INDEX

Explanations

references to application context in a programming context

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

creation of objects

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

MessageTagHelper

-0.59

 Chwiliwch

-0.54

 препратки

-0.47

 CreateTagHelper

-0.46

#+#

-0.45

 FormBuilder

-0.44

Autoritní

-0.44

 useCallback

-0.44

AutoresizingMask

-0.43

LabelTagHelper

-0.43

POSITIVE LOGITS

 financiación

0.39

 hidupnya

0.38

 terciopelo

0.38

 cuarzo

0.35

 createStore

0.35

autoreleasepool

0.35

PreferredItem

0.35

openSession

0.34

tidumbre

0.34

 détruire

0.34

Activations Density 0.010%

references to application context in a programming context

creation of objects

No Comments

No Known Activations