INDEX

Explanations

references to data recording and documentation

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

record and page numbers

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 31-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_31/width_131k/average_l0_109

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.31.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

accéder

-0.41

 onOptions

-0.39

impianto

-0.38

 Goodness

-0.36

thwaite

-0.36

 Mods

-0.34

transformers

-0.33

 OnInit

-0.32

McE

-0.32

ppers

-0.31

POSITIVE LOGITS

 record

0.65

record

0.62

ValueStyle

0.60

 RECORD

0.58

الحياه

0.58

 Record

0.57

 recording

0.57

 وتسجيلات

0.56

Recording

0.56

 MessageBoxIcon

0.56

Activations Density 0.047%

references to data recording and documentation

record and page numbers

No Comments

No Known Activations