Hugging Face's logo

C10X
/
v23

Safetensors
qwen3
Model card Files
xet
 Community
C10X commited on
Commit
d85e02a
·
verified ·
1 Parent(s): f3620a8

Upload 7 files

Browse files
Files changed (4) hide show
  1. config.json +8 -32
  2. generation_config.json +1 -1
  3. metadata.json +7 -7
  4. model.safetensors +2 -2
config.json CHANGED
@@ -7,36 +7,12 @@
7
  "bos_token_id": 0,
8
  "dtype": "float32",
9
  "eos_token_id": 4,
10
- "head_dim": 32,
11
  "hidden_act": "silu",
12
- "hidden_size": 64,
13
  "initializer_range": 0.02,
14
- "intermediate_size": 256,
15
  "layer_types": [
16
- "full_attention",
17
- "full_attention",
18
- "full_attention",
19
- "full_attention",
20
- "full_attention",
21
- "full_attention",
22
- "full_attention",
23
- "full_attention",
24
- "full_attention",
25
- "full_attention",
26
- "full_attention",
27
- "full_attention",
28
- "full_attention",
29
- "full_attention",
30
- "full_attention",
31
- "full_attention",
32
- "full_attention",
33
- "full_attention",
34
- "full_attention",
35
- "full_attention",
36
- "full_attention",
37
- "full_attention",
38
- "full_attention",
39
- "full_attention",
40
  "full_attention",
41
  "full_attention",
42
  "full_attention",
@@ -47,11 +23,11 @@
47
  "full_attention"
48
  ],
49
  "max_position_embeddings": 2048,
50
- "max_window_layers": 32,
51
  "model_type": "qwen3",
52
- "num_attention_heads": 4,
53
- "num_hidden_layers": 32,
54
- "num_key_value_heads": 2,
55
  "pad_token_id": 1,
56
  "rms_norm_eps": 1e-06,
57
  "rope_scaling": null,
@@ -59,7 +35,7 @@
59
  "sliding_window": null,
60
  "tie_word_embeddings": true,
61
  "transformers_version": "4.57.6",
62
- "use_cache": true,
63
  "use_sliding_window": false,
64
  "vocab_size": 2048
65
  }
 
7
  "bos_token_id": 0,
8
  "dtype": "float32",
9
  "eos_token_id": 4,
10
+ "head_dim": 64,
11
  "hidden_act": "silu",
12
+ "hidden_size": 512,
13
  "initializer_range": 0.02,
14
+ "intermediate_size": 2048,
15
  "layer_types": [
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
16
  "full_attention",
17
  "full_attention",
18
  "full_attention",
 
23
  "full_attention"
24
  ],
25
  "max_position_embeddings": 2048,
26
+ "max_window_layers": 8,
27
  "model_type": "qwen3",
28
+ "num_attention_heads": 8,
29
+ "num_hidden_layers": 8,
30
+ "num_key_value_heads": 4,
31
  "pad_token_id": 1,
32
  "rms_norm_eps": 1e-06,
33
  "rope_scaling": null,
 
35
  "sliding_window": null,
36
  "tie_word_embeddings": true,
37
  "transformers_version": "4.57.6",
38
+ "use_cache": false,
39
  "use_sliding_window": false,
40
  "vocab_size": 2048
41
  }
generation_config.json CHANGED
@@ -3,5 +3,5 @@
3
  "bos_token_id": 0,
4
  "eos_token_id": 4,
5
  "pad_token_id": 1,
6
- "use_cache": true
7
  }
 
3
  "bos_token_id": 0,
4
  "eos_token_id": 4,
5
  "pad_token_id": 1,
6
+ "use_cache": false
7
  }
metadata.json CHANGED
@@ -1,13 +1,13 @@
1
  {
2
- "preset": "qwen3-32layer-c10x",
3
  "family": "qwen3",
4
  "source": "inline-preset",
5
- "output_dir": "D:\\Qwen3-80m-tinystories-A\\workspace\\outputs\\qwen3-32layer-c10x",
6
- "parameters": 2496576,
7
  "vocab_size": 2048,
8
- "hidden_size": 64,
9
- "num_hidden_layers": 32,
10
- "num_attention_heads": 4,
11
- "num_key_value_heads": 2,
12
  "rope_theta": 10000.0
13
  }
 
1
  {
2
+ "preset": "qwen3-16layer-c10x",
3
  "family": "qwen3",
4
  "source": "inline-preset",
5
+ "output_dir": "D:\\Qwen3-80m-tinystories-A\\workspace\\outputs\\qwen3-16layer-c10x",
6
+ "parameters": 32515584,
7
  "vocab_size": 2048,
8
+ "hidden_size": 512,
9
+ "num_hidden_layers": 8,
10
+ "num_attention_heads": 8,
11
+ "num_key_value_heads": 4,
12
  "rope_theta": 10000.0
13
  }
model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:e223f0d607fdc649ddfa5bca27b1675e6ee75bdd65198b00450abe76cba4d283
3
- size 10023688
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:cdfe3633df9ac5c6ead66960a86a77e2999797dd40686e2caf4e856d5133c3eb
3
+ size 130072120