INDEX

Explanations

discussions around perception and impact, particularly in relation to negative influences and emotions

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

 Infórmanos

-0.44

 TextAppearance

-0.39

transQ

-0.38

GEBURTSDATUM

-0.38

parametrize

-0.37

AndPassword

-0.36

schra

-0.36

inSlope

-0.35

 Invent

-0.35

ovatel

-0.35

POSITIVE LOGITS

 Wikiseite

0.50

ACHUSET

0.44

 sedikit

0.44

 apprehensions

0.43

 toalla

0.42

 Erscheinung

0.42

Jeografia

0.42

 esigenze

0.41

TagNumber

0.41

 Nachteile

0.40

Activations Density 0.049%

discussions around perception and impact, particularly in relation to negative influences and emotions

No Comments

No Known Activations