INDEX

Explanations

references to boron and related terms in a scientific context

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

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

""")

-0.61

 Klas

-0.58

setOnAction

-0.54

Tian

-0.52

Klas

-0.52

!')

-0.50

 Tian

-0.50

 Weeknd

-0.49

````

-0.48

']?>

-0.48

POSITIVE LOGITS

Bor

0.97

BOR

0.88

Bor

0.88

bor

0.88

 Bork

0.78

 Borough

0.75

 borough

0.75

bor

0.73

 boron

0.72

 Borges

0.71

Activations Density 0.012%

references to boron and related terms in a scientific context

No Comments

No Known Activations