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

uthor

-0.69

zo

-0.69

tre

-0.68

achy

-0.68

 vacc

-0.66

aph

-0.66

 tumor

-0.66

 halluc

-0.65

dro

-0.64

 reusable

-0.64

POSITIVE LOGITS

rating

0.76

draw

0.74

ĪĴ

0.74

olor

0.72

ANCE

0.68

 Cosponsors

0.68

leaders

0.67

 Sending

0.66

 Defenders

0.65

 Discussion

0.64

Activations Density 0.000%

No Known Activations

This feature has no known activations.