INDEX

Explanations

sentiments related to experiences and conclusions drawn from situations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 21-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_21/width_16k/average_l0_129

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.21.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

 potest

-0.40

cæ

-0.39

 bré

-0.39

êques

-0.39

 rall

-0.38

shr

-0.37

 pous

-0.35

Mg

-0.35

 buta

-0.35

 Gallimard

-0.34

POSITIVE LOGITS

 ModelExpression

0.66

 ternyata

0.60

 okaza

0.58

果然

0.57

 Gegenteil

0.56

 efectivamente

0.56

archiviato

0.55

 indeed

0.55

 malah

0.54

 succeeds

0.52

Activations Density 0.841%

sentiments related to experiences and conclusions drawn from situations

No Comments

No Known Activations