INDEX

Explanations

content related to scientific measurements or analysis

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-16k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_16k/average_l0_91

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 للاسماء

-0.45

rosis

-0.45

 &___

-0.44

 Signalez

-0.43

 상세

-0.43

 تكبرها

-0.43

🟤

-0.43

ươi

-0.42

 corações

-0.42

 <>",

-0.42

POSITIVE LOGITS

 beiden

0.98

どちらも

0.98

 both

0.96

 ambos

0.93

 Both

0.92

Both

0.90

both

0.90

 entrambi

0.89

 beide

0.86

 begge

0.85

Activations Density 1.018%

content related to scientific measurements or analysis

No Comments

No Known Activations