Initial commit
Browse files- README.md +172 -0
- adapter_config.json +30 -0
- adapter_model.bin +3 -0
- adapter_model.safetensors +3 -0
- config.json +53 -0
- rng_state.pth +3 -0
- scheduler.pt +3 -0
- special_tokens_map.json +30 -0
- tokenizer.model +3 -0
- tokenizer_config.json +43 -0
- trainer_state.json +751 -0
- training_args.bin +3 -0
README.md
ADDED
|
@@ -0,0 +1,172 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
license: apache-2.0
|
| 3 |
+
library_name: peft
|
| 4 |
+
tags:
|
| 5 |
+
- generated_from_trainer
|
| 6 |
+
base_model: TheBloke/SOLAR-10.7B-Instruct-v1.0-uncensored-GPTQ
|
| 7 |
+
model-index:
|
| 8 |
+
- name: output_solor/exp_16
|
| 9 |
+
results: []
|
| 10 |
+
---
|
| 11 |
+
|
| 12 |
+
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
| 13 |
+
should probably proofread and complete it, then remove this comment. -->
|
| 14 |
+
|
| 15 |
+
[<img src="https://raw.githubusercontent.com/OpenAccess-AI-Collective/axolotl/main/image/axolotl-badge-web.png" alt="Built with Axolotl" width="200" height="32"/>](https://github.com/OpenAccess-AI-Collective/axolotl)
|
| 16 |
+
<details><summary>See axolotl config</summary>
|
| 17 |
+
|
| 18 |
+
axolotl version: `0.4.0`
|
| 19 |
+
```yaml
|
| 20 |
+
base_model: TheBloke/SOLAR-10.7B-Instruct-v1.0-uncensored-GPTQ
|
| 21 |
+
is_llama_derived_model: false
|
| 22 |
+
gptq: true
|
| 23 |
+
gptq_disable_exllama: true
|
| 24 |
+
model_type: AutoModelForCausalLM
|
| 25 |
+
tokenizer_type: LlamaTokenizer
|
| 26 |
+
tokenizer_use_fast: true
|
| 27 |
+
tokenizer_legacy: true
|
| 28 |
+
load_in_8bit: false
|
| 29 |
+
load_in_4bit: false
|
| 30 |
+
strict: false
|
| 31 |
+
push_dataset_to_hub:
|
| 32 |
+
hf_use_auth_token: true
|
| 33 |
+
datasets:
|
| 34 |
+
- path: datasets_cleansinng/datasets/helper_selector_1280_0305_v01.jsonl #Path to json dataset file in huggingface
|
| 35 |
+
#for type,conversation arguments read axolotl readme and pick what is suited for your project, I wanted a chatbot and put sharegpt and chatml
|
| 36 |
+
type:
|
| 37 |
+
system_prompt: "Instruction에 따라 적절하게 Input 데이터를 활용하여 Output 답변을 하세요. 너는 사용자 질문(Instruction)에 실시간으로 API 호출을 위한 Json 형식의 구조화된 결과를 생성하는 인공지능이야."
|
| 38 |
+
format: "[INST]### Instruction:\n{instruction}\n\n### Input:{input}\n\n[/INST]### Output: "
|
| 39 |
+
no_input_format: "[INST]### Instruction:\n{instruction}\n\n[/INST]### Output: "
|
| 40 |
+
field_instruction: Instruction
|
| 41 |
+
field_input: Input
|
| 42 |
+
field_output: Output
|
| 43 |
+
dataset_prepared_path:
|
| 44 |
+
val_set_size: 0.05
|
| 45 |
+
adapter: lora
|
| 46 |
+
lora_model_dir:
|
| 47 |
+
sequence_len: 4096
|
| 48 |
+
sample_packing:
|
| 49 |
+
lora_r: 32
|
| 50 |
+
lora_alpha: 32
|
| 51 |
+
lora_dropout: 0.05
|
| 52 |
+
lora_target_modules:
|
| 53 |
+
- k_proj
|
| 54 |
+
- o_proj
|
| 55 |
+
- q_proj
|
| 56 |
+
- v_proj
|
| 57 |
+
lora_target_linear:
|
| 58 |
+
lora_fan_in_fan_out:
|
| 59 |
+
wandb_project:
|
| 60 |
+
wandb_watch:
|
| 61 |
+
wandb_name:
|
| 62 |
+
wandb_log_model:
|
| 63 |
+
output_dir: ./output_solor/exp_16
|
| 64 |
+
gradient_accumulation_steps: 8
|
| 65 |
+
micro_batch_size: 8
|
| 66 |
+
num_epochs: 5
|
| 67 |
+
optimizer: adamw_torch
|
| 68 |
+
adam_beta2: 0.95
|
| 69 |
+
adam_eps: 0.00001
|
| 70 |
+
max_grad_norm: 1.0
|
| 71 |
+
torchdistx_path:
|
| 72 |
+
lr_scheduler: cosine
|
| 73 |
+
lr_quadratic_warmup: true
|
| 74 |
+
learning_rate: 0.0005
|
| 75 |
+
train_on_inputs: false
|
| 76 |
+
group_by_length: false
|
| 77 |
+
bf16: false
|
| 78 |
+
fp16: false
|
| 79 |
+
float16: true
|
| 80 |
+
tf32: true
|
| 81 |
+
gradient_checkpointing: true
|
| 82 |
+
early_stopping_patience:
|
| 83 |
+
resume_from_checkpoint:
|
| 84 |
+
local_rank:
|
| 85 |
+
logging_steps: 1
|
| 86 |
+
xformers_attention:
|
| 87 |
+
flash_attention:
|
| 88 |
+
sdp_attention:
|
| 89 |
+
flash_optimum:
|
| 90 |
+
warmup_steps: 100
|
| 91 |
+
evals_per_epoch: 4
|
| 92 |
+
saves_per_epoch: 1
|
| 93 |
+
debug:
|
| 94 |
+
deepspeed: deepspeed_configs/zero1.json
|
| 95 |
+
weight_decay: 0.1
|
| 96 |
+
special_tokens:
|
| 97 |
+
bos_token: "<s>"
|
| 98 |
+
eos_token: "</s>"
|
| 99 |
+
unk_token: "<unk>"
|
| 100 |
+
|
| 101 |
+
```
|
| 102 |
+
|
| 103 |
+
</details><br>
|
| 104 |
+
|
| 105 |
+
# output_solor/exp_16
|
| 106 |
+
|
| 107 |
+
This model is a fine-tuned version of [TheBloke/SOLAR-10.7B-Instruct-v1.0-uncensored-GPTQ](https://huggingface.co/TheBloke/SOLAR-10.7B-Instruct-v1.0-uncensored-GPTQ) on the None dataset.
|
| 108 |
+
It achieves the following results on the evaluation set:
|
| 109 |
+
- Loss: 0.2015
|
| 110 |
+
|
| 111 |
+
## Model description
|
| 112 |
+
|
| 113 |
+
More information needed
|
| 114 |
+
|
| 115 |
+
## Intended uses & limitations
|
| 116 |
+
|
| 117 |
+
More information needed
|
| 118 |
+
|
| 119 |
+
## Training and evaluation data
|
| 120 |
+
|
| 121 |
+
More information needed
|
| 122 |
+
|
| 123 |
+
## Training procedure
|
| 124 |
+
|
| 125 |
+
### Training hyperparameters
|
| 126 |
+
|
| 127 |
+
The following hyperparameters were used during training:
|
| 128 |
+
- learning_rate: 0.0005
|
| 129 |
+
- train_batch_size: 8
|
| 130 |
+
- eval_batch_size: 8
|
| 131 |
+
- seed: 42
|
| 132 |
+
- distributed_type: multi-GPU
|
| 133 |
+
- gradient_accumulation_steps: 8
|
| 134 |
+
- total_train_batch_size: 64
|
| 135 |
+
- optimizer: Adam with betas=(0.9,0.95) and epsilon=1e-08
|
| 136 |
+
- lr_scheduler_type: cosine
|
| 137 |
+
- lr_scheduler_warmup_steps: 100
|
| 138 |
+
- num_epochs: 5
|
| 139 |
+
|
| 140 |
+
### Training results
|
| 141 |
+
|
| 142 |
+
| Training Loss | Epoch | Step | Validation Loss |
|
| 143 |
+
|:-------------:|:-----:|:----:|:---------------:|
|
| 144 |
+
| 1.3493 | 0.05 | 1 | 1.2795 |
|
| 145 |
+
| 1.2483 | 0.26 | 5 | 1.2769 |
|
| 146 |
+
| 1.2275 | 0.53 | 10 | 1.2099 |
|
| 147 |
+
| 1.0529 | 0.79 | 15 | 1.0724 |
|
| 148 |
+
| 0.8642 | 1.05 | 20 | 0.9709 |
|
| 149 |
+
| 0.8477 | 1.32 | 25 | 0.8245 |
|
| 150 |
+
| 0.7207 | 1.58 | 30 | 0.6994 |
|
| 151 |
+
| 0.4656 | 1.84 | 35 | 0.5878 |
|
| 152 |
+
| 0.4949 | 2.11 | 40 | 0.4970 |
|
| 153 |
+
| 0.3497 | 2.37 | 45 | 0.4221 |
|
| 154 |
+
| 0.3288 | 2.63 | 50 | 0.3672 |
|
| 155 |
+
| 0.3011 | 2.89 | 55 | 0.3250 |
|
| 156 |
+
| 0.2648 | 3.16 | 60 | 0.2900 |
|
| 157 |
+
| 0.3084 | 3.42 | 65 | 0.2591 |
|
| 158 |
+
| 0.2696 | 3.68 | 70 | 0.2459 |
|
| 159 |
+
| 0.2197 | 3.95 | 75 | 0.2286 |
|
| 160 |
+
| 0.1905 | 4.21 | 80 | 0.2111 |
|
| 161 |
+
| 0.1815 | 4.47 | 85 | 0.2084 |
|
| 162 |
+
| 0.2164 | 4.74 | 90 | 0.2128 |
|
| 163 |
+
| 0.1412 | 5.0 | 95 | 0.2015 |
|
| 164 |
+
|
| 165 |
+
|
| 166 |
+
### Framework versions
|
| 167 |
+
|
| 168 |
+
- PEFT 0.9.1.dev0
|
| 169 |
+
- Transformers 4.38.0.dev0
|
| 170 |
+
- Pytorch 2.0.1+cu117
|
| 171 |
+
- Datasets 2.18.0
|
| 172 |
+
- Tokenizers 0.15.0
|
adapter_config.json
ADDED
|
@@ -0,0 +1,30 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"alpha_pattern": {},
|
| 3 |
+
"auto_mapping": null,
|
| 4 |
+
"base_model_name_or_path": "TheBloke/SOLAR-10.7B-Instruct-v1.0-uncensored-GPTQ",
|
| 5 |
+
"bias": "none",
|
| 6 |
+
"fan_in_fan_out": null,
|
| 7 |
+
"inference_mode": true,
|
| 8 |
+
"init_lora_weights": true,
|
| 9 |
+
"layers_pattern": null,
|
| 10 |
+
"layers_to_transform": null,
|
| 11 |
+
"loftq_config": {},
|
| 12 |
+
"lora_alpha": 32,
|
| 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": 32,
|
| 19 |
+
"rank_pattern": {},
|
| 20 |
+
"revision": null,
|
| 21 |
+
"target_modules": [
|
| 22 |
+
"v_proj",
|
| 23 |
+
"o_proj",
|
| 24 |
+
"k_proj",
|
| 25 |
+
"q_proj"
|
| 26 |
+
],
|
| 27 |
+
"task_type": "CAUSAL_LM",
|
| 28 |
+
"use_dora": false,
|
| 29 |
+
"use_rslora": false
|
| 30 |
+
}
|
adapter_model.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:a73d8948cbc0281d87f4edf2f31bdf4a19dd525dfcc609247c64a007af7453ec
|
| 3 |
+
size 163638973
|
adapter_model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:442249e41a7b14fe7e3d09c22d68e393a3472fb4398734124a485a3abba3fa3e
|
| 3 |
+
size 163629904
|
config.json
ADDED
|
@@ -0,0 +1,53 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "TheBloke/SOLAR-10.7B-Instruct-v1.0-uncensored-GPTQ",
|
| 3 |
+
"architectures": [
|
| 4 |
+
"LlamaForCausalLM"
|
| 5 |
+
],
|
| 6 |
+
"attention_bias": false,
|
| 7 |
+
"attention_dropout": 0.0,
|
| 8 |
+
"bos_token_id": 1,
|
| 9 |
+
"eos_token_id": 2,
|
| 10 |
+
"hidden_act": "silu",
|
| 11 |
+
"hidden_size": 4096,
|
| 12 |
+
"initializer_range": 0.02,
|
| 13 |
+
"intermediate_size": 14336,
|
| 14 |
+
"max_position_embeddings": 4096,
|
| 15 |
+
"model_type": "llama",
|
| 16 |
+
"num_attention_heads": 32,
|
| 17 |
+
"num_hidden_layers": 48,
|
| 18 |
+
"num_key_value_heads": 8,
|
| 19 |
+
"pad_token_id": 2,
|
| 20 |
+
"pretraining_tp": 1,
|
| 21 |
+
"quantization_config": {
|
| 22 |
+
"batch_size": 1,
|
| 23 |
+
"bits": 4,
|
| 24 |
+
"block_name_to_quantize": null,
|
| 25 |
+
"cache_block_outputs": true,
|
| 26 |
+
"damp_percent": 0.1,
|
| 27 |
+
"dataset": null,
|
| 28 |
+
"desc_act": true,
|
| 29 |
+
"exllama_config": {
|
| 30 |
+
"version": 1
|
| 31 |
+
},
|
| 32 |
+
"group_size": 128,
|
| 33 |
+
"max_input_length": null,
|
| 34 |
+
"model_seqlen": null,
|
| 35 |
+
"module_name_preceding_first_block": null,
|
| 36 |
+
"modules_in_block_to_quantize": null,
|
| 37 |
+
"pad_token_id": null,
|
| 38 |
+
"quant_method": "gptq",
|
| 39 |
+
"sym": true,
|
| 40 |
+
"tokenizer": null,
|
| 41 |
+
"true_sequential": true,
|
| 42 |
+
"use_cuda_fp16": false,
|
| 43 |
+
"use_exllama": false
|
| 44 |
+
},
|
| 45 |
+
"rms_norm_eps": 1e-05,
|
| 46 |
+
"rope_scaling": null,
|
| 47 |
+
"rope_theta": 10000.0,
|
| 48 |
+
"tie_word_embeddings": false,
|
| 49 |
+
"torch_dtype": "float16",
|
| 50 |
+
"transformers_version": "4.38.0.dev0",
|
| 51 |
+
"use_cache": false,
|
| 52 |
+
"vocab_size": 32000
|
| 53 |
+
}
|
rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:8f2e36934e8695cfe88ba99ce4bccbcd32a66a63f999c007176e12c26e97f01e
|
| 3 |
+
size 14575
|
scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:877e4d20600fd19d73ff925197db60020e3b1bc4858ef2aba16e8a00797785f7
|
| 3 |
+
size 627
|
special_tokens_map.json
ADDED
|
@@ -0,0 +1,30 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"bos_token": {
|
| 3 |
+
"content": "<s>",
|
| 4 |
+
"lstrip": false,
|
| 5 |
+
"normalized": false,
|
| 6 |
+
"rstrip": false,
|
| 7 |
+
"single_word": false
|
| 8 |
+
},
|
| 9 |
+
"eos_token": {
|
| 10 |
+
"content": "</s>",
|
| 11 |
+
"lstrip": false,
|
| 12 |
+
"normalized": false,
|
| 13 |
+
"rstrip": false,
|
| 14 |
+
"single_word": false
|
| 15 |
+
},
|
| 16 |
+
"pad_token": {
|
| 17 |
+
"content": "</s>",
|
| 18 |
+
"lstrip": false,
|
| 19 |
+
"normalized": false,
|
| 20 |
+
"rstrip": false,
|
| 21 |
+
"single_word": false
|
| 22 |
+
},
|
| 23 |
+
"unk_token": {
|
| 24 |
+
"content": "<unk>",
|
| 25 |
+
"lstrip": false,
|
| 26 |
+
"normalized": false,
|
| 27 |
+
"rstrip": false,
|
| 28 |
+
"single_word": false
|
| 29 |
+
}
|
| 30 |
+
}
|
tokenizer.model
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:dadfd56d766715c61d2ef780a525ab43b8e6da4de6865bda3d95fdef5e134055
|
| 3 |
+
size 493443
|
tokenizer_config.json
ADDED
|
@@ -0,0 +1,43 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 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 |
+
"bos_token": "<s>",
|
| 31 |
+
"clean_up_tokenization_spaces": false,
|
| 32 |
+
"eos_token": "</s>",
|
| 33 |
+
"legacy": true,
|
| 34 |
+
"model_max_length": 1000000000000000019884624838656,
|
| 35 |
+
"pad_token": "</s>",
|
| 36 |
+
"sp_model_kwargs": {},
|
| 37 |
+
"spaces_between_special_tokens": false,
|
| 38 |
+
"tokenizer_class": "LlamaTokenizer",
|
| 39 |
+
"trust_remote_code": false,
|
| 40 |
+
"unk_token": "<unk>",
|
| 41 |
+
"use_default_system_prompt": false,
|
| 42 |
+
"use_fast": true
|
| 43 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,751 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.2015175074338913,
|
| 3 |
+
"best_model_checkpoint": "./output_solor/exp_16/checkpoint-95",
|
| 4 |
+
"epoch": 5.0,
|
| 5 |
+
"eval_steps": 5,
|
| 6 |
+
"global_step": 95,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.05,
|
| 13 |
+
"learning_rate": 5.0000000000000004e-08,
|
| 14 |
+
"loss": 1.3493,
|
| 15 |
+
"step": 1
|
| 16 |
+
},
|
| 17 |
+
{
|
| 18 |
+
"epoch": 0.05,
|
| 19 |
+
"eval_loss": 1.2794537544250488,
|
| 20 |
+
"eval_runtime": 20.3935,
|
| 21 |
+
"eval_samples_per_second": 3.138,
|
| 22 |
+
"eval_steps_per_second": 0.392,
|
| 23 |
+
"step": 1
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.11,
|
| 27 |
+
"learning_rate": 2.0000000000000002e-07,
|
| 28 |
+
"loss": 1.301,
|
| 29 |
+
"step": 2
|
| 30 |
+
},
|
| 31 |
+
{
|
| 32 |
+
"epoch": 0.16,
|
| 33 |
+
"learning_rate": 4.5e-07,
|
| 34 |
+
"loss": 1.2596,
|
| 35 |
+
"step": 3
|
| 36 |
+
},
|
| 37 |
+
{
|
| 38 |
+
"epoch": 0.21,
|
| 39 |
+
"learning_rate": 8.000000000000001e-07,
|
| 40 |
+
"loss": 1.2522,
|
| 41 |
+
"step": 4
|
| 42 |
+
},
|
| 43 |
+
{
|
| 44 |
+
"epoch": 0.26,
|
| 45 |
+
"learning_rate": 1.2500000000000003e-06,
|
| 46 |
+
"loss": 1.2483,
|
| 47 |
+
"step": 5
|
| 48 |
+
},
|
| 49 |
+
{
|
| 50 |
+
"epoch": 0.26,
|
| 51 |
+
"eval_loss": 1.2768707275390625,
|
| 52 |
+
"eval_runtime": 20.6266,
|
| 53 |
+
"eval_samples_per_second": 3.103,
|
| 54 |
+
"eval_steps_per_second": 0.388,
|
| 55 |
+
"step": 5
|
| 56 |
+
},
|
| 57 |
+
{
|
| 58 |
+
"epoch": 0.32,
|
| 59 |
+
"learning_rate": 1.8e-06,
|
| 60 |
+
"loss": 1.3056,
|
| 61 |
+
"step": 6
|
| 62 |
+
},
|
| 63 |
+
{
|
| 64 |
+
"epoch": 0.37,
|
| 65 |
+
"learning_rate": 2.4500000000000003e-06,
|
| 66 |
+
"loss": 1.4414,
|
| 67 |
+
"step": 7
|
| 68 |
+
},
|
| 69 |
+
{
|
| 70 |
+
"epoch": 0.42,
|
| 71 |
+
"learning_rate": 3.2000000000000003e-06,
|
| 72 |
+
"loss": 1.2347,
|
| 73 |
+
"step": 8
|
| 74 |
+
},
|
| 75 |
+
{
|
| 76 |
+
"epoch": 0.47,
|
| 77 |
+
"learning_rate": 4.05e-06,
|
| 78 |
+
"loss": 1.2399,
|
| 79 |
+
"step": 9
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.53,
|
| 83 |
+
"learning_rate": 5.000000000000001e-06,
|
| 84 |
+
"loss": 1.2275,
|
| 85 |
+
"step": 10
|
| 86 |
+
},
|
| 87 |
+
{
|
| 88 |
+
"epoch": 0.53,
|
| 89 |
+
"eval_loss": 1.2099063396453857,
|
| 90 |
+
"eval_runtime": 20.6171,
|
| 91 |
+
"eval_samples_per_second": 3.104,
|
| 92 |
+
"eval_steps_per_second": 0.388,
|
| 93 |
+
"step": 10
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 0.58,
|
| 97 |
+
"learning_rate": 6.05e-06,
|
| 98 |
+
"loss": 1.154,
|
| 99 |
+
"step": 11
|
| 100 |
+
},
|
| 101 |
+
{
|
| 102 |
+
"epoch": 0.63,
|
| 103 |
+
"learning_rate": 7.2e-06,
|
| 104 |
+
"loss": 1.2325,
|
| 105 |
+
"step": 12
|
| 106 |
+
},
|
| 107 |
+
{
|
| 108 |
+
"epoch": 0.68,
|
| 109 |
+
"learning_rate": 8.45e-06,
|
| 110 |
+
"loss": 1.1022,
|
| 111 |
+
"step": 13
|
| 112 |
+
},
|
| 113 |
+
{
|
| 114 |
+
"epoch": 0.74,
|
| 115 |
+
"learning_rate": 9.800000000000001e-06,
|
| 116 |
+
"loss": 1.0295,
|
| 117 |
+
"step": 14
|
| 118 |
+
},
|
| 119 |
+
{
|
| 120 |
+
"epoch": 0.79,
|
| 121 |
+
"learning_rate": 1.1249999999999999e-05,
|
| 122 |
+
"loss": 1.0529,
|
| 123 |
+
"step": 15
|
| 124 |
+
},
|
| 125 |
+
{
|
| 126 |
+
"epoch": 0.79,
|
| 127 |
+
"eval_loss": 1.0723849534988403,
|
| 128 |
+
"eval_runtime": 20.6213,
|
| 129 |
+
"eval_samples_per_second": 3.104,
|
| 130 |
+
"eval_steps_per_second": 0.388,
|
| 131 |
+
"step": 15
|
| 132 |
+
},
|
| 133 |
+
{
|
| 134 |
+
"epoch": 0.84,
|
| 135 |
+
"learning_rate": 1.2800000000000001e-05,
|
| 136 |
+
"loss": 1.1019,
|
| 137 |
+
"step": 16
|
| 138 |
+
},
|
| 139 |
+
{
|
| 140 |
+
"epoch": 0.89,
|
| 141 |
+
"learning_rate": 1.4450000000000002e-05,
|
| 142 |
+
"loss": 1.0158,
|
| 143 |
+
"step": 17
|
| 144 |
+
},
|
| 145 |
+
{
|
| 146 |
+
"epoch": 0.95,
|
| 147 |
+
"learning_rate": 1.62e-05,
|
| 148 |
+
"loss": 1.0235,
|
| 149 |
+
"step": 18
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 1.0,
|
| 153 |
+
"learning_rate": 1.805e-05,
|
| 154 |
+
"loss": 1.1272,
|
| 155 |
+
"step": 19
|
| 156 |
+
},
|
| 157 |
+
{
|
| 158 |
+
"epoch": 1.05,
|
| 159 |
+
"learning_rate": 2.0000000000000005e-05,
|
| 160 |
+
"loss": 0.8642,
|
| 161 |
+
"step": 20
|
| 162 |
+
},
|
| 163 |
+
{
|
| 164 |
+
"epoch": 1.05,
|
| 165 |
+
"eval_loss": 0.9709420204162598,
|
| 166 |
+
"eval_runtime": 20.3773,
|
| 167 |
+
"eval_samples_per_second": 3.141,
|
| 168 |
+
"eval_steps_per_second": 0.393,
|
| 169 |
+
"step": 20
|
| 170 |
+
},
|
| 171 |
+
{
|
| 172 |
+
"epoch": 1.11,
|
| 173 |
+
"learning_rate": 2.2049999999999997e-05,
|
| 174 |
+
"loss": 0.996,
|
| 175 |
+
"step": 21
|
| 176 |
+
},
|
| 177 |
+
{
|
| 178 |
+
"epoch": 1.16,
|
| 179 |
+
"learning_rate": 2.42e-05,
|
| 180 |
+
"loss": 0.8983,
|
| 181 |
+
"step": 22
|
| 182 |
+
},
|
| 183 |
+
{
|
| 184 |
+
"epoch": 1.21,
|
| 185 |
+
"learning_rate": 2.6450000000000003e-05,
|
| 186 |
+
"loss": 0.9876,
|
| 187 |
+
"step": 23
|
| 188 |
+
},
|
| 189 |
+
{
|
| 190 |
+
"epoch": 1.26,
|
| 191 |
+
"learning_rate": 2.88e-05,
|
| 192 |
+
"loss": 0.7844,
|
| 193 |
+
"step": 24
|
| 194 |
+
},
|
| 195 |
+
{
|
| 196 |
+
"epoch": 1.32,
|
| 197 |
+
"learning_rate": 3.125e-05,
|
| 198 |
+
"loss": 0.8477,
|
| 199 |
+
"step": 25
|
| 200 |
+
},
|
| 201 |
+
{
|
| 202 |
+
"epoch": 1.32,
|
| 203 |
+
"eval_loss": 0.8244509696960449,
|
| 204 |
+
"eval_runtime": 20.6023,
|
| 205 |
+
"eval_samples_per_second": 3.106,
|
| 206 |
+
"eval_steps_per_second": 0.388,
|
| 207 |
+
"step": 25
|
| 208 |
+
},
|
| 209 |
+
{
|
| 210 |
+
"epoch": 1.37,
|
| 211 |
+
"learning_rate": 3.38e-05,
|
| 212 |
+
"loss": 0.786,
|
| 213 |
+
"step": 26
|
| 214 |
+
},
|
| 215 |
+
{
|
| 216 |
+
"epoch": 1.42,
|
| 217 |
+
"learning_rate": 3.6450000000000005e-05,
|
| 218 |
+
"loss": 0.6724,
|
| 219 |
+
"step": 27
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 1.47,
|
| 223 |
+
"learning_rate": 3.9200000000000004e-05,
|
| 224 |
+
"loss": 0.8566,
|
| 225 |
+
"step": 28
|
| 226 |
+
},
|
| 227 |
+
{
|
| 228 |
+
"epoch": 1.53,
|
| 229 |
+
"learning_rate": 4.205e-05,
|
| 230 |
+
"loss": 0.6692,
|
| 231 |
+
"step": 29
|
| 232 |
+
},
|
| 233 |
+
{
|
| 234 |
+
"epoch": 1.58,
|
| 235 |
+
"learning_rate": 4.4999999999999996e-05,
|
| 236 |
+
"loss": 0.7207,
|
| 237 |
+
"step": 30
|
| 238 |
+
},
|
| 239 |
+
{
|
| 240 |
+
"epoch": 1.58,
|
| 241 |
+
"eval_loss": 0.6993950009346008,
|
| 242 |
+
"eval_runtime": 20.6231,
|
| 243 |
+
"eval_samples_per_second": 3.103,
|
| 244 |
+
"eval_steps_per_second": 0.388,
|
| 245 |
+
"step": 30
|
| 246 |
+
},
|
| 247 |
+
{
|
| 248 |
+
"epoch": 1.63,
|
| 249 |
+
"learning_rate": 4.805e-05,
|
| 250 |
+
"loss": 0.714,
|
| 251 |
+
"step": 31
|
| 252 |
+
},
|
| 253 |
+
{
|
| 254 |
+
"epoch": 1.68,
|
| 255 |
+
"learning_rate": 5.1200000000000004e-05,
|
| 256 |
+
"loss": 0.6141,
|
| 257 |
+
"step": 32
|
| 258 |
+
},
|
| 259 |
+
{
|
| 260 |
+
"epoch": 1.74,
|
| 261 |
+
"learning_rate": 5.445000000000001e-05,
|
| 262 |
+
"loss": 0.7061,
|
| 263 |
+
"step": 33
|
| 264 |
+
},
|
| 265 |
+
{
|
| 266 |
+
"epoch": 1.79,
|
| 267 |
+
"learning_rate": 5.780000000000001e-05,
|
| 268 |
+
"loss": 0.652,
|
| 269 |
+
"step": 34
|
| 270 |
+
},
|
| 271 |
+
{
|
| 272 |
+
"epoch": 1.84,
|
| 273 |
+
"learning_rate": 6.125e-05,
|
| 274 |
+
"loss": 0.4656,
|
| 275 |
+
"step": 35
|
| 276 |
+
},
|
| 277 |
+
{
|
| 278 |
+
"epoch": 1.84,
|
| 279 |
+
"eval_loss": 0.5878281593322754,
|
| 280 |
+
"eval_runtime": 20.6201,
|
| 281 |
+
"eval_samples_per_second": 3.104,
|
| 282 |
+
"eval_steps_per_second": 0.388,
|
| 283 |
+
"step": 35
|
| 284 |
+
},
|
| 285 |
+
{
|
| 286 |
+
"epoch": 1.89,
|
| 287 |
+
"learning_rate": 6.48e-05,
|
| 288 |
+
"loss": 0.6225,
|
| 289 |
+
"step": 36
|
| 290 |
+
},
|
| 291 |
+
{
|
| 292 |
+
"epoch": 1.95,
|
| 293 |
+
"learning_rate": 6.845e-05,
|
| 294 |
+
"loss": 0.4638,
|
| 295 |
+
"step": 37
|
| 296 |
+
},
|
| 297 |
+
{
|
| 298 |
+
"epoch": 2.0,
|
| 299 |
+
"learning_rate": 7.22e-05,
|
| 300 |
+
"loss": 0.595,
|
| 301 |
+
"step": 38
|
| 302 |
+
},
|
| 303 |
+
{
|
| 304 |
+
"epoch": 2.05,
|
| 305 |
+
"learning_rate": 7.605e-05,
|
| 306 |
+
"loss": 0.5618,
|
| 307 |
+
"step": 39
|
| 308 |
+
},
|
| 309 |
+
{
|
| 310 |
+
"epoch": 2.11,
|
| 311 |
+
"learning_rate": 8.000000000000002e-05,
|
| 312 |
+
"loss": 0.4949,
|
| 313 |
+
"step": 40
|
| 314 |
+
},
|
| 315 |
+
{
|
| 316 |
+
"epoch": 2.11,
|
| 317 |
+
"eval_loss": 0.4970114827156067,
|
| 318 |
+
"eval_runtime": 20.6228,
|
| 319 |
+
"eval_samples_per_second": 3.103,
|
| 320 |
+
"eval_steps_per_second": 0.388,
|
| 321 |
+
"step": 40
|
| 322 |
+
},
|
| 323 |
+
{
|
| 324 |
+
"epoch": 2.16,
|
| 325 |
+
"learning_rate": 8.404999999999998e-05,
|
| 326 |
+
"loss": 0.5576,
|
| 327 |
+
"step": 41
|
| 328 |
+
},
|
| 329 |
+
{
|
| 330 |
+
"epoch": 2.21,
|
| 331 |
+
"learning_rate": 8.819999999999999e-05,
|
| 332 |
+
"loss": 0.5537,
|
| 333 |
+
"step": 42
|
| 334 |
+
},
|
| 335 |
+
{
|
| 336 |
+
"epoch": 2.26,
|
| 337 |
+
"learning_rate": 9.245e-05,
|
| 338 |
+
"loss": 0.3734,
|
| 339 |
+
"step": 43
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 2.32,
|
| 343 |
+
"learning_rate": 9.68e-05,
|
| 344 |
+
"loss": 0.4103,
|
| 345 |
+
"step": 44
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 2.37,
|
| 349 |
+
"learning_rate": 0.00010125000000000001,
|
| 350 |
+
"loss": 0.3497,
|
| 351 |
+
"step": 45
|
| 352 |
+
},
|
| 353 |
+
{
|
| 354 |
+
"epoch": 2.37,
|
| 355 |
+
"eval_loss": 0.42206770181655884,
|
| 356 |
+
"eval_runtime": 20.6133,
|
| 357 |
+
"eval_samples_per_second": 3.105,
|
| 358 |
+
"eval_steps_per_second": 0.388,
|
| 359 |
+
"step": 45
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"epoch": 2.42,
|
| 363 |
+
"learning_rate": 0.00010580000000000001,
|
| 364 |
+
"loss": 0.4948,
|
| 365 |
+
"step": 46
|
| 366 |
+
},
|
| 367 |
+
{
|
| 368 |
+
"epoch": 2.47,
|
| 369 |
+
"learning_rate": 0.00011045,
|
| 370 |
+
"loss": 0.2884,
|
| 371 |
+
"step": 47
|
| 372 |
+
},
|
| 373 |
+
{
|
| 374 |
+
"epoch": 2.53,
|
| 375 |
+
"learning_rate": 0.0001152,
|
| 376 |
+
"loss": 0.4234,
|
| 377 |
+
"step": 48
|
| 378 |
+
},
|
| 379 |
+
{
|
| 380 |
+
"epoch": 2.58,
|
| 381 |
+
"learning_rate": 0.00012004999999999999,
|
| 382 |
+
"loss": 0.3352,
|
| 383 |
+
"step": 49
|
| 384 |
+
},
|
| 385 |
+
{
|
| 386 |
+
"epoch": 2.63,
|
| 387 |
+
"learning_rate": 0.000125,
|
| 388 |
+
"loss": 0.3288,
|
| 389 |
+
"step": 50
|
| 390 |
+
},
|
| 391 |
+
{
|
| 392 |
+
"epoch": 2.63,
|
| 393 |
+
"eval_loss": 0.3672243356704712,
|
| 394 |
+
"eval_runtime": 20.6204,
|
| 395 |
+
"eval_samples_per_second": 3.104,
|
| 396 |
+
"eval_steps_per_second": 0.388,
|
| 397 |
+
"step": 50
|
| 398 |
+
},
|
| 399 |
+
{
|
| 400 |
+
"epoch": 2.68,
|
| 401 |
+
"learning_rate": 0.00013005,
|
| 402 |
+
"loss": 0.3481,
|
| 403 |
+
"step": 51
|
| 404 |
+
},
|
| 405 |
+
{
|
| 406 |
+
"epoch": 2.74,
|
| 407 |
+
"learning_rate": 0.0001352,
|
| 408 |
+
"loss": 0.362,
|
| 409 |
+
"step": 52
|
| 410 |
+
},
|
| 411 |
+
{
|
| 412 |
+
"epoch": 2.79,
|
| 413 |
+
"learning_rate": 0.00014045000000000003,
|
| 414 |
+
"loss": 0.344,
|
| 415 |
+
"step": 53
|
| 416 |
+
},
|
| 417 |
+
{
|
| 418 |
+
"epoch": 2.84,
|
| 419 |
+
"learning_rate": 0.00014580000000000002,
|
| 420 |
+
"loss": 0.426,
|
| 421 |
+
"step": 54
|
| 422 |
+
},
|
| 423 |
+
{
|
| 424 |
+
"epoch": 2.89,
|
| 425 |
+
"learning_rate": 0.00015125000000000002,
|
| 426 |
+
"loss": 0.3011,
|
| 427 |
+
"step": 55
|
| 428 |
+
},
|
| 429 |
+
{
|
| 430 |
+
"epoch": 2.89,
|
| 431 |
+
"eval_loss": 0.32503682374954224,
|
| 432 |
+
"eval_runtime": 20.6368,
|
| 433 |
+
"eval_samples_per_second": 3.101,
|
| 434 |
+
"eval_steps_per_second": 0.388,
|
| 435 |
+
"step": 55
|
| 436 |
+
},
|
| 437 |
+
{
|
| 438 |
+
"epoch": 2.95,
|
| 439 |
+
"learning_rate": 0.00015680000000000002,
|
| 440 |
+
"loss": 0.4011,
|
| 441 |
+
"step": 56
|
| 442 |
+
},
|
| 443 |
+
{
|
| 444 |
+
"epoch": 3.0,
|
| 445 |
+
"learning_rate": 0.00016245,
|
| 446 |
+
"loss": 0.3796,
|
| 447 |
+
"step": 57
|
| 448 |
+
},
|
| 449 |
+
{
|
| 450 |
+
"epoch": 3.05,
|
| 451 |
+
"learning_rate": 0.0001682,
|
| 452 |
+
"loss": 0.2726,
|
| 453 |
+
"step": 58
|
| 454 |
+
},
|
| 455 |
+
{
|
| 456 |
+
"epoch": 3.11,
|
| 457 |
+
"learning_rate": 0.00017404999999999998,
|
| 458 |
+
"loss": 0.3218,
|
| 459 |
+
"step": 59
|
| 460 |
+
},
|
| 461 |
+
{
|
| 462 |
+
"epoch": 3.16,
|
| 463 |
+
"learning_rate": 0.00017999999999999998,
|
| 464 |
+
"loss": 0.2648,
|
| 465 |
+
"step": 60
|
| 466 |
+
},
|
| 467 |
+
{
|
| 468 |
+
"epoch": 3.16,
|
| 469 |
+
"eval_loss": 0.29000362753868103,
|
| 470 |
+
"eval_runtime": 20.595,
|
| 471 |
+
"eval_samples_per_second": 3.108,
|
| 472 |
+
"eval_steps_per_second": 0.388,
|
| 473 |
+
"step": 60
|
| 474 |
+
},
|
| 475 |
+
{
|
| 476 |
+
"epoch": 3.21,
|
| 477 |
+
"learning_rate": 0.00018605,
|
| 478 |
+
"loss": 0.331,
|
| 479 |
+
"step": 61
|
| 480 |
+
},
|
| 481 |
+
{
|
| 482 |
+
"epoch": 3.26,
|
| 483 |
+
"learning_rate": 0.0001922,
|
| 484 |
+
"loss": 0.2442,
|
| 485 |
+
"step": 62
|
| 486 |
+
},
|
| 487 |
+
{
|
| 488 |
+
"epoch": 3.32,
|
| 489 |
+
"learning_rate": 0.00019845000000000003,
|
| 490 |
+
"loss": 0.2839,
|
| 491 |
+
"step": 63
|
| 492 |
+
},
|
| 493 |
+
{
|
| 494 |
+
"epoch": 3.37,
|
| 495 |
+
"learning_rate": 0.00020480000000000002,
|
| 496 |
+
"loss": 0.3206,
|
| 497 |
+
"step": 64
|
| 498 |
+
},
|
| 499 |
+
{
|
| 500 |
+
"epoch": 3.42,
|
| 501 |
+
"learning_rate": 0.00021125,
|
| 502 |
+
"loss": 0.3084,
|
| 503 |
+
"step": 65
|
| 504 |
+
},
|
| 505 |
+
{
|
| 506 |
+
"epoch": 3.42,
|
| 507 |
+
"eval_loss": 0.25911369919776917,
|
| 508 |
+
"eval_runtime": 20.6071,
|
| 509 |
+
"eval_samples_per_second": 3.106,
|
| 510 |
+
"eval_steps_per_second": 0.388,
|
| 511 |
+
"step": 65
|
| 512 |
+
},
|
| 513 |
+
{
|
| 514 |
+
"epoch": 3.47,
|
| 515 |
+
"learning_rate": 0.00021780000000000004,
|
| 516 |
+
"loss": 0.2479,
|
| 517 |
+
"step": 66
|
| 518 |
+
},
|
| 519 |
+
{
|
| 520 |
+
"epoch": 3.53,
|
| 521 |
+
"learning_rate": 0.00022445000000000003,
|
| 522 |
+
"loss": 0.3003,
|
| 523 |
+
"step": 67
|
| 524 |
+
},
|
| 525 |
+
{
|
| 526 |
+
"epoch": 3.58,
|
| 527 |
+
"learning_rate": 0.00023120000000000004,
|
| 528 |
+
"loss": 0.2757,
|
| 529 |
+
"step": 68
|
| 530 |
+
},
|
| 531 |
+
{
|
| 532 |
+
"epoch": 3.63,
|
| 533 |
+
"learning_rate": 0.00023804999999999996,
|
| 534 |
+
"loss": 0.1921,
|
| 535 |
+
"step": 69
|
| 536 |
+
},
|
| 537 |
+
{
|
| 538 |
+
"epoch": 3.68,
|
| 539 |
+
"learning_rate": 0.000245,
|
| 540 |
+
"loss": 0.2696,
|
| 541 |
+
"step": 70
|
| 542 |
+
},
|
| 543 |
+
{
|
| 544 |
+
"epoch": 3.68,
|
| 545 |
+
"eval_loss": 0.24587483704090118,
|
| 546 |
+
"eval_runtime": 20.6172,
|
| 547 |
+
"eval_samples_per_second": 3.104,
|
| 548 |
+
"eval_steps_per_second": 0.388,
|
| 549 |
+
"step": 70
|
| 550 |
+
},
|
| 551 |
+
{
|
| 552 |
+
"epoch": 3.74,
|
| 553 |
+
"learning_rate": 0.00025205000000000003,
|
| 554 |
+
"loss": 0.2596,
|
| 555 |
+
"step": 71
|
| 556 |
+
},
|
| 557 |
+
{
|
| 558 |
+
"epoch": 3.79,
|
| 559 |
+
"learning_rate": 0.0002592,
|
| 560 |
+
"loss": 0.2386,
|
| 561 |
+
"step": 72
|
| 562 |
+
},
|
| 563 |
+
{
|
| 564 |
+
"epoch": 3.84,
|
| 565 |
+
"learning_rate": 0.00026644999999999994,
|
| 566 |
+
"loss": 0.2281,
|
| 567 |
+
"step": 73
|
| 568 |
+
},
|
| 569 |
+
{
|
| 570 |
+
"epoch": 3.89,
|
| 571 |
+
"learning_rate": 0.0002738,
|
| 572 |
+
"loss": 0.2538,
|
| 573 |
+
"step": 74
|
| 574 |
+
},
|
| 575 |
+
{
|
| 576 |
+
"epoch": 3.95,
|
| 577 |
+
"learning_rate": 0.00028125000000000003,
|
| 578 |
+
"loss": 0.2197,
|
| 579 |
+
"step": 75
|
| 580 |
+
},
|
| 581 |
+
{
|
| 582 |
+
"epoch": 3.95,
|
| 583 |
+
"eval_loss": 0.22858355939388275,
|
| 584 |
+
"eval_runtime": 20.6122,
|
| 585 |
+
"eval_samples_per_second": 3.105,
|
| 586 |
+
"eval_steps_per_second": 0.388,
|
| 587 |
+
"step": 75
|
| 588 |
+
},
|
| 589 |
+
{
|
| 590 |
+
"epoch": 4.0,
|
| 591 |
+
"learning_rate": 0.0002888,
|
| 592 |
+
"loss": 0.182,
|
| 593 |
+
"step": 76
|
| 594 |
+
},
|
| 595 |
+
{
|
| 596 |
+
"epoch": 4.05,
|
| 597 |
+
"learning_rate": 0.00029645,
|
| 598 |
+
"loss": 0.1985,
|
| 599 |
+
"step": 77
|
| 600 |
+
},
|
| 601 |
+
{
|
| 602 |
+
"epoch": 4.11,
|
| 603 |
+
"learning_rate": 0.0003042,
|
| 604 |
+
"loss": 0.1978,
|
| 605 |
+
"step": 78
|
| 606 |
+
},
|
| 607 |
+
{
|
| 608 |
+
"epoch": 4.16,
|
| 609 |
+
"learning_rate": 0.0003120500000000001,
|
| 610 |
+
"loss": 0.2283,
|
| 611 |
+
"step": 79
|
| 612 |
+
},
|
| 613 |
+
{
|
| 614 |
+
"epoch": 4.21,
|
| 615 |
+
"learning_rate": 0.0003200000000000001,
|
| 616 |
+
"loss": 0.1905,
|
| 617 |
+
"step": 80
|
| 618 |
+
},
|
| 619 |
+
{
|
| 620 |
+
"epoch": 4.21,
|
| 621 |
+
"eval_loss": 0.21105319261550903,
|
| 622 |
+
"eval_runtime": 20.595,
|
| 623 |
+
"eval_samples_per_second": 3.108,
|
| 624 |
+
"eval_steps_per_second": 0.388,
|
| 625 |
+
"step": 80
|
| 626 |
+
},
|
| 627 |
+
{
|
| 628 |
+
"epoch": 4.26,
|
| 629 |
+
"learning_rate": 0.0003280500000000001,
|
| 630 |
+
"loss": 0.2222,
|
| 631 |
+
"step": 81
|
| 632 |
+
},
|
| 633 |
+
{
|
| 634 |
+
"epoch": 4.32,
|
| 635 |
+
"learning_rate": 0.00033619999999999993,
|
| 636 |
+
"loss": 0.2051,
|
| 637 |
+
"step": 82
|
| 638 |
+
},
|
| 639 |
+
{
|
| 640 |
+
"epoch": 4.37,
|
| 641 |
+
"learning_rate": 0.00034445,
|
| 642 |
+
"loss": 0.2013,
|
| 643 |
+
"step": 83
|
| 644 |
+
},
|
| 645 |
+
{
|
| 646 |
+
"epoch": 4.42,
|
| 647 |
+
"learning_rate": 0.00035279999999999996,
|
| 648 |
+
"loss": 0.1751,
|
| 649 |
+
"step": 84
|
| 650 |
+
},
|
| 651 |
+
{
|
| 652 |
+
"epoch": 4.47,
|
| 653 |
+
"learning_rate": 0.00036124999999999997,
|
| 654 |
+
"loss": 0.1815,
|
| 655 |
+
"step": 85
|
| 656 |
+
},
|
| 657 |
+
{
|
| 658 |
+
"epoch": 4.47,
|
| 659 |
+
"eval_loss": 0.2084398865699768,
|
| 660 |
+
"eval_runtime": 20.6383,
|
| 661 |
+
"eval_samples_per_second": 3.101,
|
| 662 |
+
"eval_steps_per_second": 0.388,
|
| 663 |
+
"step": 85
|
| 664 |
+
},
|
| 665 |
+
{
|
| 666 |
+
"epoch": 4.53,
|
| 667 |
+
"learning_rate": 0.0003698,
|
| 668 |
+
"loss": 0.1252,
|
| 669 |
+
"step": 86
|
| 670 |
+
},
|
| 671 |
+
{
|
| 672 |
+
"epoch": 4.58,
|
| 673 |
+
"learning_rate": 0.00037845,
|
| 674 |
+
"loss": 0.2354,
|
| 675 |
+
"step": 87
|
| 676 |
+
},
|
| 677 |
+
{
|
| 678 |
+
"epoch": 4.63,
|
| 679 |
+
"learning_rate": 0.0003872,
|
| 680 |
+
"loss": 0.1651,
|
| 681 |
+
"step": 88
|
| 682 |
+
},
|
| 683 |
+
{
|
| 684 |
+
"epoch": 4.68,
|
| 685 |
+
"learning_rate": 0.00039605,
|
| 686 |
+
"loss": 0.2001,
|
| 687 |
+
"step": 89
|
| 688 |
+
},
|
| 689 |
+
{
|
| 690 |
+
"epoch": 4.74,
|
| 691 |
+
"learning_rate": 0.00040500000000000003,
|
| 692 |
+
"loss": 0.2164,
|
| 693 |
+
"step": 90
|
| 694 |
+
},
|
| 695 |
+
{
|
| 696 |
+
"epoch": 4.74,
|
| 697 |
+
"eval_loss": 0.21278557181358337,
|
| 698 |
+
"eval_runtime": 20.6294,
|
| 699 |
+
"eval_samples_per_second": 3.102,
|
| 700 |
+
"eval_steps_per_second": 0.388,
|
| 701 |
+
"step": 90
|
| 702 |
+
},
|
| 703 |
+
{
|
| 704 |
+
"epoch": 4.79,
|
| 705 |
+
"learning_rate": 0.00041405000000000006,
|
| 706 |
+
"loss": 0.155,
|
| 707 |
+
"step": 91
|
| 708 |
+
},
|
| 709 |
+
{
|
| 710 |
+
"epoch": 4.84,
|
| 711 |
+
"learning_rate": 0.00042320000000000004,
|
| 712 |
+
"loss": 0.2454,
|
| 713 |
+
"step": 92
|
| 714 |
+
},
|
| 715 |
+
{
|
| 716 |
+
"epoch": 4.89,
|
| 717 |
+
"learning_rate": 0.0004324500000000001,
|
| 718 |
+
"loss": 0.1804,
|
| 719 |
+
"step": 93
|
| 720 |
+
},
|
| 721 |
+
{
|
| 722 |
+
"epoch": 4.95,
|
| 723 |
+
"learning_rate": 0.0004418,
|
| 724 |
+
"loss": 0.2498,
|
| 725 |
+
"step": 94
|
| 726 |
+
},
|
| 727 |
+
{
|
| 728 |
+
"epoch": 5.0,
|
| 729 |
+
"learning_rate": 0.00045125,
|
| 730 |
+
"loss": 0.1412,
|
| 731 |
+
"step": 95
|
| 732 |
+
},
|
| 733 |
+
{
|
| 734 |
+
"epoch": 5.0,
|
| 735 |
+
"eval_loss": 0.2015175074338913,
|
| 736 |
+
"eval_runtime": 20.5945,
|
| 737 |
+
"eval_samples_per_second": 3.108,
|
| 738 |
+
"eval_steps_per_second": 0.388,
|
| 739 |
+
"step": 95
|
| 740 |
+
}
|
| 741 |
+
],
|
| 742 |
+
"logging_steps": 1,
|
| 743 |
+
"max_steps": 95,
|
| 744 |
+
"num_input_tokens_seen": 0,
|
| 745 |
+
"num_train_epochs": 5,
|
| 746 |
+
"save_steps": 19,
|
| 747 |
+
"total_flos": 9751829478899712.0,
|
| 748 |
+
"train_batch_size": 8,
|
| 749 |
+
"trial_name": null,
|
| 750 |
+
"trial_params": null
|
| 751 |
+
}
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:b3cbfc7602d43113740000b58025b6954b5c8c01fef0cd35390c85a0eca64643
|
| 3 |
+
size 6011
|