INDEX

Explanations

phrases related to specific events and timelines

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 18-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_18/width_16k/average_l0_74

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.18.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

 summer

-0.69

 seasonally

-0.67

summer

-0.67

 year

-0.66

 seasonal

-0.65

year

-0.63

 season

-0.63

Year

-0.62

OfYear

-0.60

 Jahres

-0.60

POSITIVE LOGITS

 week

1.04

 Monday

1.02

Monday

0.96

 Tuesday

0.95

 Thursday

0.95

 monday

0.93

 days

0.93

 Friday

0.93

 Saturday

0.91

 WEEK

0.91

Activations Density 0.248%

phrases related to specific events and timelines

No Comments

No Known Activations