File size: 2,640 Bytes
cea5907
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
{
  "experiment_key": "base",
  "save_safetensors": true,
  "max_shard_size": "10GB",
  "local_rank": 0,
  "use_gradient_checkpointing": true,
  "trainer_key": "lm",
  "force_fp32": false,
  "force_fp16": false,
  "from_gptq": false,
  "huggingface_hub_token": null,
  "single_gpu": null,
  "master_port": 9994,
  "deepspeed_stage": null,
  "deepspeed_config_path": null,
  "fsdp_strategy": "",
  "fsdp_offload": true,
  "seed": 42,
  "stabilize": false,
  "norm_fp32": false,
  "path_to_env_file": "./.env",
  "prepare_dataset": true,
  "lora_hub_model_id": null,
  "lora_model_local_path": null,
  "fused_model_local_path": null,
  "fuse_after_training": false,
  "quantization_dataset_id": null,
  "quantization_max_samples": 1024,
  "quantized_model_path": "./quantized_model/",
  "quantized_hub_model_id": null,
  "quantized_hub_private_repo": true,
  "dataset_key": "soda",
  "train_local_path_to_data": "./train.jsonl",
  "eval_local_path_to_data": null,
  "shuffle": true,
  "max_eval_samples": 1000,
  "add_eval_to_train_if_no_path": false,
  "tokenizer_name_or_path": null,
  "tokenizer_use_fast": null,
  "tokenizer_padding_side": null,
  "collator_key": "lm",
  "max_length": 2048,
  "model_name_or_path": "mistralai/Mistral-7B-v0.1",
  "push_to_hub_bos_add_bos_token": false,
  "use_flash_attention_2": false,
  "trust_remote_code": true,
  "device_map": null,
  "prepare_model_for_kbit_training": true,
  "offload_folder": null,
  "load_in_8bit": false,
  "load_in_4bit": true,
  "llm_int8_threshold": 6.0,
  "llm_int8_has_fp16_weight": true,
  "bnb_4bit_use_double_quant": true,
  "bnb_4bit_quant_type": "nf4",
  "bnb_quantize_after_model_init": false,
  "gptq_bits": 4,
  "gptq_group_size": 128,
  "gptq_disable_exllama": true,
  "apply_lora": true,
  "lora_rank": 8,
  "lora_alpha": 32,
  "lora_dropout": 0.1,
  "raw_lora_target_modules": "all",
  "output_dir": "./outputs/",
  "per_device_train_batch_size": 2,
  "do_eval": false,
  "per_device_eval_batch_size": null,
  "gradient_accumulation_steps": 2,
  "eval_accumulation_steps": null,
  "eval_delay": 0,
  "eval_steps": 1000,
  "warmup_steps": 5,
  "max_steps": 501,
  "num_train_epochs": 2,
  "learning_rate": 0.0002,
  "max_grad_norm": 1.0,
  "weight_decay": 0.001,
  "label_smoothing_factor": 0.0,
  "logging_steps": 1,
  "save_steps": 25,
  "save_total_limit": 1,
  "optim": "paged_adamw_8bit",
  "push_to_hub": true,
  "hub_model_id": "nutorbit/mistral-7b-xllm-lora",
  "hub_private_repo": false,
  "neftune_noise_alpha": null,
  "project_name": null,
  "report_to_wandb": true,
  "wandb_api_key": null,
  "wandb_project": "numeval",
  "wandb_entity": "data-x"
}