INDEX

Explanations

symbols and formatting indicators in the text

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 13-gemmascope-res-65k

Configuration

google/gemma-scope-2b-pt-res/layer_13/width_65k/average_l0_75

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

65,536

Data Type

float32

Hook Name

blocks.13.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

 πως

-0.84

 autorytatywna

-0.71

Personendaten

-0.69

vrigt

-0.67

 незавершена

-0.66

Manbalar

-0.65

 Forth

-0.60

 Though

-0.60

BagConstraints

-0.59

 voilà

-0.59

POSITIVE LOGITS

1.00

 XNUMX

0.87

.;

0.75

.:

0.74

NUMX

0.68

⁇

0.65

 AssemblyTitle

0.62

!!

0.62

있는

0.61

̵

0.60

Activations Density 0.013%

symbols and formatting indicators in the text

No Comments

No Known Activations