INDEX

Explanations

elements related to programming syntax and structure

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

indices and programming constructs

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

 EconPapers

-0.45

umenter

-0.42

InitVars

-0.40

LookAnd

-0.39

PropertyChanging

-0.39

 sostenible

-0.37

 diseños

-0.36

 play

-0.35

flam

-0.35

########.

-0.34

POSITIVE LOGITS

Hentet

0.51

WriteTagHelper

0.46

 surla

0.45

المناصب

0.44

 AssemblyTitle

0.43

0.42

 муніципалі

0.41

 aapt

0.40

Espèce

0.37

طان

0.37

Activations Density 0.013%

elements related to programming syntax and structure

indices and programming constructs

No Comments

No Known Activations