INDEX

Explanations

references to structured data representation or mathematical formulations

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

"in the form"

np_acts-logits-general · gemini-2.0-flash

expressions in a specific format

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

databind

-0.51

readdir

-0.49

UnitTesting

-0.48

SpringRunner

-0.43

 kişilik

-0.42

 pati

-0.41

 StringField

-0.41

rxjs

-0.41

 McGraw

-0.40

setti

-0.40

POSITIVE LOGITS

 format

0.87

 formats

0.79

 formato

0.77

format

0.69

 Format

0.68

formato

0.64

 Formato

0.64

 Formats

0.63

 tabular

0.62

Format

0.62

Activations Density 0.560%

references to structured data representation or mathematical formulations

"in the form"

expressions in a specific format

No Comments

No Known Activations