Upload phi-samantha.yml

axolotl-config/phi-samantha.yml

@@ -0,0 +1,76 @@
+base_model: microsoft/phi-1_5
+base_model_config: microsoft/phi-1_5
+model_type: AutoModelForCausalLM
+tokenizer_type: AutoTokenizer
+is_llama_derived_model: false
+trust_remote_code: true
+
+load_in_8bit: false
+load_in_4bit: false
+strict: false
+
+datasets:
+  - path: ehartford/samantha-data
+    type: sharegpt:chat
+    data_files: samantha-1.1.json
+
+dataset_prepared_path: last_run_prepared
+val_set_size: 0.01
+output_dir: /workspace/samantha-phi
+
+sequence_len: 2048
+sample_packing: false
+pad_to_sequence_len: 
+
+adapter:
+lora_model_dir:
+lora_r:
+lora_alpha:
+lora_dropout:
+lora_target_linear:
+lora_fan_in_fan_out:
+
+wandb_project: samantha-phi
+wandb_entity:
+wandb_watch:
+wandb_run_id:
+wandb_log_model:
+
+gradient_accumulation_steps: 1
+micro_batch_size: 12
+num_epochs: 2000
+optimizer: adamw_bnb_8bit
+adam_beta2: 0.99
+adam_epsilon: 0.00001
+max_grad_norm: 1.0
+lr_scheduler: cosine
+learning_rate: 0.0002
+
+train_on_inputs: false
+group_by_length: true
+bf16: true
+fp16: false
+tf32: true
+
+gradient_checkpointing: 
+early_stopping_patience:
+resume_from_checkpoint:
+local_rank:
+logging_steps: 1
+xformers_attention: true
+flash_attention: 
+
+warmup_steps: 100
+eval_steps: 0.05
+save_steps:
+debug:
+deepspeed: deepspeed/zero2.json
+weight_decay: 0.01
+fsdp:
+fsdp_config:
+resize_token_embeddings_to_32x: true
+special_tokens:
+  bos_token: "<|endoftext|>"
+  eos_token: "<|endoftext|>"
+  unk_token: "<|endoftext|>"
+  pad_token: "<|pad|>"