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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

-0.86

nces

-0.82

redd

-0.76

govtrack

-0.74

Tur

-0.73

Fer

-0.71

FO

-0.71

]}

-0.69

Balt

-0.69

hots

-0.69

POSITIVE LOGITS

elin

0.69

ont

0.67

 priesthood

0.66

 Cind

0.64

 paran

0.63

 totem

0.60

 consum

0.60

 rehabilit

0.60

andan

0.59

 worldview

0.59

Activations Density 0.000%

No Known Activations

This feature has no known activations.