INDEX

Explanations

punctuation marks, specifically periods

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

AutoField

-0.45

 aDecoder

-0.44

 ComVisible

-0.41

 виправивши

-0.38

IsOptional

-0.34

autoreleasepool

-0.34

SwingConstants

-0.32

綜合報導

-0.31

protoimpl

-0.29

 tartalomajánló

-0.29

POSITIVE LOGITS

mybatisplus

0.63

IVEREF

0.62

cheibe

0.59

atissimo

0.59

 estadounid

0.57

 dieſer

0.56

jectives

0.55

virons

0.55

regate

0.54

 increí

0.54

Activations Density 0.001%

punctuation marks, specifically periods

No Comments

No Known Activations