metadata
library_name: peft
license: apache-2.0
base_model: EleutherAI/pythia-160m
tags:
- axolotl
- generated_from_trainer
model-index:
- name: 1049f57e-4e7b-4d20-8524-075e12b98361
results: []
See axolotl config
axolotl version: 0.4.1
adapter: lora
base_model: EleutherAI/pythia-160m
bf16: auto
chat_template: llama3
dataset_prepared_path: null
datasets:
- data_files:
- cabe0edc52f1824b_train_data.json
ds_type: json
format: custom
path: /workspace/input_data/cabe0edc52f1824b_train_data.json
type:
field_input: policy
field_instruction: redteam_query
field_output: jailbreak_query
format: '{instruction} {input}'
no_input_format: '{instruction}'
system_format: '{system}'
system_prompt: ''
debug: null
deepspeed: null
device_map: auto
do_eval: true
early_stopping_patience: 3
eval_batch_size: 4
eval_max_new_tokens: 128
eval_steps: 200
eval_table_size: null
evals_per_epoch: null
flash_attention: true
fp16: false
fsdp: null
fsdp_config: null
gradient_accumulation_steps: 4
gradient_checkpointing: false
group_by_length: true
hub_model_id: great0001/1049f57e-4e7b-4d20-8524-075e12b98361
hub_strategy: end
learning_rate: 0.0002
load_in_4bit: false
load_in_8bit: false
local_rank: null
logging_steps: 50
lora_alpha: 64
lora_dropout: 0.05
lora_fan_in_fan_out: null
lora_model_dir: null
lora_r: 32
lora_target_linear: true
lr_scheduler: constant
max_grad_norm: 1.0
max_memory:
0: 75GB
max_steps: 17977
micro_batch_size: 4
mlflow_experiment_name: /tmp/cabe0edc52f1824b_train_data.json
model_type: AutoModelForCausalLM
num_epochs: 10
optim_args:
adam_beta1: 0.9
adam_beta2: 0.95
adam_epsilon: 1e-5
optimizer: adamw_bnb_8bit
output_dir: miner_id_24
pad_to_sequence_len: true
resume_from_checkpoint: null
s2_attention: null
sample_packing: false
save_steps: 200
saves_per_epoch: null
sequence_len: 512
special_tokens:
pad_token: <|endoftext|>
strict: false
tf32: true
tokenizer_type: AutoTokenizer
train_on_inputs: false
trust_remote_code: true
val_set_size: 0.05
wandb_entity: null
wandb_mode: online
wandb_name: 2677b7fe-e0a2-439d-a729-54f9c5263ad0
wandb_project: SN56-33
wandb_run: your_name
wandb_runid: 2677b7fe-e0a2-439d-a729-54f9c5263ad0
warmup_steps: 50
weight_decay: 0.0
xformers_attention: null
1049f57e-4e7b-4d20-8524-075e12b98361
This model is a fine-tuned version of EleutherAI/pythia-160m on the None dataset. It achieves the following results on the evaluation set:
- Loss: 2.3410
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: 0.0002
- train_batch_size: 4
- eval_batch_size: 4
- seed: 42
- gradient_accumulation_steps: 4
- total_train_batch_size: 16
- optimizer: Use OptimizerNames.ADAMW_BNB with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=adam_beta1=0.9,adam_beta2=0.95,adam_epsilon=1e-5
- lr_scheduler_type: constant
- lr_scheduler_warmup_steps: 50
- training_steps: 16614
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| No log | 0.0006 | 1 | 5.0096 |
| 19.4463 | 0.1204 | 200 | 5.1624 |
| 16.3716 | 0.2407 | 400 | 3.9670 |
| 15.4725 | 0.3611 | 600 | 4.3231 |
| 13.6185 | 0.4815 | 800 | 4.5737 |
| 13.2182 | 0.6019 | 1000 | 3.3886 |
| 11.9674 | 0.7222 | 1200 | 3.3497 |
| 11.3688 | 0.8426 | 1400 | 3.2310 |
| 12.5514 | 0.9630 | 1600 | 3.9830 |
| 10.2903 | 1.0835 | 1800 | 2.4156 |
| 8.989 | 1.2039 | 2000 | 2.3129 |
| 9.2686 | 1.3243 | 2200 | 2.4183 |
| 9.0286 | 1.4446 | 2400 | 2.4323 |
| 9.8924 | 1.5650 | 2600 | 2.3410 |
Framework versions
- PEFT 0.13.2
- Transformers 4.46.0
- Pytorch 2.5.0+cu124
- Datasets 3.0.1
- Tokenizers 0.20.1