INDEX

Explanations

keywords associated with evidence and frequency related to discussions or assertions

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

Code or non-natural language context

np_acts-logits-general · gemini-2.0-flash

turnover rate

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

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

migrationBuilder

-0.66

vatar

-0.63

GEBURTSDATUM

-0.62

INSEE

-0.58

今後

-0.56

 efectivamente

-0.54

 forse

-0.53

 tadi

-0.52

 endgül

-0.52

 originally

-0.52

POSITIVE LOGITS

 whenever

0.99

Sometimes

0.92

 either

0.91

 Sometimes

0.90

 sometimes

0.89

Whenever

0.89

 Whenever

0.88

 invariably

0.88

either

0.88

Either

0.88

Activations Density 0.421%

keywords associated with evidence and frequency related to discussions or assertions

Code or non-natural language context

turnover rate

No Comments

No Known Activations

keywords associated with evidence and frequency related to discussions or assertions

Code or non-natural language context

turnover rate

No Comments

No Known Activations