INDEX

Explanations

phrases related to news and media publications

oai_token-act-pair · gpt-3.5-turbo

empty or nonexistent content markers

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 6-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.6.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.6.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

ardless

-0.53

emale

-0.47

tumblr

-0.45

 steroids

-0.44

 selves

-0.44

 blah

-0.43

 anthem

-0.42

 overwhel

-0.42

 equivalents

-0.42

 Kardash

-0.42

POSITIVE LOGITS

 Investigative

0.56

itbart

0.48

 Judicial

0.47

 Architects

0.47

heastern

0.47

Bloomberg

0.46

 Securities

0.46

 Tablet

0.46

 RAND

0.46

Media

0.45

Activations Density 1.113%

phrases related to news and media publications

empty or nonexistent content markers

No Comments

No Known Activations