INDEX

Explanations

names of places and organizations

oai_token-act-pair · gpt-3.5-turbo

proper nouns and brand names

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

 Monk

-0.72

bat

-0.68

txt

-0.64

phies

-0.63

oling

-0.63

essing

-0.63

piring

-0.63

aji

-0.63

wcs

-0.62

 Kendrick

-0.62

POSITIVE LOGITS

 affiliates

0.92

 subsidiary

0.89

 headquarters

0.89

 shareholders

0.87

 subsidiaries

0.86

 staff

0.86

CEO

0.85

 Communications

0.84

Corp

0.83

 Corporation

0.83

Activations Density 0.242%

names of places and organizations

proper nouns and brand names

No Comments

No Known Activations