INDEX

Explanations

concepts related to accuracy and ideal outcomes

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

not always

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

thâu

-0.61

دانشنامهٔ

-0.53

󠁴

-0.49

лтемелер

-0.48

Except

-0.47

except

-0.46

createClass

-0.46

arsh

-0.45

akujem

-0.44

 Except

-0.44

POSITIVE LOGITS

nor

0.92

Nor

0.72

 Tampoco

0.65

而是

0.65

 sometimes

0.64

Nor

0.64

也不是

0.63

nor

0.63

 tampoco

0.61

 Sometimes

0.57

Activations Density 0.085%

concepts related to accuracy and ideal outcomes

not always

No Comments

No Known Activations