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

 pursued

-0.68

 timet

-0.67

azo

-0.65

 deport

-0.64

 rival

-0.64

 decre

-0.63

 News

-0.62

 prevail

-0.61

 [âĢ¦]

-0.60

][

-0.60

POSITIVE LOGITS

nin

0.81

tsky

0.80

ventory

0.75

tools

0.75

strument

0.72

forth

0.71

nyder

0.70

orah

0.70

quickShipAvailable

0.69

Sharp

0.68

Activations Density 0.000%

No Known Activations

This feature has no known activations.