INDEX

Explanations

locations and organizations related to governmental and political activities

oai_token-act-pair · gpt-3.5-turbo

mentions of government entities or political organizations

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

cffff

-0.62

$.

-0.61

venge

-0.58

DonaldTrump

-0.56

 dstg

-0.55

âĹ¼

-0.55

ĸļ

-0.54

bryce

-0.53

åĤ

-0.53

performing

-0.53

POSITIVE LOGITS

 spokesman

0.93

 spokeswoman

0.87

 meanwhile

0.85

 reacted

0.85

 countered

0.78

 responded

0.78

 spokesperson

0.78

 applauded

0.74

 intervened

0.72

 also

0.70

Activations Density 0.945%

locations and organizations related to governmental and political activities

mentions of government entities or political organizations

No Comments

No Known Activations