See axolotl config

axolotl version: 0.4.1

adapter: lora
base_model: microsoft/Phi-3-mini-4k-instruct
bf16: auto
chat_template: llama3
dataset_prepared_path: null
datasets:
- data_files:
  - 6c3ac0ead1090ad2_train_data.json
  ds_type: json
  format: custom
  path: /workspace/input_data/6c3ac0ead1090ad2_train_data.json
  type:
    field_input: prefix
    field_instruction: problem
    field_output: full_solution
    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: true
fp16: null
fsdp: null
fsdp_config: null
gradient_accumulation_steps: 4
gradient_checkpointing: true
gradient_clipping: 1.0
group_by_length: false
hub_model_id: sn56/25cf62f2-d032-4bc3-8cc1-32e591405fce
hub_repo: null
hub_strategy: checkpoint
hub_token: null
learning_rate: 5.0e-05
load_in_4bit: false
load_in_8bit: false
local_rank: 0
logging_steps: 3
lora_alpha: 32
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
micro_batch_size: 8
mlflow_experiment_name: /tmp/6c3ac0ead1090ad2_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: null
sample_packing: false
saves_per_epoch: 4
seed: 3491072646
sequence_len: 1024
shuffle: true
strict: false
tf32: false
tokenizer_type: AutoTokenizer
torch_compile: true
train_on_inputs: false
trust_remote_code: true
val_set_size: 0.05
wandb_entity: sn56-miner
wandb_mode: disabled
wandb_name: null
wandb_project: god
wandb_run: e0ss
wandb_runid: null
warmup_steps: 10
weight_decay: 0.01
xformers_attention: null

25cf62f2-d032-4bc3-8cc1-32e591405fce

This model is a fine-tuned version of microsoft/Phi-3-mini-4k-instruct on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.3801

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: 5e-05
• train_batch_size: 8
• eval_batch_size: 8
• seed: 3491072646
• distributed_type: multi-GPU
• num_devices: 4
• gradient_accumulation_steps: 4
• total_train_batch_size: 128
• total_eval_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: 150

Training results

Training Loss	Epoch	Step	Validation Loss
No log	0.0025	1	0.4909
1.8792	0.0328	13	0.4459
1.6599	0.0656	26	0.4084
1.5772	0.0984	39	0.3947
1.5715	0.1312	52	0.3886
1.5352	0.1640	65	0.3850
1.5513	0.1968	78	0.3829
1.5586	0.2297	91	0.3817
1.5188	0.2625	104	0.3809
1.5139	0.2953	117	0.3804
1.4679	0.3281	130	0.3802
1.4968	0.3609	143	0.3801

Framework versions

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