INDEX

Explanations

comma-separated phrases indicating conditions or explanations

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

comma followed by a word

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

GND

-0.39

accor

-0.38

mb

-0.38

 concord

-0.37

 supp

-0.37

Ef

-0.36

tte

-0.35

 contest

-0.34

GND

-0.34

vig

-0.34

POSITIVE LOGITS

 nonUne

0.54

empêcher

0.54

 inoxydable

0.53

adaptiveStyles

0.52

 türlü

0.49

PreInfinity

0.48

 poffible

0.48

 sement

0.47

HomeAsUpEnabled

0.47

 muſt

0.46

Activations Density 0.023%

comma-separated phrases indicating conditions or explanations

comma followed by a word

No Comments

No Known Activations