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

yss

-0.66

ide

-0.65

eri

-0.63

irlf

-0.61

 Trou

-0.61

 repro

-0.60

flag

-0.59

aris

-0.59

timer

-0.58

etimes

-0.57

POSITIVE LOGITS

bilt

0.77

çīĪ

0.74

ANE

0.69

Iv

0.67

ãĤ´ãĥ³

0.67

GV

0.66

ashington

0.65

LEASE

0.64

æ©

0.63

æī

0.62

Activations Density 0.000%

No Known Activations

This feature has no known activations.