INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 tremend

-0.87

ĸļ

-0.79

 yogurt

-0.75

alid

-0.67

arlane

-0.66

icultural

-0.65

 corrid

-0.63

olate

-0.63

inement

-0.63

oxide

-0.62

POSITIVE LOGITS

 Measures

0.81

 Logged

0.73

 Provided

0.73

ÂłÂłÂłÂłÂłÂłÂłÂłÂłÂłÂłÂłÂłÂłÂłÂł

0.73

fighters

0.69

/+

0.69

 Scores

0.68

 Starts

0.68

AMA

0.66

/-

0.65

Activations Density 0.000%

No Known Activations

This feature has no known activations.