danielhanchen
commited on
Commit
•
4efb21a
1
Parent(s):
7921828
Upload config
Browse files- config.json +3 -1
config.json
CHANGED
@@ -1,5 +1,5 @@
|
|
1 |
{
|
2 |
-
"_name_or_path": "
|
3 |
"architectures": [
|
4 |
"LlamaForCausalLM"
|
5 |
],
|
@@ -17,6 +17,7 @@
|
|
17 |
"num_attention_heads": 128,
|
18 |
"num_hidden_layers": 126,
|
19 |
"num_key_value_heads": 8,
|
|
|
20 |
"pretraining_tp": 1,
|
21 |
"quantization_config": {
|
22 |
"_load_in_4bit": true,
|
@@ -45,6 +46,7 @@
|
|
45 |
"tie_word_embeddings": false,
|
46 |
"torch_dtype": "bfloat16",
|
47 |
"transformers_version": "4.44.2",
|
|
|
48 |
"use_cache": true,
|
49 |
"vocab_size": 128256
|
50 |
}
|
|
|
1 |
{
|
2 |
+
"_name_or_path": "unsloth/Meta-Llama-3.1-405B",
|
3 |
"architectures": [
|
4 |
"LlamaForCausalLM"
|
5 |
],
|
|
|
17 |
"num_attention_heads": 128,
|
18 |
"num_hidden_layers": 126,
|
19 |
"num_key_value_heads": 8,
|
20 |
+
"pad_token_id": 128004,
|
21 |
"pretraining_tp": 1,
|
22 |
"quantization_config": {
|
23 |
"_load_in_4bit": true,
|
|
|
46 |
"tie_word_embeddings": false,
|
47 |
"torch_dtype": "bfloat16",
|
48 |
"transformers_version": "4.44.2",
|
49 |
+
"unsloth_version": "2024.9",
|
50 |
"use_cache": true,
|
51 |
"vocab_size": 128256
|
52 |
}
|