SmolLM-14m-Dolma-v0.1-Proposed
This model is a fine-tuned version of allenai/OLMo-7B-0724-hf on the pico-lm/pretokenized-dolma 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: 0.003
- train_batch_size: 256
- eval_batch_size: 8
- seed: 42
- distributed_type: multi-GPU
- num_devices: 8
- total_train_batch_size: 2048
- total_eval_batch_size: 64
- optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
- lr_scheduler_type: warmup_stable_decay
- lr_scheduler_warmup_steps: 1000
- num_epochs: 1
Training results
Framework versions
- Transformers 4.47.1
- Pytorch 2.5.1
- Datasets 3.2.0
- Tokenizers 0.21.0
Model tree for Cambridge-KAIST/SmolLM-14m-Dolma-v0.1-Proposed-full
Base model
allenai/OLMo-7B-0724-hf