INDEX

Explanations

sequential indicators and descriptors related to lists or steps in an argument or explanation

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 15-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_15/width_16k/average_l0_78

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.15.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

########.

-0.68

 oprot

-0.67

smithy

-0.66

'},

-0.63

$}}

-0.63

ArrowToggle

-0.62

期刊论文

-0.61

daß

-0.59

providedIn

-0.59

ыгана

-0.58

POSITIVE LOGITS

Климат

0.77

Secondly

0.57

 Secondly

0.55

就是

0.54

TabStop

0.53

 اینکه

0.53

">//

0.51

EditorBrowsable

0.47

levance

0.47

 aspect

0.47

Activations Density 0.297%

sequential indicators and descriptors related to lists or steps in an argument or explanation

No Comments

No Known Activations