INDEX

Explanations

key phrases related to evaluating and discussing legal or procedural issues

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

followed by "another" or "an"

np_acts-logits-general · gemini-2.0-flash

one, then another

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 24-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_24/width_16k/average_l0_73

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.24.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

featureID

-0.67

Rujuakan

-0.64

 RequestMethod

-0.63

respectively

-0.58

ArrowToggle

-0.57

Cordialement

-0.57

 surla

-0.57

 defaultstate

-0.56

IsContent

-0.55

 cherchés

-0.54

POSITIVE LOGITS

 another

4.52

another

3.99

Another

3.70

 Another

3.70

 ANOTHER

3.17

 otro

2.60

 otra

2.48

另一个

2.40

Otra

2.28

另一個

2.26

Activations Density 0.800%

key phrases related to evaluating and discussing legal or procedural issues

followed by "another" or "an"

one, then another

No Comments

No Known Activations