INDEX

Explanations

situations involving deception or trickery

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

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 שוליים

-0.89

MemoryWarning

-0.75

 nakalista

-0.74

DeleteBehavior

-0.71

+#+

-0.70

#+#

-0.70

harusnya

-0.69

culously

-0.68

شهاد

-0.68

:+:

-0.68

POSITIVE LOGITS

 trap

1.81

 traps

1.58

 Trap

1.44

trap

1.43

 Traps

1.28

 TRAP

1.24

Trap

1.24

 trapping

1.20

陷阱

1.08

 trampa

1.08

Activations Density 0.265%

situations involving deception or trickery

No Comments

No Known Activations