INDEX

Explanations

negative phrasing and contrastive statements

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 15-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_15/width_16k/average_l0_78

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.15.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

AsUp

-0.69

makeText

-0.58

DispatchToProps

-0.54

Бахар

-0.45

ниципа

-0.44

XMLSchema

-0.44

spli

-0.43

but

-0.42

లి

-0.42

]<<"

-0.41

POSITIVE LOGITS

 merely

0.99

 solely

0.97

 tantum

0.91

 alone

0.91

 only

0.89

 onely

0.85

 wyłącznie

0.85

only

0.81

 exclusively

0.80

alnız

0.78

Activations Density 0.285%

negative phrasing and contrastive statements

No Comments

No Known Activations