INDEX

Explanations

references to health-related data and indicators

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

followed by forms of "to be"

np_acts-logits-general · gemini-2.0-flash

identifying what something is

np_acts-logits-general · gemini-2.5-flash-lite

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 18-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_18/width_16k/average_l0_74

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.18.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

]--;

-0.79

Efq

-0.70

<bos>

-0.69

 Majefty

-0.68

 resourceCulture

-0.64

tvguidetime

-0.64

 AspNetCore

-0.61

 Biôgrafia

-0.61

IUrlHelper

-0.61

 Monfieur

-0.60

POSITIVE LOGITS

 merupakan

1.06

是最

1.01

is

0.97

才是

0.96

 adalah

0.94

るのが

0.88

するのが

0.87

 constitue

0.87

が一番

0.86

 является

0.82

Activations Density 0.616%

references to health-related data and indicators

followed by forms of "to be"

identifying what something is

No Comments

No Known Activations