INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

éĹĺ

-0.85

isd

-0.83

assad

-0.76

ESA

-0.75

enos

-0.75

ãĤ¤

-0.71

ahon

-0.71

ouch

-0.71

ohn

-0.71

Synopsis

-0.71

POSITIVE LOGITS

artifacts

0.67

 newsletters

0.65

 Occupations

0.65

 trailing

0.64

Zip

0.62

 learners

0.61

 bribes

0.61

spa

0.61

 redes

0.60

 stereotypes

0.60

Activations Density 0.000%

No Known Activations

This feature has no known activations.