INDEX

Explanations

names, titles, and specific terms related to news, sports, and businesses

oai_token-act-pair · gpt-3.5-turbo

prominent proper nouns and their affiliations or achievements

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

Abstract

-0.79

lished

-0.71

RESULTS

-0.70

 Canaver

-0.69

 Cummings

-0.69

etheless

-0.68

enegger

-0.68

OLOGY

-0.65

Downloadha

-0.65

æĺ¯

-0.64

POSITIVE LOGITS

ona

0.71

ussie

0.70

eg

0.69

omer

0.68

od

0.68

ens

0.67

intel

0.66

odo

0.65

ont

0.64

uber

0.64

Activations Density 0.465%

names, titles, and specific terms related to news, sports, and businesses

prominent proper nouns and their affiliations or achievements

No Comments

No Known Activations