INDEX

Explanations

instances of the word "attached" and related forms indicating connection or attachment

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 15-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_15/width_16k/average_l0_78

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.15.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

 balles

-0.57

CodedInputStream

-0.54

 colspan

-0.51

thée

-0.50

addGap

-0.49

🍚

-0.48

ns

-0.46

 Commissioners

-0.45

Eun

-0.45

 Gout

-0.44

POSITIVE LOGITS

 attaching

1.47

 attachment

1.44

 Attach

1.41

 attaches

1.41

 attach

1.39

 attachments

1.39

 attached

1.35

 ATTACH

1.33

 Attachment

1.33

attached

1.27

Activations Density 0.326%

instances of the word "attached" and related forms indicating connection or attachment

No Comments

No Known Activations