9f824ef76a
* simplify the example configs to be more minimal and less daunting * drop empty s2_attention from example yamls
56 lines
1.2 KiB
YAML
56 lines
1.2 KiB
YAML
base_model: tiiuae/falcon-7b
|
|
# optionally might have model_type or tokenizer_type
|
|
model_type: AutoModelForCausalLM
|
|
tokenizer_type: AutoTokenizer
|
|
# Automatically upload checkpoint and final model to HF
|
|
# hub_model_id: username/custom_model_name
|
|
|
|
# required by falcon custom model code: https://huggingface.co/tiiuae/falcon-7b/tree/main
|
|
trust_remote_code: true
|
|
gptq: false
|
|
strict: false
|
|
push_dataset_to_hub:
|
|
datasets:
|
|
- path: teknium/GPT4-LLM-Cleaned
|
|
type: alpaca:chat
|
|
dataset_prepared_path:
|
|
val_set_size: 0.05
|
|
adapter:
|
|
lora_model_dir:
|
|
sequence_len: 2048
|
|
max_packed_sequence_len:
|
|
lora_r: 64
|
|
lora_alpha: 32
|
|
lora_dropout: 0.0
|
|
lora_target_linear: true
|
|
wandb_project:
|
|
wandb_entity:
|
|
wandb_watch:
|
|
wandb_name:
|
|
wandb_log_model:
|
|
output_dir: ./outputs/falcon-7b
|
|
batch_size: 2
|
|
micro_batch_size: 1
|
|
num_epochs: 4
|
|
optimizer: adamw_bnb_8bit
|
|
torchdistx_path:
|
|
lr_scheduler: cosine
|
|
learning_rate: 0.00003
|
|
bf16: auto
|
|
tf32: true
|
|
gradient_checkpointing: true
|
|
resume_from_checkpoint:
|
|
logging_steps: 1
|
|
xformers_attention: true
|
|
flash_attention:
|
|
gptq_groupsize:
|
|
gptq_model_v1:
|
|
warmup_steps: 40
|
|
evals_per_epoch: 4
|
|
saves_per_epoch: 1
|
|
weight_decay: 0.0
|
|
special_tokens:
|
|
pad_token: "<|endoftext|>"
|
|
bos_token: "<|endoftext|>"
|
|
eos_token: "<|endoftext|>"
|