INDEX

Explanations

numerical data and statistics within a text

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

 resourceCulture

-0.76

 мәкал

-0.54

 wireType

-0.52

 cherchés

-0.51

 ويكيپيديا

-0.50

 ObjectParameter

-0.49

EndGlobal

-0.48

EndContext

-0.47

 fronts

-0.46

LookAnd

-0.46

POSITIVE LOGITS

 graduate

0.47

 graduated

0.44

 diploma

0.43

 GRADUATE

0.42

 podjet

0.40

 Diploma

0.39

 émer

0.37

 battre

0.37

 Graduate

0.36

Diploma

0.36

Activations Density 0.009%

numerical data and statistics within a text

No Comments

No Known Activations