See axolotl config

axolotl version: 0.4.1

adapter: lora
base_model: peft-internal-testing/tiny-dummy-qwen2
bf16: auto
chat_template: llama3
dataset_prepared_path: null
datasets:
- data_files:
  - e775cbf0abe327ae_train_data.json
  ds_type: json
  format: custom
  path: /workspace/input_data/e775cbf0abe327ae_train_data.json
  type:
    field_instruction: instruction
    field_output: output
    format: '{instruction}'
    no_input_format: '{instruction}'
    system_format: '{system}'
    system_prompt: ''
debug: null
deepspeed: null
device: cuda
early_stopping_patience: null
eval_max_new_tokens: 256
eval_steps: 5
eval_table_size: null
evals_per_epoch: null
flash_attention: false
fp16: null
gradient_accumulation_steps: 4
gradient_checkpointing: true
group_by_length: false
hub_model_id: gavrilstep/71be9bf6-3848-41b5-853c-3b2e0bd5dece
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: 3
lora_alpha: 32
lora_dropout: 0.05
lora_fan_in_fan_out: null
lora_model_dir: null
lora_r: 16
lora_target_linear: true
lr_scheduler: cosine
max_memory:
  0: 75GiB
max_steps: 40
micro_batch_size: 2
mlflow_experiment_name: /tmp/e775cbf0abe327ae_train_data.json
model_type: AutoModelForCausalLM
num_epochs: 1
optimizer: adamw_torch
output_dir: miner_id_24
pad_to_sequence_len: true
resume_from_checkpoint: null
s2_attention: null
sample_packing: false
save_steps: 10
sequence_len: 1024
strict: false
tf32: false
tokenizer_type: AutoTokenizer
train_on_inputs: true
trust_remote_code: true
val_set_size: 0.05
wandb_entity: null
wandb_mode: online
wandb_name: b84adb33-79b3-4680-b053-92bfa0dc3c21
wandb_project: Gradients-On-Demand
wandb_run: your_name
wandb_runid: b84adb33-79b3-4680-b053-92bfa0dc3c21
warmup_steps: 10
weight_decay: 0.01
xformers_attention: true

71be9bf6-3848-41b5-853c-3b2e0bd5dece

This model is a fine-tuned version of peft-internal-testing/tiny-dummy-qwen2 on the None dataset. It achieves the following results on the evaluation set:

• Loss: 11.9220

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: 2
• eval_batch_size: 2
• seed: 42
• gradient_accumulation_steps: 4
• total_train_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: 10
• training_steps: 40

Training results

Training Loss	Epoch	Step	Validation Loss
No log	0.0012	1	11.9290
11.9282	0.0062	5	11.9288
11.9293	0.0124	10	11.9280
11.9267	0.0186	15	11.9266
11.9254	0.0248	20	11.9250
11.9236	0.0310	25	11.9236
11.9232	0.0372	30	11.9226
11.9234	0.0433	35	11.9221
11.9214	0.0495	40	11.9220

Framework versions

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