INDEX

Explanations

occurrences of names or identifiers related to individuals, particularly in legal or official contexts

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

sorting, saving, billing, user

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

New Auto-Interp

Top Features by Cosine Similarity

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

Configuration

google/gemma-scope-9b-it-res/layer_9/width_131k/average_l0_121

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.9.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

 exportaciones

-0.43

 invitamos

-0.42

pecabe

-0.42

 tevreden

-0.41

awtextra

-0.41

//

-0.40

berdayakan

-0.40

baguna

-0.40

katapos

-0.39

 micrófono

-0.39

POSITIVE LOGITS

Ent

0.59

 GetEnumerator

0.54

 Kent

0.52

AutoField

0.52

getStringExtra

0.52

localctx

0.51

Referanser

0.50

="@+

0.48

!(:

0.48

ent

0.47

Activations Density 0.001%

occurrences of names or identifiers related to individuals, particularly in legal or official contexts

sorting, saving, billing, user

No Comments

No Known Activations