INDEX

Explanations

repetitive occurrences of articles in various contexts

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

"the" followed by an adjective

np_acts-logits-general · gemini-2.0-flash

"the" followed by descriptors

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 31-gemmascope-res-16k

Configuration

google/gemma-scope-9b-it-res/layer_31/width_16k/average_l0_76

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.31.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

 Exactos

-0.46

 роки

-0.39

цездатний

-0.38

PreExecute

-0.36

RegressionTest

-0.36

PostExecute

-0.35

 skoro

-0.35

 виправивши

-0.35

 saites

-0.35

قایناقلار

-0.34

POSITIVE LOGITS

 widest

0.70

 broadest

0.67

 same

0.61

 equivalent

0.59

matically

0.58

 smallest

0.57

 fastest

0.56

 highest

0.56

 strongest

0.56

 gleichen

0.56

Activations Density 0.686%

repetitive occurrences of articles in various contexts

"the" followed by an adjective

"the" followed by descriptors

No Comments

No Known Activations