INDEX

Explanations

technical terms and structures related to software, data, and relationships in programming and biology

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Configuration

google/gemma-scope-9b-pt-res/layer_20/width_131k/average_l0_114

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

tvguidetime

-0.40

 ageing

-0.35

 Вікіпе

-0.35

 adjusting

-0.34

 roused

-0.34

يكب

-0.34

เจ้า

-0.33

 ADJUST

-0.33

adjust

-0.32

abestanden

-0.32

POSITIVE LOGITS

 seperate

0.48

ücksich

0.47

 doInBackground

0.45

 SEPAR

0.45

 terpisah

0.45

 Separate

0.45

separate

0.44

ſelves

0.44

 separate

0.43

الإنجليزية

0.43

Activations Density 0.227%

technical terms and structures related to software, data, and relationships in programming and biology

No Comments

No Known Activations