INDEX

Explanations

references to time, specifically focusing on the word "last" and its variations

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

ViewFeatures

-1.21

yczą

-0.93

 Pixie

-0.85

 ModelRenderer

-0.85

 Uniti

-0.85

 Diogo

-0.84

 Mercure

-0.84

Datuak

-0.83



-0.81

 Oakley

-0.81

POSITIVE LOGITS

 last

1.64

 LAST

1.59

last

1.56

 Last

1.55

LAST

1.51

Last

1.51

setLast

1.25

getLast

1.11

lastTime

1.10

 getLast

1.10

Activations Density 0.052%

references to time, specifically focusing on the word "last" and its variations

No Comments

No Known Activations