Guilherme34 commited on
Commit
cda1ab5
·
verified ·
1 Parent(s): f025b33

Update adapter_config.json

Browse files
Files changed (1) hide show
  1. adapter_config.json +24 -192
adapter_config.json CHANGED
@@ -1,193 +1,25 @@
1
  {
2
- "_name_or_path": "openbmb/MiniCPM-o-2_6",
3
- "architectures": [
4
- "MiniCPMO"
5
- ],
6
-
7
- "attention_dropout": 0.0,
8
- "bos_token_id": 151643,
9
- "eos_token_id": 151645,
10
- "hidden_act": "silu",
11
- "hidden_size": 3584,
12
- "initializer_range": 0.02,
13
- "intermediate_size": 18944,
14
- "max_position_embeddings": 32768,
15
- "max_window_layers": 28,
16
- "num_attention_heads": 28,
17
- "num_hidden_layers": 28,
18
- "num_key_value_heads": 4,
19
- "rms_norm_eps": 1e-06,
20
- "rope_theta": 1000000.0,
21
- "sliding_window": 131072,
22
- "tie_word_embeddings": false,
23
- "use_sliding_window": false,
24
- "vocab_size": 151700,
25
- "batch_vision_input": true,
26
- "drop_vision_last_layer": false,
27
- "image_size": 448,
28
-
29
- "audio_chunk_length": 1.0,
30
- "audio_config": {
31
- "_name_or_path": "openai/whisper-medium",
32
- "architectures": [
33
- "MiniCPMWhisperEncoder"
34
- ],
35
- "begin_suppress_tokens": [
36
- 220,
37
- 50257
38
- ],
39
- "bos_token_id": 50257,
40
- "d_model": 1024,
41
- "decoder_attention_heads": 16,
42
- "decoder_ffn_dim": 4096,
43
- "decoder_layers": 24,
44
- "decoder_start_token_id": 50258,
45
- "encoder_attention_heads": 16,
46
- "encoder_ffn_dim": 4096,
47
- "encoder_layers": 24,
48
- "eos_token_id": 50257,
49
- "forced_decoder_ids": [
50
- [
51
- 1,
52
- 50259
53
- ],
54
- [
55
- 2,
56
- 50359
57
- ],
58
- [
59
- 3,
60
- 50363
61
- ]
62
- ],
63
- "max_length": 448,
64
- "model_type": "whisper",
65
- "num_hidden_layers": 24,
66
- "pad_token_id": 50257,
67
- "suppress_tokens": [
68
- 1,
69
- 2,
70
- 7,
71
- 8,
72
- 9,
73
- 10,
74
- 14,
75
- 25,
76
- 26,
77
- 27,
78
- 28,
79
- 29,
80
- 31,
81
- 58,
82
- 59,
83
- 60,
84
- 61,
85
- 62,
86
- 63,
87
- 90,
88
- 91,
89
- 92,
90
- 93,
91
- 359,
92
- 503,
93
- 522,
94
- 542,
95
- 873,
96
- 893,
97
- 902,
98
- 918,
99
- 922,
100
- 931,
101
- 1350,
102
- 1853,
103
- 1982,
104
- 2460,
105
- 2627,
106
- 3246,
107
- 3253,
108
- 3268,
109
- 3536,
110
- 3846,
111
- 3961,
112
- 4183,
113
- 4667,
114
- 6585,
115
- 6647,
116
- 7273,
117
- 9061,
118
- 9383,
119
- 10428,
120
- 10929,
121
- 11938,
122
- 12033,
123
- 12331,
124
- 12562,
125
- 13793,
126
- 14157,
127
- 14635,
128
- 15265,
129
- 15618,
130
- 16553,
131
- 16604,
132
- 18362,
133
- 18956,
134
- 20075,
135
- 21675,
136
- 22520,
137
- 26130,
138
- 26161,
139
- 26435,
140
- 28279,
141
- 29464,
142
- 31650,
143
- 32302,
144
- 32470,
145
- 36865,
146
- 42863,
147
- 47425,
148
- 49870,
149
- 50254,
150
- 50258,
151
- 50358,
152
- 50359,
153
- 50360,
154
- 50361,
155
- 50362
156
- ],
157
- "torch_dtype": "float32"
158
- },
159
- "audio_pool_step": 2,
160
- "auto_map": {
161
- "AutoConfig": "configuration_minicpm.MiniCPMOConfig",
162
- "AutoModel": "modeling_minicpmo.MiniCPMO",
163
- "AutoModelForCausalLM": "modeling_minicpmo.MiniCPMO"
164
- },
165
- "chunk_input": true,
166
- "listen_speak_type": "asr",
167
- "model_type": "minicpmo",
168
- "patch_size": 14,
169
- "query_num": 64,
170
- "slice_config": {
171
- "max_slice_nums": 9,
172
- "model_type": "minicpmv"
173
- },
174
- "slice_mode": true,
175
- "torch_dtype": "bfloat16",
176
- "transformers_version": "4.44.2",
177
- "tts_config": {
178
- "model_type": "conditional_chattts",
179
- "llm_dim": 3584
180
- },
181
- "use_cache": true,
182
- "use_image_id": true,
183
- "version": 2.6,
184
- "vision_batch_size": 16,
185
- "vision_config": {
186
- "hidden_size": 1152,
187
- "image_size": 980,
188
- "intermediate_size": 4304,
189
- "model_type": "siglip_vision_model",
190
- "num_attention_heads": 16,
191
- "num_hidden_layers": 27,
192
- "patch_size": 14
193
- }
 
1
  {
2
+ "alpha_pattern": {},
3
+ "auto_mapping": null,
4
+ "base_model_name_or_path": "openbmb/MiniCPM-o-2_6",
5
+ "bias": "none",
6
+ "fan_in_fan_out": false,
7
+ "inference_mode": true,
8
+ "init_lora_weights": true,
9
+ "layer_replication": null,
10
+ "layers_pattern": null,
11
+ "layers_to_transform": null,
12
+ "loftq_config": {},
13
+ "lora_alpha": 16,
14
+ "lora_dropout": 0,
15
+ "megatron_config": null,
16
+ "megatron_core": "megatron.core",
17
+ "modules_to_save": null,
18
+ "peft_type": "LORA",
19
+ "r": 8,
20
+ "rank_pattern": {},
21
+ "revision": null,
22
+ "task_type": "CAUSAL_LM",
23
+ "use_dora": false,
24
+ "use_rslora": false
25
+ }