INDEX

Explanations

mentions of voting outcomes

oai_token-act-pair · gpt-3.5-turbo

references to voting and electoral outcomes

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

 millenn

-0.92

 Palestin

-0.91

 unden

-0.87

£ı

-0.83

 Flavoring

-0.79

 tremend

-0.76

ModLoader

-0.74

 adolesc

-0.73

OLOG

-0.72

ccording

-0.69

POSITIVE LOGITS

board

0.93

vote

0.89

room

0.86

 totals

0.85

 votes

0.85

 turnout

0.81

 Vote

0.80

 vote

0.80

 bloc

0.79

 count

0.78

Activations Density 0.027%

mentions of voting outcomes

references to voting and electoral outcomes

No Comments

No Known Activations