INDEX

Explanations

terms related to voter participation and enrollment, especially in political contexts

oai_token-act-pair · gpt-3.5-turbo

terms related to voter turnout and enrollment

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.hook_resid_pre

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

torch.float32

Hook Point

blocks.8.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

olog

-0.72

uran

-0.72

atom

-0.71

thur

-0.69

sonian

-0.68

esthetic

-0.66

 propri

-0.66

hair

-0.65

icus

-0.64

 topical

-0.64

POSITIVE LOGITS

 rates

1.09

 rate

1.03

 numbers

0.94

 Rates

0.91

 Rate

0.88

 surges

0.86

 percentages

0.84

 ratios

0.83

 Estimates

0.83

 levels

0.82

Activations Density 0.170%

terms related to voter participation and enrollment, especially in political contexts

terms related to voter turnout and enrollment

No Comments

No Known Activations