INDEX

Explanations

the word "tap" at varying levels of relevance

oai_token-act-pair · gpt-3.5-turbo

occurrences of the word "tap."

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

Reloaded

-0.78

ften

-0.74

 AMERICA

-0.72

============

-0.65

 Advent

-0.65

anchester

-0.64

IZE

-0.63

byss

-0.63

cens

-0.63

CVE

-0.62

POSITIVE LOGITS

tap

0.92

tap

0.87

Tap

0.83

deck

0.83

acity

0.82

eless

0.81

 tapping

0.80

rop

0.79

pour

0.78

ering

0.76

Activations Density 0.023%

the word "tap" at varying levels of relevance

occurrences of the word "tap."

No Comments

No Known Activations