| { |
| "architectures": [ |
| "AetherMindForCausalLM" |
| ], |
| "attention_bias": false, |
| "attention_dropout": 0.0, |
| "bos_token_id": 2, |
| "dtype": "float32", |
| "eos_token_id": 3, |
| "head_dim": 32, |
| "hidden_act": "silu", |
| "hidden_size": 256, |
| "initializer_range": 0.02, |
| "intermediate_size": 768, |
| "max_position_embeddings": 1024, |
| "model_type": "aethermind", |
| "moe_aux_loss_coeff": 0.01, |
| "moe_intermediate_size": 768, |
| "moe_layers_freq": 1, |
| "num_attention_heads": 8, |
| "num_experts": 8, |
| "num_experts_per_tok": 2, |
| "num_hidden_layers": 6, |
| "num_key_value_heads": 4, |
| "pad_token_id": 0, |
| "rms_norm_eps": 1e-06, |
| "rope_parameters": { |
| "rope_theta": 10000.0, |
| "rope_type": "default" |
| }, |
| "rope_theta": 10000.0, |
| "tie_word_embeddings": true, |
| "transformers_version": "5.17.0", |
| "use_cache": true, |
| "use_moe": false, |
| "vocab_size": 4326, |
| "auto_map": { |
| "AutoConfig": "configuration_aethermind.AetherMindConfig", |
| "AutoModel": "modeling_aethermind.AetherMindModel", |
| "AutoModelForCausalLM": "modeling_aethermind.AetherMindForCausalLM" |
| } |
| } |
