{
|
633 |
+
"epoch": 2.8,
|
634 |
+
"eval_accuracy": 0.5230263157894737,
|
635 |
+
"eval_loss": 0.725389301776886,
|
636 |
+
"eval_runtime": 16.3775,
|
637 |
+
"eval_samples_per_second": 18.562,
|
638 |
+
"eval_steps_per_second": 4.641,
|
639 |
+
"step": 420
|
640 |
+
},
|
641 |
+
{
|
642 |
+
"epoch": 2.87,
|
643 |
+
"learning_rate": 3e-05,
|
644 |
+
"loss": 0.2201,
|
645 |
+
"step": 430
|
646 |
+
},
|
647 |
+
{
|
648 |
+
"epoch": 2.87,
|
649 |
+
"eval_accuracy": 0.5296052631578947,
|
650 |
+
"eval_loss": 0.6530900001525879,
|
651 |
+
"eval_runtime": 16.3694,
|
652 |
+
"eval_samples_per_second": 18.571,
|
653 |
+
"eval_steps_per_second": 4.643,
|
654 |
+
"step": 430
|
655 |
+
},
|
656 |
+
{
|
657 |
+
"epoch": 2.94,
|
658 |
+
"learning_rate": 3e-05,
|
659 |
+
"loss": 0.2037,
|
660 |
+
"step": 440
|
661 |
+
},
|
662 |
+
{
|
663 |
+
"epoch": 2.94,
|
664 |
+
"eval_accuracy": 0.5493421052631579,
|
665 |
+
"eval_loss": 0.7827179431915283,
|
666 |
+
"eval_runtime": 16.3682,
|
667 |
+
"eval_samples_per_second": 18.573,
|
668 |
+
"eval_steps_per_second": 4.643,
|
669 |
+
"step": 440
|
670 |
+
},
|
671 |
+
{
|
672 |
+
"epoch": 2.98,
|
673 |
+
"step": 447,
|
674 |
+
"total_flos": 1.09848374069035e+17,
|
675 |
+
"train_loss": 0.43748926956381573,
|
676 |
+
"train_runtime": 1886.9121,
|
677 |
+
"train_samples_per_second": 3.809,
|
678 |
+
"train_steps_per_second": 0.237
|
679 |
+
}
|
680 |
+
],
|
681 |
+
"logging_steps": 10,
|
682 |
+
"max_steps": 447,
|
683 |
+
"num_input_tokens_seen": 0,
|
684 |
+
"num_train_epochs": 3,
|
685 |
+
"save_steps": 250,
|
686 |
+
"total_flos": 1.09848374069035e+17,
|
687 |
+
"train_batch_size": 4,
|
688 |
+
"trial_name": null,
|
689 |
+
"trial_params": null
|
690 |
+
}
|
training_args.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:103e5da4968375ab36f9593bf410c45d9e405b7af46f4c859ad873d05c786bf3
|
3 |
+
size 6712
|