| { | |
| "architectures": [ | |
| "LlamaForCausalLM" | |
| ], | |
| "attention_bias": false, | |
| "attention_dropout": 0.0, | |
| "bos_token_id": 128000, | |
| "custom_pipelines": { | |
| "quizbowl-tossup": { | |
| "default": { | |
| "model": { | |
| "pt": [ | |
| "yujiepan/llama-3-tiny-random", | |
| "main" | |
| ] | |
| } | |
| }, | |
| "impl": "simple_tiny_tossup.QBTossupPipeline", | |
| "pt": [ | |
| "AutoModelForCausalLM" | |
| ], | |
| "tf": [ | |
| "TFAutoModelForCausalLM" | |
| ], | |
| "type": "text" | |
| } | |
| }, | |
| "eos_token_id": 128001, | |
| "head_dim": 2, | |
| "hidden_act": "silu", | |
| "hidden_size": 4, | |
| "initializer_range": 0.02, | |
| "intermediate_size": 14, | |
| "max_position_embeddings": 8192, | |
| "mlp_bias": false, | |
| "model_type": "llama", | |
| "num_attention_heads": 2, | |
| "num_hidden_layers": 2, | |
| "num_key_value_heads": 1, | |
| "pretraining_tp": 1, | |
| "rms_norm_eps": 1e-05, | |
| "rope_scaling": null, | |
| "rope_theta": 500000.0, | |
| "tie_word_embeddings": false, | |
| "torch_dtype": "float32", | |
| "transformers_version": "4.52.3", | |
| "use_cache": true, | |
| "vocab_size": 128256 | |
| } | |