| { | |
| "architectures": [ | |
| "SurjoExpForCausalLM" | |
| ], | |
| "attention_bias": false, | |
| "attention_dropout": 0.0, | |
| "auto_map": { | |
| "AutoConfig": "configuration_surjo_exp.SurjoExpConfig", | |
| "AutoModelForCausalLM": "modeling_surjo_exp.SurjoExpForCausalLM" | |
| }, | |
| "bos_token_id": 0, | |
| "coda_layers": 16, | |
| "dtype": "float32", | |
| "eos_token_id": 2, | |
| "gradient_checkpointing": false, | |
| "head_dim": 72, | |
| "hidden_act": "silu", | |
| "hidden_size": 576, | |
| "initializer_range": 0.02, | |
| "intermediate_size": 1536, | |
| "max_position_embeddings": 2048, | |
| "mlp_bias": false, | |
| "model_type": "surjo_exp", | |
| "num_attention_heads": 8, | |
| "num_hidden_layers": 32, | |
| "num_key_value_heads": 2, | |
| "pad_token_id": 1, | |
| "prelude_layers": 4, | |
| "pretraining_tp": 1, | |
| "recurrent_layers": 12, | |
| "rms_norm_eps": 1e-05, | |
| "rope_parameters": { | |
| "rope_theta": 10000.0, | |
| "rope_type": "default" | |
| }, | |
| "rope_theta": 10000.0, | |
| "tie_word_embeddings": true, | |
| "transformers_version": "5.15.0", | |
| "use_cache": false, | |
| "use_flash_attn": false, | |
| "vocab_size": 32768, | |
| "xsa_projection": true | |
| } |