INDEX

Explanations

occurrences of phrases related to online activities, awards, and performance evaluation

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

Past tense verbs/participles

np_acts-logits-general · gemini-2.0-flash

verbs followed by connecting words

np_acts-logits-general · gemini-2.5-flash-lite

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 24-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_24/width_16k/average_l0_114

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.24.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

Datuak

-0.40

autorest

-0.36

scriptcase

-0.36

dagangan

-0.34

 telt

-0.34

 willst

-0.34

">//

-0.33

Sklici

-0.33

 garaje

-0.32

⟬

-0.31

POSITIVE LOGITS

 defaultstate

0.71

تقاوى

0.68

 himself

0.57

 gră

0.48

 propOrder

0.47

hObject

0.47

 pouvoit

0.46

 незавершена

0.45

 hemispheres

0.44

 riuscito

0.44

Activations Density 0.685%

occurrences of phrases related to online activities, awards, and performance evaluation

Past tense verbs/participles

verbs followed by connecting words

No Comments

No Known Activations