INDEX

Explanations

words related to data or technological specifications

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 10-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_10/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.10.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

Datuak

-0.71

Personendaten

-0.69

basicConfig

-0.68

GeneratedCode

-0.67

 jsPsych

-0.65

ArgsConstructor

-0.65

يكب

-0.64

 allAfrica

-0.63

NameInMap

-0.63

complexContent

-0.63

POSITIVE LOGITS

thâu

0.44

̍t

0.41

ieť

0.40

ืม

0.39

ndose

0.38

TNode

0.37

MathML

0.36

temos

0.36

軌道

0.36

 assurent

0.36

Activations Density 0.202%

words related to data or technological specifications

No Comments

No Known Activations