INDEX

Explanations

references to specific automobile brands and their characteristics

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 10-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_10/width_16k/average_l0_77

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.10.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

IsContent

-0.59

liflower

-0.55

 propOrder

-0.54

WEBPACK

-0.52

 lasso

-0.51

 kimdir

-0.51

 déput

-0.50

 stället

-0.50

 INSTRUMENT

-0.50

 exchanger

-0.50

POSITIVE LOGITS

 brand

0.94

 brands

0.93

 branded

0.82

brands

0.73

branded

0.73

 manufacturer

0.73

 marchio

0.72

brand

0.70

 manufacture

0.69

ブランド

0.69

Activations Density 0.256%

references to specific automobile brands and their characteristics

No Comments

No Known Activations