[whatmodel.sh](../../README.md) / Falcon-RW 1B # Falcon-RW 1B TII · released 2023-04 · 1.3B params · apache-2.0 **score 19.4** ±1.9 · #703 of 732 · from 12 benchmark results across 2 sources Score within the [LMArena (text)](https://whatmodel.sh/?set=lmarena-text&score=set) comparison set only. ## What it's good at _{● marks Falcon-RW 1B among every model on that task or test, worst to best.}

other tests (12)

test                                result  vs the others in its run  difficulty        counts  why
ARC Challenge                        28.2%  ▅····▃●···▅█ beats 3/8    easy · avg 76%       17%  sharp · too hard for it
Natural Questions (closed book)       9.3%  ▁·▁·▁▂▁●·▅▂█ beats 9/28   medium · avg 48%     14%  typical
WinoGrande                           60.7%  █··●···█▄▄·▄ beats 2/8    easy · avg 87%       12%  typical
BoolQ                                63.2%  ▄█●···▄··█▄▄ beats 3/8    easy · avg 88%       12%  at its level
HellaSwag                            46.6%  ▃▃▃··●····▃█ beats 3/8    easy · avg 81%       11%  at its level
ARC Easy                             63.3%  ▅·▃●··▃··▃·█ beats 3/8    easy · avg 87%       10%  blunt · at its level
PIQA                                 74.7%  ▄▄··▄·●·▄·██ beats 3/8    easy · avg 87%      6.4%  blunt · at its level
Social IQA                           40.5%  █●····█▄▄··▄ beats 2/7    medium · avg 60%    6.0%  blunt
OpenBookQA                           24.4%  ●······▂·▆·▂ beats 3/8    hard · avg 49%      4.4%  too hard for it · noisy
TriviaQA                             31.4%  ▁··▁█▃▃▁●▅▄█ beats 13/28  easy · avg 73%      2.9%  noisy
TruthfulQA MC2                       36.0%  ●▃█▆▁▂▅···▁▁ beats 4/31   medium · avg 57%    1.0%  blunt · at its level
HaluEval QA hallucination detecti…   42.2%  ●▁▁·▁▁·▁▁▁▁▁ beats 0/27   hard · avg 66%      0.5%  too hard for it · noisy

## How the score was worked out Each result implies a score on its own, after adjusting for how hard that test is and who ran it. The model's score is their weighted average, with a small pull toward the middle for thinly tested models. Each test's weight is its *counts* column above; here it is added up by who ran the tests. [Method](../method.md)

source                                    results  share of the score
Textbooks Are All You Need II: phi-1.5 …        8   80% ▇▇▇▇▇▇▇▇▇▇
Hallucinations Leaderboard results              4   18% ▇▇
prior: pull toward the middle                        2% ▇
weighted average                                   = 19.4

_{Licences: Textbooks Are All You Need II: phi-1.5 technical report ([CC BY 4.0](../../licenses/paper-2309.05463v1-CC-BY-4.0.txt)) · Hallucinations Leaderboard results ([Apache-2.0](https://www.apache.org/licenses/LICENSE-2.0))} ## How to run it ### Other ways to run it 1. Run it as published (BF16, needs ~2.62 GB, from [huggingface.co/tiiuae](https://huggingface.co/tiiuae/falcon-rw-1b/tree/e4b9872bb803165eb22f0a867d4e6a64d34fce19)) ```sh uv run --with torch --with 'transformers<5' python -c "from transformers import pipeline; print(pipeline('text-generation', model='tiiuae/falcon-rw-1b')('Once upon a time', max_new_tokens=40)[0]['generated_text'])" ``` Model repository and documentation: [huggingface.co/tiiuae](https://huggingface.co/tiiuae/falcon-rw-1b) [interactive version](https://whatmodel.sh/model/falcon-rw-1b) · [JSON](https://whatmodel.sh/api?format=json&all=1)