INDEX

Explanations

terms related to specific qualities or characteristics of items, namely their types, kinds, or categories

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 17-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_17/width_16k/average_l0_77

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.17.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

værende

-0.60

HasAnnotation

-0.54

жется

-0.53

באנגלית

-0.52

koku

-0.48

AddWithValue

-0.47

attamento

-0.45

WidgetItem

-0.45

 wykonania

-0.43

 다른

-0.42

POSITIVE LOGITS

 kind

1.37

 type

1.31

 kinds

1.20

 sort

1.20

 sorts

1.19

 soort

1.15

 types

1.13

type

1.11

 exact

1.04

kind

1.03

Activations Density 0.381%

terms related to specific qualities or characteristics of items, namely their types, kinds, or categories

No Comments

No Known Activations