base_model: LnL-AI/dbrx-base-converted-v2
trust_remote_code: true

load_in_8bit: false
load_in_4bit: false
strict: false

datasets:
  - path: tatsu-lab/alpaca
    type: alpaca
dataset_prepared_path: last_run_prepared
val_set_size: 0.0
output_dir: ./out

sequence_len: 512
sample_packing: false
pad_to_sequence_len: false

wandb_project:
wandb_entity:
wandb_watch:
wandb_name:
wandb_log_model:

adapter: lora
lora_model_dir:
lora_r: 8
lora_alpha: 16
lora_dropout: 0.05
# w1, w2, & v1 will hang the trainer
lora_target_modules:
  - q_proj # attn
  - k_proj # attn
  - v_proj # attn
  - out_proj # attn
  - layer # router
#  - w1
#  - w2
#  - v1

gradient_accumulation_steps: 1
micro_batch_size: 1
num_epochs: 1
optimizer: paged_adamw_8bit
lr_scheduler: cosine
learning_rate: 0.0002

train_on_inputs: false
group_by_length: false
bf16: auto
fp16:
tf32: false

gradient_checkpointing: false  # don't use with fsdp_activation_checkpointing
gradient_checkpointing_kwargs:
  use_reentrant: false
early_stopping_patience:
resume_from_checkpoint:
local_rank:
logging_steps: 1
xformers_attention:
flash_attention: true

warmup_steps: 10
evals_per_epoch:
saves_per_epoch: 1
debug:
weight_decay: 0.0
fsdp:
  - full_shard
  - auto_wrap
fsdp_config:
  fsdp_limit_all_gathers: true
  fsdp_sync_module_states: true
  fsdp_offload_params: false
  fsdp_use_orig_params: false
  fsdp_cpu_ram_efficient_loading: true
  fsdp_auto_wrap_policy: TRANSFORMER_BASED_WRAP
  fsdp_transformer_layer_cls_to_wrap: DbrxBlock
  fsdp_state_dict_type: FULL_STATE_DICT
  fsdp_activation_checkpointing: true