INDEX

Explanations

references to geographic locations and historical sites

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 10-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_10/width_16k/average_l0_113

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.10.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

 indisponible

-0.36

uanto

-0.34

 sort

-0.30

kuchen

-0.28

 conditions

-0.28

tag

-0.27

 Thron

-0.27

POSITIVE LOGITS

queryInterface

0.70

⟬

0.69

AndEndTag

0.67

setVerticalGroup

0.66

 שוליים

0.66

 queſta

0.65

<unused43>

0.65

 ſind

0.65

<unused41>

0.65

<unused17>

0.65

Activations Density 0.386%

references to geographic locations and historical sites

No Comments

No Known Activations