INDEX

Explanations

occurrences of code formatting or syntax elements

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

mathematical or code references

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 31-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_31/width_131k/average_l0_109

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.31.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>

-0.78

styleType

-0.75

yarnpkg

-0.55

verifyException

-0.53

 msglen

-0.49

requireNonNull

-0.49

 cults

-0.46

InjectAttribute

-0.46

ValueStyle

-0.46

 Wikimedijinoj

-0.45

POSITIVE LOGITS



0.56

trás

0.48

ñora

0.47

ñores

0.45

delijk

0.45

sweise

0.42

toprule

0.41

 nôtre

0.41

bjerg

0.41

'}';

0.40

Activations Density 0.051%

occurrences of code formatting or syntax elements

mathematical or code references

No Comments

No Known Activations