Adding model files
Browse files- config.json +2 -2
- generation_config.json +8 -0
- model-00001-of-00002.safetensors +3 -0
- model-00002-of-00002.safetensors +3 -0
- model.safetensors.index.json +0 -0
config.json
CHANGED
@@ -9,7 +9,7 @@
|
|
9 |
"eos_token_id": 1,
|
10 |
"head_dim": 256,
|
11 |
"hidden_act": "gelu",
|
12 |
-
"hidden_activation":
|
13 |
"hidden_size": 3072,
|
14 |
"initializer_range": 0.02,
|
15 |
"intermediate_size": 24576,
|
@@ -37,7 +37,7 @@
|
|
37 |
"rms_norm_eps": 1e-06,
|
38 |
"rope_scaling": null,
|
39 |
"rope_theta": 10000.0,
|
40 |
-
"torch_dtype": "
|
41 |
"transformers_version": "4.44.2",
|
42 |
"unsloth_version": "2024.9",
|
43 |
"use_cache": true,
|
|
|
9 |
"eos_token_id": 1,
|
10 |
"head_dim": 256,
|
11 |
"hidden_act": "gelu",
|
12 |
+
"hidden_activation": "gelu_pytorch_tanh",
|
13 |
"hidden_size": 3072,
|
14 |
"initializer_range": 0.02,
|
15 |
"intermediate_size": 24576,
|
|
|
37 |
"rms_norm_eps": 1e-06,
|
38 |
"rope_scaling": null,
|
39 |
"rope_theta": 10000.0,
|
40 |
+
"torch_dtype": "float16",
|
41 |
"transformers_version": "4.44.2",
|
42 |
"unsloth_version": "2024.9",
|
43 |
"use_cache": true,
|
generation_config.json
ADDED
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"_from_model_config": true,
|
3 |
+
"bos_token_id": 2,
|
4 |
+
"eos_token_id": 1,
|
5 |
+
"max_length": 8192,
|
6 |
+
"pad_token_id": 0,
|
7 |
+
"transformers_version": "4.44.2"
|
8 |
+
}
|
model-00001-of-00002.safetensors
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:81d3c81c91f8f86e14d4324ddc4a6bd72e7bd96cee16271d6ee1708c83678593
|
3 |
+
size 4999604049
|
model-00002-of-00002.safetensors
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:e63cd2bea6f2b47a3626e19d4fcd47dfa91e97f80cd4162720ebd3a6a40eebec
|
3 |
+
size 572543803
|
model.safetensors.index.json
ADDED
The diff for this file is too large to render.
See raw diff
|
|