INDEX

Explanations

references to the name "Watson" and related figures in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 9-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_9/width_16k/average_l0_73

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.9.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 tourmaline

-0.88

 Poincaré

-0.84

seamnă

-0.80

 Lucca

-0.79

 Arend

-0.78

Портал

-0.78

 tartalomajánló

-0.77

HX

-0.77

 Leaves

-0.76

gdx

-0.76

POSITIVE LOGITS

 numberWith

0.79

 resourceCulture

0.73

 imageNamed

0.72

 Jacobson

0.71

 DbType

0.70

 Watson

0.70

Swal

0.69

engesch

0.68

Tikang

0.67

thalene

0.66

Activations Density 1.200%

references to the name "Watson" and related figures in various contexts

No Comments

No Known Activations