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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ivery

-0.74

Slot

-0.70

Sullivan

-0.66

Shin

-0.65

BP

-0.62

 Logged

-0.61

Donnell

-0.60

heon

-0.60

SI

-0.59

 Lodge

-0.59

POSITIVE LOGITS

£ı

0.90

 Cosponsors

0.82

 unicorn

0.80

netflix

0.72

 Gutenberg

0.70

daq

0.69

 ignor

0.67

 concess

0.64

 Participant

0.62

icist

0.61

Activations Density 0.000%

No Known Activations

This feature has no known activations.