INDEX

Explanations

content related to relationships and familial connections

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 21-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_21/width_16k/average_l0_70

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.21.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

Obvious

-1.00

 ovviamente

-0.96

 seems

-0.91

 obvious

-0.89

 obviously

-0.88

 wydaje

-0.88

 natuurlijk

-0.88

 reminder

-0.87

 reportedly

-0.86

 кажется

-0.86

POSITIVE LOGITS

 indeed

1.26

indeed

1.01

 efectivamente

1.01

 inderdaad

0.87

 وأن

0.77

 للمعارف

0.75

 tatsächlich

0.74

 actually

0.74

 Indeed

0.72

 effectivement

0.67

Activations Density 1.297%

content related to relationships and familial connections

No Comments

No Known Activations