INDEX

Explanations

statements questioning the validity or reliability of claims

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

End of sentences

np_acts-logits-general · gemini-2.0-flash

your argument

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 20-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_20/width_16k/average_l0_71

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.20.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

Whew

-0.59

왠

-0.57

 kadang

-0.56

seemed

-0.56

finally

-0.55

Phew

-0.55

colgroup

-0.54

 finally

-0.54

 наконец

-0.51

 manchmal

-0.51

POSITIVE LOGITS

又不是

0.91

setViewportView

0.81

LookAnd

0.73

FTFY

0.71

/=

0.70

あなたが

0.70

 argumento

0.68

 Irrelevant

0.67

 clearly

0.66

 dumbass

0.66

Activations Density 0.844%

statements questioning the validity or reliability of claims

End of sentences

your argument

No Comments

No Known Activations