| base_model: cognitivecomputations/dolphin-2.1-mistral-7b | |
| base_model_config: cognitivecomputations/dolphin-2.1-mistral-7b | |
| model_type: MistralForCausalLM | |
| tokenizer_type: LlamaTokenizer | |
| is_mistral_derived_model: true | |
| load_in_8bit: false | |
| load_in_4bit: false | |
| strict: false | |
| datasets: | |
| - path: /workspace/datasets/prox/prox.json | |
| type: alpaca | |
| prompt_style: chatml | |
| - path: teknium/openhermes | |
| type: alpaca | |
| prompt_style: chatml | |
| hub_model_id: xenofold/prox-1.0-mistral-7b | |
| hf_use_auth_token: true | |
| dataset_prepared_path: last_run_prepared | |
| val_set_size: 0.005 | |
| output_dir: /workspace/prox-mistral-7b | |
| sequence_len: 8192 | |
| sample_packing: true | |
| pad_to_sequence_len: true | |
| wandb_project: prox | |
| wandb_entity: | |
| wandb_watch: | |
| wandb_run_id: | |
| wandb_log_model: | |
| gradient_accumulation_steps: 4 | |
| micro_batch_size: 6 | |
| num_epochs: 4 | |
| adam_beta2: 0.95 | |
| adam_epsilon: 0.00001 | |
| max_grad_norm: 1.0 | |
| lr_scheduler: cosine | |
| learning_rate: 0.000006 | |
| train_on_inputs: false | |
| group_by_length: false | |
| bf16: true | |
| fp16: false | |
| tf32: false | |
| gradient_checkpointing: true | |
| early_stopping_patience: | |
| resume_from_checkpoint: | |
| local_rank: | |
| logging_steps: 1 | |
| xformers_attention: | |
| flash_attention: true | |
| warmup_steps: 100 | |
| eval_steps: 0.05 | |
| eval_table_size: | |
| eval_table_max_new_tokens: | |
| save_steps: | |
| debug: | |
| deepspeed: deepspeed/zero2.json | |
| weight_decay: 0.1 | |
| fsdp: | |
| fsdp_config: | |
| special_tokens: | |
| bos_token: "<s>" | |
| eos_token: "<|im_end|>" | |
| unk_token: "<unk>" | |
| tokens: | |
| - "<|im_start|>" | |
| - "<|im_end|>" |