| { |
| "architectures": [ |
| "HelionForCausalLM" |
| ], |
| "model_type": "helion", |
| "torch_dtype": "bfloat16", |
| "transformers_version": "4.40.0", |
| "vocab_size": 32768, |
| "hidden_size": 4096, |
| "intermediate_size": 14336, |
| "num_hidden_layers": 32, |
| "num_attention_heads": 32, |
| "num_key_value_heads": 8, |
| "hidden_act": "silu", |
| "max_position_embeddings": 8192, |
| "initializer_range": 0.02, |
| "rms_norm_eps": 1e-6, |
| "use_cache": true, |
| "pad_token_id": null, |
| "bos_token_id": 1, |
| "eos_token_id": 2, |
| "tie_word_embeddings": false, |
| "rope_theta": 10000.0, |
| "rope_scaling": { |
| "type": "linear", |
| "factor": 1.0 |
| }, |
| "attention_bias": false, |
| "attention_dropout": 0.0, |
| "mlp_bias": false, |
| "residual_dropout": 0.0, |
| "embedding_dropout": 0.0, |
| "use_sliding_window": false, |
| "sliding_window": null, |
| "use_flash_attention_2": true, |
| "pretraining_tp": 1, |
|
|
| "use_memory_efficient_attention": true, |
| "gradient_checkpointing": false, |
| "layer_norm_type": "rms_norm", |
| "activation_function": "silu", |
| "rotary_embedding_base": 10000, |
| "rotary_pct": 1.0, |
| "tie_encoder_decoder": false, |
| "is_encoder_decoder": false, |
| "cross_attention_hidden_size": null, |
| "add_cross_attention": false, |
| "chunk_size_feed_forward": 0, |
| "output_attentions": false, |
| "output_hidden_states": false, |
| "use_return_dict": true, |
| "torchscript": false, |
| "use_bfloat16": true, |
| "pruned_heads": {}, |
| "tokenizer_class": "PreTrainedTokenizerFast", |
| "auto_map": { |
| "AutoConfig": "configuration_helion.HelionConfig", |
| "AutoModelForCausalLM": "modeling_helion.HelionForCausalLM" |
| } |
| } |
