INDEX

Explanations

terms related to relationships and connections

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

New Auto-Interp

Top Features by Cosine Similarity

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

Configuration

google/gemma-scope-9b-pt-res/layer_2/width_16k/average_l0_67

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.2.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

rawValue

-0.42

zungs

-0.38

 createState

-0.35

zni

-0.35

bitro

-0.34

颠

-0.34

 bawang

-0.32

oneofs

-0.32

 iconst

-0.31

AutoSize

-0.31

POSITIVE LOGITS

ties

1.04

0.87

 Slip

0.73

ies

0.69

slip

0.69

Slip

0.68

IES

0.68

 slip

0.66

Hochspringen

0.62

pas

0.60

Activations Density 0.556%

terms related to relationships and connections

No Comments

No Known Activations