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

cffff

-0.71

obyl

-0.70

rolet

-0.65

 Reporting

-0.63

Ago

-0.63

 Bian

-0.61

 kits

-0.59

hen

-0.59

reth

-0.59

reporting

-0.57

POSITIVE LOGITS

MU

0.76

 bells

0.72

glers

0.69

0.65

ition

0.63

OPLE

0.63

sburgh

0.61

][/

0.61

oldown

0.60

KK

0.60

Activations Density 0.000%

No Known Activations

This feature has no known activations.