INDEX

Explanations

instances of statistical or mathematical notation

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 0-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_0/width_16k/average_l0_105

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.0.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

 itſelf

-1.89

 purpoſe

-1.73

 houſe

-1.65

 Monfieur

-1.64

 themſelves

-1.61

 himſelf

-1.61

 myſelf

-1.60

 ſta

-1.58

Efq

-1.55

 ſche

-1.55

POSITIVE LOGITS

0.77

the

0.69

↵↵

0.69

is

0.68

<eos>

0.68

0.67

in

0.66

my

0.63

0.62

Activations Density 0.691%

instances of statistical or mathematical notation

No Comments

No Known Activations