INDEX

Explanations

phrases related to filling information gaps

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

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

parsedMessage

-0.66

rawDesc

-0.62

pushFollow

-0.61

 invokingState

-0.61

 للاسماء

-0.55

 zvuky

-0.55

 DbSet

-0.54

 Roskov

-0.53

 intestines

-0.52

LEncoder

-0.52

POSITIVE LOGITS

gap

2.15

 gaps

2.10

 fill

2.04

 filling

1.95

 Gaps

1.87

 Filling

1.86

gap

1.83

 fills

1.83

 Fill

1.82

Filling

1.82

Activations Density 0.331%

phrases related to filling information gaps

No Comments

No Known Activations