Views
No views yet
batch_size_training: '16'
checkpoint_type: StateDictType.FULL_STATE_DICT
dataset: orca_dolphin_100k_gpt4
dist_checkpoint_folder: fine-tuned
dist_checkpoint_root_folder: model_checkpoints
enable_fsdp: 'True'
freeze_layers: 'False'
fsdp_activation_checkpointing: 'True'
gamma: '0.85'
low_cpu_fsdp: 'True'
lr: 1e-05
micro_batch_size: '16'
mixed_precision: 'True'
model_name: meta-llama/Llama-2-70b-chat-hf
num_epochs: '1'
num_freeze_layers: '1'
num_workers_dataloader: '1'
one_gpu: 'False'
optimizer: anyprecision
output_dir: ~/llama-recipes-70b/output
peft_method: lora
pure_bf16: 'True'
quantization: 'False'
run_validation: 'True'
save_model: 'True'
save_optimizer: 'True'
seed: '42'
sharding_strategy: ShardingStrategy.FULL_SHARD
use_fast_kernels: (False,)
use_fp16: 'False'
use_peft: 'False'
val_batch_size: '16'
weight_decay: '0.0'