INDEX

Explanations

references to specific events or placements in time and space within narratives

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

ElementException

-0.56

arraycopy

-0.46

Personensuche

-0.44

 playgrounds

-0.44

 autorytatywna

-0.44

 scaven

-0.43

 miniaturka

-0.42

 Bastion

-0.41

 ſich

-0.40

Tél

-0.40

POSITIVE LOGITS

+#+

0.52

 nakalista

0.50

AndEndTag

0.48

+:+

0.45

 ویکی‌پدیای

0.43

richTextPanel

0.41

참고

0.41

UnusedPrivate

0.39

ftagPool

0.39

[*]

0.39

Activations Density 0.029%

references to specific events or placements in time and space within narratives

No Comments

No Known Activations