INDEX

Explanations

references to numerical data and awards in academic contexts

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

vail

-0.17

usan

-0.16

.scalablytyped

-0.15

emez

-0.15

ibri

-0.14

vou

-0.14

loub

-0.14

inn

-0.14

ion

-0.13

ragen

-0.13

POSITIVE LOGITS

0.16

 cont

0.15

ANNEL

0.15

é¥

0.15

Å¡ek

0.15

WithValue

0.14

aises

0.14

ÑģÑıÑĤ

0.14

 Schl

0.13

 trough

0.13

Activations Density 0.001%

references to numerical data and awards in academic contexts

No Comments

No Known Activations