INDEX

Explanations

prominent figures, such as physicists, psychologists, artists, and political commentators

oai_token-act-pair · gpt-3.5-turbo

terms related to professional roles or occupations

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

aber

-0.89

crew

-0.85

merga

-0.81

ramid

-0.81

word

-0.78

acus

-0.77

cale

-0.74

perm

-0.74

ivery

-0.74

assian

-0.72

POSITIVE LOGITS

 extraord

0.99

 Stephen

0.94

 Laura

0.90

 Shaun

0.89

 Andrew

0.89

 Tony

0.88

 Richard

0.88

 Carl

0.88

 David

0.87

 Todd

0.86

Activations Density 0.301%

prominent figures, such as physicists, psychologists, artists, and political commentators

terms related to professional roles or occupations

No Comments

No Known Activations