INDEX

Explanations

terms indicating causation or the origin of effects

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

caused by

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

ChildScrollView

-0.47

 interlocutor

-0.45

 IBOutlet

-0.44

 houſe

-0.44

endregion

-0.44

oneta

-0.43

 pleaſure

-0.43

Portail

-0.43

────────

-0.42

uito

-0.42

POSITIVE LOGITS

 Caused

1.73

caused

1.72

 caused

1.68

 causado

1.09

 spowod

0.98

 provocado

0.86

causing

0.84

Induced

0.84

 Induced

0.82

 occasioned

0.81

Activations Density 0.012%

terms indicating causation or the origin of effects

caused by

No Comments

No Known Activations