INDEX

Explanations

references to major cities or significant landmarks

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

ÙĪÙģÙĬ

-0.14

ouis

-0.13

 domestically

-0.13

raith

-0.13

elves

-0.13

Ã¦r

-0.13

 Domestic

-0.13

Å¾i

-0.13

.Dom

-0.13

 nationwide

-0.13

POSITIVE LOGITS

 world

0.74

ä¸ĸçķĮ

0.61

world

0.58

-world

0.57

 World

0.55

_world

0.55

 WORLD

0.53

World

0.52

 mundo

0.50

 ä¸ĸçķĮ

0.50

Activations Density 0.236%

references to major cities or significant landmarks

No Comments

No Known Activations