INDEX

Explanations

references to phone numbers and related concepts

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

 InputDecoration

-0.45

endcsname

-0.43

𝒞

-0.38

UNTIME

-0.37

Captor

-0.37

ecap

-0.36

Vtbl

-0.36

 intervening

-0.36

 margin

-0.36

avax

-0.35

POSITIVE LOGITS

 telephone

0.83

 phone

0.81

 téléphone

0.81

telephone

0.80

 Telephone

0.77

Telephone

0.74

 AssemblyCompany

0.74

 telefónica

0.73

 telephones

0.72

 teléfono

0.70

Activations Density 0.004%

references to phone numbers and related concepts

No Comments

No Known Activations