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

oa

-0.72

opolis

-0.66

opian

-0.66

ibaba

-0.65

ogle

-0.64

hub

-0.64

 agrees

-0.62

bot

-0.60

uria

-0.59

adesh

-0.59

POSITIVE LOGITS

ãĤ¨ãĥ«

1.03

ãĥĩãĤ£

0.92

è£

0.81

ircraft

0.80

ãĥ¯

0.78

GBT

0.77

ãĥ©ãĥ³

0.77

ãĥ¼ãĥĨãĤ£

0.77

ãģĨ

0.73

ENN

0.73

Activations Density 0.000%

No Known Activations

This feature has no known activations.