metadata

license: other
language:
  - en
tags:
  - causal-lm

`Stable LM 2 1.6B` (global_step420000)

Description

Stable LM 2 1.6B is a 1.6 billion parameter decoder-only language model pre-trained on 2 trillion tokens of diverse multilingual and code datasets for two epochs.

Usage

This branch contains the training checkpoint for Stable LM 2 1.6B at step 420,000. It is the final checkpoint taken before cooldown. We provide the following contents in the global_step420000 directory:

bf16_zero_pp_mp_rank_00_optim_states.pt: The Adam states and FP32 weights for each parameter. You will need to port this to your optimizer format when importing into your training process.
mp_rank_00_model_states.pt: The model weights following the GPT-NeoX convention.
config.yml: The pre-training configuration file for this checkpoint. Linear learning rate cooldown should be taken from lr=0.0002529 to lr=0.0.

The model weights are also converted to HuggingFace transformers format and can be loaded with the following code:

from transformers import AutoModelForCausalLM, AutoTokenizer
tokenizer = AutoTokenizer.from_pretrained("stabilityai/stablelm-2-1_6b", trust_remote_code=True)
model = AutoModelForCausalLM.from_pretrained(
  "stabilityai/stablelm-2-1_6b",
  trust_remote_code=True,
  torch_dtype="auto",
  revision="global_step420000"
)
model.cuda()

License

License: Stability AI Non-Commercial Research Community License. If you'd like to use this model for commercial products or purposes, please contact us here to learn more.

Acknowledgements

Dakota Mahan for creating the ZeRO optimizer state merging script.

Citation

@misc{StableLM-2-1.6B,
      url={[https://huggingface.co/stabilityai/stablelm-2-1_6b](https://huggingface.co/stabilityai/stablelm-2-1_6b)},
      title={Stable LM 2 1.6B},
      author={Stability AI Language Team}
}