INDEX

Explanations

names of places or people

oai_token-act-pair · gpt-3.5-turbo

words related to motels and virtual experiences

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

tones

-0.89

å£«

-0.80

GROUND

-0.78

Ô

-0.75

cknowled

-0.72

cknow

-0.71

PDATE

-0.71

ãĥ¤

-0.70

©¶æ

-0.70

pta

-0.70

POSITIVE LOGITS

itious

0.90

chell

0.87

uality

0.86

unia

0.82

ouring

0.81

olit

0.80

icket

0.79

glers

0.79

ugu

0.78

tle

0.77

Activations Density 0.028%

names of places or people

words related to motels and virtual experiences

No Comments

No Known Activations