INDEX

Explanations

terms related to formal acceptance and explicit instructions in scholarly communication

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

Follows "explicit" or "explicitly."

np_acts-logits-general · gemini-2.0-flash

explicitly, directly, specifically

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

New Auto-Interp

Top Features by Cosine Similarity

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

Configuration

google/gemma-scope-2b-pt-res/layer_16/width_16k/average_l0_78

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.16.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

 CreateTagHelper

-0.84

 specifically

-0.74

 специально

-0.72

}`).

-0.71

 Specifically

-0.70

Specifically

-0.69

utilisons

-0.68

httphttps

-0.67

 AssemblyCompany

-0.67

specifically

-0.66

POSITIVE LOGITS

적인

0.51

 genomen

0.50

 kohta

0.49

 stället

0.48

 Ecological

0.47

ιά

0.47

orship

0.45

velg

0.44

contours

0.44

nambuco

0.43

Activations Density 0.535%

terms related to formal acceptance and explicit instructions in scholarly communication

Follows "explicit" or "explicitly."

explicitly, directly, specifically

No Comments

No Known Activations