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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

stocks

-0.74

pins

-0.73

ãĢĳ

-0.66

eteenth

-0.66

ctors

-0.65

keyes

-0.64

 sliding

-0.64

vp

-0.62

itsu

-0.60

erman

-0.58

POSITIVE LOGITS

 captivity

0.71

nesty

0.69

lav

0.68

aban

0.67

Zub

0.67

eln

0.67

ratulations

0.66

HH

0.65

ategory

0.63

 cran

0.63

Activations Density 0.000%

No Known Activations

This feature has no known activations.