INDEX

Explanations

proper nouns and specific names in various contexts, particularly related to people and places

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

Tokens followed by two-letter suffixes

np_acts-logits-general · gemini-2.0-flash

proper noun prefixes and suffixes

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

 Geiſt

-0.61

✷

-0.56

𑄴

-0.54

SqlClient

-0.54

𑄨

-0.54

shadowOffset

-0.53

𑄬

-0.52

𑄮

-0.52

doPost

-0.52

 ſeyn

-0.52

POSITIVE LOGITS

stdc

0.45

ParallelGroup

0.40

saraba

0.38

chengla

0.38

 referrerpolicy

0.38

 justement

0.37

Pozdrawiam

0.34

($__

0.34

djangoproject

0.34

 pinulongan

0.34

Activations Density 0.489%

proper nouns and specific names in various contexts, particularly related to people and places

Tokens followed by two-letter suffixes

proper noun prefixes and suffixes

No Comments

No Known Activations