See axolotl config

axolotl version: 0.4.1

adapter: qlora
auto_resume_from_checkpoints: true
base_model: microsoft/Phi-3-mini-4k-instruct
bf16: auto
chat_template: llama3
dataset_prepared_path: null
dataset_processes: 6
datasets:
- data_files:
  - e2ea22acdb815831_train_data.json
  ds_type: json
  format: custom
  path: /workspace/input_data/e2ea22acdb815831_train_data.json
  type:
    field_instruction: premise
    field_output: hypothesis
    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: 200
eval_table_size: null
evals_per_epoch: null
flash_attention: true
fp16: false
fsdp: null
fsdp_config: null
gradient_accumulation_steps: 4
gradient_checkpointing: true
group_by_length: false
hub_model_id: error577/e99b6a20-be94-4148-af75-f696f7fce256
hub_repo: null
hub_strategy: checkpoint
hub_token: null
learning_rate: 0.0002
load_in_4bit: true
load_in_8bit: false
local_rank: null
logging_steps: 1
lora_alpha: 64
lora_dropout: 0.1
lora_fan_in_fan_out: null
lora_model_dir: null
lora_r: 32
lora_target_linear: true
lr_scheduler: cosine
max_grad_norm: 1.0
max_steps: null
micro_batch_size: 2
mlflow_experiment_name: /tmp/e2ea22acdb815831_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
save_steps: 200
sequence_len: 512
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: c04ebf44-479a-4275-a61c-afd0b7f9fa9a
wandb_project: Gradients-On-Demand
wandb_run: your_name
wandb_runid: c04ebf44-479a-4275-a61c-afd0b7f9fa9a
warmup_steps: 30
weight_decay: 0.0
xformers_attention: null

e99b6a20-be94-4148-af75-f696f7fce256

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: 1.6192

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
• gradient_accumulation_steps: 4
• 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
12.5501	0.0000	1	3.1170
8.1989	0.0095	200	1.6985
6.3793	0.0191	400	1.6660
8.2906	0.0286	600	1.6446
5.6457	0.0382	800	1.6340
7.2242	0.0477	1000	1.6225
6.4631	0.0573	1200	1.6280
5.7667	0.0668	1400	1.6251
8.727	0.0764	1600	1.6216
5.2037	0.0859	1800	1.6195
3.9859	0.0955	2000	1.6056
5.4473	0.1050	2200	1.6023
6.4552	0.1145	2400	1.6121
5.2552	0.1241	2600	1.6165
6.5086	0.1336	2800	1.6192

Framework versions

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