INDEX

Explanations

references to requirements or obligations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 3-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_3/width_16k/average_l0_90

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.3.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

 дописавши

-0.62

 involves

-0.46

HasIndex

-0.45

Jîn

-0.42

يميديا

-0.42

 involving

-0.42

 involve

-0.42

 presupposes

-0.42

WriteLiteral

-0.40

DIPSETTING

-0.40

POSITIVE LOGITS

 required

3.25

required

2.94

 Required

2.83

Required

2.69

 REQUIRED

2.44

REQUIRED

2.22

 needed

2.11

 Needed

1.89

needed

1.86

 necessary

1.82

Activations Density 0.137%

references to requirements or obligations

No Comments

No Known Activations