INDEX

Explanations

mentions of specific entities such as names or locations

oai_token-act-pair · gpt-3.5-turbo

acronyms or abbreviations in the text

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

oxide

-0.68

 Legions

-0.64

vae

-0.61

 Chimera

-0.60

 Jolly

-0.59

Sic

-0.58

©¶æ

-0.58

 Balk

-0.57

midt

-0.56

 Avenger

-0.56

POSITIVE LOGITS

FUL

1.00

INGS

0.97

TAIN

0.95

HAEL

0.94

 IMAGES

0.93

ICLE

0.93

AGE

0.90

RIC

0.86

hyde

0.85

IER

0.82

Activations Density 0.118%

mentions of specific entities such as names or locations

acronyms or abbreviations in the text

No Comments

No Known Activations