INDEX

Explanations

messages or communication-related words

oai_token-act-pair · gpt-3.5-turbo

references to messages being sent or received

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

culus

-0.74

riages

-0.74

 Simulator

-0.64

 stagn

-0.62

 stagnant

-0.61

 Brune

-0.61

 peas

-0.60

 liberties

-0.58

ackets

-0.58

 Railway

-0.58

POSITIVE LOGITS

mails

0.76

 WhatsApp

0.72

 greets

0.72

 greeted

0.71

 voic

0.70

 greeting

0.69

 memos

0.69

plain

0.69

mail

0.67

 messages

0.67

Activations Density 0.501%

messages or communication-related words

references to messages being sent or received

No Comments

No Known Activations