INDEX

Explanations

conditional phrases indicating potential actions or outcomes

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

conditional would

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

<bos>

-0.45

Eure

-0.41

 vicenda

-0.40

Zah

-0.40

(=)

-0.38

Zah

-0.38

referer

-0.38

SpringBootTest

-0.38

cookieParser

-0.38

catchError

-0.36

POSITIVE LOGITS

 would

1.45

would

1.36

 Would

1.29

Would

1.29

 WOULD

1.22

 will

1.01

 würde

0.91

ould

0.86

 could

0.85

 avrebbe

0.85

Activations Density 0.062%

conditional phrases indicating potential actions or outcomes

conditional would

No Comments

No Known Activations