kairos / config.json
ffurfaro's picture
Upload folder using huggingface_hub
09838da verified
Raw
History Blame Contribute Delete
1.5 kB
{
"A_init_range": [
1.0,
16.0
],
"attention_dropout": 0.0,
"attnres_block_size": 4,
"auto_map": {
"AutoConfig": "modeling.KairosConfig"
},
"head_dim": 16,
"hidden_act": "silu",
"hidden_size": 64,
"initializer_range": 0.02,
"intermediate_size": 544,
"layers_config": [
"ld",
"ld",
"ld",
"ld"
],
"linear_conv_kernel_dim": 4,
"linear_key_head_dim": 16,
"linear_num_key_heads": 4,
"linear_num_value_heads": 4,
"linear_value_head_dim": 16,
"max_position_embeddings": 4096,
"modality_scales": {
"0": [
0,
1
],
"1": [
1,
2
],
"2": [
2,
3
],
"3": [
2
],
"4": [
1
],
"5": [
0
],
"6": [
0
],
"7": [
3
]
},
"model_type": "kairos",
"moe_intermediate_size": 544,
"n_group": 1,
"n_shared_experts": 1,
"norm_topk_prob": false,
"num_attention_heads": 4,
"num_experts_per_tok": 1,
"num_hidden_layers": 4,
"num_key_value_heads": 4,
"num_local_experts": 7,
"num_modalities": 8,
"num_scales": 4,
"rms_norm_eps": 1e-06,
"rope_theta": 10000.0,
"routed_scaling_factor": 1.0,
"sliding_window_size": 128,
"slw_wsize": -1,
"stride": 3,
"text_modality_id": 0,
"time_step_floor": 0.0001,
"time_step_max": 0.1,
"time_step_min": 0.001,
"topk_group": 1,
"transformers_version": "5.15.0",
"use_cache": false,
"use_memory_gate": true,
"use_moe": true,
"vocab_size": 291
}