INDEX

Explanations

numerical references related to statistics or data in research contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 20-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_20/width_16k/average_l0_71

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.20.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

Efq

-1.10

 Monfieur

-1.08

 myſelf

-1.07

 itſelf

-1.02

 Jefus

-1.00

 ſeveral

-0.97

])));

-0.92

 Theſe

-0.90

UserScript

-0.89

 themſelves

-0.89

POSITIVE LOGITS

0.56

in

0.53

so

0.53

0.52

en

0.51

ly

0.51

lo

0.51

ra

0.50

ins

0.49

Activations Density 0.068%

numerical references related to statistics or data in research contexts

No Comments

No Known Activations