projecte-aina/Plume128k
115
1{2 "_name_or_path": "/gpfs/projects/bsc88/mt_translation/Parallel_LLM/training/saved_checkpoints/gemma128_distributed/checkpoint-570000",3 "architectures": [4 "GemmaForCausalLM"5 ],6 "attention_bias": false,7 "attention_dropout": 0.0,8 "bos_token_id": 0,9 "eos_token_id": 1,10 "head_dim": 256,11 "hidden_act": "gelu_pytorch_tanh",12 "hidden_activation": "gelu_pytorch_tanh",13 "hidden_size": 2048,14 "initializer_range": 0.02,15 "intermediate_size": 16384,16 "max_position_embeddings": 8192,17 "model_type": "gemma",18 "num_attention_heads": 8,19 "num_hidden_layers": 18,20 "num_key_value_heads": 1,21 "pad_token_id": 3,22 "rms_norm_eps": 1e-06,23 "rope_theta": 10000.0,24 "torch_dtype": "float32",25 "transformers_version": "4.41.2",26 "use_cache": true,27 "vocab_size": 12800028}29 