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

acci

-0.92

cing

-0.91

EEE

-0.87

arnaev

-0.84

 Entered

-0.82

Ago

-0.73

OR

-0.72

rase

-0.71

 Logged

-0.71

ÂłÂł

-0.69

POSITIVE LOGITS

 disproportion

0.78

 vend

0.76

 coffers

0.75

 crates

0.73

 lobbyists

0.73

 purch

0.73

 lobbyist

0.72

 purchaser

0.71

 promot

0.71

 warehouses

0.70

Activations Density 0.000%

No Known Activations

This feature has no known activations.