INDEX

Explanations

qualifiers that express uncertainty or conditionality

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 4-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_4/width_16k/average_l0_124

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.4.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ioutil

-0.58

Przypisy

-0.58

 Negara

-0.55

RenderAtEndOf

-0.52

GraphicsUnit

-0.52

VIAF

-0.52

finally

-0.50

 enfin

-0.50

øg

-0.50

 vectorielle

-0.49

POSITIVE LOGITS

 necessarily

1.33

necessarily

1.18

 necesariamente

0.93

 necessariamente

0.89

 nécessairement

0.88

 forcément

0.85

 Necess

0.82

不一定

0.81

RectangleBorder

0.77

未必

0.74

Activations Density 0.269%

qualifiers that express uncertainty or conditionality

No Comments

No Known Activations