INDEX

Explanations

references to a particular entity or term, specifically related to the term "Sham"

oai_token-act-pair · gpt-3.5-turbo

references to the word "Sham" and related terms

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

lihood

-0.67

 Prospect

-0.64

 newsletters

-0.64

 balloons

-0.63

 Jenner

-0.62

 annotation

-0.62

 strikeouts

-0.62

Eps

-0.62

 nutrit

-0.61

ĨĴ

-0.60

POSITIVE LOGITS

blers

1.18

eless

1.17

rock

1.14

aji

1.03

ash

0.97

abal

0.96

okin

0.93

arie

0.93

urai

0.92

ming

0.92

Activations Density 0.027%

references to a particular entity or term, specifically related to the term "Sham"

references to the word "Sham" and related terms

No Comments

No Known Activations