INDEX

Explanations

terms related to intensifying negative conditions or symptoms

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

şört

-0.57

 Normdatei

-0.55

 CreateTagHelper

-0.55

AutoField

-0.55

setContentView

-0.53

 ReactDOM

-0.53

 Houſe

-0.52

eseorang

-0.52

Geplaatst

-0.52

LookAnd

-0.51

POSITIVE LOGITS

 exacerbate

0.82

 exacerbated

0.81

 aggravate

0.71

 worsening

0.70

 aggravated

0.69

 exacerb

0.68

 worsen

0.68

 agrav

0.66

 exacer

0.65

 worsened

0.62

Activations Density 0.024%

terms related to intensifying negative conditions or symptoms

No Comments

No Known Activations