INDEX

Explanations

names or titles of people and entities, often associated with legal contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 12-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_12/width_16k/average_l0_82

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.12.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

󠁴

-0.89

RectangleBorder

-0.87

 snippetHide

-0.84

Personendaten

-0.83

########.

-0.81

 nakalista

-0.79

GEBURTSDATUM

-0.77

".

-0.77

MENAFN

-0.77

SharedCtor

-0.76

POSITIVE LOGITS

et

0.60

0.59

0.58

0.53

vs

0.47

<eos>

0.46

0.45

0.44

ah

0.43

Activations Density 0.980%

names or titles of people and entities, often associated with legal contexts

No Comments

No Known Activations