INDEX

Explanations

proper nouns such as names of institutions and people

oai_token-act-pair · gpt-3.5-turbo

specific roles and title attributes related to various professions and measurements

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

.)

-0.72

.):

-0.71

Parameters

-0.62

.)

-0.59

iatus

-0.57

Mos

-0.56

 Ange

-0.55

DEN

-0.55

WAR

-0.54

pard

-0.54

POSITIVE LOGITS

etc

0.73

 captcha

0.69

ilion

0.67

utils

0.65

enture

0.59

 cheat

0.57

ModLoader

0.57

vertisement

0.56

Tumblr

0.55

 inclusion

0.55

Activations Density 0.928%

proper nouns such as names of institutions and people

specific roles and title attributes related to various professions and measurements

No Comments

No Known Activations