INDEX

Explanations

descriptions of truth versus deception in various contexts

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

"truth" or "reality"

np_acts-logits-general · gemini-2.0-flash

truth, fact, reality

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 22-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_22/width_16k/average_l0_72

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.22.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

tagHelper

-0.64

ContentAsync

-0.61

SharedDtor

-0.60

 pinulongan

-0.59

PullParser

-0.58

المناصب

-0.57

CodeAttribute

-0.56

PreAuthorize

-0.56

ArrowToggle

-0.54

GOTREF

-0.54

POSITIVE LOGITS

 reality

3.05

 truth

2.69

reality

2.53

 facts

2.42

 fact

2.39

 Reality

2.37

 realities

2.27

Reality

2.24

Fact

2.23

fact

2.22

Activations Density 0.421%

descriptions of truth versus deception in various contexts

"truth" or "reality"

truth, fact, reality

No Comments

No Known Activations