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

dit

-0.84

allowed

-0.73

ahead

-0.67

toe

-0.67

%%

-0.66

 gained

-0.65

:[

-0.65

theless

-0.63

allow

-0.63

forward

-0.63

POSITIVE LOGITS

dayName

0.80

eret

0.78

esters

0.78

nikov

0.70

RIC

0.66

 CLSID

0.66

 Desk

0.66

 Relief

0.64

»Ĵ

0.63

 Meeting

0.62

Activations Density 0.000%

No Known Activations

This feature has no known activations.