Commit
•
15defba
1
Parent(s):
1321507
Model save
Browse files- README.md +84 -0
- adapter_config.json +31 -0
- adapter_model.safetensors +3 -0
- all_results.json +9 -0
- special_tokens_map.json +24 -0
- tokenizer.json +0 -0
- tokenizer_config.json +43 -0
- train_results.json +9 -0
- trainer_state.json +1254 -0
- training_args.bin +3 -0
README.md
ADDED
|
@@ -0,0 +1,84 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
license: apache-2.0
|
| 3 |
+
library_name: peft
|
| 4 |
+
tags:
|
| 5 |
+
- trl
|
| 6 |
+
- sft
|
| 7 |
+
- generated_from_trainer
|
| 8 |
+
base_model: mistralai/Mistral-7B-v0.1
|
| 9 |
+
datasets:
|
| 10 |
+
- generator
|
| 11 |
+
model-index:
|
| 12 |
+
- name: sexed-mistral-7b-sft-lora-v3
|
| 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 |
+
# sexed-mistral-7b-sft-lora-v3
|
| 20 |
+
|
| 21 |
+
This model is a fine-tuned version of [mistralai/Mistral-7B-v0.1](https://huggingface.co/mistralai/Mistral-7B-v0.1) on the generator dataset.
|
| 22 |
+
It achieves the following results on the evaluation set:
|
| 23 |
+
- Loss: 0.2857
|
| 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: 1e-05
|
| 43 |
+
- train_batch_size: 4
|
| 44 |
+
- eval_batch_size: 4
|
| 45 |
+
- seed: 42
|
| 46 |
+
- gradient_accumulation_steps: 16
|
| 47 |
+
- total_train_batch_size: 64
|
| 48 |
+
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 49 |
+
- lr_scheduler_type: cosine
|
| 50 |
+
- num_epochs: 20
|
| 51 |
+
|
| 52 |
+
### Training results
|
| 53 |
+
|
| 54 |
+
| Training Loss | Epoch | Step | Validation Loss |
|
| 55 |
+
|:-------------:|:-------:|:----:|:---------------:|
|
| 56 |
+
| 0.4906 | 0.9902 | 38 | 0.4776 |
|
| 57 |
+
| 0.3966 | 1.9805 | 76 | 0.4122 |
|
| 58 |
+
| 0.3648 | 2.9967 | 115 | 0.3658 |
|
| 59 |
+
| 0.3405 | 3.9870 | 153 | 0.3372 |
|
| 60 |
+
| 0.3153 | 4.9772 | 191 | 0.3180 |
|
| 61 |
+
| 0.2996 | 5.9935 | 230 | 0.3043 |
|
| 62 |
+
| 0.2987 | 6.9837 | 268 | 0.2969 |
|
| 63 |
+
| 0.2888 | 8.0 | 307 | 0.2923 |
|
| 64 |
+
| 0.2899 | 8.9902 | 345 | 0.2898 |
|
| 65 |
+
| 0.2873 | 9.9805 | 383 | 0.2883 |
|
| 66 |
+
| 0.2831 | 10.9967 | 422 | 0.2872 |
|
| 67 |
+
| 0.2773 | 11.9870 | 460 | 0.2866 |
|
| 68 |
+
| 0.2814 | 12.9772 | 498 | 0.2862 |
|
| 69 |
+
| 0.2781 | 13.9935 | 537 | 0.2860 |
|
| 70 |
+
| 0.2845 | 14.9837 | 575 | 0.2858 |
|
| 71 |
+
| 0.29 | 16.0 | 614 | 0.2858 |
|
| 72 |
+
| 0.2799 | 16.9902 | 652 | 0.2857 |
|
| 73 |
+
| 0.2825 | 17.9805 | 690 | 0.2857 |
|
| 74 |
+
| 0.2829 | 18.9967 | 729 | 0.2857 |
|
| 75 |
+
| 0.2801 | 19.8046 | 760 | 0.2857 |
|
| 76 |
+
|
| 77 |
+
|
| 78 |
+
### Framework versions
|
| 79 |
+
|
| 80 |
+
- PEFT 0.10.0
|
| 81 |
+
- Transformers 4.40.2
|
| 82 |
+
- Pytorch 2.1.1+cu121
|
| 83 |
+
- Datasets 2.19.1
|
| 84 |
+
- Tokenizers 0.19.1
|
adapter_config.json
ADDED
|
@@ -0,0 +1,31 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"alpha_pattern": {},
|
| 3 |
+
"auto_mapping": null,
|
| 4 |
+
"base_model_name_or_path": "mistralai/Mistral-7B-v0.1",
|
| 5 |
+
"bias": "none",
|
| 6 |
+
"fan_in_fan_out": false,
|
| 7 |
+
"inference_mode": true,
|
| 8 |
+
"init_lora_weights": true,
|
| 9 |
+
"layer_replication": null,
|
| 10 |
+
"layers_pattern": null,
|
| 11 |
+
"layers_to_transform": null,
|
| 12 |
+
"loftq_config": {},
|
| 13 |
+
"lora_alpha": 16,
|
| 14 |
+
"lora_dropout": 0.1,
|
| 15 |
+
"megatron_config": null,
|
| 16 |
+
"megatron_core": "megatron.core",
|
| 17 |
+
"modules_to_save": null,
|
| 18 |
+
"peft_type": "LORA",
|
| 19 |
+
"r": 64,
|
| 20 |
+
"rank_pattern": {},
|
| 21 |
+
"revision": null,
|
| 22 |
+
"target_modules": [
|
| 23 |
+
"q_proj",
|
| 24 |
+
"v_proj",
|
| 25 |
+
"o_proj",
|
| 26 |
+
"k_proj"
|
| 27 |
+
],
|
| 28 |
+
"task_type": "CAUSAL_LM",
|
| 29 |
+
"use_dora": false,
|
| 30 |
+
"use_rslora": false
|
| 31 |
+
}
|
adapter_model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:bfb281a28c13e74d4c00fae2a1e5564b611e98f83dce31d1ea306d96657ea750
|
| 3 |
+
size 109086672
|
all_results.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 19.80456026058632,
|
| 3 |
+
"total_flos": 4.2808751411995607e+18,
|
| 4 |
+
"train_loss": 0.32058246888612446,
|
| 5 |
+
"train_runtime": 79494.9932,
|
| 6 |
+
"train_samples": 14551,
|
| 7 |
+
"train_samples_per_second": 0.618,
|
| 8 |
+
"train_steps_per_second": 0.01
|
| 9 |
+
}
|
special_tokens_map.json
ADDED
|
@@ -0,0 +1,24 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"bos_token": {
|
| 3 |
+
"content": "<s>",
|
| 4 |
+
"lstrip": false,
|
| 5 |
+
"normalized": false,
|
| 6 |
+
"rstrip": false,
|
| 7 |
+
"single_word": false
|
| 8 |
+
},
|
| 9 |
+
"eos_token": {
|
| 10 |
+
"content": "</s>",
|
| 11 |
+
"lstrip": false,
|
| 12 |
+
"normalized": false,
|
| 13 |
+
"rstrip": false,
|
| 14 |
+
"single_word": false
|
| 15 |
+
},
|
| 16 |
+
"pad_token": "</s>",
|
| 17 |
+
"unk_token": {
|
| 18 |
+
"content": "<unk>",
|
| 19 |
+
"lstrip": false,
|
| 20 |
+
"normalized": false,
|
| 21 |
+
"rstrip": false,
|
| 22 |
+
"single_word": false
|
| 23 |
+
}
|
| 24 |
+
}
|
tokenizer.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
tokenizer_config.json
ADDED
|
@@ -0,0 +1,43 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"add_bos_token": true,
|
| 3 |
+
"add_eos_token": false,
|
| 4 |
+
"added_tokens_decoder": {
|
| 5 |
+
"0": {
|
| 6 |
+
"content": "<unk>",
|
| 7 |
+
"lstrip": false,
|
| 8 |
+
"normalized": false,
|
| 9 |
+
"rstrip": false,
|
| 10 |
+
"single_word": false,
|
| 11 |
+
"special": true
|
| 12 |
+
},
|
| 13 |
+
"1": {
|
| 14 |
+
"content": "<s>",
|
| 15 |
+
"lstrip": false,
|
| 16 |
+
"normalized": false,
|
| 17 |
+
"rstrip": false,
|
| 18 |
+
"single_word": false,
|
| 19 |
+
"special": true
|
| 20 |
+
},
|
| 21 |
+
"2": {
|
| 22 |
+
"content": "</s>",
|
| 23 |
+
"lstrip": false,
|
| 24 |
+
"normalized": false,
|
| 25 |
+
"rstrip": false,
|
| 26 |
+
"single_word": false,
|
| 27 |
+
"special": true
|
| 28 |
+
}
|
| 29 |
+
},
|
| 30 |
+
"additional_special_tokens": [],
|
| 31 |
+
"bos_token": "<s>",
|
| 32 |
+
"chat_template": "{% for message in messages %}\n{% if message['role'] == 'user' %}\n{{ '<|user|>\n' + message['content'] + eos_token }}\n{% elif message['role'] == 'system' %}\n{{ '<|system|>\n' + message['content'] + eos_token }}\n{% elif message['role'] == 'assistant' %}\n{{ '<|assistant|>\n' + message['content'] + eos_token }}\n{% endif %}\n{% if loop.last and add_generation_prompt %}\n{{ '<|assistant|>' }}\n{% endif %}\n{% endfor %}",
|
| 33 |
+
"clean_up_tokenization_spaces": false,
|
| 34 |
+
"eos_token": "</s>",
|
| 35 |
+
"legacy": true,
|
| 36 |
+
"model_max_length": 2048,
|
| 37 |
+
"pad_token": "</s>",
|
| 38 |
+
"sp_model_kwargs": {},
|
| 39 |
+
"spaces_between_special_tokens": false,
|
| 40 |
+
"tokenizer_class": "LlamaTokenizer",
|
| 41 |
+
"unk_token": "<unk>",
|
| 42 |
+
"use_default_system_prompt": false
|
| 43 |
+
}
|
train_results.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 19.80456026058632,
|
| 3 |
+
"total_flos": 4.2808751411995607e+18,
|
| 4 |
+
"train_loss": 0.32058246888612446,
|
| 5 |
+
"train_runtime": 79494.9932,
|
| 6 |
+
"train_samples": 14551,
|
| 7 |
+
"train_samples_per_second": 0.618,
|
| 8 |
+
"train_steps_per_second": 0.01
|
| 9 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,1254 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 19.80456026058632,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 760,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.13029315960912052,
|
| 13 |
+
"grad_norm": 0.91015625,
|
| 14 |
+
"learning_rate": 9.998932083939657e-06,
|
| 15 |
+
"loss": 0.7586,
|
| 16 |
+
"step": 5
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.26058631921824105,
|
| 20 |
+
"grad_norm": 0.80078125,
|
| 21 |
+
"learning_rate": 9.995728791936505e-06,
|
| 22 |
+
"loss": 0.7074,
|
| 23 |
+
"step": 10
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.39087947882736157,
|
| 27 |
+
"grad_norm": 0.67578125,
|
| 28 |
+
"learning_rate": 9.990391492329341e-06,
|
| 29 |
+
"loss": 0.6324,
|
| 30 |
+
"step": 15
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.5211726384364821,
|
| 34 |
+
"grad_norm": 0.5,
|
| 35 |
+
"learning_rate": 9.98292246503335e-06,
|
| 36 |
+
"loss": 0.5816,
|
| 37 |
+
"step": 20
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.6514657980456026,
|
| 41 |
+
"grad_norm": 0.361328125,
|
| 42 |
+
"learning_rate": 9.973324900566214e-06,
|
| 43 |
+
"loss": 0.5376,
|
| 44 |
+
"step": 25
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.7817589576547231,
|
| 48 |
+
"grad_norm": 0.2578125,
|
| 49 |
+
"learning_rate": 9.961602898685225e-06,
|
| 50 |
+
"loss": 0.5183,
|
| 51 |
+
"step": 30
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.9120521172638436,
|
| 55 |
+
"grad_norm": 0.1962890625,
|
| 56 |
+
"learning_rate": 9.947761466636014e-06,
|
| 57 |
+
"loss": 0.4906,
|
| 58 |
+
"step": 35
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.990228013029316,
|
| 62 |
+
"eval_loss": 0.4775831997394562,
|
| 63 |
+
"eval_runtime": 216.4722,
|
| 64 |
+
"eval_samples_per_second": 2.019,
|
| 65 |
+
"eval_steps_per_second": 0.508,
|
| 66 |
+
"step": 38
|
| 67 |
+
},
|
| 68 |
+
{
|
| 69 |
+
"epoch": 1.0423452768729642,
|
| 70 |
+
"grad_norm": 0.1796875,
|
| 71 |
+
"learning_rate": 9.931806517013612e-06,
|
| 72 |
+
"loss": 0.4858,
|
| 73 |
+
"step": 40
|
| 74 |
+
},
|
| 75 |
+
{
|
| 76 |
+
"epoch": 1.1726384364820848,
|
| 77 |
+
"grad_norm": 0.1669921875,
|
| 78 |
+
"learning_rate": 9.913744865236798e-06,
|
| 79 |
+
"loss": 0.464,
|
| 80 |
+
"step": 45
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 1.3029315960912053,
|
| 84 |
+
"grad_norm": 0.16015625,
|
| 85 |
+
"learning_rate": 9.893584226636773e-06,
|
| 86 |
+
"loss": 0.4706,
|
| 87 |
+
"step": 50
|
| 88 |
+
},
|
| 89 |
+
{
|
| 90 |
+
"epoch": 1.4332247557003257,
|
| 91 |
+
"grad_norm": 0.158203125,
|
| 92 |
+
"learning_rate": 9.871333213161438e-06,
|
| 93 |
+
"loss": 0.4464,
|
| 94 |
+
"step": 55
|
| 95 |
+
},
|
| 96 |
+
{
|
| 97 |
+
"epoch": 1.5635179153094463,
|
| 98 |
+
"grad_norm": 0.1611328125,
|
| 99 |
+
"learning_rate": 9.847001329696653e-06,
|
| 100 |
+
"loss": 0.4411,
|
| 101 |
+
"step": 60
|
| 102 |
+
},
|
| 103 |
+
{
|
| 104 |
+
"epoch": 1.6938110749185666,
|
| 105 |
+
"grad_norm": 0.1435546875,
|
| 106 |
+
"learning_rate": 9.820598970006068e-06,
|
| 107 |
+
"loss": 0.4344,
|
| 108 |
+
"step": 65
|
| 109 |
+
},
|
| 110 |
+
{
|
| 111 |
+
"epoch": 1.8241042345276872,
|
| 112 |
+
"grad_norm": 0.1396484375,
|
| 113 |
+
"learning_rate": 9.792137412291265e-06,
|
| 114 |
+
"loss": 0.4292,
|
| 115 |
+
"step": 70
|
| 116 |
+
},
|
| 117 |
+
{
|
| 118 |
+
"epoch": 1.9543973941368078,
|
| 119 |
+
"grad_norm": 0.1279296875,
|
| 120 |
+
"learning_rate": 9.761628814374074e-06,
|
| 121 |
+
"loss": 0.3966,
|
| 122 |
+
"step": 75
|
| 123 |
+
},
|
| 124 |
+
{
|
| 125 |
+
"epoch": 1.980456026058632,
|
| 126 |
+
"eval_loss": 0.41222962737083435,
|
| 127 |
+
"eval_runtime": 216.235,
|
| 128 |
+
"eval_samples_per_second": 2.021,
|
| 129 |
+
"eval_steps_per_second": 0.509,
|
| 130 |
+
"step": 76
|
| 131 |
+
},
|
| 132 |
+
{
|
| 133 |
+
"epoch": 2.0846905537459284,
|
| 134 |
+
"grad_norm": 0.126953125,
|
| 135 |
+
"learning_rate": 9.729086208503174e-06,
|
| 136 |
+
"loss": 0.4109,
|
| 137 |
+
"step": 80
|
| 138 |
+
},
|
| 139 |
+
{
|
| 140 |
+
"epoch": 2.214983713355049,
|
| 141 |
+
"grad_norm": 0.11181640625,
|
| 142 |
+
"learning_rate": 9.694523495787149e-06,
|
| 143 |
+
"loss": 0.3988,
|
| 144 |
+
"step": 85
|
| 145 |
+
},
|
| 146 |
+
{
|
| 147 |
+
"epoch": 2.3452768729641695,
|
| 148 |
+
"grad_norm": 0.1240234375,
|
| 149 |
+
"learning_rate": 9.657955440256396e-06,
|
| 150 |
+
"loss": 0.402,
|
| 151 |
+
"step": 90
|
| 152 |
+
},
|
| 153 |
+
{
|
| 154 |
+
"epoch": 2.47557003257329,
|
| 155 |
+
"grad_norm": 0.109375,
|
| 156 |
+
"learning_rate": 9.619397662556434e-06,
|
| 157 |
+
"loss": 0.3923,
|
| 158 |
+
"step": 95
|
| 159 |
+
},
|
| 160 |
+
{
|
| 161 |
+
"epoch": 2.6058631921824107,
|
| 162 |
+
"grad_norm": 0.107421875,
|
| 163 |
+
"learning_rate": 9.578866633275289e-06,
|
| 164 |
+
"loss": 0.383,
|
| 165 |
+
"step": 100
|
| 166 |
+
},
|
| 167 |
+
{
|
| 168 |
+
"epoch": 2.736156351791531,
|
| 169 |
+
"grad_norm": 0.11083984375,
|
| 170 |
+
"learning_rate": 9.536379665907801e-06,
|
| 171 |
+
"loss": 0.3702,
|
| 172 |
+
"step": 105
|
| 173 |
+
},
|
| 174 |
+
{
|
| 175 |
+
"epoch": 2.8664495114006514,
|
| 176 |
+
"grad_norm": 0.103515625,
|
| 177 |
+
"learning_rate": 9.491954909459895e-06,
|
| 178 |
+
"loss": 0.3757,
|
| 179 |
+
"step": 110
|
| 180 |
+
},
|
| 181 |
+
{
|
| 182 |
+
"epoch": 2.996742671009772,
|
| 183 |
+
"grad_norm": 0.1015625,
|
| 184 |
+
"learning_rate": 9.445611340695926e-06,
|
| 185 |
+
"loss": 0.3648,
|
| 186 |
+
"step": 115
|
| 187 |
+
},
|
| 188 |
+
{
|
| 189 |
+
"epoch": 2.996742671009772,
|
| 190 |
+
"eval_loss": 0.36584609746932983,
|
| 191 |
+
"eval_runtime": 216.4271,
|
| 192 |
+
"eval_samples_per_second": 2.019,
|
| 193 |
+
"eval_steps_per_second": 0.508,
|
| 194 |
+
"step": 115
|
| 195 |
+
},
|
| 196 |
+
{
|
| 197 |
+
"epoch": 3.1270358306188926,
|
| 198 |
+
"grad_norm": 0.09326171875,
|
| 199 |
+
"learning_rate": 9.397368756032445e-06,
|
| 200 |
+
"loss": 0.3651,
|
| 201 |
+
"step": 120
|
| 202 |
+
},
|
| 203 |
+
{
|
| 204 |
+
"epoch": 3.257328990228013,
|
| 205 |
+
"grad_norm": 0.10791015625,
|
| 206 |
+
"learning_rate": 9.347247763081834e-06,
|
| 207 |
+
"loss": 0.361,
|
| 208 |
+
"step": 125
|
| 209 |
+
},
|
| 210 |
+
{
|
| 211 |
+
"epoch": 3.3876221498371337,
|
| 212 |
+
"grad_norm": 0.08349609375,
|
| 213 |
+
"learning_rate": 9.295269771849426e-06,
|
| 214 |
+
"loss": 0.3513,
|
| 215 |
+
"step": 130
|
| 216 |
+
},
|
| 217 |
+
{
|
| 218 |
+
"epoch": 3.517915309446254,
|
| 219 |
+
"grad_norm": 0.1064453125,
|
| 220 |
+
"learning_rate": 9.241456985587868e-06,
|
| 221 |
+
"loss": 0.348,
|
| 222 |
+
"step": 135
|
| 223 |
+
},
|
| 224 |
+
{
|
| 225 |
+
"epoch": 3.6482084690553744,
|
| 226 |
+
"grad_norm": 0.083984375,
|
| 227 |
+
"learning_rate": 9.185832391312644e-06,
|
| 228 |
+
"loss": 0.345,
|
| 229 |
+
"step": 140
|
| 230 |
+
},
|
| 231 |
+
{
|
| 232 |
+
"epoch": 3.778501628664495,
|
| 233 |
+
"grad_norm": 0.076171875,
|
| 234 |
+
"learning_rate": 9.12841974998278e-06,
|
| 235 |
+
"loss": 0.3428,
|
| 236 |
+
"step": 145
|
| 237 |
+
},
|
| 238 |
+
{
|
| 239 |
+
"epoch": 3.9087947882736156,
|
| 240 |
+
"grad_norm": 0.076171875,
|
| 241 |
+
"learning_rate": 9.069243586350976e-06,
|
| 242 |
+
"loss": 0.3405,
|
| 243 |
+
"step": 150
|
| 244 |
+
},
|
| 245 |
+
{
|
| 246 |
+
"epoch": 3.986970684039088,
|
| 247 |
+
"eval_loss": 0.33722051978111267,
|
| 248 |
+
"eval_runtime": 216.4019,
|
| 249 |
+
"eval_samples_per_second": 2.019,
|
| 250 |
+
"eval_steps_per_second": 0.508,
|
| 251 |
+
"step": 153
|
| 252 |
+
},
|
| 253 |
+
{
|
| 254 |
+
"epoch": 4.039087947882736,
|
| 255 |
+
"grad_norm": 0.0703125,
|
| 256 |
+
"learning_rate": 9.008329178487442e-06,
|
| 257 |
+
"loss": 0.3302,
|
| 258 |
+
"step": 155
|
| 259 |
+
},
|
| 260 |
+
{
|
| 261 |
+
"epoch": 4.169381107491857,
|
| 262 |
+
"grad_norm": 0.06787109375,
|
| 263 |
+
"learning_rate": 8.94570254698197e-06,
|
| 264 |
+
"loss": 0.332,
|
| 265 |
+
"step": 160
|
| 266 |
+
},
|
| 267 |
+
{
|
| 268 |
+
"epoch": 4.299674267100977,
|
| 269 |
+
"grad_norm": 0.068359375,
|
| 270 |
+
"learning_rate": 8.881390443828788e-06,
|
| 271 |
+
"loss": 0.3214,
|
| 272 |
+
"step": 165
|
| 273 |
+
},
|
| 274 |
+
{
|
| 275 |
+
"epoch": 4.429967426710098,
|
| 276 |
+
"grad_norm": 0.06640625,
|
| 277 |
+
"learning_rate": 8.815420340999034e-06,
|
| 278 |
+
"loss": 0.3299,
|
| 279 |
+
"step": 170
|
| 280 |
+
},
|
| 281 |
+
{
|
| 282 |
+
"epoch": 4.5602605863192185,
|
| 283 |
+
"grad_norm": 0.0634765625,
|
| 284 |
+
"learning_rate": 8.747820418705632e-06,
|
| 285 |
+
"loss": 0.3304,
|
| 286 |
+
"step": 175
|
| 287 |
+
},
|
| 288 |
+
{
|
| 289 |
+
"epoch": 4.690553745928339,
|
| 290 |
+
"grad_norm": 0.08056640625,
|
| 291 |
+
"learning_rate": 8.67861955336566e-06,
|
| 292 |
+
"loss": 0.3249,
|
| 293 |
+
"step": 180
|
| 294 |
+
},
|
| 295 |
+
{
|
| 296 |
+
"epoch": 4.82084690553746,
|
| 297 |
+
"grad_norm": 0.060791015625,
|
| 298 |
+
"learning_rate": 8.607847305265312e-06,
|
| 299 |
+
"loss": 0.3199,
|
| 300 |
+
"step": 185
|
| 301 |
+
},
|
| 302 |
+
{
|
| 303 |
+
"epoch": 4.95114006514658,
|
| 304 |
+
"grad_norm": 0.059814453125,
|
| 305 |
+
"learning_rate": 8.535533905932739e-06,
|
| 306 |
+
"loss": 0.3153,
|
| 307 |
+
"step": 190
|
| 308 |
+
},
|
| 309 |
+
{
|
| 310 |
+
"epoch": 4.977198697068404,
|
| 311 |
+
"eval_loss": 0.3180455267429352,
|
| 312 |
+
"eval_runtime": 216.3986,
|
| 313 |
+
"eval_samples_per_second": 2.019,
|
| 314 |
+
"eval_steps_per_second": 0.508,
|
| 315 |
+
"step": 191
|
| 316 |
+
},
|
| 317 |
+
{
|
| 318 |
+
"epoch": 5.0814332247557,
|
| 319 |
+
"grad_norm": 0.06396484375,
|
| 320 |
+
"learning_rate": 8.461710245224149e-06,
|
| 321 |
+
"loss": 0.3223,
|
| 322 |
+
"step": 195
|
| 323 |
+
},
|
| 324 |
+
{
|
| 325 |
+
"epoch": 5.2117263843648205,
|
| 326 |
+
"grad_norm": 0.060546875,
|
| 327 |
+
"learning_rate": 8.386407858128707e-06,
|
| 328 |
+
"loss": 0.3127,
|
| 329 |
+
"step": 200
|
| 330 |
+
},
|
| 331 |
+
{
|
| 332 |
+
"epoch": 5.342019543973941,
|
| 333 |
+
"grad_norm": 0.060546875,
|
| 334 |
+
"learning_rate": 8.309658911297833e-06,
|
| 335 |
+
"loss": 0.3092,
|
| 336 |
+
"step": 205
|
| 337 |
+
},
|
| 338 |
+
{
|
| 339 |
+
"epoch": 5.472312703583062,
|
| 340 |
+
"grad_norm": 0.057861328125,
|
| 341 |
+
"learning_rate": 8.231496189304704e-06,
|
| 342 |
+
"loss": 0.3064,
|
| 343 |
+
"step": 210
|
| 344 |
+
},
|
| 345 |
+
{
|
| 346 |
+
"epoch": 5.602605863192182,
|
| 347 |
+
"grad_norm": 0.05908203125,
|
| 348 |
+
"learning_rate": 8.151953080639777e-06,
|
| 349 |
+
"loss": 0.3119,
|
| 350 |
+
"step": 215
|
| 351 |
+
},
|
| 352 |
+
{
|
| 353 |
+
"epoch": 5.732899022801303,
|
| 354 |
+
"grad_norm": 0.052734375,
|
| 355 |
+
"learning_rate": 8.071063563448341e-06,
|
| 356 |
+
"loss": 0.3064,
|
| 357 |
+
"step": 220
|
| 358 |
+
},
|
| 359 |
+
{
|
| 360 |
+
"epoch": 5.863192182410423,
|
| 361 |
+
"grad_norm": 0.0517578125,
|
| 362 |
+
"learning_rate": 7.988862191016204e-06,
|
| 363 |
+
"loss": 0.3035,
|
| 364 |
+
"step": 225
|
| 365 |
+
},
|
| 366 |
+
{
|
| 367 |
+
"epoch": 5.993485342019544,
|
| 368 |
+
"grad_norm": 0.06201171875,
|
| 369 |
+
"learning_rate": 7.905384077009693e-06,
|
| 370 |
+
"loss": 0.2996,
|
| 371 |
+
"step": 230
|
| 372 |
+
},
|
| 373 |
+
{
|
| 374 |
+
"epoch": 5.993485342019544,
|
| 375 |
+
"eval_loss": 0.30433574318885803,
|
| 376 |
+
"eval_runtime": 216.4993,
|
| 377 |
+
"eval_samples_per_second": 2.018,
|
| 378 |
+
"eval_steps_per_second": 0.508,
|
| 379 |
+
"step": 230
|
| 380 |
+
},
|
| 381 |
+
{
|
| 382 |
+
"epoch": 6.1237785016286646,
|
| 383 |
+
"grad_norm": 0.054931640625,
|
| 384 |
+
"learning_rate": 7.820664880476257e-06,
|
| 385 |
+
"loss": 0.2982,
|
| 386 |
+
"step": 235
|
| 387 |
+
},
|
| 388 |
+
{
|
| 389 |
+
"epoch": 6.254071661237785,
|
| 390 |
+
"grad_norm": 0.049072265625,
|
| 391 |
+
"learning_rate": 7.734740790612137e-06,
|
| 392 |
+
"loss": 0.3008,
|
| 393 |
+
"step": 240
|
| 394 |
+
},
|
| 395 |
+
{
|
| 396 |
+
"epoch": 6.384364820846906,
|
| 397 |
+
"grad_norm": 0.04931640625,
|
| 398 |
+
"learning_rate": 7.647648511303545e-06,
|
| 399 |
+
"loss": 0.3004,
|
| 400 |
+
"step": 245
|
| 401 |
+
},
|
| 402 |
+
{
|
| 403 |
+
"epoch": 6.514657980456026,
|
| 404 |
+
"grad_norm": 0.050537109375,
|
| 405 |
+
"learning_rate": 7.559425245448006e-06,
|
| 406 |
+
"loss": 0.2966,
|
| 407 |
+
"step": 250
|
| 408 |
+
},
|
| 409 |
+
{
|
| 410 |
+
"epoch": 6.644951140065147,
|
| 411 |
+
"grad_norm": 0.05078125,
|
| 412 |
+
"learning_rate": 7.470108679062521e-06,
|
| 413 |
+
"loss": 0.3018,
|
| 414 |
+
"step": 255
|
| 415 |
+
},
|
| 416 |
+
{
|
| 417 |
+
"epoch": 6.7752442996742674,
|
| 418 |
+
"grad_norm": 0.056396484375,
|
| 419 |
+
"learning_rate": 7.379736965185369e-06,
|
| 420 |
+
"loss": 0.2953,
|
| 421 |
+
"step": 260
|
| 422 |
+
},
|
| 423 |
+
{
|
| 424 |
+
"epoch": 6.905537459283387,
|
| 425 |
+
"grad_norm": 0.04833984375,
|
| 426 |
+
"learning_rate": 7.288348707578409e-06,
|
| 427 |
+
"loss": 0.2987,
|
| 428 |
+
"step": 265
|
| 429 |
+
},
|
| 430 |
+
{
|
| 431 |
+
"epoch": 6.9837133550488595,
|
| 432 |
+
"eval_loss": 0.2968701422214508,
|
| 433 |
+
"eval_runtime": 216.3327,
|
| 434 |
+
"eval_samples_per_second": 2.02,
|
| 435 |
+
"eval_steps_per_second": 0.508,
|
| 436 |
+
"step": 268
|
| 437 |
+
},
|
| 438 |
+
{
|
| 439 |
+
"epoch": 7.035830618892508,
|
| 440 |
+
"grad_norm": 0.048095703125,
|
| 441 |
+
"learning_rate": 7.195982944236853e-06,
|
| 442 |
+
"loss": 0.2953,
|
| 443 |
+
"step": 270
|
| 444 |
+
},
|
| 445 |
+
{
|
| 446 |
+
"epoch": 7.166123778501628,
|
| 447 |
+
"grad_norm": 0.05126953125,
|
| 448 |
+
"learning_rate": 7.102679130713538e-06,
|
| 449 |
+
"loss": 0.2936,
|
| 450 |
+
"step": 275
|
| 451 |
+
},
|
| 452 |
+
{
|
| 453 |
+
"epoch": 7.296416938110749,
|
| 454 |
+
"grad_norm": 0.05322265625,
|
| 455 |
+
"learning_rate": 7.008477123264849e-06,
|
| 456 |
+
"loss": 0.2888,
|
| 457 |
+
"step": 280
|
| 458 |
+
},
|
| 459 |
+
{
|
| 460 |
+
"epoch": 7.4267100977198695,
|
| 461 |
+
"grad_norm": 0.049072265625,
|
| 462 |
+
"learning_rate": 6.913417161825449e-06,
|
| 463 |
+
"loss": 0.2919,
|
| 464 |
+
"step": 285
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 7.55700325732899,
|
| 468 |
+
"grad_norm": 0.052490234375,
|
| 469 |
+
"learning_rate": 6.817539852819149e-06,
|
| 470 |
+
"loss": 0.2837,
|
| 471 |
+
"step": 290
|
| 472 |
+
},
|
| 473 |
+
{
|
| 474 |
+
"epoch": 7.687296416938111,
|
| 475 |
+
"grad_norm": 0.0458984375,
|
| 476 |
+
"learning_rate": 6.720886151813194e-06,
|
| 477 |
+
"loss": 0.2892,
|
| 478 |
+
"step": 295
|
| 479 |
+
},
|
| 480 |
+
{
|
| 481 |
+
"epoch": 7.817589576547231,
|
| 482 |
+
"grad_norm": 0.0478515625,
|
| 483 |
+
"learning_rate": 6.6234973460234184e-06,
|
| 484 |
+
"loss": 0.3036,
|
| 485 |
+
"step": 300
|
| 486 |
+
},
|
| 487 |
+
{
|
| 488 |
+
"epoch": 7.947882736156352,
|
| 489 |
+
"grad_norm": 0.0595703125,
|
| 490 |
+
"learning_rate": 6.525415036677745e-06,
|
| 491 |
+
"loss": 0.2888,
|
| 492 |
+
"step": 305
|
| 493 |
+
},
|
| 494 |
+
{
|
| 495 |
+
"epoch": 8.0,
|
| 496 |
+
"eval_loss": 0.2923305630683899,
|
| 497 |
+
"eval_runtime": 216.4767,
|
| 498 |
+
"eval_samples_per_second": 2.019,
|
| 499 |
+
"eval_steps_per_second": 0.508,
|
| 500 |
+
"step": 307
|
| 501 |
+
},
|
| 502 |
+
{
|
| 503 |
+
"epoch": 8.078175895765472,
|
| 504 |
+
"grad_norm": 0.04248046875,
|
| 505 |
+
"learning_rate": 6.426681121245527e-06,
|
| 506 |
+
"loss": 0.2909,
|
| 507 |
+
"step": 310
|
| 508 |
+
},
|
| 509 |
+
{
|
| 510 |
+
"epoch": 8.208469055374593,
|
| 511 |
+
"grad_norm": 0.04443359375,
|
| 512 |
+
"learning_rate": 6.327337775540362e-06,
|
| 513 |
+
"loss": 0.2893,
|
| 514 |
+
"step": 315
|
| 515 |
+
},
|
| 516 |
+
{
|
| 517 |
+
"epoch": 8.338762214983714,
|
| 518 |
+
"grad_norm": 0.042236328125,
|
| 519 |
+
"learning_rate": 6.227427435703997e-06,
|
| 520 |
+
"loss": 0.2981,
|
| 521 |
+
"step": 320
|
| 522 |
+
},
|
| 523 |
+
{
|
| 524 |
+
"epoch": 8.469055374592834,
|
| 525 |
+
"grad_norm": 0.046142578125,
|
| 526 |
+
"learning_rate": 6.126992780079032e-06,
|
| 527 |
+
"loss": 0.287,
|
| 528 |
+
"step": 325
|
| 529 |
+
},
|
| 530 |
+
{
|
| 531 |
+
"epoch": 8.599348534201955,
|
| 532 |
+
"grad_norm": 0.042236328125,
|
| 533 |
+
"learning_rate": 6.026076710978172e-06,
|
| 534 |
+
"loss": 0.2904,
|
| 535 |
+
"step": 330
|
| 536 |
+
},
|
| 537 |
+
{
|
| 538 |
+
"epoch": 8.729641693811075,
|
| 539 |
+
"grad_norm": 0.041748046875,
|
| 540 |
+
"learning_rate": 5.924722336357793e-06,
|
| 541 |
+
"loss": 0.2874,
|
| 542 |
+
"step": 335
|
| 543 |
+
},
|
| 544 |
+
{
|
| 545 |
+
"epoch": 8.859934853420196,
|
| 546 |
+
"grad_norm": 0.04443359375,
|
| 547 |
+
"learning_rate": 5.82297295140367e-06,
|
| 548 |
+
"loss": 0.2799,
|
| 549 |
+
"step": 340
|
| 550 |
+
},
|
| 551 |
+
{
|
| 552 |
+
"epoch": 8.990228013029316,
|
| 553 |
+
"grad_norm": 0.04345703125,
|
| 554 |
+
"learning_rate": 5.720872020036734e-06,
|
| 555 |
+
"loss": 0.2899,
|
| 556 |
+
"step": 345
|
| 557 |
+
},
|
| 558 |
+
{
|
| 559 |
+
"epoch": 8.990228013029316,
|
| 560 |
+
"eval_loss": 0.2898467779159546,
|
| 561 |
+
"eval_runtime": 216.4655,
|
| 562 |
+
"eval_samples_per_second": 2.019,
|
| 563 |
+
"eval_steps_per_second": 0.508,
|
| 564 |
+
"step": 345
|
| 565 |
+
},
|
| 566 |
+
{
|
| 567 |
+
"epoch": 9.120521172638437,
|
| 568 |
+
"grad_norm": 0.044189453125,
|
| 569 |
+
"learning_rate": 5.61846315634674e-06,
|
| 570 |
+
"loss": 0.2843,
|
| 571 |
+
"step": 350
|
| 572 |
+
},
|
| 573 |
+
{
|
| 574 |
+
"epoch": 9.250814332247558,
|
| 575 |
+
"grad_norm": 0.0419921875,
|
| 576 |
+
"learning_rate": 5.515790105961785e-06,
|
| 577 |
+
"loss": 0.2886,
|
| 578 |
+
"step": 355
|
| 579 |
+
},
|
| 580 |
+
{
|
| 581 |
+
"epoch": 9.381107491856678,
|
| 582 |
+
"grad_norm": 0.044677734375,
|
| 583 |
+
"learning_rate": 5.412896727361663e-06,
|
| 584 |
+
"loss": 0.2898,
|
| 585 |
+
"step": 360
|
| 586 |
+
},
|
| 587 |
+
{
|
| 588 |
+
"epoch": 9.511400651465799,
|
| 589 |
+
"grad_norm": 0.044677734375,
|
| 590 |
+
"learning_rate": 5.309826973142974e-06,
|
| 591 |
+
"loss": 0.2805,
|
| 592 |
+
"step": 365
|
| 593 |
+
},
|
| 594 |
+
{
|
| 595 |
+
"epoch": 9.64169381107492,
|
| 596 |
+
"grad_norm": 0.051513671875,
|
| 597 |
+
"learning_rate": 5.206624871244066e-06,
|
| 598 |
+
"loss": 0.2907,
|
| 599 |
+
"step": 370
|
| 600 |
+
},
|
| 601 |
+
{
|
| 602 |
+
"epoch": 9.77198697068404,
|
| 603 |
+
"grad_norm": 0.051513671875,
|
| 604 |
+
"learning_rate": 5.103334506137773e-06,
|
| 605 |
+
"loss": 0.2812,
|
| 606 |
+
"step": 375
|
| 607 |
+
},
|
| 608 |
+
{
|
| 609 |
+
"epoch": 9.90228013029316,
|
| 610 |
+
"grad_norm": 0.045654296875,
|
| 611 |
+
"learning_rate": 5e-06,
|
| 612 |
+
"loss": 0.2873,
|
| 613 |
+
"step": 380
|
| 614 |
+
},
|
| 615 |
+
{
|
| 616 |
+
"epoch": 9.980456026058633,
|
| 617 |
+
"eval_loss": 0.28830844163894653,
|
| 618 |
+
"eval_runtime": 216.3818,
|
| 619 |
+
"eval_samples_per_second": 2.02,
|
| 620 |
+
"eval_steps_per_second": 0.508,
|
| 621 |
+
"step": 383
|
| 622 |
+
},
|
| 623 |
+
{
|
| 624 |
+
"epoch": 10.03257328990228,
|
| 625 |
+
"grad_norm": 0.0400390625,
|
| 626 |
+
"learning_rate": 4.89666549386223e-06,
|
| 627 |
+
"loss": 0.2907,
|
| 628 |
+
"step": 385
|
| 629 |
+
},
|
| 630 |
+
{
|
| 631 |
+
"epoch": 10.1628664495114,
|
| 632 |
+
"grad_norm": 0.04345703125,
|
| 633 |
+
"learning_rate": 4.793375128755934e-06,
|
| 634 |
+
"loss": 0.2848,
|
| 635 |
+
"step": 390
|
| 636 |
+
},
|
| 637 |
+
{
|
| 638 |
+
"epoch": 10.29315960912052,
|
| 639 |
+
"grad_norm": 0.04345703125,
|
| 640 |
+
"learning_rate": 4.690173026857028e-06,
|
| 641 |
+
"loss": 0.2907,
|
| 642 |
+
"step": 395
|
| 643 |
+
},
|
| 644 |
+
{
|
| 645 |
+
"epoch": 10.423452768729641,
|
| 646 |
+
"grad_norm": 0.041259765625,
|
| 647 |
+
"learning_rate": 4.587103272638339e-06,
|
| 648 |
+
"loss": 0.2834,
|
| 649 |
+
"step": 400
|
| 650 |
+
},
|
| 651 |
+
{
|
| 652 |
+
"epoch": 10.553745928338762,
|
| 653 |
+
"grad_norm": 0.043701171875,
|
| 654 |
+
"learning_rate": 4.4842098940382155e-06,
|
| 655 |
+
"loss": 0.2862,
|
| 656 |
+
"step": 405
|
| 657 |
+
},
|
| 658 |
+
{
|
| 659 |
+
"epoch": 10.684039087947882,
|
| 660 |
+
"grad_norm": 0.04638671875,
|
| 661 |
+
"learning_rate": 4.381536843653262e-06,
|
| 662 |
+
"loss": 0.2832,
|
| 663 |
+
"step": 410
|
| 664 |
+
},
|
| 665 |
+
{
|
| 666 |
+
"epoch": 10.814332247557003,
|
| 667 |
+
"grad_norm": 0.0439453125,
|
| 668 |
+
"learning_rate": 4.279127979963266e-06,
|
| 669 |
+
"loss": 0.2861,
|
| 670 |
+
"step": 415
|
| 671 |
+
},
|
| 672 |
+
{
|
| 673 |
+
"epoch": 10.944625407166123,
|
| 674 |
+
"grad_norm": 0.040283203125,
|
| 675 |
+
"learning_rate": 4.17702704859633e-06,
|
| 676 |
+
"loss": 0.2831,
|
| 677 |
+
"step": 420
|
| 678 |
+
},
|
| 679 |
+
{
|
| 680 |
+
"epoch": 10.996742671009772,
|
| 681 |
+
"eval_loss": 0.2872494161128998,
|
| 682 |
+
"eval_runtime": 216.4293,
|
| 683 |
+
"eval_samples_per_second": 2.019,
|
| 684 |
+
"eval_steps_per_second": 0.508,
|
| 685 |
+
"step": 422
|
| 686 |
+
},
|
| 687 |
+
{
|
| 688 |
+
"epoch": 11.074918566775244,
|
| 689 |
+
"grad_norm": 0.03857421875,
|
| 690 |
+
"learning_rate": 4.075277663642208e-06,
|
| 691 |
+
"loss": 0.2913,
|
| 692 |
+
"step": 425
|
| 693 |
+
},
|
| 694 |
+
{
|
| 695 |
+
"epoch": 11.205211726384364,
|
| 696 |
+
"grad_norm": 0.0439453125,
|
| 697 |
+
"learning_rate": 3.973923289021829e-06,
|
| 698 |
+
"loss": 0.2877,
|
| 699 |
+
"step": 430
|
| 700 |
+
},
|
| 701 |
+
{
|
| 702 |
+
"epoch": 11.335504885993485,
|
| 703 |
+
"grad_norm": 0.05029296875,
|
| 704 |
+
"learning_rate": 3.8730072199209705e-06,
|
| 705 |
+
"loss": 0.2881,
|
| 706 |
+
"step": 435
|
| 707 |
+
},
|
| 708 |
+
{
|
| 709 |
+
"epoch": 11.465798045602606,
|
| 710 |
+
"grad_norm": 0.043212890625,
|
| 711 |
+
"learning_rate": 3.7725725642960047e-06,
|
| 712 |
+
"loss": 0.2843,
|
| 713 |
+
"step": 440
|
| 714 |
+
},
|
| 715 |
+
{
|
| 716 |
+
"epoch": 11.596091205211726,
|
| 717 |
+
"grad_norm": 0.0439453125,
|
| 718 |
+
"learning_rate": 3.67266222445964e-06,
|
| 719 |
+
"loss": 0.2822,
|
| 720 |
+
"step": 445
|
| 721 |
+
},
|
| 722 |
+
{
|
| 723 |
+
"epoch": 11.726384364820847,
|
| 724 |
+
"grad_norm": 0.040771484375,
|
| 725 |
+
"learning_rate": 3.573318878754475e-06,
|
| 726 |
+
"loss": 0.2772,
|
| 727 |
+
"step": 450
|
| 728 |
+
},
|
| 729 |
+
{
|
| 730 |
+
"epoch": 11.856677524429967,
|
| 731 |
+
"grad_norm": 0.04052734375,
|
| 732 |
+
"learning_rate": 3.4745849633222566e-06,
|
| 733 |
+
"loss": 0.2857,
|
| 734 |
+
"step": 455
|
| 735 |
+
},
|
| 736 |
+
{
|
| 737 |
+
"epoch": 11.986970684039088,
|
| 738 |
+
"grad_norm": 0.042724609375,
|
| 739 |
+
"learning_rate": 3.3765026539765832e-06,
|
| 740 |
+
"loss": 0.2773,
|
| 741 |
+
"step": 460
|
| 742 |
+
},
|
| 743 |
+
{
|
| 744 |
+
"epoch": 11.986970684039088,
|
| 745 |
+
"eval_loss": 0.286603182554245,
|
| 746 |
+
"eval_runtime": 216.2196,
|
| 747 |
+
"eval_samples_per_second": 2.021,
|
| 748 |
+
"eval_steps_per_second": 0.509,
|
| 749 |
+
"step": 460
|
| 750 |
+
},
|
| 751 |
+
{
|
| 752 |
+
"epoch": 12.117263843648209,
|
| 753 |
+
"grad_norm": 0.039794921875,
|
| 754 |
+
"learning_rate": 3.2791138481868084e-06,
|
| 755 |
+
"loss": 0.2873,
|
| 756 |
+
"step": 465
|
| 757 |
+
},
|
| 758 |
+
{
|
| 759 |
+
"epoch": 12.247557003257329,
|
| 760 |
+
"grad_norm": 0.04150390625,
|
| 761 |
+
"learning_rate": 3.1824601471808504e-06,
|
| 762 |
+
"loss": 0.2835,
|
| 763 |
+
"step": 470
|
| 764 |
+
},
|
| 765 |
+
{
|
| 766 |
+
"epoch": 12.37785016286645,
|
| 767 |
+
"grad_norm": 0.04541015625,
|
| 768 |
+
"learning_rate": 3.0865828381745515e-06,
|
| 769 |
+
"loss": 0.2849,
|
| 770 |
+
"step": 475
|
| 771 |
+
},
|
| 772 |
+
{
|
| 773 |
+
"epoch": 12.50814332247557,
|
| 774 |
+
"grad_norm": 0.04150390625,
|
| 775 |
+
"learning_rate": 2.991522876735154e-06,
|
| 776 |
+
"loss": 0.2899,
|
| 777 |
+
"step": 480
|
| 778 |
+
},
|
| 779 |
+
{
|
| 780 |
+
"epoch": 12.63843648208469,
|
| 781 |
+
"grad_norm": 0.040283203125,
|
| 782 |
+
"learning_rate": 2.8973208692864623e-06,
|
| 783 |
+
"loss": 0.2804,
|
| 784 |
+
"step": 485
|
| 785 |
+
},
|
| 786 |
+
{
|
| 787 |
+
"epoch": 12.768729641693811,
|
| 788 |
+
"grad_norm": 0.03955078125,
|
| 789 |
+
"learning_rate": 2.804017055763149e-06,
|
| 790 |
+
"loss": 0.2792,
|
| 791 |
+
"step": 490
|
| 792 |
+
},
|
| 793 |
+
{
|
| 794 |
+
"epoch": 12.899022801302932,
|
| 795 |
+
"grad_norm": 0.039794921875,
|
| 796 |
+
"learning_rate": 2.711651292421593e-06,
|
| 797 |
+
"loss": 0.2814,
|
| 798 |
+
"step": 495
|
| 799 |
+
},
|
| 800 |
+
{
|
| 801 |
+
"epoch": 12.977198697068404,
|
| 802 |
+
"eval_loss": 0.28619444370269775,
|
| 803 |
+
"eval_runtime": 216.4471,
|
| 804 |
+
"eval_samples_per_second": 2.019,
|
| 805 |
+
"eval_steps_per_second": 0.508,
|
| 806 |
+
"step": 498
|
| 807 |
+
},
|
| 808 |
+
{
|
| 809 |
+
"epoch": 13.029315960912053,
|
| 810 |
+
"grad_norm": 0.046142578125,
|
| 811 |
+
"learning_rate": 2.6202630348146323e-06,
|
| 812 |
+
"loss": 0.283,
|
| 813 |
+
"step": 500
|
| 814 |
+
},
|
| 815 |
+
{
|
| 816 |
+
"epoch": 13.159609120521173,
|
| 817 |
+
"grad_norm": 0.043212890625,
|
| 818 |
+
"learning_rate": 2.529891320937481e-06,
|
| 819 |
+
"loss": 0.2825,
|
| 820 |
+
"step": 505
|
| 821 |
+
},
|
| 822 |
+
{
|
| 823 |
+
"epoch": 13.289902280130294,
|
| 824 |
+
"grad_norm": 0.0390625,
|
| 825 |
+
"learning_rate": 2.4405747545519966e-06,
|
| 826 |
+
"loss": 0.2833,
|
| 827 |
+
"step": 510
|
| 828 |
+
},
|
| 829 |
+
{
|
| 830 |
+
"epoch": 13.420195439739414,
|
| 831 |
+
"grad_norm": 0.038818359375,
|
| 832 |
+
"learning_rate": 2.352351488696457e-06,
|
| 833 |
+
"loss": 0.2798,
|
| 834 |
+
"step": 515
|
| 835 |
+
},
|
| 836 |
+
{
|
| 837 |
+
"epoch": 13.550488599348535,
|
| 838 |
+
"grad_norm": 0.041259765625,
|
| 839 |
+
"learning_rate": 2.265259209387867e-06,
|
| 840 |
+
"loss": 0.2912,
|
| 841 |
+
"step": 520
|
| 842 |
+
},
|
| 843 |
+
{
|
| 844 |
+
"epoch": 13.680781758957655,
|
| 845 |
+
"grad_norm": 0.042236328125,
|
| 846 |
+
"learning_rate": 2.179335119523745e-06,
|
| 847 |
+
"loss": 0.2882,
|
| 848 |
+
"step": 525
|
| 849 |
+
},
|
| 850 |
+
{
|
| 851 |
+
"epoch": 13.811074918566776,
|
| 852 |
+
"grad_norm": 0.040283203125,
|
| 853 |
+
"learning_rate": 2.094615922990309e-06,
|
| 854 |
+
"loss": 0.2819,
|
| 855 |
+
"step": 530
|
| 856 |
+
},
|
| 857 |
+
{
|
| 858 |
+
"epoch": 13.941368078175895,
|
| 859 |
+
"grad_norm": 0.040283203125,
|
| 860 |
+
"learning_rate": 2.0111378089837958e-06,
|
| 861 |
+
"loss": 0.2781,
|
| 862 |
+
"step": 535
|
| 863 |
+
},
|
| 864 |
+
{
|
| 865 |
+
"epoch": 13.993485342019543,
|
| 866 |
+
"eval_loss": 0.2859518229961395,
|
| 867 |
+
"eval_runtime": 216.8194,
|
| 868 |
+
"eval_samples_per_second": 2.016,
|
| 869 |
+
"eval_steps_per_second": 0.507,
|
| 870 |
+
"step": 537
|
| 871 |
+
},
|
| 872 |
+
{
|
| 873 |
+
"epoch": 14.071661237785015,
|
| 874 |
+
"grad_norm": 0.0439453125,
|
| 875 |
+
"learning_rate": 1.928936436551661e-06,
|
| 876 |
+
"loss": 0.2838,
|
| 877 |
+
"step": 540
|
| 878 |
+
},
|
| 879 |
+
{
|
| 880 |
+
"epoch": 14.201954397394136,
|
| 881 |
+
"grad_norm": 0.039306640625,
|
| 882 |
+
"learning_rate": 1.848046919360225e-06,
|
| 883 |
+
"loss": 0.2839,
|
| 884 |
+
"step": 545
|
| 885 |
+
},
|
| 886 |
+
{
|
| 887 |
+
"epoch": 14.332247557003257,
|
| 888 |
+
"grad_norm": 0.041748046875,
|
| 889 |
+
"learning_rate": 1.7685038106952952e-06,
|
| 890 |
+
"loss": 0.279,
|
| 891 |
+
"step": 550
|
| 892 |
+
},
|
| 893 |
+
{
|
| 894 |
+
"epoch": 14.462540716612377,
|
| 895 |
+
"grad_norm": 0.040771484375,
|
| 896 |
+
"learning_rate": 1.6903410887021676e-06,
|
| 897 |
+
"loss": 0.2772,
|
| 898 |
+
"step": 555
|
| 899 |
+
},
|
| 900 |
+
{
|
| 901 |
+
"epoch": 14.592833876221498,
|
| 902 |
+
"grad_norm": 0.05224609375,
|
| 903 |
+
"learning_rate": 1.6135921418712959e-06,
|
| 904 |
+
"loss": 0.2817,
|
| 905 |
+
"step": 560
|
| 906 |
+
},
|
| 907 |
+
{
|
| 908 |
+
"epoch": 14.723127035830618,
|
| 909 |
+
"grad_norm": 0.039306640625,
|
| 910 |
+
"learning_rate": 1.5382897547758513e-06,
|
| 911 |
+
"loss": 0.2932,
|
| 912 |
+
"step": 565
|
| 913 |
+
},
|
| 914 |
+
{
|
| 915 |
+
"epoch": 14.853420195439739,
|
| 916 |
+
"grad_norm": 0.04150390625,
|
| 917 |
+
"learning_rate": 1.4644660940672628e-06,
|
| 918 |
+
"loss": 0.2834,
|
| 919 |
+
"step": 570
|
| 920 |
+
},
|
| 921 |
+
{
|
| 922 |
+
"epoch": 14.98371335504886,
|
| 923 |
+
"grad_norm": 0.038330078125,
|
| 924 |
+
"learning_rate": 1.3921526947346902e-06,
|
| 925 |
+
"loss": 0.2845,
|
| 926 |
+
"step": 575
|
| 927 |
+
},
|
| 928 |
+
{
|
| 929 |
+
"epoch": 14.98371335504886,
|
| 930 |
+
"eval_loss": 0.2858189642429352,
|
| 931 |
+
"eval_runtime": 216.4523,
|
| 932 |
+
"eval_samples_per_second": 2.019,
|
| 933 |
+
"eval_steps_per_second": 0.508,
|
| 934 |
+
"step": 575
|
| 935 |
+
},
|
| 936 |
+
{
|
| 937 |
+
"epoch": 15.11400651465798,
|
| 938 |
+
"grad_norm": 0.040771484375,
|
| 939 |
+
"learning_rate": 1.321380446634342e-06,
|
| 940 |
+
"loss": 0.2778,
|
| 941 |
+
"step": 580
|
| 942 |
+
},
|
| 943 |
+
{
|
| 944 |
+
"epoch": 15.2442996742671,
|
| 945 |
+
"grad_norm": 0.039306640625,
|
| 946 |
+
"learning_rate": 1.2521795812943704e-06,
|
| 947 |
+
"loss": 0.2812,
|
| 948 |
+
"step": 585
|
| 949 |
+
},
|
| 950 |
+
{
|
| 951 |
+
"epoch": 15.374592833876221,
|
| 952 |
+
"grad_norm": 0.03759765625,
|
| 953 |
+
"learning_rate": 1.1845796590009684e-06,
|
| 954 |
+
"loss": 0.2834,
|
| 955 |
+
"step": 590
|
| 956 |
+
},
|
| 957 |
+
{
|
| 958 |
+
"epoch": 15.504885993485342,
|
| 959 |
+
"grad_norm": 0.04150390625,
|
| 960 |
+
"learning_rate": 1.118609556171213e-06,
|
| 961 |
+
"loss": 0.2852,
|
| 962 |
+
"step": 595
|
| 963 |
+
},
|
| 964 |
+
{
|
| 965 |
+
"epoch": 15.635179153094462,
|
| 966 |
+
"grad_norm": 0.04052734375,
|
| 967 |
+
"learning_rate": 1.0542974530180327e-06,
|
| 968 |
+
"loss": 0.2869,
|
| 969 |
+
"step": 600
|
| 970 |
+
},
|
| 971 |
+
{
|
| 972 |
+
"epoch": 15.765472312703583,
|
| 973 |
+
"grad_norm": 0.048095703125,
|
| 974 |
+
"learning_rate": 9.916708215125586e-07,
|
| 975 |
+
"loss": 0.2745,
|
| 976 |
+
"step": 605
|
| 977 |
+
},
|
| 978 |
+
{
|
| 979 |
+
"epoch": 15.895765472312704,
|
| 980 |
+
"grad_norm": 0.041748046875,
|
| 981 |
+
"learning_rate": 9.307564136490255e-07,
|
| 982 |
+
"loss": 0.29,
|
| 983 |
+
"step": 610
|
| 984 |
+
},
|
| 985 |
+
{
|
| 986 |
+
"epoch": 16.0,
|
| 987 |
+
"eval_loss": 0.28575843572616577,
|
| 988 |
+
"eval_runtime": 216.8023,
|
| 989 |
+
"eval_samples_per_second": 2.016,
|
| 990 |
+
"eval_steps_per_second": 0.507,
|
| 991 |
+
"step": 614
|
| 992 |
+
},
|
| 993 |
+
{
|
| 994 |
+
"epoch": 16.026058631921824,
|
| 995 |
+
"grad_norm": 0.043212890625,
|
| 996 |
+
"learning_rate": 8.715802500172215e-07,
|
| 997 |
+
"loss": 0.2866,
|
| 998 |
+
"step": 615
|
| 999 |
+
},
|
| 1000 |
+
{
|
| 1001 |
+
"epoch": 16.156351791530945,
|
| 1002 |
+
"grad_norm": 0.03955078125,
|
| 1003 |
+
"learning_rate": 8.141676086873574e-07,
|
| 1004 |
+
"loss": 0.2881,
|
| 1005 |
+
"step": 620
|
| 1006 |
+
},
|
| 1007 |
+
{
|
| 1008 |
+
"epoch": 16.286644951140065,
|
| 1009 |
+
"grad_norm": 0.048828125,
|
| 1010 |
+
"learning_rate": 7.585430144121319e-07,
|
| 1011 |
+
"loss": 0.2846,
|
| 1012 |
+
"step": 625
|
| 1013 |
+
},
|
| 1014 |
+
{
|
| 1015 |
+
"epoch": 16.416938110749186,
|
| 1016 |
+
"grad_norm": 0.04150390625,
|
| 1017 |
+
"learning_rate": 7.047302281505735e-07,
|
| 1018 |
+
"loss": 0.2874,
|
| 1019 |
+
"step": 630
|
| 1020 |
+
},
|
| 1021 |
+
{
|
| 1022 |
+
"epoch": 16.547231270358306,
|
| 1023 |
+
"grad_norm": 0.04296875,
|
| 1024 |
+
"learning_rate": 6.527522369181655e-07,
|
| 1025 |
+
"loss": 0.2818,
|
| 1026 |
+
"step": 635
|
| 1027 |
+
},
|
| 1028 |
+
{
|
| 1029 |
+
"epoch": 16.677524429967427,
|
| 1030 |
+
"grad_norm": 0.048828125,
|
| 1031 |
+
"learning_rate": 6.026312439675553e-07,
|
| 1032 |
+
"loss": 0.2809,
|
| 1033 |
+
"step": 640
|
| 1034 |
+
},
|
| 1035 |
+
{
|
| 1036 |
+
"epoch": 16.807817589576548,
|
| 1037 |
+
"grad_norm": 0.041015625,
|
| 1038 |
+
"learning_rate": 5.543886593040737e-07,
|
| 1039 |
+
"loss": 0.2831,
|
| 1040 |
+
"step": 645
|
| 1041 |
+
},
|
| 1042 |
+
{
|
| 1043 |
+
"epoch": 16.938110749185668,
|
| 1044 |
+
"grad_norm": 0.04052734375,
|
| 1045 |
+
"learning_rate": 5.080450905401057e-07,
|
| 1046 |
+
"loss": 0.2799,
|
| 1047 |
+
"step": 650
|
| 1048 |
+
},
|
| 1049 |
+
{
|
| 1050 |
+
"epoch": 16.990228013029316,
|
| 1051 |
+
"eval_loss": 0.28573134541511536,
|
| 1052 |
+
"eval_runtime": 216.4067,
|
| 1053 |
+
"eval_samples_per_second": 2.019,
|
| 1054 |
+
"eval_steps_per_second": 0.508,
|
| 1055 |
+
"step": 652
|
| 1056 |
+
},
|
| 1057 |
+
{
|
| 1058 |
+
"epoch": 17.06840390879479,
|
| 1059 |
+
"grad_norm": 0.0439453125,
|
| 1060 |
+
"learning_rate": 4.6362033409220077e-07,
|
| 1061 |
+
"loss": 0.2828,
|
| 1062 |
+
"step": 655
|
| 1063 |
+
},
|
| 1064 |
+
{
|
| 1065 |
+
"epoch": 17.19869706840391,
|
| 1066 |
+
"grad_norm": 0.047607421875,
|
| 1067 |
+
"learning_rate": 4.211333667247125e-07,
|
| 1068 |
+
"loss": 0.2811,
|
| 1069 |
+
"step": 660
|
| 1070 |
+
},
|
| 1071 |
+
{
|
| 1072 |
+
"epoch": 17.32899022801303,
|
| 1073 |
+
"grad_norm": 0.0478515625,
|
| 1074 |
+
"learning_rate": 3.8060233744356634e-07,
|
| 1075 |
+
"loss": 0.2881,
|
| 1076 |
+
"step": 665
|
| 1077 |
+
},
|
| 1078 |
+
{
|
| 1079 |
+
"epoch": 17.45928338762215,
|
| 1080 |
+
"grad_norm": 0.051025390625,
|
| 1081 |
+
"learning_rate": 3.420445597436056e-07,
|
| 1082 |
+
"loss": 0.2926,
|
| 1083 |
+
"step": 670
|
| 1084 |
+
},
|
| 1085 |
+
{
|
| 1086 |
+
"epoch": 17.58957654723127,
|
| 1087 |
+
"grad_norm": 0.040771484375,
|
| 1088 |
+
"learning_rate": 3.0547650421285216e-07,
|
| 1089 |
+
"loss": 0.2867,
|
| 1090 |
+
"step": 675
|
| 1091 |
+
},
|
| 1092 |
+
{
|
| 1093 |
+
"epoch": 17.71986970684039,
|
| 1094 |
+
"grad_norm": 0.052001953125,
|
| 1095 |
+
"learning_rate": 2.7091379149682683e-07,
|
| 1096 |
+
"loss": 0.2752,
|
| 1097 |
+
"step": 680
|
| 1098 |
+
},
|
| 1099 |
+
{
|
| 1100 |
+
"epoch": 17.850162866449512,
|
| 1101 |
+
"grad_norm": 0.04150390625,
|
| 1102 |
+
"learning_rate": 2.3837118562592799e-07,
|
| 1103 |
+
"loss": 0.2779,
|
| 1104 |
+
"step": 685
|
| 1105 |
+
},
|
| 1106 |
+
{
|
| 1107 |
+
"epoch": 17.980456026058633,
|
| 1108 |
+
"grad_norm": 0.044921875,
|
| 1109 |
+
"learning_rate": 2.0786258770873647e-07,
|
| 1110 |
+
"loss": 0.2825,
|
| 1111 |
+
"step": 690
|
| 1112 |
+
},
|
| 1113 |
+
{
|
| 1114 |
+
"epoch": 17.980456026058633,
|
| 1115 |
+
"eval_loss": 0.28572362661361694,
|
| 1116 |
+
"eval_runtime": 216.5181,
|
| 1117 |
+
"eval_samples_per_second": 2.018,
|
| 1118 |
+
"eval_steps_per_second": 0.508,
|
| 1119 |
+
"step": 690
|
| 1120 |
+
},
|
| 1121 |
+
{
|
| 1122 |
+
"epoch": 18.110749185667753,
|
| 1123 |
+
"grad_norm": 0.044921875,
|
| 1124 |
+
"learning_rate": 1.7940102999393194e-07,
|
| 1125 |
+
"loss": 0.2817,
|
| 1126 |
+
"step": 695
|
| 1127 |
+
},
|
| 1128 |
+
{
|
| 1129 |
+
"epoch": 18.241042345276874,
|
| 1130 |
+
"grad_norm": 0.041259765625,
|
| 1131 |
+
"learning_rate": 1.5299867030334815e-07,
|
| 1132 |
+
"loss": 0.2827,
|
| 1133 |
+
"step": 700
|
| 1134 |
+
},
|
| 1135 |
+
{
|
| 1136 |
+
"epoch": 18.371335504885995,
|
| 1137 |
+
"grad_norm": 0.0439453125,
|
| 1138 |
+
"learning_rate": 1.286667868385627e-07,
|
| 1139 |
+
"loss": 0.2885,
|
| 1140 |
+
"step": 705
|
| 1141 |
+
},
|
| 1142 |
+
{
|
| 1143 |
+
"epoch": 18.501628664495115,
|
| 1144 |
+
"grad_norm": 0.05517578125,
|
| 1145 |
+
"learning_rate": 1.0641577336322761e-07,
|
| 1146 |
+
"loss": 0.2815,
|
| 1147 |
+
"step": 710
|
| 1148 |
+
},
|
| 1149 |
+
{
|
| 1150 |
+
"epoch": 18.631921824104236,
|
| 1151 |
+
"grad_norm": 0.03857421875,
|
| 1152 |
+
"learning_rate": 8.625513476320291e-08,
|
| 1153 |
+
"loss": 0.2849,
|
| 1154 |
+
"step": 715
|
| 1155 |
+
},
|
| 1156 |
+
{
|
| 1157 |
+
"epoch": 18.762214983713356,
|
| 1158 |
+
"grad_norm": 0.0419921875,
|
| 1159 |
+
"learning_rate": 6.819348298638839e-08,
|
| 1160 |
+
"loss": 0.2852,
|
| 1161 |
+
"step": 720
|
| 1162 |
+
},
|
| 1163 |
+
{
|
| 1164 |
+
"epoch": 18.892508143322477,
|
| 1165 |
+
"grad_norm": 0.0390625,
|
| 1166 |
+
"learning_rate": 5.223853336398632e-08,
|
| 1167 |
+
"loss": 0.2829,
|
| 1168 |
+
"step": 725
|
| 1169 |
+
},
|
| 1170 |
+
{
|
| 1171 |
+
"epoch": 18.996742671009773,
|
| 1172 |
+
"eval_loss": 0.28573134541511536,
|
| 1173 |
+
"eval_runtime": 216.5831,
|
| 1174 |
+
"eval_samples_per_second": 2.018,
|
| 1175 |
+
"eval_steps_per_second": 0.508,
|
| 1176 |
+
"step": 729
|
| 1177 |
+
},
|
| 1178 |
+
{
|
| 1179 |
+
"epoch": 19.022801302931597,
|
| 1180 |
+
"grad_norm": 0.0419921875,
|
| 1181 |
+
"learning_rate": 3.839710131477492e-08,
|
| 1182 |
+
"loss": 0.275,
|
| 1183 |
+
"step": 730
|
| 1184 |
+
},
|
| 1185 |
+
{
|
| 1186 |
+
"epoch": 19.153094462540718,
|
| 1187 |
+
"grad_norm": 0.04248046875,
|
| 1188 |
+
"learning_rate": 2.6675099433787212e-08,
|
| 1189 |
+
"loss": 0.2854,
|
| 1190 |
+
"step": 735
|
| 1191 |
+
},
|
| 1192 |
+
{
|
| 1193 |
+
"epoch": 19.28338762214984,
|
| 1194 |
+
"grad_norm": 0.04150390625,
|
| 1195 |
+
"learning_rate": 1.7077534966650767e-08,
|
| 1196 |
+
"loss": 0.2794,
|
| 1197 |
+
"step": 740
|
| 1198 |
+
},
|
| 1199 |
+
{
|
| 1200 |
+
"epoch": 19.41368078175896,
|
| 1201 |
+
"grad_norm": 0.03759765625,
|
| 1202 |
+
"learning_rate": 9.608507670659239e-09,
|
| 1203 |
+
"loss": 0.2782,
|
| 1204 |
+
"step": 745
|
| 1205 |
+
},
|
| 1206 |
+
{
|
| 1207 |
+
"epoch": 19.54397394136808,
|
| 1208 |
+
"grad_norm": 0.040283203125,
|
| 1209 |
+
"learning_rate": 4.2712080634949024e-09,
|
| 1210 |
+
"loss": 0.2878,
|
| 1211 |
+
"step": 750
|
| 1212 |
+
},
|
| 1213 |
+
{
|
| 1214 |
+
"epoch": 19.6742671009772,
|
| 1215 |
+
"grad_norm": 0.0439453125,
|
| 1216 |
+
"learning_rate": 1.0679160603449533e-09,
|
| 1217 |
+
"loss": 0.2894,
|
| 1218 |
+
"step": 755
|
| 1219 |
+
},
|
| 1220 |
+
{
|
| 1221 |
+
"epoch": 19.80456026058632,
|
| 1222 |
+
"grad_norm": 0.0439453125,
|
| 1223 |
+
"learning_rate": 0.0,
|
| 1224 |
+
"loss": 0.2801,
|
| 1225 |
+
"step": 760
|
| 1226 |
+
},
|
| 1227 |
+
{
|
| 1228 |
+
"epoch": 19.80456026058632,
|
| 1229 |
+
"eval_loss": 0.2857275605201721,
|
| 1230 |
+
"eval_runtime": 216.4887,
|
| 1231 |
+
"eval_samples_per_second": 2.019,
|
| 1232 |
+
"eval_steps_per_second": 0.508,
|
| 1233 |
+
"step": 760
|
| 1234 |
+
},
|
| 1235 |
+
{
|
| 1236 |
+
"epoch": 19.80456026058632,
|
| 1237 |
+
"step": 760,
|
| 1238 |
+
"total_flos": 4.2808751411995607e+18,
|
| 1239 |
+
"train_loss": 0.32058246888612446,
|
| 1240 |
+
"train_runtime": 79494.9932,
|
| 1241 |
+
"train_samples_per_second": 0.618,
|
| 1242 |
+
"train_steps_per_second": 0.01
|
| 1243 |
+
}
|
| 1244 |
+
],
|
| 1245 |
+
"logging_steps": 5,
|
| 1246 |
+
"max_steps": 760,
|
| 1247 |
+
"num_input_tokens_seen": 0,
|
| 1248 |
+
"num_train_epochs": 20,
|
| 1249 |
+
"save_steps": 500,
|
| 1250 |
+
"total_flos": 4.2808751411995607e+18,
|
| 1251 |
+
"train_batch_size": 4,
|
| 1252 |
+
"trial_name": null,
|
| 1253 |
+
"trial_params": null
|
| 1254 |
+
}
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:a406389b49c8dd312724ff4a8cf593ccb002145b214622adaa63b29ccf07a00d
|
| 3 |
+
size 5112
|