[whatmodel.sh](../../README.md) / LaMDA 422M (pre-trained) # LaMDA 422M (pre-trained) Google · released 2022-01 · 0.422B params · proprietary **score 0.7** ±3.0 · comparison only: no current run option verified · from 9 benchmark results across 1 source · † no independently run results verified Score within the [LMArena (text)](https://whatmodel.sh/?set=lmarena-text&score=set) comparison set only. ## What it's good at ● marks LaMDA 422M (pre-trained) among every model on that task or test, worst to best. 
task  vs the others          rank  counts  tests
math  ●▁▃▆██▅█▃▁▁▁   #528 of 530*    4.8%      1
math · #528 of 530 · 4.8% of its score Trails [GPT-2 Small (124M)](gpt-2-small.md), [GPT-Neo 125M](gpt-neo-125m.md), [OPT-125M](opt-125m.md) here, though they rank lower overall. 
test           result  vs the others in its run  difficulty      counts  why
GSM8K (maker)    0.4%  ●·▅▃▃▃·▃·▃▃▅ beats 0/12   easy · avg 67%    4.8%  sharp · too hard for it
other tests (8) 
test                                result  vs the others in its run  difficulty           counts  why
MAWPS (CoT-paper union) (maker)       0.9%  ●▃█···▅·▃▃▃▅ beats 0/12   very easy · avg 94%     29%  sharp · too hard for it
ASDiv (maker)                         0.8%  ●▃▃▅··▃▃·▃▃█ beats 0/12   easy · avg 83%          22%  sharp · too hard for it
SVAMP (maker)                         1.6%  ▃●█···▅·▃▃▃▅ beats 1/12   easy · avg 82%          15%  sharp · too hard for it
BIG-bench Date Understanding (mak…    1.9%  ●·▅·▅▃▅·▃▃·█ beats 0/12   medium · avg 52%        14%  too hard for it
StrategyQA (maker)                   46.4%  ●·····▅▃▃▅▃█ beats 0/12   medium · avg 70%       5.1%  too hard for it
BIG-bench Sports Understanding (m…   50.0%  ●·····▃·▁·▄▁ beats 2/12   medium · avg 81%       3.0%  sharp · too hard for it
CommonsenseQA (maker)                20.1%  ●······▁·▁▁▃ beats 2/12   easy · avg 88%         2.9%  sharp · too hard for it
AQuA-RAT (maker)                      8.3%  ●▁·▁▁·····▂▁ beats 3/12   medium · avg 48%       0.3%  sharp · too hard for it
## 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
Chain-of-Thought Prompting Elicits Reas…        9   96% ▇▇▇▇▇▇▇▇▇▇▇▇
prior: pull toward the middle                        4% ▇
weighted average                                   = 0.7
Licences: Chain-of-Thought Prompting Elicits Reasoning in Large Language Models ([CC BY 4.0](../../licenses/paper-2201.11903v6-CC-BY-4-0.txt)) ## How to run it No verified API or local download yet. [interactive version](https://whatmodel.sh/model/lamda-422m) · [JSON](https://whatmodel.sh/api?format=json&all=1)