INDEX

Explanations

variable identifiers and their assignments in a programming context

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

identifiers and numbers

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

tagHelperRunner

-0.87

AndEndTag

-0.85

ValueStyle

-0.79

ChildScrollView

-0.69

 gynhyrchwyd

-0.69

awtextra

-0.66

WireFormatLite

-0.65

GraphicsUnit

-0.65

ftagPool

-0.65

ModelAdmin

-0.64

POSITIVE LOGITS

 identifier

0.38

 identifiers

0.37

identifier

0.34

 name

0.34

identifiers

0.33

parseColor

0.32

 description

0.32

 wall

0.31

 sign

0.31

 code

0.31

Activations Density 0.045%

variable identifiers and their assignments in a programming context

identifiers and numbers

No Comments

No Known Activations