INDEX

Explanations

instances of the word "before."

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

referring to past information

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

 GenerationType

-0.54

AutoresizingMask

-0.45

̈́

-0.44

 cuán

-0.44

 kapturem

-0.44

 насељу

-0.44

➯

-0.43

GEBURTSDATUM

-0.43

EndProject

-0.42

Rhestr

-0.42

POSITIVE LOGITS

 again

0.53

 before

0.52

 BEFORE

0.51

 Before

0.50

BEFORE

0.49

 AGAIN

0.49

 Again

0.47

before

0.47

0.45

fore

0.44

Activations Density 0.011%

instances of the word "before."

referring to past information

No Comments

No Known Activations