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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Jihad

-0.76

 Bans

-0.67

âĸ¬âĸ¬

-0.63

Rah

-0.62

awa

-0.60

 invested

-0.60

 Bris

-0.60

 Haley

-0.60

Found

-0.59

%:

-0.58

POSITIVE LOGITS

ername

0.77

onut

0.74

agnetic

0.72

glers

0.71

renheit

0.70

drm

0.69

erk

0.69

rists

0.69

aido

0.68

lies

0.67

Activations Density 0.000%

No Known Activations

This feature has no known activations.