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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

opter

-0.71

burgh

-0.64

cht

-0.62

ppe

-0.61

upid

-0.61

FACE

-0.60

 Chambers

-0.59

 Tide

-0.58

idered

-0.58

 Task

-0.58

POSITIVE LOGITS

Hardware

0.76

Via

0.74

 destro

0.70

Lens

0.69

 "$:/

0.68

=-=-

0.67

 Extras

0.67

rador

0.66

addin

0.65

ollah

0.64

Activations Density 0.000%

No Known Activations

This feature has no known activations.