INDEX

Explanations

technical or formal language related to seriousness or complexity

oai_token-act-pair · gpt-3.5-turbo

adjectives describing varying degrees of seriousness or complexity

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

SEA

-0.76

FORE

-0.76

Jump

-0.71

isha

-0.71

Cola

-0.69

OVER

-0.69

DNA

-0.67

Unique

-0.65

CLE

-0.64

Ev

-0.62

POSITIVE LOGITS

 than

1.22

 versions

1.16

 forms

1.10

 ones

1.02

 aspects

0.98

than

0.96

 pursuits

0.95

 sounding

0.94

 varieties

0.92

 sibling

0.91

Activations Density 0.173%

technical or formal language related to seriousness or complexity

adjectives describing varying degrees of seriousness or complexity

No Comments

No Known Activations