See axolotl config

axolotl version: 0.4.1

adapter: lora
base_model: unsloth/gemma-2-2b
bf16: true
chat_template: llama3
dataset_prepared_path: null
datasets:
- data_files:
  - d0528b9d54249648_train_data.json
  ds_type: json
  format: custom
  path: /workspace/input_data/d0528b9d54249648_train_data.json
  type:
    field_instruction: user_prompt
    field_output: resp
    format: '{instruction}'
    no_input_format: '{instruction}'
    system_format: '{system}'
    system_prompt: ''
debug: null
deepspeed: null
early_stopping_patience: 2
eval_max_new_tokens: 128
eval_steps: 100
eval_table_size: null
flash_attention: false
fp16: null
fsdp: null
fsdp_config: null
gradient_accumulation_steps: 8
gradient_checkpointing: true
group_by_length: false
hub_model_id: Romain-XV/ae78327e-46b5-43a2-9eba-22e1a060d5df
hub_repo: null
hub_strategy: checkpoint
hub_token: null
learning_rate: 0.0002
load_best_model_at_end: true
load_in_4bit: false
load_in_8bit: false
local_rank: null
logging_steps: 1
lora_alpha: 128
lora_dropout: 0.3
lora_fan_in_fan_out: null
lora_model_dir: null
lora_r: 64
lora_target_linear: true
lr_scheduler: cosine
max_grad_norm: 1.0
max_steps: 1734
micro_batch_size: 4
mlflow_experiment_name: /tmp/d0528b9d54249648_train_data.json
model_type: AutoModelForCausalLM
num_epochs: 2
optimizer: adamw_bnb_8bit
output_dir: miner_id_24
pad_to_sequence_len: true
resume_from_checkpoint: null
s2_attention: null
sample_packing: false
save_steps: 100
sequence_len: 1024
strict: false
tf32: true
tokenizer_type: AutoTokenizer
train_on_inputs: false
trust_remote_code: true
use_rslora: true
val_set_size: 0.04
wandb_entity: null
wandb_mode: online
wandb_name: 5b3b6012-6204-4a48-bfe1-e7436b72b5a9
wandb_project: Gradients-On-Demand
wandb_run: your_name
wandb_runid: 5b3b6012-6204-4a48-bfe1-e7436b72b5a9
warmup_steps: 10
weight_decay: 0.0
xformers_attention: null

ae78327e-46b5-43a2-9eba-22e1a060d5df

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: 0.1213

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
• gradient_accumulation_steps: 8
• total_train_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: 10
• training_steps: 1734

Training results

Training Loss	Epoch	Step	Validation Loss
0.9986	0.0009	1	0.9960
0.1729	0.0921	100	0.1579
0.1433	0.1842	200	0.1625
0.1339	0.2763	300	0.1528
0.1583	0.3684	400	0.1520
0.1612	0.4605	500	0.1480
0.1726	0.5526	600	0.1428
0.1168	0.6447	700	0.1392
0.139	0.7368	800	0.1341
0.1811	0.8289	900	0.1318
0.1509	0.9210	1000	0.1283
0.0641	1.0131	1100	0.1257
0.1248	1.1052	1200	0.1249
0.0814	1.1973	1300	0.1243
0.1157	1.2894	1400	0.1225
0.1077	1.3815	1500	0.1221
0.1244	1.4736	1600	0.1217
0.0909	1.5657	1700	0.1213

Framework versions

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