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

hend

-0.64

riages

-0.64

È

-0.63

uders

-0.62

WARNING

-0.61

 Mats

-0.61

resources

-0.61

 Mous

-0.61

inois

-0.61

 hous

-0.60

POSITIVE LOGITS

icion

0.66

reditary

0.64

mopolitan

0.63

rocal

0.62

ellation

0.62

selage

0.62

ional

0.61

romeda

0.60

 unbeliev

0.60

Frag

0.60

Activations Density 0.000%

No Known Activations

This feature has no known activations.