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

agos

-0.72

ethyst

-0.66

angible

-0.63

atives

-0.61

iewicz

-0.60

 intangible

-0.58

 Olson

-0.58

 Shining

-0.58

 Quartz

-0.57

 shortages

-0.56

POSITIVE LOGITS

psc

0.73

ilib

0.68

yg

0.68

MpServer

0.67

aily

0.65

Foot

0.64

âĢ¦]

0.64

apest

0.63

EStreamFrame

0.63

 STORY

0.62

Activations Density 0.000%

No Known Activations

This feature has no known activations.