INDEX

Explanations

key terms related to specific requirements or properties in a given context

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

hanging

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

 Administrativna

-0.47

 الرياضيه

-0.47

хьтан

-0.47

 aanwezig

-0.45

 informée

-0.44

tvguidetime

-0.43

InputTagHelper

-0.43

 ब्रेकडाउन

-0.43

mbg

-0.41

 straat

-0.41

POSITIVE LOGITS

Читати

0.39

 betweenstory

0.38

>{@

0.38

ToProps

0.36

localctx

0.35

__*/

0.35

 ModelExpression

0.35

honor

0.34

omitempty

0.34

showConfirm

0.33

Activations Density 0.225%

key terms related to specific requirements or properties in a given context

hanging

No Comments

No Known Activations