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

forces

-0.71

ials

-0.62

 sounds

-0.62

 Pesh

-0.62

ahar

-0.61

>[

-0.61

wright

-0.61

offs

-0.61

 wipes

-0.60

spe

-0.60

POSITIVE LOGITS

ħĭ

0.85

çīĪ

0.82

aukee

0.82

iona

0.72

IZE

0.71

cffff

0.70

 conduc

0.69

¥ŀ

0.69

fram

0.69

TAG

0.68

Activations Density 0.000%

No Known Activations

This feature has no known activations.