INDEX

Explanations

words related to injury or trauma

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

New Auto-Interp

Top Features by Cosine Similarity

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

Configuration

google/gemma-scope-9b-pt-res/layer_2/width_16k/average_l0_67

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.2.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

EndGlobalSection

-0.57

 utafitiHapana

-0.56

Geplaatst

-0.56

исленность

-0.53

tanleria

-0.52

 makl

-0.52

 Sama

-0.51

 Мексичка

-0.49

dhury

-0.49

 abſ

-0.49

POSITIVE LOGITS

ble

0.75

__':

0.66

</thead>

0.60

Ble

0.59

ble

0.57

Ble

0.55

__":

0.55

BLE

0.54

__':

0.54

BLE

0.54

Activations Density 0.588%

words related to injury or trauma

No Comments

No Known Activations