INDEX

Explanations

specific terminologies and structures related to programming, particularly concerning matching algorithms and expressions

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

exact match or wildcard

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

PreExecute

-0.47

 fédé

-0.46

AddTagHelper

-0.44

AnchorTagHelper

-0.44

paravant

-0.44

sizeCache

-0.42

 Sünden

-0.41

 Griechen

-0.39

 anún

-0.38

 OMITBAD

-0.38

POSITIVE LOGITS

webElementGuid

0.69

exact

0.47

 conto

0.47

Match

0.45

transQ

0.44

 inflexible

0.44

 бо

0.43

 Cont

0.43

oa̍t

0.43

Cont

0.42

Activations Density 0.115%

specific terminologies and structures related to programming, particularly concerning matching algorithms and expressions

exact match or wildcard

No Comments

No Known Activations