INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.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

ANA

-0.70

ãĥ´ãĤ¡

-0.69

Õ

-0.69

Iss

-0.67

reek

-0.66

cus

-0.64

raw

-0.64

 Leban

-0.63

Ø§Ø

-0.63

DERR

-0.63

POSITIVE LOGITS

eer

0.71

reditary

0.65

iasm

0.63

eering

0.61

 faculties

0.61

 suscept

0.61

ulic

0.60

 antibodies

0.60

 vaccinated

0.59

 sands

0.58

Activations Density 0.000%

No Known Activations

This feature has no known activations.