INDEX

Explanations

instances of the word "lap" and its variations, indicating a focus on discussions related to lap-related activities or contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 26-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L26R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.26.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

onal

-0.18

anol

-0.17

------+------+

-0.15

fait

-0.15

ampton

-0.15

æĿ¯

-0.15

IEL

-0.15

ÃŃo

-0.14

PerPixel

-0.14

erior

-0.14

POSITIVE LOGITS

lace

0.25

orte

0.24

Lap

0.23

indrome

0.21

lap

0.21

LAND

0.19

land

0.18

lac

0.18

ierre

0.18

ar

0.18

Activations Density 0.007%

instances of the word "lap" and its variations, indicating a focus on discussions related to lap-related activities or contexts

No Comments

No Known Activations