INDEX

Explanations

requests for information or actions, often prefaced by "please."

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

New Auto-Interp

Top Features by Cosine Similarity

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

Configuration

google/gemma-scope-9b-it-res/layer_9/width_16k/average_l0_88

Prompts (Dashboard)

24,576 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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

multirow

-0.34

IRQn

-0.33

 nød

-0.32

 drug

-0.31

แห

-0.30

 Küste

-0.30

ないけど

-0.30

 justement

-0.30

nullptr

-0.30

 tantas

-0.30

POSITIVE LOGITS

 Infórmanos

0.86

WriteTagHelper

0.71

 please

0.71

 Please

0.69

 oprot

0.66

Please

0.65

 пожалуйста

0.64

请选择

0.63

 Meksiku

0.63

ütfen

0.60

Activations Density 0.148%

requests for information or actions, often prefaced by "please."

No Comments

No Known Activations