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

idem

-0.74

 Constantin

-0.68

 climbers

-0.65

ciating

-0.64

 swing

-0.62

 Celt

-0.62

sed

-0.62

aver

-0.62

 skelet

-0.62

 ignition

-0.61

POSITIVE LOGITS

 Orders

0.73

Stream

0.72

Rum

0.71

 Papers

0.69

 Messages

0.64

 Stats

0.64

Facebook

0.64

 Documents

0.63

 Turing

0.63

 Posts

0.63

Activations Density 0.000%

No Known Activations

This feature has no known activations.