INDEX

Explanations

terms that express relationships or connections, particularly in the context of personal pronouns and interactions

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

matches, common characters

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

-------

-0.65

 HasFactory

-0.62

VersionUID

-0.61

 HttpNotFound

-0.58

 deleteById

-0.57

 trône

-0.55

 صوتيه

-0.54

colhead

-0.53

domada

-0.53

 initComponents

-0.53

POSITIVE LOGITS

 self

0.60

 المت

0.49

 يت

0.48

 closely

0.48

 multi

0.48

 להת

0.47

 liên

0.47

ContextHolder

0.46

 focus

0.46

 connected

0.45

Activations Density 0.014%

terms that express relationships or connections, particularly in the context of personal pronouns and interactions

matches, common characters

No Comments

No Known Activations