INDEX

Explanations

mentions of radio stations

oai_token-act-pair · gpt-3.5-turbo

mentions of "Radio" and associated program identifiers

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ortium

-0.74

holders

-0.70

auri

-0.65

erest

-0.64

 deed

-0.63

holder

-0.61

uates

-0.60

enes

-0.60

jar

-0.60

ricanes

-0.59

POSITIVE LOGITS

active

1.11

ode

0.91

 Shack

0.88

Radio

0.82

actively

0.80

oute

0.79

 Radio

0.75

 Networks

0.75

ologue

0.74

omatic

0.73

Activations Density 0.018%

mentions of radio stations

mentions of "Radio" and associated program identifiers

No Comments

No Known Activations