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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

idential

-0.75

uity

-0.73

 Observer

-0.72

owe

-0.71

ossession

-0.71

 posed

-0.65

 flats

-0.65

 cock

-0.63

 creep

-0.63

 sche

-0.62

POSITIVE LOGITS

ãĤ¢ãĥ«

0.84

ãĤĮ

0.83

ãĤ¶

0.80

GGGGGGGG

0.79

èĢħ

0.78

èª

0.77

é¾įå

0.75

¿½

0.75

 ãĤµãĥ¼ãĥĨãĤ£ãĥ¯ãĥ³

0.71

NVIDIA

0.70

Activations Density 0.000%

No Known Activations

This feature has no known activations.