INDEX

Explanations

expressive phrases indicating quality or degree

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 7-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_7/width_16k/average_l0_69

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.7.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

webElementXpaths

-0.72

خرى

-0.67

 Référence

-0.63

GEBURTSDATUM

-0.63

rungsseite

-0.63

❹

-0.60

parseFrom

-0.59

новременно

-0.58

ModelForm

-0.58

 indisponible

-0.57

POSITIVE LOGITS

 aveug

0.77

few

0.74

 very

0.70

bit

0.68

Ně

0.68

Aiheesta

0.67

 väh

0.66

raisemb

0.66

 quantidade

0.66

 Lovel

0.66

Activations Density 0.451%

expressive phrases indicating quality or degree

No Comments

No Known Activations