INDEX

Explanations

references to quality and excellence 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 @ 34-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_34/width_16k/average_l0_114

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.34.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

+#+#

-0.62

Personendaten

-0.48

 Enough

-0.41

IsMutable

-0.41

 fuertemente

-0.39

CloseOperation

-0.39

">—

-0.39

 như

-0.38

цездатний

-0.38

 wichtiger

-0.38

POSITIVE LOGITS

 ever

1.49

 imaginable

1.31

 EVER

1.15

ever

1.13

 ooit

1.04

 jemals

0.98

EVER

0.96

 possible

0.94

Ever

0.93

 conceivable

0.92

Activations Density 0.280%

references to quality and excellence in various contexts

No Comments

No Known Activations