adapter: qlora
base_model: NousResearch/Yarn-Llama-2-7b-64k
bf16: auto
chat_template: llama3
dataloader_num_workers: 6
dataset_prepared_path: null
datasets:
- data_files:
  - 1e66950fe06b11fd_train_data.json
  ds_type: json
  format: custom
  path: /workspace/input_data/1e66950fe06b11fd_train_data.json
  type:
    field_input: rendered_input
    field_instruction: commands
    field_output: rendered_output
    format: '{instruction} {input}'
    no_input_format: '{instruction}'
    system_format: '{system}'
    system_prompt: ''
debug: null
deepspeed: null
early_stopping:
  patience: 3
  metric: eval_loss
  mode: min
eval_max_new_tokens: 128
eval_steps: 20
eval_table_size: null
evals_per_epoch: null
flash_attention: false
fp16: null
fsdp: null
fsdp_config: null
gradient_accumulation_steps: 16
gradient_checkpointing: true
group_by_length: false
hub_model_id: error577/e410389c-a54b-4638-9c78-f90b4d4f8539
hub_repo: null
hub_strategy: end
hub_token: null
learning_rate: 0.0003
load_in_4bit: true
load_in_8bit: false
local_rank: null
logging_steps: 1
lora_alpha: 64
lora_dropout: 0.3
lora_fan_in_fan_out: null
lora_model_dir: null
lora_r: 32
lora_target_linear: true
lr_scheduler: cosine
max_steps: 100
micro_batch_size: 2
mlflow_experiment_name: /tmp/1e66950fe06b11fd_train_data.json
model_type: AutoModelForCausalLM
num_epochs: 50
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_steps: 20
sequence_len: 512
strict: false
tf32: false
tokenizer_type: AutoTokenizer
train_on_inputs: false
trust_remote_code: true
val_set_size: 0.002
wandb_entity: null
wandb_mode: online
wandb_name: afc9a4fd-b1a5-4fca-b90e-f509c4d93ae3
wandb_project: Gradients-On-Demand
wandb_run: your_name
wandb_runid: afc9a4fd-b1a5-4fca-b90e-f509c4d93ae3
warmup_steps: 10
weight_decay: 0.0
xformers_attention: null

e410389c-a54b-4638-9c78-f90b4d4f8539

This model is a fine-tuned version of NousResearch/Yarn-Llama-2-7b-64k on the None dataset. It achieves the following results on the evaluation set:

Loss: 0.0163

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: 0.0003
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: 100

Training results

Training Loss	Epoch	Step	Validation Loss
22.8862	0.0002	1	1.3632
2.4807	0.0033	20	0.1543
1.3273	0.0067	40	0.0556
0.8466	0.0100	60	0.0328
0.3243	0.0134	80	0.0199
0.2753	0.0167	100	0.0163

Framework versions

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