INDEX

Explanations

references to new developments or discoveries

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

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

contentValues

-0.48

subplots

-0.46

 utafitiHapana

-0.45

 intptr

-0.43

 initComponents

-0.39

CPF

-0.39

 informée

-0.39

exitRule

-0.38

 Speed

-0.38

TSR

-0.37

POSITIVE LOGITS

新

0.50

AutoScaleMode

0.49

virgin

0.49

WriteTagHelper

0.47

 Autorizaciones

0.44

onAttach

0.43

 CanadaChoose

0.43

 virgin

0.43

 unexplored

0.41

 virgen

0.40

Activations Density 0.054%

references to new developments or discoveries

No Comments

No Known Activations