INDEX

Explanations

connections and relationships between various subjects or entities

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 36-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_36/width_16k/average_l0_120

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.36.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

 einerseits

-0.62

either

-0.61

 either

-0.60

 nämlich

-0.56

 både

-0.54

 entweder

-0.54

Попис

-0.53

both

-0.52

Either

-0.50

既

-0.49

POSITIVE LOGITS

 consequently

1.05

 consiguiente

1.02

 importantly

1.02

 consequent

1.02

 subsequently

1.01

 subsequent

0.98

何より

0.96

sequently

0.93

 correspondingly

0.89

 therefore

0.85

Activations Density 1.310%

connections and relationships between various subjects or entities

No Comments

No Known Activations