INDEX

Explanations

concepts related to morality and humanity

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 12-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_12/width_16k/average_l0_82

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.12.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

 feen

-0.64

 ſeveral

-0.60

 occafion

-0.60

 myſelf

-0.60

 leſs

-0.59

 fubject

-0.58

 itſelf

-0.58

 fufficient

-0.57

 مرئيه

-0.57

Efq

-0.56

POSITIVE LOGITS

+:+

0.74

tvguidetime

0.71



0.70

IVEREF

0.66

TagMode

0.65

PhysRevD

0.64

CWE

0.64

ViewFeatures

0.61

GraphicsUnit

0.59

 SwitchCompat

0.59

Activations Density 0.245%

concepts related to morality and humanity

No Comments

No Known Activations