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

ACTED

-0.72

alion

-0.68

eworks

-0.68

mma

-0.66

orius

-0.66

yang

-0.65

yon

-0.64

oche

-0.64

igraph

-0.64

POSITIVE LOGITS

tek

0.70

 exha

0.68

 newsp

0.68

ãĥĢ

0.65

osponsors

0.63

sbm

0.63

Rank

0.63

Hack

0.62

Tier

0.62

 cumbers

0.62

Activations Density 0.000%

No Known Activations

This feature has no known activations.