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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

elfth

-0.80

sole

-0.80

sea

-0.77

zzo

-0.75

zinski

-0.71

osuke

-0.69

RH

-0.68

sis

-0.68

icum

-0.67

rina

-0.67

POSITIVE LOGITS

 Standards

0.76

id

0.67

dag

0.67

 mats

0.65

 unfocusedRange

0.61

 Dangerous

0.60

 standards

0.60

 this

0.59

 stature

0.58

DC

0.57

Activations Density 0.000%

No Known Activations

This feature has no known activations.