metadata
base_model: EleutherAI/pythia-160m-deduped
library_name: peft
license: apache-2.0
tags:
- axolotl
- generated_from_trainer
model-index:
- name: pythia-160m-storytelling
results: []
See axolotl config
axolotl version: 0.4.1
base_model: EleutherAI/pythia-160m-deduped
load_in_8bit: false
datasets:
- path: jtatman/storywriting_combined_instruct
type: alpaca
dataset_prepared_path: ds-storytelling
val_set_size: 0.05
adapter: lora
lora_model_dir:
sequence_len: 2048
lora_r: 16
lora_alpha: 64
lora_dropout: 0.05
lora_target_modules:
- query_key_value
lora_target_linear:
lora_fan_in_fan_out: true # pythia/GPTNeoX lora specific
wandb_project: pythia
wandb_entity:
wandb_watch:
wandb_name: pythia-160m-storytelling
wandb_log_model:
output_dir: ./outputs/lora-alpaca-pythia-160m-storytelling
gradient_accumulation_steps: 16
micro_batch_size: 1
num_epochs: 5
learning_rate: 0.0006
lr_scheduler: cosine_with_restarts
#cosine_min_lr_ratio: 0.1
train_on_inputs: false
group_by_length: false
#bf16: auto
#fp16: true
#tf32: false
float16: true
flash_attn:
xformers_attention: true
optimizer: paged_adamw_8bit
gpu_memory_limit: 8GiB
hub_model_id: jtatman/pythia-160m-storytelling
lora_on_cpu: true
early_stopping_patience: 3
#resume_from_checkpoint: outputs/lora-alpaca-pythia-125m/checkpoint-51040
auto_resume_from_checkpoints: true
local_rank:
weight_decay: 0.1
chat_template: inst
#evals_per_epoch: 4
eval_steps: 2000
logging_steps: 1
save_steps: 2000
save_total_limit: 5
warmup_steps: 1000
pythia-160m-storytelling
This model is a fine-tuned version of EleutherAI/pythia-160m-deduped on the None dataset. It achieves the following results on the evaluation set:
- Loss: 10.3843
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: 0.0006
- train_batch_size: 1
- eval_batch_size: 1
- seed: 42
- gradient_accumulation_steps: 16
- total_train_batch_size: 16
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: cosine_with_restarts
- lr_scheduler_warmup_steps: 1000
- num_epochs: 5
Training results
Training Loss | Epoch | Step | Validation Loss |
---|---|---|---|
5.4891 | 0.0012 | 1 | 4.5640 |
8.4799 | 2.4467 | 2000 | 9.1436 |
9.9198 | 4.8944 | 4000 | 10.3843 |
Framework versions
- PEFT 0.11.1
- Transformers 4.41.2
- Pytorch 2.3.0+cu121
- Datasets 2.19.1
- Tokenizers 0.19.1