INDEX

Explanations

proper names, specifically the name "Joel," with varying intensities

oai_token-act-pair · gpt-3.5-turbo

references to the name "Joel" and its various contexts in the document

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

FACE

-0.79

dayName

-0.76

utils

-0.75

BOOK

-0.70

hib

-0.70

role

-0.68

checks

-0.66

omething

-0.65

card

-0.61

 parity

-0.61

POSITIVE LOGITS

aquin

0.95

 Joel

0.95

McH

0.89

 Wein

0.87

 Goodman

0.83

lein

0.78

 Valley

0.77

 Dixon

0.74

 Stein

0.74

 Rubin

0.74

Activations Density 0.009%

proper names, specifically the name "Joel," with varying intensities

references to the name "Joel" and its various contexts in the document

No Comments

No Known Activations