MiniMax-M3-FP8-dynamic / config.json
pstan's picture
Add files using upload-large-folder tool
9247354 verified
Raw
History Blame Contribute Delete
6.75 kB
{
"architectures": [
"MiniMaxM3SparseForConditionalGeneration"
],
"auto_map": {
"AutoConfig": "configuration_minimax_m3_vl.MiniMaxM3VLConfig"
},
"image_grid_pinpoints": "[(336, 336), (336, 672), (336, 1008), (336, 1344), (336, 1680), (336, 2016), (672, 336), (672, 672), (672, 1008), (672, 1344), (672, 1680), (672, 2016), (1008, 336), (1008, 672), (1008, 1008), (1008, 1344), (1008, 1680), (1008, 2016), (1344, 336), (1344, 672), (1344, 1008), (1344, 1344), (1344, 1680), (1344, 2016), (1680, 336), (1680, 672), (1680, 1008), (1680, 1344), (1680, 1680), (1680, 2016), (2016, 336), (2016, 672), (2016, 1008), (2016, 1344), (2016, 1680), (2016, 2016)]",
"image_seq_length": 576,
"image_token_index": 200025,
"img_token_compression_config": {
"image_token_compression_method": "patch_merge",
"spatial_merge_size": 2,
"temporal_patch_size": 2
},
"model_type": "minimax_m3_vl",
"multimodal_projector_bias": true,
"num_reward_heads": 0,
"process_image_mode": "dynamic_res",
"projector_hidden_act": "gelu",
"projector_hidden_size": 6144,
"quantization_config": {
"config_groups": {
"config_group_0": {
"format": "float-quantized",
"input_activations": {
"actorder": null,
"block_structure": null,
"dynamic": true,
"group_size": null,
"num_bits": 8,
"observer": null,
"observer_kwargs": {},
"scale_dtype": null,
"strategy": "token",
"symmetric": true,
"type": "float",
"zp_dtype": null
},
"output_activations": null,
"targets": [
"Linear"
],
"weights": {
"actorder": null,
"block_structure": null,
"dynamic": false,
"group_size": null,
"num_bits": 8,
"observer": "memoryless_minmax",
"observer_kwargs": {},
"scale_dtype": null,
"strategy": "channel",
"symmetric": true,
"type": "float",
"zp_dtype": null
}
}
},
"format": "float-quantized",
"global_compression_ratio": null,
"ignore": [
"re:.*lm_head$",
"re:.*embed_tokens$",
"re:.*vision_tower.*",
"re:.*multi_modal_projector.*",
"re:.*multimodal_projector.*",
"re:.*patch_merge_mlp.*",
"re:.*block_sparse_moe\\.gate$"
],
"kv_cache_scheme": null,
"quant_method": "compressed-tensors",
"quantization_status": "compressed",
"transform_config": {},
"version": "0.17.0"
},
"text_config": {
"architectures": [
"MiniMaxM3SparseForCausalLM"
],
"attention_output_gate": false,
"dense_intermediate_size": 12288,
"head_dim": 128,
"hidden_act": "swigluoai",
"hidden_size": 6144,
"intermediate_size": 3072,
"max_position_embeddings": 1048576,
"moe_layer_freq": [
0,
0,
0,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1
],
"n_shared_experts": 1,
"num_attention_heads": 64,
"num_experts_per_tok": 4,
"num_hidden_layers": 60,
"num_key_value_heads": 4,
"num_local_experts": 128,
"num_mtp_modules": 7,
"num_nextn_predict_layers": 1,
"partial_rotary_factor": 0.5,
"qk_norm_type": "per_head",
"rms_norm_eps": 1e-06,
"rope_theta": 5000000,
"rotary_dim": 64,
"routed_scaling_factor": 2.0,
"scoring_func": "sigmoid",
"shared_intermediate_size": 3072,
"sparse_attention_config": {
"sparse_attention_freq": [
0,
0,
0,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1
],
"sparse_block_size": 128,
"sparse_disable_index_value": [
0,
0,
0,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1,
1
],
"sparse_index_dim": 128,
"sparse_init_block": 0,
"sparse_local_block": 1,
"sparse_num_index_heads": 4,
"sparse_score_type": "max",
"sparse_topk_blocks": 16,
"use_sparse_attention": true
},
"swiglu_alpha": 1.702,
"swiglu_limit": 7.0,
"tie_word_embeddings": false,
"use_gemma_norm": true,
"use_qk_norm": true,
"use_routing_bias": true,
"vocab_size": 200064
},
"torch_dtype": "bfloat16",
"transformers_version": "4.52.4",
"video_token_index": 200026,
"vision_config": {
"attention_dropout": 0.0,
"hidden_act": "gelu",
"hidden_size": 1280,
"image_size": 2016,
"img_token_compression_config": {
"image_token_compression_method": "patch_merge",
"spatial_merge_size": 2,
"temporal_patch_size": 2
},
"initializer_factor": 1.0,
"initializer_range": 0.02,
"intermediate_size": 5120,
"layer_norm_eps": 1e-05,
"model_type": "clip_vision_model",
"num_attention_heads": 16,
"num_channels": 3,
"num_hidden_layers": 32,
"patch_size": 14,
"position_embedding_type": "rope",
"projection_dim": 6144,
"rope_mode": "3d",
"rope_theta": 10000.0,
"vision_segment_max_frames": 4,
"vocab_size": 32000
},
"vision_feature_layer": -1,
"vision_feature_select_strategy": "full"
}