llmtf_eval_k5/darumeru_cp_sent_en_params.jsonl

{
    "custom_generation_config": null,
    "model_params": {
        "model_name_or_path": "NousResearch/Meta-Llama-3-8B",
        "generation_config": {
            "bos_token_id": 128000,
            "do_sample": true,
            "eos_token_id": [
                128001,
                198,
                271
            ],
            "max_length": 8192,
            "max_new_tokens": 128,
            "pad_token_id": 128001,
            "stop_strings": [
                "\n",
                "\n\n"
            ],
            "temperature": 0.1,
            "top_k": 40,
            "top_p": 0.9,
            "transformers_version": "4.38.2",
            "trust_remote_code": [
                false
            ]
        },
        "conversation_template": {
            "system_message_template": "{content}\n",
            "user_message_template": "{content}\n",
            "bot_message_template": "{content}\n\n",
            "bot_message_template_incomplete": "{content}",
            "user_role": "user",
            "bot_role": "bot",
            "system_role": "system",
            "suffix": "",
            "add_special_tokens": false,
            "eos_token": [
                "\n",
                "\n\n"
            ],
            "global_prefix": "<|begin_of_text|>"
        },
        "load_in_8bit": false,
        "torch_dtype": "auto",
        "use_flash_attention_2": true,
        "device_map": "cuda:0",
        "use_fast_tokenizer": true,
        "leading_space": false,
        "space_token": null,
        "trust_remote_code": [
            false
        ],
        "max_model_len": 8192
    },
    "task_params": {
        "max_len": 4000,
        "few_shot_count": 5,
        "batch_size": 8,
        "max_sample_per_dataset": 10000000000000,
        "method": "generate"
    }
}