RedHatAI/whisper-small-quantized.w4a16
1216
1{2 "_name_or_path": "openai/whisper-small",3 "activation_dropout": 0.0,4 "activation_function": "gelu",5 "apply_spec_augment": false,6 "architectures": [7 "WhisperForConditionalGeneration"8 ],9 "attention_dropout": 0.0,10 "begin_suppress_tokens": null,11 "bos_token_id": 50257,12 "classifier_proj_size": 256,13 "d_model": 768,14 "decoder_attention_heads": 12,15 "decoder_ffn_dim": 3072,16 "decoder_layerdrop": 0.0,17 "decoder_layers": 12,18 "decoder_start_token_id": 50258,19 "dropout": 0.0,20 "encoder_attention_heads": 12,21 "encoder_ffn_dim": 3072,22 "encoder_layerdrop": 0.0,23 "encoder_layers": 12,24 "eos_token_id": 50257,25 "forced_decoder_ids": null,26 "init_std": 0.02,27 "is_encoder_decoder": true,28 "mask_feature_length": 10,29 "mask_feature_min_masks": 0,30 "mask_feature_prob": 0.0,31 "mask_time_length": 10,32 "mask_time_min_masks": 2,33 "mask_time_prob": 0.05,34 "max_length": null,35 "max_source_positions": 1500,36 "max_target_positions": 448,37 "median_filter_width": 7,38 "model_type": "whisper",39 "num_hidden_layers": 12,40 "num_mel_bins": 80,41 "pad_token_id": 50257,42 "quantization_config": {43 "config_groups": {44 "group_0": {45 "input_activations": null,46 "output_activations": null,47 "targets": [48 "Linear"49 ],50 "weights": {51 "actorder": "weight",52 "block_structure": null,53 "dynamic": false,54 "group_size": 64,55 "num_bits": 4,56 "observer": "minmax",57 "observer_kwargs": {},58 "strategy": "group",59 "symmetric": true,60 "type": "int"61 }62 }63 },64 "format": "pack-quantized",65 "global_compression_ratio": 2.4618563587964775,66 "ignore": [],67 "kv_cache_scheme": null,68 "quant_method": "compressed-tensors",69 "quantization_status": "compressed"70 },71 "scale_embedding": false,72 "torch_dtype": "float32",73 "transformers_version": "4.49.0",74 "use_cache": true,75 "use_weighted_layer_sum": false,76 "vocab_size": 5186577}