INDEX

Explanations

references to scientific studies and data tables

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 14-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L14R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.14.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

apos

-0.15

xDA

-0.15

afa

-0.15

hang

-0.14

](

-0.14

heim

-0.14

fo

-0.14

rego

-0.13

*</

-0.13

POSITIVE LOGITS

longleftrightarrow

0.16

.gb

0.16

 nháºŃp

0.15

ÏħÎ³

0.15

ourg

0.15

UPC

0.14

KF

0.14

eldorf

0.14

Clr

0.14

Bil

0.13

Activations Density 0.063%

references to scientific studies and data tables

No Comments

No Known Activations