INDEX

Explanations

words related to complaints or criticisms

oai_token-act-pair · gpt-3.5-turbo

words and phrases related to meanings and interpretations

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

Miller

-0.73

imoto

-0.72

INC

-0.70

 Jarvis

-0.65

atro

-0.64

 Liver

-0.64

ENTION

-0.63

Spons

-0.63

paio

-0.62

INT

-0.62

POSITIVE LOGITS

etheless

0.96

xual

0.94

egal

0.85

volent

0.83

emonic

0.79

atural

0.75

onymous

0.74

ploy

0.74

uchin

0.73

fter

0.72

Activations Density 0.057%

words related to complaints or criticisms

words and phrases related to meanings and interpretations

No Comments

No Known Activations