INDEX

Explanations

references to analytical processes and assessments

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

New Auto-Interp

Top Features by Cosine Similarity

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

Configuration

google/gemma-scope-9b-pt-res/layer_20/width_16k/average_l0_68

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

YourGuide

-0.52

omitempty

-0.52

 enfans

-0.51

 myſelf

-0.51

 teuer

-0.51

 Houſe

-0.50

 GoogleFonts

-0.50

 poffible

-0.50

HomeAsUpEnabled

-0.49

béco

-0.48

POSITIVE LOGITS

 analysis

1.27

analysis

1.12

 Analysis

1.12

 ANALYSIS

1.07

 analyze

1.01

Analysis

1.01

 analyses

0.98

 Analyse

0.98

 analyse

0.94

 Analyze

0.94

Activations Density 0.089%

references to analytical processes and assessments

No Comments

No Known Activations