INDEX

Explanations

phrases associated with navigation and giving directions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 9-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L9R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.9.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

okes

-0.15

illa

-0.15

imate

-0.14

ield

-0.14

_notifier

-0.14

Ð¸Ð»Ð¾

-0.14

anz

-0.14

berger

-0.14

-0.13

deen

-0.13

POSITIVE LOGITS

 Directions

0.17

ulings

0.17

 directions

0.17

orgia

0.16

ursos

0.16

Directions

0.15

hazi

0.15

uria

0.14

pl

0.14

 Hospitality

0.14

Activations Density 0.103%

phrases associated with navigation and giving directions

No Comments

No Known Activations