hf-transformers-bot's picture
Update tiny models for Step3p7Model
eee3f1b verified
Raw History Blame Contribute Delete
1.99 kB
{
"architectures": [
"Step3p7Model"
],
"dtype": "float32",
"image_token_id": 4,
"model_type": "step3p7",
"pad_token_id": 1,
"projector_bias": false,
"text_config": {
"attention_bias": false,
"attention_dropout": 0.0,
"bos_token_id": 0,
"eos_token_id": 128007,
"head_dim": 8,
"hidden_act": "silu",
"hidden_size": 16,
"initializer_range": 0.02,
"intermediate_size": 37,
"layer_types": [
"full_attention",
"full_attention"
],
"max_position_embeddings": 200,
"mlp_bias": false,
"mlp_layer_types": [
"dense",
"sparse"
],
"model_type": "step3p5",
"moe_intermediate_size": 8,
"moe_router_scaling_factor": 1.0,
"mtp_layer_types": null,
"mtp_mlp_layer_types": null,
"n_routed_experts": 4,
"num_attention_heads": 2,
"num_experts_per_tok": 2,
"num_hidden_layers": 2,
"num_key_value_heads": 1,
"num_nextn_predict_layers": 0,
"num_sliding_attention_heads": 2,
"pad_token_id": 1,
"per_layer_config": {},
"query_pre_attn_scalar": 8,
"rms_norm_eps": 1e-05,
"rope_parameters": {
"full_attention": {
"rope_theta": 10000.0,
"rope_type": "default"
}
},
"share_expert_dim": 8,
"sliding_window": 64,
"swiglu_limits": null,
"swiglu_limits_shared": null,
"tie_word_embeddings": false,
"use_cache": true,
"vocab_size": 128815
},
"transformers_version": "5.19.0.dev0",
"vision_config": {
"attention_dropout": 0.0,
"hidden_act": "quick_gelu",
"hidden_size": 8,
"image_size": 16,
"intermediate_size": 8,
"layer_norm_eps": 1e-05,
"layer_scale_init_value": 0.1,
"max_position_embeddings": 2704,
"mlp_ratio": 1.0,
"model_type": "step3p5_vision",
"num_attention_heads": 2,
"num_channels": 3,
"num_hidden_layers": 1,
"patch_size": 4,
"rope_parameters": {
"rope_theta": 10000.0,
"rope_type": "axial"
}
}
}