INDEX

Explanations

the word "electric" in various contexts

oai_token-act-pair · gpt-3.5-turbo

references to electric energy or electric vehicles

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

allery

-0.80

xual

-0.76

00200000

-0.72

 Haku

-0.69

xus

-0.69

 Lumpur

-0.66

 CLSID

-0.64

reads

-0.64

isSpecialOrderable

-0.64

ewski

-0.62

POSITIVE LOGITS

 propulsion

0.91

 Powered

0.84

 motors

0.82

 grids

0.80

 currents

0.80

 locom

0.76

 fences

0.76

 powered

0.74

 heater

0.73

FIG

0.72

Activations Density 0.008%

the word "electric" in various contexts

references to electric energy or electric vehicles

No Comments

No Known Activations