INDEX

Explanations

words related to mysteries and the concept of mystery itself

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

mystery novels

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

########.

-0.54

Clause

-0.51

campal

-0.49

 BorderSide

-0.48

Leeds

-0.47

Deadpool

-0.46

ftime

-0.46

 Lordships

-0.46

 getCode

-0.45

ęcz

-0.45

POSITIVE LOGITS

 mystery

1.20

 Mystery

1.07

mystery

1.05

Mystery

0.97

 mysteries

0.94

sterious

0.88

 mystère

0.85

 myster

0.82

 Myster

0.82

 misterio

0.79

Activations Density 0.004%

words related to mysteries and the concept of mystery itself

mystery novels

No Comments

No Known Activations