adapter: lora
attn_implementation: eager
base_model: openai/gpt-oss-20b
bf16: true
chat_template: llama3
cosine_min_lr_ratio: 0.3
dataloader_num_workers: 12
dataset_prepared_path: null
datasets:
- data_files:
  - 81d99e01-29b1-4f69-a23e-0dba6a297872_train_data.json
  ds_type: json
  format: custom
  path: /workspace/axolotl/data
  type:
    field_instruction: instruct
    field_output: output
    format: '{instruction}'
    no_input_format: '{instruction}'
    system_format: '{system}'
    system_prompt: ''
ddp: true
debug: null
deepspeed: null
device_map: cuda
early_stopping_patience: null
eval_max_new_tokens: 128
eval_steps: null
eval_table_size: null
evals_per_epoch: null
flash_attention: true
fp16: false
fsdp: null
fsdp_config: null
gradient_accumulation_steps: 1
gradient_checkpointing: true
gradient_checkpointing_kwargs:
  use_reentrant: false
group_by_length: true
hub_model_id: null
hub_private_repo: false
hub_repo: null
hub_strategy: checkpoint
hub_token: null
learning_rate: 0.0002
liger_fused_linear_cross_entropy: true
liger_glu_activation: true
liger_layer_norm: true
liger_rms_norm: true
liger_rope: true
load_in_4bit: false
load_in_8bit: false
local_rank: null
logging_steps: null
lora_alpha: 64
lora_dropout: 0.05
lora_fan_in_fan_out: null
lora_model_dir: null
lora_r: 32
lora_target_linear: true
loraplus_lr_embedding: 1.0e-06
loraplus_lr_ratio: 16
lr_scheduler: cosine
max_grad_norm: 1
max_steps: 4257
micro_batch_size: 12
mlflow_experiment_name: /workspace/axolotl/data/81d99e01-29b1-4f69-a23e-0dba6a297872_train_data.json
model_card: false
model_type: AutoModelForCausalLM
num_epochs: 200
optimizer: adamw_bnb_8bit
output_dir: /app/checkpoints/81d99e01-29b1-4f69-a23e-0dba6a297872/de7fae79-1bd5-4ce4-b6fa-0e39b70b101f
pad_to_sequence_len: true
plugins:
- axolotl.integrations.liger.LigerPlugin
push_every_save: true
push_to_hub: true
resume_from_checkpoint: null
rl: null
s2_attention: null
sample_packing: true
save_steps: 100
save_strategy: steps
save_total_limit: 1
saves_per_epoch: 0
sequence_len: 512
strict: false
tf32: true
tokenizer_type: AutoTokenizer
train_on_inputs: false
trl: null
trust_remote_code: false
use_flash_attention: false
use_liger: true
val_set_size: 0.0
wandb_mode: offline
wandb_name: 81d99e01-29b1-4f69-a23e-0dba6a297872_de7fae79-1bd5-4ce4-b6fa-0e39b70b101f
wandb_project: Gradients-On-Demand
wandb_run: null
wandb_runid: 81d99e01-29b1-4f69-a23e-0dba6a297872_de7fae79-1bd5-4ce4-b6fa-0e39b70b101f
warmup_steps: 200
weight_decay: 0
xformers_attention: null

app/checkpoints/81d99e01-29b1-4f69-a23e-0dba6a297872/de7fae79-1bd5-4ce4-b6fa-0e39b70b101f

This model was trained from scratch on the None dataset.

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: 12
eval_batch_size: 12
seed: 42
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: 200
training_steps: 4257

Training results

Framework versions

PEFT 0.17.0
Transformers 4.55.0
Pytorch 2.7.1+cu128
Datasets 4.0.0
Tokenizers 0.21.4

apriasmoro/de7fae79-1bd5-4ce4-b6fa-0e39b70b101f

app/checkpoints/81d99e01-29b1-4f69-a23e-0dba6a297872/de7fae79-1bd5-4ce4-b6fa-0e39b70b101f

Model description

Intended uses & limitations

Training and evaluation data

Training procedure

Training hyperparameters

Training results

Framework versions

apriasmoro/
de7fae79-1bd5-4ce4-b6fa-0e39b70b101f