INDEX

Explanations

proper nouns related to geographic locations, particularly regions or districts

oai_token-act-pair · gpt-3.5-turbo

references to geographic locations or regions

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

jas

-0.79

rays

-0.78

INESS

-0.73

ports

-0.65

sim

-0.64

alysed

-0.63

amd

-0.61

TeX

-0.60

 diff

-0.60

iku

-0.60

POSITIVE LOGITS

 Lower

3.40

 Upper

3.22

Lower

1.97

upper

1.59

 Outer

1.56

 Inner

1.38

 Higher

1.26

 Greater

1.25

 upper

1.18

 Idle

1.13

Activations Density 0.033%

proper nouns related to geographic locations, particularly regions or districts

references to geographic locations or regions

No Comments

No Known Activations