Upload tiny-clip/config.json with huggingface_hub
Browse files- tiny-clip/config.json +101 -0
tiny-clip/config.json
ADDED
@@ -0,0 +1,101 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"architectures": [
|
3 |
+
"CLIPTextModel"
|
4 |
+
],
|
5 |
+
"attention_probs_dropout_prob": 0.1,
|
6 |
+
"context_length": 1,
|
7 |
+
"hidden_act": "gelu",
|
8 |
+
"hidden_dropout_prob": 0.1,
|
9 |
+
"hidden_size": 1,
|
10 |
+
"image_resolution": 1,
|
11 |
+
"initializer_factor": 1.0,
|
12 |
+
"intermediate_size": 1,
|
13 |
+
"layer_norm_eps": 1e-05,
|
14 |
+
"logit_scale_init_value": 2.6592,
|
15 |
+
"model_type": "clip",
|
16 |
+
"num_attention_heads": 1,
|
17 |
+
"num_hidden_layers": 1,
|
18 |
+
"paddlenlp_version": null,
|
19 |
+
"projection_dim": 1,
|
20 |
+
"text_config": {
|
21 |
+
"bos_token_id": 1,
|
22 |
+
"classifier_dropout": null,
|
23 |
+
"dropout": 0.0,
|
24 |
+
"dtype": "float32",
|
25 |
+
"eos_token_id": 1,
|
26 |
+
"intermediate_size": 1,
|
27 |
+
"max_position_embeddings": 1,
|
28 |
+
"model_type": "clip_text_model",
|
29 |
+
"num_attention_heads": 1,
|
30 |
+
"num_choices": null,
|
31 |
+
"num_hidden_layers": 1,
|
32 |
+
"paddlenlp_version": null,
|
33 |
+
"projection_dim": 1,
|
34 |
+
"quantization_config": {
|
35 |
+
"act_quant_method": "abs_max",
|
36 |
+
"llm_int8_threshold": 6.0,
|
37 |
+
"quant_round_type": 0,
|
38 |
+
"quant_type": null,
|
39 |
+
"shift": false,
|
40 |
+
"shift_smooth_all_linears": false,
|
41 |
+
"smooth": false,
|
42 |
+
"weight_blocksize": 1,
|
43 |
+
"weight_double_quant": false,
|
44 |
+
"weight_double_quant_block_size": 1,
|
45 |
+
"weight_quant_method": "abs_max_channel_wise",
|
46 |
+
"weight_quantize_algo": null
|
47 |
+
},
|
48 |
+
"recompute_use_reentrant": false,
|
49 |
+
"tensor_parallel_degree": -1,
|
50 |
+
"tensor_parallel_output": false,
|
51 |
+
"tensor_parallel_rank": 0,
|
52 |
+
"top_k": 1,
|
53 |
+
"use_cache": false,
|
54 |
+
"use_flash_attention": false,
|
55 |
+
"vocab_size": 1
|
56 |
+
},
|
57 |
+
"torch_dtype": "float32",
|
58 |
+
"transformers_version": "4.33.1",
|
59 |
+
"vision_config": {
|
60 |
+
"classifier_dropout": null,
|
61 |
+
"dropout": 0.0,
|
62 |
+
"dtype": "float32",
|
63 |
+
"hidden_size": 1,
|
64 |
+
"image_size": 1,
|
65 |
+
"intermediate_size": 1,
|
66 |
+
"max_length": 1,
|
67 |
+
"model_type": "clip_vision_model",
|
68 |
+
"num_attention_heads": 1,
|
69 |
+
"num_channels": 1,
|
70 |
+
"num_choices": null,
|
71 |
+
"num_hidden_layers": 1,
|
72 |
+
"paddlenlp_version": null,
|
73 |
+
"patch_size": 1,
|
74 |
+
"projection_dim": 1,
|
75 |
+
"quantization_config": {
|
76 |
+
"act_quant_method": "abs_max",
|
77 |
+
"llm_int8_threshold": 6.0,
|
78 |
+
"quant_round_type": 0,
|
79 |
+
"quant_type": null,
|
80 |
+
"shift": false,
|
81 |
+
"shift_smooth_all_linears": false,
|
82 |
+
"smooth": false,
|
83 |
+
"weight_blocksize": 1,
|
84 |
+
"weight_double_quant": false,
|
85 |
+
"weight_double_quant_block_size": 1,
|
86 |
+
"weight_quant_method": "abs_max_channel_wise",
|
87 |
+
"weight_quantize_algo": null
|
88 |
+
},
|
89 |
+
"recompute_use_reentrant": false,
|
90 |
+
"tensor_parallel_degree": -1,
|
91 |
+
"tensor_parallel_output": false,
|
92 |
+
"tensor_parallel_rank": 0,
|
93 |
+
"top_k": 1,
|
94 |
+
"use_cache": false,
|
95 |
+
"use_flash_attention": false
|
96 |
+
},
|
97 |
+
"vision_layers": 1,
|
98 |
+
"vision_patch_size": 1,
|
99 |
+
"vision_width": 1,
|
100 |
+
"vocab_size": 1
|
101 |
+
}
|