INDEX

Explanations

JSON structure elements

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

`bf` notation for vectors

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

if

-0.51

-0.50

 grade

-0.42

 alarm

-0.42

If

-0.42

 come

-0.41

IF

-0.40

 minimum

-0.40

 guard

-0.40

 estimate

-0.39

POSITIVE LOGITS

 {...

0.85

{...

0.82

Diwedd

0.65

jsxFileName

0.63

 EconPapers

0.61

 للمعارف

0.60

NameInMap

0.59

 فريبيس

0.59

AnchorStyles

0.58

 beginnetje

0.57

Activations Density 0.001%

JSON structure elements

`bf` notation for vectors

No Comments

No Known Activations