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

REAM

-0.83

 ILCS

-0.81

REDACTED

-0.81

resso

-0.76

Alexander

-0.75

MSN

-0.71

gomery

-0.70

ANI

-0.69

OTO

-0.68

ADRA

-0.67

POSITIVE LOGITS

 Monk

0.77

 prey

0.73

 Kant

0.73

ãĥ³ãĤ¸

0.68

gets

0.67

framework

0.66

necess

0.65

fram

0.64

middle

0.64

imb

0.61

Activations Density 0.000%

No Known Activations

This feature has no known activations.