sharpenb's picture
7657f03b65fa534e57255511cb43b63f60ac673413ea8a8586ead71a2c61bb3d
0d47b82 verified
raw
history blame
932 Bytes
{
"api_key": null,
"verify_url": "http://johnrachwan.pythonanywhere.com",
"smash_config": {
"pruners": "None",
"factorizers": "None",
"quantizers": "['llm-int8']",
"compilers": "None",
"task": "text_text_generation",
"device": "cuda",
"cache_dir": "/ceph/hdd/staff/charpent/.cache/models0xj90l2n",
"batch_size": 1,
"model_name": "Nondzu/zephyr-speakleash-010-pl-3072-32-16-0.01",
"pruning_ratio": 0.0,
"n_quantization_bits": 4,
"output_deviation": 0.005,
"max_batch_size": 1,
"qtype_weight": "torch.qint8",
"qtype_activation": "torch.quint8",
"qobserver": "<class 'torch.ao.quantization.observer.MinMaxObserver'>",
"qscheme": "torch.per_tensor_symmetric",
"qconfig": "x86",
"group_size": 128,
"damp_percent": 0.1,
"save_load_fn": "bitsandbytes"
}
}