INDEX

Explanations

phrases indicating removal or exit from a situation or context

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 5-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_5/width_16k/average_l0_68

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.5.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

zé

-0.50

she

-0.45

flashdata

-0.45

-0.43

emptyset

-0.43

áculos

-0.43

物

-0.43

提

-0.42

 (;;)

-0.42

und

-0.42

POSITIVE LOGITS

CodedInputStream

0.90

 فريبيس

0.89

 صوتيه

0.81

出版年

0.77

PROMOTED

0.76

ViewImports

0.75

 IndexError

0.74

 الاطلاع

0.74

glGet

0.71

 समीक्षक

0.71

Activations Density 0.315%

phrases indicating removal or exit from a situation or context

No Comments

No Known Activations