INDEX

Explanations

syntax and structure for code-related tasks, particularly involving function definitions and return statements

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 21-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_21/width_16k/average_l0_129

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.21.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

 ujednoznacz

-0.83

Portail

-0.81

fjspx

-0.78

SBATCH

-0.73

inWeight

-0.73

 AssemblyCompany

-0.71

 nakalista

-0.70

IVEREF

-0.70

 дописавши

-0.70

WebServlet

-0.70

POSITIVE LOGITS

 thereby

0.38

<eos>

0.33

 instead

0.30

 Moreover

0.28

Another

0.28

 thus

0.27

 rather

0.27

Activations Density 0.597%

syntax and structure for code-related tasks, particularly involving function definitions and return statements

No Comments

No Known Activations