INDEX

Explanations

instances of personal involvement and experiences in various contexts

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

past actions and involvement

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

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

 zijne

-0.54

 réalité

-0.46

 navideñas

-0.44

 dignité

-0.43

holdet

-0.43

cokinetics

-0.42

 veranderen

-0.42

 nôtre

-0.42

 dobbiamo

-0.42

UrlResolution

-0.41

POSITIVE LOGITS

 witness

0.56

 attest

0.47

 involved

0.46

 nakalista

0.44

 witnesses

0.44

 inval

0.44

 involvement

0.43

 dAtA

0.43

 WITNESS

0.43

Personendaten

0.42

Activations Density 0.025%

instances of personal involvement and experiences in various contexts

past actions and involvement

No Comments

No Known Activations