adapter: lora
base_model: microsoft/Phi-3-mini-4k-instruct
bf16: true
chat_template: llama3
dataset_prepared_path: null
datasets:
- data_files:
  - 11bb3328a39885eb_train_data.json
  ds_type: json
  format: custom
  path: /workspace/input_data/11bb3328a39885eb_train_data.json
  type:
    field_instruction: query
    field_output: atom
    format: '{instruction}'
    no_input_format: '{instruction}'
    system_format: '{system}'
    system_prompt: ''
debug: null
deepspeed: null
device_map: auto
do_eval: true
early_stopping_patience: 2
eval_batch_size: 4
eval_max_new_tokens: 128
eval_steps: 150
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: true
hub_model_id: nttx/3005289b-9876-4d69-8e0d-2370a534dbb2
hub_repo: null
hub_strategy: checkpoint
hub_token: null
learning_rate: 0.0002
load_in_4bit: false
load_in_8bit: false
local_rank: null
logging_steps: 50
lora_alpha: 128
lora_dropout: 0.3
lora_fan_in_fan_out: null
lora_model_dir: null
lora_r: 64
lora_target_linear: true
lr_scheduler: cosine
max_grad_norm: 1.0
max_memory:
  0: 75GB
max_steps: 3000
micro_batch_size: 4
mlflow_experiment_name: /tmp/11bb3328a39885eb_train_data.json
model_type: AutoModelForCausalLM
num_epochs: 10
optim_args:
  adam_beta1: 0.9
  adam_beta2: 0.95
  adam_epsilon: 1e-5
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: 150
saves_per_epoch: null
sequence_len: 512
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: 7ce72720-2645-47e1-9d12-b5a347cab783
wandb_project: Gradients-On-Demand
wandb_run: your_name
wandb_runid: 7ce72720-2645-47e1-9d12-b5a347cab783
warmup_steps: 50
weight_decay: 0.0
xformers_attention: null

3005289b-9876-4d69-8e0d-2370a534dbb2

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: 0.4909

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: 4
total_train_batch_size: 16
optimizer: Use OptimizerNames.ADAMW_BNB with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=adam_beta1=0.9,adam_beta2=0.95,adam_epsilon=1e-5
lr_scheduler_type: cosine
lr_scheduler_warmup_steps: 50
training_steps: 3000

Training results

Training Loss	Epoch	Step	Validation Loss
No log	0.0009	1	4.4761
4.5239	0.1415	150	1.8525
2.9908	0.2830	300	0.6934
2.928	0.4245	450	0.6097
2.4512	0.5660	600	0.6266
2.2832	0.7075	750	0.5850
2.5273	0.8491	900	0.6573
2.2433	0.9906	1050	0.5254
1.773	1.1321	1200	0.4980
1.6777	1.2736	1350	0.4891
1.6254	1.4151	1500	0.4829
1.4815	1.5566	1650	0.4727
1.5053	1.6981	1800	0.4661
1.5059	1.8396	1950	0.4602
1.489	1.9811	2100	0.4607
1.0975	2.1226	2250	0.4909

Framework versions

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