INDEX

Explanations

terms related to sampling and retrieval in experiments or data analysis

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-mlp-16k

Configuration

google/gemma-scope-2b-pt-mlp/layer_1/width_16k/average_l0_105

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.1.hook_mlp_out

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

 nahilalakip

-0.77

发表于

-0.62

parsedMessage

-0.61

MetaObject

-0.60

/******/

-0.59

 Audiodateien

-0.59

 xoay

-0.57

ValueStyle

-0.57

脚注の使い方

-0.56

tvguidetime

-0.56

POSITIVE LOGITS

 chré

0.52

anskje

0.52

ActionTypes

0.51

 Oldest

0.48

 hereof

0.47

 Justiça

0.47

 comerciais

0.46

mú

0.46

طق

0.46

 pulizia

0.45

Activations Density 0.569%

terms related to sampling and retrieval in experiments or data analysis

No Comments

No Known Activations