INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

aughtered

-0.85

icka

-0.84

ierre

-0.82

 Tycoon

-0.77

aria

-0.76

vana

-0.72

iago

-0.72

ocrates

-0.72

iov

-0.72

inki

-0.71

POSITIVE LOGITS

 Magnetic

0.76

 Investigations

0.74

 findings

0.68

 flashlight

0.66

 prosecut

0.61

 detachment

0.60

 rigging

0.60

 magnetic

0.60

lights

0.59

 compass

0.59

Activations Density 0.000%

No Known Activations

This feature has no known activations.