INDEX

Explanations

code-related structures and syntax, particularly related to functions and methods

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 23-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_23/width_16k/average_l0_75

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.23.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

++;

-0.56

.*;

-0.54

WhereInput

-0.51

[];

-0.51

 talk

-0.49

<eos>

-0.48

++);

-0.48

();*/

-0.47

 toch

-0.47

():

-0.46

POSITIVE LOGITS

 Note

0.80

 Useful

0.80

 NOTE

0.74

 WARNING

0.72

featureID

0.69

 Typically

0.68

GOTREF

0.68

 Оно

0.67

Typically

0.67

Useful

0.67

Activations Density 0.487%

code-related structures and syntax, particularly related to functions and methods

No Comments

No Known Activations