INDEX

Explanations

technical terms and code-related contexts

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

 BorderRadius

-0.81

helves

-0.57

IContainer

-0.57

postsleuth

-0.56

醐

-0.54

 Normdatei

-0.54

ornadas

-0.52

 NSCoder

-0.52

BagConstraints

-0.51

 hablamos

-0.51

POSITIVE LOGITS

 code

1.75

code

1.33

代码

1.30

 코드

1.19

コード

1.18

 código

1.18

 Code

1.12

 codes

1.11

Code

1.07

 CODE

1.05

Activations Density 1.080%

technical terms and code-related contexts

No Comments

No Known Activations