INDEX

Explanations

phrases indicating conditions or considerations in various contexts, particularly focused on timing and planning

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

that come after "when"

np_acts-logits-general · gemini-2.0-flash

when picking, hiring, planning, choosing

np_acts-logits-general · gemini-2.5-flash-lite

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 20-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_20/width_16k/average_l0_71

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.20.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



-0.99

NameInMap

-0.81

 ivelany

-0.81

 jScrollPane

-0.79

DockStyle

-0.78

EndGlobalSection

-0.78

Enllaces

-0.77

adaptiveStyles

-0.76

ArrowToggle

-0.74

CodedInputStream

-0.72

POSITIVE LOGITS

 dealing

1.04

 planning

0.95

 considering

0.91

 deciding

0.89

 designing

0.86

 thinking

0.78

 evaluating

0.78

 contemplating

0.78

 choosing

0.78

 determining

0.78

Activations Density 0.181%

phrases indicating conditions or considerations in various contexts, particularly focused on timing and planning

that come after "when"

when picking, hiring, planning, choosing

No Comments

No Known Activations