aboutaleb commited on
Commit
dd48824
1 Parent(s): 20dac0e

Upload config.json

Browse files
Files changed (1) hide show
  1. config.json +11 -11
config.json CHANGED
@@ -1,18 +1,18 @@
1
  {
2
- "architectures": ["LLaMAModel"], // Adjust based on your model's architecture
3
  "model_type": "llama",
4
- "hidden_size": 4096, // Example value, adjust as needed
5
- "num_hidden_layers": 30, // Example value, adjust as needed
6
- "num_attention_heads": 16, // Example value, adjust as needed
7
- "intermediate_size": 11008, // Example value, adjust as needed
8
- "max_position_embeddings": 8192, // Max sequence length, as inferred from tokenizer_config.json
9
- "vocab_size": 128256, // Adjust based on your tokenizer's actual vocabulary size
10
- "activation_function": "gelu_new", // Assuming typical activation function, adjust as needed
11
  "layer_norm_eps": 1e-5,
12
  "initializer_range": 0.02,
13
- "bos_token_id": 128000, // From tokenizer_config.json
14
- "eos_token_id": 128001, // From tokenizer_config.json
15
- "pad_token_id": 128255, // From tokenizer_config.json
16
  "special_tokens_map": {
17
  "bos_token": "<|begin_of_text|>",
18
  "eos_token": "<|end_of_text|>",
 
1
  {
2
+ "architectures": ["LLaMAModel"],
3
  "model_type": "llama",
4
+ "hidden_size": 4096,
5
+ "num_hidden_layers": 30,
6
+ "num_attention_heads": 16,
7
+ "intermediate_size": 11008,
8
+ "max_position_embeddings": 8192,
9
+ "vocab_size": 128256,
10
+ "activation_function": "gelu_new",
11
  "layer_norm_eps": 1e-5,
12
  "initializer_range": 0.02,
13
+ "bos_token_id": 128000,
14
+ "eos_token_id": 128001,
15
+ "pad_token_id": 128255,
16
  "special_tokens_map": {
17
  "bos_token": "<|begin_of_text|>",
18
  "eos_token": "<|end_of_text|>",