INDEX

Explanations

programming-related terms and code structure elements

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

save/disable/add/requires/workaround

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

principalTable

-0.41

 pedagang

-0.39

rativa

-0.37

vous

-0.36

SaveChanges

-0.36

 illusions

-0.36

 namanya

-0.35

étr

-0.35

erdings

-0.35

fancy

-0.35

POSITIVE LOGITS

rungsseite

0.53

+#+#

0.53

ModelSerializer

0.52

 ujednoznacz

0.47

 Meksiku

0.46

🔕

0.46

IsMutable

0.41

WithIOException

0.39

Tikang

0.39

hnout

0.39

Activations Density 0.047%

programming-related terms and code structure elements

save/disable/add/requires/workaround

No Comments

No Known Activations