INDEX

Explanations

the abbreviation "TV" across different contexts

oai_token-act-pair · gpt-3.5-turbo

mentions of television or TV-related content

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.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

forth

-0.86

ebook

-0.83

tto

-0.70

naires

-0.69

cles

-0.68

 WhatsApp

-0.66

heet

-0.65

hens

-0.65

ific

-0.64

ussen

-0.64

POSITIVE LOGITS

trop

1.17

 Shows

1.06

 Trop

0.99

 commercials

0.98

Guide

0.95

shows

0.93

plays

0.91

 sitcom

0.88

Series

0.86

 shows

0.85

Activations Density 0.037%

the abbreviation "TV" across different contexts

mentions of television or TV-related content

No Comments

No Known Activations