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

ASE

-0.84

MpServer

-0.84

atech

-0.77

Anderson

-0.77

Cache

-0.75

FB

-0.73

nai

-0.72

aug

-0.71

èª

-0.69

 Schwar

-0.69

POSITIVE LOGITS

eless

0.80

 surn

0.71

 Citiz

0.68

ModLoader

0.67

 commod

0.67

urated

0.67

 eleph

0.66

 nodd

0.65

 RAID

0.65

 destro

0.64

Activations Density 0.000%

No Known Activations

This feature has no known activations.