See axolotl config

axolotl version: 0.4.1

adapter: lora
auto_resume_from_checkpoints: false
base_model: katuni4ka/tiny-random-codegen2
bf16: auto
chat_template: llama3
dataset_prepared_path: null
dataset_processes: 6
datasets:
- data_files:
  - 881619c188d8a836_train_data.json
  ds_type: json
  format: custom
  path: /workspace/input_data/881619c188d8a836_train_data.json
  type:
    field_input: prompt
    field_instruction: instruction
    field_output: response
    format: '{instruction} {input}'
    no_input_format: '{instruction}'
    system_format: '{system}'
    system_prompt: ''
debug: null
deepspeed: null
early_stopping_patience:
evals_per_epoch: 1
saves_per_epoch: 1
eval_max_new_tokens: 128
eval_table_size: null
evals_per_epoch: null
flash_attention: false
fp16: false
fsdp: null
fsdp_config: null
gradient_accumulation_steps: 4
gradient_checkpointing: true
group_by_length: true
hub_model_id: error577/89245e1b-674e-4a9a-8b43-84b8b262ab04
hub_repo: null
hub_strategy: checkpoint
hub_token: null
learning_rate: 0.0002
load_in_4bit: false
load_in_8bit: false
local_rank: null
logging_steps: 1
lora_alpha: 64
lora_dropout: 0.1
lora_fan_in_fan_out: null
lora_model_dir: null
lora_r: 32
lora_target_linear: true
lr_scheduler: cosine
max_grad_norm: 1.0
max_steps: null
micro_batch_size: 24
mlflow_experiment_name: /tmp/881619c188d8a836_train_data.json
model_type: AutoModelForCausalLM
num_epochs: 10
optimizer: adamw_bnb_8bit
output_dir: miner_id_24
pad_to_sequence_len: true  # Включить паддинг
resume_from_checkpoint: null
s2_attention: null
sample_packing: false
sequence_len: 256  # Убедитесь, что это значение соответствует данным
special_tokens:
  pad_token: <|endoftext|>
strict: false
tf32: false
tokenizer_type: AutoTokenizer
train_on_inputs: false
trust_remote_code: true
val_set_size: 0.005
wandb_entity: null
wandb_mode: online
wandb_name: 48b6d3b7-0547-4a96-89cf-7c557a1f7bdf
wandb_project: Gradients-On-Demand
wandb_run: your_name
wandb_runid: 48b6d3b7-0547-4a96-89cf-7c557a1f7bdf
warmup_steps: 30
weight_decay: 0.0
xformers_attention: null

89245e1b-674e-4a9a-8b43-84b8b262ab04

This model is a fine-tuned version of katuni4ka/tiny-random-codegen2 on the None dataset. It achieves the following results on the evaluation set:

• Loss: 10.6424

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: 24
• eval_batch_size: 24
• seed: 42
• gradient_accumulation_steps: 4
• total_train_batch_size: 96
• 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: 30
• num_epochs: 10

Training results

Training Loss	Epoch	Step	Validation Loss
42.6464	0.9985	513	10.6925
42.5854	1.9981	1026	10.6691
42.5595	2.9976	1539	10.6574
42.5066	3.9990	2053	10.6518
42.5097	4.9985	2566	10.6480
42.5252	5.9981	3079	10.6453
42.5139	6.9976	3592	10.6438
42.5047	7.9990	4106	10.6426
42.478	8.9985	4619	10.6423
42.5825	9.9942	5130	10.6424

Framework versions

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