| { |
| "affine_momentum": 0.9, |
| "architectures": [ |
| "NeoLLMForCausalLM" |
| ], |
| "attention_bias": false, |
| "attention_dropout": 0.1, |
| "attn_res_num_blocks": 4, |
| "auto_map": { |
| "AutoConfig": "configuration_neollm.NeoLLMConfig", |
| "AutoModel": "modeling_neollm.NeoLLMModel", |
| "AutoModelForCausalLM": "modeling_neollm.NeoLLMForCausalLM" |
| }, |
| "bos_token_id": 1, |
| "directional_routing_k": 4, |
| "directional_routing_temp": 3.0, |
| "dropout_rate": 0.1, |
| "dtype": "bfloat16", |
| "eos_token_id": 7, |
| "fan_ratio": 0.125, |
| "fan_ratio_ffn": 0.0625, |
| "generator_d_seed": 128, |
| "generator_k": 3, |
| "generator_krank": 32, |
| "generator_num_knots": 32, |
| "generator_num_modes": 8, |
| "generator_spline_degree": 2, |
| "head_dim": 64, |
| "hidden_act": "xielu", |
| "hidden_size": 512, |
| "initializer_range": 0.02, |
| "intermediate_size": 1536, |
| "jtokm_aux_loss_weight": 0.0001, |
| "jtokm_norm_eps": 1e-06, |
| "jtokm_num_experts": 4, |
| "jtokm_num_modes": 4, |
| "jtokm_top_k": 2, |
| "laurel_lr_rank": 32, |
| "lucid_attention_eps": 1e-06, |
| "max_position_embeddings": 512, |
| "mea_component_key_value_heads": 2, |
| "mea_groupnorm_eps": 1e-06, |
| "model_type": "neollm", |
| "momentum_gamma": 0.1, |
| "num_attention_heads": 8, |
| "num_hidden_layers": 12, |
| "num_key_value_heads": 2, |
| "pad_token_id": 0, |
| "partial_rotary_factor": 0.25, |
| "polynorm_exclusive": false, |
| "repo_d_p": 64, |
| "repo_start_layer": 4, |
| "rms_norm_eps": 1e-06, |
| "rope_parameters": { |
| "partial_rotary_factor": 0.25, |
| "rope_theta": 10000.0, |
| "rope_type": "default" |
| }, |
| "rope_theta": 10000.0, |
| "tie_word_embeddings": true, |
| "transformers_version": "5.5.3", |
| "use_affine_scaled_attention": true, |
| "use_attn_res": false, |
| "use_cache": false, |
| "use_directional_routing": false, |
| "use_hadamard_o_proj": true, |
| "use_jtokm": false, |
| "use_laurel": false, |
| "use_laurel_lr": false, |
| "use_laurel_rw": false, |
| "use_lucid_attention": false, |
| "use_mea_attention": false, |
| "use_momentum_attention": true, |
| "use_repo": true, |
| "use_spelling_bee_embeddings": false, |
| "use_token_generator": false, |
| "use_versatile_ffn": false, |
| "use_xsa": true, |
| "versatile_active_experts": 2, |
| "versatile_aux_loss_weight": 1e-05, |
| "versatile_gumbel_temp_decay": 0.99984, |
| "versatile_gumbel_temp_end": 0.1, |
| "versatile_gumbel_temp_start": 5.0, |
| "versatile_max_depth": 2, |
| "versatile_total_experts": 4, |
| "vocab_size": 64402, |
| "xsa_eps": 1e-06 |
| } |
|
|