INDEX

Explanations

imperative statements instructing to take action or do something

oai_token-act-pair · gpt-3.5-turbo

calls to action or ways to assist

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

TRI

-0.78

 triangle

-0.60

 behavi

-0.59

 subst

-0.58

 FORM

-0.57

 prem

-0.56

 retard

-0.56

 bureau

-0.55

ATT

-0.55

CFL

-0.54

POSITIVE LOGITS

you

1.39

you

1.37

You

1.17

You

1.12

YOU

1.12

YOU

0.99

 yourself

0.92

yout

0.91

 your

0.84

ya

0.81

Activations Density 0.157%

imperative statements instructing to take action or do something

calls to action or ways to assist

No Comments

No Known Activations