INDEX

Explanations

references to popular culture, specifically music, movies, and television shows

oai_token-act-pair · gpt-3.5-turbo

references to popular culture

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

RAW

-0.81

IVES

-0.78

IGHTS

-0.73

TPPStreamerBot

-0.73

APH

-0.70

beit

-0.67

 ACTIONS

-0.67

BILITIES

-0.67

 Cruel

-0.67

OME

-0.67

POSITIVE LOGITS

ulating

1.18

ulates

1.17

corn

1.10

ulations

1.08

ularity

1.04

ulated

1.04

ular

0.99

ulate

0.97

lar

0.91

 quiz

0.88

Activations Density 0.012%

references to popular culture, specifically music, movies, and television shows

references to popular culture

No Comments

No Known Activations