INDEX

Explanations

specific terms related to medical assessments and legal standards

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

-0.43

ical

-0.40

 Karlsson

-0.40

 proven

-0.38

 Roos

-0.37

of

-0.37

 вида

-0.37

vs

-0.36

POSITIVE LOGITS

ScopeManager

0.81

تقاوى

0.79

 lenker

0.77

الحياه

0.76

 OFDb

0.74

 Wikimedijinoj

0.73

 kasarigan

0.72

 ब्रेकडाउन

0.72

 Infórmanos

0.68

 ſever

0.67

Activations Density 0.053%

specific terms related to medical assessments and legal standards

No Comments

No Known Activations