arthurrpp commited on
Commit
2fb004d
·
verified ·
1 Parent(s): aede3fb

Upload folder using huggingface_hub

Browse files
Files changed (1) hide show
  1. config.json +86 -22
config.json CHANGED
@@ -1,31 +1,95 @@
1
  {
2
- "architectures": [
3
- "Qwen3ForCausalLM"
4
- ],
5
- "attention_bias": false,
6
- "attention_dropout": 0.0,
7
- "bos_token_id": 151643,
8
- "eos_token_id": 151643,
9
- "head_dim": 128,
10
- "hidden_act": "silu",
11
  "hidden_size": 1024,
12
- "initializer_range": 0.02,
13
  "intermediate_size": 3072,
14
- "max_position_embeddings": 32768,
15
- "max_window_layers": 28,
16
- "model_type": "qwen3",
17
- "num_attention_heads": 16,
18
  "num_hidden_layers": 28,
 
 
 
 
19
  "num_key_value_heads": 8,
 
 
 
20
  "rms_norm_eps": 1e-06,
21
- "rope_scaling": null,
22
  "rope_theta": 1000000,
23
- "sliding_window": null,
24
- "tie_word_embeddings": true,
 
 
 
 
 
25
  "torch_dtype": "float32",
26
- "transformers_version": "4.51.3",
27
- "use_cache": true,
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
28
  "use_flash_attention_2": true,
29
- "use_sliding_window": false,
30
- "vocab_size": 151936
31
- }
 
 
 
 
 
 
 
 
1
  {
2
+ "vocab_size": 151936,
3
+ "max_position_embeddings": 32768,
 
 
 
 
 
 
 
4
  "hidden_size": 1024,
 
5
  "intermediate_size": 3072,
 
 
 
 
6
  "num_hidden_layers": 28,
7
+ "num_attention_heads": 16,
8
+ "use_sliding_window": false,
9
+ "sliding_window": null,
10
+ "max_window_layers": 28,
11
  "num_key_value_heads": 8,
12
+ "head_dim": 128,
13
+ "hidden_act": "silu",
14
+ "initializer_range": 0.02,
15
  "rms_norm_eps": 1e-06,
16
+ "use_cache": true,
17
  "rope_theta": 1000000,
18
+ "rope_scaling": null,
19
+ "attention_bias": false,
20
+ "attention_dropout": 0.0,
21
+ "return_dict": true,
22
+ "output_hidden_states": false,
23
+ "output_attentions": false,
24
+ "torchscript": false,
25
  "torch_dtype": "float32",
26
+ "use_bfloat16": false,
27
+ "tf_legacy_loss": false,
28
+ "pruned_heads": {},
29
+ "tie_word_embeddings": true,
30
+ "chunk_size_feed_forward": 0,
31
+ "is_encoder_decoder": false,
32
+ "is_decoder": false,
33
+ "cross_attention_hidden_size": null,
34
+ "add_cross_attention": false,
35
+ "tie_encoder_decoder": false,
36
+ "max_length": 20,
37
+ "min_length": 0,
38
+ "do_sample": false,
39
+ "early_stopping": false,
40
+ "num_beams": 1,
41
+ "num_beam_groups": 1,
42
+ "diversity_penalty": 0.0,
43
+ "temperature": 1.0,
44
+ "top_k": 50,
45
+ "top_p": 1.0,
46
+ "typical_p": 1.0,
47
+ "repetition_penalty": 1.0,
48
+ "length_penalty": 1.0,
49
+ "no_repeat_ngram_size": 0,
50
+ "encoder_no_repeat_ngram_size": 0,
51
+ "bad_words_ids": null,
52
+ "num_return_sequences": 1,
53
+ "output_scores": false,
54
+ "return_dict_in_generate": false,
55
+ "forced_bos_token_id": null,
56
+ "forced_eos_token_id": null,
57
+ "remove_invalid_values": false,
58
+ "exponential_decay_length_penalty": null,
59
+ "suppress_tokens": null,
60
+ "begin_suppress_tokens": null,
61
+ "architectures": [
62
+ "Qwen3ForCausalLM"
63
+ ],
64
+ "finetuning_task": null,
65
+ "id2label": {
66
+ "0": "LABEL_0",
67
+ "1": "LABEL_1"
68
+ },
69
+ "label2id": {
70
+ "LABEL_0": 0,
71
+ "LABEL_1": 1
72
+ },
73
+ "tokenizer_class": null,
74
+ "prefix": null,
75
+ "bos_token_id": 151643,
76
+ "pad_token_id": null,
77
+ "eos_token_id": 151643,
78
+ "sep_token_id": null,
79
+ "decoder_start_token_id": null,
80
+ "task_specific_params": null,
81
+ "problem_type": null,
82
+ "_name_or_path": "/tmp/fix_quant_4bfxpkhw",
83
+ "transformers_version": "4.52.3",
84
+ "model_type": "qwen3",
85
  "use_flash_attention_2": true,
86
+ "quantization_config": {
87
+ "bits": 4,
88
+ "group_size": 224,
89
+ "weight_dtype": "int4",
90
+ "activation_dtype": "int8",
91
+ "quant_method": "torchao_int4",
92
+ "sym": true,
93
+ "true_sequential": true
94
+ }
95
+ }