adapter: lora
base_model: Qwen/Qwen1.5-0.5B
bf16: auto
chat_template: llama3
dataset_prepared_path: null
datasets:
- data_files:
  - b905dad0ab75693b_train_data.json
  ds_type: json
  format: custom
  path: /workspace/input_data/b905dad0ab75693b_train_data.json
  type:
    field_input: keyphrases
    field_instruction: abstract
    field_output: title
    format: '{instruction} {input}'
    no_input_format: '{instruction}'
    system_format: '{system}'
    system_prompt: ''
debug: null
deepspeed: null
eval_max_new_tokens: 128
eval_table_size: null
evals_per_epoch: 5
flash_attention: false
fp16: null
fsdp: null
fsdp_config: null
gradient_accumulation_steps: 16
gradient_checkpointing: true
group_by_length: false
hub_model_id: tuantmdev/04bccd8b-88c1-4309-b6fa-36f50aef2750
hub_repo: null
hub_strategy: checkpoint
hub_token: null
learning_rate: 2e-05
load_in_4bit: false
load_in_8bit: true
local_rank: null
logging_steps: 40
lora_alpha: 16
lora_dropout: 0.05
lora_fan_in_fan_out: null
lora_model_dir: null
lora_r: 8
lora_target_linear: true
lr_scheduler: cosine
max_steps: 50
micro_batch_size: 2
mlflow_experiment_name: /tmp/b905dad0ab75693b_train_data.json
model_type: AutoModelForCausalLM
num_epochs: 1
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_strategy: best
saves_per_epoch: 5
sequence_len: 512
strict: false
tf32: false
tokenizer_type: AutoTokenizer
train_on_inputs: false
trust_remote_code: true
val_set_size: 0.05
wandb_entity: null
wandb_mode: online
wandb_name: e527b68e-58d5-41c2-b375-f52b4f34c77f
wandb_project: Gradients-On-Demand
wandb_run: unknown
wandb_runid: e527b68e-58d5-41c2-b375-f52b4f34c77f
warmup_steps: 10
weight_decay: 0.01
xformers_attention: null

04bccd8b-88c1-4309-b6fa-36f50aef2750

This model is a fine-tuned version of Qwen/Qwen1.5-0.5B on the None dataset. It achieves the following results on the evaluation set:

Loss: 2.5474

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: 2e-05
train_batch_size: 2
eval_batch_size: 2
seed: 42
gradient_accumulation_steps: 16
total_train_batch_size: 32
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
training_steps: 50

Training results

Training Loss	Epoch	Step	Validation Loss
No log	0.0187	1	3.2242
No log	0.1871	10	3.1651
No log	0.3743	20	2.8983
No log	0.5614	30	2.6618
2.8719	0.7485	40	2.5621
2.8719	0.9357	50	2.5474

Framework versions

PEFT 0.13.2
Transformers 4.46.0
Pytorch 2.5.0+cu124
Datasets 3.0.1
Tokenizers 0.20.1