kharshita590 commited on
Commit
b7f5616
·
verified ·
1 Parent(s): de74ac7

Update config.json

Browse files
Files changed (1) hide show
  1. config.json +41 -2
config.json CHANGED
@@ -1,3 +1,42 @@
1
  {
2
- "name_or_path":"career_guidance_model_llama"
3
- }
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
  {
2
+ "name_or_path":"career_guidance_model_llama",
3
+ "architectures": [
4
+ "LlamaForCausalLM"
5
+ ],
6
+ "model_type": "llama",
7
+ "max_seq_length": 1024,
8
+ "r": 8,
9
+ "lora_alpha": 16,
10
+ "lora_dropout": 0,
11
+ "gradient_checkpointing": true,
12
+ "load_in_4bit": true,
13
+ "use_cache": true,
14
+ "vocab_size": 32000,
15
+ "tokenizer_class": "LlamaTokenizer",
16
+ "pad_token_id": 0,
17
+ "bos_token_id": 1,
18
+ "eos_token_id": 2,
19
+ "hidden_size": 4096,
20
+ "num_attention_heads": 32,
21
+ "num_hidden_layers": 32,
22
+ "intermediate_size": 11008,
23
+ "hidden_act": "silu",
24
+ "initializer_range": 0.02,
25
+ "torch_dtype": "float16",
26
+ "fp16": true,
27
+ "bf16": false,
28
+ "attention_probs_dropout_prob": 0.1,
29
+ "hidden_dropout_prob": 0.1,
30
+ "learning_rate": 0.0001,
31
+ "weight_decay": 0.05,
32
+ "lr_scheduler_type": "cosine",
33
+ "warmup_ratio": 0.03,
34
+ "optimizer_type": "adamw_8bit",
35
+ "task_specific_params": {
36
+ "text-generation": {
37
+ "temperature": 0.7,
38
+ "top_p": 0.9,
39
+ "max_new_tokens": 128
40
+ }
41
+ }
42
+ }