Built with Axolotl

See axolotl config

axolotl version: 0.4.1

adapter: lora
auto_find_batch_size: true
base_model: microsoft/phi-2
bf16: auto
chat_template: llama3
dataset_prepared_path: null
datasets:
- data_files:
  - 38064736a6270235_train_data.json
  ds_type: json
  format: custom
  path: /workspace/input_data/38064736a6270235_train_data.json
  type:
    field_input: category
    field_instruction: style
    field_output: text
    format: '{instruction} {input}'
    no_input_format: '{instruction}'
    system_format: '{system}'
    system_prompt: ''
debug: null
deepspeed: null
do_eval: true
early_stopping_patience: 3
eval_max_new_tokens: 128
eval_steps: 50
evals_per_epoch: null
flash_attention: true
fp16: false
fsdp: null
fsdp_config: null
gradient_accumulation_steps: 2
gradient_checkpointing: false
group_by_length: true
hub_model_id: lesso08/921e51cb-2cc9-4d06-aad4-ce131a131675
hub_repo: null
hub_strategy: checkpoint
hub_token: null
learning_rate: 0.000208
load_in_4bit: false
load_in_8bit: false
local_rank: null
logging_steps: 10
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: 1.0
max_steps: 500
micro_batch_size: 4
mlflow_experiment_name: /tmp/38064736a6270235_train_data.json
model_type: AutoModelForCausalLM
num_epochs: 1
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: 50
saves_per_epoch: null
seed: 80
sequence_len: 512
special_tokens:
  pad_token: <|endoftext|>
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: 74fabe7c-7e1f-4feb-91db-5cbd9154dee7
wandb_project: 08a
wandb_run: your_name
wandb_runid: 74fabe7c-7e1f-4feb-91db-5cbd9154dee7
warmup_steps: 50
weight_decay: 0.0
xformers_attention: null

921e51cb-2cc9-4d06-aad4-ce131a131675

This model is a fine-tuned version of microsoft/phi-2 on the None dataset. It achieves the following results on the evaluation set:

  • Loss: 3.2962

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: 0.000208
  • train_batch_size: 4
  • eval_batch_size: 4
  • seed: 80
  • 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: 50
  • training_steps: 500

Training results

Training Loss Epoch Step Validation Loss
No log 0.0004 1 3.8350
4.096 0.0186 50 3.5961
3.458 0.0371 100 3.3898
4.1185 0.0557 150 3.4648
3.797 0.0743 200 3.4442
3.4614 0.0929 250 3.3462
3.6412 0.1114 300 3.3367
3.6241 0.1300 350 3.3282
3.6939 0.1486 400 3.3014
3.6158 0.1672 450 3.2983
3.5623 0.1857 500 3.2962

Framework versions

  • PEFT 0.13.2
  • Transformers 4.46.0
  • Pytorch 2.5.0+cu124
  • Datasets 3.0.1
  • Tokenizers 0.20.1
Downloads last month
10
Inference Providers NEW
This model is not currently available via any of the supported Inference Providers.
The model cannot be deployed to the HF Inference API: The model has no pipeline_tag.

Model tree for lesso08/921e51cb-2cc9-4d06-aad4-ce131a131675

Base model

microsoft/phi-2
Adapter
(906)
this model