INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.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

ĪĴ

-0.93

omin

-0.73

omon

-0.67

gars

-0.60

 rode

-0.60

progress

-0.59

 wast

-0.59

 marrow

-0.59

omi

-0.58

 vain

-0.56

POSITIVE LOGITS

0.76

inian

0.69

sburg

0.65

Py

0.64

)=(

0.63

cific

0.61

Reloaded

0.60

 rehearsal

0.60

CLASSIFIED

0.60

ushima

0.59

Activations Density 0.000%

No Known Activations

This feature has no known activations.