INDEX

Explanations

assertions related to the truthfulness and reliability of information

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

Followed by "true," "accurate," or "fact."

np_acts-logits-general · gemini-2.0-flash

true or true

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



-0.62

SharedDtor

-0.61

getDescriptor

-0.53

InjectAttribute

-0.53

SourceChecksum

-0.52

ssymb

-0.52

IVEREF

-0.49

 impatiently

-0.46

 Италијани

-0.46

OrBuilder

-0.46

POSITIVE LOGITS

 truth

2.27

 true

2.11

 True

1.91

truth

1.90

true

1.88

True

1.87

 Truth

1.83

 TRUE

1.82

TRUE

1.73

Truth

1.72

Activations Density 0.404%

assertions related to the truthfulness and reliability of information

Followed by "true," "accurate," or "fact."

true or true

No Comments

No Known Activations