INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

byss

-0.81

leans

-0.80

alysed

-0.75

AME

-0.72

alde

-0.71

Cast

-0.70

raged

-0.70

poke

-0.70

ceive

-0.69

GBT

-0.68

POSITIVE LOGITS

 favors

0.66

 puff

0.65

rou

0.64

 simplest

0.63

Yu

0.62

 Skywalker

0.62

 defects

0.61

 company

0.60

 according

0.60

 companies

0.58

Activations Density 0.000%

No Known Activations

This feature has no known activations.