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

complex

-0.75

 medi

-0.71

 Despair

-0.67

SHIP

-0.66

Monitor

-0.66

 Swordsman

-0.66

hearted

-0.66

bow

-0.65

sent

-0.64

 Krug

-0.63

POSITIVE LOGITS

 rall

0.85

uana

0.82

Ħ¢

0.80

RTX

0.77

obyl

0.76

 locality

0.74

yles

0.73

reetings

0.72

namese

0.68

assic

0.67

Activations Density 0.000%

No Known Activations

This feature has no known activations.