INDEX

Explanations

proper nouns, particularly names of individuals and organizations

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-131k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

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

StoryboardSegue

-0.54

UserScript

-0.53

principalTable

-0.52

argout

-0.51

[]:

-0.48

 AssemblyTitle

-0.47

richTextPanel

-0.47

roned

-0.47

Попис

-0.47

 चीज़ों

-0.46

POSITIVE LOGITS

 plufieurs

0.56

 ainfi

0.51

ſelf

0.49

 itſelf

0.48

 démission

0.46

 Monfieur

0.44

 المعيارى

0.44

 reaſon

0.43

 pymysql

0.43

 parís

0.43

Activations Density 0.016%

proper nouns, particularly names of individuals and organizations

No Comments

No Known Activations