INDEX

Explanations

technical terms and concepts related to communication and technology

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

TokenNameLBRACE

-0.59

帖最后由

-0.53

Obrázky

-0.53

-0.52

 muß

-0.52

]);

-0.52

....)

-0.52

!!!!!!!!!!!!!!!!

-0.52

……..

-0.51

 getPassword

-0.50

POSITIVE LOGITS

˹

0.52

 yalnızca

0.49

 biomarkers

0.44

 swiftly

0.43

 brindar

0.42

 intuition

0.42

 playful

0.41

 playfully

0.41

 boosting

0.41

 boosted

0.40

Activations Density 0.059%

technical terms and concepts related to communication and technology

No Comments

No Known Activations