INDEX

Explanations

words starting with the letter "m" with a high activation value

oai_token-act-pair · gpt-3.5-turbo

occurrences of the letter "m"

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

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

 pitch

-0.82

 calcul

-0.73

 solicitation

-0.70

 tips

-0.68

 speculation

-0.66

 predictive

-0.66

 foul

-0.66

 Vera

-0.65

 briefs

-0.64

tip

-0.64

POSITIVE LOGITS

ixed

1.47

ovies

1.47

arijuana

1.41

aintain

1.36

ixture

1.35

ichael

1.34

otor

1.32

arching

1.32

igration

1.32

agn

1.30

Activations Density 0.038%

words starting with the letter "m" with a high activation value

occurrences of the letter "m"

No Comments

No Known Activations