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

ħĭ

-0.93

ļéĨĴ

-0.89

ĸļ

-0.80

 Ezek

-0.79

©¶æ¥µ

-0.77

 Revelations

-0.72

 Newsp

-0.71

 Article

-0.71

ICLE

-0.68

 nomine

-0.67

POSITIVE LOGITS

 contingent

0.70

among

0.70

"}],"

0.65

vous

0.65

 trailing

0.65

 coached

0.64

offs

0.63

cock

0.62

via

0.62

Doctors

0.61

Activations Density 0.000%

No Known Activations

This feature has no known activations.