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

onga

-0.81

ends

-0.74

ighters

-0.73

iq

-0.72

endar

-0.71

imore

-0.71

otine

-0.71

ights

-0.71

quickShipAvailable

-0.69

ighter

-0.68

POSITIVE LOGITS

 Emin

0.72

rahim

0.68

ãģķ

0.67

ä½ľ

0.63

 Canaan

0.63

QUEST

0.62

 persisted

0.60

ãĥ¼ãĤ¯

0.60

 localized

0.59

ãĥĩ

0.59

Activations Density 0.000%

No Known Activations

This feature has no known activations.