INDEX

Explanations

infinitive verbs and phrases indicating potential actions or capabilities

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

New Auto-Interp

Top Features by Cosine Similarity

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

Configuration

google/gemma-scope-2b-pt-res/layer_2/width_16k/average_l0_141

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.2.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

GEBURTSDATUM

-1.08

ⓧ

-0.91

 оригіналу

-0.81

 Majefty

-0.80

 CreateTagHelper

-0.80

 ftate

-0.79

WireFormat

-0.78

Geplaatst

-0.77

 ſtate

-0.76

TagMode

-0.74

POSITIVE LOGITS

can

1.28

Can

0.96

can

0.95

Can

0.94

 could

0.88

CAN

0.81

 dapat

0.78

Puede

0.77

 easily

0.76

สามารถ

0.76

Activations Density 0.100%

infinitive verbs and phrases indicating potential actions or capabilities

No Comments

No Known Activations