INDEX

Explanations

phrases indicating a point in time or a significant event

oai_token-act-pair · gpt-3.5-turbo Triggered by @bot

present tense verbs that indicate current actions or states

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 12-res-jb

Configuration

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

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_post

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

other

-0.56

 administrations

-0.54

iances

-0.54

 inquiries

-0.54

 suspensions

-0.53

 overlap

-0.53

 uptake

-0.52

arlane

-0.52

phas

-0.52

 weaknesses

-0.51

POSITIVE LOGITS

1.03

!!

0.97

!!!

0.95

!!!!

0.92

!!!!!

0.88

 tonight

0.86

 ðŁĳ

0.86

!:

0.86

 courtesy

0.85

!"

0.83

Activations Density 0.800%

phrases indicating a point in time or a significant event

present tense verbs that indicate current actions or states

No Comments

No Known Activations