INDEX

Explanations

occurrences of a specific character or term, mainly related to "Kerberos"

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

keratin, kernel, Kerberos

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

 pleaſure

-0.54

 Manusia

-0.53

 Hidup

-0.53

Uso

-0.51

 Cinta

-0.47

Println

-0.46

 Uang

-0.46

Aladdin

-0.46

 Plato

-0.45

 vise

-0.45

POSITIVE LOGITS

Ker

2.28

Ker

2.11

ker

1.91

KER

1.76

 kernels

1.23

ker

1.19

 Кер

1.18

 keratin

1.17

 kernel

1.16

 Kers

1.13

Activations Density 0.009%

occurrences of a specific character or term, mainly related to "Kerberos"

keratin, kernel, Kerberos

No Comments

No Known Activations