INDEX

Explanations

mentions of the name "Taylor Swift"

oai_token-act-pair · gpt-3.5-turbo

mentions of the name "Taylor."

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

PDATE

-0.89

rontal

-0.89

undai

-0.86

oard

-0.85

hemy

-0.84

cffffcc

-0.80

ntil

-0.77

cffff

-0.77

vous

-0.75

rely

-0.73

POSITIVE LOGITS

Made

1.12

 Swift

1.02

obi

0.80

 Taylor

0.78

ville

0.72

ivation

0.71

hurst

0.70

mann

0.70

ual

0.70

ying

0.69

Activations Density 0.020%

mentions of the name "Taylor Swift"

mentions of the name "Taylor."

No Comments

No Known Activations