metadata

library_name: peft
license: gemma
base_model: google/gemma-2-2b-it
tags:
  - axolotl
  - generated_from_trainer
model-index:
  - name: gemma-2-2b-it-dolly-15k
    results: []
datasets:
  - databricks/databricks-dolly-15k
pipeline_tag: text-generation

See axolotl config

axolotl version: 0.5.2

base_model: google/gemma-2-2b-it
hub_model_id: kweinmeister/gemma-2-2b-it-dolly-15k

load_in_8bit: false
load_in_4bit: true
strict: false

datasets:
  - path: databricks/databricks-dolly-15k
    type:
      field_instruction: instruction       
      field_input: context
      field_output: response
val_set_size: 0.05

sequence_len: 2048
sample_packing: true
eval_sample_packing: true
pad_to_sequence_len: true

adapter: qlora
lora_model_dir:
lora_r: 32
lora_alpha: 64
lora_dropout: 0.05
lora_target_linear: true
lora_fan_in_fan_out:

wandb_project: gemma-2-2b-it-dolly-15k
wandb_entity:
wandb_watch:
wandb_name:
wandb_log_model:

gradient_accumulation_steps: 4
micro_batch_size: 1
num_epochs: 1
optimizer: adamw_torch
lr_scheduler: cosine
learning_rate: 0.0002

train_on_inputs: false
group_by_length: false
bf16: auto
fp16:
tf32: true

gradient_checkpointing: true
gradient_checkpointing_kwargs:
  use_reentrant: true
early_stopping_patience:
resume_from_checkpoint:
local_rank:
logging_steps: 1
xformers_attention:
flash_attention: false

warmup_steps: 10
evals_per_epoch: 4
saves_per_epoch: 1
debug:
deepspeed: deepspeed_configs/zero1.json
weight_decay: 0.0
fsdp:
fsdp_config:
special_tokens:
output_dir: "/mnt/disks/gcs/axolotl/runs/google--gemma-2-2b-it-20250101-144050/out/"
dataset_prepared_path: "/mnt/disks/gcs/axolotl/last_run_prepared"

gemma-2-2b-it-dolly-15k

This model is a fine-tuned version of google/gemma-2-2b-it on the None dataset. It achieves the following results on the evaluation set:

Loss: 1.7389

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: 1
eval_batch_size: 1
seed: 42
distributed_type: multi-GPU
num_devices: 2
gradient_accumulation_steps: 4
total_train_batch_size: 8
total_eval_batch_size: 2
optimizer: Use OptimizerNames.ADAMW_TORCH 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: 10
num_epochs: 1

Training results

Training Loss	Epoch	Step	Validation Loss
5.7033	0.0061	1	5.5100
1.8197	0.2492	41	1.8752
1.6386	0.4985	82	1.7666
1.7346	0.7477	123	1.7436
1.7742	0.9970	164	1.7389

Framework versions

PEFT 0.13.2
Transformers 4.46.3
Pytorch 2.4.1+cu124
Datasets 3.1.0
Tokenizers 0.20.3