metadata
library_name: peft
license: apache-2.0
base_model: unsloth/SmolLM-360M-Instruct
tags:
- axolotl
- generated_from_trainer
model-index:
- name: f263595e-91dc-4dbc-aee8-42a3b57e2dde
results: []
See axolotl config
axolotl version: 0.4.1
adapter: lora
base_model: unsloth/SmolLM-360M-Instruct
bf16: auto
chat_template: llama3
dataset_prepared_path: null
datasets:
- data_files:
- 902ecde58c94c532_train_data.json
ds_type: json
format: custom
path: /workspace/input_data/902ecde58c94c532_train_data.json
type:
field_input: original_version
field_instruction: title
field_output: french_version
format: '{instruction} {input}'
no_input_format: '{instruction}'
system_format: '{system}'
system_prompt: ''
debug: null
deepspeed: null
device_map: auto
do_eval: true
early_stopping_patience: 3
eval_batch_size: 4
eval_max_new_tokens: 128
eval_steps: 300
eval_table_size: null
evals_per_epoch: null
flash_attention: true
fp16: false
fsdp: null
fsdp_config: null
gradient_accumulation_steps: 4
gradient_checkpointing: false
group_by_length: true
hub_model_id: robiulawaldev/f263595e-91dc-4dbc-aee8-42a3b57e2dde
hub_strategy: end
learning_rate: 0.0002
load_in_4bit: false
load_in_8bit: false
local_rank: null
logging_steps: 50
lora_alpha: 64
lora_dropout: 0.05
lora_fan_in_fan_out: null
lora_model_dir: null
lora_r: 32
lora_target_linear: true
lr_scheduler: constant
max_grad_norm: 1.0
max_memory:
0: 75GB
max_steps: 17953
micro_batch_size: 4
mlflow_experiment_name: /tmp/902ecde58c94c532_train_data.json
model_type: AutoModelForCausalLM
num_epochs: 10
optim_args:
adam_beta1: 0.9
adam_beta2: 0.95
adam_epsilon: 1e-5
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: 300
saves_per_epoch: null
sequence_len: 512
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: fb292aae-3bf8-4614-82a9-5c9ce7b3f999
wandb_project: SN56-36
wandb_run: your_name
wandb_runid: fb292aae-3bf8-4614-82a9-5c9ce7b3f999
warmup_steps: 50
weight_decay: 0.0
xformers_attention: null
f263595e-91dc-4dbc-aee8-42a3b57e2dde
This model is a fine-tuned version of unsloth/SmolLM-360M-Instruct on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.9729
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: 4
- total_train_batch_size: 16
- optimizer: Use OptimizerNames.ADAMW_BNB 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: constant
- lr_scheduler_warmup_steps: 50
- training_steps: 17953
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| No log | 0.0002 | 1 | 2.4685 |
| 1.6913 | 0.0510 | 300 | 1.6545 |
| 1.5228 | 0.1021 | 600 | 1.5083 |
| 1.4568 | 0.1531 | 900 | 1.4199 |
| 1.3707 | 0.2042 | 1200 | 1.3690 |
| 1.3347 | 0.2552 | 1500 | 1.3241 |
| 1.2919 | 0.3063 | 1800 | 1.2925 |
| 1.2462 | 0.3573 | 2100 | 1.2656 |
| 1.2175 | 0.4083 | 2400 | 1.2438 |
| 1.2624 | 0.4594 | 2700 | 1.2318 |
| 1.216 | 0.5104 | 3000 | 1.2081 |
| 1.2401 | 0.5615 | 3300 | 1.1880 |
| 1.2172 | 0.6125 | 3600 | 1.1763 |
| 1.1768 | 0.6635 | 3900 | 1.1606 |
| 1.1733 | 0.7146 | 4200 | 1.1570 |
| 1.1503 | 0.7656 | 4500 | 1.1437 |
| 1.1261 | 0.8167 | 4800 | 1.1351 |
| 1.124 | 0.8677 | 5100 | 1.1233 |
| 1.1614 | 0.9188 | 5400 | 1.1161 |
| 1.1346 | 0.9698 | 5700 | 1.1063 |
| 1.0797 | 1.0208 | 6000 | 1.1005 |
| 1.0431 | 1.0719 | 6300 | 1.0961 |
| 1.0795 | 1.1229 | 6600 | 1.0894 |
| 1.0587 | 1.1740 | 6900 | 1.0853 |
| 1.0899 | 1.2250 | 7200 | 1.0778 |
| 1.0412 | 1.2761 | 7500 | 1.0717 |
| 1.0829 | 1.3271 | 7800 | 1.0683 |
| 1.0652 | 1.3781 | 8100 | 1.0639 |
| 1.0164 | 1.4292 | 8400 | 1.0583 |
| 1.0589 | 1.4802 | 8700 | 1.0534 |
| 1.0337 | 1.5313 | 9000 | 1.0461 |
| 1.0161 | 1.5823 | 9300 | 1.0440 |
| 1.0422 | 1.6333 | 9600 | 1.0420 |
| 1.0025 | 1.6844 | 9900 | 1.0345 |
| 0.9963 | 1.7354 | 10200 | 1.0337 |
| 1.0322 | 1.7865 | 10500 | 1.0311 |
| 1.0424 | 1.8375 | 10800 | 1.0278 |
| 0.9842 | 1.8886 | 11100 | 1.0204 |
| 0.9802 | 1.9396 | 11400 | 1.0150 |
| 0.9941 | 1.9906 | 11700 | 1.0127 |
| 0.9759 | 2.0417 | 12000 | 1.0113 |
| 0.9635 | 2.0927 | 12300 | 1.0088 |
| 0.941 | 2.1438 | 12600 | 1.0050 |
| 0.9635 | 2.1948 | 12900 | 1.0041 |
| 0.9709 | 2.2459 | 13200 | 1.0028 |
| 0.9631 | 2.2969 | 13500 | 1.0006 |
| 0.9533 | 2.3479 | 13800 | 0.9955 |
| 0.9762 | 2.3990 | 14100 | 0.9941 |
| 0.9899 | 2.4500 | 14400 | 0.9924 |
| 0.9706 | 2.5011 | 14700 | 0.9898 |
| 0.9315 | 2.5521 | 15000 | 0.9859 |
| 0.9224 | 2.6031 | 15300 | 0.9868 |
| 1.0113 | 2.6542 | 15600 | 0.9829 |
| 0.9251 | 2.7052 | 15900 | 0.9817 |
| 1.0008 | 2.7563 | 16200 | 0.9793 |
| 0.9723 | 2.8073 | 16500 | 0.9787 |
| 0.9673 | 2.8584 | 16800 | 0.9739 |
| 0.9519 | 2.9094 | 17100 | 0.9714 |
| 0.9699 | 2.9604 | 17400 | 0.9713 |
| 0.93 | 3.0115 | 17700 | 0.9729 |
Framework versions
- PEFT 0.13.2
- Transformers 4.46.0
- Pytorch 2.5.0+cu124
- Datasets 3.0.1
- Tokenizers 0.20.1