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

achev

-0.78

rop

-0.69

Grade

-0.69

chem

-0.66

gars

-0.64

dogs

-0.62

OTO

-0.61

osis

-0.61

rolog

-0.61

icles

-0.60

POSITIVE LOGITS

ichick

0.65

 span

0.64

 Wond

0.61

 refinery

0.60

 Wander

0.59

staff

0.58

bull

0.58

widget

0.57

 Bull

0.57

ilial

0.57

Activations Density 0.000%

No Known Activations

This feature has no known activations.