INDEX

Explanations

phrases that indicate speculation or uncertainty about events

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

Attribution or hearsay

np_acts-logits-general · gemini-2.0-flash

reporting or alleged events

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

 obviously

-0.76

我相信

-0.62

 undeniably

-0.60

 Obviously

-0.60

imo

-0.58

Obviously

-0.58

想必

-0.57

 moim

-0.57

 denke

-0.56

 ويكيميديا

-0.56

POSITIVE LOGITS

 jopa

0.98

UnusedPrivate

0.64

NewUrlParser

0.64

DockStyle

0.64

TestBed

0.64

sizeCache

0.64

 sogar

0.63

 addirittura

0.63

 abbia

0.63

 abbiano

0.62

Activations Density 0.337%

phrases that indicate speculation or uncertainty about events

Attribution or hearsay

reporting or alleged events

No Comments

No Known Activations