tiny-clip-one / config.json
Baicai003's picture
Upload config.json with huggingface_hub
ab911bd
{
"architectures": [
"CLIPTextModel"
],
"attention_probs_dropout_prob": 0.1,
"context_length": 1,
"hidden_act": "gelu",
"hidden_dropout_prob": 0.1,
"hidden_size": 1,
"image_resolution": 1,
"initializer_factor": 1.0,
"intermediate_size": 1,
"layer_norm_eps": 1e-05,
"logit_scale_init_value": 2.6592,
"model_type": "clip",
"num_attention_heads": 1,
"num_hidden_layers": 1,
"paddlenlp_version": null,
"projection_dim": 1,
"text_config": {
"bos_token_id": 1,
"classifier_dropout": null,
"dropout": 0.0,
"dtype": "float32",
"eos_token_id": 1,
"intermediate_size": 1,
"max_position_embeddings": 1,
"model_type": "clip_text_model",
"num_attention_heads": 1,
"num_choices": null,
"num_hidden_layers": 1,
"paddlenlp_version": null,
"projection_dim": 1,
"quantization_config": {
"act_quant_method": "abs_max",
"llm_int8_threshold": 6.0,
"quant_round_type": 0,
"quant_type": null,
"shift": false,
"shift_smooth_all_linears": false,
"smooth": false,
"weight_blocksize": 1,
"weight_double_quant": false,
"weight_double_quant_block_size": 1,
"weight_quant_method": "abs_max_channel_wise",
"weight_quantize_algo": null
},
"recompute_use_reentrant": false,
"tensor_parallel_degree": -1,
"tensor_parallel_output": false,
"tensor_parallel_rank": 0,
"top_k": 1,
"use_cache": false,
"use_flash_attention": false,
"vocab_size": 1
},
"torch_dtype": "float32",
"transformers_version": "4.33.1",
"vision_config": {
"classifier_dropout": null,
"dropout": 0.0,
"dtype": "float32",
"hidden_size": 1,
"image_size": 1,
"intermediate_size": 1,
"max_length": 1,
"model_type": "clip_vision_model",
"num_attention_heads": 1,
"num_channels": 1,
"num_choices": null,
"num_hidden_layers": 1,
"paddlenlp_version": null,
"patch_size": 1,
"projection_dim": 1,
"quantization_config": {
"act_quant_method": "abs_max",
"llm_int8_threshold": 6.0,
"quant_round_type": 0,
"quant_type": null,
"shift": false,
"shift_smooth_all_linears": false,
"smooth": false,
"weight_blocksize": 1,
"weight_double_quant": false,
"weight_double_quant_block_size": 1,
"weight_quant_method": "abs_max_channel_wise",
"weight_quantize_algo": null
},
"recompute_use_reentrant": false,
"tensor_parallel_degree": -1,
"tensor_parallel_output": false,
"tensor_parallel_rank": 0,
"top_k": 1,
"use_cache": false,
"use_flash_attention": false
},
"vision_layers": 1,
"vision_patch_size": 1,
"vision_width": 1,
"vocab_size": 1
}