INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

 Keystone

-0.70

ordable

-0.66

 Caucus

-0.65

ali

-0.62

Saf

-0.61

 advoc

-0.60

berto

-0.60

anto

-0.59

Govern

-0.59

iott

-0.59

POSITIVE LOGITS

akings

0.74

croft

0.69

orius

0.65

 unlocked

0.64

que

0.64

 badge

0.63

 twitch

0.63

ronics

0.62

eanor

0.62

 lodged

0.61

Activations Density 0.000%

No Known Activations

This feature has no known activations.