INDEX

Explanations

terms related to dual functionality and selection processes

oai_token-act-pair · gpt-4o-mini

Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 1-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_1/width_16k/average_l0_69

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.1.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

IFrame

Link

Negative Logits

 navideña

-0.45

Deskripsi

-0.38

 hjælp

-0.37

Custom

-0.37

 hjelp

-0.36

št

-0.36

PDATE

-0.36

 pidana

-0.36

 custom

-0.35

 collect

-0.35

POSITIVE LOGITS

 Intention

0.66

intention

0.64

Selected

0.60

 intentionally

0.59

 intention

0.59

 intentional

0.58

 intentions

0.57

ölkerung

0.56

Opt

0.56

+#+#

0.56

Activations Density 0.258%

terms related to dual functionality and selection processes

No Known Activations