INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 dstg

-0.72

Org

-0.71

 VIDEOS

-0.69

 sidx

-0.68

²¾

-0.63

org

-0.61

 Saras

-0.59

OC

-0.58

è¦ļéĨĴ

-0.58

ãĥĻ

-0.58

POSITIVE LOGITS

atown

0.81

quel

0.71

orsi

0.71

schild

0.70

ndum

0.69

ggle

0.67

 Cortana

0.66

ference

0.65

xit

0.64

pei

0.63

Activations Density 0.000%

No Known Activations

This feature has no known activations.