INDEX

Explanations

instances where things are compared or equated

oai_token-act-pair · gpt-3.5-turbo

instances of the word "same."

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

*=-

-0.75

ases

-0.71

omics

-0.67

ohyd

-0.67

 Provided

-0.66

ommod

-0.66

acus

-0.63

bane

-0.63

HI

-0.62

anism

-0.62

POSITIVE LOGITS

 thing

0.96

 exact

0.84

 amount

0.78

ol

0.74

rity

0.72

iating

0.71

 principle

0.70

 vein

0.70

 kind

0.67

iator

0.67

Activations Density 0.041%

instances where things are compared or equated

instances of the word "same."

No Comments

No Known Activations