first commit
Browse files- config.json +5 -5
- quant_config.json +1 -1
- quantized_weights.pt +1 -1
config.json
CHANGED
@@ -1,10 +1,10 @@
|
|
1 |
{
|
2 |
-
"_name_or_path": "/
|
3 |
"architectures": [
|
4 |
-
"
|
5 |
],
|
6 |
-
"bos_token_id":
|
7 |
-
"eos_token_id":
|
8 |
"hidden_act": "silu",
|
9 |
"hidden_size": 4096,
|
10 |
"initializer_range": 0.02,
|
@@ -14,7 +14,7 @@
|
|
14 |
"model_type": "llama",
|
15 |
"num_attention_heads": 32,
|
16 |
"num_hidden_layers": 32,
|
17 |
-
"pad_token_id":
|
18 |
"rms_norm_eps": 1e-06,
|
19 |
"tie_word_embeddings": false,
|
20 |
"torch_dtype": "float16",
|
|
|
1 |
{
|
2 |
+
"_name_or_path": "huggyllama/llama-7b",
|
3 |
"architectures": [
|
4 |
+
"LlamaForCausalLM"
|
5 |
],
|
6 |
+
"bos_token_id": 1,
|
7 |
+
"eos_token_id": 2,
|
8 |
"hidden_act": "silu",
|
9 |
"hidden_size": 4096,
|
10 |
"initializer_range": 0.02,
|
|
|
14 |
"model_type": "llama",
|
15 |
"num_attention_heads": 32,
|
16 |
"num_hidden_layers": 32,
|
17 |
+
"pad_token_id": 0,
|
18 |
"rms_norm_eps": 1e-06,
|
19 |
"tie_word_embeddings": false,
|
20 |
"torch_dtype": "float16",
|
quant_config.json
CHANGED
@@ -7,5 +7,5 @@
|
|
7 |
"percdamp": 0.01,
|
8 |
"seed": 0,
|
9 |
"nearest": false,
|
10 |
-
"save": "
|
11 |
}
|
|
|
7 |
"percdamp": 0.01,
|
8 |
"seed": 0,
|
9 |
"nearest": false,
|
10 |
+
"save": "./llama-7b-4bit"
|
11 |
}
|
quantized_weights.pt
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
size 4020703149
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:66d8d284a5d786e489a4b3cda35510fa376c31005a67d4e5c92fe0d6596b5770
|
3 |
size 4020703149
|