INDEX

Explanations

passive voice constructions

oai_token-act-pair · gpt-3.5-turbo

expressions of disbelief or negation

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

ĪĴ

-0.75

 Revenge

-0.67

 Cance

-0.67

stre

-0.63

 Showdown

-0.62

 Cancel

-0.61

 Reloaded

-0.61

 Reborn

-0.60

 Fury

-0.60

 Wrong

-0.60

POSITIVE LOGITS

 bother

1.36

 realize

1.33

 realise

1.26

 bothered

1.10

 mention

1.03

 acknowledge

1.02

 admit

1.01

 noticed

1.00

 notice

0.99

 comprehend

0.98

Activations Density 0.290%

passive voice constructions

expressions of disbelief or negation

No Comments

No Known Activations