INDEX

Explanations

mentions of unusual, strange, or unconventional things

oai_token-act-pair · gpt-3.5-turbo

references to the concept of "weirdness."

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

ptive

-0.96

ptives

-0.91

vation

-0.87

cussion

-0.81

aders

-0.78

ignty

-0.78

HCR

-0.77

Interstitial

-0.77

FORE

-0.77

ULTS

-0.76

POSITIVE LOGITS

ness

1.26

nesses

1.12

ly

1.08

est

1.00

ety

0.97

oes

0.91

entimes

0.90

ening

0.81

ies

0.79

os

0.78

Activations Density 0.021%

mentions of unusual, strange, or unconventional things

references to the concept of "weirdness."

No Comments

No Known Activations