INDEX

Explanations

phrases related to significant events or actions

oai_token-act-pair · gpt-3.5-turbo

references to collective experiences or evaluations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

peria

-0.66

ĪĴ

-0.66

opes

-0.65

 timing

-0.65

scill

-0.64

idth

-0.64

label

-0.63

onso

-0.62

duction

-0.60

 timeframe

-0.60

POSITIVE LOGITS

payer

0.73

ronics

0.67

mania

0.66

POW

0.66

cks

0.65

roid

0.62

 Nobel

0.60

osaurus

0.60

 Bain

0.60

 Mead

0.60

Activations Density 0.000%

phrases related to significant events or actions

references to collective experiences or evaluations

No Comments

No Known Activations