| { | |
| "activation_function": "gelu_fast", | |
| "attn_pdrop": 0.1, | |
| "auto_map": { | |
| "AutoConfig": "configuration_flm.FLMConfig", | |
| "AutoModel": "modeling_flm.FLM", | |
| "AutoModelForCausalLM": "modeling_flm.FLM" | |
| }, | |
| "bos_token_id": 100256, | |
| "cls_token_id": 100351, | |
| "embd_pdrop": 0.1, | |
| "eos_token_id": 100256, | |
| "initializer_range": 0.02, | |
| "input_mult": 1.0, | |
| "layer_norm_epsilon": 1e-05, | |
| "model_type": "flm", | |
| "mup_base_width": 256, | |
| "mup_scale_factor": 40.0, | |
| "n_embd": 10240, | |
| "n_head": 80, | |
| "n_inner": 40960, | |
| "n_layer": 80, | |
| "n_positions": 2048, | |
| "output_mult": 2.0, | |
| "relative_encoding": "rotary", | |
| "reorder_and_upcast_attn": true, | |
| "resid_pdrop": 0.1, | |
| "rotary_theta": 10000, | |
| "rotary_use_xpos": true, | |
| "rotary_xpos_scale_base": 512, | |
| "scale_attn_by_inverse_layer_idx": true, | |
| "scale_attn_weights": true, | |
| "summary_activation": null, | |
| "summary_first_dropout": 0.1, | |
| "summary_proj_to_labels": true, | |
| "summary_type": "cls_index", | |
| "summary_use_proj": true, | |
| "tokenizer_class": "FLMTokenizer", | |
| "transformers_version": "4.30.2", | |
| "use_cache": true, | |
| "use_mup": true, | |
| "vocab_size": 100352 | |
| } | |