INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

-0.75

 Hosp

-0.71

ãĥĳ

-0.69

="#

-0.66

ËĪ

-0.63

process

-0.60

Args

-0.60

++;

-0.60

ãĤ¸

-0.59

Ã©r

-0.59

POSITIVE LOGITS

luster

0.74

soDeliveryDate

0.71

 resemble

0.68

iamond

0.68

 suspic

0.68

 repaid

0.68

 resembled

0.67

amental

0.64

tesy

0.64

 lured

0.63

Activations Density 0.000%

No Known Activations

This feature has no known activations.