INDEX

Explanations

programming constructs and references related to method implementation

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

technical terms or concepts

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

New Auto-Interp

Top Features by Cosine Similarity

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

Configuration

google/gemma-scope-9b-it-res/layer_9/width_131k/average_l0_121

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.9.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

consum

-0.34

 whose

-0.30

memory

-0.28

 Cheval

-0.27

boo

-0.27

 which

-0.27

 dür

-0.26

brid

-0.26

 bridges

-0.26

 bron

-0.25

POSITIVE LOGITS

følgelig

0.66

 Bewußt

0.65

AutoScaleMode

0.63

ſelf

0.61

WriteBarrier

0.60

BufferException

0.59

 EconPapers

0.59

 NSCoder

0.57

 Inſ

0.56

RegressionTest

0.55

Activations Density 0.367%

programming constructs and references related to method implementation

technical terms or concepts

No Comments

No Known Activations