INDEX

Explanations

references to debates and discussions

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

debate and discussion

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-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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ConstraintMaker

-0.60

 المعيارى

-0.55

 poveznice

-0.55

 <<<<<<<<<<<<<<

-0.53

expandindo

-0.52

tagHelperRunner

-0.46

])):

-0.45

 catedral

-0.44

 ujednoznacz

-0.44

ïsme

-0.44

POSITIVE LOGITS

 debat

0.50

 debate

0.47

 eloquence

0.45

 debates

0.44

 chamber

0.44

 Chamber

0.43

EventHandler

0.42

 judged

0.42

Sen

0.42

 bald

0.42

Activations Density 0.005%

references to debates and discussions

debate and discussion

No Comments

No Known Activations