INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

ãĥ¼ãĥ³

-0.79

ãĥ³ãĤ¸

-0.77

Ability

-0.72

Asset

-0.70

weights

-0.69

WF

-0.67

Phys

-0.65

QL

-0.65

fortune

-0.65

fold

-0.63

POSITIVE LOGITS

iliation

0.79

itaire

0.78

udeb

0.71

iotic

0.71

iliated

0.70

ades

0.68

ition

0.67

iotics

0.67

 ..............

0.67

ooters

0.67

Activations Density 0.000%

No Known Activations

This feature has no known activations.