INDEX

Explanations

descriptions highlighting specific aspects of a story or situation

oai_token-act-pair · gpt-3.5-turbo

phrases that refer to different aspects or components in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

incinn

-0.78

ãĤī

-0.72

 millenn

-0.65

apons

-0.64

 Klux

-0.64

Draft

-0.64

 confir

-0.63

avorite

-0.63

ternity

-0.62

ministic

-0.61

POSITIVE LOGITS

ials

0.85

icular

0.84

icularly

0.81

icle

0.78

ridge

0.77

ridges

0.73

icles

0.73

 Myster

0.72

ICLE

0.70

 Payton

0.69

Activations Density 0.051%

descriptions highlighting specific aspects of a story or situation

phrases that refer to different aspects or components in various contexts

No Comments

No Known Activations