INDEX

Explanations

dates and specific time references

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 7-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_7/width_16k/average_l0_69

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.7.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Moos

-0.72

reter

-0.65

edd

-0.63

 Steff

-0.62

 Wilk

-0.60

cdk

-0.60

 verdura

-0.60

 Wiese

-0.59

erv

-0.59

toObject

-0.59

POSITIVE LOGITS

 January

1.47

Jan

1.36

Jan

1.35

JAN

1.34

 JANUARY

1.34

 january

1.34

January

1.34

JAN

1.31

january

1.25

 Januar

1.22

Activations Density 0.056%

dates and specific time references

No Comments

No Known Activations