INDEX

Explanations

references to timing and specific sequences of events

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

after certain time periods

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

 sesuatu

-0.42

 tuot

-0.39

denkmal

-0.37

 queima

-0.35

mability

-0.34

ături

-0.34

apparence

-0.33

ändigen

-0.33

背影

-0.33

 aceler

-0.32

POSITIVE LOGITS

TypedDataSet

0.69

новништво

0.64

NameInMap

0.55

 esternos

0.55

Bakgrunnsstoff

0.54

 autorytatywna

0.52

تقاوى

0.52



0.50

Personendaten

0.50

 lenker

0.50

Activations Density 0.081%

references to timing and specific sequences of events

after certain time periods

No Comments

No Known Activations