INDEX

Explanations

positive reactions or attention in reviews, feedback, or public reception

oai_token-act-pair · gpt-3.5-turbo

terms related to gaining attention or recognition

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

 Alone

-0.80

 Relax

-0.64

wine

-0.62

 Coliseum

-0.61

arta

-0.60

 Doing

-0.60

 wired

-0.59

ways

-0.59

-|

-0.58

 Cyborg

-0.57

POSITIVE LOGITS

 widespread

1.16

 notoriety

1.08

 considerable

1.05

 traction

1.01

 applause

1.01

 attention

1.01

 rave

0.98

 popularity

0.96

 scrutiny

0.94

 approval

0.93

Activations Density 0.219%

positive reactions or attention in reviews, feedback, or public reception

terms related to gaining attention or recognition

No Comments

No Known Activations