INDEX

Explanations

references to people being in or out of town

oai_token-act-pair · gpt-3.5-turbo

mentions of being "out of town."

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

cess

-0.78

yssey

-0.78

 Accessed

-0.77

oscope

-0.71

eful

-0.70

zb

-0.70

lett

-0.70

ostic

-0.69

zes

-0.68

iom

-0.65

POSITIVE LOGITS

 speakers

0.73

 hall

0.69

 defences

0.69

Introdu

0.66

 arrivals

0.65

é¾įå¥ĳå£«

0.64

 conclud

0.63

 coffers

0.61

âĶĢâĶĢâĶĢâĶĢ

0.61

 conditioning

0.61

Activations Density 0.058%

references to people being in or out of town

mentions of being "out of town."

No Comments

No Known Activations