INDEX

Explanations

references to the act of reading or analyzing texts

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

this analysis indirectly will go

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

addGap

-0.42

addPreferredGap

-0.41

 viewWillAppear

-0.40

BorderFactory

-0.40

EDEFAULT

-0.40

ldc

-0.38

 AssemblyProduct

-0.38

 aload

-0.37

MBR

-0.36

Offic

-0.35

POSITIVE LOGITS

 GenerationType

0.64

hyrchwyd

0.54

новништво

0.51

ğraf

0.49

 EconPapers

0.48

 OMITBAD

0.47

msgTypes

0.44

lovakia

0.44

Handlung

0.44

SharedDtor

0.44

Activations Density 0.109%

references to the act of reading or analyzing texts

this analysis indirectly will go

No Comments

No Known Activations