INDEX

Explanations

phrases related to procedures, regulations, and official designations

oai_token-act-pair · gpt-3.5-turbo

prepositions and various auxiliary words indicating relationships or conditions

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

hess

-0.70

reach

-0.67

HL

-0.60

rogens

-0.58

versions

-0.58

 occup

-0.56

Manchester

-0.54

ton

-0.54

 toler

-0.53

rock

-0.53

POSITIVE LOGITS

etheless

0.98

Ö¼

0.84

ï¸ı

0.83

udder

0.71

apest

0.69

IVERS

0.69

ylon

0.68

terday

0.67

ulhu

0.67

azo

0.65

Activations Density 0.300%

phrases related to procedures, regulations, and official designations

prepositions and various auxiliary words indicating relationships or conditions

No Comments

No Known Activations