Qin Liu
commited on
Model save
Browse files- .gitattributes +1 -0
- README.md +69 -0
- adapter_config.json +31 -0
- adapter_model.safetensors +3 -0
- all_results.json +8 -0
- runs/Apr21_05-15-41_COE-CS-sv003/events.out.tfevents.1713676560.COE-CS-sv003.350271.0 +3 -0
- runs/Apr21_16-07-15_COE-CS-sv003/events.out.tfevents.1713715659.COE-CS-sv003.361169.0 +3 -0
- runs/Apr21_16-31-49_COE-CS-sv003/events.out.tfevents.1713717128.COE-CS-sv003.361979.0 +3 -0
- special_tokens_map.json +28 -0
- tokenizer.json +3 -0
- tokenizer_config.json +70 -0
- train_results.json +8 -0
- trainer_state.json +1426 -0
- training_args.bin +3 -0
.gitattributes
CHANGED
@@ -33,3 +33,4 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
|
|
33 |
*.zip filter=lfs diff=lfs merge=lfs -text
|
34 |
*.zst filter=lfs diff=lfs merge=lfs -text
|
35 |
*tfevents* filter=lfs diff=lfs merge=lfs -text
|
|
|
|
33 |
*.zip filter=lfs diff=lfs merge=lfs -text
|
34 |
*.zst filter=lfs diff=lfs merge=lfs -text
|
35 |
*tfevents* filter=lfs diff=lfs merge=lfs -text
|
36 |
+
tokenizer.json filter=lfs diff=lfs merge=lfs -text
|
README.md
ADDED
@@ -0,0 +1,69 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
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 |
+
model-index:
|
10 |
+
- name: zephyr-7b-gemma-sft-5p
|
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 |
+
# zephyr-7b-gemma-sft-5p
|
18 |
+
|
19 |
+
This model is a fine-tuned version of [google/gemma-7b](https://huggingface.co/google/gemma-7b) on the None dataset.
|
20 |
+
It achieves the following results on the evaluation set:
|
21 |
+
- Loss: 1.1230
|
22 |
+
|
23 |
+
## Model description
|
24 |
+
|
25 |
+
More information needed
|
26 |
+
|
27 |
+
## Intended uses & limitations
|
28 |
+
|
29 |
+
More information needed
|
30 |
+
|
31 |
+
## Training and evaluation data
|
32 |
+
|
33 |
+
More information needed
|
34 |
+
|
35 |
+
## Training procedure
|
36 |
+
|
37 |
+
### Training hyperparameters
|
38 |
+
|
39 |
+
The following hyperparameters were used during training:
|
40 |
+
- learning_rate: 0.0002
|
41 |
+
- train_batch_size: 8
|
42 |
+
- eval_batch_size: 8
|
43 |
+
- seed: 42
|
44 |
+
- distributed_type: multi-GPU
|
45 |
+
- num_devices: 4
|
46 |
+
- gradient_accumulation_steps: 2
|
47 |
+
- total_train_batch_size: 64
|
48 |
+
- total_eval_batch_size: 32
|
49 |
+
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
50 |
+
- lr_scheduler_type: cosine
|
51 |
+
- lr_scheduler_warmup_ratio: 0.1
|
52 |
+
- num_epochs: 3
|
53 |
+
|
54 |
+
### Training results
|
55 |
+
|
56 |
+
| Training Loss | Epoch | Step | Validation Loss |
|
57 |
+
|:-------------:|:-----:|:----:|:---------------:|
|
58 |
+
| 0.9772 | 1.0 | 326 | 1.1599 |
|
59 |
+
| 0.9241 | 2.0 | 652 | 1.1230 |
|
60 |
+
| 0.8687 | 3.0 | 978 | 1.1230 |
|
61 |
+
|
62 |
+
|
63 |
+
### Framework versions
|
64 |
+
|
65 |
+
- PEFT 0.7.1
|
66 |
+
- Transformers 4.39.0.dev0
|
67 |
+
- Pytorch 2.1.2
|
68 |
+
- Datasets 2.14.6
|
69 |
+
- Tokenizers 0.15.2
|
adapter_config.json
ADDED
@@ -0,0 +1,31 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"alpha_pattern": {},
|
3 |
+
"auto_mapping": null,
|
4 |
+
"base_model_name_or_path": "google/gemma-7b",
|
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": 8,
|
13 |
+
"lora_dropout": 0.05,
|
14 |
+
"megatron_config": null,
|
15 |
+
"megatron_core": "megatron.core",
|
16 |
+
"modules_to_save": null,
|
17 |
+
"peft_type": "LORA",
|
18 |
+
"r": 6,
|
19 |
+
"rank_pattern": {},
|
20 |
+
"revision": null,
|
21 |
+
"target_modules": [
|
22 |
+
"gate_proj",
|
23 |
+
"down_proj",
|
24 |
+
"o_proj",
|
25 |
+
"q_proj",
|
26 |
+
"v_proj",
|
27 |
+
"up_proj",
|
28 |
+
"k_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 sha256:4b3d6d830882ae12ab25b55781737f54e61a929617eb4925414841adbe0f6146
|
3 |
+
size 37555048
|
all_results.json
ADDED
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 3.0,
|
3 |
+
"train_loss": 1.2729851507214194,
|
4 |
+
"train_runtime": 11657.2618,
|
5 |
+
"train_samples": 20842,
|
6 |
+
"train_samples_per_second": 5.364,
|
7 |
+
"train_steps_per_second": 0.084
|
8 |
+
}
|
runs/Apr21_05-15-41_COE-CS-sv003/events.out.tfevents.1713676560.COE-CS-sv003.350271.0
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:f4944755fd06f81e122852a58195aca9a056c90307c9313ba97b31fcf8f29f47
|
3 |
+
size 4786
|
runs/Apr21_16-07-15_COE-CS-sv003/events.out.tfevents.1713715659.COE-CS-sv003.361169.0
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:3562b2473bc945727d305329d6a8406e96086641c62cd30881b694f6081da3fc
|
3 |
+
size 4786
|
runs/Apr21_16-31-49_COE-CS-sv003/events.out.tfevents.1713717128.COE-CS-sv003.361979.0
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:3a682f306b93d01bb336e8ff36eeea37304b059479d2511156a60cdae2088bbc
|
3 |
+
size 47205
|
special_tokens_map.json
ADDED
@@ -0,0 +1,28 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"additional_special_tokens": [
|
3 |
+
{
|
4 |
+
"content": "<|im_start|>",
|
5 |
+
"lstrip": false,
|
6 |
+
"normalized": false,
|
7 |
+
"rstrip": false,
|
8 |
+
"single_word": false
|
9 |
+
},
|
10 |
+
{
|
11 |
+
"content": "<|im_end|>",
|
12 |
+
"lstrip": false,
|
13 |
+
"normalized": false,
|
14 |
+
"rstrip": false,
|
15 |
+
"single_word": false
|
16 |
+
}
|
17 |
+
],
|
18 |
+
"bos_token": "<|im_start|>",
|
19 |
+
"eos_token": "<|im_end|>",
|
20 |
+
"pad_token": "<|im_end|>",
|
21 |
+
"unk_token": {
|
22 |
+
"content": "<unk>",
|
23 |
+
"lstrip": false,
|
24 |
+
"normalized": false,
|
25 |
+
"rstrip": false,
|
26 |
+
"single_word": false
|
27 |
+
}
|
28 |
+
}
|
tokenizer.json
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:22449cb9ef4bad0db7dd93b46ddff7ab7d6a654dd4f903e130ddb6361eac3af5
|
3 |
+
size 17477473
|
tokenizer_config.json
ADDED
@@ -0,0 +1,70 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"add_bos_token": false,
|
3 |
+
"add_eos_token": false,
|
4 |
+
"added_tokens_decoder": {
|
5 |
+
"0": {
|
6 |
+
"content": "<pad>",
|
7 |
+
"lstrip": false,
|
8 |
+
"normalized": false,
|
9 |
+
"rstrip": false,
|
10 |
+
"single_word": false,
|
11 |
+
"special": true
|
12 |
+
},
|
13 |
+
"1": {
|
14 |
+
"content": "<eos>",
|
15 |
+
"lstrip": false,
|
16 |
+
"normalized": false,
|
17 |
+
"rstrip": false,
|
18 |
+
"single_word": false,
|
19 |
+
"special": true
|
20 |
+
},
|
21 |
+
"2": {
|
22 |
+
"content": "<bos>",
|
23 |
+
"lstrip": false,
|
24 |
+
"normalized": false,
|
25 |
+
"rstrip": false,
|
26 |
+
"single_word": false,
|
27 |
+
"special": true
|
28 |
+
},
|
29 |
+
"3": {
|
30 |
+
"content": "<unk>",
|
31 |
+
"lstrip": false,
|
32 |
+
"normalized": false,
|
33 |
+
"rstrip": false,
|
34 |
+
"single_word": false,
|
35 |
+
"special": true
|
36 |
+
},
|
37 |
+
"106": {
|
38 |
+
"content": "<|im_start|>",
|
39 |
+
"lstrip": false,
|
40 |
+
"normalized": false,
|
41 |
+
"rstrip": false,
|
42 |
+
"single_word": false,
|
43 |
+
"special": true
|
44 |
+
},
|
45 |
+
"107": {
|
46 |
+
"content": "<|im_end|>",
|
47 |
+
"lstrip": false,
|
48 |
+
"normalized": false,
|
49 |
+
"rstrip": false,
|
50 |
+
"single_word": false,
|
51 |
+
"special": true
|
52 |
+
}
|
53 |
+
},
|
54 |
+
"additional_special_tokens": [
|
55 |
+
"<|im_start|>",
|
56 |
+
"<|im_end|>"
|
57 |
+
],
|
58 |
+
"bos_token": "<|im_start|>",
|
59 |
+
"chat_template": "{% for message in messages %}{{'<|im_start|>' + message['role'] + '\n' + message['content'] + '<|im_end|>' + '\n'}}{% endfor %}{% if add_generation_prompt %}{{ '<|im_start|>assistant\n' }}{% endif %}",
|
60 |
+
"clean_up_tokenization_spaces": false,
|
61 |
+
"eos_token": "<|im_end|>",
|
62 |
+
"legacy": null,
|
63 |
+
"model_max_length": 2048,
|
64 |
+
"pad_token": "<|im_end|>",
|
65 |
+
"sp_model_kwargs": {},
|
66 |
+
"spaces_between_special_tokens": false,
|
67 |
+
"tokenizer_class": "GemmaTokenizer",
|
68 |
+
"unk_token": "<unk>",
|
69 |
+
"use_default_system_prompt": false
|
70 |
+
}
|
train_results.json
ADDED
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 3.0,
|
3 |
+
"train_loss": 1.2729851507214194,
|
4 |
+
"train_runtime": 11657.2618,
|
5 |
+
"train_samples": 20842,
|
6 |
+
"train_samples_per_second": 5.364,
|
7 |
+
"train_steps_per_second": 0.084
|
8 |
+
}
|
trainer_state.json
ADDED
@@ -0,0 +1,1426 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"best_metric": null,
|
3 |
+
"best_model_checkpoint": null,
|
4 |
+
"epoch": 3.0,
|
5 |
+
"eval_steps": 500,
|
6 |
+
"global_step": 978,
|
7 |
+
"is_hyper_param_search": false,
|
8 |
+
"is_local_process_zero": true,
|
9 |
+
"is_world_process_zero": true,
|
10 |
+
"log_history": [
|
11 |
+
{
|
12 |
+
"epoch": 0.0,
|
13 |
+
"grad_norm": 71.33833396142957,
|
14 |
+
"learning_rate": 2.040816326530612e-06,
|
15 |
+
"loss": 16.3689,
|
16 |
+
"step": 1
|
17 |
+
},
|
18 |
+
{
|
19 |
+
"epoch": 0.02,
|
20 |
+
"grad_norm": 71.89278720209892,
|
21 |
+
"learning_rate": 1.0204081632653061e-05,
|
22 |
+
"loss": 16.1499,
|
23 |
+
"step": 5
|
24 |
+
},
|
25 |
+
{
|
26 |
+
"epoch": 0.03,
|
27 |
+
"grad_norm": 65.72662945467931,
|
28 |
+
"learning_rate": 2.0408163265306123e-05,
|
29 |
+
"loss": 15.7133,
|
30 |
+
"step": 10
|
31 |
+
},
|
32 |
+
{
|
33 |
+
"epoch": 0.05,
|
34 |
+
"grad_norm": 53.593107939489194,
|
35 |
+
"learning_rate": 3.061224489795919e-05,
|
36 |
+
"loss": 13.3181,
|
37 |
+
"step": 15
|
38 |
+
},
|
39 |
+
{
|
40 |
+
"epoch": 0.06,
|
41 |
+
"grad_norm": 44.297380165268656,
|
42 |
+
"learning_rate": 4.0816326530612245e-05,
|
43 |
+
"loss": 9.8593,
|
44 |
+
"step": 20
|
45 |
+
},
|
46 |
+
{
|
47 |
+
"epoch": 0.08,
|
48 |
+
"grad_norm": 19.985681179385026,
|
49 |
+
"learning_rate": 5.102040816326531e-05,
|
50 |
+
"loss": 6.3007,
|
51 |
+
"step": 25
|
52 |
+
},
|
53 |
+
{
|
54 |
+
"epoch": 0.09,
|
55 |
+
"grad_norm": 12.900325144421931,
|
56 |
+
"learning_rate": 6.122448979591838e-05,
|
57 |
+
"loss": 4.5204,
|
58 |
+
"step": 30
|
59 |
+
},
|
60 |
+
{
|
61 |
+
"epoch": 0.11,
|
62 |
+
"grad_norm": 9.486242109363502,
|
63 |
+
"learning_rate": 7.142857142857143e-05,
|
64 |
+
"loss": 3.3761,
|
65 |
+
"step": 35
|
66 |
+
},
|
67 |
+
{
|
68 |
+
"epoch": 0.12,
|
69 |
+
"grad_norm": 7.1186109989022945,
|
70 |
+
"learning_rate": 8.163265306122449e-05,
|
71 |
+
"loss": 2.3811,
|
72 |
+
"step": 40
|
73 |
+
},
|
74 |
+
{
|
75 |
+
"epoch": 0.14,
|
76 |
+
"grad_norm": 5.023418237589821,
|
77 |
+
"learning_rate": 9.183673469387756e-05,
|
78 |
+
"loss": 1.851,
|
79 |
+
"step": 45
|
80 |
+
},
|
81 |
+
{
|
82 |
+
"epoch": 0.15,
|
83 |
+
"grad_norm": 4.554190050004734,
|
84 |
+
"learning_rate": 0.00010204081632653062,
|
85 |
+
"loss": 1.6038,
|
86 |
+
"step": 50
|
87 |
+
},
|
88 |
+
{
|
89 |
+
"epoch": 0.17,
|
90 |
+
"grad_norm": 2.8804678557363896,
|
91 |
+
"learning_rate": 0.00011224489795918367,
|
92 |
+
"loss": 1.4206,
|
93 |
+
"step": 55
|
94 |
+
},
|
95 |
+
{
|
96 |
+
"epoch": 0.18,
|
97 |
+
"grad_norm": 3.9529502182289398,
|
98 |
+
"learning_rate": 0.00012244897959183676,
|
99 |
+
"loss": 1.3664,
|
100 |
+
"step": 60
|
101 |
+
},
|
102 |
+
{
|
103 |
+
"epoch": 0.2,
|
104 |
+
"grad_norm": 2.248441902031195,
|
105 |
+
"learning_rate": 0.0001326530612244898,
|
106 |
+
"loss": 1.274,
|
107 |
+
"step": 65
|
108 |
+
},
|
109 |
+
{
|
110 |
+
"epoch": 0.21,
|
111 |
+
"grad_norm": 1.4413020771523257,
|
112 |
+
"learning_rate": 0.00014285714285714287,
|
113 |
+
"loss": 1.2439,
|
114 |
+
"step": 70
|
115 |
+
},
|
116 |
+
{
|
117 |
+
"epoch": 0.23,
|
118 |
+
"grad_norm": 1.493793739973276,
|
119 |
+
"learning_rate": 0.0001530612244897959,
|
120 |
+
"loss": 1.1924,
|
121 |
+
"step": 75
|
122 |
+
},
|
123 |
+
{
|
124 |
+
"epoch": 0.25,
|
125 |
+
"grad_norm": 1.458616609136318,
|
126 |
+
"learning_rate": 0.00016326530612244898,
|
127 |
+
"loss": 1.0841,
|
128 |
+
"step": 80
|
129 |
+
},
|
130 |
+
{
|
131 |
+
"epoch": 0.26,
|
132 |
+
"grad_norm": 1.6382650720922636,
|
133 |
+
"learning_rate": 0.00017346938775510205,
|
134 |
+
"loss": 1.1002,
|
135 |
+
"step": 85
|
136 |
+
},
|
137 |
+
{
|
138 |
+
"epoch": 0.28,
|
139 |
+
"grad_norm": 1.326596598600597,
|
140 |
+
"learning_rate": 0.00018367346938775512,
|
141 |
+
"loss": 1.0797,
|
142 |
+
"step": 90
|
143 |
+
},
|
144 |
+
{
|
145 |
+
"epoch": 0.29,
|
146 |
+
"grad_norm": 1.2754617583755954,
|
147 |
+
"learning_rate": 0.00019387755102040816,
|
148 |
+
"loss": 1.1313,
|
149 |
+
"step": 95
|
150 |
+
},
|
151 |
+
{
|
152 |
+
"epoch": 0.31,
|
153 |
+
"grad_norm": 1.399995470448795,
|
154 |
+
"learning_rate": 0.00019999745104274993,
|
155 |
+
"loss": 1.0745,
|
156 |
+
"step": 100
|
157 |
+
},
|
158 |
+
{
|
159 |
+
"epoch": 0.32,
|
160 |
+
"grad_norm": 1.5754181414386559,
|
161 |
+
"learning_rate": 0.00019996877676598733,
|
162 |
+
"loss": 1.0513,
|
163 |
+
"step": 105
|
164 |
+
},
|
165 |
+
{
|
166 |
+
"epoch": 0.34,
|
167 |
+
"grad_norm": 1.9062455011264745,
|
168 |
+
"learning_rate": 0.00019990825118233957,
|
169 |
+
"loss": 1.0973,
|
170 |
+
"step": 110
|
171 |
+
},
|
172 |
+
{
|
173 |
+
"epoch": 0.35,
|
174 |
+
"grad_norm": 1.2370608192620602,
|
175 |
+
"learning_rate": 0.00019981589357601727,
|
176 |
+
"loss": 1.0174,
|
177 |
+
"step": 115
|
178 |
+
},
|
179 |
+
{
|
180 |
+
"epoch": 0.37,
|
181 |
+
"grad_norm": 1.4957785022540953,
|
182 |
+
"learning_rate": 0.0001996917333733128,
|
183 |
+
"loss": 1.0495,
|
184 |
+
"step": 120
|
185 |
+
},
|
186 |
+
{
|
187 |
+
"epoch": 0.38,
|
188 |
+
"grad_norm": 2.0317868737729614,
|
189 |
+
"learning_rate": 0.00019953581013322502,
|
190 |
+
"loss": 1.0631,
|
191 |
+
"step": 125
|
192 |
+
},
|
193 |
+
{
|
194 |
+
"epoch": 0.4,
|
195 |
+
"grad_norm": 2.3640165600919016,
|
196 |
+
"learning_rate": 0.00019934817353485501,
|
197 |
+
"loss": 1.0597,
|
198 |
+
"step": 130
|
199 |
+
},
|
200 |
+
{
|
201 |
+
"epoch": 0.41,
|
202 |
+
"grad_norm": 1.4377618154920253,
|
203 |
+
"learning_rate": 0.00019912888336157792,
|
204 |
+
"loss": 1.032,
|
205 |
+
"step": 135
|
206 |
+
},
|
207 |
+
{
|
208 |
+
"epoch": 0.43,
|
209 |
+
"grad_norm": 1.6672769945555381,
|
210 |
+
"learning_rate": 0.00019887800948199496,
|
211 |
+
"loss": 1.0587,
|
212 |
+
"step": 140
|
213 |
+
},
|
214 |
+
{
|
215 |
+
"epoch": 0.44,
|
216 |
+
"grad_norm": 1.7175054641214251,
|
217 |
+
"learning_rate": 0.00019859563182767268,
|
218 |
+
"loss": 0.9808,
|
219 |
+
"step": 145
|
220 |
+
},
|
221 |
+
{
|
222 |
+
"epoch": 0.46,
|
223 |
+
"grad_norm": 2.055095058889153,
|
224 |
+
"learning_rate": 0.00019828184036767556,
|
225 |
+
"loss": 1.0301,
|
226 |
+
"step": 150
|
227 |
+
},
|
228 |
+
{
|
229 |
+
"epoch": 0.48,
|
230 |
+
"grad_norm": 1.1983943009571292,
|
231 |
+
"learning_rate": 0.00019793673507990088,
|
232 |
+
"loss": 0.9844,
|
233 |
+
"step": 155
|
234 |
+
},
|
235 |
+
{
|
236 |
+
"epoch": 0.49,
|
237 |
+
"grad_norm": 1.3919505852905427,
|
238 |
+
"learning_rate": 0.00019756042591922434,
|
239 |
+
"loss": 1.0325,
|
240 |
+
"step": 160
|
241 |
+
},
|
242 |
+
{
|
243 |
+
"epoch": 0.51,
|
244 |
+
"grad_norm": 1.0559448337045312,
|
245 |
+
"learning_rate": 0.00019715303278246724,
|
246 |
+
"loss": 0.9838,
|
247 |
+
"step": 165
|
248 |
+
},
|
249 |
+
{
|
250 |
+
"epoch": 0.52,
|
251 |
+
"grad_norm": 1.2948165594267842,
|
252 |
+
"learning_rate": 0.00019671468547019573,
|
253 |
+
"loss": 1.0184,
|
254 |
+
"step": 170
|
255 |
+
},
|
256 |
+
{
|
257 |
+
"epoch": 0.54,
|
258 |
+
"grad_norm": 1.757117551491023,
|
259 |
+
"learning_rate": 0.00019624552364536473,
|
260 |
+
"loss": 0.9951,
|
261 |
+
"step": 175
|
262 |
+
},
|
263 |
+
{
|
264 |
+
"epoch": 0.55,
|
265 |
+
"grad_norm": 1.0463120070964065,
|
266 |
+
"learning_rate": 0.00019574569678881964,
|
267 |
+
"loss": 0.9886,
|
268 |
+
"step": 180
|
269 |
+
},
|
270 |
+
{
|
271 |
+
"epoch": 0.57,
|
272 |
+
"grad_norm": 1.3316373554522543,
|
273 |
+
"learning_rate": 0.00019521536415166978,
|
274 |
+
"loss": 1.029,
|
275 |
+
"step": 185
|
276 |
+
},
|
277 |
+
{
|
278 |
+
"epoch": 0.58,
|
279 |
+
"grad_norm": 1.206556314151547,
|
280 |
+
"learning_rate": 0.000194654694704549,
|
281 |
+
"loss": 1.0185,
|
282 |
+
"step": 190
|
283 |
+
},
|
284 |
+
{
|
285 |
+
"epoch": 0.6,
|
286 |
+
"grad_norm": 1.4497941420805274,
|
287 |
+
"learning_rate": 0.00019406386708377955,
|
288 |
+
"loss": 0.9952,
|
289 |
+
"step": 195
|
290 |
+
},
|
291 |
+
{
|
292 |
+
"epoch": 0.61,
|
293 |
+
"grad_norm": 1.5111306426761502,
|
294 |
+
"learning_rate": 0.00019344306953445633,
|
295 |
+
"loss": 0.9533,
|
296 |
+
"step": 200
|
297 |
+
},
|
298 |
+
{
|
299 |
+
"epoch": 0.63,
|
300 |
+
"grad_norm": 1.5977185668213372,
|
301 |
+
"learning_rate": 0.00019279249985046948,
|
302 |
+
"loss": 0.9654,
|
303 |
+
"step": 205
|
304 |
+
},
|
305 |
+
{
|
306 |
+
"epoch": 0.64,
|
307 |
+
"grad_norm": 1.4919134437495485,
|
308 |
+
"learning_rate": 0.000192112365311485,
|
309 |
+
"loss": 0.9859,
|
310 |
+
"step": 210
|
311 |
+
},
|
312 |
+
{
|
313 |
+
"epoch": 0.66,
|
314 |
+
"grad_norm": 1.5403119826226057,
|
315 |
+
"learning_rate": 0.00019140288261690276,
|
316 |
+
"loss": 1.0116,
|
317 |
+
"step": 215
|
318 |
+
},
|
319 |
+
{
|
320 |
+
"epoch": 0.67,
|
321 |
+
"grad_norm": 1.3349640842488235,
|
322 |
+
"learning_rate": 0.00019066427781681315,
|
323 |
+
"loss": 0.9787,
|
324 |
+
"step": 220
|
325 |
+
},
|
326 |
+
{
|
327 |
+
"epoch": 0.69,
|
328 |
+
"grad_norm": 1.3368327515154408,
|
329 |
+
"learning_rate": 0.00018989678623997503,
|
330 |
+
"loss": 0.9729,
|
331 |
+
"step": 225
|
332 |
+
},
|
333 |
+
{
|
334 |
+
"epoch": 0.71,
|
335 |
+
"grad_norm": 1.3225697881784793,
|
336 |
+
"learning_rate": 0.0001891006524188368,
|
337 |
+
"loss": 0.9581,
|
338 |
+
"step": 230
|
339 |
+
},
|
340 |
+
{
|
341 |
+
"epoch": 0.72,
|
342 |
+
"grad_norm": 1.2111399093084592,
|
343 |
+
"learning_rate": 0.00018827613001162532,
|
344 |
+
"loss": 0.9993,
|
345 |
+
"step": 235
|
346 |
+
},
|
347 |
+
{
|
348 |
+
"epoch": 0.74,
|
349 |
+
"grad_norm": 1.4099026636155603,
|
350 |
+
"learning_rate": 0.00018742348172152726,
|
351 |
+
"loss": 0.9647,
|
352 |
+
"step": 240
|
353 |
+
},
|
354 |
+
{
|
355 |
+
"epoch": 0.75,
|
356 |
+
"grad_norm": 1.2230878261166291,
|
357 |
+
"learning_rate": 0.00018654297921298863,
|
358 |
+
"loss": 0.9514,
|
359 |
+
"step": 245
|
360 |
+
},
|
361 |
+
{
|
362 |
+
"epoch": 0.77,
|
363 |
+
"grad_norm": 1.376607259051782,
|
364 |
+
"learning_rate": 0.0001856349030251589,
|
365 |
+
"loss": 1.0106,
|
366 |
+
"step": 250
|
367 |
+
},
|
368 |
+
{
|
369 |
+
"epoch": 0.78,
|
370 |
+
"grad_norm": 1.5573343816575618,
|
371 |
+
"learning_rate": 0.0001846995424825079,
|
372 |
+
"loss": 0.9622,
|
373 |
+
"step": 255
|
374 |
+
},
|
375 |
+
{
|
376 |
+
"epoch": 0.8,
|
377 |
+
"grad_norm": 1.0889975412225616,
|
378 |
+
"learning_rate": 0.00018373719560264327,
|
379 |
+
"loss": 1.0253,
|
380 |
+
"step": 260
|
381 |
+
},
|
382 |
+
{
|
383 |
+
"epoch": 0.81,
|
384 |
+
"grad_norm": 1.2977152056359567,
|
385 |
+
"learning_rate": 0.0001827481690013584,
|
386 |
+
"loss": 1.0099,
|
387 |
+
"step": 265
|
388 |
+
},
|
389 |
+
{
|
390 |
+
"epoch": 0.83,
|
391 |
+
"grad_norm": 1.3195763372373182,
|
392 |
+
"learning_rate": 0.0001817327777949407,
|
393 |
+
"loss": 0.9993,
|
394 |
+
"step": 270
|
395 |
+
},
|
396 |
+
{
|
397 |
+
"epoch": 0.84,
|
398 |
+
"grad_norm": 1.2289773734368,
|
399 |
+
"learning_rate": 0.00018069134549977172,
|
400 |
+
"loss": 0.9763,
|
401 |
+
"step": 275
|
402 |
+
},
|
403 |
+
{
|
404 |
+
"epoch": 0.86,
|
405 |
+
"grad_norm": 1.2177961464095397,
|
406 |
+
"learning_rate": 0.00017962420392925066,
|
407 |
+
"loss": 0.9791,
|
408 |
+
"step": 280
|
409 |
+
},
|
410 |
+
{
|
411 |
+
"epoch": 0.87,
|
412 |
+
"grad_norm": 1.1607055291626518,
|
413 |
+
"learning_rate": 0.00017853169308807448,
|
414 |
+
"loss": 0.9801,
|
415 |
+
"step": 285
|
416 |
+
},
|
417 |
+
{
|
418 |
+
"epoch": 0.89,
|
419 |
+
"grad_norm": 1.2705107901850918,
|
420 |
+
"learning_rate": 0.00017741416106390826,
|
421 |
+
"loss": 0.9328,
|
422 |
+
"step": 290
|
423 |
+
},
|
424 |
+
{
|
425 |
+
"epoch": 0.9,
|
426 |
+
"grad_norm": 1.3432344503611726,
|
427 |
+
"learning_rate": 0.00017627196391647982,
|
428 |
+
"loss": 0.955,
|
429 |
+
"step": 295
|
430 |
+
},
|
431 |
+
{
|
432 |
+
"epoch": 0.92,
|
433 |
+
"grad_norm": 0.9736443478187509,
|
434 |
+
"learning_rate": 0.00017510546556413498,
|
435 |
+
"loss": 1.0068,
|
436 |
+
"step": 300
|
437 |
+
},
|
438 |
+
{
|
439 |
+
"epoch": 0.94,
|
440 |
+
"grad_norm": 1.2298616794502824,
|
441 |
+
"learning_rate": 0.00017391503766788828,
|
442 |
+
"loss": 0.9365,
|
443 |
+
"step": 305
|
444 |
+
},
|
445 |
+
{
|
446 |
+
"epoch": 0.95,
|
447 |
+
"grad_norm": 1.2782580681474314,
|
448 |
+
"learning_rate": 0.00017270105951300738,
|
449 |
+
"loss": 0.9398,
|
450 |
+
"step": 310
|
451 |
+
},
|
452 |
+
{
|
453 |
+
"epoch": 0.97,
|
454 |
+
"grad_norm": 1.043708677429849,
|
455 |
+
"learning_rate": 0.0001714639178881678,
|
456 |
+
"loss": 1.0062,
|
457 |
+
"step": 315
|
458 |
+
},
|
459 |
+
{
|
460 |
+
"epoch": 0.98,
|
461 |
+
"grad_norm": 1.042074622226514,
|
462 |
+
"learning_rate": 0.00017020400696221737,
|
463 |
+
"loss": 0.9572,
|
464 |
+
"step": 320
|
465 |
+
},
|
466 |
+
{
|
467 |
+
"epoch": 1.0,
|
468 |
+
"grad_norm": 1.238677723678181,
|
469 |
+
"learning_rate": 0.00016892172815858894,
|
470 |
+
"loss": 0.9772,
|
471 |
+
"step": 325
|
472 |
+
},
|
473 |
+
{
|
474 |
+
"epoch": 1.0,
|
475 |
+
"eval_loss": 1.159922480583191,
|
476 |
+
"eval_runtime": 156.0746,
|
477 |
+
"eval_samples_per_second": 14.801,
|
478 |
+
"eval_steps_per_second": 0.468,
|
479 |
+
"step": 326
|
480 |
+
},
|
481 |
+
{
|
482 |
+
"epoch": 1.01,
|
483 |
+
"grad_norm": 1.0217319894407395,
|
484 |
+
"learning_rate": 0.00016761749002740193,
|
485 |
+
"loss": 0.946,
|
486 |
+
"step": 330
|
487 |
+
},
|
488 |
+
{
|
489 |
+
"epoch": 1.03,
|
490 |
+
"grad_norm": 1.1246569533018882,
|
491 |
+
"learning_rate": 0.00016629170811529318,
|
492 |
+
"loss": 0.918,
|
493 |
+
"step": 335
|
494 |
+
},
|
495 |
+
{
|
496 |
+
"epoch": 1.04,
|
497 |
+
"grad_norm": 1.002004590448949,
|
498 |
+
"learning_rate": 0.00016494480483301836,
|
499 |
+
"loss": 0.9497,
|
500 |
+
"step": 340
|
501 |
+
},
|
502 |
+
{
|
503 |
+
"epoch": 1.06,
|
504 |
+
"grad_norm": 1.212522253836943,
|
505 |
+
"learning_rate": 0.00016357720932086688,
|
506 |
+
"loss": 0.8865,
|
507 |
+
"step": 345
|
508 |
+
},
|
509 |
+
{
|
510 |
+
"epoch": 1.07,
|
511 |
+
"grad_norm": 1.1180679319761555,
|
512 |
+
"learning_rate": 0.00016218935731193224,
|
513 |
+
"loss": 0.9137,
|
514 |
+
"step": 350
|
515 |
+
},
|
516 |
+
{
|
517 |
+
"epoch": 1.09,
|
518 |
+
"grad_norm": 1.243690740596817,
|
519 |
+
"learning_rate": 0.00016078169099328197,
|
520 |
+
"loss": 0.928,
|
521 |
+
"step": 355
|
522 |
+
},
|
523 |
+
{
|
524 |
+
"epoch": 1.1,
|
525 |
+
"grad_norm": 1.0059412266802719,
|
526 |
+
"learning_rate": 0.00015935465886507142,
|
527 |
+
"loss": 0.9132,
|
528 |
+
"step": 360
|
529 |
+
},
|
530 |
+
{
|
531 |
+
"epoch": 1.12,
|
532 |
+
"grad_norm": 1.449853291817741,
|
533 |
+
"learning_rate": 0.0001579087155976459,
|
534 |
+
"loss": 0.9236,
|
535 |
+
"step": 365
|
536 |
+
},
|
537 |
+
{
|
538 |
+
"epoch": 1.13,
|
539 |
+
"grad_norm": 1.200969298535005,
|
540 |
+
"learning_rate": 0.00015644432188667695,
|
541 |
+
"loss": 0.9191,
|
542 |
+
"step": 370
|
543 |
+
},
|
544 |
+
{
|
545 |
+
"epoch": 1.15,
|
546 |
+
"grad_norm": 1.3541813597090206,
|
547 |
+
"learning_rate": 0.00015496194430637903,
|
548 |
+
"loss": 0.9882,
|
549 |
+
"step": 375
|
550 |
+
},
|
551 |
+
{
|
552 |
+
"epoch": 1.17,
|
553 |
+
"grad_norm": 1.2265813596883026,
|
554 |
+
"learning_rate": 0.00015346205516085306,
|
555 |
+
"loss": 0.9468,
|
556 |
+
"step": 380
|
557 |
+
},
|
558 |
+
{
|
559 |
+
"epoch": 1.18,
|
560 |
+
"grad_norm": 1.4522657029483004,
|
561 |
+
"learning_rate": 0.0001519451323336044,
|
562 |
+
"loss": 0.8801,
|
563 |
+
"step": 385
|
564 |
+
},
|
565 |
+
{
|
566 |
+
"epoch": 1.2,
|
567 |
+
"grad_norm": 1.0083988226706067,
|
568 |
+
"learning_rate": 0.0001504116591352832,
|
569 |
+
"loss": 0.9001,
|
570 |
+
"step": 390
|
571 |
+
},
|
572 |
+
{
|
573 |
+
"epoch": 1.21,
|
574 |
+
"grad_norm": 1.2286558458574148,
|
575 |
+
"learning_rate": 0.00014886212414969553,
|
576 |
+
"loss": 0.9405,
|
577 |
+
"step": 395
|
578 |
+
},
|
579 |
+
{
|
580 |
+
"epoch": 1.23,
|
581 |
+
"grad_norm": 1.269365799457322,
|
582 |
+
"learning_rate": 0.00014729702107813436,
|
583 |
+
"loss": 1.0354,
|
584 |
+
"step": 400
|
585 |
+
},
|
586 |
+
{
|
587 |
+
"epoch": 1.24,
|
588 |
+
"grad_norm": 1.5775962375858563,
|
589 |
+
"learning_rate": 0.00014571684858208044,
|
590 |
+
"loss": 0.9391,
|
591 |
+
"step": 405
|
592 |
+
},
|
593 |
+
{
|
594 |
+
"epoch": 1.26,
|
595 |
+
"grad_norm": 1.5593272986635398,
|
596 |
+
"learning_rate": 0.00014412211012432212,
|
597 |
+
"loss": 0.8933,
|
598 |
+
"step": 410
|
599 |
+
},
|
600 |
+
{
|
601 |
+
"epoch": 1.27,
|
602 |
+
"grad_norm": 2.0000926705831166,
|
603 |
+
"learning_rate": 0.00014251331380854603,
|
604 |
+
"loss": 0.9003,
|
605 |
+
"step": 415
|
606 |
+
},
|
607 |
+
{
|
608 |
+
"epoch": 1.29,
|
609 |
+
"grad_norm": 1.3265098276554033,
|
610 |
+
"learning_rate": 0.00014089097221744868,
|
611 |
+
"loss": 0.9548,
|
612 |
+
"step": 420
|
613 |
+
},
|
614 |
+
{
|
615 |
+
"epoch": 1.3,
|
616 |
+
"grad_norm": 0.9964611399554132,
|
617 |
+
"learning_rate": 0.00013925560224942144,
|
618 |
+
"loss": 0.9493,
|
619 |
+
"step": 425
|
620 |
+
},
|
621 |
+
{
|
622 |
+
"epoch": 1.32,
|
623 |
+
"grad_norm": 1.0573997683696912,
|
624 |
+
"learning_rate": 0.00013760772495385998,
|
625 |
+
"loss": 0.9212,
|
626 |
+
"step": 430
|
627 |
+
},
|
628 |
+
{
|
629 |
+
"epoch": 1.33,
|
630 |
+
"grad_norm": 1.066450746302158,
|
631 |
+
"learning_rate": 0.00013594786536515153,
|
632 |
+
"loss": 0.9229,
|
633 |
+
"step": 435
|
634 |
+
},
|
635 |
+
{
|
636 |
+
"epoch": 1.35,
|
637 |
+
"grad_norm": 1.030770134357563,
|
638 |
+
"learning_rate": 0.00013427655233539228,
|
639 |
+
"loss": 0.9116,
|
640 |
+
"step": 440
|
641 |
+
},
|
642 |
+
{
|
643 |
+
"epoch": 1.37,
|
644 |
+
"grad_norm": 1.0183082642006178,
|
645 |
+
"learning_rate": 0.00013259431836588843,
|
646 |
+
"loss": 0.9292,
|
647 |
+
"step": 445
|
648 |
+
},
|
649 |
+
{
|
650 |
+
"epoch": 1.38,
|
651 |
+
"grad_norm": 1.132995542647628,
|
652 |
+
"learning_rate": 0.00013090169943749476,
|
653 |
+
"loss": 0.9155,
|
654 |
+
"step": 450
|
655 |
+
},
|
656 |
+
{
|
657 |
+
"epoch": 1.4,
|
658 |
+
"grad_norm": 1.0945930214994195,
|
659 |
+
"learning_rate": 0.00012919923483984414,
|
660 |
+
"loss": 0.9098,
|
661 |
+
"step": 455
|
662 |
+
},
|
663 |
+
{
|
664 |
+
"epoch": 1.41,
|
665 |
+
"grad_norm": 0.7625306665500079,
|
666 |
+
"learning_rate": 0.00012748746699952338,
|
667 |
+
"loss": 0.8722,
|
668 |
+
"step": 460
|
669 |
+
},
|
670 |
+
{
|
671 |
+
"epoch": 1.43,
|
672 |
+
"grad_norm": 1.2062480398980977,
|
673 |
+
"learning_rate": 0.00012576694130724905,
|
674 |
+
"loss": 0.8827,
|
675 |
+
"step": 465
|
676 |
+
},
|
677 |
+
{
|
678 |
+
"epoch": 1.44,
|
679 |
+
"grad_norm": 1.0298307072907715,
|
680 |
+
"learning_rate": 0.00012403820594409924,
|
681 |
+
"loss": 0.9001,
|
682 |
+
"step": 470
|
683 |
+
},
|
684 |
+
{
|
685 |
+
"epoch": 1.46,
|
686 |
+
"grad_norm": 1.1672438274286052,
|
687 |
+
"learning_rate": 0.00012230181170685636,
|
688 |
+
"loss": 0.9191,
|
689 |
+
"step": 475
|
690 |
+
},
|
691 |
+
{
|
692 |
+
"epoch": 1.47,
|
693 |
+
"grad_norm": 0.9217837301990276,
|
694 |
+
"learning_rate": 0.00012055831183251607,
|
695 |
+
"loss": 0.9126,
|
696 |
+
"step": 480
|
697 |
+
},
|
698 |
+
{
|
699 |
+
"epoch": 1.49,
|
700 |
+
"grad_norm": 0.9859571731260532,
|
701 |
+
"learning_rate": 0.00011880826182201926,
|
702 |
+
"loss": 0.8884,
|
703 |
+
"step": 485
|
704 |
+
},
|
705 |
+
{
|
706 |
+
"epoch": 1.5,
|
707 |
+
"grad_norm": 0.9295473746329976,
|
708 |
+
"learning_rate": 0.0001170522192632624,
|
709 |
+
"loss": 0.9002,
|
710 |
+
"step": 490
|
711 |
+
},
|
712 |
+
{
|
713 |
+
"epoch": 1.52,
|
714 |
+
"grad_norm": 0.9211214138607775,
|
715 |
+
"learning_rate": 0.00011529074365344301,
|
716 |
+
"loss": 0.8813,
|
717 |
+
"step": 495
|
718 |
+
},
|
719 |
+
{
|
720 |
+
"epoch": 1.53,
|
721 |
+
"grad_norm": 0.9277063802264193,
|
722 |
+
"learning_rate": 0.00011352439622079689,
|
723 |
+
"loss": 0.9304,
|
724 |
+
"step": 500
|
725 |
+
},
|
726 |
+
{
|
727 |
+
"epoch": 1.55,
|
728 |
+
"grad_norm": 0.9836858148590184,
|
729 |
+
"learning_rate": 0.00011175373974578378,
|
730 |
+
"loss": 0.9336,
|
731 |
+
"step": 505
|
732 |
+
},
|
733 |
+
{
|
734 |
+
"epoch": 1.56,
|
735 |
+
"grad_norm": 0.9201365394724025,
|
736 |
+
"learning_rate": 0.00010997933838177827,
|
737 |
+
"loss": 0.9007,
|
738 |
+
"step": 510
|
739 |
+
},
|
740 |
+
{
|
741 |
+
"epoch": 1.58,
|
742 |
+
"grad_norm": 1.0294642137453307,
|
743 |
+
"learning_rate": 0.00010820175747532373,
|
744 |
+
"loss": 0.9182,
|
745 |
+
"step": 515
|
746 |
+
},
|
747 |
+
{
|
748 |
+
"epoch": 1.6,
|
749 |
+
"grad_norm": 0.9815519204802536,
|
750 |
+
"learning_rate": 0.00010642156338600551,
|
751 |
+
"loss": 0.95,
|
752 |
+
"step": 520
|
753 |
+
},
|
754 |
+
{
|
755 |
+
"epoch": 1.61,
|
756 |
+
"grad_norm": 1.00497732793873,
|
757 |
+
"learning_rate": 0.00010463932330600196,
|
758 |
+
"loss": 0.905,
|
759 |
+
"step": 525
|
760 |
+
},
|
761 |
+
{
|
762 |
+
"epoch": 1.63,
|
763 |
+
"grad_norm": 1.0176550178508887,
|
764 |
+
"learning_rate": 0.00010285560507936961,
|
765 |
+
"loss": 0.9369,
|
766 |
+
"step": 530
|
767 |
+
},
|
768 |
+
{
|
769 |
+
"epoch": 1.64,
|
770 |
+
"grad_norm": 0.9691655097650015,
|
771 |
+
"learning_rate": 0.0001010709770211212,
|
772 |
+
"loss": 0.9151,
|
773 |
+
"step": 535
|
774 |
+
},
|
775 |
+
{
|
776 |
+
"epoch": 1.66,
|
777 |
+
"grad_norm": 1.043651874834748,
|
778 |
+
"learning_rate": 9.928600773615307e-05,
|
779 |
+
"loss": 0.9027,
|
780 |
+
"step": 540
|
781 |
+
},
|
782 |
+
{
|
783 |
+
"epoch": 1.67,
|
784 |
+
"grad_norm": 1.008769921338687,
|
785 |
+
"learning_rate": 9.750126593808082e-05,
|
786 |
+
"loss": 0.8884,
|
787 |
+
"step": 545
|
788 |
+
},
|
789 |
+
{
|
790 |
+
"epoch": 1.69,
|
791 |
+
"grad_norm": 1.0140163426042983,
|
792 |
+
"learning_rate": 9.571732026803977e-05,
|
793 |
+
"loss": 0.9461,
|
794 |
+
"step": 550
|
795 |
+
},
|
796 |
+
{
|
797 |
+
"epoch": 1.7,
|
798 |
+
"grad_norm": 0.8868172626319482,
|
799 |
+
"learning_rate": 9.393473911350893e-05,
|
800 |
+
"loss": 0.9176,
|
801 |
+
"step": 555
|
802 |
+
},
|
803 |
+
{
|
804 |
+
"epoch": 1.72,
|
805 |
+
"grad_norm": 0.9481462007119397,
|
806 |
+
"learning_rate": 9.215409042721552e-05,
|
807 |
+
"loss": 0.8978,
|
808 |
+
"step": 560
|
809 |
+
},
|
810 |
+
{
|
811 |
+
"epoch": 1.73,
|
812 |
+
"grad_norm": 0.976465698662806,
|
813 |
+
"learning_rate": 9.037594154617812e-05,
|
814 |
+
"loss": 0.9459,
|
815 |
+
"step": 565
|
816 |
+
},
|
817 |
+
{
|
818 |
+
"epoch": 1.75,
|
819 |
+
"grad_norm": 1.0361077003461703,
|
820 |
+
"learning_rate": 8.860085901094595e-05,
|
821 |
+
"loss": 0.8965,
|
822 |
+
"step": 570
|
823 |
+
},
|
824 |
+
{
|
825 |
+
"epoch": 1.76,
|
826 |
+
"grad_norm": 0.9744756453463735,
|
827 |
+
"learning_rate": 8.682940838509207e-05,
|
828 |
+
"loss": 0.9142,
|
829 |
+
"step": 575
|
830 |
+
},
|
831 |
+
{
|
832 |
+
"epoch": 1.78,
|
833 |
+
"grad_norm": 1.1336224000575972,
|
834 |
+
"learning_rate": 8.50621540750175e-05,
|
835 |
+
"loss": 0.8745,
|
836 |
+
"step": 580
|
837 |
+
},
|
838 |
+
{
|
839 |
+
"epoch": 1.79,
|
840 |
+
"grad_norm": 0.9837911952285221,
|
841 |
+
"learning_rate": 8.329965915012451e-05,
|
842 |
+
"loss": 0.9572,
|
843 |
+
"step": 585
|
844 |
+
},
|
845 |
+
{
|
846 |
+
"epoch": 1.81,
|
847 |
+
"grad_norm": 1.0097764972768413,
|
848 |
+
"learning_rate": 8.154248516341548e-05,
|
849 |
+
"loss": 0.9534,
|
850 |
+
"step": 590
|
851 |
+
},
|
852 |
+
{
|
853 |
+
"epoch": 1.83,
|
854 |
+
"grad_norm": 0.9690942378398057,
|
855 |
+
"learning_rate": 7.979119197257505e-05,
|
856 |
+
"loss": 0.9227,
|
857 |
+
"step": 595
|
858 |
+
},
|
859 |
+
{
|
860 |
+
"epoch": 1.84,
|
861 |
+
"grad_norm": 1.0127982484169298,
|
862 |
+
"learning_rate": 7.804633756159259e-05,
|
863 |
+
"loss": 0.9084,
|
864 |
+
"step": 600
|
865 |
+
},
|
866 |
+
{
|
867 |
+
"epoch": 1.86,
|
868 |
+
"grad_norm": 0.7943207513052626,
|
869 |
+
"learning_rate": 7.63084778629813e-05,
|
870 |
+
"loss": 0.9248,
|
871 |
+
"step": 605
|
872 |
+
},
|
873 |
+
{
|
874 |
+
"epoch": 1.87,
|
875 |
+
"grad_norm": 0.9647056317342992,
|
876 |
+
"learning_rate": 7.457816658065134e-05,
|
877 |
+
"loss": 0.9431,
|
878 |
+
"step": 610
|
879 |
+
},
|
880 |
+
{
|
881 |
+
"epoch": 1.89,
|
882 |
+
"grad_norm": 1.1403897570938433,
|
883 |
+
"learning_rate": 7.285595501349258e-05,
|
884 |
+
"loss": 0.9263,
|
885 |
+
"step": 615
|
886 |
+
},
|
887 |
+
{
|
888 |
+
"epoch": 1.9,
|
889 |
+
"grad_norm": 1.0404617304378672,
|
890 |
+
"learning_rate": 7.114239187972416e-05,
|
891 |
+
"loss": 0.9103,
|
892 |
+
"step": 620
|
893 |
+
},
|
894 |
+
{
|
895 |
+
"epoch": 1.92,
|
896 |
+
"grad_norm": 1.128471625008042,
|
897 |
+
"learning_rate": 6.94380231420656e-05,
|
898 |
+
"loss": 0.9093,
|
899 |
+
"step": 625
|
900 |
+
},
|
901 |
+
{
|
902 |
+
"epoch": 1.93,
|
903 |
+
"grad_norm": 0.9916742401674975,
|
904 |
+
"learning_rate": 6.774339183378663e-05,
|
905 |
+
"loss": 0.9331,
|
906 |
+
"step": 630
|
907 |
+
},
|
908 |
+
{
|
909 |
+
"epoch": 1.95,
|
910 |
+
"grad_norm": 0.8859012774344858,
|
911 |
+
"learning_rate": 6.60590378856896e-05,
|
912 |
+
"loss": 0.8951,
|
913 |
+
"step": 635
|
914 |
+
},
|
915 |
+
{
|
916 |
+
"epoch": 1.96,
|
917 |
+
"grad_norm": 0.9040358847526676,
|
918 |
+
"learning_rate": 6.438549795408106e-05,
|
919 |
+
"loss": 0.9024,
|
920 |
+
"step": 640
|
921 |
+
},
|
922 |
+
{
|
923 |
+
"epoch": 1.98,
|
924 |
+
"grad_norm": 0.9311904528713643,
|
925 |
+
"learning_rate": 6.272330524978613e-05,
|
926 |
+
"loss": 0.9135,
|
927 |
+
"step": 645
|
928 |
+
},
|
929 |
+
{
|
930 |
+
"epoch": 1.99,
|
931 |
+
"grad_norm": 0.9264414330433429,
|
932 |
+
"learning_rate": 6.107298936826086e-05,
|
933 |
+
"loss": 0.9241,
|
934 |
+
"step": 650
|
935 |
+
},
|
936 |
+
{
|
937 |
+
"epoch": 2.0,
|
938 |
+
"eval_loss": 1.123016595840454,
|
939 |
+
"eval_runtime": 155.2791,
|
940 |
+
"eval_samples_per_second": 14.876,
|
941 |
+
"eval_steps_per_second": 0.47,
|
942 |
+
"step": 652
|
943 |
+
},
|
944 |
+
{
|
945 |
+
"epoch": 2.01,
|
946 |
+
"grad_norm": 2.0610368602503826,
|
947 |
+
"learning_rate": 5.9435076120856616e-05,
|
948 |
+
"loss": 0.8977,
|
949 |
+
"step": 655
|
950 |
+
},
|
951 |
+
{
|
952 |
+
"epoch": 2.02,
|
953 |
+
"grad_norm": 1.1812695122093289,
|
954 |
+
"learning_rate": 5.7810087367289744e-05,
|
955 |
+
"loss": 0.8808,
|
956 |
+
"step": 660
|
957 |
+
},
|
958 |
+
{
|
959 |
+
"epoch": 2.04,
|
960 |
+
"grad_norm": 1.1330049331313394,
|
961 |
+
"learning_rate": 5.619854084937085e-05,
|
962 |
+
"loss": 0.8697,
|
963 |
+
"step": 665
|
964 |
+
},
|
965 |
+
{
|
966 |
+
"epoch": 2.06,
|
967 |
+
"grad_norm": 1.2350886571017046,
|
968 |
+
"learning_rate": 5.4600950026045326e-05,
|
969 |
+
"loss": 0.8605,
|
970 |
+
"step": 670
|
971 |
+
},
|
972 |
+
{
|
973 |
+
"epoch": 2.07,
|
974 |
+
"grad_norm": 1.0239529925811819,
|
975 |
+
"learning_rate": 5.301782390979929e-05,
|
976 |
+
"loss": 0.9032,
|
977 |
+
"step": 675
|
978 |
+
},
|
979 |
+
{
|
980 |
+
"epoch": 2.09,
|
981 |
+
"grad_norm": 1.1559449179497776,
|
982 |
+
"learning_rate": 5.1449666904481585e-05,
|
983 |
+
"loss": 0.8576,
|
984 |
+
"step": 680
|
985 |
+
},
|
986 |
+
{
|
987 |
+
"epoch": 2.1,
|
988 |
+
"grad_norm": 1.0036439705356706,
|
989 |
+
"learning_rate": 4.989697864459452e-05,
|
990 |
+
"loss": 0.8811,
|
991 |
+
"step": 685
|
992 |
+
},
|
993 |
+
{
|
994 |
+
"epoch": 2.12,
|
995 |
+
"grad_norm": 0.9058518668863318,
|
996 |
+
"learning_rate": 4.836025383610382e-05,
|
997 |
+
"loss": 0.8795,
|
998 |
+
"step": 690
|
999 |
+
},
|
1000 |
+
{
|
1001 |
+
"epoch": 2.13,
|
1002 |
+
"grad_norm": 1.098710487990691,
|
1003 |
+
"learning_rate": 4.683998209881943e-05,
|
1004 |
+
"loss": 0.8806,
|
1005 |
+
"step": 695
|
1006 |
+
},
|
1007 |
+
{
|
1008 |
+
"epoch": 2.15,
|
1009 |
+
"grad_norm": 1.0781974748771004,
|
1010 |
+
"learning_rate": 4.5336647810396215e-05,
|
1011 |
+
"loss": 0.8598,
|
1012 |
+
"step": 700
|
1013 |
+
},
|
1014 |
+
{
|
1015 |
+
"epoch": 2.16,
|
1016 |
+
"grad_norm": 1.0706338013924643,
|
1017 |
+
"learning_rate": 4.385072995200532e-05,
|
1018 |
+
"loss": 0.8635,
|
1019 |
+
"step": 705
|
1020 |
+
},
|
1021 |
+
{
|
1022 |
+
"epoch": 2.18,
|
1023 |
+
"grad_norm": 0.9548080567529033,
|
1024 |
+
"learning_rate": 4.238270195572472e-05,
|
1025 |
+
"loss": 0.8757,
|
1026 |
+
"step": 710
|
1027 |
+
},
|
1028 |
+
{
|
1029 |
+
"epoch": 2.19,
|
1030 |
+
"grad_norm": 1.001721295347845,
|
1031 |
+
"learning_rate": 4.093303155369771e-05,
|
1032 |
+
"loss": 0.8919,
|
1033 |
+
"step": 715
|
1034 |
+
},
|
1035 |
+
{
|
1036 |
+
"epoch": 2.21,
|
1037 |
+
"grad_norm": 1.0089807158142916,
|
1038 |
+
"learning_rate": 3.9502180629107756e-05,
|
1039 |
+
"loss": 0.8483,
|
1040 |
+
"step": 720
|
1041 |
+
},
|
1042 |
+
{
|
1043 |
+
"epoch": 2.22,
|
1044 |
+
"grad_norm": 0.903535471960529,
|
1045 |
+
"learning_rate": 3.8090605069016595e-05,
|
1046 |
+
"loss": 0.9141,
|
1047 |
+
"step": 725
|
1048 |
+
},
|
1049 |
+
{
|
1050 |
+
"epoch": 2.24,
|
1051 |
+
"grad_norm": 0.9151685871279929,
|
1052 |
+
"learning_rate": 3.669875461911297e-05,
|
1053 |
+
"loss": 0.8872,
|
1054 |
+
"step": 730
|
1055 |
+
},
|
1056 |
+
{
|
1057 |
+
"epoch": 2.25,
|
1058 |
+
"grad_norm": 1.1222577897364008,
|
1059 |
+
"learning_rate": 3.53270727404179e-05,
|
1060 |
+
"loss": 0.8424,
|
1061 |
+
"step": 735
|
1062 |
+
},
|
1063 |
+
{
|
1064 |
+
"epoch": 2.27,
|
1065 |
+
"grad_norm": 0.9362543386209857,
|
1066 |
+
"learning_rate": 3.397599646799256e-05,
|
1067 |
+
"loss": 0.9175,
|
1068 |
+
"step": 740
|
1069 |
+
},
|
1070 |
+
{
|
1071 |
+
"epoch": 2.29,
|
1072 |
+
"grad_norm": 1.0462736025526818,
|
1073 |
+
"learning_rate": 3.2645956271693257e-05,
|
1074 |
+
"loss": 0.8897,
|
1075 |
+
"step": 745
|
1076 |
+
},
|
1077 |
+
{
|
1078 |
+
"epoch": 2.3,
|
1079 |
+
"grad_norm": 0.9259326236334475,
|
1080 |
+
"learning_rate": 3.133737591901864e-05,
|
1081 |
+
"loss": 0.8671,
|
1082 |
+
"step": 750
|
1083 |
+
},
|
1084 |
+
{
|
1085 |
+
"epoch": 2.32,
|
1086 |
+
"grad_norm": 0.9219118148042651,
|
1087 |
+
"learning_rate": 3.0050672340091725e-05,
|
1088 |
+
"loss": 0.9011,
|
1089 |
+
"step": 755
|
1090 |
+
},
|
1091 |
+
{
|
1092 |
+
"epoch": 2.33,
|
1093 |
+
"grad_norm": 0.8476861060157034,
|
1094 |
+
"learning_rate": 2.8786255494820835e-05,
|
1095 |
+
"loss": 0.9027,
|
1096 |
+
"step": 760
|
1097 |
+
},
|
1098 |
+
{
|
1099 |
+
"epoch": 2.35,
|
1100 |
+
"grad_norm": 0.847331969421794,
|
1101 |
+
"learning_rate": 2.754452824228132e-05,
|
1102 |
+
"loss": 0.8339,
|
1103 |
+
"step": 765
|
1104 |
+
},
|
1105 |
+
{
|
1106 |
+
"epoch": 2.36,
|
1107 |
+
"grad_norm": 0.9426248637438077,
|
1108 |
+
"learning_rate": 2.6325886212359498e-05,
|
1109 |
+
"loss": 0.9129,
|
1110 |
+
"step": 770
|
1111 |
+
},
|
1112 |
+
{
|
1113 |
+
"epoch": 2.38,
|
1114 |
+
"grad_norm": 1.0772706903168978,
|
1115 |
+
"learning_rate": 2.51307176797001e-05,
|
1116 |
+
"loss": 0.9141,
|
1117 |
+
"step": 775
|
1118 |
+
},
|
1119 |
+
{
|
1120 |
+
"epoch": 2.39,
|
1121 |
+
"grad_norm": 0.921621759803798,
|
1122 |
+
"learning_rate": 2.3959403439996907e-05,
|
1123 |
+
"loss": 0.8866,
|
1124 |
+
"step": 780
|
1125 |
+
},
|
1126 |
+
{
|
1127 |
+
"epoch": 2.41,
|
1128 |
+
"grad_norm": 1.0228032129160294,
|
1129 |
+
"learning_rate": 2.2812316688666737e-05,
|
1130 |
+
"loss": 0.8411,
|
1131 |
+
"step": 785
|
1132 |
+
},
|
1133 |
+
{
|
1134 |
+
"epoch": 2.42,
|
1135 |
+
"grad_norm": 0.9794026584875383,
|
1136 |
+
"learning_rate": 2.1689822901944457e-05,
|
1137 |
+
"loss": 0.8148,
|
1138 |
+
"step": 790
|
1139 |
+
},
|
1140 |
+
{
|
1141 |
+
"epoch": 2.44,
|
1142 |
+
"grad_norm": 1.1116773627967562,
|
1143 |
+
"learning_rate": 2.0592279720437858e-05,
|
1144 |
+
"loss": 0.9003,
|
1145 |
+
"step": 795
|
1146 |
+
},
|
1147 |
+
{
|
1148 |
+
"epoch": 2.45,
|
1149 |
+
"grad_norm": 0.9563546052669486,
|
1150 |
+
"learning_rate": 1.9520036835178668e-05,
|
1151 |
+
"loss": 0.8883,
|
1152 |
+
"step": 800
|
1153 |
+
},
|
1154 |
+
{
|
1155 |
+
"epoch": 2.47,
|
1156 |
+
"grad_norm": 0.9044636934616683,
|
1157 |
+
"learning_rate": 1.847343587620679e-05,
|
1158 |
+
"loss": 0.8769,
|
1159 |
+
"step": 805
|
1160 |
+
},
|
1161 |
+
{
|
1162 |
+
"epoch": 2.48,
|
1163 |
+
"grad_norm": 0.9221646859533941,
|
1164 |
+
"learning_rate": 1.74528103037226e-05,
|
1165 |
+
"loss": 0.8033,
|
1166 |
+
"step": 810
|
1167 |
+
},
|
1168 |
+
{
|
1169 |
+
"epoch": 2.5,
|
1170 |
+
"grad_norm": 1.0009290473276315,
|
1171 |
+
"learning_rate": 1.645848530184233e-05,
|
1172 |
+
"loss": 0.8381,
|
1173 |
+
"step": 815
|
1174 |
+
},
|
1175 |
+
{
|
1176 |
+
"epoch": 2.52,
|
1177 |
+
"grad_norm": 0.9671119344154833,
|
1178 |
+
"learning_rate": 1.5490777674990376e-05,
|
1179 |
+
"loss": 0.8645,
|
1180 |
+
"step": 820
|
1181 |
+
},
|
1182 |
+
{
|
1183 |
+
"epoch": 2.53,
|
1184 |
+
"grad_norm": 0.9041738041322639,
|
1185 |
+
"learning_rate": 1.4549995746961332e-05,
|
1186 |
+
"loss": 0.8624,
|
1187 |
+
"step": 825
|
1188 |
+
},
|
1189 |
+
{
|
1190 |
+
"epoch": 2.55,
|
1191 |
+
"grad_norm": 0.9947155263280221,
|
1192 |
+
"learning_rate": 1.3636439262684298e-05,
|
1193 |
+
"loss": 0.8862,
|
1194 |
+
"step": 830
|
1195 |
+
},
|
1196 |
+
{
|
1197 |
+
"epoch": 2.56,
|
1198 |
+
"grad_norm": 1.1355384323430027,
|
1199 |
+
"learning_rate": 1.2750399292720283e-05,
|
1200 |
+
"loss": 0.8914,
|
1201 |
+
"step": 835
|
1202 |
+
},
|
1203 |
+
{
|
1204 |
+
"epoch": 2.58,
|
1205 |
+
"grad_norm": 0.9481996675285544,
|
1206 |
+
"learning_rate": 1.1892158140523546e-05,
|
1207 |
+
"loss": 0.8962,
|
1208 |
+
"step": 840
|
1209 |
+
},
|
1210 |
+
{
|
1211 |
+
"epoch": 2.59,
|
1212 |
+
"grad_norm": 0.963461677617234,
|
1213 |
+
"learning_rate": 1.1061989252496053e-05,
|
1214 |
+
"loss": 0.8738,
|
1215 |
+
"step": 845
|
1216 |
+
},
|
1217 |
+
{
|
1218 |
+
"epoch": 2.61,
|
1219 |
+
"grad_norm": 0.9731419353271179,
|
1220 |
+
"learning_rate": 1.026015713086418e-05,
|
1221 |
+
"loss": 0.8955,
|
1222 |
+
"step": 850
|
1223 |
+
},
|
1224 |
+
{
|
1225 |
+
"epoch": 2.62,
|
1226 |
+
"grad_norm": 0.9744266823325466,
|
1227 |
+
"learning_rate": 9.486917249404815e-06,
|
1228 |
+
"loss": 0.9313,
|
1229 |
+
"step": 855
|
1230 |
+
},
|
1231 |
+
{
|
1232 |
+
"epoch": 2.64,
|
1233 |
+
"grad_norm": 0.8667111998887408,
|
1234 |
+
"learning_rate": 8.742515972048404e-06,
|
1235 |
+
"loss": 0.83,
|
1236 |
+
"step": 860
|
1237 |
+
},
|
1238 |
+
{
|
1239 |
+
"epoch": 2.65,
|
1240 |
+
"grad_norm": 1.007522159578116,
|
1241 |
+
"learning_rate": 8.027190474384127e-06,
|
1242 |
+
"loss": 0.9093,
|
1243 |
+
"step": 865
|
1244 |
+
},
|
1245 |
+
{
|
1246 |
+
"epoch": 2.67,
|
1247 |
+
"grad_norm": 0.9497316726005834,
|
1248 |
+
"learning_rate": 7.341168668092857e-06,
|
1249 |
+
"loss": 0.8619,
|
1250 |
+
"step": 870
|
1251 |
+
},
|
1252 |
+
{
|
1253 |
+
"epoch": 2.68,
|
1254 |
+
"grad_norm": 0.9890888660929842,
|
1255 |
+
"learning_rate": 6.684669128331655e-06,
|
1256 |
+
"loss": 0.8994,
|
1257 |
+
"step": 875
|
1258 |
+
},
|
1259 |
+
{
|
1260 |
+
"epoch": 2.7,
|
1261 |
+
"grad_norm": 0.9637211127886939,
|
1262 |
+
"learning_rate": 6.057901024092949e-06,
|
1263 |
+
"loss": 0.8815,
|
1264 |
+
"step": 880
|
1265 |
+
},
|
1266 |
+
{
|
1267 |
+
"epoch": 2.71,
|
1268 |
+
"grad_norm": 0.9454153835142218,
|
1269 |
+
"learning_rate": 5.461064051560705e-06,
|
1270 |
+
"loss": 0.8748,
|
1271 |
+
"step": 885
|
1272 |
+
},
|
1273 |
+
{
|
1274 |
+
"epoch": 2.73,
|
1275 |
+
"grad_norm": 0.9656325292151272,
|
1276 |
+
"learning_rate": 4.8943483704846475e-06,
|
1277 |
+
"loss": 0.9164,
|
1278 |
+
"step": 890
|
1279 |
+
},
|
1280 |
+
{
|
1281 |
+
"epoch": 2.75,
|
1282 |
+
"grad_norm": 1.167721756716136,
|
1283 |
+
"learning_rate": 4.357934543593045e-06,
|
1284 |
+
"loss": 0.8695,
|
1285 |
+
"step": 895
|
1286 |
+
},
|
1287 |
+
{
|
1288 |
+
"epoch": 2.76,
|
1289 |
+
"grad_norm": 1.0842862872661083,
|
1290 |
+
"learning_rate": 3.851993479063154e-06,
|
1291 |
+
"loss": 0.864,
|
1292 |
+
"step": 900
|
1293 |
+
},
|
1294 |
+
{
|
1295 |
+
"epoch": 2.78,
|
1296 |
+
"grad_norm": 0.9136231470411893,
|
1297 |
+
"learning_rate": 3.376686376067695e-06,
|
1298 |
+
"loss": 0.8723,
|
1299 |
+
"step": 905
|
1300 |
+
},
|
1301 |
+
{
|
1302 |
+
"epoch": 2.79,
|
1303 |
+
"grad_norm": 0.8333735277456533,
|
1304 |
+
"learning_rate": 2.9321646734147502e-06,
|
1305 |
+
"loss": 0.8476,
|
1306 |
+
"step": 910
|
1307 |
+
},
|
1308 |
+
{
|
1309 |
+
"epoch": 2.81,
|
1310 |
+
"grad_norm": 0.9013760974476497,
|
1311 |
+
"learning_rate": 2.51857000129756e-06,
|
1312 |
+
"loss": 0.8046,
|
1313 |
+
"step": 915
|
1314 |
+
},
|
1315 |
+
{
|
1316 |
+
"epoch": 2.82,
|
1317 |
+
"grad_norm": 0.9502325737857031,
|
1318 |
+
"learning_rate": 2.1360341361692517e-06,
|
1319 |
+
"loss": 0.8794,
|
1320 |
+
"step": 920
|
1321 |
+
},
|
1322 |
+
{
|
1323 |
+
"epoch": 2.84,
|
1324 |
+
"grad_norm": 1.0665384348019042,
|
1325 |
+
"learning_rate": 1.784678958757291e-06,
|
1326 |
+
"loss": 0.8471,
|
1327 |
+
"step": 925
|
1328 |
+
},
|
1329 |
+
{
|
1330 |
+
"epoch": 2.85,
|
1331 |
+
"grad_norm": 0.867152754297921,
|
1332 |
+
"learning_rate": 1.4646164152307018e-06,
|
1333 |
+
"loss": 0.8471,
|
1334 |
+
"step": 930
|
1335 |
+
},
|
1336 |
+
{
|
1337 |
+
"epoch": 2.87,
|
1338 |
+
"grad_norm": 0.8765872693616653,
|
1339 |
+
"learning_rate": 1.1759484815326294e-06,
|
1340 |
+
"loss": 0.8778,
|
1341 |
+
"step": 935
|
1342 |
+
},
|
1343 |
+
{
|
1344 |
+
"epoch": 2.88,
|
1345 |
+
"grad_norm": 0.9977478387159145,
|
1346 |
+
"learning_rate": 9.187671308895418e-07,
|
1347 |
+
"loss": 0.8814,
|
1348 |
+
"step": 940
|
1349 |
+
},
|
1350 |
+
{
|
1351 |
+
"epoch": 2.9,
|
1352 |
+
"grad_norm": 1.0928634095287688,
|
1353 |
+
"learning_rate": 6.931543045073708e-07,
|
1354 |
+
"loss": 0.8518,
|
1355 |
+
"step": 945
|
1356 |
+
},
|
1357 |
+
{
|
1358 |
+
"epoch": 2.91,
|
1359 |
+
"grad_norm": 0.9615275585544613,
|
1360 |
+
"learning_rate": 4.991818854640395e-07,
|
1361 |
+
"loss": 0.8465,
|
1362 |
+
"step": 950
|
1363 |
+
},
|
1364 |
+
{
|
1365 |
+
"epoch": 2.93,
|
1366 |
+
"grad_norm": 0.946638261056933,
|
1367 |
+
"learning_rate": 3.369116758066171e-07,
|
1368 |
+
"loss": 0.8591,
|
1369 |
+
"step": 955
|
1370 |
+
},
|
1371 |
+
{
|
1372 |
+
"epoch": 2.94,
|
1373 |
+
"grad_norm": 0.9273052910302554,
|
1374 |
+
"learning_rate": 2.0639537686037991e-07,
|
1375 |
+
"loss": 0.8652,
|
1376 |
+
"step": 960
|
1377 |
+
},
|
1378 |
+
{
|
1379 |
+
"epoch": 2.96,
|
1380 |
+
"grad_norm": 0.9396149477675896,
|
1381 |
+
"learning_rate": 1.0767457275615567e-07,
|
1382 |
+
"loss": 0.8919,
|
1383 |
+
"step": 965
|
1384 |
+
},
|
1385 |
+
{
|
1386 |
+
"epoch": 2.98,
|
1387 |
+
"grad_norm": 0.9064011249970909,
|
1388 |
+
"learning_rate": 4.078071718107701e-08,
|
1389 |
+
"loss": 0.8436,
|
1390 |
+
"step": 970
|
1391 |
+
},
|
1392 |
+
{
|
1393 |
+
"epoch": 2.99,
|
1394 |
+
"grad_norm": 0.9696335526663976,
|
1395 |
+
"learning_rate": 5.735123357042404e-09,
|
1396 |
+
"loss": 0.8687,
|
1397 |
+
"step": 975
|
1398 |
+
},
|
1399 |
+
{
|
1400 |
+
"epoch": 3.0,
|
1401 |
+
"eval_loss": 1.123045563697815,
|
1402 |
+
"eval_runtime": 155.2182,
|
1403 |
+
"eval_samples_per_second": 14.882,
|
1404 |
+
"eval_steps_per_second": 0.47,
|
1405 |
+
"step": 978
|
1406 |
+
},
|
1407 |
+
{
|
1408 |
+
"epoch": 3.0,
|
1409 |
+
"step": 978,
|
1410 |
+
"total_flos": 2513689986990080.0,
|
1411 |
+
"train_loss": 1.2729851507214194,
|
1412 |
+
"train_runtime": 11657.2618,
|
1413 |
+
"train_samples_per_second": 5.364,
|
1414 |
+
"train_steps_per_second": 0.084
|
1415 |
+
}
|
1416 |
+
],
|
1417 |
+
"logging_steps": 5,
|
1418 |
+
"max_steps": 978,
|
1419 |
+
"num_input_tokens_seen": 0,
|
1420 |
+
"num_train_epochs": 3,
|
1421 |
+
"save_steps": 500,
|
1422 |
+
"total_flos": 2513689986990080.0,
|
1423 |
+
"train_batch_size": 8,
|
1424 |
+
"trial_name": null,
|
1425 |
+
"trial_params": null
|
1426 |
+
}
|
training_args.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:dbedf4e33ee67743716fef2c59a0eb1bc0c394c1c33b0774fcc2d657c3e9c438
|
3 |
+
size 6136
|