See axolotl config

axolotl version: 0.5.2

adapter: lora
auto_find_batch_size: true
base_model: NousResearch/Hermes-3-Llama-3.1-8B
bf16: auto
chat_template: llama3
dataset_prepared_path: null
datasets:
- data_files:
  - a8730480951cb332_train_data.json
  ds_type: json
  format: custom
  path: /workspace/input_data/a8730480951cb332_train_data.json
  type:
    field_instruction: prompt
    field_output: question
    format: '{instruction}'
    no_input_format: '{instruction}'
    system_format: '{system}'
    system_prompt: ''
debug: null
deepspeed: /workspace/axolotl/configs/deepspeed_stage2.json
early_stopping: true
early_stopping_patience: 3
eval_max_new_tokens: 128
eval_sample_packing: false
eval_steps: 10
eval_strategy: steps
eval_table_size: null
flash_attention: true
fp16: false
gpu_memory_limit: 80GiB
gradient_accumulation_steps: 4
gradient_checkpointing: true
greater_is_better: false
group_by_length: true
hub_model_id: PhoenixB/b848ace4-2ded-41ec-8cea-eb1c79866ee2
hub_repo: null
hub_strategy: checkpoint
hub_token: null
learning_rate: 1e-4
liger_fused_linear_cross_entropy: true
liger_glu_activation: true
liger_layer_norm: true
liger_rms_norm: true
liger_rope: true
load_best_model_at_end: true
load_in_4bit: false
load_in_8bit: false
local_rank: null
logging_steps: 5
lora_alpha: 16
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_steps: 150
metric_for_best_model: loss
micro_batch_size: 2
mlflow_experiment_name: /tmp/a8730480951cb332_train_data.json
model_type: AutoModelForCausalLM
num_epochs: 3
optimizer: adamw_torch_fused
output_dir: miner_id_24
pad_to_sequence_len: true
plugins:
- axolotl.integrations.liger.LigerPlugin
resume_from_checkpoint: null
s2_attention: null
sample_packing: false
save_steps: 20
sequence_len: 8196
strict: false
tf32: true
tokenizer_type: AutoTokenizer
train_on_inputs: false
trust_remote_code: true
val_set_size: 0.05
wandb_entity: null
wandb_mode: online
wandb_name: 6ec2fd27-d2b8-427a-b985-e17ebac9da00
wandb_project: Gradients-On-Demand
wandb_run: your_name
wandb_runid: 6ec2fd27-d2b8-427a-b985-e17ebac9da00
warmup_steps: 20
weight_decay: 0.0

b848ace4-2ded-41ec-8cea-eb1c79866ee2

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: 0.0045

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: 2
• eval_batch_size: 2
• seed: 42
• distributed_type: multi-GPU
• num_devices: 2
• gradient_accumulation_steps: 4
• total_train_batch_size: 16
• total_eval_batch_size: 4
• optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED 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: 20
• training_steps: 150

Training results

Training Loss	Epoch	Step	Validation Loss
No log	0.0009	1	0.7479
0.2569	0.0089	10	0.7231
0.4707	0.0178	20	0.1784
0.0377	0.0267	30	0.0406
0.0149	0.0357	40	0.0138
0.0132	0.0446	50	0.0081
0.0039	0.0535	60	0.0070
0.0046	0.0624	70	0.0063
0.0033	0.0713	80	0.0052
0.0021	0.0802	90	0.0050
0.0027	0.0892	100	0.0046
0.0034	0.0981	110	0.0045
0.0022	0.1070	120	0.0047
0.0037	0.1159	130	0.0043
0.0035	0.1248	140	0.0044
0.0181	0.1337	150	0.0045

Framework versions

• PEFT 0.13.2
• Transformers 4.46.3
• Pytorch 2.5.1+cu124
• Datasets 3.1.0
• Tokenizers 0.20.3