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

 pige

-0.71

ONSORED

-0.69

raud

-0.62

 lapt

-0.62

 robbing

-0.61

©¶æ

-0.61

 Koen

-0.60

piring

-0.60

pret

-0.60

 Schn

-0.59

POSITIVE LOGITS

åº

0.75

ãĥĪ

0.71

catentry

0.71

ãĥīãĥ©

0.70

aires

0.68

åĮ

0.66

ãĥĦ

0.66

ista

0.65

Ĵ

0.64

ãĥķãĤ©

0.63

Activations Density 0.000%

No Known Activations

This feature has no known activations.