INDEX

Explanations

references to academic publications and their details

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 21-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L21R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.21.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

baum

-0.15

 yesterday

-0.15

hn

-0.15

ÑıÐ¶

-0.14

 kategor

-0.14

uby

-0.14

ech

-0.14

 till

-0.14

ado

-0.14

orf

-0.13

POSITIVE LOGITS

ãĥŃãĥ¼

0.18

 Winter

0.17

 issues

0.16

iku

0.15

 issue

0.15

ì¡±

0.15

åŃ£

0.15

igel

0.15

 dedic

0.14

entes

0.14

Activations Density 0.072%

references to academic publications and their details

No Comments

No Known Activations