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

 firewall

-0.77

 embargo

-0.76

 showc

-0.72

 weekends

-0.72

 holidays

-0.70

 extensive

-0.67

 defences

-0.65

 griev

-0.65

lander

-0.65

 overload

-0.65

POSITIVE LOGITS

 Biol

0.72

 Nanto

0.70

creation

0.69

 Stain

0.68

 IMAGES

0.67

reviewed

0.63

eers

0.63

 Levi

0.62

ngth

0.61

nick

0.60

Activations Density 0.000%

No Known Activations

This feature has no known activations.