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

ãĥĨãĤ£

-0.75

css

-0.74

ylon

-0.73

ille

-0.71

izzy

-0.69

heit

-0.68

Lev

-0.66

codes

-0.66

code

-0.66

 Liberty

-0.66

POSITIVE LOGITS

pen

0.71

 predec

0.69

 forehead

0.68

Pf

0.62

 invasion

0.60

 roundup

0.60

pora

0.58

 payday

0.58

 suspic

0.58

abl

0.57

Activations Density 0.000%

No Known Activations

This feature has no known activations.