| { |
| "_name_or_path": "stabilityai/stablelm-3b-4e1t", |
| "architectures": [ |
| "StableLMEpochForCausalLM" |
| ], |
| "attention_probs_dropout_prob": 0.0, |
| "auto_map": { |
| "AutoConfig": "configuration_stablelm_epoch.StableLMEpochConfig", |
| "AutoModelForCausalLM": "modeling_stablelm_epoch.StableLMEpochForCausalLM" |
| }, |
| "bos_token_id": 0, |
| "custom_pipelines": { |
| "text-generation": { |
| "impl": "h2oai_pipeline.H2OTextGenerationPipeline", |
| "pt": "AutoModelForCausalLM" |
| } |
| }, |
| "eos_token_id": 0, |
| "hidden_act": "silu", |
| "hidden_dropout_prob": 0.0, |
| "hidden_size": 2560, |
| "initializer_range": 0.02, |
| "intermediate_size": 6912, |
| "max_position_embeddings": 4096, |
| "model_type": "stablelm_epoch", |
| "norm_eps": 1e-05, |
| "num_attention_heads": 32, |
| "num_heads": 32, |
| "num_hidden_layers": 32, |
| "num_key_value_heads": 32, |
| "pad_token_id": 0, |
| "rope_pct": 0.25, |
| "rope_theta": 10000, |
| "rotary_scaling_factor": 1.0, |
| "tie_word_embeddings": false, |
| "torch_dtype": "float16", |
| "transformers_version": "4.34.0", |
| "use_cache": true, |
| "vocab_size": 50304 |
| } |
|
|
