INDEX

Explanations

years or dates

oai_token-act-pair · gpt-3.5-turbo

end parentheses and similar punctuation marks

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ahu

-0.83

rew

-0.68

 guarded

-0.64

anch

-0.63

rette

-0.62

 contiguous

-0.62

enta

-0.62

odo

-0.61

otine

-0.61

 cares

-0.61

POSITIVE LOGITS

 onwards

0.92

BuyableInstoreAndOnline

0.75

Reviewer

0.69

 Surviv

0.69

Interstitial

0.68

Jah

0.67

0.66

 Defin

0.66

Eag

0.65

 Ibid

0.65

Activations Density 0.107%

years or dates

end parentheses and similar punctuation marks

No Comments

No Known Activations