INDEX

Explanations

repeated mentions of specific words or phrases in various contexts, particularly in academic or formal writing

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

academic and technical descriptions

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

Билгалдахарш

-0.65

GEBURTSDATUM

-0.64

awakeFromNib

-0.57

Datuak

-0.57

Cyfeiriadau

-0.56

 nakalista

-0.55

Искәрмәләр

-0.54

extAlignment

-0.54

 للمعارف

-0.54

protoimpl

-0.54

POSITIVE LOGITS

 also

0.47

 Pacheco

0.39

 letz

0.38

 Gruber

0.38

 Nachbar

0.37

 Wolken

0.37

queles

0.37

we

0.37

now

0.36

crapers

0.36

Activations Density 0.051%

repeated mentions of specific words or phrases in various contexts, particularly in academic or formal writing

academic and technical descriptions

No Comments

No Known Activations