See axolotl config

axolotl version: 0.4.1

adapter: lora
base_model: zake7749/gemma-2-2b-it-chinese-kyara-dpo
bf16: auto
chat_template: llama3
dataloader_num_workers: 12
dataset_prepared_path: null
datasets:
- data_files:
  - 785ff67fe820a49a_train_data.json
  ds_type: json
  format: custom
  path: /workspace/input_data/785ff67fe820a49a_train_data.json
  type:
    field_instruction: init_prompt
    field_output: init_response
    format: '{instruction}'
    no_input_format: '{instruction}'
    system_format: '{system}'
    system_prompt: ''
debug: null
deepspeed: null
device_map: auto
do_eval: true
early_stopping_patience: 3
eval_batch_size: 8
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: 2
gradient_checkpointing: true
group_by_length: true
hub_model_id: cimol/f7f537b2-9316-4f38-a52b-cadb60abb76c
hub_repo: null
hub_strategy: end
hub_token: null
learning_rate: 3e-5
load_in_4bit: false
load_in_8bit: false
local_rank: null
logging_steps: 15
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: cosine
lr_scheduler_warmup_steps: 50
max_grad_norm: 1.0
max_memory:
  0: 75GB
max_steps: 1650
micro_batch_size: 8
mlflow_experiment_name: /tmp/785ff67fe820a49a_train_data.json
model_type: AutoModelForCausalLM
num_epochs: 15
optim_args:
  adam_beta1: 0.9
  adam_beta2: 0.999
  adam_epsilon: 1e-8
optimizer: adamw_torch
output_dir: miner_id_24
pad_to_sequence_len: false
resume_from_checkpoint: null
s2_attention: null
sample_packing: false
save_steps: 150
saves_per_epoch: null
seed: 17333
sequence_len: 1024
strict: false
tf32: true
tokenizer_type: AutoTokenizer
total_train_batch_size: 16
train_batch_size: 8
train_on_inputs: false
trust_remote_code: true
val_set_size: 0.05
wandb_entity: null
wandb_mode: online
wandb_name: 56600945-b760-44be-9fdc-7a38b0eec0cc
wandb_project: Gradients-On-Demand
wandb_run: your_name
wandb_runid: 56600945-b760-44be-9fdc-7a38b0eec0cc
warmup_steps: 50
weight_decay: 0.1
xformers_attention: null

f7f537b2-9316-4f38-a52b-cadb60abb76c

This model is a fine-tuned version of zake7749/gemma-2-2b-it-chinese-kyara-dpo on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.3602

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: 3e-05
• train_batch_size: 8
• eval_batch_size: 8
• seed: 17333
• gradient_accumulation_steps: 2
• total_train_batch_size: 16
• optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=adam_beta1=0.9,adam_beta2=0.999,adam_epsilon=1e-8
• lr_scheduler_type: cosine
• lr_scheduler_warmup_steps: 50
• training_steps: 1650

Training results

Training Loss	Epoch	Step	Validation Loss
No log	0.0004	1	1.8429
0.499	0.0570	150	0.5961
0.4005	0.1139	300	0.5473
0.3805	0.1709	450	0.4914
0.3526	0.2278	600	0.4572
0.3628	0.2848	750	0.4255
0.3156	0.3418	900	0.4032
0.2911	0.3987	1050	0.3818
0.2412	0.4557	1200	0.3700
0.2719	0.5126	1350	0.3641
0.2914	0.5696	1500	0.3605
0.2707	0.6265	1650	0.3602

Framework versions

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