adapter: lora
base_model: microsoft/Phi-3-mini-128k-instruct
bf16: true
chat_template: llama3
data_processes: 54
dataset_prepared_path: null
datasets:
- data_files:
  - c162e51c5b4c99d0_train_data.json
  ds_type: json
  format: custom
  path: /workspace/input_data/c162e51c5b4c99d0_train_data.json
  type:
    field_input: input
    field_instruction: instruction
    field_output: output
    format: '{instruction} {input}'
    no_input_format: '{instruction}'
    system_format: '{system}'
    system_prompt: ''
debug: null
deepspeed: null
device_map: auto
distributed_training:
  backend: nccl
  multi_gpu: true
  num_gpus: 2
do_eval: true
early_stopping_patience: 4
eval_batch_size: 16
eval_max_new_tokens: 128
eval_steps: 150
eval_table_size: null
evals_per_epoch: null
flash_attention: true
fp16: false
fsdp:
- full_shard
fsdp_config:
  backward_prefetch: BACKWARD_PRE
  cpu_offload: false
  forward_prefetch: false
  mixed_precision: bf16
  sharding_strategy: FULL_SHARD
  use_orig_params: true
gradient_accumulation_steps: 1
gradient_checkpointing: true
group_by_length: true
hub_ignore_patterns:
- README.md
- config.json
hub_model_id: cimol/79faed67-3425-4896-96fc-bd905512fd9c
hub_repo: null
hub_strategy: end
hub_token: null
learning_rate: 0.0001
load_in_4bit: false
load_in_8bit: false
local_rank: null
logging_steps: 10
lora_alpha: 128
lora_dropout: 0.1
lora_fan_in_fan_out: null
lora_model_dir: null
lora_r: 64
lora_target_linear: true
lr_scheduler: polynomial
lr_scheduler_warmup_steps: 75
max_grad_norm: 0.5
max_memory:
  0: 75GB
  1: 75GB
max_steps: 1950
micro_batch_size: 16
mlflow_experiment_name: /tmp/c162e51c5b4c99d0_train_data.json
model_type: AutoModelForCausalLM
num_epochs: 4
optim_args:
  adam_beta1: 0.9
  adam_beta2: 0.95
  adam_epsilon: 1e-8
optimizer: adamw_torch
output_dir: miner_id_24
pad_to_sequence_len: true
resume_from_checkpoint: null
s2_attention: null
sample_packing: false
save_steps: 300
saves_per_epoch: null
seed: 17333
sequence_len: 1024
strict: false
tf32: true
tokenizer_type: AutoTokenizer
total_train_batch_size: 32
train_batch_size: 32
train_on_inputs: false
trust_remote_code: true
val_set_size: 0.05
wandb_entity: null
wandb_mode: online
wandb_name: 3a017f30-2969-4026-ac3a-b3a6f5d66f55
wandb_project: Gradients-On-Demand
wandb_run: your_name
wandb_runid: 3a017f30-2969-4026-ac3a-b3a6f5d66f55
warmup_steps: 150
weight_decay: 0.01
xformers_attention: null

79faed67-3425-4896-96fc-bd905512fd9c

This model is a fine-tuned version of microsoft/Phi-3-mini-128k-instruct on the None dataset. It achieves the following results on the evaluation set:

Loss: 0.3018

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: 0.0001
train_batch_size: 16
eval_batch_size: 16
seed: 17333
distributed_type: multi-GPU
num_devices: 2
total_train_batch_size: 32
total_eval_batch_size: 32
optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=adam_beta1=0.9,adam_beta2=0.95,adam_epsilon=1e-8
lr_scheduler_type: polynomial
lr_scheduler_warmup_steps: 150
training_steps: 1950

Training results

Training Loss	Epoch	Step	Validation Loss
No log	0.0001	1	0.4837
0.4516	0.0218	150	0.3232
0.4309	0.0437	300	0.3157
0.4403	0.0655	450	0.3123
0.4072	0.0873	600	0.3098
0.4547	0.1092	750	0.3088
0.4375	0.1310	900	0.3061
0.4078	0.1528	1050	0.3062
0.4762	0.1747	1200	0.3052
0.4433	0.1965	1350	0.3030
0.4269	0.2183	1500	0.3030
0.4119	0.2402	1650	0.3025
0.4025	0.2620	1800	0.3015
0.455	0.2838	1950	0.3018

Framework versions

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