INDEX

Explanations

phrases indicating causal relationships

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

Followed by "to" or "of" indicating cause/reason

np_acts-logits-general · gemini-2.0-flash

explaining reasons or cause

np_acts-logits-general · gemini-2.5-flash-lite

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 30-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_30/width_16k/average_l0_120

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.30.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

 oneself

-0.55

HomeAsUpEnabled

-0.54

parsedMessage

-0.47

OrNull

-0.44

 jsPsych

-0.41

AsUp

-0.40

contentLoaded

-0.40

stateProvider

-0.39

enterOuterAlt

-0.39

 sánchez

-0.39

POSITIVE LOGITS

 adanya

0.75

 lack

0.75

 reasons

0.70

 dėl

0.63

 때문

0.62

Lack

0.59

lack

0.58

 alasan

0.57

 Reasons

0.56

 karena

0.56

Activations Density 0.360%

phrases indicating causal relationships

Followed by "to" or "of" indicating cause/reason

explaining reasons or cause

No Comments

No Known Activations