devadigaprathamesh commited on
Commit
516a4ed
·
verified ·
1 Parent(s): a4e4f12

Update config.json

Browse files
Files changed (1) hide show
  1. config.json +26 -62
config.json CHANGED
@@ -1,64 +1,28 @@
1
  {
2
- "app_name": "qwen-law-finetuning",
3
- "model": {
4
- "base_model_id": "Qwen/Qwen2.5-7B",
5
- "max_seq_length": 4096,
6
- "quantization": {
7
- "load_in_4bit": true,
8
- "attn_implementation": "flash_attention_2",
9
- "dtype": "bfloat16"
10
- }
11
- },
12
- "peft_config": {
13
- "r": 32,
14
- "target_modules": [
15
- "q_proj",
16
- "k_proj",
17
- "v_proj",
18
- "o_proj",
19
- "gate_proj",
20
- "up_proj",
21
- "down_proj"
22
- ],
23
- "lora_alpha": 64,
24
- "lora_dropout": 0.05,
25
- "bias": "none",
26
- "use_gradient_checkpointing": "unsloth"
27
- },
28
- "training": {
29
- "dataset": "viber1/indian-law-dataset",
30
- "output_dir": "/data/JurisQwen",
31
- "checkpoint_dir": "/data/checkpoints",
32
- "num_train_epochs": 3,
33
- "per_device_train_batch_size": 16,
34
- "gradient_accumulation_steps": 2,
35
- "optimizer": "adamw_8bit",
36
- "learning_rate": 2e-4,
37
- "weight_decay": 0.001,
38
- "lr_scheduler_type": "cosine",
39
- "warmup_ratio": 0.1,
40
- "precision": {
41
- "bf16": true,
42
- "fp16": false,
43
- "tf32": true
44
- }
45
- },
46
- "inference": {
47
- "max_new_tokens": 512
48
- },
49
- "compute": {
50
- "gpu": "A100-40GB",
51
- "training_timeout_hours": 5,
52
- "inference_timeout_minutes": 10
53
- },
54
- "prompt_template": {
55
- "user_prefix": "<|im_start|>user\n",
56
- "user_suffix": "<|im_end|>",
57
- "assistant_prefix": "<|im_start|>assistant\n",
58
- "assistant_suffix": "<|im_end|>"
59
- },
60
- "volumes": {
61
- "path": "/data",
62
- "name": "finetune-volume"
63
- }
64
  }
 
1
  {
2
+ "architectures": [
3
+ "Qwen2ForCausalLM"
4
+ ],
5
+ "attention_dropout": 0.0,
6
+ "bos_token_id": 151643,
7
+ "eos_token_id": 151643,
8
+ "hidden_act": "silu",
9
+ "hidden_size": 3584,
10
+ "initializer_range": 0.02,
11
+ "intermediate_size": 18944,
12
+ "max_position_embeddings": 131072,
13
+ "max_window_layers": 28,
14
+ "model_type": "qwen2",
15
+ "num_attention_heads": 28,
16
+ "num_hidden_layers": 28,
17
+ "num_key_value_heads": 4,
18
+ "rms_norm_eps": 1e-06,
19
+ "rope_theta": 1000000.0,
20
+ "sliding_window": 131072,
21
+ "tie_word_embeddings": false,
22
+ "torch_dtype": "bfloat16",
23
+ "transformers_version": "4.40.1",
24
+ "use_cache": true,
25
+ "use_mrope": false,
26
+ "use_sliding_window": false,
27
+ "vocab_size": 152064
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
28
  }