INDEX

Explanations

negation or terms indicating the absence of something

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 1-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_1/width_16k/average_l0_102

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.1.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

LabelTagHelper

-0.76

 vettor

-0.74

 vectorielles

-0.71

 proprement

-0.69

#+#

-0.66

曖昧さ回避

-0.65

 complètes

-0.64

 titolata

-0.64

 supérieures

-0.64

mybatisplus

-0.64

POSITIVE LOGITS

non

4.58

Non

4.16

Non

4.12

non

4.05

NON

3.72

NON

3.44

非

3.14

非

2.66

 nons

2.37

 icke

2.30

Activations Density 0.088%

negation or terms indicating the absence of something

No Comments

No Known Activations