INDEX

Explanations

proper nouns related to political figures

oai_token-act-pair · gpt-3.5-turbo

instances of political discourse or references to political figures

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

 endeavour

-0.90

 enqu

-0.79

ataka

-0.78

 realised

-0.78

 regenerate

-0.78

 practise

-0.78

iga

-0.74

 tyres

-0.73

 licence

-0.73

 colours

-0.73

POSITIVE LOGITS

Republicans

1.21

Asked

1.19

President

1.18

Republican

1.16

Polit

1.13

Democratic

1.09

Politics

1.06

Trump

1.06

Hillary

1.06

Actor

1.05

Activations Density 0.774%

proper nouns related to political figures

instances of political discourse or references to political figures

No Comments

No Known Activations