INDEX

Explanations

phrases indicating the start of a new concept or idea, often used as segues or transitions

oai_token-act-pair · gpt-3.5-turbo

conversational discourse markers and interjections

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 respons

-0.76

––

-0.74

 accordingly

-0.71

[*

-0.70

pc

-0.67

-"

-0.66

(*

-0.65

*/

-0.64

``(

-0.63

 consequently

-0.63

POSITIVE LOGITS

resa

1.07

anmar

0.90

reetings

0.90

osate

0.89

 Shot

0.83

aan

0.82

 Things

0.82

 Breat

0.81

 Known

0.80

 Tomorrow

0.80

Activations Density 0.298%

phrases indicating the start of a new concept or idea, often used as segues or transitions

conversational discourse markers and interjections

No Comments

No Known Activations