INDEX

Explanations

identifiers and references related to different data types in a structured format

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

Identifier fields

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

="@+

-0.51

tagHelperRunner

-0.39

EndContext

-0.38

BeginInit

-0.36

nonatomic

-0.35

 réussi

-0.35

setVerticalGroup

-0.31

devamını

-0.30

 successful

-0.29

{§

-0.29

POSITIVE LOGITS

IndentedString

0.60

enfone

0.57

 виправивши

0.53

tonsoft

0.51

CppMethod

0.50

denes

0.49

 Khalil

0.49

fastjson

0.47

 estekak

0.47

 attested

0.47

Activations Density 0.051%

identifiers and references related to different data types in a structured format

Identifier fields

No Comments

No Known Activations

identifiers and references related to different data types in a structured format

Identifier fields

No Comments

No Known Activations