INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

dayName

-0.72

£ı

-0.67

orget

-0.62

vals

-0.59

 postp

-0.59

osterone

-0.58

Met

-0.57

rd

-0.57

midt

-0.57

ESL

-0.57

POSITIVE LOGITS

Reloaded

0.79

 Inquiry

0.73

iners

0.73

ĳå£«

0.71

ONS

0.69

mor

0.68

adh

0.67

pring

0.66

Blade

0.66

Board

0.66

Activations Density 0.000%

No Known Activations

This feature has no known activations.