INDEX

Explanations

concepts related to simplicity and complexity in ideas or lifestyles

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

 indef

-0.30

 inac

-0.30

anzug

-0.30

 Raymond

-0.29

نما

-0.28

 toler

-0.28

 drumming

-0.28

 plá

-0.27

sb

-0.27

Daryl

-0.27

POSITIVE LOGITS

 '\\;'

0.75

SourceChecksum

0.73

 nahilalakip

0.68

 initComponents

0.68

 simplifié

0.65

Autoritní

0.62

 Infórmanos

0.59

RTEX

0.58

 complexity

0.57

FormTagHelper

0.57

Activations Density 0.037%

concepts related to simplicity and complexity in ideas or lifestyles

No Comments

No Known Activations