See axolotl config

axolotl version: 0.5.2

adapter: lora
auto_find_batch_size: true
base_model: unsloth/llama-3-8b-Instruct
bf16: auto
chat_template: llama3
dataset_prepared_path: null
datasets:
- data_files:
  - 38013dab76daf968_train_data.json
  ds_type: json
  format: custom
  path: /workspace/input_data/38013dab76daf968_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: /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/ec4d4e11-eb4e-4171-88a1-d8791ef5ccf3
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/38013dab76daf968_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: 8192
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: 77b319c4-5ecf-47e3-a09b-f0225e777067
wandb_project: Gradients-On-Demand
wandb_run: your_name
wandb_runid: 77b319c4-5ecf-47e3-a09b-f0225e777067
warmup_steps: 20
weight_decay: 0.0

ec4d4e11-eb4e-4171-88a1-d8791ef5ccf3

This model is a fine-tuned version of unsloth/llama-3-8b-Instruct on the None dataset. It achieves the following results on the evaluation set:

• Loss: 1.2136

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	0.0002	1	2.0674
1.4157	0.0015	10	2.0662
1.9105	0.0031	20	1.6511
1.3167	0.0046	30	1.4062
1.38	0.0061	40	1.3150
1.1283	0.0076	50	1.2631
1.3578	0.0092	60	1.2536
1.3771	0.0107	70	1.2546
1.2291	0.0122	80	1.2387
1.416	0.0137	90	1.2346
1.2994	0.0153	100	1.2234
1.2366	0.0168	110	1.2217
1.3202	0.0183	120	1.2177
1.1669	0.0199	130	1.2182
1.2716	0.0214	140	1.2121
1.0569	0.0229	150	1.2136

Framework versions

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