adapter: lora
base_model: katuni4ka/tiny-random-qwen1.5-moe
bf16: true
chat_template: llama3
dataset_prepared_path: null
datasets:
- data_files:
  - 16b5810173360f44_train_data.json
  ds_type: json
  format: custom
  path: /workspace/input_data/16b5810173360f44_train_data.json
  type:
    field_input: difficulty
    field_instruction: content
    field_output: java
    format: '{instruction} {input}'
    no_input_format: '{instruction}'
    system_format: '{system}'
    system_prompt: ''
debug: null
deepspeed: null
early_stopping_patience: null
eval_max_new_tokens: 256
eval_table_size: null
evals_per_epoch: 4
flash_attention: false
fp16: false
fsdp: null
fsdp_config: null
gradient_accumulation_steps: 2
gradient_checkpointing: true
group_by_length: true
hub_model_id: pedrosperber760/962accf0-33b8-4f9e-9141-56b586b99e32
hub_repo: null
hub_strategy: checkpoint
hub_token: null
learning_rate: 0.0002
load_in_4bit: true
load_in_8bit: false
local_rank: null
logging_steps: 1
lora_alpha: 64
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_steps: 10
micro_batch_size: 3
mlflow_experiment_name: /tmp/16b5810173360f44_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
sample_packing: true
saves_per_epoch: 4
sequence_len: 4096
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: 962accf0-33b8-4f9e-9141-56b586b99e32
wandb_project: Gradients-On-Demand
wandb_run: your_name
wandb_runid: 962accf0-33b8-4f9e-9141-56b586b99e32
warmup_steps: 10
weight_decay: 0.01
xformers_attention: false

962accf0-33b8-4f9e-9141-56b586b99e32

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: 11.9225

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: 3
eval_batch_size: 3
seed: 42
gradient_accumulation_steps: 2
total_train_batch_size: 6
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: 10
training_steps: 10

Training results

Training Loss	Epoch	Step	Validation Loss
11.9313	0.0142	1	11.9303
11.9315	0.0426	3	11.9297
11.9302	0.0851	6	11.9272
11.9249	0.1277	9	11.9225

Framework versions

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