INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.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

quest

-0.73

chen

-0.69

housing

-0.68

house

-0.68

houses

-0.67

hov

-0.66

sym

-0.65

emo

-0.64

iston

-0.63

lected

-0.63

POSITIVE LOGITS

pora

0.77

 forgiven

0.68

henko

0.67

 deterrence

0.61

 spend

0.61

opoulos

0.59

urt

0.59

ONEY

0.58

jad

0.58

 disperse

0.58

Activations Density 0.000%

No Known Activations

This feature has no known activations.