INDEX

Explanations

occurrences of the letter "e" in varying contexts and forms

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

i.e

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

 juge

-0.43

 Helm

-0.40

regu

-0.39

 guider

-0.39

Voto

-0.39

 StringTokenizer

-0.38

 checksum

-0.38

BH

-0.37

了不少

-0.37

cuno

-0.37

POSITIVE LOGITS

GEBURTSDATUM

0.65

QMetaType

0.60

 cioè

0.57

Jeografia

0.56

 Namely

0.56

brigens

0.56

TextAppearance

0.54

makeText

0.54

 ossia

0.54

 iaitu

0.53

Activations Density 0.050%

occurrences of the letter "e" in varying contexts and forms

i.e

No Comments

No Known Activations