INDEX

Explanations

references to relationships and familial connections

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

quantifiers and descriptors

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ScopeManager

-0.55

complexContent

-0.54

RectangleBorder

-0.54

 pleaſure

-0.49

ləş

-0.46

partiet

-0.44

 للاسماء

-0.44

 Chriftian

-0.44

StatusBadRequest

-0.44

bitField

-0.43

POSITIVE LOGITS

 كومونز

0.40

ngdoc

0.36

 onData

0.36

AddTagHelper

0.35

atik

0.35

 câte

0.34

Cyfarwyddwr

0.34

 înt

0.34

Curios

0.34

RTDA

0.33

Activations Density 0.051%

references to relationships and familial connections

quantifiers and descriptors

No Comments

No Known Activations