INDEX

Explanations

discussion of mistakes or errors

oai_token-act-pair · gpt-3.5-turbo

references to mistakes and errors

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

population

-0.78

iture

-0.70

minent

-0.68

region

-0.67

uction

-0.67

amen

-0.67

orthy

-0.65

ighth

-0.65

otor

-0.64

metry

-0.64

POSITIVE LOGITS

 mistakes

1.27

 dece

0.88

 errors

0.87

é»Ĵ

0.83

 flaws

0.81

 behavi

0.81

 mistake

0.81

 Malf

0.75

uggest

0.75

 glitches

0.74

Activations Density 0.015%

discussion of mistakes or errors

references to mistakes and errors

No Comments

No Known Activations