See axolotl config

axolotl version: 0.4.1

adapter: lora
base_model: NousResearch/Hermes-3-Llama-3.1-8B
bf16: true
chat_template: llama3
dataset_prepared_path: null
datasets:
- data_files:
  - b6b5e24c2d23bde9_train_data.json
  ds_type: json
  format: custom
  path: /workspace/input_data/b6b5e24c2d23bde9_train_data.json
  type:
    field_input: dataset
    field_instruction: task
    field_output: chosen
    format: '{instruction} {input}'
    no_input_format: '{instruction}'
    system_format: '{system}'
    system_prompt: ''
debug: null
device_map:
  ? ''
  : 0,1,2,3,4,5,6,7
early_stopping_patience: 2
eval_max_new_tokens: 128
eval_steps: 100
eval_table_size: null
flash_attention: false
gradient_accumulation_steps: 8
gradient_checkpointing: true
group_by_length: false
hub_model_id: Alphatao/07709b98-745c-4f10-8c03-ff56f7f49ac1
hub_repo: null
hub_strategy: null
hub_token: null
learning_rate: 0.0002
load_best_model_at_end: true
load_in_4bit: false
load_in_8bit: false
local_rank: null
logging_steps: 1
lora_alpha: 128
lora_dropout: 0.1
lora_fan_in_fan_out: null
lora_model_dir: null
lora_r: 64
lora_target_linear: true
lora_target_modules:
- q_proj
- k_proj
- v_proj
- o_proj
lr_scheduler: cosine
max_grad_norm: 1.0
max_steps: 840
micro_batch_size: 4
mlflow_experiment_name: /tmp/b6b5e24c2d23bde9_train_data.json
model_type: AutoModelForCausalLM
num_epochs: 2
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: 100
sequence_len: 2048
strict: false
tf32: true
tokenizer_type: AutoTokenizer
train_on_inputs: false
trust_remote_code: true
val_set_size: 0.02289146697676974
wandb_entity: null
wandb_mode: online
wandb_name: bca728df-548a-4310-bd9e-8c5e5801aad9
wandb_project: Gradients-On-Demand
wandb_run: your_name
wandb_runid: bca728df-548a-4310-bd9e-8c5e5801aad9
warmup_steps: 10
weight_decay: 0.0
xformers_attention: null

07709b98-745c-4f10-8c03-ff56f7f49ac1

This model is a fine-tuned version of NousResearch/Hermes-3-Llama-3.1-8B on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.7081

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: 4
• eval_batch_size: 4
• seed: 42
• gradient_accumulation_steps: 8
• 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: 840

Training results

Training Loss	Epoch	Step	Validation Loss
1.0732	0.0001	1	1.1708
0.7843	0.0150	100	0.7687
0.6963	0.0300	200	0.7537
0.7623	0.0450	300	0.7414
0.7152	0.0600	400	0.7314
0.6951	0.0750	500	0.7217
0.7409	0.0900	600	0.7143
0.6963	0.1050	700	0.7097
0.7063	0.1199	800	0.7081

Framework versions

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