INDEX

Explanations

comparative phrases highlighting differences in amounts or qualities

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

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 nakalista

-0.44

Rptr

-0.44

unref

-0.43

vician

-0.40

phrag

-0.38

 consigo

-0.38

 Conservancy

-0.37

 outpost

-0.36

 mộc

-0.36

ilíbrio

-0.35

POSITIVE LOGITS

脚注の使い方

0.50

 BoxFit

0.47

krat

0.44

 informée

0.41

hæng

0.41

Personendaten

0.40

 navnet

0.39

euc

0.38

endpush

0.38

 ब्रेकडाउन

0.38

Activations Density 0.190%

comparative phrases highlighting differences in amounts or qualities

No Comments

No Known Activations