INDEX

Explanations

symbols and codes frequently used in programming or computational contexts

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

key: value pairs

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

AccessorTable

-0.45

Ref

-0.40

ref

-0.36

Ref

-0.35

Ca

-0.33

يكب

-0.33

Ca

-0.32

 Once

-0.32

Reg

-0.32

ref

-0.32

POSITIVE LOGITS

 שוליים

0.77

HomeAsUpEnabled

0.66

istoitu

0.64

URLException

0.61

Tikang

0.60

abestanden

0.60

 Biôgrafia

0.59

 Wiktionnaire

0.57

Билгалдахарш

0.57

("")]

0.56

Activations Density 0.144%

symbols and codes frequently used in programming or computational contexts

key: value pairs

No Comments

No Known Activations