INDEX

Explanations

structured data or programming constructs

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 5-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_5/width_16k/average_l0_68

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.5.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

 also

-0.58

 however

-0.56

 which

-0.52

 upon

-0.51

 instead

-0.49

 additionally

-0.49

 along

-0.48

↵↵

-0.47

 furthermore

-0.47

POSITIVE LOGITS

saraba

0.99

^(@)

0.84

脚注の使い方

0.83

SBATCH

0.83

 AssemblyVersion

0.83

IUrlHelper

0.79

 myſelf

0.76

 Bioaccumulative

0.75

BibitemShut

0.73

 surla

0.73

Activations Density 0.044%

structured data or programming constructs

No Comments

No Known Activations