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

rophe

-0.75

 Mile

-0.68

eless

-0.66

Ļ½

-0.65

ife

-0.64

 Aren

-0.63

cyclopedia

-0.62

Gaw

-0.61

 Bers

-0.60

 Plot

-0.59

POSITIVE LOGITS

 SERVICE

0.79

MIC

0.77

 PRODUCT

0.67

 subpoena

0.67

=-=-=-=-=-=-=-=-

0.65

 reclaim

0.63

MIC

0.63

alian

0.62

 cens

0.61

 renewal

0.60

Activations Density 0.000%

No Known Activations

This feature has no known activations.