CoolFace
Modelpublic

datapaf/ft_starcoder2_racket

sourceHugging Faceupdated 1y agoView on Hugging Face
0likes15downloads
config.json32 linesDownload Raw Back to root
1{2  "_name_or_path": "Finetune/ft_starcoder2_racket",3  "architectures": [4    "Starcoder2ForCausalLM"5  ],6  "attention_dropout": 0.1,7  "bos_token_id": 0,8  "embedding_dropout": 0.1,9  "eos_token_id": 0,10  "hidden_act": "gelu_pytorch_tanh",11  "hidden_size": 3072,12  "initializer_range": 0.018042,13  "intermediate_size": 12288,14  "max_position_embeddings": 16384,15  "mlp_type": "default",16  "model_type": "starcoder2",17  "norm_epsilon": 1e-05,18  "norm_type": "layer_norm",19  "num_attention_heads": 24,20  "num_hidden_layers": 30,21  "num_key_value_heads": 2,22  "residual_dropout": 0.1,23  "rope_scaling": null,24  "rope_theta": 999999.4420358813,25  "sliding_window": 4096,26  "torch_dtype": "bfloat16",27  "transformers_version": "4.49.0",28  "use_bias": true,29  "use_cache": true,30  "vocab_size": 4915231}32