INDEX

Explanations

references to political positions and titles within a governmental context

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

 kasarigan

-0.69

 Wiktionnaire

-0.65

 nonUne

-0.60

UrlResolution

-0.60

évaluateur

-0.60

initWithFrame

-0.58

########.

-0.57

 OnInit

-0.56

 poveznice

-0.56

uttgart

-0.55

POSITIVE LOGITS

 Pence

0.47

epres

0.39

VP

0.36

 Vice

0.36

bajo

0.35

 quasi

0.33

 vice

0.32

Gew

0.32

 pilot

0.30

わり

0.30

Activations Density 0.010%

references to political positions and titles within a governmental context

No Comments

No Known Activations