| { | |
| "architectures": [ | |
| "GPTNeoXForCausalLM" | |
| ], | |
| "bos_token_id": 0, | |
| "eos_token_id": 0, | |
| "hidden_act": "gelu", | |
| "hidden_size": 512, | |
| "initializer_range": 0.02, | |
| "intermediate_size": 2048, | |
| "layer_norm_eps": 1e-05, | |
| "max_position_embeddings": 2048, | |
| "model_type": "gpt_neox", | |
| "num_attention_heads": 8, | |
| "num_hidden_layers": 6, | |
| "rotary_emb_base": 10000, | |
| "rotary_pct": 0.25, | |
| "tie_word_embeddings": false, | |
| "torch_dtype": "float16", | |
| "transformers_version": "4.24.0", | |
| "use_cache": true, | |
| "welcome": "\n\nYou are currently running the model `Concedo_Pythia-70M-ChatSalad`\n\nThis is a follow up finetune of Pythia-70M finetuned on the same dataset as OPT-19M-ChatSalad. It is much more coherent.\n\n", | |
| "use_parallel_residual": true, | |
| "vocab_size": 50278 | |
| } | |