INDEX

Explanations

phrases related to actions or events happening after a specific point in time

oai_token-act-pair · gpt-3.5-turbo

references to times or events occurring after something else

oai_token-act-pair · gpt-4o-mini

Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 6-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.6.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.6.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Negative Logits

foot

-0.61

enberg

-0.61

rod

-0.59

ids

-0.58

 outsider

-0.57

 privilege

-0.57

 Syndicate

-0.57

 Brett

-0.56

 predicate

-0.55

footed

-0.54

POSITIVE LOGITS

vernment

0.77

å§«

0.71

 thereafter

0.70

 proceeded

0.70

eatures

0.70

gradation

0.69

 transpired

0.69

 afterwards

0.68

 conclud

0.68

 apologize

0.67

Activations Density 0.025%

phrases related to actions or events happening after a specific point in time

references to times or events occurring after something else

No Known Activations