INDEX

Explanations

references to instructions or guidance

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

directions and instructions

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

 AssemblyVersion

-0.59

oliber

-0.54

 Gestão

-0.42

новниш

-0.41

 BANK

-0.41

 MUNICIP

-0.40

 bank

-0.40

 BANKS

-0.40

 hanem

-0.39

 Roskov

-0.39

POSITIVE LOGITS

 directions

2.17

Directions

2.17

 Directions

2.14

directions

1.97

 DIRECTIONS

1.80

DIRECTIONS

1.70

 instructions

0.88

 direcciones

0.86

 للمعارف

0.84

 Instructions

0.76

Activations Density 0.003%

references to instructions or guidance

directions and instructions

No Comments

No Known Activations