INDEX

Explanations

negative responses or expressions of disagreement

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

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

tagHelperRunner

-0.63

Билгалдахарш

-0.62

󠁣

-0.55

 мәкал

-0.50

 Wikimedijinoj

-0.47

 EconPapers

-0.46

Autoritní

-0.46

hoea

-0.45

UrlResolution

-0.43

 transfieras

-0.43

POSITIVE LOGITS

évaluateur

0.43

﹍﹍﹍

0.41

 Brunnen

0.41

ulemon

0.40

Dzi

0.40

Istorija

0.39

 Maier

0.38

 Rubin

0.37

 sibuk

0.37

ługo

0.36

Activations Density 0.008%

negative responses or expressions of disagreement

No Comments

No Known Activations