danielhanchen commited on
Commit
4864c68
1 Parent(s): b973fd3

Upload config

Browse files
Files changed (1) hide show
  1. config.json +3 -1
config.json CHANGED
@@ -1,5 +1,5 @@
1
  {
2
- "_name_or_path": "meta-llama/Meta-Llama-3-8B-Instruct",
3
  "architectures": [
4
  "LlamaForCausalLM"
5
  ],
@@ -17,6 +17,7 @@
17
  "num_attention_heads": 32,
18
  "num_hidden_layers": 32,
19
  "num_key_value_heads": 8,
 
20
  "pretraining_tp": 1,
21
  "quantization_config": {
22
  "_load_in_4bit": true,
@@ -39,6 +40,7 @@
39
  "tie_word_embeddings": false,
40
  "torch_dtype": "bfloat16",
41
  "transformers_version": "4.44.0",
 
42
  "use_cache": true,
43
  "vocab_size": 128256
44
  }
 
1
  {
2
+ "_name_or_path": "unsloth/llama-3-8b-Instruct",
3
  "architectures": [
4
  "LlamaForCausalLM"
5
  ],
 
17
  "num_attention_heads": 32,
18
  "num_hidden_layers": 32,
19
  "num_key_value_heads": 8,
20
+ "pad_token_id": 128255,
21
  "pretraining_tp": 1,
22
  "quantization_config": {
23
  "_load_in_4bit": true,
 
40
  "tie_word_embeddings": false,
41
  "torch_dtype": "bfloat16",
42
  "transformers_version": "4.44.0",
43
+ "unsloth_version": "2024.8",
44
  "use_cache": true,
45
  "vocab_size": 128256
46
  }