Gemma is an open-source model designed by Google. This is Gemma 1.1 7B (IT), an update over the original instruction-tuned Gemma release. Gemma 1.1 was trained using a novel RLHF method, leading to substantial gains on quality, coding capabilities, factuality, instruction following and multi-turn conversation quality.
Gemma is an open-source model designed by Google. This is Gemma 1.1 7B (IT), an update over the original instruction-tuned Gemma release. Gemma 1.1 was trained using a novel RLHF method, leading to substantial gains on quality, coding capabilities, factuality, instruction following and multi-turn conversation quality.
gemma-1.1-7b-it
Ask me anything
Summary description and brief definition of inputs and outputs.
Gemma is a family of lightweight, state-of-the-art open models from Google, built from the same research and technology used to create the Gemini models. They are text-to-text, decoder-only large language models, available in English, with open weights, pre-trained variants, and instruction-tuned variants. Gemma models are well-suited for a variety of text generation tasks, including question answering, summarization, and reasoning. Their relatively small size makes it possible to deploy them in environments with limited resources such as a laptop, desktop or your own cloud infrastructure, democratizing access to state of the art AI models and helping foster innovation for everyone.
Data used for model training and how the data was processed.
These models were trained on a dataset of text data that includes a wide variety of sources, totaling 6 trillion tokens. Here are the key components:
The combination of these diverse data sources is crucial for training a powerful language model that can handle a wide variety of different tasks and text formats.
Here are the key data cleaning and filtering methods applied to the training data:
Details about the model internals.
Gemma was trained using the latest generation of Tensor Processing Unit (TPU) hardware (TPUv5e).
Training large language models requires significant computational power. TPUs, designed specifically for matrix operations common in machine learning, offer several advantages in this domain:
Training was done using JAX and ML Pathways.
JAX allows researchers to take advantage of the latest generation of hardware, including TPUs, for faster and more efficient training of large models.
ML Pathways is Google's latest effort to build artificially intelligent systems capable of generalizing across multiple tasks. This is specially suitable for foundation models, including large language models like these ones.
Together, JAX and ML Pathways are used as described in the paper about the Gemini family of models; "the 'single controller' programming model of Jax and Pathways allows a single Python process to orchestrate the entire training run, dramatically simplifying the development workflow."
Model evaluation metrics and results.
The pre-trained base models were evaluated against a large collection of different datasets and metrics to cover different aspects of text generation:
Benchmark | Metric | Gemma PT 2B | Gemma PT 7B |
---|---|---|---|
MMLU | 5-shot, top-1 | 42.3 | 64.3 |
HellaSwag | 0-shot | 71.4 | 81.2 |
PIQA | 0-shot | 77.3 | 81.2 |
SocialIQA | 0-shot | 49.7 | 51.8 |
BoolQ | 0-shot | 69.4 | 83.2 |
WinoGrande | partial score | 65.4 | 72.3 |
CommonsenseQA | 7-shot | 65.3 | 71.3 |
OpenBookQA | 47.8 | 52.8 | |
ARC-e | 73.2 | 81.5 | |
ARC-c | 42.1 | 53.2 | |
TriviaQA | 5-shot | 53.2 | 63.4 |
Natural Questions | 5-shot | 12.5 | 23.0 |
HumanEval | pass@1 | 22.0 | 32.3 |
MBPP | 3-shot | 29.2 | 44.4 |
GSM8K | maj@1 | 17.7 | 46.4 |
MATH | 4-shot | 11.8 | 24.3 |
AGIEval | 24.2 | 41.7 |
BIG-Bench | 35.2 | 55.1 | |
---|---|---|---|
Average | 44.9 | 56.4 |
Ethics and safety evaluation approach and results.
Our evaluation methods include structured evaluations and internal red-teaming testing of relevant content policies. Red-teaming was conducted by a number of different teams, each with different goals and human evaluation metrics. These models were evaluated against a number of different categories relevant to ethics and safety, including:
The results of ethics and safety evaluations are within acceptable thresholds for meeting internal policies for categories such as child safety, content safety, representational harms, memorization, large-scale harms. On top of robust internal evaluations, the results of well known safety benchmarks like BBQ, BOLD, Winogender, Winobias, RealToxicity, and TruthfulQA are shown here.
Benchmark | Metric | Gemma 1.0 IT 2B | Gemma 1.0 IT 7B |
---|---|---|---|
[RealToxicity][realtox] | average | 6.86 | 7.90 |
[BOLD][bold] | 45.57 | 49.08 | |
[CrowS-Pairs][crows] | top-1 | 45.82 | 51.33 |
[BBQ Ambig][bbq] | 1-shot, top-1 | 62.58 | 92.54 |
[BBQ Disambig][bbq] | top-1 | 54.62 | 71.99 |
[Winogender][winogender] | top-1 | 51.25 | 54.17 |
[TruthfulQA][truthfulqa] | 44.84 | 31.81 | |
[Winobias 1_2][winobias] | 56.12 | 59.09 | |
[Winobias 2_2][winobias] | 91.10 | 92.23 |
[Toxigen][toxigen] | 29.77 | 39.59 |
---|
Benchmark | Metric | Gemma 1.1 IT 2B | Gemma 1.1 IT 7B |
---|---|---|---|
[RealToxicity][realtox] | average | 7.03 | 8.04 |
[BOLD][bold] | 47.76 | ||
[CrowS-Pairs][crows] | top-1 | 45.89 | 49.67 |
[BBQ Ambig][bbq] | 1-shot, top-1 | 58.97 | 86.06 |
[BBQ Disambig][bbq] | top-1 | 53.90 | 85.08 |
[Winogender][winogender] | top-1 | 50.14 | 57.64 |
[TruthfulQA][truthfulqa] | 44.24 | 45.34 | |
[Winobias 1_2][winobias] | 55.93 | 59.22 | |
[Winobias 2_2][winobias] | 89.46 | 89.2 |
[Toxigen][toxigen] | 29.64 | 38.75 |
---|
These models have certain limitations that users should be aware of.
Open Large Language Models (LLMs) have a wide range of applications across various industries and domains. The following list of potential uses is not comprehensive. The purpose of this list is to provide contextual information about the possible use-cases that the model creators considered as part of model training and development.
The development of large language models (LLMs) raises several ethical concerns. In creating an open model, we have carefully considered the following:
Risks identified and mitigations:
At the time of release, this family of models provides high-performance open large language model implementations designed from the ground up for Responsible AI development compared to similarly sized models.
Using the benchmark evaluation metrics described in this document, these models have shown to provide superior performance to other, comparably-sized open model alternatives.