INDEX

Explanations

phrases or terms that denote previous names or identities of entities

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

formerly known as

np_acts-logits-general · gemini-2.5-flash-lite

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 31-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_31/width_131k/average_l0_109

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.31.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

 AssemblyCulture

-0.71

 ExecuteAsync

-0.68

UnusedPrivate

-0.65

 poffible

-0.65

 насељу

-0.61

AsUp

-0.60

 initComponents

-0.59

 actuelles

-0.59

UnsafeEnabled

-0.59

存于互联网档案馆

-0.59

POSITIVE LOGITS

 known

0.59

of

0.45

known

0.42

</i>

0.38

 called

0.37

 KNOWN

0.37

 يعرف

0.36

 bekend

0.35

 part

0.33

 bernama

0.33

Activations Density 0.006%

phrases or terms that denote previous names or identities of entities

formerly known as

No Comments

No Known Activations