INDEX

Explanations

Various data formats including currency codes, programming code snippets, and encoded text strings, often containing the token "0" preceded by non-alphanumeric characters.

eleuther_acts_top20 · claude-3-5-sonnet-20240620 Triggered by @myinzi123

currency codes and symbols.

oai_token-act-pair · claude-3-5-sonnet-20240620 Triggered by @myinzi123

patterns or sequences of numerical values and their relationships

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-16k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_16k/average_l0_91

Prompts (Dashboard)

24,576 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

bootstrapcdn

-0.86

 nahilalakip

-0.75

 ब्रेकडाउन

-0.75

 Signalez

-0.72

󠁣

-0.69

afficheront

-0.67

writeFieldEnd

-0.63

OGND

-0.63

 Infórmanos

-0.63

FormTagHelper

-0.62

POSITIVE LOGITS

1.13

 cero

0.45

vč

0.44

 precisione

0.43

 coscienza

0.41

 extranjera

0.41

detectChanges

0.35

 supérieurs

0.35

 nowo

0.35

 gewöhnlich

0.34

Activations Density 0.044%

Various data formats including currency codes, programming code snippets, and encoded text strings, often containing the token "0" preceded by non-alphanumeric characters.

currency codes and symbols.

patterns or sequences of numerical values and their relationships

No Comments

No Known Activations