INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

kai

-0.71

lda

-0.69

McF

-0.69

fig

-0.66

ç¥ŀ

-0.65

pling

-0.64

ãĥ¼ãĥĨ

-0.64

hews

-0.63

nostic

-0.61

 connection

-0.60

POSITIVE LOGITS

 ours

0.72

 tiss

0.70

constitutional

0.68

 Obst

0.68

Front

0.68

ansom

0.66

Merit

0.66

 Pigs

0.65

etus

0.65

selves

0.64

Activations Density 0.000%

No Known Activations

This feature has no known activations.