INDEX

Explanations

conditional phrases and expressions of uncertainty or possibility

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

following "help," "lead," or similar terms

np_acts-logits-general · gemini-2.0-flash

potential and consequences

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

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 AssemblyCompany

-0.56

setVerticalGroup

-0.52

 Gunakan

-0.47

entingan

-0.45

 Tinggi

-0.45

ToUse

-0.41

especial

-0.40

resaid

-0.40

 informée

-0.40

spes

-0.39

POSITIVE LOGITS

 helps

0.70

 enables

0.68

 brings

0.65

 bring

0.60

 help

0.60

 enable

0.59

 ensures

0.57

 lead

0.56

 helped

0.55

 allows

0.54

Activations Density 0.563%

conditional phrases and expressions of uncertainty or possibility

following "help," "lead," or similar terms

potential and consequences

No Comments

No Known Activations