INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

 Finder

-0.72

Arm

-0.70

 Ambro

-0.68

elect

-0.67

Peb

-0.65

 Pent

-0.64

 Atom

-0.64

 Extensions

-0.64

 shooters

-0.63

 Alive

-0.61

POSITIVE LOGITS

olean

0.72

ossip

0.71

ibu

0.71

rived

0.69

usting

0.68

BIP

0.68

izzy

0.66

len

0.65

 reused

0.65

readable

0.64

Activations Density 0.000%

No Known Activations

This feature has no known activations.