Upload folder using huggingface_hub
#1
by
at676
- opened
- config.json +55 -0
- generation_config.json +12 -0
- model-00001-of-00007.safetensors +3 -0
- model-00002-of-00007.safetensors +3 -0
- model-00003-of-00007.safetensors +3 -0
- model-00004-of-00007.safetensors +3 -0
- model-00005-of-00007.safetensors +3 -0
- model-00006-of-00007.safetensors +3 -0
- model-00007-of-00007.safetensors +3 -0
- model.safetensors.index.json +0 -0
config.json
ADDED
@@ -0,0 +1,55 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"_name_or_path": "meta-llama/Llama-3.3-70B-Instruct",
|
3 |
+
"architectures": [
|
4 |
+
"LlamaForCausalLM"
|
5 |
+
],
|
6 |
+
"attention_bias": false,
|
7 |
+
"attention_dropout": 0.0,
|
8 |
+
"bos_token_id": 128000,
|
9 |
+
"eos_token_id": [
|
10 |
+
128001,
|
11 |
+
128008,
|
12 |
+
128009
|
13 |
+
],
|
14 |
+
"head_dim": 128,
|
15 |
+
"hidden_act": "silu",
|
16 |
+
"hidden_size": 8192,
|
17 |
+
"initializer_range": 0.02,
|
18 |
+
"intermediate_size": 28672,
|
19 |
+
"max_position_embeddings": 131072,
|
20 |
+
"mlp_bias": false,
|
21 |
+
"model_type": "llama",
|
22 |
+
"num_attention_heads": 64,
|
23 |
+
"num_hidden_layers": 80,
|
24 |
+
"num_key_value_heads": 8,
|
25 |
+
"pretraining_tp": 1,
|
26 |
+
"quip_params": {
|
27 |
+
"K": 3,
|
28 |
+
"L": 16,
|
29 |
+
"V": 2,
|
30 |
+
"codebook": "bitshift",
|
31 |
+
"codebook_version": 0,
|
32 |
+
"decode_mode": "quantlut_sym",
|
33 |
+
"skip_list": [
|
34 |
+
"0_v"
|
35 |
+
],
|
36 |
+
"split_for_tp": false,
|
37 |
+
"td_x": 16,
|
38 |
+
"td_y": 16,
|
39 |
+
"tlut_bits": 9
|
40 |
+
},
|
41 |
+
"rms_norm_eps": 1e-05,
|
42 |
+
"rope_scaling": {
|
43 |
+
"factor": 8.0,
|
44 |
+
"high_freq_factor": 4.0,
|
45 |
+
"low_freq_factor": 1.0,
|
46 |
+
"original_max_position_embeddings": 8192,
|
47 |
+
"rope_type": "llama3"
|
48 |
+
},
|
49 |
+
"rope_theta": 500000.0,
|
50 |
+
"tie_word_embeddings": false,
|
51 |
+
"torch_dtype": "bfloat16",
|
52 |
+
"transformers_version": "4.46.2",
|
53 |
+
"use_cache": true,
|
54 |
+
"vocab_size": 128256
|
55 |
+
}
|
generation_config.json
ADDED
@@ -0,0 +1,12 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"bos_token_id": 128000,
|
3 |
+
"do_sample": true,
|
4 |
+
"eos_token_id": [
|
5 |
+
128001,
|
6 |
+
128008,
|
7 |
+
128009
|
8 |
+
],
|
9 |
+
"temperature": 0.6,
|
10 |
+
"top_p": 0.9,
|
11 |
+
"transformers_version": "4.46.2"
|
12 |
+
}
|
model-00001-of-00007.safetensors
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:3bfd90cdcf248fdf1736cb3ea8da9cd73aba8ca4f5034135b7642c3133c15b38
|
3 |
+
size 4919418488
|
model-00002-of-00007.safetensors
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:5fd2264e2fafd911a5e455d39a624414d444a4b9a8a241986a5df422e1755375
|
3 |
+
size 4966090984
|
model-00003-of-00007.safetensors
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:f764ccbdae1f07be0eeb559067c4b82d287f49d4de1ddfa0a877341a464c34d7
|
3 |
+
size 4997540024
|
model-00004-of-00007.safetensors
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:2debae60ac649cdc7b65dabd9f753f36649c1a201ecd2c7405b944232e9fef35
|
3 |
+
size 4966091040
|
model-00005-of-00007.safetensors
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:d442a71828d431a62d474eee8c01e06b8054ccc33d9e1ced431a96b5e28178e2
|
3 |
+
size 4997540024
|
model-00006-of-00007.safetensors
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:44802abe6eabbcf71d7a5734830d57d388aa1e71ad0da6c84e2ae843f46ae40c
|
3 |
+
size 2980887216
|
model-00007-of-00007.safetensors
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:b0436c98c9b4f78e61f6769e588e0853179b970e2a084ae2886e2daebf1ce13d
|
3 |
+
size 2101346432
|
model.safetensors.index.json
ADDED
The diff for this file is too large to render.
See raw diff
|
|