INDEX

Explanations

terms related to dice

oai_token-act-pair · gpt-3.5-turbo

references to dice

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

REM

-0.81

SUP

-0.78

organ

-0.73

aired

-0.72

country

-0.69

owner

-0.67

sup

-0.67

ded

-0.64

gd

-0.64

obe

-0.64

POSITIVE LOGITS

 dice

1.32

 Dice

1.11

eria

0.88

 resil

0.81

acebook

0.76

ername

0.72

 Enix

0.72

 Dise

0.72

 Cutter

0.71

 poker

0.71

Activations Density 0.008%

terms related to dice

references to dice

No Comments

No Known Activations