INDEX

Explanations

proper nouns related to specific entities or concepts

oai_token-act-pair · gpt-3.5-turbo

references to specific places, people, or concepts related to identity and culture

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

.ï¿½

-0.70

.''

-0.67

___

-0.62

@#&

-0.62

quickShipAvailable

-0.62

''.

-0.60

`.

-0.59

.''.

-0.59

.<

-0.57

._

-0.56

POSITIVE LOGITS

 aboard

0.65

 navigating

0.65

 leveraging

0.65

 touting

0.64

 tasked

0.63

 gazing

0.62

 atop

0.60

 cruising

0.60

 addressing

0.59

 rescuing

0.58

Activations Density 0.950%

proper nouns related to specific entities or concepts

references to specific places, people, or concepts related to identity and culture

No Comments

No Known Activations