INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.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

cape

-0.67

Knight

-0.67

mingham

-0.66

 WATCHED

-0.66

 Meal

-0.65

Chart

-0.63

rise

-0.63

Tem

-0.62

iami

-0.62

 Overt

-0.60

POSITIVE LOGITS

rons

0.73

unda

0.69

 retarded

0.67

Vu

0.66

 greenhouse

0.66

 buggy

0.65

Phi

0.64

oppy

0.63

 reins

0.61

ELD

0.61

Activations Density 0.000%

No Known Activations

This feature has no known activations.