Catastrophic forgetting test results:
Initial evaluation loss on 1k subset of HuggingFaceTB/cosmopedia-100k dataset was 1.615, significantly more than tuning methods with smaller adaptations.
100 steps of LISA training reduced this to 1.392.
Comparison to control: cosmo-1b started out with 1.003 loss on (a different subset of) dataset, increasing to 1.024 at 100 steps.
axolotl version: 0.4.0
base_model: HuggingFaceTB/cosmo-1b
model_type: LlamaForCausalLM
tokenizer_type: LlamaTokenizer
load_in_8bit: false
load_in_4bit: false
strict: false
datasets:
- path: Vezora/Tested-22k-Python-Alpaca
type: alpaca
dataset_prepared_path: prepared-tune
val_set_size: 0.05
output_dir: ./tune-out
sequence_len: 2048
sample_packing: true
pad_to_sequence_len: true
adapter:
lora_model_dir:
lora_r:
lora_alpha:
lora_dropout:
lora_target_linear:
lora_fan_in_fan_out:
wandb_project: cosmo-python-tune
wandb_entity:
wandb_watch:
wandb_name:
wandb_log_model:
gradient_accumulation_steps: 4
micro_batch_size: 1
num_epochs: 1
optimizer: adamw_bnb_8bit
lr_scheduler: cosine
learning_rate: 0.0005
train_on_inputs: false
group_by_length: false
bf16: auto
fp16:
tf32: false
gradient_checkpointing: true
early_stopping_patience:
resume_from_checkpoint:
local_rank:
logging_steps: 1
xformers_attention:
flash_attention: true
warmup_steps: 10
evals_per_epoch: 4
saves_per_epoch: 1
debug:
deepspeed:
weight_decay: 0.0
fsdp:
fsdp_config:
special_tokens:
tune-out
This model is a fine-tuned version of HuggingFaceTB/cosmo-1b on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.2049
Model description
More information needed
Intended uses & limitations
More information needed
Training and evaluation data
More information needed
Training procedure
Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 0.0005
- train_batch_size: 1
- eval_batch_size: 1
- seed: 42
- distributed_type: multi-GPU
- num_devices: 2
- gradient_accumulation_steps: 4
- total_train_batch_size: 8
- total_eval_batch_size: 2
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: cosine
- lr_scheduler_warmup_steps: 10
- num_epochs: 1
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| 0.6623 | 0.0 | 1 | 0.6460 |
| 0.6503 | 0.25 | 238 | 0.6117 |
| 0.534 | 0.5 | 476 | 0.3380 |
| 0.3682 | 0.75 | 714 | 0.2049 |
Framework versions
- Transformers 4.40.0.dev0
- Pytorch 2.1.2+cu118
- Datasets 2.18.0
- Tokenizers 0.15.0