sdougbrown/North-Mini-Code-1.0-dflash
025
1{2 "architectures": [3 "DFlashDraftModel"4 ],5 "auto_map": {6 "": "config.DFlashSpeculatorConfig"7 },8 "aux_hidden_state_layer_ids": [9 2,10 24,11 4612 ],13 "block_size": 8,14 "draft_vocab_size": 32000,15 "dtype": "bfloat16",16 "mask_token_id": 1,17 "sample_from_anchor": false,18 "sliding_window_non_causal": false,19 "speculators_config": {20 "algorithm": "dflash",21 "default_proposal_method": "greedy",22 "proposal_methods": [23 {24 "accept_tolerance": 0.0,25 "proposal_type": "greedy",26 "speculative_tokens": 3,27 "verifier_accept_k": 128 }29 ],30 "verifier": {31 "architectures": [32 "Cohere2MoeForCausalLM"33 ],34 "name_or_path": "CohereLabs/North-Mini-Code-1.0-w4a16"35 }36 },37 "speculators_model_type": "dflash",38 "speculators_version": "0.6.0",39 "target_hidden_size": null,40 "tie_word_embeddings": false,41 "transformer_layer_config": {42 "attention_bias": false,43 "attention_dropout": 0.0,44 "bos_token_id": null,45 "eos_token_id": null,46 "head_dim": 128,47 "hidden_act": "silu",48 "hidden_size": 2048,49 "initializer_range": 0.02,50 "intermediate_size": 768,51 "layer_types": [52 "sliding_attention",53 "sliding_attention",54 "sliding_attention",55 "sliding_attention",56 "sliding_attention"57 ],58 "max_position_embeddings": 500000,59 "max_window_layers": 28,60 "model_type": "qwen3",61 "num_attention_heads": 32,62 "num_hidden_layers": 5,63 "num_key_value_heads": 4,64 "pad_token_id": null,65 "rms_norm_eps": 1e-06,66 "rope_parameters": {67 "rope_theta": 50000,68 "rope_type": "default"69 },70 "sliding_window": 2048,71 "tie_word_embeddings": false,72 "use_cache": true,73 "use_sliding_window": true,74 "vocab_size": 26214475 },76 "transformers_version": "5.13.1"77}78 