See axolotl config

axolotl version: 0.4.1

adapter: lora
base_model: unsloth/gemma-2-2b
bf16: auto
chat_template: llama3
cosine_min_lr_ratio: 0.1
data_processes: 16
dataset_prepared_path: null
datasets:
- data_files:
  - 32799483a740097f_train_data.json
  ds_type: json
  format: custom
  path: /workspace/input_data/32799483a740097f_train_data.json
  type:
    field_input: gender
    field_instruction: en
    field_output: es
    format: '{instruction} {input}'
    no_input_format: '{instruction}'
    system_format: '{system}'
    system_prompt: ''
debug: null
deepspeed: null
device_map: '{'''':torch.cuda.current_device()}'
do_eval: true
early_stopping_patience: 60
eval_batch_size: 1
eval_sample_packing: false
eval_steps: 25
evaluation_strategy: steps
flash_attention: true
fp16: null
fsdp: null
fsdp_config: null
gradient_accumulation_steps: 64
gradient_checkpointing: true
group_by_length: true
hub_model_id: sn56m3/f7a9fbea-d994-4e0e-bfbf-06206445ed8f
hub_repo: stevemonite
hub_strategy: checkpoint
hub_token: null
learning_rate: 0.0001
load_in_4bit: false
load_in_8bit: false
local_rank: null
logging_steps: 1
lora_alpha: 32
lora_dropout: 0.05
lora_fan_in_fan_out: null
lora_model_dir: null
lora_r: 16
lora_target_linear: true
lora_target_modules:
- q_proj
- v_proj
lr_scheduler: cosine
max_grad_norm: 1.0
max_memory:
  0: 70GiB
max_steps: 666
micro_batch_size: 1
mlflow_experiment_name: /tmp/32799483a740097f_train_data.json
model_type: AutoModelForCausalLM
num_epochs: 4
optim_args:
  adam_beta1: 0.9
  adam_beta2: 0.95
  adam_epsilon: 1e-5
optimizer: adamw_torch
output_dir: miner_id_24
pad_to_sequence_len: true
resume_from_checkpoint: null
s2_attention: null
sample_packing: false
save_steps: 50
save_strategy: steps
sequence_len: 2048
strict: false
tf32: false
tokenizer_type: AutoTokenizer
torch_compile: false
train_on_inputs: false
trust_remote_code: true
val_set_size: 50
wandb_entity: sn56-miner
wandb_mode: disabled
wandb_name: null
wandb_project: god
wandb_run: w2k6
wandb_runid: null
warmup_raio: 0.03
warmup_ratio: 0.05
weight_decay: 0.01
xformers_attention: null

f7a9fbea-d994-4e0e-bfbf-06206445ed8f

This model is a fine-tuned version of unsloth/gemma-2-2b on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.8542

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: 0.0001
• train_batch_size: 1
• eval_batch_size: 1
• seed: 42
• distributed_type: multi-GPU
• num_devices: 4
• gradient_accumulation_steps: 64
• total_train_batch_size: 256
• total_eval_batch_size: 4
• optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=adam_beta1=0.9,adam_beta2=0.95,adam_epsilon=1e-5
• lr_scheduler_type: cosine
• lr_scheduler_warmup_steps: 33
• training_steps: 666

Training results

Training Loss	Epoch	Step	Validation Loss
1.6422	0.0008	1	3.9106
0.8483	0.0211	25	1.0149
0.7166	0.0422	50	0.9550
0.7464	0.0633	75	0.9110
0.821	0.0844	100	0.9160
0.6771	0.1055	125	0.9025
0.7014	0.1266	150	0.8940
0.6945	0.1477	175	0.8904
0.728	0.1688	200	0.8883
0.6273	0.1898	225	0.8844
0.7142	0.2109	250	0.8767
0.7023	0.2320	275	0.8791
0.7075	0.2531	300	0.8649
0.6953	0.2742	325	0.8715
0.6841	0.2953	350	0.8684
0.6334	0.3164	375	0.8652
0.7264	0.3375	400	0.8651
0.669	0.3586	425	0.8657
0.67	0.3797	450	0.8620
0.6409	0.4008	475	0.8569
0.6531	0.4219	500	0.8557
0.6698	0.4430	525	0.8561
0.7231	0.4641	550	0.8552
0.6126	0.4852	575	0.8560
0.647	0.5063	600	0.8546
0.689	0.5273	625	0.8545
0.7106	0.5484	650	0.8542

Framework versions

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