INDEX

Explanations

numeric values, particularly those representing statistics or measurements

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

citation references

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

 arşivlendi

-0.55

 ويكيميديا

-0.48

 InputDecoration

-0.45

VIAF

-0.45

principalColumn

-0.44

 Económica

-0.44

 CanadaChoose

-0.44

jména

-0.43

 ब्रेकडाउन

-0.43

 '\\;'

-0.43

POSITIVE LOGITS

Tikang

0.49

‐

0.48

 Kins

0.47

thig

0.47

✨:

0.47

dirond

0.47

0.46

netbeans

0.45

*-

0.44

jds

0.44

Activations Density 0.019%

numeric values, particularly those representing statistics or measurements

citation references

No Comments

No Known Activations