INDEX

Explanations

names or words related to "Sandra" and "Alexander"

oai_token-act-pair · gpt-3.5-turbo

mentions of specific names and locations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.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

 Koen

-0.73

 licences

-0.71

 Mutual

-0.67

 Idlib

-0.66

lease

-0.63

 Navigation

-0.63

 Compass

-0.62

 Fighters

-0.62

dayName

-0.60

 airstrikes

-0.60

POSITIVE LOGITS

andra

3.70

andre

2.64

andro

2.18

andr

2.08

gow

1.92

 Nadu

1.10

oat

1.07

ople

1.06

abeth

1.05

vind

1.03

Activations Density 0.043%

names or words related to "Sandra" and "Alexander"

mentions of specific names and locations

No Comments

No Known Activations