INDEX

Explanations

proper names and titles, particularly with 'von'

oai_token-act-pair · gpt-3.5-turbo

instances of the name "von" 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 @ 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

gallery

-0.84

taboola

-0.78

rentice

-0.70

orative

-0.69

inates

-0.68

uyomi

-0.68

ĳå£«

-0.67

à¤¾

-0.67

à¥

-0.67

inals

-0.66

POSITIVE LOGITS

 Braun

0.85

 Frey

0.85

amins

0.73

env

0.72

hof

0.72

 Doom

0.69

der

0.69

 Karma

0.68

wald

0.68

 Schwarz

0.68

Activations Density 0.019%

proper names and titles, particularly with 'von'

instances of the name "von" in various contexts

No Comments

No Known Activations