ydy9038074's picture
Publish Modilify Mk2 Preview MLX
e4f7326 verified
Raw History Blame Contribute Delete
4.02 kB
{
"canvas_length": 256,
"channel_end_token_id": 101,
"commit_confidence_power": 1.0,
"commit_entropy_weight": 1.0,
"commit_failure_budget": 0.2,
"commit_gold_alpha": 0.4,
"commit_gold_weight": 0.2,
"commit_min_p": 0.05,
"commit_readiness_beta": 0.02,
"commit_readiness_budget_margin": 0.02,
"commit_readiness_failure_weight": 5.0,
"commit_readiness_loss_weight": 0.5,
"commit_readiness_objective": "frontier_prefix_budget_v2",
"commit_readiness_target_tokens": 16,
"commit_sequence_dim": 1024,
"commit_sequence_layers": 2,
"commit_target_confidence": 0.5,
"commit_top_k": 40,
"confidence_calibration_loss_weight": 0.01,
"eos_token_id": 1,
"initializer_range": 0.02,
"kv_cache_bucket_size": 128,
"latent_dim": 2816,
"latent_ffn_dim": 7168,
"latent_history_kv_rank": 1024,
"latent_history_length": 16,
"latent_history_views": 4,
"latent_local_attention_window": 128,
"latent_memory_slots": 256,
"latent_num_heads": 16,
"latent_num_layers": 4,
"latent_persistent_bus_unfreeze_steps": 0,
"latent_tape_probes": 4,
"latent_tape_scheme": "gdn2_spatial_probe_v1",
"latent_working_bus_unfreeze_steps": 0,
"latent_working_last_block_global": true,
"memory_architecture": "compact_gdn2_v2",
"memory_scheme": "dual_timescale_gdn2_trajectory_memory",
"model_type": "modilify_mk2",
"persistent_memory_bus": true,
"persistent_memory_write": "commit_only_transformer",
"state_schema_version": 25,
"terminal_stop_loss_weight": 0.01,
"terminal_stop_target_probability": 0.9,
"terminal_token_ids": [
106,
50
],
"text_config": {
"attention_bias": false,
"attention_dropout": 0.0,
"bos_token_id": 2,
"dtype": "bfloat16",
"eos_token_id": 1,
"final_logit_softcapping": 30.0,
"global_head_dim": 512,
"head_dim": 256,
"hidden_activation": "gelu_pytorch_tanh",
"hidden_size": 2816,
"initializer_range": 0.02,
"intermediate_size": 2112,
"layer_types": [
"sliding_attention",
"sliding_attention",
"sliding_attention",
"sliding_attention",
"sliding_attention",
"full_attention",
"sliding_attention",
"sliding_attention",
"sliding_attention",
"sliding_attention",
"sliding_attention",
"full_attention",
"sliding_attention",
"sliding_attention",
"sliding_attention",
"sliding_attention",
"sliding_attention",
"full_attention",
"sliding_attention",
"sliding_attention",
"sliding_attention",
"sliding_attention",
"sliding_attention",
"full_attention",
"sliding_attention",
"sliding_attention",
"sliding_attention",
"sliding_attention",
"sliding_attention",
"full_attention"
],
"max_position_embeddings": 262144,
"model_type": "modilify_mk2_text",
"moe_intermediate_size": 704,
"num_attention_heads": 16,
"num_experts": 128,
"num_global_key_value_heads": 2,
"num_hidden_layers": 30,
"num_key_value_heads": 8,
"pad_token_id": 0,
"rms_norm_eps": 1e-06,
"rope_parameters": {
"full_attention": {
"partial_rotary_factor": 0.25,
"rope_theta": 1000000.0,
"rope_type": "proportional"
},
"sliding_attention": {
"rope_theta": 10000.0,
"rope_type": "default"
}
},
"sliding_window": 1024,
"tie_word_embeddings": true,
"top_k_experts": 8,
"use_bidirectional_attention": null,
"vocab_size": 262144
},
"tie_word_embeddings": true,
"token_ce_normalization": "per_sample_exposure_v1",
"token_ce_supervision": "valid_canvas_v1",
"token_loss_weight": 1.0,
"training_bptt_steps": 16,
"training_prefix_cache": "incremental",
"training_scheme": "gold_prefix_shared_commit_0_256_committed_ce_calibration_causal_throughput_terminal_sft",
"transformers_version": "5.14.1",
"turn_end_token_id": 106,
"vocab_chunk_size": 32768,
"working_memory_bus": true,
"writer_slot_gate": "per_slot"
}