INDEX

Explanations

proper nouns or phrases that an individual is most recognized for

oai_token-act-pair · gpt-3.5-turbo

phrases indicating recognition or notoriety of individuals or entities

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 6-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.6.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.6.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

xit

-0.77

plet

-0.77

forestation

-0.75

otion

-0.69

escal

-0.68

amination

-0.68

otor

-0.67

mental

-0.66

ilation

-0.65

odder

-0.65

POSITIVE LOGITS

Offline

0.83

Ô

0.67

 newsp

0.67

Ø³

0.66

 landmarks

0.64

 Chap

0.63

 thous

0.61

owan

0.60

Û

0.60

 Theft

0.59

Activations Density 0.037%

proper nouns or phrases that an individual is most recognized for

phrases indicating recognition or notoriety of individuals or entities

No Comments

No Known Activations