INDEX

Explanations

references to academic articles or papers, particularly involving authors and their affiliations

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

ple

-0.14

Eli

-0.14

atri

-0.13

çĶļ

-0.13

ãģ¾ãģĻ

-0.13

Tas

-0.13

anson

-0.13

 ease

-0.13

 precious

-0.13

atus

-0.13

POSITIVE LOGITS

EW

0.16

-US

0.15

adera

0.15

lint

0.15

åįļå£«

0.15

à¥įà¤¤à¤ķ

0.14

oje

0.14

 diversified

0.14

iyeti

0.14

flater

0.14

Activations Density 0.342%

references to academic articles or papers, particularly involving authors and their affiliations

No Comments

No Known Activations