INDEX

Explanations

expressions of possibility or uncertainty regarding events or beliefs

oai_token-act-pair · gpt-4o-mini

Triggered by @bot

the auxiliary verb "to" following modal or semi-modal verbs like "would," "may," "appear," "seem," and "believed."

oai_token-act-pair · claude-4-5-haiku

Triggered by @emiglarou

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 31-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_31/width_16k/average_l0_114

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.31.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

IFrame

Link

Negative Logits

DockStyle

-0.74

исленность

-0.60

webElementXpaths

-0.54

AnchorStyles

-0.53

ValueStyle

-0.50

 InputDecoration

-0.48

Хьажоргаш

-0.47

uesia

-0.47

WebMethod

-0.47

 FetchType

-0.46

POSITIVE LOGITS

be

1.39

 have

1.02

 belong

0.89

 быть

0.82

 being

0.80

 been

0.75

 être

0.73

 essere

0.71

 være

0.71

 haberse

0.69

Activations Density 0.759%

expressions of possibility or uncertainty regarding events or beliefs

the auxiliary verb "to" following modal or semi-modal verbs like "would," "may," "appear," "seem," and "believed."

No Known Activations

expressions of possibility or uncertainty regarding events or beliefs

the auxiliary verb "to" following modal or semi-modal verbs like "would," "may," "appear," "seem," and "believed."

No Known Activations