| { |
| "architectures": [ |
| "NandiForCausalLM" |
| ], |
| "auto_map": { |
| "AutoConfig": "configuration_nandi.NandiConfig", |
| "AutoModel": "modeling_nandi.NandiModel", |
| "AutoModelForCausalLM": "modeling_nandi.NandiForCausalLM" |
| }, |
| "model_type": "nandi", |
| "vocab_size": 131072, |
| "hidden_size": 832, |
| "intermediate_size": 2496, |
| "num_hidden_layers": 16, |
| "num_attention_heads": 16, |
| "num_key_value_heads": 4, |
| "hidden_act": "silu", |
| "max_position_embeddings": 2048, |
| "initializer_range": 0.02, |
| "rms_norm_eps": 1e-05, |
| "use_cache": true, |
| "pad_token_id": null, |
| "bos_token_id": 1, |
| "eos_token_id": 0, |
| "tie_word_embeddings": true, |
| "rope_parameters": { |
| "rope_theta": 100000 |
| }, |
| "attention_bias": false, |
| "attention_dropout": 0.0, |
| "mlp_bias": false, |
| "factorized_embedding": true, |
| "embedding_rank": 196, |
| "layer_sharing": true, |
| "layer_sharing_repeats": 2, |
| "torch_dtype": "bfloat16" |
| } |
