See axolotl config

axolotl version: 0.4.1

adapter: qlora
auto_resume_from_checkpoints: true
base_model: heegyu/WizardVicuna2-13b-hf
bf16: auto
chat_template: llama3
dataloader_num_workers: 12
dataset_prepared_path: null
datasets:
- data_files:
  - afb1fdb32bdf02c6_train_data.json
  ds_type: json
  format: custom
  path: /workspace/input_data/afb1fdb32bdf02c6_train_data.json
  type:
    field_instruction: premise
    field_output: hypothesis
    format: '{instruction}'
    no_input_format: '{instruction}'
    system_format: '{system}'
    system_prompt: ''
debug: null
deepspeed: null
early_stopping_patience: 3
eval_max_new_tokens: 128
eval_steps: 100
eval_table_size: null
evals_per_epoch: null
flash_attention: false
fp16: false
fsdp: null
fsdp_config: null
gradient_accumulation_steps: 4
gradient_checkpointing: true
group_by_length: true
hub_model_id: error577/d1526c12-8a5c-45c2-8a7c-e005e4428b34
hub_repo: null
hub_strategy: checkpoint
hub_token: null
learning_rate: 0.0002
load_in_4bit: true
load_in_8bit: false
local_rank: null
logging_steps: 10
lora_alpha: 128
lora_dropout: 0.05
lora_fan_in_fan_out: null
lora_model_dir: null
lora_r: 32
lora_target_linear: true
lr_scheduler: cosine
max_grad_norm: 1.0
max_steps: null
micro_batch_size: 1
mlflow_experiment_name: /tmp/afb1fdb32bdf02c6_train_data.json
model_type: AutoModelForCausalLM
num_epochs: 3
optimizer: adamw_torch_4bit
output_dir: miner_id_24
pad_to_sequence_len: true
resume_from_checkpoint: null
s2_attention: null
sample_packing: false
save_steps: 100
sequence_len: 512
special_tokens:
  pad_token: </s>
strict: false
tf32: false
tokenizer_type: AutoTokenizer
train_on_inputs: false
trust_remote_code: true
val_set_size: 0.005
wandb_entity: null
wandb_mode: online
wandb_name: d8eda0ee-8aeb-47d4-bebc-dd3dba382021
wandb_project: Gradients-On-Demand
wandb_run: your_name
wandb_runid: d8eda0ee-8aeb-47d4-bebc-dd3dba382021
warmup_steps: 30
weight_decay: 0.0
xformers_attention: null

d1526c12-8a5c-45c2-8a7c-e005e4428b34

This model is a fine-tuned version of heegyu/WizardVicuna2-13b-hf on the None dataset. It achieves the following results on the evaluation set:

• Loss: 2.5516

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: 4
• total_train_batch_size: 4
• optimizer: Use OptimizerNames.ADAMW_TORCH_4BIT 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: 30
• num_epochs: 3

Training results

Training Loss	Epoch	Step	Validation Loss
No log	0.0003	1	4.5872
4.0918	0.0311	100	3.9953
3.8106	0.0621	200	3.5150
3.6256	0.0932	300	3.4052
3.2791	0.1242	400	3.2580
3.2456	0.1553	500	3.0634
2.9189	0.1864	600	3.0874
3.0159	0.2174	700	2.9400
3.0506	0.2485	800	2.9574
3.1942	0.2795	900	2.8810
2.7516	0.3106	1000	2.8531
2.9822	0.3417	1100	2.8689
2.7943	0.3727	1200	2.8961
2.7773	0.4038	1300	2.7702
3.0787	0.4349	1400	2.7362
2.6754	0.4659	1500	2.7145
2.882	0.4970	1600	2.6246
2.8287	0.5280	1700	2.6403
2.8178	0.5591	1800	2.5918
2.8114	0.5902	1900	2.6481
3.0178	0.6212	2000	2.5809
2.7718	0.6523	2100	2.5701
2.785	0.6833	2200	2.5290
2.8581	0.7144	2300	2.5949
2.8815	0.7455	2400	2.6250
2.9384	0.7765	2500	2.5516

Framework versions

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