| { |
| "architectures": ["ParamBharatGenForCausalLM"], |
| "model_type": "parambharatgen", |
| "auto_map":{ |
| "AutoConfig": "config_parambharatgen.ParamBharatGenConfig", |
| "AutoModelForCausalLM": "modeling_parambharatgen.ParamBharatGenForCausalLM" |
| }, |
| "attention_bias": false, |
| "attention_dropout": 0.0, |
| "bos_token_id": 2, |
| "eos_token_id": 3, |
| "head_dim": 128, |
| "hidden_act": "silu", |
| "hidden_size": 2048, |
| "initializer_range": 0.01, |
| "intermediate_size": 7168, |
| "max_position_embeddings": 2048, |
| "mlp_bias": false, |
| "num_attention_heads": 16, |
| "num_hidden_layers": 32, |
| "num_key_value_heads": 8, |
| "pretraining_tp": 1, |
| "rms_norm_eps": 1e-05, |
| "rope_scaling": null, |
| "rope_theta": 10000.0, |
| "tie_word_embeddings": false, |
| "torch_dtype": "float32", |
| "transformers_version": "4.53.1", |
| "use_cache": true, |
| "vocab_size": 256006 |
| } |
|
|
