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

ulas

-0.75

MRI

-0.72

OPLE

-0.71

resy

-0.71

rez

-0.70

alys

-0.70

nikov

-0.69

ihad

-0.68

alos

-0.68

rew

-0.67

POSITIVE LOGITS

strate

0.69

quished

0.69

middle

0.68

 Sail

0.68

 Saiyan

0.67

 Warm

0.65

 Hera

0.64

itely

0.64

 Messenger

0.63

 washed

0.62

Activations Density 0.000%

No Known Activations

This feature has no known activations.