INDEX

Explanations

references to media, publications, and editorial roles

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

New York Times

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

 ब्रेकडाउन

-0.74

MigrationBuilder

-0.63

сылкі

-0.56

脚注の使い方

-0.52

WithIOException

-0.51

Erreferentziak

-0.49

 produisons

-0.49

 betweenstory

-0.46

Билгалдахарш

-0.44

testify

-0.44

POSITIVE LOGITS

ckså

0.42

⟬

0.40

)++;

0.40

 samym

0.40

NUMX

0.38

الحياه

0.37

󠁥

0.37

الدراسه

0.37

 ontem

0.36

 cartões

0.36

Activations Density 0.097%

references to media, publications, and editorial roles

New York Times

No Comments

No Known Activations