INDEX

Explanations

phrases related to gaining insights or valuable information

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

AccessorTable

-0.67

WebElementEntity

-0.63

 referenties

-0.62

 rând

-0.58

TagMode

-0.57

ároz

-0.56

PRWEB

-0.56

 Numerade

-0.55

fony

-0.55

newtheorem

-0.54

POSITIVE LOGITS

 into

2.58

into

1.85

 Into

1.81

 INTO

1.80

Into

1.70

INTO

1.50

 vào

0.94

 insight

0.92

intos

0.84

 onto

0.77

Activations Density 0.159%

phrases related to gaining insights or valuable information

No Comments

No Known Activations