INDEX

Explanations

references to dimensions or measurements related to data structures

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

start of user turn

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

Rüyada

-0.59

Portale

-0.54

Abitanti

-0.52

 cherchés

-0.51

Personensuche

-0.49

 espectáculo

-0.49

 meille

-0.48

wapV

-0.48

onAttach

-0.47

abstractmethod

-0.47

POSITIVE LOGITS

 diện

1.05

 stratos

0.57

avanaugh

0.55

 tropo

0.52

NotEmpty

0.44

SneakyThrows

0.44

reuters

0.40

glLoadIdentity

0.40

lusconi

0.39

ci

0.38

Activations Density 0.001%

references to dimensions or measurements related to data structures

start of user turn

No Comments

No Known Activations