See axolotl config

axolotl version: 0.4.1

adapter: lora
auto_resume_from_checkpoints: false
base_model: unsloth/Phi-3.5-mini-instruct
bf16: auto
chat_template: llama3
dataset_prepared_path: null
datasets:
- data_files:
  - b4ae15917673f3a5_train_data.json
  ds_type: json
  format: custom
  path: /workspace/input_data/b4ae15917673f3a5_train_data.json
  type:
    field_instruction: instruction
    field_output: output
    format: '{instruction}'
    no_input_format: '{instruction}'
    system_format: '{system}'
    system_prompt: ''
debug: null
deepspeed: null
early_stopping_patience: 3
eval_max_new_tokens: 128
eval_steps: 100
eval_table_size: null
flash_attention: true
fp16: false
fsdp: null
fsdp_config: null
gradient_accumulation_steps: 2
gradient_checkpointing: true
group_by_length: false
hub_model_id: error577/cf1ebf8e-867f-4a25-9a84-696a66d4629e
hub_repo: null
hub_strategy: checkpoint
hub_token: null
learning_rate: 0.0002
load_in_4bit: false
load_in_8bit: true
local_rank: null
logging_steps: 1
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_grad_norm: 5.0
max_steps: null
micro_batch_size: 4
mlflow_experiment_name: /tmp/b4ae15917673f3a5_train_data.json
model_type: AutoModelForCausalLM
num_epochs: 3
optimizer: adamw_bnb_8bit
output_dir: miner_id_24
pad_to_sequence_len: false
resume_from_checkpoint: null
s2_attention: null
sample_packing: false
save_steps: 100
sequence_len: 8
strict: false
tf32: false
tokenizer_type: AutoTokenizer
train_on_inputs: false
trust_remote_code: true
val_set_size: 0.005
wandb_entity: null
wandb_mode: online
wandb_name: 61595bbc-d641-4708-aec4-c2ea3a8d9a85
wandb_project: Gradients-On-Demand
wandb_run: your_name
wandb_runid: 61595bbc-d641-4708-aec4-c2ea3a8d9a85
warmup_steps: 30
weight_decay: 0.0
xformers_attention: null

cf1ebf8e-867f-4a25-9a84-696a66d4629e

This model is a fine-tuned version of unsloth/Phi-3.5-mini-instruct on the None dataset. It achieves the following results on the evaluation set:

• Loss: 7.0293

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: 2
• total_train_batch_size: 8
• 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: 30
• num_epochs: 3

Training results

Training Loss	Epoch	Step	Validation Loss
5.9611	0.0002	1	6.0076
4.8196	0.0157	100	5.4899
6.2363	0.0314	200	5.3367
5.096	0.0471	300	5.1652
6.5592	0.0627	400	6.2290
7.2188	0.0784	500	7.9515
7.0274	0.0941	600	7.0293

Framework versions

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