INDEX

Explanations

terms related to image URLs and their handling

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

Lithuanian names

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-131k

Configuration

google/gemma-scope-9b-it-res/layer_31/width_131k/average_l0_109

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

脚注の使い方

-0.91

 queſta

-0.88

InputBorder

-0.87

 zwiſchen

-0.85

ロウィン

-0.84

﹧

-0.83

AsUp

-0.83

 myſelf

-0.82

<unused40>

-0.81

<unused20>

-0.81

POSITIVE LOGITS

ir

0.39

 cooperación

0.35

 gestión

0.32

นี่

0.28

 prevención

0.28

 límites

0.27

 acciones

0.27

 secretos

0.26

 sustancias

0.26

„

0.26

Activations Density 0.009%

terms related to image URLs and their handling

Lithuanian names

No Comments

No Known Activations