INDEX

Explanations

references to character encoding, specifically UTF formats

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 11-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_11/width_16k/average_l0_80

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.11.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

ValueStyle

-0.63

 Perang

-0.59

 deler

-0.57

aliases

-0.56

kann

-0.56

̣

-0.56

apiVersion

-0.56

blad

-0.56

 Girolamo

-0.55

POSITIVE LOGITS

utf

1.78

utf

1.77

UTF

1.48

UTF

1.30

Utf

1.25

Utf

1.11

 uitges

0.88

fono

0.69

PyExc

0.68

utm

0.68

Activations Density 0.030%

references to character encoding, specifically UTF formats

No Comments

No Known Activations