INDEX

Explanations

references to project evaluation and company reputation

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

user turn

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

 hObject

-0.47

 Roskov

-0.45

BASEPATH

-0.39

PYX

-0.37

ruptedException

-0.37

клопе

-0.37

 ✭✭

-0.36

 autorytatywna

-0.35

dz

-0.34

ngdoc

-0.34

POSITIVE LOGITS

 الحره

0.48

InjectMocks

0.48

 plainte

0.44

 Autorizaciones

0.42

 davran

0.42

 écout

0.42

 TestBed

0.41

INESS

0.41

anyakan

0.40



0.39

Activations Density 0.010%

references to project evaluation and company reputation

user turn

No Comments

No Known Activations