INDEX

Explanations

journal and publication-related metadata elements

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

ennen

-0.16

 Gomez

-0.16

orias

-0.15

coli

-0.15

 Reporter

-0.14

 Rapid

-0.14

.spi

-0.14

ÐµÐµ

-0.14

urch

-0.14

Prem

-0.14

POSITIVE LOGITS

usch

0.15

inos

0.14

lem

0.14

âĶ´

0.14

ober

0.13

-background

0.13

 kenn

0.13

ÏĦÎ¹

0.13

_submenu

0.13

oved

0.13

Activations Density 0.004%

journal and publication-related metadata elements

No Comments

No Known Activations