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

 antid

-0.70

fug

-0.70

+++

-0.64

 Bundy

-0.63

 bargaining

-0.62

etheless

-0.61

 absentee

-0.60

 unions

-0.60

VERTISEMENT

-0.59

 discretion

-0.59

POSITIVE LOGITS

eki

0.79

osit

0.71

ico

0.70

cycl

0.69

psc

0.68

aye

0.67

onen

0.67

bsite

0.67

osite

0.66

zie

0.66

Activations Density 0.000%

No Known Activations

This feature has no known activations.