See axolotl config

axolotl version: 0.4.1

base_model: katuni4ka/tiny-random-qwen1.5-moe
batch_size: 32
bf16: true
chat_template: tokenizer_default_fallback_alpaca
datasets:
- data_files:
  - 95544452e61c7393_train_data.json
  ds_type: json
  format: custom
  path: /workspace/input_data/95544452e61c7393_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: ''
eval_steps: 20
flash_attention: true
gpu_memory_limit: 80GiB
gradient_checkpointing: true
group_by_length: true
hub_model_id: willtensora/e61e89f0-854a-4922-8d25-dae435e91af0
hub_strategy: checkpoint
learning_rate: 0.0002
logging_steps: 10
lr_scheduler: cosine
max_steps: 2500
micro_batch_size: 4
model_type: AutoModelForCausalLM
optimizer: adamw_bnb_8bit
output_dir: /workspace/axolotl/configs
pad_to_sequence_len: true
resize_token_embeddings_to_32x: false
sample_packing: false
save_steps: 40
save_total_limit: 1
sequence_len: 2048
tokenizer_type: Qwen2TokenizerFast
train_on_inputs: false
trust_remote_code: true
val_set_size: 0.1
wandb_entity: ''
wandb_mode: online
wandb_name: katuni4ka/tiny-random-qwen1.5-moe-/workspace/input_data/95544452e61c7393_train_data.json
wandb_project: Gradients-On-Demand
wandb_run: your_name
wandb_runid: default
warmup_ratio: 0.05
xformers_attention: true

e61e89f0-854a-4922-8d25-dae435e91af0

This model is a fine-tuned version of katuni4ka/tiny-random-qwen1.5-moe on the None dataset. It achieves the following results on the evaluation set:

• Loss: 11.6281

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: 4
• eval_batch_size: 4
• seed: 42
• distributed_type: multi-GPU
• num_devices: 8
• total_train_batch_size: 32
• total_eval_batch_size: 32
• 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: 2
• training_steps: 40

Training results

Training Loss	Epoch	Step	Validation Loss
No log	0.0031	1	11.9223
11.7325	0.0629	20	11.6783
11.6304	0.1258	40	11.6281

Framework versions

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