INDEX

Explanations

references to character attributes and relational dynamics within narratives or discussions

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

Following "is," "are," or "were."

np_acts-logits-general · gemini-2.0-flash

instead or rather

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-16k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_16k/average_l0_91

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

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

SourceChecksum

-0.47

хьтан

-0.45

CreateMap

-0.42

toHaveBeenCalled

-0.42

OOTDTY

-0.39

dagog

-0.36

OGND

-0.35

المناصب

-0.35

UnknownFields

-0.35

 LEYENDO

-0.35

POSITIVE LOGITS

 vielmehr

0.65

むしろ

0.56

 instead

0.54

Instead

0.54

AnchorTagHelper

0.52

 Instead

0.52

あく

0.52

あくまで

0.48

而是

0.48

instead

0.47

Activations Density 0.442%

references to character attributes and relational dynamics within narratives or discussions

Following "is," "are," or "were."

instead or rather

No Comments

No Known Activations