INDEX

Explanations

legal jargon and phrases related to rules, laws, and legal reasoning

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

official wording

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



-0.75

awtextra

-0.60

出版年

-0.57

帖最后由

-0.52

 snippetHide

-0.52

StoryboardSegue

-0.50

 Paglinawan

-0.49

TypedDataSet

-0.48

expandindo

-0.48

ValueStyle

-0.47

POSITIVE LOGITS

Eſ

0.44

 propOrder

0.40

 officially

0.40

parece

0.38

 incrí

0.37

officially

0.36

Catatan

0.36

 official

0.35

 ofici

0.35

 sejauh

0.35

Activations Density 0.188%

legal jargon and phrases related to rules, laws, and legal reasoning

official wording

No Comments

No Known Activations