Text Generation
Transformers
Safetensors
mistral
openchat
C-RLFT
conversational
text-generation-inference
4-bit precision
openchat-3.5-1210-GPTQ / openchat.json
TheBloke's picture
GPTQ model commit
fb66df8
raw history blame
No virus
539 Bytes
{"local_rank": 0, "model_path": "imone/Mistral_7B_with_EOT_token", "data_prefix": "", "save_path": "", "save_every": 1, "batch_max_len": 77824, "epochs": 5, "lr": 1.2141624862852263e-05, "lr_min_ratio": 0.1, "lr_warmup_ratio": 0.05, "weight_decay": 0.1, "beta1": 0.9, "beta2": 0.95, "eps": 1e-05, "deepspeed": true, "deepspeed_config": "ochat/training_deepspeed/deepspeed_config.json", "deepscale": false, "deepscale_config": null, "deepspeed_mpi": false, "model_type": "openchat_v3.2_mistral", "device": "<non-serializable>", "epoch": 3}