See axolotl config

axolotl version: 0.5.2

adapter: lora
auto_find_batch_size: true
base_model: MNC-Jihun/Mistral-7B-AO-u0.5-b2-ver0.4
bf16: auto
chat_template: llama3
dataset_prepared_path: null
datasets:
- data_files:
  - de68f8646b22d638_train_data.json
  ds_type: json
  format: custom
  path: /workspace/input_data/de68f8646b22d638_train_data.json
  type:
    field_input: level
    field_instruction: x
    field_output: yl
    format: '{instruction} {input}'
    no_input_format: '{instruction}'
    system_format: '{system}'
    system_prompt: ''
debug: null
deepspeed: /workspace/axolotl/configs/deepspeed_stage2.json
eval_max_new_tokens: 128
eval_sample_packing: false
eval_steps: 10
eval_table_size: null
flash_attention: true
fp16: false
gpu_memory_limit: 80GiB
gradient_accumulation_steps: 4
gradient_checkpointing: false
group_by_length: true
hub_model_id: PhoenixB/d3e2643f-e7ab-494b-a191-7ec0f0d6dc4f
hub_repo: null
hub_strategy: checkpoint
hub_token: null
learning_rate: 2e-4
liger_fused_linear_cross_entropy: true
liger_glu_activation: true
liger_layer_norm: true
liger_rms_norm: true
liger_rope: 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: 30
micro_batch_size: 2
mlflow_experiment_name: /tmp/de68f8646b22d638_train_data.json
model_type: AutoModelForCausalLM
num_epochs: 3
optimizer: adamw_torch_fused
output_dir: miner_id_24
pad_to_sequence_len: false
plugins:
- axolotl.integrations.liger.LigerPlugin
resume_from_checkpoint: null
s2_attention: null
sample_packing: false
save_steps: 10
sequence_len: 32768
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: f71059d4-9bec-4672-a1c7-eac9df7bc5a3
wandb_project: Gradients-On-Demand
wandb_run: your_name
wandb_runid: f71059d4-9bec-4672-a1c7-eac9df7bc5a3
warmup_steps: 5
weight_decay: 0.0

d3e2643f-e7ab-494b-a191-7ec0f0d6dc4f

This model is a fine-tuned version of MNC-Jihun/Mistral-7B-AO-u0.5-b2-ver0.4 on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.8012

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: 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: 5
• training_steps: 30

Training results

Training Loss	Epoch	Step	Validation Loss
No log	0.0005	1	1.1129
0.8935	0.0052	10	0.8444
0.7715	0.0103	20	0.7835
0.7837	0.0155	30	0.8012

Framework versions

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