INDEX

Explanations

key characters and relationships in narratives

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

character introduction before background

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.20.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

 szóci

-0.38

RenderAtEndOf

-0.37

awtextra

-0.35

writerow

-0.35

illotine

-0.34

ativement

-0.34

 labi

-0.34

principalColumn

-0.33

 honored

-0.33

 recent

-0.33

POSITIVE LOGITS

endpush

0.54

ArrowToggle

0.51

CppMethod

0.49

󠁿

0.49

endregion

0.48

()}}

0.48

0.47

Біографія

0.46

);*/

0.46

 asistente

0.45

Activations Density 0.043%

key characters and relationships in narratives

character introduction before background

No Comments

No Known Activations