metadata
library_name: peft
license: apache-2.0
base_model: unsloth/SmolLM2-360M
tags:
- axolotl
- generated_from_trainer
model-index:
- name: 0d8f1810-af0c-4b52-973f-9fe030a29e08
results: []
See axolotl config
axolotl version: 0.4.1
adapter: lora
base_model: unsloth/SmolLM2-360M
bf16: auto
chat_template: llama3
dataset_prepared_path: null
datasets:
- data_files:
- 489ba0c0b517eaf7_train_data.json
ds_type: json
format: custom
path: /workspace/input_data/489ba0c0b517eaf7_train_data.json
type:
field_instruction: intent
field_output: example
format: '{instruction}'
no_input_format: '{instruction}'
system_format: '{system}'
system_prompt: ''
debug: null
deepspeed: null
early_stopping_patience: null
eval_max_new_tokens: 128
eval_table_size: null
evals_per_epoch: 4
flash_attention: true
fp16: null
fsdp: null
fsdp_config: null
gradient_accumulation_steps: 4
gradient_checkpointing: true
gradient_clipping: 1.0
group_by_length: false
hub_model_id: null
hub_repo: null
hub_strategy: checkpoint
hub_token: null
learning_rate: 0.0001
load_in_4bit: false
load_in_8bit: false
local_rank: 0
logging_steps: 3
lora_alpha: 16
lora_dropout: 0.05
lora_fan_in_fan_out: null
lora_model_dir: null
lora_r: 8
lora_target_linear: true
lr_scheduler: cosine
max_steps: 100
micro_batch_size: 8
mlflow_experiment_name: /tmp/489ba0c0b517eaf7_train_data.json
model_type: AutoModelForCausalLM
num_epochs: 3
optimizer: adamw_bnb_8bit
output_dir: miner_id_24
pad_to_sequence_len: true
resume_from_checkpoint: null
s2_attention: null
sample_packing: false
saves_per_epoch: 4
sequence_len: 1024
strict: false
tf32: false
tokenizer_type: AutoTokenizer
train_on_inputs: false
trust_remote_code: true
val_set_size: 0.05
wandb_entity: techspear-hub
wandb_mode: online
wandb_name: 05ecb6f6-029b-4441-9ec9-65299a737077
wandb_project: Gradients-On-Two
wandb_run: your_name
wandb_runid: 05ecb6f6-029b-4441-9ec9-65299a737077
warmup_steps: 10
weight_decay: 0.01
xformers_attention: null
0d8f1810-af0c-4b52-973f-9fe030a29e08
This model is a fine-tuned version of unsloth/SmolLM2-360M on the None dataset. It achieves the following results on the evaluation set:
- Loss: 3.5380
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: 8
- eval_batch_size: 8
- seed: 42
- gradient_accumulation_steps: 4
- 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: 100
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| No log | 0.0023 | 1 | 5.7676 |
| 5.7661 | 0.0209 | 9 | 5.7401 |
| 5.2717 | 0.0419 | 18 | 5.5136 |
| 5.1694 | 0.0628 | 27 | 5.0614 |
| 4.6917 | 0.0837 | 36 | 4.5808 |
| 4.1217 | 0.1047 | 45 | 4.1795 |
| 3.8083 | 0.1256 | 54 | 3.9013 |
| 3.6188 | 0.1465 | 63 | 3.7237 |
| 3.654 | 0.1674 | 72 | 3.6171 |
| 3.4858 | 0.1884 | 81 | 3.5632 |
| 3.5462 | 0.2093 | 90 | 3.5419 |
| 3.4723 | 0.2302 | 99 | 3.5380 |
Framework versions
- PEFT 0.13.2
- Transformers 4.46.0
- Pytorch 2.5.0+cu124
- Datasets 3.0.1
- Tokenizers 0.20.1