atharvapawar
commited on
Commit
•
91bc2d9
1
Parent(s):
6de9bc8
v2
Browse files- config.json +21 -0
config.json
CHANGED
@@ -1,5 +1,26 @@
|
|
1 |
{
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2 |
"model_type": "llama",
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
3 |
"auto_mapping": null,
|
4 |
"base_model_name_or_path": "TinyPixel/Llama-2-7B-bf16-sharded",
|
5 |
"bias": "none",
|
|
|
1 |
{
|
2 |
+
"_name_or_path": "TheBloke/Llama-2-7B-fp16",
|
3 |
+
"architectures": ["LlamaForCausalLM"],
|
4 |
+
"bos_token_id": 1,
|
5 |
+
"eos_token_id": 2,
|
6 |
+
"hidden_act": "silu",
|
7 |
+
"hidden_size": 4096,
|
8 |
+
"initializer_range": 0.02,
|
9 |
+
"intermediate_size": 11008,
|
10 |
+
"max_position_embeddings": 2048,
|
11 |
"model_type": "llama",
|
12 |
+
"num_attention_heads": 32,
|
13 |
+
"num_hidden_layers": 32,
|
14 |
+
"num_key_value_heads": 32,
|
15 |
+
"pad_token_id": 0,
|
16 |
+
"pretraining_tp": 1,
|
17 |
+
"rms_norm_eps": 1e-05,
|
18 |
+
"rope_scaling": null,
|
19 |
+
"tie_word_embeddings": false,
|
20 |
+
"torch_dtype": "bfloat16",
|
21 |
+
"transformers_version": "4.31.0",
|
22 |
+
"use_cache": true,
|
23 |
+
"vocab_size": 32000,
|
24 |
"auto_mapping": null,
|
25 |
"base_model_name_or_path": "TinyPixel/Llama-2-7B-bf16-sharded",
|
26 |
"bias": "none",
|