Silo-PDSW is a 1.3B parameter, decoder-only language model trained on data in the public domain and under permissive software licenses from the Open License Corpus (OLC).
The model is based on the LLaMA architecture as implemented in (OpenLM)[].
The model is trained with 128 A100 GPUs across 16 nodes.
Model and Training Hyperparameters
We follow the model architecture of LLaMa, and we use the GPT-NeoX-20B tokenizer, with 50432 BPE types.
During training, we use 2,048 token sequences that are packed across document boundaries, and we pre-pend a beginning-of-text token to every document.
We use weight decay of 0.1, the Adam optimizer with beta_2 of 0.95, 2,000 steps of warmup, with a cosine learning rate scheduler.
Model
#L
#H
d_model
LR
Batch
1.3B
24
16
2048
1e-3
2.6M
Training data
Silo-PDSW was trained on data in the public domain and under permissive software licenses from the Open License Corpus (OLC).
The model was trained on the following domain proportions (please see the OLC repository for more details on the data sources for each domain):
Domain
Tokens (B)
%
Code
58.9
59.1
Legal
27.1
27.2
Conversation
5.9
5.9
Math
3.5
3.5
Books
2.9
2.9
Science
1.2
1.2
News
0.2
0.2
Total
99.6
100.0
We train with early stopping for 250B tokens in total, or a little more than two epochs of training over this subset
Since the distribution of OLC is highly skewed, we perform a simple upweighting scheme where we upsample all data that accounts for less than 5% of the corpus by a factor of 3x, which we found to work well after a sweep of different settings.
Intended Uses and Limitations
This model can be used for prompting for evaluation of downstream tasks as well as text generation.
How to use
You can use this model directly with a pipeline for text generation.
python
1from transformers import pipeline
2generator = pipeline('text-generation', model="kernelmachine/silo-pdsw-1.3b", device='cuda')3generator("Hello")4[{'generated_text':"Hello, I'm a new user of Ubuntu. I'm trying to install the latest version of Ubuntu"}]
By default, generation is deterministic. In order to use the top-k sampling, please set do_sample to True.
Silo-PDSW inherits the biases and limitations of public domain data, which carry risks of toxic or otherwise unfair output, due to the prevalence of older copyright-expired text.
Silo-PDSW may also output personally identifiable information, because we did not filter that out of training data.
