base_model: Qwen/Qwen2.5-Coder-7B-Instruct
model_type: Qwen2ForCausalLM
tokenizer_type: AutoTokenizer

load_in_8bit: false
load_in_4bit: false

datasets:
  - path: felixwangg/prime_vul_plus_splitted
    type: chat_template
    split: train
test_datasets:
  - path: felixwangg/prime_vul_plus_splitted
    type: chat_template
    split: validation
dataset_prepared_path: /u901/t577wang/SecSteer-v2/axolotl-datasets/lora/Qwen2.5-Coder-7B/evol-33k-stage1-sec-stage2
val_set_size: 0
output_dir: /u901/t577wang/SecSteer-v2/axolotl-outputs/lora/Qwen2.5-Coder-7B-evol-33k-stage1-sec-stage2
sequence_len: 4096
sample_packing: false
eval_sample_packing: false
pad_to_sequence_len: true

adapter: lora
lora_model_dir: /u901/t577wang/SecSteer-v2/axolotl-outputs/lora/Qwen2.5-Coder-7B-evol-33k-stage1/checkpoint-424
lora_r: 16
lora_alpha: 16
lora_dropout: 0.05
lora_target_linear: true
merge_lora: false

wandb_project: sft-primevul-sweep-ctx-0
wandb_entity: wtkuan
wandb_watch: "false"
wandb_name: Qwen2.5-Coder-7B-evol-33k-stage1-sec-stage2
wandb_log_model: "false"


gradient_accumulation_steps: 8
micro_batch_size: 4
optimizer: adamw_torch
lr_scheduler: cosine
learning_rate: 4e-5

bf16: true
tf32: false

train_on_inputs: false
roles_to_train: ['assistant']

gradient_checkpointing: true
resume_from_checkpoint:
logging_steps: 1
flash_attention: true

num_epochs: 2
warmup_ratio: 0.1
early_stopping_patience: 1000
eval_steps: 15
save_steps: 15
save_total_limit: 1000
load_best_model_at_end: true

weight_decay: 0.02
special_tokens:

plugins:

u901/t577wang/SecSteer-v2/axolotl-outputs/lora/Qwen2.5-Coder-7B-evol-33k-stage1-sec-stage2

This model is a fine-tuned version of Qwen/Qwen2.5-Coder-7B-Instruct on the felixwangg/prime_vul_plus_splitted dataset. It achieves the following results on the evaluation set:

Loss: 0.6925
Ppl: 1.9988
Memory/max Active (gib): 38.14
Memory/max Allocated (gib): 38.14
Memory/device Reserved (gib): 51.89

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: 4e-05
train_batch_size: 4
eval_batch_size: 4
seed: 42
distributed_type: multi-GPU
num_devices: 2
gradient_accumulation_steps: 8
total_train_batch_size: 64
total_eval_batch_size: 8
optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
lr_scheduler_type: cosine
lr_scheduler_warmup_steps: 11
training_steps: 113

Training results

Training Loss	Epoch	Step	Validation Loss	Ppl	Active (gib)	Allocated (gib)	Reserved (gib)
No log	0	0	0.7698	2.1594	37.83	37.83	41.81
0.7781	0.2661	15	0.7261	2.0670	38.14	38.14	51.31
0.7302	0.5322	30	0.7072	2.0283	38.14	38.14	51.89
0.6842	0.7982	45	0.6995	2.0128	38.14	38.14	51.89
0.7187	1.0532	60	0.6958	2.0052	38.14	38.14	51.89
0.6903	1.3193	75	0.6937	2.0010	38.14	38.14	51.89
0.6705	1.5854	90	0.6928	1.9993	38.14	38.14	51.89
0.7083	1.8514	105	0.6925	1.9986	38.14	38.14	51.89
0.7126	1.9933	113	0.6925	1.9988	38.14	38.14	51.89

Framework versions

PEFT 0.19.1
Transformers 5.5.4
Pytorch 2.11.0+cu130
Datasets 4.5.0
Tokenizers 0.22.2

felixwangg/Qwen2.5-Coder-7B-evol-33k-stage1-sec-stage2

u901/t577wang/SecSteer-v2/axolotl-outputs/lora/Qwen2.5-Coder-7B-evol-33k-stage1-sec-stage2

Model description

Intended uses & limitations

Training and evaluation data

Training procedure

Training hyperparameters

Training results

Framework versions

felixwangg/
Qwen2.5-Coder-7B-evol-33k-stage1-sec-stage2