INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 McCann

-0.74

 Thatcher

-0.74

 IPCC

-0.72

 WARN

-0.70

NSA

-0.69

EPA

-0.66

 Titanic

-0.66

 Fukushima

-0.65

NSA

-0.64

DPR

-0.63

POSITIVE LOGITS

à

0.77

ONSORED

0.72

merce

0.68

ibl

0.68

ibble

0.65

ERY

0.64

atown

0.64

ible

0.64

Friend

0.64

bending

0.64

Activations Density 0.000%

No Known Activations

This feature has no known activations.