INDEX

Explanations

terms associated with measurement and evaluation

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

ArgsConstructor

-0.53

 chi̍t

-0.52

EndContext

-0.52

IntoConstraints

-0.51

hdysval

-0.50

 exemplaires

-0.48

ContentAsync

-0.48

isContained

-0.46

 resourceCulture

-0.46

Географи

-0.46

POSITIVE LOGITS

 others

0.73

SequentialGroup

0.66

 Yoshida

0.62

由此

0.60

 other

0.59

іга

0.59

 Others

0.59

OURCE

0.59

ouille

0.57

others

0.57

Activations Density 0.617%

terms associated with measurement and evaluation

No Comments

No Known Activations