INDEX

Explanations

references to saliva or related terms

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

New Auto-Interp

Top Features by Cosine Similarity

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

Configuration

google/gemma-scope-9b-pt-res/layer_2/width_16k/average_l0_67

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.2.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

 AssemblyCulture

-0.44

 Unidas

-0.39

 whereof

-0.39

diego

-0.35

Boy

-0.35

AsUp

-0.34

kkende

-0.34

lète

-0.34

DialogInterface

-0.34

 McBride

-0.34

POSITIVE LOGITS

sal

3.89

Sal

3.78

Sal

3.78

sal

3.50

SAL

3.28

SAL

3.13

 SALV

2.08

 Sali

2.06

 salted

1.93

 salic

1.92

Activations Density 0.186%

references to saliva or related terms

No Comments

No Known Activations