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

OTT

-0.69

piring

-0.69

PP

-0.67

itting

-0.66

cers

-0.66

 Recommended

-0.65

anding

-0.65

QUIRE

-0.62

chwitz

-0.61

 giveaways

-0.61

POSITIVE LOGITS

fruit

0.71

hematic

0.70

 Dynam

0.70

 istg

0.67

 Alam

0.65

cam

0.65

 Rubin

0.65

 Elon

0.65

iron

0.65

 Dartmouth

0.65

Activations Density 0.000%

No Known Activations

This feature has no known activations.