INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.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

 flights

-0.67

berus

-0.66

KE

-0.64

gotten

-0.63

INT

-0.63

GET

-0.62

onite

-0.62

Launch

-0.62

thy

-0.61

ISA

-0.61

POSITIVE LOGITS

 commod

0.71

ess

0.69

 whence

0.69

esses

0.65

 Videos

0.62

 Crim

0.59

 implicitly

0.59

Ide

0.58

 Bever

0.57

agan

0.57

Activations Density 0.000%

No Known Activations

This feature has no known activations.