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

figured

-0.70

plet

-0.70

wolves

-0.69

ricanes

-0.68

try

-0.67

ctors

-0.65

mostly

-0.64

 charge

-0.62

 Presidential

-0.59

du

-0.59

POSITIVE LOGITS

uve

0.79

Ĵ

0.78

anmar

0.76

åº

0.73

ashtra

0.69

EEK

0.68

icz

0.66

icipated

0.65

ologists

0.64

HCR

0.64

Activations Density 0.000%

No Known Activations

This feature has no known activations.