INDEX

Explanations

references to items or concepts categorized as 'junk'

oai_token-act-pair · gpt-3.5-turbo

references to "junk" in various contexts, particularly related to food, finance, and undesirable items

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

cing

-0.78

Æ

-0.77

oral

-0.74

gypt

-0.70

mberg

-0.69

ignty

-0.68

ndum

-0.66

 Lauder

-0.65

soType

-0.65

zza

-0.65

POSITIVE LOGITS

 heap

0.97

 junk

0.94

etsu

0.84

 drawer

0.84

busters

0.83

ies

0.82

ernaut

0.79

 pile

0.78

buster

0.77

ie

0.76

Activations Density 0.014%

references to items or concepts categorized as 'junk'

references to "junk" in various contexts, particularly related to food, finance, and undesirable items

No Comments

No Known Activations