INDEX

Explanations

words and characters from multiple languages

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-16k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_16k/average_l0_91

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.20.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

-------

-0.74

Portály

-0.70

 AssemblyCulture

-0.69

AISSEE

-0.68

 betweenstory

-0.67

zzleHttp

-0.67

 informée

-0.67

Autoritní

-0.66

Fordítás

-0.66

Tikang

-0.66

POSITIVE LOGITS

מ

0.44

getC

0.43

 الم

0.42

Comp

0.42

 piú

0.42

is

0.42

अ

0.42

 ప్ర

0.42

פ

0.42

 प्र

0.42

Activations Density 0.172%

words and characters from multiple languages

No Comments

No Known Activations