{ "activation_function": "swiglu", "architectures": [ "GPT2CustomLMHeadModel" ], "attn_pdrop": 0.1, "bos_token_id": 2, "dtype": "float32", "embd_pdrop": 0.1, "eos_token_id": 3, "initializer_range": 0.02, "layer_norm_epsilon": 1e-05, "model_type": "gpt2_custom", "n_embd": 512, "n_head": 8, "n_inner": 1360, "n_layer": 12, "n_positions": 512, "resid_pdrop": 0.1, "transformers_version": "5.14.1", "vocab_size": 16384, "auto_map": { "AutoConfig": "configuration_gpt2.GPT2CustomConfig", "AutoModelForCausalLM": "modeling_gpt2.GPT2CustomLMHeadModel" } }