Hugging Face's logo

haydso
/
PrimeIntellect-Reverse-Text-SFT-ONNX

ONNX
qwen3
Model card Files
xet
Community
haydso commited on
Commit
66747a7
·
verified ·
1 Parent(s): 9427e83

Upload config.json

Browse files
Files changed (1) hide show
  1. config.json +58 -50
config.json CHANGED
@@ -1,52 +1,60 @@
1
  {
2
- "model": {
3
- "bos_token_id": 151643,
4
- "context_length": 40960,
5
- "decoder": {
6
- "session_options": {
7
- "log_id": "onnxruntime-genai",
8
- "provider_options": []
9
- },
10
- "filename": "model.onnx",
11
- "head_size": 128,
12
- "hidden_size": 1024,
13
- "inputs": {
14
- "input_ids": "input_ids",
15
- "attention_mask": "attention_mask",
16
- "past_key_names": "past_key_values.%d.key",
17
- "past_value_names": "past_key_values.%d.value"
18
- },
19
- "outputs": {
20
- "logits": "logits",
21
- "present_key_names": "present.%d.key",
22
- "present_value_names": "present.%d.value"
23
- },
24
- "num_attention_heads": 16,
25
- "num_hidden_layers": 28,
26
- "num_key_value_heads": 8
27
- },
28
- "eos_token_id": [
29
- 151645,
30
- 151643
31
- ],
32
- "pad_token_id": 151643,
33
- "type": "qwen3",
34
- "vocab_size": 151936
35
- },
36
- "search": {
37
- "diversity_penalty": 0.0,
38
- "do_sample": true,
39
- "early_stopping": true,
40
- "length_penalty": 1.0,
41
- "max_length": 40960,
42
- "min_length": 0,
43
- "no_repeat_ngram_size": 0,
44
- "num_beams": 1,
45
- "num_return_sequences": 1,
46
- "past_present_share_buffer": true,
47
- "repetition_penalty": 1.0,
48
- "temperature": 0.6,
49
- "top_k": 20,
50
- "top_p": 0.95
51
- }
 
 
 
 
 
 
 
 
52
  }
 
1
  {
2
+ "architectures": [
3
+ "Qwen3ForCausalLM"
4
+ ],
5
+ "attention_bias": false,
6
+ "attention_dropout": 0.0,
7
+ "bos_token_id": 151643,
8
+ "eos_token_id": 151645,
9
+ "head_dim": 128,
10
+ "hidden_act": "silu",
11
+ "hidden_size": 1024,
12
+ "initializer_range": 0.02,
13
+ "intermediate_size": 3072,
14
+ "layer_types": [
15
+ "full_attention",
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"
43
+ ],
44
+ "max_position_embeddings": 40960,
45
+ "max_window_layers": 28,
46
+ "model_type": "qwen3",
47
+ "num_attention_heads": 16,
48
+ "num_hidden_layers": 28,
49
+ "num_key_value_heads": 8,
50
+ "rms_norm_eps": 1e-06,
51
+ "rope_scaling": null,
52
+ "rope_theta": 1000000,
53
+ "sliding_window": null,
54
+ "tie_word_embeddings": true,
55
+ "torch_dtype": "float32",
56
+ "transformers_version": "4.54.0",
57
+ "use_cache": false,
58
+ "use_sliding_window": false,
59
+ "vocab_size": 151936
60
  }