INDEX

Explanations

dates and locations in news article excerpts

oai_token-act-pair · gpt-3.5-turbo

dates and locations relevant to news events

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

Tact

-0.67

-0.60

geist

-0.60

 ourselves

-0.59

rite

-0.58

mental

-0.56

Reloaded

-0.56

 whiff

-0.56

 yourselves

-0.53

 unanswered

-0.53

POSITIVE LOGITS

 REUTERS

0.90

allery

0.82

FILE

0.77

 pose

0.74

 Courtesy

0.74

 poses

0.73

 illustration

0.71

 Â©

0.70

UTERS

0.70

./

0.70

Activations Density 0.055%

dates and locations in news article excerpts

dates and locations relevant to news events

No Comments

No Known Activations