INDEX

Explanations

proper nouns related to media and entertainment

oai_token-act-pair · gpt-3.5-turbo

mentions of the brand "Kotaku."

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

xual

-1.01

 tremend

-0.79

 confir

-0.76

mble

-0.75

utherford

-0.74

 therap

-0.74

IBLE

-0.71

 livest

-0.71

byss

-0.70

ioxide

-0.70

POSITIVE LOGITS

zeb

1.22

Kot

0.99

lin

0.90

zen

0.87

aku

0.86

arel

0.85

zon

0.84

zn

0.84

zin

0.84

oya

0.82

Activations Density 0.008%

proper nouns related to media and entertainment

mentions of the brand "Kotaku."

No Comments

No Known Activations