INDEX

Explanations

variable and function identifiers related to coding and data structures

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

variable assignments

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

 يتيمه

-0.74

 المعيارى

-0.70

/*++

-0.65

AndEndTag

-0.64

MigrationBuilder

-0.61

CopyWith

-0.60

RTSN

-0.60

ChildScrollView

-0.57

évaluateur

-0.57

MergeFrom

-0.56

POSITIVE LOGITS

 beret

0.33

is

0.33

 consists

0.32

was

0.31

berton

0.31

AMA

0.30

 verla

0.30

 SEMI

0.29

Ответ

0.29

сії

0.29

Activations Density 0.009%

variable and function identifiers related to coding and data structures

variable assignments

No Comments

No Known Activations