See axolotl config

axolotl version: 0.4.1

adapter: lora
base_model: NousResearch/Yarn-Llama-2-13b-64k
bf16: auto
chat_template: llama3
dataset_prepared_path: null
datasets:
- data_files:
  - 08a60e51b6dc37ce_train_data.json
  ds_type: json
  format: custom
  path: /workspace/input_data/08a60e51b6dc37ce_train_data.json
  type:
    field_input: input
    field_instruction: instruction
    field_output: output
    format: '{instruction} {input}'
    no_input_format: '{instruction}'
    system_format: '{system}'
    system_prompt: ''
debug: null
deepspeed: null
early_stopping_patience: null
eval_max_new_tokens: 128
eval_table_size: null
evals_per_epoch: 4
flash_attention: false
fp16: true
fsdp: null
fsdp_config: null
gradient_accumulation_steps: 4
gradient_checkpointing: true
group_by_length: false
hub_model_id: leixa/0a90eb4f-372f-4306-9cb4-322b25270120
hub_repo: null
hub_strategy: checkpoint
hub_token: null
learning_rate: 0.0001
load_in_4bit: false
load_in_8bit: false
local_rank: 0
logging_steps: 3
lora_alpha: 128
lora_dropout: 0.1
lora_fan_in_fan_out: true
lora_model_dir: null
lora_r: 64
lora_target_linear: true
lr_scheduler: cosine
max_memory:
  0: 72GB
max_steps: 50
micro_batch_size: 8
mlflow_experiment_name: /tmp/08a60e51b6dc37ce_train_data.json
model_type: AutoModelForCausalLM
num_epochs: 3
optimizer: adamw_bnb_8bit
output_dir: miner_id_24
pad_to_sequence_len: true
resume_from_checkpoint: null
s2_attention: false
sample_packing: false
saves_per_epoch: 4
sequence_len: 1024
strict: false
tf32: false
tokenizer_type: AutoTokenizer
train_on_inputs: false
trust_remote_code: true
val_set_size: 0.05
wandb_entity: leixa-personal
wandb_mode: online
wandb_name: 0a90eb4f-372f-4306-9cb4-322b25270120
wandb_project: Gradients-On-Demand
wandb_run: your_name
wandb_runid: 0a90eb4f-372f-4306-9cb4-322b25270120
warmup_steps: 10
weight_decay: 0.01
xformers_attention: null

0a90eb4f-372f-4306-9cb4-322b25270120

This model is a fine-tuned version of NousResearch/Yarn-Llama-2-13b-64k on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.3398

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: 0.0001
• train_batch_size: 8
• eval_batch_size: 8
• seed: 42
• gradient_accumulation_steps: 4
• 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	0.0003	1	0.7600
3.2519	0.0013	5	0.6099
1.8811	0.0026	10	0.4515
1.7995	0.0039	15	0.4112
1.5691	0.0052	20	0.3853
1.3646	0.0065	25	0.3685
1.3538	0.0078	30	0.3528
1.2983	0.0091	35	0.3471
1.3439	0.0104	40	0.3420
1.0076	0.0117	45	0.3402
1.5315	0.0130	50	0.3398

Framework versions

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