INDEX

Explanations

words related to handling and processing tasks or activities

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 14-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_14/width_16k/average_l0_84

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.14.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

 varandra

-0.64

 aloud

-0.62

 regeringen

-0.60

 lærer

-0.59

 alá

-0.59

 aktivitet

-0.58

Auk

-0.58

췄

-0.58

 grunn

-0.57

 decorativos

-0.57

POSITIVE LOGITS

 handled

1.43

 handling

1.37

 handles

1.35

 Handles

1.33

 handle

1.32

handled

1.26

 Handle

1.26

 dealt

1.24

handling

1.20

 Handling

1.19

Activations Density 0.143%

words related to handling and processing tasks or activities

No Comments

No Known Activations