INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.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

DOC

-0.70

Ca

-0.70

CLASSIFIED

-0.69

Blake

-0.67

PRES

-0.67

 inval

-0.66

ARB

-0.65

İĭ

-0.65

command

-0.65

classic

-0.64

POSITIVE LOGITS

istration

0.77

rish

0.67

iameter

0.63

 Notting

0.61

ves

0.61

 squats

0.59

 Trin

0.58

oples

0.58

ials

0.58

HG

0.57

Activations Density 0.000%

No Known Activations

This feature has no known activations.