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

©¶æ

-1.06

£ı

-0.89

appropri

-0.83

skirts

-0.82

 tradem

-0.79

ĳå£«

-0.78

onse

-0.77

utics

-0.74

icone

-0.74

ĸļ

-0.74

POSITIVE LOGITS

 Recommend

0.82

gow

0.67

 rack

0.66

 Trend

0.65

STL

0.65

LCS

0.64

 Dragonbound

0.64

Fed

0.63

Dj

0.63

 breeze

0.63

Activations Density 0.000%

No Known Activations

This feature has no known activations.