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

 Takeru

-0.71

OND

-0.66

linger

-0.65

 BEST

-0.64

ugu

-0.61

 recognise

-0.60

ategor

-0.59

vre

-0.59

asonic

-0.58

 politely

-0.58

POSITIVE LOGITS

¥µ

0.89

fitting

0.77

growth

0.72

cca

0.70

heny

0.69

cember

0.64

access

0.64

 Annotations

0.64

usa

0.62

 Prayer

0.61

Activations Density 0.000%

No Known Activations

This feature has no known activations.