INDEX

Explanations

terms related to arguments and debate

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

Tokens related to "argu" or "arg" or "ARG"

np_acts-logits-general · gemini-2.0-flash

argue or 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

+#+#

-0.59

makeText

-0.57

 Waray

-0.56

Cec

-0.56

__.__

-0.55

 kasarigan

-0.55

nitte

-0.54

atimes

-0.53

 jadx

-0.51

ToMany

-0.50

POSITIVE LOGITS

ative

0.79

 Arguments

0.74

 arguments

0.74

ation

0.73

arguments

0.71

 getArguments

0.69

cshtml

0.69

umentación

0.67

UMENTS

0.64

callee

0.64

Activations Density 0.096%

terms related to arguments and debate

Tokens related to "argu" or "arg" or "ARG"

argue or argument

No Comments

No Known Activations