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

chuk

-0.74

)|

-0.73

iqueness

-0.72

egu

-0.72

kson

-0.71

)."

-0.70

nant

-0.67

.,"

-0.66

eli

-0.66

ynski

-0.66

POSITIVE LOGITS

 feats

0.72

 timelines

0.72

 libraries

0.72

 hasht

0.69

 folders

0.68

æĺ

0.67

 Journals

0.64

 licenses

0.64

 breaks

0.64

 riches

0.64

Activations Density 0.000%

No Known Activations

This feature has no known activations.