INDEX

Explanations

instances where examples of various concepts or entities are provided

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

examples of

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

ſelf

-0.46

 procédures

-0.40

sterone

-0.37

 délais

-0.36

Kjelder

-0.36

awatan

-0.36

 périodes

-0.35

Robbie

-0.35

 laissant

-0.34

chriften

-0.33

POSITIVE LOGITS

 examples

0.83

oa̍t

0.70

 eksemp

0.65

Examples

0.64

 Examples

0.64

hyrchwyd

0.63

AutoScaleMode

0.61

 exemplos

0.60

 esempi

0.60

WithMany

0.59

Activations Density 0.041%

instances where examples of various concepts or entities are provided

examples of

No Comments

No Known Activations