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

MAT

-0.71

uart

-0.68

selves

-0.66

PG

-0.65

ãĤ¡

-0.65

peak

-0.63

terms

-0.63

rez

-0.62

ãĤ©

-0.61

fixed

-0.60

POSITIVE LOGITS

resents

0.73

iste

0.73

iously

0.70

Twe

0.65

iment

0.64

iments

0.62

iety

0.61

Twe

0.61

ilege

0.60

pires

0.58

Activations Density 0.000%

No Known Activations

This feature has no known activations.