INDEX

Explanations

instances of high-stakes scenarios or conditions involving legal language

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

 flotante

-0.39

 Größen

-0.37

 cromado

-0.35

 Suiza

-0.35

 independientes

-0.35

 Stufen

-0.35

 Spitze

-0.34

setPointSize

-0.33

 Verfahren

-0.33

 provas

-0.33

POSITIVE LOGITS

 للمعارف

1.01

OGND

0.87

 ویکی‌پدی

0.66



0.65

setVerticalGroup

0.63

 Infórmanos

0.63

expandindo

0.60

ंदीखरीदारी

0.60

RAG

0.60

 EconPapers

0.59

Activations Density 0.069%

instances of high-stakes scenarios or conditions involving legal language

No Comments

No Known Activations