Model save
Browse files- README.md +73 -0
- adapter_model.safetensors +1 -1
- all_results.json +9 -0
- runs/Apr21_21-04-58_dc69e3879ef9/events.out.tfevents.1713748035.dc69e3879ef9.1614.0 +2 -2
- train_results.json +9 -0
- trainer_state.json +1225 -0
README.md
ADDED
@@ -0,0 +1,73 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
---
|
2 |
+
license: gemma
|
3 |
+
library_name: peft
|
4 |
+
tags:
|
5 |
+
- trl
|
6 |
+
- sft
|
7 |
+
- generated_from_trainer
|
8 |
+
base_model: google/gemma-7b
|
9 |
+
datasets:
|
10 |
+
- generator
|
11 |
+
model-index:
|
12 |
+
- name: coding_llamaduo_result3
|
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 |
+
# coding_llamaduo_result3
|
20 |
+
|
21 |
+
This model is a fine-tuned version of [google/gemma-7b](https://huggingface.co/google/gemma-7b) on the generator dataset.
|
22 |
+
It achieves the following results on the evaluation set:
|
23 |
+
- Loss: 1.2481
|
24 |
+
|
25 |
+
## Model description
|
26 |
+
|
27 |
+
More information needed
|
28 |
+
|
29 |
+
## Intended uses & limitations
|
30 |
+
|
31 |
+
More information needed
|
32 |
+
|
33 |
+
## Training and evaluation data
|
34 |
+
|
35 |
+
More information needed
|
36 |
+
|
37 |
+
## Training procedure
|
38 |
+
|
39 |
+
### Training hyperparameters
|
40 |
+
|
41 |
+
The following hyperparameters were used during training:
|
42 |
+
- learning_rate: 0.0002
|
43 |
+
- train_batch_size: 2
|
44 |
+
- eval_batch_size: 2
|
45 |
+
- seed: 42
|
46 |
+
- distributed_type: multi-GPU
|
47 |
+
- num_devices: 2
|
48 |
+
- gradient_accumulation_steps: 2
|
49 |
+
- total_train_batch_size: 8
|
50 |
+
- total_eval_batch_size: 4
|
51 |
+
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
52 |
+
- lr_scheduler_type: cosine
|
53 |
+
- lr_scheduler_warmup_ratio: 0.1
|
54 |
+
- num_epochs: 5
|
55 |
+
|
56 |
+
### Training results
|
57 |
+
|
58 |
+
| Training Loss | Epoch | Step | Validation Loss |
|
59 |
+
|:-------------:|:-----:|:----:|:---------------:|
|
60 |
+
| 0.7425 | 1.0 | 164 | 1.1824 |
|
61 |
+
| 0.6196 | 2.0 | 328 | 1.1435 |
|
62 |
+
| 0.5124 | 3.0 | 492 | 1.1461 |
|
63 |
+
| 0.4403 | 4.0 | 656 | 1.2068 |
|
64 |
+
| 0.4072 | 5.0 | 820 | 1.2481 |
|
65 |
+
|
66 |
+
|
67 |
+
### Framework versions
|
68 |
+
|
69 |
+
- PEFT 0.7.1
|
70 |
+
- Transformers 4.40.0
|
71 |
+
- Pytorch 2.2.2+cu121
|
72 |
+
- Datasets 2.19.0
|
73 |
+
- Tokenizers 0.19.1
|
adapter_model.safetensors
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
size 100060536
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:f5cb5d2e04f3662c61c87bb8ed46bd690ca56703f41ca1fc33df3c9b54845429
|
3 |
size 100060536
|
all_results.json
ADDED
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 5.0,
|
3 |
+
"total_flos": 6.28853392705323e+17,
|
4 |
+
"train_loss": 1.6822980254161648,
|
5 |
+
"train_runtime": 5646.8844,
|
6 |
+
"train_samples": 19039,
|
7 |
+
"train_samples_per_second": 1.159,
|
8 |
+
"train_steps_per_second": 0.145
|
9 |
+
}
|
runs/Apr21_21-04-58_dc69e3879ef9/events.out.tfevents.1713748035.dc69e3879ef9.1614.0
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
-
size
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:2328979267891e1a7cb2d0c857659e22e2ef94d8567132cc9eb7b29bd4f9aad4
|
3 |
+
size 41834
|
train_results.json
ADDED
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 5.0,
|
3 |
+
"total_flos": 6.28853392705323e+17,
|
4 |
+
"train_loss": 1.6822980254161648,
|
5 |
+
"train_runtime": 5646.8844,
|
6 |
+
"train_samples": 19039,
|
7 |
+
"train_samples_per_second": 1.159,
|
8 |
+
"train_steps_per_second": 0.145
|
9 |
+
}
|
trainer_state.json
ADDED
@@ -0,0 +1,1225 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"best_metric": null,
|
3 |
+
"best_model_checkpoint": null,
|
4 |
+
"epoch": 5.0,
|
5 |
+
"eval_steps": 500,
|
6 |
+
"global_step": 820,
|
7 |
+
"is_hyper_param_search": false,
|
8 |
+
"is_local_process_zero": true,
|
9 |
+
"is_world_process_zero": true,
|
10 |
+
"log_history": [
|
11 |
+
{
|
12 |
+
"epoch": 0.006097560975609756,
|
13 |
+
"grad_norm": 470.0,
|
14 |
+
"learning_rate": 2.4390243902439027e-06,
|
15 |
+
"loss": 50.3127,
|
16 |
+
"step": 1
|
17 |
+
},
|
18 |
+
{
|
19 |
+
"epoch": 0.03048780487804878,
|
20 |
+
"grad_norm": 362.0,
|
21 |
+
"learning_rate": 1.2195121951219513e-05,
|
22 |
+
"loss": 41.9975,
|
23 |
+
"step": 5
|
24 |
+
},
|
25 |
+
{
|
26 |
+
"epoch": 0.06097560975609756,
|
27 |
+
"grad_norm": 199.0,
|
28 |
+
"learning_rate": 2.4390243902439026e-05,
|
29 |
+
"loss": 35.2694,
|
30 |
+
"step": 10
|
31 |
+
},
|
32 |
+
{
|
33 |
+
"epoch": 0.09146341463414634,
|
34 |
+
"grad_norm": 27.875,
|
35 |
+
"learning_rate": 3.6585365853658535e-05,
|
36 |
+
"loss": 22.3723,
|
37 |
+
"step": 15
|
38 |
+
},
|
39 |
+
{
|
40 |
+
"epoch": 0.12195121951219512,
|
41 |
+
"grad_norm": 21.375,
|
42 |
+
"learning_rate": 4.878048780487805e-05,
|
43 |
+
"loss": 18.6519,
|
44 |
+
"step": 20
|
45 |
+
},
|
46 |
+
{
|
47 |
+
"epoch": 0.1524390243902439,
|
48 |
+
"grad_norm": 5.8125,
|
49 |
+
"learning_rate": 6.097560975609756e-05,
|
50 |
+
"loss": 16.0212,
|
51 |
+
"step": 25
|
52 |
+
},
|
53 |
+
{
|
54 |
+
"epoch": 0.18292682926829268,
|
55 |
+
"grad_norm": 4.5625,
|
56 |
+
"learning_rate": 7.317073170731707e-05,
|
57 |
+
"loss": 15.6762,
|
58 |
+
"step": 30
|
59 |
+
},
|
60 |
+
{
|
61 |
+
"epoch": 0.21341463414634146,
|
62 |
+
"grad_norm": 5.9375,
|
63 |
+
"learning_rate": 8.53658536585366e-05,
|
64 |
+
"loss": 13.4942,
|
65 |
+
"step": 35
|
66 |
+
},
|
67 |
+
{
|
68 |
+
"epoch": 0.24390243902439024,
|
69 |
+
"grad_norm": 13.875,
|
70 |
+
"learning_rate": 9.75609756097561e-05,
|
71 |
+
"loss": 13.2206,
|
72 |
+
"step": 40
|
73 |
+
},
|
74 |
+
{
|
75 |
+
"epoch": 0.27439024390243905,
|
76 |
+
"grad_norm": 40.0,
|
77 |
+
"learning_rate": 0.00010975609756097563,
|
78 |
+
"loss": 9.5294,
|
79 |
+
"step": 45
|
80 |
+
},
|
81 |
+
{
|
82 |
+
"epoch": 0.3048780487804878,
|
83 |
+
"grad_norm": 8.0625,
|
84 |
+
"learning_rate": 0.00012195121951219512,
|
85 |
+
"loss": 2.8066,
|
86 |
+
"step": 50
|
87 |
+
},
|
88 |
+
{
|
89 |
+
"epoch": 0.3353658536585366,
|
90 |
+
"grad_norm": 2.421875,
|
91 |
+
"learning_rate": 0.00013414634146341464,
|
92 |
+
"loss": 1.5733,
|
93 |
+
"step": 55
|
94 |
+
},
|
95 |
+
{
|
96 |
+
"epoch": 0.36585365853658536,
|
97 |
+
"grad_norm": 2.0625,
|
98 |
+
"learning_rate": 0.00014634146341463414,
|
99 |
+
"loss": 1.3089,
|
100 |
+
"step": 60
|
101 |
+
},
|
102 |
+
{
|
103 |
+
"epoch": 0.39634146341463417,
|
104 |
+
"grad_norm": 3.15625,
|
105 |
+
"learning_rate": 0.00015853658536585366,
|
106 |
+
"loss": 1.1767,
|
107 |
+
"step": 65
|
108 |
+
},
|
109 |
+
{
|
110 |
+
"epoch": 0.4268292682926829,
|
111 |
+
"grad_norm": 7.0625,
|
112 |
+
"learning_rate": 0.0001707317073170732,
|
113 |
+
"loss": 1.1124,
|
114 |
+
"step": 70
|
115 |
+
},
|
116 |
+
{
|
117 |
+
"epoch": 0.4573170731707317,
|
118 |
+
"grad_norm": 3.265625,
|
119 |
+
"learning_rate": 0.0001829268292682927,
|
120 |
+
"loss": 1.0577,
|
121 |
+
"step": 75
|
122 |
+
},
|
123 |
+
{
|
124 |
+
"epoch": 0.4878048780487805,
|
125 |
+
"grad_norm": 6.59375,
|
126 |
+
"learning_rate": 0.0001951219512195122,
|
127 |
+
"loss": 0.9769,
|
128 |
+
"step": 80
|
129 |
+
},
|
130 |
+
{
|
131 |
+
"epoch": 0.5182926829268293,
|
132 |
+
"grad_norm": 10.6875,
|
133 |
+
"learning_rate": 0.00019999184556954776,
|
134 |
+
"loss": 0.9411,
|
135 |
+
"step": 85
|
136 |
+
},
|
137 |
+
{
|
138 |
+
"epoch": 0.5487804878048781,
|
139 |
+
"grad_norm": 1.171875,
|
140 |
+
"learning_rate": 0.0001999420177550043,
|
141 |
+
"loss": 0.939,
|
142 |
+
"step": 90
|
143 |
+
},
|
144 |
+
{
|
145 |
+
"epoch": 0.5792682926829268,
|
146 |
+
"grad_norm": 2.25,
|
147 |
+
"learning_rate": 0.00019984691491033906,
|
148 |
+
"loss": 0.9085,
|
149 |
+
"step": 95
|
150 |
+
},
|
151 |
+
{
|
152 |
+
"epoch": 0.6097560975609756,
|
153 |
+
"grad_norm": 0.97265625,
|
154 |
+
"learning_rate": 0.00019970658011837404,
|
155 |
+
"loss": 0.8808,
|
156 |
+
"step": 100
|
157 |
+
},
|
158 |
+
{
|
159 |
+
"epoch": 0.6402439024390244,
|
160 |
+
"grad_norm": 2.046875,
|
161 |
+
"learning_rate": 0.00019952107695258992,
|
162 |
+
"loss": 0.8201,
|
163 |
+
"step": 105
|
164 |
+
},
|
165 |
+
{
|
166 |
+
"epoch": 0.6707317073170732,
|
167 |
+
"grad_norm": 0.828125,
|
168 |
+
"learning_rate": 0.00019929048944832638,
|
169 |
+
"loss": 0.7852,
|
170 |
+
"step": 110
|
171 |
+
},
|
172 |
+
{
|
173 |
+
"epoch": 0.7012195121951219,
|
174 |
+
"grad_norm": 2.3125,
|
175 |
+
"learning_rate": 0.00019901492206471325,
|
176 |
+
"loss": 0.7695,
|
177 |
+
"step": 115
|
178 |
+
},
|
179 |
+
{
|
180 |
+
"epoch": 0.7317073170731707,
|
181 |
+
"grad_norm": 14.3125,
|
182 |
+
"learning_rate": 0.00019869449963734893,
|
183 |
+
"loss": 0.7996,
|
184 |
+
"step": 120
|
185 |
+
},
|
186 |
+
{
|
187 |
+
"epoch": 0.7621951219512195,
|
188 |
+
"grad_norm": 1.0859375,
|
189 |
+
"learning_rate": 0.00019832936732174834,
|
190 |
+
"loss": 0.7611,
|
191 |
+
"step": 125
|
192 |
+
},
|
193 |
+
{
|
194 |
+
"epoch": 0.7926829268292683,
|
195 |
+
"grad_norm": 29.0,
|
196 |
+
"learning_rate": 0.00019791969052758562,
|
197 |
+
"loss": 0.7835,
|
198 |
+
"step": 130
|
199 |
+
},
|
200 |
+
{
|
201 |
+
"epoch": 0.823170731707317,
|
202 |
+
"grad_norm": 0.6015625,
|
203 |
+
"learning_rate": 0.00019746565484376132,
|
204 |
+
"loss": 0.759,
|
205 |
+
"step": 135
|
206 |
+
},
|
207 |
+
{
|
208 |
+
"epoch": 0.8536585365853658,
|
209 |
+
"grad_norm": 0.57421875,
|
210 |
+
"learning_rate": 0.00019696746595432828,
|
211 |
+
"loss": 0.7598,
|
212 |
+
"step": 140
|
213 |
+
},
|
214 |
+
{
|
215 |
+
"epoch": 0.8841463414634146,
|
216 |
+
"grad_norm": 0.9140625,
|
217 |
+
"learning_rate": 0.0001964253495453141,
|
218 |
+
"loss": 0.7279,
|
219 |
+
"step": 145
|
220 |
+
},
|
221 |
+
{
|
222 |
+
"epoch": 0.9146341463414634,
|
223 |
+
"grad_norm": 1.484375,
|
224 |
+
"learning_rate": 0.00019583955120248237,
|
225 |
+
"loss": 0.7304,
|
226 |
+
"step": 150
|
227 |
+
},
|
228 |
+
{
|
229 |
+
"epoch": 0.9451219512195121,
|
230 |
+
"grad_norm": 0.7890625,
|
231 |
+
"learning_rate": 0.00019521033630007928,
|
232 |
+
"loss": 0.7659,
|
233 |
+
"step": 155
|
234 |
+
},
|
235 |
+
{
|
236 |
+
"epoch": 0.975609756097561,
|
237 |
+
"grad_norm": 3.03125,
|
238 |
+
"learning_rate": 0.00019453798988061535,
|
239 |
+
"loss": 0.7425,
|
240 |
+
"step": 160
|
241 |
+
},
|
242 |
+
{
|
243 |
+
"epoch": 1.0,
|
244 |
+
"eval_loss": 1.182375431060791,
|
245 |
+
"eval_runtime": 1.0783,
|
246 |
+
"eval_samples_per_second": 1.855,
|
247 |
+
"eval_steps_per_second": 0.927,
|
248 |
+
"step": 164
|
249 |
+
},
|
250 |
+
{
|
251 |
+
"epoch": 1.0060975609756098,
|
252 |
+
"grad_norm": 0.8515625,
|
253 |
+
"learning_rate": 0.00019382281652573785,
|
254 |
+
"loss": 0.6928,
|
255 |
+
"step": 165
|
256 |
+
},
|
257 |
+
{
|
258 |
+
"epoch": 1.0365853658536586,
|
259 |
+
"grad_norm": 0.9375,
|
260 |
+
"learning_rate": 0.00019306514021825118,
|
261 |
+
"loss": 0.6592,
|
262 |
+
"step": 170
|
263 |
+
},
|
264 |
+
{
|
265 |
+
"epoch": 1.0670731707317074,
|
266 |
+
"grad_norm": 0.8125,
|
267 |
+
"learning_rate": 0.00019226530419534833,
|
268 |
+
"loss": 0.6442,
|
269 |
+
"step": 175
|
270 |
+
},
|
271 |
+
{
|
272 |
+
"epoch": 1.0975609756097562,
|
273 |
+
"grad_norm": 0.72265625,
|
274 |
+
"learning_rate": 0.00019142367079312021,
|
275 |
+
"loss": 0.659,
|
276 |
+
"step": 180
|
277 |
+
},
|
278 |
+
{
|
279 |
+
"epoch": 1.1280487804878048,
|
280 |
+
"grad_norm": 0.462890625,
|
281 |
+
"learning_rate": 0.00019054062128241264,
|
282 |
+
"loss": 0.6355,
|
283 |
+
"step": 185
|
284 |
+
},
|
285 |
+
{
|
286 |
+
"epoch": 1.1585365853658536,
|
287 |
+
"grad_norm": 1.1015625,
|
288 |
+
"learning_rate": 0.00018961655569610557,
|
289 |
+
"loss": 0.6304,
|
290 |
+
"step": 190
|
291 |
+
},
|
292 |
+
{
|
293 |
+
"epoch": 1.1890243902439024,
|
294 |
+
"grad_norm": 0.5625,
|
295 |
+
"learning_rate": 0.0001886518926478932,
|
296 |
+
"loss": 0.6414,
|
297 |
+
"step": 195
|
298 |
+
},
|
299 |
+
{
|
300 |
+
"epoch": 1.2195121951219512,
|
301 |
+
"grad_norm": 0.5,
|
302 |
+
"learning_rate": 0.00018764706914264635,
|
303 |
+
"loss": 0.6429,
|
304 |
+
"step": 200
|
305 |
+
},
|
306 |
+
{
|
307 |
+
"epoch": 1.25,
|
308 |
+
"grad_norm": 1.046875,
|
309 |
+
"learning_rate": 0.00018660254037844388,
|
310 |
+
"loss": 0.6442,
|
311 |
+
"step": 205
|
312 |
+
},
|
313 |
+
{
|
314 |
+
"epoch": 1.2804878048780488,
|
315 |
+
"grad_norm": 0.73046875,
|
316 |
+
"learning_rate": 0.00018551877954036162,
|
317 |
+
"loss": 0.6394,
|
318 |
+
"step": 210
|
319 |
+
},
|
320 |
+
{
|
321 |
+
"epoch": 1.3109756097560976,
|
322 |
+
"grad_norm": 0.87109375,
|
323 |
+
"learning_rate": 0.00018439627758611385,
|
324 |
+
"loss": 0.6256,
|
325 |
+
"step": 215
|
326 |
+
},
|
327 |
+
{
|
328 |
+
"epoch": 1.3414634146341464,
|
329 |
+
"grad_norm": 0.890625,
|
330 |
+
"learning_rate": 0.00018323554302364272,
|
331 |
+
"loss": 0.6397,
|
332 |
+
"step": 220
|
333 |
+
},
|
334 |
+
{
|
335 |
+
"epoch": 1.3719512195121952,
|
336 |
+
"grad_norm": 0.96875,
|
337 |
+
"learning_rate": 0.00018203710168075788,
|
338 |
+
"loss": 0.6138,
|
339 |
+
"step": 225
|
340 |
+
},
|
341 |
+
{
|
342 |
+
"epoch": 1.4024390243902438,
|
343 |
+
"grad_norm": 1.0625,
|
344 |
+
"learning_rate": 0.0001808014964669293,
|
345 |
+
"loss": 0.6305,
|
346 |
+
"step": 230
|
347 |
+
},
|
348 |
+
{
|
349 |
+
"epoch": 1.4329268292682926,
|
350 |
+
"grad_norm": 0.51953125,
|
351 |
+
"learning_rate": 0.00017952928712734268,
|
352 |
+
"loss": 0.6213,
|
353 |
+
"step": 235
|
354 |
+
},
|
355 |
+
{
|
356 |
+
"epoch": 1.4634146341463414,
|
357 |
+
"grad_norm": 1.78125,
|
358 |
+
"learning_rate": 0.00017822104998932713,
|
359 |
+
"loss": 0.6233,
|
360 |
+
"step": 240
|
361 |
+
},
|
362 |
+
{
|
363 |
+
"epoch": 1.4939024390243902,
|
364 |
+
"grad_norm": 0.7265625,
|
365 |
+
"learning_rate": 0.00017687737770127185,
|
366 |
+
"loss": 0.6489,
|
367 |
+
"step": 245
|
368 |
+
},
|
369 |
+
{
|
370 |
+
"epoch": 1.524390243902439,
|
371 |
+
"grad_norm": 0.8984375,
|
372 |
+
"learning_rate": 0.00017549887896414851,
|
373 |
+
"loss": 0.6058,
|
374 |
+
"step": 250
|
375 |
+
},
|
376 |
+
{
|
377 |
+
"epoch": 1.5548780487804879,
|
378 |
+
"grad_norm": 1.015625,
|
379 |
+
"learning_rate": 0.0001740861782557618,
|
380 |
+
"loss": 0.6164,
|
381 |
+
"step": 255
|
382 |
+
},
|
383 |
+
{
|
384 |
+
"epoch": 1.5853658536585367,
|
385 |
+
"grad_norm": 0.73046875,
|
386 |
+
"learning_rate": 0.0001726399155478529,
|
387 |
+
"loss": 0.606,
|
388 |
+
"step": 260
|
389 |
+
},
|
390 |
+
{
|
391 |
+
"epoch": 1.6158536585365852,
|
392 |
+
"grad_norm": 0.5703125,
|
393 |
+
"learning_rate": 0.00017116074601618417,
|
394 |
+
"loss": 0.6051,
|
395 |
+
"step": 265
|
396 |
+
},
|
397 |
+
{
|
398 |
+
"epoch": 1.6463414634146343,
|
399 |
+
"grad_norm": 0.7109375,
|
400 |
+
"learning_rate": 0.0001696493397437357,
|
401 |
+
"loss": 0.6798,
|
402 |
+
"step": 270
|
403 |
+
},
|
404 |
+
{
|
405 |
+
"epoch": 1.6768292682926829,
|
406 |
+
"grad_norm": 1.6953125,
|
407 |
+
"learning_rate": 0.00016810638141714934,
|
408 |
+
"loss": 0.6205,
|
409 |
+
"step": 275
|
410 |
+
},
|
411 |
+
{
|
412 |
+
"epoch": 1.7073170731707317,
|
413 |
+
"grad_norm": 1.015625,
|
414 |
+
"learning_rate": 0.00016653257001655652,
|
415 |
+
"loss": 0.6627,
|
416 |
+
"step": 280
|
417 |
+
},
|
418 |
+
{
|
419 |
+
"epoch": 1.7378048780487805,
|
420 |
+
"grad_norm": 0.84765625,
|
421 |
+
"learning_rate": 0.0001649286184989315,
|
422 |
+
"loss": 0.6152,
|
423 |
+
"step": 285
|
424 |
+
},
|
425 |
+
{
|
426 |
+
"epoch": 1.7682926829268293,
|
427 |
+
"grad_norm": 1.21875,
|
428 |
+
"learning_rate": 0.0001632952534751122,
|
429 |
+
"loss": 0.5972,
|
430 |
+
"step": 290
|
431 |
+
},
|
432 |
+
{
|
433 |
+
"epoch": 1.798780487804878,
|
434 |
+
"grad_norm": 0.4609375,
|
435 |
+
"learning_rate": 0.00016163321488063637,
|
436 |
+
"loss": 0.6053,
|
437 |
+
"step": 295
|
438 |
+
},
|
439 |
+
{
|
440 |
+
"epoch": 1.8292682926829267,
|
441 |
+
"grad_norm": 0.423828125,
|
442 |
+
"learning_rate": 0.00015994325564054122,
|
443 |
+
"loss": 0.5965,
|
444 |
+
"step": 300
|
445 |
+
},
|
446 |
+
{
|
447 |
+
"epoch": 1.8597560975609757,
|
448 |
+
"grad_norm": 0.9296875,
|
449 |
+
"learning_rate": 0.00015822614132827837,
|
450 |
+
"loss": 0.6023,
|
451 |
+
"step": 305
|
452 |
+
},
|
453 |
+
{
|
454 |
+
"epoch": 1.8902439024390243,
|
455 |
+
"grad_norm": 0.671875,
|
456 |
+
"learning_rate": 0.00015648264981889934,
|
457 |
+
"loss": 0.5916,
|
458 |
+
"step": 310
|
459 |
+
},
|
460 |
+
{
|
461 |
+
"epoch": 1.9207317073170733,
|
462 |
+
"grad_norm": 0.68359375,
|
463 |
+
"learning_rate": 0.00015471357093666804,
|
464 |
+
"loss": 0.6082,
|
465 |
+
"step": 315
|
466 |
+
},
|
467 |
+
{
|
468 |
+
"epoch": 1.951219512195122,
|
469 |
+
"grad_norm": 0.578125,
|
470 |
+
"learning_rate": 0.00015291970609726007,
|
471 |
+
"loss": 0.6176,
|
472 |
+
"step": 320
|
473 |
+
},
|
474 |
+
{
|
475 |
+
"epoch": 1.9817073170731707,
|
476 |
+
"grad_norm": 0.55859375,
|
477 |
+
"learning_rate": 0.00015110186794471103,
|
478 |
+
"loss": 0.6196,
|
479 |
+
"step": 325
|
480 |
+
},
|
481 |
+
{
|
482 |
+
"epoch": 2.0,
|
483 |
+
"eval_loss": 1.1435465812683105,
|
484 |
+
"eval_runtime": 1.0793,
|
485 |
+
"eval_samples_per_second": 1.853,
|
486 |
+
"eval_steps_per_second": 0.927,
|
487 |
+
"step": 328
|
488 |
+
},
|
489 |
+
{
|
490 |
+
"epoch": 2.0121951219512195,
|
491 |
+
"grad_norm": 0.5703125,
|
492 |
+
"learning_rate": 0.00014926087998327837,
|
493 |
+
"loss": 0.5623,
|
494 |
+
"step": 330
|
495 |
+
},
|
496 |
+
{
|
497 |
+
"epoch": 2.042682926829268,
|
498 |
+
"grad_norm": 0.82421875,
|
499 |
+
"learning_rate": 0.00014739757620438307,
|
500 |
+
"loss": 0.5231,
|
501 |
+
"step": 335
|
502 |
+
},
|
503 |
+
{
|
504 |
+
"epoch": 2.073170731707317,
|
505 |
+
"grad_norm": 1.25,
|
506 |
+
"learning_rate": 0.0001455128007088009,
|
507 |
+
"loss": 0.5611,
|
508 |
+
"step": 340
|
509 |
+
},
|
510 |
+
{
|
511 |
+
"epoch": 2.1036585365853657,
|
512 |
+
"grad_norm": 0.5,
|
513 |
+
"learning_rate": 0.00014360740732427367,
|
514 |
+
"loss": 0.5497,
|
515 |
+
"step": 345
|
516 |
+
},
|
517 |
+
{
|
518 |
+
"epoch": 2.1341463414634148,
|
519 |
+
"grad_norm": 0.443359375,
|
520 |
+
"learning_rate": 0.00014168225921871433,
|
521 |
+
"loss": 0.5154,
|
522 |
+
"step": 350
|
523 |
+
},
|
524 |
+
{
|
525 |
+
"epoch": 2.1646341463414633,
|
526 |
+
"grad_norm": 0.498046875,
|
527 |
+
"learning_rate": 0.00013973822850918055,
|
528 |
+
"loss": 0.5307,
|
529 |
+
"step": 355
|
530 |
+
},
|
531 |
+
{
|
532 |
+
"epoch": 2.1951219512195124,
|
533 |
+
"grad_norm": 0.79296875,
|
534 |
+
"learning_rate": 0.0001377761958667946,
|
535 |
+
"loss": 0.5085,
|
536 |
+
"step": 360
|
537 |
+
},
|
538 |
+
{
|
539 |
+
"epoch": 2.225609756097561,
|
540 |
+
"grad_norm": 0.8046875,
|
541 |
+
"learning_rate": 0.00013579705011778766,
|
542 |
+
"loss": 0.5339,
|
543 |
+
"step": 365
|
544 |
+
},
|
545 |
+
{
|
546 |
+
"epoch": 2.2560975609756095,
|
547 |
+
"grad_norm": 0.7890625,
|
548 |
+
"learning_rate": 0.00013380168784085027,
|
549 |
+
"loss": 0.5263,
|
550 |
+
"step": 370
|
551 |
+
},
|
552 |
+
{
|
553 |
+
"epoch": 2.2865853658536586,
|
554 |
+
"grad_norm": 0.5234375,
|
555 |
+
"learning_rate": 0.00013179101296097035,
|
556 |
+
"loss": 0.5106,
|
557 |
+
"step": 375
|
558 |
+
},
|
559 |
+
{
|
560 |
+
"epoch": 2.317073170731707,
|
561 |
+
"grad_norm": 0.62890625,
|
562 |
+
"learning_rate": 0.00012976593633994346,
|
563 |
+
"loss": 0.5371,
|
564 |
+
"step": 380
|
565 |
+
},
|
566 |
+
{
|
567 |
+
"epoch": 2.347560975609756,
|
568 |
+
"grad_norm": 0.4140625,
|
569 |
+
"learning_rate": 0.0001277273753637408,
|
570 |
+
"loss": 0.5188,
|
571 |
+
"step": 385
|
572 |
+
},
|
573 |
+
{
|
574 |
+
"epoch": 2.3780487804878048,
|
575 |
+
"grad_norm": 0.66015625,
|
576 |
+
"learning_rate": 0.00012567625352692127,
|
577 |
+
"loss": 0.5309,
|
578 |
+
"step": 390
|
579 |
+
},
|
580 |
+
{
|
581 |
+
"epoch": 2.408536585365854,
|
582 |
+
"grad_norm": 1.21875,
|
583 |
+
"learning_rate": 0.0001236135000142765,
|
584 |
+
"loss": 0.5024,
|
585 |
+
"step": 395
|
586 |
+
},
|
587 |
+
{
|
588 |
+
"epoch": 2.4390243902439024,
|
589 |
+
"grad_norm": 0.58203125,
|
590 |
+
"learning_rate": 0.00012154004927989815,
|
591 |
+
"loss": 0.5291,
|
592 |
+
"step": 400
|
593 |
+
},
|
594 |
+
{
|
595 |
+
"epoch": 2.4695121951219514,
|
596 |
+
"grad_norm": 0.478515625,
|
597 |
+
"learning_rate": 0.00011945684062385803,
|
598 |
+
"loss": 0.5261,
|
599 |
+
"step": 405
|
600 |
+
},
|
601 |
+
{
|
602 |
+
"epoch": 2.5,
|
603 |
+
"grad_norm": 0.431640625,
|
604 |
+
"learning_rate": 0.00011736481776669306,
|
605 |
+
"loss": 0.5218,
|
606 |
+
"step": 410
|
607 |
+
},
|
608 |
+
{
|
609 |
+
"epoch": 2.5304878048780486,
|
610 |
+
"grad_norm": 0.447265625,
|
611 |
+
"learning_rate": 0.00011526492842188745,
|
612 |
+
"loss": 0.5259,
|
613 |
+
"step": 415
|
614 |
+
},
|
615 |
+
{
|
616 |
+
"epoch": 2.5609756097560976,
|
617 |
+
"grad_norm": 0.41796875,
|
618 |
+
"learning_rate": 0.0001131581238665465,
|
619 |
+
"loss": 0.5077,
|
620 |
+
"step": 420
|
621 |
+
},
|
622 |
+
{
|
623 |
+
"epoch": 2.591463414634146,
|
624 |
+
"grad_norm": 0.42578125,
|
625 |
+
"learning_rate": 0.00011104535851045539,
|
626 |
+
"loss": 0.5356,
|
627 |
+
"step": 425
|
628 |
+
},
|
629 |
+
{
|
630 |
+
"epoch": 2.6219512195121952,
|
631 |
+
"grad_norm": 0.62890625,
|
632 |
+
"learning_rate": 0.00010892758946371944,
|
633 |
+
"loss": 0.5158,
|
634 |
+
"step": 430
|
635 |
+
},
|
636 |
+
{
|
637 |
+
"epoch": 2.652439024390244,
|
638 |
+
"grad_norm": 0.46484375,
|
639 |
+
"learning_rate": 0.00010680577610318072,
|
640 |
+
"loss": 0.5297,
|
641 |
+
"step": 435
|
642 |
+
},
|
643 |
+
{
|
644 |
+
"epoch": 2.682926829268293,
|
645 |
+
"grad_norm": 0.4609375,
|
646 |
+
"learning_rate": 0.00010468087963780789,
|
647 |
+
"loss": 0.5015,
|
648 |
+
"step": 440
|
649 |
+
},
|
650 |
+
{
|
651 |
+
"epoch": 2.7134146341463414,
|
652 |
+
"grad_norm": 0.447265625,
|
653 |
+
"learning_rate": 0.00010255386267325602,
|
654 |
+
"loss": 0.5169,
|
655 |
+
"step": 445
|
656 |
+
},
|
657 |
+
{
|
658 |
+
"epoch": 2.7439024390243905,
|
659 |
+
"grad_norm": 0.6171875,
|
660 |
+
"learning_rate": 0.00010042568877579388,
|
661 |
+
"loss": 0.5034,
|
662 |
+
"step": 450
|
663 |
+
},
|
664 |
+
{
|
665 |
+
"epoch": 2.774390243902439,
|
666 |
+
"grad_norm": 0.419921875,
|
667 |
+
"learning_rate": 9.829732203579584e-05,
|
668 |
+
"loss": 0.5113,
|
669 |
+
"step": 455
|
670 |
+
},
|
671 |
+
{
|
672 |
+
"epoch": 2.8048780487804876,
|
673 |
+
"grad_norm": 0.396484375,
|
674 |
+
"learning_rate": 9.616972663099647e-05,
|
675 |
+
"loss": 0.5106,
|
676 |
+
"step": 460
|
677 |
+
},
|
678 |
+
{
|
679 |
+
"epoch": 2.8353658536585367,
|
680 |
+
"grad_norm": 0.4765625,
|
681 |
+
"learning_rate": 9.404386638970542e-05,
|
682 |
+
"loss": 0.5135,
|
683 |
+
"step": 465
|
684 |
+
},
|
685 |
+
{
|
686 |
+
"epoch": 2.8658536585365852,
|
687 |
+
"grad_norm": 0.390625,
|
688 |
+
"learning_rate": 9.192070435418079e-05,
|
689 |
+
"loss": 0.5222,
|
690 |
+
"step": 470
|
691 |
+
},
|
692 |
+
{
|
693 |
+
"epoch": 2.8963414634146343,
|
694 |
+
"grad_norm": 0.55859375,
|
695 |
+
"learning_rate": 8.980120234435849e-05,
|
696 |
+
"loss": 0.5214,
|
697 |
+
"step": 475
|
698 |
+
},
|
699 |
+
{
|
700 |
+
"epoch": 2.926829268292683,
|
701 |
+
"grad_norm": 0.6328125,
|
702 |
+
"learning_rate": 8.768632052213531e-05,
|
703 |
+
"loss": 0.5408,
|
704 |
+
"step": 480
|
705 |
+
},
|
706 |
+
{
|
707 |
+
"epoch": 2.9573170731707314,
|
708 |
+
"grad_norm": 0.490234375,
|
709 |
+
"learning_rate": 8.557701695640321e-05,
|
710 |
+
"loss": 0.5012,
|
711 |
+
"step": 485
|
712 |
+
},
|
713 |
+
{
|
714 |
+
"epoch": 2.9878048780487805,
|
715 |
+
"grad_norm": 0.41796875,
|
716 |
+
"learning_rate": 8.347424718903151e-05,
|
717 |
+
"loss": 0.5124,
|
718 |
+
"step": 490
|
719 |
+
},
|
720 |
+
{
|
721 |
+
"epoch": 3.0,
|
722 |
+
"eval_loss": 1.1460928916931152,
|
723 |
+
"eval_runtime": 1.0779,
|
724 |
+
"eval_samples_per_second": 1.855,
|
725 |
+
"eval_steps_per_second": 0.928,
|
726 |
+
"step": 492
|
727 |
+
},
|
728 |
+
{
|
729 |
+
"epoch": 3.018292682926829,
|
730 |
+
"grad_norm": 0.71484375,
|
731 |
+
"learning_rate": 8.13789638019942e-05,
|
732 |
+
"loss": 0.4767,
|
733 |
+
"step": 495
|
734 |
+
},
|
735 |
+
{
|
736 |
+
"epoch": 3.048780487804878,
|
737 |
+
"grad_norm": 0.53515625,
|
738 |
+
"learning_rate": 7.929211598583794e-05,
|
739 |
+
"loss": 0.4404,
|
740 |
+
"step": 500
|
741 |
+
},
|
742 |
+
{
|
743 |
+
"epoch": 3.0792682926829267,
|
744 |
+
"grad_norm": 0.546875,
|
745 |
+
"learning_rate": 7.721464910968627e-05,
|
746 |
+
"loss": 0.447,
|
747 |
+
"step": 505
|
748 |
+
},
|
749 |
+
{
|
750 |
+
"epoch": 3.1097560975609757,
|
751 |
+
"grad_norm": 0.40234375,
|
752 |
+
"learning_rate": 7.514750429297528e-05,
|
753 |
+
"loss": 0.4379,
|
754 |
+
"step": 510
|
755 |
+
},
|
756 |
+
{
|
757 |
+
"epoch": 3.1402439024390243,
|
758 |
+
"grad_norm": 0.59765625,
|
759 |
+
"learning_rate": 7.309161797911441e-05,
|
760 |
+
"loss": 0.4287,
|
761 |
+
"step": 515
|
762 |
+
},
|
763 |
+
{
|
764 |
+
"epoch": 3.1707317073170733,
|
765 |
+
"grad_norm": 0.4375,
|
766 |
+
"learning_rate": 7.104792151126515e-05,
|
767 |
+
"loss": 0.428,
|
768 |
+
"step": 520
|
769 |
+
},
|
770 |
+
{
|
771 |
+
"epoch": 3.201219512195122,
|
772 |
+
"grad_norm": 0.4296875,
|
773 |
+
"learning_rate": 6.901734071043071e-05,
|
774 |
+
"loss": 0.4448,
|
775 |
+
"step": 525
|
776 |
+
},
|
777 |
+
{
|
778 |
+
"epoch": 3.231707317073171,
|
779 |
+
"grad_norm": 0.4453125,
|
780 |
+
"learning_rate": 6.700079545604708e-05,
|
781 |
+
"loss": 0.4297,
|
782 |
+
"step": 530
|
783 |
+
},
|
784 |
+
{
|
785 |
+
"epoch": 3.2621951219512195,
|
786 |
+
"grad_norm": 0.435546875,
|
787 |
+
"learning_rate": 6.499919926926566e-05,
|
788 |
+
"loss": 0.423,
|
789 |
+
"step": 535
|
790 |
+
},
|
791 |
+
{
|
792 |
+
"epoch": 3.292682926829268,
|
793 |
+
"grad_norm": 0.44921875,
|
794 |
+
"learning_rate": 6.301345889911637e-05,
|
795 |
+
"loss": 0.4218,
|
796 |
+
"step": 540
|
797 |
+
},
|
798 |
+
{
|
799 |
+
"epoch": 3.323170731707317,
|
800 |
+
"grad_norm": 0.56640625,
|
801 |
+
"learning_rate": 6.104447391173858e-05,
|
802 |
+
"loss": 0.4354,
|
803 |
+
"step": 545
|
804 |
+
},
|
805 |
+
{
|
806 |
+
"epoch": 3.3536585365853657,
|
807 |
+
"grad_norm": 0.47265625,
|
808 |
+
"learning_rate": 5.909313628286601e-05,
|
809 |
+
"loss": 0.4233,
|
810 |
+
"step": 550
|
811 |
+
},
|
812 |
+
{
|
813 |
+
"epoch": 3.3841463414634148,
|
814 |
+
"grad_norm": 0.42578125,
|
815 |
+
"learning_rate": 5.716032999375006e-05,
|
816 |
+
"loss": 0.4634,
|
817 |
+
"step": 555
|
818 |
+
},
|
819 |
+
{
|
820 |
+
"epoch": 3.4146341463414633,
|
821 |
+
"grad_norm": 0.419921875,
|
822 |
+
"learning_rate": 5.524693063070492e-05,
|
823 |
+
"loss": 0.4268,
|
824 |
+
"step": 560
|
825 |
+
},
|
826 |
+
{
|
827 |
+
"epoch": 3.4451219512195124,
|
828 |
+
"grad_norm": 0.494140625,
|
829 |
+
"learning_rate": 5.335380498845559e-05,
|
830 |
+
"loss": 0.4333,
|
831 |
+
"step": 565
|
832 |
+
},
|
833 |
+
{
|
834 |
+
"epoch": 3.475609756097561,
|
835 |
+
"grad_norm": 0.46875,
|
836 |
+
"learning_rate": 5.148181067746862e-05,
|
837 |
+
"loss": 0.471,
|
838 |
+
"step": 570
|
839 |
+
},
|
840 |
+
{
|
841 |
+
"epoch": 3.5060975609756095,
|
842 |
+
"grad_norm": 0.44921875,
|
843 |
+
"learning_rate": 4.963179573544357e-05,
|
844 |
+
"loss": 0.4116,
|
845 |
+
"step": 575
|
846 |
+
},
|
847 |
+
{
|
848 |
+
"epoch": 3.5365853658536586,
|
849 |
+
"grad_norm": 0.458984375,
|
850 |
+
"learning_rate": 4.7804598243140666e-05,
|
851 |
+
"loss": 0.4301,
|
852 |
+
"step": 580
|
853 |
+
},
|
854 |
+
{
|
855 |
+
"epoch": 3.567073170731707,
|
856 |
+
"grad_norm": 0.431640625,
|
857 |
+
"learning_rate": 4.60010459447196e-05,
|
858 |
+
"loss": 0.4359,
|
859 |
+
"step": 585
|
860 |
+
},
|
861 |
+
{
|
862 |
+
"epoch": 3.597560975609756,
|
863 |
+
"grad_norm": 0.44921875,
|
864 |
+
"learning_rate": 4.422195587276058e-05,
|
865 |
+
"loss": 0.4322,
|
866 |
+
"step": 590
|
867 |
+
},
|
868 |
+
{
|
869 |
+
"epoch": 3.6280487804878048,
|
870 |
+
"grad_norm": 0.52734375,
|
871 |
+
"learning_rate": 4.2468133978137945e-05,
|
872 |
+
"loss": 0.4283,
|
873 |
+
"step": 595
|
874 |
+
},
|
875 |
+
{
|
876 |
+
"epoch": 3.658536585365854,
|
877 |
+
"grad_norm": 0.51171875,
|
878 |
+
"learning_rate": 4.0740374764914136e-05,
|
879 |
+
"loss": 0.4297,
|
880 |
+
"step": 600
|
881 |
+
},
|
882 |
+
{
|
883 |
+
"epoch": 3.6890243902439024,
|
884 |
+
"grad_norm": 0.52734375,
|
885 |
+
"learning_rate": 3.903946093041877e-05,
|
886 |
+
"loss": 0.4386,
|
887 |
+
"step": 605
|
888 |
+
},
|
889 |
+
{
|
890 |
+
"epoch": 3.7195121951219514,
|
891 |
+
"grad_norm": 0.4453125,
|
892 |
+
"learning_rate": 3.736616301067694e-05,
|
893 |
+
"loss": 0.4279,
|
894 |
+
"step": 610
|
895 |
+
},
|
896 |
+
{
|
897 |
+
"epoch": 3.75,
|
898 |
+
"grad_norm": 0.53125,
|
899 |
+
"learning_rate": 3.5721239031346066e-05,
|
900 |
+
"loss": 0.454,
|
901 |
+
"step": 615
|
902 |
+
},
|
903 |
+
{
|
904 |
+
"epoch": 3.7804878048780486,
|
905 |
+
"grad_norm": 0.447265625,
|
906 |
+
"learning_rate": 3.410543416432069e-05,
|
907 |
+
"loss": 0.4497,
|
908 |
+
"step": 620
|
909 |
+
},
|
910 |
+
{
|
911 |
+
"epoch": 3.8109756097560976,
|
912 |
+
"grad_norm": 0.48828125,
|
913 |
+
"learning_rate": 3.2519480390159806e-05,
|
914 |
+
"loss": 0.4405,
|
915 |
+
"step": 625
|
916 |
+
},
|
917 |
+
{
|
918 |
+
"epoch": 3.841463414634146,
|
919 |
+
"grad_norm": 0.55078125,
|
920 |
+
"learning_rate": 3.096409616649023e-05,
|
921 |
+
"loss": 0.4354,
|
922 |
+
"step": 630
|
923 |
+
},
|
924 |
+
{
|
925 |
+
"epoch": 3.8719512195121952,
|
926 |
+
"grad_norm": 0.443359375,
|
927 |
+
"learning_rate": 2.9439986102536043e-05,
|
928 |
+
"loss": 0.4382,
|
929 |
+
"step": 635
|
930 |
+
},
|
931 |
+
{
|
932 |
+
"epoch": 3.902439024390244,
|
933 |
+
"grad_norm": 0.416015625,
|
934 |
+
"learning_rate": 2.794784063992131e-05,
|
935 |
+
"loss": 0.4204,
|
936 |
+
"step": 640
|
937 |
+
},
|
938 |
+
{
|
939 |
+
"epoch": 3.932926829268293,
|
940 |
+
"grad_norm": 0.515625,
|
941 |
+
"learning_rate": 2.6488335739891178e-05,
|
942 |
+
"loss": 0.4199,
|
943 |
+
"step": 645
|
944 |
+
},
|
945 |
+
{
|
946 |
+
"epoch": 3.9634146341463414,
|
947 |
+
"grad_norm": 0.5234375,
|
948 |
+
"learning_rate": 2.50621325770927e-05,
|
949 |
+
"loss": 0.4275,
|
950 |
+
"step": 650
|
951 |
+
},
|
952 |
+
{
|
953 |
+
"epoch": 3.9939024390243905,
|
954 |
+
"grad_norm": 0.4765625,
|
955 |
+
"learning_rate": 2.366987724005404e-05,
|
956 |
+
"loss": 0.4403,
|
957 |
+
"step": 655
|
958 |
+
},
|
959 |
+
{
|
960 |
+
"epoch": 4.0,
|
961 |
+
"eval_loss": 1.206821322441101,
|
962 |
+
"eval_runtime": 1.0795,
|
963 |
+
"eval_samples_per_second": 1.853,
|
964 |
+
"eval_steps_per_second": 0.926,
|
965 |
+
"step": 656
|
966 |
+
},
|
967 |
+
{
|
968 |
+
"epoch": 4.024390243902439,
|
969 |
+
"grad_norm": 0.408203125,
|
970 |
+
"learning_rate": 2.2312200438498043e-05,
|
971 |
+
"loss": 0.3802,
|
972 |
+
"step": 660
|
973 |
+
},
|
974 |
+
{
|
975 |
+
"epoch": 4.054878048780488,
|
976 |
+
"grad_norm": 0.455078125,
|
977 |
+
"learning_rate": 2.0989717217622652e-05,
|
978 |
+
"loss": 0.3864,
|
979 |
+
"step": 665
|
980 |
+
},
|
981 |
+
{
|
982 |
+
"epoch": 4.085365853658536,
|
983 |
+
"grad_norm": 0.443359375,
|
984 |
+
"learning_rate": 1.9703026679477256e-05,
|
985 |
+
"loss": 0.3842,
|
986 |
+
"step": 670
|
987 |
+
},
|
988 |
+
{
|
989 |
+
"epoch": 4.115853658536586,
|
990 |
+
"grad_norm": 0.5546875,
|
991 |
+
"learning_rate": 1.8452711711561842e-05,
|
992 |
+
"loss": 0.3832,
|
993 |
+
"step": 675
|
994 |
+
},
|
995 |
+
{
|
996 |
+
"epoch": 4.146341463414634,
|
997 |
+
"grad_norm": 0.447265625,
|
998 |
+
"learning_rate": 1.7239338722771327e-05,
|
999 |
+
"loss": 0.3686,
|
1000 |
+
"step": 680
|
1001 |
+
},
|
1002 |
+
{
|
1003 |
+
"epoch": 4.176829268292683,
|
1004 |
+
"grad_norm": 0.466796875,
|
1005 |
+
"learning_rate": 1.6063457386805004e-05,
|
1006 |
+
"loss": 0.3955,
|
1007 |
+
"step": 685
|
1008 |
+
},
|
1009 |
+
{
|
1010 |
+
"epoch": 4.2073170731707314,
|
1011 |
+
"grad_norm": 0.51953125,
|
1012 |
+
"learning_rate": 1.4925600393157324e-05,
|
1013 |
+
"loss": 0.3847,
|
1014 |
+
"step": 690
|
1015 |
+
},
|
1016 |
+
{
|
1017 |
+
"epoch": 4.237804878048781,
|
1018 |
+
"grad_norm": 0.435546875,
|
1019 |
+
"learning_rate": 1.3826283205802427e-05,
|
1020 |
+
"loss": 0.3829,
|
1021 |
+
"step": 695
|
1022 |
+
},
|
1023 |
+
{
|
1024 |
+
"epoch": 4.2682926829268295,
|
1025 |
+
"grad_norm": 0.47265625,
|
1026 |
+
"learning_rate": 1.2766003829682505e-05,
|
1027 |
+
"loss": 0.3862,
|
1028 |
+
"step": 700
|
1029 |
+
},
|
1030 |
+
{
|
1031 |
+
"epoch": 4.298780487804878,
|
1032 |
+
"grad_norm": 0.4609375,
|
1033 |
+
"learning_rate": 1.1745242585104955e-05,
|
1034 |
+
"loss": 0.3965,
|
1035 |
+
"step": 705
|
1036 |
+
},
|
1037 |
+
{
|
1038 |
+
"epoch": 4.329268292682927,
|
1039 |
+
"grad_norm": 0.494140625,
|
1040 |
+
"learning_rate": 1.0764461890151112e-05,
|
1041 |
+
"loss": 0.3807,
|
1042 |
+
"step": 710
|
1043 |
+
},
|
1044 |
+
{
|
1045 |
+
"epoch": 4.359756097560975,
|
1046 |
+
"grad_norm": 0.466796875,
|
1047 |
+
"learning_rate": 9.824106051194859e-06,
|
1048 |
+
"loss": 0.3722,
|
1049 |
+
"step": 715
|
1050 |
+
},
|
1051 |
+
{
|
1052 |
+
"epoch": 4.390243902439025,
|
1053 |
+
"grad_norm": 0.451171875,
|
1054 |
+
"learning_rate": 8.924601061626048e-06,
|
1055 |
+
"loss": 0.3865,
|
1056 |
+
"step": 720
|
1057 |
+
},
|
1058 |
+
{
|
1059 |
+
"epoch": 4.420731707317073,
|
1060 |
+
"grad_norm": 0.455078125,
|
1061 |
+
"learning_rate": 8.066354408870048e-06,
|
1062 |
+
"loss": 0.3827,
|
1063 |
+
"step": 725
|
1064 |
+
},
|
1065 |
+
{
|
1066 |
+
"epoch": 4.451219512195122,
|
1067 |
+
"grad_norm": 0.470703125,
|
1068 |
+
"learning_rate": 7.249754889790539e-06,
|
1069 |
+
"loss": 0.3932,
|
1070 |
+
"step": 730
|
1071 |
+
},
|
1072 |
+
{
|
1073 |
+
"epoch": 4.4817073170731705,
|
1074 |
+
"grad_norm": 0.44140625,
|
1075 |
+
"learning_rate": 6.475172434559573e-06,
|
1076 |
+
"loss": 0.3762,
|
1077 |
+
"step": 735
|
1078 |
+
},
|
1079 |
+
{
|
1080 |
+
"epoch": 4.512195121951219,
|
1081 |
+
"grad_norm": 0.482421875,
|
1082 |
+
"learning_rate": 5.742957939074412e-06,
|
1083 |
+
"loss": 0.3825,
|
1084 |
+
"step": 740
|
1085 |
+
},
|
1086 |
+
{
|
1087 |
+
"epoch": 4.5426829268292686,
|
1088 |
+
"grad_norm": 0.435546875,
|
1089 |
+
"learning_rate": 5.0534431059970685e-06,
|
1090 |
+
"loss": 0.3846,
|
1091 |
+
"step": 745
|
1092 |
+
},
|
1093 |
+
{
|
1094 |
+
"epoch": 4.573170731707317,
|
1095 |
+
"grad_norm": 0.48828125,
|
1096 |
+
"learning_rate": 4.40694029448877e-06,
|
1097 |
+
"loss": 0.3748,
|
1098 |
+
"step": 750
|
1099 |
+
},
|
1100 |
+
{
|
1101 |
+
"epoch": 4.603658536585366,
|
1102 |
+
"grad_norm": 0.453125,
|
1103 |
+
"learning_rate": 3.803742378707198e-06,
|
1104 |
+
"loss": 0.4205,
|
1105 |
+
"step": 755
|
1106 |
+
},
|
1107 |
+
{
|
1108 |
+
"epoch": 4.634146341463414,
|
1109 |
+
"grad_norm": 0.47265625,
|
1110 |
+
"learning_rate": 3.2441226151306404e-06,
|
1111 |
+
"loss": 0.3816,
|
1112 |
+
"step": 760
|
1113 |
+
},
|
1114 |
+
{
|
1115 |
+
"epoch": 4.664634146341464,
|
1116 |
+
"grad_norm": 0.443359375,
|
1117 |
+
"learning_rate": 2.7283345187693264e-06,
|
1118 |
+
"loss": 0.3782,
|
1119 |
+
"step": 765
|
1120 |
+
},
|
1121 |
+
{
|
1122 |
+
"epoch": 4.695121951219512,
|
1123 |
+
"grad_norm": 0.52734375,
|
1124 |
+
"learning_rate": 2.256611748319792e-06,
|
1125 |
+
"loss": 0.3842,
|
1126 |
+
"step": 770
|
1127 |
+
},
|
1128 |
+
{
|
1129 |
+
"epoch": 4.725609756097561,
|
1130 |
+
"grad_norm": 0.453125,
|
1131 |
+
"learning_rate": 1.8291680003145073e-06,
|
1132 |
+
"loss": 0.4031,
|
1133 |
+
"step": 775
|
1134 |
+
},
|
1135 |
+
{
|
1136 |
+
"epoch": 4.7560975609756095,
|
1137 |
+
"grad_norm": 0.4453125,
|
1138 |
+
"learning_rate": 1.4461969123145457e-06,
|
1139 |
+
"loss": 0.3891,
|
1140 |
+
"step": 780
|
1141 |
+
},
|
1142 |
+
{
|
1143 |
+
"epoch": 4.786585365853659,
|
1144 |
+
"grad_norm": 0.443359375,
|
1145 |
+
"learning_rate": 1.107871975189234e-06,
|
1146 |
+
"loss": 0.365,
|
1147 |
+
"step": 785
|
1148 |
+
},
|
1149 |
+
{
|
1150 |
+
"epoch": 4.817073170731708,
|
1151 |
+
"grad_norm": 0.54296875,
|
1152 |
+
"learning_rate": 8.143464545226298e-07,
|
1153 |
+
"loss": 0.3849,
|
1154 |
+
"step": 790
|
1155 |
+
},
|
1156 |
+
{
|
1157 |
+
"epoch": 4.847560975609756,
|
1158 |
+
"grad_norm": 0.578125,
|
1159 |
+
"learning_rate": 5.657533211820942e-07,
|
1160 |
+
"loss": 0.3848,
|
1161 |
+
"step": 795
|
1162 |
+
},
|
1163 |
+
{
|
1164 |
+
"epoch": 4.878048780487805,
|
1165 |
+
"grad_norm": 0.50390625,
|
1166 |
+
"learning_rate": 3.622051910808666e-07,
|
1167 |
+
"loss": 0.3739,
|
1168 |
+
"step": 800
|
1169 |
+
},
|
1170 |
+
{
|
1171 |
+
"epoch": 4.908536585365853,
|
1172 |
+
"grad_norm": 0.47265625,
|
1173 |
+
"learning_rate": 2.037942741615617e-07,
|
1174 |
+
"loss": 0.392,
|
1175 |
+
"step": 805
|
1176 |
+
},
|
1177 |
+
{
|
1178 |
+
"epoch": 4.939024390243903,
|
1179 |
+
"grad_norm": 0.46484375,
|
1180 |
+
"learning_rate": 9.059233262386225e-08,
|
1181 |
+
"loss": 0.391,
|
1182 |
+
"step": 810
|
1183 |
+
},
|
1184 |
+
{
|
1185 |
+
"epoch": 4.969512195121951,
|
1186 |
+
"grad_norm": 0.419921875,
|
1187 |
+
"learning_rate": 2.2650648415334376e-08,
|
1188 |
+
"loss": 0.3862,
|
1189 |
+
"step": 815
|
1190 |
+
},
|
1191 |
+
{
|
1192 |
+
"epoch": 5.0,
|
1193 |
+
"grad_norm": 0.486328125,
|
1194 |
+
"learning_rate": 0.0,
|
1195 |
+
"loss": 0.4072,
|
1196 |
+
"step": 820
|
1197 |
+
},
|
1198 |
+
{
|
1199 |
+
"epoch": 5.0,
|
1200 |
+
"eval_loss": 1.248147964477539,
|
1201 |
+
"eval_runtime": 1.0787,
|
1202 |
+
"eval_samples_per_second": 1.854,
|
1203 |
+
"eval_steps_per_second": 0.927,
|
1204 |
+
"step": 820
|
1205 |
+
},
|
1206 |
+
{
|
1207 |
+
"epoch": 5.0,
|
1208 |
+
"step": 820,
|
1209 |
+
"total_flos": 6.28853392705323e+17,
|
1210 |
+
"train_loss": 1.6822980254161648,
|
1211 |
+
"train_runtime": 5646.8844,
|
1212 |
+
"train_samples_per_second": 1.159,
|
1213 |
+
"train_steps_per_second": 0.145
|
1214 |
+
}
|
1215 |
+
],
|
1216 |
+
"logging_steps": 5,
|
1217 |
+
"max_steps": 820,
|
1218 |
+
"num_input_tokens_seen": 0,
|
1219 |
+
"num_train_epochs": 5,
|
1220 |
+
"save_steps": 100,
|
1221 |
+
"total_flos": 6.28853392705323e+17,
|
1222 |
+
"train_batch_size": 2,
|
1223 |
+
"trial_name": null,
|
1224 |
+
"trial_params": null
|
1225 |
+
}
|