|
{ |
|
"custom_generation_config": null, |
|
"model_params": { |
|
"model_name_or_path": "/scratch/tikhomirov/workdir/data/models/ruadapt_qwen2.5_3B_ext_u48_full_lr3e4_bs256", |
|
"generation_config": { |
|
"do_sample": true, |
|
"eos_token_id": [ |
|
147075, |
|
198, |
|
271 |
|
], |
|
"max_length": 32768, |
|
"max_new_tokens": 1, |
|
"pad_token_id": 147075, |
|
"stop_strings": [ |
|
"\n", |
|
"\n\n" |
|
], |
|
"temperature": 0.1, |
|
"top_k": 40, |
|
"top_p": 0.9, |
|
"transformers_version": "4.45.2", |
|
"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": "" |
|
}, |
|
"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": 32768 |
|
}, |
|
"task_params": { |
|
"max_len": 4000, |
|
"few_shot_count": 5, |
|
"batch_size": 8, |
|
"max_sample_per_dataset": 500, |
|
"method": "calculate_logsoftmax" |
|
} |
|
} |
|
|