HuggingSara
commited on
Commit
•
c2a63d9
1
Parent(s):
5a066c1
Upload 7 files
Browse files- config.json +30 -0
- generation_config.json +7 -0
- pytorch_model-00001-of-00019.bin +3 -0
- pytorch_model-00002-of-00019.bin +3 -0
- pytorch_model-00003-of-00019.bin +3 -0
- pytorch_model-00004-of-00019.bin +3 -0
- pytorch_model-00005-of-00019.bin +3 -0
config.json
ADDED
@@ -0,0 +1,30 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"_name_or_path": "mistralai/Mixtral-8x7B-v0.1",
|
3 |
+
"architectures": [
|
4 |
+
"MixtralForCausalLM"
|
5 |
+
],
|
6 |
+
"attention_dropout": 0.0,
|
7 |
+
"bos_token_id": 1,
|
8 |
+
"eos_token_id": 2,
|
9 |
+
"hidden_act": "silu",
|
10 |
+
"hidden_size": 4096,
|
11 |
+
"initializer_range": 0.02,
|
12 |
+
"intermediate_size": 14336,
|
13 |
+
"max_position_embeddings": 32768,
|
14 |
+
"model_type": "mixtral",
|
15 |
+
"num_attention_heads": 32,
|
16 |
+
"num_experts_per_tok": 2,
|
17 |
+
"num_hidden_layers": 32,
|
18 |
+
"num_key_value_heads": 8,
|
19 |
+
"num_local_experts": 8,
|
20 |
+
"output_router_logits": true,
|
21 |
+
"rms_norm_eps": 1e-05,
|
22 |
+
"rope_theta": 1000000.0,
|
23 |
+
"router_aux_loss_coef": 0.02,
|
24 |
+
"sliding_window": null,
|
25 |
+
"tie_word_embeddings": false,
|
26 |
+
"torch_dtype": "bfloat16",
|
27 |
+
"transformers_version": "4.37.0",
|
28 |
+
"use_cache": false,
|
29 |
+
"vocab_size": 32000
|
30 |
+
}
|
generation_config.json
ADDED
@@ -0,0 +1,7 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"_from_model_config": true,
|
3 |
+
"bos_token_id": 1,
|
4 |
+
"do_sample": true,
|
5 |
+
"eos_token_id": 2,
|
6 |
+
"transformers_version": "4.37.0"
|
7 |
+
}
|
pytorch_model-00001-of-00019.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:8ad342adb98aed82bdc4d5604d0c8a06b4095865f3c24072289beebb9a25a1df
|
3 |
+
size 4892821288
|
pytorch_model-00002-of-00019.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:eb252da0231413672844511d6af7eefc0452497eb2d9946995fe7704f7c13045
|
3 |
+
size 4983016612
|
pytorch_model-00003-of-00019.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:d6cadd1c4c0d3117f7743f2b7e08f280f11a5c5a712fb881f1a0953c8d18e1cb
|
3 |
+
size 4983016696
|
pytorch_model-00004-of-00019.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:a999124b374a5f16aadc6662f32c1d7c8a7bc8c958742f7dca071499ac1df176
|
3 |
+
size 4899046246
|
pytorch_model-00005-of-00019.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:9beedfe055724d09a83d28131aa14f9bc8bb61c93658ce7e90d1654c60f0e509
|
3 |
+
size 4983016648
|