INDEX

Explanations

structured data or definitions related to types

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

punctuation and code markers

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

MemoryWarning

-0.70

تقاوى

-0.61

oredCriteria

-0.60

 onOptions

-0.60

parsedMessage

-0.56

findpost

-0.55

windowFixed

-0.55

évaluateur

-0.55

TagMode

-0.55

 snippetHide

-0.55

POSITIVE LOGITS

“

0.59

0.58

SerializedName

0.48

「

0.48

‘

0.44

0.43

 <<<<<<<<<<<<<<

0.43

„

0.41

“

0.41

Activations Density 0.078%

structured data or definitions related to types

punctuation and code markers

No Comments

No Known Activations