INDEX

Explanations

references to locations, specifically places with historical or cultural significance

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 24-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L24R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.24.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

cerebras/SlimPajama-627B

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Cities

-0.23

 cities

-0.17

Cities

-0.17

ê´ĳìĹŃìĭľ

-0.16

.metro

-0.16

usa

-0.16

dob

-0.15

 Universities

-0.15

cities

-0.15

 City

-0.14

POSITIVE LOGITS

 village

0.27

 town

0.25

-on

0.22

 villagers

0.20

 Town

0.20

 Grammar

0.19

 Village

0.19

 parish

0.19

çĶº

0.19

pton

0.18

Activations Density 0.077%

references to locations, specifically places with historical or cultural significance

No Comments

No Known Activations