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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

Vers

-0.73

 Whilst

-0.67

usters

-0.66

 Deborah

-0.65

 Fulton

-0.65

Fra

-0.64

 Schr

-0.64

 Wars

-0.63

 Fres

-0.63

 Firstly

-0.62

POSITIVE LOGITS

 taboo

0.75

ocobo

0.73

 referen

0.72

amiliar

0.70

ahime

0.69

nown

0.68

 abiding

0.67

 fantas

0.66

 anecd

0.64

ormal

0.64

Activations Density 0.000%

No Known Activations

This feature has no known activations.