INDEX

Explanations

phrases related to extensions or continuations of something

oai_token-act-pair · gpt-3.5-turbo

terms related to adaptations, extensions, or continuations of existing concepts or works

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.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

rooms

-0.67

asions

-0.67

glers

-0.62

iencies

-0.60

EMS

-0.60

Sources

-0.60

 VIDEOS

-0.58

 transitions

-0.58

 arenas

-0.57

 microphones

-0.57

POSITIVE LOGITS

 thereof

1.27

of

1.01

 reminiscent

0.79

 homage

0.79

 akin

0.71

 version

0.71

hesis

0.70

 descendant

0.70

 imitation

0.70

dayName

0.69

Activations Density 0.167%

phrases related to extensions or continuations of something

terms related to adaptations, extensions, or continuations of existing concepts or works

No Comments

No Known Activations