INDEX

Explanations

syntactic structures related to method calls and task execution in code

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

method calls or property access

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.20.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

SequentialGroup

-0.69

awtextra

-0.56

########.

-0.55

帖最后由

-0.54

transQ

-0.53

enterOuterAlt

-0.53

AndEndTag

-0.50

UrlResolution

-0.50

setVerticalGroup

-0.49

BagLayout

-0.49

POSITIVE LOGITS

 Poss

0.50

naeus

0.50

 Lebens

0.49

 auth

0.43

caped

0.42

 possessed

0.42

 possess

0.42

 posses

0.42

blica

0.42

 represented

0.41

Activations Density 0.023%

syntactic structures related to method calls and task execution in code

method calls or property access

No Comments

No Known Activations