INDEX

Explanations

references to locations and geographical entities

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 29-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L29R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.29.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

âĢ¦the

-0.17

âĢ¦but

-0.16

âĢ¦it

-0.16

âĢ¦and

-0.15

Úĺ

-0.14

.âĢ¦

-0.14

âĢ¦I

-0.14

 âĢ¦↵

-0.14

Tup

-0.14

 âĢ¦↵↵

-0.13

POSITIVE LOGITS

region

0.17

#Region

0.17

 Ð¾Ð±Ð»Ð°ÑģÑĤÐ¸

0.17

 region

0.17

 counties

0.17

 Counties

0.17

jylland

0.16

 provinces

0.16

vincia

0.15

olland

0.15

Activations Density 0.199%

references to locations and geographical entities

No Comments

No Known Activations