da5c2431-03f2-4e43-81c9-00474ba27c3e

<!-- This model card has been generated automatically according to the information the Trainer had access to. You should probably proofread and complete it, then remove this comment. -->

<img src="https://raw.githubusercontent.com/axolotl-ai-cloud/axolotl/main/image/axolotl-badge-web.png" alt="Built with Axolotl" width="200" height="32"/>

<details><summary>See axolotl config</summary>

axolotl version: 0.4.1

adam_epsilon: 1e-6
adapter: lora
base_model: zake7749/gemma-2-2b-it-chinese-kyara-dpo
bf16: true
chat_template: llama3
dataloader_num_workers: 0
dataloader_pin_memory: false
dataset_prepared_path: null
datasets:
- data_files:
  - 8f8574b1c18c056f_train_data.json
  ds_type: json
  field: prompt
  path: /workspace/input_data/
  split: train
  type: completion
debug: null
eval_max_new_tokens: 128
eval_table_size: null
evals_per_epoch: 2
flash_attention: false
fp16: false
gradient_accumulation_steps: 4
gradient_checkpointing: true
gradient_clipping: 1.0
group_by_length: false
hub_model_id: CheapsetZero/da5c2431-03f2-4e43-81c9-00474ba27c3e
hub_strategy: checkpoint
learning_rate: 0.0002
load_in_4bit: false
load_in_8bit: false
local_rank: null
logging_steps: 10
lora_alpha: 64
lora_dropout: 0.05
lora_r: 32
lora_target_linear: true
lr_scheduler: constant_with_warmup
max_steps: 4000
micro_batch_size: 16
mlflow_experiment_name: /tmp/8f8574b1c18c056f_train_data.json
model_type: AutoModelForCausalLM
optimizer: adamw_torch
output_dir: miner_id_24
pad_to_sequence_len: true
sample_packing: false
save_safetensors: true
save_total_limit: 3
saves_per_epoch: 2
sequence_len: 512
strict: false
tf32: true
tokenizer_type: AutoTokenizer
torch_compile: false
train_on_inputs: false
trl:
  beta: 0.1
  clip_rewards: true
  max_completion_length: 256
  normalize_rewards: true
  num_generations: 8
  reward_clip_value: 10.0
  reward_funcs:
  - rewards_31adf3bc-017a-4973-a03b-37f27f9f783c.reward_low_unique_words_percentage
  - rewards_31adf3bc-017a-4973-a03b-37f27f9f783c.reward_high_unique_words_percentage
  - rewards_31adf3bc-017a-4973-a03b-37f27f9f783c.reward_low_readability
  - rewards_31adf3bc-017a-4973-a03b-37f27f9f783c.reward_reasoning_keywords
  reward_weights:
  - 9.626289578183636
  - 0.947766253701412
  - 0.5053492281078087
  - 1.796308387812069
  temperature: 1.0
  use_vllm: false
trust_remote_code: true
val_set_size: 0.05
wandb_mode: online
wandb_name: 31adf3bc-017a-4973-a03b-37f27f9f783c
wandb_project: Gradients-On-Demand
wandb_run: your_name
wandb_runid: 31adf3bc-017a-4973-a03b-37f27f9f783c
warmup_steps: 100
weight_decay: 0.01
xformers_attention: false

</details><br>

da5c2431-03f2-4e43-81c9-00474ba27c3e

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: 1.6745

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: 16
• eval_batch_size: 16
• seed: 42
• gradient_accumulation_steps: 4
• total_train_batch_size: 64
• optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-06 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: constant_with_warmup
• lr_scheduler_warmup_steps: 100
• training_steps: 871

Training results

Framework versions

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