INDEX

Explanations

specific terminology related to processing and handling complex information or data

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

technical terms and analysis

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

 installé

-0.42

 utafitiHapana

-0.40

</thead>

-0.40

 Ressource

-0.39

niosek

-0.39

engarkan

-0.38

ностран

-0.36

 '\\;'

-0.36

発展

-0.35

yelidikan

-0.35

POSITIVE LOGITS

 BoxFit

0.52

GOTREF

0.49

gebob

0.46

PYX

0.46

TestingModule

0.45

AsUp

0.44

 jsPsych

0.42

 становника

0.42

ViewChild

0.41

 ProtoMessage

0.41

Activations Density 0.254%

specific terminology related to processing and handling complex information or data

technical terms and analysis

No Comments

No Known Activations