base_model: deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B
# 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

trust_remote_code: true

load_in_8bit: false
load_in_4bit: true
strict: false

datasets:
  - path: length_human_train.jsonl
    type: alpaca
dataset_prepared_path:
val_set_size: 0.05
output_dir: /data/user_data/jiewenh/saved_models/DeepSeek-R1-Distill-Qwen-1.5B_test

sequence_len: 2048  
sample_packing: false
pad_to_sequence_len:

adapter: qlora
lora_model_dir:
lora_r: 32
lora_alpha: 16
lora_dropout: 0.05
lora_target_linear: true
lora_fan_in_fan_out:

wandb_project: "DeepSeek-R1-Distill-Qwen-1.5B_test"
wandb_entity:
wandb_watch:
wandb_name:
wandb_log_model:

gradient_accumulation_steps: 2
micro_batch_size: 1
num_epochs: 1
optimizer: adamw_bnb_8bit
lr_scheduler: cosine
learning_rate: 0.0002

train_on_inputs: false
group_by_length: false
bf16: auto
fp16:
tf32: false

gradient_checkpointing: false
early_stopping_patience:
resume_from_checkpoint:
local_rank:
logging_steps: 1
xformers_attention:
flash_attention:

warmup_steps: 10
evals_per_epoch: 0
eval_table_size:
eval_max_new_tokens: 128
saves_per_epoch: 1
debug:
deepspeed:
weight_decay: 0.0
fsdp:
fsdp_config:
special_tokens:

Improver-DeepSeek-R1-Distill-Qwen-1.5B_test

This model is a fine-tuned version of deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B on the length_human_train.jsonl dataset. It achieves the following results on the evaluation set:

Loss: 0.2914

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: 1
eval_batch_size: 1
seed: 42
gradient_accumulation_steps: 2
total_train_batch_size: 2
optimizer: Use OptimizerNames.ADAMW_BNB 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: 10
num_epochs: 1

Training results

Training Loss	Epoch	Step	Validation Loss
0.1058	0.9996	1379	0.2914

Framework versions

PEFT 0.14.0
Transformers 4.47.1
Pytorch 2.5.1
Datasets 3.1.0
Tokenizers 0.21.0

nutPace/Improver-DeepSeek-R1-Distill-Qwen-1.5B

Improver-DeepSeek-R1-Distill-Qwen-1.5B_test

Model description

Intended uses & limitations

Training and evaluation data

Training procedure

Training hyperparameters

Training results

Framework versions

nutPace/
Improver-DeepSeek-R1-Distill-Qwen-1.5B