Bochkov/ab_ext_binary16
0568
1{2 "architectures": [3 "AttnExtForCausalLM"4 ],5 "attention_bias": false,6 "auto_map": {7 "AutoConfig": "configuration_attn_ext.AttnExtConfig",8 "AutoModelForCausalLM": "modeling_attn_ext.AttnExtForCausalLM"9 },10 "binary_dim": 16,11 "binary_encoding": "zero_one",12 "binary_repeat": 128,13 "binary_scale": 1.0,14 "block_size": 2048,15 "bos_token_id": 0,16 "code_seed": 12345,17 "d_model": 2048,18 "dropout": 0.0,19 "eos_token_id": 0,20 "ffn_multiplier": 4.0,21 "head_dim": 64,22 "hidden_size": 2048,23 "initializer_range": 0.02,24 "input_mode": "binary16",25 "is_decoder": true,26 "is_encoder_decoder": false,27 "max_position_embeddings": 2048,28 "min_col_weight": 4,29 "min_row_weight": 4,30 "mlp_bias": false,31 "model_type": "attn_ext",32 "multiple_of": 256,33 "n_head": 32,34 "n_layer": 24,35 "num_attention_heads": 32,36 "num_hidden_layers": 24,37 "pad_token_id": null,38 "rms_norm_eps": 1e-05,39 "rope_theta": 10000.0,40 "tie_word_embeddings": false,41 "transformers_version": null,42 "use_cache": false,43 "vocab_size": 4915244}45 