INDEX

Explanations

references to proof and evidence in various contexts

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

Попис

-0.47

Demikian

-0.37

yet

-0.36

Yet

-0.35

ItemBackground

-0.35

 carreira

-0.35

 spoloč

-0.34

 tear

-0.33

 kardeş

-0.33

 çek

-0.32

POSITIVE LOGITS

 preuve

0.60

Zeneca

0.56

 Nachweis

0.52

evidence

0.52

DataAnnotations

0.52

Proven

0.51

Prove

0.50

pexpr

0.50

 proof

0.50

AnchorTagHelper

0.50

Activations Density 0.030%

references to proof and evidence in various contexts

No Comments

No Known Activations