INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.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

%%%%

-0.71

 defective

-0.68

EStreamFrame

-0.67

allo

-0.67

©¶æ

-0.65

 Titan

-0.64

 Investig

-0.64

 Hilbert

-0.64

 Rampage

-0.63

 Demon

-0.63

POSITIVE LOGITS

é£

0.74

channelAvailability

0.72

chant

0.70

ABLE

0.69

AMES

0.67

Standard

0.67

UTC

0.66

 cert

0.65

KNOWN

0.65

å®

0.64

Activations Density 0.000%

No Known Activations

This feature has no known activations.