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

CHR

-0.76

NAT

-0.71

 Macedonia

-0.68

 dialog

-0.65

idine

-0.63

 NATO

-0.61

KP

-0.61

scl

-0.61

kWh

-0.61

 banter

-0.61

POSITIVE LOGITS

anting

0.73

glomer

0.71

irming

0.70

itsu

0.68

testing

0.68

ieu

0.68

aeda

0.68

quartered

0.68

fitting

0.67

emale

0.66

Activations Density 0.000%

No Known Activations

This feature has no known activations.