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

 Logged

-0.75

 Judges

-0.73

 Views

-0.65

 diam

-0.63

entin

-0.62

DERR

-0.62

 âĢİ

-0.62

Ans

-0.61

pmwiki

-0.59

 Tend

-0.59

POSITIVE LOGITS

Ò

0.79

OX

0.71

atern

0.69

ktop

0.67

"],

0.65

asio

0.64

ospels

0.63

ãĥ¼ãĥĨ

0.62

GV

0.62

Ger

0.61

Activations Density 0.000%

No Known Activations

This feature has no known activations.